Embed Size (px)
Citation preview
1
MICROBIAL COMMUNITY COMPOSITION AND DYNAMICS OF MOVING BED BIOFILM 1
REACTOR SYSTEMS TREATING MUNICIPAL SEWAGE 2
3
Running Title: Microbial communities of moving bed biofilm reactors 4
5
Kristi Biswas and Susan J. Turner 6
The Centre for Microbial Innovation, School of Biological Sciences, The University of 7
Auckland, Private Bag 92019, Auckland, New Zealand. 8
Correspondence to: Susan J. Turner, Email: [email protected], Phone: +64 9 3737599 9
Ext 82573, Fax: +64 9 3737416 10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
Copyright © 2011, American Society for Microbiology and/or the Listed Authors/Institutions. All Rights Reserved.Appl. Environ. Microbiol. doi:10.1128/AEM.06570-11 AEM Accepts, published online ahead of print on 2 December 2011
on January 22, 2018 by guesthttp://aem
.asm.org/
Dow
nloaded from
2
ABSTRACT 27
Moving Bed Biofilm Reactor (MBBR) systems are increasingly used for municipal and 28
industrial wastewater treatment, yet in contrast to activated sludge (AS) systems, little is 29
known about their constituent microbial communities. This study investigated the community 30
composition of two municipal MBBR wastewater treatment plants (WWTPs) in Wellington, 31
New Zealand. Monthly samples comprising biofilm and suspended biomass were collected 32
over a 12 month period. Bacterial and archaeal community composition was determined using 33
a full cycle community approach including analysis of 16S rRNA gene libraries, Fluorescence 34
in-situ Hybridization (FISH) and Automated Ribosomal Intergenic Spacer Analysis (ARISA). 35
Differences in microbial community structure and abundance were observed between the two 36
WWTPs and between biofilm and suspended biomass. Biofilms from both plants were 37
dominated by Clostridia and sulphate-reducing members of the Deltaproteobacteria (SRBs). 38
FISH analyses indicated morphological differences in the Deltaproteobacteria detected at the 39
two plants and also revealed distinctive clustering between SRBs and members of the 40
Methanosarcinales, which were the only Archaea detected and were present in low 41
abundance (< 5%). Biovolume estimates of the SRBs were higher for biofilm samples from 42
one of the WWTPs which receives both domestic and industrial waste, and is influenced by 43
seawater infiltration. The suspended communities from both plants were diverse, and 44
dominated by aerobic members of the Gamma- and Betaproteobacteria. This study 45
represents the first detailed analysis of microbial communities in full-scale MBBR systems and 46
indicates that this process selects for distinctive biofilm and planktonic communities, both of 47
which differ from those found in conventional AS systems. 48
49
on January 22, 2018 by guesthttp://aem
.asm.org/
Dow
nloaded from
3
INTRODUCTION 50
The Moving Bed Biofilm Reactor (MBBR) system was developed in the late 1980’s for the 51
treatment of domestic and industrial wastewaters. These systems are now operating in more 52
than 22 countries (including New Zealand) and range from large to small scale wastewater 53
treatment plants (WWTPs) (35). The MBBR process combines features of both fixed-growth 54
and activated sludge (AS) systems in that the microbial community is largely retained within 55
the reactor as a biofilm on suspended carriers, with a smaller planktonic fraction being 56
present in suspension as free-floating cells or small flocs. MBBR technology offers a number 57
of advantages over conventional technologies for treating waste including a high effluent 58
quality, no bulking problems and lower cost. Other advantages of fixed biofilm growth include 59
the decoupling of biomass retention from hydraulic retention time leading to longer sludge 60
ages and low waste sludge volumes (5, 37). 61
62
Studies on the microbial community composition of conventional activated sludge systems 63
indicate that the community is typically dominated by aerobic or facultatively anaerobic 64
heterotrophic bacteria belonging to the Betaproteobacteria (41). It is unclear whether similar 65
communities are found in MBBR processes as there have been no microbiological studies on 66
full-scale wastewater systems. Differences in microbial communities might be expected given 67
that MBBR systems support development of microbial biofilms within which 68
microenvironments support the growth of both anaerobic and aerobic organisms within the 69
same ecosystem (19). Molecular techniques such as FISH and denaturing gel gradient 70
electrophoresis (DGGE) have demonstrated the presence of organisms associated with 71
simultaneous nitrification and denitrification within a lab-scale continuous-flow MBBR system 72
treating wastewater under aerobic conditions (16). The presence of a reduced oxygen 73
gradient within biofilms has the potential to also support the growth of anaerobic ammonium-74
on January 22, 2018 by guesthttp://aem
.asm.org/
Dow
nloaded from
4
oxidizing bacteria (anammox) (47). Anammox is a process that is of significant recent interest 75
in the field of wastewater treatment because of the energy efficiencies that can be achieved in 76
the conversion of ammonium and nitrite to nitrogen gas under anaerobic conditions (46). 77
78
Biofilm development is a key process in the establishment of an effective MBBR process. To 79
maintain effective gas and nutrient transfer the ideal biofilm is relatively thin and evenly 80
distributed over the carrier surface (31). This can be influenced by turbulence in the reactors 81
which also influences substrate and oxygen transfer. Aside from these few key factors the 82
effects of other operational parameters and influent composition on microbial community 83
structure and function within MBBR systems are poorly understood. 84
85
The aim of this study was to investigate the microbial communities in Wellington’s Moa Point 86
(MP) and Karori WWTPs and thus to provide the first comprehensive insight into the key 87
microbial groups in full-scale MBBR systems. Approximately monthly samples, consisting of 88
suspended biomass and biofilm scraped from carriers, were collected over a 12 month period 89
from the two plants. A full cycle community analysis approach including analysis of 16S rRNA 90
gene libraries, Fluorescence in-situ Hybridization (FISH) and Automated Ribosomal Intergenic 91
Spacer Analysis (ARISA), was used to determine bacterial and archaeal community 92
composition and dynamics. Total dissolved sulfides and dry/wet weight of biofilm adhering to 93
carriers were also determined. Microbial community analysis was also performed on a sample 94
from a conventional floc-based activated sludge system for comparison. 95
96
MATERIALS AND METHODS 97
Sample sites 98
Sampling was carried out at Wellington’s MP and Karori WWTPs. The MBBR reactors at both 99
plants contained suspended polyethylene carriers (K1 media, AnoxKaldnesTM) comprising 100
on January 22, 2018 by guesthttp://aem
.asm.org/
Dow
nloaded from
5
30–50% of reactor volume. Reactor samples, comprising suspended K1 carriers with 101
adherent biofilm, were collected once a month over a year from three MBBR reactors at MP 102
