INTERNATIONALJOURNAL OF SYSTEMATICBACTERIOLOGY, July 1990, p. 268-272 Vol. 40, No. 3 0020-7713/90/030268-05$02.OO/O Copyright 0 1990, International Union of Microbiological Societies

Clostridium aldrichii sp. nov., a Cellulolytic Mesophile Inhabiting a Wood-Fermenting Anaerobic Digester?

JUNCHANG C. YANG,l* DAVID P. CHYNOWETH,l DONNA S. WILLIAMS,, AND ANMING L12$ Bioprocess Engineering Research Laboratory, Agricultural Engineering Department,' and Microbiology and Cell Science Department,2 University of Florida, Gainesville, Florida 3261 1

An anaerobic, mesophilic, spore-forming, cellulolytic bacterium was repeatedly isolated from a wood- fermenting anaerobic digester. Cells of this organism were gram-positive rods, motile with a bundle of polar flagella, and formed subterminal oblong spores. The colonies in agar had an irregular shape with many platelike structures and were greyish white. Cellulose, xylan, and cellobiose served as substrates for growth. Acetate, propionate, butyrate, isobutyrate, isovalerate, lactate, succinate, H,, and CO, were products of cellobiose fermentation. The optimal temperature and pH for growth were 35°C and 7, respectively. The DNA composition was 40 mol% G+C. The name Clostridium aldn'chii sp. nov. is proposed. The type strain is P-1 (= OGI 112, = ATCC 49358).

Woody biomass generally was not considered to be bio- 10 ml; resazurin (0.5 mg/ml), 1 ml; L-cysteine hydrochloride, degradable under anaerobic conditions until Zeikus and 0.5 g; Na,S . 9H20, 0.2 g. The agar concentration was 1.5% Ward (22) showed that a methane fermentation occurs in the (wthol). The pH was adjusted to 7.0 with NaHCO,. The heartwood of live trees and Chynoweth and co-workers (4, medium was prepared and dispensed with continuous flush- 9) showed that the carbohydrates of poplar and other hard- ing of the gas mixture mentioned above. Medium was woods are degraded to methane by a domestic sludge dispensed into serum tubes (1) (18 by 150 mm; Bellco Glass, digester inoculum without pretreatment other than reduction Inc., catalog no. 2048-00150) or serum bottles, according to of particle size to the millimeter range. These observations need, stoppered with butyl rubber stoppers (Bellco, catalog have stimulated research to determine the organisms and no. 2048-11800), sealed with aluminum seals (Wheaton In- interactions involved with the possible goal of improving the dustries; 20-mm tear-off, catalog no. 224193), and sterilized extent and rate of these depolymerization reactions. Sleat at 121°C for 20 min. and co-workers (18, 19) have isolated two new cellulolytic, The cellulose slurry was prepared as follows: 30 g of Clostridium species from wood methane fermentation. Our Whatman no. 1 filter paper was ground with 1,000 ml of laboratory has begun to isolate predominant cellulolytic distilled water for 72 h in a 5-liter alumina grinding jar with pectinolytic, and xylanolytic bacteria from poplar-fed wood pebbles (20 to 30 mm in diameter). The cellulose slurry was digesters. The first of these, a new cellulolytic Clostridium added (50 ml/liter) to the basic medium to make the cellulose species, has been repeatedly isolated from high dilutions of medium or cellulose agar medium (CAM). The amount of the the samples over a period of 2 years and is described below. cellulose slurry added was varied according to the purpose The name Clostridium aldrichii is proposed for this organ- of the culture. The cellulose concentration in the medium ism. was typically in the range of 0.2 to 0.4% (wthol). A Whatman no. 1 filter paper strip (50 by 5 mm) was sub- MATERIALS AND METHODS merged in the 4.5-ml basal medium tube as filter paper strip Samples and treatments. Strict anaerobic methods (7) were medium (CSM) to determine the highest dilution of the used for sample collection, medium preparation, inocula- mixed population exhibiting cellulolytic activity. tion, and identification. Samples were collected from the Other substrates, including cellobiose, pectin, xylan, and effluent of a continuously stirred tank, bench-scale, meso- glucose, were added to the basal medium at a final concen- philic (35°C)) poplar-fed digester. The effluent was blended tration of 1% (wthol). The media and methods of Holdeman for less than 1 min in a homogenizer under oxygen-free et al. (6) were used for biochemical tests and fermentation N,-CO, (80:20). The treated sample was diluted in basal product analysis. medium up to lo-'' in order to exceed the highest dilution Isolation, purification, and enumeration. Serial dilutions of with cellulolytic activity. The first sample was collected 65 sample (0.5 ml) were inoculated in triplicate into CAM (4.5 days after the digester was started. ml per tube) for roll tubes and CSM for cellulolytic activity Culture media. The basal medium contained (per liter): detection. The tubes were incubated at 35°C until colonies NH,C1,2 g; KCl, 0.5 g; K,HP04,1.65 g; MgSO4 * 7H,O, 0.5 surrounded by clear zones appeared in CAM or until the g; Trypticase (BBL Microbiology Systems), 0.5 g; yeast submerged filter paper strips were digested in the CSM. extract (BBL), 0.5 g; clarified rumen fluid, 200 ml; clarified Cellulolytic colonies in the highest-dilution roll tube were woody-biomass digester sludge, 100 ml; mineral solution picked and suspended in cellulose medium, which was used "aCl(6 g), (NH4),S04 (6 g), CaCl, (0.6 g), H,O (1,000 mOl, to inoculate CAM roll tubes. The tubes were incubated at 150 ml; trace mineral solution (l),10 ml; vitamin solution (l), 35°C until cellulolytic colonies appeared again. This transfer procedure was repeated several times until a pure culture * Corresponding author. was obtained. To confirm the purity, the isolate was trans- t Florida Agricultural Experiment Station Journal Series no. ferred to cellobiose or glucose agar medium in roll tubes and R-00436. incubated at 35°C. After ascertaining that only one morpho- $ Present address: Chengdu Institute of Biology, Academia Sin- logical type of colony appeared, a single colony was picked ica, Chengdu, Sichuan, People's Republic of China. and transferred back to CAM roll tubes to check its cellu-

