MANUSCRIPT ES&T Version 03
Exploring methane oxidizing communities for the co-
metabolic degradation of micropollutants
Running title: Co-metabolic degradation of sulfamethoxazole and benzotriazole by mixed
methane oxidizing communities
Jessica Benner1, Delfien De Smet1, Adrian Ho1, Frederiek-Maarten Kerckhof1, Lynn
Vanhaecke2, Kim Heylen3, Nico Boon1*
1Laboratory of Microbial Ecology and Technology (LabMET), Department of Biochemical
and Microbial Technology, Faculty of Bioscience Engineering, Ghent University, Coupure
Links 653, B-9000 Gent, Belgium
2 Laboratory of Chemical Analysis, Department of Veterinary Public Health and Food Safety,
Faculty of Veterinary Medicine, Ghent University, Salisburylaan 133, B-9820, Merelbeke,
Belgium
3Laboratory of Microbiology, Department of Biochemistry and Microbiology, Faculty of
Sciences, Ghent University, K.L. Ledeganckstraat 35, B-9000 Ghent, Belgium
* Correspondence to: Nico Boon, Ghent University; Faculty of Bioscience Engineering;
Laboratory of Microbial Ecology and Technology (LabMET); Coupure Links 653; B-9000
Gent, Belgium; phone: +32 (0)9 264 59 76; fax: +32 (0)9 264 62 48; E-mail:
[email protected]; Webpage: www.labmet.Ugent.be.
1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
ABSTRACT
Methane oxidizing cultures were enriched from five different inocula to be used for co-
metabolic degradation of micropollutants. In a first screening, 18 different compounds were
tested for degradation with the obtained cultures as well as with four pure methane oxidizing
bacteria (MOB) strains. The selected compounds are relevant in drinking water source
contamination, including pharmaceuticals, chemical additives, pesticides and their
degradation products. All enriched cultures were successful in the degradation of at least 4
different pollutants, but the compounds which were consistently degraded were
sulfamethoxazole (SMX) and benzotriazole (BTZ). The pure MOB cultures exhibited less
degradation potential, but SMX and BTZ were also degraded by 3 of the 4 tested pure strains.
For the enrichment cultures, which also contained heterotrophs , it was necessary to validate
which group were the main degraders of these compounds.. Addition of acetylene, a specific
methane monooxygenase (MMO) inhibitor, revealed that SMX and BTZ were primarily
degraded co-metabolically by the MOB.
For MOB, copper (Cu2+) concentration is an important factor, as someMOB have the ability to
express a soluble MMO (sMMO) if Cu2+ concentration is low. The sMMO has been described
to have a broader substrate range, possessing the ability to co-metabolically degrade
pollutants, even those with aromatic structures. This study also investigated the influence of
Cu2+ concentration on co-metabolic degradation of SMX and BTZ.
2
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
INTRODUCTION
Micropollutants, i.e. pollutants in concentrations of ng up to µg/L, often enter surface waters
through effluents of wastewater treatment plants, as many of these anthropogenic compounds
are not efficiently removed by the conventional treatment {Benner, 2013 #218}. The
ecological impact, as well as the impact on human health is difficult to estimate. Bank
filtration and percolation of water through the soil can lead to contamination of groundwater.
Groundwater as well as surface water are the most important sources for the production of
drinking water.
Apart from screening a wide spectrum of micropollutants in this study, specific tests were
done on the degradation of sulfamethoxazole (SMX) and benzotriazole (BTZ). SMX is an
antibiotic detected in wastewater treatment plant (WWTP) effluents and surface water, or
even in some source waters for drinking water (REF). Tests of aquifers have shown that it had
the same breakthrough behavior as a bromide tracer, which shows how easily this compound
can reach groundwater and with this drinking water sources {Barber, 2009 #126}. It is known
as a recalcitrant compound for its very limited biodegradability. {Liu, 2011 #117}. BTZ is
used in a variety of application such as de-icing of planes, but also as additive to plastics and
it has also been found in several water bodies, including ground water and drinking water
sources {Benner, 2013 #218}. However, as typical micropollutants the concentrations of BTZ
and SMX detected in the different water bodies were usually below the µg/L range. These
very low concentrations of pollutants might lower the chance of biological degradation
substantially, as there is hardly the need for present communities to specialize in degradation
as low concentration hardly ever pose an acute effect on organisms. The ability to degrade the
compounds can in most cases not be accounted as an evolutionary advantage, so enrichment
of specialized strains in unlikely. Next to this metabolic degradation, some organisms are also
3
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
capable to co-metabolize compounds, i.e. to degrade something without using it as energy or
carbon source. In this case, the low concentrations might even be advantageous. There would
be less competition between the primary substrate of the bacteria and the pollutant for the
reactive sites of the enzyme {Alvarez-Cohen, 2001 #220}.
