8
Click here to load reader
Click here to load reader
P
Eec
Da
b
a
ARRAA
KLFBESL
1
dmamcmagw
iL
h1
ARTICLE IN PRESSG ModelRBI-10171; No. of Pages 8
Process Biochemistry xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Process Biochemistry
jo ur nal home p age: www.elsev ier .com/ locate /procbio
valuation of lactic acid bacterial strains of boza for theirxopolysaccharide and enzyme production as a potential adjunctulture
ilek Heperkana,∗, Ceren Daskaya-Dikmena, Banu Bayramb
Istanbul Technical University, Faculty of Chemical and Metallurgical Engineering, Department of Food Engineering, 34469 Maslak, Istanbul, TurkeyTUBITAK, Marmara Research Center Food Institute, 41470 Kocaeli, Turkey
r t i c l e i n f o
rticle history:eceived 6 December 2013eceived in revised form 4 June 2014ccepted 5 June 2014vailable online xxx
eywords:actic acid bacteriaT-IRozaxopolysaccharidetreptococcus macedonicusactobacillus paracasei
a b s t r a c t
Boza is a non-alcoholic beverage obtained from fermented cereals. Thirteen lactic acid bacteria (LAB),previously isolated from boza were identified and evaluated to determine the various technological prop-erties for selecting appropriate strains as adjunct culture in boza. Each isolate was checked for purity,Gram-stained and tested for the catalase and oxidase activity and then subjected to identification bypolymerase chain reaction (PCR) with partial 16S rRNA gene sequencing. The tests for carbohydratefermentation and enzyme profiles were carried out with the API 50 CHL and API ZYM galleries, respec-tively. Exopolysaccharide (EPS) production of strains was determined by Fourier transform infraredspectroscopy (FT-IR) and quantified by the phenol sulphuric acid method. To our best knowledge, this isthe first study reporting on Lactococcus garvieae (E32), Pediococcus parvulus (E42) and Streptococcus mace-donicus (A15) in boza. All strains, except S. macedonicus (A15) produced EPS. Leuconostoc citreum (E55) andLactococcus lactis (A47) were the highest EPS producing strains, yielding 2.39 ± 0.49 and 1.98 ± 0.23 g/Lof EPS, respectively. Lactobacillus paracasei (D41), Lactobacillus plantarum (B2), Lactococcus lactis (F39)
and among low-EPS producing strains Lactobacillus coryniformis (C55), L. paracasei (E8), and P. parvu-lus (E42) were evaluated to be promising candidates as potential adjunct culture in boza. The varietyof enzyme production was also concern. Lc. garvieae (E32) was found to produce the largest variety ofenzymes among the strains. FT-IR spectroscopy can be used for the assessment of EPS production bymicroorganisms reliably and accurately.© 2014 Elsevier Ltd. All rights reserved.
. Introduction
Boza is a cereal-based fermented beverage that can be pro-uced by boiling coarsely ground cereals such as millet, rice, wheat,aize and cracked wheat in water. The ratio of cereal and water is
pproximately 1–5. The boiling process is completed until the rawaterial becomes thick soup-like consistency. The slurry is then
ooled, distributed to the containers, and finally subjected to fer-entation process. The fermentation process takes one to two days
t 20–22 ◦C. Boza is consumed without any treatment by all age-roups including children and elderly particulary in autumn andinter. Cinnamon powders are sprinkled over boza for better flavor.
Please cite this article in press as: Heperkan D, et al. Evaluation of lacenzyme production as a potential adjunct culture. Process Biochem (2
Several members of the lactic acid bacteria (LAB) participaten boza fermentation itself as actively growing bacteria includingactobacillus, Lactococcus, Leuconostoc, Pediococcus, Enterococcus
∗ Corresponding author. Tel.: +90 212 2856041; fax: +90 212 2852925.E-mail address: [email protected] (D. Heperkan).
ttp://dx.doi.org/10.1016/j.procbio.2014.06.012359-5113/© 2014 Elsevier Ltd. All rights reserved.
and Weissella [1–4]. Since these microorganisms have complexnutritional requirements [5], cereals and cereal-based productspossess the nutrients necessary for LAB requirements, and theyare accepted as suitable environments to maintain the growth andvitality of the bacteria [6].
Exopolysaccharides (EPS) are polymers with a high molecularweight that are produced in situ by microorganisms in fermentedmilk products and are evaluated as an alternative food additives[7,8]. EPS, synthesized by certain lactic acid bacteria, Streptococcus,Lactobacillus, Lactococcus and Leuconostoc are GRAS status [7]. Theyare soluble in water and form a gel as the viscosity of food increases[8].
There are limited products produced by fermented cereals forhuman consumption. Boza is one of the well-known fermentedcereal-based beverage in Turkey, Balkan countries and South Russia
tic acid bacterial strains of boza for their exopolysaccharide and014), http://dx.doi.org/10.1016/j.procbio.2014.06.012
[9]. In boza fermentation, the use of starter cultures is not common.The manufacturers use a back slopping technique with indigenousstrains. A portion of a prior fermentation is used as an inoculumfor the following batch. However, boza spoils quickly, and it is
ARTICLE ING ModelPRBI-10171; No. of Pages 8
2 D. Heperkan et al. / Process Bioche
Table 1Physico-chemical properties of the boza samples.
Physico-chemical properties Boza samples
1a 2b 3a
Dry matter (%) 22.74 ± 3.22 14.39 ± 1.31 24.07 ± 2.45Total sugar (%) 22.62 ± 2.11 6.63 ± 1.00 16.00 ± 1.98Protein (%) 0.46 ± 0.01 0.65 ± 0.07 0.56 ± 0.05Ash (%) 0.44 ± 0.05 0.17 ± 0.01 0.24 ± 0.03Total acidity (%) 0.25 ± 0.03 0.30 ± 0.05 0.27 ± 0.02pH 3.95 ± 0.45 3.68 ± 0.25 3.47 ± 0.58Water activity 0.94 ± 0.04 0.92 ± 0.08 0.91 ± 0.13
vdabhsu
qcdsescmisTd
2
2
fboPLLAnifW
TT
a Sample 1 and 3 produced by wheat and maize.b Sample 2 produced by millet.
ery susceptible temperature differences between day and nighturing transporting. Because of short shelf-life (only 2–3 weeks)nd unsuitable for refrigeration, boza may contain pathogenicacteria, produce undesired fermentation products such as alco-ol Thus the quality and sensory characteristics of boza is nottable and sometimes producing abdominal discomfort and flat-lence.
