Volume 64, Issue 4, October 2020 - Johnson Matthey ...

Johnson Matthey’s international journal of research exploring science and technology in industrial applications

www.technology.matthey.com

Volume 64, Issue 4, October 2020Published by Johnson Matthey

ISSN 2056-5135

© Copyright 2020 Johnson Matthey

Johnson Matthey Technology Review is published by Johnson Matthey Plc.

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License. You may share, copy and redistribute the material in any medium or format for any lawful purpose. You must give appropriate credit to the author and publisher. You may not use the material for commercial purposes without prior permission. You may not distribute modifi ed material without prior permission.

The rights of users under exceptions and limitations, such as fair use and fair dealing, are not aff ected by the CC licenses.



Johnson Matthey’s international journal of research exploring science and technology in industrial applications

Contents Volume 64, Issue 4, October 2020

394 Guest Editorial: Breaking Down Barriers and Borrowing from Biology By Tom Sturgeon

396 Preparation and Evaluation of a Composite Filler Micro-Embedded with Pseudomonas putida for the Biodegradation of Toluene

By Yuxi Yan, Rencheng Zhu and Shunyi Li

407 UnlockingtheFullEvolutionaryPotentialofArtificialMetalloenzymesThrough Direct Metal-Protein Coordination

By George S. Biggs, Oskar James Klein, Sally R. Boss and Paul D. Barker

419 ReductionofBiofilmFormationonCoolingTowerHeatExchangersusing Nano-silica Coating

By Irfan Turetgen

425 A Mini-Review of Shape-Memory Polymer-Based Materials ByMathewJ.HaskewandJohnG.Hardy

443 Application of Chitosan-Encapsulated Orange Oil onto Footwear Insock Leathers

ByBuketYılmazandHüseyinAtaKaravana

452 BacterialCommunityCompositioninProducedWaterofDiyarbakırOilFieldsin Turkey

ByTuğçeTüccar,EsraIlhan-SungurandGerardMuyzer

466 TheBiotechnologicalPotentialsofBacteriaIsolatedfromParsıkCave,Turkey ByBegümÇandiroğluandNihalDoğruözGüngör

480 Antibacterial Potential of Six Lichen Species against Enterococcus durans from Leather Industry

ByDidemBerber,İpekTürkmenoğluandNüzhetCenkSesal

489 TheDestructiveEffectsofExtremelyHalophilicArchaealStrainson Sheepskins, and Proposals for Remedial Curing Processes

By Meral Birbir, Pinar Caglayan and Yasar Birbir

504 JohnsonMattheyHighlights 507 AntibioticandHeavyMetalResistantBacteriaIsolatedfromAegeanSeaWater andSedimentinGüllükBay,Turkey

ByGülşenAltuğ,MineÇardak,PelinSalihaÇiftçiTüretken,SametKalkanandSevanGürün

526 “Nanomaterials and Environmental Biotechnology” AbookreviewbyMartinHayes

529 Biocatalytic Reduction of Activated Cinnamic Acid Derivatives By Samantha Staniland, Tommaso Angelini, Ahir Pushpanath, Amin Bornadel, ElinaSiirola,SerenaBisagni,AntonioZanotti-GerosaandBeatrizDomínguez


https://doi.org/10.1595/205651320X15954136194837 Johnson Matthey Technol. Rev., 2020, 64, (4), 394–395

394 © 2020 Johnson Matthey

Introduction

As humans, we seem to desire structure, relationships and laws to understand the universe. Through increased understanding, we can solve the problems and challenges that we perceive. This method and the output are given the label of science. At its best, science provides exquisite understanding, life-changing solutions or sometimes both. The downside of the structures and rules we

impose is that they can create inertia. Because the structure or rule served a purpose in the past, we can be more willing to stand by it blindly than openly seek the understanding or solutions we truly desire; a dynamic seen in the natural and social sciences alike and revealing more about human nature than the universe. One such structure is that of the disciplines within science. We should challenge ourselves to be very clear on the purpose of any structures we adhere to and be ready to remove barriers that get in the way of progress. One such example is uncovering the fertile ground of interdisciplinary research. In recent years interdisciplinary research has been of increasing importance across the sciences. Volume 64 of the Johnson Matthey Technology Review started a celebration of interdisciplinary science by looking at when chemistry collaborates with physics (1) and in this issue, we will celebrate the cross-disciplinary contributions of biology with other fields. This wide-ranging issue explores topics such as:

what we can continue to learn from organisms in unusual environments; how we might leverage biology in artificial situations; and even how we manage the interface between human-made, controlled systems and the outside world. In particular, the diversity of industrial applications is striking. Some are familiar to Johnson Matthey and this journal such as fine chemical synthesis, while

others, such as hides and textiles, show that as boundaries within science are removed, previously distant industries will have much to learn from one another.

Themes on Interdisciplinary Science

I have reflected on three themes as this issue has come together. There are numerous examples on each theme and I would challenge the reader to think “what next?” for each:1. Interdisciplinary understanding coming into

biology; for example, computational methods and coding which go hand-in-hand with the biological understanding required for directed evolution of proteins

2. Interdisciplinary understanding coming from biology; for example, improved understanding of biochemical pathways and the relevant biological structures being coupled with synthetic chemistry understanding to allow much more targeted small molecule therapeutics to be designed

3. Platform technologies; for example, clustered regularly interspaced short palindromic repeats and CRISPR-associated protein 9 (CRISPR-Cas9) genome editing where you can custom design the edit while following standardised procedures.

This third theme is perhaps the most important as it turns niche expertise into something accessible to scientists across fields. Understanding the technology may be beneficial but is not a prerequisite to accessing it. Biology follows favourably in the footsteps of computing in producing such platform technologies and it is an attribute that perhaps we should value and prioritise more in other fields. To expand on this theme, it is exciting to look both backwards and forwards to the contributions made possible by platform technologies from the field of

Guest Editorial

Breaking Down Barriers and Borrowing from Biology


https://doi.org/10.1595/205651320X15954136194837 Johnson Matthey Technol. Rev., 2020, 64, (4)

biology. Often these point back to unlocking our understanding of the structure and function of DNA at a molecular level and have resulted in some of the most impactful scientific contributions of the last 50 years or so. Our health has been a significant beneficiary of these advances with cancer drugs providing an illustrative case study. Looking back, we can see recent classes of therapeutics that were significantly enabled by this flow of understanding and platform technologies such as tyrosine kinase inhibitors and antibody-based therapies (2). Most importantly, patient outcomes have improved substantially in part, thanks to these therapies (3). Looking to the future, gene and cell therapies

appear to be following a similar pattern and will hopefully deliver similar patient benefits. Outside of cancer treatments and healthcare, we can see many industries set to benefit from being able to access biological understanding and technologies. This is particularly as we seek to learn from biology and reduce our impact on the planet by using materials and energy in keeping with what Earth can sustain.

Conclusions

As you read through this issue, I hope you enjoy reading something outside of your current field. I would take you back to my earlier challenge and see if you can gain any greater insights by not seeing the separation between your field and those of the authors. Rather, question what you can leverage, what you can learn and what next?

TOM STURGEONImmaterial Ltd, 25 Cambridge Science Park,

Milton Road, Cambridge, CB4 0FW, UKEmail: [email protected]

References

1. A. Smith, Johnson Matthey Technol. Rev., 2020, 64, (2), 101

2. T. A. Baudino, Curr. Drug Discov. Technol., 2015, 12, (1), 3

3. M. Quaresma, M. P. Coleman and B. Rachet, The Lancet, 2015, 385, (9974), 1206




Yuxi YanSchool of Ecology and Environment, Zhengzhou University, Zhengzhou 450001, China; College of Chemical Engineering, Zhengzhou University, Zhengzhou 450001, China

Rencheng Zhu, Shunyi Li*School of Ecology and Environment, Zhengzhou University, Zhengzhou 450001, China

*Email: [email protected]

The main objective of this study was to evaluate the performance of a self-developed filler micro-embedded with Pseudomonas putida (P. putida) for toluene removal in a biofilter under various loading rates. The results show that the biofilter could reach 85% removal efficiency (RE) on the eighth day and remain above 90% RE when the empty bed residence time (EBRT) was 18 s and the inlet loading was not higher than 41.4 g m–3 h–1. Moreover, the biofilter could tolerate substantial transient shock loadings. After two shut-down experiments, the removal efficiency could be restored to above 80% after a recovery period of three days and six days, respectively. Sequence analysis of the 16S rRNA gene of fillers in four operating periods revealed that the highly efficient bacterial colonies in fillers mainly included Firmicutes, Actinobacteria and Proteobacteria and that the abundance of Bacteroidetes increased significantly during the re-start period.

1. Introduction

The massive discharge of volatile organic compounds (VOCs) has a great negative impact

on the environment (1). Toluene is a common pollutant in VOCs and is produced in a large number of industrial activities, such as chemical refining and dye processing. Toluene stimulates skin and mucosa, and when it reaches a certain high concentration, it also causes paralysis of the human nervous system. Compared with photocatalysis and chemical oxidation, using a biofilter to remove VOCs is more economical and environmentally friendly (2). More important is that it does not produce secondary pollution. The key element to ensure the removal capacity of the biofilter is the preparation of the filler. As a carrier for the transfer of pollutants, the filler can provide a suitable growth environment for microorganisms (3). Micro-embedding technology is a method which

uses physical or chemical methods to keep microorganisms in a defined space, ensuring microorganisms with high activity. The principle of using micro-embedding technology to degrade VOCs is to use a hollow porous membrane to intercept microorganisms inside the filler. The pore size of the hollow porous membrane is smaller than that of microbial cells, so that microorganisms can be embedded. The VOCs can enter the interior of the embedded carrier freely due to the small particle size, and the degradation products can flow out of the carrier through the pore size (4, 5).A large number of studies have been carried out

on different types of fillers. Chen et al. (6) used a two-layer biofilter filled with new mixed packing materials to remove hydrogen sulfide gas. Dumont et al. (7) prepared a nutritional slow-release filler (UP20) to biodegrade H2S. In the above studies, the fillers were not embedded with microorganisms. The concentration of microorganisms in the filler was small, and the removal efficiency of the biofilter was low in the start-up phase, resulting in a longer

Preparation and Evaluation of a Composite Filler Micro-Embedded with Pseudomonas putida for the Biodegradation of ToluenePreparation of composite filler with high toluene removal efficiency



start-up period. Zhu et al. (8) used a composite packing material with functional microorganisms to remove H2S. However, toluene does not biodegrade easily due to the presence of a benzene ring. Zuo et al. (9) found that engineered P. putida could simultaneously degrade organophosphates, pyrethroids and carbamates. Muñoz et al. (10) studied the long-term performance and stability of P. putida in a toluene removal bioreactor. The above studies have found that P. putida is highly effective in degrading organics containing benzene rings.However, there is a lack of studies on filler micro-

embedded P. putida for toluene biodegradation. Existing problems with biofilters packed with fillers include bed clogging, low biomass concentration and pressure drops. These problems become more prominent when the biofilter is operated under high VOC loading rates or long-term operation (11). For example, Ryu et al. (12) found that the benzene removal efficiency of a well-designed biofilter decreased from greater than 90% to approximately 75% after 27 days of operation due to clogging caused by the excess growth of biomass.The main objective of this study was to evaluate

the performance of a self-developed filler micro-embedded with P. putida for toluene removal under various inlet loading rates. The variations in start-up period, pressure drop, biomass concentration and tolerance to transient shock loading were monitored throughout the experiments. Special attention was paid to the analysis of the microbial community attached to these fillers and to monitoring the evolution of the microbial community in various periods.

2. Material and Methods

2.1 Preparation of Filler

The composite filler was mainly composed of polyvinyl alcohol, sodium alginate, polypropylene fibre, decomposed plants, calcium carbonate and activated carbon. First, polyvinyl alcohol and sodium alginate, as the embedding and protective agents, were heated, dissolved and cooled to 35°C. Then polypropylene fibre as the skeleton, decomposed plants as nutrients and calcium carbonate as the pH buffer were added into the liquid agent, respectively. Additionally, activated carbon and P. putida BRJC1032 (screening from the activated sludge) were mixed with above agents to increase the physical adsorption capacity and biodegradation capacity of toluene. After that, the mixtures were stirred in a container for 15 min and extruded to spherical particles. Finally, these particles were

cross-linked in boric acid-calcium chloride solution and dried at room temperature for 24 h. Taking the mechanical strength as a single variable factor, the proportions of polyvinyl alcohol, sodium alginate and polypropylene fibre were adjusted to obtain the optimal ratio. After many tests and modifying the design, the optimum proportions of each component of the filler were determined as follows: polyvinyl alcohol accounted for 30%~36%, sodium alginate accounted for 12%~18%, polypropylene fibre accounted for 4%~8%, decomposed plants accounted for 15%~25%, calcium carbonate accounted for 15%~25%, activated carbon accounted for 4%~10% and P. putida accounted for 0.5%~1.5% (13). The schematic pictures of the size and the composition of the composite filler can be seen in Figure S1 and Figure S2 in the Supplementary Information.

2.2 Experimental Setup

The experimental system used in this experiment is shown in Figure 1. Three biofilters were constructed with transparent organic glass pipes. Each biofilter consisted of three modules (each module is 105 mm in inner diameter and 500 mm in height), and all of them were filled with 300 mm composite fillers. A sampling port was set in the top of each module. Toluene gas was prepared by mixing fresh air with pure toluene in a mix chamber, and then introduced into the bottom of each biofilter through the three models in sequence.Three biofilters, namely biofilter 1 (BF1), biofilter

2 (BF2) and biofilter 3 (BF3), were used in this experiment to evaluate the start-up performance. BF1 was packed with the composite filler micro-embedded with P. putida, and both BF2 and BF3 were packed with the sterilised fillers without any microorganisms. However, the nutrient solution used for BF2 at the start-up period was mixed with the P. putida suspension and the microbial concentration of the suspension was the same as that of the P. putida suspension added in the preparation of the composite filler in BF1. Specially, nutrient solution (0.11 K2HPO4, 0.04 KH2PO4, 0.025 NH4Cl, 0.067 MgSO4, 0.036 CaCl2, 0.25 FeCl3, 0.03 MnSO4, 0.04 ZnSO4, 0.03 (NH4)2Mo7O4·4H2O; unit: g l–1; adjusted to pH = 7.0 with NaOH) for microorganism growth was sprayed into the filler bed from the top of three biofilters throughout the experiment. The nutrient solution was intermittently sprayed onto the top of the three biofilters with a spray intensity of 1.5 l h–1 by a peristaltic pump for one hour out of every three hours and the nutrient solution was changed every seven days.



2.3 Toluene Concentration Analysis

The determination of toluene concentration was carried out by adsorption of activated carbon and desorption of carbon disulfide, and then the toluene gas was injected into a gas chromatograph (GC-2014, Shimadzu, Japan) equipped with a packed column (free fatty acid phase (FFAP) capillary column, 30 m × 0.25 mm × 0.25 μm) and a flame ionisation detector (FID). The gas chromatography nitrogen was used as the carrier gas with a flow rate of 1 ml min–1. Temperatures of the injection port, column and detection port were set to 150°C, 65°C and 150°C, respectively. Gas samples were collected from the inlet and outlet of the biofilter with a gas-tight syringe and injected into the GC daily (14). Data were obtained from the workstation by automatic comparison of the peak area of the inlet and outlet samples with the baseline of toluene. The performance of the biofilter was evaluated in terms of (%) RE and the elimination capacity (EC) as a function of toluene loading. The RE and EC were calculated as in Equations (i)–(iii):

Removal efficiency =−( )

×C CCin out

out

100% (i)

Inlet loading =×Q CV

in (ii)

Elimination capability =× −Q C CVin out( ) (iii)

where the Cin and Cout are the inlet and outlet toluene concentration (mg m–3), the V is the

volume of the whole biofilter (l) and Q is the gas flow rate (l min–1).

2.4 Physical and Chemical Property Analysis

The specific surface area and the porosity of the filler were measured by a surface area analyser (Gemini® VII 2390, Micromeritics®, USA). Solid samples were filtered and the pH value of the filtrate was detected using a Bioblock 90431 electrode connected to a C-835 Bioblock multiparameter analyser (Fisher Scientific, France).The mechanical strength of the composite filler was

measured by using a compressive strength-testing instrument (YHKC-2A, Taizhou Yinhe Instrument Plant, China). The pressure drop of the packed bed was measured using a digital pressure gauge (testo 510, Testo SE & Co KGaA, Germany) connecting two ends from the inlet and outlet. The pressure gauge had a measuring range of 0–100 kPa, a resolution of 1 Pa and an accuracy of ±0.3 Pa. The saturated moisture content: some packing

fillers were chosen randomly and immersed into distilled water for 2 h to adsorb as much water as possible. Then the packing fillers were removed and placed in a vacuum oven (DZF6050, Yiheng Scientific Instrument Co Ltd, China) at 105°C for at least 12 h until its weight remained stable. The concentration of microorganisms in the filler

was determined by plate counting. Approximately 10 g fillers were taken out homogeneously from the three modules of the running biofilter, and then put into a conical flask with 90 ml distilled

Fig. 1. Schematic diagram of the experimental set-up

Air pump

Valve

Flowmeter

Toluene

Mixingchamber

Gas pump

Samplingpoints

Gas outlet

Gas inlet Drainage

Peristaltic pump

Circulation cistern



water. After that, the mixture was shaken in a thermostatic shaker bath for 2 h at 25°C to obtain the liquid containing microorganisms. Next, a series of solutions were prepared by different dilution factors (1, 10, 102, 103, 104 and 105 times). Each 0.1 ml solution was taken and inoculated into three types of plate cultures (beef-protein, Rose Bengal medium and Gause’s No.1 medium) for bacteria, fungi and actinomycetes, respectively. The plates were placed in a biochemical incubator (CLIN-250, Tianjin Huabei Experimental Instrument Co Ltd, China) for 2–7 days at 28°C. Finally, the number of microorganism colonies in each plate was counted. Moreover, all the glass vessels used in this experiment were sterilised by using a seating automatic electro-thermal pressure steam steriliser (Model ZDX-35B, Shanghai Medical Instrument Manufactory, China) (15, 16).

2.5 DNA Extraction and Sequencing

Approximately 10 g fillers were randomly sampled from the lowest module of BF1 system at the 25th day, 65th day, 95th day and 145th day. Then the samples were sealed with aluminium foil and frozen at –4°C in a fridge.Microbial DNA was extracted from the above four

samples using the E.Z.N.A.® soil DNA Kit (Omega Bio-tek Inc, USA) according to the manufacturer’s protocols. The final DNA concentration and purification were determined by a NanoDropTM 2000 UV-vis spectrophotometer (Thermo ScientificTM, USA), and DNA quality was checked by 1% agarose gel electrophoresis. Polymerase chain reaction (PCR) was conducted according to the following: 3 min of denaturation at 95°C, 27 cycles

of 30 s at 95°C, 30 s of annealing at 55°C, 45 s of elongation at 72°C and a final extension at 72°C for 10 min. PCR was performed in triplicate in 20 μl mixtures containing 4 μl of 5 × FastPfu Buffer, 2 μl of 2.5 mM deoxyribonucleotide triphosphates (dNTPs), 0.8 μl of each primer (5 μM), 0.4 μl of FastPfu Polymerase and 10 ng of template DNA. The resulting PCR products were extracted from a 2% agarose gel, further purified using the Axygen® AxyPrep DNA Gel Extraction Kit (Corning Inc, USA) and quantified using QuantiFluor®-ST fluorometer (Promega, UK) according to the manufacturer’s protocol (16).Purified amplicons were pooled in equimolar

fashion and paired-end sequenced on a MiSeq platform (Illumina Inc, USA) according to the standard protocols established by Shanghai Majorbio Bio-Pharm Technology Co Ltd (Shanghai, China). The acquired sequences were compared with 16S rRNA gene sequences in the National Center for Biotechnology Information (NCBI) database.

3. Results and Discussion

3.1 Physicochemical Properties of the Filler

Physicochemical properties of the experimental filler used in this study and some other materials from the references are listed in Table I (13). As shown in Table I, the experimental filler is spherical with a diameter of approximately 10 mm. The bulk density of the experimental filler is approximately 271 kg m–3, similar to that of pine bark, and lighter than most of the reference fillers. The mechanical strength is greater than that of pine bark but

Table I Physicochemical Properties of the Fillers

Filler Size, mm

Bulk density, kg m–3

Mechanical strength, N pH

Saturated moisture content, %

Porosity rate, %

Specific surface area, m2 g–1

Organic matter rate, %

Experimental filler 10 ± 2 271 ± 17 153 ± 5 7.0 55.3 ± 3 13 ± 2 1.32 53 ± 4

Pine barka (17) — 244 — 5.7 56.3 59.9 18.39 98.2

Lava rocka (17) — 591 — 5.9 28.9 65.4 2.77 0.6

UP20 (7) 7 920 — 6.9 47 — — —

Composite filler (8) 12 471 427 10.5 49 38 3.91 —

Slow-release filler (16) 50 164 — 7.9 46.7 88 — —

aDue to the irregular shape of pine bark and lava rock, there is no corresponding size data. Other “—” data show that the author did not determine its physicochemical property data in the corresponding literature



smaller than that of volcanic stone (>500 N) (17). The porosity rate is approximately 13%, which is significantly smaller than other fillers and helps toluene to better contact microorganisms in the filler when entering the biofilter (18, 19). The initial pH of the filler is 7.0 ± 0.2. The specific surface area is approximately 1.3 ± 0.1 m2 g–1, which is similar to that of lava rock and composite filler. Compared with lava rock, UP20 and slow-release filler (7, 8, 16), the saturated moisture content and organic matter rate are higher, which can provide water and nutrients for microorganisms in fillers. In addition, the decomposed plant fibre contained within the filler can provide nutrients for microbial growth during experimental operation (20). The selected microbial source added to the filler was P. putida, and the activated carbon was contained in fillers, which can adsorb toluene quickly, promoting toluene to enter the biofilter. The filler in the biofilter did not appear to have deformation, accumulation or other phenomena after operating approximately 150 days. The results indicated that the fillers had favourable properties as biofilter media, and maintained characteristics under long-term operation.

3.2 Start-up Performance

The removal efficiency of the three biofilters during the start-up period is presented in Figure 2. Three biofilters, operated at low toluene concentrations (100–120 mg m–3) and an EBRT of 35 s, demonstrated different removal performance for toluene at the start-up period. The removal

efficiency of BF1 increased from the initial 40% to 80%, and stabilised between 82% and 85% after the eighth day (21, 22). The removal efficiency of BF2 showed a downward trend in the first few days and then rose to approximately 85% at the 14th day. The removal efficiency of BF3 gradually declined from the beginning, and it decreased to almost zero on the 16th–18th days (14, 23). The results showed that fillers embedded with activated carbon and polypropylene fibres have a certain adsorption capacity. However, the removal efficiency was gradually reduced when the filler reached adsorption saturation, as shown in the BF3 trend line in Figure 2. For the same reason, the BF2 line also showed a downward trend at the beginning. Due to the substantial growth of microorganisms, the subsequent removal efficiency gradually increased as shown in the BF2 trend line. Compared with BF2, the fillers in BF1 embedded with P. putida showed unique degradation of toluene at the beginning. The filler-embedded microorganisms entered the working state faster than those cultured with the bacterial solution. These results indicated that the biofilter packed with the composite fillers prepared by micro-embedding could be quickly started up and the microorganisms in the biofilter could well utilise toluene as the carbon source (22).

3.3 Continuous Biodegradation Performance

Toluene continuous removal experiments were performed in three phases based on controlling the

Fig. 2. Removal performances of BF1 (packed with the fillers micro-embedded with P. putida), BF2 and BF3 during the start-up period

100

80

60

Rem

oval

effic

ienc

y, %

40

20

0 2 4 6 8

Time, d

10

BF3BF1BF2

12 14 16 18



EBRT of BF1 to 35 s (Phase 1, day 10 to day 49), 18 s (Phase 2, day 50 to day 80) and 12 s (Phase 3, day 81 to day 110). The results of these experimental stages (Figure 3) are described below. Initially, the biofilter was operated at a low loading rate of toluene (10.5 g m–3 h–1) corresponding to a low inlet concentration (100–120 mg m–3) and high EBRT (35 s) to facilitate proper microbial growth and establish steady-state conditions (8, 23). Steady state was achieved on the 10th day of operation, which was evident from the constant value of the removal efficiency (83%). On the 18th day, the inlet concentration increased to 200 mg m–3, the removal efficiency was almost stable at 88% after a slight decrease. On the 28th day, the inlet concentration increased to 400 mg m–3, and the removal efficiency dropped rapidly to 72% and finally stabilised at 90% after five days of continuous operation. However, when the inlet concentration was controlled at 800 mg m–3, the removal efficiency did not reach a correspondingly high state (less than 80%). In Phase 1, the initial rapid increase within 90% of RE may be due to some extent to competition among microorganisms in the filter unit (14, 21, 23). Again, in Phase 2, the inlet loading rate was

increased and maintained at 81.2 g m–3 h–1 with a corresponding EBRT of 18 s, and the toluene inlet concentration varied between 100 mg m–3 and 400 mg m–3. The removal efficiency reached a maximum when the inlet loading rate was less than 41.4 g m–3 h–1 and was stable above 90%. However, the removal efficiency was only slightly decreased and then stabilised close to 86% at the end of this phase. This result might be attributed to the decrease in residence time of toluene in the

biofilter. At a higher flow rate, the contact time between the toluene and the microorganisms in the fillers was shortened and that resulted in deterioration of the biodegradation ability of the filter bed, leading to lower removal efficiency (24). Similarly, in Phase 3, the toluene inlet concentration increased from 100 mg m–3 to 400 mg m–3, and the intake load increased to 123.3 g m–3 h–1 with a corresponding EBRT of 12 s. During this phase, the removal efficiency of toluene gradually decreased to 80%, and no significant improvement in removal efficiency was observed (17, 22). Elimination capacity, another important indicator of

the biofilter, was also used to assess the ability of the biofilter in terms of toluene removal. Figure 4 demonstrates the relationship of elimination capacity upon the inlet loading. It could be seen from Figure 4 that the elimination capacity presented a slow increase with the increase of inlet loading rates. The maximum elimination capacity of the biofilter was 101 g m–3 h–1, which is better than other typical biofilters. For example, Zhu et al. (10) used composite packing materials to remove H2S and observed a maximum elimination capacity of 65 g m–3 h–1. Liu et al. (18) reported compost-based biofilter with a maximum elimination capacity of 50 g m–3 h–1 for toluene.The concentration of toluene in the nutrient

solution was 0.3 ± 0.1 g l–1 (the saturated solubility of toluene in water was 0.5 ± 0.1 g l–1). This may be due to the short contact time between toluene and the nutrient solution. In addition, part of the toluene dissolved in the nutrient solution was utilised by the filler with circulation of the nutrient solution.

1000

800

600

400

200

0

20

40

60Inlet concentrationOutlet concentrationRemoval efficiency

Removal efficiency, %

80

EBRT = 35s EBRT = 18s EBRT = 12s100

0 10 20 30 40 50Time, d

Tolu

ene

conc

entr

atio

n, m

g m

–3

60 70 80 90 100 110

Fig. 3. Time course of the inlet and outlet concentration and the removal efficiency of BF1



The above results showed that a sudden increase in the inlet loading will cause the removal rate to decrease within a certain period of time. As the experiment proceeds, the system will gradually return to a higher removal rate. When the microorganisms grew under suitable conditions, the recovery ability of the system also increased. However, when the inlet loading rate was too high, the degradation ability of the microorganisms was exceeded, resulting in a relatively low removal rate. After entering the biofilter, toluene is first adsorbed by activated carbon and biofilms in the filler, and

then biodegraded by microorganisms in the filler. A certain amount of toluene will be dissolved in the nutrient solution, but with the circulation of the nutrient solution, part of the toluene will be degraded by the microorganisms in the filler again.

3.4 Tolerance for Transient Shock Loading

To test the ability of the biofilter to resist sharp load change, two interference-shutdown experiments were operated after running for 114 days.

Fig. 4. Toluene elimination capacity of BF1 versus the inlet loading

140

140

120

120

100

100

80

y = 07989x + 2.9602R2 = 0.995

60

Inlet loading, g m–3 h–1

Elim

inat

ion

capa

bilit

y, g

m–3

h–1

40

40 60 80

20

200

Fig. 5. Performance evaluation during shutdown and restart periods of BF1 under transient shock loading

1000

800

600

400

200

0

20

40

60

Inlet concentrationOutlet concentrationRemoval efficiency

Removal efficiency, %

80

7 days3 days 100

1100

115 120 125 130 135

Time, d

Tolu

ene

conc

entr

atio

n, m

g m

–3

140 145



Figure 5 shows the performance evaluation during shutdown and restart periods of BF1 under transient shock loading. When the inlet toluene concentration decreased from 400 mg m–3 to 200 mg m–3, the removal efficiency increased to 90%. Then, the biofilter was subjected to a three-day shutdown experiment and the removal efficiency was restored to 81.2% after running three days. Compared with the shutdown experiments of Singh and Wang (22, 23), the interrupt experiment in this study better reflects the change of flow in actual operation. In the second experiment, when the inlet toluene concentration increased from 400 mg m–3 to 800 mg m–3, the removal efficiency decreased drastically to 62%, and time for the RE to reach at 80.9% was only six days after seven days of shutdown operation. This result clearly indicates that a certain amount of toluene absorbed in activated carbon was supplied to the microorganisms during the shutdown operation of the system, and the microbial activity was maintained; in addition, the decomposed plant fibres also provided a carbon source for the microorganisms, as found by Jorge and Livington (25).

3.5 Biomass Concentration and Pressure Drop in the Biofilter

The attached growth biomass concentration and pressure inside the device were measured during 1–60 days in the biofilter, as shown in Figure 6. The pressure drop increased more obviously from 56 Pa to 373 Pa. The biomass concentration in the biofilter gradually increased from 5 × 104 colony

forming units (CFU) g–1 (the filler was placed in the refrigerator for 1 month, and the biomass concentration was reduced to 5 × 104 CFU g–1) to 4 × 108 CFU g–1 on the 60th day, which was consistent with the trend in the pressure drop (24, 26). The above result indicates that the increase in system pressure drop was mainly due to the rapid growth in microbial biofilm formation and inlet loading rates. The efficient growth and reproduction of microbial biomass played an important role in the efficient operation of the system and the growth of the microorganisms affected the pressure drop across the packed bed and the ease with which the packed bed was clogged. Low biomass reduces the removal efficiency. In contrast, excess biomass reduces the space required for gas and liquid to pass through the biofilter, which leads to an increase in the system pressure drop (27). Although the biomass concentrations in the biofilter increased and the porosity of the system was reduced, this process did not cause blockage of the system and had no significant effect on the removal performance.

4. Bacterial Community Analysis

To explore the bacterial communities in the biomass attached to BF1, genetic sequencing analyses were carried out. Sequencing of 16S rRNA genes amplified from the active bacterial communities during the operational stages revealed 21 phyla, 41 classes, 96 orders, 184 families and 347 genera (28, 29). The community analysis at phylum level of the fillers is shown in Figure 7. The four operational stages were sampled at the

500

400

300

200Pres

sure

, Pa

Biom

ass concentration, CFU

g–1

100

0 10 20 30

Time, d

PressureBiomass concentration

40 50 60104

105

106

107

108

109

1010 Fig. 6. Biomass concentration and pressure drop changes in BF1 during the first 60 days



25th day, the 65th day, the 95th day and the 145th day, where the 25th day, the 65th day and the 95th day had a different EBRT and the same inlet toluene concentration, and the 145th day was after two interference-shutdown experiments. The dominant phyla were Firmicutes (63.4 ± 8.7%), followed by Actinobacteria (14.6 ± 3.9%) and Proteobacteria (10.1 ± 4.2%). With decreased EBRT, the abundance of Firmicutes remained high, but the abundance of Actinobacteria decreased, and the abundance of Proteobacteria increased. This is mainly due to a reduction in residence time leading to the inability of microorganisms to fully utilise toluene, and a reduction in the carbon source leading to a change in the proportion of microorganisms (30). After two interference-shutdown experiments, the abundance of Bacteroidetes increased and the normal microecological balance was broken, which indicated that Bacteroidetes is a sensitive biological indicator, similar to the results found by Wolińska (31). Using this indicator (the increase in Bacteroidetes), it can be judged whether the biofilter is in an unstable state, which would provide some guidance for practical engineering applications.In the four operational periods, few Pseudomonas

(abundance less than 1%, as shown in Figure S3) were found in the sampling of the above four periods. As the inlet loading rate increased, the abundance

of Pseudomonas genus increased from 4.7 × 10–4 to 1.9 × 10–3. After two intervention-shutdown experiments, the abundance of Pseudomonas genus decreased to 8.5 × 10–5, which indicates that the biofilter was not in a sterile environment and that there are other microorganisms competing with the P. putida added to the filler. When the environmental conditions and the nutrients in the biofilter became unsuitable for the added microorganisms and were suitable for other microorganisms, the other microorganisms were activated and enriched (32). However, in the start-up phase, the biofilter embedded with P. putida started quickly, and the removal efficiency of toluene remained high, which indicated that the added P. putida contributed to the efficient operation of the biofilter (33). These results indicated that the biomass could maintain itself by microbial community changes, and the rapid re-adaptation of the biofilter could contribute to the activity retention of its biomass during the starvation period.

5. Conclusions

A composite filler micro-embedded with P. putida was prepared and evaluated for the biodegradation of toluene. The biofilter packed with the fillers could start up quickly with 85% RE on the eighth day, and tolerate substantial transient shock loadings. The RE of the biofilter remained above 90% when

Community barplot analysis

Day_25

Day_65

Day_95

Sam

ples

Day_145

0 0.2 0.4 0.6 0.8 1Percent of community abundance on Phylum level

FirmicutesActinobacteriaProteobacteriaBacteroidetesChloroflexiGemmatimonadetesOthers

Fig. 7. Bacterial community analysis of the fillers sampled at the 25th day, the 65th day, the 95th day and the 145th day in BF1



the EBRT was 18 s and the intake load was not higher than 41.4 g m–3 h–1. In the experimental period of 145 days, no filter plugging phenomenon was observed. Moreover, the high removal efficiency and elimination capacity contributed to rich bacterial communities for the efficient biodegradation of toluene. The communities mainly included Firmicutes, Actinobacteria and Proteobacteria, and the abundance of Bacteroidetes increased significantly during the recovery period. The composite filler exhibited favourable physicochemical properties in this experiment and its practicability in industrial engineering should be further investigated.

Acknowledgments

The authors would like to acknowledge the support of the National Natural Science Foundation of China (No. U1304216), the Science and Technology Plan of He’nan Province, China (No. 122102310366), the University Key Research Project of He’nan Province, China (No. 19A610002 and 19A150010), and the China Postdoctoral Science Foundation (No. 2018M632794).

References

1. M.-C. Delhoménie and M. Heitz, Crit. Rev. Biotechnol., 2005, 25, (1–2), 53

2. R. Underhill, R. J. Lewis, S. J. Freakley, M. Douthwaite, P. J. Miedziak, O. Akdim, J. K. Edwards and G. J. Hutchings, Johnson Matthey Technol. Rev., 2018, 62, (4), 417

3. Y. J. Tham, P. A. Latif, A. M. Abdullah, A. Shamala-Devi and Y. H. Taufiq-Yap, Bioresour. Technol., 2011, 102, (2), 724

4. E. R. Rene, B. T. Mohammad, M. C. Veiga and C. Kennes, Bioresour. Technol., 2012, 116, 204

5. Y. Deng, F. Yang, C. Deng, J. Yang, J. Jia and H. Yuan, Appl. Biochem. Biotechnol., 2017, 183, (3), 893

6. Y. Chen, X. Wang, S. He, S. Zhu and S. Shen, J. Environ. Manage., 2016, 165, 11

7. E. Dumont and Y. Andrès, J. Chem. Technol. Biotechnol., 2010, 85, (3), 429

8. R. Zhu, S. Li, X. Bao and É. Dumont, Sci. Rep., 2017, 7, 42241

9. Z. Zuo, T. Gong, Y. Che, R. Liu, P. Xu, H. Jian, C. Qiao, C. Song and C. Yang, Biodegradation, 2015, 26, (3), 223

10. R. Muñoz, M. Hernández, A. Segura, J. Gouveia, A. Rojas, J. L. Ramos and S. Villaverde, Appl. Microbiol. Biotechnol., 2009, 83, (1), 189

11. J. V. Littlejohns, K. B. McAuley and A. J. Daugulis, J. Hazard. Mater., 2010, 175, (1–3), 872

12. H. W. Ryu, K.-S. Cho and D. J. Chung, Bioresour. Technol., 2010, 101, (6), 1745

13. Y. Nie, R. Zhu, S. Li, S. Li, M. Wang and Y. Yan, Chinese J. Environ. Eng., 2019, 13, (3), 678

14. X. Chen, W. Qian, L. Kong, Y. Xiong and S. Tian, Biochem. Eng. J., 2015, 98, 56

15. W.-F. Yang, H.-J. Hsing, Y.-C. Yang and J.-Y. Shyng, J. Hazard. Mater., 2007, 148, (3), 653

16. R. Zhu, S. Li, Z. Wu and É. Dumont, Environ. Technol., 2017, 38, (8), 945

17. Y. Luo, S. Li, H. Ma and Y. Wang, Trans. Chinese Soc. Agric. Eng., 2017, 33, (12), 218 (in Chinese)

18. Y. Liu, X. Quan, Y. Sun, J. Chen, D. Xue and J. S. Chung, J. Hazard. Mater., 2002, 95, (1–2), 199

19. R. Logares, S. Sunagawa, G. Salazar, F. M. Cornejo-Castillo, I. Ferrera, H. Sarmento, P. Hingamp, H. Ogata, C. de Vargas, G. Lima-Mendez, J. Raes, J. Poulain, O. Jaillon, P. Wincker, S. Kandels-Lewis, E. Karsenti, P. Bork and S. G. Acinas, Environ. Microbiol., 2014, 16, (9), 2659

20. J. Zhang, L. Li and J. Liu, Biochem. Eng. J., 2017, 118, 105

21. Q. Hu, C. Wang and K. Huang, Chem. Eng. J., 2015, 279, 689

22. K. Singh, B. S. Giri, A. Sahi, S. R. Geed, M. K. Kureel, S. Singh, S. K. Dubey, B. N. Rai, S. Kumar, S. N. Upadhyay and R. S. Singh, Bioresour. Technol., 2017, 242, 351

23. M. Wang, S. Xu, S. Li and R. Zhu, J. Ind. Eng. Chem., 2019, 75, 224

24. Y. Ding, W. Wu, Z. Han and Y. Chen, Biochem. Eng. J., 2008, 38, (2), 248

25. F. Abbasian, R. Lockington, M. Megharaj and R. Naidu, Appl. Biochem. Biotechnol., 2016, 178, (2), 224

26. J. Song and K. A. Kinney, Biotechnol. Bioeng., 2000, 68, (5), 508

27. Y. Hajizadeh, M.-M. Amin and I. Parseh, J. Ind. Eng. Chem., 2018, 62, 418

28. H. Li, S. Huang, Z. Wei, P. Chen and Y. Zhang, Sci. Total Environ., 2016, 562, 533

29. H. Liu, S.-J. Wang, J.-J. Zhang, H. Dai, H. Tang and N.-Y. Zhou, Appl. Environ. Microbiol., 2011, 77, (13), 4547



30. L. Bergdoll, E. Point, F. Bayman and D. Picot, Biochim. Biophys. Acta., 2012, 1817, S138

31. A. Wolińska, A. Kuźniar, U. Zielenkiewicz, D. Izak, A. Szafranek-Nakonieczna, A. Banach and M. Błaszczyk, Appl. Soil Ecol., 2017, 119, 128

32. S. R. Geed, M. K. Kureel, A. K. Shukla, R. S. Singh and B. N. Rai, Resour. Eff. Technol., 2016, 2, (1), S3

33. M. Kumar, B. S. Giri, K.-H. Kim, R. P. Singh, E. R. Rene, M. E. López, B. N. Rai, H. Singh, D. Prasad and R. S. Singh, Bioresour. Technol., 2019, 285, 121317

The Authors

Yuxi Yan received a bachelor’s degree from Zhengzhou University, China, in 2018 and is currently studying for a master’s degree at Zhengzhou University. His research interests include the biodegradation of VOCs.

Rencheng Zhu received his PhD from Nanjing University of Aeronautics and Astronautics, China, in 2017 and currently serves as an associate professor at Zhengzhou University. His research interests include the governance of VOCs and the characteristics of automobile exhaust emissions.

Shunyi Li received his PhD from Sun Yat-sen University, China, in 2005. He is currently an executive director of the Henan Environmental Protection Federation, China, and a professor at Zhengzhou University. His research interests include the management of VOCs and the management of odorous gases.




George S. Biggs, Oskar James Klein, Sally R. Boss*, Paul D. Barker**Department of Chemistry, University of Cambridge, Lensfield Road, Cambridge, CB2 1EW, UK

Email: *[email protected]; **[email protected]

Generation of artificial metalloenzymes (ArMs) has gained much inspiration from the general understanding of natural metalloenzymes. Over the last decade, a multitude of methods generating transition metal-protein hybrids have been developed and many of these new-to-nature constructs catalyse reactions previously reserved for the realm of synthetic chemistry. This perspective will focus on ArMs incorporating 4d and 5d transition metals. It aims to summarise the significant advances made to date and asks whether there are chemical strategies, used in nature to optimise metal catalysts, that have yet to be fully recognised in the synthetic enzyme world, particularly whether artificial enzymes produced to date fully take advantage of the structural and energetic context provided by the protein. Further, the argument is put forward that, based on precedence, in the majority of naturally evolved metalloenzymes the direct coordination bonding between the metal and the protein scaffold is integral to catalysis. Therefore, the protein can attenuate metal activity by positioning ligand atoms in the form of amino acids, as well as making non-covalent contributions to catalysis, through intermolecular interactions that pre-organise substrates and

stabilise transition states. This highlights the often neglected but crucial element of natural systems that is the energetic contribution towards activating metal centres through protein fold energy. Finally, general principles needed for a different approach to the formation of ArMs are set out, utilising direct coordination inspired by the activation of an organometallic cofactor upon protein binding. This methodology, observed in nature, delivers true interdependence between metal and protein. When combined with the ability to efficiently evolve enzymes, new problems in catalysis could be addressed in a faster and more specific manner than with simpler small molecule catalysts.

1. Introduction

Metalloenzymes have been prominent in the field of enzyme engineering since its emergence some 40 years ago, at the birth of protein and enzyme engineering (1, 2). Metal ions or cofactors in solution have an intrinsic chemistry that can be catalytic and these are accessible to detailed mechanistic study. These properties mean that co-localisation of substrate and metal within a peptidic scaffold can be sufficient in forming an ArM, without further influence from the protein on the catalytic mechanism. With the advent of modern protein engineering and design technologies, ArMs were developed by incorporating metal binding sites in or adjacent to hydrophobic pockets. While the resulting ArMs were active, they often displayed low efficiency and specificity. Therefore, directed evolution (i.e. iterative rounds of mutagenesis and selection for activity, Figure 1) has become a key step in

Unlocking the Full Evolutionary Potential of Artificial Metalloenzymes Through Direct Metal-Protein CoordinationA review of recent advances for catalyst development



creating enzymes with new and useful properties. The choice of starting point for such a forced evolution campaign, in this case the metal-protein complex formed initially, is of great importance. Since any particular enzyme follows a unique evolutionary trajectory as new mutations move it along the fitness landscape towards (potentially local) maxima, choice of the starting point may directly predetermine the result. By nature of the selection process, it is further possible, that trajectories leading to the global maximum fitness fall beneath the cut-off limit for further evolution, becoming inaccessible. For instance, a mutation introduced in the first round of mutagenesis may lead to a destabilisation of the protein at assay conditions, causing that initial variant to be discarded through selection. However, a compensating mutation to that variant in a subsequent round of mutagenesis could result in an enzyme which is stable, active and closer to a global fitness maximum. Finally, not every method of generating ArMs may be compatible with current methods for directed evolution and therefore limit the extent of evolution that can be achieved.In this perspective, different routes towards

ArMs are considered in the context of the starting

protein scaffold as well as the type of catalytic centre and reactions involved. Advances in ArMs have recently been reviewed and the reader is referred to these for further details of the strategies used to find new systems (3–5). This article aims to provide an overview of the strengths and weakness of these different approaches and to provide a perspective of some challenges that remain.

2. Why Do We Want New Artificial Metalloenzymes?

One particular area that will greatly impact chemical production on this planet is synthetic biology. Replacing synthetic catalysts, acting on petrochemical feedstocks in non-aqueous solvents, with biocatalytic systems working in water with simple carbon neutral feedstocks (carbon dioxide even?) is clearly highly desirable. But why engineer new enzymes, particularly using expensive and relatively scarce transition metals, when the ability to find new catalysts amongst gene products from all corners of the biological world has developed at staggering pace (6–8)? As a consequence of the latter, any target chemical can conceivably be obtained by recombining pre-existing metabolic

Activity assay

Selection and DNA recovery

Mutagenesis

Gene library

Expression

Metal modification

Fig. 1. The general overview of a directed evolution campaign for ArMs. The Darwinian algorithm can be reproduced in the laboratory, greatly increasing the speed of evolution. Mutagenesis methods introduce mutations with various levels of randomness, depending on the method used, to the starting point gene, forming a gene library. This library can then be expressed in a manner that couples expression products and genetic sequence information to yield the different proteins. Upon addition of the metal cofactor, the ArMs are formed and can be selected for improved variants in regard to desired parameters (reaction rates, yield, stereoselectivity, stability). The metal modification step must itself clearly be efficient and high yielding to avoid limiting the library size at that stage. The sequence information of the improved candidates is recovered and can be subjected to further rounds of directed evolution



pathways (9). What will new and unnatural metalloenzymes provide?One clear feature is orthogonality: the objective

of introducing functionality into a cell that has no counterpart in the natural world could provide chemistry that biology cannot currently catalyse, alkene metathesis for example. As there is a limit to the number of additional transformations a viable cell will perform, these orthogonal reactions may allow access to much shorter, and therefore more efficient, pathways. If not for a synthetic purpose, one could also imagine orthogonal catalytic chemistry providing a diagnostic or reporter output without interference from the host endogenous processes. For it to be truly orthogonal, it is difficult to imagine evolving a new enzyme based around metals already abundant in nature and already used as catalysts in biology. The transition metals used by nature are very carefully controlled by acquisition and regulatory networks that ensure catalytic metal ions are not free to operate outside the endogenous metabolism. Therefore, there is significant advantage in trying to introduce metals that biology currently has no evolved means of metabolising. This work therefore focuses primarily on non-biological transition metal cofactors as a route to introducing novel orthogonal activity into a biologically viable system.

3. Evolutionary Routes to Optimised Artificial Metalloenzymes

Natural evolution has provided numerous examples of metal ions used by enzymes for a plethora of different catalytic purposes. Rigorous mechanistic and structural biochemistry has advanced understanding of the mechanistic detail of metalloenzyme activity significantly, to the point that a few underpinning principles can be identified, linking protein structure and thermodynamics to catalytic activity of metal centres. Together with the knowledge garnered from extensive research on transition metal catalysts, it is possible to establish key properties desirable for novel ArMs.

3.1 Considerations on Protein-Substrate Interactions

As mentioned above, the ability of enzymes to organise reactants cooperatively can in itself give rise to enhanced activity over background rates in solution and in highly evolved systems this may even be the greatest factor driving increased reaction rates. It is important to realise that while

metal-substrate proximity may be enough to confer reactivity, directional metal-substrate orbital overlap also plays a crucial role in activating the substrate to react. Indeed, it is via the formation of metal ligand, including metal substrate, molecular orbitals that the substrate chemistry is attenuated by the presence of the metal and that catalytic reactivity can be achieved. Significant computational advancements have been made in the in silico design of catalytic metal binding sites (10, 11) and the mechanistic understanding of reported ArMs (12–15). However, given the lack of reliable parameters for defining transition metal bonding, and the immense complexity of the many low energy interactions that determine the coupling of protein folding to the binding of small molecules, it is beyond current computational capabilities to predict what primary sequence and cofactors are necessary to achieve the optimal arrangement for metal catalysis. It therefore becomes important to have a malleable, promiscuous starting system that can be used to sample a large space of different structures (16). Hence, while choosing proteins with well-defined properties and unique structures has some advantages from a design point of view, starting points that do not fold into one specific structure may be desirable, since they are not as closely constrained by any one particular energy well. For similar reasons, in choosing a particular chemical strategy for introducing a metal cofactor into the protein, it becomes essential to use a method that allows for high throughput selection or screening (17).

3.2 Considerations on Metal Chemistry in Proteins

In addition to sampling sequence space to optimise the geometrical factor, protein evolution offers the unique possibility of sampling transition metal chemistry by poising the metal in energised states. In small molecule transition metal catalysis, ligands will arrange around the metal centre to maximise bonding interactions and reach a thermodynamic minimum. In order to maintain the ligand exchange necessary for catalysis, some ligands tend to be weakly bonding, with the presence of strongly bonding ligands (for instance water or hydroxide) being a major factor in catalyst poisoning. In enzymes however, the intramolecular bonds generated within the whole protein scaffold can be used to place and maintain coordinating atoms from amino acids. These interactions can be seen as the second coordination sphere, shaping



the metal complex and potentially leaving the first sphere ligand atoms in a suboptimal configuration around the metal centre so that the energy of the resulting complex is not at a minimum on the coordination energy landscape. The stabilisation of this complex is made possible by the favourable intramolecular peptidic interactions (i.e. protein fold energy) offsetting the steric and electronic distortion of the optimum geometry (18). These energised, or entatic, states have a reactivity that is not easily realised in conventional, synthetic metal catalysts, if it is possible at all (19). This effect is most easily visualised by considering the common biological process of activation of inert cofactors by alteration of coordination upon binding to their respective apoenzymes. For instance, on their own the cobalt metallo-organic cofactor, vitamin B12 and methionine synthase are catalytically inert; upon protein-cofactor binding and coordination of the cobalt centre to a specific histidine, methyl transfer activity is unmasked with great control and substrate specificity (20–22). Applying this principle, it can be envisioned that even with the limited donor atoms available to proteins, a vast number of different complexes with different chemistries can be accessed, because the exact positioning as well as characteristics of the ligands dictate metal properties such as electron density, redox potential, Lewis acidity and ligand exchange rates. Further, the metal cofactor does not need to be a bare metal ion but could be incorporated with other ligands already attached. Interaction between these ligands (for instance π–π stacking with an arene ligand) and the protein can be relayed to the metal centre and allow for an even finer tuning of the metal centre. Again, current possibilities for design are insufficient to predict these effects which can be very subtle, highlighting the need for biochemical high throughput screening methods.

3.3 The Optimal Method of ArM Formation

The above considerations define a range of requirements for potential methods of forming ArMs. Primarily, there needs to be a direct connection between the protein scaffold and the metal ion in the form of at least one coordination bond, not only for localisation but also for poising the metal reactivity. As will be detailed below, most of the successful methods of generating ArMs published to date are efficient but rely on fully saturated, catalytically active cofactors such as commercial transition metal catalysts

decorated with a linker moiety. These cannot make use of the protein fold energy to optimise the chemical process of catalysis, a potential factor in why directed evolution campaigns of ArMs have been of limited success. Whereas improvements in enantioselectivity and turnover number have been reported, which can be traced to substrate binding and the hydrophobic micro-environment respectively (23–25), significant increases in the chemical turnover rate (in many systems characterised by the initial kcat) from the free cofactor to the formed ArM have so far been limited. Small changes in kcat can be explained by organisational effects and indirect interactions with the substrate orbitals, such as charge compensation. As demonstrated by Hilvert et al., significant increases in kcat have been shown to be possible by fine tuning the actual centre of reaction, which is the first coordination sphere of the metal complex (26). From the perspective of the protein scaffold, the formation of an entatic state requires the peptide to be at least partially folded before binding the metal. The more defined the fold, the greater the ability of the fold to energise the metal complex. This is in contrast to the desirable dynamic system for the evolutionary process. A potential compromise can be struck by using a starting scaffold that is partially folded as the apoprotein and upon cofactor binding rigidifies to a completely folded form. The initial folding energy can be used to poise the metal in an activated state, while the folding process occurring during cofactor binding allows for the system to adapt during directed evolution. Once the ArM becomes more specialised after rounds of evolution, the apoprotein will probably approach a more fully folded form, yielding an ArM after cofactor addition that is less promiscuous but contains a more energised and active metal centre.To summarise, the number of different complex

chemical factors required of ArMs demand the use of directed evolution in order to form enzymes with industrially and medically relevant properties. In order to ensure a high level of engineerability, an optimal methodology for combining 4d and 5d metals starts with a highly promiscuous and malleable holoprotein that further has dative bonds between the metal ion and the peptidic moieties. A further point considering the cofactor attachment point is that the cofactor should be in a deep cleft within the protein topology rather than at the surface. This is to allow the protein to maximise substrate binding and secondary transition state stabilising effects, as well as



second sphere interactions influencing the metal complex.

4. Strategies for Generating Artificial Metalloenzymes

ArMs are generated either from the combination of an unnatural transition metal cofactor being introduced into a protein scaffold or a natural metalloprotein being evolved in a laboratory to enhance or alter its natural catalytic reactivity. A detailed review of the field of the directed evolution of natural metalloproteins is out of the scope of this perspective. However, the engineering and evolutionary approaches developed by Frances Arnold and applied to haem metalloproteins (for example, cytochrome P450) are particularly noteworthy and applicable when evolving unnatural metal-protein hybrid catalysts (27–29). Four successful strategies have been employed to localise an unnatural metal to a well-defined location within a protein matrix.

4.1 Metal Ion Substitution in Natural Enzymes

Natural metal cofactors can be found in proteins encapsulated by ligands supplied by the protein or with non-protein ligands also coordinated. This enables two different methods of metal substitution: (a) substituting the metal ion in a protein defined coordination site; or (b) substituting the metal ion in a natural metal-organic cofactor (such as haem) (Figure 2).Many ArMs have been generated by substituting

the catalytic Zn(II) ion located in a His3 binding site of carbonic anhydrase with different metals, for example, Coleman et al. reported esterase activity of a Co(II) substituted carbonic anhydrase (30). Replacement with different Rh(I) species has also

been explored, with catalytic hydrogenation (31) and hydroformylation (32) demonstrated. However, these rhodium metalloenzymes have a much slower activity than commercial small molecule rhodium catalysts alone. Although in these examples it is demonstrated that unnatural metal complexes can coordinate to the natural Zn(II) binding site, relatively low catalytic activity is observed. The highly evolved zinc binding site contains a complex secondary sphere architecture, in order to modulate the Lewis acidity of zinc. The chemically different demands for rhodium catalysed hydrogenation and hydroformylation reactions will therefore not be met in this system. Further, evolution of such a specialised system may be difficult.Hartwig et al. reported taking the metal-

organic cofactor haem and substituting iron for a range of different 4d and 5d metals (including rhodium, ruthenium, iridium and silver) (33). In one particularly comprehensive example, an Ir(Me) porphyrin was incorporated into the cytochrome P450 enzyme CYP119 and catalytic functionalisation of C–H bonds to C–C bonds by carbene insertion was demonstrated, capable of high stereospecificity (25). Evolutionary campaigns on this artificial iridium metalloenzyme generated variants with an impressive 4000-fold increase in catalytic efficiency (defined by the kcat/KM), with kinetic parameters and selectivities matching those of native enzymes. These parameters highlight the potential of this attachment method, and in particular the advantages of introducing exogenous metal cofactors with non-protein ligands remaining coordinated upon ArM formation.In this case, the mutations made to this iridium

CYP119 metalloenzyme have greatly optimised the binding and pre-organisation of the substrate for catalysis, lowering the value for KM, (Figure 3). In this system there is no direct iridium-protein coordination; the iridium metal is coordinatively saturated by four haem nitrogens, one methyl ligand and coordination to the substrate. Therefore, the moderate increase in kcat cannot have come through an electronic (through bond) contribution to catalysis from amino acid side chain ligands and protein fold energy but must arise from other minor contributions as discussed in the previous section. Another limitation of such a system is that it does not allow for the metal to interact with more than one substrate at a time, an essential feature of many interesting organometallic transformations such as metathesis.

M1

M2

M1

M2

Add

Remove

Fig. 2. Schematic representation of metal ion substitution in natural enzymes. The natural cofactor (red) can be substituted with a suitable unnatural cofactor (blue). This may include the bare metal ion or larger cofactors such as haem



M

M M

Bare metal cofactor

20-fold increase in kcat

Protein 4000-fold increase in kcat/KM

WT ArM system kcat = n/a KM ≥ 5 mM

Evolved ArM kcat = 45.8 min–1 KM = 0.17 mM

Fig. 3. Comparisons between the activities of a bare cofactor and ArM before and after directed evolution. The data in this figure are taken from the work of Hartwig et al. (25). This elegant study is a good example of the issues encountered when using fully substituted artificial cofactors, even in highly optimised systems. Whereas directed evolution was able to achieve an impressive 4000-fold increase in kcat/KM, the actual chemical kcat was only moderately enhanced when compared to the cofactor in solution. This can be explained by the enzyme evolving to more strongly bind the substrate and optimise the orientation of the substrate-metal complex. However, as there is no direct metal-protein interface, directed evolution cannot influence the metal chemistry, capping the chemical potential at that observed for the free cofactor in solution

4.2 Supramolecular, Non-Covalent Binding of Tagged ComplexesThere are many specific complexes between proteins and small molecules which are well understood and have very high affinity. ArMs have therefore been generated where a catalytic metal complex has been attached to a small molecule with high affinity for a protein target (Figure 4). This means of localising the new cofactor into a protein scaffold has been widely explored. Building on the work of Wilson and Whitesides in the 1970s (34), Ward and coworkers have assembled ArMs based on the high supramolecular affinity of small molecule biotinylated metal catalysts for the protein streptavidin. As many as 12 different catalytic transformations have been performed by these metal-streptavidin hybrids, including ruthenium-catalysed olefin metathesis (17), ruthenium-catalysed deallylation (35), iridium-catalysed transfer hydrogenation (24) and dirhodium-catalysed cyclopropanation (36), all in vivo.This strategy has also been employed in ArMs

that were reported by Tanaka et al. for potential therapeutic application. In this example, a coumarin derivative tagged with a ruthenium

metathesis catalyst was localised to a hydrophobic binding site in human serum albumin. The metalloenzyme was directed to cancerous tissue (through specific glycosylation) and a pro-drug was administered which upon metathesis induced cellular death (37).One key benefit of supramolecular assembly

is apparent in the examples described above, and that is that the conjugation between metal and protein is robust enough to be performed in complex cellular environments. Furthermore, unlike covalent attachment, supramolecular assembly can be a reversible process, which allows

Fig. 4. Schematic representation of supramolecular, non-covalent binding of tagged complexes. The metal cofactor (red) is localised by non-covalent interaction between a ligand bound recognition group (blue) and the protein

M

M



for component recycling. In a recent report of Duhme-Klair et al. catalytic transfer hydrogenation is demonstrated from a siderophore-protein combination that enables strong but redox-reversible catalyst anchoring (38). All current examples of ArMs generated by supramolecular assembly do, however, rely on the assembly of proteins with known, highly catalytically active metal complexes. As discussed previously, using complexes which maintain their ligand set during ArM formation does not allow the metal complex to be subjected to evolutionary pressures limiting the evolutionary potential.

4.3 Covalent Anchoring Through Metal Ligands

Covalent anchoring relies on using a chemical reaction to covalently link a protein side chain to a strong ligand for a metal (Figure 5). Covalent anchoring methods can be split into two broad categories: (a) modification of a natural amino acid side chain (for example cysteine, lysine or tyrosine), via a nucleophilic–electrophilic reaction and (b) coupling through a genetically encoded unnatural amino acid (UAA).There is a resurgence in research for developing

novel bioconjugation and protein modification techniques of natural amino acids (such as cysteine, lysine or tyrosine) (39, 40). Generating ArMs through cysteine modification is attractive due to the high nucleophilicity and rarity of free cysteines allowing for greater control of reactivity. Salmain and coworkers have modified the free Cys25 in the cysteine protease papain, using a variety of ruthenium, rhenium and rhodium complexes all functionalised with either a maleimide or chloroacetamide group (41–43).The pioneering work of the Schultz laboratory

enabled incorporation of UAAs into protein

scaffolds (44). Since then, the most successful generation of ArMs involving a covalent linkage to an UAA were reported by Lewis et al. and involve a reaction between an alkyne-substituted dirhodium catalyst and a genetically encoded L-4-azidophenylalanine residue through strain-promoted azide-alkyne cycloaddition (SPAAC) (45–47). Hypothetically, UAAs could be encoded into a specific residue of most proteins; here, the protein scaffold selected was a β-barrel prolyl oligopeptidase and the resulting metalloenzymes generated catalysed olefin cyclopropanation.The effectiveness of introducing UAA via stop

codon methodology is that theoretically the same conjugation technology is applicable to many different proteins to generate diverse ArMs through a specific, fast and irreversible covalent conjugation. Beside commonly relying on pre-formed metal complexes, an overarching issue of covalent attachment and supramolecular assembly is that the protein scaffold is used predominantly as an auxiliary providing a chiral and hydrophobic micro-environment. Further, many reported methods utilise a long flexible linker between the point of attachment and the metal complex which could remove the catalytic centre from the very interactions needed for the protein to exert an influence on transition states.

