Molecular analysis of hot spring microbial mats to study bacterial diversity and physiology by Stephen Charles Nold A thesis submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Microbiology Montana State University © Copyright by Stephen Charles Nold (1996) Abstract: Molecular studies investigating 16S rRNA sequence diversity in cyanobacterial mat communities inhabiting hot springs in Yellowstone National Park have revealed that these communities contain numerous uncultivated microbial species. Here, attempts were made to cultivate from one of these mat communities the aerobic chemoorganotrophic bacteria whose 16S rRNA nucleotide sequences were previously observed using molecular retrieval techniques. By utilizing serial dilution enrichment culture and a variety of enrichment conditions, a diversity of bacterial isolates was obtained. 16S rRNA sequence analysis revealed seven genotypically distinct isolates, including Thermus, proteobacterial, and Gram positive representatives. However, only one of these isolates, a β-proteobacterium, contained a 16S rRNA sequence previously observed in Octopus Spring mat. These results illustrate the differing views of microbial community composition which cultivation and molecular techniques provide, and demonstrate the problems encountered when using cultivation approaches to associate microbial activity with bacterial populations whose 16S rRNA sequences were detected in natural samples. One cultivation-independent approach to associate bacterial activity with retrieved 16S rRNA sequence types would be to selectively capture rRNA molecules synthesized by actively growing microorganisms incubated in the presence of a radiolabeled substrate, then quantify the relative extent of radiolabel incorporation into specific 16S rRNA molecules. Initial studies investigating the feasibility of this approach revealed that although logarithmically growing cyanobacterial cells incorporated photosynthetically fixed 14CO2 into rRNA, cyanobacteria inhabiting hot spring mats predominately incorporated 14CO2 into polyglucose during periods of illumination (between 77% and 85% of total incorporated carbon). Although photosynthetically active, the cyanobacteria of these mat communities do not appear to be rapidly growing, since only limited synthesis of growth-related macromolecules was detected. The fate of polyglucose reserves was investigated by allowing mat cyanobacteria to photoassimilate 14CO2 into polyglucose, then transferring samples to the dark, anaerobic conditions which mat communities experience at night. Radiolabel in the polysaccharide fraction decreased 74.7% after 12 hours dark incubation, of which 58.5% was recovered in radiolabeled fermentation products (i.e. [14C]acetate, 14CO2, and [14C]propionate). These results indicate tightly coupled carbon fixation and fermentative processes, and the potential for significant carbon transfer from primary producers to heterotrophic members of these cyanobacterial mat communities.  MOLECULAR ANALYSIS OF HOT SPRING MICROBIAL MATS TO STUDY BACTERIAL DIVERSITY AND PHYSIOLOGY by Stephen Charles Nold A thesis submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Microbiology MONTANA STATE UNIVERSITY-BOZEMAN Bozeman, Montana September 1996 ^113 ii APPROVAL of a thesis submitted by Stephen Charles Nold This thesis has been read by each member of the thesis committee and has been found to be satisfactory regarding content, English usage, format, citations, bibliographic style, and consistency, and is ready for submission to the College of Graduate Studies. Dr. David M. Ward (Signature) Date Approved for the Department of Microbiology Dr. Al J. Jesaitis (Signati Approved for the College of Graduate Studies Dr. Robert L. Brown (Signature) Date Date iii STATEMENT OF PERMISSION TO USE In presenting this thesis in partial fulfillment of the requirements for a doctoral degree at Montana State University-Bozeman, I agree that the Library shall make it available to borrowers under rules of the Library. I further agree that copying of this thesis is allowable only for scholarly purposes, consistent with "fair use" as prescribed in the U.S. Copyright Law. Requests for extensive copying or reproduction of this thesis should be referred to University Microfilms International, 300 North Zeeb Road, Ann Arbor, Michigan 48106, to whom I have granted "the exclusive right to reproduce and distribute my dissertation in and from microform along with the non-exclusive right to reproduce and distribute my abstract in any format in whole or in part." Signature _ Date f ^ iv ACKNOWLEDGEMENTS I sincerely thank Dr. Dave Ward for his enthusiasm and wellspring of ideas, and imparting to me an appreciation for the kind of science that "makes people stand up and take notice." I am also grateful for the critical discussions and encouragement I received from the members of my graduate committee; Drs. Cliff Bond, Martin Teintze, Gill Geesey, and Keith Cooksey. I would also like to express thanks to my co-workers Mike Ferris, Niels Ramsing, Michael Friederich, Mary Bateson, Sjila Santegoeds,. and Niels-Peter Revsbech.. Daily contact with these scientists improved both the quality of my life and the quality of my science. Garth James, Joe Sears, Marcia Riesselman, and Dan Siemsen generously shared their time and expertise, and for their efforts are gratefully acknowledged. I reserve special appreciation and grateful thanks for the love and support given to me by my wife, Susan Lindahl. This work was supported by a grant from the U.S. National Aeronautics and Space Administration (NAGW-2764). V TABLE OF CONTENTS Chapter Page 1. MOLECULAR ANALYSIS OF HOT SPRING MICROBIAL MATS: A GENERAL IN TRO D U CTIO N ......................................................................I Cultivation to Associate Microbial Activity with Microbial Diversity. . 3 rRNA Synthesis to Monitor Activity of Microbial Populations . . . 9 Hypotheses . 15 References Cited................................................................. ... . . . . 16 2. DIVERSE THERMUS SPECIES INHABIT A SINGLE HOT SPRING MICROBIAL MAT ............................................................................................20 Introduction.............................................................................. ' . . . 20 Materials and Methods ................................................................................21 Cultivation of Thermus Iso lates............................................................. 21 Characterization of 16S RNA S e q u e n c e s ........................................... 22 Results and D iscussion............................................................................... 23 References Cited........................................................................................... 30 3. CULTIVATION OF AEROBIC CHEMOORGANOTROPHIC PROTEOBACTERIA AND GRAM POSITIVE BACTERIA FROM A HOT SPRING MICROBIAL M A T .................................................. 32 Introduction..................................................................................................32 Materials and M e th o d s............................................................................... 34 Cultivation of Isolates . 34 Characterization of 16S RNA S e q u e n c e s ............................................35 Results........................... 37 D isc u ss io n ..................................................................................................44 References Cited........................................................................ .5 0 4 * 4. PHOTOSYNTHATE PARTITIONING AND FERMENTATION IN HOT SPRING MICROBIAL MAT COMMUNITIES......................................53 Introduction..................................................................................................53 Materials and M e th o d s............................................................................... 56 Mat Samples and C u ltu re s ................................... 56 Radiolabeling.................................... ' ................................................... 57 Nucleic Acid A n a ly s is ..........................................................................59 Polysaccharide Id e n tif ic a tio n ............................................................. 59 Protein A na ly sis ..................................................................................... 61 Lipid A n a l y s i s ............................................... 62 Photosynthate P a rtitio n in g ................................................................... 63 Headspace Gas Analysis.............................. 63 Volatile Fatty Acid D e te c t io n ............................................................. 64 TABLE OF CONTENTS (Continued) Chapter Page R adioassays......................................................-.................................. 65 Results........................................ 65 Nucleic Acid Synthesis . . 67 Radiolabeling with 6 * * * * * * * 14CO2 . . . 67 Radiolabeling with 32PO42" ............................................................. 67 Effect of Environmental Manipulations on rRNA Synthesis . .'70 Identification of Radiolabeled Material in the Nucleic Acid E x t r a c t ................................................................... 70 Protein Synthesis..................................................................................... 72 Lipid S y n th e s is ..................................................................................... 72 14CO2 Partitioning into Cellular C o m p o n en ts ..................................... 76 Polysaccharide Fermentation................................................................... 76 [14CjAcetate Partitioning into Cellular C o m p o n e n ts ..........................79 D isc u ss io n ....................................................................... 80 References Cited............................................................................................86 5. PHYSIOLOGICAL SUCCESSION AFTER DISTURBANCE OF A HOT SPRING CYANOBACTERIAL M AT.................................................. 91 Introduction............................................... 91 Materials and M e th o d s............................. 93 Results'. . ; ......................................................................................... .9 4 D isc u ss io n ....................................................................... 96 References Cited..................................................... 100 6. MOLECULAR ANALYSIS OF HOT SPRING MICROBIAL MATS TO STUDY BACTERIAL DIVERSITY AND PHYSIOLOGY: A S U M M A R Y ................................................................................................ 101 Aerobic Chemoorganotrophic Bacterial Diversity . . . . . . 101 Fate of Photosynthetically Fixed C a rb o n ................................................ 109 Validity of H ypotheses..................................................... ...... . . 112 Hypothesis I ..........................................................................................112 Hypothesis II ............................................................ 113 Hypothesis I I I ............................. 113 Hypothesis IV............................................................ 114 References C ited............................................................ . . . . 116 vi vii LIST OF TABLES Table Page I. Bacteria Known to Inhabit Octopus Spring Which Have or Could Have Aerobic Chemoorganotrophic Metabolisms Based on Cultivation and Molecular S t u d i e s ........................................................................ . 5- 2. Thermus Strains Cultivated from the 50-55°C Octopus Spring Cyanobacterial Mat C o m m u n ity ...................................................... . 24 3. 16S rRNA Sequence Similarities of Selected Thermus Species . 25 4. Bacterial Isolates Cultivated from the 50 to 55°C Octopus Spring Cyanobacterial Mat Community Sampled During October and November 1992 .................................................................................... . 38 5. 