APPuED MIcRoBIoLoGY, Sept., 1966Copyright © 1966 American Society for Microbiology

Vol. 14, No. 5Printed in U.S.A.

Importance of the Isovalerate Carboxylation Pathwayof Leucine Biosynthesis in the RumenMILTON J. ALLISON, JERRY A. BUCKLIN, AND I. M. ROBINSON

National Animal Disease Laboratory, Animal Disease and Parasite Research Division, U.S. DepartmentofAgriculture, Ames, Iowa

Received for publication 6 May 1966

ABSTRACT

ALLISON, MILTON J. (National Animal Disease Laboratory, Ames, Iowa), JERRYA. BUCKLIN, AND I. M. ROBINSON. Importance of the isovalerate carboxylationpathway of leucine biosynthesis in the rumen. Appl. Microbiol. 14:807-814. 1966.-Certain anaerobic ruminal bacteria synthesize the leucine carbon skeleton by use ofa pathway different from that described in other microorganisms. These organismscarboxylate the intact carbon skeleton of isovalerate, synthesizing leucine-2-C04from isovalerate-J-C14. Strains of Bacteroides ruminicola and Peptostreptococcuselsdenii were like Ruminococcus flavefaciens in that they incorporated appreciableamounts of C14 from isovalerate-J-C'4 into cellular protein and in that the onlylabeled amino acid found was leucine. The specific activity of ,B-isopropylmalatedehydrogenase in extracts from R. flavefaciens and from the mixed bacterial popula-tion from the rumen was very low as compared with the specific activity of thisenzyme in extracts from Escherichia coli. This suggests that the pathway of leucinebiosynthesis that operates in many aerobic and facultative microorganisms is notthe major pathway in rumen bacteria. This was supported by the finding thatafter fermentation of whole rumen contents with acetate-2-C'4, leucine from thebacterial cells had a specific activity lower than one would expect if acetate wasincorporated directly into carbons 1 and 2 of leucine.

A number of microorganisms utilize a commonpathway for biosynthesis of the leucine carbonskeleton. Ehrensvard et al. (17) showed that inyeast the carboxyl carbon of leucine was derivedfrom the carboxyl carbon of acetate. Later,Abelson and co-workers (1, 2, 34) and Strassmanet al. (37) presented evidence that indicated thata-ketoisovalerate condensed with a 2-carbonfragment coming from acetate, and that thisproduct, a-isopropylmalate, was subsequentlyrearranged and decarboxylated to yield a-keto-isocaproate, which by transamination could beconverted to leucine. The operation of this path-way has since been demonstrated in a number ofaerobic and facultative microorganisms (28). InRuminococcus flavefaciens, an anaerobic cellulose-digesting bacterium from the rumen, anotherpathway for biosynthesis of leucine operates. Inthis organism, and in whole rumen contentsincubated in vitro, isovalerate is carboxylatedby some unknown mechanism to yield the carbonskeleton of leucine (4, 5).The present investigation was conducted in

an attempt to assess the relative importance ofthese biosynthetic pathways in the rumen.

MATERIALS AND METHODSThe pure cultures of rumen bacteria studied were

isolated and described by Bryant and Robinson (13).All organisms were grown by use of a modification ofthe anaerobic culture technique described by Hungate(25).

Assay for ,1-isopropylmalate dehydrogenase. Cellsof Ruminococcusflavefaciens, strain C94, were grownon a medium similar to the RGCA medium of Bryantand Burkey (11), except that glucose and agar weredeleted. To maintain anaerobiosis, the medium wasbubbled slowly with O2-free CO2 while it was cooledin ice water immediately after it was removed fromthe autoclave. Cysteine-HCI and Na2CO3 solutionswere added to the cooled medium and Na2S-9H2O(0.5 mg/ml, sterilized under N2) was added just priorto inoculation. Cells from 2 liters of medium wereharvested by centrifugation after growth for 17 hr tooptical density (OD) 0.39 (600 m, in 18-mm cuvettes)and were washed once with anaerobic dilution solution(11).

Escherichia coli strain B was grown in medium Cof Roberts et al. (34). Cells were harvested by cen-trifugation after growth from OD 0.02 to 0.63 in 5.5hr. Cells were washed once with 0.85%0 NaCl.