Methane oxidizing bacteria (MOB) are known to oxidize components in addition to their
main growth substrate (methane, CH4), using methane monooxygenase (MMO), the key
enzyme for methane oxidation. Methanotrophic bacteria are grouped in three ‘Types”: Type I
methanotroph (e.g. Methylomonas) are known to fix carbon via the ribulose monophosphate
cycle, whereas the Type II methanotrophs (e.g. Methylocystis) use the serine cycle. Species
belonging to Type X (e.g. Methylococcus capsulatus) can show attributes of both other types
{Hanson, 1996 #207}. MOB can have two different types of MMO, either the particulate,
membrane associated particulate MMO (pMMO) or the soluble MMO (sMMO) which is
located in the cytoplasm. Nearly all known MOB can express pMMO, but only some have the
ability to express sMMO {Semrau, 2010 #130}. Especially for pollutant removal purposes
this differentiation is crucial, as studies have shown that these enzymes do not only differ in
location inside the cells, but also in substrate affinity and specificity{Semrau, 2011 #164}.
For oxidation of methane this means that sMMO has in general a higher turnover, but pMMO
has a lower affinity, so it can use also low concentrations of methane {Lee, 2006 #196}. In
general, sMMO is considered less specific, i.e. alkane with a structure of up to 8 carbon atoms
as well as aromatic structure have shown to be oxidized, but with pMMO only C-5 alkanes
were found to be oxidized and no aromatic compounds. However, in field conditions no
active sMMO could yet be identified but pMMO has found to be the active enzyme, e.g. in
TCE bioremediation studies {Lee, 2006 #196}.
The aim of this study was, to see whether a wide range of micropollutants can be removed by
mixed MOB cultures and to determine whether underlying pathway is due to co-metabolism.
4
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
In addition, influence of Cu2+ concentration and CH4 limitation were tested.
5
1
MATERIAL AND METHODS
1.1. ENRICHMENT OF METHANE OXIDIZING CONSORTIA
In 125 ml gas tight serum bottles, 30 ml of nitrate mineral salts (NMS) medium (see SIx) was
supplemented to the different inocula. After addition of 20% CH4 to the headspace they were
placed on a rotary shaker at 28°C for incubation. During enrichment CH4 and O2 content were
measured with a Compact GC (Global Analyser Solutions, Breda, The Netherlands), equipped
with a Molsieve 5A pre-column and Porabond column. Concentrations of gases were
determined by means of a thermal conductivity detector. Whenever either of the gases was
below 3 % (v/v) extra gas was added to the headspace. Whenever an enrichment showed high
turbidity or floc formation, fresh media was inoculated with 10% of this culture (one
enrichment cycle). Before starting any degradation test, this was at least done 5 times and the
culture was maintained either by successive cultivation as described above or stored at 4 °C,
followed by at least one further enrichment cycle before usage of culture for any degradation
tests.
Inocula originating from a top soil (SOIL), from rice growth substrate (RICE), from a biofilter
used at a drinking water production plant (BF) and from aerobic activated sludge of two
independent waste water treatment plants (WWTP1 and WWTP2) were used (more
information on location in SI). Enrichments were performed under three different enrichment
conditions: they were either grown in absence (no-MP) or presence of two distinct groups of
selected micropollutants (MixA; mainly composed of pesticides and pesticide metabolites:
2,6-dichlorobenzamide (BAM), bentazone (BNZ), mecoprop (MCP), linuron (LIN),
isoproturon(IPR), atrazine (ATZ); MixB; composed of chemical additives and pharmaceutical
active substances: benzotriazole (BTZ), methyl-benzotriazole (CH3-BTZ), 5,6-dimethyl-1H-
benzotriazole(dCH3-BTZ), 5-chlorobenzotriazole (Cl-BTZ), carbamazepine (CBZ),
6
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
primidone (PMD), diclofenac (DCF), sulfamethoxazole (SMX), iopamidol (IPA), iopromide
(IPR), diatrizoate (DIA), phenazone (PHZ)).
In addition, after all cultures were enriched, a mixture of all enrichments was made by mixing
equal volumes. This culture was then also used for degradation tests.
To test the influence of Cu2+ on the degradation capacity, the SOILno-MP culture was also
cultivated in NMS media which had no CuSO4 supplementation (SOILnoCu). The culture was
maintained as described above and test were only done after it was transferred at least 5 times,
always into fresh NMS media without Cu2+.
1.2. HIGH THROUGHPUT TEST ON CONCENTRATION INFLUENCE ON GROWTH INHIBITION
Media was prepared with the mix of IPR, BTZ, SMX, CBZ, BNZ, MCP and DCF at 0, 0.1,
0.2, 0.5 1, 2 and 5 mg/L per pollutant and inoculated on well plates with the following
cultures: WWTP1MixB, WWTP1no-MP, BFMixA, and BFno-MP. At day 3, 5 and 7 after incubation
under an atmosphere of 50% CH4, growth was followed up by measuring optical density at
620 nm with a Tecan sunrise well plate reader (Tecan Infinite M200 Pro; Tecan UK, Reading,
United Kingdom). Inhibition was expressed in relative term to the culture grown at 0 mg/L
pollutant.
1.3. DNA EXTRACTION,AND 16S RRNA GENE-DGGE ANALYSIS,
DNA-extraction was based on the Modified FastDNA® Spin kit (Q-BIO gene) (Details in
SIx) and 16 rRNA –gene based polymerase chain reaction (PCR) was conducted as
previously described by Boon et. al {Boon, 2000 #211}. The used primers, sequences as well
as temperature program can be found in TableS x.