16S rRNA Sequence Similarities of Octopus Spring Isolates and Selected Proteobacterial and Gram Positive Sequence Types . . 41 6. 14CO2 Uptake in Various Hot Spring Cyanobacterial Mat Samples and > Logarithmically Growing Synechococcus Isolate Cl Cells . 66 7. 14CO2 Partitioning Among Molecular Fractions in Logarithmically Growing Synechococcus Isolate Cl Cultures and Hot Spring Cyanobacterial Mat Samples Incubated under Light and Dark Conditions . 77 8. [14CjAcetate Partitioning among Molecular Fractions in Octopus Spring and Clearwater Springs Site D Cyanobacterial Mat Samples Incubated under Light and Dark C o n d itio n s ...................................................... . 80 9. Bacteria Known to Inhabit Octopus Spring Which Have .or Could Have Aerobic Chemoorganotrophic Metabolisms Based on Cultivation and Molecular Studies......................................... .......................................... 103 viii LIST OF FIGURES 1. Distance Matrix Phylogenetic Tree of the Thermus and Deinococcus Lines of Descent Inferred from Full 16S rRNA Sequence Data . . .2 6 2. Effect of Temperature on Growth Rates of Octopus Spring Thermus I s o l a te s ................................................ .... 28 Figure Page 3. Distance Matrix Phylogenetic Tree Showing the Placement of 16S rRNA Sequences of Aerobic Chemoorganotrophic Isolates Cultivated from the Octopus Spring Mat Community Relative to those of Representatives of the Major Bacterial Lines of Descent . .' . . .4 0 4. . Distance Matrix Phylogenetic Tree Showing the Placement of 16S rRNA Sequences of Cultivated and Cloned Proteobacterial Octopus Spring Mat Cyanobacteria! Mat Populations Relative to those of Representatives of the Major Proteobacterial Lines of Descent (a, |3, y, 5, e ) ........................................................................ ...... 42 5. Composition of Nucleic Acid Extracts from a Logarithmically Growing Synechococcus Culture (Cl) and Octopus Spring Cyanobacterial Mat Community (Mat) after Radiolabeling with 14CO2 in the Light . . 68 6. Autoradiogram of Polyacrylamide Gel Containing Nucleic Acid Extract from Octopus Spring Cyanobacterial Mat Community after Radiolabeling with 32PO42" in the Light (32P Mat) and 14C-Labeled Synechococcus Isolate Cl Nucleic Acid Extract (14C Cl) . . . .6 9 7. (A), Hydrolyzed and TMS-derivitized Glycogen (top panel) and Octopus Spring Mat Nucleic Acid Extract (bottom panel) Samples Analyzed by Gas Chromatography. (B), Mass Spectra of Peaks with Retention Times of 838 and 833 Seconds iii Glycogen and Mat Samples, Respectively..................................................................................................71 8. Autoradiogram of Polyacrylamide Gel Containing Enzymatically Treated Octopus Spring Mat Nucleic Acid Extract After Radiolabeling with 14CO2 in the L i g h t ........................ ' ...................................................... 73' ix 9. Figure 10. 1 1 . 12. 13. LIST OF FIGURES (Continued) Composition of Proteins Solubilized from Whole-Cell Extracts of a y Logarithmically Growing Synechococcus Isolate Cl Culture (Cl) and Octopus Spring Cyanobacterial Mat Community (Mat) after Page Radiolabeling with 14CO2 in the L ig h t ..................................................... 74 Composition of Lipids Extracted from a Logarithmically Growing Synechococcus. Isolate Cl Culture (Cl) and Octopus Spring Cyanobacterial Mat Community (Mat) after Radiolabeling with 14CO2 in the L ig h t ............................................................ '. . . 75 (A), Changes in 14C Detected in Polysaccharide and Protein Cellular Fractions and Acetate, CO2, and Propionate Fermentation Products in Clearwater Springs Site D Mat Cores Shifted from a 3 Hour Light Incubation in the Presence of 14CO2 to a O,.3, 6, 12, and 24 Hour Dark Anaerobic Incubation. (B), with Bromoethanesulfonic Acid Added to Inhibit Methanogenesis; (C), Formalin Killed Control. . ..............................78 Carbon Incorporation (top panel) and Carbon Incorporation into Polysaccharide, Protein, Lipid, and Low Molecular Weight Metabolite Cellular Fractions (bottom panel). (A), Undisturbed Mat; (B), Scraped Mat 0, 5, 12, and 21 Days After D is tu rb a n c e ........................................................... ...... 95 Conceptual Model of Carbon Flux through Primary Producers (Synechococcus spp.) to Heterotrophs (Chloroflexus spp.) in Hot Spring Cyanobacterial Mat C om m unities.............................. HO ABSTRACT Molecular studies investigating 16S rRNA sequence diversity in cyanobacterial mat communities inhabiting hot springs in Yellowstone National Park have revealed that these communities contain numerous uncultivated microbial species. Here, attempts were made to cultivate from one of these mat communities the aerobic chemoorganotrophic bacteria whose 16S rRNA nucleotide sequences were previously observed using molecular retrieval techniques. By utilizing serial dilution enrichment culture and a variety of enrichment conditions, a diversity of bacterial isolates was obtained. 16S rRNA sequence analysis revealed seven genotypically distinct isolates, including Thermus, proteobacterial, and Gram positive representatives. However, only one of these isolates, a (3-proteobacterium, contained a 16S rRNA sequence previously observed in Octopus Spring mat. These results illustrate the differing views of microbial community composition which cultivation and molecular techniques provide, and demonstrate the problems encountered when using cultivation approaches to associate microbial activity with bacterial populations whose 16S rRNA sequences were detected in natural samples. One cultivation-independent approach to associate bacterial activity with retrieved 16S rRNA sequence types would be to selectively capture rRNA molecules synthesized by actively growing microorganisms incubated in the presence of a radiolabeled substrate, then quantify the relative extent of radiolabel incorporation into specific 16S rRNA molecules. Initial studies investigating the feasibility of this approach revealed that although logarithmically growing cyanobacterial cells incorporated photosynthetically fixed 14CO2 into rRNA, cyanobacteria inhabiting hot spring mats predominately incorporated 14CO2 into polyglucose during periods of illumination (between 77% and 85% of total incorporated carbon). Although photosynthetically active, the cyanobacteria of these mat communities do not appear to be rapidly growing, since only limited synthesis of growth-related macromolecules was detected. The fate of polyglucose reserves was investigated by allowing mat cyanobacteria to photoassimilate 14CO2 into polyglucose, then transferring samples to the dark, anaerobic conditions which mat communities experience at night. Radiolabel in the polysaccharide fraction decreased 74.7% after 12 hours dark incubation, of which 58.5% was recovered in radiolabeled fermentation products (i.e. [14CJacetate, 14CO2, and [14CJpropionate). These results indicate tightly coupled carbon fixation and fermentative processes, and the potential for significant carbon transfer from primary producers to heterotrophic members of these cyanobacterial mat communities. I CHAPTER I MOLECULAR ANALYSIS OF HOT SPRING MICROBIAL MATS: A GENERAL INTRODUCTION Studies employing the techniques of molecular biology have revolutionized our understanding of the microbial world. The realization that the information contained in nucleic acid and protein sequences can be used to reconstruct molecular evolutionary history (55) has led to extensive comparative studies of molecular evolution. One molecule in particular, the small subunit ribosomal RN A, has been central to our understanding of evolutionary relationships among microbial species (51). Comparison of small subunit ribosomal RNA nucleotide sequences has allowed construction of a universal phytogeny based on genetic relationships among organisms without reliance on phenotypic traits (31,52). This universal "tree of life" divides life on this planet into three distinct primary groups, the domains Bacteria, Archaea, and Eukarya (51,53). Two of these domains, the Bacteria and the Archaea, are microbial, and the differences separating these domains are more significant than those which distinguish the traditional kingdoms (i.e. plants and animals) from one another (53). Molecular studies have also revealed that natural habitats harbor a great diversity of undescribed microbial species. Studies of DNA-DNA reassociation of nucleic acids directly extracted from soils showed that DNA from this habitat is extremely heterogeneous (41). The authors of this study deduced that soil DNA 2 exhibits complexity comparable to ca. 4000 completely different bacterial genomes, most of which are from uncultivated microbial community members. Studies investigating the genotypic diversity of small subunit ribosomal RNA (more specifically, 16S rRNA) sequence types retrieved from natural microbial communities lead to a similar conclusion:, that microbial diversity is both very great and easily surpasses the diversity of validlly described microbial species cultivated from natural habitats (1,14,16,43,46). The diversity of uncultivated community members is both scientifically interesting and potentially economically valuable, since microbial physiological diversity has been a source of new natural products used in pharmaceutical, biotechnology, and industrial microbiology applications. However, the physiological diversity of uncultivated community members is largely unknown, since few pure cultures of these species have been obtained for phenotypic characterization. Molecular analyses have also aided our understanding of the physiological activity exhibited by microorganisms as they occur in nature. Studies investigating elemental composition, patterns of photosynthetically fixed carbon allocation among molecular classes, and synthesis of individual macromolecules have shown that phytoplankton communities are both photosynthetically active and exhibit growth at or near maximal rates (18,19,22,25,26). Not all microbial communities are rapidly growing, however. Novitsky (30) observed a high ATP content combined with a low rate of nucleic acid synthesis in a marine sediment microbial community, indicating that the microbial cells were active, but not rapidly dividing. In other studies, the effects of disturbance and environmental perturbation on molecular synthesis have been investigated by analyzing phospholipid and sterol biomarkers in marine sediment 3 communities (12,13,24). These studies illustrate the utility of molecular analyses to investigate microbial response to environmental change. These advances in our knowledge of microbial evolution, diversity, and activity have led to the questions addressed by this thesis. I have chosen the laminated cyanobacterial mat communities inhabiting mildly alkaline silicious hot springs in Yellowstone National Park as units of study. The microbial mat community inhabiting one hot spring in particular, Octopus Spring, has been intensively studied in an effort to make fundamental observations of microbial species composition and physiological activity (5,43-45). This existing information base allows the construction of testable hypotheses concerning the activities which occur in microbial communities and the identity of the microorganisms which perform those activities. The theme which unifies the investigations presented in this thesis is the goal of associating microbial activity with microbial diversity. More specifically, the unifying goal of this thesis is to associate microbial activities which occur in hot spring cyanobacterial mat communities with microbial populations which share identical 16S rRNA sequences. The remainder of this chapter introduces the experiments which were performed to accomplish this goal, and highlights the hypotheses which were tested by those experiments. Cultivation to Associate Microbial Activity with Microbial Diversity Selective enrichment culture techniques have been used for more than a century to obtain naturally occurring microorganisms for study in pure culture. The 4 microbiologists who originally developed these techniques cautioned that the methods may select for organisms which are best adapted to the enrichment culture environment, but which may not be the dominant organisms in nature (42,50). Comparison of the 16S rRNA sequences of organisms cultivated from Octopus Spring mat to 16S rRNA sequences detected in the mat using molecular retrieval techniques reveals that enrichment culture does indeed fail to cultivate the dominant microorganisms whose rRNAs are detected in natural microbial communities. To illustrate this point, bacteria which were cultivated from Octopus Spring mat and which exhibit an aerobic chemoorganotrophic type of metabolism appear in Table I, column I. These five species include representatives from the green non-sulfur bacteria, Thermus, and planctomyces lines of descent. Molecular retrieval approaches have revealed many 16S rRNA sequence types representing microorganisms which might exhibit aerobic chemoorganotrophic metabolisms (Table I, column 2). Some of the 13 unique 16S rRNA sequence types retrieved from this community are related to . members of the green non-sulfur, green sulfur, and proteobacterial lines of descent, while others do not readily cluster into known phylogenetic groups. However, Table I clearly shows that the species detected by molecular retrieval and cultivation approaches are completely different, illustrating the conflicting views of microbial community composition provided by these two techniques. Similar observations are repeated in other physiologically related groups in the Octopus Spring mat (11,43,44,47,48), as well as other microbial communities (16,37,39). Clearly, our knowledge of microbial diversity based On organisms cultivated from natural environments is incomplete. 