268 VOL.40,1990 C. ALDRZCHZZ SP. NOV., A CELLULOLYTIC MESOPHILE 269

FIG. 1. (A) Ultrathin section of P-1 cells, showing gram-positive-type cell wall structures (an enlarged cross section is shown at bottom right). (B) Ultrathin section showing the round particle between two cells. lolytic activity. The colony and cell morphologies from the a Balzers model MED 010 vacuum evaporator. Grids were secondary growth of the clear-zoned colonies in CAM were examined by transmission electron microscopy, compared and were identical to those of the previous isolate Physiological and biochemical identifications. Basal me- (8). dium with cellobiose was used for the measurements of Numbers of cellulolytic bacteria were estimated from the optimal growth temperature and pH. Cell density was deter- reciprocal of the highest dilution of the inoculated samples in mined by measurement of optical densities of cultures with a which the filter paper strip was digested in CSM and in which spectrophotometer (Perkin-Elmer model 35) at 610 nm. clear-zoned colonies appeared in CAM. Optimal temperature and pH measurements were deter- Morphological studies and microscopy. Morphological ob- mined twice, and each measurement was made in triplicate. servations of the cells from 24- to 48-h growth broth cultures Nutrient requirements were tested in the medium with and single colonies were made by both light microscopy cellobiose as the only carbon and energy source and with (Nikon Labophot equipped with phase optics, epifluores- different combinations of yeast extract, Trypticase, clarified cence, and an automatic photomicrographic Nikon FX-35A rumen fluid, digester sludge, and vitamin solution. Two- camera with UFX-I1 control) and electron microscopy (Phil- day-old cultures grown on cellobiose basal agar slants were ips EM-301 transmission electron microscope). Gram stains, washed with the salt medium to serve as inocula for these the KOH test (S), and the ~danine-4-nitroanilide(LANA) tests. The fermentation products were analyzed after incu- test (2) were used to complement the Gram reaction. Me- bation for 2 days in PY basal medium (6) with cellobiose as dium without Trypticase, yeast extract, clarified rumen fluid, the substrate. Biochemical tests were performed in PY basal and digester sludge but with the vitamin solution and cello- medium with carbohydrates, enzymes, and other substrates. biose was used to induce spore formation. A fresh culture (1 Fermentation product analyses and biochemical tests were to 2 days) in cellobiose medium was used for electron done in the laboratory of W.E. C. and L. V. H. Moore, microscopic observation of flagella. Virginia Polytechnic Institute and State University, Blacks- To prepare specimens for electron microscopy, glutaral- burg. dehyde was injected into the fresh broth culture to a final Analysis of DNA base composition. Resuspended cells from concentration of approximately 1.3%. Cells were immedi- a centrifuged pellet were treated with 5% (voVvol) Triton ately centrifuged, and the supernatant was replaced with 1% X-100, and DNA was extracted and purified by the method glutaraldehyde in 0.025 M sodium cacodylate buffer (pH of Marmur (13). DNA base compositions were determined 7.2). Cells were postfixed in buffered 1% osmium tetroxide from the buoyant density of DNA in CsCl gradients by the followed by 1% aqueous uranyl acetate, dehydrated in a method of Preston and Boone (16). DNAs from Micrococcus graded ethanol to acetone series, and embedded in Spurr lysodeikticus and Escherichia coli were used as standards. low-viscosity resin. Sections were cut with a diamond knife The mol% G+C of the DNA was calculated by the method of on an LKB Ultrotome I11 and poststained with uranyl Schildkraut et al. (17). acetate and lead citrate. For observation of flagella, cells were gently pipetted from the fresh culture and placed on RESULTS AND DISCUSSION Formvar-coated 200-mesh copper grids which had been precoated with 0.1% bacitracin to improve surface wettabil- The isolate was rod shaped, gram positive, motile, formed ity. Cells were allowed to settle for 3 to 5 min, and the excess endospores, and obligately anaerobic but was unable to medium was drawn off. The grids were washed with water to reduce sulfate to sulfide. These differential characters indi- remove soluble medium components and shadowed at a 30" cate that the isolate can be placed in the genus Clostridium angle with a carbon-platinum source in an electron gun using (3). It was unlike previously described mesophilic, cellu- 270 YANG ET AL. INT. J. SYST.BACTERIOL.

