3123

1Presented at a symposium titled “Manipulation of RumenFermentation from the Perspective of the Forage and Feed Sourceand the Rumen Environment” at the ASAS 75th Southern Sect.Mtg., January, 1998, Little Rock, AR.

2Supported by the Univ. of Georgia Agric. Exp. Sta.3Corresponding address: 312 Animal and Dairy Sci. Complex.Received February 17, 1998.Accepted July 17, 1998.

Manipulation of Ruminal Fermentation with Organic Acids: A Review1,2

Scott A. Martin3

Department of Animal and Dairy Science, University of Georgia, Athens 30602-2771

ABSTRACT: The dicarboxylic acids aspartate,fumarate, and malate stimulate lactate utilization bythe predominant ruminal bacterium, Selenomonasruminantium. Malate stimulates lactate uptake by S.ruminantium more than does aspartate or fumarate,and it seems that malate and sodium are involved instimulating lactate utilization by this bacterium.Based on the ability of S. ruminantium to grow onmalate in the presence of extracellular hydrogen andproduce succinate, malate may be acting as anelectron sink for hydrogen in the succinate-propionatepathway used by S. ruminantium. Incorporation ofDL-malate into soluble starch and cracked cornfermentations with mixed ruminal microorganismschanged final pH, CH4, and VFA in a manneranalogous to ionophore effects. When compared witheither dicarboxylic acids or monensin alone, dicarbox-ylic acid plus monensin addition to cracked cornincubations stimulated the mixed ruminal microor-ganism fermentation to produce more propionate, less

lactate, and increased final pH. Reduced lactateconcentrations in dicarboxylic acid- and monensin-treated incubations most likely represents an additiveeffect of decreased lactate production by monensin-sensitive bacteria (i.e., Streptococcus bovis) andincreased lactate utilization by the monensin-resistantS. ruminantium. The inclusion of malate as a feedadditive into the diets of ruminants is currently noteconomically feasible; however, forages rich in organicacids might serve as vehicles for providing malate toruminants. When five alfalfa varieties and threebermudagrass hay varieties were surveyed for malatecontent, the concentration of malate in both plantspecies declined as maturity increased. However, after42 d of maturity, the concentration of malate in bothforages ranged between 1.9 and 4.5% of the DM. Theseresults suggest that the incorporation of foragevarieties that are high in malate may include malateeconomically into the diet and reduce losses associatedwith ruminal acidosis.

Key Words: Rumen, Bacteria, Dicarboxylic Acids, Fermentation

1998 American Society of Animal Science. All rights reserved. J. Anim. Sci. 1998. 76:3123–3132

Introduction

A goal of ruminant microbiologists and nutritionistsis to manipulate the ruminal microbial ecosystem toimprove the efficiency of converting feed to productsconsumable by humans. Much of the research in thepast 20 to 25 yr has focused on the effects ofantimicrobial compounds on ruminal fermentation.Specifically, ionophore antibiotics are used in feedlotdiets to reduce energy losses associated with methano-genesis in the rumen (Russell and Strobel, 1989).These compounds seem to inhibit hydrogen-producingmicroorganisms and Gram-positive lactate-producing

bacteria such as Streptococcus bovis (Russell, 1987;Russell and Strobel, 1989). Reductions in hydrogenproduction reduce ruminal methanogenesis and im-prove feed utilization by increasing the amount ofmetabolizable energy available to the animal aspropionate (Bergen and Bates, 1984; Russell andStrobel, 1989).

Scientists have recently become interested in evalu-ating alternative means for manipulating gastrointes-tinal microflora in livestock. Their motivation comesfrom increasing public scrutiny about the use ofantibiotics in the animal feed industry. However,compared with the efforts to detail the effects ofantimicrobial compounds on ruminal fermentation,little research has been conducted to evaluate alterna-tives to antimicrobial compounds. In the past 10 yr,interest has increased in direct-fed microbial ( DFM)products, and research has been conducted to examinethe effects of DFM on ruminant performance. Some ofthese products have shown promise in favorablyaltering ruminal fermentation and improving animalperformance, but the effects have been variable and

MARTIN3124

Figure 1. Effects of aspartate (Asp), fumarate (Fum),and malate (Mal) on lactate uptake by whole cells ofSelenomonas ruminantium (Nisbet and Martin, 1990).

Table 1. Effect of increasing concentrations ofL-malate on L-lactate uptake by Selenomonas

ruminantium (Nisbet and Martin, 1991)

aNanomoles per milligram of protein per minute, mean ± SD.*P < .05.

L-Malate, mM Specific activitya

0 .8 ± .04.03 1.4 ± .03*.06 2.7 ± .13*.12 3.2 ± .13*.48 5.6 ± .13*

2.5 8.5 ± .22*5.0 10.4 ± 1.40*

10.0 12.2 ± .09*

inconsistent (Martin and Nisbet, 1992). Organic acidscan stimulate the growth of the prominent ruminalbacterium, Selenomonas ruminantium, can favorablyalter the mixed ruminal microorganism fermentation,and can improve the performance of feedlot steers(Nisbet and Martin, 1990, 1991, 1993, 1994; Martinand Streeter, 1995; Callaway and Martin, 1996;Martin et al., 1999). Therefore, the objective of thispaper is to provide an overview of research withorganic acids and discuss their applications for beefand dairy cattle.

Pure Culture Studies

Batch Culture Experiments

Selenomonas ruminantium is a common Gram-negative ruminal bacterium that can account for up to51% of the total viable bacterial counts in the rumen(Caldwell and Bryant, 1966). This bacterium cangrow under a variety of dietary conditions and fermentmany different soluble carbohydrates (Hungate,1966). When S. ruminantium is grown in batchculture with glucose, homolactic fermentation occurs(Hobson, 1965). However, after the glucose isdepleted from the medium, S. ruminantium thenutilizes lactate as a C and energy source (Scheifingeret al., 1975). Only some strains of S. ruminantium(subspecies lactilytica) are able to ferment lactate(Stewart and Bryant, 1988).