(designated as M1, M2, & M3) and two reactors from Karori (designated K1 & K2) treatment 103
plants. For comparison, mixed liquor from a conventional activated sludge system was 104
collected from a large municipal WWTP in northern New Zealand. Samples were collected in 105
one liter bottles and transported refrigerated overnight to the laboratory. 106
107
Physical and Chemical analyses 108
Dry and wet weight determinations were made on biofilm from five K1 carriers from each 109
sample. To determine the wet weight, carriers with adherent biofilm were blotted dry on tissue 110
paper for 2 min and then weighed. The samples were then transferred to a desiccator and 111
dried for one week, then re-weighed. Control samples comprising five unused carriers were 112
also subjected to desiccation and then weighed. Dry and wet weight determinations were 113
made following subtraction of the average weight of the control carriers. The statistical 114
significance of differences between samples was determined using a T-Test (unequal 115
variance). To measure the suspended biomass, 1mL of sample was pelleted by centrifugation 116
at 13,000xg for 10 min in a 1.5mL microfuge tube. The supernatant was carefully removed, 117
the wet weight determined, and the opened tube subjected to desiccation, then weighed as 118
described above. 119
Total dissolved sulfide was measured immediately upon receipt of samples by a colourimetric 120
method as outlined previously (13). A standard curve was prepared using varying 121
concentrations of sodium sulfide. The absorbance of copper sulfide precipitate for each 122
sample was measured at 460nm in a UV-vis Spectrophotometer. 123
124
on January 22, 2018 by guesthttp://aem
.asm.org/
Dow
nloaded from
6
DNA extraction 125
Total genomic DNA was extracted from biomass using a phosphate, SDS, chloroform-bead 126
beater method as described previously (44). Extracted DNA was eluted in 50 µl of DNase-free 127
water and stored at -20oC until further analysis. 128
129
Automated Ribosomal Intergenic Spacer Analysis (ARISA) 130
Automated Ribosomal Intergenic Spacer Analysis (ARISA) was used as a rapid method to 131
profile bacterial community structure and to make comparisons between monthly samples. 132
The intergenic spacer region between the 16S and 23S rRNA genes was amplified using two 133
universal bacterial primers SDBact and LDBact (33), in a PCR reaction described previously 134
(21). The fluorescently-labelled products were purified using a QIAquick® PCR purification Kit 135
(Qiagen) and analyzed along with an internal LIZ1200 standard on a 3130XL Capillary 136
Genetic Analyzer using a 50 cm capillary (Applied Biosystems Ltd., NZ). 137
Results from ARISA were analyzed using Genemapper software (v 3.7) to create bacterial 138
community profiles for each sample. Multi-dimensional scaling (MDS) plots were constructed 139
from community profile data using Primer 6 software (v 6.1.6). Manhattan distance was 140
chosen as the measure between samples on the MDS plots. 141
142
16S rRNA GENE ANALYSIS 143
Clone library analysis 144
Cloning and RFLP analysis of PCR-amplifed 16S rRNA genes was performed as generally 145
described previously (7). Forward and reverse primers for PCR amplification of bacterial 16S 146
rRNA genes were 5’- AGRGTTTGATCMTGGCTCAG-3’ and 5’- GKTACCTTGTTACGACTT-147
3’ respectively (39). Primers used for analysis of archaeal 16S rRNA gene sequences were 148
5’-TTCCGGTTGATCCYGCCGA-3’(Arch21F) and 5’-YCCGGCGTTGAMTCCAATT-3’ 149
(Arch958R) (15). Cloned inserts were recovered by PCR amplification using the vector-150
on January 22, 2018 by guesthttp://aem
.asm.org/
Dow
nloaded from
7
specific primers PGEM-F (5’-GGCGGTCGCGGGAATTGATT-3’) and PGEM-R (5’-151
GCCGCGAATTCAACTAGTTGATT-3’) (1). 152
153
Restriction Fragment Length Polymorphism (RFLP) of clones 154
Restriction endonuclease Hae III (Invitrogen) digestion was used to generate RFLP profiles 155
for archaeal clones to investigate diversity prior to selection of clones for sequencing as 156
previously described (7). The resulting products were resolved and visualized by 157
Polyacrylamide Gel Electrophoresis (PAGE) through 6% non-denaturing gels as described 158
previously (38). Gels were run at 120 V for 70 min and then stained with ethidium bromide. 159
Unique RFLP profiles were designated as operational taxonomic units (OTUs) and 160
representative clones selected for sequencing. 161
162
Sequencing and data analysis 163
PCR-amplified inserts from representative clones were purified and sequenced in one 164
direction using the aforementioned vector-specific (pGEM-F) primer. Purification and 165
sequencing was performed under contract by Macrogen Inc (Kumchun-Ku, Seoul, South 166
Korea) using a 3730 x 1 DNA Analyzer (Applied Biosystems). A total of 96 clones were 167
sequenced from each bacterial clone library, whereas one clone representing each OTU was 168
sequenced from the Archaea libraries. 169
Sequence data was processed using the high-throughput pipeline available through the 170
Ribosomal Database Project II (http://rdp.cme.msu.edu) (49) and also compared with the 171
GenBank database sequences (http://www.ncbi.nlm.nih.gov) using nucleotide-nucleotide 172
BLAST (2). 173
174
Phylogenetic analysis 175
on January 22, 2018 by guesthttp://aem
.asm.org/
Dow
nloaded from
8
Sequences were screened for chimeras using Mallard software (version 1.02; School of 176
Biosciences, Cardiff University [http://www.bioinformatics-toolkit.org/Mallard/index.html]) (6) 177
and unreliable sequences eliminated from further analysis. The remaining partial and full 178
length 16S rRNA sequences were aligned using the SINA web alignment tool, SILVA 179
(www.arb-silva.de/aligner) (32). These sequences were imported into the ARB software (25) 180
along with the SSU ref database (SILVA version 106) to construct phylogenetic trees. The 181
SSU ref database contains more than half a million curated full length (>1200 bp) 16S rRNA 182
sequences that have been previously published or uploaded to public databases (such as 183
GenBank). Phylogenetic trees were constructed using Maximum Likelihood methods. 184
185
Fluorescence In Situ Hybridization (FISH) 186
FISH was used in combination with confocal laser scanning microscopy to examine samples 187
for the presence of methanogens and SRBs (4, 27). Samples were fixed within 12 h of sample 188
collection using 4% paraformaldehyde as described previously (27), and stored at -20oC in a 189
1:1 PBS and ethanol mixture awaiting further analysis. Slides were prepared from fixed 190
samples by placing 20µL of sample into 6 mm diameter wells on Teflon ® coated slides 191
(ProSciTech) which were then air dried. Samples were dehydrated by immersion in 50%, 80% 192
and 98% ethanol respectively for 3 min each. Hybridization was carried out in a 50m Falcon 193
tube chamber at 46oC for 2 h. Oligonucleotide probes were derived from published sequences 194
with a 5’-fluorescent label (specified in Table 1) and synthesized by Thermo Fisher Scientific 195
(Germany). Each well was hybridized with buffer (0.9 M NaCl, 0.01% SDS, 20mM Tris-HCl, 196