DGGE (Denaturing Gradient Gel Electrophoresis) based on the protocol of Muyzer et al.
(1993) {Muyzer, 1993 #212} was performed using the INGENYphorU System (Ingeny
International BV, The Netherlands). PCR fragments were loaded onto 8% (w/v)
polyacrylamide gels in TAE buffer (20 mM Tris, 10 mM acetate, 0.5 mM EDTA pH 7.4). The
7
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
polyacrylamide gels were made with denaturing gradients ranging from 45% to 60% (where
100% denaturant contains 7 M urea and 40% formamide). The electrophoresis was run for 16
hours at 60°C and 120 V. Staining and analysis of the gels was performed as described
previously {Boon, 2000 #211}. The normalization and analysis of DGGE gel patterns was
done with the BioNumerics software 5.10 (Applied Maths, Sint-Martens-Latem, Belgium).
1.4. HIGH-THROUGHPUT SCREENING FOR POSSIBLE DEGRADATION OF MICROPOLLUTANTS
96-well plates were used for testing all combinations of the 15 enriched cultures and the
originally 18 selected compounds. Medium containing only one of the pollutants at
concentration of 100 µg/L was inoculated with 5% of the cultures and incubated for 1 week at
28°C under an atmosphere of 50% CH4. Optical density was measured right after inoculation
as well as at the end of the experiment as growth control. At the end of the tests samples were
taken and conserved as described below. As the small volumes did not allow a sampling right
after addition of inoculum, a control plate was prepared where instead of inoculum, the same
volume of pure NMS-media was added and placed in the same incubator to correct for
possible evaporation losses. Equal test setup was also applied for pure strains.
1.5. PURE MOB STRAINS
The following pure methane oxidizing bacteria strains were used for degradation studies:
Methylococcus capsulatus (NCIMB 11853T; Type X-MOB; MOB 1), Methylosinus
trichosporium (NCIMB 11131T, Type II-MOB; MOB 2) Methylocystis parvus (NCIMB
11129T, Type II-MOB; MOB 3), an isolate closest to type strain Methylomonas methanica
(98,6%) from Hoefman et al. {Hoefman, 2012 #200} (R-45374 = LMG 26616, Type I-MOB,
able to express sMMO; MOB 4). Until usage, they were cultivated on diluted nitrate mineral
salt (dNMS)-agar (see SI) plates in a sterile jar containing a 50% CH4-atmosphere. Liquid
suspensions of the cultures were prepared in sterile dNMS-medium prior to inoculation of
8
1
2
3
4
5
6
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
well plates.
1.6. ANALYSIS OF MICROPOLLUTANTS WITH LC-MS
In case of batch incubation tests, 300 µL sample were taken right after inoculation as well as
at the end of the incubation, for screening test the complete content of a well was used as
sample. To conserve the samples, 10 µL of 20 % formic acid was added and the mixture was
then filtered with a 0.20 µm syringe filter (Chromafil® PET-20/15 MS). Until analysis,
samples were stored at -20°C.
The compounds were all analysed in a single method using a Thermo Exactive mass
spectrometer (ThermoFish Scientific, Bremen, Duitsland) which uses exact mass
measurements for identification (Orbitrap detector).. The detailed method and specific
measurement parameters can be found in the SI (SIx).
1.7. DEGRADATION OF MICROPOLLUTANTS IN BATCH TESTS
General batch test with selected micropollutants
For more detailed degradation studies, incubations of 10 % inoculum in NMS medium were
performed in presence of IPR, BTZ, SMX, CBZ, BNZ, MCP and DCF (100 µg/L per
compound) with the selected cultures: RICEno-MP ,WWTP2MixB, BFMixA, RICEMixA, WWTP1MixB
as well as a mix of all enriched cultures. To determine the fractions of total degradation which
were due to heterotrophic activity or adsorption, 3 treatments were tested (each time in
independent triplicates): Treatment A was total degradation at 20 % CH4 in the headspace,
treatment B contained additional 2% C2H2 in the headspace as known inhibitor of the methane
monooxigenase (MMO) and in treatment C 0.16 g/L chloramphenicol (dissolved in DMSO)
was added to stop all bacterial activity. Incubation were done under the same conditions as the
enrichments and CH4 removal as well as O2 levels in the headspace were followed by CGC
measurement during one week. Whenever CH4 or O2 was below 3 %, additional CH4 or O2
was added.
9
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
The very same test setup was used for a test where SOILnoCu was inoculated with 10 µg/L per
compound of all originally selected compounds.
1.8. DEGRADATION BEHAVIOR TEST DURING LIMITATION OF PRIMARY SUBSTRATE
Using the same experimental setup as in the test with 100 µg/L per pollutant, BFMixA was
chosen to be exposed to 0, 2 and 20 % of CH4 in headspace during the degradation
experiment. The culture was first grown for 3 days at 20%, then it was divided into the test
setups in triplicates to ensure the same starting community as well as similar biomass content.
After 4 days the 20 % tests had depleted all CH4, which was then again added to result again
in 20% CH4 in the headspace.
