APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Dec. 1998, p. 4877–4882 0099-2240/98/$04.0010 Copyright © 1998, American Society for Microbiology. All Rights Reserved.
Vol. 64, No. 12
Assessment of Changes in Microbial Community Structure during Operation of an Ammonia Biofilter with Molecular Tools Y. SAKANO
AND
L. KERKHOF*
Department of Environmental Sciences and Institute of Marine and Coastal Sciences, Cook College, Rutgers, The State University of New Jersey, New Brunswick, New Jersey 08903-0231 Received 10 June 1998/Accepted 14 September 1998
Biofiltration has been used for two decades to remove odors and various volatile organic and inorganic compounds in contaminated off-gas streams. Although biofiltration is widely practiced, there have been few studies of the bacteria responsible for the removal of air contaminants in biofilters. In this study, molecular techniques were used to identify bacteria in a laboratory-scale ammonia biofilter. Both 16S rRNA and ammonia monooxygenase (amoA) genes were used to characterize the heterotrophic and ammonia-oxidizing bacteria collected from the biofilter during a 102-day experiment. The overall diversity of the heterotrophic microbial population appeared to decrease by 38% at the end of the experiment. The community structure of the heterotrophic population also shifted from predominantly members of two subdivisions of the Proteobacteria (the beta and gamma subdivisions) to members of one subdivision (the gamma subdivision). An overall decrease in the diversity of ammonia monooxygenase genes was not observed. However, a shift from groups dominated by organisms containing Nitrosomonas-like and Nitrosospira-like amoA genes to groups dominated by organisms containing only Nitrosospira-like amoA genes was observed. In addition, a new amoA gene was discovered. This new gene is the first freshwater amoA gene that is closely affiliated with Nitrosococcus oceanus and the particulate methane monooxygenase gene from the methane oxidizers belonging to the gamma subdivision of the Proteobacteria. microbial oxidation of ammonia to nitrate). Small-subunit rRNA genes were used to track communities of heterotrophic bacteria at three times during a 102-day experiment in the NH3 biofilter. In addition, the ammonia monooxygenase gene (amoA) was used as the target gene to identify microorganisms that were responsible for ammonia removal. (All ammonia-oxidizing bacteria bear the amoA gene, which catalyzes the first step in the nitrification pathway.) Large decreases in heterotrophic bacterial diversity during operation of the filter were observed, shifts in the community structure of both heterotrophic and ammonia-oxidizing bacteria were documented, and a novel amoA gene closely related to the Nitrosococcus oceanus gene was discovered during this study. In the future, studies such as the one described here will be crucial for characterization of microbial communities, for biofilter optimization, and for development of reliability measures for biofilters in other applications.
Biofiltration is a technique which removes toxic compounds (organic and inorganic compounds), odors, and volatile organic compounds from contaminated air. Since the 1920s, odorous compounds (e.g., H2S) have been removed by biofilters at a variety of wastewater treatment plants. Biofilters have also been used at solid waste processing plants and food processing plants for several decades (10, 31). In Europe, biofilters have been used to remove volatile organic compounds and odorous compounds since the late 1970s. Currently, there is a growing interest in the applications of biofiltration techniques in a variety of other settings (4). Reactor designs, filter materials, and other factors involved in biofilter function (e.g., moisture content, temperature, pH, O2 concentration, salt concentration) have been investigated (10, 24, 26). However, it has been suggested that monitoring microbial populations is a way to optimize biofilter performance (31), and studies on microbial ecology in bioreactors are needed (26, 29). Recently, molecular techniques have been used to study microbial populations in a wastewater-activated sludge reactor (25), a wastewater-trickling filter reactor (19), a toluene-degrading biofilter (13), and a phenol-degrading batch reactor (20). Unfortunately, these studies focused on single samples from the bioreactors, and little is known about population changes in microbial communities through time. In this study, we characterized microbial community structure in a laboratory-scale ammonia biofilter under development at Rutgers University. This biofilter was designed to remove ammonia through the process of nitrification (i.e.,
