Vol. 54, No. 8APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Aug. 1988, p. 1917-19220099-2240/88/081917-06$02.00/0Copyright © 1988, American Society for Microbiology

Esterase Activities in Butyrivibrio fibrisolvens StrainsROBERT B. HESPELL* AND P. J. O'BRYAN-SHAH

Northern Regional Research Center, Agricultural Research Service, U.S. Department of Agriculture,1815 North University Street, Peoria, Illinois 61604

Received 11 January 1988/Accepted 5 May 1988

Thirty strains of Butyrivibrio fibrisolvens isolated in diverse geographical locations were examined foresterase activity by using naphthyl esters of acetate, butyrate, caprylate, laurate, and palmitate. All strainspossessed some esterase activity, and high levels of activity were observed with strains 49, Hl7c, S2, AcTF2,and LM8/1B. Esterase activity also was detected in other ruminal bacteria (Bacteroides ruminicola, Seleno-monas ruminantium, Ruminobacter amylophilus, and Streptococcus bovis). For all B. fibrisolvens strains tested,naphthyl fatty acid esterase activity paralleled culture growth and was predominantly cell associated. Withstrains 49, CF4c, and S2, the activity was retained by protoplasts made from whole cells. Esterase activity wasdetected with all strains when grown on glucose, and some strains showed higher activity levels when grown onother substrates (larchwood xylan or citrus pectin). When nitrophenyl esters of fatty acids were used tomeasure esterase activity, generally four- to sevenfold-higher activity levels were detected, and with a numberof strains substantial levels were found in the culture fluid. Cultures of these strains (H17c, NOR37, Dl, andD30g) contained xylanase and acetyl xylan esterase activities, neither of which was associated to any greatextent with the cells. Acetyl xylan esterase has not been previously detected in ruminal bacteria and may beimportant to overall digestion of forage by these organisms.

Butyrivibriofibrisolvens is a bacterial species ubiquitouslyfound in the gastrointestinal tracts of mammals. The specieswas initially described by using strains isolated from rumencontents (5) and was subsequently found also in fecal mate-rial from rabbits, horses, pigs, and humans (4, 17). Strains ofthis species characteristically are curved rods that arestrictly anaerobic and usually are motile by one polar orsubpolar flagellum. They stain gram negatively, but have agram-positive cell wall structure, and they produce butyricacid as a major fermentation product. All strains are saccha-rolytic, and individual strains have been isolated that canferment cellulose, xylan, starch, and other plant cell poly-saccharides. In addition, over half the strains produce anextracellular protease and utilize proteins and/or peptides asnitrogen sources for growth (8).

Plant triglycerides and other lipid like materials are notextensively degraded in the rumen, but are subjected topartial hydrolysis. Strains of B. fibrisolvens have been iso-lated that are capable of hydrolyzing saponins (11), tributy-rin (15), or galactolipids (12). Strain 53 was shown to have anesterase activity toward hydrolysis of esters of short-chainfatty acids (15). Strain S2 was found to possess phospholi-pase and galactolipase activities (13). However, most strainshave not been examined for esterase activities.A number of previous studies have clearly shown that B.

fibrisolvens strains are quite variable in phenotypic traits.Recent studies from our laboratory have indicated that B.fibrisolvens is a collection of genetically diverse strains thatdiffer in the G+C content of their DNAs, in the hybridizationlevels between DNAs (B. Mannarelli and R. B. Hespell,Abstr. Annu. Meet. Am. Soc. Microbiol. 1987, 1-139, p.195), and in the composition of the extracellular polysaccha-rides produced by these strains (19). These data suggest thatthere may be several different species of this bacterium. This

* Corresponding author.

study was conducted to examine various Butyrivibrio strainsisolated from diverse sources and to hence determinewhether esterase activity is a common property of thesestrains and to characterize the activity in terms of substratespecificity and other properties.

MATERIALS AND METHODS

Abbreviations. The following abbreviations are used in thetext: NX ester, naphthyl ester; NPX ester, p-nitrophenylester; X is acetate (A), butyrate (B), caprylate (C), laurate(L), or palmitate (P).

