Production of Trans-10 , Cis-12 Conjugated Linoleic Acid by Megasphaera Elsdenii YJ-4 : Physiological Roles in the Rumen

Megaspahera elsdenii YJ-4, which was previously isolated as a producer of trans-10, cis-12 CLA, was studied for its carbon source on the CLA production. M. elsdenii YJ-4, was incubated with glucose and lactose, and cultured in batch and continuous culture systems with linoleic acid at various pHs to investigate CLA production. Batch cultures of the ruminal bacterium, M. elsdenii YJ4, were resistant to stearic acid and linoleic acid, and little growth inhibition was observed even when the fatty acid concentration in the culture was as much as 4 mg ml. Stationary phase batch cultures (0.25 mg bacterial protein ml) that had been grown on lactate and incubated with linoleic acid (0.20 mg ml) produced approximately 12 μg trans-10, cis-12 CLA mg protein and little cis-9, trans-11 CLA was detected. Some linoleic acid was converted to hydrogenated products (chiefly stearic acid), but these fatty acids were less than 5 μg mg bacterial protein. Stationary phase batch cultures that had been grown on glucose produced at least 3-fold less trans-10, cis-12 CLA than ones grown on lactate. Cells from lactate-limited continuous cultures produced less trans-10, cis-12 CLA than those from batch culture, but only if the pH was greater than 6.4. When the pH of the lactate-limited continuous cultures was lower than 6.4, trans10, cis-12 CLA and hydrogenated products declined. Cells from glucose-limited continuous cultures produced less trans-10, cis-12 CLA and hydrogenated products than the cells that had been limited by lactate, but pH had little impact on this production. These results support the idea that M. elsdenii YJ-4 could be one of the major producers of trans-10, cis-12 CLA which causes cows to produce milk with a low fat content. (Asian-Aust. J. Anim. Sci. 2005. Vol 18, No. 10 : 1425-1429)


INTRODUCTION
Conjugated linoleic acid (CLA) is a collective term for different positional and geometric isomers of octadecanoic acid with double bonds in a conjugate configuration.In recent years, research has shown that CLA can inhibit chemically-induced tumors, prevent atherosclerosis and improve the protein to fat ratio of experimental animals (Ha et al., 1987;Lee et al., 1994a, b;Dougan et al., 1997;Choi et al., 2002).CLA is produced in ruminants consuming a diet rich in polyunsaturated fatty acids.The presence of CLA in dairy product is mainly due to isomerization of linoleic acid and linolenic acid during the biohydrogenation by microorganisms in the rumen via a characteristic process called biohydrogenation which is influenced by various dietary factors (Kim, 2003;Wang and Song 2003;Wang et al., 2003).Kepler and Tove (1967) showed that the first step involved in isomerization that created a conjugated fatty acid.It was shown that B. fibrisolvens A38 accumulated cis-9, trans-11 CLA when the substrate fat concentration was high enough to inhibit growth and CLA reduction to further saturated fatty acids (Kim et al., 2000).B. fibrisolvens is very sensitive to unsaturated fatty acids (Henderson, 1973), and previous work indicated that B. fibrisolvens A38 did not produce cis-9, trans-11 CLA from linoleic acid until its concentration was high enough to inhibit the growth (Kim et al., 2000).These results are consistent with idea that cis-9, trans-11 CLA was an intermediate in the conversion of unsaturated fatty acids to less toxic hydrogenated products.
In the 1960's, Davis et al. (1964) noted that dietary oil supplements accentuated the milk fat depression of cattle fed low fiber rations, but abomasal infusions of the same oil did not.These results suggested that milk fat depression was associated with changes in ruminal metabolism rather than a direct effect of the oils on the mammary gland per se.These authors also noted that oil supplements increased the amount of trans-C18:1 and other isomers of unsaturated fatty acids in the rumen.They concluded that these "unacceptable substrates" might be "rejected" by the mammary gland.
More recently, Gaynor et al. (1994) examined the effect of trans-versus cis-C18:1 on the milk fat yield of dairy cattle.Neither fatty acids inhibited stearoyl-CoA desaturase or reduced milk fat synthesis.However, Baumgard et al. (2001) demonstrated that abomasal infusion of trans-10, cis-12 CLA caused a significant decrease in milk fat yield.These latter results indicated that trans-10, cis-12 CLA could cause a depression of milk fat, but until recently bacteria responsible for this production had not been identified (Dewhurst et al., 2001;Kim et al., 2002).Fibrolytic rumen bacteria such as B. fibrisolvens A38 are typified by high sensitivity to low pH caused by high concentrate diet which selects for starch fermenting ruminal bacteria that produce lactate, but much of this lactate is converted to acetate and propionate by other rumen bacteria, Megasphaera elsdenii (Counotte et al., 1981).We hypothesized that lactate utilizing rumen bacteria produce trans-10, cis-12 CLA and the production could be increased when ruminal fluid was enriched with lactate.
Previous work showed that stationary cultures of M. elsdenii YJ-4 that had been grown on lactate produced trans-10, cis-12 CLA and smaller amounts of cis-9, trans-11, but the effects of other culture conditions were not examined.The following experiments described the effects of substrate and pH on the CLA production of M. elsdenii YJ-4.