Ruminal contents were obtained from a heifer 2hr after it had been fed an alfalfa-hay ration. Thematerial was strained through cheesecloth and cen-

807

on April 24, 2020 by guest

http://aem.asm

.org/D

ownloaded from

ALLISON, BUCKLIN, AND ROBINSON

trifuged at 365 X g for 3 min to remove most of theprotozoa, some bacteria, and large plant particles; thesupernatant fluid was then centrifuged at 15,000 X g

for 15 min. The precipitate, called "mixed rumen

bacteria," was tested for enzyme activity.Cell-free extracts were prepared by treatment of

cells suspended in 0.067 M phosphate buffer (pH 7.2)for 20 and 40 min at 4 C with a 20-kc MSE ultrasonicoscillator (Measuring and Scientific Equipment, Ltd.,London, England). Cell debris was removed by cen-

trifugation at 18,000 X g for 20 min at 4 C, and thesupernatant fluid was tested for its capacity to producea-ketoisocaproate from fl-isopropylmalate by use ofthe procedure described by Burns, Umbarger, andGross (16). Glutamic-pyruvic transaminase activitywas determined with use of reagents and methodsfrom Sigma Chemical Co., St. Louis, Mo. Proteinwas measured by the method of Lowry et al. (26).

Aminio acid biosynthesis from acetate. Rumen con-

tents from a sheep (150-lb wether on an alfalfa-hayration) were obtained through a ruminal fistula. A5-g amount of the ingesta was incubated under CO2in the presence of 50,uc of acetate-2-C14 in a rubber-stoppered, side-arm test tube, connected in series byrubber tubing to traps containing 2 N HCl and 3 NNaOH. The acetate-2-C04 (Calbiochem) was dissolvedin 0.5 ml of water and had a specific activity of 20.5mc/mmole. After fermentation for 60 min in oneexperiment and 90 min in another, the fermentationmixture was strained through two layers of cheese-cloth with pressure applied manually. The solids wereresuspended twice in 9 ml of anaerobic dilution solu-tion, and the filtrates which passed through cheese-cloth were pooled. The filtrate was centrifuged at30 X g for 5 min to remove protozoa and large plantparticles, and the bacteria in the supernatant fluidwere precipitated by centrifugation at 18,000 X g for15 min. The bacteria were fractionated by use of,essentially, the methods of Roberts et al. (34).The microbial protein was hydrolyzed under reflux

in 20% (v/v) hydrochloric acid for 22 hr at 112 C.The hydrochloric acid was removed by repeated con-centration of the solution under reduced pressure atless than 60 C on a rotary evaporation apparatus.Neutral and acidic amino acids in the protein hy-drolysates were separated by column chromatographyon ion-exchange resins by use of methods similar tothose of Moore et al. (29) and Zacharius and Talley(44). Basic amino acids were separated by use of amethod similar to that described by Theurer (38).Amino acids in protein hydrolysates were also

separated by paper chromatography with develop-ment in two dimensions by use of the method de-scribed by Wolfe (43). Systems used for paper chroma-tographic separation in one dimension included thebutanol-acetic acid system described by Smith (35)and the benzyl alcohol-butanol system at pH 8.4described by McFarren (27). The latter system per-mitted separation of leucine from isoleucine.

Leucine biosynithesis from isovalerate by pure cul-tures. Incorporation of isovalerate-1-C14 by purecultures of rumen bacteria was studied by use ofmethods similar to those used previously (4). Themedium used to grow the pure cultures was the syn-

thetic medium of Bryant and Robinson (12) withthe following modifications. Glucose was added atthe same concentration as cellobiose, whereas thecasein hydrolysate was deleted. Methionine (3 X10-1 M) and hemin (0.002 mg/ml) were added to mediaused to grow Bacteroides ruminicola. An amino acidmixture (serine, alanine, aspartic acid, arginine-HCl,histidine *HCl * H20, lysine-HCl, glycine, proline,threonine, and methionine, all at 0.1 mg/ml; glutamicacid, 0.2 mg/ml; and tyrosine, 0.017 mg/ml) wasadded to media used to culture Peptostreptococcuselsdenii. The culture volume was 10 ml, and eachtube contained approximately 5 ,uc of isovalerate-J-C'4 (Calbiochem). The specific activity of the iso-valerate-l-C'4 was calculated to be 0.6 /Ac/,Amole,except for cultures of P. elsdenii and B. ruminicolastrain 23, in which the specific activity was 5.8 Acl/Amole.

Radioactivity was measured by use of a liquidscintillation spectrometer with the xylene, dioxane,Cellosolve (XDC) scintillation medium described byBruno and Christian (9). Counting efficiency wasdetermined by adding internal standard (toluene-C14).Radioactive areas on paper chromatograms werelocated by use of a windowless gas-flow Geiger-Mullerpaper-strip scanning system and by preparing radio-autographs.

RESULTS

Assay for f-isopropylmalic acid dehydrogenase.A comparison of the specific activities of ,B-iso-propylmalic acid dehydrogenase in cell-freeextracts of E. coli, R. flavefaciens, and mixedrumen bacteria is given in Table 1. The concentra-tion of the enzyme in the extract from E. coli wasmuch greater than the concentration in eitherthe extract from mixed rumen bacteria or R.flavefaciens. The presence of the dehydrogenasein R. flavefaciens is questionable, because, evenin undiluted cell extracts, differences betweencontrols and test samples were difficult to meas-ure.

