Vol. 55, No. 10

Sodium-Dependent Transport of Branched-Chain Amino Acids by aMonensin-Sensitive Ruminal Peptostreptococcus

GUANGJIONG CHEN' AND JAMES B. RUSSELL' 2*

Department ofAnimal Science, Cornell University,' and Agricultural Research Service, U.S. Department ofAgriculture,2Ithaca, New York 14853

Received 27 March 1989/Accepted 18 July 1989

A recently isolated ruminal peptostreptococcus which produced large amounts of branched-chain volatilefatty acids grew rapidly with leucine as an energy source in the presence but not the absence of Na. Leucinetransport could be driven by an artificial membrane potential (Al) only when Na was available, and a chemicalgradient of Na+ (AuNa') also drove uptake. Because Na+ was taken up with leucine and a ZApH could notserve as a driving force (with or without Na), it appeared that leucine was transported in symport with Na+.The leucine carrier could use Li as well as Na and had a single binding site for Na+. The Km for Na was 5.2mM, and the Km and Vmax for leucine were 77 ,uM and 328 nmol/mg of protein per min, respectively. Sincevaline and isoleucine competitively inhibited (Kls of 90 and 49 ,uM, respectively) leucine transport, it appearedthat the peptostreptococcus used a common carrier for branched-chain amino acids. Valine or isoleucine wastaken up rapidly, but little ammonia was produced if they were provided individually. The lack of ammoniacould be explained by an accumulation of reducing equivalents. The ionophore, monensin, inhibited growth,but leucine was taken up and deaminated at a slow rate. Monensin caused a loss of K, an increase in Na, a slightincrease in A*, and a decrease in intracellular pH. The inhibition of growth was consistent with a largedecrease in ATP. The capacity of the ruminal peptostreptococcus to transport and deaminate branched-chainamino acids and its sensitivity to monensin clarifies the observation that monensin decreases branched-chainvolatile fatty acid production in vivo.

For many years it had been assumed that most, if not all,of the predominant bacteria had been isolated from therumen, but these species could not account for in vivo ratesof amino acid degradation (2). Recently, isolated ammonia-producing ruminal bacteria (6, 21) grew rapidly with aminoacids or peptides as their sole energy source and werepresent in significant numbers in vivo. One of them, apeptostreptococcus, produced large amounts of branched-chain volatile fatty acids, and these acids arose from thefermentation of branched-chain amino acids (5).Monensin is often used as a feed additive in beef cattle,

and it inhibits amino acid deamination (9, 27). The reductionin wasteful amino acid fermentation is desirable, but mo-

nensin can also decrease branched-chain volatile fatty acidproduction (13, 17). These acids are required by ruminalcellulolytic bacteria, and a deficiency can inhibit cellulosedigestion (1). Because branched-chain volatile fatty acidscan have a significant effect on feed digestion (23), wedecided to examine the metabolism of branched-chain aminoacids by the ruminal peptostreptococcus in greater detail.The rumen is an Na-rich environment (approximately 90

mM), and results indicated that the peptostreptococcus tookup leucine, valine, and isoleucine by a common carrierwhich was dependent on Na. The carrier had a single site forNa (Km of 5.2 mM) and could be driven by an Na+ gradientin the absence of either a A/i or ZApH. Growth was inhibitedby monensin, an Na+-H+ antiporter, but this effect appearedto be more closely correlated with a decrease in ATP than a

decrease in transport activity or proton-motive force.

MATERIALS AND METHODSCells. The ruminal peptostreptococcus strain was used and

this bacterium has previously been described (5, 21). Cells

* Corresponding author.

were grown anaerobically in medium containing (per liter)292 mg of K2HPO4, 292 mg of KH2PO4, 480 mg of Na2SO4,480 mg of NaCl, 100 mg of MgSO4 - 7H20, 64 mg ofCaCl2. H20, 600 mg of cysteine hydrochloride, 15 g ofCasamino Acids (Difco Laboratories, Detroit, Mich.), vita-mins, and microminerals (7). The medium pH was 6.7, andincubation temperature was 39°C. The peptostreptococcuswas also inoculated into Na-deficient medium which con-tained K salts instead of Na salts, 1 g of purified amino acidsper liter with a composition similar to Casamino Acids as a

carbon source, and 15 g of leucine per liter as an energysource. Na dependency was ascertained by supplementingthis Na-deficient medium with 100 mM NaCl. Batch cultureswere grown overnight to an optical density of approximately1.0 (600 nm, 1 cm). Cells were harvested by centrifugation(11,000 x g, 5 min, 25°C) and washed twice in potassiumphosphate buffer (100 mM, pH 6.5). Supernatants and somecell pellets were stored at -15°C for further analysis. Mem-brane vesicles were prepared as previously described (22).