5 Table I. Bacteria known to inhabit Octopus Spring which have or could have aerobic chemoorganotrophic metabolisms based on cultivation and molecular studies3. 168 rRNA Sequence Previously Cultivated Retrieved Green non-sulfur bacteria Chloroflexus aurantiacus Y-400-fl Thermomicrobium roseum type C OS-V-L-20 Thermus/Deinococcus Group Thermus sp. OS-Ramaley-4 Thermus aquaticus YT-I Planctomyces Isosphaera pallida ISlB Proteobacteria Alpha subdivision type O Beta subdivision type G type N type R Green sulfur-like bacteria W type E type M os-m-9 Uncertain Affiliation type L type D type F OP-I-2 “Adapted from reference (44). : 6 Careful assessment of the enrichment culture techniques used by previous investigators to cultivate aerobic chemoorganotrophic bacteria from hot spring cyanobacterial mats may help to explain the failure of these methods to recover the numerically abundant microbial populations (i.e those microbial species whose 16S rRNA sequences were detected using molecular retrieval approaches). In general, isolates were obtained by directly streaking mat material onto solidified media (17,21,32) or by the direct addition of undiluted inoculum to enrichment flasks (6) containing relatively high levels of carbon substrates (0.1% to 3% tryptone and yeast extract) (6,21,32) and incubating at high temperatures (VO0C) regardless of the temperature of the collection site (6,21). There are several potential problems with these methods. Incubating cultures at temperatures which are different than the sample collection site may select against microorganisms adapted for optimal growth at collection site temperatures. Strain purification by picking isolated colonies requires growth on solidified media; if a bacterial strain is incapable of colonial growth, this species will not appear in culture collections. Directly plating mat inoculum onto solidified media precludes attempts to measure the relative abundance of the cultivated organisms, and directly adding undiluted inoculum to enrichment flasks may promote culture overgrowth by numerically insignificant species. Recently, researchers attempting to cultivate microorganisms from seawater have successfully obtained / isolates of oligotrophic ultramicrobacteria (i.e. bacteria smaller than 2 pm which are adapted to low organic carbon concentrations) using serial dilution enrichment culture techniques (7,38). These researchers inoculated a series of flasks containing unamended sterile seawater with a serially diluted inoculum source, initially resulting 7 in 10-fold fewer microorganisms in each enrichment flask. The isolates obtained from very high (106-fold) dilutions resembled the majority of the microorganisms in the original seawater. These experiments illustrate two important points. First, .. employment of serial dilution enrichment culture techniques may lead to successful cultivation of the more numerically abundant microorganisms from natural samples, and second, not all bacteria are adapted to copiotrophic conditions (i.e. high organic carbon concentrations). Thus, enrichments containing relatively high concentrations of organic carbon may select against microorganisms adapted to oligotrophic conditions. Finally, although microbial mat communities are characterized by extreme environmental gradients (33,34), conditions in culture media are remarkably homogenous. The environmental homogeneity which characterizes most culture media may limit the diversity of cultivated species by failing to provide a range of conditions from which the numerically abundant microorganisms may select for growth (8,54), The first research objective of this thesis was to attempt to overcome some of the existing problems with enrichment culture techniques in order to cultivate the numerically abundant aerobic chemoorganotrOphic bacteria from Octopus Spring mat (see hypothesis I, page 15). Serial dilution enrichment culture was used to provide a measure of the relative abundance of the isolates obtained, and to avoid culture overgrowth by numerically insignificant species (11). Enrichments were also performed under more natural conditions (e.g. incubating at the temperature of sample collection, and using more relevant carbon substrates and concentrations). By successfully cultivating the microorganisms whose 16S rRNA sequences were retrieved from Octopus Spring mat, and associating those sequence types with the 8 aerobic chemoorganotrophic metabolism, the major goal of this thesis would be addressed. Namely, a microbial activity (aerobic chemoorganotrophy) would be associated with microbial populations whose 16S rRNA sequence types are detected in natural mat samples. During these investigations, a diversity of bacterial isolates was identified and characterized. Analysis of 16S rRNA sequence types revealed seven genotypically distinct isolates, including representatives belonging to the Thermus, proteobacterial, and Gram positive lines of descent. Characterization of the relative abundance and growth characteristics of the cultivated Thermus isolates led to the conclusion that Thermus distribution may be controlled by specialization to temperature, a condition which varies in hot spring habitats (Chapter 2). Phenotypically distinct Gram positive isolates exhibited identical 16S rRNA nucleotide sequence through a variable region of the molecule, indicating the conserved nature of bacterial diversity estimates based on 16S rRNA sequence information (Chapter 3). However, only one of the seven isolates, a (3-proteobacterium, contained a 16S rRNA sequence identical to a sequence type previously detected in Octopus Spring mat using molecular retrieval techniques (Chapter 3). By combining these results with a closely related study of species composition in similar enrichment cultures before strain purification (36), it becomes possible to speculate about the causes of the failure of enrichment culture techniques to recover the numerically abundant microorganisms from natural microbial communities (Chapter 6). Clearly, there exists a need to cultivate the microorganisms which actually occur in natural habitats. However, even if abundant populations were successfully 9 cultivated, it would be difficult to extrapolate the phenotypes observed' in pure culture to the activities , the cultivated microorganisms exhibit in nature. For example, photoheterotrophic carbon uptake (the light-induced incorporation of organic carbon) by the filamentous Chloroflexus-Iike hot spring cyanobacterial mat inhabitants is considered to be an important mechanism of carbon transfer from the primary producers {Synechococcus spp.) to the primary consumers (the Chloroflexus-Iike cells) in these communities (2,35). However, cultivation conditions may be sufficiently different that an ecologically important phenotype such as photoheterotrophy may not be expressed in pure culture. When Chloroflexus cells were first cultivated under dark, aerobic conditions, the photoheterotrophic metabolic capability of this bacterium was not fully appreciated (4,5). Likewise, demonstration of a phenotype in pure culture does not necessarily demonstrate an ecologically important activity. There appears a need for cultivation-independent methods to assess the activity exhibited by genetically related microbial populations. Ideally, such a method would accurately measure microbial activity in situ, and successfully associate microbial activity with a 16S rRNA sequence type. One possible solution to this challenge may come from the techniques of molecular biology. rRNA Synthesis to Monitor Activity of Microbial Populations Ribosomal RNA exhibits several properties that make it uniquely suited as a species-specific indicator of bacterial activity. The nucleotide sequence of rRNA contains conserved and variable regions, allowing design of oligodeoxynucleotide 10 hybridization probes specific to phylogenetic groups and individual species (23). Synthesis of ribosomes and ribosomal RNA is also proportional to growth rate. Studies using Escherichia coli as a model organism showed remarkable growth rate- dependent regulation of ribosomal component synthesis (15,29). Using fluorescently labelled rRNA-targeted oligodeoxynucleotide probes, DeLong et al. (9) showed a similar result; as growth rate of E. coli increased, probe response increased proportionally. I hypothesized that by providing a microbial community with a specific radiolabeled substrate, only those populations which are actively growing and utilizing that substrate would incorporate radiolabel into rRNA. By assaying the relative amounts of radiolabel incorporation into group- or species-specific rRNA sequence types, it should be possible to address the major goal of this thesis: to monitor microbial activity (by measuring radiolabel incorporation into rRNA) and associate that activity with microbial populations whose 16S rRNA sequence types are detected in natural mat samples. Obtaining radiolabeled rRNA should not be difficult if microbial populations are growing. However, separating specific 16S rRNA sequence types from a mixture containing diverse rRNA sequence types (such as would be found in a microbial community) requires specialized molecular techniques. Selectively retrieving sequence-specific rRNA types from a mixture of community rRNAs has been accomplished using oligodeoxynucleotide hybridization- based capture probe technology (20,27,28,40). This method requires the specific hybridization of a biotinylated oligodeoxynucleotide probe to a target 16S rRNA nucleotide sequence. The probe/target rRNA hybrid is then "captured" onto a magnetic bead particle by high affinity binding which occurs between the biotin 11 moiety on the probe and strepavidin molecules which are covalently attached to the magnetic bead. Separation of target rRNA from non-target community rRNAs is accomplished by washing the uncaptured rRNA from the magnetic bead/capture probe/target RNA complex. Thus, capture probes can be used to separate individual 16S rRNA sequence types to assay the extent of radiolabel incorporation into specific rRNAs. Successful capture probe retrieval of radiolabeled rRNA sequence types requires that rRNA is synthesized in detectable quantities. Before investing the time to develop capture probe methods, the extent of radiolabel incorporation into rRNA in Octopus Spring mat was investigated. Since