TABLE 1. Relative specific activities of 3-isopropyl-malate dehydrogenase in extracts of Escherichiacoli, Ruminococcus flavefaciens, and mixedrumen bacteria

EnzymaticSource of extract' specific Ratio

activityb

R. flavefaciens ........... 0.05 1Mixed rumen bacteria.... 0.4 8E. coli ................ 12.1 240

a Extracts prepared by sonic treatment for 20min.

b Units per milligram of protein. One unit ofenzyme is that required to form 1 ,umole of a-keto-isocaproate. The assay system was that describedby Burns et al. (16).

808 APPL. MICROBIOL.

on April 24, 2020 by guest

http://aem.asm

.org/D

ownloaded from

LEUCINE BIOSYNTHESIS IN THE RUMEN

Failure to demonstrate higher levels of thedehydrogenase in extracts of rumen bacteria wasnot due to presence of an inhibitor, since theseextracts did not inhibit formation of the keto acidwhen added to an E. coli extract. The activityof extracts prepared by sonic treatment for 20 or40 min was similar.A measurement was made of glutamic-pyruvic

transaminase activity in the extracts of E. coliand mixed rumen bacteria. The ratio of specificactivities of the transaminase in these extractswas 5.2 (E. coli) :1 (mixed rumen bacteria). Thisis in contrast to a ratio of 30:1 for the specificactivity of the ,B-isopropylmalate dehydrogenase.

Incorporation of acetate-2-C'4 into amino acids.The distribution of C14 after incubation of wholerumen contents 1 hr in vitro with acetate-2-C'4is given in Table 2. The amount of C14 remainingwith the washed plant fibers was not determined.

TABLE 2. Distribution of radioactivity afterinicubation of rumen contentsa in vitro with

acetate-2-C'4Material Count/minb

Suspension from washed fibers....... 10,600,000Protozoal fraction (ppt 30 X g) ......... 28,000Bacterial fraction (ppt 25,000 X g)......60,600Supernatant fluid. . 10, 300,000

Bacterial fractionEthyl alcohol-ether extract ........ ...... 29,400Hot trichloroacetic acid extract ........ 9,100Wash of hot trichloroacetic acid ppt ...... 900Hot trichloroacetic acid ppt .21,200a A 5-g amount of whole rumen contents in-

cubated under CO2 for 1 hr at 39 C.b Counted with a thin-window Geiger-Muller coun-

ter at infinite thinness.

Most of the radioactivity in free, low molecularweight organic acids was still present as acetate;however, some condensation of acetate to producebutyrate appears to have occurred (Table 3).The particulate fraction that was precipitated bycentrifugation at 30 X g for 5 min was notfractionated, but the bacterial fraction (thatprecipitated at 18,000 X g for 15 min) was frac-tionated (Table 2). The lipid and nonlipid mate-rials in the ethyl alcohol-ether extract wereseparated by the method of Folch et al. (19), and78% of the C14 in the extract was in the washedlipid portion.When a hydrolysate of the hot trichloroacetic

acid-precipitated "protein" fraction was placedon a Dowex 50 H+ column (34), 98% of theradioactivity recovered was in the fraction elutedwith NH40H. The specific activities of aminoacids in the protein were determined after asimilar in vitro fermentation for 90 min. Measure-ments of the distribution of radioactivity weresimilar after separation of amino acids by eitherpaper or column chromatography. Figures 1

TABLE 3. Distribution of radioactivity in extracellu-lar low molecular weight acids after incuibationof rumen contents with acetate-2-C'4 fbr I hr

PercentageAcida of C14

recoveredb

Butyric and higher acids ................. 4.8Propionic acid ........................ 2.2Acetic acid ........................ 91 .0Formic, lactic, and succinic acid ......... 2.0

a Acids separated by the column chromato-graphic method of Wiseman and Irvin (42).bAverage of measurements from duplicate

chromatographic separations.

45 50 55 60 65 70 75 80 85 90 95 100 105 110 115 120 125 130 135 I14 145 i, 1 15,i, iOV 1Ou0TUBE NUMBER

FIG. 1. Chromatographic separation of acidic and neutral amino acids of the protein hydrolysate from rumenbacteria after incubation of whole rumen contents in vitro for 90 min in the presence of acetate-2-C'4.

809VOL. 14X 1966

on April 24, 2020 by guest

http://aem.asm

.org/D

ownloaded from

ALLISON, BUCKLIN, AND ROBINSON

and 2 show the separation of amino acids on theion exchange resin columns, and the measurementof C14 in the amino acids. Measurements of thespecific activities of the amino acids and therelative amounts of C14 in various amino acidsare given in Table 4. Most of the radioactivityin the amino acids was in alanine, aspartic acid,and glutamic acid.

