You are on page 1of 9

INTERNATIONAL JOURNAL OF SYSTEMATIC BACTERIOLOGY, July 1986, p. 435-443 0020-77131861030435-09$02.

OO/O Copyright 0 1986, International Union of Microbiological Societies

Vol. 36, No. 3

Isolation and Characterization of Acidothermus cellulolyticus gem nov., sp. nov., a New Genus of Thermophilic, Acidophilic, Cellulolytic Bacteria
A. MOHAGHEGHI,' K. GROHMANN,' M. HIMMEL,l L. LEIGHTON,' AND D. M. UPDEGRAFF2* Fermentation Section, Biotechnology Branch, Solar Fuels Division, Solar Energy Research Institute, and Department of Chemistry and Geochemistry, Colorado School of Mines,2 Golden, Colorado 80401

'

Twelve isolates of thermophilic, acidophilic, cellulolytic bacteria were obtained from three different primary enrichment cultures from acidic hot springs at Yellowstone National Park, Wyo. The three isolates which had the highest cellulolytic activity, as shown by the diameter of clearing zones surrounding colonies on cellulose agar plates, were selected for intensive study. All were gram-variable, nonsporulating aerobic rods which formed no pigment. They grew at 37 to 65"C, with optimum growth at 55C. The pH range for growth was 3.5 to 7, with an optimum pH of 5. The guanine-plus-cytosine content of the deoxyribonucleic acid was 60.7 k 0.6 mol%. The organisms are resistant to penicillin G at 100 pglml. They share several important features with Thermus strains, namely heterotrophic, aerobic, and thermophilic mode of growth; morphological features; sensitivity to lysozyme; and presence of catalase. They differ in other important aspects, such as the pattern of carbon sources utilized for growth, the pH and temperature profiles of growth, the pattern of sensitivity to antibiotics, the guanine-plus-cytosine content of DNA, the composition of amino acids in the cell walls, and the structure of the cell walls. Thermus species are very sensitive to penicillin G, whereas our strains are resistant. Our strains also are different, in important respects, from the genus Thermomicrobium. Therefore, we designate the organism Acidothermus ceZZuZoZyticus gen. nov., sp. nov., of which the type strain is our strain 11B, ATCC 43068.

This research effort was undertaken as part of a research program on the production of fuel alcohol from cellulosic biomass materials. The driving hypothesis is that a thermophilic, acidophilic, cellulolytic bacterium can be grown in coculture with other thermotolerant microorganisms to produce ethanol at a high rate from cellulosic wastes. Alternatively, the cellulolytic organism could be used to produce cellulase, which could be added to a cellulose-containing culture of an organism which ferments glucose to ethanol. Most of the thermophilic, cellulolytic microorganisms isolated to date are sporeforming anaerobes belonging to the genus Clostridium. None of these are acidophilic, and all have an optimum pH near 7. Thus, they would not grow well under the acidic conditions (pH 3 to 5 ) prevalent in an ethanolic fermentation. Aerobic, thermophilic, cellulolytic microorganisms include several species of fungi (10) and a few species of filamentous bacteria belonging to the family Actinomycetaceae (33). Among the fungi, Myceliophthora thermophila is of considerable interest, since it grows well on cellulose in submerged culture at 50C and pH 4.5 and synthesizes cellulolytic enzymes (10). Thus, the organism is acidophilic, as well as cellulolytic and thermophilic. Rosenberg (24) screened 21 species of thermophilic and thermotolerant fungi for the ability to degrade cellulose and found that 13 species had this ability. It was not determined whether these organisms could grow well in submerged culture under thermophilic and acidophilic conditions and produce significant amounts of cellulase. Aspergillus

* Corresponding

author.

fumigatus has been shown to grow well on cellulose in submerged culture at 45C and pH 4.6 and to produce considerable amounts of cellulase (27). Among the thermophilic, aerobic bacteria, only a few actinomycetes are actively cellulolytic, notably Thermomonospora curvata (28) and Thermoactinomyces cellulosae (1, 29). However, neither of these organisms is acidophilic. The acidic hot springs at Yellowstone National Park, Wyo., provide an excellent source for the isolation of acidophilic thermophilic bacteria. The distribution of pH among the springs at Yellowstone is bimodal, with a large group of springs in the highly acidic range, pH 2 to 5 , and another group in the near-neutral to alkaline range, pH 6 to 10. Brock (2) isolated Thermus aquaticus, Bacillus acidocaldarius, and Sulfolobus species from Yellowstone springs, and Jackson et al. (9) isolated Thermomicrobium species. All of these organisms are obligate thermophiles, and B . acidocaldarius and Sulfolobus are acidaphilic; however, none of these are known to be cellulolytic. Since the major objective of the screening program was to isolate microorganisms for potential application in the fermentation of cellulosic substrates, the following major constraints were placed on enrichment conditions and properties of these isolates: pH range, 3 to 6 (optimum pH of 5); temperature range, 50 to 60C; and secretion of the cellulase enzyme complex. Selection for specific pH range was done to match the pH range at which many industrial fermentations (e.g., ethanol production) are carried out. Furthermore, we assumed that pH profiles of cellulase enzymes produced by such microorganisms would be similar to pH profiles for growth of these microorganisms. The temperature selected for enrichment is slightly higher than the highest temperature (50C) at which yeasts seem to be able
435

436

MOHAGHEGHI ET AL.