Early research showed that S. ruminantium strainHD4 requires L-aspartate, CO2, p-aminobenzoic acid,and biotin for growth in a lactate-salts medium(Linehan et al., 1978). In addition, the requirementfor L-aspartate can be replaced by L-malate orfumarate. Because little work had been done in thisarea since the report by Linehan et al. (1978), mylaboratory initiated studies to evaluate the effects ofaspartate, fumarate, and malate on growth andlactate uptake by S. ruminantium HD4 (Nisbet andMartin, 1990). The growth of S. ruminantium HD4 inmedium that contained DL-lactate was stimulatedapproximately twofold by 10 mM L-aspartate,fumarate, or L-malate after 24 h of incubation (Nisbetand Martin, 1990). Both L-aspartate and fumarateincreased L-lactate uptake over 4-fold, and L-malatestimulated uptake over 10-fold (Nisbet and Martin,1990; Figure 1). In addition, different concentrationsof L-malate (.03 to 10 mM) stimulated L-lactateuptake by S. ruminantium HD4 in a dose-responsefashion (Nisbet and Martin, 1991; Table 1). D-Lactateuptake by S. ruminantium HD4 was also stimulatedby 10 mM L-malate (Nisbet and Martin, 1993).

Sodium concentrations between 25 and 100 mMstimulated L-lactate uptake by S. ruminantium HD4in the presence of 10 mM L-malate, whereas uptake inthe absence of L-malate was low regardless of the Na+

concentration (Nisbet and Martin, 1994; Figure 2).These results suggest that L-malate and Na+ playroles in stimulating L-lactate utilization by S.ruminantium HD4. Sodium is the predominant cationin the rumen, and concentrations range between 90 to150 mM (Durand and Kawashima, 1980).

L-Lactate uptake by cells grown in batch culturewas stimulated by 10 mM L-malate at extracellularpH values between 4.0 and 8.0 (Nisbet and Martin,1994). Because L-malate had little effect on L-lactateuptake for cells grown on soluble carbohydratescompared with lactate-grown cells, it seems that thestimulation due to malate is inducible (Nisbet andMartin, 1994). Based on sensitivity to several meta-bolic inhibitors, it has been suggested that proton

ORGANIC ACIDS AND RUMINAL FERMENTATION 3125

Figure 2. Effect of different concentrations of NaCl onL-lactate uptake by whole cells of Selenomonasruminantium strain HD4 in the absence (open bars) andpresence (shaded bars) of 10 mM L-malate (Nisbet andMartin, 1994).

gradients may be involved in L-lactate uptake instrain HD4, but more research is needed to defini-tively characterize the membrane bioenergetics in-volved in the transport of lactate (Nisbet and Martin,1994).

A more recent isolate, Selenomonas ruminantiumstrain H18, also requires organic acids to grow onlactate (Strobel and Russell, 1991a). However, strainH18 differs from strain HD4 in that lactate could beused to support growth only when Na+ and aspartatewere added to the medium. When malate or fumaratereplaced aspartate, Na+ was not required (Strobel andRussell, 1991a). Involvement of a proton symport inthe transport of fumarate and malate in strain H18has been suggested (H. J. Strobel, personal communi-cation).

Because they are intermediates of the citric acidcycle, fumarate and malate—four-carbon dicarboxylicacids—are commonly found in biological tissues (Leh-ninger, 1975). Some strictly anaerobic bacteria use areductive or reverse citric acid cycle known as thesuccinate-propionate pathway to synthesize succinateand(or) propionate (Gottschalk, 1986). Both malateand fumarate are key intermediates in the succinate-propionate pathway, and S. ruminantium uses thispathway (Gottschalk, 1986). The fact that dicarbox-ylic acids, especially malate and fumarate, stimulatelactate utilization is consistent with the presence ofthis pathway in this ruminal anaerobe.

Linehan et al. (1978) hypothesized that aspartate,fumarate, and malate probably stimulate the growthof strain HD4 on lactate because they overcome thedeficiency of oxaloacetate associated with gluconeo-genesis, either by converting to it (in the case offumarate and malate) or by sparing it (in the case ofaspartate). Growth of strain HD4 on lactate resultedin lower cellular carbohydrate concentrations whencompared with glucose-grown cells, and there waslittle change in protein (Martin and Park, 1996).Incorporation of either dicarboxylic acid into themedium with lactate resulted in protein and carbohy-drate concentrations as high or higher than lactate-grown cells. When strain HD4 was grown on eachdicarboxylic acid plus 1 atm of H2, protein contentremained fairly high. However, compared with cellsgrown on dicarboxylic acids plus lactate, there was alarge decrease in carbohydrate levels. Because theaddition of dicarboxylic acids to lactate mediumtended to increase cellular carbohydrate levels, itseems likely that this increase in carbohydrate maylead to higher oxaloacetate concentrations within thecell. Therefore, these results support the hypothesisthat dicarboxylic acids stimulate the growth of S.ruminantium on lactate because they overcome thedeficiency of oxaloacetate associated with gluconeo-genesis (Linehan et al., 1978).

In addition to overcoming a deficiency of oxaloace-tate, malate may also provide an electron sink for H2that allows for increased lactate utilization by S.ruminantium HD4 (Nisbet and Martin, 1991). WhenS. ruminantium HD4 and H18 were incubated in amedium that contained aspartate, fumarate, or malatein the presence of 1 atm of H2, significant growth wasobserved (Martin and Park, 1996). Succinate was themajor end product of these fermentations. When cellswere incubated with lactate plus 1 atm of H2, growthwas minimal, and little lactate was fermented (Mar-tin and Park, 1996). These results suggest that ifreducing equivalents (H2) accumulate in the medium,the S. ruminantium lactate dehydrogenase cannotconvert lactate to pyruvate. Therefore, the observationthat both S. ruminantium strains are unable toferment lactate in the presence of H2 but that they canferment aspartate, fumarate, and malate in thepresence of H2 supports the suggestion that malate(organic acids) may provide an electron sink for H2that allows for increased lactate utilization by S.ruminantium.