Synthesis of leucine from isovalerate by pure

OPTICAL DE

COUNTS / MI

30 35 40 45 50 55TUBE NUMBEF

2.01.5

1.2:NSITY - 1.

INUTES 1.0

0.9

0.8z

0.7 w

0.6

w - 0.5 2zZ

- .4'z .400.3

TI IIII 0.?

0 . 0~~~.100 105 110 115

FIG. 2. Chromatographic separation of basic aminoacids of the protein hydrolysate from rumen bacteriaafter incubation of whole rumen contents in vitro for90 min in the presence ofacetate-2-C'4.

TABLE 4. Radioactivity in amino acids from rumen

bacteria incubated with acetate-2-C'4 for 90 mina

Percentage Disinte- SpecificAmino acid of total C14 gration per activityin amino min per relative

acid ;mole to alanine

Aspartic acid. 19.2 3,430 0.59

Threonine . 5.4 2,460 0.42Serine..4....... .9 2,330 0.40Glutamic acid 17.4 4,800 0.83

Proline ............... 2.5 1,730 0.30Glycine ............... 1.0 260 0.04Alanine ............... 20.6 5,800 1.0Cystine < nil

Valine ................ 7.7 2,930 0.50Diaminopimelic acid.. 1.7 12,110 2.1

Methionine ........... 1.4 2,350 0.40Isoleucine.......... ... 3.9 1,890 0.33Leucine ............... 6.6 3,230 0.56Tyrosine .............. <1 860 0.15Phenylalanine......... <1 90 0.02

Lysine ................ 6.0 5,060 0.87Histidine .... .... .. .. <1 400 0.07Arginine .............. <1 1,220 0.21

-The specific activity of the acetate-2-C"4 in thefermentation mixture at zero-time was 3.9 X 106 dis-integrations per min per jAmole.

TABLE 5. Distribution of C14 in cells ofBacteroidesruminicola and Peptostreptococcus elsdenii aftergrowth in media containing isovalerate-l-Cl4

B. ruminicolaP. elsdenii

Determination strainStrain Strain B-159cGA33a 23b

Disintegrations per min inWhole culture ......... 12,000,000 12,200, 000 11, 000, 000Cells ................. 3,370,000 7,710,000 876,000

Percentage of C14 in cellfractions

Cold trichloroaceticacid extract ....... .. 3.1 1.5 1.4

Ethyl alcohol ether ex-tract ............... 43.5 34.2 2.8

Hot trichloroacetic acidextract ............. 3.5 3.5 3.5

Wash of protein ....... 6.2 7.4 2.0Protein ............... 43.8 53.4 90.2

a Cells fractionated after growth from OD (at 600 myA in18-mm tubes) 0.14 to 1.1.

b Cells fractionated after growth from OD 0.07 to 1.4.c Cells fractionated after growth from OD 0.01 to 0.5.

cultures. The extent of incorporation of iso-valerate-J-C'4 and the distribution of C14 incellular fractions of B. ruminicola and P. elsdeniiare given in Table 5. Amino acids in proteinhydrolysates were separated by paper chroma-tography (27, 35), and, with both methods, theradioactivity migrated with leucine. Severalother strains of rumen bacteria that were culturedin media containing isovalerate-J-C"4 did notincorporate appreciable amounts of the radio-activity. These were Eubacterium ruminantiumB4, Selenomonas ruminantium GA192, Lachno-spira multiparus D32, and Butyrivibrio fibrisolvensDI.

DISCUSSIONThe failure to demonstrate appreciable levels

of ,B-isopropylmalate dehydrogenase in extractsfrom mixed ruminal bacteria or R. flavefacienssuggests that the major pathway for leucinesynthesis by these bacteria does not involve ,B-isopropylmalate as an intermediate. From theseresults alone, however, one cannot exclude thepossibility that ,B-isopropylmalate dehydrogenaseis present and functional in ruminal bacteria butis less stable or requires conditions other thanthose used in assay of enzymes from E. coli orSalmonella typhimurium (16). The latter possi-bility does not appear to be the logical explana-tion, however, in view of results which indicatethat the specific activity of bacterial leucine wasnot particularly high after incubation of themixed population in the presence of acetate-2-C'4.

