*For offprint requests: RGoodman@facstaff.wisc.edu

SUMMARY

Our overall goals are to develop new techniques for environmental monitoring and risk assessment based on a census of microbial communities and to determine how this census changes in response to environmental changes. The use of microbes as environmental indicators requires methods that are sensitive to subtle changes in the composition of bacterial communities and that allow identification of the organisms responsible for those changes. However, recent studies have determined that environmental assemblages of microorganisms contain an astonishing diversity of as yet undescribed organisms. To assess the vast diversity of uncultured organisms in environmental samples demands new biological and computational techniques that can accommodate this complexity. We are currently developing a molecular hybridization method, which utilizes an array of oligonucleotides derived from small subunit (SSU) rRNA gene sequences, for analysis of microbial communities. An initial hybridization array is currently being developed for analysis of soil microbial communities. To assist in the design of oligonucleotides for this initial array, we employed a molecular phylogenetic technique to obtain SSU rRNA gene sequences from soil. Soil microbial communities were sampled directly by isolation of total DNA from soil and SSU rRNA genes were amplified by polymerase chain reaction (PCR). In this report, we present the sequence and phylogenetic analyses of 179 cloned SSU rRNA genes isolated from an agricultural soil. Of this number, 144 clones were amplified using primers specific SSU rRNA genes of Bacteria. The remaining 35 clones, were amplified using primers specific for SSU rRNA genes of Archaea. These analyses identified a wide diversity of procaryotes in soil, some of which form novel lineages. The SSU rRNA gene sequences obtained from soil will be used for designing oligonucleotides for the hybridization array and for assessing the composition and diversity of soil and rhizosphere microbial communities.

Key words: Small subunit rRNA, oligonucleotide array, microbial diversity, environmental monitoring, molecular phylogeny

INTRODUCTION

We are currently developing a novel approach to risk assessment that is based on using microbes as indicators of ecosystem change. Microorganisms are likely to be sensitive indicators of environmental change due to their short life cycles, their physiological diversity, and their omnipresence in ecosystems. The use of microorganisms as environmental indicators requires methods that are sensitive to subtle changes in the composition of microbial communities and that also provide a signature of specific microbes that are associated with that change. Our approach utilizes the phylogenetic properties of SSU rRNA gene sequences. A hybridization array will be constructed using oligonucleotides derived from SSU rRNA gene sequences in existing databases and from novel sequences isolated from environmental samples. PCR amplification and labeling of microbial SSU rRNA genes from total DNA isolated from environmental samples will be applied to the array to determine patterns of hybridization. The resulting patterns will provide information about the diversity of the microbial community and its response to environmental change. Use of this technology could address the critical need for risk assessment of environmental change associated with introduction of transgenic and non-native organisms for a wide range of applications in agriculture and environmental remediation.

The hybridization array technology can be used to assess the composition and diversity of microbial communities in any environment. However, we are designing an initial hybridization array for the analysis of soil and rhizosphere communities. To aid in design of oligonucleotides for this initial array, we have amplified and cloned SSU rRNA gene sequences from members of both procaryotic phylogenetic domains, Archaea and Bacteria, in a soil microbial community. Phylogenetic analyses of these cloned SSU rRNA gene sequences revealed a remarkable diversity of procaryotes in soil, many of which belong to novel lineages.

MATERIALS AND METHODS

Extraction and purification of total DNA from soil sample. A subsurface (2-10 cm) soil sample was collected in August 1995, from the West Madison Agricultural Research Station, Madison, WI, frozen on dry ice, and processed immediately. Total DNA was isolated by three different extraction methods which were designed to isolate total DNA from a variety of cell types and were based on the method described by Porteus et al. (Porteus et al., 1994). Total DNA was purified by sequential organic extractions and ultrafiltration using Microcon-100 microconcentrators (Amicon, Inc., Beverly, MA). Total DNA obtained by these methods was suitable as template for PCR amplification of SSU rRNA gene sequences from the soil community.

Amplification of SSU rRNA genes in the soil microbial community. Purified total DNA was used as template for PCR amplification of procaryotic SSU rRNA gene sequences. The oligonucleotide primer sequences used for amplification of the SSU rRNA gene sequences were 23FPL (Barns et al., 1994) (5'-GCGGATCCGCGGCCGCTGCA GAYCTGGTYGATYCTGCC-3'), 27F-Not I (5'-GCGGATCCGCGGCCGCAGAGTTT GATCMTGGCTCAG-3'), and 1492r (Lane, 1991) (5'-GGYTACCTTGTTAACGACTT-3'), where R is a purine, Y is a pyrimidine, and M is either an A or C. The underlined sequences in oligonucleotide primers 23FPL and 27F-Not I correspond to a Not I recognition site that was used for directional cloning of the amplified SSU rRNA gene sequences. The primer sets 23FPL and 1492r or 27-Not I and 1492r were designed to amplify selectively nearly full-length (1.4 kb) SSU rRNA genes from most Archaea or Bacteria, respectively.

