Effects of Monensin and Fish Oil on Conjugated Linoleic Acid Production by Rumen Microbes in Holstein Cows Fed Diets Supplemented with Soybean Oil and Sodium Bicarbonate *

The present study was conducted with four ruminally canulated Holstein cows to observe the effects of monensin or fish oil on diet fermentation and production of conjugated linoleic acids (CLAs) in the rumen when fed diets supplemented with soybean oil and sodium bicarbonate. Cows of the control treatment were fed a basal diet (CON) consisting of 60% commercial concentrate and 40% chopped rye grass hay. Cows of other treatments were fed the same diet as CON, but the concentrate was supplemented with 7% of soybean oil and 0.5% of sodium bicarbonate (SO-B), SO-B supplemented with monensin (30 ppm, SO-BM) or concentrate supplemented with 6.3% of soybean oil, 0.5% of sodium-bicarbonate, 30 ppm of monensin and 0.7% of fish oil (SO-BMF). Dry matter (DM) intake of the cows was significantly (p<0.011) reduced by feeding the SO-BMF diet compared to the other diets which did not differ in DM intake. Whole tract digestibility of major dietary components was significantly (p<0.004-0.027) higher for SO-BMF than the other supplement-containing diets. Dietary supplements did not clearly affect rumen pH and ammonia concentrations compared to the CON diet. Significantly reduced (p<0.05) total VFA concentration was obtained by the addition of fish oil to the diet (SO-BMF) compared to other diets. No differences, however, were obtained in major VFA proportions as well as in total VFA between the supplemented diets. The SO-BM diet increased (p<0.01-0.05) the concentrations of trans-11 C18:1 and linoleic acid in rumen fluid. Total CLA concentration was also increased by the feeding of SO-B and SO-BM diets during early fermentation times (up to 3 h) postfeeding. Cis-9, trans-11 CLA concentration in rumen fluid was highest (p<0.05) for SO-B up to 1 h while the highest (p<0.01) value for SO-BM occurred at 3 h post-feeding. An increased trans-10, cis-12 CLA concentration was obtained from the SO-B and SO-BM diets at 1 and 3 h post feeding compared to the other diets. Supplementation of oils with monensin and sodium bicarbonate increased (p<0.05) the proportions of C18:1 and CLA in the plasma of cows, but the effect of monensin and/or fish oil was limited to trans-10, cis-12 CLA. (


INTRODUCTION
There has been an increasing interest in natural nutrients of ruminant products that have health benefits for humans, and one of them is conjugated linoleic acid (CLA).CLA has several important physiological functions, such as anticarcinogenic (Ha et al., 1987;Albright et al., 2005) and anti-atherogenic (Lee et al., 1994) effects, immunemodulation (Whigham et al., 2002), and reduction of body fat (Cook et al., 1993).These fatty acids are generated naturally in the rumen or mammary gland of ruminants.
Various attempts to increase CLA content in ruminant products have included the application of lipid source.High levels of linoleic and linolenic acids in the form of oil seeds or oils are known to enhance the formation of CLA in the rumen fluid (Wang et al., 2002), milk (Dhiman et al., 2000;Loor et al., 2005) or meat (Choi et al., 2006;Wang et al., 2006).
Because the ruminal environment is subject to change by dietary manipulation it is possible that changes in pH of rumen fluid will have an effect on CLA formation (Wang et al., 2002).Wang and Song (2003) reported that formation of cis-9, trans-11 CLA by rumen bacteria increased with the increment of pH when ground oilseeds were incubated in vitro.In addition, ionophoric antibiotics have been known to inhibit the ruminal hydrogenation of unsaturated fatty acids.Monensin, for example, inhibited the growth of Butyrivibrio fibrisolvens, a very active Gram positive bacterium in hydrogenation, and resulted in an increased proportion of cis9, trans11-CLA with decreased proportion of stearic acid when incubated with linoleic acid (C 18:2 ) in vitro (Fellner et al., 1997).Unsaturated fatty acids were accumulated by presence of salinomycin in the rumen fluid and duodenal digesta of sheep (Kobayashi et al., 1992).Wang et al. (2006) also observed an increased proportion of the cis9, trans11-CLA from Korean native steers when the concentrate was supplemented with monensin.
Meanwhile, eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA), which are rich in fish oil, might be resistant to ruminal hydrogenation in vitro and thus could increase CLA contents (Ashes et al., 1992).Wang et al. (2005) also observed in vitro an increased proportion of cis-9, trans-11 CLA from the supplementation of fish oil.
However, although each of these dietary manipulations had some effect on CLA production in the rumen and ruminant products, the effect was relatively small.The current in vivo study, therefore, was conducted to determine the effects of monensin and fish oil on the formation of CLA isomers (cis-9, trans-11 and cis-10, trans-12 C 18:2 ) and fermentation characteristics in the rumen of Holstein cows when fed a diet supplemented with C 18:2 -rich soybean oil and sodium bicarbonate.

