APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Nov. 1978, P. 643-6490099-2240/78/0036-0643$02.00/0Copyright © 1978 American Society for Microbiology

Vol. 36, No. 5

Printed in U.S.A.

Factors Affecting the Activity of Cellulases Isolated from theRumen Digesta of Sheep

G. L. FRANCIS,* J. M. GAWTHORNE,t AND G. B. STORER

Division ofHuman Nutrition, Commonwealth Scientific and Industrial Research Organization, Adelaide,South Australia 5000

Received for publication 14 August 1978

Sodium phosphate buffer was used to extract cellulases from the plant solidsfraction ofrumen contents. The mixed cellulase preparation had maximal activityat pH 6.9 and 490C. The Vm. and the apparent Km for wheaten hay cellulosewere 19.8 glucose units/min and 6.35 mg/ml, respectively, and for microcrystallinecellulose (Sigmacell) at the same enzyme concentration, they were 33 glucoseunits/min and 27.5 mg/ml, respectively. For these assays a glucose unit was

defined as nanomoles of glucose plus twice the nanomoles of cellobiose. Consid-eration of thermodynamic and kinetic data suggested that the hydrolysis of a

relatively labile arabino-xylan comprising 3% of the wheaten hay cellulose was

dependent on prior removal of the protecting fl-1,4-glucose chains at the outersurface of the cellulose preparation. Sequential removal of structural polysac-charides from the plant cell wall rendered the latter more susceptible to cellulaseactivity. Cellulase activity was stimulated by increasing the concentration ofphosphate from 5 to 50 mM. The stimulation was magnified in the presence ofcell-free rumen fluid. Cellulase activity was not stimulated by calcium, magne-

sium, iron, zinc, manganese, copper, or cobalt ions and was unaffected by thechelators ethylenediaminetetraacetic acid and ethyleneglycol-bis(,8-aminoethylether)-N,N'-tetraacetic acid. O-phenanthroline inhibited activity by 30 to 50%,but this may have been due to nonchelate properties. Anaerobic conditions or

thiol protective agents were not essential for either the activity or stability of thecellulases during assay. An ultrafiltrable inhibitor of cellulase activity was de-tected in cell-free rumen fluid.

Cellulose digestion in the rumen has beenshown to involve concerted hydrolysis by a num-ber of enzymes (8, 9). Frequently these enzymeshave been characterized by their activity to-wards soluble cellulose derivatives rather thaninsoluble "native" celluloses. Few workers haveused substrates equivalent to the cellulose pres-ent in pasture grasses, and conclusions drawnfrom experiments using the soluble cellulose de-rivatives, e.g., carboxymethyl-cellulose, areprobably less valid than if native substrates areused. The cellulases that occur in rumen con-tents have not been investigated in detail be-cause of difficulties in both isolating an activepreparation and the insensitivity of assays usinginsoluble native celluloses. The pioneering workof King (6, 9, 10) is still the most comprehensivestudy available.

Cellulases are extracellular enzymes and aretherefore subject to factors present in the fluidenvironment that may modify their activity.

t Present address: School of Veterinary Science, MurdochUniversity, Murdoch, Western Australia 6153.

Thus, the pH, ionic strength, temperature, redoxpotential, cations, or anions in the environmentmay interact in a complex way to determine thecellulase activity at a given time. In addition,cellulose substrates vary in physical structuresuch as the degree of crystallinity and length offibrils, and this also influences the rate of cellu-lose hydrolysis. In this paper we report the re-sults of experiments in which the effect of someof these factors on cellulase activity was testedutilizing the sensitivity and specificity offered bya carbohydrase enzyme assay based on gas-liq-uid chromatography (G. B. Storer, J. M. Gaw-thorne, and G. L. Francis, Anal. Biochem., inpress).