Transport assays. Washed cells (or vesicles) were treatedwith valinomycin (5 ,uM) and incubated in 100 mM potas-sium phosphate (pH 6.5) to load them with K. K-loaded cellswere diluted 50-fold into 100 mM choline or sodium phos-phate (pH 6.5) to create an artificial Ai or a Au, and achemical gradient of Na+ (AuNa'). The effect of Na wasmonitored by diluting K-loaded cells into choline phosphatebuffer which was supplemented with NaCl (0 to 100 mM).pH experiments were conducted with cells which wereincubated in potassium phosphate (pH 4.5 to 7.5) and dilutedinto the same pH. An artificial ZApH was created by dilutingK (100 mM)- and acetate (80 mM, pH 6.5)-loaded cells intopotassium phosphate (100 mM, pH 6.5). Cells were some-times treated with monensin (5 ,uM) during K loading anddiluted into 100 mM sodium phosphate.

Transport was initiated by adding 4 pAl of concentrated

2658

APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Oct. 1989, p. 2658-26630099-2240/89/102658-06$02.00/0Copyright © 1989, American Society for Microbiology

on April 29, 2016 by guest

http://aem.asm

.org/D

ownloaded from

Na-DEPENDENT TRANSPORT BY RUMINAL PEPTOSTREPTOCOCCUS

cells (40 ,ug of protein) to 200 pI of phosphate buffer(described above). After 0 to 60 s, transport was terminatedby adding 2 ml of 100 mM ice-cold LiCl or 100 mM KCl andfiltering through cellulose nitrate membrane filters (0.45-,umpore size). The filters were washed once with 2 ml of 100 mMLiCl and dried for 20 min at 105°C. Filters were then countedby liquid scintillation. Competition experiments were con-ducted with a 100-fold excess of unlabeled amino acid.The peptostreptococcus transported branched-chain

amino acids very rapidly, and the rate of transport over 5 swas proportional to cell protein as long as the concentrationwas less than 500 ,ug/ml. Assays were always conducted withless than 250 pug of protein per ml for a period of 5 s. The ratewas not linear for more than 5 s, even if the cell suspensionswere diluted, and it appeared that this lack of linearity wasrelated to a rapid dissipation of the artificial AiJ.

Proton-motive force. Samples (2 ml) of an exponentiallygrowing culture were incubated anaerobically with [7-'4C]benzoate (1.00 puCi, 10 ,uCi/,imol), [1,2-'4C]polyethyleneglycol (1.00 puCi, 0.1 ,uCi/mg), 3H20 (1.00 mCi, 0.25 ,Ci/mg)or tetraphenylphosphonium bromide (1.00 puCi, 30 to 40,Ci/,mol) for 5 min. Cultures were then transferred tomicrocentrifuge tubes (1.5 ml) which contained 0.35 g ofsilicon oil (equal-part mixture of Dexter Hysol 550 andDexter Hysol 560; Hysol Co., Olean, N.Y.) and had beenplaced in an anaerobic glove box for 24 h. The tubes werecentrifuged (13,000 x g, 5 min, 22°C), and 20-pl samples ofsupernatant were removed for scintillation counting. Thetubes were then frozen (-15°C), and the bottoms containingcell pellets were removed with a pair of dog nail clippers.Pellets were dissolved in scintillation fluid and counted.Intracellular volume (1.5 plI/mg of protein) was computedfrom the difference in ['4C]polyethylene glycol and 3H20.ApH was determined from the distribution of ["4C]benzoatebetween the medium and pellet by using the Henderson-Hasselbalch equation. Akp was calculated from the uptake of[14C]tetraphenylphosphonium bromide according to theNernst relationship. Nonspecific tetraphenylphosphoniumbromide binding was estimated from cells which had beentreated with valinomycin and nigericin (5 puM).Sodium and potassium. Cells which had been centrifuged