INT. J. SYST.BACTERIOL.

to grow (17, 25); a similar assumption, with respect to temperature profiles of cellulolytic enzymes and growth temperature, was made.
MATERIALS AND METHODS Bacterial strains. The 12 isolates described were obtained from the upper Norris Geyser basin area in Yellowstone National Park. The acidic springs selected had temperatures of 45 to 65C and a pH range of 4 to 5.5. The strains were isolated by procedures described below. All of the isolates formed similar colony types on cellulose agar and D-cellobiose agar, and all showed identical morphology. The type strain, 11B, has been deposited with the American Type Culture Collection, Rockville, Md., with the number 43068. All strains have been successfully maintained both in a freeze-dried state and frozen at -70C after the addition of 77 pl of dimethyl sulfoxide per ml of culture suspension. Culture media. Enrichment cultures were prepared with a low-phosphate basal salts medium (LPBM) which contained the following, in grams per liter: NH4C1, 1.0; KH2P04, 1.0; N a 2 H P 0 4 . 7 H 2 0 , 0.1; M g S 0 4 . 7 H 2 0 , 0 . 2 ; a n d CaC12 . 2H20, 0.2. This medium was also supplemented with the following, in grams per liter: yeast extract (Difco Laboratories, Detroit, Mich.), 0.5; D-cellobiose, 0.5; and cellulosic substrate, 5 . The cellulosic substrates used included filter paper strips (Whatman number 1;Whatman Ltd., England), Sigmacell 50 powdered cellulose (Sigma Chemical Co., St. Louis, Mo.), carboxymethyl cellulose (46 MF; Hercules Inc., Wilmington, Del.), and amorphous phosphoric acidtreated swollen cellulose (28). Solid media contained 25 g of Bacto-Agar (Difco) per liter. All media were adjusted to pH 5.2 and sterilized by autoclaving at 121C for 20 min. Isolation conditions. Portions (9-ml each) of sterile LPBM containing 0.05% yeast extract, 1%Sigmacell 50, and filter paper strips were put up in capped culture tubes. Each tube was inoculated with 1 ml of a slurry of mud and decaying vegetation from acidic (pH 4.5 to 5.5) hot (45 to 65C) springs. Tubes were placed in a field incubator within a few minutes of collection and incubated for 1 to 2 weeks. Samples showing bacterial growth or signs of cellulose decomposition were subcultured to fresh tubes of the same medium. Pure culture isolation. The filter paper strips served as an indicator of cellulolytic activity and in some instances provided a solid surface for the growth of colonies of cellulolytic bacteria. Samples of the degrading paper strips were teased off with an inoculating loop and streaked on plates of LPBM supplemented with 0.05% yeast extract and 0.5% amorphous (phosphoric acid-swollen) cellulose (3 1). Cellulolytic colonies were identified by the presence of a clear zone around the colonies. These colonies were generally found to contain more than a single species; therefore, they were streaked on carboxymethyl cellulose and D-cellobiose agar containing LPBM supplemented with yeast extract. Colonies were picked from these plates and streaked out again on the swollen cellulose agar for the isolation of pure cultures. Microscopy. All of the cultures were repeatedly examined (magnification, X 1,000) under phase-contrast illumination with a Nikon Biophot microscope. All isolated pure cultures were examined when growth first appeared, and at intervals of 3 or 4 days thereafter for up to 2 weeks, to search for morphological developmental changes such as endospore formation, motility, or rod-to-coccus changes such as those reported for Arthrobacter species.

Cells of strains 10B, 11A, and l l B T , grown to logarithmic phase in LPBM-yeast extract-D-cellobiose medium, were examined by electron microscopy. The cells were pelleted by centrifugation and washed twice with 0.1 M N-2hydroxyethylpiperazine-N'-2-ethanesulfonic acid (HEPES) buffer (pH 6.8). The pellets were suspended in 4% glutaraldehyde (electron microscopy grade; Sigma) fixative, buffered with HEPES buffer, and incubated at 20C for 1.5 h. Samples were then washed with three changes of HEPES buffer, set in Noble M agar, cut into 1-mm cubes, postfixed in 1%buffered osmium tetroxide for 2 h at 20"C, and washed again in HEPES buffer. The samples were then stained for 2 h in 2% uranyl acetate and washed three times in HEPES buffer. Samples were dehydrated with an ethyl alcohol series (25 to 100% ethyl alcohol), followed by dehydration with an ethanol-propylene oxide series (100% ethyl alcohol to 100% propylene oxide), after which the samples were embedded in Epon-812 resin. After curing at 60C for 48 h, the samples were sectioned on a Reichardt Ultracut model E-4 ultramicrotome. These thin sections (60 nm each) were stained with a 2% solution of uranyl acetate and stained again with Reynolds lead citrate solution. Sections were examined with a Phillips EM-400T transmission electron microscope operated under standard conditons. Samples to be examined by scanning electron microscopy were fixed with glutaraldehyde and osmium tetroxide, mounted on polylysine-coated cover slips, dehydrated with an acetone elution series, subjected to critical point drying, and coated with gold-palladium. The Au-Pd coating was approximately 19 nm thick. Samples were examined with a Hitachi HHS-2R electron microscope at a 5-mm working distance with 15" of tilt. Magnifications were calibrated with commercial replicas made from ruled diffraction gratings with 15,240 lines per in. (1 in. = 2.54 cm) (number 64005; Ladd Research Industries, Burlington, Vt.). Cell growth. Cell growth in fiquid cultures containing only soluble substrates was measured by determination of absorbance at 640 nm with a Spectronic 20 (Bausch & Lomb, Inc., Rochester, N.Y .) spectrophotometer and, in some cases, with a Petroff-Hauser bacteria-counting chamber and a phase-contrast microscope. Cultures containing cellulose were extremely turbid; in this case, growth was estimated by a modified Lowry protein assay (15). The cell pellets were suspended in Lowry reagent A, and cellulose and cell debris were removed by centrifugation after lysis induced by heating to 60C for 20 min. The lysates were cooled and centrifuged, the supernatants were mixed with Lowry reagent B, and the chromophore was developed as usual. The good agreement between these two widely different methods is shown in Fig. 1. DNA composition analysis. Deoxyribonucleic acid (DNA) was isolated by the procedure of Marmur (16), modified. Cells were lysed by a lysozyme-ethylene diaminetetraacetate (EDTA)-sodium dodecyl sulfate (SDS) treatment. Lysates were then extracted twice with phenol, and the DNA was precipitated from aqueous layers by the addition of 2 volumes of ethanol. After centrifugation, the pellets were dissolved in 50 mM Tris chloride-5 mM EDTA buffer (pH 8 .O) and treated successively with ribonuclease A (Sigma) and protease K (Sigma). The density of solutions was adjusted to 1.68 g/ml by the addition of cesium chloride, and DNA was purified by two cycles of equilibrium density centrifugation with a Beckman VTi 50 rotor at 41,000 rpm for 24 h in a Beckman L8-70 ultracentrifuge. Purified DNA solutions were dialyzed against 0.1 x standard saline-citrate buffer (SSC) ( I x SSC is 0.15 M NaCl plus 0.015 M sodium

