For information about Web accessibility, please contact the Webmaster at

Haverford College

Marian E. Koshland Integrated Natural Sciences Center

16S rRNA sequences reveal numerous uncultured microoganisms in a natural community

Letters to Nature

Reproduced from a Letters to Nature article: Ward, D.M., Weller, R., Bateson, M.M. 16S rRNA sequences reveal numerous uncultured microorganisms in a natural community. Nature, 345: 63-65 (1990)

David M. Ward, Roland Weller & Mary M. Bateson Department of Microbiology, Montana State University, Bozeman, Montana 59717, USA

16S rRNA sequences reveal numerous uncultured microoganisms in a natural community

Letters to Nature

Reproduced from a Letters to Nature article: Ward, D.M., Weller, R., Bateson, M.M. 16S rRNA sequences reveal numerous uncultured microorganisms in a natural community. Nature, 345: 63-65 (1990)

David M. Ward, Roland Weller & Mary M. Bateson Department of Microbiology, Montana State University, Bozeman, Montana 59717, USA

MICROBIOLOGISTS have been constrained in their efforts to describe the compositions of natural microbial communities using traditional methods.  Few microorganisms have sufficiently distinctive morphology to be recognized by microscopy.  Culture-dependent methods are biased, as a microorganism can be cultivated only after its physiological niche is perceived and duplicated experimentally.  It is therefore widely believed that fewer than 20% of the extant microorganisms have been discovered1,2, and that culture methods are inadequate for studying microbial community composition3-7.  In view of the physiological and phylogenetic diversity among microorganisms8, speculation that 80% or more of microbes remain undiscovered raises the question of how well we know the Earth's biota and its biochemical potential.  We have performed a culture-independent analysis of the composition of a well-studied hot spring microbial community, using a common but distinctive cellular component, 16S ribosomal RNA.  Our results confirm speculations about the diversity of uncultured microorganisms it contains.

We developed a cloning method for retrieving naturally occurring 16S rRNA sequences9 that is more selective than methods previously proposed for recovering 16S rRNA genes10,11.  The selectivity is based on the synthesis of complementary cDNA from 16S rRNA.  Cloned 16S rcDNA sequences from the community are then compared to 16S rRNA sequences of microorganisms isolated from this or similar habitats.

We studied 16S rcDNA sequences from the 55°C cyanobacterial mat of Octopus Spring, Yellowstone National Park.  This totally microbial community should have a restricted12 and relatively constant13 composition.  It has been well-characterized with respect to component microorganisms by microscopic and culture methods13; a single cyanobacterium, Synechococcus lividus, and a photosynthetic bacterium, Chloroflexus aurantiacus, are thought to be responsible for production of the mat, and several bacteria which may participate in mat decomposition have been isolated.  The mat has also been a rich source of phylogenetically interesting bacteria8Studies of lipid cellular components have indicated, however, the presence of as yet undescribed community members14,15.

Detailed analysis of primary and secondary structure provides convincing evidence that the cloned 16S rRNA sequences correspond to 16S rRNA sequences.  For example, Fig. 1 shows how primary sequence conservation and compensatory mutations in double-stranded regions accommodate the folding of a 16S rcDNA sequence recovered from the mat to match a secondary structure reported for the Escherichia coli 16S rRNA.  Most of the sequence similarity data we present are from a conservative, restricted analysis which limits sequence comparison to regions in which nucleotides can be unambiguously aligned.  This ensures that differences are not due to misalignment within highly variable sequence regions16,17.  The variable regions were included in unrestricted comparisons as a more rigorous test of identity between sequences.

Similarity values among 16S rcDNA sequences retrieved from the Octopus Spring mat (Table 1) reveal eight distinct sequence types (A-H) which must be contributed by eight different community members.  Five sequences are identical in both restricted and unrestricted analysis (type-A).  Another sequence (type-B) is highly related, but eleven differences distinguish it from type-A sequences.  Two other sequence types (C and D) are revealed as pairs which are also identical in restricted and unrestricted analysis.  Four additional sequence types (E, F, G and H) are represented only once.