through silicon oil (as described above) were dissolved in 3N HNO3 (24 h) and analyzed for K and Na with a Perkin-Elmer atomic absorption spectrophotometer (The Perkin-Elmer Corp., Norwalk, Conn.). K and Na concentrationswere calculated from the ratio of intracellular volume toprotein (1.5 pdl/mg), and these values were corrected forextracellular K and Na contamination. Na uptake was mea-sured with a combination Na electrode (MicroelectrodesInc., Londonderry, N.H.).ATP determinations. Cells (0.16 mg of protein) were ex-

tracted for 20 min in 0.5 ml of ice-cold 14% perchloric acidsupplemented with 9 mM EDTA and then were centrifuged(13,000 x g, 5 min, 22°C). A portion of the extract (1 ml) wasneutralized with 0.5 ml of KOH-KHCO3 (1 M each) andquickly frozen. The neutralized extracts were thawed, cen-trifuged (13,000 x g, 5 min, 22°C), and assayed for ATP bythe firefly luciferase method (15). Neutralized extracts werediluted 50-fold with 40 mM Tris containing 2 mM EDTA, 10mM MgCl2, and 0.1% bovine serum albumin (pH 7.75). Theluciferase reaction was initiated by adding 100 pld of apurified luciferine-luciferase mix to 100 pld of diluted extractaccording to the recommendations of the supplier (SigmaChemical Co., St. Louis, Mo.). Light output was immedi-ately measured with a luminometer (model 1250; LKB

CD0

z

-J

-)

0-0

1.0l

0.8

0.6

0.4

0.2

0.00 5 10 15 20 25

TIME (h)FIG. 1. The growth of the ruminal peptostreptococcus in sodi-

um-deficient medium with (-) and without (0) 100 mM sodiumchloride.

Instruments, Inc., Gaithersburg, Md.), and ATP served as

the standard.Products. Cell protein was measured by the method of

Lowry et al. (14) after the cells had been heated to 100°C in0.2 N NaOH for 15 min. Ammonia was assayed by thecolorimetric method of Chaney and Marbach (4). Volatilefatty acids from metaphosphoric acid-treated (6% [wt/vol],final concentration) supernatant samples were measuredwith a Gow Mac model 580 flame ionization gas chromato-graph equipped with a Supelco 1220 column (1% H3PO4,100/120 mesh).

Materials. 14C-labeled amino acids were obtained fromAmersham Corp., Arlington Heights, Ill. All chemicals andreagents were reagent grade.

RESULTS

Effect of sodium on growth and transport. When theruminal peptostreptococcus was incubated in an Na-defi-cient medium with purified amino acids (1 g/liter) as a carbonsource and 15 g of leucine per liter as an energy source, thecells grew very slowly (less than 0.02/h), and there was littleincrease in optical density (18-mm tubes, 600 nm) even after24 h (Fig. 1). However, if NaCl (100 mM) was added, thebacteria grew more rapidly (0.17/h), and the final opticaldensity at 24 h was more than 0.9 (Fig. 1).When K-loaded cells were diluted into choline phosphate

to create an artificial A4 in absence of Na, no uptake ofleucine was detected (Fig. 2). However, rapid transport ofleucine was observed if NaCl or LiCl was added to cholinephosphate, and transport could be driven by a chemicalgradient of Na+ (AuNa') in the absence of an artificial Alp.An artificial ZApH created by acetate diffusion could notserve as a driving force for leucine transport with or withoutNa. No transport was observed in the absence of either an

artificial A4i or a AuNa'. Similar results were obtained withisoleucine and valine (data not shown).When the rate of leucine transport was measured at Na+

concentrations ranging from 0 to 100 mM, there was littleincrease at concentrations greater than 20 mM (Fig. 3b). AnEadie-Hofstee plot (v/S versus v) yielded a Km for Na of 5.2mM (Fig. 3a, inset), while a Hill plot indicated that thecarrier had only one binding site for Na+ (Fig. 3b). Na+