MATERIALS AND METHODS

The strategy for maximal yield of enzyme was es-tablished by preliminary experiments which showedthat 90% of the extractable cellulase activity of sheeprumen contents was absorbed on the digesting plantmaterial, and the residual activity was distributedbetween the cell-free rumen fluid (3%) and the surfaceof bacteria (7%). Thus, the plant solids fraction of

643

on March 28, 2020 by guest

http://aem.asm

.org/D

ownloaded from

644 FRANCIS, GAWTHORNE, AND STORER

rumen contents together with its entrapped bacteriawas used as the source of enzyme in the presentstudies. Rumen digesta was collected from fistulatedmerino sheep (19) fed a daily ration of 750 g of choppedwheaten hay and 250 g of chopped lucerne hay givenas 24 equal-sized portions hourly. Representative sam-ples of rumen digesta were obtained and cooled to 4°Cin a vessel that contained sufficient dithiothreitol(DTT) to give a final concentration of approximately1 mM. Unless otherwise specified, all subsequent op-erations were carried out below 4°C in the presence of1 mM DTT. Rumen contents (2.2 kg) were manuallyexpressed through two layers of terylene voile of 0.2-mm pore size to yield 550 g of squeezed solids. Thiswas divided into four equal portions that were eachsuspended in 300 ml of 20 mM sodium phosphatebuffer, pH 6.9. The mixture was shaken vigorously for10 min and strained through the terylene voile. Thisprocedure was repeated four more times, using freshsodium phosphate buffer for each extraction. All'ex-tracts were combined and centrifuged at 20,000 x gfor 30 min. The supernatant was removed, and 650 gof analytical-grade ammonium sulfate per liter wasslowly dissolved with stirring. Precipitation was al-lowed to continue overnight under an atmosphere ofhigh-purity nitrogen. The precipitate was recoveredby centrifugation, redissolved in 200 ml of 20 mMsodium phosphate buffer (pH 6.9), and dialyzed for 29h against 2 liters of degassed glass-distilled watercontaining 1 mM DTT. The dialysis was repeated twomore times to remove salts from the enzyme prepa-ration. The dialyzed preparation was then centrifugedat 80,000 x g for 1 h, and the supernatant was lyoph-ilized. The dried enzyme was stored at -15°C.Enzyme assay. The enzyme preparation was as-

sayed for cellulase activity as previously described(Storer et al., in press). Hemicellulase activity wasdetermined similarly, since the conditions used for gaschromatography gave separate peaks for glucose, cel-lobiose, xylose, arabinose, galactose, and mannose,allowing quantitation of the individual sugars released.Unless otherwise stated, the reaction mixtures con-tained 0.75 mg of freeze-dried enzyme and 30 mg ofsubstrate in 2 ml of sodium phosphate buffer (pH 6.9)with 1 mM DTT. After incubation for 30 min, thereaction was stopped by adding 10 ml of redistilledethanol, and soluble sugars were identified and mea-sured by gas chromatography (Storer et al., in press).All assays were performed in duplicate. Cellulose sub-strates used were either Sigmacell 20 (Sigma ChemicalCo., St. Louis, Mo.), a microcrystalline cellulose withapproximately 20-,um-diameter particle size, orwheaten hay cellulose isolated from the feed fed tosheep used in these experiments (Storer et al., inpress). Before use in assays, the substrates were sus-pended in phosphate buffer and degassed under vac-uum to ensure wetting of the particle surfaces. Enzymeactivity was expressed as nanomoles of the particularsugar released per minute or as nanomoles of glucoseunits released per minute (nanomoles of glucose plustwice the nanomoles of cellobiose). Protein was mea-sured by the method of Lowry et al. (14). Specificactivity was expressed as nanomoles of glucose unitsreleased per minute per milligram of protein. Sub-strate composition was determined by gas chromato-graphic analysis (Storer et al., in press) of the products

APPL. ENVIRON. MICROBIOL.

of acid hydrolysis (18).Experimental details. Substrate concentration

and temperature effects on cellulase activity wereexamined in 50 mM phosphate buffer (pH 6.9), asindicated in the figure legends.

Modification of cellulase activity by changes inphosphate concentration was tested by first dissolvingthe enzyme in 5 mM sodium phosphate (pH 6.9), andaliquots were then combined with substrate in a set ofincubation vessels containing various concentrationsof sodium phosphate buffer, pH 6.9. To examine theeffect of phosphate concentration on cellulase activityin the presence of rumen fluid, enzyme and substratewere incubated in reaction mixtures that had 1.8 ml ofrumen fluid and 0.2 ml of sodium phosphate buffercalculated to give the desired final concentration ofphosphate at pH 6.9. Cell-free rumen fluid was ob-tained by straining rumen contents through terylenevoile and centrifuging the fluid at 100,000 x g for 60min. The supernatant was filtered through a 0.45-,imfilter before use.

