INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Soil and litter contain steep oxygen (O₂) gradients and anoxic microzones (51, 57). Acetate is the most abundant organic acid in soil extracts (18, 52, 54), and the anaerobic acetate-forming capacities of soils and litter are likely linked to oxidative acetate-consuming processes (32-35, 58). Thus, acetate might be an important intermediate in the turnover of carbon in terrestrial ecosystems and serve as a trophic link at oxic-anoxic interfaces in soil and litter (12, 35). Soils and litter harbor facultative and strict anaerobes capable of producing acetate (35, 43); however, the interactions between the microorganisms involved in these trophic relationships are not well resolved. Indeed, although it is well established that the microflora of soils facilitate both aerobic and anaerobic processes, information on the coexistence and interaction of the organisms associated with these fundamentally different processes is limited.

Acetogenic bacteria are strict anaerobes that engage the acetyl coenzyme A (acetyl-CoA) Wood-Ljungdahl pathway for the reductive synthesis of acetyl-CoA from CO₂ and have been isolated mostly from sediments or gastrointestinal tracts (11, 49). Although soil is not a strictly anoxic habitat, acetogens are, nonetheless, the most enumerable of strict anaerobes in soil and litter (35, 43). The capacity of soils to form acetate from H₂-CO₂ is enhanced by high temperatures (32, 58), suggesting that soils that are subject to elevated temperatures might harbor thermophilic acetogens. During efforts to isolate thermophilic acetogens from such soils (21, 23), a thermophilic coculture of an acetogen (ET-5a) and a fermentative microaerophile (ET-5b) was obtained. The main objectives of this study were to characterize these two thermophilic organisms and to resolve the trophic-level basis of their interaction.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Soil collection. Surface soil (the first 3 cm of depth) was collected from a grassy garden in Hurghada, Egypt. Soil was transported to the laboratory and stored at 5°C for 4 weeks prior to use. The soil exhibited an approximate pH of 7.4 and a dry weight of 96.9%; the total carbon content and organic carbon content of the soil approximated 23.0 and 10.8 g (kg [dry weight] of soil¹), respectively.

Medium composition and growth conditions. The anoxic, carbonate-buffered, undefined (U) medium contained yeast extract, vitamins, trace metals, reducer (sodium sulfide and cysteine hydrochloride), and resazurin (redox indicator) (7). The defined (D) medium was U medium without yeast extract. U and D media were dispensed under CO₂ into 27-ml culture tubes (7 ml of medium per tube) or 1-liter infusion bottles (500 ml of medium per bottle, used for preparation of cell extracts), which were then sealed and autoclaved; the pH was approximately 6.7. Tryptic soy broth (TSB) medium contained 28 g of TSB liter¹; anoxic TSB medium had a 100% N₂ gas phase. The reduction of iron was determined by assessing the growth-dependent production of white Fe(II) precipitates in medium formulated for the growth of Fe(III)-reducing bacteria (4). Culture tubes and bottles were incubated in a horizontal, static position. Unless otherwise indicated, the temperature of incubation was 55°C.

Enrichment cultures. Soil samples were brought into a Mecaplex (Grenchen, Switzerland) O₂-free chamber (100% N₂ gas phase; room temperature) and added to anoxic medium (approximately 5 g [wet weight] of soil per 45 ml of D medium in a 150-ml infusion bottle). The medium was supplemented with vanillate (5 mM); the gas phase was H₂-CO₂ (a ratio of 1:3 at approximately 30 kPa of overpressure). Enrichments were incubated at 55°C and subsequently streaked onto medium solidified with 1% Gelrite (Carl Roth GmbH, Karlsruhe, Germany). Subsequent enrichments in medium with carbon monoxide (CO) utilized a CO-CO₂ gas phase (a ratio of 1:3 at approximately 30 kPa of overpressure). Isolated colonies were transferred to liquid medium, cultivated, and assayed for substrate utilization and product formation.

Transmission electron microscopy. Cells were negatively stained with uranyl acetate or phosphotungstic acid (56). For thin-section preparations, cells were fixed in glutaraldehyde-OsO₄ and prepared according to a standard protocol (55). Thin sections were stained with 2% (wt/vol) aqueous uranyl acetate and lead citrate (46). Specimens were observed with a Zeiss CEM 902A (Oberkochen, Germany).