FIG. 2. Cell morphology of strain P-1 visualized by light microscopy. (A) Vegetative cells after 24 to 48 h of growth in basal medium with cellobiose as the substrate; a round particle (arrow) can be seen between two cells (B) Spores of strain P-1 (arrow).

lolytic clostridia in growth substrates, fermentative prod- ville, for his contributions to ultrastructural research on the ucts, morphology, and immunospecificity . Therefore, we strictly anaerobic bacteria) cells are easily decolorized in propose a new species, Clostridium aldrichii, as follows. Gram stains, but KOH and LANA tests are negative. Cell Clostridium aldrichii sp. nov. Clostridium aldrichii (Al. wall structure is that of gram-positive bacterium (Fig. 1A). dri’chii, N.L. gen. n., of Aldrich, named for Henry C. Cells are 0.5 to 1.0 Fm by 3 to 5 Fm, straight or slightly Aldrich, a professor of the University of Florida, Gaines- curved with rounded ends (Fig. 2). Pleomorphism is evident,

FIG. 3. Carbon-platinum-shadowed whole-cell preparation; bundled flagella can be seen at both ends of the paired cells. VOL. 40, 1990 C. ALDRZCHZZ SP. NOV., A CELLULOLYTIC MESOPHILE 271

FIG. 4. Colonial morphology of strain P-1 in CAM with clear zones around each colony. Arrow indicates the surface colony. with some cells up to 8 to 12 Fm in length; occasionally, DNA base composition. The G+C content of strain P-1 some cells are swollen at both attached ends of paired cells. DNA is 40 mol%. Cells are arranged mostly in pairs but some are single, or in The type strain of C. afdrichii is strain P-1 (= OGI 112, = short chains (Fig. 2). Round particles are sometimes seen ATCC 49358 [OGI, Collection of Methanogenic Archaeo- between two connected cells (Fig. 1B and 2A). Young cells bacteria of Oregon Graduate Institute of Science and Tech- of strain P-1 are motile, with a bundle of flagella at one end nology]). It was isolated in 1986 from a 3-month-old, meso- of the cell (Fig. 3). Oblong spores occur subterminally and philic , poplar wood-fed continuously stirred anaerobic are formed after 2 weeks in a medium containing salts, trace digester. It has been considered a predominant cellulolytic vitamins, and cellobiose as the only carbon source at 35°C. bacterium, since it inhabited (at lo6 to 10’) this digester for at Mature spores are 0.5 to 1.0 Fm by 1 to 2 pm (Fig. 2B). least 2 years. Spores are viable after being heated at 80°C for 10 min. Enumeration of strain P-1.Cellulolytic activities in CSM Morphologically identical colonies appear in CAM 4 to 6 (digestion of the immersed part of the filter paper strip; Fig. days after transfer of strain P-1 from medium to medium and 5) and CAM (colonies with clear zones; Fig. 4) correspond- within 24 h in the medium with cellobiose as the substrate. ingly appear in the same or nearly the same dilution in The surface colonies are round, smooth, entire, soft, jelly- isolations, and the highest dilutions showing positive cellu- like, semitransparent or greyish white, and 1 to 2 mm in lolytic activity fall in the range of to lo-’. diameter. Clear zones caused by cellulolytic activity around Distinguishing characteristics. Clostridium aldrichii P-1 is colonies are usually 4 to 6 mm in diameter, and the widths of considered a new species on the basis of the following the clear zones from the edge of the colonies are 2 to 3 mm. evidence. Most mesophilic, cellulolytic Cfostridium species In roll tubes, surface colonies often flow down along the agar which have been reported (10-12, 14, 15, 18, 19) grow and surface to form elongated colonies; the clear zones subse- produce acids in glucose; strain P-1 does not produce acids quently appear around the elongated growth. Subsurface from glucose. No immunological cross-reactions were ob- colonies are irregular in shape, with many platelike struc- served between the polyclonal antibody of C. afdrichii P-1 tures in different dimensions; they are greyish white and and 14 other known bacterial species (including 11 cellu- more opaque than surface colonies (Fig. 4). This distinctive colony morphology and associated clear zone facilitate rec- ognition of colonies of strain P-1. Strain P-1 is a strict anaerobe. The optimal temperature for growth is 35”C, with a range between 20 and 45°C. The optimal pH for growth is 7, and the range is 6.2 to 7.8. P-1 does not require rumen fluid, digester sludge, Trypticase, yeast extract, and added trace vitamins for growth but grows more slowly without these additions. Cellulose, xylan, and cellobiose serve as growth substrates, and esculin is hydro- lyzed. Amygdalin, arabinose, erythritol, fructose, glucose, glycogen, inositol, lactose, maltose, mannitol, mannose, melezitose, melibiose, pectin, raffinose, rhamnose, ribose, salicin, sorbitol, starch, sucrose, trehalose, and xylose are not utilized. Gelatin is not liquified, and milk is not clotted. Indole and H,S are not produced. The isolate does not reduce nitrate to nitrite, and the lecithinase A test is nega- tive. The fermentation products from cellobiose are acetic, FIG. 5. Filter paper strips in CSM digested in dilutions up to propionic, isobutyric, butyric, isovaleric, lactic, and SUC- lo-’ of the sample inocula. Higher dilutions (right three tubes) did cinic acids, hydrogen, and carbon dioxide. not digest filter paper strips. 272 YANG ET AL. INT.J. SYST.BACTERIOL.