MATERIALS AND METHODS Sampling. Samples were collected from an ammonia biofilter under development at New Jersey–NASA-Sponsored Center of Research and Training (NSCORT). Briefly, perlite (an inert silica matrix) was inoculated with nitrifying activated sludge and leaf compost to create a biofilter having a biomass concentration of 0.5 to 1 g of biomass/kg (dry weight) of perlite. An airstream containing 20 ppm of ammonium and 500 to 600 ppm of CO2 was applied to the biofilter. Composite samples (150 g) were collected at the beginning of the experiment (day 0) and on days 15, 21, 28, 35, and 102. The samples were stored frozen (220°C) until they were processed in the laboratory. The system removed between 70 and 100% of the nitrogen during the course of the experiment (7a). The microbial populations in the day 0, 15, and 102 samples were assayed with molecular tools as described below. DNA extraction and purification. DNA was extracted by using a modified phenol-chloroform procedure (8). A total of 2.4 g of biofilter matrix was used for extraction. The contents of eight tubes containing 300 mg of perlite from a single time point were crushed, and the contents of each tube were suspended in 300 ml of buffer (cold 50 mM glucose–10 mM EDTA–25 mM Tris [pH 8.0]). The
* Corresponding author. Mailing address: Institute of Marine and Coastal Sciences, Rutgers, The State University of New Jersey, 71 Dudley Rd., New Brunswick, NJ 08901-8521. Phone: (732) 932-6555, ext. 335. Fax: (732) 932-6520. E-mail:
[email protected]. 4877
4878
SAKANO AND KERKHOF
APPL. ENVIRON. MICROBIOL.
FIG. 1. Rarefaction curves plotting the number of unique 16S rRNA clones versus the number of individual 16S rRNA clones as determined by restriction enzyme digestion. The clonal libraries obtained on day 0 (E), day 15 (‚), and day 102 (h) were examined. samples were frozen rapidly in liquid nitrogen and thawed at 37°C four times. One hundred microliters of a lysozyme solution (10 mg/ml) was added to each sample, and the preparation was incubated at room temperature for 5 min on a rotator. Fifty microliters of a 50% sodium dodecyl sulfate solution was added to lyse the cells. The lysate was extracted twice with 800 ml of a phenol-chloroformisoamyl alcohol mixture (25:24:1). The nucleic acids were precipitated by adding 50 ml of 3.0 M sodium acetate, 2 ml of glycogen (20 mg/ml), and 1 ml of 100% ethanol. The DNA was pelleted by centrifugation (16,000 3 g) at 4°C for 20 min. The eight nucleic acid pellets were sequentially resuspended in 500 ml (total volume) of 13 TE (10 mM Tris [pH 7.5], 1 mM EDTA) by adding the 13 TE to the first tube, letting the pellet dissolve, and transferring the entire contents to the next tube. The concentrated extract was then purified by cesium chloride ultracentrifugation (30). The concentration and size of the extracted DNA were determined by 1% (wt/vol) agarose gel electrophoresis in 13 TAE (40 mM Tris, 5 mM sodium acetate, 1 mM EDTA [pH 7.8]). PCR amplification. (i) 16S rRNA genes. Hot-start PCR amplification was performed with a model 2400 DNA thermal cycler (Perkin-Elmer, Foster City, Calif.) by using eubacterial primers 27F (59 cua cua cua cua AAG GAG GTG WTC CAR CC 39) and 1525R (59 cau cau cau cau AGA GTT TGA TCC TGG CTC 39) (9) bearing a uracil-rich end for cloning (see below). Less than 10 ng of template DNA, 15 pM primer, and 1 to 2 U of Taq polymerase were used for