VOL. 36, 1986

ACIDOTHERMCJS CELLULOLYTICUS GEN. NOV., SP. NOV.

437

citrate), and the DNA melting temperature was determined with a Gilford spectrophotometer equipped with the thermal programming accessory. T, was determined with Escherichia coli B DNA (Sigma) as a standard as suggested by Owen and Hill (21). The guanine-plus-cytosine (G + C) content of the DNA was also determined by analytical ultracentrifugation with a Beckman model E ultracentrifuge by the methods of Vinograd and Hearst (32) and Szybalski (30). By these methods, the apparent buoyant density of the DNA sample is determined from its equilibrium position and from that of an internal standard DNA ( E . coli B) in the ultracentrifuge cell. This condition was found, from serial photographs, to take approximately 42 to 48 h at 44,000 rpm in an AN-D rotor with RTIC temperature control at 25C. DNA samples were prepared in 1 mM Tris-buffered solutions of 56.0% cesium chloride (5.65 M) to an optical density at 260 nm equal to about 0.02. DNA G + C composition was estimated with the relationships described by De Ley (6). Amino acid analysis of cell wall preparations. Cells were grown to mid-logarithmic phase in a batch culture on 1% glucose in LPBM and harvested by low-speed centrifugation (7,000 rpm, GSA rotor) in a Sorvall RC5B centrifuge at 4C for 15 min. The cell pellets were washed twice by centrifugation in deionized water. A portion of the pellets was lyophilized for amino acid analysis. The remaining pelleted fractions were suspended in 50 ml of deionized water and homogenized in a Braun homogenizer at 3,000 rpm for 8 min at 4C. Glass beads (450 to 500-pm diameter; Sigma) were used to homogenize the cells. No visible lysis occurred during this step. Cell suspensions were filtered and suspended in 150 ml of 2% SDS-25 mM EDTA . Na3 solution. Light microscopy confirmed that lysis of the cells occurred instantly at room temperature. Also, the viscosity of the suspensions increased dramatically. The cell suspensions were placed for 30 min into a water bath held at 60C to complete the lysis, and the lysates were centrifuged at 8,000 rpm in a GSA rotor at 25C. The pellets were suspended in 20 ml of SDS-EDTA solution and stirred at room temperature overnight. Suspensions were maintained at 60C for 30 min and pelleted again by centrifugation. The pellets were then washed once with 0.2% SDS solution and twice with deionized water, and fractions were then lyophilized for amino acid analysis. The remaining pellet fractions were suspended in 45 ml of 50 mM sodium phosphate buffer (pH 7.8) with 20 mg of crystalline ribonuclease A (Sigma) and incubated for 2 h at 37C. The mixtures were then incubated at 37C for 6 h with 50 mg of trypsin (1:250, vol/vol; Difco) and pelleted by centrifugation. The pellets were suspended in 1% SDS solution, heated to 60C for 30 min, and pelleted again. Finally, the pellets were washed twice with deionized water and lyophilized. After dialysis with water and lyophilization, 62-pg samples of the cell wall preparation were dissolved in 5OO-pl portions of 4.7 N HC1 containing 0.05% (vol/vol) phenol and 0.5% (vol/uol) P-mercaptoethanol. The tubes were evacuated, sealed, and then hydrolyzed at 110C for 20 h. Each hydrolysis was performed in duplicate for each hydrolysis time. The amino acids and amino sugars were analyzed with a Beckman model 121 CL amino acid analyzer. The amino acid composition was not normalized for losses or apparent increases during hydrolysis. Nutritional studies and biochemical tests. The nutritional requirements of the three selected strains were determined by adding the various nutritional supplements to mineral base LPBM supplemented with trace elements at the following final concentrations, in milligrams per liter: FeS04 . 7 H20, 0.5; MnS04 . H 2 0 , 0.1; Na2Mo04 . 2H20, 0.1; CuC12,