VOL. 55, 1989 2659

on April 29, 2016 by guest

http://aem.asm

.org/D

ownloaded from

2660 CHEN AND RUSSELL

AV* AuNaorLi12 0.8- * AuNa0X.. & m(+Na)cm 0.6 X Z4A (+ or- Na)

_ |_ _ n~~~~1 No force0

E 0.4

z 0.2

Lj 0.0~

0 20 40 60TIME (sec)

FIG. 2. Transport of leucine by cells which were treated withvalinomycin (5 p.M), loaded with 100 mM potassium phosphate, anddiluted 50-fold into phosphate buffers containing 100 mM choline(A) or 100 mM choline plus 100 mM NaCl or 100 mM LiCl (0). Cellswere loaded with 100 mM potassium phosphate and diluted into 100mM potassium phosphate plus 100 mM NaCl (M). Cells were loadedwith 100 mM potassium phosphate plus 100 mM NaCl and werediluted into 100 mM sodium phosphate (A) or diluted into 100 mMpotassium phosphate plus 100 mM NaCl (O). Cells were also loadedwith 100 mM potassium phosphate and 80 mM acetate or with 100mM potassium phosphate, 100 mM NaCl, and 80 mM acetate andwere diluted into 100 mM potassium phosphate or 100 mM potas-sium phosphate plus 100 mM NaCl to create an artificial ZApH in theabsence or presence of Na, respectively (0). The pH was always6.5.

uptake in the presence of leucine supported the assumptionthat leucine was transported in symport with Na+ (Fig. 4).

Effects of pH on leucine transport. When cells were treatedwith valinomycin, loaded with K (pH 6.5), and diluted intosodium phosphate at pH levels ranging from 5.0 to 7.0, therewas little effect on leucine transport (Fig. 5a). Even at pH 4.5the uptake rate decreased only 25%. However, if cells wereloaded (30 min) at the same pH as they were diluted, therewas a significant decline in the rate of uptake at pH levelsless than 6.0 (Fig. Sb). Since the assay period over which alinear rate could be detected (5 s) was very short, the initialstudies may not have reflected true changes in extracellularpH, intracellular pH, or the stability of the carrier.

Isoleucine and valine. When the peptostreptococcus wasprovided with 40 mM leucine, more than 70% was fermentedduring the 24-h incubation and the ratio of isovalerate toisocaproate was approximately 0.5 (Table 1). Little valinewas deaminated, but if 10 mM leucine was also provided,isobutyrate production increased more than seven times andvirtually all of the leucine was converted to isocaproate.When leucine was increased further, isobutyrate reached amaximal concentration of approximately 9 mM, some of theleucine was converted to isovalerate, and the cultures were

no longer substrate limited. Total fermentation acid produc-tion was consistent with ammonia production. Based on

theoretical pathways of branched-chain amino acid fermen-tation, the ratio of isobutyrate plus isovalerate to isocaproateshould have been 0.5 (Fig. 6), a value which was in closeagreement with the experimental results (Table 1). Similarexperiments could not be conducted with isoleucine becausestandard methods of gas-liquid or high-pressure liquid chro-matography are unable to separate isovalerate from 2-meth-ylbutyrate.

*E 100

Eo 4

E_ 8

C

uJ

L-

0 20 40 60 80 100 120

NeCl (mM)

2

x0E

0

I

0

0.0 0.5 1.0 1.5 2.0 2.5

Log SFIG. 3. (a) Effect of added NaCl on leucine transport by cells

loaded with potassium and diluted 50-fold into choline phosphate(pH 6.5). An Eadie-Hofstee (v/S versus v) plot is shown in the inset.(b) A Hill plot of the data shown.