Even though the above studies documented theeffects of organic acids on lactate utilization by S.ruminantium, little information was available describ-ing lactate plus malate utilization by this bacterium.Therefore, we conducted a study to evaluate factorsaffecting utilization of these two organic acids by bothS. ruminantium HD4 and H18 (Evans and Martin,1997). When strains HD4 and H18 were grown inbatch culture on DL-lactate and DL-malate, bothstrains co-utilized both organic acids for the initial 20

MARTIN3126

to 24 h of incubation, and acetate, propionate, andsuccinate accumulated. However, when malate andsuccinate concentrations reached 7 mM, malate utili-zation ceased and, with strain H18, there was acomplete cessation of DL-lactate utilization. Thereason for the accumulation of succinate by bothstrains in the presence of malate is unclear. Becauseclarified ruminal fluid can serve as a source ofvitamins and unidentified growth factors, 20% clari-fied ruminal fluid was added to the fermentation ofDL-lactate plus DL-malate by both S. ruminantiumstrains. However, succinate still accumulated in themedium (Evans and Martin, 1997).

Experiments were conducted to evaluate the effectsof adding mixed ruminal bacteria to the lactate plusmalate fermentation by strain H18 (Evans andMartin, 1997). Prior to the mixed bacterial inocula-tion at 30 h, lactate and malate utilization by strainH18 resulted in the concomitant production of acetate,propionate, and succinate as well as in the inhibitionof lactate and malate utilization at 18 h. Following theaddition of mixed ruminal bacteria to the fermenta-tion at 30 h, succinate, lactate, and malate wereutilized, and acetate and propionate concentrationsincreased. Even though these results do not exclude S.ruminantium strains as being responsible for suc-cinate decarboxylation within the rumen, other rumi-nal bacteria or the mixed population may be moreactive in this capacity than either S. ruminantiumstrain alone. In addition, the mixed bacterial popula-tion may provide growth factors not provided in ourmedium that are required to decarboxylate succinate.

Selenomonas ruminantium has been implicated asbeing the microorganism primarily responsible forsuccinate decarboxylation within the rumen (Strobeland Russell, 1991b). Other members of the genera arealso capable of succinate decarboxylation. A newlyisolated species, Selenomonas acidaminovorans, isable to grow solely on succinate, and decarboxylationof this organic acid is thought to result in a sodiumefflux and formation of a sodium gradient (Guang-sheng et al., 1992). However, organisms even moreactive in succinate decarboxylation seem to exist inthe rumen. Schwartzia succinivorans and Suc-ciniclasticum ruminis are both capable of utilizingsuccinate as the sole energy source (van Gylswyk,1995; van Gylswyk et al., 1997).

Continuous Culture Experiments

In general, the liquid dilution rate within therumen can range from .05 to .10 h−1 (Hungate, 1966).Most studies aimed at evaluating lactate utilization byS. ruminantium have been conducted in batch culture,and few experiments have been performed in continu-ous culture. Therefore, experiments were conducted toexamine the effect of extracellular pH, lactate concen-tration, and malate addition on the growth of S.ruminantium HD4 in continuous culture (Evans and

Martin, 1997). When S. ruminantium HD4 was grownon lactate at pH 6.8, the primary end products wereacetate and propionate with all concentrations oflactate (Table 2). Little succinate or malate wasproduced. Even though not all lactate was utilized, aslactate concentration increased there was a cor-responding increase in optical density at 600 nm( OD600) , protein, and carbohydrate. These resultssuggested that lactate was limiting growth. Lactateutilization ranged between 35 and 50%.

Selenomonas ruminantium HD4 was unable togrow (culture washout) on 6 mM lactate whenextracellular pH was reduced to 5.5 (Table 2). Growthdid occur on 30 or 54 mM lactate at this pH, andacetate and propionate were the main end productsthat were produced. Little malate or succinate ac-cumulated. Bacterial protein and OD600 increased aslactate concentration increased, and there was adecrease in cellular carbohydrate. When 8 mM malatewas added to the growth medium, strain HD4 wasable to grow on 6 mM lactate at pH 5.5 and 80% of thelactate was utilized. Acetate, propionate, and suc-cinate were the primary fermentation productsproduced with all three lactate concentrations in thepresence of malate. Malate addition increased theamount of lactate utilized, OD600, and concentrationsof protein and cellular carbohydrate synthesized bystrain HD4. Lactate utilization ranged between 77and 80% in the presence of malate compared with only40 and 70% in the absence of malate. Malateutilization ranged between 51 and 64%. Similar effectswere seen when strain HD4 was grown at a dilutionrate of .10 h−1 (Evans and Martin, 1997).

When domestic ruminants (beef and dairy cattle)are fed diets high in rapidly fermentable carbohy-drates (i.e., cereal grains), lactate can accumulateand decrease ruminal pH (Slyter, 1976; Russell andHino, 1985; Russell and Strobel, 1989). Lactateconcentrations as high as 29 mM have been observedwith these types of diets (Counotte et al., 1981). Iflactate concentrations remain elevated, ruminal pHdrops below 6.0, and this leads to a variety ofmicrobial and physiological problems, includingdecreased fiber digestion, decreased digesta turnover,decreased salivation, rumen ulceration, founder, anddeath (Slyter, 1976; Russell and Hino, 1985).

Based on our continuous culture results, it seemsthat malate enhances the ability of strain HD4 togrow on all three lactate concentrations at an extracel-lular pH of 5.5 (Table 2). These results are consistentwith the observation that malate treatment increasedfinal pH and decreased lactate concentrations inmixed ruminal microorganism fermentations ofcracked corn and soluble starch (Martin and Streeter,1995; Callaway and Martin, 1996). Therefore, addingmalate to the diets of ruminants fed high levels ofrapidly fermentable carbohydrates may improve theability of S. ruminantium HD4 to utilize lactate at pH≤ 6.0.

ORGANIC ACIDS AND RUMINAL FERMENTATION 3127

Table 2. Effect of extracellular pH, lactate concentration, and malate on growth of Selenomonas ruminantiumstrain HD4 in continuous culture (dilution rate = .05 h−1) (Evans and Martin, 1997)

aValues in parentheses represent the percentage of substrate utilized.bCulture washout.