Animals and diets
The metabolic study was conducted as a 4×4 Latin square design with 4 ruminally cannulated non-lactating Holstein cows (640±37 kg) for 4 dietary treatments.The cows were fed a basal diet consisting of 60% commercial concentrate and 40% chopped rye grass hay (DM basis) without any supplements (CON).The cows were also fed the same basal diet as CON cows, but the concentrate was supplemented with 7% of soybean oil and 0.5% of sodiumbicarbonate (SO-B), SO-B diet supplemented with monensin (30 ppm of concentrate, SO-BM) or SO-B diet supplemented with 6.3% soybean oil, 0.5% sodium bicarbonate, 30 ppm monensin and 0.7% fish oil (SO-BMF, Table 1).Experimental diets were prepared at 3 day intervals.Concentrate constituents and nutrient composition of the total diets are shown in Table 1, and fatty acid compositions of soybean oil and fish oil are shown in Table 2.

Feeding management
The cows were housed in metabolism crates, and were fed 13 kg (DM basis) of the mixed diets of concentrate and forage twice daily (08:00 and 18:00) in equal amounts, and were allowed free access to water and mineral blocks.Between experimental periods, the cows were allowed free walking in outdoor grounds for 7 days.The study was conducted for 12 weeks in total, with 10 days for diet adaptation and 6 days for the sampling in each experimental period.

Measurements and analysis
Feed residues were collected before morning feeding (08:00) daily to estimate feed intake and digestibility.Ruminal contents were collected on three days during each period from three sites in the rumen immediately before morning feeding and at 1, 3, 6 and 9 h post-feeding, and were squeezed to collect the rumen fluid at each sampling time.The pH of rumen fluid was measured instantly, and after being strained through 4 layers of cheese cloth 5 ml rumen fluid was collected for ammonia and volatile fatty acid (VFA) analysis.All samples collected were kept frozen at -20°C until analyzed.Feces were also collected for three days after the collection of ruminal contents to estimate whole tract digestibility of the major dietary components.
Blood was collected from the jugular vein with a vacutainer (Becton Dickinson) containing sodium heparin from cows at 4 h after feeding on the same days as fecal collection.Blood samples were centrifuged at 3,000×g for 10 min., and plasma was transferred to 30 ml screw-cap tubes and kept frozen at -70°C until analyzed.
Proximal analyses of the diets and feces followed the methods of AOAC (1995).NDF content was estimated by the method of Van Soest et al. (1991).Ammonia concentration was determined by the method of Fawcett and Scott (1960) using a spectrophotometer (DU-650).Rumen fluid (4 ml) was mixed with 1 ml 25% phosphoric acid and 0.5 ml pivalic acid solution (1%, w/v) as an internal standard.The mixed solution was centrifuged at 15,000×g for 15 min.and the supernatant was used to determine the concentration and composition of VFA using gas chromatography (GC, HP 5890II, Hewlett Packard Co.).Lipid was extracted from rumen fluid and plasma using a mixture of organic solvents (one part hexane, three parts isopropanol, one part acetone) purchased from Fisher Co. (Fair Lawn, NJ, USA).The suspensions were then centrifuged at 1,000×g for 3 min at 20°C.The solvent (top) layer was removed and flushed with nitrogen gas until dried.Methylation of the fatty acids followed the method of Lepage and Roy (1986) prior to injecting into the GC.A fused silica capillary column (100 m×0.25 mm, i.d.×0.20 μm thickness, Supelco, SP TM -2560; USA) was used.Cis-9, trans-11 and trans-10, cis-12 CLA isomers (Sigma, USA) were used to identify and quantify each CLA isomer.Other FA standards were obtained from Supelco Co. (18920, USA).Tridecanoic acid (C 13:0 ) was used as an internal standard and all CLA isomers and other FAs in rumen fluid were quantified using FA standards.