1.9. STATISTICAL ANALYSIS OF CORRELATION BETWEEN ABSOLUTE CH4 OXIDATION AND
POLLUTANT DEGRADATION
In order to see whether this is a significant linear correlation different statistical approaches
were used. At first, a person Pearson correlation was used, but in order to do so, the tested
parameters need to follow normal distribution. This was tested with several different methods
and most of the data sets turned out not to be normally distributed (SI…). In that case Pearson
correlation cannot be applied, so instead a Spearman correlation was used. (Table 3)
1.10. PMOA-BASED MICROARRAY ANALYSIS
The DNA extracts were used as template for the diagnostic microarray analysis targeting the pmoA
gene (a subunit of the genes transcribing the pMMO) as described in detail before (Bodrossy et al.,
2003) with minor modifications {Ho, 2011 #226}. The pmoA gene was amplified using the
A189f/T7_A682r primer combination which has an extensive coverage of the methanotroph inventory.
Microarray data was analysed in R ver.2.10.0 (R Development Core, 2012) and visualized as a
heatmap using heatmap.2 implemented in gplots ver.2.7.4. The specificity and coverage of the probes
have been given in Ho et al. (2013){Ho, 2013 #205}.
10
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
RESULTS
Growth inhibition of elevated concentrations of pollutants
To investigate at which concentration of pollutants the communities would be inhibited and
whether they might have been able to adapt, growth was followed up in a media containing a
pollutant-mix (IPR, BTZ, SMX, CBZ, BNZ, MCP and DCF) at 0, 0.1, 0.2, 0.5 1, 2 and
5 mg/L per pollutant. Test were performed on 96-well plates with 4 cultures representing the
conditions enriched in presence of Mix A-pollutants or in presence of Mix B compared to the
same inoculum enriched without pollutants present. In Figure 1 the relative inhibition, i.e. the
increase or decrease of optical density of the culture exposed to the pollutant mixtures in
comparison to the non-exposed ones are shown.
After incubation of 7 days of RICEno-MP and RICEMixA, inhibition was detected for RICEMixA at
a much higher concentration than for the inoculum enriched without pollutants present. This
is even more pronounced when comparing the WWTP1 inoculum. None of the applied
concentration caused inhibition for WWTP1MixB. At 5 mg/L WWTP1MixB even shows better
growth than the non-exposed control.
High throughput degradation screening
To assess the degradation potential of all the different enrichments for the different
compounds, a high throughput incubation based on 96-well plates was used. As given in
Table 1, all enriched cultures were successful in the degradation of at least 4 different
pollutants. As the screening test can only give a rough idea about removal efficiencies, only
values above 20 % removal were accounted for positive removal. In total 11 pollutants could
be removed for ≥ 20%. When comparing the degradation success rate with the enrichment
background of the cultures, no increase in performance could be seen by enrichments done in
presence of the tested pollutants. Overall, only slight differences in the compounds which are
11
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
removed as well as in the extent of removal in the different cultures could be observed. With
an exception (MOB 2), all tested cultures removed approximately 90% of SMX . In addition,
most cultures were also able to remove BTZ (and derivatives), often also to a high extent. In
order to assess the underlying process of the removal, some compounds were selected either
based on the high extent of removal or their relevance as recalcitrant micropollutants for more
detailed batch incubation tests.
Batch incubation with selected compounds: elucidating the degradation pathways
As degradation achieved in the mixed cultures can also be due to metabolic activity of the
present heterotrophic bacteria in the community or adsorption onto biomass, further tests were
performed in batch scale. To distinguish the different processes, three treatments were tested:
Treatment A was the total degradation (sum of abiotic, heterotrophic and co-metabolic (by
MOB)) at 20 % CH4 in the headspace, treatment B contained additional 2% acetylene (C2H2)
in the headspace as inhibitor of the MMO {Bedard, 1989 #191}and in treatment C 0.16 g/L
chloramphenicol was added to inhibit any present bacteria. If the amount of compound lost
due to adsorption (treatment C) is subtracted of the one from treatment B, degradation based
on heterotrophic activity can be determined. If the degradation efficiencies in tests of
treatment A result in higher values than treatment B, the difference can be accounted for co-
metabolic oxidation by the present MOBs. Tests were performed with a selection of 5
enrichments , i.e. RICEno-MP ,WWTP2MixB, BFMixA, RICEMixA, WWTP1MixB as well as a mix of
all enriched cultures (Table 2). To combine pollutant concentration which high enough to be
analyze directly without enrichment (100 µg/L each pollutant) and to limit the total carbon
load in the test only 7 compounds, i.e. IPR, BTZ, SMX, CBZ, BNZ, MCP and DCF were
added. The compounds were selected based on either the results of the screening test (IPR,
BTZ, SMX) or relevance for drinking water sources (CBZ, BNZ, MCP, DCF).
12
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
After 7 days, BFMixA showed a removal of 41 ± 21 % of IPR in treatment A, but no significant
removal in treatment B or C was detected. However, all cultures showed a significant removal
of BTZ as well as SMX for the treatments (Table xx). SMX was always removed by more
than 90%. BTZ was degraded best by RICEno-MP and WWTP1MixB with 73 ± 15 % and 76± 8%,
respectively in total degradation tests. In Table xx removal efficiencies of the different
treatments and inocula are given.