formamide at optimal concentrations for each probe) and 1µL of 50ng/µL probe. Following 197
incubation, slides were immersed in pre-heated (48oC) wash buffer (20mM Tris-HCl, 5mM 198
EDTA, NaCl at optimal concentrations for each probe) and air dried in the dark. Salts were 199
removed from the slides by washing in ice-cold distilled water. DAPI (10µg/µL) was then 200
on January 22, 2018 by guesthttp://aem
.asm.org/
Dow
nloaded from
9
applied as a universal DNA stain. Slides were incubated in the dark for 5 min then rinsed with 201
distilled water and air dried. 202
203
Confocal microscopy 204
FISH slides were viewed using an FV1000 Olympus confocal laser scanning microscope 205
(CLSM) equipped with a 100x oil immersion objective lens. Excitation of FITC, Cy3 and Cy5 206
was performed at 488 nm (Ar-laser), 543 nm (He-Ne laser), and 635 nm (red diode laser), 207
respectively. Images were viewed by using an FV10-ASW2.0 (Olympus) viewer. 208
For quantification, at least five z-series of 1 µm depth (6-10 sections) were made for each 209
sludge sample with a universal stain (DAPI) and a group-specific probe (MSMX860 or 210
Delta495a). The percentage biovolume was calculated by using DAIME software (Version 1.2, 211
http://www.microbial-ecology.net/daime) (14). 212
213
RESULTS 214
Sampling was carried out at Wellington’s MP and Karori WWTPs. Both plants are configured 215
to include biological treatment in the form of aerated moving bed bioreactors. Solids are 216
removed by contact stabilization and clarification. MP receives household and industrial 217
waste, including that from an abattoir, and has an average dry weather flow of 822 L/s with a 218
BOD of 0.23 kg/m3. The Karori WWTP receives only domestic waste with an average dry 219
weather inflow of 20 L/s with a BOD of 0.37 kg/m3. Dissolved oxygen levels were regulated at 220
similar levels between the two MBBR facilities of interest, with an annual average of 1.56 221
mg/L for MP and 1.45 mg/L for Karori. Elevated conductivity levels (yearly average 4.072 222
mS/cm), indicative of seawater infiltration, are consistently detected at the MP plant while 223
levels in influents reaching the Karori plant are low and typical of domestic effluents. A total of 224
11 approximately-monthly samples were collected from each of the two treatment plants over 225
on January 22, 2018 by guesthttp://aem
.asm.org/
Dow
nloaded from
10
a period of 12 months. Samples were taken from the aeration tank of the MBBRs and 226
comprised K1 carriers with adherent biofilm and reactor fluid. 227
228
General characteristics of biofilm and suspended biomass 229
The colour and odour of biofilms attached to K1 media differed between sites. At MP the 230
biofilms were black in colour and had a sulfurous odor. In contrast, biofilm samples from the 231
Karori WWTP were consistently greyish-brown and had no obvious odour. A comparison of 232
biofilm quantity was made by determining the wet and dry weight of biofilms from each MBBR 233
for all time points. No significant differences (p = 0.124) were observed between the dry 234
weight of the biofilm samples from MP (0.029 ± SD 0.033 g, n=8) and Karori (0.013 ± SD 235
0.006 g, n=8). Similarly, no significant differences were observed between the wet weights of 236
biofilm from the two sites (p = 0.379). P-values < 0.05 were considered significant. 237
To provide an estimate of the amount of suspended biomass, dry and wet weight 238
determinations were also made on pellets prepared from 1 mL of supernatant fluids. 239
Supernatant dry weight values were low (<0.021 g/mL) for all samples. No significant 240
differences (p = 0.427) were found between samples from the two MBBR systems. 241
Total dissolved sulfides were detected in all samples from the MP site with values ranging 242
between 1.5 to 13 mM. In contrast sulfide was not detected (<1 mM) in samples collected 243
from the Karori plant. Significant differences (p = 0.00012) were observed in sulphide values 244
between the two MBBR sampling sites. 245
246
BACTERIAL COMMUNITY ANALYSIS 247
ARISA 248
ARISA was used as a rapid method to profile and compare the bacterial community structure 249
in both biofilms and suspended biomass. Multi-dimensional scaling (MDS) was used to 250
investigate differences between ARISA community profiles over time and between treatment 251
on January 22, 2018 by guesthttp://aem
.asm.org/
Dow
nloaded from
11
plants. These data indicated differences between the two WWTPs and also between biofilm 252
and suspended communities (Fig. 1). 253
254
16S rRNA gene analysis 255
To determine the bacterial community composition, clone libraries of PCR-amplified 16S 256
rRNA genes were prepared and sequenced from biofilm scraped from carriers and from 257
suspended biomass. This analysis was performed on reactor samples collected at three time 258
points (March-2010, August-2010, January-2011). Figure 2 presents a summary of phyla 259
detected for a representative sample from each plant. Phylogenetic trees constructed using 260
the entire sequence dataset are presented for all phyla (Fig. 3), the Deltaproteobacteria (Fig. 261
4) and the Epsilonproteobacteria (Fig. 5). A phylogenetic tree of sequences aligning to the 262
Firmicutes is presented in Appendix 1. 263
Each treatment plant yielded a characteristic community composition that was consistent 264
between reactors and time points (Fig. 2A). Biofilm communities from both plants showed 265
limited bacterial diversity (Shannon-Wiener index of 0.93 - 1.18) and were dominated by 266
Firmicutes (40-44% of clones). Phylogenetic alignment of sequences indicated a broad 267
diversity within the Firmicutes (Appendix I) although the majority of clones represented 268
members of the Clostridia (38% of clones). Strictly anaerobic, sulfate-reducing bacteria (SRB) 269
belonging to the class Deltaproteobacteria were the second most abundant group, comprising 270
29.4% of the biofilm clone library at MP and 24.7% at Karori. Members of Desulfobacterales 271
(11-19 %), Syntrophobacterales (8-10%) and Desulfovibrionales (0.5-1.5%) were the most 272
abundant SRBs found within these samples (Fig. 4). The biofilms from Karori WWTP differed 273
due to an elevated incidence of Beta- and Gammaproteobacteria (17.6% of clones), which 274
were both in lower abundance (<7.4%) in the biofilm community from MP. A number of other 275
organisms including representatives of the phyla Bacteroidetes, Synergistes, Planctomycetes, 276
Verrucomicrobia, and Acidobacteria, as well as various unclassified bacteria, were also 277
on January 22, 2018 by guesthttp://aem
.asm.org/
Dow
nloaded from
12
detected at low abundance (<5% of clones) in the biofilm samples from both plants. Analysis 278
of the archaeal community from the two MBBR systems indicated a very limited diversity, with 279
only one RFLP pattern detected among the 20 clones screened. Sequence analysis indicated 280
that this pattern represented organisms from the methanogen order Methanosarcinales. 281
In contrast to the biofilm samples, the suspended biomass from both MBBR systems (Fig. 2B) 282