4.4 Direct Activation by Metal Coordination to Protein Side Chains

Dative ArMs have one or more coordination bonds directly from the metal to a Lewis basic amino acid residue (His, Cys, Ser, Glu, Asp) on the protein scaffold (Figure 6). The protein therefore has a direct electronic influence on the reactivity at the metal centre. The active hybrid molecule is formed by substitution reactions from a precursor metal species and the apoprotein. This allows

Fig. 5. Schematic representation of covalent anchoring through metal ligands. The metal cofactor (red) is attached to the protein by a reaction forming a covalent bond, for instance nucleophile (Nu) attacking an electrophile (E)

ME

E

MNu Nu

Fig. 6. Schematic representation of cofactor attachment via direct activation by metal coordination to protein side chains. The free metal cofactor (red) attaches to Lewis basic residues on the protein (LB) via ligand substitution reactions, forming a new metal-protein complex (blue)

M M

LB LB LB LB



for potentially very clean reaction conditions for assembly of the metal-protein complex. Although advances have been made, the complexity of these metal-protein binding processes remain a major challenge for the design of competently folded and catalytically active metalloproteins from scratch. It is important to distinguish between metalloenzymes where coordination to the metal is provided only by amino acid sidechains, substrates and solvents, and those in which the metal brings its own specific ligands with it. The latter, metal cofactors would be artificial versions of commonly encountered natural examples such as haem, vitamin B12 and molybdopterin which are (bio)synthesised separately and bind to the protein through both non-covalent interactions and coordination. As pointed out above, their activity is defined by the other ligands they carry to an active site as well as the coordination by the protein.Degrado and coworkers have pioneered the

design of a number of synthetic proteins which directly coordinate bare metal atoms or metal cofactors (10, 48). For example, in some of the earliest work, the His3-Zn(II) binding motif found in carbonic anhydrase was introduced into a designed four helical bundle protein, and hydrolytic activity was observed (49). More recently, de novo design has been coupled with directed evolutionary approaches to generate an artificial zinc metalloenzyme capable of accelerating ester cleavage with un-paralleled catalytic efficiency (kcat/KM of 106 M–1 s–1) (26).In a range of studies (13, 50–52), Roelfes and

coworkers use amber stop codon technology to introduce the UAA (2,2′-bipyridin-5yl)alanine into a range of protein scaffolds. Upon addition of different bare metal ions, they were able to obtain ArMs catalysing the Friedel-Crafts alkylation of indoles, enantioselective metallohydration and the stabilisation of a semiquinone radical. By the use of sophisticated computational design, the group was able to introduce beneficial point mutations in many of the novel hybrid molecules, improving both enantioselectivity and yield. The advances in stop codon technology to introduce UAAs, especially in the context of directed evolution, make their use a promising option and provides an enticing method for expanding the ligand set available to the protein scaffold (53–55). In another study, Reetz and coworkers computationally designed a Cu(II) ion binding site into the thermostable protein imidazole glycerol phosphate synthase (56). The resulting ArM was able to catalyse the Diels-Alder cycloaddition of an azachalcone and

cyclopentadiene with medium selectivity, however, to our knowledge no subsequent directed evolution experiments have been reported.In contrast to these examples of forming the

complete coordination sphere by binding a bare metal to the apoprotein state of the ArM, to the best of our knowledge there are only very few examples of adding exogenous metal complexes (particularly 4d and 5d metal complexes) as precursor cofactors which then show catalytic activity upon direct coordination to a protein (57). This is a particularly attractive methodology as the challenges of taking unnatural ligands such as arenes, carbenes and phosphanes into biology become opportunities for expanding the repertoire of chemistries available for catalysis. Controlling the ligand exchange behaviour of 4d and 5d metal complexes with protein side chain ligands is challenging, not least because coordination bonds between ligands and heavier metals are often stronger than their 3d counterparts and hence exchange rates are slower. This, however, remains an exciting area of research due to the catalytic diversity demonstrated by many 4d and 5d metal complexes. In this specific area our own work has focused upon ruthenium complexes and their ligand exchange behaviour with biological systems, laying the foundation for future work into ArMs with direct metal-protein coordination (58, 59).

5. Summary and Outlook

Significant advances in the incorporation of organometallic complexes into proteins in order to generate ArMs have been made. The studies highlighted above reliably create hybrid molecules where the stability and turnover number of the metal centre is higher than the comparable small molecule organometallic complex in aqueous solution. Maybe unsurprisingly, the propensity for side reactions and catalyst decomposition is lowered once the complex is in a hydrophobic protein environment, already showcasing the usefulness of these hybrid systems. However, the question remains, as to whether or not these strategies make full use of the protein component. The unique and numerous demands of ArMs call for a highly integrated approach. To date, most of the work described in the literature attempts to exploit the chemistry of metal ions and their complexes in a protein scaffold but with limited influence from the protein on any catalytic activity because metal-protein coordination is largely indirect and so cooperativity is limited.



The potential for synthetic organometallic chemistry to deliver cofactors which utilise ligand chemistry not available to naturally evolved systems can vastly expand the orthogonal catalysis available in synthetic biological applications. Using such molecules to embed novel metal-peptide hybrid complexes in protein scaffolds allows for three-dimensional and electronic control around the metal centre that reduces the need for intricate synthetic catalyst generation. Instead, control of the steric and electronic environment around the metal ion can be delivered via the protein coordination sphere, particularly where a direct coordination bond is used to anchor the metal ion to the protein. When combined with the ability to efficiently evolve enzymes, a sophisticated organometallic precursor complex together with a suitable apoprotein could potentially give rise to a number of diverse reactivities. Therefore, new problems in catalysis could be addressed in a faster and more specific manner than with small molecule catalysts. Together with non-covalent contributions to catalysis and the intermolecular interactions that pre-organise substrates and stabilise transition states, such a system contains many readily evolvable components.The majority of protein scaffolds selected for

ArM construction have been chosen because of their apparent engineerability. However, in most cases the focus seems to lie solely on the peptidic component with little consideration for evolution of the metal complex. Although methods of selection and directed evolution have been applied, these are often operating on already well-defined protein scaffolds that carry an abiotic cofactor but not a direct protein-metal complex, which inevitably limits the scope for evolution. Arguably it is desirable, therefore, to select for a promiscuous and versatile protein starting point which is not constrained by one energy minima but instead can potentially offer numerous distinct metal-binding environments, both in terms of direct coordination and through secondary, intramolecular spheres of influence, ultimately generating differential catalytic ArM activity.Performing catalysis with exogenous metal

complexes within cellular environments has enormous potential applications in medicinal chemistry and synthetic biology. Given the potential difficulties associated with cell-uptake, minimising deactivation, overcoming toxicity of exogenous metal ions and precise localisation of metal cofactors in cells, the idea of using traditionally inert organometallic complexes has obvious advantages

in that reactive promiscuity is reduced. As pointed out above, such complexes would be designed to have a latent catalytic activity which emerges once the metal complex is bound to a protein. The design challenges raised by this approach are not just as a result of a need to control the electronic and three-dimensional steric coordination sphere of the metal ion, but also to limit ligand exchange processes, restricting lability of a precursor complex (in the cellular milieu) until it reaches a specific protein target. Since the metal-ligand exchange processes for 4d and 5d metal complexes are typically slow, they are particularly attractive from this point of view but are hard to predict ab initio.

6. Conclusion

In conclusion, in order to optimise the chemistry and biochemistry of ArMs, directed evolutionary campaigns coupled with high throughput screening methods rather than individually-designed synthetic strategies are much more likely to generate optimised orthogonal catalysts for new and efficient metabolic processes. Direct coordination between metal ions and enzymes is essential in order to deliver truly interdependent systems, ideally where entatic states deliver enhanced reactivity, efficiency and selectivity that cannot easily be replicated in conventional, synthetic metal catalysis. Going forward, methods of generating ArMs should be evaluated and developed for both their ability to be used in directed evolution procedures and the extent to which the protein scaffold participates in the activity of the metal complex.

Acknowledgements

George Biggs is supported by the Engineering and Physical Sciences Research Council (EPSRC) (EP/N509620/1) and Peterhouse, University of Cambridge, UK. Oskar James Klein is supported by the EPSRC (EP/R513180/1). Sally Boss and Paul Barker thank the Department of Chemistry, University of Cambridge. We thank Florian Hollfelder for deep discussions.

References

1. G. Winter, A. R. Fersht, A. J. Wilkinson, M. Zoller and M. Smith, Nature, 1982, 299, (5885), 756

2. A. R. Fersht, J-P. Shi, J. Knill-Jones, D. M. Lowe, A. J. Wilkinson, D. M. Blow, P. Brick, P. Carter, M. M. Y. Waye and G. Winter, Nature, 1985, 314, (6008), 235



3. F. Schwizer, Y. Okamoto, T. Heinisch, Y. Gu, M. M. Pellizzoni, V. Lebrun, R. Reuter, V. Köhler, J. C. Lewis and T. R. Ward, Chem. Rev., 2018, 118, (1), 142

4. M. Jeschek, S. Panke and T. R. Ward, Trends Biotechnol., 2018, 36, (1), 60

5. J. G. Rebelein and T. R. Ward, Curr. Opin. Biotechnol., 2018, 53, 106

6. J. D. Tyzack, A. J. M. Ribeiro, N. Borkakoti and J. M. Thornton, ACS Synth. Biol., 2019, 8, (11), 2494

7. A. Bhushan, P. J. Egli, E. E. Peters, M. F. Freeman and J. Piel, Nat. Chem., 2019, 11, (10), 931

8. M. C. Wilson, T. Mori, C. Rückert, A. R. Uria, M. J. Helf, K. Takada, C. Gernert, U. A. E. Steffens, N. Heycke, S. Schmitt, C. Rinke, E. J. N. Helfrich, A. O. Brachmann, C. Gurgui, T. Wakimoto, M. Kracht, M. Crüsemann, U. Hentschel, I. Abe, S. Matsunaga, J. Kalinowski, H. Takeyama and J. Piel, Nature, 2014, 506, (7486), 58

9. A. Casini, F. Y. Chang, R. Eluere, A. M. King, E. M. Young, Q. M. Dudley, A. Karim, K. Pratt, C. Bristol, A. Forget, A. Ghodasara, R. Warden-Rothman, R. Gan, A. Cristofaro, A. E. Borujeni, M.-H. Ryu, J. Li, Y.-C. Kwon, H. Wang, E. Tatsis, C. Rodriguez-Lopez, S. O’Connor, M. H. Medema, M. A. Fischbach, M. C. Jewett, C. Voigt and D. B. Gordon, J. Am. Chem. Soc., 2018, 140, (12), 4302

10. W. F. DeGrado, C. M. Summa, V. Pavone, F. Nastri and A. Lombardi, Annu. Rev. Biochem., 1999, 68, 779

11. L. Falivene, Z. Cao, A. Petta, L. Serra, A. Poater, R. Oliva, V. Scarano and L. Cavallo, Nat. Chem., 2019, 11, (10), 872

12. L. Alonso-Cotchico, J. Rodrıguez-Guerra, A. Lledós and J.-D. Maréchal, Acc. Chem. Res., 2020, 53, (4), 896

13. I. Drienovská, L. Alonso-Cotchico, P. Vidossich, A. Lledós, J.-D. Maréchal and G. Roelfes, Chem. Sci., 2017, 8, (10), 7228

14. L. Alonso-Cotchico, G. Sciortino, P. Vidossich, J. Rodríguez-Guerra Pedregal, I. Drienovská, G. Roelfes, A. Lledós and J.-D. Maréchal, ACS Catal., 2019, 9, (5), 4616

15. H. Feng, X. Guo, H. Zhang, L. Chen, P. Yin, C. Chen, X. Duan, X. Zhang and M. Wei, Phys. Chem. Chem. Phys., 2019, 21, (42), 23408

16. R. B. Leveson-Gower, C. Mayer and G. Roelfes, Nat. Rev. Chem., 2019, 3, (12) 687

17. M. Jeschek, R. Reuter, T. Heinisch, C. Trindler, J. Klehr, S. Panke and T. R. Ward, Nature, 2016, 537, (7622), 661

18. W. R. Hagen, Metallomics, 2019, 11, (11), 1768

19. B. L. Vallee and R. J. Williams, Proc. Natl. Acad.

Sci., 1968, 59, (2), 498

20. C. L. Drennan, S. Huang, J. T. Drummond, R. G. Matthews and M. L. Lidwig, Science, 1994, 266, (5191), 1669

21. M. L. Ludwig, C. L. Drennan and R. G. Matthews, Structure, 1996, 4, (5), 505

22. R. Banerjee and S. W. Ragsdale, Annu. Rev. Biochem., 2003, 72, 209

23. M. T. Reetz, Acc. Chem. Res., 2019, 52, (2), 336

24. J. Zhao, J. G. Rebelein, H. Mallin, C. Trindler, M. M. Pellizzoni and T. R. Ward, J. Am. Chem. Soc., 2018, 140, (41), 13171

25. P. Dydio, H. M. Key, A. Nazarenko, J. Y.-E. Rha, V. Seyedkazemi, D. S. Clark and J. F. Hartwig, Science, 2016, 354, (6308), 102

26. S. Studer, D. A. Hansen, Z. L. Pianowski, P. R. E. Mittl, A. Debon, S. L. Guffy, B. S. Der, B. Kuhlman and D. Hilvert, Science, 2018, 362, (6420), 1285

27. P. S. Coelho, E. M. Brustad, A. Kannan and F. H. Arnold, Science, 2013, 339, (6117), 307

28. F. H. Arnold, Acc. Chem. Res., 1998, 31, (3), 125

29. F. H. Arnold, Angew. Chem., Int. Ed., 2018, 57, (16), 4143

30. J. E. Coleman, Nature, 1967, 214, (5084), 193

31. Q. Jing, K. Okrasa and R. J. Kazlauskas, Chem. Eur. J., 2009, 15, (6), 1370

32. Q. Jing and R. J. Kazlauskas, ChemCatChem, 2010, 2, (8), 953

33. H. M. Key, P. Dydio, D. S. Clark and J. F. Hartwig, Nature, 2016, 534, (7608), 534

34. M. E. Wilson and G. M. Whitesides, J. Am. Chem. Soc., 1978, 100, (1) 306

35. T. Heinisch, F. Schwizer, B. Garabedian, E. Csibra, M. Jeschek, J. Vallapurackal, V. B. Pinheiro, P. Marlière, S. Panke and T. R. Ward, Chem. Sci., 2018, 9, (24), 5383

36. A. D. Liang, J. Serrano-Plana, R. L. Peterson and T. R. Ward, Acc. Chem. Res., 2019, 52, (3), 585

37. S. Eda, I. Nasibullin, K. Vong, N. Kudo, M. Yoshida, A. Kurbangalieva and K. Tanaka, Nat. Catal., 2019, 2, (9), 780

38. D. J. Raines, J. E. Clarke, E. V. Blagova, E. J. Dodson, K. S. Wilson and A.-K. Duhme-Klair, Nat. Catal., 2018, 1, (9), 680

39. J. N. deGruyter, L. R. Malins and P. S. Baran, Biochemistry, 2017, 56, (30), 3863

40. E. A. Hoyt, P. M. S. D. Cal, B. L. Oliveira and G. J. L. Bernardes, Nat. Rev. Chem., 2019, 3, (3) 147



41. P. Haquette, B. Talbi, S. Canaguier, S. Dagorne, C. Fosse, A. Martel, G. Jaouen and M. Salmain, Tetrahedron Lett., 2008, 49, (31), 4670

42. P. Haquette, M. Salmain, K. Svedlung, A. Martel, B. Rudolf, J. Zakrzewski, S. Cordier, T. Roisnel, C. Fosse and G. Jaouen, ChemBioChem, 2007, 8, (2), 224

43. N. Madern, N. Queyriaux, A. Chevalley, M. Ghasemi, O. Nicolotti, I. Ciofini, G. F. Mangiatordi and M. Salmain, J. Mol. Catal. B: Enzym., 2015, 122, 314

44. C. J. Noren, S. J. Anthony-Cahill, M. C. Griffith and P. G. Schultz, Science, 1989, 244, (4901), 182

45. J. C. Lewis, Curr. Opin. Chem. Biol., 2015, 25, 27

46. H. Yang, P. Srivastava, C. Zhang and J. C. Lewis, ChemBioChem, 2014, 15, (2), 223

47. P. Srivastava, H. Yang, K. Ellis-Guardiola and J. C. Lewis, Nat. Commun., 2015, 6, (1), 7789

48. A. Lombardi, F. Pirro, O. Maglio, M. Chino and W. F. DeGrado, Acc. Chem. Res., 2019, 52, (5), 1148

49. T. Handel and W. F. DeGrado, J. Am. Chem. Soc., 1990, 112, (18), 6710

50. I. Drienovská, A. Rioz-Martínez, A. Draksharapu and G. Roelfes, Chem. Sci., 2015, 6, (1), 770

51. N. Ségaud, I. Drienovská, J. Chen, W. R. Browne

and G. Roelfes, Inorg. Chem., 2017, 56, (21),

13293

52. M. Bersellini and G. Roelfes, Org. Biomol. Chem.,

2017, 15, (14), 3069

53. Q. Wang, A. R. Parrish and L. Wang, Chem. Biol.,

2009, 16, (3), 323

54. C. Mayer, C. Dulson, E. Reddem, A.-M. W. H.

Thunnissen and G. Roelfes, Angew. Chem. Int.

Ed., 2019, 58, (7), 2083

55. A. Fallah-Araghi, J.-C. Baret, M. Ryckelynck and

A. D. Griffiths, Lab Chip, 2012, 12, (5), 882

56. J. Podtetenieff, A. Taglieber, E. Bill, E. J. Reijerse

and M. T. Reetz, Angew. Chem. Int. Ed., 2010,

49, (30), 5151

57. J. de Jesús Cázares-Marinero, C. Przybylski and

M. Salmain, Eur. J. Inorg. Chem., 2018, (12),

1383

58. T. G. Scrase, M. J. O’Neill, A. J. Peel, P. W. Senior,

P. D. Matthews, H. Shi, S. R. Boss and P. D. Barker,

Inorg. Chem., 2015, 54, (7), 3118

59. G. S. Biggs, M. J. O’Neill, P. Carames Mendez,

T. G. Scrase, Y. Lin, A. M. Bin-Maarof, A. D. Bond,

S. R. Boss and P. D. Barker, Dalton Trans., 2019,

48, (20), 6910

The Authors

George Biggs completed an MChem in Chemistry at the University of Bath, UK, in 2016. He is now a PhD student in the Department of Chemistry at the University of Cambridge. Supervised by Paul Barker and Sally Boss, his project is focused on understanding the reactivity of Ru(II) arene complexes with proteins for the development of novel ArMs.

Oskar James Klein obtained an MSc in Chemistry from the University of Cambridge in 2019, where he remains as a PhD student in the Department of Chemistry. Supervised by Paul Barker and Sally Boss and in collaboration with Professor Florian Hollfelder his project tries to develop a high throughout methodology for the formation and evolution of novel ArMs.



Sally Boss studied for an MSci in Chemistry at the University of Bristol, UK, and Heidelberg University, Germany, before moving to the University of Cambridge to begin a PhD on the synthesis and reactivity of Lewis acidic, heterobimetallic main group compounds. Shortly after obtaining her PhD in 2005, she was appointed to a joint College Lectureship in the Department of Chemistry and at Churchill College, University of Cambridge. Her time is split between teaching and research and her specific interest is in improving the utility of heavy metals in biology by careful design of complexes, targeted direction of metal-cofactors to protein targets and using spectroscopy to understand how they behave in situ.

Paul Barker is Senior Lecturer at the University of Cambridge, Department of Chemistry, and a Fellow of Downing College, University of Cambridge. His research has always been at the interface between inorganic chemistry and biology. It started in the field of electron transfer proteins studied by biophysical methods and mutagenesis, in the early days of protein engineering. After two, independent Medical Research Council (MRC), UK, and Biotechnology and Biological Sciences Research Council (BBSRC), UK, fellowships in Cambridge he joined the Chemistry faculty and has been combining protein engineering with synthesis and self-assembly for the purposes of generating novel protein based electronic and catalytic systems. His current interests span protein design and evolution, self-assembling materials and synthesis of organometallic complexes.




Irfan TuretgenBasic and Industrial Microbiology Section, Department of Biology, Faculty of Science, Istanbul University, 34134 Vezneciler, Istanbul, Turkey

Email: [email protected]

Cooling towers are industrial cooling units operating to dissipate heat. As with any surface in contact with aqueous systems, biofilm formation appears on the surface of heat exchangers. Although biofilm formation on plastic tower fill in wet cooling towers has been studied widely, no studies were found regarding biofilm formation on steel heat exchangers in closed-loop systems. In this study, heat exchangers were coated with nano-silica, which is known to reduce the formation of biofilm. Natural biofilm formation was monitored for six months. Biofouling was examined monthly using epifluorescence microscopy by assessing the numbers of live and dead bacteria. It was observed that the biofilm layer formed on the nano-silica coated heat exchanger surfaces was significantly lower than on the control surfaces. 3 log microbial reduction was recorded on coated surfaces in the first month. After six months, total biomass on control surfaces reached 1.28 × 1012 cell cm–2, while the biomass on nano-silica coated surfaces was 6.3 × 104 cell cm–2.

1. Introduction

A cooling tower is a heat dissipation unit which cools bulk water in industrial systems. Cooling

towers provide cooling by spraying the heated water coming from the system onto a fill material and rejecting the heat to the open atmosphere (1). The cooled water returns to a basin to recirculate again through the system. Common uses of wet cooling towers include air conditioner systems, manufacturing facilities, telecommunication devices and power plants. Such man-made installations provide an ideal environment for bacterial growth similar to an incubator, supported by water temperatures ranging between 24°C and 38°C (2–4). The heated water comes from the source to the heat exchanger that allows the exchange of heat between two liquids at different temperatures by indirect contact inside water jacketed tubes (5). Wet cooling towers providing cool water

for heating, ventilating and air conditioning (HVAC) systems are known to be subject to contamination. Organic and inorganic substances in bulk water are deposited on the water contact surfaces, reducing the heat transfer significantly and threatening the operating stability of the whole system. Established biofilms offer cleaning challenges because they are resistant to most chemical and physical cleaning protocols and they also reduce the heat transfer efficiency (6, 7). HVAC systems are responsible for about half of the energy consumed in modern buildings and industrial facilities. Therefore, biofouling is always a significant issue for heat exchangers and should be taken into account during heat exchanger design and production. As a solution, altering the surface properties could be an effective approach to reduce biofouling in such hard-to-reach articulated systems (2, 8, 9).

Reduction of Biofilm Formation on Cooling Tower Heat Exchangers using Nano-silica Coating Environmentally sustainable antifouling coating demonstrated on stainless steel heat exchanger tubes



A biofilm layer is a community formed by bacterial cells living in a polymeric matrix that they produce; a functional partnership adhered to a living or inanimate surface organised by microorganisms in a dense exopolymer matrix. The capability of microbes to stick to a substratum and to produce a biofilm layer has great significance in a diversity of cooling towers, where fouling can act as a perpetual source of contamination. Biofilm layer must be kept to a minimum in order to prolong the operating life of man-made water systems and facilitate control of pathogens. Disinfectants may be used for this purpose (4, 9).Industrial cooling towers can be manufactured

from different materials. Generally, towers are made of reinforced concrete or fibreglass, stainless steel, wood or reinforced plastic sheets. The fill material is generally made of plastic sheets (polypropylene, polyethylene or polyvinylchloride) where heat dissipation occurs. For corrosion resistance, towers are specially treated, painted and covered with a protective film layer (7). In the case of corrosive water or atmospheric conditions, the use of plastic towers is recommended. But heat exchanger units are made of stainless steel or copper for better thermal conduction (5). The critical issue that affects cooling is the aggregation of deposits over the heat exchanger surfaces which includes biofouling. Conventional steel heat exchangers may have corrosion or deposits may have formed on the heat exchanger tubes. Both of these factors reduce the heat transfer rate (10). To solve this problem, novel anti-fouling coatings are considered. Nano-silica can be used in the form of liquid composites in many matrices as coating materials. Nano-silica is used in the textile and automotive industries because of its self-cleaning, abrasion resistant, hydrophobic and oleophobic features. It is known that nano-silica is able to create low-cost, hard and tough coatings which are resistant to wear and weathering (11). Although biofilm formation on plastic fill surfaces

in wet cooling towers has been studied widely, no studies were found on biofilm formation on steel heat exchangers in cooling towers. As coating of heat exchangers is not common, the aim of the current work was to limit tenacious biofouling on heat exchangers using a nano-silica coating, which will lead to longer material life, better cooling of water and less clogging in closed-loop systems.

Materials and Method

A brand new fabricated closed-loop cooling tower was monitored for six months. A real size, fully working closed-loop cooling tower system was kept in operation by the manufacturer during the experimental period at the factory test laboratory. The system was filled with distributed network water. Regular blowdown was implemented to limit the concentration of dissolved solids. In a circulation rig, hot process water was kept separate from the cooling water in a closed-loop system (Figure 1). For the experiments, a half portion of the stainless steel (316 SS) heat exchanger tubes were coated with nano-silica and the other part was left without coating. Coating was done by coaxial electrospraying before assembly and left to cure in air for 24 hours. Coaxial electrospraying has several implicit advantages such as high encapsulation efficiency and uniform particle distribution. The coating thickness was between 4–6 µm. Before coating, the heat exchanger tubes were sprayed with 96% ethanol to remove any dirt, oil or grime. This application made the bonding of the coating stronger. Silica in powder form is hydrophilic. To produce

hydrophobic nano-silica, the silica particles were transformed by fluorination to confer hydrophobicity. The final particle size was about 40 nm. The aqueous form of the nano-silica coating contains ethanol as solvent to keep it in liquid form before

Heat exchanger

Fan

Hot water inlet

Cooled water outlet

Fig. 1. Schematic view of the cooling tower and heat exchanger



use. The final nano-silica product was supplied by a local company. After curing, the coating was solid on the surfaces, and no colour change, shedding or weight loss were observed on any of the coated test surfaces after the experimental period. The stability of the coating was tested in a different study by the present author (9) and the mean overall adhesion capability of the coating was recorded as 1.6 using a pull-off adhesion tester, which matches very well with the general rating of adhesion. Water was circulated over the stainless steel (316 SS) heat exchanger tubes, where natural biofilm formation was allowed to occur. Sampling of the biofilm required dismantling the outer shell of the heat exchanger unit every month. The system temperature water was kept constant at 37°C using an electrical heating unit to eliminate temperature fluctuation which might influence biofilm formation over time.Pipe segments were cut monthly from the heat

exchanger using an angle grinder, kept in a container filled with system water and brought quickly to the laboratory for analysis. LIVE/DEAD® BacLightTM Bacterial Viability Kit (InvitrogenTM, Thermo Fisher Scientific, USA) dye was added immediately to cover the surfaces completely to stain the actively respiring and dead bacteria. After 15 min, the surfaces were rinsed with sterile bi-distilled water to remove unattached cells, air dried, covered with immersion oil and cover slip, then examined in the dark. This was repeated every month until the study finished at the sixth month. An epifluorescence microscope (Eclipse 80i, Nikon Instruments Inc, Japan) was used to visualise the biofilm cells in situ. The camera enables counting and taking images of bacteria on solid surfaces, with the signals displayed on the computer monitor. Counting and recording were

carried out using special software (NIS-Elements, Nikon Instruments Inc, Japan). Signals obtained from 20 randomly selected regions were recorded. Images were saved for later analysis. The LIVE/DEAD® kit stains dead cells red and

live cells green in colour. The LIVE/DEAD® test kit contains two DNA-binding dyes, propidium iodide and SYTO® 9. These dyes differ in their spectral properties and their ability to enter the living bacterial cell. The first dye in the kit is SYTO® 9, which can pass through the membrane of all bacteria and stain the cells green. Propidium iodide only enters into cells with a damaged cell membrane, allowing them to appear red under fluorescent light. The number of viable and dead bacteria on surfaces can be determined in a single step using a dual emission filter cube (Chroma Technology GmbH, Germany).For both parameters over the six-month duration

of the experiment, the difference between the average bacterial numbers were compared by two-way analysis of variance. A follow-up post-hoc analysis was done in order to determine differences. The difference was considered significant when p < 0.05. SPSS® Version 18.0 software (IBM Corp, USA) was used for the statistical analyses.

Results and Discussion

The bacterial numbers from the LIVE/DEAD® test kit were analysed in situ on the surfaces using the manufacturer’s software during the experimental period for six months. The results are given in Table I. The number of signals per cm2 were calculated using the magnification factor. Since the raw data were too scattered, the values are given in the logarithmic (log10) base for better

Table I Numbers with Standard Deviation of Live-Dead Bacteria Counted on Heat Exchanger Surfaces

Nano-silica coated test surfaces, cell cm–2 Uncoated control surfaces, cell cm–2

Months Dead (log10) Live (log10) Total (log10) Dead (log10) Live (log10) Total (log10)

1 3.6 ± 0.07 3.6 ± 0.05 4.6 ± 0.09 6.8 ± 0.11 6.0 ± 0.08 8.1 ± 0.14

2 3.3 ± 0.09 3.6 ± 0.10 4.3 ± 0.12 7.7 ± 0.13 7.5 ± 0.11 9.7 ± 0.16

3 3.3 ± 0.04 3.7 ± 0.07 4.2 ± 0.08 6.9 ± 0.11 8.0 ± 0.14 10.1 ± 0.09

4 3.0 ± 0.02 3.8 ± 0.04 4.5 ± 0.10 7.0 ± 0.13 8.2 ± 0.12 11.5 ± 0.17

5 3.1 ± 0.09 3.8 ± 0.10 4.7 ± 0.11 7.5 ± 0.15 8.1 ± 0.15 11.9 ± 0.17

6 3.2 ± 0.08 3.9 ± 0.06 4.8 ± 0.12 7.8 ± 0.17 9.1 ± 0.18 12.1 ± 0.21



comparison. The logarithmic reduction was clearly significant starting from the first sampling.The total bacterial numbers on coated tubes

were recorded as 49,090 cell cm–2 after the initial month, and 13,016,957 cell cm–2 on uncoated surfaces after the first month. The results distinctly showed that this type of coating reduces biofouling formation on heat exchanger surfaces from the start of the experimental set-up. The numbers of surface associated bacteria on uncoated control tubes gradually increased and reached 1.28 × 1012 cell cm–2 after the sixth month, at which time the biomass on nano-silica coated tubes was 6.3 × 104 cell cm–2. No significant rise (p < 0.05) of bacterial numbers on nano-coated heat exchanger tubes was recorded during the six-month period in terms of total biofilm counts. This outcome demonstrates that a nano-silica coating can clearly reduce the bacterial biofilm layers on coated heat exchanger surfaces.As expected, nano-silica coating slowed down

the adhesion and colonisation of bacteria on the substrata thanks to its strong hydrophobic properties. The pH, dissolved oxygen, total dissolved matter and temperature values of the water in the system during the six-month test period were recorded and are given in Table II. The values in Table II were important to monitor circulating water due to the blowdown regime.It is known that even with conventional cleaning

and disinfection regimens, there is a problem fighting against biofilm formation and development of microbial resistance (12). Based on previous studies conducted in this field (13, 14), it is impossible to eliminate the formation of biofilm layers on surfaces, but biofilm formation can be reduced (9, 15, 16). For this purpose, it is possible to modify surfaces with different coatings. The nano-hydrophobic coating changes the surface

properties of the material and supports less biofilm formation (16–18). Hydrophobic coatings limit the wettability of the surface, making it difficult for organic and inorganic matter or microorganisms to adhere; and even if they do, they can easily be detached from the surface by physical factors such as laminar or turbulent water shear stress (19). The issue of antimicrobial coatings has been

extensively studied (20–24). The problem with these products is development of bacterial resistance against the agent (11, 25). Even antibiotic-containing coatings have been reported to promote biofilm formation (26). Silver compounds combined with silica, silane and titanium coatings in particular gave antimicrobial properties but the problem of toxicity in medical devices was mentioned (27). In industrial use, the resistance of microorganisms is at the top of the list as a disadvantage (28). In addition, silver compounds in water systems will reach the aquatic environment and appear as a separate environmental problem. It is also emphasised that anti-biofilm coatings

are very important for preventing the formation of a biofilm layer at an early stage (29, 30). However, studies conducted to date are mostly aimed at solving clinical problems and have been done in vitro with pure cultures (15, 17, 18, 31–33). Using monospecies biofilms is a sterile approach and cannot represent mixed cultures in the natural environment and their interaction with each other. Sol-gel products and superhydrophobic coatings which are more strongly water repellent (31, 34) have also been tried. It was observed that the life of these coatings was not as long as hydrophobic coatings. On the other hand, the high cost of superhydrophobic coatings was a drawback. Contrary to hydrophobic coatings, some hydrophilic coatings were also found to be effective against biofouling. Holberg et al. (8) reported that

Table II pH, Dissolved Oxygen, Total Dissolved Solids and Temperature Values of Circulating Water in the Systema

Months pH Dissolved oxygen, mg l–1 Total dissolved solids, ppm Temperature, °C

1 7.33 7.40 110 37

2 7.48 7.54 113 37

3 7.28 7.34 109 37

4 7.24 7.24 108 37

5 7.30 7.55 107 37

6 7.36 7.39 110 37

aThe numerical data were the average of three consecutive measurements



biocide-free silicone coatings showed promising real-life performance on fresh water-cooled heat exchangers and also performed well in laboratory tests.Ding et al. (35) tested an environmentally friendly

antifouling coating product, butenolide, which was designed for controlled release from biodegradable polyurethane. The anti-fouling effect was shown by in situ tests. The main target was marine biofouling, especially larval settlement on surfaces. Since the adhesion of fouling organisms relies on a microbial biofilm layer, inhibition of primer settlement is crucial. Hu et al. (36) sprayed bacterial-anti-adhesive modified polystyrene microspheres to construct bacterially-anti-adhesive surfaces. It can be used on any surface thanks to the lotus effect. It was reported as robust and durable on surfaces. Similar surface engineering strategies focus on altering the physicochemical properties of the material surface. In general, reduced efficacy of regular disinfectants leads to progress in development of antimicrobial surfaces and coatings (37, 38).

ConclusionThis is the first report of a nano-silica coating on a stainless steel cooling tower heat exchanger. The study showed that the nano-silica coating significantly reduced bacterial fouling on surfaces. There are many similar surfaces with biofouling problems which have contact with water and require a solution. Nano-silica has proven to be effective at reducing the formation of biofilms on surfaces and can be applied as a cost-effective, effortless, non-toxic, readily available material. Due to growing restraints on environmental release of biocidal agents and the growing restrictions on the use of disinfectants in man-made water systems, as well as demand to decrease the cost of system maintenance, different ways to limit biofilms in man-made water systems hold much expectation.

Acknowledgments This study was supported by ‘Research Fund of the Istanbul University’. Project number: 29220.

References

1. F. Di Pippo, L. Di Gregorio, R. Congestri, V. Tandoi and S. Rossetti, FEMS Microbiol Ecol., 2018, 94, (5), fiy044

2. S. K. R. Namasivayam, A. L. Francis, R. S. A. Bharani and C. V. Nachiyar, J. Clean. Prod., 2019, 231, 872

3. H.-F. Tsao, U. Scheikl, C. Herbold, A. Indra, J. Walochnik and M. Horn, Water Res., 2019, 159, 464

4. I. Türetgen, Biofouling: J. Bioadhes. Biofilm Res., 2004, 20, (2), 81

5. Z. Nourani, A. Naserbegi, Sh. Tayyebi and M. Aghaie, Therm. Sci. Eng. Prog., 2019, 14, 100406

6. A. Zaza, N. E. Laadel, E. G. Bennouna, Y. El Hammami and M. T. Janan, Energy Proc., 2019, 157, 1230

7. T. V. Wagner, J. R. Parsons, H. H. M. Rijnaarts, P. de Voogt and A. A. M. Langenhoff, J. Haz. Mater., 2020, 384, 121314

8. S. Holberg, R. Losada, F. H. Blaikie, H. H. W. B. Hansen, S. Soreau and R. C. A. Onderwater, Mater. Today Comm., 2020, 22, 100750

9. I. Türetgen, Water SA, 2015, 41, (3), 295

10. M. Lemouari, M. Boumaza and A. Kaabi, Energy, 2011, 36, (10), 5815

11. M. Malaki, Y. Hashemzadeh and M. Karevan, Prog. Org. Coat., 2016, 101, 477

12. M. Simões, L. C. Simões and M. J. Vieira, Food Sci. Technol., 2010, 43, (4), 573

13. I. W. Sutherland, Microbiology, 2001, 147, (1), 3

14. C. Gómez-Suárez, J. Pasma, A. J. van der Borden, J. Wingender, H.-C. Flemming, H. J. Busscher and H. C. van der Mei, Microbiology, 2002, 148, (4), 1161

15. N. M. Dat, L. D. Manh, D. Hamanaka, D. V. Hung, F. Tanaka and T. Uchino, Food Control, 2014, 42, 94

16. M. Pasmore, P. Todd, B. Pfeifer, M. Rhodes and C. N. Bowman, Biofouling: J. Bioadhes. Biofilm Res., 2002, 18, (1), 65

17. K. Naik and M. Kowshik, Mater. Sci. Eng., 2014, 34, 62

18. J. Azeredo and R. Oliveira, ‘Biofilm Characteristics: Biofilm Formation: The Role of Hydrophobicity and Exopolymers in Initial Adhesion and Biofilm Formation’, in “Biofilms in Medicine, Industry and Environmental Biotechnology”, eds. P. Lens, A. P. Moran, T. Mahony, P. Stoodley and V. O’Flaherty, Pt. 1, Section 1, IWA Publishing, London, UK, 2003, pp 16–32

19. B. Arkles, ‘Hydrophobicity, Hydrophilicity and Silanes’, Paint and Coatings Industry Magazine, 2006, 22, (10), 114

20. V. Antoci, C. S. Adams, J. Parvizi, H. M. Davidson R. J. Composto, T. A. Freeman, E, Wickstrom, P. Ducheyne, D. Jungkind, I. M. Shapiro and N. J. Hickok, Biomaterials, 2008, 29, (35), 4684



21. A. L. Casey, L. A. Mermel, P. Nightingale and T. S. J. Elliott, Lancet: Infect. Dis., 2008, 8, (12), 763

22. S. A. McConnell, P. O. Gubbins and E. J. Anaissie, Clin. Infect. Dis., 2003, 37, (1), 65

23. J. C. Hockenhull, K. M. Dwan, G. W. Smith, C. L. Gamble, A. Boland, T. J. Walley and R. C. Dickson, Crit. Care Med., 2009, 37, (2), 702

24. B. S. Niël-Weise, T. Stijnen and P. J. van den Broek, Intens. Care Med., 2007, 33, (12), 2058

25. A. K. Epstein, B. Pokroya, A. Seminara and J. Aizenberg, Proc. Natl. Acad. Sci., 2011, 108, (3), 995

26. L. R. Hoffman, D. A. D’Argenio, M. J. MacCoss, Z. Zhang, R. A. Jones and S. I. Miller, Nature, 2005, 436, (7054), 1171

27. P. Jena, S. Mohanty, R. Mallick, B. Jacob and A. Sonawane, Int. J. Nanomed., 2012, 7, 1805

28. H. Du, T.-M. Lo, J. Sitompul and M. W. Chang, Biochem. Biophys. Res. Comm., 2012, 424, (4), 657

29. L. D. Renner and D. B. Weibel, MRS Bull., 2011, 36, (5), 347

30. M. Chen, Q. Yu and H. Sun, Int. J. Mol. Sci., 2013, 14, (9), 18488

31. L. J. Zhong, L. Q. Pang, L. M. Che, X. E. Wu and X. D. Chen, Coll. Surf. B: Biointerfaces, 2013, 111, 252

32. A. Pagedar, J. Singh and V. K. Batish, J. Basic Microbiol., 2010, 50, (S1), S98

33. A. Okada, T. Nikaido, M. Ikeda, K. Okada, J. Yamauchi, R. M. Foxton, H. Sawada, J. Tagami and K. Matin, Dental Mater. J., 2008, 27, (4), 565

34. B. J. Privett, J. Youn, S. A. Hong, J. Lee, J. Han, J. H. Shin and M. H. Schoenfisch, Langmuir, 2011, 27, (15), 9597

35. W. Ding, C. Ma, W. Zhang, H. Chiang, C. Tam, Y. Xu, G. Zhang and P.-Y. Qian, Biofouling: J. Bioadhes. Biofilm Res., 2018, 34, (1), 111

36. J. Hu, J. Lin, Y. Zhang, Z. Lin, Z. Qiao, Z. Liu, W. Yang, X. Liu, M. Dong and Z. Guo, J. Mater. Chem. A, 2019, 7, (45), 26039

37. P. S. V. V. S. Narayana and P. S. V. V. Sirihari, Regen. Eng. Transl. Med., 2019

38. S. Achinas, N. Charalampogiannis and G. W. Euverink, Appl. Sci., 2019, 9, (4), 2801

The Author

Irfan Turetgen is a Full Professor at the Department of Biology, Faculty of Science, Istanbul University, Turkey. He received his PhD degree in Environmental Microbiology from Istanbul University. His major area of research is disinfection of heterotrophic biofilms in man-made water systems, Legionella ecology in cooling towers and anti-biofilm coatings. He has authored research papers published in reputed journals and conference proceedings. Recently he is supervising thesis projects as a mentor. His current interest is modelling man-made water systems to mimic their function at laboratory scale and test disinfectants under conditions as close as possible to real life.




Mathew J. Haskew, John G. Hardy*Department of Chemistry and Materials Science Institute, Faraday Building, Lancaster University, Lancaster, LA1 4YB, UK


Shape-memory polymers (SMPs) enable the production of stimuli-responsive polymer-based materials with the ability to undergo a large recoverable deformation upon the application of an external stimulus. Academic and industrial research interest in the shape-memory effects (SMEs) of these SMP-based materials is growing for task-specific applications. This mini-review covers interesting aspects of SMP-based materials, their properties, how they may be investigated and highlights examples of the potential applications of these materials.

Introduction

SMEs refers to the ability of the material to memorise a shape and materials that possess these properties have a multitude of exciting technical and medical applications (1–14). For materials such as alloys this is commonly in a one-way SME (7, 15), however, there are a variety of materials that are capable of reverting to their permanent shape or original state upon exposure to a stimulus (such as a temperature change) or indeed multiple stimuli (16). SMP-based materials have been widely investigated since the 1980s because of the abundance of potential applications imparted by their interesting properties (for instance, stimuli-responsiveness and ability to change shape), which can lead to technological innovation and the generation of new high value products for technical and medical applications (1, 17–19).

The reversible transformation of SMPs functions by primary crosslinking net points (hard segments) memorising and determining the permanent shape, and secondary switching segments (soft segments) with a transition (Ttrans) to reduce strain stress and hold the temporary shape. Below the Ttrans, the material will be in its permanent shape and be stiffer than when Ttrans is reached and the SMPs are more malleable and can be deformed into a desired shape (usually through application of an external force). The deformed state is maintained once the external force has been removed and the system is no longer at or above Ttrans. SMPs revert to their original state once the Ttrans conditions are met. This process describes the SME pathway of SMP-based materials that are thermally-induced (albeit not for some light or chemical-induced systems).While most SMP-based materials hold a single

permanent shape and a single temporary shape, recent advances in SMP technology have allowed the generation of multiple-shaped-memory materials with different stimuli responses (light or chemical) (16, 20, 21). An interesting example of this is a triple shaped-memory material generated by combining two dual SMPs with different glass transition temperatures (Tg) (22, 23), where the SMPs switch from one temporary shape to another at the first Ttrans, and then back to the permanent shape at another, higher activation temperature (22).SMPs have a large range of properties from stable

to biodegradable and transient, elastic to rigid or soft to hard, depending on the structural units that constitute the SMP. Consequently, SMPs not only respond to temperature (24) and magnetism (25) like shape-memory alloys (SMAs) (26), but also to moisture (27), electricity (28), light (29) and chemical stimuli (such as a pH change) (30). Moreover, there are other principles of SME; for instance, a thermal-responsive SMP can proceed

A Mini-Review of Shape-Memory Polymer-Based MaterialsStimuli-responsive shape-memory polymers



via a Diels-Alder reaction (chemical crosslinking/reversible covalent bonds) (31). SMPs tend to have much milder processing conditions than SMAs (<200°C, low pressure), have a greater extent of deformation (strain more than 200% for most materials) and tend to be based on cheap starting materials with simple synthetic procedures (12, 32). After the term ‘shape-memory’ was first proposed by Vernon in 1941 (32), the significance of SMPs was not fully realised until the 1960s, when crosslinked polyethylene (PE) was used to make heat-shrinkable tubes and films (33). Significant investment in the development of SMPs began in the 1980s (34) with rapid progress realised in the last decade, particularly with a view to the generation of shape-memory materials with exciting and versatile features.

Shape-Memory Polymer Function

Two important quantities used to describe SMEs are the strain recovery rate (Rr) and the strain fixity rate (Rf). Rr describes the ability of a material to memorise its permanent shape, while Rf describes the ability of switching segments to fix the mechanical deformation. Rr is calculated using Equation (i):

Rr(N) = × 100% (i) εm(N) – εp(N–1)

εm(N) – εp(N)

where N is the cycle number, εm is the maximum strain imposed on the material and εp is the strain of the sample after recovery. Rf is calculated using Equation (ii):

Rf(N) = × 100% (ii) εm(N)εu(N)

where εu is the strain in the fixed temporary shape. SMPs respond to specific stimuli through changes in their macroscopic properties (for example, shape) (26). The polymer network underlying active movement involves a dual system, one that is highly elastic and another that can reduce the stiffness upon application of a certain stimulus. The latter system incorporates either molecular switches or stimulus sensitive domains (35). Their shape-memory feature is a result of the combination of the polymer’s architecture, and a programming procedure that enables the formation of a temporary shape. Net points consist of covalent bonds or intermolecular interactions and the SMP’s

hard segments form the net points that link the soft segments (acting as a fixed phase), whereas the soft segments work as the molecular switches (acting as a reversible phase). The fixed phase prevents free flow of the surrounding polymer chains upon the application of stress. The reversible phase, on the other hand, undergoes deformation in a shape-memory cycle and is responsible for elasticity. For example, if the Ttrans is Tg, the micro-Brownian motion of the network chains is fixed at low temperature (below Tg) and will be switched back on at high temperature (above Tg), recovering its original state. When Ttrans is the crystal melting temperature (Tm), the switching segments crystallise at low temperature (below Tm), and then recover their original state at high temperature (above Tm). In addition, Tg normally extends over a broader temperature range compared to Tm, which tends to have relatively sharper transitions in most cases (26). Moreover, after the exposure to a specific stimulus and the Ttrans is achieved, the strain energy in the deformed state is released, resulting in the shape recovery phenomenon. The general process of this SME for SMPs is depicted in Figure 1, wherein the polymer network structure is either chemically or physically crosslinked and the switching units are made from a semi-crystalline or amorphous phase.Shape-memory behaviour can be demonstrated

in various polymer systems that are significantly different in molecular structure and morphology. SME mechanisms differ according to the specific SMP(s); for instance, the SME mechanism of the chemically crosslinked semi-crystalline PE SMP. The crystalline phase, with a Ttrans being Tm, is used as the molecular switching unit providing shape fixity. The chemically crosslinked PE network memorises the permanent shape after deformation upon heating (12, 36, 37), and the mechanism of the thermally-induced shape-memory PE (SMPE) is depicted in Figure 2.The associated modulus of elasticity is dictated

by configurational entropy reduction that occurs with deformation of the constituent chains and is therefore often termed entropy elasticity. For T>Ttrans (Tg, Tm or other), polymer networks exhibit super-elasticity wherein the polymer chain segments between crosslink points can deform quite freely and are prone to being twisted randomly via rotations about backbone bonds, maintaining a maximum entropy and minimum internal energy as macroscopic deformation occurs (12). The classic prediction from rubber



elastic theory is that the resulting elastic shear modulus (G) is proportional to both crosslink density and temperature (Equation (iii)):

G = vKBT = (iii) MC

pRT

where ν is the number density of network chains, p the mass density, R the universal gas constant and MC the molecular weight between crosslinks. From a macroscopic viewpoint, the SME in SMPs can be graphically represented in three-dimensions (3D). Tensile strain vs. temperature and tensile stress (for example, elongation) is depicted in Figure 3.Using the shape-memory strain-temperature-

stress relationship description in Figure 3, the features of SMPs that allow for good shape-memory behaviour include: a sharp transition that can be used to quickly fix the temporary shape at

low temperature, and the ability to trigger shape recovery at high temperature; super-elasticity above Ttrans that leads to the eventual shape recovery and avoids residual strain (permanent deformation); and complete and rapid fixing of the temporary shape by immobilising the polymeric chains without creep thereafter (12, 37). Thus far, the SME models describing how SMPs recover their original state prominently involve thermo-responsive SMPs. However, careful design of the polymers allows the opportunity for SMPs to possess different stimuli responses and applications.

Shape-Memory Polymer Triggers

A multitude of different triggers for SMEs and SMPs exist. However, an in-depth review is outside the scope of this mini-review, and therefore a few examples are highlighted below.

Fig. 1. (a) The general SME mechanism of SMPs; (b) thermally-responsive SMP

+ ∆conditions

+ ∆conditions

– ∆conditions

SMP

SMP

SMP SMP

SMP

Permanentshape

T<Ttrans

Deformation stage, mayrequire an external forcee.g. (elongation undertensile stress T>Ttrans)

Deformedtemporary shape

T>Ttrans

Permanentshape, shape

recoveryT>Ttrans

Deformed temporaryshape, if force appliedthen the external force

is removedT<Ttrans

External applied force,bending of polymer

Strain energystored

Strain energy released,polymers shape recovery

Heating, a hotplate

Cooling, taken offhot plate

Heating, a hotplate

Original stateT<Tg/Tm

T>Tg/Tm Deformed stateT<Tg/Tm

T>Tg/Tm Recovered shape,original state

(a)

(b)



Thermally-Induced Shape-Memory Polymer

It is possible to generate thermally-induced SMEs in a variety of materials (18–20, 38–40), however a comprehensive overview is outside the scope of this mini-review. As previously discussed, the SME of SMPs can be thermally-induced, and these SMPs are the most common (26). Figure 1 depicts a general overview of the SME mechanism of

SMPs, with a schematic of the SME mechanism for thermally-induced SMPs with Tg (amorphous cases) and Tm (crystalline cases). Figure 2 presents a specific example of the SME mechanism for SMPE with the Ttrans being Tm. In addition, advanced thermomechanical constitutive models have been used to study the materials’ behaviour (for example strain-temperature-stress development with time) in a very accurate way (41). By applying these models to SME mechanistic studies and the detailed characterisation of the SMPs (crosslinks, intermolecular and intramolecular interactions involving the SMPs) (12), a deeper understanding of the SME of SMPs can be achieved, which has proven beneficial for the development of new SMPs and their proposed applications (31). For example, poly(ε-caprolactone) (PCL), typically a biodegradable polymer, has been reported to possess high shape fixity and recovery. This was achieved by integrating reversible bonds within the PCL polymer network via the Diels-Alder addition of 1,2,4-triazoline-3,5-dione (TAD)-anthracene and Alder-ene addition of TAD-indole (42). These PCL SMPs were reported to attain recovery ratios greater than 99% (43). Furthermore, a dual-functional (self-healing and shape-memory) polymer network was achieved by crosslinking a polydimethylsiloxane (PDMS) polymer containing dense carboxylate groups (100% mol) (PDMS-COOH) with small amount of poly(ethylene glycol) diglycidyl ether

Fig. 2. Molecular model of the thermally-induced SME mechanism of crosslinked SMPE

PE cross-linkedby radiation

T>Tm

T>Tm

T<Tm

Heating at orabove Tm

Force ofstretching

Stretched at orabove Tm

Cooled at stretchedshape

Fig. 3. A general 3D plot of an SMP during a thermomechanical shape-memory cycle

200

150

100

50

0

Strain removal

Shape fixity

DeformationRecovery

Str

ain

(ε),

%

2040

6080 100

120 00.5

1.01.5

2.0

3.02.5

Stress

(σ),

MPa

Temperature (T), ˚C

εu

εm

εp*



(PEGDGE) (44). This SMP (PDMS-COO-E) actuates at body temperature (37°C) with possible strain ca. 200% and shape recovery ratios at 98.06%. In addition, a 25 mm × 4 mm × 1 mm sample cut into two separate pieces healed (the two pieces become one whole piece with no evidence of a cut) when the two cut surfaces were brought into contact after 6 h at 25°C. Thus, the unique material, PDMS-COO-E, may have a wide range of applications in many fields, including wearable electronics, biomedical devices and four-dimensional (4D) printing (1, 19). Interestingly, the material was also reported to possess a greater than 85% light transmittance (425 nm to 700 nm) (44), therefore PDMS-COO-E has potential applications in transparent electronic devices. Figure 4 illustrates the possible SME mechanism of PDMS-COO-E. The short PDMS linear chains are crosslinked by chemical covalent interactions and abundant hydrogen bonds into a 3D network. The covalent crosslinked networks of PDMS-COO-E maintain the permanent shape and resilience, whereas, at ca. 37°C the weak hydrogen bonds are broken, and the dynamics of polymer chains increase, resulting in recovering the permanent shape. Meanwhile, a large number of hydrogen bonds enable the samples to heal at temperature without external stimulus (44).

Light-Induced Shape-Memory Polymer

It is possible to generate light-induced SMEs in a variety of materials (18–20, 38, 40, 45), however a comprehensive overview is outside the scope of this mini-review. Light-activated SMPs (LASMPs) (46) typically use photothermal or photochemical (photocrosslinking or photocleavage) triggers for SMEs. For instance, photothermal LASMPs typically employ photo-absorber molecules and particles that convert light to heat, thereby increasing the temperature at the desired region within the LASMP. Photochemical LASMPs incorporate photosensitive molecules to create or cleave bonds during irradiation with light, imparting potentially very swift SMEs (47, 48). It is possible to improve the response time of SMPs by increasing the thermal conductivity with various conductive additives (49). However, the heating and cooling of materials with substantial thickness takes time, which can be minimised by using light to trigger transitions in LASMPs (46). It is also possible to generate multistimuli-responsive materials using components of the materials that respond to different wavelengths of light (for example, one wavelength of light to induce photocrosslinking, while a second wavelength of light cleaves bonds). It is possible to produce materials that can be reversibly switched between an elastomer and a rigid polymer employing polymers containing cinnamic groups (48) that can be fixed into pre-determined shapes utilising ultraviolet (UV) light illumination (>260 nm), and then recovered their original state when exposed to UV light at a different wavelength (<260 nm) (49). Figure 5 depicts one example of the process of LASMPs shape recoverability.

Electrically-Induced Shape-Memory Polymer

It is possible to generate electrically-induced SMEs in a variety of materials (18, 20, 50–55), however a comprehensive overview is outside the scope of this mini-review. A variety of electrically conductive materials including organic electronic materials (including conductive polymers such as polypyrrole (PPy) (28, 56–58) and carbon nanotubes (CNTs) (59, 60)) and inorganic electronic materials (such as alloys, metals (61) and silver nanowires (NWs)), have been incorporated in materials displaying SMEs to impart swift triggers to the SMEs, enabling a variety of interesting applications.

Fig. 4. The possible mechanism about shape memory effect of PDMS-COO-E polymer. Reprinted with permission from MDPI (44)

Cooling reshape

Heatingrecovery

Stretched

= PDMS = PEGGE

=O H O

O H O= –COOH



Highlighting some of the potential of electrically-induced SMEs, electrically-induced SMP composites incorporating shape-memory polyurethane (SMPU) and Ag NWs in a bilayer structure exhibits flexibility and electrical conductivity (62–64), which may find applications as capacitive sensors, healable transparent conductors and wearable electronics (65). In such materials the Ag NWs are randomly distributed on the surface layer of the composite to form a conductive percolating network that retains conductivity (200 Ω sq–1) after a 12% elongation. However, continual increase in elongation causes a dramatic increase to the composites’ resistance value and the eventual loss of electrical conductivity (66). When the material (deformed or in its original state), is connected to a typical circuit, a low voltage of 1.5 V was enough to activate a light-emitting diode (LED) (65). The composites possessing a higher Ag NW content exhibited a higher recovery ratio and reached the maximum recovery speed quicker (66). It was assumed that all the heat from electrical (Joule) heating was absorbed by the sample, i.e. no convective loss (67). Therefore, the composites with higher Ag NW content had a lower resistance value and the heating effectiveness was promoted. Heat initiates the thermal Ttrans of the SMPU leading to an improved shape recovery, and voltages as low as 5 V reverted bent composites to their original state within 3 s (66). This represents a good example of a multifunctional SMP and demonstrates the potential

of SMP designs driving technological innovation. A schematic of the composite is shown in Figure 6.Polymeric blend SMPs can be constructed from

two immiscible polymeric matrices. The shape-recovery of these systems can be controlled with relative ease by varying the ratio of the polymer blends (68). However, this process may have adverse effects on shape-memory characteristics and diminish the material’s performance, thereby limiting potential applications. On the other hand, SMP functionality may also be enhanced with other capabilities. For instance, it was recently reported that a new hybrid SMP was developed by combining single-walled CNTs (SWCNT) into a poly(lactic acid) (PLA) and thermoplastic polyurethane (TPU) SMP system, containing poly(ethylene glycol) (PEG) plasticiser (68). By incorporating PEG, the hybrid SMP composite achieved a lower temperature Tg (for example, 10 wt% of PEG lowered Tg of the PLA/TPU sample from 60°C to 40°C), meanwhile enhancing the dispersion of SWCNT (for instance, even at 4 wt% of SWCNT loading, 100% SMP tensile strain was possible, much greater than previously reported electrically-induced SMP studies, i.e. 12% discussed previously). In addition, the presence of the SWCNT can stabilise the SMP system and enhance its shape-fixity after deformations at room temperature conditions (68). Furthermore, the material was capable of a conductivity above 10–7 S cm–1, which can be considered conductive, as documented (68). The PLA/TPU SMP composite (2 wt% SWCNT and

Fig. 5. Schematic of an example of the SME function of LASMPs

hv2

hv1

Shape recovery

Permanent shape, elastomer

External force applied(e.g. elongation under

tensile stress)

Deformed temporary shape with external

force removed,rigid polymer

= elastomer network cross-link

= photo-cross-link site

= polymer chain backbone

A

B

C

D



10 wt% PEG) also achieved shape-recovery, via Joule heating derived from electricity, in 80 s when currents of 125 mA were applied. The high stiffness of SWCNT filler results in decreasing shape-recovery performance because of the hindrance on the polymer chain movements (68). As a result, under room temperature stretching, the Rf and Rr values obtained were ca. 80% and 65%, respectively. Therefore, when its shape-recoverability is compared to other SMPs (shape-recovery ratios being upwards of 98%), the material is lacking. However, the hybrid SMP composite does possess electroactive ability, thus a trade-off relationship between shape-memory/recovery and electroactive ability needs to be carefully considered when designing similar materials.