Substrate concentration Residual substrate, optical density at 600 nm (OD600) , and end products

Extracellular Lactate, Malate, Lactate, Malate, Protein,Carbohy-

drate, Succinate, Acetate, Propionate,pH mM mM mM mM OD600 mg/L mg/L mM mM mM

6.8 6 0 3.0 (50)a .1 .121 31.9 17.7 .1 2.3 1.96.8 35 0 18.1 (48) .1 .322 59.8 23.9 .3 5.4 8.26.8 69 0 44.6 (35) .1 .351 64.5 27.1 0 5.6 9.85.5 6 0 —b — — — — — — —5.5 30 0 8.8 (70) .2 .489 80.9 60.2 .1 9.9 16.55.5 54 0 32.1 (40) 0 .514 95.0 35.6 .1 7.9 16.35.5 6 8 1.1 (80) 3.2 (62) .320 59.1 35.3 3.6 4.5 2.75.5 28 8 6.3 (78) 4.0 (51) .624 154 54.4 3.7 10.7 14.25.5 57 8 12.8 (77) 2.8 (64) .726 138.0 66.8 1.9 13.1 20.7

Mixed Culture Studies

Based on the stimulation of lactate utilization bydicarboxylic acids in S. ruminantium and becauseinformation was limited detailing the effects of organicacids on ruminal fermentation, experiments wereconducted to evaluate the effects of L-aspartate,fumarate, and DL-malate on in vitro mixed ruminalmicroorganism fermentation (Martin and Streeter,1995; Callaway and Martin, 1996). Fermentation ofcracked corn in the presence of 8 or 12 mM DL-malateresulted in an increase in final pH and propionateconcentration (Martin and Streeter, 1995). Total VFAtended to increase, and final concentrations of L-lactate numerically decreased. In the case of solublestarch, 8 and 12 mM DL-malate caused a decrease inCH4 concentration. When only ruminal fluid (noadded anaerobic medium) was used as the inoculumrather than 20% ruminal fluid medium, similarresults for final pH, propionate, L-lactate, and totalVFA were observed for soluble starch and cornincubations treated with DL-malate. These changes infermentation products are analogous to ionophoreeffects (Russell and Strobel, 1989).

To determine whether organic acids plus monensinhave an additive effect on ruminal fermentation, theeffects of organic acid (L-aspartate, fumarate, or DL-malate) and monensin treatment on in vitro mixedruminal microorganism fermentation were evaluated(Callaway and Martin, 1996). Monensin (5 ppm)altered the fermentation as expected by decreasingCH4 concentrations and increasing propionate concen-trations. The addition of 8 and 12 mM organic acids tocracked corn fermentations increased final pH, tendedto increase total gas production and CO2 concentra-tion, and decreased the acetate:propionate ratio.Organic acids tended to decrease CH4 concentrationsand H2 concentration was not altered. DL-Malateaddition at all levels reduced L-lactate accumulation.

Monensin is thought to alter ruminal fermentationby inhibiting Gram-positive bacteria, specificallyStreptococcus bovis, that are responsible for much ofthe ruminal lactate, ammonia, and H2 production(Dennis et al., 1981; Russell and Strobel, 1989). Thereduction of ruminal lactate production is thought tohelp increase ruminal pH owing to the low pKa oflactate compared with the VFA normally associatedwith ruminal fermentation (pKa 3.9 vs 4.8 foracetate) (Russell and Hino, 1985). Therefore, due toinhibition of lactate production by Streptococcus bovisand increased lactate utilization by monensin-resis-tant Selenomonas ruminantium, it was hypothesizedthat an additive effect of organic acid and monensintreatment on mixed ruminal microorganism fermenta-tion might occur. An additive effect was observed withincubations that contained DL-malate and monensinas well as fumarate and monensin (Callaway andMartin, 1996).

Lactate levels were lowest in incubations contain-ing DL-malate and monensin; however, final pH washighest in incubations containing only DL-malate(Callaway and Martin, 1996). Therefore, despite thedecrease in lactate accumulation in organic acid-treated fermentations, not all of the increase in finalpH can be attributed to a reduction of lactate levels.Sodium bicarbonate has been incorporated into dairycattle diets to buffer ruminal pH for over 20 yr. Somebelieve that bicarbonate increases the buffering capac-ity of ruminal fluid, in part, by increasing the amountof dissolved CO2. In addition to stimulating lactateutilization by S. ruminantium, organic acid addition tomixed ruminal microorganism fermentations in-creased CO2 concentration in vitro (Callaway andMartin, 1996). Carbon dioxide is an end product oflactate fermentation to propionate via the succinate-propionate pathway that is utilized by S. ruminan-tium (Gottschalk, 1986). Therefore, organic acidtreatment may act to increase ruminal pH by a dualmechanism of increased lactate utilization and CO2

MARTIN3128

production by S. ruminantium. Even though the gasanalysis did not always show an increase in CO2concentration, the total gas measured was found to becomposed of 95% CO2 and 5% CH4 (Callaway andMartin, 1996). Perhaps the method used to quantifyCO2 (thermal conductivity gas chromatography) didnot possess the sensitivity needed to detect changes inthe amount of CO2 produced, and perhaps the increasein total gas production was primarily composed ofCO2. When S. ruminantium is grown in medium thatcontains lactate plus organic acids, we routinelyobserve more gas pressure in these cultures comparedwith lactate-only cultures. Therefore, it seems thatstimulation of S. ruminantium in mixed culture byorganic acid addition helps to increase ruminal pH byincreasing lactate utilization as well as the concentra-tion of CO2.

As detailed above, organic acids have been sug-gested to act as an electron sink for S. ruminantium(Nisbet and Martin, 1991; Martin and Park, 1996).Treatment of mixed ruminal microorganism fermenta-tions with DL-malate yielded responses similar tothose of ionophores (i.e., increased propionate,decreased CH4, and decreased lactate), suggestingthat organic acids have an effect on electron flow(Martin and Streeter, 1995). Ionophore effects closelyassociated with electron redistribution (decreasedlactate and increased propionate) were enhanced byorganic acid treatment (Callaway and Martin, 1996).Therefore, by providing an electron sink in the form oforganic acids, the effects of monensin are enhanced insome cases.