810 APPL. MICROBIOL.

on April 24, 2020 by guest

http://aem.asm

.org/D

ownloaded from

LEUCINE BIOSYNTHESIS IN THE RUMEN

Earlier studies demonstrated that R.flavefacienssynthesized leucine, with use of isovalerate, evenwhen an adequate quantity of exogenous leucinewas in the growth medium. Furthermore, exoge-nous leucine or other amino acids contributedto only a small amount of the amino acids inbacterial protein (5). Thus, it does not seemprobable that the deficiency of,3-isopropylmalatedehydrogenase in R. flavefaciens was due torepression of synthesis of the enzyme by therelatively small amount of exogenous leucinewhich might have been present in the rumen fluidof the growth medium.

Nutritional studies have shown that acetateis important for growth of many species of rumenbacteria (5, 10, 13, 21). Since ruminal micro-organisms are accustomed to growth in an en-vironment containing a relatively high concentra-tion of acetate, it is not surprising that exogenousacetate carbon is used for synthesis of cellularmaterial. Previous work has shown that acetatecarbon was used by rumen bacteria in aminoacid biosynthesis (5, 23), but the extent of in-corporation into various individual amino acidswas not given.

In E. coli, Neurospora crassa, and Torulopsisutilis cells grown with glucose as a main sourceof carbon, acetate carbon is incorporated mainlyinto lipids, leucine, and amino acids derived fromKrebs cycle intermediates, the "aspartic acidfamily" (aspartic acid, lysine, diaminopimelicacid, threonine, methionine, and isoleucine) andthe "glutamic acid family" [glutamic acid, proline,and arginine (34)]. Major differences betweenthe above pattern and the distribution of C'4(Table 4) are the extensive incorporation ofmethyl-carbon of acetate into alanine by therumen anaerobes, and the finding that leucinewas only moderately labeled.The amount of diaminopimelic acid in the

protein hydrolysate from rumen bacteria wasrelatively small, so that quantitation was difficult.Thus, the high specific activity of diaminopimelicacid given in Table 4 represents a less accuratemeasurement than that obtained with otheramino acids. From the data in Fig. 1 and 2 it canbe seen that the methyl carbon of acetate wasnot used, or was only slightly used, in synthesisof glycine, cystine, tyrosine, phenylalanine,histidine, and arginine.The specific activity of the leucine synthesized

during incubation of ruminal ingesta with acetate-2-C14 was about the same as the specific activityof valine. This suggests that carbon from acetatewas incorporated into carbons other than carbons1 and 2 of leucine. R. flavefaciens did not in-

corporate C14 from acetate-2-C'4 into eitherleucine or valine (4). In several aerobic micro-organisms which incorporate acetate as a unitinto carbons 1 and 2 of leucine, little C'4 fromacetate was found in valine, whereas leucine hada relatively high specific activity (34). Certainnonruminal anaerobic bacteria also differ fromR. flavefaciens and mixed rumen bacteria withregard to incorporation of carbon from acetateinto leucine. Chlorobium thiosulfatophilum grownin acetate-2-Cl4 (20) and Clostridium kluyveriigrown in acetate-i-C'4 (39) incorporated moreC'4 into leucine than into any other amino acid.

Extensive fixation of CO2 into amino acidsin rumen microorganisms has been demonstratedby several workers (22, 24, 30). R. flavefaciensfixed C02, but not formate, into the carboxylcarbon of leucine (4), but the mechanism of thecarboxylation was not determined. The synthesismay be analogous to the reductive, acyl-coen-zyme A, and ferredoxin-dependent pyruvatesynthesis system that has recently been demon-strated in Clostridium pasteurianum (7), C.kluyverii (36), and in certain photosyntheticanaerobic bacteria (7, 18). The high specificactivity of alanine (Table 4) suggests that acetatecarboxylation reactions similar to those demon-strated in these anaerobes may operate in therumen. Analysis of the position of C14 in alaninewould further support or disprove this hypothesis.The discovery of organisms other than R.

flavefaciens that utilize isovalerate for synthesisof leucine supports the supposition that this is asignificant biosynthetic pathway in the rumen.This is especially true in the case of B. ruminicola,which is among the more numerous and versatileof rumen microorganisms. There are markeddifferences among strains of this species, andit has been separated into two subspecies. Thetwo strains used here are the type strains of eachsubspecies (15). This species appears to be im-portant in protein catabolism in the rumen, andone strain produced a labeled 5-carbon acid(presumably isovalerate) during growth in mediacontaining leucine-2-C"4 (8). Apparently thisspecies can catabolize leucine to produce iso-valerate, which may be utilized subsequently insynthesis of cellular leucine. Ammomna can serveas the main source of nitrogen for growth of B.ruminicola if methionine and cysteine are present.Peptides, but not free amino acids, can replaceammonia as the major nitrogen source (32).The difference in uptake of isovalerate between

B. ruminicola strains (Table 5) may be becauseof difference in specific activity of the isovalerate-J-C'4 in the culture medium. The relatively largequantity of C'4 in the ethyl alcohol-ether extracts