Bacterial strains and growth conditions. All strains usedwere previously characterized and were from our culturecollection or from M. P. Bryant (Department of AnimalSciences, University of Illinois, Urbana). The originalsources of the Butyrivibrio strains have been indicated in aprevious publication (19). Unless indicated otherwise, allstrains were grown on RGM medium containing glucoseunder an atmosphere of 20% carbon dioxide-80% nitrogen at37°C (14). All cultures were grown to mid- to late logarithmicgrowth phase prior to experimental use. Growth was mea-sured by optical density values at 660 nm. Cell suspensionswere made from cell pellets collected from cultures bycentrifugation (16,000 x g, 20 min, 15°C) and suspended inoxygen-free phosphate buffer (40 mM [pH 7.0]).

Assays of enzymatic activities. Esterase activity was mea-sured by two separate assay methods involving the hydrol-ysis of either NX esters or NPX esters of fatty acids, adaptedfrom the procedures of Lanz and Williams (15) and Westlakeet al. (20), respectively. The NX ester assay mixture con-tained 0.5 ml of 0.2 M sodium phosphate buffer (pH 7.0),1.75 ml of distilled water, 0.15 ml of 5.0 mM naphthyl fattyacid ester (dissolved in methanol), and 0.1 ml of enzymesource. The assay was initiated by adding the source ofenzymatic activity and then incubating the mixture at 37°C

1917

on October 24, 2020 by guest

http://aem.asm

.org/D

ownloaded from

1918 HESPELL AND O'BRYAN-SHAH

TABLE 1. Esterase activity of ruminal microorganisms

Activity of following esterase':Bacterial species

NA NB NC NL NP NPA NPB NPL NPP

B. fibrisolvens 49 15.0 9.7 6.8 2.2 1.6 78.0 135.0 _b 21.0S. bovis JB1 6.7 10.8 1.5 - - 165.0 306.0 - 66.0R. amylophilus H18 8.5 12.0 1.6 - - 132.0 228.0 - -

B. ruminicola 23 1.7 - - - - 51.0 288.0 - -

B. ruminicola B14 8.1 0.7 2.0 0.0 0.0 69.0 231.0S. ruminantium HD4 10.1 18.7 11.2 1.1 - 219.0 225.0 - -

S. ruminantium D - - - - - 246.0 240.0 - -

a Activity is expressed as nanomoles of NX or NPX hydrolyzed per minute per culture optical density at 660 nm.b -, Not measured.

for 10 min. After termination by addition of 0.5 ml of FastGarnet GBC (5 mg/ml in 10% [wt/vol] sodium dodecylsulfate) and incubation at room temperature for 15 min, theoptical density at 560 nm was measured. The standard curvewas prepared by using a-naphthol. The NPX ester assaymixture contained 2.0 ml of 0.1 M sodium phosphate buffer(pH 7.0) in 1 mM dithiothreitol), 0.01 ml of 0.25 M p-nitrophenyl fatty acid ester (dissolved in dimethylforma-mide), and 0.5 ml of enzyme source. The assay was initiatedby adding the enzyme source and then incubating the mix-ture at 37°C for 15 min. The optical density at 405 nm wasmeasured immediately. The standard curve was prepared byusing p-nitrophenyl phosphate. With both assay proceduresthe esterase activity was linear over the time of incubationwhen cell suspensions with optical density values at 660 nmof 0.1 to 0.50 were used. Assays were run in the presence ofair, since no changes in activity levels were noted by when anitrogen atmosphere was used with about six differentstrains. In most cases activities were normalized as nano-moles hydrolyzed per minute per culture optical density at650 nm to correct for various culture cell densities and forthe inability to accurately measure the trace protein levels (5to 15 ,ug/ml) in the culture fluids.