All of the studies reviewed up to now have dealtwith examining the effects of organic acids on ruminalmicroorganism fermentation. Experiments have alsobeen conducted to determine the effects of cellobioseand monensin on the in vitro fermentation of all threeorganic acids by mixed ruminal bacteria (Callawayand Martin, 1997). Cellobiose addition to organic acidfermentations increased the rate of organic acidutilization by the mixed bacterial population. TotalVFA concentrations were increased by cellobioseaddition to all fermentations. A lag period ( ≤ 4 h)occurred in monensin-treated fermentations beforeorganic acids were utilized; however, total VFA wereincreased and the acetate:propionate ratio wasdecreased by monensin treatment. When cellobioseand monensin were added together, propionateproduction and organic acid utilization were increased.Disappearance rates of organic acids and concentra-tions of total VFA were found to be highest, and theacetate:propionate ratio was the lowest, in incubationstreated with cellobiose plus monensin. Therefore, thebeneficial effects (i.e., increased total VFA andpropionate concentrations) of organic acids on fermen-tation by mixed ruminal bacteria are apparentlyenhanced by the addition of cellobiose and monensin.

In Vivo Studies

Limited in vivo research has been conducted toevaluate the effects of organic acids on ruminantperformance. Kung et al. (1982) reported that feeding140 g of malate per day increased milk persistency inlactating cows and increased total VFA during earlylactation. Feeding malate to Holstein bull calvesimproved ADG and feed efficiency but had little effecton blood serum constituents (Sanson and Stallcup,1984). Even though in vitro studies have shown thatDL-malate favorably alters ruminal fermentation(Martin and Streeter, 1995; Callaway and Martin,1996), little information is available detailing theeffects of DL-malate on beef cattle performance.Therefore, the effects of DL-malate on ruminal traitsas well as effects of supplementing finishing diets withthis organic acid on feedlot cattle performance weredetermined (Streeter et al., 1994; Martin et al.,1999).

To determine the effects of DL-malate on ruminalmetabolism, four steers (485 ± 24.8 kg) equipped withruminal cannulas were fed an 80% rolled grain (75%corn:25% wheat) diet twice daily with a DMI equal to2.0% of BW (Streeter et al., 1994; Martin et al.,1999). DL-Malate was infused into the rumen on twoconsecutive days in 500 mL of phosphate buffer toprovide 0, 27, 54, or 80 g of DL-malate/d or finalruminal concentrations of 0, 4, 8, or 12 mM with anassumed ruminal volume of 50 L. When only 500 mLof phosphate buffer was infused into the rumen (0mM DL-malate), ruminal pH declined to 5.49 at 1 hafter infusion and increased to 6.29 at 12 h (Figure3). These results are consistent with the pH valuesassociated with subclinical acidosis (Britton andStock, 1986; Nocek, 1997; Owens et al., 1998). Therewas no decrease ( P < .05) in ruminal pH at 1 h afterinfusion of all three concentrations of DL-malate.When compared with the 0-mM DL-malate treatment,ruminal pH was also higher ( P < .05) at 2 and 4 h inthe presence of all three DL-malate concentrations.Ruminal pH was always greater than 6.0 in thepresence of 12 mM DL-malate over the 12-h samplingperiod. DL-Malate treatment linearly decreased ( P <.10) total VFA and tended to linearly increase ( P =.10) acetate concentration. There was no treatmenteffect on propionate, butyrate, or L-lactate concentra-tions or the acetate:propionate ratio. Ruminal concen-trations of acetate, butyrate, and total VFA wereincreased by malate treatment in early-lactation dairycows, but malate had no effect on ruminal pH (Kunget al., 1982). However, ruminal fluid samples werecollected by stomach tube, and ruminal pH valuesranged between 6.9 and 7.1 in the study by Kung et al.(1982). These results suggest that either the diet didnot induce acidosis or the ruminal fluid samples werecontaminated with saliva. Kung et al. (1982) alsoreported that malate treatment had no effect on VFAconcentrations in midlactation cows.

ORGANIC ACIDS AND RUMINAL FERMENTATION 3129

Figure 3. Effect of DL-malate treatment on ruminalpH. Treatments were 0 mM DL-malate (…), 4 mM DL-malate (ÿ), 8 mM DL-malate (⁄), and 12 mM DL-malate(π). Error bars represent standard error of the mean(Streeter et al., 1994; Martin et al., 1999).

The lack of significant accumulation of L-lactate isconsistent with the etiology of subclinical acidosis(Nocek, 1997; Owens et al., 1998). It has beensuggested that total acid load and not just lactatealone is responsible for acidosis (Owens et al., 1998).In our study, total VFA concentration in the controlanimals was 157.9 mM, and increasing amounts ofDL-malate reduced ( P < .10) these concentrations(Martin et al., 1999). However, it is unlikely that thisdecrease in total VFA concentrations was solelyresponsible for the increase in ruminal pH. In additionto stimulating lactate utilization by S. ruminantium,as detailed above, organic acid addition to in vitromixed ruminal microorganism fermentations in-creased CO2 concentrations (Callaway and Martin,1996). Therefore, malate treatment may act toincrease the pH of ruminal contents by a dualmechanism of increased lactate utilization and CO2production by S. ruminantium.

Three finishing studies were conducted to deter-mine the effects of feeding DL-malate on growth rateand feed efficiency (Martin et al., 1999). In a98-d experiment, 33 crossbred steers were randomlyallotted to three DL-malate levels (0, 40, and 80 g/d).Steers were fed a rolled corn-based diet twice daily.After 84 d on feed, feed efficiency (gain:feed) tendedto be improved with more DL-malate (linear, P < .10)and was 8.1% greater ( P < .05) for DL-malatecompared with control. Average daily gain linearly

increased ( P < .05) with more DL-malate and was8.6% greater ( P = .10) for DL-malate than for thecontrol. After 98 d on feed, ADG was linearly ( P = .09)increased by DL-malate, with the greatest increaseoccurring with 80 g of DL-malate. In the secondperformance study, 27 Angus steers were randomlyallotted to three DL-malate concentrations (0, 60, and120 g/d). Steers were fed diets used in the98-d experiment twice daily, with DL-malate includedduring the 10-d step-up period. During the10-d step-up period, feed efficiency (gain:feed) andADG linearly increased ( P < .05) with more DL-malate. DL-Malate had little effect on steer and heiferperformance or plasma constituents in a 113-d finish-ing study. Furthermore, DL-malate had no effect oncarcass characteristics in all three finishing studies.Collectively, these results suggest that feeding DL-malate to cattle that are consuming high-grain dietsmay alleviate subclinical acidosis and improve animalperformance in two out of three finishing studies.