The textural characteristics of boza are very important for theuality of final product. If necessary, hot water after cooking pro-ess can be added to modify boza viscosity. Boza is a highly viscousrink, but gel formation is not desired. Thus, the strain used fortarter culture in boza should not be an EPS producer. Microbialnzyme activities during fermentation process are largely respon-ible for the formation of flavor compounds, because cereals areooked for boza production. The aim of this study was to deter-ine the lactic acid bacteria in boza for selecting appropriate
ndigenous strains as adjunct/starter cultures. EPS production ofelected strains was studied to establish one of selection criteria.herefore, the enzymatic properties of selected strains were alsoetermined.
. Materials and methods
.1. Materials
Thirteen strains of lactic acid bacteria (LAB) previously isolatedrom boza obtained from three major boza manufacturers in Istan-ul were used in this study. The physico-chemical characteristicsf boza samples were shown in Table 1. Strains were identified byCR using partial 16S rRNA gene sequencing as explained below.actobacillus (L.) paracasei (D41, E8), Lactobacillus plantarum (B2),actococcus (Lc.) lactis (A47, F39), Leuconostoc (Ln.) citreum (E55,31, B56) and Streptococcus macedonicus (A15) (recent proposed
Please cite this article in press as: Heperkan D, et al. Evaluation of lacenzyme production as a potential adjunct culture. Process Biochem (2
ame by Schlegel et al. [10] S. gallolyticus subsp. macedonicus) weresolated from boza made of wheat and maize. Whereas, L. coryni-ormis (C55), Lc. garvieae (E32), Pediococcus (P.) parvulus (E42) and
eissella (W.) confusa (C19) were isolated from boza made of millet
able 2he raw materials used in boza production, the media and the temperatures used for the
LAB species Raw materials
Lactobacillus coryniformis (C55) Millet
Lactobacillus paracasei (D41, E8) Millet; maize and wheat
Lactobacillus plantarum (B2) Millet; maize and wheat
Lactococcus garvieae (E32) Millet
Lactococcus lactis (A47, F39) Millet; maize and wheat
Leuconostoc citreum (E55, A31, B56) Millet; maize and wheat
Pediococcus parvulus (E42) Millet
Streptococcus macedonicus (A15) Millet; maize and wheat
Weissella confusa (C19) Millet
a Rogosa agar (Merck).b de Man, Rogosa and Sharp (MRS) agar (Merck).c M17 agar (Merck).
PRESSmistry xxx (2014) xxx–xxx
only. The culture media and the temperatures used for isolation ofeach strain and the raw materials used for boza production wereshown in Table 2.
2.2. The physico-chemical characteristics of boza samples
The physico-chemical analyses conducted for the samples areas follows. Dry matter was determined with a forced draft oven.Protein, ash, pH was analyzed with a Kjeldhal 1030 protein appa-ratus, combustion at 550 ◦C, and a Jenway 3010 type pH meter,respectively. Total sugar contents along with Turkish Standard[11], water activity (aw) was determined by a Decagon PawKit(AquaLab, Washington, USA). Also, total titritable acidity wasanalyzed and expressed as % lactic acid by the method from Thyya-garaja et al. [12]. All analyses were carried out with duplicatesamples.
2.3. Isolation and enumeration of LAB
Culture-dependent methods were used for the determinationof lactic biodiversity in boza. Lactic acid bacteria were isolatedand enumerated from boza by using three different culture medianamely de Man, Rogosa and Sharpe (MRS) agar (Merck, Darmstadt,Germany), Rogosa agar (Merck, Darmstadt, Germany) and M17 agar(Merck, Darmstadt, Germany) and two different incubation tem-peratures such as at 30 ◦C and 37 ◦C for 48 h. 25 g boza sample wasadded to 225 ml sterile phosphate buffer solution in aseptic condi-tions and suitable dilutions were made. 0.1 ml of the sample wasinoculated onto three different culture media.
2.4. Identification of lactic acid bacteria by PCR
Cultures were subcultured, to check purity. Also, cultures wereGram-stained and tested for catalase and oxidase activity, andthen all sample identification were confirmed by PCR. The reac-tions of carbohydrate fermentation were determined by API50CHLwith Apilab Plus software (BioMerieux, Istanbul, Turkey). In thisstudy, partial 16S rRNA gene sequencing was used with the uni-versal primers: forward 5′ CCG TCA ATT CCT TTG AGT TT 3′ andreverse 3′ AGA GTT TGA TCC TGG CTC AG 5′ [13]. The result-ing PCR product was sequenced; and sequence similarity searchwas carried out with BLAST (Basic Local Alignment Search Tool).Comparisons of reference strain and isolates sequences were per-formed using Clustal W alignment algorithm. The dendogramconstructed (Neighbor-joining) from the partial 16S sequencesof isolates and matching sequences from the reference strainsfound in databases (NCBI gene bank) is presented in Fig. 1. Their
tic acid bacterial strains of boza for their exopolysaccharide and014), http://dx.doi.org/10.1016/j.procbio.2014.06.012
accession numbers are as follows: L. coryniformis (NR029018.1),L. paracasei (JQ680426), L. plantarum (KC454277.1), Lc. garvieae(AB598960), Lc. lactis (HE805077.1), Ln. citreum (DQ489736.1), P.parvulus (NR029136), S. macedonicus (NR074404.1) and W. confusa
isolation of each strain.
Isolation medium Temperature (◦C)
Rogosaa 30Rogosa 37MRSb 37MRS 37M17c 37MRS 37MRS 30MRS 37Rogosa 37
ARTICLE IN PRESSG ModelPRBI-10171; No. of Pages 8
D. Heperkan et al. / Process Biochemistry xxx (2014) xxx–xxx 3
E8
Lactobacillus paracasei
D4 1
E42 Pediococcus parvulus
C55
Lactobacillus coryniformis
B2
Lactobacillus plantarum
C1 9
Weissella con fus a
B56
A31
E55
Leuconostoc citreum
A15
Strep tococcu s maced onicu s
E32
Lactococcus garvieae
A47
F39
Lactococcus lactis
100
95
98
95
99
88
100
100
88
100
100
10068
50
97
0.005
F ces of
(
(w
2
fwbrTiccptmw
S
2
Zcri
ig. 1. The dendogram constructed (Neighbor-joining) from the partial 16S sequenNCBI gene bank).