Statistical analysis
The results obtained were subjected to least squares analysis of variance according to the general linear models procedure of SAS (1985) and significances were compared by S-N-K Test (Steel and Torrie, 1980).
The SO-B and SO-BM diets increased (p<0.001-0.05) the ruminal concentrations of both C 18:0 (Figure 3) at all sampling times and C 18:1 (Figure 4) at 3, 6 and 9 h post-  feeding compared to the CON and SO-BMF diets.The SO-BM diet also increased (p<0.01-0.05) the concentrations in rumen fluid of trans-11 C 18:1 (Figure 5) and C 18:2 (Figure 6) at most sampling times post-feeding.Total CLA concentration was also increased by the feeding of SO-B and SO-BM diets during early fermentation times (up to 3 h) post-feeding (Figure 7).Cis-9, trans-11 CLA concentration was highest (p<0.05) for SO-B up to 1 h while the highest (p<0.01)value for SO-BM occured at 3 h in rumen fluid (Figure 8).Increased trans-10, cis-12 CLA concentration in rumen fluid was obtained from SO-B and SO-BM diets (Figure 9) at 1 and 3 h post feeding compared to other diets.Supplementation of oils with monensin and sodium bicarbonate increased (p<0.005) the proportions of C 18:1 , total CLA and both isomers of cis-9, trans-11 CLA and trans-10, cis-12 CLA in the plasma of cows (Table 5), but the effect of monensin and/or fish oil was greater for trans-10, cis-12 CLA than for cis-9, trans-11 CLA.The supplement also slightly increased the proportions of unsaturated fatty acid and polyunsaturated fatty acid in the plasma.