each reaction. The amplification program was as follows: one cycle consisting of 94°C for 5 min, followed by 30 cycles consisting of 94°C for 0.5 min, 55°C for 0.5 min, and 72°C for 1.5 min and a final extension step consisting of 72°C for 10 min. (ii) Ammonia monooxygenase (amoA) genes. Two primers, primers A189 (59 cau cau cau cau GGN GAC TGG GAC TTC TGG 39) and A682 (59 cua cua cua cua GAA SGC NGA GAA GAA SGC 39), were used to detect ammonia oxidizers (7). The reaction conditions were the same as those described above. The amplification program was as follows: one cycle consisting of 94°C for 5 min, followed by 28 cycles consisting of 94°C for 0.5 min, 56°C for 0.5 min, and 72°C for 1 min and a final extension step consisting of 72°C for 7 min. Cloning. The amplified 16S rRNA and amoA genes were purified with a Geneclean kit (Bio 101, La Jolla, Calif.) by following the manufacturer’s instructions. The PCR products (22 cycles) were ligated by using the CLONEAMP pAMP1 system (Life Technologies, Gaithersburg, Md.) as recommended by the manufacturer and were transformed into high-efficiency competent cells (Promega, Madison, Wis.) at a DNA concentration of ,6 ng template21. Unique clones were identified by HaeIII (Promega) restriction digestion. Plasmid DNA from transformants that produced unique restriction patterns were repurified by using a FlexiPrep kit (Pharmacia, Piscataway, N.J.) for sequencing. Sequence analysis. DNA sequences were determined by using a model ABI 373A automated sequencer (Perkin-Elmer/ABI, Foster City, Calif.). Primary sequences were analyzed by using the Auto Assembler and SeqNavigator ABI software programs, as well as BLASTN (2). For the 16S rRNA clonal libraries, a neighbor-joining tree with the Jukes-Cantor correction was reconstructed with the Genetic Data Environment (21) (data not shown). Similarity ranks for smallsubunit rRNA clones were examined by using the Ribosomal Database Project method (11). For the amoA gene clonal libraries, a maximum-likelihood tree was reconstructed (100 replicates bootstraps) by using the fastDNAml program, and a distance similarity matrix for nucleotides and amino acids was constructed by using the neighbor-joining distance method in the Genetic Data Environment (21). The accession numbers of the gene sequences used in this study are as follows: Nitrosomonas europaea, L08050; Nitrosomonas eutropha, U51630; Nitrosospira multiformis, U89833; Nitrosospira sp., X90821; Nitrosospira sp. strain N39-19, AF006692; Nitrosospira briensis, U76553; Nitrosospira tenuis, U76552; Nitrosococcus oceanus, U96611; Methylococcus capsulatus, U94337; Methylobacter albus, U31654; Methylomonas methanica, U31653; Methylosinus trichosporium, U31650; and Methylocystis parvus, U31651. Clone designations. The biofilter amoA gene clones were designated as follows: BAXY, where X is the day(s) on which the sample(s) was obtained (A, day 0; B, day 15; C, day 102; D, days 0, 15, and 102) and Y is the isolate number. The biofilter heterotrophic (16S rRNA) gene clones were designated as follows: BHXY, where X is the day(s) on which the sample(s) was obtained (A, day 0; B, day 15; C, day 102; D, days 0, 15, and 102) and Y is the isolate number. Nucleotide sequence accession numbers. The nucleotide sequences determined in this study have been deposited in the GenBank database. The accession numbers for the amoA gene sequences are AF070983 through AF070987, and
TABLE 1. Comparison of sequences in 16S rRNA clonal libraries obtained on days 0, 15, and 102 to sequences in the Ribosomal Database Project database and changes in community structure over time Taxon
Proteobacteria Alpha subdivision Beta subdivision
Gamma subdivision
Epsilon subdivision Gram-positive bacteria Unknown
a
Clone
Most similar organism
Saba
BHA19 BHA5 BHB12 BHA16 BHA17 BHA18 BHB9 BHD21 BHA13 BHA7 BHB6 BHC6 BHC9 BHD15 BHB4 BHA9 BHA10 BHB5 BHD3 BHA14 BHC12
Sphingomonas sp. strain SS86 Bordetella avium Bordetella bronchiseptica S-1 Bordetella bronchiseptica S-1 Comamonas testosteroni RH 1104 Comamonas testosteroni RH 1104 Zoogloea ramigera Oceanospirillum kriegii Pseudomonas putida Xanthomonas campestris Environmental strain SMK2321 Marinomonas vaga Symbiont of Bathymodiolus thermophilus gills Campylobacter sp. Helicobacter pullorum Eubacterium multiforme Actinosynnema mirum Environmental strain MC25 Isosphaera sp. strain Schlesner 666 Isosphaera sp. strain Schlesner 666 Strain SRB2068
0.571 0.753 0.679 0.785 0.611 0.941 0.765 0.522 0.978 0.698 0.578 0.524 0.536 0.524 0.558 0.378 0.477 0.694 0.607 0.566 0.709
Presence on: Day 0
Day 15
Day 102
1 1 2 1 1 1 2 1 1 1 2 2 2 1 2 1 1 2 1 1 2
2 2 1 2 2 1 1 1 2 1 1 2 2 1 1 2 2 1 1 2 2
2 2 2 2 2 2 2 1 2 2 1 1 1 1 2 2 2 2 1 2 1
Sab values indicate the levels of homology to sequences in the database; a value of 0.9 or greater indicates a close match to a sequence in the database.