Organism and culture conditions
The isolation, characterization and 16S rDNA sequence of M. elsdenii YJ-4 was previously described (Kim et al., 2002).Batch cultures were grown anaerobically at 39°C in a basal medium that contained (per liter), 290 mg K 2 HPO 4 , 240 mg KH 2 PO 4 , 480 mg (NH 4 ) 2 SO 4 , 480 mg NaCl, 100 mg MgSO 4 , 64 mg CaCl 2 , 0.5 g yeast extract, 1.0 g Trypticase, 0.6 g cysteine hydrochloride and 1 mM each volatile fatty acids (isovalerate, isobutyrate, 2methylbutyrate, and valerate).The basal medium was dispensed anaerobically into tubes (18×150 mm) that were sealed with butyl rubber stoppers and capped with aluminum seals.M. elsdenii YJ-4 was also grown with glucose (2 mg/ml) or lactate (4 mg/ml) in continuous culture under O 2 -free CO 2 (170 ml culture vessel, 39°C).Yeast extract concentration in the medium of continuous culture was increased to 2.5 g/L to ensure better cell growth.The pH of the continuous culture was decreased by adding concentrated hydrochloride to the medium reservoir.Samples were not taken until at least 4 culture vessel volumes of medium had passed through the continuous culture vessel (98% turnover).Growth was monitored by measuring optical density (600 nm, 1 cm path length, Gilford Spectrophotometer, Oberlin, OH) and the relationship between optical density and cell protein was 200 µg protein ml -1 optical density -1 .

Fatty acid analyses
The linoleic acid (100 mg in 10 ml water with 200 mg bovine serum albumin ml -;1 ; Sigma Chemical Co.St. Louis, MO) was filter-sterilized (pore size 0.22 µm), and was added to stationary phase pure cultures to achieve a final concentration of 200 µg linoleic acid ml -1 .The bovine serum albumin ensured that the linoleic acid remained in culture media (Kim et al., 2000).Bacterial cell suspensions (10 ml) taken from batch or continuous cultures were incubated with 200 µg linoleic acid ml -1 (39°C, 30 min) and immediately extracted with a mixture of organic solvents three times (4 ml, 1 part hexane, 3 parts isopropanol, 1 part acetone, 1 min using a vortex mixer).The suspensions were centrifuged (1,000×g, 3 min, 20°C), the solvent layer (top) was flushed with nitrogen until dry.The fatty acids were dissolved in toluene (1 ml) and methylated as previously described by Kim and Liu (1999).Fatty acid methyl esters were separated by a Supelcowax-10 fused silica capillary column (60 m×0.32 mm, 0.5 µm film thickness; Supelco, Inc., Bellefonte, PA) using a Hewlett Packard model HP5890 gas chromatograph equipped with a flame ionization detector and model HP3392 integrator.The conditions were: 2.4 ml min -1 helium flow; injector 200°C; detector 250°C; the column chamber temperature was initially 40°C (5 min), the column temperature was increased to 220°C at 20°C/min and held for 30 min.
Sample (1 µl) containing 0.5-5 µg of linoleic acid or CLA was injected into the column in a splitless mode.Heptadecanoic acid (C17:0) was used as an internal standard and cis-9, trans-11 octadecadienoic acid and trans-10, cis-12 octadecadienoic acid were used as a CLA standard (>98% purity Matreya Inc. Pleasant Gap, PA).The recovery of CLA was 83% and C17:0 was 80%.A known standard mixture of fatty acids (Sigma) was used to identify other fatty acids.This protocol was able to separate eight isomers of linoleic acid, but it could not differentiate cis, trans versus trans, cis configurations in the same position.