1.0 -

1
-

loo

0.8
0

80

0,
! -

E
a ,

2 0
Ln

c .-

0.6 -

60

m
Ln

I I-

0.4 -

-40

c3

c
c3

0.2 -

- 20

to-

0.05; CoCl2 . 5 H 2 0 , 0.05; Z n S 0 4 . 7 H 2 0 , 0.05; and Na2B407. 10H20,0.05. Major nutrients were added at 0.5 to 1%concentration. Each 5-ml tube of medium was inoculated with 0.1 ml of a washed cell suspension in distilled water. Routine biochemical tests were carried out as described in the Manual of Methods for General Bacteriology (7) and elsewhere (13). Volatile fatty acids were determined with a HewlettPackard model 5840A gas chromatograph equipped with a flame ionization detector and a Carbopack C/0.3% CW 20M/0.1% H3P04 packed column (6 ft [l ft = 30.48 cm] by 2 mm; Supelco). The column, injection port, and detector temperatures were maintained at 125, 145, and 250"C, respectively. Nitrogen served as the carrier gas. The column was calibrated with volatile fatty acid standards obtained from Supelco. Nonvolatile fatty acids were determined by high-pressure liquid chromatography using the HPX-87H column from Bio-Rad Laboratories. The high-pressure liquid chromatography system consisted of a Beckman model llOB pump, a Beckman-Altex model 210 injection valve, and a Waters model 450 variable wavelength detector adjusted to 220 nm. The mobile phase was 10% (vollvol) acetonitrile in 0.0007 N H2SO4. Nonvolatile fatty acid standards were obtained from Sigma. Effects of temperature and pH on growth. The relationships among growth rate, temperature, and pH were determined by chemostat studies in Biogen and Bioflo (New Brunswick Scientific Co., Inc., Edison, N.J.) fermentors with 2-liter reactors charged with 1.2 liter of LPBM supplemented with 1g of yeast extract per liter and 10 g of D-cellobiose per liter. Antibiotic resistance. Antibiotic resistance tests were carried out in LPBM with 0.1% yeast extract and 0.5% Dcellobiose. The antibiotics were dissolved in water or an appropriate solvent as stock solutions, filter sterilized, and

438

MOHAGHEGHI ET AL.

INT. J. SYST.BACTERIOL.

FIG. 2. Scanning electron micrograph of strain l l B T cultures showing flocculated cells. Bar, 1 km. added to the culture medium at the following concentrations, in micrograms per ml: 0.1, 1,10,25,50, and 100. Tubes were inoculated and incubated at 55C for 72 h. Growth was followed by measurement of the absorbance at 640 nm. Measurement of cellulase activities. Cellulase activity in culture supernatants was measured by a filter paper assay (14) in 25 mM sodium acetate buffer (pH 4 . 8 ) . Samples were incubated for 1 h at 55 and 1 h at 65C. Reducing sugars formed from cellulose were determined by a dinitrosalicylic acid method (18). Reducing sugars produced by cellulose hydrolysis were characterized further by high-pressure size exclusion chromatography (8). Induction of cellulase enzymes. Initial studies on cellulase induction were conducted in batch cultures with shake flasks of LPBM supplemented with 0.5% Sigmacell 50 powdered cellulose or 0.5% amorphous cellulose at 55C. Cell growth and cellulase activities were monitored as described above. These tests were followed up by batch experiments in the Biogen fermentor set at 55C and maintained at pH 5.0 by the addition of sodium hydroxide.
RESULTS Enrichment and isolation. Pure cultures were isolated from the enrichment cultures grown on powdered cellulose media as described in Materials and Methods. The presence of persistently contaminating bacteria necessitated repeated streaking on two different kinds of agar. A total of 12 strains were isolated from three different enrichments from acidic hot springs (pH range, 4 to 5.5; temperature range, 50 to

60C). All of the isolates formed identical colony types on LPBM-yeast extract-cellulose , D-cellobiose, and carboxymethyl cellulose agars. All produced cellulase , as shown by clear zones on amorphous cellulose agar, and all were similar morphologically. Three strains, 10B, 11A, and llBT, which had the highest cellulolytic activity as determined by clearing zone assay, were selected for intensive study. Cultural characteristics. All 12 strains formed small (1to 3 mm), smooth, circular, entire, creamy white colonies on agar plates containing LPBM-yeast extract-D-cellobiose and amorphous cellulose. Transfers to LPBM plus yeast extract agar with added carboxymethyl cellulose, Dcellobiose, or D-glucose yielded the same colony types. Morphology. Upon initial isolation on cellulose agar, all strains formed slender rods and long, slender filaments, ca. 0.4 by 5 to 20 Fm. Upon repeated transfer, many of the strains formed a higher proportion of shorter rods, 0.5 by 2 to 5 pm, and fewer long filaments. In liquid media, both rods (0.5 x 2 to 5 pm) and filaments were present in a ratio which varied from strain to strain, as did the propensity for flocculation. At no time were spores or other resting bodies produced by any of the pure cultures. Motility was not observed, nor were flagella seen in electron micrographs. The large rotund bodies described for T . aquaticus by Brock and Freeze (4) were never observed, Gram stains were usually gram negative but occasionally gram positive. No clear relationship between phases of growth and staining properties of these strains has been observed. Scanning and transmission electron micrographs of strain

VOL. 36, 1986

ACIDOTHERMUS CELLULOLYTICUS GEN. NOV., SP. NOV.

439

FIG. 3. Transmission electron micrograph of a thin section of strain l l B T cells. The specimen was fixed with glutaraldehyde-osmium tetroxide and stained with uranyl acetate and lead citrate. Bar, 0.1 pm.

l l B T are shown in Fig. 2 and 3, respectively. Other electron micrographs (data not shown) demonstrated identical cell wall structures for strains 10B and 1lA. Mesosomelike structures and internal cell membranes are clearly visible in Fig. 3. Also, the outermost side of the cell wall appears to have a thin, electron-dense layer. Although the plasma membrane is well defined, no outer cell membranes were seen in thin sections of all three strains at any magnification. Thus, the cell walls are distinctly different from the typical gram-netative cell walls of T . aquaticus (4) and Thermomicrobium species (9, 23). Influence of temperature and pH on growth rate. The relationships between temperature and specific growth rate in LPBM plus 0.1% yeast extract and 0.5% D-cellobiose maintained at pH 5.5 and aerated with 1 volume of air per volume of fermenter per min are shown in Fig. 4. All three strains grew between 37 and 70C. From multiple experiments, the average temperature optima for growth were found to be 60, 55, and 50C for strains 10B, llBT, and 11A, respectively. The maximum specific growth rates (k) were about 0.15 per h for all strains, which is equivalent to a generation time of 6.7 h. The effect of pH on growth determined in the chemostat with the same medium at 55C is illustrated in Fig. 5. Strains 10B and l l B T grew between pH 3 and pH 7, whereas strain 11A grew from pH 4 to pH 7. The average pH optimum from