Water-Induced Shape-Memory Polymer

It is possible to generate water-induced SMEs in a variety of materials (18, 20, 38, 39, 69–72),

however a comprehensive overview is outside the scope of this mini-review. Water is an important stimulus due to the fact it is abundant in a multitude of different environments, non-toxic and safe for a variety of applications.An interesting example highlighting the

potential of such materials is based on strong and flexible composite films (73) utilising the combination of a flexible interpenetrating polyol-borate network (74) and electroactive PPy (75, 76) that exchange water with the environment resulting in film expansion or contraction. The free-standing multi-functional SMP films were prepared by electropolymerisation of pyrrole in the presence of the polyol-borate complex (composed of pentaerythritol ethoxylate (PEE) coordinated to boron(III)) (74), wherein the interpenetrating network enables water-gradient-induced displacement, converting chemical potential energy in water gradients to mechanical work (73), and results in adaptation of the

Fig. 6. (a) transmission electron microscopy (TEM) image of Ag NWs; (b) atomic force microscopy (AFM) image of Ag NWs; (c) schematic illustration of composites fabrication process; (d) the LED turned on as the composite was applied with voltages (the inset shows the circuit connecting with the composites). Reprinted with permission. Copyright 2014 Elsevier (66)

(a) (b)

412.7 nm500 nm6.7

20

20 um

15

15

5

10

15

20 um

10

105

Release substrate Release substrate

Spin coatingAgNWs

AgNWs

Coating SMPlayer

Peel off

Release substrateSMP layer

5

(c) (d)

Power

LEDCompositeelectrode

LED

Conductivecomposite

SMP layer

AgNW layerAgNW layer



architecture in response to an environmental condition change (i.e. sorption and desorption of water which drives the SME process, as depicted in Figure 7). The design of the water-responsive PPy-PEE composites was creatively applied to prepare actuators and generators driven by water gradients. The film actuator can generate contractile stress up to 27 MPa, lift objects 380 times heavier than itself and transport cargo 10 times heavier than itself (73). An assembled generator associating the actuator with a piezoelectric element driven by water gradients, outputs alternating electricity at ca. 0.3 Hz, with a peak voltage of ca. 1 V (73). The electrical energy can be stored in capacitors that could power micro and nanoelectronic devices (73). The SME mechanism for this SMP differs to that of Figure 1 and Figure 2, utilising water as the shape-memory trigger for Ttrans, and the original and deformed state interchange automatically via water sorption and desorption states. However, the shape-memory phenomenon remains the same, further demonstrating the potential of SMP designs driving technological innovation.

pH-Induced Shape-Memory Polymer

It is possible to generate pH-induced SMEs in a variety of materials (18, 20, 38, 77–80), however a comprehensive overview is outside the scope of this mini-review. An example of the interesting properties of such pH-responsive SMPs and their composites is produced by blending poly(ethylene glycol)-poly(ε-caprolactone)-based polyurethane (PECU) with functionalised cellulose nanocrystals (CNCs) displaying pH responsive pyridine moieties (CNC-C6H4NO2) (81, 82). At high pH values the pyridine is deprotonated, facilitating hydrogen bonding interactions between the pyridine groups and hydroxyl moieties on the cellulose, whereas at low pH values, the protonation of the pyridine moieties diminishes these interactions. By comparison, carboxylic acid functionalised cellulose nanocrystals (CNC-CO2H) responded to pH variation in the opposite manner (83–85). When the functionalised CNCs were combined with PECU polymer matrix to form a nanocomposite network, the mechanical properties of PECU were improved along with the pH-responsiveness of CNCs (85).

Fig. 7. Design and performance of a water-gradient–driven generator: (a) the assembly of a piezoelectric polyvinylidene fluoride (PVDF) element with a PEE-PPy actuator to form the generator; (b) the connection of the generator with a 10 MW resistor as load; (c) the configuration of the rectifying circuit and charge storage capacitor; (d) the generator’s output voltage onto the 10 MW resistor; (e) voltage across a capacitor when being charged by the generator. The inset shows a stepwise increase in the capacitor voltage accompanying each cycle of the energy conversion process. Reprinted with permission. Copyright 2013 The American Association for the Advancement of Science (73)

Metalelectrode

Insulatinglayer

PVDF

Wire

PEE-PPy

Voutput

Vcap

Vrec

RL

VG

1.5

1.0

0 0

0.7

(a)

(d) (e)

(b) (c)

0.6

0.5

0.4

0.3

0.2

0.1

20

Volta

ge,

V

Volta

ge,

V

Volta

ge,

V

40 60Time, s Time, s

Time, s

80 100 100 200

0.55

0.50290 310

300 400120

0.5

0.0

–0.5

–1.0

–1.5

G

G



The percolated network of pH-sensitive CNC in the polymer matrix served as the switching units for the shape-memory composite, the SME process of this material is depicted in Figure 8 (81, 82). The CNC serves as the switching unit of the SMP composite within the matrix of PECU which is physically crosslinked and microphase separated to yield the net points. Such pH-responsive shape-memory nanocomposites have promise in the design of biomaterials for biomedical applications (for example, SMP-based drug delivery systems

triggered by transition along the digestive tract) (83).

Magnetically-Induced Shape-Memory Polymer

It is possible to generate magnetically-induced SMEs in a variety of materials (18, 20, 38, 86– 88), however a comprehensive overview is outside the scope of this mini-review. The SMP devices discussed thus far are being researched with potential application into wearable electronics, nanoelectronics (such as actuators), biomaterials and biomedical devices (1, 18, 19). However, in some instances (such as medical devices) a key challenge is the design and implementation of a safe and effective method of actuating a variety of device geometries in vivo. As previously discussed, a pH-triggered SMP design can be potentially effective when utilised as drug delivery devices, when the target environment has a substantial pH difference (for instance, the digestive system) (83). However, the development of electrically and thermally-triggered devices that safely operate in vivo is difficult due to the (generally) high temperatures these SMPs can reach (relative to biological systems). For instance, the electroactive PLA-TPU SMP composite (2 wt% SWCNT and 10 wt% PEG) reaches temperatures greater than 70°C in 80 s as shape-recovery is achieved (68).An alternative method of achieving actuation is

inductive heating by loading ferromagnetic particles into an SMP system and exposing the doped device to an alternating electromagnetic field (89), benefiting from the innate thermoregulation offered by a ferromagnetic material’s Curie temperature (Tc, at which a ferromagnetic material becomes paramagnetic, losing its ability to generate heat via a hysteresis loss mechanism) (90). By using particle sizes and materials that will heat mainly via a magnetic hysteresis loss mechanism over an eddy current mechanism, it is possible to have an innate thermoregulation mechanism that limits the maximum achievable temperature to Tc (89). Therefore, by selecting ferromagnetic particle materials with a Tc within safe medical limits, Curie thermoregulation eliminates the danger of overheating and the need for a feedback system to monitor implanted device temperatures (89). However, this technology is not only useful when applied to medical devices. Other useful applications include remote activation in which wires or connections to SMP devices could be eliminated, simplifying the design and reducing possible points

Fig. 8. Schematic representation of the pH-responsive shape-memory materials, which rely on hydrogen bonding switching mechanism in the interactions between cellulose nanocrystals (CNC–C6H4NO2) within polymer matrix upon immersion in hydrochloric acid solution (pH = 4) or sodium hydroxide solution (pH = 8). Reprinted with permission. Copyright 2015 American Chemical Society (81)

DeformationpH = 8 pH = 4

pH = 4 pH = 8

Force

Fixity

Recovery

PECU matrix in solution Hydrogen bonding

C

OCNC–C6H4NO2 CNC NH++



of failure. An example of this method of actuation involves the incorporation of 10% by volume nickel zinc ferrites (for example C2050 (Ceramic Magnetics Inc, USA) and CMD5005 (Ceramic Magnetics Inc), particle sizes ca. 50 µm with spherical shapes) with an ester-based thermoset polyurethane (PU) SMP, MP5510 (SMP Technologies Inc, Japan) (Tg of 55°C) (89). The magnetic field utilised to achieve shape-recovery was a copper-wound solenoid coil with a 2.54 cm diameter, 7.62 cm length and with a total of 7.5 turns. The unit possessed an adjustable power setting capable of outputting 27 W to 1500 W at between 10 MHz and 15 MHz frequency (note: this high frequency may induce eddy currents in the tissue, causing undesirable direct heating of the human body in medical applications) (91). However, an alternating magnetic field of 12.2 MHz and approx. 400 A m–1 (centre of the inductive coil) at room temperature was used for actuation to demonstrate proof of concept for the device. It was also reported that clinically useable frequencies (50 kHz to 100 kHz) (92) should still be effective (89), albeit this could result at a different quantitative level (i.e. shape-recovery and memory performance may be reduced). Furthermore, C2050 and CMD5005 possess a Tc of 340°C and 130°C, respectively. These temperatures exceed physiological limits and are therefore not practical for medical devices currently, however, these doped SMP

composites did not exceed temperatures above the respective Ni Zn particle Tc values, signifying a thermoregulation characteristic. In addition, it was stated that the 10% volume of Ni Zn particles did not impact the SMPs shape-memory properties significantly (89). The Tg increased from 55°C to 61.4°C and the shape-recovery of a flower and foam-based device was achieved within 15 s to 25 s, at a temperature range of 23°C to 78.6°C. The potential applications for this device are illustrated in Figure 9. Optimisation of this device/design is still required before it can be considered clinically viable, however, this SMP composite highlights very interesting characteristics, remote activation (via magnetic fields inducing thermally-triggered actuation) and thermoregulation (via Tc temperature of the material being employed).

Shape-Memory Polymer Classification

As highlighted above, SMP materials are diverse and respond to many different external stimuli (including temperature, light, electricity, water, pH and electromagnetic fields) by a variety of mechanisms. Although SMPs can be classified based on their composition and structure, stimulus and shape-memory function, their classification can be difficult, as organising these polymeric smart materials into one or two simple categories

Fig. 9. SMP devices used to evaluate feasibility of actuation by inductive heating: (a) flower shaped device shown in collapsed and actuated form; (b) SMP foam device shown in collapsed and actuated form. Reproduced with permission. Copyright 2006 IEEE Transactions on Biomedical Engineering (89)

1.5 mm

(a)

(b)

5 mm



is an over-simplification of their abilities and characteristics (93). SMPs are considered to consist of net points and

molecular switches or stimuli sensitive domains. These net points can be achieved by covalent bonds (chemically crosslinked) or intermolecular interactions (physically crosslinked). Chemically crosslinked SMPs involve suitable crosslinking chemistry and are referred to as thermosets (94, 95). Physically crosslinked SMPs involve a polymer morphology consisting of at least two segregated domains and are referred to as thermoplastics (96). The network chains of the SMP can be either amorphous or crystalline and therefore, the Ttrans is either a Tg or Tm. The network architectures are thought to be constructed through crosslinking net points, with polymer segments connecting adjacent net points. The strongly crosslinked architectures ensure the polymer can maintain a stable shape on the macroscopic level (93). Thermoplastic polymers exhibit a more reversible nature (97), meaning the physical crosslinked net points can be disrupted and reformed with relative ease. The interconnection of the individual polymer chains in a physically crosslinked network is achieved by the formation of crystalline or glassy phases. For thermoset polymers, the individual polymer chains are connected by covalent bonds and are therefore more stable than physically crosslinking networks and show an irreversible nature (98–100).Regarding thermo-responsive SMPs, they can

be classified according to the nature of their permanent net points and the Ttrans related to the switching domains into four different categories: (a) physically crosslinked thermoplastics, Ttrans = Tg; (b) physically crosslinked thermoplastics, Ttrans = Tm; (c) chemically crosslinked amorphous polymers, Ttrans = Tg; (d) chemically crosslinked semi-crystalline polymer networks Ttrans = Tm (93).

Thermoplastic Shape-Memory Polymers

For the physically crosslinked SMPs, the formation of a phase-segregated morphology is the fundamental mechanism behind the thermally-induced SME of these materials (93, 99). One phase provides the physical crosslinks while the other acts as a molecular switch. They can be further classified into linear polymers, branched polymers or a polymer complex. Linear SMPs may consist of block copolymers and high molecular weight polymers, the typical physically crosslinked SMP is linear block copolymers, such as PU. In

polyesterurethanes (PEU), oligourethane segments are the hard-elastic segments, while polyester serves as the switching segment (99).

Thermoset Shape-Memory Polymers

For chemically crosslinked SMPs, two methods are commonly used to synthesise covalently crosslinked networks (36, 41). The first method relies on addition of a multi-functional crosslinker during polymerisation (41), whereas the second method relies on the subsequent crosslinking of a linear or branched polymer (36). The networks are formed based on many different polymer backbones. Covalently crosslinked SMPs possess chemically interconnected structures determining the original macroscopic shape. The switching segments of these materials are generally the network chains between net points, and a Ttrans of the polymer segments is used as the shape-memory switch. The chemical, thermal, mechanical and shape-memory properties are determined by the reaction conditions, curing times, the type and length of the network chains and the crosslinking density (35). Comparing physically crosslinked SMPs with chemically crosslinked SMPs, the chemically crosslinked SMPs often show less creep, thus, any irreversible deformation of the polymer during shape recovery is less. This is because covalent crosslinked networks are more stable than physical crosslinked networks. As a result, chemically crosslinked SMPs usually show better chemical, thermal, mechanical and shape-memory properties than physically crosslinked SMPs (96). For example, the shape recovery ratio of thermoplastic SMPU is usually in the range of 90% to 95% within 20 shape recovery cycles, and the elastic modulus is between 0.5 GPa and 2.5 GPa at room temperature (26). Additionally, when exposed to air, it is sensitive to moisture and therefore possesses unstable mechanical properties. In contrast, an epoxy SMP shows better overall performance as a shape-memory material. The shape recovery ratio typically reaches 98–100%, the elastic modulus between 2 GPa and 4.5 GPa, and it is generally stable in the presence of moisture (26). Thermoplastic SMPs (such as SMPU) are mostly researched and used as functional materials at a small scale, such as for biomaterials (30, 97). However, thermosetting SMPs (for example styrene-based SMP (SSMP) and epoxy SMPs) are generally used for structural materials, such as space deployable structures and automobile actuators (97, 98).



Shape-Memory Functionality

The approaches to designing different shape-memory functions become more abundant as scientists and engineers better understand the SME mechanism of SMPs. For instance, discussed thus far are examples of SMPs with polymeric blends, addition of crosslinking species, incorporation of electroactive and ferromagnetic substances. All of which enhances an SMP device functionality, enabling unique and interesting characteristics which can be tailored to a plethora of applications (for example, self-healing and wearable electronics, drug delivery and implantable medical devices) (101–110). Further still, one-way SMEs, two-way SMEs (such as dual shape PPy-PEE, discussed previously), triple SMEs, multiple SMEs and even temperature-memory effects (TMEs) have been widely investigated in SMPs (34). As the types of SMP materials increasingly diversify, two and even three different types of shape-memory functions can be achieved simultaneously in the same SMP material (34, 111). These types of materials can usually be achieved when combining different SMPs possessing different properties. A schematic of one-way, two-way, dual shape and triple shape functionality SMPs is shown in

Figure 10, and an integrated insight into the classification of SMPs is shown in Figure 11.An example of a selective triple shape

multicomposite SMP was documented to incorporate a neat SSMP (112) and two SSMP composites (113). One incorporated iron(II, III) oxide nanoparticles while the other CNT nanoparticles. This unique SMP composite successfully possessed three different regions within the sample: neat SSMP, SSMP-Fe3O4 and SSMP-CNT. Because of this, the material also possessed distinct shape-memory capabilities with different triggers. For instance, the material was documented undergoing a three-step shape-memory recovery process, subjected to an alternating magnetic field of 30 kHz, a radio frequency (RF) field of 13.56 MHz and direct oven heating at 130°C (113). Furthermore, the Rf and Rr for the original shape to the first temporary shape (and back to the original shape) was reported at 93% and 93%, respectively. Meanwhile, the Rf and Rr for the first temporary shape to the second temporary shape (and back to the first temporary shape) was at 95% and 99%, respectively (113). The SME mechanism for this multicomposite is represented in Figure 12 and it was concluded that this unique material has promising characteristics to be used in biomimetic materials. Examples of

Fig. 10. The varying shape-memory functionality of SMPs

One-way SME

Dual shape

Triple shape

Originalshape

Deformedshape

maintained

Deformedshape1

maintained Deformedshape2

maintained

Deformedshape1 Deformed

shape2

>Ttrans

>Ttrans

>Ttrans1 >Ttrans2

>Ttrans2

>Ttrans2

<Ttrans2>Ttrans1

>Ttrans1

<Ttrans1

>Ttrans

<TtransOriginalshape

Originalshape

Deformedshape

Originalshape

Two-way SME



applications of SMP-based materials and their composites are highlighted in Table I.

Conclusion

As the understanding of SMPs continually develops among the academic and industrial communities, the generation of new and potentially innovative SMPs will be more rapid while we realise the full potential of these materials. SMPs are one of the most interesting of polymer classes within the field of functional polymers. In addition, SMP composites can enhance the already impressive capabilities of

SMPs by imparting new functional characteristics, broadening the potential applications of these materials and enabling a multipurpose material. SMPs and their composites are capable of industrially important applications (examples of which include: self-healing (101–104), generators driven by water gradients (73), sensors (72), task-specific medical devices (18, 105) and wearable electronics (106–110), a few examples of which are highlighted in Table I. The literature published to date de-risks investment from governments and industry to raise the technology readiness levels towards products on the market.

Fig. 11. The classification of SMPs based on composition and structure, stimulus triggers and the possible type of shape-memory functions

Composition and structure

Block copolymer

Supramolecular polymerELECTRICITY

WATER SENSITIVE

TEMPERATURE

Stimulus

Thermo-sensitive

Water-sensitive

Redox-sensitive

Light-sensitive

ONE-WAY SME

TWO-WAY SME

TRIPLE SHAPE SME

MULTI-SHAPE SME

MULTI-FUNCTIONALITY

Shape-memory function type

MAGNETIC

OXIDATION-REDUCTION

LIGHT/RADIATION

Polymer blend/composites

Cross-linked homopolymer

Polymer IPN/semi-IPN

e.g. PPy-PEEdiscussed in

Figure 7



Fig. 12. Schematic of the selective shape recoveries of the multicomposite SSMP induced by alternating magnetic field heating, RF field heating and oven heating, respectively (an, bn and cn stand for the n sections of SSMP–Fe3O4, neat SSMP and SSMP–CNT, respectively). Reproduced by permission of The Royal Society of Chemistry. Copyright 2015 The Royal Society of Chemistry (113)

a1 a2 c2c1b1 b2

a1 a2 c2c1

c

b1 b2

a1

a1

a2

a2

c2

c2

c1

c1

b1

b1

b2

b2

a1 a2 c2c1

b1 b2

a

SSMP-Fe3O4

Original shape

Temporary shape C

Temporary shape A

Temporary shape B

Recovered shape SSMP-CNT

Deformedat 130�C

Fixed at20�C

Oven heating130�C

RF field heating

13.56 MHz

Alternating magneticfield heating of 30 kHz

Neat SSMP

b

an, bn, cn

Table I Examples of Applications of SMP-Based Materials and Their CompositesApplication ReferencesActuators (for example, for generators) (73)

Biomedical devices (such as drug delivery systems, expanding foam and endovascular thrombectomy device) (44, 83, 89)

Multipurpose/multifunctionality (for example, self-healing, biocompatible, body temperature actuation and selective triple shape-memory) (44, 113)

Thermoregulators (89, 90)

Wearable electronics (65, 68)

Acknowledgements

We acknowledge the Faculty of Science and Technology, Lancaster University, UK, for an Early Career Internal Grant and The Royal Society, UK, for a Research Grant (RG160449). We acknowledge the Engineering and Physical Sciences Research Council (EPSRC), UK, for an EPSRC First Grant (EP/R003823/1) and a Pathfinder Grant from the EPSRC Centre for Innovative Manufacturing in Large Area Electronics (EP/K03099X/1). We also thank the Biotechnology and Biological Sciences Research Council (BBSRC), UK, for financial support of

research aligned with some of the topics discussed in this review, specifically the Glycoscience Tools for Biotechnology and Bioenergy (IBCarb) Network in Industrial Biotechnology and Bioenergy (NIBB, BB/L013762/1); the BBSRC FoodWasteNet NIBB (BB/L0137971/1), the BBSRC From Plants to Products (P2P) NIBB (BB/L013819/1) and the Lignocellulosic Biorefinery Network (LBNet) NIBB (BB/L013738/1).

References

1. A. Lendlein and O. E. C. Gould, Nat. Rev. Mater., 2019, 4, (2), 116



2. T. Biggs, M. B. Cortie, M. J. Witcomb and L. A. Cornish, Platinum Metals Rev., 2003, 47, (4), 142

3. D. Kapoor, Johnson Matthey Technol. Rev., 2017, 61, (1), 66

4. Y. V. Kudriavtsev and E. L. Semenova, Platinum Metals Rev., 2014, 58, (1), 20

5. R. Oshima, S. Muto and T. Hamada, Platinum Metals Rev., 1988, 32, (3), 110

6. J. M. Jani, M. Leary, A. Subic and M. A. Gibson, Mater. Des., 2014, 56, 1078

7. C. Naresh, P. S. C. Bose and C. S. P. Rao, ‘Shape Memory Alloys: A State of Art Review’, International Conference on Advances in Materials and Manufacturing Applications (IConAMMA-2016), Bangalore, India, 14th–16th July, 2016, IOP Conference Series: Materials Science and Engineering, Vol. 149, IOP Publishing Ltd, Bristol, UK, 2016

8. D. Patil and G. Song, Smart Mater. Struct., 2017, 26, (9), 093002

9. N. Ma, Y. Lu, J. He and H. Dai, J. Text. Inst., 2019, 110, (6), 950

10. C. Wen, X. Yu, W. Zeng, S. Zhao, L. Wang, G. Wan, S. Huang, H. Grover and Z. Chen, AIMS Mater. Sci., 2018, 5, (4), 559

11. W. M. Huang, Z. Ding, C. C. Wang, J. Wei, Y. Zhao and H. Purnawali, Mater. Today, 2010, 13, (7–8), 54

12. C. Liu, H. Qin and P. T. Mather, J. Mater. Chem., 2007, 17, (16), 1543

13. Y. Liu, H. Du, L. Liu and J. Leng, Smart Mater. Struct., 2014, 23, (2), 023001

14. W. Sokolowski, A. Metcalfe, S. Hayashi, L. Yahia and J. Raymond, Biomed. Mater., 2007, 2, (1), S23

15. P. K. Kumar and D. C. Lagoudas, ‘Introduction to Shape Memory Alloys’, in “Shape Memory Alloys”, ed. D. C. Lagoudas, Springer, Boston, USA, 2008, pp. 1–51

16. K. Yu, T. Xie, J. Leng, Y. Ding and H. J. Qi, Soft Matter, 2012, 8, (20), 5687

17. J. G. Hardy, M. Palma, S. J. Wind and M. J. Biggs, Adv. Mater., 2016, 28, (27), 5717

18. K. Wang, S. Strandman and X. X. Zhu, Front. Chem. Sci. Eng., 2017, 11, 143

19. M. Behl and A. Lendlein, Mater. Today, 2007, 10, (4), 20

20. H. Meng and G. Li, Polymer, 2013, 54, (9), 2199

21. L. Sun, W. M. Huang, Z. Ding, Y. Zhao, C. C. Wang, H. Purnawali and C. Tang, Mater. Des., 2012, 33, 577

22. I. Bellin, S. Kelch, R. Langer and A. Lendlein, Proc. Natl. Acad. Sci., 2006, 103, (48), 18043

23. F. Pilate, A. Toncheva, P. Dubois and J.-M. Raquez, Eur. Polym. J., 2016, 80, 268

24. F. Ji, Y. Zhu, J. Hu, Y. Liu, L.-Y. Yeung and G. Ye, Smart Mater. Struct., 2006, 15, (6), 1547

25. R. Mohr, K. Kratz, T. Weigel, M. Lucka-Gabor, M. Moneke and A. Lendlein, Proc. Natl. Acad. Sci., 2006, 103, (10), 3540

26. J. Leng, X. Lan, Y. Liu and S. Du, Prog. Mater. Sci., 2011, 56, (7), 1077

27. B. Yang, W. M. Huang, C. Li and L. Li, Polymer, 2006, 47, (4), 1348

28. N. G. Sahoo, Y. C. Jung and J. W. Cho, Mater. Manuf. Processes, 2007, 22, (4), 419

29. A. Lendlein, H. Jiang, O. Jünger and R. Langer, Nature, 2005, 434, 879

30. A. Lendlein and R. Langer, Science, 2002, 296, (5573), 1673

31. K. Inoue, M. Yamashiro and M. Iji, J. Appl. Polym. Sci., 2009, 112, (2), 876

32. L. B. Vernon and H. M. Vernon, The Vernon Benshoff Company, ‘Process of Manufacturing Articles of Thermoplastic Synthetic Resins’, US Patent 2,234,993; 1941

33. W. C. Rainer, E. M. Redding, J. J. Hitov, A. W. Sloan and W. D. Stewart, W. R. Grace & Co, ‘Polyethylene Product and Process’, US Patent 3,144,398; 1964

34. J. Hu, Y. Zhu, H. Huang and J. Lu, Prog. Polym. Sci., 2012, 37, (12), 1720

35. A. Lendlein and S. Kelch, Angew. Chem. Int. Ed., 2002, 41, (12), 2034

36. S. Ota, Radiat. Phys. Chem., 1981, 18, (1–2), 81

37. Q. Zhao, H. J. Qi and T. Xie, Prog. Polym. Sci., 2015, 49–50, 79

38. X. Wu, W. M. Huang, Y. Zhao, Z. Ding, C. Tang and J. Zhang, Polymers, 2013, 5, (4), 1169

39. Y. Bai, J. Zhang and X. Chen, ACS Appl. Mater. Interfaces, 2018, 10, (16), 14017

40. K. Yu, Y. Liu and J. Leng, RSC Adv., 2014, 4, (6), 2961

41. I. A. Rousseau and P. T. Mather, J. Am. Chem. Soc., 2003, 125, (50), 15300

42. J. Caprasse, T. Defize, R. Riva and C. Jérôme, ‘Comparative Study of PCL Shape-Memory Networks with Diels-Alder or Alder-ene Adducts’, Advanced Functional Polymers for Medicine (AFPM), Montpellier, France, 16th–18th May, 2018



43. T. Defize, R. Riva, J.-M. Raquez, P. Dubois, C. Jérôme and M. Alexandre, Macromol. Rapid Commun., 2011, 32, (16), 1264

44. H.-Y. Lai, H.-Q. Wang, J.-C. Lai and C.-H. Li, Molecules, 2019, 24, (18), 3224

45. D. Iqbal and M. H. Samiullah, Materials, 2013, 6, (1), 116

46. D. J. Maitland, M. F. Metzger, D. Schumann, A. Lee and T. S. Wilson, Lasers Surg. Med., 2002, 30, (1), 1

47. H. Xie, K.-K. Yang and Y.-Z. Wang, Prog. Polym. Sci., 2019, 95, 32

48. Z. Yuan, A. Muliana and K. R. Rajagopal, Math. Mech. Solids, 2016, 22, (5), 1116

49. E. Havens, E. A. Snyder and T. H. Tong, ‘Light-Activated Shape Memory Polymers and Associated Applications’, SPIE Smart Structures and Materials + Nondestructive Evaluation and Health Monitoring, San Diego, California, USA, 5th May, 2005, “Smart Structures and Materials 2005: Industrial and Commercial Applications of Smart Structures Technologies”, Vol. 5762, Society of Photo-Optical Instrumentation Engineers (SPIE), Bellingham, USA, 2005, 8 pp

50. Y. Liu, H. Lv, X. Lan, J. Leng and S. Du, Compos. Sci. Technol., 2009, 69, (13), 2064

51. H. Lu, Y. Yao and L. Lin, Pigm. Resin Technol., 2014, 34, (1), 26

52. J. Alam, A. Khan, M. Alam and R. Mohan, Materials, 2015, 8, (9), 6391

53. J. Zhang, X. Ke, G. Gou, J. Seidel, B. Xiang, P. Yu, W.-I. Liang, A. M. Minor, Y. Chu, G. Van Tendeloo, X. Ren and R. Ramesh, Nat. Commun., 2013, 4, 2768

54. X. Gong, L. Liu, Y. Liu and J. Leng, Smart Mater. Struct., 2016, 25, (3), 035036

55. J. Zhou, H. Li, R. Tian, R. Dugnani, H. Lu, Y. Chen, Y. Guo, H. Duan and H. Liu, Sci. Rep., 2017, 7, 5535

56. S.-K. Lee, S.-J. Lee, H.-J. An, S.-E. Cha, J.-K. Chang, B. Kim and J. J. Pak, ‘Biomedical Applications of Electroactive Polymers and Shape-Memory Alloys’, SPIE’s 9th Annual International Symposium on Smart Structures and Materials, San Diego, USA, 11th July, 2002, “Smart Structures and Materials 2002: Electroactive Polymer Actuators and Devices (EAPAD)”, Vol. 4695, Society of Photo-Optical Instrumentation Engineers (SPIE), Bellingham, USA, 2002, 15 pp

57. N. G. Sahoo, Y. C. Jung, H. J. Yoo and J. W. Cho, Compos. Sci. Technol., 2007, 67, (9), 1920

58. N. G. Sahoo, Y. C. Jung, N. S. Goo and J. W. Cho, Macromol. Mater. Eng., 2005, 290, (11), 1049

59. G. Zhou, H. Zhang, S. Xu, X. Gui, H. Wei, J. Leng, N. Koratkar and J. Zhong, Sci. Rep., 2016, 6, 24148

60. M. Dahmardeh, M. S. M. Ali, T. Saleh, T. M. Hian, M. V. Moghaddam, A. Nojeh and K. Takahata, Phys. Status Solidi A, 2013, 210, (4), 631

61. H. B. Gilbert and R. J. Webster, IEEE Robot. Autom. Lett., 2016, 1, (1), 98

62. M. Xie, L. Wang, J. Ge, B. Guo and P. X. Ma, ACS Appl. Mater. Interfaces, 2015, 7, (12), 6772

63. H. Tanaka and K. Honda, J. Polym. Sci. Pol. Chem., 1977, 15, (11), 2685

64. H. Luo, J. Hu and Y. Zhu, Mater. Letters, 2012, 89, 172

65. C. Gong, J. Liang, W. Hu, X. Niu, S. Ma, H. T. Hahn and Q. Pei, Adv. Mater., 2013, 25, (30), 4186

66. H. Luo, Z. Li, G. Yi, X. Zu, H. Wang, Y. Wang, H. Huang, J. Hu, Z. Liang and B. Zhong, Mater. Letters, 2014, 134, 172

67. T. Akter and W. S. Kim, ACS Appl. Mater. Interfaces, 2012, 4, (4), 1855

68. Y.-C. Sun, M. Chu, M. Huang, O. Hegazi and H. E. Naguib, Macromol. Mater. Eng., 2019, 304, (10), 1900196

69. Y. Guo, Z. Lv, Y. Huo, L. Sun, S. Chen, Z. Liu, C. He, X. Bi, X. Fan and Z. You, J. Mater. Chem. B, 2019, 7, (1), 123

70. I. T. Garces, S. Aslanzadeh, Y. Boluk and C. Ayranci, Materials, 2019, 12, (2), 244

71. K. Fan, W. M. Huang, C. C. Wang, Z. Ding, Y. Zhao, H. Purnawali, K.C. Liew and L. X. Zheng, eXPRESS Polym. Lett., 2011, 5, (5), 409

72. L. Sun, T. X. Wang, H. M. Chen, A. V. Salvekar, B. S. Naveen, Q. Xu, Y. Weng, X. Guo, Y. Chen and W. M. Huang, Polymers, 2019, 11, (6), 1049

73. M. Ma, L. Guo, D. G. Anderson and R. Langer, Science, 2013, 339, (6116), 186

74. M. Shibayama, M. Sato, Y. Kimura, H. Fujiwara and S. Nomura, Polymer, 1988, 29, (2), 336

75. E. Smela, Adv. Mater., 2003, 15, (6), 481

76. R. H. Baughman, Science, 2005, 308, (5718), 63

77. Q. Song, H. Chen, S. Zhou, K. Zhao, B. Wang and P. Hu, Polym. Chem., 2016, 7, (9), 1739

78. H. Xiao, C. Ma, X. Le, L. Wang, W. Lu, P. Theato, T. Hu, J. Zhang and T. Chen, Polymers, 2017, 9, (4), 138

79. X.-J. Han, Z.-Q. Dong, M.-M. Fan, Y. Liu, J.-H. Li, Y.-F. Wang, Q.-J. Yuan, B.-J. Li and S. Zhang, Macromol. Rapid Commun., 2012, 33, (12), 1055

80. J. Li, Q. Duan, E. Zhang and J. Wang, Adv. Mater. Sci. Eng., 2018, 7453698



81. Y. Li, H. Chen, D. Liu, W. Wang, Y. Liu and S. Zhou, ACS Appl. Mater. Interfaces, 2015, 7, (23), 12988

82. T. Wu, Y. Su and B. Chen, ChemPhysChem, 2014, 15, (13), 2794

83. H. Chen, Y. Li, Y. Liu, T. Gong, L. Wang and S. Zhou, Polym. Chem., 2014, 5, (17), 5168

84. K. H. M. Kan, J. Li, K. Wijesekera and E. D. Cranston, Biomacromolecules, 2013, 14, (9), 3130

85. A. E. Way, L. Hsu, K. Shanmuganathan, C. Weder and S. J. Rowan, ACS Macro. Lett., 2012, 1, (8), 1001

86. N. Gabdullin and S. H. Khan, ‘Review of Properties of Magnetic Shape Memory (MSM) Alloys and MSM Actuator Designs’, 2014 Joint IMEKO TC1-TC7-TC13 Symposium: Measurement Science Behind Safety and Security, Madeira, Portugal, 3rd–5th September, 2014, Journal of Physics: Conference Series, Vol. 588, IOP Publishing Ltd, Bristol, UK, 2015, 6 pp

87. E. Faran and D. Shilo, Exp. Tech., 2016, 40, (3) 1005

88. J. Karger-Kocsis and S. Kéki, Polymers, 2018, 10, (1), 34

89. P. R. Buckley, G. H. McKinley, T. S. Wilson, W. Small, W. J. Benett, J. P. Bearinger, M. W. McElfresh and D. J. Maitland, IEEE Trans. Biomed. Eng., 2006, 53, (10), 2075

90. A. Goldman, “Modern Ferrite Technology”, Springer Verlag, New York, USA, 2006

91. P. R. Stauffer, T. C Cetas, A. M. Fletcher, D. W. Deyoung, M. W. Dewhirst, J. R. Oleson and R. B. Roemer, IEEE Trans. Biomed. Eng., 1984, BME-31, (1), 76

92. A. Jordan, R. Scholz, P. Wust, H. Fähling and R. Felix, J. Magn. Magn. Mater., 1999, 201, (1–3), 413

93. X. Fu, Y. Yuan, Z. Liu, P. Yan, C. Zhou and J. Lei, Eur. Polym. J., 2017, 93, 307

94. A. M. Kushner, J. D. Vossler, G. A. Williams and Z. Guan, J. Am. Chem. Soc., 2009, 131, (25), 8766

95. A. Li, J. Fan and G. Li, J. Mater. Chem. A, 2018, 6, (24), 11479

96. F. Xie, L. Huang, J. Leng and Y. Liu, J. Intell. Mater. Syst. Struct., 2016, 27, (18), 2433

97. S. Kelch, S. Steuer, A. M. Schmidt and A. Lendlein, Biomacromolecules, 2007, 8, (3), 1018

98. J. Leng, H. Lu, Y. Liu, W. M. Huang and S. Du, MRS Bull., 2009, 34, (11), 848

99. B. K. Kim, S. Y. Lee and M. Xu, Polymer, 1996, 37, (26), 5781

100. S. Rimdusit, M. Lohwerathama, K. Hemvichian, P. Kasemsiri and I. Dueramae, Smart Mater. Struct., 2013, 22, (7), 075033

101. H. Luo, H. Wang, H. Zhou, X. Zhou, J. Hu, G. Yi, Z. Hao and W. Lin, Appl. Sci., 2018, 8, (3), 392

102. J. Zhang, M. Huo, M. Li, T. Li, N. Li, J. Zhou and J. Jiang, Polymer, 2018, 134, 35

103. G. Ji, P. Zhang, J. Nji, M. John and G. Li, ‘11 - Shape Memory Polymer-Based Self-Healing Composites’, in “Recent Advances in Smart Self-healing Polymers and Composites”, eds. G. Li and H. Meng, Woodhead Publishing Series in Composites Science and Engineering, ch. 11, Woodhead Publishing, Cambridge, UK, 2015, pp 293–363

104. A. V. Menon, G. Madras and S. Bose, Polym. Chem., 2019, 10, (32), 4370

105. A. Lendlein, M. Behl, B. Hiebl and C. Wischke, Expert Rev. Med. Devices, 2010, 7, (3), 357

106. M. P. Gaj, A. Wei, C. Fuentes-Hernandez, Y. Zhang, R. Reit, W. Voit, S. R. Marder and B. Kippelen, Org. Electron., 2015, 25, 151

107. S. J. Park and C. H. Park, Sci. Rep., 2019, 9, 9157

108. S. Thakur ‘Shape Memory Polymers for Smart Textile Applications’, in “Textiles for Advanced Applications”, eds. B. Kumar and S. Thakur, ch. 12, Intech, Rijeka, Croatia, 2017, 432 pp

109. L. Li, P. Shi, Li Hua, J. An, Y. Gong, R. Chen, C. Yu, W. Hua, F. Xiu, J. Zhou, G. Gao, Z. Jin, G. Sun and W. Huang, Nanoscale, 2018, 10, (1), 118

110. Y. Huang, M. Zhu, Z. Pei, Q. Xue, Y. Huang and C. Zhi, J. Mater. Chem. A, 2016, 4, (4), 1290

111. S. Ahn, P. Deshmukh and R. M. Kasi, Macromolecules, 2010, 43, (17), 7330

112. J. Leng, D. Zhang, Y. Liu, K. Yu and X. Lan, Appl. Phys. Lett., 2010, 96, 111905

113. W. Li, Y. Liu and J. Leng, J. Mater. Chem. A, 2015, 3, (48), 24532



The Authors

Mathew John Haskew received his BSc in Chemistry from Lancaster University, UK, and subsequently undertook an MSc by research on SMP-based materials (with John Hardy at Lancaster University). He is currently undertaking a PhD in Engineering with Samuel Murphy and John Hardy at Lancaster University. His PhD involves the development of biodegradable biomaterials, and a combination of computational modelling and experimental validation of their efficacy.

John George Hardy received his MSci and PhD in Chemistry from the University of Bristol, UK, and the University of York, UK, respectively. Thereafter he undertook postdoctoral research in Biochemistry, Biomedical Engineering, Materials Science and Pharmacy (in France, Germany, Northern Ireland and the USA) before returning to the UK to lead a research group developing stimuli-responsive materials for technical and medical applications.




Buket YılmazDepartment of Materials Science and Engineering, Graduate School of Natural and Applied Science, Ege University, 35100 Bornova-Izmir, Turkey

Hüseyin Ata Karavana*Department of Leather Engineering, Faculty of Engineering, Ege University, 35100 Bornova-Izmir, Turkey; Department of Materials Science and Engineering, Graduate School of Natural and Applied Science, Ege University, 35100 Bornova-Izmir, Turkey

*Email: [email protected]; [email protected]

The purpose of this study was to devise an antibacterial treatment for footwear insock leathers. Orange oil-loaded chitosan microparticles were utilised for this purpose. Emulsion formulations with different ratios were prepared, and from these formulations microparticles were manufactured using a spray drying technique. Microparticles obtained in this way were applied to the insock leathers using a finishing process. Successful encapsulation was confirmed by ultraviolet-visible (UV-vis) spectrophotometry, Fourier transform infrared (FTIR) spectroscopy and scanning electron microscopy (SEM) techniques. The microparticles exhibited highly spheroid shape with a size range of 3–5 µm. Microparticle encapsulation efficiencies ranged from 79.41% ± 3.36% to 86.60% ± 1.13%. After performing microbiological tests and in vitro release studies on the insock leathers, it was determined that the prepared microparticles are able to perform core material delivery. Also, successful microparticle application resulted in these

leathers acquiring antibacterial properties. The products and process are biodegradable, nontoxic and biocompatible.

1. Introduction

Footwear is the most commonly worn apparel in daily life, and its design features must prioritise anatomy, comfort and hygiene. For this reason, it is important to develop sustainable improvements to footwear’s functional properties.Footwear that carries the body’s weight during

the day can affect foot health physically, chemically and microbiologically. Continuous contact with the external environment exposes footwear to microorganisms during normal use. All sorts of footwear play a role in the transport, spread and contamination of pathogenic or non-pathogenic microorganisms (1).There are different microorganisms in every part

of the human body. Sweat is regularly secreted from the body under normal conditions. It contains 98% water and urea, uric acid, fatty acid, lactic acid and sulfates (2). The feet have more sweat glands than other parts of the body. Sweat secreted from feet during the usage of footwear is decomposed by means of foot microbiota; as a result, bad odours emerge in footwear. Brevibacterium linens, Staphylococcus epidermidis, Staphylococcus aureus and Escherichia coli are some microorganisms that make footwear unhygienic. As a result of the breakdown of amino acids in sweat and skin by these microorganisms, bad odour arises in feet, socks and footwear (3, 4).Nowadays, in addition to individual foot care

and hygiene for odour prevention in shoes, commercial materials with various deodorising and antimicrobial effects are also employed (5).

Application of Chitosan-Encapsulated Orange Oil onto Footwear Insock LeathersSpray drying technique for an environmentally sustainable antibacterial formulation



Footwear insock is a thin layer of materials put into the shoe after manufacture to cover the insole. It directly contacts the sole of the wearer’s foot and can provide a more sanitary environment when specially treated for antimicrobial purposes (6).Spray drying is an advantageous way to encapsulate

active substances and essential oils. Spray drying is a common and accepted encapsulation method for industrial applications. With this method, it is possible to mass produce capsules. The distribution of particles is uniform (7–10).Microencapsulation technology has been used for

the application of orange oil to textiles and leathers, being an economically viable, fast and efficient method by combining core and shell materials, desirable perceptual and functional characteristics, and also allowing functional substances to be released in a controlled manner. This technique has also been used to microencapsulate a wide range of active, functional, sensitive or volatile substances (11–14). Tea tree oil containing melamine formaldehyde microcapsules, essential oils (eucalyptus, lavender or oregano), polyurethane dispersions containing photoactive antimicrobial agents, zinc oxide and silver nanoparticles are some substances that protect upper leathers from the harmful propagation of microorganisms (3, 15). In addition, aromas confined to microcapsules are also used to prevent bad odours in footwear (16, 17). Application of antibacterial and aromatic materials onto footwear insocks to control bad odours is good for foot hygiene and desired shoe comfort.The use of orange oil presents as an ecological

alternative to synthetic chemicals, attracting the attention of the scientific community to the development of eco-friendly antimicrobials. In this study, microparticles were produced by a spray drying method after the emulsions with orange oil and chitosan were prepared in different ratios. Microparticles manufactured in this way were then transferred to the surface of the footwear insock leathers using a finishing process. Afterward, some tests and analyses were performed on microparticle coated footwear insock leather samples to evaluate the effectiveness of the microparticles, their presence on the leather surface and their antimicrobial properties.

2. Experiment

Pharmaceutical grade cold pressed orange oil was donated from Ephesus, Turkey. Chitosan was purchased from Acros OrganicsTM (Belgium).

Analytical grade chemicals were used in the analyses. Insock leathers, without dye and ready for experimental application, were donated from Ata Dilek Leather (Izmir, Turkey).For the microparticle production, chitosan (shell

material) was added to 1% w/w aqueous acetic acid for preparing the chitosan solution. This solution was stirred at 45°C by using a magnetic stirrer until wholly dissolved. During the pre-emulsion preparation, orange oil (core material) was gradually mixed into the chitosan solution and stirred for 1 min at 10,000 rpm. The surfactants as compound emulsifiers used for pre-emulsion preparations were Tween 40 with Span 20 at the ratio of 8:2 w/w. Then, the microparticles were prepared by using an SD Basic spray dryer (Labplant, UK) with nozzle diameter of 0.5 mm. The orange oil to chitosan ratios in the four encapsulating compounds came to 1:1, 1:1.33, 1:1.67 and 1:2 w/w. The ingredients of the formulations in the spray-drying process are shown in Table I. Homogeneous emulsions were fed to the spray dryer under the following conditions: pump speed 12 ml min–1, outlet air temperature 114°C and inlet air temperature 175°C.The microparticles’ morphology was examined by

a Quanta 250 FEG scanning electron microscope (FEI, USA) at 2 kV accelerating voltage. Before coating in an argon atmosphere with gold-palladium by a K550X sputter coating machine (Quorum Emitech, UK), the samples were mounted onto an aluminium stub. The grain side of leathers coated with microparticles was examined by a TM1000 tabletop scanning electron microscope (Hitachi, Japan) after coating with gold-palladium.The FTIR spectra of the spray-dried microparticles

and leathers with microparticles were procured by a Spectrum 100 FTIR attenuated total reflectance (ATR) spectrometer (PerkinElmer, USA). The measurements were made using four scans with a resolution of 4 cm–1 between 4000 cm–1 to 650 cm–1 wavenumber ranges at room temperature.Encapsulation efficiencies of microparticles were

calculated as the amount of orange oil (core material) encapsulated in the microparticles. The

Table I Composition of the FormulationsFormulation code Orange oil:chitosan,

w/wT3 1:1

T4 1:1.33

T5 1:1.67

T6 1:2



encapsulation efficiency was calculated using Equation (i) (18).

%EE = × 100 (i)

A solvent extraction method was used to determine total oil content. A 0.1 g measurement of orange oil loaded microparticles was dissolved in 10 ml of 1% acetic acid solution at room temperature for 45 min. Released orange oil which was obtained from the completely dissolved microparticles was placed in a beaker containing 50 ml n-hexane for extraction 45 min. So as to determine the total amount of orange oil in the microparticles, this extract was filtered through a syringe filter (0.22 μm). Orange oil content in the filtrate was measured using a UV-1800 UV-vis spectrophotometer (Shimadzu, Japan) at 202 nm in triplicate. Surface oil content was also determined by the same solvent extraction method described above, except for a dissolving process in 1% acetic acid solution (11).In vitro release studies of microparticles and

microparticle loaded leathers were carried out at a speed of 100 rpm in phosphate-buffered saline (PBS) and methanol at 37°C. 1 mg orange oil loaded microparticles was suspended in beakers containing 4 ml methanol and 16 ml of PBS. Insock leathers with 2.5 cm2 area were placed in beakers containing 16 ml methanol and 16 ml of PBS for in vitro release studies of microparticle loaded leathers. At suitable time intervals, the medium in the beakers was filtered through a 0.22 μm syringe filter. Sink conditions were maintained in the receptor compartment during in vitro release studies. The released amount of orange oil was analysed by UV method, as previously described, for 5 h. Experiments were performed five times.A spraying pistol with nozzle diameter of 0.5 mm

was used to apply microparticles to the insock leathers during the finishing process. Spray-dried microparticles were added to the finishing recipe as 20 g m–2 (19). The basic finishing recipe for the insock leathers is given in Table II (20).The efficacy of microparticle coated insock leathers

against test microorganisms Staphylococcus aureus ATCC® 6538TM, Escherichia coli ATCC® 25922TM, Candida albicans ATCC® 10231TM, Klebsiella pneumoniae ATCC® 4352TM and Bacillus subtilis ATCC® 6633TM was examined by agar disc diffusion method (21–24). Test microorganisms

were placed into an incubator for incubation at 37°C for 18 h in the Mueller Hinton broth (MHB) medium. Then, microorganisms were inoculated in petri dishes containing 105 colony forming unit (CFU) ml–1 of Mueller Hinton agar (MHA) medium. Next, microparticle coated insock leather samples with 12.7 mm diameter were placed into the petri dishes (20, 25). All petri dishes were placed into an incubator for incubation at 37°C for 24 h, and inhibition zones were measured to determine antibacterial activity.


In this study, orange oil microparticles were successfully prepared by spray drying method. This method is a simple, viable method to obtain microparticles, suitable to prevent active substance biological activity loss, avoiding exposure to elevated heating and to organic solvents.

3.1 Surface Appearance of Microparticles and Microparticle Coated Insock Leathers

A scanning electron microscope was used to examine the morphology of the spray-dried microparticles. SEM micrographs revealed that all microparticle formulations have a highly spheroid shape with a morphology approximating an orange peel effect. Microparticles of non-uniform size were observed with clear distinction between shell and core materials. These shape features indicate that orange oil is spread on the surface of the microparticles. The morphology of spray-dried microparticle formulations is shown in Figure 1. Particle morphology (surface, size and distribution) was not affected by the polymer ratio or core:shell ratio. It is observed that there was formation of microcapsules, but they have stuck one to another and an agglomerate of microcapsules occurred. Microparticles with similar morphology were also

B (total amount of oil content – the amount of

surface oil content)

Table II Basic Finishing Recipe Applied to Insock Leathers

Materials Amount, part PracticeWater 100 3 × Spray

Anionic wax 50

Non-ionic aliphatic polyurethane binder

25

Orange oil loaded microparticles 12

A (total amount of oil content)



obtained in other spray drying experiments carried out using natural polymeric mixtures (11, 26, 27).We also examined the surface appearance of

microparticle-free and microparticle-coated insock leather samples. The different microparticle formulations were clearly observed on insock leather surfaces after successful application of the finishing process. Micrographs of insock leather surfaces are shown in Figure 2.After the finishing process, the presence of

microparticles on the insock leather can be seen very clearly for all formulations. The images indicate that the fixation was successfully achieved. Hence, the leather samples preserve the capsule content even after the finishing process.

3.2 Fourier Transform Infrared Spectroscopy Studies

Interactivity between the core material and shell material usually leads to characteristic alterations in the FTIR spectra. FTIR spectra of chitosan, orange oil, microparticles and insock leather samples are shown in Figure 3 and Figure 4. Characteristic peaks at 1029 cm–1, 1149 cm–1, 1373 cm–1, 1419 cm–1, 1585 cm–1, 2867 cm–1 and 3362 cm–1 were demonstrated in the FTIR spectrum of chitosan (Figure 3). The peak at 3362 cm–1 (OH and NH2 stretching) was attributed to the amino group of chitosan. An intense absorption peak was seen at 2867–2922 cm–1 owing to C–H stretching in all spectra. The peak at 1585 cm–1 was attributed to N–H bending of the NH3

+ functional group present in the chitosan (28, 29). The peak at 1373 cm–1 confirmed the presence of an amide III band in the chitosan. The C–O–C stretching resulted from the spectra at 1149 cm–1 and 1029 cm–1. The spectrum at 660 cm–1 was attributed to stretching vibration of pyranoside ring (30–34).The FTIR spectrum of the orange oil showed

the distinctive bands of D-limonene, which is the primary constituent in orange oil (Figure 3). Especially, the bands between 2919–2834 cm–1 were attributed to the C–H stretching vibrations in –CH–, –CH2– and –CH3. The spectrum at 2965 cm–1 was attributed to the stretching vibrations of =C–H. The band 1644 cm–1 was attributed to the stretching vibrations of C=C. The band seen at 1435 cm–1 was attributed to the C–H bending vibrations in –CH–, –CH2– and –CH3. The peaks at 885 cm–1 and 797 cm–1 were attributed to the bending vibrations (out of plane) in =CH2 and =C(R)–H, respectively. The band at 1376 cm–1 was also attributed to the C–H bending vibrations in –CH3 (mostly used to describe the existence of methyl) (35, 36).As seen in Figure 3, most bands in the FTIR spectra

of the microparticles belonged to chitosan, which indicated that orange oil droplets were trapped in chitosan (shell material) and that distinctive band of orange oil vanished or declined. Evidently, the free vibrations of orange oil molecules were blocked by the chitosan because of physical interactions such as van der Waals or electrostatic interaction. Furthermore, the intensity of microparticle peaks on the FTIR spectrum was lower than that of chitosan because of the interaction between orange oil and chitosan. The FTIR spectra of microparticles demonstrated the C–H bending vibrations of –CH3 at 1376 cm–1, except the =C(R)–H bending

(a) (b)

(c) (d)

(e) (f)

(g) (h)

2 μm

2 μm

2 μm

2 μm

10 μm

10 μm

10 μm

10 μm

3.78

0 μm

3.173 μm4.523 μm

5.330 μm

Fig. 1. SEM micrographs of microparticle formulations: (a) T3 formulation, 50,000 × magnification; (b) T3 formulation, 10,000 × magnification; (c) T4 formulation, 50,000 × magnification; (d) T4 formulation, 10,000 × magnification; (e) T5 formulation, 50,000 × magnification; (f) T5 formulation, 10,000 × magnification; (g) T6 formulation, 50,000 × magnification; (h) T6 formulation, 10,000 × magnification



(a) (b) (c)

(d) (e)

30 μm 30 μm100 μm

100 μm 100 μm

Fig. 2. SEM micrographs of the insock leather after finishing process: (a) microparticle free; (b) T3 formulation; (c) T4 formulation; (d) T5 formulation; (e) T6 formulation

4000 3600 3200 2800 2400 2000 1800 1600 1400 1200 1000 800 650cm–1

T, %

Chitosan

Orange oil

T3

T4

T5

T6

3362.83 2867.87 1585.84 1419.871373.85 1149.73

1029.40

892.61 660.57

758.92

797.80

885.61

914.86956.93

1051.921016.921147.91

1644.851435.80

1376.862834.88

2919.81

2965.87

3295.79 2922.80

3283.79 2874.81

3288.76 2876.79

3351.78 2875.81

1556.771511.79

1407.751151.73

1067.531028.53

888.72

752.71887.69

1028.49

1151.72

1376.76

1510.801557.79

888.66888.69

1027.451063.46

1151.701408.74

1510.791557.76

1593.821509.83

1376.791151.72

Fig. 3. The FTIR spectrum of the chitosan, orange oil and four different microparticle formulations (T3, T4, T5 and T6)

vibration at 797 cm–1, which was presumably due to the fact of the D-limonene ring being covered with chitosan (14, 36).Figure 4 show that bands between

1535–1547 cm–1 attributed to the NH band of chitosan, did not appear in the blank leather sample (34). Similarly, it was determined that IR band vibration at 1095 cm–1 was observed in the microparticle loaded leathers but absent from the blank leather. That was evidence of the presence of terpenoid, a component in orange oil (37).

3.3 Encapsulation Efficiency

Orange oil loaded microparticles were produced with a high orange oil encapsulation efficiency. The encapsulation efficiency of T3, T4, T5 and T6

formulations were determined as 79.41% ± 3.36%, 81.28% ± 1.69%, 83.56% ± 0.66% and 86.60% ± 1.13%, respectively. A great deal of encapsulated orange oil is preferred. These results showed that the microparticles’ encapsulation efficiency is affected by the core:shell ratio. Increasing the chitosan weight resulted in more encapsulated orange oil, i.e. high encapsulation efficiency. This is an effect similar to the oil:polymer ratio given by Li and associates in their 2013 study (11).

3.4 In Vitro Release Studies of Microparticles and Microparticle Coated Insock Leathers

Figure 5 shows the in vitro release behaviours of orange oil released from microparticles in four



different formulations. The quantity of released orange oil was measured at 202 nm in PBS at different times. Previous experiments used PBS as an in vitro release and diffusion medium for topical applications (38, 39). Oil release from microparticulate systems occurs via different mechanisms including diffusion, desorption, disintegration and surface erosion (40).The typical release pattern of the spray dried

microparticles is characterised by a small initial burst release and a sustained release rate following that. It can be seen that orange oil release from microparticles gradually increased over time with exposure to PBS, which indicates that the orange oil disintegrated swiftly in PBS. This circumstance is presumably owed to the fact that PBS is slightly alkaline; chitosan is inclined to dissolve in slightly alkaline solution. Nonetheless, it can be seen that the release rate was not affected by chitosan concentration in the formulations. Figure 5 graphs release behaviour as a function of orange oil concentration, which was independent from chitosan concentration.The in vitro release results of the leathers

impregnated with orange oil loaded microparticles in pH 7.4 PBS at 37°C are presented in Figure 6. This line graph shows controlled release behaviour from leather treated with all formulations. Orange oil trapped inside the microparticles caused sustained release up to 24 h. When the formulations are compared to each other, we see the oil release ratio of insock leathers was affected by polymer concentration. High polymer concentration caused a slow release ratio of orange oil.

3.5 Microbiologic Studies on Microparticle Coated Insock Leathers

Table III shows microbiologic test results of insock leathers treated with four microparticle

4000 3600 3200 2800 2400 2000 1800 1600 1400 1200 1000 800 650cm–1

T, %

3309.92

2917.62

2849.72 1728.461643.83

1552.831464.76

1379.801237.63

1159.51

1024.72852.85

720.79

3285.90

2916.722850.79

1730.66

1644.831541.81

1470.811379.80 1241.78

1155.66

1024.553309.93

2916.662850.76 1728.54

1644.801547.81

1464.78

1380.81

1237.66

1159.59

1095.681024.67

851.84717.78

3307.93

2916.682850.77 1728.49

1644.811544.80

1464.781380.79

1237.641158.55

1095.661024.65

851.83718.78

3309.94

2917.742850.82 1535.83

1463.791380.80

1236.611157.52

1095.651023.64

851.83732.78

Blank leather

T3

T4

T5

T6

Fig. 4. The FTIR spectrum of footwear insock leather without microparticles (blank leather) and with four different microparticles formulations (T3, T4, T5 and T6)

908070605040302010

0

C,

%

30 60 90 120 150 180 210 240 270 300Time, min

T3

T4

T5

T6

Fig. 5. In vitro release of orange oil loaded microparticles

60

50

40

30

20

10

0

C,

%

500 1000 1500Time, min

T3

T4

T5

T6

Fig. 6. In vitro release of microparticle-coated insock leathers



formulations. An important revelation is that the test microorganisms did not grow on these leather samples. However, in some test groups, a meagre antimicrobial inhibition zone around the insock leather samples meant that orange oil diffusion did not occur. There is a visible zone on Candida albicans in all formulations. T3 and T4 formulations, whose orange oil releases are higher in 24 h, look more effective against Escherichia coli. The antimicrobial activity is dependent on chitosan’s inherent behaviour and orange oil present on leather samples. When the inhibition zones in the T6 formulation are examined, it can be seen that orange oil found in the insock leather samples is more effective than the natural behaviour of chitosan on antimicrobial activity. The antimicrobial effect can be considered as proliferation or non-proliferation in the area under the insock leather samples. This effect is also expressed as contact inhibition. No proliferation was observed on the contact surface of the insock leathers, that is, on the surface where it touches the medium and microorganism. Also, any proliferation on the surfaces or edges of insock leather samples was not observed. There was no difference between leather formulations on the antimicrobial test.

4. Conclusion

During the usage of footwear, perspiration and bacterial activity negatively impact foot health and generate bad odours from both the feet and footwear. Shoe production using natural and non-toxic materials that prevent or inhibit bad odours and bacterial activity is one solution to this hygienic problem. Likewise, the successful application of microparticles that release for a long time on footwear insock leather is an important alternative to existing toxic products.Our research found that emulsions with orange

oil and chitosan have natural antibacterial activity. These emulsions, when successfully converted into encapsulated powders by a spray drying method, produce a core-shell material. SEM images showed how an effective finishing process was used to apply laboratory produced microparticles to the surface of footwear insock leather. Microbiological tests performed on microparticle coated leathers proved that footwear insock leathers were fortified with antibacterial properties.These findings demonstrate that application of

orange oil-chitosan microparticles onto footwear insock leather surfaces is an alternative natural method to control hygiene and eliminate bad

Table III Microbiologic Test Results of the Microparticle-Loaded Insock Leathers

Formulation codes

Bacillus subtilis ATCC® 6633TM

Candida albicans ATCC® 10231TM

Escherichia coli ATCC® 25922TM

Klebsiella pneumoniae ATCC® 4352TM

Staphylococcus aureus ATCC® 6538TM

T3

T4

T5

T6



odours. Non-toxic, functional, leather shoes can incorporate such natural materials in their manufacture and maintenance. This production improvement would thus contribute to people’s foot health, hygiene and comfort.

Acknowledgements

The authors would like to thank the Scientific Research Projects Coordination Unit, Ege University, Turkey (Project No: 17FBE006) for financial support and the Turkish Prime Ministry’s State Planning Organisation (Project No: 07DPT001) for equipment provision.