When the peptostreptococcus was incubated with[14C]leucine and a 100-fold excess of nonradiolabeled glu-tamine, phenylalanine, serine, threonine, tryptophan, oralanine, little reduction of leucine transport (less than 10%)was observed (data not shown). Isoleucine and valine com-petitively inhibited leucine uptake (Fig. 7), and Ki values forisoleucine and valine were 49 and 90 FM, respectively.There was some difference between the Vmax and Km values

-%21

o

9-

0x

0

0

Leu

1"

_m

10 20 30

TIME (min)FIG. 4. Effect of leucine (1 mM) on the uptake of sodium by the

ruminal peptostreptococcus cells (6.9 mg of cell protein) at pH 7.

a

1.5

>01.5Ien I .0-> 0.5~

0.0 I I

0 4 8V

APPL. ENVIRON. MICROBIOL.

on April 29, 2016 by guest

http://aem.asm

.org/D

ownloaded from

Na-DEPENDENT TRANSPORT BY RUMINAL PEPTOSTREPTOCOCCUS

a

I I I I

4 5 6 7 84 5 6 7

FIG. 6. Theoretical pathways and stoichiometries of branched-chain amino acid fermentation by the ruminal peptostreptococcus.

8pH pH

FIG. 5. Effect of pH on the transport of leucine by cells whichwere loaded with potassium and diluted 50-fold into sodium phos-phate (pH 4.5 to 7.0). (a) Cells were loaded at pH 6.5. (b) Cells wereloaded at the same pH as they were assayed.

for leucine, valine, and isoleucine (Table 2), but the Km toVmax ratios were virtually the same (approximately 0.2,uM/nmol per mg of protein per min).

Effect of monensin. When the peptostreptococcus wasincubated with 15 g of Casamino Acids per liter, monensinaddition (5 ,uM) caused an almost immediate cessation ofgrowth but ammonia production continued (data not shown).Since cells which were pretreated with monensin (5 ,uM) andincubated with 100 mM leucine produced nearly half asmuch ammonia as untreated controls, the inhibition ofgrowth could not be explained solely on the basis of trans-port.When untreated cultures (15 g of Casamino Acids per liter)

were incubated with [14C]tetraphenylphosphonium bromideor benzoate to estimate the proton-motive force (Ap), A*was 91 mV and there was little ZApH (Table 3). Monensincaused a decrease in K, an increase in Na, a reversal ofintracellular pH, and a small increase in A+. The inhibition ofgrowth was concomitant with a large decline in ATP.

DISCUSSION

When leucine was provided as the sole energy source, thepeptostreptococcus was unable to grow in the absence of Naand the lack of growth could be explained by the effect ofNaon transport. Na dependency was supported by the obser-

TABLE 1. Effect of leucine on the deamination and fermentationof valine by the ruminal peptostreptococcusa

Values (mM) for: Ratio ofValine Leucine -IB + IV(MM) (MM) Isobu- Isova- Isoca- TB+tIVb

tyrate lerate proate NH3 to Icb

0 40 0 9.3 18.1 26.0 0.5140 0 0.9 0.0 0.0 0.540 10 6.6 0.2 10.0 14.3 0.6840 20 9.2 0.9 17.5 26.5 0.5840 30 9.2 2.8 23.4 33.1 0.5140 40 8.5 4.3 23.8 33.0 0.53

a Washed cells (150 mg of protein per liter) were incubated for 24 h witheach of the substrates or combinations of substrates.

b Ratio of isobutyrate (IB) plus isovalerate (IV) to isocaproate (IC).

vations that (i) leucine transport could be driven by anartificial Atp only when Na was present, (ii) a chemicalgradient of Na+ (AuNa') alone was sufficient to drivetransport, and (iii) Na+ was taken up in the presence ofleucine. Since ZApH could not serve as a driving force evenwhen Na was present, it appeared that leucine was trans-ported in symport with Na+ in response to either a AI orAuNa'. Competitive inhibition by valine and isoleucineindicated that the peptostreptococcus had a common carrierfor branched-chain amino acids. This carrier had a singlebinding site for Na+, a Km for Na of 5.2 mM, and couldutilize Li as well as Na. The Vmax from branched-chainamino acids ranged from 194 to 328 nmol/mg of protein permin.