multiple experiments was approximately 5.0 for strains 10B and llBT; strain 11A had a broad pH optimum, from 5 to 6. Cultures growing in liquid media with D-glucose or Dcellobiose produce organic acids, and the pH may drop from 5 to 6 to as low as 3.4, resulting in the death of the culture. Succinic, maleic, and citric acids were found in growth supernatants at concentrations less than 0.1% (wt/wt); acetic acid was found only in trace quantities. Nutritional studies. The nutritional studies, summarized in Table 1, show that many common nutrients, including cellulose, xylan, starch, D-cellobiose, D-maltose, D-raffinose, sucrose, D-trehalose, D-glucose, D-galactose, D-mannose, sorbitol, and D-mannitol, are capable of serving as a sole source of energy and carbon, with ammonium ion as the nitrogen source for all three strains studied. A few nutrients supported growth of one or more of the three strains, and a few others (acetate, DL-lactate, citrate, and pectin) failed to support growth of any strain. The vitamin supplement, containing a mixture of vitamins, failed to stimulate growth. Amino acids, including vitamin-free Casamino Acids (Difco), asparagine, and glutamic acid, were definitely stimulatory. In fact, 0.1% Casamino Acids alone supported fair growth, as did 0.1% yeast extract plus 0.1% tryptone. There was no growth in nutrient broth or on nutrient agar at pH 5.5, and not only did citrate and acetate fail to support growth, but they also proved to be inhibitory at 0.01 M

440

MOHAGHEGHI ET AL.

INT. J. SYST.BACTERIOL.

TABLE 1. Nutritional studies


Substrate Utilization by strainasc: 10B 11A llBTC

Starch D-Trehalose D-Fructose D-Xylose L- Arabinose Glycerol


?, slight growth; -, no growth. The following substrates permitted growth (+) of all three strains: cellulose, xylan, D-cellobiose, D-maltose, D-raffinose, sucrose, D-glucose, Dgalactose, D-mannose, D-sorbitol, and D-mannitol. The following substrates permitted no growth (-) of all three strains: acetate, citrate, DL-lactate, and pectin. T, Type strain.

" +, Positive growth;

- 20

40

60

80

TEMPERATURE ("C)

FIG. 4. Plot of specific growth rates (p) at different temperatures. pH was maintained at 5.5.

concentration. Therefore, the organisms are prototrophic, requiring no vitamins or accessory growth factors. Biochemical tests. The following routine biochemical tests were carried out at 55C in liquid or solid media adjusted to the pH range 5 to 6: motility, catalase, oxidase, sulfur oxidation, hydrogen sulfide production, nitrate reduction, gelatin liquefaction, casein hydrolysis, growth on egg yolk, indole production, hemolysis of blood agar, methyl red test, Voges-Proskauer test, growth in litmus milk, and citrate utilization (Simmons citrate). No variability in response of the three strains to these tests was observed. The catalase test was positive, indole production was weakly positive, and all other tests yielded negative results. Shake cultures in LPBM plus 0.1% yeast extract and 0.1% D-glucose or D-cellobiose plus 7 g of purified agar (Oxoid Ltd., London, England) per ml yielded growth only near the surface of the agars when incubated in air; no growth was observed in nitrogen, indicating that the organisms are obligate aerobes. Salt tolerance tests carried out with strains 10B, 11A, and l l B T in LPBM-yeast extract-glucose medium showed that growth was equally good with 0, 0.5, 1.0, 1.5, and 2.0% added NaC1, but growth was poor at 4% added NaC1. Thus, the organisms are moderately osmotolerant.
20

Antibiotic resistance tests. Resistance to 12 antibiotics was determined in liquid cultures with the following concentrations of antibiotics, in micrograms per ml: 0.1, 1, 10, 25, 50, and 100. The concentrations of individual antibiotics at which no growth of all three strains occurred were as follows, in micrograms per ml: vancomycin, 1; ampicillin and chloramphenicol, 10; bacitracin, 25; polymyxin B and rifampin, 50; and erythromycin and kanamycin, 100. All three strains were also resistant to penicillin G at 100 pg/ml. Variable response was observed for actinomycin D, which prevented growth of strains 10B and 11A at 10 pg/ml and of strain l l BT at 25 kg/ml, and for tetracycline, which prevented growth of strains 10B and 11A at 25 pg/ml and of strain l l B T at 10 pglml. Amino acid analysis of cell wall preparations. The amino acid and amino sugar compositions of hydrolysates of purified cell wall fractions are presented in Table 2. The amino acids cysteine, cystine, and tryptophan are not listed in the table, since they are, to a large extent, destroyed by the
TABLE 2. Amino acid composition of cell wall hydrolysates
Amino acid in hydro1ysate Amino acid composition (nmoY100 pl of purified cell wall fraction hydrolystate) of strain: 10B-1" 10B-2" 11A llBT

15
0

?\

X
'L
v

0
c

1 0

5
PH

Aspartate Threonine Serine Muramic acidb Glutamate Proline Glycine Alanine Valine Diaminopimelic acid Methionine Isoleucine Leucine Tyrosine Phen ylalanine Glucosamine Histidine Lysine Arginine

1.8 1.7 7.1 3.2 8.1 1.8 3.4 8.8 2.5 10.5 0.7 1.2 2.8 0.9 1.3 7.9 ND' 1.5 1.6

2.6 2.3 9.6 6.1 9.7 2.1 4.5 10.6 3.O 20.1 0.4 1.2 3.5 1.2 1.7 14.2 1.1 2.9 2.5
~~

2.7 2.3 8.3 5.7 8.7 2.0 4.2 10.4 2.5 10.4 0.9 1.3 3.9 1.o 1.7 13.7 1.0 1.7 2.3
~~

5.9 5.6 13.0 14.1 4.0 10.2 20.5 6.2 14.7 2.0 4.6 9.4 2.3 3.5 19.2 1.6 2.5 5.1

5.3

FIG. 5. Plot of specific growth rates at different pH values. The temperature was maintained at 55C.

formed during the last step of purification. A severe destruction (up to 75%) of muramic acid occurred during dilute acid hydrolysis. ND, Not detected.