Acetyl xylan esterase activity was measured by the re-lease of acetate from acetyl birchwood xylan (see below) asthe substrate. The assay mixture contained 0.125 ml of 100mM phosphate buffer (pH 6.8), 0.125 ml of 10 mM dithio-threitol, 0.50 ml of acetyl birchwood xylan (25 mg/ml), and0.50 ml of sample. The assay was initiated by addition of thesample. After incubation at 37°C for 30 or 60 min, a 0.20-mlportion was removed and added to 0.60 ml of ice-coldethanol; the mixture was incubated on ice for 30 min toprecipitate the xylan, which was removed by centrifugation(10,000 x g at 4°C for 10 min). The supernatant fluid wasremoved, frozen, and lyophilized to dryness. The residuewas suspended in 0.10 ml of water, acidified by addition of0.90 ml of 10% (vol/vol) phosphoric acid, and immediatelyassayed for acetate by gas-liquid chromatography (seebelow).The xylanase activities of various samples were measured

by using the orcinol method to detect the release of pentosesfrom larchwood xylan as described previously (14). Resultsof all assays of enzymatic activities are the average ofduplicate or triplicate experiments.Formation of protoplasts. Cell pellets were suspended to 1/

10 of the culture volume in wash buffer [10 mM piperazine-N,N'-bis(2-ethanesulfonic acid) (PIPES; pH 6.8), 25 mMsodium chloride, 10 mM magnesium chloride] containing10% (wt/vol) sucrose after being washed once by centrifu-gation in wash buffer. Lysozyme was added to a finalconcentration of 1 mg/ml, and the suspension was incubated

at 37°C until about 90% of the cells formed protoplasts (30 to45 min). When necessary, protoplasts were centrifuged(10,000 x g, 5 min, 4°C) and the resultant pellet wassuspended in sucrose-wash buffer.Xylan sources. Larchwood and oatspelt xylans were pur-

chased commercially (Sigma Chemical Co., St. Louis, Mo.).Birchwood xylan was prepared from birchwood chips (ob-tained from T. Jefferies, U.S. Department of AgricultureForest Products Laboratory, Madison, Wis.) by autoclavingequal volumes of wood and water twice for 10 min. After thewood had been filtered out, the aqueous extract was lyoph-ilized to dryness, suspended in a minimal amount of water,and exhaustively dialyzed against water in tubing of about6,000- to 8,000-molecular-weight cutoff size. The sugar-freexylan preparation was then lyophilized to dryness, yieldingabout 0.1% of the initial wood weight. A portion of thebirchwood was digested with 0.1 M NaOH for 48 h at 37°C),and, from measurements of the acetate released, the xylanwas found to contain about 11% acetate by weight. Theacetate was measured by gas-liquid chromatography with aHewlett-Packard 5890A gas chromatograph equipped with a30-m megabore DBWax column. The column was run iso-thermally at 125°C with argon as the carrier gas (flow rate, 15ml/h). The injector and flame ionization detector were keptat 200 and 220°C, respectively.Other reagents. All other chemicals or biochemicals were

reagent grade or better and were purchased from Sigma.

RESULTS

A survey of major ruminal bacteria indicated that a varietyof species possessed some esterase activity, including Buty-rivibrio, Bacteroides, Selenomonas, and Streptococcusstrains (Table 1). All tested strains were capable of hydro-lyzing NA, NPA, NB, and NPB esters, and a few speciescould hydrolyze esters with longer fatty acid chains. Gener-ally, much higher activities were observed with NPX estersas substrates.The changes in total culture esterase activity were moni-

tored as a function of the culture growth stage. With B.fibrisolvens 49 and CF4c the esterase activity increased withcell growth until the stationary growth phase was reached,after which the activity remained fairly stable (Table 2). Theproduction of extracellular proteases is constitutive (8) andfollowed the same pattern as seen for the esterase activity.Both the esterase and protease activities in the cultureremained at stable levels after growth ceased. Similar effectsof growth on esterase activity were found for B. fibrisolvens12, S2, IL631, and CE51, as measured by using NA, NPA,NB, NPB, NL, and NPL ester.