rates of cyanobaeterial oxygenic photosynthesis are very high (34), and previous studies have suggested that cyanobaeterial populations are growing (5,10) (see below), photoautotrophic carbon incorporation by Synechococcus species was chosen as a test case to maximize the amount of radiolabel incorporation into rRNA. Therefore, 14CO2 was provided as a substrate to predominately radiolabel the photosynthetically active Synechococcus cyanobaeterial species. Although 14CO2 was readily incorporated into cellular material, no incorporation of 14CO2 into rRNA could be detected in Octopus Spring mat (Chapter 4). This observation was somewhat unexpected, given what is known about the growth, activity, and interactions of hot spring cyanobaeterial mat inhabitants. Hot spring cyanobaeterial mats are active and dynamic microbial communities. Cyanobaeterial gross primary productivity calculated from measured oxygen production rates is 20 to 40 mmol CO2 m"2 hr"1 (34). For comparison, these values are greater than the carbon incorporated by the entire photic zone of the world’s most 12 productive lakes (calculated from data provided in Table 15-9 of reference (49)). Other experiments indicate that hot spring cyanobacterial mats are growing. By measuring the decrease of Synechococcus cells over time after darkening a mat. Brock (5) estimated cyanobacterial productivity to be 5.7 x IO11 to 1.6 x IO12 cells m"2 day"1. Accretion rates of mat material above a silicon carbide layer sifted onto the mat surface provide further evidence for mat growth. Long-term (I year) measurements of organic material accretion above silicon carbide layers indicated growth rates of 18 to 45 pm day"1 (10). There is also evidence for carbon transfer between primary producers and primary consumers in hot spring mat communities. Photosynthetically active Synechococcus species photoexcrete glycolic acid (up to 7% of total photosynthate) which is readily incorporated by the Chloroflexus-Iike primary consumers in the mats (3). Under dark, anaerobic conditions, fermentation products are produced at a rate of ca. 10 mmol' acetate and propionate m"2 hr'1 (calculated from (2)). Acetate and propionate accumulate in mats overnight and are photohetero- trophically incorporated by Chloroflexus-Iike cells the following day (2,35). Terminal anaerobic processes occurring in the mats include methanogenesis (methane production rates are ca. 0.8 mmol CH4 m"2 hr"1, (2,14)) and possibly acetogenesis, the anaerobic conversion of hydrogen and carbon dioxide to acetate (2,3). Clearly, cyanobacterial primary production fuels active carbon transfer within microbial mat communities. Mat communities also appear to be growing, which is why the observation that 14CO2 is not readily incorporated into rRNA was unexpected. The second research objective of this thesis was to investigate the fate of photosynthetically fixed carbon in hot spring cyanobacterial mat communities (see 13 hypotheses II to IV, page 15). This research objective represents a significant divergence from the major goal of this thesis. However, the observation that rRNA was not synthesized by photosynthetically active mat cyanobacteria leads to the possibility that these populations may not be rapidly growing (hypothesis II). The physiological ecology of a microbial community which is very active, but may not be rapidly growing was sufficiently unique to justify this change in research direction. It was hoped that investigations into the fate of photosynthetically fixed carbon would change our perceptions about the activity of cyanobacterial mat inhabitants. Further, it was hoped that conditions under which rRNA synthesis did occur would be discovered, thus permitting use of the capture probe approach. These investigations revealed that mat cyanobacteria do not allocate significant amounts of photosynthetically fixed carbon into growth-related molecules such as rRNA, protein, and lipid. Instead, photosynthate is stored in the form of polyglucose during periods of illumination (Chapter 4). These results lead to the conclusion that although mat cyanobacteria are photosynthetically active, they do not appear to be rapidly growing. Mat cyanobacteria were shown to ferment polyglucose reserves under the dark, anaerobic conditions which mats experience at night (Chapter 4). Although unanticipated, this finding helped to achieve the unifying goal of this thesis by associating fermentative activities with the genotypically related Synechococcus cyanobacterial populations. Cyanobacterial fermentation has been demonstrated in pure cultures, but has not until now been demonstrated in situ. These results indicate the potential for massive carbon transfer between the Synechococcus primary producers and the Chloroflexus-Iike primary consumers in these mat communities. 14 There remain significant questions regarding the basis for this symbiotic relationship (Chapter 6). Although mat communities normally synthesize only limited amounts of rRNA, disturbance to the mat community resulted in patterns of molecular synthesis which more closely resembled logarithmically growing cyanobacterial cells (Chapter 5). These results indicate the potential for successful application of capture probe technologies when mat communities are undergoing post-disturbance recolonization. The experiments described in chapters 4 and 5 have changed our perception of the physiological ecology of hot spring cyanobacterial mat inhabitants, and have allowed the construction of a conceptual model which describes the fate of photosynthetically fixed carbon in these mat communities (Chapter 6). 15 Hypotheses The introductory material presented in this chapter allows the construction of testable hypotheses concerning the activity of hot spring cyanobacteria! mat inhabitants and the identity of the microorganisms which perform those activities. Hypothesis I was tested by experiments described in Chapters 2 and 3, hypotheses II and HI were tested by experiments described in Chapter 4, and hypothesis IV was tested by experiments described in Chapter 5. I. Application of cultivation techniques which employ more rational enrichment culture approaches will lead to the cultivation of aerobic chemoorganotrophic ■ bacteria whose 16S rRNA sequences have been previously observed in Octopus Spring mat. II. Since cyanobacterial cells inhabiting hot spring mat communities are photosynthetically active and appear to be rapidly growing, it should be possible to detect synthesis of growth-related molecules such as ribosomal RNA and protein. III. Cyanobacteria ferment photoautotrophically fixed polyglucose under the dark, anaerobic conditions hot spring cyanobacterial mats experience at night. IV. IV. Cyanobacterial biomass lost from mat communities during disturbance events is replaced by growth and cellular division. 16 References Cited 1. Amann, R., W. Ludwig, and K. H. Schleifer. 1995. Phylogenetic identification and in situ detection of individual microbial cells without cultivation. Microbiol. Rev. 59:143-169. 2. Anderson, K. L., T. A. Tayne, and D. M. Ward. 1987. Formation and fate of fermentation products in hot spring cyanobacterial mats. Appl. Environ. Microbiol. 53:2343-2352. 3. Bateson, M. M. and D. M. Ward. 1988. Photoexcretion and fate of glycolate in a hot spring cyanobacterial mat. Appl. Environ. Microbiol. 54:1738-1743. 4. Brock, T. D. 1969. Vertical zonation in hot spring algal mats. Phycologia 8:201-205. 5. Brock, T. D. 1978. Thermophilic Microorganisms and Life at High Temperatures, Springer-Verlag, New York. 6. Brock, T. D. and H. Freeze. 1969. Thermus aquaticus gen. n. and sp. n., a non-sporulating extreme thermophile. J. Bacteriol. 98:289-297. 7. Button, D. K., F. Schut, P. Quang, R. Martin, and B. R. Robertson. 1993. Viability and isolation of marine bacteria by dilution culture: theory, procedures, and initial results. Appl. Environ. Microbiol. 59:881-891. 8. Caldwell, D. E. and P. Hirsch. 1973. Growth of microorganisms in two-dimensional steady-state diffusion gradients. Can. J. Microbiol. 19:53-58. 9. DeLong, E. F., G. S. Wickham, and N. R. Pace. 1989. Phylogenetic stains: ribosomal RNA'-based probes for the identification of single cells. Science 243:1360-1363. 10. Doemel, W. N. and T. D. Brock. 1977. Structure, growth, and decomposition of laminated algal-bacterial mat's in alkaline hot springs. Appl. Environ. Microbiol. 34:433-452. 11. Ferris, M. J., A. L. Ruff-Roberts, E. D. Kopczynski, M. M. Bateson, and D. M. Ward. 1996. Enrichment culture and microscopy conceal diverse thermophilic Synechococcus populations in a single hot spring mat habitat. Appl. Environ. Microbiol. 62:1045-1050. 12. Findlay, R. H., M. B. Trexler, and D. C. White. 1990. Response of a benthic microbial community to biotic disturbance. Mar. Ecol. Prog. Ser. 62:135-148. 13. Findlay, R. H., L. Watling, and L. M. Mayer. 1995. Environmental impact of salmon net-pen culture on marine benthic communities in Maine: a case study. Estuaries 18:145-179. 14. Fuhrman, J. A., K. McCallum, and A. A. Davis. 1993. Phylogenetic diversity of subsurface marine microbial communities from the Atlantic and Pacific Oceans. Appl. Environ. Microbiol. 59:1294-1302. 15. Causing, K. 1977. Regulation of ribosome production in Escherichia- coli: synthesis and stability of ribosomal KNA and of ribosomal protein messenger RNA at different growth rates. J. Mol. Biol. 115:335-354. 16. Giovannoni, S. J., T. B. Britschgi, C. L. Moyer, and K. G. Field. 1990. Genetic diversity in 17 Sargasso Sea bacterioplankton. Nature 344:60-63. 17. Giovannoni, S. J., E. Schabtach, and R. W. Castenholz. 1987. Isosphaera pallida, gen. and comb, nov., a gliding, budding eubacterium from hot springs. Arch. Microbiol. 147:276-284. 18. Glover, H. E. and A. E. Smith. 1988. Diel patterns of carbon incorporation into biochemical constituents of Synechococcus spp. and larger algae in the Northwest Atlantic Ocean. Mar. Biol. 97:259-267. 19. Goldman, J. C., J. J. McCarthy, and D. G. Peavey. 1979. Growth rate influence on the chemical composition of phytoplankton in oceanic waters. Nature 279:210-215. 20. Hunsaker, W. R., H. Badri, M. Lombardo, and M. L. Collins. 1989. Nucleic acid hybridization assays employing dA-tailed capture probes. Anal. Biochem. 181:360-370. 21. Jackson, J. J., R. F. Ramaley, and W. G. Meinschein. 1973. Thermomicrobium, a new genus of extremely thermophilic bacteria. Int. J. Sys. Bacteriol. 23:28-36. 22. Kramer, J. G. and F. L. Singleton. 1993. Measurement of rRNA variations in natural communities of microorganisms in the Southeastern U.S. continental shelf. Appl. Environ. Microbiol. 59:2430-2436. 23. Manz, W., R. Amann, W. Ludwig, M. Wagner, and K. H. Schleifer. 1992. Phylogenetic oligodeoxynucleotide probes for the major subclasses of proteobacteria: problems and solutions. System. Appl. Microbiol. 15:593-600. 24. Mayer, L. M., D. F. Schick, R. H. Findlay, and D. L. Rice. 1991. Effects of commercial dragging on sedimentary organic matter. Marine Environmental Research 31:249-261. 25. Morris, I. 1981. Photosynthetic products, physiological state, and phytoplankton growth. Can. Bull. Fish. Aquat. Sciences 210:83-102. 26. Morris, I., A. E. Smith, and H. E. Glover. 1981. Products of photosynthesis in phytoplankton off the Orinoco River and in the Caribbean Sea. Limnol. Oceanogr. 26:1034-1044. 27. Morrissey, D. V. and M. L. Collins. 1989. Nucleic acid hybridization assays employing dA-tailed capture probes. Single capture methods. Mol. Cell. Probes 3:189-207. 28. Morrissey, D. V., M. Lombardo, J. K. Eldridge, K. R. Kearney, E. P. Groody, and M. L. Collins. 1989. Nucleic acid hybridization assays employing dA-tailed capture probes. Anal. Biochem. 181:345-359. 29. Nomura, M., R. Course, and G. Baughman. 1984. Regulation of the synthesis of ribosomes and ribosomal components. Ann. Rev. Biochem. 53:75-117. 30. Novitsky, J. A. 1987. Microbial growth rates and biomass production in a marine sediment: evidence for a very active but mostly nongrowing community. Appl. Environ. Microbiol. 53:2368-2372. 31. Olsen, G. J. and C. R. Woese. 1993. Ribosomal RNA: a key to phytogeny. The FASEB Journal 7:113-123. 