" Preparations 10B-1 and 10B-2 correspond to two different pellets which

VOL. 36, 1986

ACIDOTHERMUS CELLULOL YTICUS GEN. NOV., SP. NOV.

441

TABLE 3. Properties of different species and strains of the genera Thermus, Thermomicrobium, and Acidothermus
Species Pigmentation Nutrition Temperature range for growth pc) pH range for growth/ DNA G +C optimum p~ (mol ratio o/o) Reference(s)

T. aquaticus YT-1 Thermus spp strain X-1 T. jlavus AT-62 T. thermophilus HB-8 T. aquaticus (many strains)

Yellow White Yellow Yelow

Some white, 47-85 ND some yellow Requires yeast extract 35-70160 T. ruber Red 8.Qb Thermomicrobium sp. (ATCC 27502) Pink Prototroph 70-75b 8.3b A. cellulolyticus l l B T White Prototroph 37-7OJ55-60 3-715.5
a

ND" Requires vitamins and amino acids Requires vitamins and amino acids ND

Prototrop h

40-79170 40-80170 40-80170 47-85165-72

6-9.5J7.5-7.8 ND (adjusted to 8) 6-917.0-7.5 5.1-9.617 . O-7.5

65.4-67.4 64 66.6 69

4 23 25 20

ND
66 64

5
11, 12 9 This work

60.7 2 0.6

ND, Not determined. Optimum.

hydrolysis steps. The enrichment in amino acids common in peptidoglycans and other cell wall components can be clearly seen. Diaminopimelic acid, glucosamine, and muramic acid were notably enriched. The amino acids which can be ascribed to the cell wall peptidoglycan on the basis of their ratio to glutamic acid (>0.5), which is a known component of peptidoglycans in most gram-negative and grampositive bacteria (19, 26), are serine, alanine, and diaminopimelic acid, and perhaps glycine and leucine in strain llBT. The amounts of alanine, diaminopimelic acid, and glucosamine seemed to be consistently higher than glutamic acid. Other dibasic amino acids which are common in cell walls of bacteria were either not detected (e.g., ornithine) or partially removed by the treatment (e.g., lysine). Galactosamine was also not detected.

400

0.5

E c .+
( u

0.4

300

:
I

0,
L

-!=
E

Base composition of DNA. G+C composition of DNA isolated from strains 10B, 11A, and llBT was determined by thermal melting studies and isopycnic banding centrifugation. No strain specificity was apparent, as the G+C content was found to be 61.1 f 0.2 mol% G+C and 60.3 5 0.3 mol% G+C from T, values of 81.8 2 0.1"C and buoyant density values of 1.7196 5 0.0003 g/ml, respectively, for the three strains of Acidothermus cellulolyticus. Induction of cellulase activity. All three strains produced active cellulase enzymes when grown on powdered or amorphous cellulose in liquid cultures. An example of an induction of cellulolytic activity secreted by the batch culture of strain l l B Tgrown aerobically in the liquid LPBM containing 0.5% powdered cellulose (Sigmacell 50) as a carbon source at 55C and pH 5.0 is shown in Fig. 6. The cellulolytic activity was secreted mainly during the late logarithmic and early stationary phases of growth, and the pattern of induction was similar for all three strains. The cellulase enzyme complexes can completely degrade microcrystalline cellulose (Sigmacell 50 and a-cellulose [Sigma]); this indicates that the full complement of cellulolytic enzymes is being secreted. The products of cellulose hydrolysis are D-glucose and D-cellobiose. No higher oligosaccharides were detected by high-performance size exclusion chromatography. A detailed study on the induction and characterization of cellulase enzymes produced by these microorganisms will appear in a future publication. DISCUSSION

:
Q,

-0 0,

0.3

p 2 !
~

2 200
m

aJ m 0

I
0
LK

0.2

3 Q,

c3

100

0.1

I 0

> != 2

1 0

20 30 TIME (hours)

40

FIG. 6. Correlation of strain l l B T cell growth and the production of cellulase activity found in portions taken from the culture supernatant and subjected to incubation with cellulose substrate at 55 and 65C.

As a result of screening, several isolates of thermophilic, cellulolytic bacteria growing in the acidic pH range were obtained. The microorganisms we isolated are a result of selective conditions applied during enrichment and isolation and do not represent major microbial populations in geothermal hot springs, The number of thermophilic microorganisms which were discarded during the isolation process far exceeds the number of isolates we obtained. It should be noted, however, that no thermophilic fungi, yeasts, or actinomycetes were observed during enrichment and isolation. The majority of bacteria we observed during the initial stages of isolation were sporulating rods resembling bacteria of the genus Bacillus. The second surprising result was a uniformity of cellulolytic strains we isolated from various springs. The isolated thermophiles are all nonpigmented, gramvariable rods. These morphological features and the heterotrophic, obligately aerobic mode of growth are similar

442

MOHAGHEGHI ET AL.

INT. J . SYST.BACTERIOL.