Streptococcus cremoris also has a common carrier forbranched-chain amino acids, but transport is coupled to H+rather than Na+ (10). Streptococcus bovis has an Na-dependent transport system for neutral amino acids whichhas two binding sites for Na+ and has a Km for Na of 30 mM(22). Escherichia coli has a variety of Na-dependent trans-porters, and many of them can use Li as well as Na (11, 25,26). The Na-coupled serine-threonine transport system of E.coli has a Km for Na of 21 i.aM (12), a value 250 times lowerthan the one determined for the peptostreptococcus (dis-cussed above).The use of membrane vesicles has enhanced our under-

standing of transport mechanisms in bacteria because the

£

E1._

-icI-0.

E0ECl

6

4

2

0

V (nmoVmg protein/min)FIG. 7. An Eadie-Hofstee plot of leucine transport with either

isoleucine or valine as competitive inhibitors. Isoleucine and valinewere provided at 50 .lM, and the Ki values were 49 and 90 ,uM,respectively.

b12 -

10 -

8 -

6 -

4 -

2 -

0-

ADP+ Pi ATPc

.SE

9-

0EC

CA

Ei3E~

YALI NE(ISOLEUCI NE)

2 NADH 2 NAD4 LEULCINE

ISOBUTYRATE(2 METHYLBUTYRATE)

4 4ISOCAPROATE

4 ISOYALERATE

r & a--% uo s.- h -"2 NAD 2 NADH

LEUCI NE A

ADP+ Pi ATP

0 200 400 600

VOL. 55, 1989 2661

4

-T

on April 29, 2016 by guest

http://aem.asm

.org/D

ownloaded from

2662 CHEN AND RUSSELL

TABLE 2. The Michaelis constants of the branched-chain aminoacid carrier which were estimated from an Eadie-Hofstee plot and

the square of the regression coefficient (r2)

Amino acid Km vmax Ratio of Km

Leucine 77 328 0.23 0.93Valine 55 243 0.22 0.93Isoleucine 33 194 0.17 0.94

a Ratios are given as micromolars per (nanomoles per milligram of proteinper minute).

kinetics are not confounded by intracellular metabolism. Wewere able to prepare membrane vesicles of the peptostrep-tococcus, but the length of time needed for protoplastformation was very long (>1.5 h), even if the cultures werepretreated with penicillin. Since previous work showed thatthe peptostreptococcus contained 80% protein (5), it is notall that surprising that the cell wall was resistant to lysozymeand mutanolysin. Some osmotically sensitive vesicles wereobtained, but they were unable to transport branched-chainamino acids via a K diffusion potential (data not shown).This inactivity could have been related to the time neededfor protoplast formation and exposure to hydrolytic en-zymes.

Previous studies showed that the ruminal peptostrepto-coccus deaminated leucine at a very rapid rate and thatisoleucine and valine were fermented slowly (5). However,leucine, isoleucine, and valine were all taken up rapidly by acommon carrier, and differences in ammonia productioncould not be explained by a transport limitation. Leucine isdeaminated by a dual pathway in which reducing equivalentsresulting from isovalerate formation are used in isocaproateproduction (Fig. 6). Valine and isoleucine are converted toisobutyrate and 2-methylbutyrate, respectively, by path-ways which contain two sites that generate reducing equiv-alents and furnish no means of reducing-equivalent disposal.When leucine was provided with valine, reducing equiva-lents arising from isobutyrate production were used in iso-caproate formation, and the rate of valine deaminationincreased markedly (Table 1). Based on the observation thatthis species produces small amounts of hydrogen and thatinterspecies hydrogen transfer to ruminal methanogens pro-vides another means of reducing-equivalent disposal (3), the

TABLE 3. The effect of monensin on the membrane potential,ATP, and intracellular cations of the ruminal peptostreptococcusa

Values for:Parameter

Control Monensin

pH 6.90 6.90pHi 6.92 6.77ZApH (mV) -1 +8Alp (mV) 91 105Ap (mV) 92 97Ke (mM)" 7.4 7.4Ki (mM)b 437 133Nae (mM)b 100 100Nai (mM)b 622 1560ATP (mM) 9.9 3.3

a Cells were grown with 15 g of Casamino Acids per liter. Control cells wereharvested at an optical density of 0.68. Monensin-treated cells were harvested1 h later at a similar optical density.

b Abbreviations: e, extracellular; i, intracellular. The ratio of intracellularvolume to protein was 1.5 Fd/mg of protein.

ruminal peptostreptococcus could play an important role inthe degradation of ruminal valine and isoleucine as well asleucine.