JQ805675.1). The cultures were lyophilized in 20% fat-free milkith 8% saccharose and kept at 4 ◦C for further studies.
.5. Determination of cultures resistance to freeze-drying
Thirteen LAB strains were investigated for their resistance toreeze-drying. Ten mL of MRS broth containing 48 h-old culturesere harvested by centrifugation at 5578 rpm for 15 min followed
y washing twice with quarter-strength Ringers solution; and thenesuspended in 20% reconstituted skim milk with 8% saccharose.he cultures were kept at −80 ◦C for 4 h and freeze-dried for 24 hn Freeze Dryer (Christ Alpha 1–2 LDplus, Germany). Initial cellount was determined by performing serial dilutions of 48h-oldultures at 37 ◦C in quarter-strength Ringers solution followed bylating onto or brine by using selective media such as MRS agar, andhen enumerated. After freeze-drying, survival of cells was deter-
ined by the same method explained above. The survival ratesere determined using a following equation:
urvival rate (%) = 100 × Viable cell counts after freeze-dryinginitial cell counts
.6. Determination of enzyme activity of LAB by API ZYM
Enzyme activity of selected cultures was determined by API
Please cite this article in press as: Heperkan D, et al. Evaluation of lacenzyme production as a potential adjunct culture. Process Biochem (2
YM. Enzyme activity was graded from 0 to 5 based on the colorhart (BioMerieux, Istanbul, Turkey). The results were reported aseactions of following category; no activity (0), low activity (1),ntermediate activity (2–3) and high activity (4–5).
the isolates and the matching sequences from reference strains found in databases
2.7. Determination of exopolysaccharide production of LAB byFT-IR
Twenty four hour cultures of lactic acid bacteria were inocu-lated to MRS broth and incubated at 37 ◦C for 24 h. After incubation,the cultures were harvested by centrifugation at 5578 rpm for15 min. The supernatants were sterilized by filtration with 0.22 �mmembranes (Whatman FP 30/0.2 CA-S, Dassel, Germany). Eachsupernatant was freeze dried (Christ Freeze Dryer Alpha 1–2LDplus, Germany), acidified with acetic acid. The EPS produc-tion was characterized by FT-IR (Perkin Elmer Spectrum 65 FT-IRspectrometer, Massachusetts, USA) equipped with attenuated totalreflectance (ATR) sampling accessory. Four scans between 1400 and800 cm−1 with a resolution of 4 cm−1 were taken in absorbancemode for each sample.
2.8. Determination of exopolysaccharide by phenol sulphuric acidmethod
The amount of EPS was determined by same methods as Smiti-nont et al. [14] and Van Geel-Schutten et al. [15] used. Cultureswere incubated in MRS broth at 37 ◦C for 24 h followed by adding0.1 mL of MRS broth to 10 mL modified MRS (m-MRS) whichcontains 100 g/L of saccharose (sugars are sterilized separately).These mixed cultures were incubated for 3 days at 37 ◦C underanaerobic conditions. After that, the cultures were centrifuged for
tic acid bacterial strains of boza for their exopolysaccharide and014), http://dx.doi.org/10.1016/j.procbio.2014.06.012
20 min at 6000 rpm (Hettich, Universal 16A) to remove the cells.Three volumes of ethanol (Riedel-de Häen) were added to super-natant and kept for 24 h at 4 ◦C for precipitation. This culturewas centrifuged again for 20 min at 6000 rpm. After removal of
IN PRESSG ModelP
4 Biochemistry xxx (2014) xxx–xxx
sdwdUc0
3
3
timsa2osoaoTTa
tT
3
sgt(
i(Lawiiotfwbiramfiocpeam
Table 3LAB species isolated from boza.
LAB species Methods of identification Reference
L. brevis API50CHL + PCR [2]API50CHL + Ribo Printer [3]
L. coryniformis API50CHL + PCR †a
L. fermentum API50CHL + PCR [2]L. graminis API50CHL + Ribo Printer [3]L. paracasei subsp. paracasei API 50CHL + PCR [2]
API50CHL + Ribo Printer [3]API50CHL + PCR †
L. paraplantarum API50CHL + Ribo Printer [3]L. pentosus API50CHL + PCR [2]L. plantarum API50CHL + PCR [2]
API50CHL + Ribo Printer [3]API50CHL + PCR [19]API50CHL + PCR †
L. rhamnosus API50CHL + PCR [2]Lactococcus lactis subsp. lactis PCR [20]Lc. garvieae API50CHL + PCR †Leuconostoc citreum API50CHL + Ribo Printer [3]
API50CHL + PCR †Ln. lactis API50CHL + PCR [19]Pediococcus spp. API50CHL + Ribo Printer [3]Pediococcus parvulus API50CHL + PCR †Streptococcus macedonicus API50CHL + PCR †Weissella confusa API50CHL + PCR †a From this study.
Table 4Quantities of EPS produced by LAB strains isolated from boza.
LAB Strains Amount of EPS (g/L)
Lactobacillus coryniformis (C55) 0.83 ± 0.16Lactobacillus paracasei (D41) –ND
Lactobacillus paracasei (E8) 0.88 ± 0.13Lactobacillus plantarum (B2) –Lactococcus garvieae (E32) –Lactococcus lactis (A47) 1.98 ± 0.23Lactococcus lactis (F39) –Leuconostoc citreum (E55) 2.39 ± 0.49Leuconostoc citreum (A31) 0.87 ± 0.11Leuconostoc citreum (B56) 0.85 ± 0.05Pediococcus parvulus (E42) 1.29 ± 0.05Streptococcus macedonicus (A15) –
ARTICLERBI-10171; No. of Pages 8
D. Heperkan et al. / Process
upernatant, the precipitate was dissolved in 10 mL of water andialysed (cellulose dialysis tube, Sigma D-9652) against distilledater at 4 ◦C for 24 h. Exopolysaccharide (total carbohydrate) pro-uction was determined by phenol sulphuric acid method (T80V-Vis Spectrophotometer, PG Instrument Ltd. UK) against a glu-ose standard [16], which was prepared with concentrations of–800 �g/mL. All analyses were carried out with triplicate samples.