VOL. 64, 1998
CHANGES IN MICROBIAL COMMUNITY STRUCTURE
4879
rRNA genes was observed with the day 0, 15, and 102 samples (data not shown) suggesting that there was little or no inhibition of PCR by DNA extracts at these times. No amplification was observed with the day 28 sample for some reason, and this sample was not analyzed further. For the samples that were successfully amplified, between 150 and 1,000 colonies were obtained from each clonal library by using an aliquot of the PCR mixture. The frequencies of unique clones in the libraries were used to estimate the microbial diversity in the system (a rarefaction curve was obtained by restriction enzyme analysis) (Fig. 1). A clear shift in microbial diversity was observed from day 0 to day 102, and roughly 38% of the unique clones disappeared during this time. In order to determine whether the individual members of the heterotrophic community also changed, we performed a sequence analysis of the cloned 16S rRNA genes. Table 1 shows the results of similarity searches performed by using the Ribosomal Datbase Project (11) for the heterotrophic population. This sequence analysis revealed that there was some duplication (99% identity) of the various restriction fragment length polymorphism patterns in the clonal libraries, and the clones were assumed to represent different copies of the rRNA operons within a single genome. Therefore, the data in Table 1 differ from the total number of unique clones shown in Fig. 1 (i.e., 13 clones were obtained from the day 0 sample, 9 clones were obtained from the day 15 sample, and 8 clones were obtained from the day 102 sample). The phylogenetic tree reconstruction results confirmed the positions of all of the 16S rRNA clones in their respective phyla with Sab values (data not shown). The heterotrophic community was comprised mostly of members of Proteobacteria phyla along with six representatives loosely affiliated with the gram-positive phyla or the Isosphaera group. In the proteobac-
FIG. 2. Consensus phylogenetic tree based on ammonia monooxygenase genes as determined by the maximum-likelihood method and 100 iterations (483-bp alignment). The numbers at the nodes are the percentages of bootstrap trees within similar topologies; only values greater than 50% are shown. Biofilter clones BAB33, BAC5, BAD34, BAC6, and BAA8 were obtained in this study. the accession numbers for the 16S rRNA gene sequences are AF090535 through AF090553.
RESULTS Purification of genomic DNA from the perlite matrix yielded 50 to 150 ng of DNA per 2.4 g of perlite mixture (data not shown). This DNA was used to amplify the 16S rRNA and amoA genes as described above. Strong amplification of 16S
TABLE 2. Distance-similarity matrix for 480 bp of amoA and pmoA nucleotides and 160 predicted amino acids encoded by amoA and pmoA
a
68 69 67 67 68 87 70 64 68 27 27 39 27 35 24 21
88 85 85 81 69 83 81 84 28 33 51 37 37 36 43
85 83 96 91 86 85 69 87 83 86 26 34 54 38 34 40 43
84 83 96 96 83 87 68 89 84 87 28 36 51 31 31 35 40
83 81 94 94 94 84 69 84 79 84 24 32 51 37 37 35 39
83 80 94 94 96 94 70 95 89 91 30 33 52 34 35 37 37
73 67 71 31 24 43 30 35 29 29
91 92 30 32 54 35 38 37 39
90 23 24 48 29 30 33 35
83 81 96 95 97 94 98 83 99 94 28 36 54 37 35 39 40
49 46 48 47 48 47 46 49 46 44 47 68 49 51 50 41 45
52 48 53 51 52 49 50 51 50 47 51 76 46 54 50 39 36
56 53 57 56 55 54 54 55 54 51 55 65 62 64 62 57 61
The values on the upper right are levels of amino acid similarity, and the values on the lower left are levels of nucleotide similarity.