Preparation of cell extract and enzyme assays. Cells were cultivated in U medium on the substrate indicated and harvested in early stationary phase. Cells were lysed in anoxic lysozyme buffer (38), and cell extracts were prepared under anoxic conditions (31). The enzyme assay buffer was 100 mM Tris hydrochloride (pH 8.5) containing benzyl viologen (1 mM) and dithiothreitol (1 mM); the assay temperature was 55°C. To determine the specific activities of CO dehydrogenase, formate dehydrogenase, and hydrogenase, assay tubes were supplemented with CO (100% gas phase), sodium formate (5 mM), or H₂ (100% gas phase), respectively (10).

Membrane preparation and redox difference spectra. Membranes were prepared from cell extracts by ultracentrifugation under aerobic conditions (14, 19). Washed membranes were reduced with sodium dithionite, and reduced-minus-oxidized (oxidized indicates that the membranes in the reference cuvette were not reduced with sodium dithionite) spectra were obtained with a Uvikon 930 (Kontron Instruments, Milan, Italy) double-beam recording spectrophotometer at room temperature (19).

G+C content. Cells were washed with 50 mM phosphate buffer (pH 7.0) and DNA was extracted by the NaOH method (1). The G+C content was determined by high-performance liquid chromatography (42).

16S rRNA gene sequence. The 16S rRNA gene sequences of ET-5a and ET-5b were determined by direct sequencing of the PCR-amplified 16S rRNA genes; 1,557 and 1,426 nucleotides were sequenced for ET-5a and ET-5b, respectively. Genomic DNA extraction, PCR-mediated amplification of the 16S rRNA gene, and purification of the PCR products were performed according to published protocols (45). For ET-5a, purified PCR products were sequenced with a Sequi-Gen GT sequencer (Bio-Rad Laboratories GmbH, Munich, Germany). For ET-5b, purified PCR products were sequenced with an ABI PRISM Dye Terminator Cycle Sequencing Ready Reaction kit and a 373A DNA sequencer (Applied Biosystems, Foster City, Calif.).

Construction of the dendrogram for ET-5b. The stretch of 1,426 nucleotides of the 16S rRNA gene of ET-5b (positions 18 to 1438 of the Escherichia coli sequence [3]) was initially aligned to the sequences of the ARB database (54a). Following determination of the approximate position within the radiation of bacterial phyla, the sequence of strain ET-5b was transferred to the German Collection of Microorganisms and Cells (DSMZ; Braunschweig, Germany) database of members of the Clostridium-Bacillus subphylum with the AE2 editor (39). Evolutionary distances were calculated by the method of Jukes and Cantor (29). Phylogenetic dendrograms were constructed according to the method of DeSoete (8) and by the neighbor-joining method contained in the PHYLIP software package (16, 48). Bootstrap analysis was used to evaluate the tree topology of the neighbor-joining data by performing 500 resamplings (15).

Additional analytical methods. Growth and cell dry weights were determined as previously described (7). When cultivated in oxic medium (at an initial gas phase of 17% O₂) and anoxic U medium supplemented with 10 mM glucose, a culture optical density (at 660 nm) of 1 corresponded to 0.34 and 0.37 mg (dry weight) of cells liter¹, respectively. Protein was determined by dye staining and colorimetric analysis (2). The amounts of substrates and products present in culture fluids and gas phases were determined by high-performance liquid chromatography and gas chromatography (7, 22, 30, 40). The concentration of gases represents the combined total of both the liquid and gas phases. Soil pH was determined in 1:2.5 suspensions of soil in 0.02 M CaCl₂, and soil dry weight was obtained by weighing before and after drying at 105°C for 16 h. Total carbon of oven-dried (65°C), homogenized organic matter was quantitated with an element analyzer (CHN-O-Rapid; Foss-Heraeus, Hanau, Germany). In this study, no distinction is made between CO₂ and its salt forms and between organic acids and their salt forms. All results are representative of replicate experiments.

Nucleotide sequence accession numbers. The 16S rRNA gene sequences of ET-5a and ET-5b have been deposited in the EMBL nucleotide sequence database (Cambridge, United Kingdom) under accession no. AJ242494 and AJ242495, respectively.