For the study of the effect of sodium chloride con-centration on cellulase activity, the enzyme was dis-solved in 10 mM phosphate buffer (pH 6.9), and thesubstrate was suspended in 20 mM phosphate buffer(pH 6.9) that contained various levels of sodium chlo-ride between 10 and 500 mM. One-milliliter volumesof each were mixed and incubated for 30 min.

Stability of the enzyme at different levels of phos-phate was tested using standard quantities of enzymeand substrate in 20, 50, or 100 mM sodium phosphate,pH 6.9. Each enzyme solution was assayed after prein-cubation at 39°C for 0, 30, 60, or 90 min.The importance ofDTT for activity and short-term

stability of the cellulases was investigated by assayingthe enzyme in 50 mM phosphate buffer (pH 6.9) inthe presence or absence of 1 mM DTT before andafter a 2-h incubation that was carried out at 39°C inthe presence or absence of DTT.

Modulation of enzyme activity by added metal ionswas tested using ferric, calcium, magnesium, cobalt,copper, manganese, and zinc chlorides dissolved sep-arately in 20 mM phosphate buffer at 0.4 mM andcombining 0.5 ml of each solution with 0.5 ml ofdouble-strength enzyme solution in a reaction tubethat contained 30 mg of substrate suspended in 1 mlof 20 mM sodium phosphate buffer, pH 6.9. Thechelators ethyleneglycol-bis(fB-aminoethyl ether)-N,N'-tetraacetic acid, ethylenediaminetetraaceticacid, and O-phenanthroline were titrated to pH 6.9with 3 M sodium hydroxide and used at a final con-centration of 5 mM in enzyme assays.

Cellulase activity response to the addition of anultrafiltrate of rumen fluid was examined by replacing0.8 ml of sodium phosphate buffer with 0.8 ml ofultrafiltrate in a standard cellulase assay. The ultraf-iltrate was prepared by filtering cell-free rumen fluidthrough a noncellulosic filter (Amicon, UM-05; molec-ular weight exclusion, 500).

RESULTS AND DISCUSSIONEnzyme preparation. A 0.27-g amount of

crude enzyme was extracted per kg of rumencontent (1.09 g/kg of squeezed plant solids). Thespecific activity of this material was 22 glucose

on March 28, 2020 by guest

http://aem.asm

.org/D

ownloaded from

VOL. 36, 1978

units/min per mg of crude enzyme (35.7 glucoseunits/min per mg of protein) when either cellu-lose substrate was used in the assay. However,the ratio of cellobiose to glucose produced was0.15 with Sigmacell and 0.29 with wheaten haycellulose. Cellulase activity was linear for morethan 30 min with both substrates under standardassay conditions (Storer et al., in press).Substrate concentration. Even though cel-

lulose is an insoluble substrate, the cellulolysisappeared to conform to Michaelis-Menten ki-netics under the conditions used. Linear Eadie-Hofstee plots were obtained (Fig. 1). Forwheaten hay cellulose, the V,,. and apparentKm were 19.8 glucose units/min and 6.35 mg/ml,respectively, and for Sigmacell they were 33glucose units/min and 27.5 mg/ml, respectively.For complex systems such as this involving anumber of intermediate reactions, the Vm,. andapparent Km describe the rate-limiting step (5).Thus, the apparent Km values, expressed as mil-ligrams substrate per milliliter indicate that atthe rate-limiting step the cellulases have agreater affinity for the wheaten hay-derived cel-lulose than the recrystallized Sigmacell. How-ever, this conclusion must remain tentative untilthe ratio of accessible surface area to weight ofthese two substrates is knowvn, because it istheoretically more accurate to express substrateconcentration as accessible surface area per mil-liliter rather than weight per milliliter for thederivation of an affinity constant (11, 15). Thecalculated VmXlx is unaffected by the surface area-to-weight ratio; consequently, at nonlimiting

35'

C

GE

En

Qo 20

1-4

CE 15

n

o 10

0

0 0.5 1.0 1.5 2.0 2.5 3.0

Vel./[S] as nmol glucose units per min per mg substrate

FIG. 1. Hofsteeplot for the calculation of V.. andKm for (0) wheaten hay cellulose and (0) Sigmacellcellulose. Each reaction tube contained 0.75 mg ofcrude enzyme incubated with increasing amounts ofsubstrate.