VOL. 14, 1966 811

on April 24, 2020 by guest

http://aem.asm

.org/D

ownloaded from

ALLISON, BUCKLIN, AND ROBINSON

of B. ruminicola cells suggests that isovaleratewas used in biosynthesis of 15- and 17-carbonbranched-chain fatty acids or aldehydes, whichare major lipid components of ruminococci (6),B. succinogenes, and a Borrelia species from therumen (40, 41). Since isovalerate was not in-corporated into P. elsdenii lipids to a similarextent, it is suggested that 15- or 17-carbon iso-acids and aldehydes are not significant compo-nents of P. elsdenii cells. Media used in the presentstudy to test for incorporation of isovalerate didnot contain leucine. Although R. flavefaciensused isovalerate to synthesize leucine even whenthe medium contained casein hydrolysate (5),it is not known whether B. ruminicola and P.elsdenii would behave similarly. Unlike R.flavefaciens, neither B. ruminicola nor P. elsdeniirequires isovalerate or other branched-chainfatty acids for growth in media containing en-zymatic casein hydrolysates (13). Growth ofboth strains of B. ruminicola, however, wasgreatly stimulated by volatile fatty acids, in-cluding isovalerate, and evidence was given thatthese acids replaced or partially replaced thegrowth stimulatory effect of casein hydrolysate.

Other amino acids appear to be synthesizedin the rumen via carboxylation reactions similarto those operative in synthesis of leucine fromisovalerate. Valine is synthesized from isobutyrate(4, 40), phenylalanine is synthesized from phenyl-acetate (3), tryptophan is synthesized from indole-3-acetic acid (Allison, unpublished data), and,probably, isoleucine is synthesized from 2-methylbutyrate. These acid precursors of the aminoacids are produced from the same amino acidsduring ruminal fermentation, and the branched-chain volatile fatty acids are usually present inhigh concentrations relative to the concentrationof the free amino acids.Many rumen bacteria are able to grow with

ammonia as the principal source of nitrogen. Anumber of species do not utilize exogenous aminoacids efficiently (5, 14) and synthesize most ofthe nitrogenous components of their cells by useof ammonia, even when grown in a medium richin organic nitrogen (14). The importance ofmicrobial biosynthesis of amino acids in therumen, as opposed to assimilation of free aminoacids, is further suggested by the studies ofPortugal (33), who found that free glutamic acidhad a rapid turnover (t½i = 72-78 sec). Heconcluded that a rather small proportion of theglutamic acid of the free pool, and consequentlyof the diet, was directly incorporated into micro-bial protein, and that there was considerable denovo synthesis of glutamic acid from other

sources of carbon. It seems likely that this istrue for other amino acids also.Our studies indicate that the carboxylation of

isovalerate to produce leucine is an importantreaction in the rumen, but the relative importanceof this pathway and other pathways is not known.Ruminal organisms which do not utilize exoge-nous isovalerate carbon for leucine synthesisinclude strains of B. succinogenes and a Borreliasp. (40, 41), R. albus (6), and the strains of E.ruminantium, S. ruminantium, L. multiparus, andB. fibrisolvens included in this study. Those whichutilize isovalerate for synthesis of leucine includestrains of R. flavefaciens, B. ruminicola, andP. elsdenii. Other important ruminal bacteriahave not been tested; however, even if such in-formation were known, it would be impossible toestimate the quantitative importance of this path-way in the rumen from pure culture studies. Thelimitations of measurements of enzymatic specificactivity as an index of the relative importance ofmetabolic pathways in studies with the mixedruminal population have been discussed (31).Measurement of the kinetics of isovalerate in-corporation into leucine by use of radiotracersis hampered by difficulties in measurement ofisovalerate concentration in the presence of 2-methylbutyrate. Development of satisfactorygas chromatographic methods for separation ofthese isomers will aid future studies.

ACKNOWLEDGMENTS

We acknowledge the generous gift of a sample of,3-isopropylmalate from R. 0. Burns and H. E.Umbarger. We also appreciate the assistance of J.Poley in the separation of amino acids by columnchromatography, and of H. M. Cook in the prepara-tion of figures.

LITERATURE CITE1. ABELSON, P. H. 1954. Amino acid biosynthesis in

Escherichia coli: isotopic competition with C14glucose. J. Biol. Chem. 206:335-343.

2. ABELSON, P. H., AND H. J. VOGEL. 1955. Aminoacid biosynthesis in Torulopsis utilis andNeurospora crassa. J. Biol. Chem. 213:335-364.

3. ALLISON, M. J. 1965. Phenylalanine biosynthesisfrom phenylacetic acid by anaerobic bacteriafrom the rumen. Biochem. Biophys. Res.Commun. 18:30-35.