APPL. ENVIRON. MICROBIOL.

on October 24, 2020 by guest

http://aem.asm

.org/D

ownloaded from

ESTERASES OF BUTYRIVIBRIO FIBRISOLVENS 1919

TABLE 2. Production of esterase or protease withgrowth of B. fibrisolvens 49 or CF4c

Strain 49 Strain CF4cTime(h) Growtha Esterase Growth Protease Esterase

productionb productionc production

0 0.01 0.00 0.00 0.00 0.002 0.10 0.90 0.07 0.02 9.183 0.24 9.72 0.18 0.39 15.664 0.44 23.67 0.34 0.04 26.825 0.79 27.72 0.53 0.09 35.466 1.03 29.34 0.69 0.10 38.978 1.08 44.64 0.62 0.10 40.8624 0.76 46.98 0.58 0.10 40.95

a Growth is expressed as the optical density at 660 nm.b Esterase production is expressed as nanomoles of nitrophenylbutyrate

hydrolyzed per minute per culture optical density at 660 nm.c Protease production is expressed as milligrams of azocasein hydrolyzed

per minute per milliliter of culture.

The effects of additions of different metal ions and metalchelators on total culture esterase activity as measured byNA ester hydrolysis were examined by using B. fibrisolvensCF4c, E9a, and 49. Addition of calcium chloride (0.1 to 1.0mM) decreased activity by 10 to 30%. The addition ofmagnesium chloride or manganese chloride at similar levelsgenerally had no effect. However, when sodium EDTA wasadded at 0.1 or 10.0 mM, a 25 to 50% or a 40 to 80% loss ofactivity, repectively, was observed. When the pH of theassay mixture was varied over the normal range found in therumen, pH 5.8 to 7.0, the esterase activity varied less than10%. With cultures of strains CF4, S2, or 49, preincubationof the culture sample at 42 or 60°C for 30 min resulted in a 10or 75% decline in esterase activity, repectively. Changingthe assay temperature to either 22 or 60°C also caused abouta 30 to 35% decline in activity. In addition, no changes inactivity were noted when assays with the above strains wererun with oxygen-free asssay components and a nitrogenatmosphere.

All previous esterase activities were measured with cul-ture samples, and the results reflected total culture activities.More detailed studies on the distribution of esterase activitybetween cells and culture fluids were done with B. fibrisol-vens 49, CF4c, and S2 (Table 3). The known lipolytic strainS2 produced the most activity. With all three strains, most ofthe NB esterase activity was cell associated, and littleactivity was detectable in the culture fluids. When wholecells were converted to protoplasts with lysozyme plussucrose as an osmotic stabilizer, esterase activity was re-tained by the protoplast suspension. When the protoplastswere gently centrifuged and resuspended in fresh sucrosebuffer, the washed protoplasts still retained much esterase-activity, whereas the protoplast supernatant fluid containedno detectable activity (Table 3). However, an unexplainablesubstantial loss of overall activity occurred with this lastprocedure.The majority of B. fibrisolvens strains are xylanolytic and

often can grow on a variety of xylans having different sugarcompositions (14). When several strains that grow well onxylans were grown on larchwood xylan, the amount ofesterase activity produced increased significantly withstrains D16f and D30g, but not with strains 787 or X6C61(Table 4). When the strains were grown on citrus pectin,which contains ester-linked methyl groups, only strainX6C61 showed increased activity levels. In almost all in-stances, the majority of the NA or NB esterase activities

were cell associated, regardless of the carbon source usedfor growth. In addition, the resuspended cells often showedactivity levels equal to or greater than those measured withthe total culture.A survey was conducted of a large number of B. fibrisol-