was dominated by a more diverse group of aerobic organisms from the Alphaproteobacteria 283
(Rhizobiales and Rhodobacterales), Gammaproteobacteria (Pseudomonadales and 284
Aeromonadales) and Betaproteobacteria (Burkholderiales and Rhodocyclales). Members of 285
the Clostridia, comprising majority of the Fimicutes, were also present at MP (14% of clones) 286
and Karori (40% of clones) treatment plants. In addition, MP samples had an elevated level of 287
Epsilonproteobacteria (54% of clones) which were affiliated with the Campylobacteraceae and 288
aligned most closely to Arcobacter spp (Fig. 5). Though a diversity of sequences were 289
observed, the majority aligned most closely to clones obtained from estuarine and river 290
environments or to Arcobacter nitrofigilis. 291
The clone library prepared for comparative purposes from a mixed liquor sample from a 292
conventional AS plant (Fig. 2A) was more diverse and dominated by members of the 293
Betaproteobacteria (28%) and Bacteroidetes (25%). 294
295
FISH analysis 296
FISH was used as a PCR-independent approach to validate the 16S rRNA gene library 297
results, and to investigate the spatial distribution and abundance of Archaea using a 298
Methanosarcinales probe and putative sulfate reducing bacteria using a Deltaproteobacteria 299
probe. FISH analysis was performed on samples collected between Mar-2010 to Jun-2010, 300
Sept-2010, Dec-2010 and Jan-2011. 301
The distinctive black biofilm from MP revealed colloidal clusters of Deltaproteobacteria buried 302
within the biofilm structures (Fig. 6A). Based on the clone library results, these are likely to be 303
on January 22, 2018 by guesthttp://aem
.asm.org/
Dow
nloaded from
13
sulfate reducing bacteria. Samples from Karori also showed the presence of 304
Deltaproteobacteria but these were fewer in abundance and were filamentous (Fig. 6B). 305
Biovolume analysis confirmed the occurrence of Deltaproteobacteria at both treatment plants, 306
with higher volumes recorded in samples from MP WWTP (Fig. 6). However, there was no 307
significant (p = 0.282) difference observed between the biovolume of Deltaproteobacteria 308
between the two study sites. 309
Methanosarcinales were observed in low numbers (<5% biovolume) in biofilms from the two 310
MBBR systems. However, in samples from the MP plant these were observed in distinctive 311
clusters around cells hybridising with the Deltaproteobacteria probe, which represent putative 312
SRBs (Fig. 6A). 313
314
DISCUSSION 315
Since their invention, MBBR systems have been used to treat both municipal and a wide 316
range of industrial wastes including pulp and paper (43), cheese factory waste (36), and 317
phenolic wastewater (11). Previous studies (35, 51) have reported on the physical and 318
engineering aspects of MBBR systems while the composition of the microbial communities 319
responsible for the biological activity has received no attention in full-scale systems treating 320
municipal wastewater. This study set out to investigate the prokaryotic community dynamics 321
in a parallel study of two MBBR treatment plants over a 12 month period. The MBBR 322
treatment systems included in this study are situated in the same city and are operated under 323
similar parameters to treat urban municipal wastewater. The major operational difference 324
between the two plants is that the Karori plant receives effluent from a largely residential 325
catchment, while the MP plant services a much larger catchment that includes mixed urban 326
and industrial uses. 327
328
Biofilm communities 329
on January 22, 2018 by guesthttp://aem
.asm.org/
Dow
nloaded from
14
Differences in bacterial community profiles between the two WWTPs were expected due to 330
the different sources of waste entering these systems. The results of this study indicated no 331
difference in the amount of biomass retained on carriers, but distinct differences were noted in 332
both colouration and dissolved sulfide concentrations. The observation that MP biofilms were 333
black in colour and sulfurous in odour is consistent with the detection of appreciable levels of 334
dissolved sulfides at this site. The detection of sulfides at MP indicates the possible activity of 335
anaerobic SRBs, which are commonly found in anoxic systems (8, 29). Sulfate is reduced to 336
sulfide by SRBs, resulting in a pungent sulfurous odour and upon reaction with metals yields 337
black precipitates. 338
Despite differences in the colour of biofilms from the two sites, both yielded bacterial 16S 339
rRNA gene libraries that were dominated by anaerobes, notably Clostridia and members of 340
the Desulfobacterales and Syntrophobacterales – known sulfate-reducing bacteria (29). Both 341
plants also showed similar communities of Archaea, which were dominated by 342
Methanosarcinales. Validation of these results was performed by FISH using probes targeted 343
to the Methanosarcinales and the Deltaproteobacteria. Although the Deltaproteobacteria 344
probe is not specific for sulfate-reducing bacteria, these were the only members of the class 345
detected in the clone libraries. It was therefore presumed that cells hybridizing to the 346
Deltaproteobacteria probe were SRBs. Biovolume analysis confirmed the presence of SRBs 347
in both sample sets but indicated a higher abundance in the MP site and different cell 348
morphology to that seen in the Karori samples. These differences may go some way towards 349
explaining the higher levels of sulfides detected at the MP site. It is also possible that 350
differences in influent composition influence the community structure and function in these 351
biofilms. In addition to domestic wastewater, the MP plant receives industrial waste from an 352
abattoir, expected to carry large amoun ts of fats, oils and greases (28). The presence of long 353
chain fatty acids (LCFA) derived from the degradation of lipids (45) and low water 354
temperatures have been shown to select for SRBs in a wastewater treatment system (9, 22). 355
on January 22, 2018 by guesthttp://aem
.asm.org/
Dow
nloaded from
15
Enhanced SRB activity would also require access to excess sulfate as an electron acceptor. 356
The source of this in the MP plant has not yet been clearly established, although seawater 357
infiltration is suspected in this system, as indicated by elevated conductivity levels. Seawater 358
is known to contain appreciable concentrations of sulfate and high concentrations have been 359
found in an MBBR system treating waste from a marine aquarium (20). 360
FISH and biovolume analysis confirmed the presence of members of the Methanosarcinales 361
in approximately equal abundance at both sites. This analysis also revealed a clear 362
association of these methanogens with Deltaproteobacteria in the MP samples. The majority 363
of SRBs described in the past are putatively free-living (29), but recent studies of the marine 364
environment using FISH have revealed consortia of SRBs and methanogens (10, 18), 365
suggesting a symbiotic bacterial-archaeal interaction. The physiological basis of this 366
symbiosis is still a matter of conjecture. Although methanogens were detected at the Karori 367