TABLE 4. Sensitivity of Therrnus species to antibiotics and inhibitors


Sensitivity of": Antibiotic (&ml)
T. aquaticusb T' f

l a ~ ~ ~ ~ ~ T. ~ thermophilusd ~ s o ' i d
ND~IND/ ND/ND

Actinomycin D 1 Penicillin 1
10

ND
-

Novobiocin
10

100
Chloramphenicol 10 100 Streptomycin
10
-

ND +I+ -I+ ND/+ ND/-

ND
ND

ND

100
Tetracycline 10
100

ND

+/+
-I*

Cycloserine
10 100

SDS 10 100
a

+/+ -/+

- at 144 pg/ml

For symbols, see Table 1, footnote a. ND, Not determined. Solid medium; reference 4. Reference 25. Liquid medium; reference 20.

to those reported for Thermus and Thermomicrobium strains which were isolated from similar enviroments (4, 9, 20, 22). The characteristic features of Thermus and Thermomicrobium species are summarized in Tables 3 and 4 and are compared with our isolates. Our isolates share with Thermus and Thermomicrobium strains several important features, namely morphology, sensitivity to lysozyme, and presence of catalase. They differ in other important aspects, such as the pattern of carbon sources utilized for growth, pH and temperature profiles of growth, the pattern of sensitivity to antibiotics, the G+C content of DNA, the composition of amino acids in cell walls, and the structure of the cell walls. Also, Thermus spp. are very sensitive to penicillin G, to which our strains are resistant at 100 pg/ml, and the dibasic amino acid in cell walls of Thermus spp. is ornithine (22), whereas in our strains it is diaminopimelic acid. Serine is another amino acid in cell walls of our isolates which is not present in the cell walls of Thermus spp. Finally, we could not detect D-galactosamine, which is present in cell walls of Thermus spp. The cell wall structure as revealed by transmission electron microscopy resembles the cell wall structure of gram-positive bacteria since it lacks an outer membrane, which is an important characteristic of gram-negative bacteria of the genus Thermus (3). The genus Thermomicrobium is morphologically distinctly different from our isolates since it forms small, short, pleomorphic cells rather than the long slender filaments of our organism. Therefore, me must conclude that the strains we isolated

belong to a new species in a new genus of thermophilic bacteria. Based on the experimental results presented above, we believe that a new genus should be created to accommodate this organism, for which we propose the name Acidothermus. The type species of the genus is A . cellulolyticus. The type strain is our 11B, ATCC 43068. Description of Acidothermus gen. nov. (L. adj, acidus acid; Gr. n. thermus heat; M. L. masc. n. Acidothermus acid and heat [loving]). Morphology: slender rods and long, slender filaments, 0.4 by 5 to 20 ym, which have rounded ends. Does not form endospores, flagella, or rotund bodies. Nonmotile. Gram variable, but usually stains gram negative. Thin sections show no outer cell membranes and thus are distinctly different from the typical gram-negative cell walls of Thermus and Thermomicrobium species. The principal constituents of purified cell walls are diaminopimelic acid, glucosamine, muramic acid, serine, and alanine. Colony characteristics: on LPBM mineral salts agar with 0.5 g of yeast extract per liter and 5.0 g of D-glucose or D-cellobiose per liter, creamy white, smooth, circular, entire, 1 to 3 mm. Liquid cultures: on same medium lacking agar show a moderate uniform turbidity. Some strains tend to flocculate and sediment out after 3 to 10 days of incubation. Physiology: obligate aerobes, prototrophic; grow on several carbohydrates, including D-glucose and D-cellobiose. Temperature characteristics: thermophilic, growth range 37 to 70C. pH requirements: acidophilic; pH range for growth, 3.5 to 7. DNA base ratio, 60.7 2 0.6 mol% G+C. Source: acidic hot springs at Yellowstone National Park, pH 4 to 5.5, 45 to 65C. The type species is A . cellulolyticus. Description of A . cellulolyticus sp. nov. (L. M. cellulosa cellulose, lytic Gr. v. break up; cellulolyticus cellulose dissolving). Morphology: as described for the genus. Colony characteristics: as described for the genus. Temperature characteristics: grow from 37 to 70"C, with optima between 50 and 60C. pH requirements: pH range 3.5 to 7, with an optimum near 5.0. Physiology: obligate aerobes; prototrophic, grow on Dglucose, D-cellobiose, cellulose, xylan, D-galactose, maltose, sucrose, raffinose, D-mannose, D-mannitol, or Dsorbitol as sole source of carbon and energy, with ammonium ion or amino acids as a source of nitrogen. Grows on Casamino Acids (0.1%) plus tryptone (0.1%); no growth on nutrient broth, acetate, lactate, citrate, or pectin. Citrate and acetate inhibitory at 0.01 M. Resistant t o penicillin G at 100 k.g/ml; sensitive to vancomycin and lysozyme. Catalase positive. Actively digests cellulose. The type strain is strain 11B (ATCC 43068). Description of the type strain. Strain 11B (ATCC 43068T) has all of the characteristics given for the species. In addition, it utilizes D-fructose, L-arabinose, and glycerol as a sole source of carbon, and it appears to be slightly more resistant to actinomycin D and slightly more sensitive to tetracycline than strains 10B and 11A.
ACKNOWLEDGMENTS

We thank T. J. Beveridge for performing thin-section transmission electron microscopy and M. P. Tucker for assisting with DNA thermal-melting analysis.

VOL. 36, 1986

ACIDOTHERMUS CELLULOLYTICUS GEN. NOV., SP. NOV.