Culture accession numbers. Cultures of ET-5a and ET-5b have been deposited at the DSMZ under accession no. 12797 and 12793, respectively.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Isolation of ET-5a and ET-5b. A colony (ET-5) that was picked from solidified medium and transferred into liquid medium was subsequently shown to grow both aerobically on saccharides and anaerobically with H₂-CO₂; under the latter condition, H₂-CO₂ was converted acetogenically to acetate. Since no acetogen has been shown to be capable of aerobic growth, it was suspected that ET-5 consisted of more than one organism. This possibility was evaluated by growing ET-5 in both oxic and anoxic media and subsequently preparing oxic or anoxic dilution series (1:10 in U medium), respectively, from these cultures. A strict anaerobe was obtained from the highest growth-positive dilution of an anoxic CO-CO₂ dilution series (U medium) and was designated ET-5a; this rod-shaped, spore-forming organism grew acetogenically with both H₂-CO₂ and CO-CO₂. An organism capable of both aerobic and anaerobic growth was obtained from the highest growth-positive dilution of an oxic cellobiose dilution series (U medium) and was designated ET-5b; this rod-shaped organism did not grow acetogenically with H₂-CO₂ or CO-CO₂. The purities of ET-5a and ET-5b were assured by repeated isolation on solidified medium.

Phylogenetic analyses of ET-5a and ET-5b. The 16S rRNA gene sequence of ET-5a was 99.8% similar to that of Moorella thermoacetica. The morphology, substrate range, and product profile of ET-5a were very similar to those of M. thermoacetica (data not shown), confirming that ET-5a is a new strain of this classic thermophilic acetogen.

View larger version (38K): [in this window] [in a new window]   FIG. 1.   Phylogenetic tree of strain ET-5b, representative organisms from the Bacillus subgroup, and two representatives of the Clostridium subgroup of the Clostridium-Bacillus line of descent. Sequence accession numbers and strain numbers are indicated; the strain number of M. thermoacetica is not available from the Ribosomal Database Project (39). Numbers within the dendrogram indicate the percentages of occurrence of the branching order in 500 bootstrapped trees (only values of 70 and above are shown). The sequence of Clostridium botulinum served as a root. The scale bar represents 10 nucleotide substitutions per 100 nucleotides.

Morphology and ultrastructure of ET-5b. Cells of ET-5b were approximately 2.5 × 0.5 µm and stained weakly gram positive. Colonies on solidified medium were beige. Cells were motile, and the helical flagellar core was surrounded by a flexible, discontinuous sheath that was morphologically separate from the core (Fig. 2A to C). Flagella were inserted laterally (Fig. 2D). Cells appeared to be enveloped by a capsule (Fig. 2D), and the S-layer was composed of hexagonal subunits (53) (Fig. 2E). Thin sections of ET-5b revealed both outer and cytoplasmic membranes (Fig. 2F), indicating that cells contained a periplasm (26). Spores were not apparent.

View larger version (153K): [in this window] [in a new window]   FIG. 2.   Electron micrographs of ET-5b. ET-5b is shown negatively stained with 2% uranyl acetate (pH 4.6) (A, B, C, and E), negatively stained with 2% phosphotungstic acid (pH 7.0) (D), and by ultrathin section (F). Panel C is a high magnification of the flagellar sheath separated from the core; the large arrow in panel D points to the fibrillar structures surrounding the cell in the capsular domain. Abbreviations: F, flagella; C, flagellar core; SH, flagellar sheath; H, flagellar hook; S, surface layer; OM, outer membrane; CM, cytoplasmic membrane. Bars are in micrometers.

Effect of O₂ on the growth of ET-5b. Elevated amounts of O₂ impaired the growth of ET-5b (Fig. 3). Older cultures often did not grow when transferred into oxic medium. ET-5b consumed O₂, and the resazurin in oxic medium was reduced after O₂ was consumed. In U medium lacking reducer, growth did not occur when the gas phase contained 21% O₂ but did occur when the gas phase contained 5% O₂. Cells grew rapidly in anoxic TSB medium but did not grow in TSB medium when culture bottles were lightly sparged with filter-sterilized air. Thus, cultures of ET-5b were more easily maintainable under anoxic conditions or conditions with limited amounts of O₂.