PROPERTIES OF RUMEN CELLULASES 645

substrate concentrations the regenerated cellu-lose of Sigmacell is the more active surface forcellulolysis. This suggests that factors otherthan a difference in surface area are involved indetermining the observed cellulase activity.Wheaten hay cellulose contained an arabino-xy-lan which was hydrolyzed to arabinose and xy-lose by the hemicellulases present in the crudeenzyme preparation. For xylose release the Vmel,and apparent Km were 25.6 rmol/min and 50mg/ml, whereas for arabinose they were 7nmol/min and 51 mg/ml, respectively.Only 1.4% of the substrate present was hydro-

lyzed to glucose and cellobiose in establishingthe conformity to Michaelis-Menten kinetics.Thus, the possibility that the measured activitywas due to hydrolysis of a labile glucan presentat low concentration was tested. Incubation ofwheaten hay cellulose with porcine a-amylasefailed to release any detectable sugars, indicatingthat hydrolysis of an a-1,4-glucan was not in-volved. Furthermore, digestion of wheaten haycellulose with a commercial Aspergillus nigercellulase to convert 13% of the substrate to glu-cose did not reduce the initial reaction rate ofsuch predigested substrate when reincubatedwith fresh enzyme under identical conditions. Infact, it slightly activated the substrate towardshydrolysis. This indicates that a unique readilyhydrolyzable component is not present in theprepared wheaten hay cellulose nor responsiblefor an erroneous initial rate measurement. How-ever, it reinforces the view that the substrateshould be defined.

Effect of temperature. The optimum tem-perature for cellulase activity was 49°C for bothsubstrates under the assay conditions used. Hy-drolysis of the arabino-xylan component ofwheaten hay cellulose was maximal at 43°C (Fig.2). The apparent activation energies obtainedfrom Arrhenius plots (Fig.,3) were 56,900 J/molof glucose released for Sigmacell and 61,100J/mol of glucose for wheaten hay cellulose. Xy-lose release from wheaten hay cellulose also hadan activation energy of 61,100 J/mol. All Ar-rhenius plots showed an inflection point at 35°C.These last two observations may be related tothe nature of the substrate as discussed below.Substrate range. A variety of carbohydrases

constitute the crude enzyme preparation usedhere, enabling the susceptibility of plant cell wallpolysaccharides to enzymatic hydrolysis to betested. Sequential removal of the structural pol-ysaccharides from the cell wall had a dramaticeffect on its susceptibility to enzymatic hydrol-ysis (Table 1). Whole plant cell walls were rela-tively resistant, and removal of the pectic sub-stances did not significantly alter this. Ligninremoval had a dramatic effect on the release ofsoluble sugars, especially xylose and arabinose.

on March 28, 2020 by guest

http://aem.asm

.org/D

ownloaded from

646 FRANCIS, GAWTHORNE, AND STORER

Glucose release from increased cellulase activitywas not observed until after the hemicelluloseswere separated from cellulose. The hemicellu-loses were readily attacked, releasing xylose as

25 -

20

C

15

0)

-50

25 30 35 40 45 50 55Temperature (°C)

FIG. 2. Effect of temperature on the rate ofreleaseof (C1) glucose units, (0) xylose, and (0) arabinosefrom wheaten hay cellulose, and (U) glucose unitsfrom Sigmacell cellulose.

the major product, which is reduced in thebranched form of hemicellulose B as expectedfrom previous studies (4). Individual sugars werenot always released at a rate proportional totheir relative abundance in the substrate.

1.2

1.0E

a 0.8co

E 1.0

(._2 1.2a)a}0

-J 1.4

0

l

_-

-0.2L_310 315 320 325 330 335 340

1 X 105K 1

FIG. 3. Arrhenius plots to calculate the apparentactivation energy for the hydrolytic release of (U)glucose (56,900 J/mol), from Sigmacell, and (E) glu-cose units (66,100 JImol), (0) glucose (61,100 J/mol),(0) xylose (61,100 J/mol), and (A) cellobiose (74,500Jlmol), from wheaten hay cellulose, by the crudeenzyme preparation.