None of the eight Octopus Spring 16S rcDNA sequence types is identical with a 16S rRNA sequence of any microorganism isolated from the mat or other similar geothermal habitat (Table 2).  The highest similarity is between type-C sequences and the sequence of C. aurantiacus (94.7% in restricted analysis), but the lack of identity is verified by a lower similarity value (85.9%) in unrestricted analysis.  Consequently, we believe that the eight sequence types we have recovered from the mat are contributed by microbial inhabitants which have not been cultured from this community.  Our confidence in this interpretation comes from several lines of reasoning.  First, the redundancy of type-A, type-C and type-D 16S rcDNA sequence determination18-20 demonstrate the high fidelity of these methods in copying 16S rRNA sequence information.  Second, the method is equally faithful in recovering 16S rcDNA sequences of isolated bacteria grown in pure culture20.  Third, the 16S rcDNA sequence data provide convincing evidence, at both the primary and secondary structure level, that similarity values are based on unambiguous nucleotide differences occurring at nonrandom positions in

the molecule (mainly in variable regions as expected, and with compensatory differences at complementary positions in double-stranded regions, see Fig. 1).  Finally, we have made similar observations in other Octopus Spring 16S rcDNA libraries.

One valuable application of the use of 16S rRNA sequences in characterizing microbial communities is to provide insight into the phylogenetic types of uncultured community members, as well as a basis for designing hybridization probes to study their autecology 10,11,21-23.  Our data show that all eight mat sequence types are of eubacterial rather than archaebacterial origin, as similarity values with eubacterial isolates are typical of inter- or intraphylum difference whereas similarity values with Methanobacterium thermoautotrophicum, the only archaebacterial isolate from the mat, are typical or interkingdom difference.  Two sequence types can be placed within specific C. aurantiacus sequences suggests that these sequences are contributed by a member of the green non-sulphur eubacterial phylum.  The type-H sequence, while dissimilar to sequences of mat isolates, exhibits 94.3% similarity with the sequence for Spirochaeta halophila, suggesting it is contributed by an as yet uncultivated spirochaete inhabitant of the mat.  These affinities are further supported by preliminary distance matrix tree analysis and by nucleotide and oligonucleotide signatures characteristic of members of these phyla8,24.  Type-A and ?B sequences show 90.6% to 91.9% similarity with sequences representative of the cyanobacterial (S. lividus) or Gram-positive (Clostridium thermoautotrophicum) eubacterial phyla.  Whereas nucleotide and oligonucleotide signatures do not resolve placement into one of these phyla, the sequences show greater similarity with that of another cyanobacterium, Anacystis nidulans25 (93.2% and 94.5% in restricted analysis for type-A and ?B sequences, respectively).  We will explore further the possibility of a cyanobacterial origin of these sequences, as this would challenge the notion derived from microscopic and enrichment culture evidence that S. lividus is the sole cyanobacterium building the mat.  Other sequence types bear too little similarity with known eubacterial phyla to presently enable further assignment; some of these may represent novel phyla.

The ecological message is clear?this well-studied microbial mat contains a large number of community members which have never been cultured from this habitat.  The same is almost certainly true of most, if not all, microbial communities, as our knowledge of them is similarly based on culture-dependent methods.  Analysis of marine microflora by 16S rRNA-based methods has led to similar observations26.  Further work will be required before we can know more precisely the percentage of extant microorganisms pure culture collections represent.  In this community the number of new inhabitants so far suggested by our method is comparable to the number of cultured inhabitants is likely to be, as speculated, quite high.  This new culture-independent method is beginning to reveal some of the unexplored diversity within the microbial world and should eventually provide a more objective approach by which to understand the composition of natural microbial communities27.

Received 19 January; accepted 5 March, 1990.