4. ALLISON, M. J., AND M. P. BRYANT. 1963. Bio-synthesis of branched-chain amino acids frombranched-chain fatty acids by rumen bacteria.Arch. Biochem. Biophys. 101:269-277.

5. ALLISON, M. J., M. P. BRYANT, AND R. N.DOErSCH. 1962. Studies on the metabolic func-tion of branched-chain volatile fatty acids,growth factors for ruminococci. I. Incorpora-tion of isovalerate into leucine. J. Bacteriol.83:523-532.

812 APPL. MICROBIOL.

on April 24, 2020 by guest

http://aem.asm

.org/D

ownloaded from

LEUCINE BIOSYNTHESIS IN THE RUMEN

6. ALLISON, M. J., M. P. BRYANT, I. KATZ, ANDM. KEENEY. 1962. Studies on the metabolicfunction of branched-chain volatile fatty acids,growth factors for ruminococci. II. Biosynthesisof higher branched-chain fatty acids andaldehydes. J. Bacteriol. 83:1084-1093.

7. BACHOFEN, R., B. B. BUCHANAN, AND D. I.ARNON. 1964. Ferredoxin as a reductant inpyruvate synthesis by a bacterial extract. Proc.Natl. Acad. Sci. U.S. 51:690-694.

8. BLADEN, H. A., M. P. BRYANT, AND R. N.DOETSCH. 1961. Production of isovaleric acidfrom leucine by Bacteroides ruminicola. J.Dairy Sci. 44:173-174.

9. BRUNO, G. A., AND J. E. CHRISTIAN. 1961. Deter-mination of carbon-14 in aqueous bicarbonatesolutions by liquid scintillation techniques.Application to biological fluids. Anal. Chem.33:1216-1218.

10. BRYANT, M. P. 1965. Rumen methanogenicbacteria. In R. W. Dougherty, R. S. Allen, W.Burroughs, N. L. Jacobson, and A. D. McGil-liard [ed.], Physiology of digestion in theruminant. Butterworth & Co., Ltd., London.

11. BRYANT, M. P., AND L. A. BURKEY. 1953. Cul-tural methods and some characteristics of someof the more numerous groups of bacteria inthe bovine rumen. J. Dairy Sci. 36:205-217.

12. BRYANT, M. P., AND I. M. ROBINSON. 1961. Somenutritional requirements of the genus Rumino-coccus. Appl. Microbiol. 9:91-95.

13. BRYANT, M. P., AND I. M. ROBINSON. 1962. Somenutritional characteristics of predominantculturable ruminal bacteria. J. Bacteriol. 84:605-614.

14. BRYANT, M. P., AND I. M. ROBINSON. 1963.Apparent incorporation of ammonia andamino acid carbon during growth of selectedspecies of ruminal bacteria. J. Dairy Sci.46:150-154.

15. BRYANT, M. P., N. SMALL, C. BOUMA, AND H.CHU. 1958. Bacteroides ruminicola n. sp. andSucciuiimonas amylolytica the new genus andspecies. Species of succinic acid-producinganaerobic bacteria of the bovine rumen. J.Bacteriol. 76:15-23.

16. BURNS, R. 0., H. E. UMBARGER, AND S. R. GROSS.1963. The biosynthesis of leucine. III. Theconversion of a-hydroxy-j-carboxyisocaproateto a-ketoisocaproate. Biochemistry 2:1053-1058.

17. EHRENSVXRD, G., L. REIO, E. SALUSTE, ANDR. J. STJERNHOLM. 1951. Acetic acid metabolismin Torulopsis utilis. III. Metabolic connectionbetween acetic acid and various amino acids.J. Biol. Chem. 189:93-108.

18. EVANS, M. C. W., AND B. B. BUCHANAN. 1965.Photoreduction of ferredoxin and its use incarbon dioxide fixation by a subcellular systemfrom a photosynthetic bacterium. Proc. Natl.Acad. Sci. U.S. 53:1420-1425.

19. FOLCH, J., M. LEES, AND G. H. SLOANE-STANLEY.1957. A simple method for the isolation and

purification of total lipides from animal tissue.J. Biol. Chem. 226:497-509.

20. HOARE, D. S., AND J. GIBSON. 1964. Photo-assimilation of acetate and the biosynthesis ofamino acids by Chlorobium thiosulphatophilum.Biochem. J. 91 :546-559.

21. HOBSON, P. N., S. D. MANN, AND W. SMITH.1962. Growth factors for Selenomonas ruminan-tium. Nature 198:213.

22. HOLLER, H., J. HARMEYER, AND W. v. ENGEL-HARDT. 1963. Zum Einban von 14C in dieAminosauren des Pansenbakterien-und Pansen-protozoen-Eirvesses bei der Ziege nach intra-ruminaler Gabe von Natrium-'4C-Carbonat.Zentr. Veterinaermed. 10:713-722.