. Results and discussion
.1. The properties of boza samples and strains
The physico-chemical properties of boza samples obtained fromhree different producing companies are given in Table 1. Accord-ng to the National Standard for boza [11], the quantity of dry
atter must be at least 20% and total sugar content of saccharosehould be 10%. Physico-chemical properties of all samples were inccordance with the standard values with exception of the sample
(14.39 ± 1.31% dry matter and 6.63 ± 1.00% total sugar content)btained by using millet as raw material. The water activity of theamples was between 0.91 and 0.94 (Table 1). Average LAB countsbtained from boza samples at 2 different incubation temperaturesnd using 3 different media. The averages for LAB counts carriedut at 30 ◦C, and 37 ◦C in MRS agar were 9.36 and 9.79 log cfu/g.he averages for counts were 9.13 and 9.15 log cfu/g in M17 agar.he averages for LAB counts were 9.17 and 8.93 log cfu/g in Rogosagar.
The raw materials used in boza production, the media and theemperatures used for the isolation of each strain are shown inable 2.
.2. Identification of lactic acid bacteria
A total of 13 Gram-positive, catalase and oxidase negativetrains were used for partial 16S rRNA gene sequencing. The dendo-ram constructed from the partial 16S sequences of the isolates andhe matching sequences from reference strains found in databasesNCBI Gene bank) is presented in Fig. 1.
Based on identification by molecular methods the strains weredentified as follows: Lactobacillus coryniformis (C55), L. paracaseiD41, E8), L. plantarum (B2), Lc. garvieae (E32), Lc. lactis (A47, F39),n. citreum (E55, A31, B56), P. parvulus (E42), S. macedonicus (A15)nd W. confusa (C19). Lactic acid bacteria isolated from boza world-ide excluding classical microbiological methods used for their
dentification is shown in Table 3. The lactic microbiota determinedn different studies are different occasionally in Turkey as well asther countries [1–3,17–20]. Among lactic acid bacteria, L. plan-arum, Lc. lactis, L. brevis, L. fermentum and W. confusa, are the mostrequently cited species followed by Pediococcus species world-ide. Leuconostocs are also commonly found as a group of LAB in
oza. However, Streptococcus was the least frequently found speciesn boza (Table 3). Indeed, to our knowledge, this is the first studyeporting on S. macedonicus (A15) together with Lc. garvieae (E32)nd P. parvulus (E42) in boza. Even if Hancıoglu and Karapınar [21]entioned L. coryniformis in boza, this is the first report of L. coryni-
ormis (C55) determination by molecular methods. The differencesn lactic microbiota could be attributed by raw material, methodsf making boza, and isolation and identification of bacteria. It isonsidered that the use of different cereals for manufacturing boza
Please cite this article in press as: Heperkan D, et al. Evaluation of lacenzyme production as a potential adjunct culture. Process Biochem (2
layed a significant role in the variety and number of isolates. Forxample, L. coryniformis (C55), Lc. garvieae (E32), P. parvulus (E42)nd W. confusa (C19) were isolated from the boza only made ofillet.
Weissella confusa (C19) 0.93 ± 0.06
ND, not detected.
3.3. The enzymatic profiles of LAB strains
The enzymatic profile of strains is shown in Table 5. The strainstested showed one or more type of enzyme activity. None ofthe LAB species tested in this study showed definite lipase ortrypsin activity; while all species showed high acid phosphataseactivity. Lc. garvieae (E32) showed the widest variety of enzymesamong the species. The weakest enzyme activity was observedin leuconostocs (E55, A31, B56), S. macedonicus (A15) and in W.confusa (C19). All lactococci strains (A47, F39, E32), L. paracasei(D41, E8) and L. plantarum (B2) showed esterase and esteraselipase activities. Results showed a high aminopeptidase activi-ties for L. coryniformis (C55), L. paracasei (D41, E8), L. plantarum(B2), Lc. garvieae (E32), Lc. lactis (A47) and P. parvulus (E42)and no protease activities except Lc. garvieae (E32) and Lc. lactis(A47). Within 3 leuconostocs, only two showed esterase and �-glucosidase activity. L. coryniformis (C55), L. paracasei (D41, E8), L.plantarum (B2), Lc. garvieae (E32), Lc. lactis (A47) and P. parvulus(E42) showed very high activity for all aminopeptidases. Amongproteases, trypsin, which operates most effectively in alcaline pH
tic acid bacterial strains of boza for their exopolysaccharide and014), http://dx.doi.org/10.1016/j.procbio.2014.06.012
values, was not detected or detected in low levels in the teststrains of boza, a fairly acidic product. Chymotrypsin was pro-duced only by Lc. garvieae (E32) and Lc. lactis (A47), the latter have
ARTICLE IN PRESSG ModelPRBI-10171; No. of Pages 8
D. Heperkan et al. / Process Biochemistry xxx (2014) xxx–xxx 5
Fig. 2. FT-IR spectra of �-glucan bands and C–O–C; C–O stretching vibrations and glycosidic linkages of the exopolysaccharide in supernatant of LAB strains: Ln. citreum( asei (El
hsag�pfoAtPmpaals
3
ms((thTriiCagbncpt1Ld
E55), P. parvulus (E42), Lc. lactis (A47), W. confusa (C19), Lc. garvieae (E32), L. paracactis (F39), L. paracasei (D41) and S. macedonicus (A15).
igh activity. In terms of carbohydrate sources, none of specieshowed high �-mannosidase, �-glucuronidase or �-fucosidasectivity except Lc. garvieae (E32). Only L. plantarum (B2) and Lc.arvieae (E32) showed both �- and �-galactosidase activities. High-glucosidase and aminopeptidase activities were detected for L.aracasei strains (D41, E8); whereas, no �-mannosidase and �-ucosidase activities were observed. Enzyme activities of strainsbserved in this study is in accordance with the literature [22–24].lthough limited information was available for enzymatic activi-
ies of strains including L. paracasei (D41, E8), Lc. garvieae (E32),. parvulus (E42) and S. macedonicus (A15), enzyme profiles ofost of species tested have similar profiles to those in literature
ublished previously. The presence of high �- and �-glucosidasectivities and low activities toward other carbon sources suchs mannose, fructose and glucuronides suggest that species iso-ated from boza prefer glucose for their carbon and energyource.