View larger version (20K): [in this window] [in a new window]   FIG. 3.   Effect of O2 on the growth of ET-5b in U medium supplemented with 10 mM glucose. The initial concentrations (%) of O2 in the gas phase (the remaining gas was N2) were 0 (), 5 (), 12 (), 21 (), 28 (), 34 (), and 44 (+). Tubes containing O2 were initially oxidized; the resazurin in all growth-positive tubes was reduced during growth, indicating that O2 was consumed in those tubes. Culture tubes were inoculated with exponentially growing cells from an anaerobic culture.

Doubling times and effects of temperature and pH on the growth of ET-5b. Growth occurred at 37 to 65°C in anoxic U medium supplemented with glucose. The optimum temperature for growth was 55 to 60°C; growth was not observed at 30 or 70°C. Cultures did not grow when subjected to 100°C for 5 min, further demonstrating that ET-5b did not form spores. Growth occurred at pH 5.5 to 7.7 in anoxic TSB medium supplemented with glucose. The optimum pH for growth was 6.5 to 7.0; growth was not observed at pHs 5.1 and 8.1. Doubling times under both anoxic and O₂-limited conditions were very similar (Fig. 3) and were approximately 1.5 to 2 h.

Substrate range of ET-5b under anoxic and oxic conditions. Cultures of ET-5b were maintainable in anoxic D medium supplemented with either cellobiose or glucose; thus, yeast extract was not required for growth. In D medium, the following substrates supported growth under both oxic (i.e., with an initial gas phase of 21% O₂) and anoxic conditions: stachyose, raffinose, cellobiose, maltose, sucrose, lactose, galactose, fructose, glucose, mannose, and xylose. Anaerobic growth on stachyose occurred only after several days of incubation; this extended lag phase did not occur when the inoculum was derived from anaerobic raffinose cultures. Succinate, acetate, and formate were growth supportive only under oxic conditions. ET-5b did not grow aerobically or anaerobically with the following substrates: cellulose, arabinose, gluconate, glyoxylate, lactate, pyruvate, oxalate, ethanol, catechol, protocatechuate, vanillate, CO, H₂, vanillate plus CO, and vanillate plus H₂.

Effect of O₂ on the soluble product profiles of ET-5b. When ET-5b was grown on acetate, succinate, or formate under oxic conditions, substrates were consumed and no soluble products were detected (Table 1 and data not shown). Likewise, the amount of carbon recovered in the soluble products from stachyose, raffinose, xylose, and other saccharides (see above) under oxic conditions was significantly lower than that obtained under anoxic conditions (Table 1 and data not shown). In addition, the amount of biomass formed under oxic conditions was greater than that obtained under anoxic conditions (Table 1). These results indicated that ET-5b was capable of oxidizing substrates to CO₂ via O₂-dependent respiration.

Cytochrome content of ET-5b. Particulate (i.e., membrane-associated) and soluble b-type cytochromes were detected in anaerobically cultivated cells of ET-5b (Fig. 4). When membranes were prepared from cells cultivated under oxic conditions, the absorption maxima at 428 and 559 nm shifted to 433 and 560 nm, respectively.

View larger version (23K): [in this window] [in a new window]   FIG. 4.   Difference spectra of soluble (A) and particulate (B) material obtained from cells of ET-5b cultivated anaerobically on glucose in U medium. Vertical bars indicate the relative scale for the change in absorbance.

Dynamics of product formation. When grown anaerobically on cellobiose in D medium, ET-5b produced acetate, succinate, ethanol, lactate, formate, and H₂ simultaneously during exponential growth (Fig. 5A). Acetate, succinate, ethanol, and lactate reached stable end concentrations simultaneously with the complete consumption of cellobiose and the onset of the stationary phase. The consumption of formate in the stationary phase was concommitant to the continued production of H₂ (Fig. 5A), suggesting that stationary-phase cells contained formate-hydrogen lyase. Neither CO nor CH₄ was detected.

View larger version (23K): [in this window] [in a new window]   FIG. 5.   Cellobiose-dependent product profiles of ET-5b cultivated in D medium in the absence (A) and presence (B) of a high initial concentration of O2 (approximately 21% of the initial gas phase). In panel B, the phase of maximal growth and the period of maximal O2 consumption is enclosed in the broken-line box. Inocula were derived from maintained anoxic (A) and oxic (B) cultures. Symbols: , growth; , cellobiose; , acetate; , succinate; , formate; +, lactate; , ethanol; , H2; , O2 (mM × 0.34; initial concentration approximated 28 mM).