TABLE 1. Comparison of C. aldrichii and 2. Carlone, G. M., M. J. Valadez, and M. J. Pickett. 1983. Methods C.polysaccharolyticum for distinguishing gram-positive from gram-negative bacteria. J. Clin. Microbiol. 16:1157-1159. C. polysaccharo- Characteristic" C. aldrichii 3. Cato, E. P., W. L. George, and S. M. Finegold. 1986. Genus lyticum Clostridium Prazmowski 1880, 23*=, p. 1141-1200. In P. H. A. Cell morphology Rods (0.5 to 1 pm Rods (0.8 pm by Sneath, N. S. Mair, M. E. Sharpe, and J. G. Holt (ed.), by 3 to 5 pm) 3 to 6 pm) Bergey's manual of systematic bacteriology, vol. 2. The Spores Oblong (0.5 to 1 pm OvaVspherical Williams & Wilkins Co., Baltimore. by 1 to 2 pm); (1.2 by 0.8 pm); 4. Chynoweth, D. P., and D. E. Jerger. 1985. Anaerobic digestion subterminal terminalhbter- of woody Biomass. Dev. Ind. Microbiol. 265235-246. minal 5. Halebian, S., B. Harris, S. M. Finegold, and R. D. Rolfe. 1981. Flagella Polar bundles Peritrichous Rapid method that aids in distinguishing gram-positive from Acids from glucose - - gram-negative anaerobic bacteria. J. Clin. Microbiol. 13:444- Gelatin hydrolysis - 448. Esculin hydrolysis + + 6. Holdeman, L. V., E. P. Cato, and W. E. C. Moore (ed.). 1977. Anaerobe laboratory manual, 4th ed. Virginia Polytechnic In- Acid production from: stitute and State University, Blacksburg. Cellobiose 7. Hungate, R. E. 1969. A roll tube method for cultivation of strict Xylan anaerobes, p. 117-132. In J. R. Norris and D. W. Ribbons (ed.), Starch Methods in microbiology, vol. 3B. Academic Press, Inc., New Maltose Y ork . L-Arabinose 8. Hungate, R. E. 1950. The anaerobic mesophilic cellulolytic Xylose bacteria. Bacteriol. Rev. 14:149. 9. Jerger, D. E., M. Novil, and D. P. Chynoweth. 1982. Bioconver- Fermentation products sion of woody biomass as a renewable source of energy. from cellobiose: Biotechnol. Bioeng. Symp. 12:233-248. Formate 10. Kelly, W. J., R. V. Asmundson, and D. H. Hoperon. 1987. Acetate Isolation and characterization of a strictly anaerobic, cellu- Propionate lolytic spore former: Clostridium chartatabidum sp. nov. Arch. Butyrate Microbiol. 147:169-173. Isobutyrate 11. Leschine, S. B., and E. Canale-Parola. 1983. Mesophilic cellu- Lactate lolytic clostridia from freshwater environments. Appl. Environ. Microbiol. 46:72&737. Mol% G+C 40 42 12. Madden, R. H., M. J. Bryder, and N. J. Poole. 1982. Isolation Source of isolates Wood digester Sheep rumen and characterization of an anaerobic, cellulolytic bacterium, Clostridium papyrosolvens sp. nov. Int. J. Syst. Bacteriol. a Methods used for biochemical tests and fermentative product detections 32:87-91. were the same (6) under different laboratory conditions, except DNA base 13. Marmur, J. 1961. A procedure for the isolation of deoxyribo- ratio measurements. Symbols: +, positive reaction; -, negative reaction; w, nucleic acid from micro-organisms. J. Mol. Biol. 3:208-218. weak reaction. 