Organic acids, specifically citrate and malate,stimulate salivation in the human mouth (Clark etal., 1961; Hay et al., 1962; Slack et al., 1964; Annerothet al., 1980). These organic acids have been used intablets to stimulate salivation in order to removedebris from teeth. To my knowledge, no studies havebeen conducted to evaluate the effect of malate onsalivation in ruminants. However, in addition tostimulating lactate utilization and CO2 production byS. ruminantium in the rumen, it might also bepossible that adding malate to the diet stimulatessalivation in ruminants. Increasing salivary flow tothe rumen would help buffer the rumen and alleviateacidosis. Research is needed to address the role ofmalate in ruminant salivation.

Forages and Organic Acids

Because the cost of supplementing diets with DL-malate is estimated to range between $.09 to $.19/dper animal under feedlot conditions (Streeter et al.,1994; Martin et al., 1999), the inclusion of malate as afeed additive in the diets of ruminants may not beeconomically feasible. Plants are a rich source ofnutrients that can be utilized by both the ruminantanimal and mixed ruminal microbial population.Intermediates of the citric acid cycle accumulate inplant tissue and may represent as much as 10% of theDM of grasses (Stout et al., 1967; Bohman et al.,1983). Malate can amount to as much as 1.5% of theDM of mature grasses (Bohman et al., 1983).Therefore, forages that are high in organic acids mightprovide a vehicle for the inclusion of malate inproduction ruminant diets. To address this possibleuse of forages, a study was conducted to determine theconcentrations of malate present in five alfalfa varie-ties (‘Alfagraze’, ‘Apollo Supreme’, ‘Cimarron’, ‘Crock-ett’, and ‘Magnum III’) and three bermudagrass hay

MARTIN3130

Figure 4. Regression analysis of malate concentration(percentage of DM) in five alfalfa varieties (Callaway etal., 1997).

(‘Coastal’, ‘Tifton-78’, and ‘Tifton-85’) varieties atdifferent stages of maturity (Callaway et al., 1997).

Samples were collected from replicated plots (n =3) of each alfalfa variety at 9, 18, 28, 35, and 42 d ofmaturity; bermudagrass hay samples were compositedfrom six bales of each variety from two cuttings stagedto be harvested at 27 and 41 d of maturity (Callawayet al., 1997). Alfalfa samples were collected fresh fromlate June to August 1994. Bermudagrass hay sampleswere obtained from well-established .81-ha pastures ofeach variety (Coastal, Tifton-78, and Tifton-85) at theUniversity of Georgia Coastal Plain Station at Tifton.Pastures were mowed and fertilized (24-6-12 of N-P-K; 336 kg/ha) on May 11, 1994. Each pasture wasmowed and the residue removed on June 7, half ofeach pasture was mowed and the residue removed onJune 21, and 4 to 6 wk-maturity hays were harvestedon July 18 and baled on July 19.

As maturity increased, malate concentrationdeclined in both plant species. Regression analysisindicated that malate concentrations in the alfalfavarieties decreased with increasing maturity at differ-ent rates (Figure 4). Cimarron had the highest rate ofdecline, followed by Apollo Supreme, Alfagraze, Crock-ett, and Magnum III. No differences in malateconcentrations were observed at 9 d among the fivealfalfa varieties. Malate concentrations were numeri-cally higher in two alfalfa varieties (Crockett andMagnum III) at 35 and 42 d of maturity than in allother alfalfa varieties. Malate concentrations in ber-mudagrass (41 d) were lower than malate concentra-

tions in all alfalfa varieties at 42 d of maturity. Malatedeclined with increasing maturity in the Coastal andTifton-78 varieties.

Based on the presence of malate in forages, it is ofinterest to determine the potential ruminal concentra-tion of malate after consumption of these forages byproduction ruminants. Given that malate concentra-tions ranged between 2.9 and 4.5% of alfalfa DM at 42d of maturity, one can estimate how much malatemight be available in the rumen of a dairy cow fedalfalfa at 42 d of maturity (Callaway et al., 1997). Ifthe amount of alfalfa consumed per day is 6.0 kg in atotal mixed ration (West et al., 1997) and the malateconcentration in Alfagraze is 2.9%, the animal con-sumes 174 g of malate. If the ruminal volume isapproximately 70 L, then the intraruminal concentra-tion of malate would be 2.5 g/L or 18.6 mM (molecularmass of malate is 134.1 g/L; Sigma Chemical Co., St.Louis, MO). Using similar calculations and 4.5% ofDM malate, Magnum III would provide 29 mMintraruminal malate. Malate concentrations between.03 and 10 mM increased lactate uptake by S.ruminantium in a dose-response fashion, and the 10mM level was the most stimulatory (Nisbet andMartin, 1991). Therefore, all alfalfa varieties seem toprovide adequate concentrations of malate to max-imally stimulate lactate utilization by S. ruminan-tium. However, because of ruminal dilution rate aswell as malate utilization by ruminal microorganisms(Russell and Van Soest, 1984; Callaway and Martin,1997), it is unlikely that all malate would be readilyavailable. In addition, release of this organic acid maybe dependent on at least some plant cell-wall diges-tion. In vitro studies have shown that 7.5 mM malateis completely fermented within 10 to 24 h by mixedruminal microorganisms (Russell and Van Soest,1984; Callaway and Martin, 1997). Collectively, theseresults suggest that selecting for and incorporatingforage varieties that are high in malate into theruminant diet may provide a vehicle for economicallyincluding malate in the diet.

Final Thoughts

Compared with the amount of research, both in vivoand in vitro, that has been conducted with other feedadditives, little research has been conducted withorganic acids. Given the increasing concern and(or)criticism regarding the feeding of antibiotics and thepotential for selection of antibiotic-resistant bacteriaand resistance transfer to pathogens (Nikolich et al.,1994), alternatives to antimicrobial compounds needto be investigated. Organic acids potentially providean alternative to currently used antimicrobial com-pounds by stimulating rather than inhibiting specificruminal microbial populations. By initially studyingthe effects of dicarboxylic acids on lactate utilizationby S. ruminantium, we have progressed to mixed

ORGANIC ACIDS AND RUMINAL FERMENTATION 3131

culture fermentations, and limited in vivo studiesseem promising. This approach to understanding thedetails of the ruminal fermentation is needed ifnutritionists and microbiologists hope to successfullymanipulate ruminal microorganisms in the future.