TABLE 1. Digestion of wheaten hay cell wall fractionsa% Composition in: Total

sugar re-Substrate type Substrate Enzyme hydrolysate leased

(nmol/Glc Xyl Ara Man Gal Glc Xyl Ara Man Gal min)

Whole cell walls - - - - - 70 30 ND ND ND 2.30Whole cell walls less - - - - - 64 36 ND ND ND 1.99

pectinHolocellulose (above, 53 41 6.0 ND ND 16 73 11 ND ND 21.4

less lignin)Hemicellulose A 2.2 90 7.9 ND 0.33 9.9 80 9.1 ND 0.67 79.7Hemicellulose B 7.2 73 16 0.36 3.4 8.4 79 11 ND 2.0 121Hemicellulose, 2.2 50 38 0.39 8.9 12.2 69 17 ND 1.1 76.2branched B

Hemicellulose, linear B 8.0 76 14 0.25 2.3 8.7 79 11 ND 1.9 126Hemicellulose C - - - - - 22 61 17 ND ND 11.7Cellulose 96.9 2.47 0.65 ND ND 72 22 6.0 ND ND 15.3

a Standard assay conditions were used for all substrates, and percent composition was calculated on a weightbasis. Abbreviations: Glc, glucose; Xyl, xylose; Ara, arabinose; Man, mannose; Gal, galactose; ND, not detectable.

Not measured.Substrates were prepared as previously described (2).

APPL. ENVIRON. MICROBIOL.

1.4r

on March 28, 2020 by guest

http://aem.asm

.org/D

ownloaded from

PROPERTIES OF RUMEN CELLULASES

Nonglucan polysaccharides of cellulose prep-arations are thought to be intertwined with theouter glucan chain of microfibrillar cellulose(16). The thermodynamic and kinetic data abovesuggest that hydrolysis of the wheaten hay cel-lulose arabino-xylan is dependent on prior re-moval of protecting glucan chains. Besides anequivalence of apparent Km values for arabinoseand xylose release, the activation energies forthe rate-limiting step in glucose and xylose re-lease were the same. Furthermore, the rates ofxylose and arabinose release from wheaten haycellulose mirror their relative abundance in thesubstrate. This was not the case for other hem-icellulose substrates tested (Table 2). Overall,the above results indicate the complexity of het-eropolymer digestion by a mixed population ofcarbohydrases such as exists in the rumen.

Effect of phosphate buffer and sodiumchloride on cellulase activity. Increasing theconcentration of phosphate buffer from 5 to1,000 mM affected cellulase activity (Fig. 4).Phosphate buffer added in the physiologicalrange 5 to 50 mM stimulated cellulase activity,with a maximum at 50 mM for both substrates.Since sodium chloride at ionic strengths up tothat of 100 mM sodium phosphate buffer had anegligible effect on cellulase activity, this stim-ulatory effect of phosphate buffer was due tophosphate ions rather than sodium ions or non-specific effects of ionic strength. Although phos-phate salts are known to stimulate cellulosedigestion by rumen microorganisms in vitro (3,7), a direct effect of phosphate on the activity ofcellulolytic enzymes has not been previously re-ported. Stimulation at the unphysiological phos-phate concentrations above 200 mM was notfurther investigated in these experiments andwas probably due to nonspecific effects of highionic strength on the surface properties of thesubstrate.

TABLE 2. Comparison of substrate composition andproducts of enzyme hydrolysisa

Ratio of xylose to arabi-nose in:

Substrate typeb EnzymeSubstrate hydroly-

sate

Cellulose 3.7 3.7Hemicellulose A 11.3 8.8Hemicellulose B 4.4 7.3Hemicellulose, branched B 1.3 4.0Hemicellulose, linear B 5.4 7.3

a Standard assay conditions were used for all sub-strates.

b Substrate was prepared as previously described(2).

14c

En

a)cn0-

.0)

Ec

C-)

>10a

0 100 200 300 400 500 1000Final phosphate concentration (mM)

FIG. 4. Effect ofphosphate buffer concentration oncellulase activity measured as glucose units releasedper minute from (0) Sigmacell and (0) wheaten haycellulose.