52 49 55 54 53 51 52 51 52 49 53 67 66 76 72 48 48
50 47 52 51 50 49 49 49 50 47 51 62 63 73 83 43 41
46 42 48 48 47 46 47 45 46 43 47 60 56 64 67 62 87
Methylosinus trichosporium
Methylocystis parvus
Methylomonas methanica
Methylobacter albus
Methylococcus capsulatus
78 77 92 90 92 90 92 79 94
Clone BAC6
82 82 96 96 97 96 98 83
Nitrosococcus oceanus
98 96 86 84 83 83 82
Clone BAC5
Nitrosospira multiformis
Nitrosospira sp.
Nitrosospira tenuis
Nitrosospira sp. strain N3919
Nitrosospira briensis
85 83
Clone BAD34
97
Clone BAB33
85 69 70 70 70 71 92 72 67 71 32 27 44 32 36 32 30
Clone BAA8
Nitrosomonas europaea Nitrosomonas eutropha Nitrosospira briensis Nitrosospira sp. strain N39-19 Nitrosospira tenuis Nitrosospira sp. Nitrosospira multiformis Clone BAA8 Clone BAB33 Clone BAD34 Clone BAC5 Nitrosococcus oceanus Clone BAC6 Methylococcus capsulatus Methylobacter albus Methylomonas methanica Methylocystis parvus Methylosinus trichosporium
Nitrosomonas eutropha
Taxon
Nitrosomonas europaea
% Similarity toa:
45 43 48 48 47 46 46 46 46 44 46 61 57 65 64 61 88
4880
SAKANO AND KERKHOF
APPL. ENVIRON. MICROBIOL.
FIG. 3. Partial alignment of the predicted amino acids encoded by amoA (ammonia-oxidizing bacteria) and pmoA (methane-oxidizing bacteria). The residues conserved in all sequences are highlighted. The conserved amoA residues are enclosed in shaded boxes. Abbreviations: Nmeur, Nitrosomonas europaea; Nmeut, Nitrosomonas eutropha; Nsmul, Nitrosospira multiformis; Nssp., Nitrosospira sp.; NsN39, Nitrosospira sp. strain N39-19; Nsbri, Nitrosospira briensis; Nsten, Nitrosospira tenuis; Ncoce, Nitrosococcus oceanus; Mcap, Methylococcus capsulatus; Malu, Methylobacter albus; Mmet, Methylomonas methanica; Mtri, Methylosinus trichosporium; Mpar, Methylocystis parvus; A8, clone BAA8; B33, clone BAB33; D34, clone BAD34; C5, clone BAC5; C6, clone BAC6.
terial group, a shift from predominantly members of the b and g subdivisions at day 0 to members of the g subdivision at day 102 occurred. Only the following three members of the heterotrophic community were found throughout the study: clone BHD21 (g subdivision), clone BHD15 (ε subdivision), and clone BHD3 (affiliation unknown). The remaining small rRNA clones were observed only at one or two times. There were no known nitrifiers in the 16S rRNA gene li-
brary. Therefore, we amplified a portion of the amoA gene (7) to monitor the nitrifying population during the 102-day experiment. An analysis of the amoA clonal libraries in which rarefaction curves were used did not reveal a change in the diversity of the nitrifying populations over time (data not shown). However, a reconstructed phylogenetic tree of amoA sequences revealed that there were major shifts in the clonal amoA gene population structure during the study (Fig. 2). Only
VOL. 64, 1998
CHANGES IN MICROBIAL COMMUNITY STRUCTURE
one sequence (clone BAD34) was found throughout the study. This clone was most similar to amoA genes recovered from Nitrosospira species. The following two additional amoA sequences affiliated with the Nitrosomonas europaea clade were found on day 0: clone BAA7 (identified as Nitrosomonas europaea) and clone BAA8. Within 2 weeks, Nitrosomonas-like amoA genes were no longer detected, and the ammonia-oxidizing population appeared to be composed of bacteria with amoA genes most similar to those of Nitrosospira species; a new clone (clone BAB33) was detected that was not present in the day 0 library. By the end of the experiment, an additional Nitrosospira-like amoA gene (clone BAC5) was detected, as was a new amoA gene (clone BAC6) that exhibited 68% identity to the Nitrosococcus oceanus gene. The BAC6 gene sequence was also similar to particulate methane monooxygenase (pmoA) gene sequences. Table 2 is a distance similarity matrix in which the nucleotide and predicted amino acid sequences of the amoA genes