site, these were not observed in association with the filamentous SRBs, further supporting the 368
notion that this community differs from that found in the MP samples. Filamentous SRBs 369
belonging to the genus Desulfonema have been detected previously in marine and freshwater 370
sediments (17, 50). Members of this genus exhibit gliding motility which is presumed to confer 371
the ability to move along gradients within dense mats and avoid predation by grazing 372
protozoa. These features would be equally advantagous in both MBBR biofilms, though 373
filamentous SRBs were only detected in the Karori samples. This further supports the notion 374
that the biofilm communities in these two systems are influenced by external factors which 375
may include differences in influent composition. Indeed, other studies have confirmed that 376
specialized groups of microorganisms develop over time in response to changes in complex 377
organic matter entering the system (40). 378
379
Suspended biomass 380
on January 22, 2018 by guesthttp://aem
.asm.org/
Dow
nloaded from
16
Clone library analysis indicated that the suspended fraction was dominated by aerobic 381
members of the Alpha-, Beta-, Gamma- and Epsilonproteobacteria. Members of the Clostridia 382
were also present but in much lower abundance than in biofilm samples, possibly resulting 383
from biofilm detachment processes which are known to occur periodically in such systems 384
(23). A notable difference between the two plants was the high abundance of sequences in 385
the MP clone library that aligned to Arcobacteria spp including Arcobacter nitrofigilis. The 386
genus Arcobacter was proposed in 1991 (48) to describe a group of aero-tolerant members of 387
the Campylobacteraceae and has been of increasing interest as an emerging zoonotic 388
pathogen. The genus includes species isolated from a diverse range of aquatic and terrestrial 389
environments as well as the faeces and reproductive tracts of domestic livestock, sewage and 390
abbatoir effluents. The genus type species, Arcobacter nitrofigilis, is a nitrogen-fixing 391
organism that was isolated from the root zone of a salt-marsh plant although not all species 392
are salt tolerant (12). The reason for the abundance of Arcobacter in the MP suspension 393
samples remains unclear although the MP plant receives effluent from a local abbatoir and it 394
is possible that they derive from this source. 395
Consistent differences between the planktonic and biofilm communities are indicated by the 396
MDS analysis of the ARISA results. These differences may be explained by the different 397
conditions that prevail within these structured environments. The relatively short hydraulic 398
retention time within the MBBR system selects for organisms that have a high growth rate in 399
order to withstand wash-out and to be retained in the reactor. The bulk liquid phase is also 400
aerated, supporting aerobic metabolism. While members of the Clostridia were detected in the 401
suspended phase, the majority of taxa identified in the clone library analysis were from 402
putatively fast-growing aerobic species including members of the Burkholderiales, 403
Rhodocyclales, Aeromonadales, Pseudomonadales, and Rhodobacterales. 404
405
Comparison with activated sludge 406
on January 22, 2018 by guesthttp://aem
.asm.org/
Dow
nloaded from
17
Microbial communities in activated sludge systems have been widely discussed in the 407
literature as the majority of modern wastewater treatment facilities utilize this technology (42). 408
Activated sludge typically comprises free-swimming planktonic cells and aggregated flocs that 409
can be maintained under both aerobic and anoxic conditions. Bacterial communities in AS 410
plants treating industrial and municipal wastewater are typically dominated by 411
Betaproteobacteria, followed by Alpha- and Gammaproteobacteria. Other groups of bacteria 412
found in low abundance are Bacteroidetes and Firmicutes (30, 41). The clone library prepared 413
from a conventional AS mixed liquor sample in this study yeilded results consistent with 414
studies reported in the literature and suggests that both the attached biofilm and suspended 415
communities in MBBR systems differ substantially from that found in conventional AS. 416
417
In summary, this study provides the first detailed investigation of key microbial groups in full-418
scale MBBR systems treating municipal wastewater. The results indicate that MBBR 419
communities differ substantially from those in conventional AS systems by selecting for two 420
distinct bacterial communities: a biofilm community that is dominated by anaerobes and a 421
suspended community that includes fast-growing aerobic bacteria. The observation, from 422
FISH analyses, of a close association between Methanosarcinales and putative SRBs in the 423
MP biofilms implys a functional relationship between these two groups of microorganisms. 424
Further studies that include consideration of both the prokaryotic and eukaryotic communities 425
are required to elucidate the nature of these microbial relationships and to develop food web 426
models that will underpin manipulation of the system for optimal performance. 427
428
ACKNOWLEDGEMENTS 429
The authors would like to acknowledge the assistance of United Water Limited staff from Moa 430
Point and Karori WWTPs for provision of samples and support for this study. 431
432
on January 22, 2018 by guesthttp://aem
.asm.org/
Dow
nloaded from
18
REFERENCES 433
1. Aislabie, J., S. Jordan, J. Ayton, J. L. Klassen, G. M. Barker, and S. Turner. 2009. 434
Bacterial diversity associated with ornithogenic soil of the Ross Sea region, Antarctica. 435
Can. J. Microbiol. 55: 21–36. 436
2. Altschul, S. F., W. Gish, W. Miller, E. W. Myers, and D. J. Lipman. 1990. Basic local 437
alignment search tool. J. Mol. Biol. 215: 403-410. 438
3. Amann, R., L. Krumholz, and D. A. Stahl. 1990. Fluorescent-oligonucleotide probing of 439
whole cells for determinative, phylogenetic, and environmental studies in microbiology. J 440
Bacteriol. 172: 762-770. 441
4. Amann, R., W. Ludwig, and K. Scheleifer. 1995. Phylogenetic identification and in situ 442
detection of individual microbial cells without cultivation. Microbiol. Rev. 59: 143-169. 443
5. Anderottola, G., P. Foladori, M. Ragazzi, and F. Tatáno. 2000. Experimental 444
comparison between MBBR and activated sludge system for the treatment of municipal 445
wastewater. Wat. Sci. Tech. 41: 375-382. 446
6. Ashelford, K. E., N. A. Chuzhanova, J. C. Fry, A. J., Jones, and A. J. Weightman. 447
2006. New Screening software shows that most recent large 16S rRNA gene clone 448
libraries contain chimeras. Appl. Envion. Microbiol. 72: 5734-5741. 449
7. Ayton, J., J. Aislabie, G. Barker, D. Saul, and S. Turner. 2010. Crenarchaeota affiliated 450
with group 1.1b are prevalent in coastal mineral soils of the Ross Sea region of Antarctica. 451
Environ. Microbiol. 12: 689-703. 452
8. Barton, L. L., and G. D. Fauque. 2009. Biochemistry, physiology and biotechnology of 453
sulfate-reducing bacteria. Adv. Appl. Microbiol. 68: 41-98. 454