DISCUSSION
The present study was focused to maximize CLA production in the rumen with the combination of various dietary factors.Soybean oil has been useful as a source of C 18:2 which is one of the two major precursors of CLA (Wang et al., 2002), and sodium bicarbonate acts for pH control in rumen fluid (Wang and Song, 2003).Supplementation of monensin (Fellner et al., 1997;Sauer et al., 1998;Wang et al., 2006) or fish oil (Ashes et al., 1992;Wang et al. (2005) has resulted in an increased ruminal CLA production.
Addition of supplements including oil to the diet decreased DM intake and the reduction was greater for the cattle fed the fish oil-supplemented diet (SO-BMF, Table 3), despite the addition level being only 0.7% (Table 1).The intake of SO-BMF by cows was only approximately 71% of the mean intake of other diets (Table 3).The reduced DM intake of SO-BMF might occur due to the sickening smell of fish oil as observed in Korean native steers fed a diet supplemented with fish oil (Wang et al., 2006), and affected many observations from total VFA production and whole tract digestibility of nutrients to CLA contents.Overall patterns of ruminal pH and ammonia-N concentration in the rumen after feeding looked normal, but the ruminal pH might not have responded much to the addition of sodium bicarbonate (0.5% of concentrate) since the total diet consisted of 60% concentrate and 40% forage and its range was 6.2 to 7.1 during the indicated sampling times for all the diets (Figure 1).Decreased ammonia concentration in rumen fluid from the supplemented diets (Figure 2) might be due to the oil reducing the rate of ruminal fermentation (Wang and Song, 2001).Lowered total VFA concentration in rumen fluid from SO-BMF could be due to the reduced  feed intake in comparison with the other diets (Table 4).No differences in the proportions of major VFA indicated that oil supplementation at the level of 7% to the concentrate may not interfere with ruminal fermentation of the total diet.Another possible reason for no difference in VFA proportion compared to the CON diet may be, in part, due to the addition of sodium bicarbonate to the oil supplemented diets.
The lowest contents of C18-fatty acids on the CON diet were certainly due to no addition of oil while monensin (30 ppm in concentrate) in the SO-BM diet caused slight increases in the content of these fatty acids (Figure 3 to Figure 6) in rumen fluid compared to other diets.These results indicate that monensin may stimulate some hydrogenation activity by rumen microbes although its action is limited to a certain stage of the fermentation process.However, Fellner et al. (1997) reported that ionophoric antibiotics inhibit the growth of Butyrivibrio fibrisolvens which is involved in ruminal hydrogenation.Kobayashi et al. (1992) reported that salinomycin supplementation resulted in accumulation of unsaturated fatty acids in rumen fluid and duodenal digesta of sheep.
The lowest contents of CLA from the CON diet among all diets were also due to the absence of supplemented oil (Table 7 to Table 9).The monensin effect on CLA production by rumen microbes has been controversial.Monensin increased the proportion of cis-9, trans11-CLA in vitro (Fellner et al., 1997) and in vivo (Sauer et al., 1998;Wang et al., 2006), whereas Dhiman et al. (1996) and Chouinard et al. (1998) observed no effect of monensin on CLA content in milk fat.Wang et al. (2005) found only a small effect of monensin (10 ppm, w/v) when incubated with safflower in vitro.In the present study, total CLA content in rumen fluid was not greatly affected by monensin.The effect of monensin might differ depending on the duration.A monensin effect (SO-BM) was not apparent in cis-9, trans-11 CLA content until 3h post-feeding compared to other diets, while trans-10, cis-12 CLA content was not influenced by monensin.The possibility of a timedependent effect of monensin was suggested by Wang et al. (2005) who observed the highest cis-9, trans-11 CLA content from the longest incubation (12 h) in vitro.
Meanwhile, fish oil has been applied to enhance CLA production in rumen fluid.Ashes et al. (1992) postulated, based on an in vitro study, that EPA and DHA in fish oil might be resistant to ruminal biohydrogenation and thus could increase CLA content.In another in vitro study, Wang et al. (2005) observed an increased proportion of cis-9, trans-11 CLA from the supplementation of fish oil without interference in fermentation.Loor et al. (2005) reported that the fish oil effect on CLA production in the rumen may be related to reduced populations of ruminal bacteria associated not only with the final reduction step but with the overall process of biohydrogenation.They reported that sunflower oil with fish oil resulted in increased flows of cis-9, trans-11 CLA and trans-10, cis-12 CLA.However, in the present study, a fish oil effect (SO-BMF) on CLA production could not be properly evaluated nor compared among diets since feed intake by these cows was only approximately 71% of the mean intake of other diets (Table 3).Although fish oil helps to enhance ruminal CLA production, the mode of action still remains to be elucidated through further studies.
Unlike the concentration of fatty acid in rumen fluid, its profile in plasma was expressed as a proportion (%).Increased proportions of c9-C 18:1 , CLA and UFA in the plasma of cows were mostly due to the soybean oil supplementation of the diet (Table 5).An effect of monensin and/or fish oil on the trans-10, cis-12 CLA was not big enough and these supplements did not even affect the cis-9, trans-11 CLA proportion in plasma.There is much evidence that dietary supplementation of C 18 -rich plant oil enhances the CLA and UFA proportions.Bell and Kennelly (2000) observed increased cis-9, trans-11 CLA and UFA proportions in milk from cows fed a diet supplemented with plant oil.Supplementation of soybean oil also slightly increased CLA and UFA proportions in both plasma and subcutaneous fat of sheep (Choi et al., 2007).Feeding a diet supplemented with rolled flaxseed (15% of the concentrate) increased UFA proportion in milk compared to the control diet (Chung et al., 1996).Increased UFA by an oil-enriched diet is beneficial to health (Grummer, 1991).Although effects of monensin and fish oil are known to limit microbial action in the rumen, Wang et al. (2006) observed that supplementation of mixed oil (C 18:2 -rich safflower oil and soybean oil) and fish oil with monensin to the diet slightly increased CLA and UFA proportions in both intramuscular-and subcutaneous fats in Korean native steers.This effect might be due to the transfer of increased CLA content in the rumen to the blood as influenced by monensin and/or fish oil.
Based on the results obtained from the present study, the supplements did not clearly affect the fermentation of diet in the rumen.Effects of monensin and fish oil in soybean oil supplemented diets on CLA production were small, and might appear differently depending on CLA isomers under the condition of the present study.Although it has some effect, fish oil may not be practical to apply for the enhancement of CLA production in vivo due to the intake problem.

Figure 1 .
Figure 1.pH of rumen fluid in Holstein cows when fed soybean oil with buffer.

Figure 2 .
Figure 2. Ammonia-N concentration (mg/100 ml) in rumen fluid of Holstein cows when fed soybean oil with buffer.

Table 1 .
Ingredients of concentrate and nutrient composition of the total diets (DM basis)

Table 3 .
Supplementation effects of monensin and fish oil on nutrient intake (kg/day) by Holstein cows when fed soybean oil with buffer Refer to Table1. 2 Standard error of the mean. 3Probability levels.

Table 4 .
Total VFA concentration and individual VFA proportions in rumen fluid of Holstein cows when fed soybean oil with buffer

Table 5 .
Additive effect of monensin and fish oil on fatty acid composition (%) in the plasma of Hostein cows when fed soybean oil with 1 Referred to Table1. 2 Standard error of the mean. 3Probability levels.