Malate can be purchased in bulk quantities, and, atcurrent prices, the use of malate as a feed additive (80g/d per animal) is estimated to cost $.09 to $.19/d perhead under feedlot conditions. At present, this costmay prohibit the inclusion of malate in the diets offeedlot or dairy cattle. However, by selecting for andincorporating forage varieties that are high in malateinto the ruminant diet, ruminal fermentation might beimproved. Perhaps the high-malate concentrationsobserved in alfalfa account for some of the benefitsassociated with this forage. More research is needed toevaluate whether ensiling has any effect on foragemalate concentrations.

According to Owens et al. (1998), acute andsubacute acidosis continues to plague feedlotruminants. It has been conservatively estimated thatmonetary losses attributed to ruminal acidosis in U.S.feedlots are between $60 to $100 million/yr (Martin,1992). Economic losses associated with acidosis in thedairy industry are not available, but, because of thehigh-energy demands by high-producing dairy cows,these losses are probably significant (Hinders, 1995;Nocek, 1997). Owing to the growing focus on foodsafety by national media and consumers, it is impera-tive that microbiologists and nutritionists explorealternatives to antimicrobial compounds to addressthe acidosis problem. Based on the results from ourstudies, an organic acid, like malate, would be adesirable alternative because there is no risk ofdeveloping antibiotic resistance or having unwantedresidues appear in either meat or milk products.

Implications

Initial research characterized the effects of organicacids (aspartate, fumarate, and malate) on lactateutilization by the predominant ruminal bacterium,Selenomonas ruminantium. Based on this pure-cultureresearch, progress has been made in demonstratingthat some of these organic acids, primarily malate,alter mixed ruminal microorganism fermentation in amanner that is similar to that of ionophores. Malatewas also effective in reducing the drop in ruminal pHnormally seen 1 to 2 h after feeding a high-grain dietand improved cattle performance. Therefore, sup-plementing finishing diets or high-producing dairycattle diets with malate might be effective in reducingsubclinical acidosis.

Literature Cited

Anneroth, G., G. Nordenram, and S. Bengtsson. 1980. Effect ofsaliva stimulants (Hybrin and malic acid) on cervical rootsurfaces in vitro. Scand. J. Dent. Res. 88:214−218.

Bergen, W. G., and D. B. Bates. 1984. Ionophores: Their effect onproduction efficiency and mode of action. J. Anim. Sci. 58:1465−1483.

Bohman, V. R., F. P. Horn, B. A. Stewart, A. C. Mathers, and D. L.Grunes. 1983. Wheat pasture poisoning. I. An evaluation ofcereal pastures as related to tetany in beef cows. J. Anim. Sci.57:1352−1363.

Britton, R. A., and R. A. Stock. 1986. Acidosis, rate of starchdigestion and intake. In: Feed Intake Symp. Proc. pp 125−137.Stillwater, OK.

Caldwell, D. R., and M. P. Bryant. 1966. Medium without rumenfluid for nonselective enumeration and isolation of rumen bac-teria. Appl. Microbiol. 14:794−801.

Callaway, T. R., and S. A. Martin. 1996. Effects of organic acid andmonensin treatment on in vitro mixed ruminal microorganismfermentation of cracked corn. J. Anim. Sci. 74:1982−1989.

Callaway, T. R., and S. A. Martin. 1997. Effects of cellobiose andmonensin on in vitro fermentation of organic acids by mixedruminal bacteria. J. Dairy Sci. 80:1126−1135.

Callaway, T. R., S. A. Martin, J. L. Wampler, N. S. Hill, and G. M.Hill. 1997. Malate content of forage varieties commonly fed tocattle. J. Dairy Sci. 80:1651−1655.

Clark, R., D. I. Hay, C. J. Schram, and B. J. Wagg. 1961. Removal ofcarbohydrate debris from the teeth by salivary stimulation. Br.Dent. J. 111:244−248.

Counotte, G.H.M., R.A. Prins, R.H.A.M. Janssen, and M.J.A. DeBie.1981. Role of Megasphaera elsdenii in the fermentation of DL-[213]lactate in the rumen of dairy cattle. Appl. Environ.Microbiol. 42:649−655.

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

Durand, M., and R. Kawashima. 1980. Influence of minerals inrumen microbial digestion. In: Y. Ruckebusch and P. Thivend(Ed.) Digestive Physiology and Metabolism in Ruminants. pp375−408. AVI Publishing, Westport, CT.

Evans, J. D., and S. A. Martin. 1997. Factors affecting lactate andmalate utilization by Selenomonas ruminantium. Appl. Envi-ron. Microbiol. 63:4853−4858.

Gottschalk, G. 1986. Bacterial Metabolism (2nd Ed.). Springer-Verlag, New York.

Guangsheng, C., C. M. Plugge, W. Roelofson, F. P. Houwen, andA.J.M. Stams. 1992. Selenomonas acidaminovorans sp. nov., aversatile thermophilic proton-reducing anaerobe able to growby decarboxylation of succinate to propionate. Arch. Microbiol.157:169−175.

Hay, D. I., B.R.W. Pinsent, C. J. Schram, and B. J. Wagg. 1962. Theprotective effect of calcium and phosphate ions against aciderosion of dental enamel and dentine. Br. Dent. J. 112:283−287.

Hinders, R. 1995. Rumen acidosis concerns increase as per cow milkproduction rises. Feedstuffs 67:11.

Hobson, P. N. 1965. Continuous culture of some anaerobic andfacultative anaerobic rumen bacteria. J. Gen. Microbiol. 38:167−180.

Hungate, R. E. 1966. The Rumen and Its Microbes. Academic Press,New York.