.4. Exopolysaccharide production of the LAB strains
Exopolysaccharide (EPS) production of the LAB strains deter-ined by FT-IR is shown in Fig. 2. Among the 12 out of 13
trains were found to produce EPS except S. macedonicus (A15)Fig. 2). FT-IR spectroscopy is based on the use of vibrationalstretching and bending) spectroscopy method that represent func-ional groups of molecule [25]. The functional groups of EPS showigh absorbancies in the region 1200–900 cm−1 in FT-IR spectra.hus, EPS productions of microorganisms were identified accu-ately by FT-IR. As shown in Fig. 2, the band near 1370 cm−1
s related to characteristic peak for �-glucan [26]. Wide stretch-ng between 1200 and 1000 cm−1 correspond to carbohydrate–O–C; C–O stretching vibrations as shown in Fig. 2 [27]. Thebsorption bands between 872 and 812 cm−1 show � and �-lycosidic linkages [28]. EPS production of strains was quantifiedy the phenol sulphuric acid method is shown in Table 4. Theumbers of EPS positive strains were 8 out of 13 (Table 4). Leu-onostoc citreum (E55) and Lc. lactis (A47) were the highest EPSroducing strains, yielding 2.39 ± 0.49 and 1.98 ± 0.23 g/L, respec-
Please cite this article in press as: Heperkan D, et al. Evaluation of lacenzyme production as a potential adjunct culture. Process Biochem (2
ively. The rest of 6 strains produced between 0.85 ± 0.06 and.29 ± 0.05 g/L EPS. Although EPS production of L. paracasei (D41),. plantarum (B2), Lc. garvieae (E32) and Lc. lactis (F39) wasetected by FT-IR, they were not detected with the phenol sulphuric
8), Ln. citreum (A31), Ln. citreum (B56), L. coryniformis (C55), L. plantarum (B2), Lc.
acid method. This could be attributed to the supernatants werelyophilized therefore more concentrated samples were used in FT-IR.
The results of EPS production capabilities of strains were inagreement with the literature [29–32] except S. macedonicus (A15)and L. coryniformis (C55). Based on our best knowledge, this is thefirst report of L. coryniformis (C55) produced EPS. The yield of EPSproduced by LAB depends on factors such as nutrients, tempera-ture, pH, amount of oxygen, incubation period. Modified MRS isa more effective medium for determining strains with a high EPSproduction potential [33]. Laws and Marshall [34] investigated therelevance of EPS to the rheological properties in fermented milk byusing nuclear magnetic resonance (NMR). More recently Palombaet al. [31] used NMR to characterize the EPS by LAB. Ln. citreumwas the highest EPS-producing species in their study similar tothis study [31]. For the optimization of the boza production, EPS-negative or low-EPS producer LAB were selected as adjunct/startercultures. Therefore, the cultures were checked for their capabilityto produce EPS only.
Boza is a highly viscous drink, but gel formation is not desired.The viscosity is modified by adding hot water if necessary bymanufacturer before adding starter cultures. Therefore, EPS pro-ducing ability of a particular strain does not give additionalbenefit to boza manufacturer. The advantages and disadvantagesof EPS-negative and low-EPS producer strains were discussedbelow.
S. macedonicus is frequently associated with traditional dairyproducts isolated from naturally fermented cheese [35]. AlthoughS. macedonicus is a promising strain for a starter culture it isphylogenetically related to streptococci associated with diseases.Therefore, the pathogenicity status of S. macedonicus is ambivalent,raising concerns about the safety of its use as a starter culture infood fermentations [36].
L. coryniformis found in fermented products of dairy origin suchas cheese and natural green fermented olives [37,38]. Some strainhad strong inhibitory activity in vitro against the molds; however,a weaker activity was observed against the yeasts [39] which arevery beneficial to cereal-based products such as boza. Therefore, L.
tic acid bacterial strains of boza for their exopolysaccharide and014), http://dx.doi.org/10.1016/j.procbio.2014.06.012
coryniformis (C55) can be a promising strain for starter culture inboza.
Lactobacillus paracasei found in cheeses and natural green fer-mented olives [40,41]. Recently, L. paracasei have more attention
Please cite this article in press as: Heperkan D, et al. Evaluation of lacenzyme production as a potential adjunct culture. Process Biochem (2
ARTICLE ING ModelPRBI-10171; No. of Pages 8
6 D. Heperkan et al. / Process BiocheTa
ble
5En
zym
atic
acti
viti
es
of
the
LAB
stra
ins
isol
ated
from
boza
.
Enzy
mes
Lact
obac
illus
cory
nifo
rmis
(C55
)
L.
para
case
i(D
41)
L.
para
case
i(E
8)L. pl
anta
rum
(B2)
Lact
ococ
cus
garv
ieae
(E32
)Lc
. lac
tis
(A47
)Lc
. lac
tis
(F39
)Le
ucon
osto
cci
treu
m
(E55
)Ln
. cit
reum
(A31
)Ln
. cit
reum
(B56
)Pe
dioc
occu
spa
rvul
us
(E42
)St
rept
ococ
cus
mac
edon
icus
(A15
)
Wei
ssel
laco
nfus
a (C
19)
Alk
alin
e
ph
osp
hat
ase
–ND
–
–
–
3
4
3
3
–
–
–
–
3Es
tera
se
–
4
–
4
3
4
3
–
3
–
–
–
–Es
tera
se
lip
ase
3
4
3
4
4
3
3
–
–
–
4
–
–Li
pas
e
–
–
–
–
–
–
–
–
–
–
–
–
–Le
uci
ne
aryl
amid
ase
5
5
5
5
5
5
5
3
–
4
5
4
–V
alin
e
aryl
amid
ase
5
5
3
4
5
4
–
3
–
–
5
3 –
Cys
tin
e
aryl
amid
ase
3
3
3
4
4
3
–
–
–
–
4
2 –
Tryp
sin
–
–
–
–
–
–
–
–
–
–
–
– –
�-C
hym
otyr
psi
n
–
–
–
–
3
5
–
–
–
–
–
–
–A
cid
ph
osp
hat
ase
3
5
3
4
4
5
5
4
3
3
4
4
5N
aph
thol
-AS-
BI-
ph
osp
hoh
ydro
lase
3
3
4
3
3
3
3
3
–
–
3
–
3
�-G
alac
tosi
das
e
3
–
–
5
3
–
–
–
–
–
– –
–�
-Gal
acto
sid
ase
–
4
–
5
5
–
–
–
–
–
– –
3�
-Glu
curo
nid
ase
–
–
–
–
3
–
–
–
–
–
–
–
–�
-Glu
cosi
das
e
5
4
5
5
4
4
–
–
5
4
5
–
–�
-Glu
cosi
das
e
3
3
5
–
5
4
–
–
–
–
5
–
–N
-ace
tyl-
�-
glu
cosa
min
idas
e–
–
4
–
5
–
–
–
–
– –
–
–
�-M
ann
osid
ase
–
–
–
–
3
–
–
–
–
– –
–
–�
-Fu
cosi
das
e
–
–
–
–
4
–
–
–
–
–
–
–
–
ND
, not
det
ecte
d.