Cellulases were stable to 30 min of incubationat 39°C in phosphate buffers between 20 and100 mM. Even incubating the enzyme for 90 minbefore assay resulted in only a 3% loss of activityin 20 mM phosphate buffer, increasing to 10%loss of activity in 100 mM phosphate buffer.Stimulation of cellulase activity by phos-

phate in the presence of cell-free rumenfluid. The effect of phosphate concentration oncellulase activity was greatly magnified in thepresence of rumen fluid (Fig. 5). Increasing thephosphate concentration from 18 to 90 mM inthe presence of rumen fluid increased cellulaseactivity approximately threefold with wheatenhay cellulose and fourfold with Sigmacell. Wedo not known the mechanisms of this action, butpotentiation of the phosphate effect by rumenfluid would occur if sodium phosphates inacti-vated inhibitors in the fluid in addition to stim-ulating cellulose activity.

Inhibition of cellulase activity by rumenfluid at constant phosphate concentration.Figure 6 shows that cellulase activity was in-hibited up to 70% by increasing the concentra-tion of cell-free rumen fluid in the assay from 0to 90% (vol/vol). The degree of inhibition wassimilar for both cellulose substrates. Rumenfluid that had been passed through an ultrafilterto exclude compounds with molecular weights

AC4AVOL. 36, 1978

on March 28, 2020 by guest

http://aem.asm

.org/D

ownloaded from

648 FRANCIS, GAWTHORNE, AND STORER

5)

0.

Ec

U,

u

0

U)

0 20 40 60 80 100 120 140 160Final phosphate concentration (mM)

FIG. 5. Effect of phosphate concentration on cel-lulase activity measured against (0, 0) wheaten haychaff and (-, 0) Sigmacell celluloses in the absence(circles) or presence (squares) of centrifuged rumenfluid.

greater than 500 inhibited to the same extent asunfiltered cell-free fluid. This indicates that theinhibitor was a small molecule and was removedby the extensive dialysis used in the preparationof crude enzyme. Other workers have also de-tected inhibitors of cellulase activity in prepa-rations from rumen contents (9, 10, 12, 13), butthe nature of these inhibitors has not been de-termined.Metals and chelators. None of the ions

tested had a significant effect on cellulase activ-ity. The chelators ethyleneglycol-bis(,8-amino-ethyl ether)-N,N'-tetraacetic acid and ethyl-enediaminetetraacetic acid at concentr&tions of5 mM also had no significant effect on cellulaseactivity, but similar concentrations of O-phe-nanthroline caused 47% inhibition when whea-ten hay cellulose was the substrate and 29%inhibition with Sigmacell. The reported effectsof added metal ions on the activity of rumencellulases have been variable, particularly forferric ions (11, 12). Absence of any effect in thepresent experiments is supported by the nonin-hibition of the chelators EDTA and EGTA. 0-

phenanthroline inhibition may have been due toproperties other than specific chelation of Zn2+and Fe2".DTT. The anaerobic environment of rumen

contents together with the low redox potential(8) was the rationale for including the thiol

c

E

n

0 _

0 2

2

0 20 40 60 80 100% rumen fluid

FIG. 6. Effect of different levels of rumen fluid ata constant phosphate concentration on the cellulaseactivity towards (0) wheaten hay cellulose and (0)Sigmacell cellulose.

protective agent DTT in all solutions used inthe current work. Published results of experi-ments with cellulases ioslated from the rumenor of pure cultures of rumen microorganismsdiffer in their sensitivity to oxygen and the pro-tection obtained with cysteine or other reducingagents (10, 12, 20). In our experiments additionofDTT at a concentration of 1 mM in the assayhad no effect on cellulase activity, and neitherdid the exclusion of oxygen by flushing withhigh-purity nitrogen. Preincubation of enzymefor 2 h at 390C in the presence ofDTT decreasedcellulase activity by 12%, but preincubationwithout DTT had no effect. We have not ex-amined the role of DTT during enzyme extrac-tion, but in view of the lack of effect during assayand the observation that extracellular proteinsof microbial origin are usually devoid of cysteine(16), its inclusion at that stage may be unneces-sary.Conclusion. The cellulases of rumen micro-

organisms are extracellular enzymes and aretherefore subject to the action of stimulators orinhibitors in the surrounding fluid. Interactionbetween these factors is probably complex andmay be dramatically altered when the host ani-mal consumes food or water. Determination ofthe microenvironment in which cellulases func-tion in vivo would assist the evaluation of factorsthat affect cellulose hydrolysis in vitro. It hasbeen suggested, for example, that hydrolytic en-zymes of rumen bacteria are retained within the

APPL. ENVIRON. MICROBIOL.

on March 28, 2020 by guest

http://aem.asm

.org/D

ownloaded from

PROPERTIES OF RUMEN CELLULASES 649

confines of a diffuse extracellular coat whichcontrols the entry of factors that may affect theactivity of these enzymes (1).The nature of the feedstuff also determines

the extent of nutritional utilization by the ani-mal. An indication of how structural and chem-ical relationships affect this was obtained in theexperiments reported above. Despite the prob-ably extracellular source of our enzyme, theresults obtained reinforce the view that the deg-radative capacity of the rumen microflora isdependent on the synergistic action of the freeand cell-associated hydrolases present.