  1. Wolfe, R. S. in The Prokaryotes (eds Starr, M.P., Stolp, H., Truper. H.G., Balows. A. & Schlegel. (H.G.) v-vi (Springer-Verlag, Berlin. 1981).
  2. Wayne, L. G. et al. Int. J. syst. Bacteriol. 37,4630464 (1987).
  3. Atlas, R. M. in Current Perspectives in Microbial Ecology (eds Klug, M.J. & Reddy, C.A.) 540-545 (Am. Soc. Microbiology, Washington, DC. 1984).
  4. Williams, S. T., Goodfellow, M & Vickers, J. C. Symp. Soc. gen. Microbiol. 36, 219-256 (1984).
  5. Staley, J. T. in Microbiology-1980 (ed Schlessenger, D.) 321-322 (Am. Soc. Microbiology. Washington DC, 1980).
  6. Brock, T. D. symp. Soc. gen. Microbiol. 41, 1-17 (1987).
  7. Rosswall, T. & Kvillner, E. Adv. microb. Ecol. 1, 1-48 (1978).
  8. Woese, C. R. Microbiol. Rev. 51, 221-271 (1987).
  9. Weller, R. & D. M. Ward, Appl. environ. Microbiol. 55, 1818-1822 (1989).
  10. Pace, N. R., Stahl, D. A., Lane, D. J. & Oslen, G. J. Adv. microb. Ecol. 9, 1-55 (1986).
  11. Olsen, G. J., Lane, D. J., Giovannoni, S. J. & Pace, N. R. A. Rev. Microbiol. 40, 337-365 (1986).
  12. Brock, T. D Thermophilic Mcrioorganisms and Life at High Temperatures (Springer. Berlin. 1978).
  13. Ward, D. M., Tayne, T. A., Anderson, K. L. & Bateson, M. M. Symp. Soc. gen. Microbiol. 41, 179-210 (1987).
  14. Ward, D. M., Brassell, S. C. & Eglinton, G. Nature 318, 656-659 (1985).
  15. Ward, D. M. et al. in Microbial Mats: Physiological Ecology of Benthic Microbial Commnities (eds Cohen, Y. & Rosenberg, E.) 439-454 (Am. Soc. Microbiol., Washington DC, 1989).
  16. Olsen, G. J., Cold Spring Harb. Symp. quant. Biol. 52, 829-837 (1987).
  17. Olsen, G. J. Meth. Enzym. 164, 793-812 (1988).
  18. Bateson, M. M., Wiegel, J. & Ward, D. M. Syst. appl. Microbiol. 12, 1-7 (1989).
  19. Bateson, M. M., Thibault, K. J. & Ward, D. M. Syst. appl. Microbiol. Rev. (in the press).
  20. Ward, D. M., Weller, R. & Bateson, M.M FEMS Microbiol. Rev. (in the press).
  21. Stahl, D. A., Flesher, B., Mansfield, H. R. & Montgomery, L. Appl. environ. Microbiol. 54, 1079.1084 (1988).
  22. Giovannoni, S. J., DeLong, E., Olsen, G. J. & Pace, N. R. J. Bacteriol. 170, 720-726 (1988).
  23. DeLong, E. F., Wickham, G. S. & Pace, N. R. Science 243, 1360-1363 (1989).
  24. Woese, C. R., Stackebrandt, E., Macke, T. J. & Fox, G. E. System. appl. Microbiol. 6, 143-151 (1985).
  25. Tomoika, N. & Sugiura, M. Molec. Gen Genet. 191, 46-50 (1983).
  26. Giovannoni, S., Britschgi, T. & Feld, K., Nature 345, 60-63 (1990).
  27. Ward, D.M. in Structure and Function of Biofilms (eds Characklis, W.G. & Wilderer, P.A.) 145-163 (Chichester, UK. 1989).
  28. Oyaizu, H., Debrunner-Vossbrinck, B., Mandelco, L., Studier, J. A. & Woese, C. R. System appl. Microbiol. 9, 47-53 (1987).
  29. Gutell, R. R., Weiser, B., Woese, C. R. & Noller, H. F. Prog. Nucleic Acid Res. 32, 155-216 (1985).
  30. Gray, M. W., Sankoff, D. & Cedergren, R. J. Nucleic Acids Res. 12, 5837-5852 (1984).