23. HOOVER, W. H., E. M. KESLER, AND R. J. FLIPSE.1963. Carbon sources for in vitro proteinsynthesis by rumen bacteria. J. Dairy Sci.46:733-739.

24. HUHTANEN, C. N., F. J. CARLETON, AND H. R.ROBERTS. 1954. Carbon dioxide utilization byrumen microorganisms. J. Bacteriol. 68:749-751.

25. HUNGATE, R. E. 1950. The anaerobic mesophiliccellulolytic bacteria. Bacteriol. Rev. 14:1-49.

26. LOWRY, 0. H., N. J. ROSEBROUGH, A. L. FARR,AND R. J. RANDALL. 1951. Protein measure-ment with the Folin phenol reagent. J. Biol.Chem. 193:265-275.

27. McFARREN, E. F. 1951. Buffered filter paperchromatography of the amino acids. Anal.Chem. 23:168-174.

28. MEISTER, A. 1965. Biochemistry of the aminoacids, vol. 2, 2nd ed. Academic Press, Inc.,New York.

29. MOORE, S., D. H. SPACKMAN, AND W. H. STEIN.1958. Chromatography of amino acids onsulfonated polystyrene resins; an improvedsystem. Anal. Chem. 30:1185-1190.

30. OTAGAKI, K. K., A. L. BLACK, H. Goss, ANDM. KLEIBER. 1955. In vitro studies with rumenmicroorganisms using carbon-14 labeled casein,glutamic acid, leucine and carbonate. J. Agr.Food Chem. 3:948-951.

31. PALMQUIST, D. L., AND R. L. BALDWIN. 1966.Enzymatic techniques for the study of pathwaysof carbohydrate utilization in the rumen. Appl.Microbiol. 14:60-69.

32. PITTMAN, K. A., AND M. P. BRYANT. 1964. Pep-tides and other nitrogen sources for growth ofBacteroides ruminicola. J. Bacteriol. 88:401-410.

33. PORTUGAL, A. V. 1963. Some aspects of proteinand amino acid metabolism in the rumen ofthe sheep. Ph.D. Thesis, Aberdeen Univ.,Aberdeen, Scotland.

34. ROBERTS, R. B., P. H. ABELSON, D. B. COWIE,E. T. BOLTON, AND R. J. BRITTON. 1957. Studiesof biosynthesis in Escherichia coli. CarnegieInst. Wash. Publ. 607.

35. SMITH, I. 1960. Chromatographic and electro-phoretic techniques, vol. 1. Interscience Pub-lishers, Inc., New York.

36. STERN, J. R. 1965. Role of cofactors in pyruvateoxidation and synthesis by extracts of Clos-

VOL. 14, 1966 813

on April 24, 2020 by guest

http://aem.asm

.org/D

ownloaded from

814 ALLISON, BUCKL

tridium kluyveri. In A. San Pietry [ed.], Non-heme iron proteins. Antioch Press, YellowSprings, Ohio.

37. STRASSMAN, M., L. A. LOCKE, A. J. THOMAS, ANDS. WEINHOUSE. 1955. A study of leucine bio-synthesis in Torulopsis utilis. Science 121:303-304.

38. THEURER, C. B. 1962. Influence of protein sourceson performance and plasma amino acid patternsof ruminants. Ph.D. Thesis, Iowa State Univ.,Ames.

39. TOMLINSON, N., AND H. A. BARKER. 1954. Carbondioxide and acetate utilization by Clostridiumkluyveri. I. Influence of nutritional conditionson utilization patterns. J. Biol. Chem. 209:585-595.

40. WEGNER, G. H. 1962. The metabolic fate of fatty

IN4, AND ROBINSON APPL. MICROBIOL.

acids required by certain rumen bacteria.Ph.D. Thesis, Univ. Wisconsin, Madison.

41. WEGNER, G. H., AND E. M. FOSTER. 1963. In-corporation of isobutyrate and valerate intocellular plasmalogen by Bacteroides succino-genes. J. Bacteriol. 85:53-61.

42. WISEMAN, H. G., AND H. M. IRVIN. 1957. Deter-mination of organic acids in silage. J. Agr.Food Chem. 5:213-215.

43. WOLFE, M. 1957. The quantitative determinationof amino acids by paper chromatography: asolvent to replace phenol. Biochim. Biophys.Acta 23:186-191.

44. ZACHARIUS, R. M., AND E. A. TALLEY. 1962.Elution behavior of naturally occurring nin-hydrin positive compounds during ion exchangechromatography. Anal. Chem. 34:1551-1556.

on April 24, 2020 by guest

http://aem.asm

.org/D

ownloaded from