No appreciable difference was observed in the degradation efficiency between incubations
supplemented with high (15.7 µmol/L) or low (0.157 µmol/L) Cu2+ concentrations (data not
shown).
Correlation of absolute CH4 consumption and BTZ/SMX oxidation
As nearly all degradation test showed high variability in both CH4 consumption as well as the
pollutant oxidation efficiency between the triplicates, test were repeated with only BTZ or
SMX present. Variation between independent triplicates were still observed and high standard
deviations for the calculated removal efficiencies were obtained. Nevertheless, BTZ/SMX
consumption and methane uptake were linearly correlated, and consistent between the
replicates (e.g. Figure S1x). Table 3 shows the results of the Spearman correlation tests
together with the p values, indicating significant correlation of the two values for all tested
data sets. These sets were taken from degradation tests where either BTZ or SMX were tested
individually, some with Cu2+ being present and some without Cu2+ and samples were taken at
several time points during incubation. The positive correlation shows that SMX and BTZ
degradation coincides to the CH4 oxidation activity of a culture.
Degradation batch test at 10 µg/L each compound
In the same batch incubation setup using SOILno-CU as inoculum, degradation with a starting
concentration of 10 µg/L per compound was investigated. As this lowered the total carbon
13
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
load, this test could be performed with all compounds originally used for the screening test.
Total degradation, heterotrophic as well as adsorption controls were analyzed after an
incubation of 7 days. CH4 oxidation profiles showed the same activity as in the test with
higher pollutant concentration, but the removal efficiencies differed. First of all, positive
degradation in total as well as heterotrophic treatment from 20-80 % could be seen for
compounds which so far did not show significant removal (Figure 5). This was the case for
DIA, PHZ, PMD, Cl-BTZ and CBZ. Also BTZ and SMX showed removal, but in this case
only SMX, Cl-BTZ and PMD showed a significant fraction being oxidized by co-metabolism.
Communities’ characteristics obtained from enrichments:
All enriched cultures consumed CH4 as shown in the methane depletion curve, thus MOB
were present in all cultures (see SI.x). With each enrichment step, the CH4-oxidation rate
increased, so that finally the added CH4 was depleted after 2-3 days of incubation in
comparison to 7-12 days for the original inoculation. The three different conditions (no-MP,
MixA, MixB) did not result in differences of CH4 oxidation rates (data not shown). This
indicates that the MOB communities of the cultures were not directly influenced by the
presence of micropollutants. Differences in community structure between the enriched
cultures were investigated by PCR-DGGE based on the 16 S rRNA gene. The chosen PCR-
condition and primers only distinguished between different bacteria in general, hence
including heterotrophs and MOB. Each of the 15 tested cultures gave a different pattern, so
inocula and enrichments technique influence the structure evolvement in all cases (Figure SI
1).
Microarray analysis of enriched MOB cultures and mixture of cultures
In contrast to the highly diverse DGGE profiles (Figure 2), the pmoA-based diagnostic
microarray analysis {Bodrossy, 2003 #206}, specifically targeting the aerobic methanotrophs,
14
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
showed similar community structure across all tested cultures as well as the mixture from all
the enrichments. All positive hybridization signals correspond to probes indicative for either
Methylocystis-like or Type II methanotrophs . Apart from some differences in the relative
abundance, the MOB community composition was consistent in the cultures. As the mixture
of all enrichments did not show appreciable signal for other MOB, it was assumed that none
of the cultures contained any significant amounts of MOB belonging to Type I or Type X.
Type II methanotrophs, including Methylocystis are known to have the potential to express
sMMO under Cu2+ limited conditions {Semrau, 2010 #130}. Previous studies have shown that
sMMO has a broader substrate range as well as a higher CH4 turnover. Therefore, further test
were performed to evaluate whether limitation of Cu2+ would lead to an increase of co-
metabolic degradation activity.
Batch test in Cu2+ limited condition
SOILnoCu was enriched in a medium not containing any Cu2+ for at least 3 enrichment cycles
before used for degradation tests with SMX. Comparing heterotrophic and abiotic controls for
this culture ruled out a significant loss of compound due to either adsorption or heterotrophic
degradation (data not shown). With this knowledge kinetic degradation test of SMX were
performed, both in presence as well as absence of Cu2+ in order to evaluate influences on the
co-metabolic degradation rate.CH4 oxidation as well as pollutants degradation were monitored
at several time points during an incubation of 8 days.
The addition of 18.8 µmol/L Cu2+ led to a prolonged lag-phase in CH4 oxidation of around 2-3
days before any CH4 oxidation was observed (Figure 3, maybe SI?). For the degradation of
SMX it could also be seen that addition of Cu2+ resulted in a slower removal as the incubation
without additional Cu2+ showed removal of 60 ± 2 % within less than 3 days, where in
presence of Cu2+ only 15± 7 % were removed. However, once the CH4 oxidation also started
15
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
(between day 3 and 4) SMX was degraded with and without Cu2+ to a similar extent of 80 ±
10 %. and 67 ± 9 %, respectively.