ACKNOWLEDGEMENTS. We thank W. G. Weisberg and C. R. Woese for the S. halophila sequence and Gary Oslen, Gijs Kuenen, Geoffrey Eglinton, Mary Allen, Jean Starkey, Jim Cutler, Sandy Ewald and Alyson Ruff for their comments on the manuscript.  We appreciate the assistance of the US National Park Service in Yellowstone National Park.  This work was supported by the National Science Foundation.

Dr. Ward interview:

Dr. David Ward, Professor, Department of Land Resources and Environmental Sciences, Montana State University

1.How did Dr. Ward become interested in studying cyanobacterial mat communities?

Dr. Ward's interest in ecology and natural history began on camping trips to the Adirondacks that his parents took him on throughout his childhood. Many years later as a graduate student at the University of Wisconsin-Madison, Dr. Ward chose to work in Tom Brock's lab (** ) because of his interest in microbial ecology. While at the University of Wisconsin, Dr. Ward studied oil biofuel degradation in the aquatic community of Lake Mendota (see for more details). After taking Dr. G. Fred Lee's course in the water chemistry of Lake Mendota, Dr. Lee's natural view of the lake environment began to greatly influenced Dr. Ward and change his perspective on the work he was doing. Dr. Ward began comparing the chemistry of the lake to the chemistry of the culture media he was using and he observed large differences between the two. Although Dr. Ward was then nearing the end of his thesis, this natural view of microbial communities became extremely influential in his future research.

Dr. Ward realized that a large unresolved problem in the field of microbiology involved the idea that nature was full of things that had not been cultivated in a laboratory setting and with that idea in mind Dr. Ward became interested in the growing field of using lipid biomarkers to infer the types of bacteria present in a community. Also, while working in Dr. Brock's lab, other students in the lab were traveling to Yellowstone National Park to study the microbial mat community and Dr. Ward began to view the unique microbial mat community of Yellowstone as a great model in which to try to study lipid biomarkers as well an ideal extreme environment in which to test the idea that nature was full of bacteria that had yet to be cultured in the lab.

2.What were some of the most challenging parts of the experiments conducted in the paper?

One of the most challenging parts of the experiment was designing a methodology to isolate the 16S rRNA gene from the samples. Dr. Norman Pace, who first introduced Dr. Ward to the idea of examining microbial diversity via the 16S rRNA gene, was trying to clone the gene using a shot-gun cloning approach (see Dr. Pace's lab website: Dr. Pace's approach, though, required isolating the 16S rRNA gene among a pool of thousands of other genes, which was very labor-intensive. Dr. Stephen Giovannoni (whose paper was published back-to-back with Dr. Ward's: wanted to use PCR to amplify the 16S rRNA gene. Dr. Ward's approach, though, was to capture the natural amplification of the 16S rRNA gene in vivo, use reverse transcriptase to make cDNA, and then sequence the more stable cDNA copy of the gene.

The challenge in the RNA approach was being able to successfully reverse-transcribe the RNA. Dr. Ward describes a couple of intense weeks discussing troubleshooting options for the RNA extraction procedure while going on daily runs at the Montana State University indoor track with Roland Weller, a PhD student in his lab. Together, through their discussions, they were ultimately successful in modifying the protocol to extract cDNA to send for sequencing.

The time between when Dr. Ward began his lab and when a paper was published regarding the 16S rRNA gene was seven years, which is longer than the duration of a typical grant length. During this seven-year period, though, the National Science Foundation (NSF) (** refunded Dr. Ward's work twice, which allowed Dr. Ward to continue the 16S rRNA work without publishing a paper on it until 1990. Without NSF's willingness to continue to fund his work, the development of the culture-independent approach to cultivating bacteria might have been put in jeopardy.

3.Were the results Dr. Ward obtained surprising to him in any way? Why or why not?