Continuous culture studies indicated that the peptostrep-tococcus was sensitive to growth at low pH (5), and thissensitivity was consistent with the effect of pH on transport(Fig. Sb). It is well documented that low pH in the rumen isdetrimental to the growth of ruminal cellulolytic bacteria(23). While the direct effect of pH on cellulolytic bacteriacannot be discounted (19), branched-chain volatile fatty acidrequirements should be considered. The negative impact oflow pH on the peptostreptococcus could lead to a deficiencyof branched-chain volatile fatty acids and an inhibition ofcellulolysis.Monensin is an ionophore which is effective primarily

against gram-positive bacteria (18). For many years it wasassumed that Megasphaera elsdenii was the primary pro-ducer of ruminal branched-chain volatile fatty acids (16), butits resistance to monensin (8) could not explain the negativerelationship between monensin and decreased branched-chain volatile fatty acids (17). The peptostreptococcus wassensitive to monensin, and this sensitivity may explain theobservation that exogenous branched-chain volatile fattyacids overcame the negative effect of monensin on milkproduction (L. L. Coutinho, P. F. Machado, R. M. Cook,Abstr. 82nd Annu. Meet. Am. Dairy Sci. Assoc. 1987, abstr.no. P339, p. 217).The effect of monensin on the peptostreptococcus was

consistent with a general mechanism of ionophore action inruminal bacteria (20). K was depleted, and there was aninflux of H+ (reversal of ApH) and subsequent increase inNa (Table 3). The Na content of untreated cells was veryhigh (>600 mM), but recent work showed that S. bovis,another ruminal bacterium, also had a large pool of boundNa (24). Since an ionophore should have little effect on Nabinding to the cell wall, the Na+ gradient was probablyreversed by monensin. A reversed Na+ gradient across thecell membrane could have impeded leucine transport andammonia production, but A+, another driving force foruptake, was not significantly affected by the electroneutralaction of this antiporter. Growth inhibition could most easilybe explained by a large decrease in ATP and excessiveATPase activity which would be needed to expel incomingH+ or Na+.

ACKNOWLEDGMENT

This work was supported by the U.S. Dairy Forage ResearchCenter, Madison, Wis.

LITERATURE CITED1. Allison, M. J., and M. P. Bryant. 1958. Volatile fatty acid

growth factor for cellulolytic cocci of bovine rumen. Science128:4741-475.

2. Bladen, H. A., M. P. Bryant, and R. N. Doetsch. 1961. A studyof bacterial species from the rumen which produce ammoniafrom protein hydrolyzate. Appl. Microbiol. 9:175-180.

3. Bryant, M. P., and M. J. Wolin. 1975. Rumen bacteria and theirmetabolic interactions, p. 297-306. In T. Hasegawa (ed.), Pro-ceedings of the First Intersectional Congress of the Int. Assoc.Microbiol. Soc., vol. 2. Developmental microbiology, ecology.Science Council of Japan, Tokyo.

4. Chaney, A. L., and E. P. Marbach. 1962. Modified reagents fordetermination of urea and ammonia. Clin. Chem. 8:130-132.

5. Chen, G., and J. B. Russell. 1988. Fermentation of peptides andamino acids by a monensin-sensitive ruminal peptostreptococ-cus. Appl. Environ. Microbiol. 54:2742-2749.

6. Chen, G., and J. B. Russell. 1989. More monensin-sensitive,

APPL. ENVIRON. MICROIBIOL.

on April 29, 2016 by guest

http://aem.asm

.org/D

ownloaded from

Na-DEPENDENT TRANSPORT BY RUMINAL PEPTOSTREPTOCOCCUS

ammonia-producing bacteria from the rumen. Appl. Environ.Microbiol. 55:1052-1057.

7. Cotta, M. A., and J. B. Russell. 1982. Effect of peptides andamino acids on effciency of rumen bacterial protein synthesis incontinuous culture. J. Dairy Sci. 65:226-234.

8. Dennis, S. M., T. G. Nagaraja, and E. E. Bartley. 1981. Effectsof lasalocid or monensin on lactate-producing or -using rumenbacteria. J. Anim. Sci. 52:418-426.