9. Ben-Dov, E., A. Kushmaro, and A. Brenner. 2009. Long-term surveillance of sulfate-455
reducing bacteria in highly saline industrial wastewater evaporation ponds. Saline Systems 456
5: 1-5. 457
on January 22, 2018 by guesthttp://aem
.asm.org/
Dow
nloaded from
19
10. Boetius, A., K. Ravenschlag, C. J. Schubert, D. Rickert, F. Widdel, A. Gieseke, et al. 458
2000. A marine microbial consortium apparently mediating anaerobic oxidation of 459
methane. Nature 407: 623-626. 460
11. Borghei, S. M., and S. H. Hosseini. 2004. The treatment of phenolic wastewater using a 461
moving bed biofilm reactor. Process Biochem. 39: 1177-1181. 462
12. Collado, L., M. J. Figueras. 2011. Taxonomy, epidemiology, and clinical relevance of the 463
genus Arcobacter. Clin. Microbiol. Rev. 24: 174-192. 464
13. Cord-Ruwisch, R. 1985. A quick method for the determination of dissolved and 465
precipitated sulfides in cultures of sulfate-reducing bacteria. J. Microbiol. Methods 4: 33-466
36. 467
14. Daims, H., S. Lüker, M. Wagner. 2006. diame, a novel image analysis program for 468
microbial eology and biofilm research. Environ. Microbiol. 8: 200-213. 469
15. Delong, E. F. 1992. Archaea in coastal marine environments. Proc. Natl. Acad. Sci. USA 470
89: 5685-5689. 471
16. Fu, B., X. Lia, L. Ding, and H. Ren. 2010. Characterization of microbial community in an 472
aerobic moving bed biofilm reactor applied for simultaneous nitrification and denitrification. 473
World J. Microbiol. Biotechnol. 26: 1981-1990. 474
17. Fukui, M., A. Teske, B. Abmus, G. Muyzer, and F. Widdel. 1999. Physiology, 475
phylogenetic relationships, and ecology of filamentous sulfate-reducing bacteria (genus 476
Desulfonema). Arch. Microbiol. 172: 193-203. 477
18. Knittel, K., A. Boetius, A. L. H. Eilers, K. Lochte, O. P. P. Linke, and R. Amann. 2003. 478
Activity, distribution, and diversity of sulfate reducers and other bacteria in sediments 479
above gas hydrates (Cascadia Margin, Oregon). Geomicrobiol. J. 20: 269-294. 480
19. Kolenbrander, P. E. 2000. Oral mircobial communities: Biofilms, interactions, and genetic 481
systems. Annu. Rev. Microbiol. 54: 413-437. 482
on January 22, 2018 by guesthttp://aem
.asm.org/
Dow
nloaded from
20
20. Labelle, M-A., P. Juteau, M. Jolicoeur, R. Villemur, S. Parent, and Y. Comeau. 2005. 483
Seawater denitrification in a closed mesocosm by a submerged moving bed biofilm 484
reactor. Water Res. 39: 3409-3417. 485
21. Lear, G., and G. D. Lewis. 2009. Impact of catchment land use on bacterial communities 486
within stream biofilms. Ecol. Indic. 9: 848-855. 487
22. Leloup, J., F. Petit, D. Boust, J. Deloffre, G. Bally, O. Clarisse, et al. 2005. Dynamics 488
of sulfate-reducing microorganisms (dsrAB genes) in two contrasting mudflats of the Seine 489
estuary (France). Microb. Ecol. 50: 307-314. 490
23. Lewandowski, Z., H. Beyenal, J. Myers, and D. Stookey. 2007. The effect of 491
detachment on biofilm structure and activity: the oscillating pattern of biofilm accumulation. 492
Water Sci. Technol. 55: 429-436. 493
24. Loy, A., A. Lehner, N. Lee, J. Adamcyzk, H. Meier, J. Ernest, et al. 2002. 494
Oligonucleotide microarray of 16S rRNA gene-based detection of all recognised lineages 495
of sulfate-reducing prokaryotes in the environment. Appl. Environ. Microbiol. 70: 6998-496
7009. 497
25. Ludwig, W., O. Strunk, R. Westram, L. Richter, H. Meier, Yadhukumar, et al. 2004. 498
ARB: a software environment for sequence data. Nuc. Acids Res. 32: 1363-1371. 499
26. Macalady, J. L., E. H. Lyon, B. Koffman, L. K. Albertson, K. Meyer, S. Galdenzi, et al. 500
2006. Dominant microbial populations in limestone corroding stream biofilms, Frasassi 501
cave system, Italy. Appl. Environ. Microbiol. 72: 5596-5609. 502
27. Manz, W., R. Amann, W. Ludwig, M. Wagner, and K. H. Schleifer. 1992. Phylogenetic 503
oligodeoxynucleotide probes for the major subclasses of proteobacteria: problems and 504
solutions. Syst. Appl. Microbiol. 15: 593-600. 505
28. Mittal, G. S. 2006. Treatment of wastewater from abbattoirs before land application- a 506
review. Bioresource Technol. 97: 1119-1135. 507
on January 22, 2018 by guesthttp://aem
.asm.org/
Dow
nloaded from
21
29. Muyzer, G., and A. J. M. Stams. 2008. The ecology and biotechnology of sulphate-508
reducing bacteria. Nature Rev. Microbiol. 6: 441-454. 509
30. Nielsen, P.H., T. R. Thomsen, and J. L. Nielsen. 2004. Bacterial composition of 510
activated sludge-importance for floc and sludge properties. Water Sci. Technol. 49: 51-58. 511
31. Ødegaard, H., B. Rusten, and J. Suljudalen. 1999. The development of the moving bed 512
biofilm process from idea to commercial product. Eur. Water. Manage. 2: 36-43. 513
32. Pruesse, E., C. Quast, K. Knittel, B. Fuchs, W. Ludwig, J. Peplies, and F. O. 514
Gl�ckner. 2007. SILVA: a comprehensive online resource for quality checked and aligned 515
ribosomal RNA sequence data compatible with ARB. Nuc. Acids Res. 35: 7188-7196. 516
33. Ranjard, L., F. Poly, J. C. Lata, C. Mougal, J. Thioulouse, and S. Nazaret. 2001. 517
Characterization of bacterial and fungal soil communities by automated ribosomal 518
intergenic spacer analysis fingerprints: Biological and methodological variability. Appl. 519
Environ. Microbiol. 67: 4479-4487. 520
34. Raskin, L., L. K. Poulsen, D. R. Noguera, B. E. Rittmann, and D. A. Stahl. 1994. 521
Quantification of methanogenic groups in anaerobic biological reactors by oligonucleotide 522
probe hybridisation. Appl. Environ. Microbiol. 60: 1241-1248. 523
35. Rusten, B., B. Eikebrokk, Y. Ulgenes, and E. Lyngren. 2006. Design and operations of 524
the Kaldnes moving bed biofilm reactors. Aquacult. Eng. 34: 322-331. 525
36. Rusten, B., J. G. Siljudalen, and H. Strand. 1996 Upgrading of a biological-chemical 526
treatment plant for cheese factory wastewater. Water Sci. Technol. 14: 41-49. 527
37. Rusten, B., O. Kolkinn, and H. Ødegaard. 1997. Moving bed biofilm reactors and 528
chemical precipitation for high efficiency treatment of wastewater from small communities. 529
Wat. Sci. Tech. 35: 71-79. 530
38. Sambrook, J., and D. Russell. 2001. Molecular cloning. New York: Cold Spring habour. 531
on January 22, 2018 by guesthttp://aem
.asm.org/
Dow
nloaded from
22
39. Saul, D., J. Aislabie, C. Brown, L. Harris, and J. Foght. 2005. Hydrocarbon 532
contamination changes the bacterial diversity of soil from around Scott Base, Antarctica. 533
FEMS Microbiol. Ecol. 53: 141-155. 534
40. Schink, B. 1997. Energetics of syntrophic cooperation in methanogenic degradation. 535
Microbiol. Mol. Biol. Rev. 61: 262-280. 536
41. Schmid, M., A. Thill, U. Purkhold, M. Walcher, J. Y. Bottero, and P. Ginestet, et al. 537
2003. Characterization of activated sludge flocs by confocal lazer scanning microscopy 538
and image analysis. Water Res. 37: 2043-2052. 539
42. Seviour, R., and Nielsen, P. H. 2010. Microbial communities in Activated sludge plants, 540
p. 95-126. In Seviour, R., and Nielsen, P. H.(ed.), Microbial ecology of activated sludge, 541
IWA publishing, Norfolk, UK. 542
43. Sigrun, J. J., A. R. Jukka, and H. Ødegaard. 2002. Aerobic moving bed biofilm reactor 543
treating thermomechanical pulping whitewater under thermophilic conditions. Water Res. 544