Although there were people in the field at the time who speculated that culturing techniques were highly imperfect based on the observation that direct bacterial counts were always a few orders of magnitude higher than cultured bacteria counts, Dr. Ward said that he had placed too much faith in the internal standards of the Yellowstone microbial mat community generated via culturing techniques. He did not anticipate the extent to which the bacteria cultured from the community in the lab differed from the natural diversity of bacteria present in the mat community detected via culture-independent techniques. None of the sequences that were isolated via the culture-independent technique were the same as those well-studied from the Yellowstone microbial mat community via culture-dependent techniques.

4.How has this paper affected Dr. Ward's subsequent work?

After Dr. Ward's paper on 16S rRNA sequences (annotated above), the work he pursued immediately afterwards involved asking whether different sequence variants of cyanobacteria identified by the culture-independent technique were closely related species or members of a single cyanobacterial species. Dr. Ward also became interested in placing the organisms in their ecological context. His work has revealed that some organisms with identical 16S rRNA sequences inhabit different ecological niches, which reveals that in many cases the molecular resolution of the 16S rRNA gene is not fine enough. (16S rRNA sequences are now estimated to lump together about 10-20 different bacterial species on the average)

Using the thermal springs present in Yellowstone National Park, Dr. Ward has measured the distribution of bacteria over the thermal gradient range of 72°C to 50°C. Over this 22°C range, significantly different distributions of bacteria are present at various temperatures. To read more about the paper published from his lab that describes the diversity of the bacteria as a function of temperature environment see Ward/Ferris et al. 1997. (

Dr. Ward is currently still interested in understanding not only who is present in the thermal hot spring environments of Yellowstone National Park, but also who is where within the environment. His lab is currently focused on how different species of bacterial in the thermal hot springs adapt to varying light intensities, which vary with depth and seasonally.

5.How has the field changed since the paper was published?

Since the paper was published in 1990, the number of entries into GenBank of environmental clones has increased exponentially. By 1997, the number of environmental clones entered into GenBank surpassed the number of entries of cultured microorganisms (Rappé et al, 2003).

It is evident that Dr. Ward has made a profound impact onto the field, with his 1990 paper currently having been cited by over 1200 other published papers in the field. Although Dr. Ward is very proud to have made such a large contribution to the field with his 16S rRNA work, he believes that most people do not understand that when they are using the 16S rRNA gene to identify bacterial species in actuality they are undoubtedly grouping together a set of species. He emphasizes that it is important for his colleagues to put the bacteria they are working with in ecological context because each species has an important and unique role in its environment.

6.Are there any pitfalls to 16S rRNA sequencing? Are there better alternatives (eg metagenome sequencing)?

An important pitfall of 16S rRNA sequencing is that its resolution is limited: it lumps too many different bacterial species together. Species of bacteria can be more closely related than previously thought so the field currently needs to use techniques that offer more molecular resolution to differentiate species from one another. One way to go about doing that would be to use a different gene found in bacteria that has evolved more rapidly than the 16S rRNA gene.

Dr. Ward describes how metagenomics also does not provide investigators with more molecular resolution to differentiate species because looking at many genes at the same time complicates the analysis. Dr. Ward does recommend using iTag sequencing (bar code analysis) to try to understand species diversity. With iTag sequencing the DNA of organisms is extracted and amplified by PCR. The primers used in the PCR have a specific barcode attached to them and after the amplified DNA picks up the tag, the DNA/tagged primer complexes are placed on a chip to bind to their particular barcode associated region. The ratio of sequence binding to a particular barcode region on the chip compared to regions of the chip associated with other barcodes allows investigators to gauge relative levels of abundance present in the population as a whole. Using iTag sequencing, if one suspects that two different sequences are the same species, he/she can alter an environmental variable equally for both organisms with those sequences and observe whether the level of abundance of the two sequences on the iTag chip co-varies. One would suspect that if the two sequences were the same species, the sequences would co-vary and co-occur together.

7.Your paper was published back-to-back with another paper about the Sargasso Sea bacterioplankton community, do you think that influenced the interpretation of your paper? How so?