vens strains for esterase activity by using napthyl esters ofshort- to long-chain fatty acids (Table 5). Esterase activityvaried over a range of about 10-fold between strains, asmeasured by NA ester hydrolysis. All strains were able tohydrolyze esters of fatty acids from C2 (acetate) to C16(palmitate), but there was a general decrease in activity withincreased carbon chain length of the substrate used (data forNC and NP esters not shown). In general, about 70% of thetotal culture NX esterase activities were associated with thecells. As noted above (Table 1), esterase activity could alsobe measured by hydrolysis of nitrophenyl esters of fattyacids. When the same cultures were assayed in this manner,some different patterns of esterase activity were observed(Table 6). For many strains, esterase activities were abouttwo- to sevenfold higher than those observed with NXesters, but the same general decrease in activity with longer-chain fatty acids was seen. The addition of 20% methanol(used in the NX ester assays) to the NPX ester assaymixtures did not result in decreased activities. In addition,substantial amounts of NPX esterase activity were detectedin the culture fluids, and this observation was particularlyevident for strains CF3a, CF3b, NOR37, E21c, CE51, andCE52 (Table 6 versus Table 5).The increased esterase activities measured with NPX

esters suggested that the culture fluid contained some ester-ase activity that was not adequately measured by use ofNXesters as substrates. This possibility was explored furtherwith B. fibrisolvens 49, H17c, and NOR37 grown on RGMmedium containing glucose as the energy substrate (Table 7).All three cultures showed esterase activity as noted above.All three cultures possessed xylanase activity toward larch-wood and birchwood xylan. Acetyl xylan esterase activity,as measured by release of acetate from birchwood acetylxylan, was detected with strains H17c and NOR37. Verylittle or none of this activity was detected in the cells of thesestrains, and considerable activity was found in the culturefluid. Similar results were also found for strains Dl and D30g(data not shown). Only trace levels of acetyl xylan esteraseactivity were found with strain 49.

DISCUSSION

These data show that a number of major ruminal bacteriahave some esterase activity (Table 1) and that all B. fibrisol-vens strains examined displayed high esterase activities asmeasured by hydrolysis of NX or NPX esters of fatty acids

TABLE 3. Distribution of esterase activity in B. fibrisolvens49, CF4c, and S2

Esterase activity" in strain:Fraction

49 CF4c S2

Total culture 31.50 36.33 149.70Culture fluid 2.52 1.23 4.50Cells 15.84 29.40 89.91Protoplast suspension 29.70 22.05 87.69Protoplast fluid 0.00 0.00 0.00Resuspended protoplasts 5.82 19.26 36.63

aExpressed as nanomoles of NB hydrolyzed per minute per culture opticaldensity at 660 nm.

VOL. 54, 1988

on October 24, 2020 by guest

http://aem.asm

.org/D

ownloaded from

1920 HESPELL AND O'BRYAN-SHAH

TABLE 4. Effect of carbon source on esterase of B. fibrisolvens strains

Esterase activitya in strain:

Growth substrate Assay D16f D30g 787 X6C61substrate ___________ __________ ___________ ___________

T F C T F C T F C T F C

Glucose NA 39 3 27 6 1 6 92 25 76 45 0 49NB 12 0 11 8 0 32 20 4 10 10 0 18

Larchwood xylan NA 189 29 306 82 27 71 103 34 49 44 18 70NB 35 14 25 23 5 25 33 10 24 28 5 13

Citrus pectin NA 33 89 115 92 29 66 81 31 76 175 35 147NB 18 3 28 44 16 0 46 17 35 13 20 0

a Activity is expressed as nanomoles ofNA or NB hydrolyzed per minute per culture optical density at 660 nm. T, F, C, Total culture, culture fluid, and cells,respectively.

(Tables 5 and 6). In almost all cases the measured activitydeclined with increasing carbon chain length of the fattyacid used as the substrate, and with some strains thedecreases were directly proportional to the chain length.These decreases in activity could result from lower substrateturnovers with the enzyme per se, from decreased solubili-ties of longer-chain fatty acid substrates, or from botheffects. The decreased activities were observed with bothNX or NPX esters, as well as with various ruminal bacteria(Table 1).With B. fibrisolvens strains, esterase activities occurred in

proportion to growth (Table 2) and paralleled extracellularprotease formation. With most strains growth on variouscarbon sources did not dramatically alter the levels ofesterase activity, although with some strains 3- to 10-foldincreases were noted (Table 4). These data are consistentwith at least NX esterase activity being constitutively pro-duced by B. fibrisolvens strains.The NX esterase activities for most strains were found to