Pollutant degradation under limitation of methane
To investigate influence of available CH4 on the degradation efficiency of the investigated
compounds a new batch test was setup, but this time BFMixA was exposed to 0, 2 and 20 % of
CH4 in headspace. The culture was first grown for 3 days at 20%, and subsequently divided
into the test setups in triplicates. The relative CH4 removal during this test is shown in Figure
x.
Surprisingly, the 2 % sample did not show CH4 removal during incubation After incubation
for 8 days under CH4 limitation, all cultures were again exposed to 20 % CH4 and all still
showed very good CH4 removal capacity. The cultures which were exposed to the starvation
period show a slightly slower CH4 oxidation potential, but as the removal could not be
normalized to the present biomass, it is very likely that the cultures which were incubated at
20% CH4 during all the experiment contained more biomass (Figure 4).
After 3 and 7 days of incubation, degradation efficiencies were analyzed and no significant
removal for any other compound but BTZ was found (SMX analysis failed due to technical
problems). The removal efficiencies for BTZ at the two time points of all three CH4
concentrations are given in Table 4 (or Figure x). After 3 days of incubation, highest removal
was observed in the tests which had 20 % CH4 in the headspace with 94 ± 4 %, but after 7
days the efficiencies of test at 2 and 0% CH4 increase up to 84 ± x and 74 ± x %, respectively.
Screening pure cultures
As further confirmation of co-metabolic degradation, 4 pure MOB cultures (type x: MOB 1;
type II: MOB 2 and 3; type I: MOB 4) were exposed to the tested pollutants in the 96 well-
plate screening test (Table 1). In the case of the pure cultures, MOB 1, 3 and 4 showed very
16
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
high removal for SMX and MOB 1 and 4 removed more than 20% for BTZ. Only some of the
BTZ derivatives and IPR showed removal in few cases higher than 20%.. The degradation
pattern achieved by this screening test did not show differences between MOB belonging to
different groups.
17
1
2
3
4
5
6
DISCUSSION
The different growth inhibition for the pre-exposed cultures indicate that enrichment in
presence of pollutants caused the total community to adapt, leading to more resistance against
higher concentrations of the selected compounds. However, in the case of the MOB
community, the presence of pollutants did not show a significant effect. On the one hand, this
shows that the compounds do no inhibit the bacteria but also that very likely no adaptation of
the culture is necessary to implement them for degradation of these pollutants if co-
metabolism is the major pathway. In this case, the rather uniform MOB community might
explain the similarity in the degradation capacity of the different enrichments, even though
the total communities differed as shown in the DGGE analysis. The MOB community
structure was most probably shaped by the general enrichment conditions, such as high CH4
concentration and temperature. Also, Begonja et. al {Begonja, 2001 #104} have found a
predominance of Methylocystis-like species after enriching under similar conditions (i.e. high
CH4 concentration, limited O2 concentration and relative high incubation temperature).
The degradation tests with the selected pollutants at a starting concentration of 100 µg/L
identified co-metabolism as main pathway for the degradation of BTZ and SMX (table 2).
This was supported by the linear relationship of CH4 consumption and the degradation of both
compounds (Table 3). Limitation of methane resulted in decreased pollutant degradation.
However, even if no CH4 oxidation was taking place, BTZ was still removed by more than
70% after 7 days. As the MMO enzyme should still be present from the growth period prior to
the limitation-test, this shows that periods of CH4 limitation should not immediately stop
pollutant removal, and that the communities would be able to overcome shortages.
Influence of Cu2+
The original screening as well as the degradation batch tests were performed with a Cu2+
18
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
concentration of 15.7 µmol/L which is >15 times higher than concentration reported to inhibit
sMMO expression and activity (approximately 1 µmol/L){Begonja, 2001 #104}.
Unfortunately, quantitative sMMO analysis based on oxidation of naphthalene was not
feasible with the mixed cultures; we cannot exclude naphthalene oxidation by the
heterotrophs. In addition, protein analysis would overestimate the total protein content by
including the heterotrophic organisms. However, the higher Cu2+ concentration during
enrichment and the tests should eliminate any possible sMMO formation in the cultures,
indicating the relevance of the pMMO which may be differentially expressed among the
MOB upon copper amendment (Ho et al., 2013). Hence, assuming that the sMMO is not
active at high Cu2+ concentration while CH4 oxidation was detected, our results indicate the
predominance and activity of the pMMO. The chemical structure of SMX and BTZ (SI x)
both contain aromatic and heterocyclic structures. So far, literature has only reported of
oxidation of aromatic substrates by sMMO, and pMMO so far found not to be able to catalyze
oxidation of aromatic substrates{Semrau, 2011 #164}. This is the first report of oxidation of
aromatic compounds by MOB where sMMO cannot have been the active enzyme. Oxidation
most probably took place at the aromatic structure as in contrast to SMX, BTZ has no side
chains. As degradation of SMX and BTZ was also possible with the pure culture MOB 3,
Methylocycstis parvus, it is confirmed that pMMO can catalyze the oxidation of aromatic
structures, as this organism only able to express pMMO {del Cerro, 2012 #209}.