identified in this study are compared. The levels of nucleotide similarity and amino acid similarity obtained in this study ranged from 23 to 95% and from 44 to 98%, respectively. An alignment of the amino acids in parts of the amoA- and pmoAencoded proteins is presented in Fig. 3. A total of 39 of 161 amino acid residues were conserved in all sequences. The number of conserved residues increased to 52 when just amoA proteins were compared. DISCUSSION Bioreactors are increasing in popularity as a means of remediating waste streams. However, little is known about the bacterial population structure of biofilters since most research treats the biofilter system as a “black box.” Furthermore, much of the microbiological research done previously involved culturing isolates, and many bacteria in complex systems are now widely believed to be as yet unculturable (3, 23, 28). In this study, the population structure and dynamics of an ammonia biofilter were examined with molecular tools. A large (38%) decrease in heterotrophic diversity and significant changes in the microbial community were observed by the end of the experiment. In addition, our preliminary molecular profiling analysis in which 16S rRNA genes were used did not detect known nitrifying bacteria. In previous research on nitrifiers in Antarctic lakes workers found that they were not able to directly amplify 16S rRNA genes from nitrifiers without a twostage enrichment procedure for nitrifier rRNA genes (27). To circumvent this problem, we targeted the ammonia oxidizers by amplifying the ammonia monooxygenase (amoA) genes (7, 17). Although changes in amoA gene diversity were not observed over time, structural shifts in the amoA gene population analogous to the changes observed in the heterotrophic populations occurred. Four of the five amoA gene sequences detected in our biofilter were novel. One amoA sequence (clone BAC6) is the first amoA gene sequence recovered from a freshwater system that is closely affiliated with the amoA gene of a member of the g Proteobacteria. The maximum-likelihood tree and distance similarity matrix demonstrated that the amoA genes of clone BAC6 and Nitrosococcus oceanus are more closely related to the pmoA genes of methane oxidizers than to amoA genes of the other ammonia oxidizers. At present, we do not know the phylogenetic affiliation of the nitrifier bearing the novel amoA gene. Furthermore, our data support the concept that many of the model organisms traditionally used to study nitrogen cycling may not be the indigenous bacteria important in any particular system (18). One possible explanation for the results which we obtained
4881
is that the differences which we measured with the molecular tools were the result of methodological artifacts. However, we believe that this explanation is unlikely. Although the 16S rRNA gene characterization method for identifying bacteria is becoming routine in many labs, potential biases in the traditional clone and sequence approach do exist. In particular, DNA extraction procedures can miss entire groups that are difficult to lyse, such as gram-positive organisms. In addition, large amounts of template DNA, high cycle numbers during PCR amplification, and large amounts of transforming DNA are used to maximize the number of colonies obtained during cloning. However, this strategy may confound attempts at analysis due to PCR (1, 5, 16, 22) or cloning biases resulting from the asymptotic transformation of Escherichia coli at DNA masses of .10 ng (6). Therefore, we took steps in our study to minimize the biases inherent in the traditional approach. The extraction procedure which we used has been shown to quantitatively recover nucleic acids from easily lysed bacteria, such as Pseudomonas stutzeri Zobell (8), and can successfully isolate DNA from organisms that are more difficult to lyse, such as gram-positive