be associated predominantly with the cells and not with the

TABLE 5. Esterase activity (NX esters) ofB. fibrisolvens strains

NA esterase NB esterase NL esterase

Strain activitya activity activity

T F C T F C T F C

H17c 60 15 43 31 9 32 _b _ _12 5 1 4 8 3 6 2 1 2Dl 26 3 16 16 0 9 3 0 2AcTF2 30 0 20 11 9 8 3 1 4CF3 22 2 18 12 1 8 4 0 3CF3a 4 1 4 11 1 5 1 0 0CF3c 13 2 10 4 0 12 10 0 7CF4c 23 2 20 12 3 12 6 0 2NOR37 18 0 22 6 2 7 6 0 2IL631 6 0 5 6 0 5 5 0 4E9a 14 6 13 11 6 14 5 1 10E21c 21 0 16 22 0 23 4 0 2S2 88 88 2 68 74 5 46 19 0C14 28 3 22 14 1 11 13 0 2LM8/1B 30 5 35 37 10 26 9 0 3H4a 23 3 25 12 17 14 23 0 23CE51 13 0 9 10 0 7 7 0 4CE52 11 0 6 8 0 4 6 0 5

a Activity is expressed as nanomoles hydrolyzed per minute per milliliterper culture optical density at 660 nm. T, F, C, Total culture, culture fluid, andcells, respectively.b-, Not measured.

culture fluid (Tables 3 to 5). Although low levels werefrequently detected in the culture fluid, these activities mayhave arisen from a small amount of cell lysis. When wholecells were converted to protoplasts, NX esterase activitywas retained by the protoplasts, possibly associated with thecell membrane. The observation that both short- and long-chain fatty acid NX esters were hydrolyzed suggests that thecellular NX esterase may reflect a lipase(s). These findingsare in agreement with the properties and location of phos-pholipase and galactolipase activities studied in detail withB. fibrisolvens S2 (13) and with the hydrolysis of triglycer-ides by B. fibrisolvens 53 (15). Although these previousstudies on strain 53 showed no effects of metal ions onesterase activity, we observed EDTA inhibition of activitywith a number of other strains, suggesting that some metalions may be needed for optimal activity.

TABLE 6. Esterase activity (NPX esters) ofB. fibrisolvens strains

NPA esterase NPB esterase NPP esteraseStrain activitya activity activity

T F C T F C T F C

H17c 72 74 15 60 37 19 _b _ _12 23 0 3 107 75 77 16 0 7Dl 60 18 37 175 59 162 12 0 11AcTF2 17 2 7 98 60 56 12 2 11CF2d 4 0 0 61 34 42 40 8 35CF3 134 81 129 109 65 69 39 11 32CF3a 260 264 1 98 72 2 44 24 20CF3c 208 238 12 150 133 31 17 15 0CF4c 117 41 56 21 0 4 3 0 0NOR37 33 15 17 72 44 32 11 0 17ARD22a 390 340 48 153 110 8 0 0 0E9a 31 21 23 48 46 41 18 0 11E21c 56 45 34 86 51 34 47 8 38S2 272 44 160 88 34 9 0 0 0C14 242 61 0 144 0 72 19 20 25R28 210 194 10 100 53 16 4 0 0LM8/1B 54 44 53 87 41 54 42 4 40H4a 27 3 5 120 82 79 137 2 11HlOb 31 35 2 80 70 16 31 31 5B385-1 123 70 81 80 33 64 40 5 41CE51 84 35 58 55 3 37 26 0 13CE52 81 43 35 43 2 27 16 0 4

a Activity is expressed in nanomoles per minute per milliliter per cultureoptical density at 660 nm. T, F, C, Total culture, culture fluid, and cells,respectively.

b -, Not measured.