For cultures grown under Cu2+ limitation it is difficult to estimate whether in that case sMMO
has been expressed. From the microarray test it is not clear whether the type II methanotroph
enriched are able to express sMMO or not. Methanotrophs are known for the capability to
sequester Cu2+ with a compound called methanobactin {Semrau, 2013 #204} and maybe it is
possible that the stored Cu2+ was sufficient to sustain pMMO activity also under Cu2+
limitation. However, the occurrence of a lag-phase after addition of excess Cu2+ (Figure 3),
19
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
indicates that there is time necessary to adapt. One possibility would be that a switch from
sMMO to pMMO occurs and the lag-phase is due to the time needed to express enough
pMMO to be active. As SMX degradation is again occurring once the CH4 oxidation has
started again, this could indicate that both sMMO and pMMO are capable of co-metabolic
degradation of SMX. However, to support this hypothesis, sMMO activity would need to be
proven by a direct test which cannot be interfered by the present heterotrophic bacteria.
In the tests where only 10 µg/L were used as starting concentration degradation of several
compounds was found to be due to the metabolic activity of heterotrophic bacteria (Figure 5).
Even if co-metabolism plays only a minor role at low pollutant concentration, this can serve
as an example for a beneficial symbiosis of active heterotrophs and MOBs. The heterotrophic
members of this community will not survive on the low pollutant concentration but on the
compounds the MOB are leaking {Hesselsoe, 2005 #223} {van der Ha, 2013 #224}. It is
possible that this effect could not be observed at the higher pollutant concentration due to the
high standard deviations of the analysis. Another possibility could be to a very high affinity of
the heterotrophs towards the pollutants (only very small amounts of pollutant can possibly be
degraded) or occurrence of higher competition of pollutants among each other.
In conclusion, to our knowledge, this study demonstrates for the first time, the successful
cometabolic degradation of BTZ and SMX by methane-oxidizing cultures. Furthermore, tests
at lower micropollutant concentration indicated that these cultures might be able to degrade a
broader range of micropollutants. As most of the contaminated waters usually contain a
mixture of several compounds, this could be an advantage in comparison to other approaches
which use single strains specialized to degrade specific compounds. The survival of a diverse
community based on the oxidation of CH4 might open opportunities for application also in
oligotrophic condition, such as drinking water purification {Hesselsoe, 2005 #223}.
However, so far this study has only shown the proof of principle and in order to evaluate a
20
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
possible implementation of the cultures into regeneration of either WWTP effluents, surface
waters or drinking water sources, there is still a long way ahead. The degradation of
environmental relevant concentrations of all of the compounds need to be investigated and the
efficiency of the treatment at lower temperatures as well as lower CH4 concentrations need to
be done. It still needs to be investigated in which form these cultures can be implemented, e.g.
whether it is possible to integrate them into biofilms on biological active filter systems and
how sufficient CH4 transfer can be realized. Some ground waters which serve as source for
drinking water, have CH4 already present in the water {Darling, 2006 #225}. It should be
investigated whether the amounts would be sufficient to sustain a MOB culture which then
can co-metabolically oxidize present pollutants or help sustaining a pollutant degrading
heterotrophic community.
Apart from this, it should be noted that this approach is an oxidation step, which might lead to
the formation of unknown oxidation by-products with unpredictable properties. However,
studies have shown that oxidation often leads to a higher bioavailability and in a microbial
community other members might be capable of further degradation of the oxidized
products{Hrsak, 2000 #105} {Higgins, 1980 #221}(Ho et al., In press), further supportinghe
importance of co-cultures (MOB-heterotrophs; Ho et al., In press) than mono-cultures
(MOB). Formation, possible accumulation or further degradation of the transformation
products originating from the co-metabolic oxidation still needs to be investigated.
21
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
ACKNOWLEDGMENTS
This work was supported by the FP7 EU-project BIOTREAT (Contract number 266039; call
FP7- KBBE-2010.3.5.01). AH and FMK were supported by research grants from the
Geconcerteerde Onderzoeksactie (GOA) of Ghent University (BOF09/GOA/005).
We thank XXX for the useful suggestions, Tim Lacoere for assistance during the molecular
work, and …. for critically reading the manuscript.
22
1
2
3
4
5
6
REFERENCES
23
1
2
3
Tables: max 3
Table 1: Removal percentages of screening test of 15 MOB cultures in 96-well plates with
selected micropollutants. Green ≥90%, yellow ≥ 10% (no significant degradation) red no
detectable degradation. Yellow background: enrichment in presence of compounds of Mix B;
blue background: enrichment in presence or compounds of Mix A.
24
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
Iopa
mid
ol
diat
rizoa
te
iopr
omid
e
phen
azon
e
prim
idon
e
dicl
ofen
ac
SMX
CB
Z
BT
Z
BA
M
5-m
ethy
l BT
Z
5,6-
dim
ethB
TZ
5-ch
loro
BT
Z
bent
azon
e
isop
rotu
ron
atra
zine
mec
opro
p
linur
on
SOILMixB
BFMixB
RICEMixB
WWTP1MixB
WWTP2MixB
SOILMixA
BFMixA
RICEMixA
WWTP1MixA
WWTP2MixA
SOILno-MP
BFno-MP
RICEno-MP
WWTP1no-MP
WWTP2no-MP
Pure MOB 1
Pure MOB 2
Pure MOB 3
Pure MOB 4
25
1
2
Table 2: Degradation efficiencies of treatment A (total degradation), B (heterotrophic activity)
and C (abiotic control, adsorption) for RICEno-MP, RICEMIXA, BFMixA, WWTP2MixB, WWTP1no-
MP and a mix of all enriched cultures are given.