Kung, L., Jr., J. T. Huber, J. D. Krummrey, L. Allison, and R. M.Cook. 1982. Influence of adding malic acid to dairy cattlerations on milk production, rumen volatile acids, digestibility,and nitrogen utilization. J. Dairy Sci. 65:1170−1174.

Lehninger, A. L. 1975. Biochemistry (2nd Ed.). Worth Publishers,New York.

Linehan, B., C. C. Scheifinger, and M. J. Wolin. 1978. Nutritionalrequirements of Selenomonas ruminantium for growth on lac-tate, glycerol, or glucose. Appl. Environ. Microbiol. 35:317−322.

Martin, S. A. 1992. Potential for manipulating gastrointestinalmicroflora: Alternatives to antibiotics and ionophores? In: Proc.29th Annu. Ga. Nutr. Conf. Feed Ind., Atlanta. Univ. of Geor-gia, Athens. pp 28−38.

Martin, S. A., and D. J. Nisbet. 1992. Effect of direct-fed microbialson rumen microbial fermentation. J. Dairy Sci. 75:1736−1744.

MARTIN3132

Martin, S. A., and C. M. Park. 1996. Effect of extracellular hydrogenon organic acid utilization by the ruminal bacterium Selenomo-nas ruminantium. Curr. Microbiol. 32:327−331.

Martin, S. A., and M. N. Streeter. 1995. Effect of malate on in vitromixed ruminal microorganism fermentation. J. Anim. Sci. 73:2141−2145.

Martin, S. A., M. N. Streeter, D. J. Nisbet, G. M. Hill, and S. E.Williams. 1999. Effects of DL-malate on ruminal metabolismand performance of cattle fed a high concentrate diet. J. Anim.Sci. 77:(In press).

Nikolich, M. P., G. Hong, N. B. Shoemaker, and A. A. Salyers. 1994.Evidence for natural horizontal transfer of tetQ between bac-teria that normally colonize humans and bacteria that normallycolonize livestock. Appl. Environ. Microbiol. 60:3255−3260.

Nisbet, D. J., and S. A. Martin. 1990. Effect of dicarboxylic acids andAspergillus oryzae fermentation extract on lactate uptake bythe ruminal bacterium Selenomonas ruminantium. Appl. Envi-ron. Microbiol. 56:3515−3518.

Nisbet, D. J., and S. A. Martin. 1991. Effect of a Saccharomycescerevisiae culture on lactate utilization by the ruminal bac-terium Selenomonas ruminantium. J. Anim. Sci. 69:4628−4633.

Nisbet, D. J., and S. A. Martin. 1993. Effects of fumarate, L-malate,and an Aspergillus oryzae fermentation extract on D-lactateutilization by the ruminal bacterium Selenomonas ruminan-tium. Curr. Microbiol. 26:133−136.

Nisbet, D. J., and S. A. Martin. 1994. Factors affecting L-lactateutilization by Selenomonas ruminantium. J. Anim. Sci. 72:1355−1361.

Nocek, J. E. 1997. Bovine acidosis: Implications on laminitis. J.Dairy Sci. 80:1005−1028.

Owens, F. N., D. S. Secrist, W. J. Hill, and D. R. Gill. 1998. Acidosisin cattle: A review. J. Anim. Sci. 76:275−286.

Russell, J. B. 1987. A proposed mechanism of monensin action ininhibiting ruminal bacterial growth: Effects on ion flux andprotonmotive force. J. Anim. Sci. 64:1519−1525.

Russell, J. B., and T. Hino. 1985. Regulation of lactate production inStreptococcus bovis: A spiraling effect that contributes to ru-men acidosis. J. Dairy Sci. 68:1712−1721.

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

Russell, J. B., and P. J. Van Soest. 1984. In vitro ruminal fermenta-tion of organic acids common in forage. Appl. Environ.Microbiol. 47:155−159.

Sanson, D. W., and O. T. Stallcup. 1984. Growth response andserum constituents of Holstein bulls fed malic acid. Nutr. Rep.Int. 30:1261−1267.

Scheifinger, C. C., M. J. Latham, and M. J. Wolin. 1975. Relation-ship of lactate dehydrogenase specificity and growth rate tolactate metabolism by Selenomonas ruminantium. Appl.Microbiol. 30:916−921.

Slack, G. L., E. Millward, and W. J. Martin. 1964. The effect oftablets stimulating salivary flow on the incidence of dentalcaries. Br. Dent. J. 116:105−108.

Slyter, L. L. 1976. Influence of acidosis on rumen function. J. Anim.Sci. 43:910−929.

Stewart, C. S., and M. P. Bryant. 1988. The rumen bacteria. In: P.N. Hobson (Ed.) The Rumen Microbial Ecosystem. pp 21−75.Elsevier Science Publishing, New York.

Stout, P. R., J. Brownell, and R. G. Burau. 1967. Occurrences oftrans-aconitate in range forage species. Agron. J. 59:21−24.

Streeter, M. N., D. J. Nisbet, S. A. Martin, and S. E. Williams. 1994.Effect of malate on ruminal metabolism and performance ofsteers fed a high concentrate diet. J. Anim. Sci. 72(Suppl. 1):384 (Abstr.).

Strobel, H. J., and J. B. Russell. 1991a. Role of sodium in the growthof a ruminal selenomonad. Appl. Environ. Microbiol. 57:1663−1668.

Strobel, H. J., and J. B. Russell. 1991b. Succinate transport by aruminal selenomonad and its regulation by carbohydrate avail-ability and osmotic strength. Appl. Environ. Microbiol. 57:248−254.

van Gylswyk, N. O. 1995. Succiniclasticum ruminis gen. nov., sp.nov., a ruminal bacterium converting succinate to propionate asthe sole energy-yielding mechanism. Int. J. Syst. Bacteriol. 45:297−300.

van Gylswyk, N. O., H. Hippe, and F. A. Rainey. 1997. Schwartziasuccinivorans gen. nov., sp. nov., another ruminal bacteriumutilizing succinate as the sole energy source. Int. J. Syst.Bacteriol. 47:155−159.

West, J. W., G. M. Hill, R. N. Gates, and B. G. Mullinix. 1997.Effects of dietary forage source and amount of forage additionon intake, milk yield, and digestion for lactating dairy cows. J.Dairy Sci. 80:1656−1665.