References

1. V. K. Özkan, M. T. Uzun and M. T. Gündoğan, J. Fungus, 2018, 9, (2), 182

2. H. Yalçın, B. Özkalp, ‘Importance of Body Hygiene and New Developments in Wound Care’, 4th National Sterilization Disinfection Congress, 20th–24th April, 2005, Samsun, Turkey, Bilimsel Tıp Publishing House, Ankara, Turkey, 287–308

3. Z. Majidnia, A. Idris and P. Valipour, J. Teknol., 2013, 60, (1), 5

4. P. Velmurugan, M. Cho, S.-M. Lee, J.-H. Park, S. Bae and B.-T. Oh, Carbohydr. Polym., 2014, 106, 319

5. M. M. Sánchez-Navarro, M. A. Pérez-Limiñana, N. Cuesta-Garrote, M. I. Maestre-López, M. Bertazzo, M. A. Martínez-Sánchez and F. Arán-Aís, ‘Latest Developments in Antimicrobial Functional Materials for Footwear’, in “Microbial Pathogens and Strategies for Combating Them: Science, Technology and Education”, ed. A. Méndez-Vilas, Vol. 1, Formatex, New York City, USA, 2013, pp. 102–113

6. Z. Tülek, ‘A Research Over Side Industry Products in the Shoe Manufacturing’, Master Thesis, Social Science Institute, Istanbul Arel University, Turkey, 2016, 126 pp

7. M. Gohel, R. K. Parikh, S. A. Nagori, A. V. Gandhi, M. S. Shroff, P. K. Patel, C. S. Gandhi, V. Patel, N. Y. Bhagat, S. D. Poptani, S. R. Kharadi, R. Pandya and T. C. Patel, Pharma. Rev., 2009, 7, (5), 1

8. B. N. Estevinho, F. Rocha, L. Santos and A. Alves, Trends Food Sci. Technol., 2013, 31, (2), 138

9. D. Santos, A. C. Maurício, V. Sencadas, J. D. Santos, M. H. Fernandes and P. S. Gomes, ‘Spray Drying: An Overview’, in “Biomaterials: Physics and Chemistry”, ed. R. Pignatello, Ch. 2, InTechOpen, London, UK, 2018, pp. 9–35

10. O. Tomazelli Júnior, F. Kuhn, P. J. M. Padilha, L. R. M. Vicente, S. W. Costa, A. A. Boligon,

J. Scapinello, C. N. Nesi, J. Dal Magro and S. L. Castellví, Braz. J. Biol., 2018, 78, (2), 311

11. Y. Li, L. Ai, W. Yokoyama, C. F. Shoemaker, D. Wei, J. Ma and F. Zhong, J. Agric. Food Chem., 2013, 61, (13), 3311

12. I. Gönülşen, M. Sariişik, G. Erkan and S. Okur, Tekst. ve Mühendis, 2016, 23, (101), 21

13. W. Rossi, M. Bonet-Aracil, E. Bou-Belda, J. Gisbert-Payá, K. Wilson and L. Roldo, IOP Conf. Ser. Mater. Sci. Eng., 2017, 254, (2), 022007

14. P. Velmurugan, V. Ganeshan, N. F. Nishter and R. R. Jonnalagadda, Surf. Interfac., 2017, 9, 124

15. I. P. Fernandes, J. S. Amaral, V. Pinto, M. J. Ferreira and M. F. Barreiro, Carbohydr. Polym., 2013, 98, (1), 1229

16. M. M. Sánchez-Navarro, M. A. Pérez-Limiñana, F. Arán-Aís and C. Orgilés-Barceló, Polymer Int., 2015, 64, (10), 1458

17. C. Torres-Alvarez, A. Núñez González, J. Rodríguez, S. Castillo, C. Leos-Rivas and J. G. Báez-González, CyTA-J. Food, 2017, 15, (1), 129

18. N. Suwannateep, S. Wanichwecharungruang, S. F. Haag, S. Devahastin, N. Groth, J. W. Fluhr, J. Lademann and M. C. Meinke, Eur. J. Pharm. Biopharm., 2012, 82, (3), 485

19. M. Kleban, J. Weisser, F. Koch and W. Schwaiger, Bayer Corp, ‘Leather Finished with Scent-Containing Microcapsules’, US Patent Appl. 2002/198,392

20. F. Yalcin, H. A. Karavana, S. Rencber and S. Y. Karavana, J. Am. Leather Chem. Assoc., 2020, 115, (3), 79

21. G. M. Caputo, P. R. Cavanagh, J. S. Ulbrecht, G. W. Gibbons and A. W. Karchmer, N. Engl. J. Med., 1994, 331, (13), 854

22. J. H. Calhoun, K. A. Overgaard, C. M. Stevens, J. P. F. Dowling and J. T. Mader, Adv. Skin Wound Care, 2002, 15, (1), 31

23. B. Örmen, N. Türker, I. Vardar, N.A. Coşkun, F. Kaptan, S. Ural, S. El and M. Türker, Turkish J. Infect., 2007, 21, (2), 65

24. İ. Yaşa, N. Lkhagvajav, M. Koizhaiganova, E. Çelik and Ö. Sarı, World J. Microbiol. Biotechnol., 2012, 28, (7), 2531

25. H. A. Karavana, S. Rencber, S. Y. Karavana and F. Yalcin, ‘Encapsulated Chlorhexidine Digluconate Usage on the Diabetic Footwear Lining Leathers’, 6th International Conference on Advanced Materials and Systems (ICAMS), 20th–22nd October, 2016, Bucharest, Romania, CERTEX, Bucharest, Romania, 257–262

26. A. Soottitantawat, H. Yoshii, T. Furuta, M. Ohgawara, P. Forssell, R. Partanen, K. Poutanen and P. Linko, J. Agric. Food Chem., 2004, 52, (5), 1269



27. A. Javid, Z. A. Raza, T. Hussain and A. Rehman, J. Microencapsul., 2014, 31, (5), 461

28. F.-L. Mi, H.-W. Sung, S.-S. Shyu, C.-C. Su and C.-K. Peng, Polymer, 2003, 44, (21), 6521

29. D. R. Bhumkar and V. B. Pokharkar, AAPS PharmSciTech, 2006, 7, (2), E138

30. R. Yoksan, J. Jirawutthiwongchai and K. Arpo, Coll. Surf. B: Biointer., 2010, 76, (1), 292

31. J. Jingou, H. Shilei, L. Weiqi, W. Danjun, W. Tengfei and X. Yi, Coll. Surf. B: Biointer., 2011, 83, (1), 103

32. L. Keawchaoon and R. Yoksan, Coll. Surf. B: Biointer., 2011, 84, (1), 163

33. S. F. Hosseini, M. Zandi, M. Rezaei and F. Farahmandghavi, Carbohydr. Polym., 2013, 95, (1), 50

34. A. K. T. Chang, R. R. Frias, L. V Alvarez, U. G. Bigol and J. P. M. D. Guzman, Biocatal. Agric. Biotechnol., 2019, 17, 189

35. D. Cava, J. M. Lagarón, A. López-Rubio, R. Catalá and R. Gavara, Polym. Test., 2004, 23, (5), 551

36. D. Li, H. Wu, W. Huang, L. Guo and H. Dou, Eur. J. Lipid Sci. Technol., 2018, 120, (9), 1700521

37. H. Boughendjioua and S. Djeddi, Am. J. Opt. Photonics, 2017, 5, (3), 30

38. T. Şenyiğit, F. Sonvico, S. Barbieri, Ö. Özer, P. Santi and P. Colombo, J. Control. Release, 2010, 142, (3), 368

39. S. T. Tanrıverdi and Ö. Özer, Eur. J. Pharm. Sci., 2013, 48, (4–5), 628

40. S. Hariharan, V. Bhardwaj, I. Bala, J. Sitterberg, U. Bakowsky and M. N. V. Ravi Kumar, Pharm. Res., 2006, 23, (1), 184

The Authors

Buket Yılmaz graduated from Chemical Engineering, Faculty of Engineering, Anadolu University, Turkey, in 2015. For a period during her undergraduate education, she benefited from the FARABİ exchange programme for further chemical engineering studies at Ege University. Yılmaz won and completed a competitive internship at Turkey’s two leading companies involved in polymers and food production. In 2016, she started her master’s degree at Ege University’s Institute of Science, Materials Science and Engineering. Her scientific expertise has also been employed by the private sector in sales and in the quality control unit of a food production enterprise.

Hüseyin Ata Karavana graduated from the Leather Technology Department, Faculty of Agriculture, Ege University, Turkey. He earned his MSc degree in Leather Technology in 2001 from that institution’s Graduate School of Natural and Applied Science. From 2006 to 2007 he continued his studies as an Erasmus student in the Department of Footwear Engineering and Hygiene at the Tomas Bata University’s Faculty of Technology (Zlin, Czech Republic). Karavana completed his PhD degree in Leather Engineering at Ege University in 2008. Karavana currently serves as Associate Professor in the Department of Leather Engineering at Ege University’s Faculty of Engineering. His research interests are in all manner of leather and footwear engineering including plastic composites, microencapsulation, leather quality and control, footwear quality and control.




Tuğçe Tüccar*Department of Biology, Institute of Graduate Studies in Sciences, Istanbul University, 34134, Vezneciler, Istanbul, Turkey

Esra Ilhan-SungurDepartment of Biology, Faculty of Science, Istanbul University, 34134, Vezneciler, Istanbul, Turkey

Gerard MuyzerDepartment of Biotechnology, Delft University of Technology, van der Maasweg 9, 2629 HZ Delft, The Netherlands; Microbial Systems Ecology, Department of Freshwater and Marine Microbiology, Institute for Biodiversity and Ecosystem Dynamics, University of Amsterdam, PO Box 94240, 1090 GE, Amsterdam, The Netherlands


Oil fields harbour a wide variety of microorganisms with different metabolic capabilities. To examine the microbial ecology of petroleum reservoirs, a molecular-based approach was used to assess the composition of bacterial communities in produced water of Diyarbakır oil fields in Turkey. Denaturing gradient gel electrophoresis (DGGE) of polymerase chain reaction (PCR)-amplified 16S rRNA gene fragments was performed to characterise the bacterial community structure of produced water samples and to identify predominant community members after sequencing of separated DGGE bands. The majority of bacterial sequences retrieved from DGGE analysis of produced water samples belonged to unclassified bacteria (50%).

Among the classified bacteria, Proteobacteria (29.2%), Firmicutes (8.3%), Bacteroidetes (8.3%) and Actinobacteria (4.2%) groups were identified. Pseudomonas was the dominant genus detected in the produced water samples. The results of this research provide, for the first time, insight into the complexity of microbial communities in the Diyarbakır oil reservoirs and their dominant constituents.

1. Introduction

Although much progress has been made in the use of renewable energy in recent years, fossil fuels (especially oil and gas) still meet most of the global energy demand, and they will continue to be the dominant source of energy worldwide over the next few decades (1).Petroleum is a naturally occurring material found

in various geological formations (reservoirs) worldwide. Crude oil, the liquid part of petroleum, is primarily composed of hydrocarbons (2). However, it may also include compounds of nitrogen, sulfur, oxygen and metals (3). Because crude oil in reservoirs is found as a mixture containing varying constituents and proportions, each crude oil has its own unique properties. The most important specified properties are density and sulfur content (4). The density of crude oil is reported in terms of American Petroleum Institute (API) gravity (specific gravity). Based on the API gravity, crude oils can be classified into light, medium, heavy and extra heavy oils (3). Depending on the amount of sulfur content (elemental sulfur or sulfur compounds such as hydrogen sulfide), the crude oil is categorised as ‘sweet’ or ‘sour’. In addition to chemical composition and physical properties, crude oil typically is also identified by underground

Bacterial Community Composition in Produced Water of Diyarbakır Oil Fields in Turkey Bacterial communities in produced waters of south-eastern Turkey reported in detail for the first time



reservoir (4). Reservoir characteristics (depth, temperature, pressure and other factors) vary significantly from one location to another, even in the same geologic formation (5, 6). The fact that microbial community composition and reservoir conditions vary dramatically not only between the different geographical areas, but also among different oil fields in the same region, makes each oil reservoir ecosystem unique.Despite the extreme environmental conditions

in the oil-bearing formations (i.e. anoxic, high temperature, high salinity), many microorganisms are capable of surviving in the oil and water phases of the oil wells (7, 8). Oil fields harbour mainly facultative aerobic and strictly anaerobic microorganisms due to the low redox potential in the reservoirs (8). These ecosystems contain different types of microbial communities (such as mesophiles, thermophiles and halophiles) which adapt to the reservoir conditions (9). Bacterial and archaeal groups identified in oil fields include sulfate-reducing bacteria (10), sulfur-oxidising bacteria (11), methanogens (12), fermentative microorganisms (13), acetogens (14), nitrate reducers (15), manganese and iron reducers (16) and hydrocarbon degraders (17). Among these microbes, sulfate-reducing bacteria have attracted much attention due to their detrimental effects such as reservoir souring and biocorrosion (7). In addition, different members of the oil microbial community are involved in syntrophic interactions. Fermenting bacteria and methanogenic archaea are involved in methanogenic hydrocarbon biodegradation through their close syntrophic associations (18). This microbial process is undesirable in oil reservoirs because it causes a decrease in oil quality and value (19). Syntrophic microorganisms in oil reservoirs also play important roles in the global biogeochemical cycling of sulfur, carbon and nitrogen. For instance, sulfate-reducing bacteria and sulfur-oxidising bacteria, the key drivers in sulfur transformations, are involved in the sulfur cycle (11). Thus, knowledge of the microbial groups and microbial dynamics in oil fields enable us to obtain detailed insights into the microbial ecology of oil associated environments. Understanding the microbial ecology of oil reservoirs

is crucial to the petroleum industry because the success of oilfield operations is strongly influenced by the activity of microorganisms. Oil microbes with different metabolic capabilities have significant negative and positive impacts on the petroleum resources and the extraction processes (7). Microbial activity may lead to severe problems such as

reservoir souring and microbial corrosion. Reservoir souring, which is characterised by an increase in production of H2S in the reservoir fluids, most commonly occurs when sulfidogenic microorganisms reduce sulfate to sulfide, a toxic and corrosive product (20). Undesirable accumulation of sulfide minerals in reservoirs is one of the major challenging problems in oil production because it causes plugging of reservoirs, decreasing the oil quality and value and increasing the refining costs. Moreover, exposure to H2S can be dangerous in terms of worker health and safety due to its high toxicity. Additionally, the produced H2S promotes corrosion of the metallic equipment and structures used for oil production and processing (21). Another destructive phenomenon is biocorrosion, which is defined as microbial attack on the surface of the metal infrastructure leading to disruption of the material (22). In addition to sulfate-reducing bacteria, which play a major role in biocorrosion, other corrosive microbes, such as acetogenic bacteria and methanogenic archaea, are also associated with corrosion failures (23). Biocorrosion is a great concern because it leads to loss of material, large economic losses and safety issues in the oil industry (24). In contrast, hydrocarbon-degrading bacteria may be used for environmental clean-up processes (6). Bacterial degradation of hydrocarbons was carried out by both aerobic (for example, Rhodococcus sp., Sphingomonas sp., Pseudomonas putida, Pseudomonas stutzeri, Acinetobacter sp.) and anaerobic bacteria (such as Fe(III)-reducing bacteria, sulfate-reducing bacteria) (6, 17). Furthermore, microbial products such as biopolymers and biosurfactants can be used for facilitating oil movement in a widely used technology, known as microbial enhanced oil recovery (MEOR) (1). Compared with other conventional oil recovery techniques, MEOR has advantages such as low cost, wide application, high efficiency and low environmental pollution (25). Therefore, diversity, metabolic processes and habitat conditions of microbial communities in oil reservoirs should be investigated, so that their negative effects can be decreased and their positive effects can be exploited.This study aimed to determine the bacterial

community composition and to identify the predominant community members in produced water from oil fields located in the Diyarbakır region in Turkey. To this end, we used PCR-DGGE to analyse 20 produced water samples from the Diyarbakır region. There are limited studies on produced water from the Diyarbakır region and this paper represents the only in situ study available. The results of this study provide not only new



data about the microbial ecology of the Diyarbakır oil fields, but also information on the bacterial populations which may have potential roles in terms of increasing or decreasing the efficiency of industrial applications.

2. Materials and Methods

2.1 Sampling Procedure

The sampling site, the Diyarbakır region, is located at the boundary of the Anatolian plate and the Middle Eastern oil region in south-eastern Turkey. A total of 20 crude oil samples (B1, B6, B8, B14, B23, B32, B56, GK8, GS6, GS15, M3, K2, K3, K32, K35, K44, S4, S15, Y18 and Y30) consisting of an oil/water mixture were collected from the production wells of Diyarbakır oil fields (Figure 1). These wells produced oils withdrawn from the oil sandstone deposits (depths from 1600 m to 2620 m, API gravity from 24.3° to 42.3°, water content around 94%, an average pH of 7.0 and salinity from 2966 mg l−1 to 26,961 mg l−1). The samples were aseptically taken at the wellhead and put into sterile 500 ml serum bottles sealed with rubber stoppers and aluminium caps. The samples were shipped at ambient temperature. Upon arrival at the laboratory, the samples were immediately analysed. All samples were treated within 48 h after collection. Decantation was used to separate produced water from the oil/water mixture.

2.2 DNA Extraction

Bacteria in the produced water samples were collected by filtration over 0.20 μm pore size polyamide filters (Sartolon®, Sartorius AG, Germany). Genomic DNA was extracted with the UltraClean® Microbial DNA isolation kit (MO BIO Laboratories Inc, USA) according to the manufacturer’s protocol.

2.3 Polymerase Chain Reaction Amplification

Extracted DNA was used as the template for PCR amplification of partial 16S rRNA fragments. Primer pair consisting of 341F with a GC clamp and 907R was used for DGGE analysis (26). A 40-base GC clamp was used to prevent complete denaturation of the fragment during DGGE (27). Due to the low DNA yield, a two-step PCR strategy

was used. At the first step, a real-time PCR (quantitative PCR, qPCR) approach was applied to the produced water samples. The reaction mixture in a final volume of 22.5 µl contained 0.2 µl of each primer, 12.5 µl iQTM SYBR® Green Supermix (Bio-Rad Laboratories Inc, USA), 9.6 µl RNase-Free Water (Qiagen, Germany) and 0.5 µl DNA template. qPCR was performed in iCycler iQTM Real-Time PCR Detection System (Bio-Rad Laboratories Inc, USA) using the following conditions: 5 min at 95°C; 40 cycles of 95°C for 30 s, 57°C for 40 s,

B32

0 10 20 km

B23B56 B14

K3K2

K32

K44K35

B6B8 S15

GS15S4

GK8

GS6Y30

Y18

M3

B1

NB32

Black sea

Yerevan

BeirutDamascus

Fig. 1. Sampling locations in Diyarbakır region. Produced water samples were collected from 20 different oil wells © Maphill / Creative Commons Attribution-NoDerivatives (CC BY-ND)

37º00' 40º20'

36º20'

39º40'

37º00' 40º20' 43º40'

43º40'



72°C for 40 s and 80°C for 25 s; and a final 72°C for 10 min. In the qPCR method, after each cycle, a signal was formed. By observing the signals for each sample, PCR products could be detected. The reaction was terminated when the desired amount of product was reached. At the second step, a conventional PCR approach was applied to the qPCR products. Reaction mixture in a final volume of 25 µl contained 0.2 µl of each primer, 12.5 µl Taq PCR Master Mix (Qiagen, Germany), 9.6 µl RNase-Free Water (Qiagen, Germany) and 0.5 µl DNA template. The PCR was performed in TGradient thermocycler (Biometra, Germany) using the following conditions: 5 min at 95°C; 12 cycles of 95°C for 30 s, 57°C for 40 s and 72°C for 40 s; and a final 74°C for 30 min.

2.4 Denaturing Gradient Gel Electrophoresis

The DCodeTM system (Bio-Rad Laboratories, USA) was used for DGGE analysis. 25 µl of each PCR product (200–300 ng) were loaded onto 6% polyacrylamide gels (w/v) containing gradients of 20% to 70% denaturants (urea/formamide). The gels were run for 16 h at 100 V and 60°C in 1× Tris-acetate-EDTA buffer. After completion of electrophoresis, the gels were stained with SYBR® Gold Nucleic Acid Gel Stain (InvitrogenTM, Thermo Fisher Scientific, USA) for 20 min, visualised and photographed. Selected predominant DGGE bands were excised, eluted in 40 µl of 1× Tris buffer (pH 8) for 2 d at 4°C and re-amplified with 25 cycles as described above. Reaction mixture in a final volume of 25 µl contained 0.125 µl of primer 341F, 0.125 µl of primer 907R, 12.5 µl of Taq PCR Master Mix, 9.75 µl of ultra-pure water and 0.5 µl of template. The PCR products were quantified on a 1.5% (w/v) agarose gel and then sequenced by Macrogen Inc (Seoul, South Korea).

2.5 Comparative Sequence Analysis

The resulting sequences were first aligned and edited using CodonCode Aligner software (CodonCode Corp, USA). Then they were compared to sequences stored in the database GenBank® using the National Center for Biotechnology Information (NCBI) Basic Local Alignment Search Tool (BLAST®) (28, 29). All obtained partial 16S rRNA gene sequences were deposited in GenBank® database under the following accession numbers: KF720792 - KF720796, KF720798, KF720801 - KF720802, KF720804, KF720806 - KF720808, KF720810 - KF720811, KF720814, KF720818,

KF720820, KF720823, KF720825 - KF720826, KF720828, KF720830 - KF720832, KF720839, KF720844, KF720852, KF720855, KF720858, KF720872, KF720877, KF720882 - KF720884, KF720886 - KF720889, KF720891, KF720893 - KF720894, KF720896 and KF720903.

3. Results

3.1 Molecular Analysis of Bacterial Communities

Bacterial DNA isolation could only be achieved for 16 (B1, B8, B6, B14, B23, B32, B56, GS6, GK8, K35, K44, M3, S4, S15, Y18, Y30) of the 20 produced water samples. Because the water phase could not be separated from the oil phase for the other four produced water samples, DNA could not be extracted from these samples. The extracted DNA was used as template DNA for the amplification of 16S rRNA gene fragment. Unfortunately, direct PCR with bacterial primers did not yield a product from any of the produced water samples. For this reason, a two-step PCR was applied: the first step was a qPCR to increase the concentration of genetic material to measurable amounts (30), while the second step was a normal PCR to obtain enough material for DGGE analysis. For produced water samples, a total of 113 DGGE gel bands were analysed, but only 69 bands yielded sequences of satisfactory quality (Figure 2).

(a) (b) (c)

B32 B6 B14 B23 S4 GK8 GS6 M3 B56 B1 B8 K35 Y30 Y18 S15

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Fig. 2. DGGE profiles of 16S rRNA gene fragments amplified from produced water samples. See legend to Figure 1. (a) 1, B32; 2, B6; 3, B14; 4, B23; 5, S4, 6, GK8; (b) 7, GS6; 8, M3; (c) 9, B56; 10, B1; 11, B8; 12, K35; 13, Y30; 14, Y14; 15, S15



Comparative sequence analysis of the DGGE bands indicated that 50% of the bacterial sequences belonged to ‘unclassified bacteria’. Among the classified bacteria, members of the phyla Proteobacteria, Bacteroidetes, Firmicutes and Actinobacteria, and the classes Alphaproteobacteria, Betaproteobacteria, Gammaproteobacteria, Sphingobacteriia, Bacilli and Actinobacteria were identified (Figure 3).

3.1.1 Proteobacteria

Proteobacteria was the dominant phylum, comprising 29.2% of the total sequences retrieved from the produced water samples (Figure 3). The sequences B6_19 and B14_37 shared 100% and 99% identity with uncultured bacteria (EU044497 and JF421153, respectively) (Table I). The sequence B32_3 was distantly (94%) related to a moderately thermophilic bacterium Phenylobacterium lituiforme, a member of Alphaproteobacteria (31). Within Betaproteobacteria, the sequence represented by B32_55 was identified (98%) as Aquincola sp. THE-49 (JN128637), isolated from water reservoir (published only in GenBank®). The produced water contained different members of the class Gammaproteobacteria. DGGE bands B32_4, B14_35, S4_70, GS6_2 and GK8_79 were affiliated (100%, 94%, 99%, 93%, 99%, respectively) to Pseudomonas stutzeri (Table I), a non-fluorescent denitrifying bacterium (32). The sequence from band B1_20 showed a 100% similarity to Acinetobacter sp. VKPM 2838 (Table I). The genus

Acinetobacter comprises important soil organisms where they contribute to the mineralisation of aromatic compounds and they are suited to exploitation for biotechnological purposes, such as biodegradation (33). B8_30 was related (96%) to Marinobacter sp. Trimyema-2, a thermophilic strain that was isolated from the hydrothermally heated sea floor at Vulcano Island, Italy (34). Members of the genus Marinobacter were also identified in the production water retrieved from a Dutch oil field (35). The sequence from B32_2 was distantly related (93%) to Thermithiobacillus sp. ParkerM (HM173631) that is moderately thermophilic and obligately chemolithoautotrophic on reduced inorganic sulfur compounds (36). Another member of the class Gammaproteobacteria was close to the sequence of uncultured hydrocarbon seep bacterium (91% similarity) (AF154088) (Table I).

3.1.2 Bacteroidetes

8.3% of the sequences detected among the produced water samples fell into Bacteroidetes (Figure 3). The sequence of band B6_14 was affiliated to unclassified Chitinophagaceae. It shared 99% identity with Chitinophagaceae bacterium F1 (AB535716), isolated from compost (Table I). DGGE bands B6_16, S4_67 and B8_27 were identified (92% to 99% sequence identity) as uncultured Bacteroidetes bacteria (Table I). The sequences from S4_67 and B8_27 were related to uncultured bacteria that were taught as members of biocorroding microbiota colonising on steel surfaces immerged in coastal seawater (37).

Actinobacteria 4.2%

Firmicutes 8.3%

Bacteroidetes 8.3%

Proteobacteria 29.2%

Unclassified bacteria

50%

(a) (b)

Actinobacteria 15%

Sphingobacteriia 8%

Bacilli 8%

Betaproteobacteria 8%

Gammaproteobacteria 38%

Alphaproteobacteria 23%

Fig. 3. Phylogenetic distribution of the 16S rRNA sequences of produced water samples from the Diyarbakır oil wells at: (a) the phylum level; and (b) the class level



Tab

le I

Ph

ylog

enet

ic A

ffilia

tion

s of

Bac

teri

al S

equ

ence

s R

etri

eved

fro

m P

rod

uce

d W

ater

Sam

ple

s B

ased

on

16

S r

RN

A A

nal

ysis

an

d

Gen

Ban

k® A

cces

sion

Nu

mb

ers

Ass

ign

ed t

o th

ese

Seq

uen

ces

Wel

l n

o.D

GG

E b

and

A

cces

sion

n

um

ber

Clo

sest

BLA

ST®

mat

chB

LAS

T®

acce

ssio

n

nu

mb

er

Sim

ilari

ty,

%P

hyl

um

Cla

ssIs

olat

ion

sou

rce

B1

B1_

20KF7

2088

3Aci

neto

bact

er s

p. V

KPM

283

8JF

8913

9010

0Pr

oteo

bact

eria

Gam

ma

prot

eoba

cter

ia—

B1_

23KF7

2089

1Aer

ibac

illus

pal

lidus

str

ain

MCM

B-8

86

JN70

1188

89Fi

rmic

utes

Bac

illi

Petr

oleu

m r

eser

voir

B1_

24KF7

2089

3U

ncul

ture

d Fi

rmic

utes

bac

terium

HM

0419

4298

Firm

icut

es—

Prod

uced

flui

d

B6

B6_

14KF7

2081

8Chi

tinop

haga

ceae

bac

terium

F1

AB53

5716

99Bac

tero

idet

es S

phin

goba

cter

iiaCom

post

B6_

16KF7

2083

0U

ncul

ture

d Bac

tero

idet

es

bact

eriu

mFR

8714

1392

Bac

tero

idet

es—

Tota

l cop

epod

ex

trac

ts

B6_

17KF7

2079

3U

ncul

ture

d ba

cter

ium

G

Q25

9593

98

——

Bio

reac

tor

B6_

19KF7

2080

2U

ncul

ture

d Sph

ingo

mon

as s

p.EU

0444

9710

0Pr

oteo

bact

eria

Alp

ha

prot

eoba

cter

iaSoi

l

B6_

20KF7

2080

8U

ncul

ture

d Fi

rmic

utes

bac

terium

EU19

4836

96

Firm

icut

es

—Cha

rles

ton

Har

bor

sedi

men

t

B6_

26KF7

2079

8Cor

ioba

cter

iace

ae b

acte

rium

en

rich

men

t cu

lture

clo

ne B

3113

HQ

1330

2910

0Act

inob

acte

ria

Act

inob

acte

ria

Cru

de o

il co

ntam

inat

ed s

oil

B8

B8_

27KF7

2088

2U

ncul

ture

d Bac

tero

idet

es

bact

eriu

mEF

4914

3099

Bac

tero

idet

es—

Ste

el s

urfa

ces

imm

erge

d in

mar

ine

wat

er

B8_

29KF7

2088

7U

ncul

ture

d ba

cter

ium

FJ62

8289

96—

—

Bra

ckis

h w

ater

fro

m

anox

ic f

jord

Niti

nat

Lake

at

dept

h of

50

m

B8_

30KF7

2088

8M

arin

obac

ter

sp.

Trim

yem

a-2

AJ2

9252

896

Prot

eoba

cter

iaG

amm

a pr

oteo

bact

eria

The

hydr

othe

rmal

ly

heat

ed s

ea fl

oor

(Con

tinue

d)



Wel

l n

o.D

GG

E b

and

A

cces

sion

n

um

ber

Clo

sest

BLA

ST®

mat

chB

LAS

T®

acce

ssio

n

nu

mb

er

Sim

ilari

ty,

%P

hyl

um

Cla

ssIs

olat

ion

sou

rce

B1

4

B14

_35

KF7

2080

4Ps

eudo

mon

as s

tutz

eri

HQ

1897

5594

Prot

eoba

cter

iaG

amm

a pr

oteo

bact

eria

Wat

er/s

oil m

ix p

ile

of s

ampl

es f

rom

oil

wel

ls

B14

_37

KF7

2081

4U

ncul

ture

d Cae

nisp

irill

um s

p.

clon

e Pp

ss_M

a27

JF42

1153

99Pr

oteo

bact

eria

Alp

ha

prot

eoba

cter

ia

Petr

oleu

m-

cont

amin

ated

sa

line-

alka

li so

il w

ith

phyt

orem

edia

tion

B14

_38

KF7

2082

0U

ncul

ture

d ba

cter

ium

FN

4295

3598

——

Was

tew

ater

of

oil

refin

ery

trea

tmen

t pl

ant

B14

_39

KF7

2082

5G

eorg

enia

dae

guen

sis

HQ

2461

6310

0Act

inob

acte

ria

Act

inob

acte

ria

Act

ivat

ed s

ludg

e fr

om in

dust

rial

w

aste

wat

er

trea

tmen

t

B14

_41

KF7

2079

4U

ncul

ture

d ba

cter

ium

GQ

4570

2596

——

Rhi

zosp

here

B2

3

B23

_52

KF7

2081

0U

ncul

ture

d ba

cter

ium

FN40

1244

99—

—D

omes

tic t

oile

t bi

ofilm

B23

_55

KF7

2082

6Aqu

inco

la s

p. T

HE-

49JN

1286

3798

Prot

eoba

cter

iaBet

a pr

oteo

bact

eria

Wat

er r

eser

voir

B23

_56

KF7

2083

1U

ncul

ture

d ba

cter

ium

HM

9211

4499

——

Gro

undw

ater

fro

m

drin

king

wat

er

trea

tmen

t pl

ant

B3

2

B32

_1KF7

2079

2U

ncul

ture

d so

il ba

cter

ium

AY22

1598

99—

—M

etal

and

hy

droc

arbo

n co

ntam

inat

ed s

oil

B32

_2KF7

2079

6Th

erm

ithio

baci

llus

sp.

Park

erM

HM

1736

3193

Prot

eoba

cter

iaG

amm

a pr

oteo

bact

eria

—

B32

_3KF7

2080

1Ph

enyl

obac

terium

litu

iform

eAY

5348

8794

Prot

eoba

cter

iaAlp

ha

prot

eoba

cter

iaSub

surf

ace

aqui

fer

B32

_4KF7

2080

7Ps

eudo

mon

as s

tutz

eri

FJ34

5693

100

Prot

eoba

cter

iaG

amm

a pr

oteo

bact

eria

Are

a co

ntam

inat

ed

by c

rude

oil

and

chem

ical

s

B5

6B56

_17

KF7

2090

3U

ncul

ture

d Fi

rmic

utes

bac

terium

HM

0419

4297

Firm

icut

es—

Prod

uced

flui

d

(Con

tinue

d)



Wel

l n

o.D

GG

E b

and

A

cces

sion

n

um

ber

Clo

sest

BLA

ST®

mat

chB

LAS

T®

acce

ssio

n

nu

mb

er

Sim

ilari

ty,

%P

hyl

um

Cla

ssIs

olat

ion

sou

rce

GK

8G

K8_

79KF7

2082

8Ps

eudo

mon

as s

tutz

eri

JF72

7663

99Pr

oteo

bact

eria

Gam

ma

prot

eoba

cter

ia

Petr

oleu

m-

cont

amin

ated

sa

line-

alka

li so

ils

GS

6

GS6_

1KF7

2083

2U

ncul

ture

d ba

cter

ium

JN03

0519

99—

—Fi

ssur

e w

ater

co

llect

ed f

rom

a

bore

hole

GS6_

2KF7

2083

9Ps

eudo

mon

as s

tutz

eri

JN22

8329

93Pr

oteo

bact

eria

Gam

ma

prot

eoba

cter

ia—

GS6_

4KF7

2085

2U

ncul

ture

d ba

cter

ium

JF49

7820

90—

—Act

ivat

ed s

ludg

e

GS6_

5KF7

2085

8U

ncul

ture

d m

arin

e ba

cter

ium

FM21

1087

90—

—M

icro

cosm

ex

perim

ent

M3

M3_

28KF7

2085

5U

ncul

ture

d ba

cter

ium

PH

OS-

HE3

1AF3

1443

099

——

Bat

ch r

eact

or

M3_

31KF7

2087

2U

ncul

ture

d ba

cter

ium

HM

9211

4498

——

Gro

undw

ater

fro

m

drin

king

wat

er

trea

tmen

t pl

ant

M3_

32KF7

2087

7U

ncul

ture

d ba

cter

ium

HQ

5386

3999

——

Bul

king

act

ivat

ed

slud

ge

M3_

34KF7

2084

4U

ncul

ture

d ba

cter

ium

AB23

1448

99—

—En

hanc

ed b

iolo

gica

l ph

osph

orus

rem

oval

(E

BPR

) sl

udge

K3

5

K35

_44

KF7

2088

4U

ncul

ture

d hy

droc

arbo

n se

ep

bact

eriu

m B

PC02

8AF1

5408

891

Prot

eoba

cter

iaG

amm

a pr

oteo

bact

eria

Hyd

roca

rbon

see

p se

dim

ent

K35

_46

KF7

2088

9U

ncul

ture

d ba

cter

ium

HM

9211

4498

——

Gro

undw

ater

fro

m

drin

king

wat

er

trea

tmen

t pl

ant

K35

_48

KF7

2089

4U

ncul

ture

d ba

cter

ium

FJ62

3379

97—

—Bat

ch r

eact

or

K35

_49

KF7

2089

6U

ncul

ture

d ba

cter

ium

AB23

1448

99—

—EB

PR s

ludg

e (Con

tinue

d)



Wel

l n

o.D

GG

E b

and

A

cces

sion

n

um

ber

Clo

sest

BLA

ST®

mat

chB

LAS

T®

acce

ssio

n

nu

mb

er

Sim

ilari

ty,

%P

hyl

um

Cla

ssIs

olat

ion

sou

rce

S4

S4_

67KF7

2080

6U

ncul

ture

d Bac

tero

idet

es

bact

eriu

mEF

4914

3092

Bac

tero

idet

es—

Ste

el s

urfa

ces

imm

erse

d in

mar

ine

wat

er

S4_

68KF7

2081

1U

ncul

ture

d ba

cter

ium

FJ62

8289

96—

—Bra

ckis

h w

ater

fro

m

anox

ic f

jord

S4_

70KF7

2082

3Ps

eudo

mon

as s

tutz

eri

JN22

8329

99Pr

oteo

bact

eria

Gam

ma

prot

eoba

cter

ia—

S4_

73KF7

2079

5U

ncul

ture

d ba

cter

ium

JF51

4265

100

——

Sea

S1

5S15

_60

KF7

2088

6U

ncul

ture

d hy

droc

arbo

n se

ep

bact

eriu

m B

PC02

8AF1

5408

891

Prot

eoba

cter

iaG

amm

a pr

oteo

bact

eria

Hyd

roca

rbon

see

p se

dim

ent

Y1

8Y1

8_70

KF7

2088

9U

ncul

ture

d ba

cter

ium

HM

9211

4498

——

Gro

undw

ater

fro

m

drin

king

wat

er

trea

tmen

t pl

ant

Y3

0

Y30_

66KF7

2088

9U

ncul

ture

d ba

cter

ium

HM

9211

4498

——

Gro

undw

ater

fro

m

drin

king

wat

er

trea

tmen

t pl

ant

Y30_

68KF7

2089

4U

ncul

ture

d ba

cter

ium

FJ62

3379

97—

—Bat

ch r

eact

or

Y30_

69KF7

2089

6U

ncul

ture

d ba

cter

ium

AB23

1448

99—

—EB

PR s

ludg

e

3.1.3 Firmicutes

Sequences belonging to members of Firmicutes accounted for 8.3% of the bacteria in the produced water (Figure 3). DGGE band B1_23 was distantly related (89%) to Aeribacillus pallidus strain MCM B-886 (JN701188), isolated from petroleum reservoir (published only in GenBank®) (Table I). In addition, different strains of Aeribacillus pallidus (with sequence similarity values from 98% to 99.6%) were isolated previously from various geothermal sites of Turkey (38). DGGE band B6_20 was distantly related (96%) to an uncultured Firmicutes bacterium, isolated from marine sediment in Charleston, South Carolina, USA (39). B56_17 and B1_24 were affiliated (97% and 98%, respectively) to an uncultured Firmicutes bacterium (Table I), detected in produced fluid from non-water-flooded high-temperature reservoir of the Niibori oilfield, Japan (40).

3.1.4 Actinobacteria

The phylum Actinobacteria comprised 4.2% of the bacterial community recovered from the produced water (Figure 3). DGGE band B6_26 displayed 100% sequence similarity to Coriobacteriaceae bacterium enrichment culture clone B3113 (HQ133029) isolated from crude oil contaminated soil of Shengli oil fields, China (41). The sequence B14_39 was closely related (100% similarity) to an aerobic bacterial strain Georgenia daeguensis 2C6-43, isolated from an activated sludge sample collected from an industrial wastewater treatment plant in Daegu, South Korea (42). Although little is known about the presence of G. daeguensis in oil associated environments, it was reported that different strains of G. daeguensis were isolated from hydrocarbon contaminated soil of an industrial zone and oil-saturated soil under laboratory conditions (43–45).

4. Discussion

In order to increase our knowledge about microbial diversity, culture-dependent



and molecular-based approaches are used for describing the diversity of microbes. Molecular-based approaches such as PCR-DGGE methodology, which is a useful tool for monitoring the genetic diversity of complex microbial populations (26), provide valuable information about the microbial community structure and dynamics in nature. For these reasons, PCR-DGGE fingerprinting analysis of environmental samples was used in this study. The choice of appropriate primers for PCR

amplification is a crucial step to accurately characterise the microbial communities. In this study, primer pair (341F-GC/907R), targeting the V3-V5 region of the 16S rRNA gene fragment, was selected due to its suitability for DGGE analysis of bacterial populations in environmental samples (26). This primer pair designed by Muyzer et al. (27, 28) has been used predominantly for microbial community analysis (26).The DNA yield obtained from produced water

samples was very low. It is known that crude oil samples contain low amounts of biomass which makes DNA isolation difficult to achieve (46). In this study, the permit included taking up to 500 ml of oil/water mixture from each sampling point so that only ca. 25 ml of each produced water sample could be obtained. In this scope, the low sample volumes of produced water separated from the oil/water mixture may be a reason for the low amount of DNA. It was reported in other studies that higher sample volumes (100–4000 ml) of produced water were used for DNA isolation (35, 47–50). The low DNA yield affected the efficiency of the PCR technique and for this reason, a two-step PCR was applied to the produced water samples. Thus, a sufficient amount of PCR product for DGGE for the produced water samples could be obtained. Bacterial communities associated with the

produced waters was analysed by the PCR-DGGE approach. Although numerous bands were visible on the DGGE gel, only dominant bands could be excised and sequenced. Most of the sequences retrieved from produced water samples were related to unclassified bacteria. Different studies on oil reservoir microbiota have also shown that oil fields harbour new and still unidentified microbial species. For example, Lenchi et al. described microbial communities in production and injection waters from the Algerian oil fields. In their study, they detected that a large number of unclassified bacterial and archaeal sequences were found in the water samples (51). Furthermore, uncultured bacteria such as uncultured Sphingomonas sp. and uncultured Caenispirillum sp. clone

Ppss_Ma27 were detected in our study. This result is consistent with the fact that the vast majority of microorganisms are uncultured and do not grow under laboratory conditions as stated by Lewis et al. (52). In order to isolate more microbes, an appropriate identification laboratory protocol should be followed. At this point, different strategies such as mimicking natural conditions via decreased nutrient, extended incubation times, the modification of isolating media formulations and different incubation parameters (for example, temperature) were suggested for the cultivation of microorganisms (53). For instance, pollutant degrader Sphingomonas, which seemed to be previously uncultured by nutrient-rich methods, could be isolated from crude oil contaminated soil by using an in situ method that mimics the original environment (54). In addition, culture-dependent investigation should also be supported by molecular techniques. Based on the sequences, organisms related to

known mesophilic bacteria were predominant in the produced water samples. In addition, some organisms related to thermophilic bacteria (Aeribacillus pallidus, Marinobacter sp. Trimyema-2, Phenylobacterium lituiforme and Thermithiobacillus sp.) were also identified. Bacteria having different metabolic capabilities (denitrifying, biodegrading and sulfur removing bacteria) were also detected. In addition, bacteria which may cause biocorrosion on steel surfaces were detected. The dominant bacterial phylum was the

Proteobacteria. The members of this phylum were also frequently found in many other studies on microbial diversity of oil field produced waters (55–58). Moreover, it was stated that Proteobacteria are ubiquitous in oil reservoirs over all temperature ranges (59).In this study, among the detected genera in

produced water samples that potentially contain hydrocarbon degrading bacteria were Aeribacillus, Acinetobacter, Sphingomonas, Marinobacter and Phenylobacterium. It has been known for years that the species belonging to these genera are capable of degrading hydrocarbons (6, 17, 60, 61). In addition, G. daeguensis, a hydrocarbonoclastic bacterium, was detected in produced water sample with a 100% sequence similarity. G. daeguensis has also been demonstrated as a potential microbe for bioremediation due to its hydrocarbon degradation ability (44). Further investigations are needed because our current knowledge of the metabolic capability of G. daeguensis is limited. Moreover, sulfur-oxidising Thermithiobacillus sp.



was also identified in produced water sample. Sulfur-oxidising bacteria, which oxidise the sulfur compounds produced by the activity of sulfate-reducing bacteria in oil reservoirs, may play a key role in the oil industry because they can be utilised to resolve processing problems such as reservoir souring (11). Pseudomonas was the dominant genus

detected among the produced water samples. Pseudomonas stutzeri was the species identified in five produced water samples. P. stutzeri was previously isolated not only from formation water, produced from the petroleum wells in Adıyaman (62), but also oil-contaminated soils in Batman petroleum refinery, Turkey (63). These two areas are close to the Diyarbakır region from where the samples in this study were collected and these findings show that P. stutzeri is distributed widely in south-eastern Turkey. In other different geographical areas, this species was also isolated from oil-associated environments, such as oil field production water (64), oil sludge (65) and oil contaminated soil (66). However, although P. stutzeri is often isolated from oil reservoirs, the origin of P. stutzeri in oil reservoirs is a debatable issue. Because oil reservoirs have low redox potentials and contain little oxygen, anaerobic microorganisms are considered as truly indigenous to oil reservoirs (67). In this regard, it is believed that P. stutzeri, most of whose strains are aerobes, is an exogenous organism inoculated into oil reservoirs during the oil production processes. Even if strains of P. stutzeri are introduced into oil reservoirs with injected fluids, they should adapt to the physicochemical characteristics of the reservoir to survive. At this point, it has been proposed that extreme reservoir conditions may act as special factors for the evolution of P. stutzeri, thereby forming mutant strains (68). Furthermore, P. stutzeri, being found in a wide variety of habitats, is known for its diverse metabolism. Some strains of P. stutzeri are capable of denitrification, degradation of aromatic compounds and nitrogen fixation (32). These metabolic features make P. stutzeri highly attractive for biotechnological processes, such as reservoir souring control (69), microbial enhanced oil recovery (64) and bioremediation of oil-polluted environments (65).In undisturbed oil reservoirs, microorganisms

are found in different phases such as reservoir fluid containing crude oil and formation water, and rock surfaces. While planktonic microbes thrive in the water phase, sessile microbes may attach to

oil or rock surfaces (59). In addition, biofilm may form on the metal surfaces of the pipes in the oil-producing wells (70). Oil microbiome studies focus mainly on the analysis of the water phase due to its easy sampling. However, it should be noted that the water phase itself contains only a minor portion of the microbes found in the oil reservoir (59). On the other hand, the sampling of sessile microbes is likely to be more challenging (70).

5. Conclusion

This study reported for the first time the bacterial community composition of produced water from Diyarbakır oil reservoirs as obtained by DGGE analysis of PCR-amplified 16S rRNA gene fragments. DGGE analysis of produced water samples demonstrated that the majority of the bacterial sequences belonged to unclassified bacteria, indicating that oil reservoirs harbour still undescribed microbial species. Among the classified bacteria, the members of Proteobacteria were more abundant. Pseudomonas was the dominant genus detected in the produced water. Although the members of Pseudomonas were known as exogenous organisms inoculated into oil reservoirs, Pseudomonas stutzeri was found in five produced water samples. Bacteria having different metabolic capabilities (denitrifying, biodegrading and sulfur removing bacteria) were also detected. It can be stated that the metabolic capacities of these bacteria make them potential candidates for utilising in biodegradation, bioremediation, the improvement of oil quality and oil recovery processes. The knowledge of the bacterial community composition in oil reservoirs of the Diyarbakır region obtained in this study will be of great interest for both scientific research and applications in the oil industry. To build on the data presented in this study, metagenomic analyses should be performed to explore the undescribed microbes.

Acknowledgements

This work was supported by ‘Research Fund of Istanbul University’ (Project number: 28699). Tuğçe Tüccar was awarded an Erasmus LLP Scholarship. Esra Ilhan-Sungur was awarded a Post-doctoral Research Scholarship by the Scientific and Technological Research Council of Turkey (TUBITAK-BIDEB). We thank the Turkish Petroleum Corporation for permission to collect samples, and Ender Taptık and Hasan Kaya for their assistance



with the sample collection. We thank Ben Abbas for his technical assistance. We acknowledge Hakan Hosgormez for his helpful comments and suggestions.

References

1. C. Nikolova and T. Gutierrez, Front. Microbiol., 2020, 10, 2996

2. H. Dembicki, ‘Introduction’, in “Practical Petroleum Geochemistry for Exploration and Production”, Ch. 1, Elsevier Inc, Amsterdam, The Netherlands, 2017, pp. 1–17

3. J. G. Speight, “Handbook of Industrial Hydrocarbon Processes”, 2nd Edn., Elsevier Inc, Cambridge, USA, 2020, 786 pp

4. S. Romanow-Garcia and H. L. Hoffman, ‘Petroleum and Its Products’, in “Kent and Riegel’s Handbook of Industrial Chemistry and Biotechnology”, Vol. 1, 11th Edn., Springer Science and Business Media LLC, New York, USA, 2007, pp. 801–842

5. A. Satter and G. M. Iqbal, “Reservoir Engineering: The Fundamentals, Simulation, and Management of Conventional and Unconventional Recoveries”, Elsevier Inc, Waltham, USA, 2016, 472 pp

6. S. J. Varjani, Bioresour. Technol., 2017, 223, 277

7. S. J. Varjani and E. Gnansounou, Bioresour. Technol., 2017, 245, (A), 1258

8. N. Youssef, M. S. Elshahed and M. J. McInerney, ‘Microbial Processes in Oil Fields: Culprits, Problems, and Opportunities’, in “Advances in Applied Microbiology”, eds. A. I. Laskin, S. Sariaslani and G. M. Gadd, Ch. 6, Elsevier Inc, San Diego, USA, 2009, pp. 141–251

9. M. Magot, O. Basso, C. Tardy-Jacquenod and P. Caumette, Int. J. Syst. Evol. Microbiol., 2004, 54, (5), 1693

10. A. Hussain, A. Hasan, A. Javid and J. I. Qazi, 3 Biotech, 2016, 6, (2), 119

11. H. Tian, P. Gao, Z. Chen, Y. Li, Y. Li, Y. Wang, J. Zhou, G. Li and T. Ma, Front. Microbiol., 2017, 8, 143

12. C. Berdugo-Clavijo and L. M. Gieg, Front. Microbiol., 2014, 5, 197

13. S. Kh. Bidzhieva, D. Sh. Sokolova, T. P. Tourova and T. N. Nazina, Microbiology, 2018, 87, (6), 757

14. J.-F. Liu, X.-B. Sun, G.-C. Yang, S. M. Mbadinga, J.-D. Gu and B.-Z. Mu, Front. Microbiol., 2015, 6, 236

15. C. C. Okoro and O. O. Amund, Petrol. Sci. Technol., 2018, 36, (4), 293

16. S. Tamazawa, D. Mayumi, H. Mochimaru, S. Sakata, H. Maeda, T. Wakayama, M. Ikarashi,

Y. Kamagata and H. Tamaki, Int. J. Syst. Evol. Microbiol., 2017, 67, (10), 3982

17. J. D. Van Hamme, A. Singh and O. P. Ward, Microbiol. Mol. Biol. Rev., 2003, 67, (4), 503

18. S. Che and Y. Men, J. Ind. Microbiol. Biotechnol., 2019, 46, (9–10), 1343

19. N. Jiménez, H. H. Richnow, C. Vogt, T. Treude and M. Krüger, J. Mol. Microbiol. Biotechnol., 2016, 26, (1–3), 227

20. Y. Xue and G. Voordouw, Front. Microbiol., 2015, 6, 1387

21. D. Enning and J. Garrelfs, Appl. Environ. Microbiol., 2014, 80, (4), 1226

22. J. Telegdi, A. Shaban and L. Trif, ‘Corrosion Mechanisms: Current Knowledge, Gaps and Future Research: Microbiologically Influenced Corrosion (MIC)’, in “Trends in Oil and Gas Corrosion Research and Technologies: Production and Transmission”, ed. A. M. El-Sherik, Part 3, Ch. 8, Elsevier Ltd, Duxford, UK, 2017, pp. 191–214

23. S. Kato, Microb. Biotechnol., 2016, 9, (2), 141

24. R. F. Wright, P. Lu, J. Devkota, F. Lu, M. Ziomek-Moroz and P. R. Ohodnicki, Sensors, 2019, 19, (18), 3964

25. M. Safdel, M. A. Anbaz, A. Daryasafar and M. Jamialahmadi, Renew. Sustain. Energy Rev., 2017, 74, 159

26. S. J. Green, M. B. Leigh and J. D. Neufeld, ‘Denaturing Gradient Gel Electrophoresis (DGGE) for Microbial Community Analysis’, in “Hydrocarbon and Lipid Microbiology Protocols: Microbial Quantitation, Community Profiling and Array Approaches”, eds. T. J. McGenity, K. N. Timmis and B. Nogales, Springer-Verlag, Berlin, Germany, 2017, pp. 77–100

27. G. Muyzer, E. C. de Waal and A. G. Uitterlinden, Appl. Environ. Microbiol., 1993, 59, (3), 695

28. G. Muyzer, A. Teske, C. O. Wirsen and H. W. Jannasch, Arch. Microbiol., 1995, 164, 165

29. Basic Local Alignment Search Tool (BLAST®), National Center for Biotechnology Information, US National Library of Medicine, Bethesda, MD, USA, 18th June, 2020

30. T. L. Skovhus, N. B. Ramsing, C. Holmström, S. Kjelleberg and I. Dahllöf, Appl. Environ. Microbiol., 2004, 70, (4), 2373

31. S. Kanso and B. K. C. Patel, Int. J. Syst. Evol. Microbiol., 2004, 54, (6), 2141

32. J. Lalucat, A. Bennasar, R. Bosch, E. Garcia-Valdés and N. J. Palleroni, Microbiol. Mol. Biol. Rev., 2006, 70, (2), 510

33. D. L. Gutnick and H. Bach, ‘Potential Application of Acinetobacter in Biotechnology’, in “Acinetobacter:



Molecular Biology”, ed. U. Gerischer, Ch. 9, Caister Academic Press, Caister, UK, 2008, pp. 231–264

34. M. Baumgartner, K. O. Stetter and W. Foissner, J. Eukaryot. Microbiol., 2002, 49, (3), 227

35. G. M. van der Kraan, J. Bruining, B. P. Lomans, M. C. M. van Loosdrecht and G. Muyzer, FEMS Microbiol. Ecol., 2010, 71, (3), 428

36. D. P. Kelly and A. P. Wood, Int. J. Syst. Evol. Microbiol., 2000, 50, (2), 511

37. H. Dang, R. Chen, L. Wang, S. Shao, L. Dai, Y. Ye, L. Guo, G. Huang and M. G. Klotz, Environ. Microbiol., 2011, 13, (11), 3059

38. A. C. Cihan, B. Ozcan, N. Tekin and C. Cokmus, World J. Microbiol. Biotechnol., 2011, 27, (11), 2683

39. B. J. Mathis, C. W. Marshall, C. E. Milliken, R. S. Makkar, S. E. Creager and H. D. May, Appl. Microbiol. Biotechnol., 2008, 78, (1), 147

40. H. Kobayashi, K. Endo, S. Sakata, D. Mayumi, H. Kawaguchi,, M. Ikarashi, Y. Miyagawa, H. Maeda and K. Sato, J. Biosci. Bioeng., 2012, 113, (2), 204

41. L. Cheng, J. Rui, Q. Li, H. Zhang and Y. Lu, FEMS Microbiol. Ecol., 2013, 83, (3), 757

42. S. G. Woo, Y. Cui, M.-S. Kang, L. Jin, K. K. Kim, S. T. Lee, M. Lee and J. Park, Int. J. Syst. Evol. Microbiol., 2012, 62, (7), 1703

43. Y. Toptaş, M. Çelikdemir, C. Tuncer, Y. B. Şahin, P. A. Çelik, N. Burnak, A. Çabuk and V. Bütün, Turkish J. Biochem., 2016, 41, (5), 338

44. N. Ali, N. Dashti, M. Khanafer, H. Al-Awadhi and S. Radwan, Sci. Rep., 2020, 10, 1116

45. S. S. Radwan, D. M. Al-Mailem and M. K. Kansour, Sci. Rep., 2019, 9, 19508

46. T. B. P. Oldenburg, S. R. Larter, J. J. Adams, M. Clements, C. Hubert, A. K. Rowan, A. Brown, I. M. Head, A. A. Grigoriyan, G. Voordouw and M. Fustic, Anal. Chem., 2009, 81, (10), 4130

47. V. J. Orphan, L. T. Taylor, D. Hafenbradl and E. F. Delong, Appl. Environ. Microbiol., 2000, 66, (2), 700

48. H. Li, S.-Z. Yang, B.-Z. Mu, Z.-F. Rong and J. Zhang, FEMS Microbiol. Ecol., 2007, 60, (1), 74

49. J. Wang, T. Ma, L. Zhao, J. Lv, G. Li, F. Liang and R. Liu, World J. Microbiol. Biotechnol., 2008, 24, (9), 1981

50. R. Kumaraswamy, S. Ebert, M. R. Gray, P. M. Fedorak and J. M. Foght, Appl. Microbiol. Biotechnol., 2010, 89, (6), 2027

51. N. Lenchi, Ö. İnceoğlu, S. Kebbouche-Gana, M. L. Gana, M. Llirós, P. Servais and T. Garcia-Armisen, PLoS One, 2013, 8, (6), e66588

52. K. Lewis, S. Epstein, A. D’Onofrio and L. L. Ling, J. Antibiot., 2010, 63, (8), 468

53. A. Bodor, N. Bounedjoum, G. E. Vincze, Á. Erdeiné Kis, K. Laczi, G. Bende, Á. Szilágyi, T. Kovács, K. Perei and G. Rákhely, Rev. Environ. Sci. Bio/Technol., 2020, 19, (1), 1

54. H. Zhao, Y. Zhang, X. Xiao, G. Li, Y. Zhao and Y. Liang, Int. Biodeterior. Biodegrad., 2017, 117, 269

55. J. You, G. Wu, F. Ren, Q. Chang, B. Yu, Y. Xue and B. Mu, Appl. Microbiol. Biotechnol., 2016, 100, (3), 1469

56. W.-F. Song, J.-W. Wang, Y.-C. Yan, L.-Y. An, F. Zhang, L. Wang, Y. Xu, M.-Z. Tian, Y. Nie and X.-L. Wu, Int. Biodet. Biodeg., 2018, 132, 18

57. X. Wang, X. Li, L. Yu, L. Huang, J. Xiu, W. Lin and Y. Zhang, Sci. Total Environ., 2019, 653, 872

58. T. Tüccar, E. Ilhan-Sungur, B. Abbas and G. Muyzer, Anaerobe, 2019, 59, 19

59. M. Pannekens, L. Kroll, H. Müller, F. T. Mbow and R. U. Meckenstock, New Biotechnol., 2019, 49, 1

60. J. Eberspächer and F. Lingens, ‘The Genus Phenylobacterium’, in “The Prokaryotes”, eds. M. Dworkin, S. Falkow, E. Rosenberg, K. H. Schleifer and E. Stackebrandt, Vol. 5, 3rd Edn., Springer Science and Business Media LLC, New York, USA, 2006, pp. 250–256

61. G. T. Mehetre, S. G. Dastager and M. S. Dharne, Sci. Total Environ., 2019, 679, 52

62. Ç. Babaarslan, A. Tekeli and T. Mehmetoğlu, Energy Sources, 2003, 25, (7), 733

63. T. Kaya, ‘Çeşitli endüstriyel atık maddelerde bazı mikroorganızmaların yüzey aktif özelliklerinin incelenmesi’ [‘Research of Surface Active Properties of Some Microorganisms in Various Industrial Wastes’], Masters Thesis, Biology Department, Gazi University, Ankara, Turkey, 28th April, 2008, 126 pp

64. F. Zhao, C. Guo, Q. Cui, Q. Hao, J. Xiu, S. Han and Y. Zhang, Carbohyd. Polym., 2018, 199, 375

65. A. Afifi, H. Motamedi, B. Alizadeh and H. Leilavi, Environ. Experi. Biol., 2015, 13, (1), 13

66. Y. Anwar, A. A. El-Hanafy, J. S. M. Sabir, S. M. S. Al-Garni, K. Al-Ghamdi, H. Almehdar and M. Waqas, Polycyc. Aromat. Comp., 2020, 40, (1), 135

67. M. Magot, ‘Indigenous Microbial Communities in Oil Fields’, in “Petroleum Microbiology”, eds. B. Ollivier and M. Magot, ASM Press, Washington, DC, USA, 2005, pp. 21–34

68. F. Zhang, Y.-H. She, I. M. Banat, L.-J. Chai, L.-Q. Huang, S.-J. Yi, Z.-L. Wang, H.-L. Dong and D.-J. Hou, MicrobiologyOpen, 2014, 3, (4), 446

69. F. Fan, B. Zhang, P. L. Morrill and T. Husain, RSC Adv., 2018, 8, (47), 26596



70. X. Zhu and M. A. Al-Moniee, ‘Corrosion Inhibitors – Advancements in Testing: Molecular Microbiology Techniques’, in “Trends in Oil and Gas Corrosion

Research and Technologies: Production and Transmission”, ed. A. M. El-Sherik, Part 4, Ch. 19, Elsevier Ltd, Duxford, UK, 2017, pp. 513–536

The Authors

Tuğçe Tüccar is a PhD candidate in Fundamental and Industrial Microbiology at Istanbul University, Turkey. She received her Bachelor’s degree in Biology from Middle East Technical University, Turkey, in 2009. She obtained her Master’s degree in Fundamental and Industrial Microbiology from Istanbul University, Turkey, in 2011. Her dissertation was on investigation of sulfate-reducing bacteria in petroleum samples. She was awarded an Erasmus LLP Scholarship and conducted her research work at Delft University of Technology, The Netherlands. Areas of interest are microbial ecology, microbial genetics, petroleum microbiology and microbial corrosion.