PRESSmistry xxx (2014) xxx–xxx
to be a potential probiotic LAB culture both in fermented cheeseand olive [40,41]. Due to the number of positive properties, L.paracasei (D41, E8) can be promising strain for starter culture inboza.
L. plantarum is the most common species found in fermentedfood ecosystem including fermented cereals. This can be explainedby L. plantarum ables to produce transporters and enzymes involvedin carbohydrate fermentation [42,43]. L. plantarum (B2) strain iso-lated from boza in this study has high �-galactosidase activity, theabsence of protease and strong peptidase activities are positivedesirable properties of this strain to become a good candidate asan adjunct culture in boza.
Lactococcus genus is composed of eight species and subspecies.Among them, Lc. garvieae are mostly involved in human pathol-ogy [44]. The route of infection can be ingestion of contaminatedraw fish, grilled fish or cheese [45,46]. The pathogenic status ofLc. garvieae limits its usage as starter culture. In contrast with Lc.garvieae, Lc. lactis is a GRAS (generally regarded as safe) microor-ganism that is widely used for cheese manufacturing. Along withacidifying activity, lactococci contribute to textural and organolep-tic quality of fermented product [47]. In addition to the role ofstarter culture bacteria for fermented foods, LAB demonstrated thecapacity to improve both allergic responses in murine models [48].Lactococci are capable of fermenting lactose and citrate to form CO2.These gases also result in development of off-flavors [49]. In caseof Lc. lactis, care of process control must be taken due to possiblegas defects in products as well as undesirable flavor profiles. SinceLc. lactis (F39) was lower EPS producing strain than Lc. lactis (A47),Lc. lactis (F39) may be promising candidate for starter culture.
Ln. citreum strains found in boza in this study were all EPSproducers, one of them (E55) was the highest EPS producer. Leu-conostocs are only suggested if there is a specific problems in somefermented foods such as prevention of syneresis in yogurt [50].Therefore these strains are not considered as starter culture bacte-ria in boza.
Among LAB in fermented foods, pediococci are less found gen-era. Some types of cereal based fermented products of Africanorigin contain pediococci [51]. Recently, the bifidogenic effect ofEPS-producing P. parvulus in oat based fermented products wasreported [31]. In this study, the strain is low EPS-producer and highpeptidase activity. Thus, P. parvulus (E42) strain may help to pre-vent bitterness defects, occurred in boza occasionally and it can bepromising species for starter culture in boza.
Although some strains of W. confusa have emerged as oppor-tunistic pathogens for humans and animals, W. confusa have beenused in fermented food production [52]. It was reported as attrac-tive species due to its diverse biotechnological applications insourdoughs without strong acidification [53]. W. confusa (C19)strain isolated from boza was a low (<1 g/L) EPS producer, how-ever this strain was not considered as a potential starter culturebacterium for boza production, due its pathogenic character.
4. Conclusion
In this study, the isolation step followed by strain characteri-zation, survival of strains upon freeze-drying, and determinationof properties of EPS production as well as enzyme activities hasbeen completed. FT-IR spectroscopy can be used reliably and accu-rately for the assessment of EPS production by microorganisms.L. paracasei (D41), L. plantarum (B2), Lc. lactis (F39) and among
tic acid bacterial strains of boza for their exopolysaccharide and014), http://dx.doi.org/10.1016/j.procbio.2014.06.012
low-EPS producing strains, L. coryniformis (C55), L. paracasei (E8),and P. parvulus (E42) were evaluated to be promising strains asadjunct cultures for boza. Future studies should cover pheno-typic characteristics and other technological properties including
ING ModelP
Bioche
ds
A
E(
R
[
[[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
ARTICLERBI-10171; No. of Pages 8
D. Heperkan et al. / Process
etermination of antibacterial and antiyeast properties of strainshould be studied in detail before commercial application.
cknowledgement
The authors greatly appreciated critical reviews of the use ofnglish and valuable comments and corrections of Dr. Bosoon ParkUSDA, ARS, USA).
eferences
[1] Zorba M, Hancioglu O, Genc M, Karapinar M, Ova G. The use of starter culturein the fermentation of boza, a traditional Turkish beverage. Process Biochem2003;38:1405–11.
[2] Botes A, Todorov SD, von Mollendorff JW, Botha A, Dicks LMT. Identifica-tion of lactic acid bacteria and yeast from boza. Process Biochem 2007;42:267–70.
[3] Kivanc M, Yilmaz M, Cakir E. Isolation and identification of lactic acid bacteriafrom boza, and their microbial activity against several reporter strains. Turk JBiol 2011;35:313–24.
[4] Altay F, Karbancıoglu-Güler F, Daskaya-Dikmen C, Heperkan D. A review ontraditional Turkish fermented non-alcoholic beverages: microbiota, fermen-tation process and quality characteristics. Int J Food Microbiol 2012;167:44–56.
[5] Hebert EM, Raya RR, De Giori GS. Nutritional requirements and nitrogen-dependent regulation of proteinase activity of Lactobacillus helveticus CRL 1062.Appl Environ Microbiol 2000;66:5316–21.