APPL. ENVIRON. MICROBIOL.

on October 24, 2020 by guest

http://aem.asm

.org/D

ownloaded from

ESTERASES OF BUTYRIVIBRIO FIBRISOLVENS 1921

TABLE 7. Esterase, xylanase, and acetyl xylan esterase activities of B. fibrisolvens strains

Esterase activityb on: Xylanase activityc on: Acetyl xylan esteraseStrain Fractiona activityd on birch

NPA NPB NA NB Larch xylan Birch xylan acetyl xylan

49 T 59 66 33 21 2.92 0.83 0.07F 42 13 10 4 1.79 1.35 0.00C 18 29 25 10 0.34 0.11 0.08

H17c T 260 115 60 31 6.42 4.30 4.01F 200 119 43 9 11.04 3.12 2.04C 38 62 15 11 1.27 0.00 0.16

NOR37 T 148 84 26 22 1.19 2.35 1.54F 131 43 25 7 1.68 1.76 1.28C 51 27 26 12 0.12 0.00 0.00

a T, F, C, Total culture, culture fluid, and cells, respectively.b Esterase activity is expressed as nanomoles per minute per culture optical density at 660 nm.c Xylanase activity is expressed as micromoles per minute per culture optical density at 660 nm.d Acetyl xylan esterase is expressed as micromoles of acetate formed per minute per culture optical density at 660 nm.

Substantial levels of NPX esterase activities were foundboth in the culture fluid and with the cells (Table 6). Theculture fluid activity was quite low when using NX esters formany strains (Table 6 versus 5). These results suggest that asecond esterase activity was present that was associatedmainly with the culture fluid. This activity may, in part, bedue to the extracellular protease produced by many B.fibrisolvens strains. However, attempts to use syntheticsubstrates involving NP or NPX esters of amino acids tomeasure this protease activity have been unsuccessful (8). Asecond possibility is that the culture fluid esterase activity isassociated with the presence of an acetyl xylan esterase.Two strains, H17c and NOR37, which possessed high extra-cellular esterase activities were also found to produce anextracellular acetyl xylan esterase activity in addition toxylanase activity (Table 7). Similar results were found forstrains Dl and D30g.Recently the presence of acetyl xylan esterase activities in

fungi (2) and yeasts (16) has been reported. The existence ofenzymatic activity that releases acetate from acetylatedxylans, i.e., acetyl xylan esterase, in B. fibrisolvens is notentirely surprising, since most strains are highly xylanolytic.The cell walls of higher plants have been shown to contain0-acetyl groups with the acetyl content equivalent to about2% of the plant dry weight (1). Measurements of ruminaldigestion of perennial or Italian rye grass cell walls haveshown that acetylated xylose residues have low digestibilityand may have an inhibitory effect on the overall digestion ofxylans (18). However, the digestibilities of these cell wallpreparations were markedly increased when the acetylgroups were removed by anhydrous sodium ethoxide treat-ment. Recent studies have shown that a variety of othergrasses and straws contain hemicelluloses having alkali-labile substituents on the xylose (0-2 and/or 0-3) andarabinose (0-5) residues, of which acetyl groups couldaccount for 50 to 70% of the substitutions (6). Although theprecise roles of the acetyl xylan esterase and other esteraseactivities of B. fibrisolvens strains in the degradation of plantcell walls, hemicelluloses, and xylans are unknown at thistime, it would seem that these activities complement thexylanase activity to enhance overall degradation, as sug-gested by some preliminary studies with fungal systems (3).We are currently exploring this possibility by purifying theseenzymes from selected B. fibrisolvens strains.

LITERATURE CITED

1. Bacon, J. S. D., A. H. Gordon, and E. J. Morris. 1975. Acetylgroups in cell-wall preparations from higher plants. Biochem. J.149:485-487.

2. Beily, P. 1985. Microbial xylanolytic systems. Trends Biotech-nol. 3:286-290.

3. Biely, P., C. R. MacKenzie, J. Puls, and H. Schneider. 1986.Cooperativity of esterases and xylanases in the enzymaticdegradation of acetyl xylan. Bio/Technology 4:731-733.