443

This work was supported by the Office of Alcohol Fuels of the Department of Energy under WPA no. 349.
LITERATURE CITED 1. Bellamy, W. D. 1977. Cellulose and lignocellulose digestion by thermophilic actinomyces for single-cell protein production. Dev. Ind. Microbiol. 18:249-254. 2. Brock, T. D. 1978. Thermophilic microorganisms and life at high temperatures. Springer-Verlag, New York. 3. Brock, T. D., and M. R. Edwards. 1970. Fine structure of Thermus aquaticus, an extreme thermophile. J. Bacteriol. 104:509-517. 4. Brock, T. D., and H. Freeze. 1969. Thermus aquaticus gen. n. and sp. n., a nonsporulating extreme thermophile. J. Bacteriol. 98:289-297. 5. Brock, T. D., and I. Yoder. 1971. Thermal pollution of a small river by a large university: bacteriological studies. Proc. Indiana Acad. Sci. 80:183-187. 6. De Ley, J. 1970. Reexamination of the association between melting point, buoyant density, and chemical base composition of deoxyribonucleic acid. J. Bacteriol. 101:738-754. . , R. G. E. Murray, R. N. Costilow, E. W. Nester, 7. Gerhardt, P W. A. Wood, N. R. Krieg, and G. B. Phillips (ed.). 1981. Manual of methods for general bacteriology. American Society for Microbiology, Washington, D.C. 8. Himmel, M. E., M. D. Tucker, and K. K. Oh. 1983. Applications of high performance size exclusion chromatography in biomass processing. Biotechnol. Bioeng. Symp. 13583-595. 9. Jackson, T. J., R. F. Ramaley, and W. G. Meinschein. 1973. Therrnomicrobiurn, a new genus of extremely thermophilic bacteria. Int. J. Syst. Bacteriol. 23:28-36. 10. Loginova, L. G., L. A. Belyakova, E. P. Guzkova, I. K. Yusupova, L. G. Burdenko, and L. M. Seregina. 1983. Thermophilic, cellulose-decomposing M y c e l i o p h t h o r a thermophila. Mikrobiologiya 52:605-608. 11. Loginova, L. G., and L. A. Egorova. 1975. An obligately thermophilic bacterium Therrnus ruber from hot springs in Kamchatka. Mikrbiologiya 44:661-665. 12. Loginova, L. G., L. A. Egorova, R. S. Golovacheva, and L. M. Seregina. 1984. Thermus ruher sp. nov., nom. rev. Int. J. Syst. Bacteriol. 34:49&499. 13. MacFaddin, J. F. 1976. Biochemical tests for identification of medical bacteria. The Williams & Wilkins Co., Baltimore. 14. Mandels, M., and R. E. Andreotti. 1976. Measurement of saccharify ing cellulase. Biotechnol. Bioeng . Symp. 6: 17-34. 15. Markwell, M. A. R., S. Hass, N. Tolbert, and L. Bieber. 1981. Protein determination in membrane and lipoprotein samples: manual and automated procedures. Methods Enzymol. 72:296303. 16. Marmur, J. 1961. A procedure for the isolation of deoxyribonucleic acid from microorganisms. J. Mol. Biol. 3:208-218.

17. McCracken, L. D., and C. H. Gong. 1982. Fermentation of cellulose and hemicellulose carbohydrates by thermotolerant yeasts. Biotechnol. Bioeng. Symp. 12:91-102. 18. Miller, G. L. 1959. Use of dinitrosalicylic acid reagent for determination of reducing sugar. Anal. Chem. 31:426-428. 19. Mirelman, D. 1979. Biosynthesis and assembly of cell wall peptidoglycan, p. 115-168. In M. Inouye (ed.), Bacterial outer membranes. John Wiley & Sons, Inc., New York. 20. Oshima, T., and K. Imahori. 1974. Description of Therrnus therrnophilus (Yoshida and Oshima) comb. nov., a nonsporulating thermophilic bacterium from a Japanese thermal spa. Int. J. Syst. Bacteriol. 24:102-112. 21. Owen, R. J., and L. R. Hill. 1979. The estimation of base compositions, base pairing, and genome sizes of bacterial DNAs, p. 277-296. In F. A. Skinner and D. W. Lovelock (ed.), Identification methods for microbiologists. The Society for Applied Bacteriology Technical Series number 14. Academic Press, Inc., New York. 22. Pask-Hughes, R. D., and R. A. D. Williams. 1977. Yellowpigmented strains of Thermus spp. from Icelandic hotsprings. J. Gen. Microbiol. 102:375-383. 23. Ramaley, R. F., and J . Hixson. 1970. Isolation of a nonpigmented, thermophilic bacterium similar to Therrnus aquaticus. J. Bacteriol. 103527-528. 24. Rosenberg, S. L. 1978. Cellulose and lignocellulose degradation by thermophilic and thermotolerant fungi. Mycologia 70: 1-13. 25. Saiki, T., R. Kimura, and K. Arima. 1972. Isolation and characterization of extremely thermophilic bacteria from hpt springs. Agric. Biol. Chem. 36:2357-2366. 26. Schleifer, K. H., and 0. Kandler. 1972. Peptidoglycan types of bacterial cell walls and their taxonomic implications. Bacteriol. Rev. 36:407477. 27. Shaker, H. M., M. A. Farid, and A. I. El-Diwang. 1984. Optimization of the composition of the nutrient medium for cellulase and protein biosynthesis by thermophilic Aspergillus furnigatus, NRC 272. Enzyme Microb. Technol. 6:212-216. 28. Stutzenberger, F. J. 1979. Degradation of cellulosic substances by Thermornonospora curvata. Biotechnol. Bioeng. 2199-913. 29. Su, T. M., and J. Paulavicius. 1975. Enzymatic saccharification of cellulose by thermophilic actinomyces. Appl. Polym. Symp. 28:221-236. 30. Szybalski, W. 1968. Equilibrium sedimentation of viruses, nucleic acids and other macromolecules in density gradients. Fractions 1:l-15. 31. Tansey, M. R. 1971. Agar diffusion assay of cellulolytic ability of thermophilic fungi. Arch. Microbiol. 77:l-11. 32. Vinograd, J., and J. E. Hearst. 1962. Equilibrium sedimentation of macromolecules and viruses in a density gradient. Fortschr. Chem. Org. Naturst. 20:327422. 33. Zeikus, J. G. 1979. Thermophilic bacteria: ecology, physiology and technology. Enzyme Microb. Technel. 1:243-252.