[6] Charalampopoulos D, Wang R, Pandiella SS, Webb C. Application of cerealand cereal components in functional foods: a review. Int J Food Microbiol2002;79:131–41.
[7] Boels IC, van Kranenburg R, Hugenholtz J, Kleerebezem M, de Vos WM.Sugar catabolism and its impact on the biosynthesis and engineering ofexopolysaccharide production in lactic acid bacteria. Int Dairy J 2001;11:723–32.
[8] De Vuyst L, De Vin F, Vaningelgem F, Degeest B. Recent developments in thebiosynthesis and applications of heteropolysaccharides from lactic acid bacte-ria. Int Dairy J 2001;11:687–707.
[9] Arici M, Daglioglu O. Boza A lactic acid fermented cereal beverage as a tradi-tional Turkish food. Food Rev Int 2002;18:39–48.
10] Schlegel L, Grimont F, Ageron E, Grimont PAD, Bouvet A. Reappraisal of the tax-onomy of the Streptococcus bovis/Streptococcus equinus complex and relatedspecies: description of Streptococcus gallolyticus subsp. gallolyticus subsp. nov.,S. gallolyticus subsp. macedonicus subsp. nov. and S. gallolyticus subsp. pasteuri-anus subsp. nov. Int J Syst Evol Microbiol 2003;53:631–45.
11] TS, 1992. Boza, Institute of Turkish Standards, No. 9778, Ankara.12] Thyyagaraja N, Otani H, Hosono A. Studies on microbiological changes during
the fermentation of ‘Idly’. LWT 1992;25:77–9.13] Beasley S, Saris S, Per EJ. Nisin producing Lactococcus lactis strains isolated from
human milk. Appl Environ Microbiol 2004;8:5051–3.14] Smitinont T, Tansakul C, Tanasupawat S, Keeratipibul S, Nvarini L, Bosco M,
et al. Exopolysaccharide producing lactic acid bacteria strains from traditionalthai fermented foods: isolation, identification and exopolysaccharide charac-terization. Int J Food Microbiol 1999;51:105–11.
15] Van Geel-Schutten GH, Flesch F, ten Brink B, Smith MR, Dijkhuizen L. Screeningand characterization of Lactobacillus strains producing large amounts ofexopolysaccharides. Appl Microbiol Biotechnol 1998;50:697–703.
16] Dubois M, Gilles KA, Hamilton JK, Rebers PA, Smith F. Colorimetric methodfor determination of sugars and related substances. Anal Chem 1956;28:350–6.
17] Gotcheva V, Pandiella SS, Angelov A, Roshkova ZG, Webb C. Microflora identifi-cation of the Bulgarian cereal-based fermented beverage Boza. Process Biochem2000;36:127–30.
18] Todorov SD, Dicks LMT. Pediocin ST18, an anti-listerial bacteriocin produced byPediococcus pentosaceus ST18 isolated from boza, a traditional cereal beveragefrom Bulgaria. Process Biochem 2005;40:365–70.
19] Todorov SD. Diversity of bacteriocinogenic lactic acid bacteria isolatedfrom boza, a cereal-based fermented beverage from Bulgaria. Food Control2010;21:1011–21.
20] Akkoc N, Ghamat A, Akcelik M. Optimisation of bacteriocin production of Lacto-coccus lactis subsp. lactis MA23, a strain isolated from boza. Int J Dairy Technol2011;64:425–32.
21] Hancıoglu O, Karapınar M. Microflora of boza, a traditional Turkish beverage.Int J Food Microbiol 1997;35:271–4.
22] Mathara JM, Schillinger U, Kutima PM, Mbugua SK, Holzapfel WH. Isolation,identification and characterisation of the dominant microorganisms of kulenaoto: the Maasai traditional fermented milk in Kenya. Int J Food Microbiol
Please cite this article in press as: Heperkan D, et al. Evaluation of lacenzyme production as a potential adjunct culture. Process Biochem (2
2004;94:269–78.23] Papamanoli E, Tzanetakis N, Litopoulou-Tzanetaki E, Kotzekidou P. Char-
acterization of lactic acid bacteria isolated from a Grek dry-fermentedsausage in respect of their technological and probiotic properties. Meat Sci2003;65:859–67.
[
PRESSmistry xxx (2014) xxx–xxx 7
24] Samolada M, Litopoulou-Tzanetaki E, Xanthopoulos V, Tzanetakis N. Changes inmicrobial flora during manufacture of a traditional fermented milk from ewe’smilk. Food Microbiol 1998;15:43–50.
25] Silverstein RM, Webster FX, Kiemle DJ. Spectrometric identification oforganic compounds. 7th ed. New Jersey: John Wiley&Sons Inc.; 2011. p.72–98.
26] Steluti RM, Giese EC, Piggato MM, Sumiya AFG, Covizzi LG, Job AE, et al.Comparison of Botryosphaeran production by the ascomyceteous fungusBotryosphaeria sp., grown on different carbohydrate carbon sources, and theirpartial structural features. J Basic Microbiol 2004;44:480–6.
27] Wang LL, Wang LF, Ren XM, Ye XD, Li WW, Yuan SJ, et al. pH dependenceof structure and surface properties of microbial EPS. Environ Sci Technol2012;46:737–44.
28] Ismail B, Nampoothiri KM. Production, purification and structural characteri-zation of an exopolysaccharide produced by a probiotic Lactobacillus plantarumMTCC 9510. Arch Microbiol 2010;192:1049–57.
29] Ruas-Madiedo P, Hugenholtz J, Zoon P. An overview of the functional-ity of exopolysaccharides produced by lactic acid bacteria. Int Dairy J2002;12:163–71.
30] Tieking M, Korakli M, Ehrmann MA, Gänzle MG, Vogel RF. In situ productionof exopolysaccharides during sourdough fermentation by cereal and intestinalisolates of lactic acid bacteria. Appl Environ Microbiol 2003;69:945–52.
31] Palomba S, Cavella S, Torrieri E, Piccolo A, Mazzel P, Blaiotta G, et al.Polyphasic screening, homopolysaccharide composition, and viscoelasticbehaviour of wheat sourdough from a Leuconostoc lactis and Lactobacilluscurvatus exopolisaccharide-producing starter culture. Appl Environ Microbiol2012;78:2737–47.