4. Brown, D. W., and W. E. C. Moore. 1960. Distribution ofButyrivibriofibrisolvens in nature. J. Dairy Sci. 43:1570-1574.

5. Bryant, M. P., and N. Small. 1956. The anaerobic monotrichousbutyric acid-producing curved rod-shaped bacteria in the ru-men. J. Bacteriol. 72:16-21.

6. Chesson, A., A. H. Gordon, and J. A. Lomax. 1983. Substituentgroups linked by alkali-labile bonds to arabinose and xyloseresidues of legume, grass, and cereal straw cell walls and theirfate during digestion by rumen microorganisms. J. Sci. FoodAgric. 34:1330-1340.

7. Coen, J. A., and B. A. Dehority. 1970. Degradation and utiliza-tion of hemicellulose from intact forages by pure cultures ofrumen bacteria. Appl. Microbiol. 20:362-368.

8. Cotta, M. A., and R. B. Hespell. 1986. Proteolytic activity of theruminal bacterium Butyrivibrio fibrisolvens. Appl. Environ.Microbiol. 52:51-58.

9. Dehority, B. A. 1965. Degradation and utilization of isolatedhemicelluloses by pure cultures of cellulolytic rumen bacteria.J. Bacteriol. 89:1515-1520.

10. Dehority, B. A. 1967. Rate of isolated hemicellulose degradationand utilization by pure cultures of rumen bacteria. Appl. Micro-biol. 15:987-993.

11. Gutierrez, J., R. E. Davis, and I. L. Lindahl. 1959. Character-istics of saponin-utilizing bacteria from the rumen of cattle.Appl. Microbiol. 5:304-308.

12. Hazlewood, G., and R. M. C. Dawson. 1979. Characteristics of alipolytic and fatty acid-requiring Butyrivibrio sp. isolated fromthe ovine rumen. J. Gen. Microbiol. 112:15-27.

13. Hazlewood, G. P., K. Y. Cho, R. M. C. Dawson, and E. A.Munn. 1983. Subcellular fractionation of the gram-negativerumen bacterium, Butyrivibrio S2, by protoplast formation, andlocalization of lipolytic enzymes in the plasma membrane. J.Appl. Bacteriol. 55:337-347.

14. Hespell, R. B., R. Wolf, and R. J. Bothast. 1987. Fermentation ofxylans by Butyrivibrio fibrisolvens and other ruminal bacterialspecies. Appl. Environ. Microbiol. 53:2849-2853.

15. Lanz, W. W., and P. P. Williams. 1973. Characterization ofesterases produced by a ruminal bacterium identified as Butyri-vibriofibrisolvens. J. Bacteriol. 113:1170-1176.

VOL. 54, 1988

on October 24, 2020 by guest

http://aem.asm

.org/D

ownloaded from

1922 HESPELL AND O'BRYAN-SHAH

16. Lee, H., R. J. B. To, R. K. Latta, P. Biely, and H. Schneider.1987. Some properties of extracellular acetylxylan esteraseproduced by the yeast Rhodotorula mucilaginosa. Appl. Envi-ron. Microbiol. 53:2831-2834.

17. Moore, W. E. C., and L. V. Holdeman. 1974. Human fecal flora:the normal flora of 20 Japanese Hawaiians. Appl. Microbiol. 27:961-979.

18. Morris, E. J., and J. S. D. Bacon. 1977. The fate of acetyl groups

APPL. ENVIRON. MICROBIOL.

and sugar components during the digestion of grass cell walls insheep. J. Agric. Sci. 89:327-340.

19. Stack, R. J. 1988. Neutral sugar composition of extracellularpolysaccharides produced by strains of Butyrivibrio fibrisol-vens. Appl. Environ. Microbiol. 54:878-883.

20. Westlake, K., R. I. Mackie, and M. F. Dutton. 1987. T-2 toxinmetabolism by ruminal bacteria and its effect on their growth.Appl. Environ. Microbiol. 53:587-592.

on October 24, 2020 by guest

http://aem.asm

.org/D

ownloaded from