Esra Ilhan-Sungur is professor in the Biology Department at Istanbul University, Turkey, since 2018. A key focus of her research is microbiologically induced corrosion and its prevention. Further research interests lie in the area of anaerobic bacteria (especially sulfate-reducing bacteria), petroleum microbiology, microbial diversity and ecology, microbial genetics and biofilm. She was awarded a postdoctoral research scholarship by the Scientific and Technological Research Council of Turkey (TUBITAK-BIDEB) and worked as a guest researcher at Delft University of Technology.

Gerard Muyzer is Professor in Microbial Systems Ecology at the University of Amsterdam, The Netherlands. He is studying the structure, function and dynamics of microbial communities, their role in biogeochemical cycles and their application in biotechnological processes. For this he is using a systems biology approach in which he combines experimental work, the use of state-of-the-art omics techniques, and mathematical modelling. He is mainly focusing on the microbial sulfur cycle, and in particular on sulfur bacteria that are present in natural ecosystems (such as soda lakes, stratified lakes, rhizosphere of seagrasses) as well as man-made ecosystems, such as full-scale bioreactors removing toxic sulfur compounds from wastewater.




Begüm Çandiroğlu**Institute of Graduate Studies in Sciences, Istanbul University, Balabanaga Mah. Sehzadebasi Cd., 34134 Vezneciler, Fatih-Istanbul, Turkey

Nihal Doğruöz Güngör*Department of Biology, Faculty of Science, Istanbul University, Balabanaga Mah. Sehzadebasi Cd., 34134 Vezneciler, Fatih-Istanbul, Turkey

Email: *[email protected]; **[email protected]

Since cave ecosystems have extraordinary environmental conditions, these ecosystems offer opportunities for microbiological studies. In this study, cultivable bacteria isolated from Parsık cave, Turkey, were investigated regarding enzyme profiles, antibiotic resistance and potential for production of antimicrobial agents. The metabolic properties of 321 bacterial isolates were determined. The most produced enzyme by the isolates was found to be tyrosine arylamidase. The enzymatic reactions of the bacteria showed that Parsık cave isolates have high aminopeptidase activity. The highest antibiotic resistance frequency of the isolates was 38.6% against ampicillin. While the isolates displayed variable inhibition rates against tested pathogenic microorganisms, they showed the highest inhibition against Candida albicans. The results show that the bacteria isolated from Parsık cave have potential for further studies related to biotechnological applications. The study findings contribute increased knowledge on metabolic peculiarities of bacteria isolated from cave ecosystems.

1. Introduction

Caves are dark environments with high humidity, low nutrients, stable temperature and high mineral diversity. They are natural geological formations constituting ecological niches for microorganisms (1). Each cave is singular in its physical, chemical, biological and ecological factors. These conditions contribute to the formation of unique microbial communities in every cave. Moreover, caves contain some unique microorganisms which lead to rock weathering process and biomineralisation by carrying out various enzymatic reactions as a result of their metabolism. These microorganisms play an important and major role in the formation of cave structures such as stalactites, stalagmites, cave pearls and curtains (2–5). Studies have shown that cave isolates have biotechnological and industrial applications such as microplastic degradation (6), biological treatment of metal contaminated soil and groundwater (7) and use in self-healing concrete (8).The insufficient nutrient levels in caves stimulate

competition among microorganisms by forcing them to develop survival strategies such as producing high amounts of exopolymeric substances, enzymes and antimicrobial metabolites. Hence, caves could be considered as incomparable environments for the discovery of new antibiotics and production of novel enzymes (9–11). Since microorganisms have the capacity to

produce a high quantity of stable enzymes in a short period of time, they become the preferred source of industrial enzymes. Microbial enzymes are used in the clinical field for diagnosis, treatment, biochemical tests and monitoring of various diseases. Furthermore, cave microbial enzymes are

The Biotechnological Potentials of Bacteria Isolated from Parsık Cave, TurkeyMeasuring the enzyme profiles, antibiotic resistance and antimicrobial activity in bacteria



used in biotechnological and industrial fields such as biodegradation, recycling of waste (12), purification and dirt or waste-dissolving products. It is reported that enzymes from microorganisms isolated from cold cave or ocean environments offer economic benefits and contribute to energy conservation due to their activation at low temperatures (13, 14).Apart from the importance of enzymes isolated

from cave microorganisms, it is interesting to investigate the potential of producing new antimicrobial agents. Since the World Health Organization pointed out the need for new antibiotics because of increasing microbial resistance (15), studies in this field are multiplying and many cave isolates producing antimicrobial substances have been discovered. Cervimycin A, B, C and D from Streptomyces tendae strain HKI 0179 isolated from Grotta dei Cervi in Italy (16), Xiakemycin A from Streptomyces sp. CC8-201 isolated from Chongqing City karst soil in China (17), and Hypogeamicin A, B, C and D from Nonomuraea specus isolated from Hardin’s cave system in Tennessee, USA (18) were the first produced and purified bioactive substances from microorganisms of caves situated in different geographical regions. Bacteria in environments far away from human

influence are not expected to have antibiotic resistance. However, studies have shown that bacteria isolated from such environments do have antibiotic resistance. Some bacteria have resistance genes by

which they can produce neutralising or detoxifying products which act against microorganisms in the same environment. This explains the imperative production of antibiotics in these bacteria. Since the resistance and antimicrobial biosynthesis genes are often linked and coregulated, antibiotic resistance in environmental bacteria remains a major indicator of antibiotic production, as is the case of bacteria isolated from soil (19, 20). Therefore, it is important to establish antibiotic resistance profiles as well as the antibacterial properties of bacteria.This study has two main goals:

• Detection of enzyme profiles of the isolates and determination of isolates that have potential uses in biotechnology

• Investigation of antimicrobial agents and antibiotic resistance of cave bacteria.

2. Experimental

2.1 Studying Area and Sampling

Parsık cave is located in Izmit-Aksığın village (Global Positioning System (GPS) coordinates 40° 37’ 50.1060”N, 29° 57’ 56.5056”E), in the north-west of Turkey. It is a horizontal cave with a length of 778 m and a depth of 166 m. There is an intense water inlet in Parsık cave throughout four seasons. Samples were taken from water, soil and surface formations (‘moonmilk’) (Figure 1). The selected

Fig. 1. Map of Parsık cave (red dots show the sampling areas) from the Anatolian Speleology Association, Turkey

Plan

0 m

0 cm

6 m 12 m

10 cm5 cm

MANYETIK KUZEYNorth

B

A

C



sampling zones are the sole area away from the entrance area, trip and running water pathway. Although Parsık cave is not a show cave, it is open to cavers and researchers.Surface formation samples were collected

by sterile swabs under aseptic conditions and cultivated on starch casein agar (SCA), inorganic salt-starch agar (ISP4), soil extract agar (SEA) and Actinomycetes isolation agar (AIA-G) in duplicate for each region. Once the plates reached the laboratory, they were incubated aerobically for a period of 5–30 days at 20°C (21). All water and soil samples were taken in sterile sample containers.

2.2 Physicochemical Measurements of Sampling Areas

Humidity and temperature values of the sampling areas were measured by a portable temperature/humidity meter. In addition, the temperature, conductivity, amount of dissolved substances and pH values of the sampled water sources were measured during sampling and recorded by a HQ40D digital two channel multimeter (Hach Lange GmbH, Germany).

2.3 Total (Live/Dead) Bacteria Number

The redox dye 5-cyano-2,3-ditolyl-tetrazolium chloride (CTC) was used together with the DNA-binding fluorescent dye 4’,6-diamidino-2-phenylindole (DAPI) to determine the total number of bacterial cells and the viable count of bacteria which actively respire. The concept is to distinguish between the metabolically active cells and the dead cells present in each of the water and soil samples. The experimentation procedure is the same as previously described by Güngör and Yurudu (22).

2.4 Enumeration and Isolation of Culturable Aerobic Heterotrophic Bacteria

1 l of water samples were condensed by using polyamide filters of 0.22 µm pore size. Filters were re-suspended in 20 ml of sterile physiological saline water. 1 g of the soil samples was homogenised in 9 ml of sterile physiological saline water. All samples were cultivated using the 10-fold serial dilution method. Diluted samples were cultured on tap water agar (TWA) and Reasoner’s 2A agar (R2A) for enumeration and isolation of bacteria from water and soil samples. In addition, bacterial

isolation from soil samples was on SCA, ISP4, AIA-G, SEA and 1/2 tryptic soy agar (TSA) media, and that from water samples was on 1/2 TSA only. Plates were incubated aerobically for a period of

5–30 days at 20°C (21). At the end of incubation, plates which contained between 30 and 300 colonies were considered for both soil and water samples. Colonies which appeared different were selected for identification, then stored at –86°C for subsequent uses.

2.5 Identification of Cave Isolates and Their Enzymatic Reactions

Cave isolates were identified through biochemical tests performed in the VITEK® 2 system (bioMérieux SA, France). One of the three formats of this system is the VITEK® 2 Compact 30 which focuses mainly on the industrial microbiology-testing environment. Based on this industrial software, three reagent cards of VITEK® 2 Compact 30, named Gram-negative fermenting and non-fermenting bacilli (GN), Gram-positive cocci and non-spore-forming bacilli (GP) and Gram-positive spore-forming bacilli (BCL), were used to characterise the isolated bacteria following the procedure and data given by the system manufacturers. Reagent cards are based on established biochemical methods and developed substrates (23). The results of biochemical reactions were interpreted to establish enzymatic profiles of isolates.

2.6 Ability of Cave Bacteria to Produce Antimicrobial Materials

The ability of Bacilli or Actinobacteria to produce antimicrobial agents was tested on standard strains of fungi species of Candida albicans (ATCC® 10231TM) and bacterial species of Escherichia coli (ATCC® 8739TM), Pseudomonas aeruginosa (ATCC® 9027TM), Staphylococcus aureus (ATCC® 6538TM), Bacillus subtilis (ATCC® 6633TM), Staphylococcus epidermidis (ATCC® 12228TM), Klebsiella pneumoniae (ATCC® 4352TM), Enterococcus hirae (ATCC® 10541TM), vancomycin-resistant Enterococcus faecalis (VRE) (ATCC® 51299TM) and methicillin-resistant Staphylococcus aureus (MRSA) (ATCC® 33591TM).Bacterial suspensions containing 3 × 108 cells ml–1

of the selected isolates were prepared. 2.5 μl of each suspension were incubated on Mueller Hinton Agar (MHA) plates at 20°C for 24 h. After incubation, all media in which bacterial colonies were observed, were exposed to ultraviolet (UV) radiation in an



open laminar flow cabinet. Therefore, the vitality of the bacteria was destroyed. 1.5 × 108 cells ml–1 of 24 h fresh cultures of the standard strains were prepared. 100 μl of each suspension was mixed with TSA medium at 45°C. Subsequently, it was poured into the previously UV exposed plates, then incubated for 24 hours at 37°C after solidification. At the end of the incubation period, the growth of the standard bacteria in the TSA was investigated and the zone diameters were measured (24).

2.7 Susceptibility to Antibiotics

The sensitivity of 101 selected isolates to antibiotics was examined by using the disc diffusion method of Kirby-Bauer (21) in which 10 antibiotics were used: piperacillin (100 µg), erythromycin (15 µg), vancomycin (30 µg), ampicillin (10 µg), neomycin (10 µg), gentamycin (10 µg), chloramphenicol (30 µg), tetracycline (10 µg), rifampicin (30 µg) and ofloxacin (10 µg). The incubation conditions were 24 h at 20°C. Escherichia coli (ATCC® 8739TM), Pseudomonas aeruginosa (ATCC® 9027TM) and Staphylococcus aureus (ATCC® 6538TM) were tested against the same antibiotics as control microorganisms (25).


3.1 Physicochemical Measurements of Sampling Areas

Temperature, pH, conductivity and hardness values of water samples are shown in Table I. The air temperature of the sample areas A, B, C (Figure 1) was determined. The temperature of area A was 9.8°C and that of B and C were determined as 9.4°C. The moisture value was evaluated as 93% in all these areas. The Parsık cave resembles most cave systems with its high level of humidity and stable air temperature (26, 27). It was determined that the pH and hardness values of the waters at points B and C were higher than those at point A.

These details highlight the differences in chemical environment that may exist within the cave areas.

3.2 Number of Determined Total (Live/Dead) Bacteria

The highest vitality percentage of bacteria isolated in soil samples was found in samples from point B with 38.7%, whereas the highest vitality percentage in the water samples was found in samples from point C with 26.3% (Table II). In cave environments, it is observed that bacteria can survive metabolically but cannot be cultured. This is because bacteria enter a viable but nonculturable cell form under extreme environmental conditions such as low or high temperature, nutrient deficiency, osmolarity and light. In addition, cave microorganisms obtain their energy from the cave atmosphere or the cave surfaces to which they are attached (28, 29).

3.3 Number and Classification of Culturable Aerobic Heterotrophic Bacteria

SCA, ISP4, SEA and AIA-G have been used especially in surface and soil samples to increase the probability of isolating bacteria belonging to phylum Actinobacteria, which have an extremely high potentials in terms of antimicrobial production (30). TWA and R2A medium were used for both isolation and counting of other bacterial groups. Apart from these media, 1/2 TSA was used for isolation of other bacterial groups from all samples. The cave environment in general is oligotrophic and these media provide a similar environment to the culturable cave bacteria. The number of culturable aerobic heterotrophic bacteria from water and soil samples obtained from R2A and TWA media is given in Figure 2. When the bacterial counts of water and soil

samples in R2A and TWA media were examined, the highest bacterial numbers were found in R2A medium. These results were evaluated statistically using the Kruskal-Wallis test. The p value was found to be 0.09 and no significant difference was found between the numbers of bacteria grown on the R2A and TWA media. In a study conducted in 2014 (31), the efficiency of various media (SEA, TWA, SCA, TSA) was compared to their suitability for bacterial counting. Efficient results for both isolation and counting were obtained in TWA. A total of 372 bacteria were isolated from all

samples. VITEK® analysis was applied to only 321 bacteria which had different characteristics in

Table I Physicochemical Measurements of Water Samples

Measured values Water sample areasA point B point C point

pH 8.2 9.8 9.8

Hardness, ppt 0.107 0.145 0.145

Conductivity, mS 0.22 0.30 0.30

Temperature, °C 10 9.2 9.2



culture-based analyses. The results of the systematic classification of the bacteria were compiled by biochemical analysis using the VITEK® 2 Compact 30 automated system. Actinobacteria (33%) was determined to be the dominant phylum in this study while the other determined phyla were Firmicutes (25%) and Proteobacteria (16%). In our previous work in Kadıini cave in Turkey, the dominant phylum was Firmicutes (86%), followed by Proteobacteria (12%) and Actinobacteria (2%) respectively (32). In addition, in the study done by Tomova et al. (33), Proteobacteria (51.45%) were found to be the dominant phylum in the samples taken from the Magura cave, Bulgaria, followed by Actinobacteria (43.68%) and Bacteroidetes (3.88%). Although the bacterial habitat of each cave is specific, Proteobacteria, Actinobacteria, Bacteroidetes and Firmicutes are the most identified groups in culture-based microbiological studies in caves (34–36). In our study, Firmicutes was the most common

phylum in soil samples with a rate of 33%, while the most common phylum determined in surface

and water isolates was Actinobacteria with 36% and 35% respectively. Considering all the samples, at the class level, Actinobacteria was the most dominant with 33%, while Bacilli with 23% was detected as the second dominant class. It was demonstrated through previous studies that Actinobacteria existed mainly in cave walls, soil, sediment and on speleothem surfaces, which might have considerably contributed to the formation of cave structures and the biomineralisation in the cave ecosystems (4–37). Actinobacteria as well as Firmicutes are frequent among the microbial population inhabiting the caves. Compared to the Proteobacteria group, Firmicutes are more resistant to stress caused by dehydration as well as restriction of nutrients (37). Contrary to our findings for Parsık cave, Proteobacteria are a dominant group in heterotrophic bacterial communities in many caves (33, 34, 38–40). In the current study, Proteobacteria were determined respectively as 10%, 21% and 17% in the surface, water and soil samples. The dominant classes of this phylum were found to be Gammaproteobacteria and Alphaproteobacteria with 9.2% and 6.4%, respectively. In our previous study in Kadıini cave, Alphaproteobacteria were detected at 2%, while Gammaproteobacteria were at 9% (32). The phylum Proteobacteria, having a key role in biogeochemical cycles, and being abundant in samples from cave sediment, soil, dripping water and cave surface, is a cosmopolitan bacterial group (37).

3.4 Enzymatic Reactions of Parsık Cave Bacteria

Enzymatic reactions of microorganisms give us ideas of their metabolic activities which are related to their environment. The biochemical tests of our isolates in the VITEK® system were not only useful for bacterial identification but

Fig. 2. Number of aerobic heterotrophic bacteria that can be cultured from water and soil samples (TA = soil sample A; TB = soil sample B; TC = soil sample C; SA = water sample A; SB = water sample B; SC = water sample C)

7

6

5

CFU

ml–

1 lo

g, 1

0

4

3

2

1

0TA TB TC

Points of samplesSA SB SC

R2A

TWA

log(

10)

Table II Number of Bacteria in Water and Soil Samples According to DAPI/CTC MethodSamples Total number of signals, cm2 Vitality, %

CTC Total

SA 406,505,880 2,947,167,630 13.8

SB 135,501,960 643,634,310 21

SC 508,132,350 1,930,902,930 26.3

TA 1,084,015,680 5,318,451,930 20.3

TB 8,130,117,600 21,002,803,800 38.7

TC 6,097,588,200 20,664,048,900 29.5



also to provide more information about nutrients in Parsık cave. In addition, results of these tests were used to evaluate the potentials of the isolates for biotechnological uses in terms of their enzyme production. 76 Gram-negative bacteria, 194 Gram-positive and 51 Gram-positive spore forming bacteria have been tested using the GN, GP and BCL cards respectively in the VITEK® 2 compact device, and results are given in Figure 3, Figure 4 and Figure 5 respectively. Most of the isolates displayed peptidase (arylamidase) while only Gram-negative bacteria (less than 10%) showed lipolytic activity. In the study conducted in Gumki cave, India, 75.5% of bacteria produced lipase, 47% were amylase producers and 24% produced protease (41). Another study screening cave bacteria for enzyme production found 40% lipase and 87% protease producers (33). This variation in enzymatic profiles in cave bacteria reinforces the idea that every cave is unique.The high activity of amino acids arylamidase

determined in our tested isolates indicates their potential for protein catabolism (42). The phyla Firmicutes (31%) and Actinobacteria (30.7%) produced the highest amounts of arylamidases identified among the tested isolates. 85.52%, 65.97% and 82.35% of Gram-negative, Gram-

positive and Gram-positive spore forming bacilli revealed tyrosine-arylamidase activity. Tyrosine is a non-essential amino acid which is synthesised through phenylalanine hydrolysis. It plays a major role in most enzyme synthesis as reported by Kalkan and Altuğ (42), since it is the phosphate and sulfate receptor of protein kinase during protein synthesis. It is also used to reinforce the activity of proteins as demonstrated in a study conducted in thrombin inhibitors showing that tyrosine sulfation could open a way for the development of an anti-thrombotic drug (43). Hence, tyrosine arylamidase has a valuable role in biotechnology since it contributes to the liberation of the amino acid tyrosine.Enzymes like leucine arylamidase have been

reported to be important in food processing industries and the treatment of waste products (44, 45). The degradation of leucine and other amino acids results in volatile molecules responsible for the flavours of some foods like meat products as reported by Papamanoli et al. (46) and Lee et al. (44). In addition, a study showed the roles of bacteria in conversion of paper mill sludge demonstrating the important contribution of amino acid peptidase with leucine arylamidase (45). In our study, 81.95% and 88.23% of Gram-positive

Fig. 3. Biochemical properties of Gram-negative bacteria. Tests for Gram-negative bacteria by VITEK® 2 Compact 30 micro device. See Glossary in main text for explanation of terms

70

60

50

APP

AAD

OPy

rAIA

RL

dCEL

BG

AL

BN

AG

AG

LTp

dGLU

GG

TO

FFBG

LUdM

AL

dMAN

dMN

EBXYL

BAla

pPr

oA LIP

PLE

TyrA

URE

dSO

RSA

CdT

AG

dTRE

CIT

MN

T5K

GIL

ATk

AG

LUSU

CT

NAG

AAG

AL

PHO

SG

lvA

OD

CLD

CIH

ISa

CM

T01

29R

GG

AA

IMLT

aEL

LM

40

30

20

10

0

Gr– bacteria

O



bacteria and Gram-positive bacilli showed leucine arylamidase activity. This enzyme was the second most produced enzyme, after the tyrosine arylamidase, by our isolates. Bacteria which can

produce this enzyme could be used directly or indirectly by using their enzymes in both composting of sludge and fermentation of food products such as meat and dairy products.

Fig. 5. Biochemical properties of Gram-positive Bacilli bacteria. Tests for Gram-positive Bacilli bacteria by VITEK® 2 Compact 30 micro device. See Glossary in main text for explanation of terms

LysA

LeuA

PheA

APP

A

AM

AN

Asp

A

INO

BG

AL

CD

EX

AG

LU

ProA

ELLM

BXYL

AG

AL

Ala

A

IRH

A

PHC

dMIZ

dRIB

ESC

OLD TTZ

NC6.

5%NAG

Gr(+) bacilli bacteria

60

50

40

30

20

10

0

PyrA

BN

AG

dGLU

BG

LU

dGAL

GLY

G

Gly

AdM

AN

dMN

E

PLE

TyrA

dTAG

MTE INU

PVAT

E

KAN

POLY

B R

POLY

B_R

NC6.

5

Fig. 4. Biochemical properties of Gram-positive bacteria. Tests for Gram-positive bacteria by VITEK® 2 Compact 30 micro device. See Glossary in main text for explanation of terms

60

AM

YPI

PLC

dXYL

AD

H1

APP

A

AM

AN

Asp

A

PyrA

BG

AL

CD

EX

AG

LU

LeuA

ProA

PHO

S

BG

AR

dMAL

dMAN

dMN

EM

BdG PU

LdR

AF

SAL

SAC

AD

H2s

OPT

O

BG

URr

BG

UR

AG

AL

Ala

A

BACI

ILAT

k

POLY

B

TyrA

URE

dSO

R

dGAL

dRIB

dTRE

NO

VO

NC6.

5

NAG

LAC

0129

R

40

20

100

120

140

160

180

80

0

Gr+ bacteria

O

POLY

B_R



VITEK® results have showed that some Parsık cave isolates exhibit beta-galactosidase activity which is the more expressed carbohydrate hydrolase in this study. Considering the whole of the tested isolates, most of the bacteria producing beta-galactosidase belong to the Firmicutes phylum with 40.6%, while only 10.9% of beta-galactosidase producers were classified under the phylum Proteobacteria. The main role of the beta-galactosidase enzyme is

to convert lactose into monosaccharides. Glucose and galactose resulting from this reaction not only contribute to the development of the cell but can also be used in dairy product processing (44, 47). This enzyme is important since it solves the problem of human lactose intolerance. The hydrolysation of lactose by this enzyme results in molecules like galactooligosaccharides which have health benefits as prebiotics (47). Moreover, breakdown of some sugars like D-mannose, D-mannitol and D-glucose by fermentation was reported, especially in Gram-negative bacteria. Lipolytic activity was also observed in some of our

isolated Gram-negative bacteria (less than 10%). Even if it was produced by a minimum number of isolates, the activity of lipase was fully expressed by bacteria belonging to the phylum Proteobacteria. This class of enzymes which is used in hydrolysation of lipids could be important in bioremediation since it could participate in oil degradation. Sharma et al. reported that microbial lipases are best for biodiesel production (48). Since they can use all types of free fatty acids and glycerides, they exhibit a high activity, thermostability, alcohol resistance, less reaction time as well as less production inhibition (48). Other enzymes were produced by some of the bacteria in Parsık caves. Further studies should be carried out to clarify them and assess their biotechnological uses.

3.5 Antimicrobial Agent Production Capability

Microorganisms with broad-spectrum bioactive components, antifungal and antibacterial agents in cave-specific habitats are common in these extreme environments (17). In our study, a total of 129 cave bacteria were tested for their antimicrobial effect against nine different standard bacterial strains and one fungal strain. Experiments have shown that 10 of the selected bacteria (six from Actinobacteria class, four from Bacilli class) have antimicrobial effects against the standard strains. Parsık cave isolates displayed variable inhibition

rates against the tested microorganisms and

the highest inhibition rate was observed against Candida albicans. Some of our cave isolates have been found to have inhibitory effects against S. aureus, S. epidermidis, VRE and P. aeruginosa. The zone diameters of cave bacteria with antimicrobial properties against tested microorganisms are shown in Table III. In our study, the isolate which affected S.

epidermidis belongs to the Bacilli class and those which inhibit VRE and S. aureus belong to the Actinobacteria class. Some studies have shown that bacteria with antimicrobial activity inhabiting karst caves are often from the Actinobacteria class (30, 31). However, cave bacteria belonging to phyla Proteobacteria, Firmicutes (especially Bacilli class) and Bacteroides were determined to have antimicrobial and bioactive substances. Thus, approximately 50% of the bacteria isolated from the Magura cave, Bulgaria were detected to inhibit the increase of P. aeruginosa (33). Cave bacteria inhibiting MRSA and VRE clinical strains were determined in a study on Actinomycetes isolated from 19 different caves in Turkey (30). Certainly the bacteria belonging to the class Actinobacteria are the best known in terms of antimicrobial material synthesis, but the isolation of bacteria belonging to the other classes is very important especially in karst environments.

3.6 Determination of Antibiotic Resistance Profiles of Isolated Bacteria

Antibiotic resistant bacteria are widespread in several environments. In this study, resistance to 10 different antibiotics of 101 bacteria (76% Gram-positive; 25% Gram-negative) selected from the cave isolates was investigated. Isolates with a metabolic reaction rate of at least 95% similarities to the data in the VITEK® database were selected.When the antibiotic resistance profiles of the

isolates were examined, 7% of the bacteria belonging to the cave were resistant to all antibiotics. The highest number of bacteria showed resistance against ampicillin with a rate of 38.6%. In addition, 35.6% of the isolates showed resistance against two or more antibiotics. Antibiotic resistance profiles of Gram-positive and

Gram-negative cave isolates are shown in Figure 6. The lowest resistance was observed to rifampicin (9% for Gram-positive and 8% for Gram-negative). In parallel with our study, it was determined that all the Pajsarjeva jama, Slovenia, isolates were sensitive to rifampicin (49). Likewise, low levels



Tab

le I

II A

nti

mic

rob

ial A

gen

t P

rod

uct

ion

Ab

ility

Isol

ates

/cl

asse

s of

b

acte

ria

Res

ult

ing

zon

e d

iam

eter

s, m

m

E. c

oli

E. f

aeca

lisB

. su

bti

lisS

. au

reu

sS

. ep

ider

mid

isM

RS

AV

RE

P.

aeru

gin

osa

K.

pn

eum

onia

eC

and

ida

alb

ican

s

SA

22

/ A

ctin

obac

teri

a–

––

––

–9

––

–

TA4

4/

Act

inob

acte

ria

––

––

––

–16

––

TA1

2/

Act

inob

acte

ria

––

––

––

–13

.5–

–

SA

56

/ A

ctin

omyc

etes

––

–13

–

––

––

–

TB4

8/

Bac

illi

––

––

––

––

–13

SB

1/

Bac

illi

––

––

––

––

–30

TA6

2/

Act

inom

ycet

es–

––

––

––

––

24

TB2

7/

Bac

illi

––

––

15

––

––

–

SC

3/

Bac

illi

––

––

––

––

–30

TB6

4/

Act

inob

acte

ria

––

––

––

––

–13

An

tib

ioti

cs

Pip

erac

illin

1124

1913

209

3116

22N

D

Van

com

ycin

–16

138

1021

18–

–N

D

Gen

tam

icin

116.

516

813

20–

1122

ND

Tetr

acyc

line

10–

168

14–

145

19N

D

Rif

amp

icin

716

1710

1835

288

15N

D

Ofl

oxac

in15

1821

1315

3130

2335

ND

ND

= n

ot d

eter

min

ed



of resistance to ofloxacin, which is a DNA/RNA synthesis inhibitor like rifampicin, were observed in Parsık cave isolates (11% in Gram-positive and 12% in Gram-negative). The resistance rate of Pajsarjeva jama isolates to erythromycin was 73.6% for Gram-negative and 39% for Gram-positive bacteria. The resistance of Parsik cave isolates to erythromycin was determined at lower levels of 20% and 21% for Gram-negative and Gram-positive bacteria respectively. The levels of resistance to protein synthesis inhibitors other than erythromycin (gentamycin, neomycin, tetracycline and chloramphenicol) were determined to range from 12% to 20% for both Gram-positive

and Gram-negative bacteria. Contrary to our study, Lavoie et al. (50) showed that cave isolates were highly resistant to gentamycin, neomycin and chloramphenicol antibiotics (33–66% for Gram-negative bacteria and 61–83% for Gram-positive bacteria).Furthermore, the lowest resistance to cell wall

synthesis inhibitors was observed in piperacillin for both Gram-positive (12%) and Gram-negative (16%) bacteria. In the study conducted by Lavoie et al. (50), the resistance to piperacillin, compared to other antibiotic resistance, was found to be lower (average 37.5%).

Fig. 6. Resistance levels of Parsık cave bacteria against various antibiotics which are grouped according to their mode of action: (a) Gram-positive isolates; (b) Gram-negative isolates

30

35(a)

(b)

Gram-positive isolates

Gram-negative isolates

Piper

acilli

n

Vanc

omyc

in

Ampic

illin

Neomyc

in

Gentamyc

in

Tetra

cycli

ne

Erythr

omyc

in

Antibiotics

Antibiotics

Chlor

amph

enico

l

Rifam

picin

Ofloxa

cin

20

10

15

25

5

0

Perc

enta

ge o

f ba

cter

ia

show

ing

resi

stan

cePe

rcen

tage

of

bact

eria

sh

owin

g re

sist

ance

70

60

50

40

30

20

10

0

Piper

acilli

n

Vanc

omyc

in

Ampic

illin

Neomyc

in

Gentamyc

in

Tetra

cycli

ne

Erythr

omyc

in

Chlor

amph

enico

l

Rifam

picin

Ofloxa

cin

Cell wall synthesis inhibitors

30S ribosome synthesis inhibitors

50S ribosome synthesis inhibitors

DNA/RNA synthesis inhibitors



In our study, considering the cell wall synthesis inhibitors (vancomycin and ampicillin), Gram-negative bacteria were found to be more resistant than the Gram-positive ones. Similar to our study, Avguštin et al. (49) revealed that cave Gram-negative isolates showed higher resistance to ampicillin.According to VITEK® results, except ampicillin

and vancomycin, Actinobacteria were determined to be the most resistant (47–70%) phylum to all tested antibiotics. The highest resistance to ampicillin and vancomycin was observed in the phylum Proteobacteria. Like the microbial diversity of caves, antibiotic resistance is also variable. While the antibiotic resistance rates were high, no isolate producing antimicrobial agent was detected in the study conducted by Lavoie et al. (50). However, one of the antibiotic resistance hypotheses in caves is that there is a high rate of antibiotic resistance in the presence of microorganisms producing antimicrobial agents. Studies have shown that bacteria having antibiotic genes can also produce antimicrobial agents (51, 52). In our study, it was found that 50% of the isolates producing antimicrobial agents were resistant to at least two antibiotics. Therefore, study of bacterial antibiotic resistance may contribute to the development of new antibiotics. To clarify this issue, studies in this issue should be continued.

4. Further Work

In our future studies, we are planning to purify and use the enzymes and antimicrobial substances

produced by these isolates. Apart from the potential of bacteria isolated from Parsık cave to produce enzymes and antimicrobial agents, it is planned to determine their potential use in biodegradation, self-healing concrete and production of antimicrobial drugs against phytopathogens and entomopathogens.

5. Conclusion

The microorganisms attached to the specific environmental conditions of caves are important in terms of exploring their uses and specific features. This study was the first microbiological study in Parsık cave. It has been demonstrated that enzymes such as arylamidases, carbohydrate hydrolases and lipases found in bacteria isolated from Parsık cave can be important in industrial as well as clinical fields. In addition, some of our isolates have shown antimicrobial activity and can contribute to the development of new antibiotics. Antibiotic resistance profiles of Parsık cave isolates should be clarified in further studies through studies of their genes.

Acknowledgements

The author would like to thank the Anatolian Speleology Association, Turkey, and Nahdhoit Ahamada Rachid, Department of Fundamental and Industrial Microbiology of Istanbul University, for their contributions. The author also thanks Istanbul University Scientific Project Unit, Turkey, (BAP Project No FYL-2016-20759 and FHZ-2017-26457) for financial support.

Glossary

Term Definition5KG 5-keto-D-gluconate

ADH1 arginine dihydrolase 1

ADH2s arginine dihydrolase 2

ADO Adonitol

AGAL alpha-galactosidase

AGLTp glutamyl arylamidase pNA

AGLU alpha-glucosidase

AIA-G Actinomycetes isolation agar

AlaA alanine arylamidase

AMAN alpha-mannosidase

AMY D-amygdalin

APPA Ala-Phe-Pro-arylamidase

AspA L-aspartate arylamidase

BACI bacitracin resistance

BAlap beta-alanine arylamidase pNA

BCL Gram-positive spore-forming bacilli

BGAL beta-galactosidase

BGAR beta-galactopyranosidase

BGLU beta-glucosidase

BGUR/ BGURr beta-glucuronidase

BNAG beta-N-acetyl-glucosaminidase

BXYL beta-xylosidase

CDEX cyclodextrin

CIT citrate (sodium)



CMT coumarate

CTC 5-cyano-2,3-ditolyl-tetrazolium chloride

DAPI 4’,6-diamidino-2-phenylindole

dCEL D-cellobiose

dGAL D-galactose

dGLU D-glucose

dMAL D-maltose

dMAN D-mannitol

dMLZ D-melezitose

dMNE D-mannose

dRAF D-raffinose

dRIB D-ribose

dSOR D-sorbitol

dTAG D-tagatose

dTRE D-trehalose

dXYL D-xylose

ELLM Ellman

ESC esculin hydrolysis

GGAA Glu-Gly-Arg-arylamidase

GGT gamma-glutamyl-transferase

GlyA glycine arylamidase

GLYG glycogen

GN Gram-negative fermenting and non-fermenting bacilli

GP Gram-positive cocci and non-spore-forming bacilli

GPS Global Positioning System

IARL L-arabitol

IHISa L-histidine assimilation

ILATk L-lactate alkalinisation

IMLTa L-malate assimilation

INO myo-inositol

INU inulin

IRHA L-rhamnose

ISP4 inorganic salt-starch agar

KAN kanamycin resistance

LAC lactose

LDC lysine decarboxylase

LeuA leucine arylamidase

LIP lipaseLysA L-lysine-arylamidaseMBdG methyl-beta-D-glucopyranosideMHA Mueller Hinton AgarMNT malonate

MRSA methicillin-resistant Staphylococcus aureus

MTE maltotrioseNAG N-acetyl-D-glucosamineNAGA beta-N-acetyl-galactosaminidaseNC6.5 growth in 6.5% NaClNOVO novobiocin resistanceO129R O/129 resistance (comp. vibrio.)ODC ornithine decarboxylaseOFF fermentation/glucoseOLD oleandomycin resistanceOPTO optochin resistancePHC phosphoryl cholinePheA phenylalanine arylamidasePHOS phosphatasePIPLC phosphatidylinositol phospholipase CPLE PalatinoseTM

POLYB_R polymyxin B resistanceProA L-proline arylamidasePUL pullulanPVATE pyruvatePyrA L-pyrrolidonyl-arylamidaseR2A Reasoner’s 2A agarSAC saccharose/sucroseSAL salicinSCA starch casein agarSEA soil extract agarSUCT succinate alkalinisationTSA 1/2 tryptic soy agarTTZ tetrazolium redTWA tap water agarTyrA tyrosine arylamidaseURE ureaseUV ultraviolet

VRE vancomycin-resistant Enterococcus faecalis

References

1. C. Schabereiter-Gurtner, C. Saiz-Jimenez, G. Piñar, W. Lubitz and S. Rölleke, FEMS Microbiol. Ecol., 2004, 47, (2), 235

2. H. A. Barton, J. Caves Karst Stud., 2006, 68, (2), 43

3. L. C. Kelly, C. S. Cockell, A. Herrera-Belaroussi, Y. Piceno, G. Andersen, T. DeSantis, E. Brodie, T. Thorsteinsson, V. Marteinsson, F. Poly and X. LeRoux, Microb. Ecol., 2011, 62, (1), 69

4. S. Cuezva, A. Fernandez-Cortes, E. Porca, L. Pašić, V. Jurado, M. Hernandez-Marine, P. Serrano-Ortiz, B. Hermosin, J. C. Cañaveras, S. Sanchez-Moral



and C. Saiz-Jimenez, FEMS Microbiol. Ecol., 2012, 81, (1), 281

5. A. Rusznyák, D. M. Akob, S. Nietzsche, K. Eusterhues, K. U. Totsche, T. R. Neu, T. Frosch, J. Popp, R. Keiner, J. Geletneky, L. Katzschmann, E.-D. Schulze and K. Küsel, Appl. Environ. Microbiol., 2012, 78, (4), 1157

6. S. U. U. Jamil, S. Zada, I. Khan, W. Sajjad, M. Rafiq, A. A. Shah and F. Hasan, J. Caves Karst Stud., 2017, 79, (1), 73

7. N. K. Dhami, M. E. C. Quirin and A. Mukherjee, Ecol. Eng., 2017, 103, (A), 106

8. A. Nugroho, A. Sumarno, L. N. Ngeljaratan, D. Zulfiana, N. P. R. A. Krishanti, T. Triastutil and E. Widodo, J. Kim. Terap. Indones., 2019, 21, (1), 7

9. A. A. Elmanama, M. T. Alhour, J. Adv. Sci. Eng. Res., 2013, 3, (4), 388

10. S. Krishnapriya, D. L. Venkatesh Babu, G. Prince Arulraj, Microbiol Res., 2015, 174, 48

11. A. I. Omoregie, ‘Isolation, Identification and Characterisation of Urease-Producing Bacteria from Limestone Caves of Sarawak’, in ‘Characterization of Ureolytic Bacteria Isolated from Limestone Caves of Sarawak and Evaluation of their Efficiency in Biocementation’, Master of Science (Research) Thesis, Chapter 2, Faculty of Engineering, Computing and Science, Swinburne University of Technology, Melbourne, Australia, 2016, pp 41–92

12. S. Mondal and D. Palit, ‘Effective Role of Microorganism in Waste Management and Environmental Sustainability’, in “Sustainable Agriculture, Forest and Environmental Management”, Eds. M. K. Jhariya, A. Banerjee, R. S. Meena and D. K. Yadav, Springer Nature Singapore Pte Ltd, Singapore, 2019, pp 485–515

13. C. Gerday, M. Aittaleb, M. Bentahir, J. P. Chessa, P. Claverie, T. Collins, S. D’Amico, J. Dumont, G. Garsoux, D. Georlette, A. Hoyoux, T. Lonhienne, M.-A. Meuwis and G. Feller, Trends Biotechnol., 2000, 18, (3), 103

14. M. S. Cabeza, F. L. Baca, E. M. Puntes, F. Loto, M. D. Baigorí and V. I. Morata, Food Technol. Biotech., 2011, 49, (2), 187

15. Z. Kmietowicz, Brit. Med. J., 2017, 358, j4430

16. K. Herold, F. A. Gollmick, I. Groth, M. Roth, K.-D. Menzel, U. Möllmann, U. Gräfe and C. Hertweck, Chem. Eur. J., 2005, 11, (19), 5523

17. Z. Jiang, L. Guo, C. Chen, S. Liu, L. Zhang, S. Dai, Q. He, X. You, X. Hu, L. Tuo, W. Jiang and C. Sun, J. Antibiot., 2015, 68, (12), 771

18. D. K. Derewacz, C. R. McNees, G. Scalmani, C. L. Covington, G. Shanmugam, L. J. Marnett, P. L. Polavarapu and B. O. Bachmann, J. Nat. Prod., 2014, 77, (8), 1759

19. V. M. D’Costa, C. E. King, L. Kalan, M. Morar, W. W. L. Sung, C. Schwartz, D. Froese, G. Zazula, F. Calmels, R. Debruyne, G. B. Golding, H. N. Poinar and G. D. Wright, Nature, 2011, 477, (7365), 457

20. J. R. Nodwell, J. Bacteriol., 2007, 189, (10), 3683

21. S. Massa, M. Caruso, F. Trovatelli and M. Tosques, World J. Microbiol. Biotechnol., 1998, 14, (5), 727

22. N. Doğruöz Güngör and N. Ö. Şanlı Yürüdü, “The Battle Against Microbial Pathogens: Basic Science, Technological Advances and Educational Programs”, ed. A. Méndez-Vilas, Vol. 5, Formatex, Badajoz, Spain, 2015, pp 923–929

23. D. H. Pincus, ‘Microbial Identification Using the bioMérieux VITEK® 2 System’, in “Encyclopedia of Rapid Microbiological Methods”, ed. M. J. Miller, Parenteral Drug Association, Bethesda, USA, 2006, pp 1–32

24. K. B. Ritchie, M. Schwarz, J. Mueller, V. A. Lapacek, D. Merselis, C. J. Walsh and C. A. Luer, Front. Microbiol., 2017, 8, 1050

25. M. J. Ferraro, M. A. Wikler, W. A. Craig, M. N. Dudley, G. M. Eliopoulos, D. W. Hecht, J. Hindler, L. Barth Reller, A. T. Sheldon, J. M. Swenson, F. C. Tenover, R. T. Testa and M. P. Weinstein, ‘Performance Standards for Antimicrobial Disk Susceptibility Tests: Approved Standard’, 8th Edn., M2-A8, Vol. 23, No. 1, National Committee for Clinical Laboratory Standards (NCCLS), Wayne, USA, 2003, 63 pp

26. S. Leuko, K. Koskinen, L. Sanna, I. M. D’Angeli, J. De Waele, P. Marcia, C. Moissl-Eichinger and P. Rettberg, PLoS One, 2017, 12, (7), e0180700

27. K. H. Lavoie, A. S. Winter, K. J. H. Read, E. M. Hughes, M. N. Spilde and D. E. Northup, PloS One, 2017, 12, (2), e0169339

28. H. A. Barton, M. R. Taylor and N. R. Pace, Geomicrobiol. J., 2004, 21, (1), 11

29. M. K. Chelius and J. C. Moore, Geomicrobiol. J., 2004, 21, (2), 123

30. S. Yücel and M. Yamaç, Pak. J. Pharm. Sci., 2010, 23, (1), 1

31. B. H. Velikonja, R. Tkavc and L. Pašić, Int. J. Speleol., 2014, 43, (1), 45

32. N. Doğruöz-Güngör, B. Çandıroğlu, G. Altuğ, J. Caves Karst Stud., 2020, 82, (2), 106

33. I. Tomova, I. Lazarkevich, A. Tomova, M. Kambourova and E. Vasileva-Tonkova, Int. J. Speleol., 2013, 42, (1), 65

34. J. P. Zhou, Y. Q. Gu, C. S. Zou and M. H. Mo, J. Microbiol., 2007, 45, (2), 105

35. H. A. Barton and V. Jurado, Microbe, 2007, 2, 132

36. V. Jurado, E. Porca, S. Cuezva, A. Fernandez-Cortes, S. Sanches-Moral and C. Saiz-Jimenez, Sci. Total Environ., 2010, 408, (17), 3632



37. K. Tomczyk-Żak and U. Zielenkiewicz, Geomicrobiol J., 2016, 33, (1), 20

38. M. Yasir, Braz. J. Microbiol., 2018, 49, (2), 248

39. V. Ivanova, I. Tomova, A. Kamburov, A. Tomova, E. Vasileva-Tonkova, M. Kambourova, J. Caves Karst Stud., 2013, 75, (3), 218

40. C. Schabereiter-Gurtner, C. Saiz-Jimenez, G. Piñar, W. Lubitz and S. Rölleke, FEMS Microbiol. Lett., 2002, 211, 7

41. R. Rautela, S. Rawat, R. Rawat, P. Verma, A. B. Bhatt, Environ. Conserv. J., 2017, 18, (3), 115

42. S. Kalkan, and G. Altuğ, Environ. Monit. Assess., 2020, 192, (6), 356

43. R. E. Thompson, X. Liu, J. Ripoll-Rozada, N. Alonso-García, B. L. Parker, P. J. B. Pereira and R. J. Payne, Nature Chem., 2017, 9, (9), 909

44. N. K. Lee, J. Y. Hong, S. H. Yi, S. P. Hong, J. E. Lee and H. D. Paik, J. Funct. Foods, 2019, 58, 324

45. R. K. Ganguly, and S. K. Chakraborty, J. Environ. Health Sci. Eng., 2018, 16, (2), 205

46. E. Papamanoli, N. Tzanetakis, E. Litopoulou-Tzanetaki and P. Kotzekidou, Meat Sci., 2003, 65, (2), 859

47. J. R. Xavier, K. V. Ramana, R. K. Sharma, J. Food Biochem., 2018, 42, (5), e12564

48. A. Sharma, Shadiya, T. Sharma, R. Kumar, K. Meena, S. S. Kanwar, ‘Biodiesel and the Potential Role of Microbial Lipases in Its Production’, in “Microbial Technology for the Welfare of Society: Microorganisms for Sustainability”, ed. P. K. Arora, Vol. 17, Springer Nature Singapore Pte Ltd, Singapore, 2019, pp. 83–89

49. J. A. Avguštin, P. Petrič and L. Pašić, Int. J. Speleol., 2019, 48, (3), 295

50. K. Lavoie, T. Ruhumbika, A. Bawa, A. Whitney and J. De Ondarza, Diversity, 2017, 9, (4), 42

51. M. K. Gibson, B. Wang, S. Ahmadi, C.-A. D. Burnham, P. I. Tarr, B. B. Warner and G. Dantas, Nature Microbiol., 2016, 1, (4), 16024

52. A. C. Pawlowski, W. Wang, K. Koteva, H. A. Barton, A. G. McArthur and G. D. Wright, Nature Commun., 2016, 7, 13803

The Authors

Nihal Doğruöz Güngör is an Associate Professor in the Department of Fundamental and Industrial Microbiology at Istanbul University, Turkey. She obtained her doctorate at Istanbul University in 2008, focusing on microbiological corrosion of copper. Her research interests include cave microbiology, antimicrobial activities of bacteria, microbial corrosion and biotechnology.

Begüm Çandıroğlu is a biologist from Istanbul University. In 2014 she obtained her MSc in microbiology in the Department of Fundamental and Industrial Microbiology at the same university. She studied cave microbiology and worked with cave bacteria that live in soil, water and cave walls. She is interested in biotechnology, enzymatic reactions of bacteria and antibiotic resistance.




Didem Berber*Department of Biology, Faculty of Arts and Sciences, Marmara University, Istanbul, Turkey; Gastronomy and Culinary Arts Department, Faculty of Fine Arts, Maltepe University, Marmara Eğitim Köyü, Istanbul, Turkey

İpek TürkmenoğluBiology Department, Institute of Pure and Applied Sciences, Marmara University, Istanbul, Turkey

Nüzhet Cenk SesalDepartment of Biology, Faculty of Arts and Sciences, Marmara University, Istanbul, Turkey


Antibacterial resistant bacteria are a significant problem in the hide or skin soaking process due to their destructive properties on finished leather. Lichens may be a solution to overcome this resistance problem. Enterococcus durans (99.86%) was isolated from soak liquor samples. For screening of possible antibacterial effects of lichen acetone extracts, six lichen species (Hypogymnia tubulosa, H. physodes, Evernia divaricata, Pseudevernia furfuracea, Parmelia sulcata and Usnea sp.) were examined by nine-fold dilution against E. durans. H. tubulosa, H. physodes and E. divaricata extracts showed antibacterial effects at the concentrations of 240 µg ml–1, 120 µg ml–1 and 60 µg ml–1 whereas the extracts of P. furfuracea had an antibacterial effect at 240 µg ml–1 and 120 µg ml–1. On the other hand, P. sulcata had no antibacterial

effect. The most successful lichen extract was determined to be Usnea sp. at the concentrations of 240 µg ml–1, 120 µg ml–1, 60 µg ml–1, 30 µg ml–1 and 15 µg ml–1. In conclusion, lichen extracts seem to have potential antibacterial efficacies against E. durans.

1. Introduction

The leather industry produces and exports high-quality products with high added value to the world market. However, several bacterial problems during leather-making processes are reflected in finished products and lead to economic losses. After the flaying process in slaughterhouses, microflora on hide or skin surfaces change due to bacterial contamination originating from faeces, air, dust or the animal skin itself and some bacteria easily colonise (1–4). The soaking process is the first tannery operation

that recovers water loss during raw hide or skin curing applications. There are some criteria to be taken into consideration during the soaking process of raw hides or skins. Especially prolonged soaking provides a convenient milieu for bacterial activity and damage to hides or skins may occur. Due to reduced salt content and high protein and lipid constituents, hides or skins become defenceless against bacterial attacks in the soaking process (5–8). It has been reported that the number of bacterial populations in soak liquors may be up to 105 colony forming unit (CFU) ml–1 (5). But in a previous study, it was demonstrated that total bacterial numbers were considerably higher than 105 CFU ml–1 in soak liquor samples (9). The adverse effects of the soaking process on the hide quality originate from

Antibacterial Potential of Six Lichen Species against Enterococcus durans from Leather IndustryEvaluation of acetone extracts obtained from several lichen species as alternative natural antibacterial agents



degradative enzymatic properties of bacteria such as protease and lipase activities. These enzymatic activities can irreversibly affect the structure of hide or skin substances that cannot be fixed at the subsequent stages of hide processing (10). High numbers of bacteria with protease and lipase activities cause unwanted defects such as hair-slip, putrefaction, grain peeling, loose grain, holes on the hides or skins and light stains on the suede surface (1, 3, 11–15). Antibiotics are used in various industries as

well as in the treatment of diseases. The World Health Organization declared that antimicrobial resistance in most countries and industrial sectors has increased dramatically (16, 17). The emergence of antibiotic-resistant bacteria due to improperly used antibiotics in humans, animals and agriculture has been reported in the literature (17). In the leather industry, to control bacterial numbers and their degradative properties on hides or skins, various antibacterial agents are utilised during the soaking process of beam house operations. The normal microflora in animals comprises many harmless bacteria but any of them may become resistant to commonly utilised antibacterial agents due to intrinsic or acquired resistance (17, 18). The resistant bacteria may survive despite bactericides and may transfer their resistance properties to others through horizontal gene transfer (5, 9, 18). Bactericides may remain ineffective against proteolytic and lipolytic bacteria in soak liquors because of high organic content in soak liquors (9, 19). The existence of many non-halophilic bacteria was demonstrated in the presence of an antimicrobial agent at twofold increased concentration (0.8 g l–1) (19). This finding emphasises the antibacterial resistance of bacteria in the soaking process. More recently, it was reported that antimicrobial agents used in the soaking process could not control multidrug-resistant Enterobacteriaceae from soaked sheepskins and cattle hides treated with an antibacterial agent (20).Over the past decades, it has been suggested

that alternative compounds from natural resources may overcome the antimicrobial resistance of many bacteria. Previously, the potential of lichen derived extracts from P. furfuracea (L.) Zopf was reported in the leather industry (21). Lichens are symbiotic organisms between a fungus and one or more algae or cyanobacteria. They synthesise unique secondary metabolites that cannot be synthesised by higher plants (22, 23). Secondary metabolites

from numerous lichen extracts have been reported to have biological activities such as antibacterial activity against Gram-positive and Gram-negative bacteria (24–27). It has been reported that approximately 2000 of the 20,000 lichen species in the world are in Turkish lichen mycota. There are many studies evaluating the bioactivities of lichen species in Turkey against different bacterial species (25–27). In the previous study, the acetone extracts of H. physodes, E. divaricata, P. furfuracea and Usnea sp. at different concentrations were tested on some Bacillus species which were isolated from soak liquor samples. These extracts were detected to have potential antibacterial effects (28). From this point, lichen species may have potential

antibacterial efficacies against various antibacterial-resistant bacterial strains in the soaking process which cannot be exterminated by antimicrobial agents. Therefore, the antibacterial effects of acetone extracts of lichen species H. tubulosa, H. physodes, E. divaricata, P. furfuracea, P. sulcata and Usnea sp. against Isolate 1 (E. durans), which has protease and lipase acitivities, was evaluated in the present study.


2.1 Sample Collection

Three soak liquor samples were collected from Istanbul Leather Organized Industrial Zone, Tuzla, Istanbul, Turkey. These samples were immediately placed into sterile sample bags and carried on ice during transportation. Direct and serial dilutions were spread onto nutrient agar plates. The morphologically different colony was picked up to obtain the pure culture of the isolate and was numbered as Isolate 1.

2.2 Biochemical and Molecular Analyses

Gram staining, catalase, oxidase, lipase and protease activities were examined. Protease activity of Isolate 1 was examined on gelatin agar medium containing 2% gelatin (w/v). The agar plates were flooded with Frazier solution following 24 h incubation. Clear zones around the colonies were evaluated as positive for protease activity. Lipase activity was tested on Tween® 80 agar medium containing 1% (w/v) Tween® 80. After incubation, opaque zones around the colonies were accepted as evidence of lipase activity (29, 30).



Genomic DNA of Isolate 1 which was determined to have protease and lipase activities were extracted by phenol/chloroform extraction and ethanol precipitation. DNA isolation was confirmed by agarose gel electrophoresis. DNA samples were stored at −20°C until use. The 16S rRNA gene was amplified by polymerase chain reaction (PCR) with the universal bacterial primers 27F (5-AGAGTTTGATCMTGGCTCAG) and 1492R (5-TACCTTGTTACGACTT). Negative control was included in PCR amplifications. PCR amplification was carried out by an initial denaturation at 95°C for 4 min, followed by 30 cycles at 95°C for 1 min, 57°C for 1 min and 73°C for 1 min. The reactions were finished by a final extension at 73°C for 7 min. The PCR products were also monitored by agarose gel electrophoresis. These products were purified by GeneJETTM Gel Extraction Kit (Thermo ScientificTM, Thermo Fisher Scientific, USA). These purified samples were analysed by Medsantek Ltd Co, Istanbul, Turkey. The 16S rRNA sequence contigs were generated by the software ChromasPro version 2.1.8 (Technelysium Pty Ltd, Australia). Then, consensus sequences were exported in FASTA format for each sample for data analysis. These sequences were compared with sequences in the National Center for Biotechnology Information (NCBI) using the Basic Local Alignment Search Tool (BLAST®) search program.

2.3 Lichen Samples

The lichen samples belonging to H. tubulosa, H. physodes, E. divaricata, P. furfuracea, P. sulcata and Usnea sp. were collected from fir trees of Kastamonu province in the north-west of Turkey. They were identified through classical taxonomical methods by microscopic examination. H. tubulosa, H. physodes, E. divaricata,

P. furfuracea, P. sulcata and Usnea sp: Turkey, Kastamonu province, Kapaklı Village, 41.24492, 34.18330, G. Çobanoğlu.

2.4 Extraction of Lichen Samples

The experiment steps included washing, drying in air, weighing, pulverising by liquid nitrogen, adding acetone (ACS, ISO, Reag. Ph. Eur.), keeping in a dark place for 24 h followed by filtration through filter paper. Then, the evaporation of acetone in a rotary evaporator was performed and crude lichen acetone extracts were obtained (27).

2.5 Determination of Antibacterial Efficacies of Lichen Samples

The test isolate was grown on Tryptic soy agar media at 37°C for 24 h. The tests were performed in 96-well CELLSTAR®, F-bottom microplates with lid (Greiner Bio-One GmbH, Austria). Tryptic soy broth was added to each well and nine-fold serial dilutions of the acetone extracts of H. tubulosa, H. physodes, E. divaricata, P. furfuracea, P. sulcata and Usnea sp. were made. Final concentrations of all lichen extracts were 240 µg ml–1, 120 µg ml– 1, 60 µg ml–1, 30 µg ml–1, 15 µg ml–1, 7.5 µg ml– 1, 3.75 µg ml–1, 1.9 µg ml–1 and 0.9 µg ml–1. Overnight culture of the isolate was added to obtain a total volume of 100 µl with an optical density (OD) 600 nm of 0.01. The experiments included untreated and blank controls. The tests were performed in three replicates. Bacterial growth ratios at an OD 600 nm were measured using CytationTM 3 Multi-Mode microplate reader (BioTek Instruments Inc, USA).


In the present study, Isolate 1, which was obtained from soak liquor samples collected from different tanneries in Istanbul Leather Organized Industrial Zone, Turkey, was identified by biochemical and molecular techniques. To our knowledge, there is no study on the antibacterial efficacies of lichen extracts against E. durans from soak liquor samples. For the first time, H. tubulosa, H. physodes, E. divaricata, P. furfuracea, P. sulcata and Usnea sp. acetone extracts were examined against E. durans isolated from soak liquor samples.Isolate 1 was Gram-positive, oxidase and

catalase-negative, protease and lipase positive. The degradative protease and lipase activities of bacteria have an important role in the production of high-quality leather. There are many studies focused on protease and lipase activities of halophilic, extremely halophilic and non-halophilic bacteria on hides or skins in the literature. McLaughlin and Highberger reported that bacterial strains with proteolytic activity were present in high percentages on salt-cured goat skins (31). The proteolytic and lipolytic activities of halophilic and extremely halophilic bacteria were also reported in previous studies. Birbir reported that 91% of 35 salt-cured skins had halophilic bacteria and 67% of 85 extremely halophilic bacterial strains had proteolytic activities (32). Bailey and Birbir



detected that 98% of 131 brine-cured skin samples had extremely halophilic microorganisms and 94% of 332 isolates from these samples showed proteolytic activity (12). Bitlisli et al. demonstrated that 53–74% of halophilic bacteria from salt-cured sheepskins had proteolytic activity and 47–62% of them had lipolytic activity (33). There are also several studies revealing the proteolytic and lipolytic activities of non-halophilic bacteria from soak liquor samples. Veyselova et al. showed proteolytic activity of some bacteria belonging to the genera Enterobacter, Pseudomonas, Enterococcus, Lactococcus, Aerococcus, Vibrio, Kocuria, Staphylococcus and Micrococcus and lipolytic activity of B. licheniformis, B. pumilus, P. luteola and E. cloacae from soak liquor samples (10). In molecular analyses, the tested isolate was

identified by comparative partial 16S rRNA gene sequence analysis with the sequences deposited in the GenBank® database via the BLAST® program. The Isolate 1 had similarities with E. durans CMGB-120 (99.86%, GenBank® accession number MF348232.1). The existence of Enterococcus species was previously reported from hides or skins in the leather industry (6, 34). It is well known that Enterococcus species are common in surface water, soil, vegetables and animal products and they are naturally commensal members of gut microflora of human and warm-blooded animals. Enterococcus avium, E. casseliflavus, E. durans, E. faecalis, E. faecium and E. gallinarum have been isolated from salted hide samples (34). Furthermore, despite increasing the concentration of antimicrobial agents containing didecyl dimethyl ammonium chloride from 0.4 g l–1 to 0.8 g l–1, several bacteria including E. avium and E. faecium were reported from soak liquor samples (19). These results suggest that some Enterococcus species may come from salted hides and can survive in soak liquor samples even in the presence of antibacterial agents. Fluckey et al. isolated 279 Enterococcus isolates from faecal and hide samples. Among them, 169 isolates were detected to be E. durans by biochemical tests (35). E. durans is mostly found in pre-ruminant calves and young chickens and can survive in moderately harsh conditions such as various temperature ranges, pH degrees and salt concentrations as well as detergents (36–38). Similarly to our results, the proteolytic and lipolytic activities of E. durans were also demonstrated in previous studies. Aslan and Birbir detected that six E. durans isolates had proteolytic and lipolytic activities (34). In this regard, Isolate 1 may have the potential to cause

several unwanted defects on finished products due to its enzymatic activities.Antibacterial agents that are commonly used in

the soaking process seem to be ineffective due to random or insufficient application and lead to antimicrobial-resistant bacteria in soak liquors (12, 19). From this point, we can suggest that E. durans from salted hides or skins could not be exterminated by curing methods and also in the soaking process despite the use of antibacterial agents. There are several studies focused on the determination of effective concentrations of several antimicrobial agents against various species of bacteria. Both the ineffectiveness of antibacterial agents in some cases and possible harmful and toxic effects for the environment and human health of some synthetic antimicrobial agents were emphasised in the literature (19, 21). In this respect, the need for safer, more ecological and effective materials has come into prominence for the leather industry. In the previous study, the potential antibacterial effects of acetone extracts of H. physodes, E. divaricata, P. furfuracea and Usnea sp. at the concentrations of 240 µg ml–1, 120 µg ml–1, 60 µg ml–1 and 30 µg ml–1 were demonstrated against Bacillus toyonensis, B. mojavensis, B. subtilis, B. amyloliquefaciens, B. velezensis, B. cereus and B. licheniformis which were isolated from soak liquor samples (28). In respect to these findings, we suggested that H. tubulosa, H. physodes, E. divaricata, P. furfuracea, P. sulcata and Usnea sp. acetone extracts may have antibacterial potential against E. durans which has protease and lipase activities. According to our results, the acetone extracts of

P. sulcata had no antibacterial effect at all tested concentrations against E. durans (Figure 1). On the other hand, we observed a considerable

antibacterial effect for the acetone extracts of H. tubulosa and H. physodes against E. durans. High inhibitory effects of these tested extracts for the growth of E. durans (above 50% inhibition) were detected at the concentrations of 240 µg ml–1, 120 µg ml–1 and 60 µg ml–1 with inhibition ratios of 82.54%, 79.53% and 79.98% for H. tubulosa, and 86.8%, 78.2%, 77.75% for H. physodes, respectively (Figures 2 and 3). The acetone extracts of P. furfuracea also had

antibacterial effect against E. durans at the concentrations of 240 µg ml–1 and 120 µg ml–1 by the inhibition percentages of 80.63% and 85.2%. The other tested concentrations had also inhibitory effects on the tested bacteria but the inhibition ratios recorded were below 50% (Figure 4).