36: 1067-1075. 545
44. Smith, B. Y., S. J. Turner, and K. A. Rogers. 2003. Opal-A and associated microbes 546
from Wairakei, New Zealand: the first 300 days. Mineralogical Magazine 67: 563-579. 547
45. Sousa, D. Z., J. I. Alves, M. M. Alves, H. Smidt, and A. J. M. Stams. 2009. Effects of 548
sulphate methanogenic communities that degrade unsaturated and saturated long-chain 549
fatty acids (LCFA). Environ. Microbiol. 11: 68-80. 550
46. Strous, M., J. A. Fuerst, E. H. M. Kramer, S. Logemann, G. Muyzer, K. T. van de Pas-551
Schoonen, et al. 1999. Missing lithotroph identified as new Planctomycete. Nature. 400: 552
446-449. 553
47. Tal, Y., J. E. M. Watts, S. B. Schreier, K. R. Sowers, and H. J. Schreier. 2003. 554
Characterization of the microbial community and nitrogen transformation processes 555
associated with moving bed bioreactors in a closed recirculated mariculture system. 556
Aquaculture 215: 187-202. 557
on January 22, 2018 by guesthttp://aem
.asm.org/
Dow
nloaded from
23
48. Vandamme, P., E. Falsen, R. Rossau, B. Hoste, P. Segers, and R. Tytgat et al. 1991. 558
Revision of Campylobacter, Helicobacter, and Wolinella taxonomy: emendation of generic 559
descriptions and proposal of Arcobacter gen. nov. Int. J. Syst. Bacteriol. 41: 88-103. 560
49. Wang, Q., G. M. Garrity, J. M. Tiedje, and J. R. Cole. 2007. Naïve Bayesian classifier for 561
rapid assignment of rRNA sequences into the new bacterial taxonomy. Appl. Environ. 562
Microbiol. 73: 5261-5267. 563
50. Widdel, F. 1983. Methods for enrichment and pure culture isolation of filamentous gliding 564
sulfate-reducing bacteria. Arch. Microbiol. 134: 282–285. 565
51. Yang, S., F. Yang, Z. Fu, and R. Lei. 2009. Comparison between a moving bed 566
membrane bioreactor and a conventional membrane bioreactor on organic carbon and 567
nitrogen removal. Bioresource Technol. 100: 2369-2374. 568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
on January 22, 2018 by guesthttp://aem
.asm.org/
Dow
nloaded from
24
FIGURE LEGENDS 584
Figure 1: Multidimensional scaling (MDS) plot of ARISA data representing bacterial 585
communities in samples collected over 12 months from MP and Karori WWTPs. Each symbol 586
within the graph represents a bacterial community for one given time point. Minimum stress = 587
0.1. 588
Bacterial communities from Moa Point (black) and Karori (grey) samples have formed two 589
distinct clusters. Evidence of biofilm communities (enclosed within the black circles) 590
segregating away from the suspended biomass is also displayed. 591
592
Figure 2: Composition of 16S clone libraries in biofilm (A) and suspended biomass (B) from 593
Moa point and Karori reactors. Samples were collected in March-2010. For comparison, a 594
clone library prepared from conventional AS mixed liquor (C) is also shown. Each colour on 595
the graph represents the dominant Phyla or Class found within the community. 596
597
Figure 3: Maximum likelihood tree showing alignment of sequences from clone libraries 598
prepared from biofilm and suspended communities. Numbers within brackets indicate 599
percentage of clones from the MP biofilm, Karori biofilm, MP suspended and Karori 600
suspended samples respectively. These numbers differ to those presented in figure 2 as they 601
indicate percentages over three time points (March 2010, Aug 2010, Jan 2011). The scale 602
bar indicates 10% sequence divergence. 603
604
Figure 4: Maximum likelihood tree showing alignment of sequences from clone libraries to the 605
sulfate reducing members of the Deltaproteobacteria. Clone sequences are presented in bold 606
and identified by sample type, date, and clone reference respectively. Sample type identifiers: 607
M1, Moa Point biofilm; MS1 Moa Point suspended biomass; K1, Karori biofilm; KS1, Karori 608
suspended biomass. Numbers in brackets after the sequence identifier indicate number of 609
on January 22, 2018 by guesthttp://aem
.asm.org/
Dow
nloaded from
25
clones within the cluster. Shaded boxes indicate diversity within a group of clones. Dotted 610
lines indicate partial sequences (450 – 750 bp). Open and filled circles indicate clades 611
supported by bootstrap value of ≥75% and ≥90% respectively. Horizontal bars indicate groups 612
within the Syntrophomonadales (Syn), Desulfobacterales (Dsb) and Desufovirbionales (Dsv). 613
Outgroup consists of sequences from other bacterial phyla. 614
615
Figure 5: Maximum likelihood tree showing alignment of sequences from clone libraries to the 616
Epsilonproteobacteria. Vertical bars with brackets in bold, indicate groups within the genus 617
Arcobacter. Details are same as those provided in Figure 3 and 4. 618
619
Figure 6: Biofilm samples scraped from carriers (as seen in column 1) were hybridized with 620
fluorescently labeled probes targeting methanogenic Archaea (MSMX860-red), 621
Deltaproteobacteria (DELTA495a-cyan), and eubacteria (EUB338-green). DAPI (blue) was 622
used as a universal DNA stain. Clusters of Deltaproteobacteria (cyan-arrowed ‘s’) were 623
abundant in black biofilm from MP (A). Clusters of Deltaproteobacteria from Karori were 624
present as filaments (B). Methanogenic Archaea (arrowed ‘m’) are visible in both samples. 625
Biovolume estimates of methanogens and Deltaproteobacteria relative to DAPI stained 626
material are shown in graphs for both samples. 627
on January 22, 2018 by guesthttp://aem
.asm.org/
Dow
nloaded from
Table 1: Fluorescence in situ hybridisation (FISH) probe sequences.
Probe
name
Target group Probe sequence
5’ 3’
Label Reference
EUB338 Eubacteria GCTGCCTCCCGTAGGAGT
CY3 3
MSMX860 Methanosarcinales
GGCTCGCTTCACGGCTTCCCT
CY3 34
DELTA 495a
Most Deltaproteobacteria
and Gemmatimonadetes
AGTTAGCCGGTGCTTCCT
CY5 24
cDELTA 495a
Competitor for DELTA495a
AGTTAGCCGGTGCTTCTT
- 26
on January 22, 2018 by guesthttp://aem
.asm.org/
Dow
nloaded from
Moa PointKarori
Resemblance: D7 Manhattan distance2D Stress: 0.21Resemblance: D7 Manhattan distance
on January 22, 2018 by guesthttp://aem
.asm.org/
Dow
nloaded from
(A) (C)
MP Biofilm Karori Biofilm Conventional AS
(B)
Alphaproteobacteria
Gammaproteobacteria Betaproteobacteria
Deltaproteobacteria Epsilonproteobacteria Firmicutes
MP planktonic Karori planktonic
Bacteroidetes Others
on January 22, 2018 by guesthttp://aem
.asm.org/
Dow
nloaded from
Synergistes(5.6;1.1;0;2.8)
Firmicutes(43.9;40.8;17.8;53.6)
Planctomycetes(0.4;1.1;0;0.6)
Actinobacteria(0.7;0;0;2.2)
Cyanobacteria(0;0;0.6;0)
( ; ; ; )Acidobacteria(1.5;0.7;0;0)
Verrucomicrobia(0;0.7;0;0)
Bacteroidetes
BD1-5(0.4;0;0;0)
Epsilon-(3.7;0.4;54;0.6)
Delta-(29.4;24.7;0;11.7)BRC1
(0;0.7;0;0)
Chlorobi(0;0.7;0;0)
(3.3;4.9;4.6;1.7)
Alpha-(3.7;6.4;1.7;9.5)Beta/Gamma-
(7.4;17.6;21.3;17.3)
0.10
Proteobacteria
on January 22, 2018 by guesthttp://aem
.asm.org/
Dow
nloaded from
M1_190310_H7; M1(5);KS1(5)M1_190310_C12; M1(1);KS1(1)
Marine sediment clone, EU700163Forested wetland clone, AF523958
M1_260810_A4; M1(1);K1(1);KS1(1)Freshwater pond clone, DQ676381
Desulfomonile tiedjei, AM086646M1_260810_C4
Microbial mat clone, AM490653K1_260810_A1
K1_190310_H9Biphenyl degrading clone, EU651865
K1_190310_B3c; K1(4)K1_260810_A9
Mi bi l t l AM490740Syn.