32] Lindström C, Xu J, Öste R, Holst O, Molin G. Oral administration oflive exopolysaccharide-producing Pediococcus parvulus, but not purifiedexopolysaccharide suppressed Enterobacteriaceae without affecting bacterialdiversity in caecum of mice. Appl Environ Microbiol 2013;79:5030–7.
33] Degeest B, Vaningelgem F, De Vuyst L. Microbial physiology, fermentationkinetics and process engineering of heteropolysaccharide production by lacticacid bacteria. Int Dairy J 2001;11:747–57.
34] Laws AP, Marshall VM. The relevance of exopolysaccharides to the rheologicalproperties in milk fermented with ropy strains of lactic acid bacteria. Int DairyJ 2001;11:709–21.
35] Tsakalidou E, Zoidou E, Pot B, Wassill L, Ludwig W, Devriese LA, et al. Identifica-tion of streptococci from Greek Kasseri cheese and description of Streptococcusmacedonicus sp. nov. Int J Syst Bacteriol 1998;48:519–27.
36] Papadimitriou K, Ferreira S, Papandreou NC, Mavrogonatou E, Supply P, Pot B,et al. Complete genome sequence of the dairy isolate Streptococcus macedonicusACA-DC 198. J Bacteriol 2012;194:1838–9.
37] Masoud W, Vogensen FK, Lillevang S, Al-Soud WA, Sorensen SJ, Jakobsen M.The fate of indigenous microbiota, starter cultures, Escherichia coli, Listeriainnocua and Staphylococcus aureus in Danish raw milk and cheeses determinedby pyrosequencing and quantitative real rime (qRT)-PCR. Int J Food Microbiol2012;153:192–202.
38] Randazzo CL, Ribbera A, Pitino I, Romeo FV, Caggia C. Diversity of bacterialpopulation of table olives assessed by PCR-DGGE analysis. Food Microbiol2012;32:87–96.
39] Magnusson J, Schnürer J. Lactobacillus coryniformis subsp. coryniformis strain Si3produces a broad-spectrum proteinaceous antifungal compound. Appl EnvironMicrobiol 2001;67:1–5.
40] Van Hoorde K, Van Leuven I, Dirinck P, Heyndrickx M, Coudijzer K, Vandamme P,et al. Selection, application and monitoring of Lactobacillus paracasei strains asadjunct cultures in the production of Gouda-type cheeses. Int J Food Microbiol2010;144:226–35.
41] Argyri AA, Zoumpopoulou G, Karatzas KAG, Tsakalidou E, Nychas GJE, PanagouEZ, et al. Selection of potential probiotic lactic acid bacteria from fermentedolives by in vitro tests. Food Microbiol 2013;33:282–91.
42] Boekhorst J, Siezen RJ, Zwahlen MC, Vilanova D, Pridmore RD, Mercenier A, et al.The complete genomes of Lactobacillus plantarum and Lactobacillus johnsoniireveal extensive differences in chromosome organization and gene content.Microbiology+ 2004;150:3601–11.
43] De Vuyst L, Vrancken G, Ravyts F, Rimaux T, Weckx S. Biodiversity, ecologi-cal determinants, and metabolic exploitation of sourdough microbiota. FoodMicrobiol 2009;26:666–75.
44] Boone DR, Castelnhoz RW, Garrity GM. Bergey’s manual of systematic bacteri-ology. 2nd ed. New York: Springer; 2001.
45] Fortina MG, Ricci G, Foschino R, Picozzi C, Dolci P, Zeppa G, et al. Phenotypic typ-ing, technological properties and safety aspects of Lactococcus garvieae strainsfrom dairy environments. J Appl Microbiol 2007;103:445–53.
46] Russo G, Iannetta M, D’Abramo A, Mascellino MT, Pantosti A, Erario L, et al. Lac-tococcus garvieae endocarditis in a patient with colonic diverticulosis: first casereport in Italy and review of the literature. New Microbiol 2012;35:495–501.
47] Roces C, Campelo AB, Veiga P, Pinto JPC, Rodríguez A, Beatriz Martínez B. Contri-bution of the CesR-regulated genes llmg0169 and llmg2164-2163 to Lactococcuslactis fitness. Int J Food Microbiol 2009;133:279–85.
48] Mei HC, Liu YW, Yi-Chin Chiang YC, Chao SH, Mei NW, Liu YW, et al.Immunomodulatory activity of Lactococcus lactis A17 from Taiwan fermentedcabbage in OVA-sensitized BALB/c mice. Evid Based Complement Alternat Med
tic acid bacterial strains of boza for their exopolysaccharide and014), http://dx.doi.org/10.1016/j.procbio.2014.06.012
2013;2013:287803.49] O’Sullivan DJ, Giblin L, McSweeney LH, Sheehan J, Cotter PD. Nucleic acid-
based approaches to investigate microbial-related cheese quality defects. FrontMicrobiol 2013;4:1, http://dx.doi.org/10.3389/fmicb.2013.00001.
ING ModelP
8 Bioche
[
[
[
ARTICLERBI-10171; No. of Pages 8
D. Heperkan et al. / Process
50] Wacher-Rodarte C, Galvan MV, Farres A, Gallardo F, Marshall VME, Garcia-
Please cite this article in press as: Heperkan D, et al. Evaluation of lacenzyme production as a potential adjunct culture. Process Biochem (2
Garibay M. Yogurt production from reconstituted skimm ilk powders usingdifferent polymer and non-polymer forming starter cultures. J Dairy Res1993;60:247–54.
51] Holzapfel WH. Appropriate starter culture technologies for small-scale fermen-tation in developing countries. Int J Food Microbiol 2002;75:197–212.
[
PRESSmistry xxx (2014) xxx–xxx
52] Fusco V, Quero GM, Stea G, Morea M, Visconti A. Novel PCR-based identifica-
tic acid bacterial strains of boza for their exopolysaccharide and014), http://dx.doi.org/10.1016/j.procbio.2014.06.012
tion of Weissella confusa using an AFLP-derived marker. Int J Food Microbiol2011;145:437–43.
53] Katina K, Maina NH, Juvonen R, Flander L, Johansson L, Virkki L, et al. In situproduction and analysis of Weissella confusa dextran in wheat sourdough. FoodMicrobiol 2009;26:734–43.