Potential antibacterial efficacy was also detected for the acetone extracts of E. divaricata against E. durans. At the concentration of 240 µg ml–1, we detected 91% inhibition on the bacterial growth. Antibacterial effects were observed at the concentrations of 120 µg ml–1 and 60 µg ml–1 with inhibition ratios of 81% and 79% (Figure 5). Usnea sp. acetone extract was determined to be the

most successful among the tested lichen extracts. 240 µg ml–1, 120 µg ml–1, 60 µg ml–1, 30 µg ml–1 and 15 µg ml–1 of the extracts belonging to Usnea sp. had an antibacterial effect above 80% inhibition. The inhibition ratios at these concentrations were similar and recorded as 88.7%, 84.2%, 92%, 87.8%

and 89.5% respectively. Furthermore, a 58.1% inhibition ratio was noted for the concentration of 7.5 µg ml–1 (Figure 6). All data showed that the acetone extracts of

H. tubulosa, H. physodes, P. furfuracea, E. divaricata and Usnea sp. had potential antibacterial efficacies at varying concentrations against E. durans. Usnea sp. acetone extracts were found to have a stronger inhibitory effect on the bacterial growth of E. durans, even at a low concentration of 15 µg ml–1 (89.5% inhibition) compared to other extracts. These results emphasise the potential of lichens to be utilised as an antibacterial agent in the leather industry. Further studies are needed

Fig. 1. Antibacterial effect of acetone extracts of P. sulcata against E. durans from soak liquor samples

2.5

1.5

0.5

Concentration, µg ml–1

Opt

ical

den

sity

, O

D

0

1.0

2.0

Untre

ated

120 60 30 15 7.

53.

75 1.9

0.9

240

Fig. 2. Antibacterial effect of acetone extracts of H. tubulosa against E. durans from soak liquor samples

1.5

0.5

Opt

ical

den

sity

, O

D

0

1.0

2.0


Untre

ated

120 60 30 15 7.

53.

75 1.9

0.9

240



to detect potential compounds of these lichen species and then these compounds may be used in formulations in the industry.

4. Conclusions

In the leather industry, bacteria with proteolytic and lipolytic activities are important in terms of finished product quality. In this study, we tried to answer the question of whether acetone extracts of six lichen species (H. tubulosa, H. physodes, P. sulcata,

P. furfuracea, E. divaricata and Usnea sp.) have antibacterial effects against E. durans with protease and lipase properties. Whereas P. sulcata did not have any antibacterial efficacy against E. durans, other tested extracts were successful depending on the lichen species and concentrations applied. The acetone extracts of Usnea sp. had the highest antibacterial efficacy. The potential antibacterial efficacies of several lichen species suggest that compound(s) extracted from lichens as natural resources may be used in the leather industry. We

1.5

0.5Opt

ical

den

sity

, O

D

0

1.0

2.0


Untre

ated

120 60 30 15 7.

53.

75 1.9

0.9

240

Fig. 3. Antibacterial effect of acetone extracts of H. physodes against E. durans from soak liquor samples

Fig. 4. Antibacterial effect of acetone extracts of P. furfuracea against E. durans from soak liquor samples

1.5

0.5Opt

ical

den

sity

, O

D

0

1.0

2.0


Untre

ated

120 60 30 15 7.

53.

75 1.9

0.9

240



believe that more comprehensive studies about their unique chemical compounds will provide new insight to utilise them in this sector.

Acknowledgement

The authors are grateful to Gülşah Çobanoğlu Özyiğitoğlu for taxonomic identification of lichen species tested. The authors would like to thank Arhun Ali Balkan, Ayla Yıldız and Barış Gökalsın

(Marmara University) for sharing their experiences about the experiments.

References

1. S. Dahl, J. Am. Leather Chem. Assoc., 1956, 51, 103

2. D. Solaiman, R. Ashby, M. Birbir and P. Caglayan, J. Am. Leather Chem. Assoc., 2016, 111, (10), 358

Fig. 5. Antibacterial effect of acetone extracts of E. divaricata against E. durans from soak liquor samples

1.5

0.5Opt

ical

den

sity

, O

D

0

1.0

2.0

Untre

ated


120 60 30 15 7.

53.

75 1.9

0.9

240

Fig. 6. Antibacterial effect of acetone extracts of Usnea sp. against E. durans from soak liquor samples

1.5

0.5Opt

ical

den

sity

, O

D

0

1.0

2.0


Untre

ated

120 60 30 15 7.

53.

75 1.9

0.9

240



3. M. Birbir and A. Ilgaz, J. Soc. Leather Technol. Chem., 1996, 80, (5), 147

4. Y. Birbir, N. Dolek, M. Birbir and P. Caglayan, Rom. Biotechnol. Lett., 2015, 20, (1), 10123

5. R. Rangarajan, T. D. Didato and S. Bryant, J. Am. Leather Chem. Assoc., 2003, 98, (12), 477

6. A. Orlita, Int. Biodet. Biodeg., 2004, 53, (3), 157

7. M. Birbir, Y. Birbir, E. Yilmaz and P. Caglayan, Int. J. Biosci. Biochem. Bioinform., 2016, 6, (4), 121

8. J. Wu, L. Zhao, X. Liu, W. Chen and H. Gu, J. Clean. Prod., 2017, 148, 158

9. D. Berber and M. Birbir, J. Am. Leather Chem. Assoc., 2010, 105, (10), 320

10. C. Veyselova, M. Birbir and D. Berber, J. Soc. Leather Technol. Chem., 2013, 97, (4), 166

11. P. Caglayan, M. Birbir, C. Sánchez-Porro, A. Ventosa and Y. Birbir, J. Am. Leather Chem. Assoc., 2018, 113, (2), 41

12. D. G. Bailey and M. Birbir, J. Am. Leather Chem. Assoc., 1993, 88, 285

13. H. Anderson, J. Soc. Leather Trade. Chem., 1949, 33, 250

14. B. M. Haines, J. Am. Leather Chem. Assoc., 1984, 79, (8), 319

15. J. J. Tancous, W. T. Roddy and F. O’Flaherty, “Skin, Hide and Leather Defects”, The Western Hills Publishing Company, Ohio, USA, 1959

16. ‘Antibiotic Resistance’, World Health Organization, Geneva, Switzerland, 31st July, 2020

17. M. Birbir, K. Ulusoy and P. Caglayan, J. Am. Leather Chem. Assoc., 2016, 111, (9), 334

18. Y. Birbir, G. Uğur and M. Birbir, J. Electrostat., 2008, 66, (7–8), 355

19. D. Berber, M. Birbir and H. Hacioglu, J. Am. Leather Chem. Assoc., 2010, 105, (11), 354

20. M. Birbir, E. Yazici and P. Çağlayan, J. Soc. Leather Technol. Chem., 2019, 103, 6

21. M. F. Türkan, A. Aslan, A. N. Yapıcı, B. Meriçli Yapıcı and S. T. Bilgi, Tekstil ve Konfeksiyon, 2013, 23, (2), 176

22. T. H. Nash III, “Lichen Biology”, 2nd Edn., Cambridge University Press, Cambridge, UK, 2008, p. 303

23. K. Molnár and E. Farkas, Z. Naturforsch. C., 2010, 65, (3–4), 157

24. B. Paudel, H. D. Bhattarai, J. S. Lee, S. G. Hong, H. W. Shin and J. H. Yim, Phytother. Res., 2008, 22, (9), 1269

25. G. Çobanoğlu, C. Sesal, B. Gökmen and S. Çakar, South West. J. Hortic. Biol. Environ., 2010, 1, (2), 153

26. G. Çobanoğlu, C. Sesal, B. Açıkgöz and İ. Karaltı, Mod. Phytomorphol., 2016, 10, 19

27. B. Gökalsın, D. Berber, G. Ç. Özyiğitoğlu, E. Yeşilada and N. C. Sesal, Plant Biosyst., 2019

28. D. Berber, J. Am. Leather Chem. Assoc., 2020, 115, (3), 96

29. P. Caglayan, M. Birbir, C. Sánchez-Porro and A. Ventosa, Turk. J. Biochem., 2018, 43, (3), 312

30. P. Caglayan, M. Birbir, C. Sánchez-Porro and A. Ventosa, J. Am. Leather Chem. Assoc., 2017, 112, (6), 207

31. G. D. McLaughlin and J. H. Highberger, J. Am. Leather Chem. Assoc., 1926, 21, 280

32. M. Birbir, J. Turk. Microbiol. Soc., 1997, 27, 68

33. B. O. Bitlisli, H. A. Karavana, B. Basaran, O. Sarı, I. Yasa and M. Birbir, J. Am. Leather Chem. Assoc., 2004, 99, (12), 494

34. E. Aslan and M. Birbir, J. Am. Leather Chem. Assoc., 2011, 106, (12), 372

35. W. M. Fluckey, G. H. Loneragan, R. D. Warner, A. Echeverry and M. M. Brashears, J. Food Protect., 2009, 72, (4), 766

36. B. D. Shepard and M. S. Gilmore, Microb. Infect., 2002, 4, (2), 215

37. D. M. F. Amaral, L. F. Silva, S. N. Casarotti, L. C. S. Nascimento and A. L. B. Penna, J. Dairy Sci., 2017, 100, (2), 933

38. A. P. G. Frazzon, B. A. Gama, V. Hermes, C. G. Bierhals, R. I. Pereira, A. G. Guedes, P. A. d’Azevedo and J. Frazzon, World J. Microbiol. Biotechnol., 2010, 26, (2), 365

The Authors

Didem Berber received her MSc degree from the Pediatric Allergy-Immunology Department, School of Medicine, Marmara University, Turkey, in 2003 and PhD from the Department of Biology, Faculty of Arts and Sciences, Marmara University in 2010. She has been studying as postdoctoral researcher in the same department from 2016 up to date. She contributed to projects (European Cooperation in Science and Technology (COST) and other bilateral collaboration projects) on bacterial quorum sensing and biofilm inhibition. Her research topics are hide microbiology, environmental microbiology, antimicrobial agents, fungi, quorum sensing and biofilm formation.



İpek Türkmenoğlu graduated from the Biology Department, Atatürk Faculty of Education, Marmara University in 2012. She is continuing to the master’s programme and she is studying as a scholarship researcher with the support of Scientific and Technological Research Council of Turkey (TÜBİTAK) on the determination and utilisation of species-specific allosteric inhibition zones in glycolytic enzymes in pharmaceutical design. Her research topics are hide microbiology, environmental microbiology, antimicrobial agents, quorum sensing and biofilm formation.

Nüzhet Cenk Sesal graduated from the Biology Department, Atatürk Faculty of Education, Marmara University. He has been working at the Department of Biology, Faculty of Arts and Sciences, Marmara University since 2001. His research area is molecular microbiology. He has been working as a principal investigator, researcher, and consultant in national and international projects, especially about molecular diversity, environmental microbiology, antimicrobial agents, quorum sensing and biofilm formation.




Meral Birbir*, Pinar CaglayanDivision of Plant Diseases and Microbiology, Department of Biology, Faculty of Arts and Sciences, Marmara University, Göztepe Campus, 34722 Kadıköy, Istanbul, Turkey

Yasar Birbir Department of Electrical and Electronics Engineering, Faculty of Technology, Marmara University, Göztepe Campus, 34722 Kadıköy, Istanbul, Turkey


Proteolytic and lipolytic extremely halophilic archaea found in curing salt may contaminate skins during the brine curing process and damage skin structure. In the present study, three proteolytic and lipolytic extremely halophilic archaea were isolated from deteriorated salted sheepskins and characterised using conventional and molecular methods. Each test strain (Haloarcula salaria AT1, Halobacterium salinarum 22T6, Haloarcula tradensis 7T3), a mixed culture of these strains and the mixed culture treated with 1.5 A direct current (DC) were used for brine curing processes of fresh sheepskins and examined during 47 days of storage to evaluate the degree of destruction wreaked by these microorganisms. Both organoleptic properties and scanning electron microscopy (SEM) images of sheepskins proved that each separate test strain and the mixed culture caused serious damage. However, the mixed culture of strains treated with electric current did not damage sheepskin structure. Therefore, we highly recommend sterilisation of

brine using DC to prevent archaeal damage on cured hides and skins in the leather industry.

1. Introduction

Extremely halophilic archaea have been found in hypersaline salt lakes, salterns, salt mines, salted foods and salted hides. There have been numerous studies on the presence of extremely halophilic archaea in these hypersaline environments (1–12). Due to the high salt requirements of extreme halophiles (15–30% NaCl), these microorganisms have been denominated as extremely halophilic archaea (13, 14). Cells of Haloarchaea staining Gram-negatively are irregular rods, cocci, pleomorphic rods, cups, irregular disks, flattened disks, irregular triangles, rectangles and squares (2, 5, 15). Chemoorganotroph extremely halophilic archaea, which can be motile or non-motile, grow aerobically and use different amino acids. Colonies of these microorganisms are pink, red and orange due to C50-carotenoid pigments called bacterioruberins (15, 16). Observation of red or violet discolorations on

the flesh side of salted hides and skins is the key for detecting extremely halophilic archaea in the leather industry. These discolorations are a sign of bacterial deterioration of hides and skins (17, 18). Previous experiments reported that microorganisms in curing salts and raceway brines contaminated hides and skins and caused red heat (10). The brine cured hides and skins were often stored in hot warehouses, trucks or ships, and these high temperature conditions, combined with moisture, offer an ideal medium for proteolytic extremely halophilic archaea to grow and potentially digest collagen fibres in the hides and skins (10).

The Destructive Effects of Extremely Halophilic Archaeal Strains on Sheepskins, and Proposals for Remedial Curing ProcessesUse of sterile brine or direct electric current to prevent red heat damage on salted sheepskins



Extremely halophilic archaea (102–105 colony forming units (CFU) g–1), proteolytic (102–104

CFU g–1) and lipolytic (102–104 CFU g–1) extremely halophilic archaea were detected in 40 curing salt samples collected from different tanneries in Turkey (19). Almost all salted hides and skins contained extremely halophilic archaea, proteolytic and lipolytic extremely halophilic archaea originating in the curing salt. Extremely halophilic archaea were also detected on 94% of 131 brine-cured cattle hides collected from USA, 91% of 35 salted hides cured in France and Russia and all salted hides cured in Turkey, Greece, the UK, USA, Serbia, Bulgaria, Russia, South Africa and Australia (20–22). Five extremely halophilic archaeal species, Halorubrum saccharovorum, Halorubrum tebenquichense, Halorubrum lacusprofundi, Natrinema pallidum and Natrinema gari were isolated from five salted hides originating in England and Australia (22). Also, 101 extremely halophilic archaeal strains (Halorubrum tebenquichense, Halorubrum saccharovorum, Halorubrum kocurii, Halorubrum terrestre, Halorubrum lipolyticum, Halococcus dombrowskii, Halococcus qingdaonensis, Halococcus morrhuae, Natrinema pellirubrum, Natrinema versiforme, Halostagnicola larsenii and Haloterrigena saccharevitans) were isolated from four salted sheepskin samples (Spain) exhibiting bad odour, a slimy layer, hair slip, red and yellow discolorations (23). Moreover, 28 extremely halophilic archaeal strains (Natrialba aegyptia, Halovivax asiaticus, Halococcus morrhuae, Halococcus thailandensis, Natrinema pallidum, Halococcus dombrowskii, Halomicrobium zhouii, Natronococcus jeotgali, Haloterrigena thermotolerans, Natrinema versiforme and Halobacterium noricense) were isolated from eight salted hide and skin samples from Turkey, Iraq, Turkmenistan and Kazakhstan (24). While there are many reports that detect the

presence of extremely halophilic archaea on salted hides and skins (10, 17, 20–25), the destructive effects of these microorganisms on salted hides have been studied much less (25, 26). In our previous investigation, we found that extremely halophilic archaeal strains, isolated from hides brine cured in the USA, damaged grain the surface of hides at 41°C after 49 days (25). An experiment with extremely halophilic Haloferax gibbonsii (ATCC 33959TM) and Haloarcula hispanica (ATCC 33960TM) obtained from American Type Culture Collection (ATCC), USA, demonstrated that Haloferax gibbonsii caused hair slip, loss of hide substance and deterioration of brine cured hide after 45 days at 40°C (26).

The adverse effects of extremely halophilic archaeal hide isolates and ATCC strains of extremely halophilic archaea on brine cured hides have been reported in these studies, respectively (25, 26). However, the destructive effects of salted sheepskin strains of extremely halophilic Haloarcula salaria, Halobacterium salinarum and Haloarcula tradensis on brine cured sheepskins have not been examined yet. Therefore, the aim of this study was to examine adverse effects of proteolytic and lipolytic archaeal sheepskin strains (Haloarcula salaria AT1, Halobacterium salinarum 22T6, Haloarcula tradensis 7T3) and the mixed culture of these strains on sheepskins during a 47-day storage period at 33°C. We also investigated effective curing methods to prevent the destructive effects of these microorganisms on sheepskins. Additionally, we evaluated pH values, ash contents, moisture contents, salt saturations, total counts of extremely halophilic archaea and organoleptic properties of the brine cured sheepskin samples during different storage periods to determine the brine curing procedure’s efficiency and the test microorganisms’ adverse effects of on sheepskins.


2.1 Isolation of Extremely Halophilic Archaeal Strains from Deteriorated Salted Sheepskins

Two deteriorated salted sheepskins containing red discolorations were collected from two tanneries in the Istanbul Leather Organized Industrial Zone (40°52′39.7″N,29°20′25.3″E) in Tuzla, Turkey. The samples were immediately placed into sterile sample bags and transported on ice to the laboratory. Then, 20 g of the salt-pack cured sheepskin samples were weighed and separately soaked in flasks containing 180 ml 30% NaCl (Merck KGaA, Germany) solution. The flasks were placed into a shaking incubator at 90 rpm, 24°C for 3 h. The suspension of the skin was diluted with sterile physiological saline water (30% NaCl). An aliquot of 100 µl each of direct and serial skin suspension dilutions was spread onto the surface of modified Brown agar media containing (per litre): 1 g CaCl2·H2O, 2 g KCl, 20 g MgSO4·7H2O, 3 g trisodium citrate, 250 g NaCl, 5 g yeast extract, 20 g agar, pH 7 (5, 27). The plates were incubated at 39°C for 10 days. Following incubation, red pigmented colonies on the agar media were selected and restreaked several times to obtain pure cultures. A total of 22 isolates were obtained from the sheepskins and then, these strains were



examined the proteolytic and lipolytic activities. Proteolytic activity of each strain was detected on gelatin agar medium containing 2% gelatin. After incubation, clear zones around the colonies on the gelatin agar medium indicated protease production (5, 10). Lipolytic activity of each strain was screened on Tween® 80 agar medium containing 1% Tween® 80. After growth was obtained, opaque zones around the colonies were interpreted as positive lipase activity (5). In the present study three red pigmented proteolytic and lipolytic strains (AT1, 22T6 and 7T3) were obtained from two salted sheepskins and these strains were used in the present study.

2.2 Phenotypic Characteristics of Test Strains

Exponentially growing pure cultures of three strains designated as AT1, 22T6 and 7T3 were used in all experiments. First, the strains’ salt requirement and salt tolerance were examined on Brown agar plates containing different salt concentrations (0%, 0.5%, 3%, 5%, 7.5%, 10%, 12.5%, 15%, 20%, 25% and 30%) (27). After detection of optimum salt concentration for each strain, pH and temperature ranges for growth of each strain (AT1, 22T6, 7T3) were respectively examined at Brown agar plates with different pH values (pH 4, pH 5, pH 6, pH 7, pH 7.5, pH 8, pH 9, pH 10, pH 11 and pH 12) and different temperatures (4°C, 10°C, 15°C, 24°C, 28°C, 35°C, 37°C, 39°C, 45°C, 50°C, 55°C, 60°C) according to the methods described in Proposed Minimal Standards for Description of New Taxa in the Order Halobacteriales (28). Based on the pH, and temperature range of each test strain, the optimal pH and growth temperature of each test strain were determined. Pigmentation, size, margin, elevation and

opacity of colonies of the strains grown on Brown agar media were examined under optimal growth conditions (28). Cell morphology, cell length, cell width and motility of each strain were examined using both light microscopy and electron microscopy. Microscopic observation of each strain was made by using freshly prepared wet mount (28). For SEM observations, 20 ml of each test strain were separately passed through 0.2 μm pore size cellulose nitrate membrane filter placed in the stainless steel funnel via vacuum pump (Sartorius AG, Germany). The archaeal cells of each strain trapped on the membrane filters were observed under SEM (QuantaTM 450 FEG (FEI, USA)). Gram staining was performed with acetic acid-fixed slides (28–30). Catalase

and oxidase activities, indole production, methyl red test, H2S and NH3 productions of each strain were investigated according to the procedures described previously (4, 28, 31). Furthermore, each strain’s caseinase activity was determined on the agar medium containing 2% skim milk. After incubation, clear zones around the colonies were evidence of positive caseinase activity (4). Urease production was investigated on Christensen urea agar medium. The tubes were examined for pink or red colour change in the medium after seven days of incubation (28, 31). β-galactosidase activity was screened in test tubes containing ortho-nitrophenyl-β-galactoside (ONPG) discs and 1 ml of sterile saline water (30% NaCl). The yellow colour formation in the test tube was accepted as positive β-galactosidase activity (5, 31). Amino acid utilisation of each strain was examined in the test medium containing 1% amino acid, 0.5% beef extract, 0.5% peptone, 0.05% dextrose, 0.0005% cresol red, 0.001% bromocresol purple, 0.0005% pyridoxal and saline water (30% NaCl). Purple colour formation in the test tube containing archaeal culture was accepted as a positive test after 10 days incubation period at 39°C (31).

2.3 Amplification and Sequencing of 16S rRNA Genes of Test Strains

Chromosomal DNA was isolated by QIAamp DNA Mini Kit (Qiagen, Germany) and purified by QIAquick PCR Purification Kit (Qiagen, Germany) according to the manufacturer’s directions. The 16S rRNA genes of the strains were amplified by polymerase chain reaction (PCR) using forward primer 21F and reverse primer 1492R (32). The 16S rRNA gene sequences of three strains (AT1, 22T6 and 7T3) were determined by IONTEK Laboratory (Turkey). The sequences of these strains were analysed using ChromasPro v.2.1.8 software (Technelysium, Australia) and then compared with the sequence on the EZBioCloud Database (ChunLab, South Korea) (33).

2.4 Preparation of Test Strains and Sheepskin Samples for Brine Curing Treatments

2.4.1 Preparation of Strains and Cultures Used in Brine Curing Processes

Pure cultures of each test strain (AT1, 22T6, 7T3) were separately grown in liquid Brown test medium containing 30% NaCl for 10 days at 39°C.



Each archaeal cell suspension’s turbidity was adjusted to 0.5 McFarland standard (108 CFU ml–1) using densitometer (DEN-1, BIOSAN, Latvia). Each cell suspension was diluted in sterile saline solution (30% NaCl) to adjust the cell suspension to 107 CFU ml–1. In addition, mixed cultures of these strains (107 CFU ml–1) were prepared. Then, 20 ml of each test strain, 20 ml of the mixed culture were used in the brine curing solutions of T1–T4 (Table I).To prepare brine curing solution containing

electrically inactivated mixed culture (T5), 20 ml of the mixed culture containing AT1, 22T6, 7T3 strains (107 CFU ml–1) were placed into the electrolysis cell consisting of a glass beaker having two internally attached platinum wire electrodes and 180 ml of sterile brine solution (30% NaCl) (34, 35). To detect the archaeal numbers of the mixed

culture in the electrolysis cell before the electric current application, 100 µl of the test medium was removed from the electrolysis cell and diluted to 10–2–10–4 using sterile 30% NaCl solution. The diluted archaeal suspensions were spread over the Brown agar media. Then, 1.5 A DC was applied to the electrolysis cell for 22 min (Figure 1). A 100 µl quantity of test medium was removed from the cell at intervals of 1 min, 4 min, 7 min, 10 min, 13 min, 16 min, 19 min and 22 min of electric current application. Direct and diluted suspensions of electrically inactivated the mixed culture were spread over Brown agar media. All inoculated Brown media were incubated for 10 days at 39°C, and colonies on the agar plates were counted. This test medium was used for curing process of the sheepskin (T5) after 22 min of electric current application on the mixed culture (Table I).

Table I Protocol for Brine Curing Treatments of Sheepskins Brine Curing CompositionsControl Treatments 59.5 g sheepskin sample + 200 ml sterile brine solution

T1 59.5 g sheepskin sample + 180 ml sterile brine solution + 20 ml strain AT1 (107 CFU ml–1)

T2 59.5 g sheepskin sample + 180 ml sterile brine solution + 20 ml strain 22T6 (107 CFU ml–1)

T3 59.5 g sheepskin sample + 180 ml sterile brine solution + 20 ml strain 7T3 (107 CFU ml–1)

T4 59.5 g sheepskin sample + 180 ml sterile brine solution + 20 ml mixed culture (107 CFU ml–1)

T5 59.5 g sheepskin sample + 180 ml sterile brine solution containing 20 ml electrically inactivated mixed culture

Inverter switch

220 V

50 Hz AC

Input

Voltmeter

AC-DC

Fuse

DC output 0–220 VAC output

R Mp

DC AC

Main switch

100 0Variac

10 2

Electrolysis cell

Insulated layer

Test medium

DC ammeter AC ammeter

A A

Platinum electrodes

Anode Cathode

+ –

Fig. 1. Electrolysis cell system used 1.5 A DC treatment in this study (R: phase, Mp: ground)



2.4.2 Preparation of Sheepskin Samples for Brine Curing Treatments

One freshly slaughtered, de-fleshed whole sheepskin sample was obtained from a slaughterhouse in Istanbul, Turkey. Then, the sheepskin sample was immediately placed into sterile sample bag and transported on ice to the laboratory. The sheepskin was cut into six pieces perpendicular to backbone, from backbone to belly. Next, we carried out the following six treatments for brine curing of the sheepskin samples. In each treatment, sterile 30% NaCl (Merck KGaA) solution was used. In all treatments, a 400% float of the brine solution (238 g of the brines without test strain, with each test strain or mixed culture/59.5 g of sheepskin) was used (25). Sterile 30% NaCl solution containing the sheepskin sample was used as Control. The sheepskin samples (T1–T4) were separately placed in a glass beaker containing the brine solution, each test strain or mixed culture (T1–T4, Table I). In the Treatment 5, the sheepskin sample was placed in a glass beaker containing the brine solution with electrically inactivated mixed culture (T5, Table I).The curing processes of all sheepskins were

carried out the protocol described in Table I. The sheepskin samples were separately cured in the brine solutions at 90 rpm for 18 h at 24°C. After the curing processes, all sheepskins were taken from the brine solutions and stored for 47 days at 33°C.

2.5 Determination of Extremely Halophilic Archaeal Counts in Curing Solutions and Cured Sheepskin Samples To determine total counts of extremely halophilic archaea in the curing solutions before the curing processes, 100 µl of the test medium was removed from the each curing solution and diluted to 10–2–10–4 using sterile 30% NaCl solution. The diluted archaeal suspensions were spread over the Brown agar media. In addition, subsequent to each brine curing process detailed above (T1–T5), the suspensions of cured sheepskin samples were prepared at intervals of 5 days, 16 days, 28 days and 47 days of storage. 2 g of each skin sample were put into a flask containing 18 ml sterile 30% NaCl solution and incubated for 1 h at 24°C and 100 rpm. Direct and serial dilutions of the suspensions were spread onto the surface of Brown agar media. All inoculated Brown media were incubated at 39°C for 10 days and the colonies grown on the test media were counted.

2.6 Determination of pH, Moisture Content, Ash Content and Salt Saturation of Cured Sheepskin Samples

After curing processes, 5 g of the sheepskins were cut and placed into flasks containing 100 ml of sterile distilled water. The flasks were placed in a shaking incubator for 1 h at 100 rpm and then pH was measured with a pH meter. Hairs and dirt on the samples were removed to properly determine the samples’ moisture content. 3 g of the samples were placed into an oven at 102°C for 6 h. The dried samples were weighed, returned to the oven for 1 h, and then were weighed again. The drying procedure was repeated until the first dry weight was equal to the second dry weight. The samples were put into a desiccator for 30 min to cool. Next, we calculated the skins’ moisture contents (20, 21). The dry sheepskins samples were placed in ceramic crucibles and ashed in a muffle furnace at 600°C for 8 h. After cooling, the samples were weighed to determine ash content. Moisture content, ash content and salt saturations of skin samples were calculated according to the aforementioned methods (30, 36). The pH value, ash content, moisture content and salt saturation of all cured sheepskin samples were examined at different storage periods.

2.7 Organoleptic Examination of Brine Cured Sheepskin Samples During Storage Periods

All cured sheepskin samples were examined organoleptically (hair slip, deterioration of skins, bad odour, sticky appearance, red heat, hole formation) during different storage periods.

2.8 Preparation of Sheepskin Samples for Scanning Electron Microscopy Observation

After a 47-day storage period, the sheepskin samples were prepared for SEM observation. The samples were fixed in 4% glutaraldehyde solution prepared in 0.1 M phosphate buffer (pH 7.2) for 30 min. The samples were washed three times with 0.1 M phosphate buffer for 10 min and were treated with 1% OsO4 prepared in 0.1 M phosphate buffer at room temperature for 1 h. The samples were washed two times in sterile distilled water for 10 min. Then, the water in the sheepskins was gradually removed by 35%, 50%, 75%, 95% and absolute ethanol. The mixtures



of ethanol-hexamethyldisilazane (ethanol-HMDS) [1:1 (v/v)] (1 × 30 min), ethanol-HMDS [1:2 (v/v)] (1 × 30 min) and HMDS (2 × 30 min) were used for air drying process. After drying, HMDS was poured from petri dishes and the samples were placed in a desiccator for 12 h. Later, the sheepskin samples were examined under SEM (QuantaTM 450 FEG) using sample stub with double-sided sticky tape (37).


3.1 Isolation and Selection of Test Strains from Sheepskins

A total of 22 red coloured strains were isolated from two deteriorated salted sheepskin samples obtained from two tanneries in the Istanbul Leather Organized Industrial Zone in Tuzla, Turkey. While nine, seven and three strains respectively produced protease, lipase, both protease and lipase, three strains did not produce either lipase or protease enzymes. The red coloured three extremely halophilic strains producing both protease and

lipase enzymes were selected and used as test strains (AT1, 22T6 and 7T3) in the present study.

3.2 Phenotypic Characteristics of Test Strains

Strains AT1, 22T6 and 7T3 grew at 15–30% NaCl, 15–30% NaCl, 20–30% NaCl concentrations, respectively. Optimum salt concentrations of strains AT1, 22T6 and 7T3 were determined as 25% NaCl. Hence, these strains were accepted as extremely halophilic archaea. The pH and temperature ranges for growth of strains AT1, 22T6 and 7T3 were respectively found as pH 6–11 and 20–50°C, pH 6–11 and 15–55°C, pH 5–11 and 15–55°C. All extremely halophilic archaeal strains optimally grew at 39°C and pH 7. The colony pigmentation, size, margin, elevation and opacity of strains AT1, 22T6, 7T3 were respectively observed as: red, 0.6– 2 mm, entire, convex, translucent; red, 1–2 mm, entire, convex, translucent; red, 0.8–1.9 mm, entire, convex, translucent. The cells of strains AT1 (Figure 2(a)) and 7T3 (Figure 2(c)) were non-motile, extremely pleomorphic (triangle, square,

5 μm 10 μm

5 μm

(a) (b)

(c)

Fig. 2. SEM micrographs of pleomorphic test strains of (a) Haloarcula salaria (AT1) cells; (b) Halobacterium salinarum (22T6) cells; (c) Haloarcula tradensis (7T3) cells trapped on the membrane filter



irregular disk, short rod). The cells of strains AT1 and 7T3 were approximately 0.4–1.3 µm × 0.4–2.0 µm and 0.3–0.7 µm × 0.3–4 µm, respectively. The cells of strain 22T6 (Figure 2(b)) were motile, pleomorphic rods, approximately 0.5–1.2 µm × 3.2–6.6 µm. All strains were Gram-negative (Table II). While all strains showed positive catalase, oxidase, protease, lipase activities, indole production, the methyl red, caseinase, urease and β-galactosidase reactions of all strains were negative. The strains did not produce H2S and NH3 (Table II). Our experimental results showed that Haloarcula

salaria (AT1), Halobacterium salinarum (22T6), Haloarcula tradensis (7T3) strains have protease activities which can breakdown proteins in corium

of sheepskin causing loss of skin substance. When the protein structure of salted skins is broken down by proteolytic extremely halophilic archaea, these microorganisms can utilise some amino acids as a source of carbon, nitrogen and energy. Haloarcula salaria AT1 and Halobacterium salinarum 22T6 utilised most of the amino acids examined. While Haloarcula salaria AT1, Halobacterium salinarum 22T6 utilised 17 amino acids, Haloarcula tradensis 7T3 used only three amino acids (Table III). In another study, the liquid test media containing calfskin samples, 30% NaCl and proteolytic red and pink strains of the extremely halophilic archaea were separately prepared to show disintegration of the skin proteins. After an incubation period,

Table II Phenotypic Characteristics of Haloarcula salaria, Halobacterium salinarum, Haloarcula tradensis

Characteristics Haloarcula salaria Halobacterium salinarum Haloarcula tradensisStrain code AT1 22T6 7T3

Motility Non-motile Motile Non-motile

Cell morphology Extremely pleomorphic Pleomorphic rods Extremely pleomorphic

Cell width, µm 0.4–1.3 0.5–1.2 0.3–0.7

Cell length, µm 0.4–2 3.2–6.6 0.3–4

Gram staining Negative Negative Negative

Pigmentation Red Red Red

Colony size, mm 0.6–2 1–2 0.8–1.9

Colony margin Entire Entire Entire

Colony elevation Convex Convex Convex

Colony opacity Translucent Translucent Translucent

NaCl concentration, % 15–30 15–30 20–30

pH range 6–11 6–11 5–11

Temperature range, °C 20–50 15–55 15–55

Optimum NaCl 25 25 25

Optimum Temperature, °C 39 39 39

Optimum pH range 7 7 7

Catalase activity + + +

Oxidase activity + + +

Methyl red reaction – – –

Caseinase activity – – –

Urease activity – – –

β-galactosidase activity – – –

Indole production – – –

H2S production – – –

NH3 production – – –

Protease activity + + +a

Lipase activity + + +a Haloarcula tradensis (7T3) showed weak protease activity



decomposition of the skin samples in the media was detected by visual observation. While contents of asparagine, threonine, serine, glutamine, proline, glycine, alanine, valine, isoleucine, leucine, phenylalanine, lysine and arginine in the test tubes were detected at high levels, contents of methionine, tyrosine and histidine were low (10). Phenotypic features of extremely halophilic AT1,

7T3 and 22T6 strains detected in this study were fairly similar to phenotypic features of Haloarcula salaria, Haloarcula tradensis and Halobacterium salinarum isolated by other researchers (15, 38, 39).

3.3 16S rRNA Gene Sequences of Test Strains

The phylogenetic analysis revealed that three strains shared highly similar identities with their closest phylogenetic relatives. Strains AT1, 22T6, 7T3 were respectively assigned to Haloarcula salaria (98.36%-1344 base pairs), Halobacterium salinarum (99.78%-1345 base pairs), Haloarcula tradensis (98.37%-1355 base pairs). The gene sequence data of the strains AT1, 22T6, 7T3 were respectively deposited in GenBank® (National

Center for Biotechnology Information, USA) under accession numbers as MN585896, MN585803, MN585804. In our previous study, extremely halophilic

archaeal strains were isolated from Tuz Lake and its salterns (5). In Turkish leather industry, curing salt is mostly obtained from Tuz Lake and its salterns. Hence, we suspect that contaminations of our sheepskin samples with Haloarcula salaria AT1, Halobacterium salinarum 22T6 and Haloarcula tradensis 7T3 were due to the curing salt obtained from Tuz Lake and its salterns.

3.4 Extremely Halophilic Archaeal Counts in Curing Solutions Before Curing

In the study carried out with 25 salted sheepskin samples (Australia, Bulgaria, Dubai, Greece, Israel, Kuwait, South Africa, Turkey, USA) and 25 salted goat skin samples (Australia, Turkey, Bulgaria, Israel, South Africa, Russia, China, France), proteolytic extremely halophilic archaea and lipolytic extremely halophilic archaea were detected as 102–105 CFU g–1; 102–106 CFU g–1

Table III Utilisation of Amino Acids by Strains

Amino acids Haloarcula salaria (AT1) Halobacterium salinarum (22T6)

Haloarcula tradensis (7T3)

L-arginine + + +

L-cysteine – – –

L-glycine + + –

L-alanine + + –

L-tyrosine + + –

L-proline + + –

L-hydroxyproline + + –

L-glutamic acid – – –

L-methionine + + –

L-serine + + –

L-isoleucine + + –

myo-inositol + + –

L-lysine + + +

L-phenylalanine + + –

L-leucine + – –

L-valine + + –

L-threonine + + –

L-ornithine + + –

L-histidine + + +

L-aspartic acid – – –

L-cystine – + –



and 102–106 CFU g–1; 102–106 CFU g–1 on salted sheepskins and goat skins, respectively (40). The highest number of proteolytic and lipolytic extremely halophilic archaea on the salted skins was found as 106 CFU g–1 (40). Therefore, the archaeal cell numbers of test strains in the brine curing solutions were adjusted to 106 CFU ml–1. Before the curing processes of sheepskins, while the archaeal cell numbers in the brine solutions of Treatments 1, 3 and 4 were detected as 2.1 × 106 CFU ml–1, the archaeal cell numbers in the brine solution of Treatment 2 was detected as 2.2 × 106 CFU ml–1. The archaeal cell numbers in the mixed culture was

detected as 2.1 × 106 CFU ml–1 in the electrolysis cell before 1.5 A DC application. While the archaeal cell numbers in the mixed culture were reduced

from 2.1 × 106 CFU ml–1 to 3.2 × 105 CFU ml–1 after 1 min of DC treatment, the cell numbers of 1.24 × 102 CFU ml–1 was detected after 4 min of DC treatment. All archaeal cells in the mixed culture were completely killed in 7 min of DC treatment. In the present study, log10 value of the mixed culture of extremely halophilic archaea in the brine solution before the DC treatment was 6.32. After 1 min, 4 min and 7 min of 1.5 A DC treatment; 0.82, 4.23 and 6.32 log10 reduction values (CFU ml–1) of the mixed culture in the brine were detected, respectively. Temperature and pH of the electrolysis cell were

respectively measured as 31°C and pH 6 prior to the electric current treatment. After treating the brine solution with the electric current, the temperature of the brine was adjusted to 24°C for using in

Table IV pH, Ash Content, Moisture Content and Salt Saturation Values, Total Extremely Halophilic Archaeal Counts of the Sheepskin Samples After Different Storage Periods

Experiment pH Ash content, %

Moisture content, %

Salt saturation, %

Total count of extremely halophilic archaea

After 5 days Control 7.55 20 55 >100 0

T1 6.72 24 50 >100 2.0 × 107

T2 6.59 23 50 >100 3.4 × 107

T3 6.65 21 57 >100 2.2 × 107

T4 6.53 26 52 >100 3.8 × 107

T5 7.80 21 57 >100 0


T1 6.52 30 47 >100 3.0 × 107

T2 6.70 27 51 >100 6.0 × 107

T3 6.65 22 50 >100 3.4 × 107

T4 6.85 32 46 >100 8.4 × 107

T5 7.32 23 55 >100 0


T1 7.70 29 40 >100 1.2 × 107

T2 7.52 29 43 >100 2.0 × 107

T3 7.36 33 44 >100 2.0 × 107

T4 7.51 32 39 >100 3.4 × 107

T5 7.81 29 46 >100 0


T1 7.58 34 26 >100 1.0 × 107

T2 7.47 34 35 >100 1.8 × 107

T3 7.31 44 24 >100 1.7 × 107

T4 7.60 37 38 >100 2.0 × 107

T5 7.64 33 33 >100 0



curing process of sheepskin in the Treatment 5. While the temperature and pH of the test medium respectively increased from 31°C to 41°C and from pH 6 to pH 8.5 during the electric current treatment, voltage values slightly decreased from 4.7 V to 4.3 V.We also demonstrated the inactivation

of extremely halophilic strains via DC and alternating electric current (AC) in our previous studies (35, 41, 42). A 0.5 A DC was applied for 30 min to several strains of extremely halophilic archaea (107 CFU ml–1) isolated from Tuz Lake, Kaldırım and Kayacık salterns (35). While the mixed culture of extremely halophilic archaea was exterminated in 10 min, protease producing extremely halophilic archaea were killed in 5 min. However, lipase or lipase and protease producing extremely halophilic archaea were exterminated in 20 min (35). In another experiment, lipase and protease producing extremely halophilic strains (105–106 CFU ml–1), separately grown in liquid Brown media, were inactivated by a 10 min treatment with 0.5 A DC (41). It was also detected that 1 min of 2 A AC treatment was enough to kill extremely halophilic archaea found in brine solution (102–104 CFU ml–1). When 2 A AC was applied to lipolytic extremely halophilic archaea, proteolytic extremely halophilic archaea, both proteolytic and lipolytic extremely halophilic archaea, and a mixed culture of these strains (106 CFU ml–1), all test microorganisms found in 25% NaCl solution were exterminated in 5 min (42).

3.5 Extremely Halophilic Archaeal Counts on Cured Sheepskin Samples During Storage

After the curing processes of sheepskins, we did not detect any extremely halophilic archaea on the sheepskin sample cured with the sterile brine solution (Control) and the sheepskin sample cured with the brine solution containing electrically inactivated mixed culture (T5) during the all storage periods. While extremely halophilic archaeal numbers on

both skin samples cured with each strain and the skin sample cured with mixed cultures of the strains slowly increased from 106 CFU ml–1 to 107 CFU during five days and 16 days storage periods, the numbers of these strains slowly decreased 28 days and 47 days storage periods due to attachment of these cells to sheepskins (Table IV).

3.6 pH, Moisture Content, Ash Content and Salt Saturation Values of Cured Sheepskin Samples

After the curing processes of skins, pH values of the sheepskin samples were measured as pH 7.35 for Control; pH 6.89 for T1; pH 7.09 for T2; pH 7.05 for T3; pH 7.16 for T4; pH 8.05 for T5. While salt saturation values of all cured sheepskins were higher than 100% during all storage periods, pH, ash content and moisture content values changed during different storage periods. pH, ash content and moisture content values of the cured skins were detected between pH 6.52–7.81, 20–44%, 24–57%, respectively (Table IV).Moisture, minimum and maximum ash contents,

salt saturation values of adequately cured salted hides were suggested as 40–48%, 14–48%, higher than 85%, respectively (36). Due to detection of high moisture content in all samples (between 50–57%) after five days storage, sterile salt was added to all sheepskins to reduce their moisture contents according to curing procedure described in the previous study (43). While all skin samples reached the suggested moisture content values (39–46%) after 28 days, the suggested saturation values were detected after five days. The samples’ lowest moisture content values were detected after 47 days. Ash contents of all skins (20–44%) were close to suggested values (36). While the skins’ pH values changed during storage periods, all values were found sufficient to support the growth of extremely halophilic strains (Table IV). The pH, moisture content, ash content and salt saturation values detected in this study were also consistent with pH range (pH 6.53–8.01), moisture content (32–68%), ash content (12–30%) and salt saturation (58–100%) values of 25 salted sheepskin samples determined in the previous experiment (40).

3.7 Organoleptic Characteristics of Brine Cured Sheepskin Samples During Storage

While hair slip and bad odour were detected on the sheepskin samples cured with each strain and the mixed culture after five days at 33°C, sticky appearance and red heat were observed on the cured sheepskin samples after 16 days (T1–T4, Figure 3). In addition to the aforementioned organoleptic properties, hole formations were observed on these sheepskin samples after 28 days. However,



we did not detect any organoleptic properties on sheepskin samples cured with sterile brine and the brine treated with 1.5 A DC (Control and T5, Figure 3). In another study, the commercially cured

hides stored one year in the USA were also examined for proteolytic activity of extremely halophilic archaea. Experimental results of that study showed that the flesh side of hides containing extremely halophilic archaea had pink discolorations called red heat. When these hides were incubated at 35°C–40°C, bad odour, hair slip and severe grain damage were detected. Damaged grain surfaces were observed on leather made from these hides (10). In another experiment researchers emphasised that temperatures of the brines and hides should be maintained below 20°C to prevent growth of extremely halophilic archaea (44).

3.8 Scanning Electron Microscopy Observation of Mixed Culture and Treated Sheepskin Samples

Figure 4 shows extremely halophilic archaeal cells of the mixed culture on 0.2 µm pore-size cellulose nitrate membrane filter in pleomorphic shapes such as triangle, square, irregular disk and rod. As seen in the SEM micrograph, 1.5 A DC treatment

significantly debilitated structural integrity of the cells in the mixed culture trapped on the filter (Figure 5). The SEM images clearly showed that electric current application damaged cell structures of each strain in the mixed culture (Figure 5). As seen in Figure 6, the sterile brine curing process protected the sheepskin against microbial damage during 47 days of storage. Attachment of Haloarcula salaria AT1,

Halobacterium salinarum 22T6 and Haloarcula tradensis 7T3 to corium fibres and the consequent destructive effects on sheepskins are seen in Figures 7–10. Haloarcula salaria AT1, Halobacterium salinarum 22T6 and the mixed culture of the strains caused fibres in the corium to split and weaken (Figures 7, 8 and 10). In contrast with the skin samples treated with Haloarcula salaria AT1, Halobacterium salinarum 22T6, skin sample treated with Haloarcula tradensis 7T3 had compact appearance, although the shredding of the fibres was still present in corium (Figure 9). That damage was due to the proteolytic activities of these microorganisms.Figure 11 clearly shows that the curing process of

sheepskin with the brine containing mixed culture treated with DC prevented extremely halophilic archaea from contaminating the sheepskin and furthermore protected the skin very well against microbial damage during a long storage period.

(a) (b) (c)

(d) (e) (f)

Fig. 3. Organoleptic characteristics of brine cured sheepskin samples after 16 days storage period: (a) Control, sheepskin sample cured with sterile brine (30% NaCl); (b) T1, sheepskin sample cured with brine containing Haloarcula salaria AT1; (c) T2, sheepskin sample cured with brine containing Halobacterium salinarum 22T6; (d) T3, sheepskin sample cured with brine containing Haloarcula tradensis 7T3; (e) T4, sheepskin sample cured with brine containing mixed culture; (f) T5, sheepskin sample cured with brine containing electrically inactivated mixed culture



20 μm

Fig. 4. SEM micrograph of mixed culture of undamaged pleomorphic cells of Haloarcula salaria (AT1), Halobacterium salinarum (22T6) and Haloarcula tradensis (7T3) trapped on the membrane filter

10 μm

Fig. 5. SEM micrograph of mixed culture of damaged Haloarcula salaria (AT1), Halobacterium salinarum (22T6) and Haloarcula tradensis (7T3) cells treated with 1.5 A DC trapped on the membrane filter

50 μm

Fig. 6. SEM micrograph of the longitudinal section of undamaged sheepskin structure treated with sterile brine (Control) stored for 47 days at 33°C

30 μm

Fig. 7. SEM micrograph of the longitudinal section of damaged corium layer of sheepskin treated with Haloarcula salaria (AT1) stored for 47 days at 33°C

40 μm

Fig. 8. SEM micrograph of the longitudinal section of damaged corium layer of sheepskin treated with Halobacterium salinarum (22T6) stored for 47 days at 33°C

50 μm

Fig. 9. SEM micrograph of the longitudinal section of damaged corium layer of sheepskin treated with Haloarcula tradensis (7T3) stored for 47 days at 33°C

The present study proved that organoleptic changes detected in the sheepskins were closely related to proteolytic and lipolytic activities of extremely halophilic archaeal strains on the skin.

Electron micrographs also showed that each test isolate and a mixed culture of extremely halophilic strains destroyed the skins’ collagen fibres. We did not detect any difference when assessing the



efficacy of sterile brine and electrically treated brine curing processes of sheepskin samples throughout 47 days. We did not observe any damage to the compactness of sheepskin structure cured with both the sterile brine and electrically treated brine containing the mixed culture. Both methods were found very effective for preventing archaeal growth and damage on the brine cured sheepskins. Our results were consistent with those of other

experimental studies on the extremely halophilic strains and culture collection strains of extremely halophilic archaea (25, 26). In our previous experiment, SEM images showed that hides cured with proteolytic extremely halophilic archaeal strains had red heat and severe grain damage after 49 days of storage at 41°C (25). In another study, the cured hides with extremely halophilic Haloferax gibbonsii (ATCC 33959TM) exhibited hair loss, thinner and flaccid structure; these consequences of deterioration and loss of hide substance. The open fibre structure was also detected in the corium of the hide inoculated with Haloferax gibbonsii (27). The SEM images showed that the fibre structures of hide were broken down into the smaller fibres after 43 days (27).

4. Conclusion

This is the first study that detects adverse effects of characterised extremely halophilic archaeal strains on brine cured sheepskins with SEM. SEM images proved that proteolytic and lipolytic Haloarcula salaria AT1, Halobacterium salinarum 22T6, Haloarcula tradensis 7T3 caused corium fibres to split apart in cured sheepskins after 47 days in storage. The mixed culture of proteolytic and lipolytic extremely halophilic archaea originating in curing salt can be exterminated with

application of 1.5 A DC treatment in 7 min. Our experimental results proved that a curing process using sterile brine or brine treated with electric current prevented red heat and deterioration of sheepskins during long storage periods. Therefore, an environmentally friendly, easy, cheap, very simple electrolysis system is a very attractive alternative for brine disinfection: (a) it kills different species of microorganisms including proteolytic and lipolytic extremely halophilic archaea; (b) it prevents development of resistant strains in leather factories; (c) it kills very effectively the aggregated microorganisms found in the brine containing high organic substances; (d) it can achieve a reduction factor of more than 6 log10 for proteolytic and lipolytic extremely halophilic archaea; and (e) it has irreversibly lethal action. Hence, we suggest using this effective brine disinfection system in the leather industry after sufficient insulation and grounding are provided by an electrical engineer.

Acknowledgements

This study is dedicated to our wonderful teachers who have contributed so much to science and taught us to love this discipline. We thank Yeşim Müge Şahin, Demet Sezgin Mansuroğlu (Istanbul Arel University, Turkey) and Aslıhan Çetinbaş Genç (Marmara University) for the SEM micrographs. We are also grateful to Martin Louis Duncan (Virginia Union University, USA) for his critical reading of the manuscript.

References

1. F. Rodriguez-Valera, F. Ruiz-Berraquero and A. Ramos-Cormenzana, Microb. Ecol., 1981, 7, (3), 235

50 μm

Fig. 10. SEM micrograph of the longitudinal section of damaged corium layer of sheepskin treated with mixed culture of Haloarcula salaria (AT1), Halobacterium salinarum (22T6), Haloarcula tradensis (7T3) stored for 47 days at 33°C

100 μm

Fig. 11. SEM micrograph of the longitudinal section of undamaged sheepskin structure treated with electrically inactivated mixed culture stored for 47 days at 33°C



2. D. R. Arahal, F. E. Dewhirst, B. J. Paster, B. E. Volcani and A. Ventosa, Appl. Environ. Microbiol., 1996, 62, (10), 3779

3. A. Oren, A. Ventosa, M. C. Gutiérrez and M. Kamekura, Int. J. Syst. Bacteriol., 1999, 49, (3), 1149

4. M. Birbir, A. Ogan, B. Calli and B. Mertoglu, World J. Microbiol. Biotechnol., 2004, 20, (6), 613

5. M. Birbir, B. Calli, B. Mertoglu, R. E. Bardavid, A. Oren, M. N. Ogmen and A. Ogan, World J. Microbiol. Biotechnol., 2007, 23, (3), 309

6. R. Ghai, L. Pašić, A. B. Fernández, A.-B. Martin-Cuadrado, C. M. Mizuno, K. D. McMahon, R. T. Papke, R. Stepanauskas, B. Rodriguez-Brito, F. Rohwer, C. Sánchez-Porro, A. Ventosa and F. Rodríguez-Valera, Sci. Rep., 2011, 1, 135

7. A. Ventosa, A. B. Fernández, M. J. León, C. Sánchez-Porro and F. Rodriguez-Valera, Extremophiles, 2014, 18, (5), 811

8. A. B. Fernández, R. Ghai, A.-B. Martin-Cuadrado, C. Sánchez-Porro, F. Rodriguez-Valera and A. Ventosa, FEMS Microbiol. Ecol., 2014, 88, (3), 623

9. A. Ventosa, R. R. de la Haba, C. Sánchez-Porro and R. T. Papke, Curr. Opin. Microbiol., 2015, 25, 80

10. W. E. Kallenberger, ‘Halophilic Bacteria in Hide Curing’, PhD Thesis, Division of Graduate Studies and Research of the University of Cincinnati, Department of Basic Science Tanning Research of the Collage of Arts and Science, Cincinnati, USA, 6th August, 1985, p. 79

11. M. A. Amoozegar, M. Siroosi, S. Atashgahi, H. Smidt and A. Ventosa, Microbiology, 2017, 163, (5), 623

12. P. Corral, M. A. Amoozegar and A. Ventosa, Mar. Drugs, 2020, 18, (1), 33

13. D. J. Kushner, ‘Life in High Salt and Solute Concentrations’, in “Microbial Life in Extreme Environments”, ed. D. J. Kushner, Academic Press, London, UK, 1978, p. 318

14. D. J. Kushner, ‘Growth and Nutrition of Halophilic Bacteria’, in “The Biology of Halophilic Bacteria”, eds. R. H. Vreeland, L. I. Hochstein, 1st Edn, CRC Press Inc, Boca Raton, Florida, USA, 1992, p. 87

15. W. D. Grant, M. Kamekura, T. J. McGenity and A. Ventosa, ‘Order I. Halobacteriales Grant and Larsen 1989b, 495VP’, in “Bergey’s Manual® of Systematic Bacteriology: The Archaea and the Deeply Branching and Phototrophic Bacteria”, eds. D. R. Boone, R. W. Castenholz and G. M. Garrity, 2nd Edn., Springer-Verlag, New York, USA, 2001, Vol. 1, p. 294

16. M. T. Madigan, K. S. Bender, D. H. Buckley, W. M. Sattley and D. A. Stahl, “Brock Biology

of Microorganisms”, 15th Edn., Global Edition, Pearson Education Ltd, London, UK, 2019

17. M. Bergmann, J. Soc. Leather Technol. Chem., 1929, 11, 599

18. L. S. Stuart, R. W. Frey and L. H. James, ‘Microbiological Studies of Salt in Relation to the Reddening of Salted Hides’, Technical Bulletin No. 383, United States Department of Agriculture, Washington, DC, USA, September, 1933, 32 pp

19. D. Berber and M. Birbir, J. Am. Leather Chem. Assoc., 2010, 105, (10), 320

20. D. G. Bailey and M. Birbir, J. Am. Leather Chem. Assoc., 1993, 88 (8), 285

21. M. Birbir, J. Turk. Microbiol. Soc., 1997, 27, 68

22. P. Caglayan, C. Sánchez Porro, A. Ventosa and M. Birbir, “Characterization of Moderately and Extremely Halophilic Microorganisms from Salt-pack Cured Hides”, 9th International Congress Extremophiles, Seville, Spain, 10th–13th September, 2012, pp. 128

23. C. Akpolat, A. Ventosa, M. Birbir, C. Sánchez-Porro and P. Caglayan, J. Am. Leather Chem. Assoc., 2015, 110, (7), 211

24. S. T. Bilgi, B. Meriçli Yapici and İ. Karaboz, J. Am. Leather Chem. Assoc., 2015, 110, (2), 33

25. D. G. Bailey and M. Birbir, J. Am. Leather Chem. Assoc., 1996, 91, (2), 47

26. R. H. Vreeland, S. Angelini and D. G. Bailey, J. Am. Leather Chem. Assoc., 1998, 93 (4), 121

27. A. D. Brown, Biochim. Biophys. Acta, 1963, 75, 425

28. A. Oren, A. Ventosa and W. D. Grant, Int. J. Syst. Bacteriol., 1997, 47, (1) 233

29. H. P. Dussault, J. Bacteriol., 1955, 70, (4), 484

30. M. Birbir and A. Ilgaz, J. Soc. Leather Technol. Chem., 1996, 80, (5), 147

31. T. R. Johnson and C. L. Case, “Laboratory Experiments in Microbiology”, 9th Edn., Pearson Benjamin Cummings, San Francisco, CA, USA, 2010

32. E. F. Delong, Proc. Natl. Acad. Sci. USA, 1992, 89, (12), 5685

33. S.-H. Yoon, S.-M. Ha, S. Kwon, J. Lim, Y. Kim, H. Seo and J. Chun, Int. J. Syst. Evol. Microbiol., 2017, 67, (5), 1613

34. J.-C. Park, M. S. Lee, D. H. Lee, B. J. Park, D.-W. Han, M. Uzawa and K. Takatori, Appl. Environ. Microbiol., 2003, 69, (4), 2405

35. Y. Birbir and M. Birbir, J. Electrostat., 2006, 64, (12), 791

36. D. G. Bailey, J. Am. Leather Chem. Assoc., 2003, 98, 308

37. M. Das Murtey and P. Ramasamy, ‘Sample Preparations for Scanning Electron Microscopy – Life



Sciences’, in “Modern Electron Microscopy in Physical and Life Sciences”, ed. M. Janecek, InTech Open Ltd, London, UK, 2016, pp. 162–185

38. S. Namwong, S. Tanasupawat, T. Kudo and T. Itoh, Int. J. Syst. Evol. Microbiol., 2011, 61, (2), 231

39. L. Ducharme, A. T. Matheson, M. Yaguchi and L. P. Visentin, Can. J. Microbiol., 1972, 18, (8), 1349

40. P. Caglayan, M. Birbir, C. Sánchez-Porro, A. Ventosa and Y. Birbir, J. Am. Leather Chem. Assoc., 2018, 113, (2), 41

41. Y. Birbir, D. Degirmenci and M. Birbir, J. Electrostat., 2008, 66, (7–8), 388

42. Y. Birbir, S. Anik, M. Birbir and P. Caglayan, Johnson Matthey Technol. Rev., 2015, 59, (2), 109

43. C. A. Money and U. Adminis, ‘Curing of Hides and Skins: Alternative Methods’, Meat Technology Update, 94/3, Australian Meat Technology (AMT), Queensland, Australia, 1994

44. W. E. Kallenberger and R. M. Lollar, J. Am. Leather Chem. Assoc., 1986, 81, 248

The Authors

Meral Birbir graduated from the Biology Department, Ataturk Faculty of Education, Marmara University, Turkey. She received her MSc and PhD Degrees in Biology (especially microbiology) from the Institute of Pure and Applied Sciences, Marmara University. Professor Birbir has been working in the Biology Department of Marmara University since 1985. She was a research scientist at the Department of Pathology and Microbiology, Veterinary Medical School, Purdue University, USA (1990) and Hides and Leather Department of the United States Department of Agriculture (1992–1993). Her research interests are halophilic microorganisms, hide/skin microbiology, antimicrobial agents, electric current applications on microorganisms and microbial communities in hypersaline environments.