9. Dinius, D. A., M. E. Simpson, and P. B. Marsh. 1976. Effect ofmonensin fed with forage on digestion and the ruminal ecosys-tem of steers. J. Anim. Sci. 42:229-234.

10. Driessen, A. J. M., S. dejong, and W. N. Konings. 1987.Transport of branched-chain amino acids in membrane vesiclesof Streptococcus cremoris. J. Bacteriol. 169:5193-5200.

11. Fujimura, T., I. Yamato, and Y. Anraku. 1983. Mechanism ofglutamate transport in Escherichia coli B. 2. Kinetics of gluta-mate transport driven by artificially imposed proton and sodiumion gradient across the cytoplasmic membrane. Biochemistry22:1959-1965.

12. Hama, H., T. Shimamoto, M. Tsuda, and T. Tsuchiya. 1987.Properties of a Na+-coupled serine-threonine transport systemin Escherichia coli. Biochim. Biophys. Acta 905:231-239.

13. Hino, T., and J. B. Russell. 1985. Effect of reducing-equivalentdisposal and NADH/NAD on deamination of amino acids byintact rumen microorganisms and their cell extracts. Appl.Environ. Microbiol. 50:1368-1374.

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

15. Lundin, A., and A. Thore. 1975. Comparison of methods forextraction of bacterial adenine nucleotides determined by fireflyassay. Appl. Environ. Microbiol. 30:713-721.

16. Miura, H., M. Horiguchi, and T. Matsumoto. 1980. Nutritionalinterdependence among rumen bacteria, Bacteroides amylo-philus, Megasphaera elsdenii, and Ruminococcus albus. Appl.

Environ. Microbiol. 40:294-300.17. Richardson, L. F., A. P. Raun, E. L. Potter, C. 0. Cooley, and

R. P. Rathmacher. 1976. Effect of monensin on rumen fermen-tation in vitro and in vivo. J. Anim. Sci. 43:657-664.

18. Russell, J. B. 1987. A proposed model of monensin action ininhibiting rumen bacterial growth: effect on ion flux and proton-motive force. J. Anim. Sci. 64:1519-1525.

19. Russell, J. B. 1987. Effect of extracellular pH on growth andproton motive force of Bacteroides succinogens, a cellulolyticruminal bacterium. Appl. Environ. Microbiol. 53:2379-2383.

20. Russell, J. B., and H. J. Strobel. 1989. Effect of ionophores onruminal fermentation. Appl. Environ. Microbiol. 55:1-6.

21. Russell, J. B., H. J. Strobel, and G. Chen. 1988. Enrichment andisolation of a ruminal bacterium with a very high specificactivity of ammonia production. Appl. Environ. Microbiol.54:872-877.

22. Russell, J. B., H. J. Strobel, A. J. M. Driessen, and W. N.Konings. 1988. Sodium-dependent transport of neutral aminoacids by whole cells and membrane vesicles of Streptococcusbovis, a ruminal bacterium. J. Bacteriol. 170:3531-3536.

23. Stewart, C. S. 1977. Factors affecting the cellulolytic activity ofrumen contents. Appl. Environ. Microbiol. 33:497-502.

24. Strobel, H. J., and J. B. Russell. 1989. Non-proton-motive-force-dependent sodium efflux from the ruminal bacteriumStreptococcus bovis: bound versus free pools. Appl. Environ.Microbiol. 55:2664-2668.

25. Tsuchiya, T., M. Oho, and S. Shiota-Niiya. 1983. Lithiumion-sugar cotransport via the melibiose transport system inEscherichia coli. J. Biol. Chem. 258:12765-12767.

26. Tsuchiya, T., Y. Yamane, S. Shiota, and T. Kawasaki. 1984.Cotransport of proline and Li' in Escherichia coli. FEBS Lett.168:327-330.

27. Van Nevel, C. J., and D. I. Demeyer. 1977. Effect of monensinon rumen metabolism in vitro. Appl. Environ. Microbiol. 34:251-257.

VOL. 55, 1989 2663

on April 29, 2016 by guest

http://aem.asm

.org/D

ownloaded from