32. Pierson, B. K. and R. W. Castenholz. 1974. A phototrophic gliding filamentous bacterium of hot springs, Chloroflexus aurantiacus, gen. and sp. nov. Arch. Microbiol. 100:5-24. 18 33. Revsbech, N. P. and B. B. Jorgensen. 1986. Microelectrodes: Their use in microbial ecology, p. 293-352. In K.C. Marshall (ed.), Advances in Microbial Ecology, 9th ed. Plenum Press, New York. 34. Revsbech, N. P. and D. M. Ward. 1984. Microelectrode studies of interstitial water chemistry and photosynthetic activity in a hot spring microbial mat. Appl. Environ. Microbiol. 48:270-275. 35. Sandbeck, K. A. and D. M. Ward. 1981. Fate of immediate methane precursors in low-sulfate, hot-spring algal-bacterial mats. Appl. Environ. Microbiol. 41:775-782. 36. Santegoeds, C. M., S. C. Mold, and D. M. Ward. 1996. Denaturing gradient gel electrophoresis used to monitor the enrichment culture of aerobic chemoorganotrophic bacteria from a hot spring cyanobacterial mat. Submitted to Appl. Environ. Microbiol. 37. Schmidt, T. M., E. F. DeLong, and N. R. Pace. 1991. Analysis of a marine picoplankton community by 16S rRNA gene cloning and sequencing. J. Bacteriol. 173:4371-4378. 38. Schut, F., E. J. de Vries, J. C. Gottschal, B. R. Robertson, W. Harder, R. A. Prins, and D. K. Button. 1993. Isolation of typical marine bacteria by dilution culture: growth, maintenance, and characteristics of isolates under laboratory conditions. Appl. Environ. Microbiol. 59:2150-2160. 39. Stackebrandt, E., W. Liesack, and B. M. Goebel. 1993. Bacterial diversity in a soil sample from a subtropical Australian environment as determined by 16S rDNA analysis. The FASEB Journal 7:232-236. 40. Thompson, J., R. Solomon, M. Pellegrino, K. Sakai, M. Lewin, M. Feild, M. Castrovinci, L. Sacramone, and D. Gillespie. 1989. A noise-free molecular hybridization procedure for measuring RNA in cell lysates. Anal. Biochem. 181:371-378. 41. Torsvik, V., J. Goksoyr, and F. L. Daae. 1990. High diversity in DNA of soil bacteria. Appl. Environ. Microbiol. 56:782-787. 42. van Niel, C. B. 1949. The "Delft school" and the rise of general microbiology. Bacteriol. Rev. 13:161-174. 43. Ward, D. M., M. M. Bateson, R. Weller, and A. L. Ruff-Roberts. 1992. Ribosomal RNA analysis of microorganisms as they occur in nature, p. 219-286. In K.C. Marshall (ed.), Advances in Microbial Ecology, Plenum Press, New York. 44. Ward, D. M., M. J. Ferris, S. C. NoId, M. M. Bateson, E. D. Kopczynski, and A. L. Ruff-Roberts. 1994. Species diversity in hot spring microbial mats as revealed by both molecular and enrichment culture approaches-relationship between biodiversity and community structure, p. 33-44. In L.J. Stal and P. Caumette (ed.), Microbial Mats: Structure, Development and Environmental Significance, Springer-Verlag, Heidelberg. 45. Ward, D. M., T. A. Tayne, K. A. Anderson, and M. M. Bateson. 1987. Community structure and interactions among community members in hot spring cyanobacterial mats, p. 179-210. In M. Fletcher, T.R.G. Gray, and J.G. Jones (ed.), Ecology of Microbial Communities, Cambridge University Press, Cambridge. 46. Ward, D. M., R. Weller, and M. M. Bateson. 1990. 16S rRNA sequences reveal numerous uncultured microorganisms in a natural community. Nature 344:63-65. 19 47. Weller, R., M. M. Bateson, B. K. Heimbuch, E. D. Kopczynski, and D. M. Ward. 1992. Uncultivated cyanobacteria, Chloroflexus-like. inhabitants, and spirochete-like inhabitants of a hot spring microbial mat. Appl. Environ. Microbiol. 58:3964-2969. 48. Weller, R., J, W. Weller, and D. M. Ward. 1991. 16S rRNA sequences of uncultivated hot spring cyanobacterial mat inhabitants retrieved as randomly primed cDNA. Appl. Environ. Microbiol. 57:1146-1151. . 49. Wetzel, R. G. 1983. Limnology, Saunders College Publishing, New York. 50. Winogradsky, S. 1949. Microbiologie du sol, problemes et methodes, Barneoud Freres, France, 51. Woese, C. R. 1987. Bacterial evolution. Microbiol. Rev. 51:221-271. 52. Woese, C. R. 1994. There must be a prokaryote somewhere: Microbiology’s search for itself. Microbiol. Rev. 58:1-9. 53. Woese, C. R., O. Kandler, and M. L. WheeIis. 1990. Towards a natural system of organisms: proposal for the domains archaea, bacteria, and eucarya. Proc. Natl. Acad. Sci. 87:4576-4579. 54. Wolfaardt, G. M., J. R. Lawrence, M. J. Hendry, R. D. Robarts, and D. E. Caldwell. 1993. Development of steady-state diffusion gradients for the cultivation of degradative microbial consortia. Appl. Environ. Microbiol. 59:2388-2396. 55. Zuckerkandl, E. and L. Pauling. 1965. Molecules as documents of evolutionary history. J. Theoret. Biol. 8:357-366. 20 CHAPTER 2 DIVERSE THERMUS SPECIES INHABIT A SINGLE HOT SPRING MICROBIAL MAT1 Introduction Recent studies of the laminated cyanobacterial mat community located in Octopus Spring (Yellowstone National Park, Wyoming) have focussed on species diversity and evolution (24,26) and microbial community ecology (18,23). To further understand the diversity of aerobic chemoorganotrophic bacteria in the Octopus Spring mat we attempted to cultivate numerically abundant species using a variety of different strategies. Enrichment conditions were designed to favor recovery of species adapted - to substrates and temperatures which reflect prevailing environmental resources and conditions. Inoculum was serially diluted to extinction to provide an estimate of relative abundance of the species cultivated (21) and to eliminate rapid overgrowth by numerically insignificant species (5). In the course of this work we isolated and characterized several Thermus strains. Our results reveal a diversity of Thermus isolates within one hot spring microbial community whose distribution may be 1This study has'been published as: Nold, S. C. and D. M. Ward. 1995. Diverse Thermus Species Inhabit a Single Hot Spring Microbial Mat. System. Appl. Microbiol. 18:274-278. 21 controlled by specialization to environmental features such as temperature that vary in the habitat. Materials and Methods Cultivation of Thermus Isolates Samples were collected from the top I cm of the cyanobacterial mat community located in the shoulder region (50-55°C, pH 8.5) of Octopus Spring (2) on 30 September 1992 and 29 October 1992. Samples were kept between 46°C and 50°C for 3 hours in transit to the laboratory, then homogenized with a Dounce tissue homogenizer and serially diluted (1:10) in sterile enrichment medium (30 September 1992) or medium D (29 October 1992) (17) before inoculation. Enumeration of Synechococcus cells by direct microscopic count was performed on appropriate _ dilutions using a Petroff-Hausser counting chamber. A standard procedure to cultivate Thermus species (0.1% w/v tryptone, 0.1% w/v yeast extract (TYE) in Castenholz medium D, incubation temperature VO0C) (17) was compared to the use of the same basal medium with substrates that seemed more rational replacing TYE and incubating at lower temperatures (details provided in Table 2). Substrates included glycolate, a compound shown to be excreted by thermophilic cyanobacteria in Octopus Spring mat (I), and solidified autoclaved mat homogenate as carbon sources (Table 2). Liquid media (TYE, glycolate) were incubated with shaking (150 rpm) until turbid, then transferred to solid media containing 3% agar for isolation. Mat agar was spread inoculated with 100 pL homogenized and diluted sample and placed in a static 22 incubator at 50°C. Growth rates were measured at different temperatures in duplicate, using sidearm flasks containing 50 mL of medium optimized for each isolate (3.0% TYE, pH 8.2 for isolates ac-1 and ac-7; 1.0% TYE, pH 7.5 for isolates ac-2 and ac-17). Klett absorbance readings were taken during logarithmic growth, compared to a standard curve, and doublings per hour were calculated. Characterization of 16S rRNA Sequences Harvested cells were lysed according to an established enzymatic protocol (25). Nucleic acids were extracted and the 16S rDNA gene was amplified using the polymerase chain reaction, cloned, and sequenced as described by Kopczynski et al. (9). Full-length 16S rRNA sequences {Escherichia coli positions 28-1483) for isolates ac-1, ac-2, ac-7, and ac-17 (GenBank accession numbers L37520, L37521, L37522, and L37523, respectively) were aligned relative to existing Thermus sequences (see Saul et al. (20) for a full description of strains and GenBank accession numbers), Deinococcus radiodurans, and E. coli, using the sequence editor provided by the Ribosomal Database Project (RDP) (10). D. radiodurans and E. coli sequences were derived from the Ribosomal Database Project on the anonymous ftp server at the University of Illinois in Urbana, Illinois updated on August I, 1993. Similarity values were calculated and phylogenetic trees constructed from evolutionary distances using the evolutionary distance matrix algorithm as described by Olsen (14) and available through the RDP. 23 Results and Discussion The most numerically abundant Thermus isolates were ac-17 and ac-14, which survived dilution to IO"6 and IO"5, respectively (Table 2). These were enriched on glycolate medium at 50°C. All other isolates were obtained from IO"2 to IO"3 dilutions in TYE media or on mat agar incubated at SO0C or 70°C. Characterization of the 16S rRNAs of Octopus Spring Thermus isolates revealed four unique sequence types (sequences 1-4, Table 3). Isolates ac-6 and ac-14 were found to be identical to isolate ac-17 over 469 and 1369 bases through several variable regions, respectively, so no further sequence analysis of these isolates was performed. Although isolates ac-17 and ac-14 show identical 16S rRNA sequences, they exhibit stable differences in pigmentation (Table 2). Pigment variation in T. ruber has been previously described (6,13). Isolate ac-1 is nearly identical to Thermus strains ZHGI and ZHGIB cultivated from hot springs in Iceland (7). Isolate ac-7 is similar to T. aquaticus YT-I and Thermus sp. YSPED, two isolates obtained from hot springs in Yellowstone National Park (3,7). Strain YSPID was previously cultivated from Octopus Spring. Isolates ac-2 and ac-17 are very similar but not identical, and most similar to T. ruber, which was originally cultivated from hot springs in Russia (12). A distance matrix phylogenetic tree including these strains (Figure I) was quite similar to that produced by Saul et al. (20). The Octopus Spring isolates clustered within existing clades. Geographic isolation has been suggested to be important in determining Thermus distribution (8,19) using numerical classifications which assign organisms to Table 2. Thermus strains cultivated from the 50-55°C Octopus Spring cyanobacterial mat community Strain SampUng Date Medium Incubation Temperature (0C) Pigmentation Highest Dilution Total Cell Counts3 ac-17 29 Oct 1992 Glycolateb 50 Orange IO"6 5.3 x IO8 ac-14 29 Oct 1992 Glycolate 50 Red IO"5 5.3 x IO8 ac-2 30 Sept 1992 TYEc 50 Red IO"3 1.8 x IO9 ac-7 29 Oct 1992 Mat Agard 50 Yellow IO"3 5.3 x IO8 ac^l 30 Sept 1992 TYE 70 Yellow IO"3 1.8 x IO9 ac-6 29 Oct 1992 Mat Agar 50 Orange IO"2 5.3 x IO8 aSynechococcus cells/ml mat homogenate bO. I % w/v glycolic acid in Castenholz medium D amended with 1/3 v/v Octopus Spring water1 cO. I % w/v tiyptone + 0.1% w/v yeast extract in Castenholz medium Dt d10% v/v mat homogenate with 3% w/v agar in Castenholz medium D amended with 1/3 v/v Octopus Spring water' TpH of all media adjusted to 8.2 before autoclaving Table 3. 