Pinar Caglayan graduated from the Biology Department, Ataturk Faculty of Education, Marmara University. She received her MSc and PhD Degrees in Biology from the Institute of Pure and Applied Sciences, Marmara University. She was an Erasmus student in the Department of Microbiology and Parasitology, Faculty of Pharmacy, Sevilla University, Spain (2008–2009). She has been working as a research and teaching assistant at the Division of Plant Diseases and Microbiology, Marmara University since 2011. Her research interests are moderately halophilic bacteria, extremely halophilic archaea, antimicrobial agents, hide microbiology and electric current applications on microorganisms.

Yasar Birbir received his BSc Degree from Gazi University, Turkey, and MSc and PhD Degrees in Electrical Education from Marmara University. He has been working at Marmara University since 1983. He attended the World Bank Industrial Training Project at Indiana and Purdue Universities, USA (1989–1990). He worked as a visiting research scientist at the Electrical and Computer Engineering Department of Drexel University, USA (1992–1993). He has been working as a Professor at the Technology Faculty, Department of Electrical Engineering, Marmara University. His current interests are power electronic converters and drivers, electromagnetic filtering processes in industry and applications of electric currents for inactivation of different microorganisms.




Johnson Matthey HighlightsA selection of recent publications by Johnson Matthey R&D staff and collaborators

A Stochastic Approach to Model Chemical Looping CombustionM. A. Schnellmann, G. Williams and J. S. Dennis, Powder Technol., 2020, 365, 39

Using two coupled fluidised-bed reactors, a stochastic model for reactor-regenerator systems was established. As a result the stochastic model was able to achieve the simulation of the circulating fluidised bed with acceptable precision. This model had been used before to comprehend how sensitive a chemical looping combustion (CLC) process is when other factors, for example the nature of gas-solid reactions, are included. To show the stochastic model has value for simulation and optimisation formations of CLC it is applied here with methane fuel gas in a laboratory-scale circulating fluidised bed.

ZSM-5 Additive Deactivation with Nickel and Vanadium Metals in the Fluid Catalytic Cracking (FCC) ProcessA. A. Gusev, A. C. Psarras, K. S. Triantafyllidis, A. A. Lappas, P. A. Diddams and I. A. Vasalos, Ind. Eng. Chem. Res., 2020, 59, (6), 2631

This article explores properties of ZSM-5 additives and the role of nickel and vanadium in fluid catalytic cracking (FCC). Loadings of 4000 ppm and 12,000 ppm nickel and vanadium were found. There was deactivation of ZSM-5 in a cyclic deactivation unit and vacuum gas oil (VGO) reacted with nickel and vanadium naphthenates when cracking-regeneration reactions were undertaken. There was an even distribution of nickel across a particle where characterisation of deactivated ZSM-5 additives by nitrogen physiosorption, SEM and pyridine FTIR techniques were used. The disparity was small in the Brønsted acidity, but when nickel and vanadium were added there was a rise in Lewis acidity. Further tests were carried out with VGO and where butylene increased,

propylene selectivity had a decrease in response to the metals.

Ga2.52V2.48O7.33(OH)0.67, A Synthetic Member of the Nolanite/Akdalaite-Type Family of Oxyhydroxides Containing Trivalent Vanadium D. S. Cook, M. R. Lees, J. M. Fisher, D. Thompsett and R. I. Walton, J. Solid State Chem., 2020, 288, 121396

Powder neutron diffraction shows oxyhydroxide Ga2.52V2.48O7.33(OH)0.67 prepared by reaction between gallium metal and Na3VO4 in a 1:1 monoethanolamine:water mixture at 240ºC demonstrates the material is isostructural with nolanite and akdalaite (Figure 1). Rietveld refinement was undertaken against the data showing all vanadium is octahedrally coordinated. Vanadium’s oxidation state is close to V3+ when vanadium K-edge XANES spectroscopy is used. There is dehydration around 300ºC (oxide Ga2.52V2.48O8 is produced and has a larger amount of V4+) followed by decomposition at 500ºC. While both materials seem to follow the Curie-Weiss law at high temperatures, this is not so at low temperature. No reducing gas atmospheres are required in the preparation of V(III) oxides.

Fig. 1. Reprinted from D. S. Cook et al., J. Solid State Chem., 2020, 288, 121396, Copyright (2020), with permission from Elsevier

1 µm



Hierarchical ZSM-5 Catalysts: The Effect of Different Intracrystalline Pore Dimensions on Catalyst Deactivation Behaviour in the MTO Reaction T. Weissenberger, A. G. F. Machoke, J. Bauer, R. Dotzel, J. L. Casci, M. Hartmann and W. Schwieger, ChemCatChem, 2020, 12, (9), 2461

ZSM-5 zeolites used as methanol-to-olefins (MTO) catalysts were studied to determine the effect of intracrystalline pore systems in different combinations. Intracrystalline mesopores, intracrystalline macropores and a novel ZSM-5 type zeolite with both intracrystalline meso and macropores were used. There was a prolonged catalyst lifetime with the hierarchical catalysts unlike microporous only ZSM-5 catalyst. Using the intracrystalline mesopores and intracrystalline macropores as catalysts resulted in the ZSM-5 catalyst lasting up to three times longer. There were a number of important outcomes which were also noted including how mesopores and macropores effect catalyst deactivation. Overall, the study demonstrates intracrystalline macropores (alone or in combination with mesopores) significantly enhance the ZSM-5 catalytic performance in the MTO reaction.

Accelerating Pharmaceutical Development via Metal-Mediated Bond FormationC. J. Borths and S. D. Walker, Israel J. Chem., 2020, 60, (3–4), 340

This article presents work being undertaken related to progress with metal-mediated carbon–carbon and carbon–heteroatom bond forming processes. The viewpoint of a current drug portfolio is used. Several case studies are discussed from the authors’ laboratories looking at synthetic challenges plus prospects offered by pharmaceutically pertinent platforms. In several instances, available synthetic methods are challenged by target structures. This drives development permitting the acceleration of technology for drug development. There is also a discussion about metal-mediated processes at large scale.

Facile Synthesis of Precious-Metal Single-Site Catalysts Using Organic Solvents X. Sun, S. R. Dawson, T. E. Parmentier, G. Malta, T. E. Davies, Q. He, L. Lu, D. J. Morgan, N. Carthey, P. Johnston, S. A. Kondrat, S. J. Freakley, C. J. Kiely and G. J. Hutchings, Nature Chem., 2020, 12, (6), 560

In many catalytic reactions, high activity and selectivity can be demonstrated by single-site catalysts. Creation of low metal loadings or a variety of metal species can be achieved by impregnation, for example, from strongly oxidising aqueous solutions. This study shows an atomic distribution of cationic metal species is achieved

with an impregnation of metal precursors onto activated carbon from a low boiling point solvent. Single-site gold, palladium, ruthenium and platinum catalysts supported on carbon were prepared in a facile method. In addition, it is shown that a single-site gold on carbon catalyst for acetylene hydrochlorination can be produced by this method.

Efficient and Selective Solvent-Free Homogeneous Hydrogenation of Aldehydes Under Mild Reaction Conditions Using [RuCl2(dppb)(ampy)] A. Zanotti-Gerosa, T. Angelini and S. Roseblade, Tetrahedron Lett., 2020, 61, (13), 151677

Using commercial grade aldehydes, effective, solvent-free homogeneous hydrogenation of aldehydes was undertaken with catalysts [RuCl2(dppb)(ampy)] and [RuCl2(dppf)(ampy)]. This gave high conversion to the related alcohols using molar catalyst loadings of 10,000/1– 50,000/1. Aldehydes can be reduced preventing byproducts being formed with the minimum of waste which has led to a solvent-free protocol being established. This gives a straightforward hydrogenation technique for reduction of aldehydes to alcohols and commercial grade aldehydes require no further purification.

Innovation in Fischer–Tropsch: Developing Fundamental Understanding to Support Commercial OpportunitiesM. Peacock, J. Paterson, L. Reed, S. Davies, S. Carter, A. Coe and J. Clarkson, Top. Catal., 2020, 63, (3–4), 328

The BP-Johnson Matthey proprietary Fischer-Tropsch technology and advanced CANSTM reactor and catalyst system are detailed. It provides improved heat transfer, reduced pressure drop and higher productivity and subsequently less financial expenditure. A clear understanding of how catalysts behave is crucial to finding a catalyst stable during its use and life. This report presents a study on catalyst activation on different catalyst supports and combines in situ techniques and reactor testing. Logical and systematic catalyst programmes are crucial for their development and are discussed in the results. Also, catalyst understanding, optimisation and development in combination with the novel CANSTM reactor design can maximise potential.

Cu/M:ZnO (M = Mg, Al, Cu) Colloidal Nanocatalysts for the Solution Hydrogenation of Carbon Dioxide to MethanolA. H. M. Leung, A. García-Trenco, A. Phanopoulos, A. Regoutz, M. E. Schuster, S. D. Pike, M. S. P. Shaffer and C. K. Williams, J. Mater. Chem. A, 2020, 8, (22), 11282

A synthesis is undertaken using controlled hydrolysis of a mixture of organometallic precursors



for doped-ZnO NPs capped with dioctylphosphinate ligands. Following substitutional doping and after hydrolysis, colloidal nanoparticles (2–3 nm) were characterised. Doped-ZnO nanoparticles and colloidal Cu(0) nanoparticles in solution were applied for hydrogenation catalysis of CO2 to methanol in a liquid-phase continuous flow stirred tank reactor under the following conditions: 210ºC, 50 bar, CO2:H2 = 1:3, 150 ml min−1, mesitylene, 20 h. Higher rates are displayed for all catalyst systems with respect to methanol production compared to a benchmark catalyst. There is better stability. There was around double the activity for Al(III)-doped nanocatalyst. Mg(II) doping outperforms the benchmark catalyst but was worse compared to undoped ZnO. There is an implication that Al(III) migrates to the catalyst surface, and is proposed to enable stabilisation of the catalytic ZnO/Cu interfaces.

N-Functionalised Imidazoles as Stabilisers for Metal Nanoparticles in Catalysis and Anion BindingC. J. Serpell, J. Cookson and P. D. Beer, ChemistryOpen, 2020, 9, (6), 683

The physicochemical properties of metal NPs are discrete from bulk and molecular metal species. Consequently, this delivers opportunities in areas like catalysis and sensing, for example. The surface of the NPs usually need to be protected to hamper aggregation. However, access to the surface can also be blocked by these coatings preventing the ability to benefit from their uncommon properties. The article shows that palladium, platinum, gold and silver NPs can be stabilised by alkyl imidazoles. It also outlines the limits of their synthesis. Proof-of-principle in catalysis and anion binding is established showing that the ligands deliver a level of surface protection.

In Situ K-edge X-ray Absorption Spectroscopy of the Ligand Environment of Single-Site Au/C Catalysts During Acetylene HydrochlorinationG. Malta, S. A. Kondrat, S. J. Freakley, D. J. Morgan, E. K. Gibson, P. P. Wells, M. Aramini, D. Gianolio, P. B. J. Thompson, P. Johnston and G. J. Hutchings,

Chem. Sci., 2020, 11, (27), 7040

The environmental impact of acetylene hydrochlorination was substantially reduced by replacing HgCl2/C with Au/C as a catalyst. Atomically dispersed cationic gold species are the catalytically active site. There have been limited studies which look at the ligand environment around the metal centre. This study uses K-edge soft XAS. Three separate chlorine species are identified and how they evolve in the reaction is demonstrated. Au–S interactions are established in catalysts prepared using thiosulfate precursors. The catalysts display evidence of high stability towards reduction to inactive metal NPs. Gas switching experiments made clear this stability. C2H2 on its own did not particularly change the gold electronic structure and the thiosulfate catalyst was not deactivated.

Optimization of Biomass Pyrolysis Vapor Upgrading Using a Laminar Entrained-Flow Reactor SystemB. Peterson, C. Engtrakul, T. J. Evans, K. Iisa, M. J. Watson, M. W. Jarvis, D. J. Robichaud, C. Mukarakate and M. R. Nimlos, Energy Fuels, 2020, 34, (5), 6030

To obtain understanding of commercial scale ex situ catalytic fast-pyrolysis (CFP) a customised bench-scale continuous-flow catalytic fast-pyrolysis CFP reactor system was built. The study successfully carried out CFP of pine over two commercial zeolite catalysts. The transmission of pyrolysis vapours to the vapour-phase upgrader was optimised to limit secondary thermal cracking and preserve carbon in the ex situ CFP process. Products attained were comparable to those from fixed bed and fluidised bed reactor systems and entrained-flow riser reactor systems. Experiments that duplicated the process provided a good average mass balance closure and comparable trends in deactivation of catalyst were seen in the laminar entrained-flow reactor system. There was a decreasing catalyst-to-biomass ratio. Optimised conditions suggest a feasible option for CFP of pine. For the two catalysts tested, minor variances were detected.




Gülşen Altuğ*Department of Marine Biology, Faculty of Aquatic Sciences, Istanbul University, Balabanağa Mahallesi Ordu Caddesi No 8, Laleli, Fatih, Istanbul, 34134, Turkey

Mine ÇardakDepartment of Fisheries Technology, Faculty of Çanakkale Applied Sciences, Çanakkale Onsekiz Mart University, Terzioğlu Campus, Çanakkale, 17020, Turkey

Pelin Saliha Çiftçi TüretkenDepartment of Marine Biology, Faculty of Aquatic Sciences, Istanbul University, Balabanağa Mahallesi Ordu Caddesi No 8, Laleli, Fatih, Istanbul, 34134, Turkey

Samet KalkanDepartment of Marine Biology, Faculty of Fisheries and Aquatic Sciences, Recep Tayyip Erdoğan University, Zihni Derin Campus, Rize, 53100, Turkey

Sevan GürünDepartment of Marine Biology, Faculty of Aquatic Sciences, Istanbul University, Balabanağa Mahallesi Ordu Caddesi No 8, Laleli, Fatih, Istanbul, 34134, Turkey


Heavy metal and antibiotic-resistant bacteria have potential for environmental bioremediation applications. Resistant bacteria were investigated

in sediment and seawater samples taken from the Aegean Sea, Turkey, between 2011 and 2013. Bioindicator bacteria in seawater samples were tested using the membrane filtration technique. The spread plate technique and VITEK® 2 Compact 30 micro identification system were used for heterotrophic aerobic bacteria in the samples. The minimum inhibition concentration method was used for heavy metal-resistant bacteria. Antibiotic-resistant bacteria were tested using the disk diffusion method. All bacteria isolated from sediment samples showed 100% resistance to rifampicin, sulfonamide, tetracycline and ampicillin. 98% of isolates were resistant against nitrofurantoin and oxytetracycline. Higher antibiotic and heavy metal resistance was recorded in bacteria isolated from sediment than seawater samples. The highest levels of bacterial metal resistance were recorded against copper (58.3%), zinc (33.8%), lead (32.1%), chromium (31%) and iron (25.2%). The results show that antibiotic and heavy metal resistance in bacteria from sediment and seawater can be observed as responses to environmental influences including pollution in marine areas.

1. Introduction

In the era of Industry 4.0, with global climate change, increasing population and developing technology, the spread of heavy metal pollutants in aquatic areas is increasing. Bacterial resistance and metal accumulation capability are common phenomena that can be exploited for the bioremediation of the environment, hence these resistant bacteria may be potential candidates for biotechnological applications. Despite the risks caused by antibiotic-

Antibiotic and Heavy Metal Resistant Bacteria Isolated from Aegean Sea Water and Sediment in Güllük Bay, TurkeyQuantifying the resistance of identified bacteria species with potential for environmental remediation applications



resistant bacteria, heavy metal-resistant bacteria can be used in detoxification processes to convert a toxic form to a harmless form of a substance by developing biotransformation mechanisms. Bioremediation studies have been carried out to identify candidate species (1–4).In recent years, the increase in pollution by toxic

compounds and heavy metals in marine areas makes it increasingly important to study the relationships between bacteria and toxic compounds. Studies related to the transformation of compounds into different forms via bacterial metabolic processes for the removal of toxic substances from the environment have gained importance. Detection of bacteria that are resistant to heavy metals in natural environments constitutes the first step to provide data for remediation studies.Bacteria are some of the most important

components in marine ecosystems. Since bacteria adapt to new conditions created by environmental variables around them, knowledge of bacteria provides data in terms of defining environmental factors, public health status and ecosystem function. Marine areas are exposed to domestic and industrial wastes depending on local technology levels and population. Many xenobiotic micro pollutants, antibiotic derivatives and metabolites reach the sea from human activity. This concerning issue is considered an important factor for global health with respect to the evolution and detection of antibiotic resistance in bacterial pathogens (5). Since the spread of antimicrobial resistance is not restricted by phylogenetic, geographic or ecological borders, studies describing regional status of bacterial resistance in natural areas are important.Antibiotic resistance can spread rapidly among

bacterial species (6). It is known that the occurrence of antibiotic-resistant bacteria in natural environments reduces the effectiveness of antibiotics in the treatment of infectious diseases. Due to the increasing global resistance of bacteria against antibiotics, humanity is constantly being forced to develop new antibiotic derivatives. This vicious circle is one of the most important problems of our age and poses a threat for the future. Thus, it is important to know the resistance levels of bacteria and to produce regional antibiotic resistance profiles in natural areas. Aquatic environments constitute a way to disseminate not only antibiotic-resistant bacteria but also the resistant genes in natural bacterial habitats (7). It has been well documented that the aquatic environment is a potential reservoir of antibiotic-resistant bacteria, furthermore the prevalence and

persistence of antibiotic resistance in bacterial pathogens is a threat and a source of considerable concern to public health (8–15). It is known that environmental factors such as overpopulation, livestock farming, insufficient drainage and sanitation infrastructure may provide hotspots for environmental antibiotic-resistant bacteria transmission (16).In aquatic environments, antibiotic-resistant

bacteria can be accompanied by heavy metal-resistant bacteria that are often induced by the presence of metal caused by anthropogenic activities and environmental factors (16, 17). Heavy metals are introduced into the marine environment in different ways. Accumulation in sediment can affect aquatic life negatively for a long time. Bacteria that will take part in the transformation of heavy metal salts into harmless forms must be resistant to the heavy metals. Bacteria that cannot adapt to the changes metabolically will be eliminated and therefore various pollution inputs accumulated in the sediment will affect the composition of microbial diversity. Sediments containing harmful, inorganic or organic particles are relatively heterogeneous in terms of physical, chemical and biological properties and are an important source of heavy metal contamination (11). It has been reported that microplastics mediate the spread of metal- and antibiotic-resistant pathogens due to their ability to adsorb various pollutants (18, 19). Bacteria resistant to heavy metals in marine areas have developed various resistance mechanisms to counteract heavy metal stress. Only bacteria that can withstand the current heavy metal concentration can survive in these areas. Heavy metals accumulate in biota via food chains

and are transferred between organisms in marine environments. This cumulative process, named biomagnification, is higher in the sea than in terrestrial environments (15, 20) and this implies significant effects of heavy metal pollution in marine areas. On the other hand, heavy metal-resistant bacteria can play a role in detoxification by converting a toxic form into a harmless form through biotransformation mechanisms that develop in natural environments. These mechanisms include the formation and sequestration of heavy metals in complexes and the reduction of a metal to a less toxic species (21, 22). Metal-resistant bacteria have developed very

efficient and varying mechanisms for tolerating high levels of toxic metals and thus they carry an important potential for controlling heavy metal pollution (23). In many prokaryotes, it has been



shown that the mechanism for resisting heavy metals develops over time. This process has been studied in species such as Escherichia coli and Staphylococcus aureus. It is reported that many different species of Pseudomonas, Bacillus, Enterobacter, Providencia and Chryseobacterium are efficient for reducing heavy metals (1–4).It is known that the occurrence of bacteria

resistant to antibiotics and heavy metal salts in the sea is related to the pollutants present in the environment. For the reasons highlighted above, it is important to determine the profile of antibiotic and heavy metal-resistant bacteria in marine environments. Marine areas which have different environmental inputs present novel media for bacterial studies. For the present study, the Güllük Bay of the

Aegean Sea, Turkey, was chosen since it is a dynamic area due to marine transportation, seasonal population growth depending on tourism, aquaculture, recreational and agricultural activities and terrestrial pollution inputs transported from rivers. Probable faecal source analysis conducted in

Güllük Bay showed that the primary source of the detected bacteriological pollution is anthropogenic (24). A significant part of domestic wastewater in the region collects in sealed septic tanks. It is possible for the wastewater to reach the sea by mixing the sedimentary septic tanks with groundwater. Chemical and biological studies (24– 33) confirm that regional pollutants have reached Güllük Bay. It is well known that sewage transported via

domestic wastewater carries antibiotics to marine environments. This has an effect on metabolic capabilities of bacteria in marine environments. For example, β-lactam antibiotic derivatives used for human infection treatment may enter marine environments via domestic wastewater. Bacteria may obtain resistance via intercellular contact mostly using a conjugation mechanism (34). The existence of antibiotic-resistant bacteria is an indicator of domestic pollution. Furthermore, antibiotic-resistant bacteria may cause a vicious cycle. This problem has grown in recent years due to systematic use of antibiotics in animal husbandry and overuse of antibiotics (35, 36). The frequencies of heavy metal-resistant bacteria

and antibiotic-resistant bacteria were investigated in seawater and sediment samples collected from Güllük Bay in the period between May 2011 and February 2013.

2. Material and Methods

2.1 Sampling Area

Güllük Bay is an important location due to its natural resources. The region is open to different environmental influences and inputs due to tourism, port activities, marine transportation, domestic and industrial wastes and fish farms. The bay is also affected by the presence of Sarıçay Creek, Kazıklı Port, Güllük Port and Akbük Port (24–26). Fish farms were operated in Güllük Bay until 2008. Although they have been relocated away from the coastal regions to an offshore area, the indirect effects of this long-time pollution may have contributed to the sediment. The export of feldspar and bauxite from the region

has been conducted from ports within the borders of Güllük Bay. The port is mainly used by dry cargo and other cargo-type ships. It is reported that an annual average of 800,000 tonnes of ballast water is transported to the bay from 157 different ports. The amount of ballast water carried is reported as: 68% from the Mediterranean, 21% from the Aegean Sea, 7% from the Sea of Marmara, 2% from the Atlantic Ocean and 1% from the Black Sea and Red Sea, respectively (37). The operation of many tourism-oriented boats in Güllük Bay is also among the possible polluters of the bay due to bilge water and wastewater. More than half of Turkey’s sea bream and sea bass production was in farms operating in the coastal areas of the Güllük Bay for many years. These farms have been operating in the offshore areas of the region for the past 10 years. The domestic wastewater of the human population, reaching approximately 50,000 around the region in the summer months, and the wastes of small industrial establishments such as yogurt, yeast and olive oil producers that directly reach streams are the other main sources of pollution in Güllük Bay. The population of the Bodrum peninsula, which is 25,000 in winter, can reach 1,500,000 in summer (27). The change in the population between the seasons was among the biggest pollution sources according to the terrestrial bioindicator bacteria distribution in coastal areas in the region (24, 26). Sampling stations were selected to represent

tourist areas (G1, G5, G7, G8); harbours (G4, G6); fresh water entry-exit points of the Sarıçay Creek (G9); fish farms (G11, G12, G13); and the deepest point in the bay as a reference station (G14). Figure 1 shows the location of Güllük Bay and the sampling stations.



2.2 Sampling

Seawater and sediment samples were collected from 12 different sampling stations in Güllük Bay between May 2011 and February 2013. Three units of seawater samples were taken from each station at surface (0–30 cm), mid-point and bottom-point water (Figure 1). In each sampling process covering 12 stations, 36 seawater samples were collected. In the spring and summer months monthly, at other times seasonally, a total of 432 seawater samples were collected in the period between May 2011 and February 2013. The seawater samples were collected using a Nansen bottle that was cleaned with acid (10% HCl in distilled water), sterilised with alcohol (50:50, v/v) and rinsed with sterile water. The seawater samples were then transferred into brown sterile glass bottles and transported to the laboratory as a cold chain. Surface sediment samples were collected using

Ekman grab (HYDRO-BIOS Apparatebau GmbH, Germany, 15 × 15) from the sampling stations which have various depths from 8 m to 66 m (Figure 1). A total of 144 units of sediment samples were collected during the two-year study from 12 stations (one from each station). The sediment samples were transferred into sterile zip seal bags from Ekman grab and transferred in the cold chain to the laboratory.

2.3 Bacterial Isolation and Identification

Bacterial heavy metal and antibiotic resistance were tested in heterotrophic aerobic bacteria isolated from seawater and sediment samples. Heavy metal and antibiotic resistance of indicator bacteria (faecal coliform, coliform and faecal Streptococcus) isolated from the seawater samples were also tested.

2.3.1 Seawater Samples

Indicator bacteria and heterotrophic aerobic bacteria analyses were performed on the seawater samples. The membrane filtration technique was used to detect indicator bacteria. A sample containing 300 ml seawater was diluted serially (10–5 dilution) and filtered through membrane filters (0.45 µm, Sartorius AG, Göttingen, Germany). The filters were placed on m-Endo, m-FC and m-Azide media (Sartorius AG). The plates were incubated for 24 h (at 37 ± 0.1°C; at 44 ± 0.1°C for m-FC). Brown-red colonies growing on the azide medium were considered as suspicious faecal Streptococcus, blue colonies growing on the m-FC medium as suspicious faecal coliform and yellow-green colonies with yellow-metallic gloss on the m-Endo medium as suspicious coliform. Cytochrome oxidase test (API® 20 Strep, bioMérieux, France) was performed on suspicious

Code Mid point, m Max. depth, m

G1 25 50

G4 25 50

G5 10 20

G6 11 22

G7 14 28

G8 8 16

G9 4 8

G10 18 37

G11 5 10

G12 25 50

G13 12 25

G14 33 66

G1

G4

G6G5

G7

G8

G9

G10

G11G12

G13

G14

Black Sea

TURKEY

Marmara Sea

Aegean Sea

Mediterranean Sea

Fig. 1. Location of Güllük Bay and seawater (0–30 cm surface, mid-point and bottom-point) and sediment sampling stations



coliform colonies and oxidase negative colonies were evaluated numerically. Cytochrome oxidase (API® 20 Strep, bioMérieux) and indole tests were performed on the suspicious colonies of faecal coliform.Colonies with oxidase negative and indole positive

results were evaluated as faecal coliform. Suspicious Streptococcus colonies, to which the catalase test was applied (1 ml, 3% H2O2), were incubated on Bile Esculin Agar (BEA) for 18 h at 37°C for esculin hydrolysis and 40% bile resistance control. Blackening in the medium and the formation of black shadow around the colony, positive of esculin hydrolysis, and the number of colonies showing growth in the medium were evaluated as 40% bile resistant, and catalase negative and breeding colonies in BEA were evaluated as faecal Streptococcus. Counted colonies were multiplied by the 10–5 dilution factor to determine the number of colony forming units (CFU) 100 ml–1 in the original sample (38).The spread plate technique was used for

heterotrophic aerobic bacteria analyses in seawater. Seawater samples 0.1 ml with 10–5 dilution were used for duplicate spreading on the DifcoTM Marine Agar 2216 (Becton, Dickinson and Company, USA) and the plates were incubated for five days at 22 ± 0.1°C. At the end of the incubating period, counted colonies were multiplied by the 10–5 dilution factor to determine the number of CFU ml–1 in the original sample. An average of 10 different colonies were picked and restreaked several times to obtain pure cultures. The pure isolates were Gram-stained. For identification of spore-forming bacilli, the isolates were stained with Indian ink according to the negative staining technique and were evaluated using a light microscope (Nikon E110, Nikon, Japan). The isolates were then tested using Gram-negative fermenting and non-fermenting bacilli (GN), Gram-positive cocci and non-spore-forming bacilli (GP) and Gram-positive spore-forming bacilli (BCL) cards in the automated micro identification system VITEK® 2 Compact 30 (bioMérieux) (39).

2.3.2 Sediment Samples

The spread plate technique was used for heterotrophic aerobic bacteria analyses in sediment samples. Each sediment sample was mixed and homogenised. Then 1 g sample was taken from each and serially diluted with sterile commercial seawater. 0.1 ml samples of 10–5 dilutions were taken and spread on DifcoTM Marine Agar 2216. The plates were incubated for five days at 22 ± 0.1°C. Growing colonies were evaluated as CFU g–1 (40).

Further processes related to heterotrophic bacteria identification were continued by using VITEK® 2 Compact 30 similarly to the seawater samples described above.

2.4 Bacterial Resistance Against Antibiotics

The antibiotic resistance of the isolates was examined by the Kirby–Bauer method with slight modifications. Two or three colonies of each isolate were suspended with 5 ml of DifcoTM Marine Broth 2216 and diluted with sterile water against the 0.5 McFarland turbidity standard to approximately 106 cells ml–1 and swabbed as 2 ml on DifcoTM Marine Agar 2216. Antibiotic discs (Oxoid, UK) containing ampicillin (10 µg), nitrofurantoin (300 µg), oxytetracycline (30 µg), sulfonamide (300 µg), rifampicin (2 µg), tetracycline (10 µg) and tetracycline (30 µg) were incubated for two to three days at 37°C. The results were interpreted according to the guidelines of the Clinical Laboratory Standard Institute (CLSI) (41). All isolates that showed resistance were classified as ‘resistant’. Other isolates that did not show resistance were classified as ‘sensitive’ or ‘susceptible’.

2.4.1 Multiple Antibiotic Resistance

The multiple antibiotic resistance (MAR) index of a given sample was calculated by the equation: a/ (bc), where a represents the aggregate antibiotic resistance score of all isolates from a sample; b is the total number of isolates; and c is the number of isolates from a sample (42). Bacterial isolates that displayed resistance to three or more antibiotic agents were designated as multiple antibiotic resistant (ranging from two to 10).

2.5 Bacterial Resistance Against Heavy Metal Salts

Different concentrations (50 μg ml–1, 100 μg ml–1, 150 μg ml–1, 200 μg ml–1 and 250 μg ml–1) of heavy metal salts (FeSO4, ZnSO4, CuSO4, Cr2(SO4)3 and Pb(NO3)2) were used to test the bacteria resistivity against iron, zinc, copper, chromium and lead. The microdilution method was followed with minor modifications to determine the resistance of isolates to heavy metals (43). Stock solutions of metal salts prepared in distilled water were sterilised by filtration (0.20 μm). In U-well microtiter plates, serial dilutions of heavy metals were prepared and then each well was inoculated with bacteria



inoculation. The OxoidTM Turbidometer (Thermo Fisher Scientific Inc, USA) provides the inoculum density standardisation for 0.5 McFarland which is necessary to ensure accurate reproducible results. Before the addition of bacterial inoculation, no precipitation was seen. The plates were incubated at 37°C for 24 h and then examined for visual turbidity. The lowest concentration of the metal salt, at which growth was inhibited (indicated by lack of turbidity), was taken as the minimum inhibitory concentration (MIC) (44) Samples of 10 μl were drawn from each well without turbidity and were subcultured on agar plates to determine bactericidal concentration. Reference strains of Escherichia coli (ATCC®

25922TM), Salmonella enterica (ATCC® 2577TM) and Staphylococcus epidermidis (ATCC® 12228TM) which are susceptible to Cu2+, Zn2+, Pb2+, Cr2+ and Fe3+ and metal-free plates were used in the control tests to evaluate the viability of the strains and culture media. All of the experiments were carried out in triplicate.

3. Results

3.1 Bacterial Resistance Against Antibiotics

Table I shows the antibiotic-resistant, intermediate or susceptible bacteria species isolated from the seawater and sediment samples in this study.Bacterial species isolated from the seawater samples

showed considerable resistance to rifampicin (98%), sulfonamide (98%) and ampicillin (76%) and considerable sensitivity to tetracycline-30 µg (52%), tetracycline-10 µg (39%) and oxytetracycline (33%). Almost all the bacterial species isolated from sediment samples showed resistance to rifampicin (100%), sulfonamide (100%), ampicillin (100%), nitrofurantoin (98%), tetracycline-30 µg (100%), tetracycline-10 µg (100%) and oxytetracycline (98%) while they showed almost no sensitivity to antibiotics except nitrofurantoin (2%) and oxytetracycline (2%). Pseudomonas aeruginosa (24%) and Sphingomonas paucimobilis (20%), isolated from seawater samples, showed higher resistance to antibiotics than did Raoultella oxytica, Staphylococcus xylosus, Kocuria kristinae, Aeromonas salmonicida and Proteus vulgaris strains. On the contrary, Aeromonas caviae, Alicyclobacillus acidoterrestris, Brevundimonas diminuta, Chryseobacterium indologenes, Lactococcus garvieae, Neisseria animaloris, Pseudomonas aeruginosa, Serratia marcescens,

Shewanella algae and Vibrio parahaemolyticus isolates from the sediment samples showed resistance to all antibiotics (Table I). The highest number of antibiotic-resistant

bacteria were detected from the sediment samples. The frequency of resistant bacteria (%) to oxytetracycline (30 µg), nitrofurantoin (300 µg), rifampicin (2 µg), tetracycline (10 µg), tetracycline (30 µg), sulfonamide (300 µg) and ampicillin (10 µg) from the seawater and sediment samples are shown in Figure 2. The frequencies of antibiotic resistance in bacteria species from seawater and sediment samples are shown in Figure 3.A total of 258 and 158 isolates were tested

against antibiotics from seawater and sediment samples, respectively. The frequencies of resistance against seven antibiotics in bacteria species isolated from the seawater samples were recorded as 49% in Gammaproteobacteria, 22% in Αlphaproteobacteria, 3% in Betaproteobacteria, 14% in Bacilli, 8% in Flavobacteriia and 4% in Actinomycetales. The resistance frequencies against seven antibiotics in bacteria isolated from the sediment samples were recorded as 43% in Gammaproteobacteria, 34% in Bacilli, 7% in Αlphaproteobacteria, 7% in Betaproteobacteria, 7% in Flavobacteriia and 2% in Actinomycetales.

3.2 Multiple Antibiotic Resistance Indexes

The MAR index was calculated for each of the antibiotic-resistant bacteria. If the MAR index is lower than 0.2, it shows a non-point based source of pollution and if it is higher than 0.2 it shows point-based pollution and a high risk of contamination by excessive antibiotic presence (23). Table II shows the MAR indexes.The MAR indexes of the study showed possible

exposure of these bacterial isolates to the tested antibiotics. The MAR index of bacteria isolated from all stations around fish farm areas (0.0576) was 2.6 times greater than the MAR index for the combined non-fish farm areas (0.022).

3.3 Bacterial Resistance Against Heavy Metals

The frequencies of heavy metal resistance in the bacteria species isolated from the seawater samples were recorded as 76.72% in Gammaproteobacteria, 71.82% in Αlphaproteobacteria, 80.01% in Bacilli, 56.92% in Flavobacteriia and 75% in



Tab

le I

An

tib

ioti

c R

esis

tan

t, I

nte

rmed

iate

or

Su

scep

tib

le B

acte

ria

Sp

ecie

s Is

olat

ed f

rom

Sea

wat

er a

nd

Sed

imen

t

SampleO

rder

/cl

ass

test

ed (

%)

Bac

teri

al is

olat

es

test

ed (

n)

An

tib

ioti

csa

AM

(1

0 µ

g)

TE (30

µg

)S (3

00

µg

)TE (1

0 µ

g)

RD

(2 µ

g)

F/M

(30

0 µ

g)

OT

(30

µg

)Seawater

Prot

eoba

cter

ia/

Alp

ha

prot

eoba

cter

ia(2

7%)

Bre

vund

imon

as

dim

inut

a (3

)

R:

66.7

%I:

33.

3%S:

0.0%

R:

33.3

%I:

0.0

%S:

66.7

%

R:

100%

I: 0

.0%

S:

0.0%

R:

33.3

%I:

33.

3%S:

33.3

%

R:

100%

I: 0

.0%

S:

0.0%

R:

66.7

%I:

0.0

%S:

33.3

%

R:

66.7

%I:

33.

3%S:

0.0%

Bre

vund

imon

as

vesi

cula

ris

(4)

R:

75%

I: 0

.0%

S:

25%

R:

50%

I: 0

.0%

S:

50%

R:

100%

I: 0

.0%

S:

0.0%

R:

50%

I: 0

.0%

S:

50%

R:

100%

I: 0

.0%

S:

0.0%

R:

50%

I: 0

.0%

S:

50%

R:

25%

I: 0

.0%

S:

75%

Sph

ingo

mon

as

pauc

imob

ilis

(38)

R:

71.0

5%I:

2.6

3%S:

26.3

1%

R:

31.5

7%I:

5.2

6%S:

63.1

5%

R:

97.3

8%I:

0.0

%S:

2.63

%

R:

42.1

0%I:

7.8

9%S:

50%

R:

97.3

6%I:

0.0

%S:

2.63

%

R:

60.5

2%I:

0.0

%S:

39.4

7%

R:

26.3

1%I:

34.

21%

S:

39.4

7%

Sph

ingo

mon

as

thal

poph

ilum

(4)

R:

50%

I: 0

.0%

S:

50%

R:

0.0%

I: 0

.0%

S:

100%

R:

100%

I: 0

.0%

S:

0.0%

R:

0%I:

25%

S:

75%

R:

100%

I: 0

.0%

S:

0.0%

R:

75%

I: 0

%S:

25%

R:

25%

I: 2

5%S:

50%

Prot

eoba

cter

ia/

Bet

a pr

oteo

bact

eria

(%)

Bur

khol

deria

cepa

cia

(3)

R:

100%

I: 0

.0%

S:

0.0%

R:

0.0%

I: 0

.0%

S:

100%

R:

100%

I: 0

.0%

S:

0.0%

R:

0.0%

I: 0

.0%

S:

100%

R:

100%

I: 0

.0%

S:

0.0%

R:

0.0%

I: 0

.0%

S:

100%

R:

0.0%

I: 0

.0%

S:

100%

Bur

khol

deria

mal

lei

(3)

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

Nei

sser

ia a

nim

alor

is

(3)

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

Prot

eoba

cter

ia/

Gam

ma

prot

eoba

cter

ia(5

3%)

Aci

neto

bact

er lw

offii

(3)

R:

100%

I: 0

.0%

S:

0.0%

R:

0.0%

I: 0

.0%

S:

100%

R:

100%

I: 0

.0%

S:

0.0%

R:

0.0%

I: 0

.0%

S:

100%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

0.0%

I: 1

00%

S:

0.0%

Aer

omon

as h

ydro

phila

(4

)

R:

100%

I: 0

.0%

S:

0.0%

R:

50%

I: 0

%S:

50%

R:

100%

I: 0

.0%

S:

0.0%

R:

50%

I: 0

.0%

S:

50%

R:

50%

I: 0

.0%

S:

50%

R:

50%

I: 0

.0%

S:

50%

R:

0.0%

I: 5

0%S:

50%

Aer

omon

as

salm

onic

ida

(4)

R:

50%

I: 0

.0%

S:

50%

R:

50%

I: 0

.0%

S:

50%

R:

100%

I: 0

.0%

S:

0.0%

R:

50%

I: 0

.0%

S:

50%

R:

100%

I: 0

.0%

S:

0.0%

R:

50%

I: 0

.0%

S:

50%

R:

50%

I: 0

.0%

S:

50%

Aer

omon

as s

obria

(3)

R:

100%

I: 0

.0%

S:

0.0%

R:

0.0%

I: 0

.0%

S:

100%

R:

100%

I: 0

.0%

S:

0.0%

R:

0.0%

I: 0

.0%

S:

100%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

0.0%

I: 1

00%

S:

0.0%

Aer

omon

as v

eron

ii (3

)R:

100%

I: 0

.0%

S:

0.0%

R:

0.0%

I: 0

.0%

S:

100%

R:

100%

I: 0

.0%

S:

0.0%

R:

0.0%

I: 0

.0%

S:

100%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

0.0%

I: 0

.0%

S:

100%



Sample

Ord

er/

clas

s te

sted

(%

)

Bac

teri

al is

olat

es

test

ed (

n)

AM

(1

0 µ

g)

An

tib

ioti

cs

TE (30

µg

)S (3

00

µg

)TE (1

0 µ

g)

RD

(2 µ

g)

F/M

(30

0 µ

g)

OT

(30

µg

)Seawater

Citr

obac

ter

sedl

akii

(3)

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

Cro

noba

cter

du

blin

ensi

s su

bsp.

la

usan

nens

is (

3)

R:

0.0%

I: 0

.0%

S:

100%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

Ente

roba

cter

ae

roge

nes

(6)

R:

33.3

%I:

33.

3%S:

33.3

%

R:

33.3

%I:

0.0

%S:

66.6

%

R:

66.6

%I:

0.0

%S:

33.3

%

R:

66.6

%I:

0.0

%S:

33.3

%

R:

100%

I: 0

.0%

S:

0.0%

R:

33.3

%I:

0.0

%S:

66.6

%

R:

33.3

%I:

66.

6%S:

0.0%

Ente

roba

cter

clo

acae

su

bsp.

dis

solv

ens

(4)

R:

100%

I: 0

.0%

S:

0.0%

R:

0.0%

I: 1

00%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

0.0%

I: 0

.0%

S:

100%

Ente

roba

cter

clo

acae

(4

)

R:

0.0%

I: 5

0%S:

50%

R:

0.0%

I: 0

.0%

S:

100%

R:

0.0%

I: 1

00%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

0.0%

I: 0

.0%

S:

100%

Ente

roba

cter

clo

acae

co

mpl

ex (

4)

R:

100%

I: 0

.0%

S:

0.0%

R:

50%

I: 0

.0%

S:

50%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

50%

I: 5

0%S:

0%

Esch

eric

hia

coli

(30)

R:

78.5

%I:

10.

7%S:

10.7

%

R:

71.4

%I:

10.

7%S:

17.8

%

R:

92.9

%I:

3.5

%S:

3.5%

R:

89.3

%I:

0.0

%S:

10.7

%

R:

100.

0%I:

0.0

%S:

0.0%

R:

100.

0%I:

0.0

%S:

0.0%

R:

75%

I: 0

.0%

S:

25%

Kle

bsie

lla p

neum

onia

e su

bsp.

oza

enae

(3)

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

Past

eure

lla c

anis

(3)

R:

100%

I: 0

.0%

S:

0.0%

R:

0.0%

I: 0

.0%

S:

100%

R:

100%

I: 0

.0%

S:

0.0%

R:

0.0%

I: 0

.0%

S:

100%

R:

100%

I: 0

.0%

S:

0.0%

R:

0.0%

I: 0

.0%

S:

100%

R:

0.0%

I: 0

.0%

S:

100%

Prot

eus

vulg

aris

gro

up

Prot

eus

penn

eri (

3)

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

Pseu

dom

onas

ae

rugi

nosa

(13

)

R:

100%

I: 0

.0%

S:

0.0%

R:

90.9

1%I:

0.0

%S:

9.09

%

R:

100%

I: 0

.0%

S:

0.0%

R:

90.9

1%I:

0.0

%S:

9.09

%

R:

100%

I: 0

.0%

S:

0.0%

R:

90.9

1%I:

0.0

%S:

9.09

%

R:

90.9

1%I:

0.0

%S:

9.09

%

Rao

ulte

lla

orni

thin

olyt

ica

(3)

R:

100%

I: 0

.0%

S:

0.0%

R:

0.0%

I: 0

.0%

S:

100%

R:

100%

I: 0

.0%

S:

0.0%

R:

0.0%

I: 0

.0%

S:

100%

R:

100%

I: 0

.0%

S:

0.0%

R:

0.0%

I: 0

.0%

S:

100%

R:

0.0%

I: 0

.0%

S:

100%

Rao

ulte

lla y

tica

(3)

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%



Sample

Ord

er/

clas

s te

sted

(%

)B

acte

rial

isol

ates

te

sted

(n

)

An

tib

ioti

cs

AM

(10

µg

)TE (3

0 µ

g)

S (30

0 µ

g)

TE (10

µg

)R

D(2

µg

)F/

M(3

00

µg

)O

T(3

0 µ

g)

Seawater

Ser

ratia

mar

cesc

ens

(5)

R:

100%

I: 0

.0%

S:

0.0%

R:

66.6

%I:

0.0

%S:

33.4

%

R:

100%

I: 0

.0%

S:

0.0%

R:

66.6

%I:

0.0

%S:

33.4

%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

66.6

%I:

0.0

%S:

33.4

%

She

wan

ella

pu

tref

acie

ns (

13)

R:

81.8

1%I:

0.0

%S:

18.1

8%

R:

45.4

5%I:

18.

18%

S:

36.3

6%

R:

100%

I: 0

.0%

S:

0.0%

R:

72.7

2%I:

0.0

%S:

27.2

7%

R:

100%

I: 0

.0%

S:

0.0%

R:

63.6

3%I:

0.0

%S:

36.3

6%

R:

54.5

4%I:

36.

36%

S:

9.06

%

Ste

notr

opho

mon

as

mal

toph

ilia

(7)

R:

80%

I: 0

.0%

S:

20%

R:

40%

I: 0

.0%

S:

60%

R:

100%

I: 0

.0%

S:

0.0%

R:

20%

I: 5

0%S:

50%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

20%

I: 0

.0%

S:

80%

Vib

rio

vuln

ificu

s (4

)R:

50%

I: 0

.0%

S:

50%

R:

0.0%

I: 0

.0%

S:

100%

R:

100%

I: 0

.0%

S:

0.0%

R:

0.0%

I: 0

.0%

S:

100%

R:

100%

I: 0

.0%

S:

0.0%

R:

0.0%

I: 0

.0%

S:

100%

R:

0.0%

I: 5

0%S:

50%

Ente

roco

ccus

fae

cium

(3

)

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

Firm

icut

es/

Bac

illi

(9%

)

Alic

yclo

baci

llus

acid

ocal

darius

(3)

R:

0.0%

I: 1

00%

S:

0.0%

R:

0.0%

I: 0

.0%

S:

100%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

0.0%

I: 1

00%

S:

0.0%

Bac

illus

cer

eus

(7)

R:

100%

I: 0

.0%

S:

0.0%

R:

60%

I: 0

.0%

S:

40%

R:

100%

I: 0

.0%

S:

0.0%

R:

80%

I: 2

0%S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

60%

I: 2

0%S:

20%

Bac

illus

pum

ilus

(5)

R:

66.7

%I:

0.0

%S:

33.3

%

R:

66.7

%I:

0.0

%S:

33.3

%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

66.7

%I:

0.0

%S:

33.3

%

R:

66.7

%I:

0.0

%S:

33.3

%

Sta

phyl

ococ

cus

xylo

sus

(3)

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

Sta

phyl

ococ

cus

aure

us (

3)

R:

0.0%

I: 0

.0%

S:

100%

R:

0.0%

I: 0

.0%

S:

100%

R:

0.0%

I: 1

00%

S:

100%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

0.0%

I: 0

.0%

S:

0.0%

Sta

phyl

ococ

cus

war

neri (

5)

R:

33.3

%I:

0.0

%S:

66.7

%

R:

33.4

%I:

0.0

%S:

66.7

%

R:

66.7

%I:

0.0

%S:

33.3

%

R:

66.7

%I:

0.0

%S:

33.3

%

R:

33.3

%I:

66.

7%S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

33.3

%I:

0.0

%S:

66.7

%

Bac

tero

idet

es/

Flav

obac

teriia

(8%

)

Chr

yseo

bact

eriu

m

indo

loge

nes

(13)

R:

54.5

4%I:

18.

18%

S:

27.2

7%

R:

36.3

6%I:

0.0

%S:

63.6

3%

R:

100%

I: 0

.0%

S:

0.0%

R:

45.4

5%I:

0.0

%S:

54.5

4%

R:

100%

I: 0

.0%

S:

0.0%

R:

54.5

4%I:

0.0

%S:

45.4

5%

R:

45.4

5%I:

18.

18%

S:

36.3

6%

Myr

oide

s sp

p. (

5)R:

100%

I: 0

.0%

S:

0.0%

R:

66.6

%I:

0.0

%S:

33.3

%

R:

100%

I: 0

.0%

S:

0.0%

R:

66.6

%I:

0.0

%S:

33.3

%

R:

100%

I: 0

.0%

S:

0.0%

R:

66.6

%I:

0.0

%S:

33.3

%

R:

66.6

%I:

0.0

%S:

33.3

%



Sample

Ord

er/

clas

s te

sted

(%

)B

acte

rial

isol

ates

te

sted

(n

)

An

tib

ioti

cs

AM

(10

µg

)TE (3

0 µ

g)

S (30

0 µ

g)

TE (10

µg

)R

D(2

µg

)F/

M(3

00

µg

)O

T(3

0 µ

g)

Seawater

Act

inob

acte

ria/

Act

inom

ycet

ales

(3%

)

Der

mac

occu

s ni

shin

omiy

aens

is (

3)

R:

0.0%

I: 0

.0%

S:

100%

R:

0.0%

I: 0

.0%

S:

100%

R:

100%

I: 0

.0%

S:

0.0%

R:

0.0%

I: 0

.0%

S:

100%

R:

100%

I: 0

.0%

S:

0.0%

R:

0.0%

I: 0

.0%

S:

100%

R:

0.0%

I: 0

.0%

S:

100%

Koc

uria

krist

inae

(4)

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

50%

I: 0

.0%

S:

50%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

50%

I: 0

.0%

S:

50%

Koc

uria

var

ians

(3)

R:

0.0%

I: 1

00%

S:

0.0%

R:

0.0%

I: 0

.0%

S:

100%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

0.0%

I: 0

.0%

S:

100%

R:

0.0%

I: 1

000%

S:

0.0%

Mic

roco

ccus

lute

us (

4)R:

50%

I: 0

.0%

S:

50%

R:

50%

I: 0

.0%

S:

50%

R:

100%

I: 0

.0%

S:

0.0%

R:

50%

I: 0

.0%

S:

50%

R:

100%

I: 0

.0%

S:

0.0%

R:

50%

I: 0

.0%

S:

50%

R:

0.0%

I: 1

00%

S:

0.0%

Sediment

Prot

eoba

cter

ia/

Alp

ha

prot

eoba

cter

ia(7

%)

Bre

vund

imon

as

dim

inut

a (1

)

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

Sph

ingo

mon

as

pauc

imob

ilis

(1)

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

Sph

ingo

mon

as

thal

poph

ilum

(1)

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

Prot

eoba

cter

ia/

Bet

a pr

oteo

bact

eria

(7%

)

Nei

sser

ia a

nim

alor

is

(3)

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

Chr

omob

acte

rium

vi

olac

eum

(1)

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

Prot

eoba

cter

ia/

Gam

ma

prot

eoba

cter

ia(4

3%)

Aer

omon

as c

avia

e (1

)R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

Aer

omon

as s

obria

(1)

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

Pseu

dom

onas

ae

rugi

nosa

(1)

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

Ser

ratia

mar

cesc

ens

(5)

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%



Sample

Ord

er/

clas

s te

sted

(%

)B

acte

rial

isol

ates

te

sted

(n

)

An

tib

ioti

cs

AM

(1

0 µ

g)

TE (

30

µg

)S

(3

00

µg

)TE

(1

0 µ

g)

RD

(2

µg

)F/

M (

30

0 µ

g)

OT

(30

µg

)Sediment

She

wan

ella

alg

ae (

15)

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

She

wan

ella

pu

tref

acie

ns (

11)

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

Vib

rio

algi

noly

ticus

(1

4)

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

66.7

%I:

0.0

%S:

33.3

%

R:

66.7

%I:

0.0

%S:

33.3

%

Vib

rio

fluvi

alis

(11

)R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

Vib

rio

para

haem

olyt

icus

(1

3)

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

Vib

rio

vuln

ificu

s (1

1)R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

Firm

icut

es/

Bac

illi

(34%

)

Alic

yclo

baci

llus

acid

oter

rest

ris

(11)

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

Bac

illus

cer

eus

(23)

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

Bac

illus

pum

ilus

(11)

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

Lact

ococ

cus

garv

ieae

(1

3)

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

Bac

tero

idet

es/

Flav

obac

teriia

(7%

)

Chr

yseo

bact

eriu

m

indo

loge

nes

(12)

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

Myr

oide

s sp

p. (

12)

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

Act

inob

acte

ria/

Act

inom

ycet

ales

(2%

)M

icro

cocc

us ly

lae

(11)

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

R:

100%

I: 0

.0%

S:

0.0%

a Am

pici

llin

(AM

, 10

µg)

, ni

trof

uran

toin

(F/

M,

300

µg),

oxy

tetr

acyc

line

(OT,

30

µg),

sul

fona

mid

e (S

, 30

0 µg

), r

ifam

pici

n (R

D,

2 µg

), t

etra

cycl

ine

(TE,

10

µg)

and

tetr

acyc

line

(TE,

30

µg).

Res

ista

nt

(R),

inte

rmed

iate

(I)

or

susc

eptib

le (

S)



Table II Multiple Antibiotic Resistance Indexes and Resistance Ratios

Number of antibiotics to which bacteria show resistance MAR Index Resistance, % p-value

1 0.0052 0.75187 0.3635

2 0.0104 18.0451 0.6513

3 0.0022 12.7819 0.1323

4 0.0294 12.7819 0.1325

5 0.048 9.7744 0.3635

6 0.0576 14.2857 0.4084

7 0.0208 31.57894 0.1234

Ampicillin, 10 µg

Sulfonamide, 300 µg

Oxytetracycline, 30 µg

Tetracycline, 30 µg

Tetracycline, 10 µg

Rifampicin, 2 µg

Nitrofurantoin, 300 µg

0 20 40 60 80 100Resistance, %

Fig. 2. The frequency of bacteria resistant to specific antibiotics (%) in the seawater and sediment samples

Sediment Seawater

0 20 40 60Resistance, %

0 10 20 30 40 50Resistance, %

(a) (b)

Betaproteobacteria

Actinomycetales

Flavobacteriia

Bacilli

Alphaproteobacteria

Gammaproteobacteria

Actinomycetales

Flavobacteriia

Betaproteobacteria

Alphaproteobacteria

Bacilli

Gammaproteobacteria

Fig 3. The frequencies of antibiotic resistance in bacteria species from (a) seawater samples and (b) sediment samples

3%

4%

8%

14%

22%

49%

2%

7%

7%

7%

34%

43%

Actinomycetales. The frequencies of resistance to Cu2+, Zn2+, Pb2+, Cr2+ and Fe2+ were detected as an average of 58.3%, 33.8%, 32.1%, 31.0% and 25.2% respectively in 258 bacterial strains isolated from seawater samples.The frequencies of heavy metal resistance in

bacteria species isolated from the sediment samples were recorded as 100% in Αlphaproteobacteria, 100% in Betaproteobacteria, 97.5% in Flavobacteriia, 95% in Gammaproteobacteria, 72.5% in Bacilli and 66.6% in Actinomycetales. The frequencies of resistance to Cu2+, Zn2+, Pb2+, Cr2+

and Fe2+ were detected as an average of 33.3%, 30.3%, 25.5%, 35.3% and 28.4% respectively in 158 strains isolated from the sediment samples.The frequencies of heavy metal-resistant bacteria

isolated from sediment samples were higher than the frequencies of heavy metal-resistant bacteria isolated from the seawater samples. Table III shows the heavy metal resistance in bacteria isolates from seawater and sediment in Güllük Bay.The MICs of the isolates ranged from 0.004 mM

to 2.5 mM. The isolates from sediment samples obtained from stations close to fish farms showed



higher frequency of resistance against chromium, copper and zinc than other stations. The highest resistance (MIC value: 2.5 mM) was displayed against Cr+ by all isolates. Bacillus isolates showed a higher resistance to chromium, lead and copper than Pseudomonas isolates, and Vibrio isolates showed higher resistance to zinc, copper and chromium than Escherichia coli. Tolerance to the maximum MIC (>2.5 mM) for chromium was 10.1% for Bacillus and 0.8% for Pseudomonas isolates. Bacillus isolates from sediment samples showed higher resistance to chromium, lead, iron and copper than Klebsiella spp. and Escherichia coli strains from seawater samples. Similarly, Shewanella spp. and Serratia spp. strains from the sediment samples also showed higher resistance than the species mentioned above.