Statistical analyses
Data were presented as mean±standard error for at least three replications as a mean values and standard errors of the means.Statistical analyses were conducted using SigmaStat program (Version 8.0; Jandel Corp. San Rafael, CA).Differences among treatments were determined using the Student t test (p<0.05).

RESULTS AND DISCUSSION
M. elsdenii YJ-4 batch cultures that were grown with either lactate or glucose as a carbon source were relatively resistant to linoleic acid, which serves as substrate for CLA production, and stearic acid, the final product of biohydrogenation.A decrease in growth was not observed until the concentration was 4 mg ml -1 (approximately 14 mM) (Figure 1).This indicated that M. elsdenii YJ-4 was highly resistant to long chain fatty acids.Indeed, M. elsdenii YJ-4 grew rapidly even if the concentration of linoleic acid was 7-fold higher than the amount needed to completely inhibit the growth of B. fibrisolvens A38 (Kim et al., 2000).Based on these results, it is unlikely that the biohydrogenation reaction of M. elsdenii is a detoxification reaction per se, but the reductases of biohydrogenation would serve as sink for reducing equivalents.
M. elsdenii strains are closely related to low G+C Gram positive-bacteria (Stackebrandt et al., 1985), but this bacterium has an outer membrane and is resistant to some potentially toxic compounds (e.g.ionophore and monensin) (Callaway et al., 1999).It has been suggested that unsaturated fatty acids exhibit the inhibitory effect on the growth of various bacteria (Thompson et al., 1994;Kim et al., 2000).Moreover, reducing equivalent disposal can be a problem for anaerobic bacteria in the rumen where oxygen is not available as a terminal electron acceptor.Bacteria with membrane bound hydrogenases transfer reducing equivalents to methanogenic species via interspecies hydrogen transfer.However, some anaerobes must use alternative schemes of oxidation (e.g., lactate and alcohol dehydrogenase), and these schemes ultimately decrease the availability of ATP (Wolin et al., 1997).
Stationary phase M. elsdenii YJ-4 batch cultures (0.25 mg bacterial protein ml -1 ) that had been grown on lactate and were incubated with linoleic acid (0.20 mg ml -1 ) produced approximately 12 µg trans-10, cis-12 CLA mg protein -1 (Figure 1), and little cis-9, trans-11 CLA was detected (<2 µg mg protein -1 ) (data not shown).Some linoleic acid was converted to hydrogenated products (chiefly stearic acid), but these acids were less than 5 µg mg bacterial protein -1 .Stationary phase batch cultures (0.21 mg bacterial protein ml -1 ) that had been grown on glucose produced at least 3-fold less trans-10, cis-12 CLA than those that had been grown on lactate (Figure 2; p<0.05).The glucose-grown cells tended to produce less hydrogenated products than those grown on lactate, but this trend was not statistically significant (p>0.05).
M. elsdenii is capable of producing hydrogen (Hungate, 1966), but lactate is catabolized via a pathway that employs acrylyl-CoA reductase (dehydrase) (Baldwin et al., 1965).Brockman and Wood (1975) provided evidence that this latter reaction was linked to an electron-transferring flavoprotein, but the involvement of this system in biohydrogenation is not known.Glucose carbon is also converted to propionate, but in this case additional reducing equivalents are generated from the Embden Meyerhoff Parnas scheme.
The M. elsdenii YJ-4 batch cultures had an initial pH of approximately 6.7, and neither fermentation caused a significant (>0.1 unit) change in final pH (data not shown).Because the batch cultures were buffered by bicarbonate, the pH was difficult to adjust and regulate, but the pH of continuous cultures (dilution rate 0.1 h -1 ) could be gradually decreased by the addition of concentrated HCl to the medium reservoir.The continuous cultures reached a steady state after approximately 98% of the original culture medium had been diluted from the culture vessel.Bacterial cells from glucose-limited continuous cultures produced less trans-10, cis-12 CLA (Figure 3b) than those from batch culture (Figure 2) when the