bacteria from marine sediment systems (15). As for PCR and cloning artifacts, in most previous studies the researchers utilized high template concentrations (.200 ng of template) and high numbers of cycles (.28 cycles) for the PCR. In this study, we used minimal template concentrations (,10 ng of genomic DNA), lower numbers of cycles (20 to 25 cycles), and low amounts of transforming DNA (,6 ng) to create our clonal libraries. In addition, all amplification reactions and cloning of the different samples were performed at the same time with the same reaction mixtures. In conclusion, further work is needed to understand the relationship between biofilter performance and microbial community dynamics. Currently, the influence of microbial diversity and the influence of shifts in community structure, if any, on the rates of ammonia conversion in the biofilters which we used are unclear. However, the molecular methods described in this study can shed light on the ecology of the unculturable organisms and their population changes within our bioreactor over time. An understanding of microbial populations in biofilters is important since biodiversity may play a major role in enhancing bioreactor predictability and reliability (12, 14). Finally, microbial profiling may provide an early warning system that can be used to predict degradation of the biological components far sooner than loss of function. ACKNOWLEDGMENTS This research was supported in part by funds from Rutgers University. We thank the Waste Processing Team of NJ-NSCORT at Rutgers University for providing samples and analytical data. We also thank David Scala and Mary Voytek for assistance in sequencing and data analyses. We are also indebted to anonymous reviewers and Jerome Kukor; their comments significantly improved the manuscript. REFERENCES 1. Alard, P., O. Lantz, M. Sebagh, C. F. Calvo, D. Weill, G. Chavanel, A. Senik, and B. Charpentier. 1993. A versatile ELISA-PCR assay for mRNA quantitation from a few cells. BioTechniques 15:730–737. 2. Altschul, S. F., W. Gish, W. Miller, E. W. Myers, and D. J. Lipman. 1990. Basic local alignment search tool. J. Mol. Biol. 215:403–410. 3. Amann, R. I., W. Ludwig, and K. H. Schleifer. 1995. Phylogenetic identification and in situ detection of individual microbial cells without cultivation. Microbiol. Rev. 59:143–169. 4. Bohn, H. 1992. Consider biofiltration for decontaminating gases. Chem. Eng. Prog. 88:34–40. 5. Farrelly, V., F. A. Rainey, and E. Stackebrandt. 1995. Effect of genome size and rrn gene copy number on PCR amplification of 16S rRNA genes from a
4882
SAKANO AND KERKHOF
mixture of bacterial species. Appl. Environ. Microbiol. 61:2798–2801. 6. Hanahan, D. 1983. Studies on transformation of Escherichia coli with plasmids. J. Mol. Biol. 166:557–580. 7. Holmes, A. J., A. Costello, M. E. Lidstrom, and J. C. Murrell. 1995. Evidence that particulate methane monooxygenase and ammonia monooxygenase may be evolutionarily related. FEMS Microbiol. Lett. 132:203–208. 7a.Joshi, J. A., et al. Unpublished data. 8. Kerkhof, L., and B. B. Ward. 1993. Comparison of nucleic acid hybridization and fluorometry for measurement of the relationship between RNA/DNA ratio and growth rate in a marine bacterium. Appl. Environ. Microbiol. 59: 1303–1309. 9. Lane, D. J. 1991. 16S/23S rRNA sequencing, p. 115–175. In E. Stackebrandt and M. Goodfellow (ed.), Nucleic acid techniques in bacterial systematics. John Wiley & Sons Ltd., Chichester, England. 10. Leson, G., and A. M. Winer. 1991. Biofiltration: an innovative air pollution control technology for VOC emissions. J. Air Waste Manag. Assoc. 41:1045– 1054. 11. Maidak, B. L., G. J. Olsen, N. Larsen, R. Overbeek, M. J. McCaughey, and C. R. Woese. 1997. The RDP (Ribosomal Database Project). Nucleic Acids Res. 25:109–111. 12. McGrady-Steed, J., P. M. Harris, and P. J. Morin. 