PCR amplification was performed with a Robocycler Gradient 96 Temperature Cycler (Stratagene, La Jolla, CA). The cycling parameters were: 1 min denaturation at 94C followed by 30 cycles of 94C for 30 sec, 55C for 1.5 min, and 72C for 2.5 min. PCR amplification reaction conditions were: 10 ng of template, 40 pmoles of each primer, 200 M of each dNTP (Boehringer Mannheim Corp., Indianapolis, IN), 1.25 U Pfu DNA polymerase (Stratagene, La Jolla, CA), and 1X Pfu DNA polymerase reaction buffer in a 50 l reaction volume. Positive and negative control reactions were performed for both primer sets. Positive control reactions contained either 20 ng of Sulfolobus acidocaldarius or 10 ng of Escherichia coli genomic DNA. Negative control reactions contained no exogenous template. Amplified products of the expected size (1.4 kb) were observed with the positive control reactions using both primer sets. No amplified products were observed in the negative control reactions.

Cloning of amplified SSU rRNA genes sequences. Amplified products from four independent PCR reactions were pooled, purified with QIAquick PCR purification columns (Qiagen Inc., Chatsworth, CA), digested with the restriction enzyme Not I (Stratagene, La Jolla, CA), and again purified by spin column purification. Digested products were directionally cloned by ligation to Not I- Hinc II digested pGEM11Zf(+) (Promega Corp, Madison, WI). Aliquots of the ligation mixture were transformed into Escherichia coli DH5a by electroporation using a Gene Pulser Transfection Apparatus (BIORAD, Richmond, CA). Transformants were screened for -complementation on MacKonkey agar (DIFCO laboratories, Detroit, MI) supplemented with ampicillin (100 g/ml).

Nucleotide sequencing of cloned SSU rRNA genes. The nucleotide sequences of the cloned SSU rRNA genes were determined using an automated DNA sequencer (model 377; Applied Biosystems, Foster City, CA). Plasmid DNA was purified with QIAprep plasmid purification kits (Qiagen Inc., Chatsworth, CA) and used as template for PCR cycle sequencing with Prism Ready Reaction Dyedeoxy Terminator Sequencing Kit (Applied Biosystems, Foster City, CA). Oligonucleotide primers used for sequence analysis of the cloned SSU rRNA gene sequences were pUC/M13 reverse (Promega Corp., Madison, WI) and a SSU rRNA-specific primer (Lane, 1991), 519r (5'-GWATTACCGCGGCKGCTG-3'), where W is an A or T and K is a G or T. Use of these sequencing primers resulted in the determination of approximately 450 bp of the 5' terminus of the SSU rRNA gene sequence. Nucleotide sequences of the cloned SSU rRNA gene sequences were obtained by sequencing both template strands at least twice.

Sequence analysis of cloned SSU rRNA gene sequences. Analyses of the cloned SSU rRNA gene sequences were performed using SIMILARITY_RANK and ALIGN_SEQUENCE from the Ribosomal Database Project (RDP) (release date of May 18, 1995) at the University of Illinois in Urbana (Maidak et al., 1996). These analyses were used to determine that the cloned sequences were SSU rRNA gene sequences and to estimate the degree of similarity to other SSU rRNA gene sequences in the Procaryotic SSU rRNA Database at the RDP. Nucleotide sequences of the cloned SSU rRNA gene sequences were manually aligned into putative secondary structures and were also submitted to the CHECK_CHIMERA program at the RDP to detect the presence of chimeric artifacts.

Phylogenetic analysis of cloned SSU rRNA gene sequences. The cloned SSU rRNA gene sequences were manually aligned to SSU rRNA gene sequences from the RDP based on primary and secondary structure considerations using the Genetic Data Environment (GDE) multiple sequence editor distributed by the RDP. Phylogenetic analyses were restricted to nucleotide positions that were unambiguously alignable in all sequences. Phylogenetic analyses using neighbor joining and parsimony methods were performed using the Phylogenetic Inference Package (PHYLIP) (version 3.57c, J. Felsenstein, University of Washington, Seattle). Maximum likelihood analyses were performed using FASTDNAML (version 1.1.1; distributed by RDP). Bootstrap analysis (Felsenstein, 1985) was used to provide confidence estimates for phylogenetic tree topologies.

RESULTS

To assess the diversity of procaryotes in soil, total DNA was isolated directly from soil and SSU rRNA gene sequences from Archaea and Bacteria were selectively amplified by PCR and cloned. Partial nucleotide sequence information was obtained for 179 cloned SSU rRNA gene sequences (144 clones of Bacteria, 35 clones of Archaea). This sequence information was used to determine that the cloned inserts were SSU rRNA gene sequences and to identify unique SSU rRNA gene sequences within the clone collection.

Sequence and phylogenetic analyses of cloned SSU gene sequences of Archaea. Of the 35 cloned archaeal sequences we examined, 18 were unique within our clone collection. These cloned sequences had the highest similarity (78-84% nucleotide sequence identity) to members of the planktonic Archaea, which is placed in the Crenarchaeota, one of the two kingdoms of Archaea. The archaeal clone sequences in our collection were more similar to each other (the least having 88% nucleotide sequence identity) than to any database sequences. Nucleotide signature and phylogenetic analyses determined that the archaeal sequences obtained from soil formed a tightly-clustered, deeply-diverging group which is affiliated to the planktonic Archaea (Figure 1).