Microbial mat clone, AM490740M1_190310_A2; M1(1);KS1(2)Desulfomonile limimaris, AF230531M1_180111_F12; M1(2)
Cold seep clone, FM165239
M1_190310_D9; KS1(1)Microbial mat clone, DQ154835
Desulforhabdus sp. DDT, EF442978M1_190310_H11c; M1(2);KS1(1)
Desulforhabdus amnigena, X83274K1_260810_D10
Hot spring sediment clone, FJ638562Syntrophobacter sulfatireducens, AY651787
K1_190310_A3c; M1(3); K1(5); KS1(1)
Cave biofilm clone, FJ716466Desulfarculus baarsii, AF418174
K1_190310_C8K1 260810 F11
y
_ _Stream biofilm clone, DQ415831
K1_190310_B2cSinkhole clone, FJ484450
K1_260810_G1; K1(1)
K1_260810_F2c; K1(1)Wetland sediment clone, FJ516878
Peat swamp clone, GQ402620
K1_190310_C10c; K1(2)Desulfobacca acetoxidans, AF002671
K1_180111_D10c; K1(1)Anaerobic reactor clone, FJ462058
Sinkhole clone, FJ484421Syntrophus aciditrophicus, CP000252
M1_180111_F8c; M1(7);KS1(2)Desulfococcus multivorans, AF418173
Desulfonema magnum U45989
Syn.
Desulfonema magnum, U45989CandidatusMagnetoglobusmulticellularis, EF014726
K1_190310_B10c; K1(3)Lake sediment clone, HQ330650
K1_260810_G10Microbial mat clone, AM490744
K1_190310_E7; K1(1)Wetland isolate, FJ517061
Desulfosarcina variabilis, M34407M1_180111_H9cM1_260810_C1; M1(1)
Marine sediment clone, GU584771Antarctic sediment clone, AY177788
KS1_190310_A7; M1(1)Mangrove soil clone, DQ811807
Oil degrading isolate, AF177428M1_180111_B5
M1 180111 B12c; M1(2)M1_180111_B12c; M1(2)Wetland sediment clone, FJ516922Desulfobacterium indolicum, AJ237607
M1_180111_D8; K1(2)Lake clone, EF520539
M1_180111_G2; K1(2)Column water clone, EU981251
M1_260810_H11M1_180111_A9; M1(1)
Desulfatirhabdium butyrativorans, DQ146482Desulfobacter curvatus, AF418175
M1_260810_G10Wastewater clone, EU234152
Desulfobacula toluolica, AJ441316
M1_190310_C2; KS1(1)Intertidal sediment clone, GU197414
M1 260810 D9
Dsb.M1_260810_D9
Desulforhopalus vacuolatus, L42613
M1_180111_F3; M1(2)Desulforhopalus singaporensis, AF118453
M1_260810_D4; M1(3); K1(7)
KS1_190310_G1; M1(5); KS1(1)Desulfobacterium catecholicum, AJ237602Marine sediment isolate, AY771933
KS1_190310_D2; M1(2)
M1_180111_A7; M1(8)Desulfotalea psychrophila , CR522870
K1_260810_H4Wastewater clone, HM584333
M1_180111_F11c; M1(1)
K1 260810 A2c; M1(1); K1(2); KS1(1)K1_260810_A2c; M1(1); K1(2); KS1(1)
K1_180111_D2c; K1(2)
Freshwater sediment clone, DQ831535K1_260810_A10
Desulfocapsa thiozymogenes, X95181K1_180111_E4; K1(2)
M1_260810_E11Desulfofustis glycolicus, X99707
M1_180111_E10cLake clone, HM128150
Desulfobulbus propionicus, AY548789K1_260810_G4c; K1(1)
Brackish sediment clone, DQ831533K1_190310_D7
Reed bed reactor clone, AB240351Desulfurivibrio alkaliphilus, EF422413
K1_180111_G8cStream biofilm clone, DQ133926
Lake clone, AJ389622
Stream biofilm clone, DQ133926Peredibacter starrii, AF084852
K1_190310_B9; M1(2)KS1_260810_H3cLeech gut clone, DQ355176
Desulfovibrio cuneatus, X99501M1_260810_C8c
Microbial mat clone, EU246265Desulfovibrio indonesiensis, Y09504
Desulfomicrobium baculatum, AF030438M1_260810_B4cDesulfurella multipotens, Y16943
0.10
Dsv.
Outgroup
on January 22, 2018 by guesthttp://aem
.asm.org/
Dow
nloaded from
MS3_260810_B1c; MS3(32)
River clone, FN428747River estuary clone, DQ234101
MS1_190310_E9c; MS(7); M1(1); K1(1)
Arcobacter nitrofigilis, CP001999
MS3_260810_H3; MS(3)
MS3_260810_A10cWater column clone, GQ340204
Seawater filtration clone, HM591442Seawater isolate, AB496655
MS3_260810_D1; MS3(4); M1(1)
Arcobacter thereius, AY314753, 1409MS1_190310_C5; MS1(2)
Activated sludge clone, Z94001Arcobacter cryaerophilus, L14624
Anaerobic sludge clone, CU924753M1_180111_C12
MS1_190310_H6 Leachate sediment clone, HQ183859
Arcobacter skirrowii, L14625Activated sludge clone, EU104239
Arcobacter
MS1_190310_H3c(16); M1(1)Mangrove clone, DQ234115
MS1_190310_H10; MS1(16)Arcobacter cibarius, AJ607391
Arcobacter butzleri, AY621116MS3_260810_H7 (4); M1(1); KS1(1)
Dolphin blowhole clone, FJ959747
M1_180111_A3c; M1(3); MS1(1)
Hydrothermal vent clone, AJ441203Stream biofilm clone, EF467591
Sulfurovum lithotrophicum, AB091292L k di t l GQ472365
Outgroup
Lake sediment clone, GQ472365
M1_260810_H9cSulfuricurvum kujiense, AB080645
Sulfurimonas autotrophica, AB088431Helicobacter pylori, M88157
MS1_190310_G5cLake isolate, FJ612321
Sulfurospirillum barnesii, U41564MS3_260810_D7
Campylobacter jejuni subsp. jejuni, L04315
0.10
on January 22, 2018 by guesthttp://aem
.asm.org/
Dow
nloaded from
Carriers FISH image % biovolume g
Poin
t
15
20
25
s
(A)
Moa
P
0
5
10
Methanosarcinales δ-proteobacteria
m
ri
15
20
25
s
m
(B)
Kar
o
0
5
10
Methanosarcinales δ-proteobacteria
on January 22, 2018 by guesthttp://aem
.asm.org/
Dow
nloaded from