16S rRNA sequence similarities of selected Thermus species3 I 2 3 4 5 6 7 8 9 10 11 12 I. Thermus sp. ac-1 65 216 220 I .60 60 72 88 212 292 382 2. Thermus sp. ac-7 95.4 - 220 224 64 14 19 68 94 214 315 390 3. Thermus sp. ac-2 84.9 84.6 - 21 218 216 212 218 225 23 284 396 4. Thermus sp. ac-17 84.7 84.4 98.8 - 222 222 218 221 229 35 290 405 5. Thermus sp. ZHGIb 99.9 95.5 84.8 84.6 - 57 57 73 89 214 293 386 6. Thermus sp. YSPIDb 95.8 98.8 84.9 84.6 94.0 - 14 58 81 206 308 379 7. T. aquaticus YT-Ib 95.8 98.7 85.2 84.9 94.0 99.0 - 57 86 204 308 384 8. T. thermophilus HBSb 94.9 95.2 84.8 84.7 94.9 95.9 96.0 - 81 214 307 388 9. T. filiformish 93.8 93.4 84.3 84.1 93.8 94.3 94.0 94.3 - 217 295 387 10. T. ruberh 85.2 85.0 98.2 97.9 85.1 85.6 85.7 85.1 84.9 - 284 398 11. Deinococcus 79.6 78.0 80.0 79.7 79.6 78.5 78.5 78.6 79.5 80.0 408 radiodurans0 12. Escherichia coif ■ 74.1 73.6 73.2 72.8 73.9 74.4 74.1 73.8 73.9 73.1 72.4 3Values on the lower left are percent sequence similarities based on all available sequence data, values on the upper right are the absolute number of unambiguous nucleotide differences, many of which occurred as complimentary nucleotide substitutions in double-stranded regions. bStrains and nucleotide sequences as found in (10) 0Nucleotide sequences were derived from the RDP (20). I-TokS rTokS T-TokSO rW28 r - T . filiformis r Rt41A T351 -OK6 r FijiS —HS1 LfHBS 1 HB27 “T. fIavu^ ----------- NMX2 ZFI ZHGIB — ac-1 ZHGI —T. aquaticus YT-1 j-----ac-7 T-YSPID ___ I—ac-2 I— ac-17 — T. ruber New Zealand New Zealand New Zealand New Zealand New Zealand New Zealand New Zealand New Zealand Pacific New Zealand Japan Japan Japan USA Iceland Iceland USA Iceland USA USA USA USA USA Russia ■D. radiodurans Figure I. Distance matrix phylogenetic tree of the Thermus and Deinococcus lines of descent inferred from full 16S rRNA sequence data. Boldface denotes Octopus Spring aerobic chemoorganotrophic isolates. Remaining sequences and geographic origins are as described by Saul et al. (20). The tree was rooted to the 16S rRNA sequence of Escherichia coli. Scale bar represents 0.02 fixed point mutations per sequence position. 27 taxa according to phenotypic characteristics. Based on the geographic location of Thermus isolates used to construct a 16S rRNA-based phylogenetic tree, Saul et al. (20) also hypothesized a geographic basis for Thermus distribution. In contrast, within one hot spring we find a diversity of Thermus species that segregate into several major clades of the phylogenetic tree (Figure I). Our results are consistent with the observation that T ruber has been cultivated from diverse geographic sources when incubation temperatures were lowered to 50-60°C (4,6,11,22). Octopus Spring Thermus isolates belonging to distinct clades displayed optimal growth rates at different temperatures (Figure 2). T. aquaticus-like isolates ac-7 and ac-1 had temperature optima of 65°C and 70°C, respectively, while T. ruber-like isolates ac-17 and ac-2 displayed optimal growth rates at 50°C. Differences in temperature adaptations between T. ruber and T aquaticus have been previously reported (3,11). Abundance of Thermus spp. within the 50-55°C Octopus Spring mat appears to be related to temperature adaptation. Low temperature adapted isolate ac-17 was more numerically abundant in the 50-55°C mat (surviving a IO"6 dilution) than were high temperature adapted isolates ac-7 and ac-1 (surviving IO'3 dilutions). Ramaley and Bitzinger (16) also observed differential dominance of differently pigmented Thermus strains in a man-made thermal gradient. Specialization to temperature has been shown in other thermophilic genera (15), and may represent an evolutionary strategy driving diversity and community structure in thermal environments (24). Perhaps other Thermus populations exist in Octopus Spring mat that are specialized to different D ou bl in gs p er H ou r 28 *- ac-2 ac-17 Temperature (0C) Figure 2. Effect of temperature on growth rates of Octopus Spring Thermus isolates. Error bars indicate standard deviation, n=2. 29 parameters such as substrate and pH. Our results do not exclude the possibility that, geographic barriers may limit dispersal, thereby affecting distribution of some Thermus species. Indeed, major clades of the phylogenetic tree are composed of organisms cultivated from geographically distinct locations (e.g. T. filiformis and T. aquaticus clades are only known to contain organisms cultivated from New Zealand and Yellowstone National Park, respectively). However, our results do suggest that for some Thermus species, adaptation to local environmental conditions might help explain population distributions within the environmental gradients found in hot spring habitats. 30 References Cited 1. Bateson, M. M. and D. M. Ward. 1988. Photoexcretion and fate of glycolate in a hot spring cyanobacteria! mat. AppL Environ. Microbiol. 54:1738-1743. 2. Brock, T. D. 1978. Thermophilic Microorganisms and Life at High Temperatures, Springer-Verlag, New York. 3. Brock, T. D. and H. Freeze. 1969. Thermus aquaticus gen. n. and sp. n., a non-sporulating extreme thermophile. J. Bacteriol. 98:289-297. 4. Donato, M. M., E. A. Seleiro, and M. S. daCosta. 1990. Polar lipid and fatty acid composition of strains of the genus Thermus. System. Appl. Microbiol. 13:234-239. 5. Ferris, M. J., A. L. Ruff-Roberts, E. D. Kopczynski, M. M. Bateson, and D. M. Ward. 1996. Enrichment culture and microscopy conceal diverse thermophilic Synechococcus populations in a single, hot spring mat habitat. Appl. Environ. Microbiol. 62:1045-1050. 6. Hensel, R., W. Demharter, O. Handler, R. M. Kroppenstedt, and E. Stackebrandt. 1986. Chemotaxonomic and molecular-genetic studies of the genus Thermus: evidence for a phylogenetic relationship of Thermus aquaticus and Thermus ruber to the genus Deinococcus. Int. J. Sys. Bacteriol. 36:444-453. 7. Hudson, J. A. 1986. The taxonomy and ecology of the genus Thermus. Ph.D. Thesis, University of Waikato, Hamilton, New Zealand. 8. Hudson, J. A., H. W. Morgan, and R. M. Daniel. 1989. Numerical classification of Thermus isolates from globally distributed hot springs. System. Appl. Microbiol. 11:250-256. 9. Kopczynski, E. D., M. M. Bateson, and D. M. Ward. 1994. Recognition of chimeric small-subunit ribosomal DNAs composed of genes from uncultivated microorganisms. Appl. Environ. Microbiol. . 60:746-748. 10. Larsen, N., G. J. Olsen, B. L. Maidak, M. J. McCaughey, R. Overbeek, T. J. Macke, T. L. Marsh, and C. R. Woese. 1993. The ribosomal database project. Nucleic Acids Res. 21 (Supplement):3021-3022. 11. Loginova, L. G. and L. A. Egorova. 1975. An obligately thermophilic bacterium Thermus ruber from hot springs in Kamchatka. Mikrobiologiya 44:661-665. 12. Loginova, L. G., L. A. Egorova, R. S. Golovacheva, and L. M. Seregina. 1984. Thermus ruber sp. nov., nom. rev. Int. J. Sys. Bacteriol. 34:498-499. 13. Loginova, L. G., G. I. Khraptsova, T. I. Bogdanova, L. A. Egorova, and L. M. Seregina. 1978. A thermophilic bacterium Thermus ruber producing a bright orange pigment. Mikrobiologiya 47:561-562. 14 14. Olsen, G. J. 1988. Phylogenetic analysis using ribosomal RNA. Methods Enzymol. 164:793-812. 31 15. Peary, J. A. and R. W. Castenholz. 1964. Temperature strains of a thermophilic blue-green alga. Nature 202:720-721. 16. Ramaley, R. F. and K. Bitzinger. 1975. Types and distribution of obligate thermophilic bacteria in man-made and natural thermal gradients. Appl. Environ. Microbiol. 30:152-155. 17. Ramaley, R. F . and J . Hixson. 1970. Isolation of a nonpigmented, thermophilic bacterium sim ilar to Thermus aquaticus. J. Bacteriol. 103:527-528. 18. Ruff-Roberts, A. L., J. G. Kuenen, and D. M. Ward. 1994. Distribution of cultivated and uncultivated cyanobacteria and Chloroflexus-Yike bacteria in hot spring microbial mats. Appl. Environ. Microbiol. 60:697-704. 19. Santos, M. A., R. A. D. Williams, and M. S. daCosta. 1989. Numerical taxonomy of Thermus isolates from hot springs in Portugal. System. Appl. Microbiol. 12:310-315. 20. Saul, D. J., R. A. Rodrigo, R. A. Reeves, L. C. Williams, K. M. Borges, H. W. Morgan, and P. L. Bergquist. 1993. Phytogeny of twenty Thermus isolates constructed from 16S rRNA gene sequence data. Int. J. Sys. Bacteriol. 43:754-760. 21. Schut, F., E. J. de Vries, J. C. Gottschal, B. R. Robertson, W. Harder, R. A. Prins, and D. K. Button. 1993. Isolation of typical marine bacteria by dilution culture: growth, maintenance, and characteristics of isolates under laboratory conditions. Appl. Environ. Microbiol. 59:2150-2160. ; 22. Sharp, R. J. and R. A. D. Williams. 1988. Properties of Thermus ruber strains isolated from Icelandic hot springs and DNA:DNA homology of Thermus ruber and Thermus aquaticus. Appl. Environ. Microbiol. 54:2049-2053. 23. Ward, D. M., M. M. Bateson, R. Weller, and A. L. Ruff-Roberts. 1992. Ribosomal RNA analysis of microorganisms as they occur in nature, p.219-286. In K.C. Marshall (ed.), Advances in Microbial Ecology, Plenum Press, New York. 24. Ward, D. M., M. J. Ferris, S. € . Nold, M. M. Bateson, E. D. Kopczynski, and A. L. Ruff-Roberts. 1994. Species diversity in hot spring microbial mats as revealed by both molecular and enrichment culture approaches-relationship between biodiversity and community structure, p.33-44. In LJ. Stal and P. Caumette (ed.), Microbial Mats: Structure, Development and Environmental Significance, Springer-Verlag, Heidelberg. 25. Ward, D. M., A. L. Ruff-Roberts, and R. Weller. 1995. Methods for extracting RNA or ribosomes from microbial mats and cultivated microorganisms, 1.2.3:1-14. In A.D.L. Akkermans, J.D. van Elsas, and FJ. De Bruijn (ed.), Molecular Microbial Ecology Manual, Kluwer Academic Publishers, Dordrecht, The Netherlands. 26. Weller, R., M. M. Bateson, B. K. Heimbuch, E. D. Kopczynski, and D. M. Ward. 1992. Uncultivated cyanobacteria, Chloroflexus-Iikc inhabitants, and spirochete-like inhabitants of a hot spring microbial mat. Appl. Environ. Microbiol. 58:3964-2969. 32 CHAPTER 3 CULTIVATION OF AEROBIC CHEMOORGANOTROPHIC PROTEOBACTERIA AND GRAM POSITIVE BACTERIA FROM A HOT SPRING MICROBIAL MAT1 Introduction Recent studies investigating microbial species diversity in the Octopus Spring cyanobacteria! mat community have revealed a marked disparity between the native 16S rRNA sequence types observed in the mat using molecular retrieval techniques and the 16S rRNA' sequences of aerobic chemoorganotrophic bacteria cultivated from this and other geothermal habitats (34,35,38). Sequences retrieved from the Octopus Spring mat which may belong to organisms exhibiting aerobic chemoorganotrophic metabolic capabilities include planctomycete, proteobacterial, and Gram positive bacterial representatives, as well as relatives of green sulfur and green nonsulfur bacteria (35) [See Table I]. However, characterizing the metabolic capabilities of the bacteria which contain retrieved 16S rRNA sequence types is difficult without first cultivating these organisms. 1This study has been accepted for publication in Applied and Environmental Microbiology as: Nold, S. C., E. D. Kopczynski, and D. M. Ward. Cultivation of Aerobic Chemoorganotrophic Proteobacteria and Gram Positive Bacteria from a Hot Spring Microbial Mat. 33 Previous efforts to cultivate aerobic chemoorganotrophic bacteria from alkaline silicious hot spring cyanobacterial mats including Octopus Spring mat have yielded Bacillus (5), Thermomicrobium (19), Thermus (6), and Chloroflexus (4) isolates. These studies utilized similar cultivation techniques characterized by relatively high organic substrate concentrations (usually 0.1% w/v tryptone and yeast extract in liquid medium). An exception was the cultivation of the oligotrophic bacterium Isosphaera pallida using culture media devoid of organic substrates (15). Enrichments were often conducted at high incubation temperatures (70°C) regardless of the temperature of the collection site and the existence of temperature-adapted strains (23). Isolates were obtained by directly streaking mat material onto solidified media or by the direct addition of undiluted inoculum to enrichment flasks. Either of these methods preclude attempts to measure the relative abundance of the organisms cultivated. Since the development of enrichment culture techniques, microbiologists have suspected that these methods may select for the organisms which are best adapted to the enrichment culture environment, but which may not be the dominant organisms in nature (33,41). Our investigations resulted from our suspicion that the selectivity of the enrichment culture environment may explain the discrepancy between cultivated and naturally occurring populations detected by molecular retrieval techniques. In this study we attempted to cultivate more numerically abundant aerobic chemoorganotrophic bacteria from the 50 to 55°C region of Octopus Spring cyanobacterial mat community located in Yellowstone National Park. We utilized serial dilution enrichment culture (8,31) to provide a relative measure of the . 