1997. Biodiversity regulates ecosystem predictability. Nature 390:162–165. 13. Moller, S., A. R. Pedersen, L. K. Poulsen, E. Arvin, and S. Molin. 1996. Activity and three-dimensional distribution of toluene-degrading Pseudomonas putida in a multispecies biofilm assessed by quantitative in situ hybridization and scanning confocal laser microscopy. Appl. Environ. Microbiol. 62:4632–4640. 14. Naeem, S., and S. Li. 1997. Biodiversity enhances ecosystem reliability. Nature 390:507–509. 15. Phelps, C. D., L. J. Kerkhof, and L. Y. Young. Molecular characterization of sulfate-reducing consortium which mineralizes benzene. FEMS Microbiol. Lett., in press. 16. Reysenbach, A. L., L. J. Giver, G. S. Wickham, and N. R. Pace. 1992. Differential amplification of rRNA genes by polymerase chain reaction. Appl. Environ. Microbiol. 58:3417–3418. 17. Rotthauwe, J. H., K. P. Witzel, and W. Liesack. 1997. The ammonia monooxygenase structural gene amoA as a functional marker—molecular finescale analysis of natural ammonia-oxidizing populations. Appl. Environ. Microbiol. 63:4704–4712. 18. Scala, D., and L. Kerkhof. 1998. Three nitrous oxide reductase (nosZ) genes from marine sediments and development of PCR primers for identification
APPL. ENVIRON. MICROBIOL. of denitrifiers. FEMS Microbiol. Lett. 162:61–68. 19. Schramm, A., L. H. Larsen, N. P. Revsbech, N. B. Ramsing, R. Amann, and K. H. Schleifer. 1996. Structure and function of a nitrifying biofilm as determined by in situ hybridization and the use of microelectrodes. Appl. Environ. Microbiol. 62:4641–4647. 20. Selvaratnam, S., B. A. Schoedel, B. L. McFarland, and C. F. Kulpa. 1997. Application of the polymerase chain reaction (PCR) and reverse transcriptase/PCR for determining the fate of phenol-degrading Pseudomonas putida ATCC 11172 in a bioaugmented sequencing batch reactor. Appl. Microbiol. Biotechnol. 47:236–240. 21. Smith, S. W., R. Overbeek, C. R. Woese, W. Gilbert, and P. M. Gillevet. 1994. The genetic data environment, an expandable GUI for multiple sequence analysis. Comput. Appl. Biosci. 10:671–675. 22. Suzuki, M. T., and S. J. Giovannoni. 1996. Bias caused by template annealing in the amplification of mixtures of 16S rRNA genes by PCR. Appl. Environ. Microbiol. 62:625–630. 23. Torsvik, V., R. Sorheim, and J. Goksoyr. 1996. Total bacterial diversity in soil and sediment communities: a review. J. Indust. Microbiol. 17:170–178. 24. Utkin, I. B., M. M. Yakimov, E. I. Kozlyak, and I. S. Rogozhin. 1990. Biological air purification methods. Appl. Biochem. Microbiol. 25:613–620. 25. Vainio, E. J., A. Moilanen, T. T. Koivula, D. H. Bamford, and J. Hantula. 1997. Comparison of partial 16S rRNA gene sequences obtained from activated sludge bacteria. Appl. Microbiol. Biotechnol. 48:73–79. 26. Van Groenestijn, J. W., and P. G. M. Hesselink. 1993. Biotechniques for air pollution control. Biodegradation 4:283–301. 27. Voytek, M. A., and B. B. Ward. 1995. Detection of ammonium-oxidizing bacteria of the beta subclass of the class Proteobacteria in aquatic samples with the PCR. Appl. Environ. Microbiol. 61:1444–1450. 28. Ward, B. B. 1990. Kinetics of ammonia oxidation by a marine nitrifying bacterium: methane as a substrate analogue. Microb. Ecol. 19:211–225. 29. Weckhuysen, B., L. Vriens, and H. Verachtert. 1994. Biotreatment of ammonia- and butanal-containing waste gases. Appl. Microbiol. Biotechnol. 42: 147–152. 30. Weeks, D. P., N. Beerman, and O. M. Griffith. 1986. A small-scale five-hour procedure for isolating multiple samples of CsCl-purified DNA: application to isolations from mammalian, insect, higher plant, algal, yeast, and bacterial sources. Anal. Biochem. 152:376–385. 31. Williams, T. O., and F. C. Millers. 1993. Composting facility odor control using biofilters, p. 262–281. In H. A. Hoitink and H. M. Keener (ed.), Science and engineering of composting. Renaissance Publishing, Columbus, Ohio.