pH was greater than 6.3 (p>0.05).When the pH of the lactate-limited continuous cultures was less than 6.3, the activity of the cells to produce trans-10, cis-12 CLA declined by half at pH 5.6.The ability of the cells to produce hydrogenated products also decreased (p<0.05), and the correlation coefficient was high (r 2 = 0.86).Cells from glucose-limited continuous cultures produced less trans-10, cis-12 CLA than the cells that had been limited by lactate (Figure 3a) (p<0.05), and pH had less of an impact compared to that of lactate-limited continuous cultures.The ability of cells from glucoselimited continuous cultures to produce trans-10, cis-12 CLA was greatest if the pH was 6.1, but the difference between either pH 6.8 and 6.1 or 5.4 and 6.1 was less than 1-fold.Continuous culture pH did not influence the ability of the glucose-limited cells to produce hydrogenated products (Figure 3b) (p>0.05).
Fermented silages can contain as much as 10% lactate, and lactate can also be an end-product of ruminal starch and sugar fermentation.Because lactate is a relatively strong acid, ruminal lactate production can cause a significant decrease in ruminal pH.M. elsdenii counteracts this decrease by converting lactate to volatile fatty acids that have a significantly higher pK a value (3.9 versus ≥4.8), and several groups have attempted to use ruminal inoculations with M. elsdenii as a tool for increasing ruminal pH.The impact of pH on ruminal fatty acid metabolism has not been examined in a systematic fashion, but it has long been recognized that milk fat depression is generally correlated with an increased concentration of starch in the ration and a reduction in pH (Davis et al., 1964).Based on these earlier results, we had originally thought that acidic pH might enhance the ability of M. elsdenii YJ-4 to produce trans-10, cis-12 CLA, but an opposite effect was observed.When M. elsdenii YJ-4 was grown in continuous culture at pH values less than 6.4, the specific rate of CLA production decreased almost 100%.
The success of M. elsdenii in the rumen has been linked with tolerance in low pH conditions and its ability to use lactate (Counotte et al., 1981;Miwa et al., 1997), and M. elsdenii YJ-4 batch cultures that had been grown on lactate had more than a two-fold greater ability to produce trans-10, cis-12 CLA than those grown on glucose.The use of batch cultures to model ruminal fermentation is confounded by the fact that substrate concentrations and growth rates are higher in vitro than in vivo.The rumen tends to operate as a natural continuous culture device (Hungate, 1966), and our results indicated that lactate-limited continuous cultures of M. elsdenii YJ-4 also produced more trans-10, cis-12 CLA than those limited by glucose.The observation that grain feeding promotes the growth of M. elsdenii in vivo (Counotte et al., 1981) is consistent with our results that this species survive acidic rumen environment and thus could play some roles in milk fat depression.
We were not able to separate all the isomers of linoleic acid, especially cis, trans versus trans, cis configurations in the same position in this study.Further work will be clearly needed to study the effect of each isomer production mechanism as well as to define the role of M. elsdenii in ruminal CLA production in vivo.

Figure 1 .Figure 2 .
Figure1.Inhibition of the cell growth of M. elsdenii YJ-4 by fatty acids.M. elsdenii YJ-4 cells were incubated for 24 h with glucose (20 mM) and stearic acid (•) or linoleic acid (○) at concentrations from 0 to 8 mg ml -1 and maximal growth was measured at 600 nm during the incubation.

Figure 3 .
Figure3.The production of trans-10, cis-12 CLA (•) and hydrogenated products (HP: C18:0 and trans-C18:1) (○) by stationary phase M. elsdenii YJ-4 cells that had been grown in continuous culture with lactate (40 mM).pH was decreased by the addition of HCl to the medium reservoir.The production of trans-10, cis-12 CLA and biohydrogenated products (BP: C18:0 and trans-C18:1) by stationary phase M. elsdenii YJ-4 cells that had been grown in continuous culture with glucose (20 mM) is shown in part b.