26
1
2
3
4
TreatmentTotal
degradation [%]
Heterotrophic
control [%]
Abiotic
control [%]Cometabolic
Total
degradation [%]
Heterotrophic
control [%]
Abiotic
control [%]Cometabolic
inocula SMX BTZ
RICEno-MP 100 ± 10 40 ± 10 33 ± 2 60 % 70± 20 12 ± 5 20 ± 10 58 %
WWTP2MixB 97± 8 42 ± 8 49 ± 1 45 % 60 ± 10 38 ± 8 10 ± 10 22 %
BFMixA 90 ± 10 47 ± 5 29 ± 3 43 % 50 ± 10 20 ± 30 0 ± 20 30 %
RICEMixA 95 ± 2 0 ± 20 0 ± 20 95 % 50 ± 10 0 ± 20 0 ± 30 50 %
WWTP1MixB 97 ± 4 0 ± 10 0 ± 60 97 % 76 ± 8 0 ± 20 0 ± 10 76 %
Mix of all cultures
98 ± 1 0 ± 40 0 ± 10 98 % 34 ± 5 0 ± 10 0 ± 20 34 %
27
1
23
Table 3 Spearman correlation and p value for CH4 consumption versus BTZ or SMX
oxidation
Dataset Spearman rho p-value
Prelim. BTZ test I [100 µg/L] no Cu2+ * 1.000 0.003
Prelim. BTZ test I[100 µg/L] no Cu2+ ** 0.783 0.004
BTZ test [250 µg/L] no Cu2+ 0.965 0.000
BTZ test [250 µg/L] 18.8 µmol Cu2+/L 0.685 0.035
SMX test [250 µg/L] no Cu2+ 0.525 0.016
SMX test [250 µg/L] 18.8 µmol Cu2+/L 0.818 0.002
* Test done on 25/9/12 (so far data not shown)
** Test done on 16/10/12 (so far data not shown)
Table 4:
20% CH4 2% CH4 0% CH4
Day
394 ± 4 62 ± x 29 ± x
Day
798 ± x4 84 ± x 74 ± x
28
1
2
3
4
5
6
7
8
9
Figures: max 4-5
29
1
2D Graph 3
mg/L per pollutant
0.1 0.2 0.5 1.0 2.0 5.0
% in
hibi
tion
-80
-60
-40
-20
0
20
40
60
80
100
day 3 day 5 day 7
RICEMixA
-40
-20
0
20
40
60
80
100
RICEno-MP
-20
0
20
40
60
80
100
WWTP1no-MP
-140
-120
-100
-80
-60
-40
-20
0
20
40
60WWTP1MixA
30
Figure 1. Growth inhibition in presence of different concentration of the batch-test
pollutant Mix measured as optical density in 96-wellplates. Error bars represent the
standard deviation of independent triplicates.
31
1
2
3
Figure 2: 16 S rRNA-gene based DGGEs of MOB culture used in former degradation batch tests (A) compared to pmoA-based microarray (B)
analysis showing the diversity of the methanotrophic community in the starting material and in response to amendments with micro-pollutants.
32
1
2
3
The microarray analysis is visualized as a heatmap, and shown here as the average of triplicate DNA extractions from independent batch
incubations. The intended probe specificity has been given elsewhere (Table S1; Ho et al., 2013). Red and blue in the color key denote high, and
no signal intensities, respectively. The microarray probes cover known type I and type II methanotrophs (Bodrossy et al., 2003). Probes
indicative for amoA (a gene encoding for ammonia monoxygenase), pmoA2 (an isozyme of the pmoA; Baani and Liesack (2008));
verrucomicrobial methanotrophs, and other environmental clusters without a defined function (methane or ammonium oxidizers) are grouped as
‘Others’.
33
1
2
3
4
5
6
7
time [h]
0 50 100 150 200
c/C
o C
H4
-0,2
0,0
0,2
0,4
0,6
0,8
1,0
% re
mov
al S
MX
-20
0
20
40
60
80
100
c/c0 CH4 SMX c/c0 CH4 SMX + Cu2+ %removal SMX % removal SMX + Cu2+
Figure 3: c/c0 of oxidized CH4 versus the removal % of SMX (start concentration 250 µ/L)
during 8 days batch test with 18.8 µmol Cu2+/L or without Cu2+.
Figure 4 Percentage CH4 removed during incubation with 0%, 2% and 20% CH4 in
headspace.
34
1
2
3
4
5
6
7
DIAPHZ
PMDBTZ
SMX
DCH3-BTZ
Cl-BTZ
CBZATZ
MCP
% re
mov
al
0
20
40
60
80
100
120TotalHeterotrophic controlAbiotic control
Figure 5: Removal of compounds in batch test with starting concentration of 10 µg/L per
compound; including controls for heterotrophic degradation as well as adsorption to biomass.
35
1
2
3
4