14. Palop, M. L., S. Valles, F. Pinaga, and A. Flors. 1989. Isolation and characterization of an anaerobic, cellulolytic bacterium, Clostridium celerecrescens sp. nov. Int. J. Syst. Bacteriol. lolytic species) by using an enzyme-linked immunosorbent 39:68-71. assay technique (an immunological study of C. aldrichii P-1 15. Petitdemange, E., F. Caillet, J. Giallo, and C. Gaudin. 1984. will be reported elsewhere). Clostridium cellulolyticum sp. nov., a cellulolytic, mesophilic Only one reported cellulolytic, mesophilic Clostridiurn species from decayed grass. Int. J. Syst. Bacteriol. 34:155-159. species, Clostridiurn polysaccharolyticurn (20, 21), does not 16. Preston, J. F., and D. R. Boone. 1972. Analytical determination produce acids from glucose. The important characteristics of of the buoyant density of DNA in acrylamide gels after prepar- the two species are compared in Table 1. Although the DNA ative CsCl gradient centrifugation. FEBS Lett. 37:321-324. base ratios are similar, the arrangement of the flagella, acid 17. Schildkraut, C. L., J. Marmur, and P. Doty. 1962. Determina- tion of the base composition of deoxyribonucleic acid from its production from different carbohydrates, and the differences buoyant density in CsCl. J. Mol. Biol. 4:430-443. in fermentation products from cellobiose are the primary 18. Sleat, R., and R. A. Mah. 1985. Clostridium populeti sp. nov., a bases for separating them into different species. cellulolytic species from a woody-biomass digester. Int. J. Syst. Bacteriol. 35160-163. ACKNOWLEDGMENTS 19. Sleat, R., R. A. Mah, and R. Robinson. 1984. Isolation and characterization of an anaerobic, cellulolytic Bacterium, Clos- We gratefully acknowledge L. V. H. and W. E. C. Moore of the tridium cellulovorans sp. nov. Appl. Environ. Microbiol. 48: Department of Anaerobic Microbiology, Virginia Polytechnic Insti- 88-93. tute and State University, Blacksburg, for biochemical tests and 20. van Gylswyk, N. 0. 1980. Fusobacterium polysaccharolyticum fermentative product analyses. sp. nov., a gram negative rod from the rumen that produces This work was supported by the Gas Research Institute, New butyrate and ferments cellulose and starch. J. Gen. Microbiol. York State Energy Development Authority, and New York Gas 116:157-163. Group under contract no. 5086-226-1199. 21. van Gylswyk, N. O., E. J. Morris, and H. J. Els. 1980. Sporu- lation and cell wall structure of Clostridium polysaccharolyti- LITERATURE CITED cum comb. nov. (formerly Fusobacterium polysaccharolyti- 1. Balch, W. E., G. E. Fox, L. J. Magrum, C. R. Woese, and R. S. cum). J. Gen. Microbiol. 121:491493. Wolfe. 1979. Methanogens: reevaluation of a unique biological 22. Zeikus, J. G., and J. C. Ward. 1974. Methane formation in living group. Microbiol. Rev. 43:260-296. trees: a microbial origin. Science 184:1181-1183.