34 abundance of the isolates obtained and to avoid culture overgrowth by numerically insignificant species (14). We also enriched under seemingly more natural conditions (e.g. incubating at the temperature of sample collection, using more relevant substrates known to be present in the habitat such as mat material and glycolic acid (2), and using lower substrate concentrations). Previously reported results from this study (22) revealed a diversity of Thermus isolates cultivated from Octopus Spring mat whose distribution may be controlled by specialization to different temperatures which occur within the habitat. Here, we report the cultivation and 16S rRNA sequence characterization of phenotypically and phylogenetically distinct proteobacterial and Gram positive aerobic chemoorganotrophic bacteria from the Octopus Spring mat. Materials and Methods Cultivation of Isolates Procedures for sample collection and enrichment culture conditions were performed as described in Nold and Ward (22). Briefly, cyanobacterial mat samples were collected in September and October 1992 and November 1993 from the shoulder region of Octopus Spring (50 to 55°C, pH 8.5). Samples were kept between 46°C and SO0C for 3 hours in transit to the laboratory, then homogenized with a Dounce tissue homogenizer and serially diluted (1:10) in sterile medium D (9) before inoculation. Carbon sources included glycolic acid (GLD), casein (CND), solidified autoclaved mat homogenate (MTD), and a standard substrate used to cultivate Thermus species (6), 35 tryptone and yeast extract in medium D (TYD); details provided in Table 4. The sample for TYD enrichments was collected 30 September 1992, and the sample for MTD, CND, and GLD enrichments was collected 29 October 1992. On a later sampling date (9 November 1993) 0.1%, 0.01%, and 0.001% (w/v) tryptone and yeast extract were used as carbon sources. Liquid enrichments (50 ml GLD, TYD, or CND in 300 ml shake flasks) were inoculated with 5 ml serially diluted mat homogenate and incubated at 50°C with shaking (150 rpm) until turbid, then transferred to solidified medium containing 3% agar for isolation. Solidified mat homogenate was spread inoculated with 100 pi serially diluted inoculum and placed in a static incubator at 50°C. Individual colonies which exhibited unique and stable phenotypic properties (colony color, cell morphology, spore formation, motility) were re-streaked for purification and perpetuated from each medium type! Enumeration of cyanobacterial (Synechococcus spp.) cells by direct microscopic count was performed on appropriate dilutions of mat homogenate using a Petroff-Hausser counting chamber. Total direct counts of cyanobacterial cells in undiluted homogenized mat inocula were 5.3 x IO8 unicells ml"1 (TYD) and 1.8 x IO9 unicells ml'1 (GLD, CND, MTD). The abundance of each isolate is reported relative to the number of Synechococcus cells present in the diluted inoculum source used to inoculate the flask in which the isolate was observed. Characterization of 16S rRNA Sequences 16S rRNA sequence data were generated for each phenotypically unique isolate which grew from the most highly diluted inoculum in each enrichment type. 36 Harvested cells were lysed according to an established enzymatic protocol (36). Nucleic acids were extracted and the 16S rRNA gene was amplified using the polymerase chain reaction, cloned, and full-length {Escherichia coli positions 28-1483) (7) 16S rRNA sequence data were generated for isolates ac-15, ac-16, and ac-18 (GenBank accession numbers U46749, U46748, and U46747, respectively) as described by Kopczynski et al. (20). Partial 16S rRNA sequence data were generated for the remaining isolates by directly sequencing PCR products using the Sequenase PCR Product Sequencing Kit (United States Biochemical, Cleveland, OH) according to the manufacturer’s directions. The 16S rRNA sequence we report here which was retrieved directly from the mat (Octopus Spring type R, GenBank accession number U46750) was obtained from Octopus Spring library V-L using previously described methods (39). This sequence was previously reported as clone OS-V-L-28 (35), but was not characterized. Sequences were aligned and similarity values were calculated using the SeqEdit program version 3.0.4 provided by the Ribosomal Database Project (RDP) at the University of Illinois (21). All available sequence data, including ambiguous bases and alignment gaps, were included in similarity calculations. Sequences were analyzed for potential chimeric structures using the Check_Chimera program available through the RDP. Phylogenetic trees were constructed using the programs DNADIST, SEQBOOT, FITCH, and CONSENSE from the Phylogenetic Inference Package (PHYLIP) version 3.57c (11) with representative sequences derived from the RDP. 37 Results The outcome of enrichment culture experiments conducted on samples collected in September and October 1992 is summarized in Table 4. While phenotypically diverse bacterial isolates were obtained from TYD, MTD, and GLD enrichments, the enrichment containing casein as a carbon source yielded only one isolate (ac-12). Growth occurred to different dilutions depending upon medium type. Enrichments containing glycolate as a carbon source yielded an isolate (ac-21) which was cultivated from a IO8-Told dilution, which initially contained 9.0 x IO2 Synechococcus cells. TYD, CND, and MTD enrichments all yielded less abundant isolates cultivated from IO3- to IO3Told dilutions, initially containing 2.7 x IO7 to 9.0 x IO5 Synechococcus cells, respectively. The extent to which the inoculum was diluted before incubation affected the type of growth observed in liquid enrichment cultures. Less abundant populations (IO2- to IO3Told dilutions) exhibited turbid growth, while more abundant populations growing in GLD enrichments (106- to IO8- fold dilutions) were characterized by filamentous organisms growing as faint orange pellicles detected at the interface between air and water in the shaken enrichment flask. Isolation of these sheathed filamentous organisms (isolates ac-19, ac-20, and ac- 21) was achieved by streaking pellicle samples onto solidified medium D pre­ inoculated with Synechococcus isolate Cl (14). Subsequent transfer of colonies to liquid enrichments failed, thereby precluding 16S rRNA sequence characterization of these isolates. Table 4. Bacterial isolates cultivated from the 50 to 55°C Octopus Spring cyanobacterial mat community sampled during October and November 1992 Isolate Mediumb Highest Dilution Distinguishing Characteristics 16S rRNA Sequence Type Identical to: ac-3 TYDc IO"2 Motile coccobacillus ac-15 ac-4 TYD IO"3 Brown coccobacillus ac-18 ac-5 TYD IO"3 Yellow sporulating rod ac-18 ac-8 MTDd IO"4 Motile coccobacillus ac-15 ac-9 MTD IO"4 White sporulating rod ac-18 ac-10 MTD IO"5 Motile rod ac-16 ac-11 MTD IO"5 Sheathed orange rod, 0.5 x 5 pm N.D. ac-12 CNDe IO"5 Motile rod ac-16 ac-13 GLDf IO"4 Yellow sporulating rod ac-18 ac-15a- GLD IO"6 Motile coccobacillus ac-15 ac-16a GLD IO"6 Motile rod ac-16 ac-18* GLD IO"7 Colorless coccobacillus ac-18 ac-19 GLD IO"7 Sheathed orange trichome, 0.5 x 200pm N.D. ac-20 GLD IO"7 Sheathed orange trichome, I x 10-50 pm N.D. ac-21 GLD 10"8 Sheathed orange trichome, 0.6 x 10-50 pm N.D. aDenotes isolate from which full-length 16S rDNA sequence data are available. bpH of all media adjusted to 8.2 before autoclaving. cTYD, 0.1% (w/v) tryptone + 0.1% (w/v) yeast extract in Castenholz medium D. dMTD, 10% (v/v) Cyanobacterial mat homogenate with 3% (w/v) agar in Castenholz medium D amended with 1/3 (v/v) Octopus Spring water. eCND, 4% (w/v) casein in Castenholz medium D amended with 1/3 (v/v) Octopus Spring water. fGLD, 0.1% (w/v) glycolic acid in Castenholz medium D amended with 1/3 (v/v) Octopus Spring water. N.D. no data 39 In enrichment culture experiments conducted on samples collected in November 1993, carbon source concentration did not influence the extent of growth (all TYD concentrations exhibited growth to a IO8-Told dilution), but evidence of growth was observed sooner in the more dilute enrichments (0.001% and 0.01% TYD, 6 days) than in 0.1% TYD (44 days). Growth at the highest dilutions resembled the orange filamentous pellicle described above, and characterization attempts similarly failed. Isolates clustered into three phylogenetically distinct groups: proteobacteria, Gram positive bacteria, and Thermus (previously reported in (22)) (Figure 3). Isolate ac-15 contains a 16S rRNA nucleotide sequence which is identical to the previously retrieved |3-proteobacterial Octopus Spring type N sequence (34) over the 277 nucleotides available for comparison (Table 5). Isolate ac-16 and the retrieved but previously uncharacterized Octopus Spring type R sequence also displayed similarity to proteobacterial sequences (Table 5). Isolates ac-15 and ac-16, and the Octopus Spring type R sequence are compared to other mat proteobacterial sequence types and representatives of the major proteobacterial lines of descent in Figure 4. Analysis of diagnostic secondary structure (E. coli positions 140 to 223) and diagnostic nucleotide signatures (E. coli positions 50, 108, 124, 640, 690, 722, 760, 812, 871, 929, 947, and 1234) (42), as well as sequence similarity (Table 5) and tree results (Figure 4) support the inference that isolates ac-15, ac-16, and the retrieved Octopus Spring type R sequence belong to the (3 subdivision of the proteobacteria. Isolate ac-18 exhibited a 16S rRNA nucleotide sequence nearly identical to the sequence from Bacillus 40 r 28 100 68 100 - Isolate ac-16 — S p irillum vo lu tans Isolate ac-15=O.S. type N E sch e rich ia co li A g ro b a c te riu m tum e fa c ie ns ------- H e lic o b a c te r p y lo r i D e su lfo v ib rio d e su lfu rica n s F ib ro b a c te r in te s tina le s ■ S p iroch ae ta ha loph ila ------------------ P la n c to m y c e s s ta le y i 100 ---------- S yn ech ococcu s sp. str. 6301 — A rth ro b a c te r g lo b ifo rm is Isolate ac-18 B a c illu s fla vo the rm u s — C h lo rob ium v ib rio fo rm e B acte rio ide s fra g ilis ------------ C h lo ro flexu s a u ra n tia cu s ■ D e in oco ccus rad io du ran s 44f Isolate ac-2 100 I 88 100 Therm us ru b e r Isolate ac-17 a ticu s3T T h e rm u s a q u L Isolate ac-7 L Isolate ac-1 T herm o toga m aritim a Wm GS Pl Cy GN Th Figure 3. Distance matrix phylogenetic tree showing the placement of 16S rRNA sequences of aerobic chemoorganotrophic isolates cultivated from the Octopus Spring mat community relative to those of representatives of the major Bacterial lines of descent. Figure legend continued on page 43. Table 5. 16S rRNA sequence similarities of Octopus Spring isolates and selected proteobacterial and Gram positive sequence types* % similarity\number of unambiguous differences with sequence Sequence I 2 3 4 5 6 7 8 9 10 Proteobacteria I. Isolate ac-15 (1392) 0 172 142 66 152 177 365 363 385 2. O.S. type N (277) 99.8 - 29 22 27 28 44 65 65 79 3. Isolate ac-16 (1456) 87.7 89.7 161 60 87 185 357 358 391 4. O.S. type R (1346) 88.9 91.9 88.0 - 56 148 181 333 334 349 5. O.S. type G (588) 88.8 90.2 89.8 90.6 - 70 93 ' 120 120 145 6. Azoarcus denitrificans (1458) 89.1 90.0 93.9 89.0 88.2 - 193 367 365 402 7. O.S. type O (728) 76.0 77.6 75.0 75.0 79.8 73.8 - 170 169 211 Gram positive bacteria 8. Isolate ac-18 (1478) 74.6 76.8 76.1 75.9 79.9 75.5 77.1 6 341 9. Bacillus flavothermus (1477) 74.7 76.8 76.0 75.8 79.9 75.6 77.2 99.6 - 340 10. Thermotoga maritima (1481) 73.4 72.6 74.0 74.9 75.9 73.3 71.7 77.2 77.2 - “Values in the lower left are percent sequence similarities based on all available sequence data (number of nucleotides available for comparison in parentheses); Values in the upper right are the absolute number of unambiguous nucleotide substitutions. 