View larger version (28K): [in this window] [in a new window]   FIG. 6.   Cellobiose-dependent product profiles of ET-5b cultivated in nonreduced U medium in the presence of a low initial concentration of O2 (approximately 5% of the initial gas phase). The phase of maximal growth and the period of maximal O2 consumption is enclosed in the broken-line box. Symbols: , growth; , cellobiose; , acetate; , succinate; , formate; +, lactate; , ethanol; , H2; , O2.

Trophic interaction of ET-5a and ET-5b. ET-5a did not grow with stachyose, raffinose, or cellobiose, but it grew acetogenically at the expense of lactate, formate, or H₂ (Table 2 and data not shown). Although ET-5a grew rapidly on fructose, growth on glucose was marginal. When ET-5a and ET-5b were cultivated together on cellobiose, the products lactate, formate, and H₂ remained at relatively low levels throughout growth (Fig. 7). The end concentrations of these products were either nondetectable or minimal when ET-5a and ET-5b were cocultured on stachyose, raffinose, cellobiose, or glucose (Table 2 and data not shown).

View larger version (24K): [in this window] [in a new window]   FIG. 7.   Cellobiose-dependent product profiles of a coculture of ET-5a and ET-5b in anoxic U medium. Symbols: , growth; , cellobiose; , acetate; , succinate; , formate; +, lactate; , ethanol; , H2.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

In the dendrogram generated with the ARB program, ET-5b was placed as an individual and deeply rooting line of descent next to the genera Bacillus, Paenibacillus, Brevibacillus, Thermoactinomyces, and Aneurinibacillus, as well as the closely related genus Oxalophagus. When the ET-5b 16S rRNA gene sequence was compared to those in the more extensive DSMZ database, the isolated position of this sequence as a separate subline of the Bacillus genus proper was confirmed (Fig. 1). The generation of various dendrograms with changing sequence composition by the neighbor-joining method and alternative algorithms (8) led to slightly changing branching points; however, ET-5b was not located within the radiation of a reference genus. Similarity values between the sequences of strain ET-5b and reference organisms from the Clostridium-Bacillus lineage were below 88%, a value that is at least 5% lower than the intrageneric values of the genera listed above. The phylogenetic position of strain ET-5b between the moderately thermophilic species Alicyclobacillus acidoterrestris (viable at 45°C) and Thermoactinomyces vulgaris (viable at 50°C) is not supported by high bootstrap values, confirming the low statistical significance of its branching point among the lineages of gram-positive bacteria with low moles percent G+C content. These findings demonstrate that ET-5b represents the nucleus of a new genus, and the following name is proposed for the type strain: Thermicanus aegyptius ET-5b (Therm.i.ca'nus ae.gyp'ti.us).

T. aegyptius ET-5b grows at the expense of a broad range of substrates, including oligosaccharides such as stachyose, and prefers anoxic conditions or conditions with limited amounts of O₂. As such, T. aegyptius ET-5b might be best described as a thermophilic fermentative microaerophile. Under oxic conditions, T. aegyptius ET-5b is able to utilize certain products that it produces under anoxic conditions. As an organism that resides in an environment prone to fluctuations in O₂ and organic carbon levels, these factors likely contribute to the competitiveness of T. aegyptius ET-5b under in situ conditions.

T. aegyptius ET-5b contained a membranous b-type cytochrome when grown anaerobically. The shift in the absorption maxima of the membranous chromophores of cells cultivated in the presence of O₂ indicated that redox conditions might influence the production of dissimilar cytochromes by T. aegyptius ET-5b. The absorption maxima of the soluble and particulate material were also dissimilar (Fig. 4). However, it cannot be unequivocally stated that these differences in absoption maxima were attributable to different cytochromes. Nonetheless, the differential expression of cytochromes in aerobically and anaerobically cultivated cells of facultative aerobes is well established. For example, E. coli produces different b-type cytochromes in response to changes in the availability of O₂ (5). In M. thermoacetica, a membranous b-type cytochrome that is involved in the flow of reductant during acetogenesis is not expressed when cells are dissimilating nitrate (12, 19). Since T. aegyptius ET-5b contains a periplasm, the soluble cytochrome detected might be localized in the periplasm rather than in the cytoplasm. Resolving the function(s) of the cytochrome(s) of T. aegyptius ET-5b during growth under oxic and anoxic conditions will require further study.