Sequence and phylogenetic analyses of cloned SSU gene sequences of Bacteria. The collection of cloned SSU rRNA gene sequences of Bacteria was far more diverse than that of our archaeal clone collection. Out of 144 cloned sequences examined, 135 were unique within this collection. None of the cloned sequences were identical to any previously identified SSU rRNA gene sequence and most had low similarity (78-80% nucleotide sequence identity) to database sequences. Phylogenetic analyses of these cloned sequences determined that approximately half of the unique clone sequences could be placed into four known phyla of the domain Bacteria (Figure 2). These phyla are the High G+C Gram positive, the Low G+C Gram positive, the Proteobacteria, and Cytophaga/Flexibacter/Bacteroides phyla. The other half of the cloned sequences could not be confidently placed into any of the known phyla of Bacteria, indicating that these sequences constitute new lineages of Bacteria

DISCUSSION

As shown by this report, soil microbial communities are far more complex than can be demonstrated by standard methods such as culturing. The molecular identification of novel lineages of Archaea and Bacteria in soil provides a wealth of previously unknown SSU rRNA sequence information. This molecular information is useful for designing oligonucleotides for the hybridization array and for assessing the soil microbial diversity and the response of the soil microbial communities to environmental change. We will utilize the hybridization array technology to assess the changes in the soil microbial community in response to treatment of soil with agricultural pesticides, heavy metals, a bacterial agent for crop disease control, and the introduction of a transgenic plant. For the latter treatment, we are using a line of Russet Burbank potato transgenic for SAR8.2d, which is a member of a tobacco gene family involved in the Systemic Acquired Resistance response. A study by us showed that tobacco transgenic for SAR8.2d are resistant to the oomycete pathogen Pythium torulosum. Field trials with the transgenic potato line have been performed. To determine if expression of the transgene affects the composition and diversity of rhizosphere microbial communities, we are using culturing and molecular analyses. Culturing analysis did not detect significant changes in the compostion and diversity of rhizosphere communities between transgenic and control lines. A molecular analysis of these rhizosphere communities, which involves use of a hybridization array, is currently underway.

ACKNOWLEDGMENTS

We gratefully acknowledge James S. Ireland for assistance with the computational analyses. This work was supported by a National Institutes of Health postdoctoral fellowship to S.B.B., an Environmental Protection Agency cooperative agreement (CR822902-01-0), a grant from the U.S. Department of Agriculture CSEES Biotechnology Risk Assessment Program (94-39210-0559), and a grant from the McKnight Foundation. Computational analyses were made possible in part by instrumentation provided by an Industrial and Economic Development and Research Grant, University-Industry Relations Program at the University of Wisconsin-Madison.

REFERENCES

Barns SM, Fundyga RE, Jeffries MW, Pace NR. 1994. Remarkable archaeal diversity detected in a Yellowstone National Park hot spring. Proc. Natl. Acad. Sci. USA 91: 1609-1613.

Felsenstein J. 1985. Confidence limits on phylogenies: an approach using the bootstrap. Evolution 39: 783-791.

Lane DJ. 1991. 16S/23S rRNA sequencing. In: Stackebrandt, E and Goodfellow M. (eds) Nucleic Acid Techniques in Bacterial Systematics. John Wiley and Sons,Chichester, New York, Brisbane, Toronto, Singapore, pp 115-176.

Maidak BL, Olsen GJ, Larsen N., Overbeek R., McCaughey MJ, Woese CR. 1996. The Ribosomal Database Project (RDP). Nuc Acids Res 24: 82-85.

Porteus, LA, Armstrong AJ, Seidler RJ, Watrud LS. 1994. An effective method to extract DNA from environmental samples for polymerse chain reaction amplification and DNA fingerprint analysis. Curr. Microbiol. 29: 301-307.

Figure 1. Inferred unrooted phylogenetic tree of archaeal SSU rRNA gene sequences cloned from soil, illustrating close affiliation of these sequences (designated SCA clones) with those of the planktonic Crenarchaeota (Group I marine sequence). Tree was inferred by neighbor joining analysis of 637 homologous positions of sequence from each organism or clone. Scale bar represents 10 mutations per 100-nt of homologous sequence positions. The percentage of 100 bootstrap resamplings that support some of the major topological elements in neighbor joining (above line) and parsimony (below line) analyses is indicated.

Figure 2. Inferred unrooted phylogenetic tree of bacterial SSU rRNA gene sequences isolated from soil. Cloned sequences that had an affiliation to known phyla of Bacteria are designated by the phylum name. Groups of cloned sequences that could not be assigned to any of the known phyla of Bacteria are designated as Groups I-VII. Tree was inferred by neighbor joining analysis of 276 homologous positions of sequence from each organism or clone. Scale bar represents 10 mutations per 100-nt of homologous sequence positions.