Although Dr. Ward and Dr. Giovannoni were both together for a short time in Dr. Norman Pace's lab (Dr. Ward on sabbatical and Dr. Giovannoni as a visiting graduate student), they approached sequencing the 16S rRNA gene from different angles (see question 2 for more details). Drs. Pace, Ward, and Giovannoni all perceived the importance of understanding how representative the species cultured in lab were to their respective environments, but Drs. Ward and Giovannoni did not coordinate submitting their papers to Nature together. Dr. Ward believes that the fact that Nature received both their papers around the same time helped to get their papers accepted and published by Nature because both papers reported the same fundamental findings from two very different environments which was something the journal could not ignore. (see Dr. Gary Olsen's News and Views article for more details,

8.What kind of advice would you give to people going into your field today?

Dr. Ward advises people going into the field of microbiology and research today to not necessarily wait until a method is perfect to use it. The method used to sequence the 16S rRNA gene in the paper was definitely not perfect. The method was finicky and had to troubleshoot frequently, but Dr. Ward and his colleagues were still be able to use it to advance the field.

Dr. Ward also advises young scientists to work hard, but also emphasizes the importance of taking breaks and having interests outside the field of science. The importance of a balanced lifestyle is not only important, according to Dr. Ward, but also necessary to be able to work in an effective and efficient manner in a research setting.

9. When you're not doing lab-related things, what kind of activities do you enjoy doing in your spare time?

Dr. Ward enjoys spending time with his family and living with them in Montana. He enjoys hiking, backpacking, and skiing on the nearby slopes in the winter and he does many of these activities with members of his family. He also likes reading non-fiction as well as scuba diving. Dr. Ward enjoys scuba diving because he enjoys observing the beauty and the diversity of life present in the ocean.

About Me


My name is Jordana Bloom and I will be graduating this May (Haverford College Class of 2014). As a Biology major and Psychology minor, I enjoyed many of the classes I took in those departments. While at Haverford, I have had the opportunity to play on the Haverford College Women's Varsity Tennis Team, be a Peer Awareness Facilitator in the Customs Program, conduct research in the Biology Department, as well as be a Guided Learning Group Instructor and Peer Tutor for the Biology Department.

I became interested in the project after learning about it in Dr. Okeke's Microbiology course and I became interested in the particular topic of my project after my Junior Year Superlab course with Drs. Okeke and Wilson. I have enjoyed this project immensely as it has allowed me to learn more about 16S rRNA sequencing and to make a noteworthy research paper into an annotated and readily available teaching tool. I learned even more than I imagined I would while conducting the project and as a describe in the Acknowledgments section, I could not be more appreciative of Dr. Okeke and Mr. Moore for their assistance with the project as well as with Dr. Ward for being willing to meet with me and the Haverford College Biology Department for funding the project.

I am proud to say that next year I will be attending Cornell University's PhD program in Biochemistry, Cell and Molecular Biology where I know the skills I learned while conducting this project, and my Haverford education as a whole, will allow me to make the most of my time there.


Thanks to Dr. Ward: I would like to thank Dr. Ward for meeting with me to assist with enriching the annotation of his paper. It was an honor to meet with a scientist as distinguished and influential to the field of microbiology as Dr. Ward. I learned much more from Dr. Ward than the invaluable insight into his paper. I learned how to think with an ecological perspective and about how to be a scientist, and by the end of my short visit, I knew that I would treasure the opportunity I had to visit him for many years to come.

Thanks to Professor Okeke: I would like to thank Professor Okeke for introducing me to this project and for being a supportive mentor. I always enjoyed our meetings regarding the project and finishing this project is definitely a bittersweet moment for me.

Thanks to Mr. Moore: I would like to thank Mr. Moore for using the time he could have spent on other projects to work on the technical aspects of mine. Without his help, the project would not be in its current format for people to access and view.

Thanks to the Haverford College Biology Department: I would like to thank the Haverford College Biology Department for supporting the project. The support of the Department allowed me to visit Bozeman, Montana to interview Dr. Ward in-person.