80 100 88 46 24 98 — Zoogloea ramigera — Azoarcus denitrificans — Spirillum volutans Isolate ac-16 Isolate ac-15=O.S. type N ------------------------Rubrivivax gelatinosus O S. Type R ----- O.S. Type G ------------------------------Escherichia coli - O.S. Type O Agrobacterium tumefaciens --------------------------------------------------------------------------------------------- Helicobacter pylori ------------------------------------------------------------------------------------------------------Desulfovibrio desulfuricans -------------------------------------------------------------------------------------------------- Thermotoga maritima P y $ Figure 4. Distance matrix phylogenetic tree showing the placement of 16S rRNA sequences of cultivated and cloned proteobacterial Octopus Spring cyanobacterial mat populations relative to those of representatives of the major proteobacterial lines of descent (a, (3, y, 8, e). Boldface denotes proteobacterial sequences reported in this paper. The consensus values at the nodes indicate the number of times the group consisting of the species to the right of the node occurred among 100 trees inferred from the bootstrapped data set sampled by analysis of nucleotides which align with Escherichia coli 16S rRNA positions 802-825, 875-886, 1046-1114, 1157-1250, 1287-1392. Scale bar represents 0.01 fixed point mutations per sequence position. 43 Figure 3 legend (continued). Representatives were chosen from the following lines: Pr, proteobacteria; F, Fibwbacter; Sp, spirochetes and relatives; PI, Planctomyces and relatives; Cy, cyanobacteria; G+, Gram positive; GS, green sulfur bacteria; CE, Flexibacter-Cytophaga-Bacteriodes\ GN, green non-sulfur bacteria and relatives; DT, Deinococcus-Thermus subdivision of the green non-sulfur bacteria; Th, Thermotogales. Thermus sequences reported by Nold and Ward (22) are included to exhibit the full phylogenetic range of the aerobic chemoorganotrophic bacteria cultivated during these studies (isolates denoted by boldface type). The consensus values at the nodes indicate the number of times the group consisting of the species to the right of the node occurred among 100 trees inferred from the bootstrapped data set sampled by analysis of restricted nucleotide positions (39) which were common to all sequences (at least 898 nucleotides). This tree was rooted using the 16S rRNA sequence of Methanobacterium thermoautotrophicum. Scale bar represents 0.05 fixed point mutations per sequence position. flavothermus, a group 5 Bacillus in the low G+C subdivision of the Gram positive bacteria (25) cultivated from a hot spring habitat (16) (Figure 3, Table 5). We found no evidence of chimera formation in the 16S rRNA sequences of any of the isolates using Check_Chimera (low maximum improvement scores and lack of peakedness) and secondary structural analyses, but detecting chimeration can be problematic (20,28). Stable phenotypic differences were displayed by isolates ac-4, ac-5 (which was phenotypically similar to ac-13), ac-9, and ac-18 (Table 4), yet all exhibit Gram positive 16S rRNA nucleotide sequences identical to isolate ac-18 through the region 1086-1295, which includes the V9 variable region of the molecule (nucleotides 1110- 1276). Only two of these isolates (isolates ac-5 and ac-13) exhibited phenotypic similarity to B. flavothermus (i.e. rod-shaped morphology, spore formation, and yellow colony color) (16). Other isolates displayed similar morphology and identical (3 proteobacterial 16S rRNA sequence types. A motile rod (isolates ac-10, ac-12, and ac- 44 16) and a motile coccobacillus (isolates ac-3, ac-8, and ac-15) were isolated from different dilutions and substrate types and displayed identical 16S rRNA nucleotide sequences through nucleotides 1087-1289 and 1085-1280, respectively. Discussion Here we present the cultivation and full-length 16S rRNA sequence characterization of an organism whose sequence was previously observed in the Octopus Spring mat using molecular retrieval techniques. We detect no unambiguous 16S rRNA nucleotide differences between isolate ac-15 and the previously retrieved sequence fragment Octopus Spring type-N (Table 5). It appears that the type-N population corresponds to an aerobic chemoorganotrophic motile coccobacillus of (3- proteobacterial descent which is able to utilize glycolic acid, tryptone and yeast extract, and mat homogenate as carbon sources. Other examples of bacterial populations detected by both cultivation and 16S rRNA sequence retrieval are rare; Ferris et al. (14) successfully cultivated a relevant thermophilic cyanobacterium from this habitat, and Huber et al. (17) obtained an archaeal isolate whose 16S rRNA was previously detected. Isolates ac-15=O.S. type N and ac-16 are the first proteobacterial isolates to be cultivated from Octopus Spring mat community (35). These isolates are phylogenetically related to members of the (3 subdivision of the proteobacteria, a physiologically diverse clustering of organisms which display nitrogen fixation 45 (18,26,40), aerobic and anaerobic respiration (32), and photoautotrophic and photoheterotrophic capabilities (40). While isolates ac-15=O.S. type N and ac-16 clearly exhibit aerobic chemoorganotrophic metabolic capabilities, they may, like other (3-proteobacteria, display physiological versatility. Thus, their metabolic functions exhibited in Octopus Spring mat community are currently unknown. Four of the five known proteobacterial sequences detected in the Octopus Spring mat are most similar to members of the (3 subdivision. We have observed a similar pattern of multiple representatives within one phylogenetic type in the cyanobacterial, green sulfur, green non-sulfur, and ThermuslDeinococcus lines of descent (12,22,30). One possible explanation for this recurring pattern could be that progenitor bacteria within a phylogenetic group became specialized to conditions which vary in the habitat, resulting in subsequent evolutionary radiation and the observed diversity of modern 16S rRNA types (37,43). Isolate ac-18 contains the first thermophilic Bacz1ZZMj1-Iike 16S rRNA sequence observed in the Octopus Spring mat (35). Isolates ac-4, ac-5 (and ac-13), ac-9, and ac-18 are phenotypically distinct (Table 4), yet share this 16S rRNA nucleotide sequence at least through the V9 variable region (nucleotides 1239-1298). This observation could be due to either the highly conserved nature of the 16S rRNA molecule (42), or to undetected differences in other regions of the molecule. Since only limited data were obtained from those isolates which exhibited identical 16S rRNA sequences (207 to 222 nucleotides), we cannot reject this latter possibility." Sequence similarity within phenotypically diverse thermophilic Gram positive bacteria I 46 has been previously observed (3). Caution should be applied when interpreting bacterial diversity detected by a conservative genetic marker, since populations exhibiting identical 16S rRNA nucleotide sequences may contain highly related yet phenotypieally distinct members. By observing the extent of growth from a serially diluted inoculum, an estimate of the relative abundance of cultivated strains can be obtained (8). The highest dilution in which we observed isolate ac-15=O.S. type N (106-fold) originally contained almost five orders of magnitude more Synechococcus cells in the inoculum (9.0 x IO4), indicating the relative numerical insignificance of this cultivated species in native mat material. If we can assume that the frequency at which a sequence is observed in cloning libraries reflects the abundance of that sequence in nature, then the low abundance of isolate ac-15 in mat homogenate may explain why the Octopus Spring type N sequence was detected only once using molecular retrieval techniques (34). Isolates ac-16 and ac-18 were cultivated from similar dilutions (IO6- and IO7- fold, respectively), but these populations have not been previously observed in cloning libraries constructed from Octopus Spring mat nucleic acid. Isolates ac-19, ac-20, and ac-21, which exhibited growth in the highest dilutions (IO7- and 108-fold), could not be grown to sufficient quantity for sequence analysis. These organisms grew as sheathed trichomes similar to the Chloroflexus species described by Pierson and Castenholz (24), and also obtained from aerobic chemoorganotrophic enrichments by Brock (4). Glycolate has previously been identified as a substrate for aerobic chemoorganotrophic metabolism in Octopus Spring 47 mat. Under illuminated conditions, glycolate is excreted by photosynthetically active Synechococcus cells and is readily incorporated by filamentous Chloroflexus-Iike organisms (2). This result confirms the importance of glycolate as a carbon source for aerobic chemoorganotrophy in Octopus Spring mat, and illustrates the importance of using ecologically relevant carbon substrates to cultivate the more numerically abundant bacterial species. Combining cultivation and molecular retrieval approaches has allowed us to confirm the Chloroflexus-Yike nature of the sheathed trichomes which grow to the highest dilutions in GLD medium (30). The other carbon substrates (i.e. CND, TYD, and MTD) only yielded growth to a 105-fold dilution. Further growth in these enrichments may have been inhibited by the inappropriateness of the carbon source provided or the relatively high substrate concentrations provided in these enrichments. To test the hypothesis that substrate concentration influences the extent of growth in serial dilution enrichment culture, 0.1%, 0.01%, and 0.001% tryptone and yeast extract were provided in separate dilution series. The observation that substrate concentration did not influence the extent of growth (all substrate concentrations yielded growth to a IO8-Told dilution) was unexpected, since the levels of soluble organic substrates in densely populated cyanobacterial mats may be quite low, favoring organisms adapted to low substrate concentrations. The aerobic mat heterotroph Isosphaera pallida was successfully cultivated using oligotrophic substrate concentrations (i.e. unamended mineral salts medium) (15). The relatively high substrate concentrations provided in the enrichments reported in this paper may have selectively recovered only those 48 populations adapted to high organic carbon concentrations. These results allow the comparison of cloning and cultivation techniques as methods to describe microbial diversity in natural environments. Although attempts were made to cultivate the more numerically abundant aerobic chemoorganotrophic bacteria from Octopus Spring mat by utilizing serial dilution enrichment culture and possibly more natural incubation conditions, we still failed to cultivate most organisms whose 16S rRNA sequences were previously retrieved. Since samples for cultivation and molecular cloning experiments were not collected simultaneously, this failure may have been due to seasonal bacterial population variation. However, recent studies of 16S rRNA sequence type variation in Octopus Spring mat have shown remarkable seasonal stability of bacterial populations (13). Alternatively, our failure to cultivate organisms with retrieved 16S rRNA sequences may indicate either that the retrieved \ sequences do not correspond to an aerobic chemoorganotrophic metabolism or that we did not cultivate the relevant organisms. The inability of enrichment culture to recover predominant populations is well documented (I), and isolation techniques which require growth as a colony on solid media may further limit retrieval of relevant organisms (29,37), so simplified species diversity using cultivation methods is not unexpected. However, 16S rRNA sequence retrieval methods may also underestimate species diversity by only detecting those species whose nucleic acids are readily cloned or PCR amplified (10,27). Although we expect the 16S rRNA of numerically abundant organisms to appear in cloning libraries, these techniques may suffer from a lack of sensitivity, resulting in the inability to detect less abundant populations such as I those presented in this paper. 50 References Cited 1. Amann, R., W. Ludwig, and K. H. Schleifer. 1995. Phylogenetic identification and in situ detection of individual microbial cells without cultivation. Microbiol. Rev. 59:143-169. 2. Bateson, M. M. and D. M. Ward. 1988. Photoexcretion and fate of glycolate in a hot spring cyanobacterial mat. Appl. Environ. Microbiol. 54:1738-1743. 3. Bateson, M. M., J. Wiegel