The acetogen M. thermoacetica ET-5a grew commensally with T. aegyptius ET-5b via the interspecies transfer of H₂, formate, and lactate (Fig. 8). In contrast to T. aegyptius ET-5b, M. thermoacetica ET-5a did not utilize oligosaccharides; thus, these two organisms would not be in direct competition for these primary substrates. Indeed, the substrate profiles of these two organisms are quite different. Under certain conditions, the interspecies transfer of reductant between strict anaerobes is a syntrophic process. For example, the capacity of Syntrophomonas wolfei to grow at the expense of butyrate is thermodynamically possible only via the syntrophic transfer of H₂ to a methanogen (41). Likewise, various H₂-consuming methanogens and sulfate reducers can be cocultured syntrophically with Acetobacterium woodii or other acetogens via the interspecies transfer of H₂ (25, 60). The pectin fermenter Lachnospira multiparus and the acetogen Eubacterium limosum (which is unable to utilize pectin) inhabit the rumen and interact trophically via the interspecies transfer of methanol, ethanol, formate, lactate, and H₂ (47).

View larger version (26K): [in this window] [in a new window]   FIG. 8.   Scheme illustrating the postulated trophic interaction of T. aegyptius ET-5b and M. thermoacetica ET-5a.

As has been documented with certain facultative anaerobes (24, 59), cultures of T. aegyptius ET-5b produced fermentation products concomitant with the consumption of O₂. In such cultures, whether some cells of T. aegyptius ET-5b were strictly respiring O₂ while others were strictly fermentative is not known. The production of lactate and formate by T. aegyptius ET-5b under oxic conditions (Fig. 4B) suggests that a trophic interaction between T. aegyptius ET-5b and M. thermoacetica ET-5a might occur in the presence of O₂ (conceived in the scheme shown in Fig. 8 as a zone of changing conditions). This possibility assumes that M. thermoacetica ET-5a can tolerate minimal levels of O₂. Although O₂ is classically considered to be toxic to strict anaerobes, certain sulfate reducers and methanogens can consume at least minimal levels of O₂ and survive periods of oxidation (6, 9, 36), and acetogens in prairie soil can withstand drying under oxic conditions (58). Acetogenic bacteria can be readily enriched and enumerated from well-drained, aerated soils and litter (35, 43, 58), and a new acetogen, Sporomusa silvacetica, was recently isolated from forest soil (31); these findings attest to the ability of acetogens to cope with periodic in situ fluctuations in O₂ levels. The sensitivity of T. aegyptius ET-5b to large amounts of O₂ suggests that it would function optimally in microzones prone to anoxia or minimal aeration, thus increasing the likelihood that T. aegyptius ET-5b could reside proximally to M. thermoacetica ET-5a under in situ conditions.

It is not improbable that the survival strategy of acetogens in soils is at least partly coupled to their trophic interactions with facultative microaerophiles. Under chemostatic conditions, a sulfate-reducing bacterium (Desulfovibrio HL21) and a facultatively anaerobic Vibrio species were maintained in coculture under low-O₂ conditions (24). In addition, the obligate anaerobe Veillonella alcalescens can coexist with the obligate aerobe Comamonas testosteroni in the presence of low levels of O₂ (20). Such observations suggest that O₂-consuming organisms lessen the O₂-dependent inhibition of anaerobes. Thus, facultative microaerophiles might not only produce products that can be used commensally by acetogens but also minimize the level of O₂ in microzones inhabitated by acetogens. It has been proposed that the acetate formed in anaerobic microsites of soils is primarily subject to oxidation via aerobic or other respiratory processes (12, 32, 58). Since T. aegyptius ET-5b grew aerobically on acetate, the acetate produced by M. thermoacetica ET-5a might be subject to O₂-dependent oxidation by its microaerophilic partner under certain conditions, thus further benefiting the acetogen via the consumption of incoming O₂. Previous studies have documented the capacity of aerobic organisms to oxidize the fermentation products of an obligate anaerobe when such organisms are cocultured under low-O₂ conditions (20).