LITERATURE CITED1. Akin, D. E., D. P. Burdick, and G. E. Michaels. 1974.

Rumen bacterial interrelationships with plant tissueduring degradation revealed by transmission electronmicroscopy. Appl. Microbiol. 27:1149-1156.

2. Blake, J. D., and G. N. Richards. 1970. Polysaccharidesof tropical pasture herbage. Aust. J. Chem.23:2353-2360.

3. Evans, J. L., and G. K. David. 1966. Influence of sulfur,molybdenum, phosphorus and copper interrelationshipsin cattle upon cellulose digestion in vivo and in vitro. J.Anim. Sci. 25:1014-1019.

4. Gaillard, B. D. E., R. W. Bailey, and R. T. J. Clarke.1965. The action of rumen bacterial hemicellulases onpasture plant hemicellulose fractions. J. Agric. Sci.64:449-454.

5. Gibson, K. D. 1953. True and apparent activation ener-gies of enzymic reactions. Biochim. Biophys. Acta10:221-229.

6. Gill, J. W., and K. W. King. 1957. Characteristics of freerumen cellulases. J. Agric. Food Chem. 5:363-367.

7. Hall, 0. G., H. D. Baxter, and C. S. Hobbs. 1961. Effectof phosphorus in different chemical forms on in vitro

cellulose digestion by rumen microorganisms. J. Anim.Sci. 20:817-819.

8. Hungate, R. F. 1966. The rumen and its microbes, p. 203.Academic Press Inc., New York.

9. King, K. W. 1956. Basic properties of the dextrinizingcellulases from the rumen of cattle. Va. Agric. Exp. Stn.Tech. Bull. 127:3-16.

10. King, K. W. 1959. Activation and cell-surface localizationof certain fB-glucosidases of the ruminal flora. J. DairySci. 42:1848-1856.

11. King, K. W. 1966. Enzymic degradation of crystallinehydrocellulose. Biochem. Biophys. Res. Commun.24:295-298.

12. Krishnamurti, C. R., and W. D. Kitts. 1969. Preparationand properties of cellulases from rumen microorga-nisms. Can. J. Biochem. 15:1373-1379.

13. Leatherwood, J. M. 1965. Cellulase from Ruminococcusalbus and mixed rumen microorganisms. Appl. Micro-biol. 13:771-775.

15. Lowry, 0. H., N. J. Rosebrough, A. L. Farr, and R. J.Randall. 1951. Protein measurement with the Folinphenol reagent. J. Biol. Chem. 193:265-275.

15. McLaren, A. D. 1963. Enzyme reactions in structurallyrestricted systems. IV. The digestion of insoluble sub-strates by hydrolytic enzymes. Enzymologia26:237-246.

16. Northcote, D. H. 1972. Chemistry of the plant cell wall.Ann. Rev. Plant Physiol. 23:113-132.

17. Pollock, M. R. 1962. Exoenzymes, p. 121-178. In I. C.Gunsalas and R. Y. Stanier (ed.), The bacteria, vol. 4.Academic Press Inc., New York.

18. Sloneker, J. H. 1971. Determination of cellulose andapparent hemicellulose in plant tissue by gas-liquidchromatography. Anal. Biochem. 43:539-546.

19. Smith, R. M., and H. R. Marston. 1970. Production,absorption, distribution and excretion of vitamin B,2 insheep. Br. J. Nutr. 24:857-891.

20. Smith, W. F., I. Yu, and R. E. Hungate. 1973. Factorsaffecting cellulolysis by Ruminococcus albus. J. Bacte-riol. 114:729-737.

VOL. 36, 1978

on March 28, 2020 by guest

http://aem.asm

.org/D

ownloaded from