4. Discussion

Indicator bacteria levels reported in Güllük Bay and the presence of pathogenic bacteria (25, 26) support the relationship between the resistance data detected in the current study with bacteriological pollution levels. In the present study, bioindicator bacteria showing human-induced pollution input isolated from seawater had the highest frequency of resistance against nitrofurantoin (100%) and sulfonamide (95%). Sulfonamides were the first antibiotics developed for clinical use. Sulfonamides have been widely used to treat bacterial and protozoan infections in humans, domestic animals and fish since their introduction to clinical practice in 1935 (45–47). The results of higher resistance against sulfonamide in the present study were similar to the findings of sulfonamide resistance

in another study (48). For example, there were significant increases in numbers of bacteria resistant to oxytetracycline, oxolinic acid and florfenicol in sediments from an aquaculture site compared with those from a non-aquaculture control site. Interestingly, in another study a similar number of antibiotic-resistant bacteria were isolated from aquaculture and non-aquaculture sites (49). Gram-negative bacteria (predominantly Plesiomonas shigelloides and Aeromonas hydrophila) were isolated from aquaculture ponds in the south-eastern USA and it was reported that antibiotic resistance to tetracycline, oxytetracycline, chloramphenicol, ampicillin and nitrofurantoin were higher in antibiotic-treated ponds compared to non-treated rivers (50). It was determined that bacteria isolated from Sopot Beach, Poland, were resistant to ampicillin (51). A high percentage of bacteria were reported as resistant to streptomycin (100%), cefazolin (89.8%), ampicillin (83.7%) and trimethoprim-sulfamethoxazole (69.4%), whereas a low percentage of bacteria were resistant to cefepime (12.3%) and meropenem (14.3%) in the aquaculture region of İskenderun Bay, Turkey (52). In the current study, higher numbers of sulfonamide,

rifampicin and ampicillin-resistant bacteria were recorded in the stations around aquaculture areas than other stations. Sphingomonas paucimobilis, Escherichia coli and Enterobacter cloacae isolated from both seawater and sediment at the stations around aquaculture areas had the highest levels of antibiotic resistance. The development of resistant pathogens in aquaculture environments is well documented (53, 54) and evidence of transfer of resistance encoding plasmids between

Table III Heavy Metal Resistance in Bacteria Species from Seawater and Sediment in Güllük Bay, Turkey

Heavy metals

Samplingsides

Metal concentrations, µg ml–1 Isolates Resistant isolates

0.8 1.6 3.1 6.5 12.5 25 50 100 200 >200 n n %

Cu2+ Seawater 3 3 3 7 8 9 10 11 6 – 258 149 58.3

Sediment 7 4 9 3 6 17 17 29 11 – 158 53 33.3

Zn2+ Seawater 3 6 4 9 8 7 11 8 2 – 258 86 33.8

Sediment 4 2 2 8 4 19 36 22 6 – 158 48 30.3

Pb2+ Seawater 1 8 2 7 7 10 11 12 2 – 258 82 32.1

Sediment 1 9 4 4 3 18 13 23 17 – 158 40 25.5

Cr2+ Seawater 3 2 7 4 6 7 12 6 4 16 258 79 31.0

Sediment 3 4 5 8 4 9 11 10 27 32 158 56 35.3

Fe2+ Seawater 1 2 – – 9 24 15 6 – – 258 67 25.2

Sediment 3 13 3 5 9 15 24 29 – – 158 45 28.4

Total number of tested isolates 416



aquaculture environments and humans has been presented recently (55). It has been reported that antibiotic-resistant bacteria are present in a seafood ecosystem where antibiotics have never been used (56). This is interesting in terms of showing that aquaculture areas may be adversely affected by the presence of environmental antibiotic-resistant bacteria. In the present study, a high percentage of the

bacteria Sphingomonas paucimobilis were isolated, which was especially prevalent in Güllük Bay. The natural habitat of Sphingomonas has not been defined, but it is widely distributed in the natural environment especially in water and soil (57). The second most prevalent species were Escherichia coli and Enterobacter cloacae. Escherichia coli is an indicator of faecal contamination in aquatic environments. Enterobacter cloacae is the most frequent species associated with nosocomial infections along with Klebsiella pneumoniae that is a growing problem in human healthcare. The highest number of Bacillus cereus was isolated from the sediment underneath fish farms. A few Bacilli of marine origin have been reported to produce unusual metabolites different from those isolated from terrestrial bacteria (58). Due to the ubiquity and ability of the Bacillus species to survive under difficult circumstances, Bacillus strains are considered to be species of certain habitats (59, 60). In the current study, Bacillus pumilus, B. thuringiensis, B. mycoides and B. cereus were isolated from the sediment samples of the stations around fish farms.The high frequency of resistance among bacterial

isolates in the present study confirms the earlier reports regarding the role of antimicrobial use that plays a role in selecting antibiotic-resistant bacteria in water and aquatic sediments (46–52). Many previous studies have shown that the increases in antibiotic resistance in human medicine, agriculture and aquaculture are directly related to the amounts of antimicrobials used (61–65).Infections caused by antibiotic-resistant bacteria

are one of the most important public health concerns worldwide. Currently, MARs have been reported in a wide range of human pathogenic or opportunistic bacteria such as Vibrio sp. (66), Klebsiella pneumoniae (67), Salmonella sp. (68), Pseudomonas aeruginosa and also in pathogens (69, 70). Reservoirs of antibiotic resistance can interact between different ecological systems and potential transfer of resistant bacteria or resistant genes from animals to humans may occur through the food chain (70). In the current study, the MAR index of multiple antibiotic-resistant bacteria

was found to be 2.6 times greater in the stations around fish farm areas (0.057) than the other stations (0.022).Marine sediments offer more informative results

than seawater about environmental pollution due to the accumulation of various pollutants at the bottom of the sea, therefore analysis of sediments is widely used in tests. The association of microorganisms with sediment particles is one of the primary factors in assessing microbial fate in aquatic systems. In this study, the bacteria isolated from sediment in all samples showed a higher resistance rate than bacteria isolated from seawater. Detection of higher antibiotic resistance in sediment bacteria than bacteria isolated from seawater showed that sediment bacteria were exposed to more antibiotics. Natural ecosystems containing high concentrations of heavy metals are also frequent. Heavy metal resistance genes are commonly found in environmental bacteria (71). The resistance to seven heavy metals has been reported in the order Cu > Mn > Ni > Zn > Pb > Cd > Fe for seawater bacteria isolated from the Golden Horn, Istanbul, Turkey (17). Heavy metal resistance in bacteria found in seawater from the Mediterranean has been reported as Cd > Cu > Cr = Pb > Mn; in Karataş, Turkey Cd > Cu > Cr = Mn > Pb; and İskenderun Bay, Cu > Cd > Mn > Cr > Pb (72).In the present study, resistance to five different

heavy metals (Zn2+, Pb2+, Cu2+, Cr3+ and Fe3+) were investigated for all isolates. Trends in heavy metal resistance vary depending on the sample sites: Güllük Bay, fish farm water column: Cu > Zn > Pb > Cr > Fe; sediment: Cr > Cu> Zn > Fe > Pb. Frequency of bacteria resistance to heavy metals shows the direct effects of metal pollution. Neisseria animaloris, Aeromonas caviae and Bacillus cereus isolated from sediment samples were the most tolerant of all the heavy metal salts. Chryseobacterium indologenes displayed the highest degree of sensitivity to all metal salts while Lactococcus garvieae showed the highest degree of sensitivity to Zn2+, Pb2+, Cu2+ and Fe3+. Kocuria kristinae, Escherichia coli and Acinetobacter lwoffii, which were isolated from the seawater underneath the fish farm, displayed similar sensitivities to all tested heavy metal salts. Resistances to heavy metals for Aeromonas and Pseudomonas isolates were similar to those from İskenderun Bay, with cadmium, 35.0% and 56.5%; copper, 98.3% and 75.4%; chromium, 38.3% and 31.9%; lead, 1.7% and 7.2%; manganese, 43.3% and 44.9%; and zinc 35.0% and 41.3%, respectively (72).



Both Gram-positive and Gram-negative bacteria can resist heavy metals (73). Resistance to toxic metals in bacteria probably reflects the level of environmental contamination with these substances and it may be related to the concentration of bacteria (74). The present project found heavy metal pollution in Güllük Bay sediment samples at all stations. In the sediment samples, the heavy metal contents were reported at varying rates: between 1 μg g–1 and 209 μg g–1 for lead; 10 μg g–1 and 259 μg g–1 for zinc; 1 μg g–1 and 59 μg g–1 for copper; 0.1 μg g–1 and 46 μg g–1 for chromium; <0.01 μg g–1 and 2.8 μg g–1 for cadmium; <0.01 μg g–1 and 0.4 μg g–1 for arsenic; and 0.6% and 5.9% for aluminium, respectively. The region was defined according to cadmium, lead and zinc levels as moderately polluted. Recorded high metal values were evaluated as an indicator of domestic and industrial inputs, carried via Sarıçay Creek, port operations and tourism activities within Güllük Bay (75). In the current study, the high frequencies of heavy metal-resistant bacteria detected in the sediment samples support this data. Bacterial heavy metal resistance detected in the study may depend on many factors. A possible explanation for differences in heavy metal resistance is the proximity of Güllük Bay to iron-steel factories. Additionally, Güllük Harbour is a serious pollution source. It was reported that 2862-unit ships carried 4.8 million tonnes of ballast water to Güllük Harbour during 2007–2012 (37). Another potential source of increased resistance may be the discharge of thermal power plants located 107 km, 46 km and 39 km away from Güllük Bay. The effects of thermal power plant discharge on the accumulation of heavy metals have been reported in other studies (29, 75). The association between antibiotic resistance

and resistance to heavy metals is quite common in the same organism. The increasing numbers of antibiotic and heavy metal-resistant bacteria could be a result of gene transfer activities demonstrating that industrial pollution most likely selects for antibiotic resistance and vice versa (58). In this study, similarly, the most antibiotic-resistant bacteria such as Sphingomonas paucimobilis, Escherichia coli and Enterobacter cloacae were also resistant to heavy metals. Metal-resistant isolates from Güllük Bay also showed high resistance to sulfonamide, rifampicin and ampicillin. Bacteria from different sources such as humans, animals and soil can transfer or exchange their resistance genes. At the same time, water contaminated with antibiotics, disinfectants,

pesticides and heavy metals might encourage selection and result in antibiotic and heavy metal resistance. Marine environmental conditions are extremely dynamic compared to the terrestrial environment, allowing bacteria to bring resistance mechanisms they have developed together while being adapted to the varying conditions. This makes the isolation of various bacteria useful to assess environmental pollution and provides a pathway to possible solutions to remove pollution from marine environments. For bacteria to take part in the transformation of any heavy metal salt into a harmless form, those bacteria must firstly be resistant to the heavy metal; thus the data related to frequency of metal resistant bacteria can provide knowledge on the continual accumulation or transformation of heavy metals in the marine environment. The findings of the current study provide

data regarding the distribution of heavy metal- and antibiotic-resistant bacteria in seawater and sediment samples of Güllük Bay, Aegean Sea, Turkey. As a result, preliminary data on candidate bacteria will offer opportunities for further studies on the elimination of heavy metal contamination by the detection of heavy metal-resistant bacteria.

5. Conclusions

Analyses of the presence of antibiotic resistance in bacteria provide knowledge on pollution sources such as septic systems on regional ecosystems. Since antibiotic-resistant bacteria can affect pathogen virulence, these pollution sources can induce pathogens and can create health risks for both humans and the ecosystem. In the present study, bacteria resistant to antibiotics and heavy metals in seawater and sediment were investigated. The bacterial information obtained provides essential data for identifying the regional distribution of resistant bacteria. Levels of resistance against heavy metals and antibiotics in bacteria isolated from seawater and sediments of the Aegean Sea were quantified. Bacteria isolated from Güllük Bay sediment were resistant to all antibiotics tested and exhibited higher resistance than those isolated from seawater. The frequency of antibiotic-resistant bacteria was higher around fish farms and near the exit of Sarıçay Creek. The widespread resistances of indicator bacteria to antibiotics suggest the presence of anthropogenic influences due to domestic waste and maritime transport.



In order for bacteria to take part in the transformation of heavy metal salts into harmless forms, they must initially be resistant to heavy metals. The frequency of resistance thus provides information regarding the continual accumulation or transformation of heavy metal salts in the marine environment. The findings of the present research have shown the existing contamination status of Güllük Bay via heavy metal and antibiotic resistance tests. The study region is under pressure of pollution as stated in previous research (25, 26, 75) and the bacterial resistance data of the current study showed that there is a prevalence of resistant bacteria in the region that may be due to indirect effects of environmental dynamics and pollution. In this study, the presence of higher levels

of resistant bacteria in sediment compared to seawater may indicate the presence of microplastics in the sediment as well as the probability that the sediment is a suitable medium for accumulation of metals and antibiotics. Further studies on this subject will provide detailed data on the spread of antibiotic- and metal-resistant bacteria in marine sediments. The present study showed bacterial responses

to environmental stress and influences in terms of antibiotic and heavy metal resistance both in sediment and seawater samples at Güllük Bay, Turkey. These findings highlight the necessity of holistic assessments with a ‘one health’ approach and the need to control bacteria entering marine areas due to human activities, considering the contributions of resistant bacteria to global distribution. The data may also provide a useful resource to help identify strains of bacteria for environmental remediation applications.

Acknowledgments

The authors wish to thank the Scientific and Technical Research Council of Turkey (TÜBITAK, project number: 110Y243, 2011) and Istanbul University Scientific Research Project Unit (İÜ BAP Project/19347) for their financial support.

References

1. D. A. Rouch, B. T. O. Lee and A. P. Morby, J. Ind. Microbiol., 1995, 14, (2), 132

2. S. Congeevaram, S. Dhanarani, J. Park, M. Dexilin and K. Thamaraiselvi, J. Hazard. Mater., 2007, 146, (1–2), 270

3. M. R. Bruins, S. Kapil and F. W. Oehme, Ecotoxicol. Environ. Saf., 2000, 45, (3), 198

4. E. E. Bestawy, S. Helmy, H. Hussien, M. Fahmy and R. Amer, Appl. Water Sci., 2012, 3, (1), 181

5. H. K. Allen, J. Donato, H. H. Wang, K. A. Cloud-Hansen, J. Davies and J. Handelsman, Nat. Rev. Microbiol., 2010, 8, (4), 251

6. H. H. Wang and D. W. Schaffner, Appl. Environ. Microbiol., 2011, 77, (20), 7093

7. N. Rosenblatt-Farrell, Environ. Health Perspect., 2009, 117, (6), A244

8. K. Kümmerer, J. Antimicrob. Chemother., 2004, 54, (2), 311

9. S. Kim and D. S. Aga, J. Toxicol. Environ. Health: Part B, 2007, 10, (8), 559

10. A. J. Watkinson, G. B. Micalizzi, G. M. Graham, J. B. Bates and S. D. Costanzo, Appl. Environ. Microbiol., 2007, 73, (17), 5667

11. J. L. Caplin, G. W. Hanlon and H. D. Taylor, Environ. Microbiol., 2008, 10, (4), 885

12. L. Nonaka, K. Ikeno and S. Suzuki, Microbes Environ., 2007, 22, (4), 355

13. P. T. P. Hoa, L. Nonaka, P. Hung Viet and S. Suzuki, Sci. Total Environ., 2008, 405, (1–3), 377

14. D. I. Andersson and D. Hughes, Nat. Rev. Microbiol., 2010, 8, (4), 260

15. S. Squadrone, Environ. Monit. Assess., 2020, 192, (4), 238

16. M. L. Nadimpalli, S. J. Marks, M. C. Montealegre, R. H. Gilman, M. J. Pajuelo, M. Saito, P. Tsukayama, S. M. Njenga, J. Kiiru, J. Swarthout, M. A. Islam, T. R. Julian and A. J. Pickering, Nat. Microbiol., 2020, 5, (6), 787

17. G. Altug and N. Balkis, Environ. Monit. Assess., 2009, 149, (1–4), 61

18. M. Imran, K. R. Das and M. M. Naik, Chemosphere, 2019, 215, 846

19. P. Laganà, G. Caruso, I. Corsi, E. Bergami, V. Venuti, D. Majolino, R. La Ferla, M. Azzaro and S. Cappello, Int. J. Hyg. Environ. Health, 2019, 222, (1), 89

20. G. Lunde, Environ. Health Perspect., 1977, 19, 47

21. A. Hernandez, R. P. Mellado and J. L. Martinez, Appl. Environ. Microbiol., 1998, 64, (11), 4317

22. L. D. Rasmussen and S. J. Sørensen, Curr. Microbiol., 1998, 36, (5), 291

23. D. H. Nies, Appl. Microbiol. Biotechnol., 1999, 51, (6), 730

24. G. Altuğ, M. Çardak and P. S. Ciftci, J. Fish. Aquat. Sci., 2007, 22, (23), 39 (in Turkish)

25. G. Altuğ, M. Çardak, P. S. Ciftci, S. Gürün and S. Kalkan, ‘Bacterial Diversity in Güllük Bay’, Tübitak Project Workshop, 10th May, 2013, Güllük, Muğla, Turkey, ed. G. Altuğ, Istanbul University, Turkey, 2013, pp. 3–7 (in Turkish)



26. S. Kalkan and G. Altuğ, Mar. Pollut. Bull., 2015, 95, (1), 380

27. H. Yıldız, H. M. Doğan and Ö. Urla, Tarla Bit. Merk. Arş. Ens. Derg., 2002, 11, (1–2), 142 (in Turkish)

28. A. Demirak, A. Balci, Ö. Dalman and M. Tüfekçi, Water Air Soil Pollut., 2005, 162, (1–4), 171

29. A. Baba, A. Kaya and Y. K. Birsoy, Water Air Soil Pollut., 2003, 149, (1–4), 93

30. P. Çiftçi G. Altuğ, M. Çardak and S. Gürün, ‘Distribution of Indicator Bacteria in Recreational and Fish Farming Areas of Gulluk Bay, Turkey’, Effective Utilization of Ocean Resources and Future Maritime Industries, Tokyo, Japan, 2nd– 11th November, 2011, Tokyo University of Marine Science and Technology (TUMSAT), Japan, p. 25

31. A. Demirak, A. Balci, and M. Tüfekçi, Environ. Monit. Assess., 2006, 123, (1–3), 1

32. İ. Atılgan and Ö.Egemen, Ege J. Fish. Aquat. Sci., 2001, 18, (1–2), 225

33. G. Yucel-Gier, I. Pazi and F. Kucuksezgin, Turk. J. Fish. Aquat. Sci., 2013, 13, (4), 737

34. E. Marti, E. Variatza and J. L. Balcazar, Trends Microbiol., 2014, 22, (1), 36

35. K. P. Acharya and R. T. Wilson, Front. Med., 2019, 6, 105

36. S. A. Kraemer, A. Ramachandran and G. G. Perron, Microorganisms, 2019, 7, (6), 180

37. A. Olgun, ‘Evaluation in Terms of Vessels and Ballast Waters entering Güllük Bay’, Bacteriology of Güllük Bay, Tübitak Project Workshop, 10th May, 2013, Güllük, Muğla, Turkey, ed. G. Altuğ, Istanbul University, Turkey, pp. 33–37 (in Turkish)

38. “Standard Methods for the Examination of Water and Waste Water”, eds. E. W. Rice, R. B. Baird, A. D. Eaton and L. S. Clesceri, 22nd Edn., American Public Health Association, American Water Works Association, Water Environment Federation, 2012

39. D. H. Pincus, ‘Microbial Identification Using the bioMérıeux VITEK® 2 System’, in “Encyclopedia of Rapid Microbiological Methods”, ed. M. J. Miller, Vol. 2, Parenteral Drug Association, Bethesda, USA, 2006, pp. 1–32

40. P. A. Sobecky, T. J. Mincer, M. C. Chang, A. Toukdarian and D. R. Helinski, Appl. Environ. Microbiol., 1998, 64, (8), 2822

41. M. A. Wikler, F. R. Cockerill, W. A. Craig, M. N. Dudley, G. M. Eliopoulos, D. W. Hecht, J. F. Hindler, D. E. Low, D. J. Sheehan, F. C. Tenover, J. D. Turnidge, M. P. Weinstein, B. L. Zimmer, M. J. Ferraro and J. M. Swenson, ‘Performance Standards for Antimicrobial Susceptibility Testing: Sixteenth Informational Supplement’, M100– S16, Clinical and Laboratory Standards Institute, Wayne, USA, 2006, 192 pp

42. P. H. Krumperman, Appl. Environ. Microbiol., 1983, 46, (1), 165

43. J. A. Washington and V. L. Sutter, ‘Dilution Test Procedures’, in “Manual of Clinical Microbiology”, eds. E. H. Lennette A. Balows W. J. Hausler and J. P. Truant, American Society for Microbiology, Washington, DC, USA, 1981, pp 549–555

44. A. D. Geiselbrecht, R. P. Herwig, J. W. Deming and J. T. Staley, Appl. Environ. Microbiol., 1996, 62, (9), 3344

45. V. Perreten and P. Boerlin, Antimicrob. Agents Chemother., 2003, 47, (3), 1169

46. T. X. Le, Y. Munekage and S. Kato, Sci. Total Environ., 2005, 349, (1–3), 95

47. M. T. Blahna, C. A. Zalewski, J. Reuer, G. Kahlmeter, B. Foxman and C. F. Marrs, J. Antimicrob. Chemother., 2006, 57, (4), 666

48. Y. Agersø and A. Petersen, J. Antimicrob. Chemother., 2007, 59, (1), 23

49. A. H. Buschmann, A. Tomova, A. López, M. A. Maldonado, L. A. Henriquez, L. Ivanova, F. Moy, H. P. Godfrey and F. C. Cabello, PLoS One, 2012, 7, (8), e42724

50. J. Kerry, R. Coyne, D. Gilroy, M. Hiney and P. Smith, Aquaculture, 1996, 145, (1–4), 31

51. F. Matyar, A. Kaya and S. Dinçer, Sci. Total Environ., 2008, 407, (1), 279

52. Z. J. Mudryk, Mar. Pollut. Bull., 2005, 50, (1), 80

53. H. Sørum, Acta Vet. Scand. Suppl., 1999, 92, 29

54. V. Inglis, ‘Antibacterial Chemotherapy in Aquaculture: Review of Practice, Associated Risks and Need for Action’, Use of Chemicals in Aquaculture in Asia, Tigbauan, Philippines, 20th–22nd May, 1996, “Proceedings of the Meeting on the Use of Chemicals in Aquaculture in Asia”, eds. J. R. Arthur, C. R. Larilla-Pitogo and R. P. Subasinghe, Southeast Asian Fisheries Development Center Aquaculture Department, Tigbauan, Philippines, 2000, pp. 7–22

55. G. Rhodes, G. Huys, J. Swings, P. McGann, M. Hiney, P. Smith and R. W. Pickup, Appl. Environ. Microbiol., 2000, 66, (9), 3883

56. Y. Huang, L. Zhang, L. Tiu and H. H. Wang, Front. Microbiol., 2015, 6, 914

57. P.-R. Hsueh, L.-J. Teng, P.-C. Yang, Y.-C. Chen, H.-J. Pan, S.-W. Ho and K.-T. Luh, Clin. Infect. Dis., 1998, 26, (3), 676

58. C. Baker-Austin, M. S. Wright, R. Stepanauskas and J. V. McArthur, Trends Microbiol., 2006, 14, (4), 176

59. P. R. Jensen and W. Fenical, Annu. Rev. Microbiol., 1994, 48, 559

60. D. Claus and R. C. W. Berkeley, ‘Genus Bacillus, Cohn 1872’, in “Bergey’s Manual of Systematic



The Authors

Gülşen Altuğ is a professor in the Department of Marine Biology of the Faculty of Aquatic Science at Istanbul University, Turkey. Her research focuses on marine bacteriology, including bacterial diversity and micro-geographical variations, clinical, industrial and ecological uses of marine isolates, bacterial pollution, epibiotic bacterial communities and anti-bacterial characteristics, bacterial remediation (oil degrading capacity of marine isolates) and resistant bacterial isolates against heavy metals and antibiotics. She is also the inventing founder of the biotechnology company named BIYOTEK15 R&D Training and Consulting Industry and Trade Ltd Company in Entertech of Istanbul University Technocity.

Mine Çardak is an associate professor at Çanakkale Onsekiz Mart University, School of Çanakkale Applied Sciences, Department of Fisheries Technology, Turkey. Her researches focus on marine bacteriology, bacterial resistance against heavy metals and antibiotics, bacterial pollution and biotechnology. She has worked as a scientist since 2000.

Pelin Saliha Çiftçi Türetken is a researcher at İstanbul University, Faculty of Aquatic Sciences, Department of Marine Biology. Her research focus on marine bacteriology, bacterial remediation, bacterial resistance and biotechnology. She has a PhD degree in marine biology. She has worked as an academic at university since 2005.

Bacteriology” eds. P. H. A. Sneath , N. S. Mair, M. E. Sharpe and J. G. Holt, Vol. 2, Williams and Wilkins Co, Baltimore, USA, 1986, pp 1105–1139

61. S. B. Levy and B. Marshall, Nat. Med., 2004, 10, (12), S122

62. D. G. Capone, D. P. Weston, V. Miller and C. Shoemaker, Aquaculture, 1996, 145, (1–4), 55

63. F. M. Aarestrup, Basic Clin. Pharmacol. Toxicol., 2005, 96, (4), 271

64. N. van de Sande-Bruinsma, H. Grundmann, D. Verloo, E. Tiemersma, J. Monen, H. Goossens and M. Ferech, Emerg. Infect. Dis., 2008, 14, (11), 1722

65. E. Gullberg, S. Cao, O. G. Berg, C. Ilbäck, L. Sandegren, D. Hughes and D. I. Andersson, PLoS Pathog., 2011, 7, (7), e1002158

66. C. P. Randall, A. Gupta, N. Jackson, D. Busse and A. J. O’Neill, J. Antimicrob. Chemother., 2015, 70, (4), 1037

67. L. A. M. Carneiro, A. P. S. Silva, V. L. C. Merquior and M. L. P. Queiroz, FEMS Microbiol. Lett., 2003, 228, (2), 175

68. L. P. Randall, S. W. Cooles, M. K. Osborn, L. J. V. Piddock and M. J. Woodward, J. Antimicrob. Chemother., 2004, 53, (2), 208

69. A. S. Schmidt, M. S. Bruun, I. Dalsgaard and J. L. Larsen, Appl. Environ. Microbiol., 2001, 67, (12), 5675

70. M. Teuber, Curr. Opin. Microbiol., 2001, 4, (5), 493

71. S. Silver and L. T. Phung, Annu. Rev. Microbiol., 1996, 50, 753

72. F. Matyar, T. Akkan, Y. Uçak and B. Eraslan, Environ. Monit. Assess., 2010, 167, (1–4), 309

73. S. Silver and M. Walderhaug, Microbiol. Rev., 1992, 56, (1), 195

74. H. Aiking, A. Stijnman, C. van Garderen, H. van Heerikhuizen and J. van ’t Riet, Appl. Environ. Microbiol., 1984, 47, (2), 374

75. N. Çağlar, A. Aksu and G. Altuğ, Geol. Bull. Turkey, 2020, 63, (1), 117



Samet Kalkan has a PhD degree from Istanbul University, Institute of Graduate Studies in Science and Engineering, Department of Marine Biology. He currently works as a doctor scientist at Recep Tayyip Erdogan University- Faculty of Fisheries, Department of Marine Biology, Turkey. He has worked as academic at university since 2010. His main researches focus on marine bacteria, bacterial diversity, bacterial pollution, resistant bacteria against heavy metals-antibiotics, also marine biotechnology. He has scientific abroad experiences in Italy and Portugal.

Sevan Gürün graduated with a degree in Biology from Istanbul University. He has a PhD degree from Istanbul University, Institute of Graduate Studies in Science and Engineering, Department of Marine Biology. He worked as a researcher in various scientific projects. He has been working as a researcher in a private company since 2016. His expertise focuses on bacterial diversity, marine bacteria, bacterial pollution, bacterial biotechnology, resistant bacteria against heavy metals and antibiotics.




Reviewed by Martin HayesJohnson Matthey, 28 Cambridge Science Park, Milton Road, Cambridge, CB4 0FP, UK

Email: [email protected]

Introduction

This book is a fascinating account of how nanoparticles and nanotechnology are increasingly employed in a diverse array of applications ranging from plant growth to food packaging, biosensing, enzyme immobilisation and more. The book is divided into 20 chapters, each dealing with a specific application of nanotechnology and written by a different group of eminent academics from Indian universities and research institutes. Each chapter is a review of its own topic area and the

literature citations at the end of each chapter make it easy for the reader to use this book as a reference volume from which further in-depth reading can be pursued by following the cited literature.

Safety of Nanoparticles in Plants and Packaging

Chapter 1 deals with the phytotoxicity of nanoparticles in plants which can lead to both positive and negative outcomes. For example, improvement in germination rate and growth have been reported in seeds of rice exposed to carbon nanotubes; on the other hand, toxicity has also been widely reported, including studies

on aluminium oxide and zinc oxide nanoparticles hindering root growth rate. Chapter 2 considers an entirely separate but

equally important area of use of nanomaterials in food packaging. It concludes that “it is essential to perform safety assessment of nanomaterials before their application in food packaging or processing” and provide a citation on how to do this using a “decision tree”.

Biosensors from Nanobiotechnology

Chapter 7 introduces how biosensors derived from nanobiotechnology can be used to monitor the environment and gain information relating to its health and the detrimental effects that modernisation and industrialisation have had on the planet. Biosensors need to be specific, rapid, sensitive and cost-effective. The advent of nanotechnology and biosensors has made this possible and the authors of this chapter (Gupta and Kakkar) explore the different types of biosensors that have been developed over recent years. The authors give a brief explanation of how different types of sensors work using a combination of bio-recognition components and different transduction principles. Types include: (a) immunosensors; (b) enzymatic biosensors; (c) whole-cell based sensors; (d) biosensors; (e) genosensors; (f) aptasensors and (g) biomimetic biosensors. The role of the transducer is to convert the biochemical response into an analysable and measurable signal. The outputs can be electrochemical, optical, piezoelectric, thermometric or magnetic.

“Nanomaterials and Environmental Biotechnology”Edited by Indu Bhushan (Shri Mata Vaishno Devi University, India), Vivek Kumar Singh (Shri Mata Vaishno Devi University, India), Durgesh Kumar Tripathi (Amity University, India), Nanotechnology in the Life Sciences Series, Springer Nature Switzerland AG, Cham, Switzerland, 2020, 434 pp, ISBN: 978-3-030-34543-3, £129.99, €155.99, US$179.99



Enzyme Immobilisation

Chapter 10 tackles the interesting area of enzyme immobilisation and the use of chitosan nanoparticles therein. The biopolymer’s distinct physicochemical properties have been described to offer an excellent microenvironment for enzyme immobilisation through adsorption, covalent binding or cross-linking, to achieve desirable enzymatic activity and stability. On the other hand, nanoparticles as materials of enhanced properties, owing to their high surface to volume ratio, have been introduced as attractive candidates for enzyme immobilisation. The chapter briefly discussed various methods for the preparation of chitosan nanoparticles for enzyme immobilisation including reverse micelle, coprecipitation, ionotropic gelation and ionic or emulsion cross-linking methods. Different methods for enzyme immobilisation such as support binding, cross-linking and entrapment, as well as different materials used as supports have been explained too. This section is then closed by presenting some examples for immobilisation of different enzyme families (for example, α-amylase, β-galactosidase, cellulase, laccase, lipase or protease) through applying chitosan nanoparticles.

Solid Lipid Nanoparticles

Chapter 13 offers an overview of solid lipid nanoparticles (SLN) as pharmaceuticals delivery systems whereas Chapter 19 gives a review of the most commonly used nanocarriers for drug delivery systems, with a focus on vesicular, polymeric and inorganic carriers.SLN are lipid-based formulations, containing

typically non-toxic biodegradable polymers forming a solid hydrophobic core suspended in an aqueous phase, the whole structure being stabilised by surfactants. The therapeutic agent is dissolved or dispersed in the solid lipid core, the SLN being suitable for incorporation and delivery of both hydrophilic and hydrophobic drugs. SLN present significant advantages over conventional drug delivery systems, including but not limited to biocompatibility and bioavailability, reduced drug leakage and increased physical stability of the drugs. In addition, they have been used successfully in various drug delivery techniques.Novel applications of SLN as drug carriers are

described in the field of gene therapy, peptide drug delivery and vaccines. SLN production methods use low mechanical force, allowing successful incorporation and delivery of nucleic acids in gene therapy. Overall, SLN are promising alternatives

to traditional drug delivery systems, offering multiple advantages in terms of drug delivery and bioavailability, as well as being economically efficient and easy to produce on scale.

FDA-Approved Nanomedicines

An extensive summary of FDA-approved nanomedicines is included in Table 19.1 (Chapter 19), which also summarises the advantages of these specific formulations. The main types of nanocarriers described in Chapter 19 are vesicular carriers (liposomes and niosomes), polymeric nanoparticles and inorganic carriers (silica, gold and calcium nanoparticles). Liposomes and solid lipid nanoparticles (see also Chapter 13) are suitable for the delivery of drugs by any route, either oral or parenteral and can be used with both hydrophilic and lipophilic drugs. Their main advantages reside in protecting labile drugs, limited toxicity and a sustainable targeted release of the drug.Inorganic nanocarriers exhibit higher stability and

resistance to microbial growth, while having a low toxicity and allowing facile surface modifications. Mesoporous silica nanoparticles allow encapsulation of the therapeutic agent and targeted delivery to tumour cells in cancer therapy. Gold nanoparticles are biocompatible and bio-inert and have been successfully used in covalent conjugation with protein antigens in developing vaccines for cancer immunotherapy. Calcium phosphate nanoparticles are excellent candidates for developing ceramic-based carriers for peptide drugs prone to degradation, such as insulin.

Summary

In conclusion, I consider this book to be a positive contribution to the biotechnology literature, although I do not recommend reading this book sequentially from Page 1 as the variety of topics introduced is too great and each individual topic is not explored in depth. It is best used (and deserves recommendation) as a reference source from which each chapter can be used as the starting point to a more in-depth study or review of a particular topic. There are some negative aspects of the presentation of this work which do, unfortunately, detract from its enjoyment. These are exemplified in the poor quality of the diagrams, the grammatical errors and the somewhat odd references of Chapter 7.Overall, though, this book is a positive addition to

the biotechnology reference bookshelf.



"Nanomaterials and Environmental Biotechnology"

The Reviewer

Martin Hayes is Biotechnology Lead at Johnson Matthey based in Cambridge, UK. He has worked with Johnson Matthey since 1997 and has held multiple research, development and customer-facing technical roles across the company. He holds a PhD in heterogeneous catalysis and is interested in the application of biology as a technology to realise circular chemical processing and accelerate the transition to “Net Zero”.




Samantha Staniland, Tommaso Angelini, Ahir Pushpanath, Amin Bornadel, Elina Siirola, Serena Bisagni, Antonio Zanotti-Gerosa, Beatriz Domínguez*Johnson Matthey, 260 Cambridge Science Park, Milton Road, Cambridge, CB4 0WE, UK


The asymmetric reduction of C=C double bonds is a sought-after chemical transformation to obtain chiral molecules used in the synthesis of fine chemicals. Biocatalytic C=C double bond reduction is a particularly interesting transformation complementary to more established chemocatalytic methods. The enzymes capable of catalysing this reaction are called ene-reductases (ENEs). For the reaction to take place, ENEs need an electron withdrawing group (EWG) in conjugation with the double bond. Especially favourable EWGs are carbonyls and nitro groups; other EWGs, such as carboxylic acids, esters or nitriles, often give poor results. In this work, a substrate engineering strategy is proposed whereby a simple transformation of the carboxylic acid into a fluorinated ester or a cyclic imide allows to increase the ability of ENEs to reduce the conjugated double bond. Up to complete conversion of the substrates tested was observed with enzymes ENE-105 and *ENE-69.

1. Introduction

The use of enzymes for the asymmetric reduction of activated C=C double bonds can be a viable and straightforward alternative to asymmetric

hydrogenation. Traditionally, whole cell microorganisms were used for this purpose but a recent increase in the number of isolated and characterised ENEs means that recombinantly-expressed enzyme preparations are now generally favoured over whole cells, as a number of recent publications demonstrate (1–10).Double bond ‘activation’ to facilitate ENEs mediated

reduction can be achieved in many cases by alpha substituted functional groups including aldehydes, ketones or nitro moieties. Carboxylate derivatives (such as esters, lactones and anhydrides) can also act as activating groups but their ability to sufficiently activate the C=C bond in the absence of other groups is less evident (11, 12). The traditional approach in these cases is to turn to chemocatalytic hydrogenation (see (13–15) for reviews focused on industrial applications). Herein we describe a new approach to activate α,β-unsaturated carboxylic acids for the reduction with ENEs using a substrate engineering approach.

2. Experimental

2.1 General

All reagents and solvents were purchased from Sigma-Aldrich and Alfa Aesar, Thermo Fisher Scientific. They were of the highest available purity and were used without further purification. 1H nuclear magnetic resonance (NMR) spectra were recorded using a Bruker 400 MHz Avance III HD equipped with SMART probe (Bruker Corporation, USA) where spectra are referenced to deuterated chloroform (CDCl3) 7.26 ppm, shifts are recorded in parts per million and J values in hertz. The NMR results can be found in the Supplementary Information.

Biocatalytic Reduction of Activated Cinnamic Acid DerivativesAsymmetric reduction of C=C double bonds using Johnson Matthey enzymes



2.2 Enzyme Preparations

Genes coding for Johnson Matthey, ENEs (ENE-101, ENE-102, ENE-103, ENE-104, ENE-105, *ENE-69 and GDH-101) were ordered codon-optimised from GeneArt (Thermo Fisher Scientific) and cloned into T5 vector pJEx401 (ATUM). Enzymes were expressed recombinantly in Escherichia coli BL21 in both shake flasks and fed batch fermentations, whereby induction was carried out with isopropyl β-D thiogalactopyranoside (IPTG) at 30°C. Harvested biomass was resuspended in 100 mM potassium phosphate buffer (pH 7) and cells were broken up either by sonication or homogenisation. The so-obtained cell lysate was clarified by centrifugation and filtrated prior to lyophilisation. Protein expression was assessed by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and chromatographic activity assays.Enzymes ERED-103, ERED-110, ERED-112,

ERED-207, ERED-P1-A04, ERED-P1-E04 and ERED-P1-H09 were purchased from Codexis.

2.3 2,2,2-Trifluoroethyl Cinnamate (3a) and 3-Phenyl-Acrylic Acid 2,2,2-Trifluoro-1-Trifluoromethyl-Ethyl Ester (5a)Cinnamic acid 1a (5 g, 33.75 mmol) and oxalyl chloride (2.85 ml, 33.75 mmol) in dichloromethane (5 ml) were stirred at 25°C for 2 h before adding the fluorinated alcohol-trifluoro ethanol for 3a (2.47 ml, 33.75 mmol) and 1,1,1,3,3,3-hexafluoropropan-2-ol for 5a (3.50 ml, 33.75 mmol). The reaction was then stirred at room temperature overnight before being quenched by addition of saturated aqueous NaHCO3 (20 ml) and extracted with dichloromethane (2 × 20 ml), dried over MgSO4, filtered and concentrated under reduced pressure to afford the corresponding fluorinated esters 3a and 5a in quantitative yield.

2.4 3-Phenyl-Acrylic Acid 2,2,3,3,4,4,4-Heptafluoro-Butyl Ester (6a) and (Perfluorophenyl)Methyl Cinnamate (7a)Cinnamoyl chloride (0.75 g, 4.50 mmol) and the corresponding fluorinated alcohols – 2,2,3,3,4,4,4-heptafluorobutan-1-ol for 6a (0.98 g, 4.50 mmol) and pentafluroro benzyl alcohol for 7a (0.89 g, 4.50 mmol) – in dichloromethane (2.5 ml) were stirred at room temperature overnight. The reaction

was then quenched by addition of saturated aqueous NaHCO3 (20 ml) and extracted with dichloromethane (2 × 20 ml), dried over MgSO4, filtered and concentrated under reduced pressure to afford the corresponding fluorinated esters 6a and 7a in 95% to 99% yield.

2.5 1-Cinnamoylpyrrolidin-2-one (9a)

Cinnamoyl chloride (5 g, 30.01 mmol), pyrrolidinone (2.3 ml, 36.01 mmol) and triethylamine (13 ml, 90.03 mmol) in dichloromethane (50 ml) were stirred at room temperature overnight. The reaction was quenched by addition of water (20 ml), the organic layer was separated and washed with saturated aqueous NaCl (20 ml), dried over MgSO4, filtered and concentrated under reduced pressure to afford 9a in 81% yield.

2.6 3-Cinnamoyloxazolidin-2-one (8a)

Cinnamic acid 1a (5 g, 33.56 mmol) and oxalyl chloride (2.85 ml, 33.56 mmol) in dichloromethane (5 ml) were stirred at room temperature overnight before removing the solvent under reduced pressure. The reaction crude was dissolved in anhydrous tetrahydrofuran (THF) (20 ml) and n-butyllithium (1.6 M in hexane, 21 ml, 33.56 mmol, one equivalent) was added dropwise over 30 min. The cinnamoyl chloride solution was then added dropwise to a solution of oxazolidinone (2.92 g, 33.56 mmol) in anhydrous THF (100 ml) at 0°C before stirring at room temperature overnight. The reaction was quenched with water (50 ml), extracted with ethyl acetate (EtOAc) (2 × 100 ml), washed with saturated aqueous NaHCO3 (20 ml) and saturated aqueous NaCl (20 ml). The solvent was removed under reduced pressure and the solid was recrystallised from a 1:1 mixture EtOAc:heptane (20 ml). The solid was filtered and washed with hexane (10 ml) to give crystals of 8a in 80% yield.

2.7 (E)-1-(2-Methyl-3-Phenylacryloyl)Pyrrolidin-2-one (10a) and (E)-1-(2,3-Diphenylacryloyl)Pyrrolidin-2-one (11a)

(E)-2-methyl-3-phenylacrylic acid (5 g, 30.86 mmol) was converted to the corresponding



acid chloride by addition of oxalyl chloride (1.4 ml, 30.86 mmol) in dichloromethane (5 ml). The reaction was stirred at room temperature for 3 h. Pyrrolidinone (2.82 ml, 37.03 mmol) and triethylamine (13 ml, 92.58 mmol) were added before stirring the reaction overnight. The reaction was quenched by addition of water (20 ml) and saturated aqueous NaCl (20 ml). The solvent was removed under reduced pressure and the solid was dissolved in EtOAc and treated with activated charcoal (1 g), filtered through Celite® and concentrated. The solid was recrystallised from heptane (10 ml) to give 10a in 55% yield.Following an identical procedure, 11a was

synthesised in 53% yield from (E)-2,3-diphenylacrylic acid (10 g, 44.64 mmol).

2.8 Small Scale Screening Reactions

Substrates 1a–9a (0.025 mmol) and enzymes ENE-101, ENE-102, ENE-103, ENE-104, ENE-105 or *ENE-69 (2.5 mg), were added to reaction vials containing 500 µl of aqueous media at pH 7 (250 mM potassium phosphate buffer pH 7, 1.1 mM NAD(P)+, 100 mM D-glucose, 10 U ml–1 GDH-101) to give a final concentration of substrate of 50 mM. The vials were shaken at 400 rpm, 30°C for 18 h. For high-performance liquid chromatography (HPLC) analysis, the reactions were quenched with acetonitrile (MeCN) (1 ml), vortexed, centrifuged and aliquoted. For gas chromatography (GC) analysis, samples were extracted with EtOAc (2 × 0.5 ml), dried over MgSO4 and analysed directly. For NMR analysis, the reactions were extracted with CDCl3 and analysed directly.

2.9 Preparative Scale Screening Reactions

Reactions were scaled up using three-neck round bottom flask equipped with stir bar and pH titrator (10 M NaOH). To the flask was weighed 100–500 mg substrate (40–100 mM final concentration) and 5 mg ml–1 enzyme which was suspended in aqueous media at pH 7 (250 mM potassium phosphate buffer pH 7, 1.1 mM NAD(P)+, 100–200 mM D-glucose (two equivalent), 10 U ml–1 GDH-101) the reactions were stirred at 30°C, 400 rpm for 18 h.

2.10 Analytical Methods

HPLC analysis of conversion was conducted on an 1260 Infinity II LC system (Agilent, USA) using a

C18 SunFire Column (Waters Corporation, USA, 150 × 4.6 mm, 3.5 µm) with an isocratic method (MeCN:Water, 30:70 + 0.1% trifluoroacetic acid) and a flow rate of 1 ml min–1. Chiral HPLC analysis was performed on a Varian

ProStar series (Agilent) with a CHIRALCEL® OD-H column (Chiral Technologies, USA, 250 × 4.6 mm, 5 µm) with an isocratic method A (heptane:isopropyl alcohol (IPA), 88:12) and a flow rate of 1 ml min–1 or isocratic method B (heptane:IPA, 98:2). GC analysis of conversion was performed on

a Varian CP-3800 (Agilent) using γ-DEX™ 225 capillary column (Sigma-Aldrich, 30 m × 0.25 mm × 0.25 μm) and using helium as carrier gas. Percentage conversion was measured by integration of the product peak in the GC (uncorrected area under curve (AUC)), values below 100% indicate that unreacted starting material was detected. No side products were detected in any of the reported reactions. GC program parameters: injector 250°C, flame ionization detector (FID) 250°C, 80°C for 3 min then 5°C min–1 up to 160°C, hold 1 min (total time 20 min), constant flow 5 ml min–1.


It has been found that a particular ENE in Johnson Matthey’s collection, a homologue from the tobacco ENE reductase fold (16), ENE-105, was capable of reducing methyl ester 2a (Figure 1), albeit in a very low yield of 3% (Entry 2, Table I). By comparison, cinnamic acid 1a was a poor substrate and showed no conversion to the reduced product 1b at pH 7.0 (Entry 1, Table I). The pKa of cinnamic acid 1a is 4.4 and therefore, at pH 7.0, the carboxylic acid should be deprotonated affecting its ability to bind to the enzyme active site. This observation is in line with other literature examples where carboxylates were found to be poor activating groups (17). Encouraged by this initial result, we turned our efforts towards the use of more activated esters. It was envisaged that converting the alkyl chain in the ester moiety to a more EWG could lead to an increase in double bond activation. A similar approach has been reported previously by BASF SE for the lipase-catalysed kinetic resolution of racemic amines and alcohols, where the choice of acylating agent proved critical (18). We chose trifluoroethyl ester 3a as a starting point which was reduced by ENE-105 and *ENE-69 in 6% and 12% conversion respectively (Entry 3, Table I) suggesting that the addition of an EWG had a positive activating-effect on the reduction. To consolidate this theory, ethyl ester 4a was tested



with the novel ENEs; only a trace of reduction was observed <0.5% (Entry 4, Table I).Other EWGs such as hexafluoroethyl in

compound 5a, heptafluorobutyl in 6a and pentafluorobenzyl in 7a could also activate the double bond in the same way, so 5a, 6a and 7a were prepared by reacting cinnamoyl chloride with the corresponding fluorinated alcohols and these substrates were subsequently tested with the ENEs. Hexafluoro 5a was not reduced by ENE-105 or *ENE-69 (Entry 5, Table I), instead, a significant amount of hydrolysis product (cinnamic acid 1a, 10%) was observed. Heptafluorobutyl 6a and pentafluoro 7a were poor activating groups with 6a showing only a trace amount of product 6b (Entry 6, Table I) and 7a giving no conversion (Entry 7, Table I). With only limited success with the fluorinated

activating groups, our efforts turned towards

cyclic imides since activated substrates 8a and 9a have been shown to be highly activated towards Michael addition reactions (19, 20, 21) (Figure 2). Compounds 8a and 9a were synthesised and tested with enzymes ENE-105 and *ENE-69. Pleasingly, oxazolidinone 8a was successfully reduced by both ENEs (51% and 39% conversion to 8b, Entry 1, Table II) and pyrrolidinone 9a was reduced to 9b in >95% conversion (Entry 2, Table II), proving to be an excellent activating group. The 1H NMR shift of the alkene proton alpha to the carbonyl for pyrrolidinone 9a is shifted down field (7.92 ppm) compared to cinnamic acid 1a (6.46 ppm), therefore supporting the electron-withdrawing nature of the activating group. The enzymes were then tested for their ability to

reduce α-substituted cinnamic acid derivatives such as α-methyl 10a and α-phenyl 11a (Figure 3). Encouragingly, the tri-substituted double bond in 10a was reduced to 10b in >95% conversion by 1H NMR analysis (Entry 2, Table III). However, bulkier substrate 11a, was not tolerated so well on an analytical scale due to solubility issues causing mass-transfer limitations (Entry 3, Table III). The reaction was repeated on a larger scale with stirring (Entry 4, Table III) and >95% conversion

1b R = OH2b R = OCH33b R = OCH2CF34b R = OCH2CH35b R = OCH(CF3)26b R = OCH2CF2CF2CF37b R = OCH2C6F5

[a]O

R

Fig. 1. Reduction of cinnamic acid and cinnamoyl esters. [a] = 1–7a (50 mM concentration), ENE-105 or ENE-69 (5 mg ml–1), 500 µl buffer (250 mM KPi, pH 7, 1.1 mM NAD(P)+, 100 mM D-glucose, 10 U ml–1 GDH-101), 400 rpm, 30°C, 18 h

Table I Reduction of Cinnamoyl Esters at 50 mM Substrate Concentration, pH 7, 30°C, 18 h

Conversion, %a Entry Substrate ENE-105 *ENE-691b 1a 0 0

2 2a 3 1

3 3a 6 12

4 4a <0.5 <0.5

5c 5a 0 0

6d 6a <0.5 <0.5

7d 7a 0 0aIntegration of the product peak in the GC (uncorrected AUC), values below 100% indicate that unreacted starting material was detected; no side products were detected for these reactionsbIntegration of the product peak in the HPLC (achiral method, uncorrected AUC), values below 100% indicate that unreacted starting material was detected; no side products were detected for these reactionsc10% cinnamic acid observedd Conversion calculated by 1H NMR

Table II Reduction of Cinnamoyl Cyclic Imide Derivatives at 50 mM Substrate Concentration, pH 7, 30°C, 18 h

Conversion, %Entry Substrate ENE-105 *ENE-691a 8a 51 39

2b 9a >95 >95aIntegration of the product peak in the HPLC (chiral method, uncorrected AUC), values below 100% indicate that unreacted starting material was detected; no side products were detected for these reactionsbConversion calculated by 1H NMR

1a R = OH2a R = OCH33a R = OCH2CF34a R = OCH2CH35a R = OCH(CF3)26a R = OCH2CF2CF2CF37a R = OCH2C6F5

O

R



was achieved. 10b and 11b were obtained as racemic mixtures.With a successful activating group found, the

reaction was repeated on a preparative scale to test reproducibility and scalability (Table IV). Pyrrolidinone 9a was successfully reduced using enzyme ENE-105 at 130 mg scale with the desired product 9b being obtained in 95% conversion by 1H NMR (Entry 1, Table IV). 72% conversion to 10b was achieved after 20 h (Entry 3, Table IV) on the reduction of pyrrolidinone 10a at 500 mg scale.Having found enzymes in Johnson Matthey’s

collection that could successfully reduce masked

carboxylic acids, other commercially available enzymes were tested as a comparison on the reduction of 10a (Table V). Six enzymes from Johnson Matthey collection (Entries 3 to 6, Table V) and seven enzymes purchased from Codexis (Entries 7 to 13, Table V) were compared with ENE-105 and ENE-69* (Entries 1 and 2, Table V). It was found that, despite the extra activation of the C=C double bond, none of the tested enzymes could reduce cinnamic acid derivative 10a, highlighting the unique ability of ENE-105 and *ENE-69 within the focused library (13 enzymes) screened. In summary, we have shown that cinnamic acid

derivatives activated as fluorinated esters or as cyclic imides can be reduced using Johnson Matthey enzymes ENE-105 or *ENE-69. The concept of ‘substrate engineering’ as opposed to ‘enzyme engineering’, offers a complimentary and faster approach to developing a bioprocess, making difficult transformations possible. The reduced products can be subsequently converted to the parent carboxylic acids by LiOH hydrolysis (22, 23) and the potential re-use of these activating groups will be investigated in the future. It is envisaged that the work will lead to further examples of activated acids or esters being reduced by ENEs.

Fig. 2. Cinnamoyl cyclic imide derivatives. [a] = 8a–9a (50 mM concentration), ENE-105 or ENE-69 (5 mg ml–1), 500 µl buffer (250 mM KPi, pH 7, 1.1 mM NAD(P)+, 100 mM D-glucose, 10 U ml–1 GDH-101), 400 rpm, 30°C, 18 h

O

R

O

R[a]

O O O O

OON N N NR = R =

8a 9a 8b 9b

Fig. 3. Reduction of α-substituted cinnamoyl pyrrolidinones. [a] = 9a–11a (40–100 mM concentration), ENE-105 or ENE-69 (5 mg ml–1), buffer (250 mM KPi, pH 7, 1.1 mM NAD(P)+, two equivalent D-glucose, 10 U ml–1 GDH-101), 400 rpm, 30°C, 18 h

[a]O O O O

N NR R

9a R = H10a R = Me11a R = Ph

9b R = H10b R = Me11b R = Ph

Table III Reduction of α-Substituted Cinnamoyl Pyrrolidinones at 50 mM Substrate Concentration, pH 7, 30°C, 18 h

Conversion, %a

Entry Substrate ENE-105 *ENE-691 9a >95 >95

2 10a >95 >95

3 11a 24 2

9b 11a >95 –aConversion calculated by 1H NMRb100 mg scale with stirring



4. Conclusions

The biocatalysed reduction of the double bond of cinnamic acid derivatives is strongly influenced by the nature of the EWG. While no conversion was observed on the biocatalysed reduction of cinnamic acid 1a, an enzyme in Johnson Matthey’s collection, ENE-105, was capable of reducing methyl ester derivative 2a in low conversion. By replacing the alkyl chain in the ester moiety by a more EWG, such as fluorinated alkanes, and in the presence of enzymes ENE-105 and *ENE-69, we were able to significantly increase conversion to the reduced product. Furthermore, other electronegative derivatives such as cyclic imides proved to be even better activating groups, allowing the reduction of challenging substituted double bonds such as substrates 10a and 11a.In summary, by ‘masking’ the carboxylic acid

moiety into a fluorinated alkyl ester or a cyclic imide, following a straightforward synthetic

procedure, and in combination with the right enzyme, it was possible to biocatalytically reduce the conjugated double bond of cinnamic acid and substituted derivatives.

References

1. B. Dominguez, U. Schell, S. Bisagni and T. Kalthoff, Johnson Matthey Technol. Rev., 2016, 60, (4), 243

2. M. Hall, C. Stueckler, H. Ehammer, E. Pointner, G. Oberdorfer, K. Gruber, B. Hauer, R. Stuermer, W. Kroutil, P. Macheroux and K. Faber, Adv. Synth. Catal., 2008, 350, (3), 411

3. M. Hall, C. Stueckler, W. Kroutil, P. Macheroux and K. Faber, Angew. Chem., Int. Ed., 2007, 46, (21), 3934

4. J. F. Chaparro-Riggers, T. A. Rogers, E. Vazquez-Figueroa, K. M. Polizzi and A. S. Bommarius, Adv. Synth. Catal., 2007, 349, (8–9), 1521

5. A. Müller, B. Hauer and B. Rosche, Biotechnol. Bioeng., 2007, 98, (1), 22

6. M. A. Swiderska and J. D. Stewart, J. Mol. Catal. B: Enzym., 2006, 42, (1–2), 52

7. D. Dobrijevic, L. Benhamou, A. E. Aliev, D. Méndez-Sánchez, N. Dawson, D. Baud, N. Tappertzhofen, T. S. Moody, C. A. Orengo, H. C. Hailes and J. M. Ward, RSC Adv., 2019, 9, (63), 36608

8. H. S. Toogood and N. S. Scrutton, ACS Catal., 2018, 8, (4), 3532

9. G. Brown, T. S. Moody, M. Smyth, S. J. C. Taylor, ‘Almac: An Industrial Perspective of Ene Reductase (ERED) Biocatalysis’, in “Biocatalysis: An Industrial Perspective”, eds. G. de Gonzalo and P. Domínguez de María, ch. 8, Royal Society of Chemistry, London, UK, 2018, pp. 229–256

10. D. Mangan, I. Miskelly and T. S. Moody, Adv. Synth. Catal., 2012, 354, (11–12), 2185

11. R. Stuermer, B. Hauer, M. Hall and K. Faber, Curr. Opin. Chem. Biol., 2007, 11, (2), 203

12. Y. Kawai, M. Hayashi, Y. Inaba, K. Saitou and A. Ohno, Tetrahedron Lett., 1998, 39, (29), 5225

13. H. U. Blaser, B. Pugin and F. Spindler, ‘Asymmetric Hydrogenation’, in “Topics in Organometallic

Table IV Reduction of Cinnamoyl Pyrrolidinones by ENE-105 at pH 7b and 30°CEntry Substrate Scale, mg Concentration, mM Time, h Conversion, %a 1 9a 130 40 16 >95

2 10a 500 100 4 33

3 10a 500 100 20 72aConversion calculated by 1H NMRbpH controlled with NaOH titration (10 M)

Table V Reduction of Cinnamoyl Pyrrolidinone 10a at 50 mM Substrate Concentration, pH 7, 30°C, 18 h

Entry Enzyme Conversion, %a

1 ENE-105 >95

2 *ENE-69 >95

3 ENE-101 <0.5

4 ENE-102 1

5 ENE-103 0

6 ENE-104 0

7 ERED-103 0

8 ERED-110 0.5

9 ERED-112 0

10 ERED-207 <0.5

11 ERED-P1-A04 1

12 ERED-P1-E04 0

13 ERED-P1-H09 0aConversion calculated by 1H NMR



Chemistry: Organometallics as Catalysts in the Fine Chemical Industry”, eds., M. Beller, H. U. Blaser, Vol. 42, Springer-Verlag, Berlin, Germany, 2012, pp. 65–102

14. D. J. Ager, A. H. M. de Vries and J. G. de Vries, Chem. Soc. Rev., 2012, 41, (8), 3340

15. C. S. G. Seo and R. H. Morris, Organometallics, 2018, 38, (1), 47

16. D. J. Mansell, H. S. Toogood, J. Waller, J. M. X. Hughes, C. W. Levy, J. M. Gardiner and N. S. Scrutton, ACS Catal., 2013, 3, (3), 370

17. J. Waller, H. S. Toogood, V. Karuppiah, N. J. W. Rattray, D. J. Mansell, D. Leys, J. M. Gardiner, A. Fryszkowska, S. T. Ahmed, R. Bandichhor, G. P. Reddy and N. S. Scrutton, Org. Biomol. Chem., 2017, 15, (20), 4440

18. M. Breuer, K. Ditrich, T. Habicher, B. Hauer, M. Keßeler, R. Stürmer and T. Zelinski, Angew. Chem., Int. Ed., 2004, 43, (7), 788

19. D. Monge, H. Jiang and Y. Alvarez-Casao, Chem. Eur. J., 2015, 21, (12), 4494

20. V. A. Soloshonok, C. Cai, V. J. Hruby, L. Van Meervelt and T. Yamazaki, J. Org. Chem., 2000, 65, (20), 6688

21. T. Inokuma, Y. Hoashi and Y. Takemoto, J. Am. Chem. Soc., 2006, 128, (29), 9413

22. D. A. Evans, T. C. Britton, R. L. Dorow and J. F. Dellaria, J. Am. Chem. Soc., 1986, 108, (20), 6395

23. T. F. Woiwode and T. J. Wandless, J. Org. Chem., 1999, 64, (20), 7670

The Authors

Samantha Staniland graduated from The University of Manchester, UK, in 2011 with an MChem in Chemistry with Industrial Experience, while carrying out her industrial placement at Pfizer, UK, in Medicinal Chemistry. In 2011–2015, Sam did a PhD in the groups of Professor Jonathan Clayden and Professor Nicholas Turner on the biocatalytic asymmetric synthesis of atropisomers. Sam joined Johnson Matthey in 2015 as a research chemist in catalysis.

Tommaso Angelini completed his PhD in Chemical Science in 2010 from University of Perugia, Italy, working on the development of environmentally friendly synthetic protocols. During his postdoctoral studies, he finalised his work designing new continuous flow devices for the use of solid supported catalyst in low E-Factor transformations. Later, he gained experience in developing active pharmaceutical ingredient (API) production process at Procos (Italy). In 2015, he joined Johnson Matthey as Research Chemist, designing new enantioselective synthetic process for the preparation of APIs. He is now a Research Expert at Evotec Verona (Italy), working on the production of preclinical and Phase 1 API candidates.

Ahir Pushpanath obtained his PhD in Birkbeck College (University of London, UK) working on the engineering of enzymes for industrial biofuel production. With a biochemistry background, he specialises in the use of bioinformatics and computational biology in the rational design of new enzyme variants. Ahir joined Johnson Matthey in 2013 as a Senior Biologist and was instrumental in demonstrating the utility of computational techniques for rapid enzyme discovery through genome mining, in silico design and targeted enzyme engineering. He currently leads the enzyme development arm of biocatalysis, continuing to develop faster, more effective methods for ‘predictive biocatalysis’.



Amin Bornadel studied chemical engineering and received a PhD in biotechnology from Lund University in Sweden. For postdoctoral work, Amin went to Germany, where he carried out research within biocatalysis at University of Dresden and Technical University of Hamburg. In 2016, Amin joined Johnson Matthey to work as a biocatalysis researcher. He is currently a senior scientist working in the Biotech team.

Elina Siirola completed her PhD in 2012 from the University of Graz, Austria, where she worked on biocatalytic C=C bond hydrolysis. After a postdoctoral position in enzyme engineering at the Max Planck Institute for Coal Research, Germany, she joined Johnson Matthey in 2013, where she worked on biocatalysis research and development (R&D). Since 2017 she is a Principal Scientist in the Bioreactions group at Novartis Pharma in Basel, Switzerland.

Serena Bisagni completed her MSc in Industrial Biotechnology from the University of Pavia, Italy, in 2010 and then moved to Lund University, Sweden, for her postgraduate studies. In 2014 she obtained her PhD in Biotechnology in which she focused on the identification of new Baeyer-Villiger monooxygenases for fine chemicals synthesis within the Marie Curie Innovative Training Networks (ITN) ‘Biotrains’. In 2015 Serena joined Johnson Matthey. Her main interests are enzyme screening for synthesis of active pharmaceutical ingredients and fine chemicals and identification of novel biocatalysts.

Antonio Zanotti-Gerosa studied in Milano, Italy, completing his PhD in 1994 (organometallic chemistry). His academic experience include secondments to Imperial College, UK (Professor S. V. Ley), Nagoya University, Japan (Professor R. Noyori) and postdoctoral research at the University of Lausanne, Switzerland (Professor C. Floriani). Since 1997 he has been working on industrial applications of homogeneous catalysis. In 2003 he joined Johnson Matthey and, as R&D Director, he is leading the chemocatalysis group in the Cambridge laboratories.

Beatriz Domínguez gained her PhD in Synthetic Organic Chemistry from the University of Vigo, Spain, and then moved to the UK where she worked with Professor Tom Brown at the University of Southampton, UK, and with Professor Guy Lloyd-Jones at the University of Bristol, UK. In 2002 she joined Synetix, soon to become Johnson Matthey Catalysts and Chiral Technologies and has worked at Johnson Matthey’s facilities in Cambridge since. Beatriz has gained broad experience in the application of metal catalysis and biocatalysis, working closely with fine chemicals companies to deliver optimal catalysts for chemical processes.

Johnson Matthey Technology Review is Johnson Matthey’s international journal of research exploring science and technology in industrial applications


Editorial team

Manager Dan CarterEditor Sara ColesEditorial Assistant Yasmin StephensSenior Information Officer Elisabeth Riley

Johnson Matthey Technology ReviewJohnson Matthey PlcOrchard RoadRoystonSG8 5HEUKTel +44 (0)1763 253 000Email [email protected]


Date post:	18-Jan-2023
Category:	Documents
Upload:	khangminh22
View:	0 times
Download:	0 times

Volume 64, Issue 4, October 2020 - Johnson Matthey ...

Documents