M. thermoacetica grows very poorly on ethanol under acetogenic conditions but grows rapidly on ethanol when dissimilating nitrate (19), suggesting that ethanol produced by T. aegyptius ET-5b might also be used commensally by M. thermoacetica in the presence of nitrate. Furthermore, H₂- and formate-dependent cell yields of M. thermoacetica are significantly enhanced under nitrate-dissimilating conditions (19). Thus, the trophic interaction of M. thermoacetica ET-5a and T. aegyptius ET-5b would theoretically be more dynamic than the results depicted in Fig. 8 if nitrate were available. Determining how closely T. aegyptius ET-5b and M. thermoacetica ET-5a are associated under in situ conditions would provide further insight into their capacity to interact in soil.

M. thermoacetica is the classic acetogen which H. G. Wood and L. G. Ljungdahl used to resolve the acetyl-CoA pathway (61). It is the best-characterized acetogen to date, and much of the information available on the physiology and enzymology of acetogenesis is based on work done with this organism (13, 37, 44). M. thermoacetica was originally isolated from horse manure as a glucose-fermenting heterotroph (17) but was later shown to (i) contain hydrogenase (10) and be lithoautotrophic on both H₂-CO₂ and CO-CO₂ (7); (ii) preferentially dissimilate nitrate to ammonium (19, 50); (iii) utilize a diverse array of substrates, including carboxylic acids, alcohols, oxalate, glyoxylate, glycolate, and numerous methoxylated aromatic compounds (12, 13); and (iv) utilize the carboxyl groups of certain aromatic compounds in the reductive synthesis of acetate (27, 28). The isolation of M. thermoacetica ET-5a from Egyptian soil as a trophic partner of a facultative microaerophile was unexpected and accentuates the fact that very little is known about the ecology of this historically important acetogen. The occurrence of M. thermoacetica in Kansan and Egyptian soils (references 21 and 23 and unpublished data) indicates that the organism is a soil microbe that is geographically widespread. Current studies are focused on resolving the effects of O₂ on the interaction and ecophysiology of M. thermoacetica ET-5a and T. aegyptius ET-5b, and the co-occurrence of these two thermophiles in well-aerated, high-temperature soils.

Description of Thermicanus aegyptius ET-5b^T gen. nov., sp. nov. Thermicanus aegyptius ET-5b^T gen. nov., sp. nov. (DSMZ 12793^T) (therm.i.ca'nus. Gr adj. thermos, hot; Gr. adj. ikanos, M.L. icanus, capable; M.L. Thermicanus, the capable thermophile; ae.gyp'ti.us. L. adj. aegyptius, Egyptian or from Egypt). A motile, weakly gram-positive, thermophilic, facultative microaerophile isolated from Egyptian soil. Cells are rod shaped and have an S-layer and outer and cytoplasmic membranes; spores not apparent. Growth is optimal at 55 to 60°C and pH 6.5 to 7; doubling times are approximately 1.5 to 2 h. High levels of O₂ impair growth; prefers anoxic or microaerophilic conditions. Substrates include stachyose, raffinose, maltose, sucrose, cellobiose, lactose, galactose, glucose, fructose, mannose, and xylose. Acetate, formate, and succinate are utilized only under oxic conditions. Does not grow on cellulose, arabinose, gluconate, glyoxylate, lactate, pyruvate, oxalate, ethanol, catechol, protocatechuate, vanillate, CO, and H₂. Fermentation products are acetate, succinate, ethanol, formate, lactate, and H₂; fermentation products are also formed under oxic conditions. Nitrate, sulfate, and thiosulfate are not dissimilated; Fe³⁺ is reduced to Fe²⁺ as a side reaction. Particulate and soluble fractions have b-type cytochrome(s). G+C content is 50.3 mol%. Most closely related to Paenibacillus, Bacillus, and Oxalophagus, with the nearest 16S rRNA gene sequence similarity value being approximately 88%. The thermophilic acetogen M. thermoacetica ET-5a grows commensally with T. aegyptius ET-5b^T on oligosaccharides via the interspecies transfer of H₂, formate, and lactate.