INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

CONJUGATED LINOLEIC ACIDS (CLA) are octadecadienoic acids that are found predominately in food products derived from ruminant animals. They have been implicated in a wide range of beneficial effects, including anticarcinogenic, antidiabetic, and immune stimulation (2, 18). CLA have also been shown to reduce body fat and alter nutrient partitioning in growing animals, including mice and pigs (10, 11, 26, 27, 33). However, all of these biological effects have been observed using dietary supplements that contain a variety of CLA isomers, and effects of specific isomers are unknown.

CLA administration also affects lipid synthesis in lactating cows, resulting in marked reductions in milk fat secretion (5, 6, 20). Several dietary situations, such as high-concentrate, low-fiber diets or increasing intake of plant oils, also cause a depression in milk fat secretion in dairy cows, and the mechanism has been postulated to involve specific fatty acid isomers arising from rumen biohydrogenation (9). The predominant CLA in ruminant fat is the cis-9,trans-11 isomer. It originates in part from CLA produced by rumen bacteria as an intermediate in the biohydrogenation of linoleic acid. A portion also comes from tissue synthesis of CLA by ⁹-desaturase conversion of trans-11 C_18:1, an intermediate in rumen biohydrogenation of several polyunsaturated fatty acids (14). Milk fat content of trans-10 C_18:1 is increased during dietary-induced milk fat depression (15), and rumen biohydrogenation of trans-10,cis-12 CLA is the putative source of trans-10 C_18:1 (14). On the basis of this, we hypothesized that the specific CLA isomers that cause a reduction in milk fat synthesis in lactating dairy cows are those containing a trans-10 double bond. Our objective was to compare the effects of two relatively pure CLA isomers, cis-9,trans-11 and trans-10,cis-12, on milk fat synthesis. Animal performance and blood variables associated with lipid metabolism were also monitored.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

All procedures involving animals were approved by the Cornell University Institutional Animal Care and Use Committee. Three lactating multiparous Holstein cows fitted with rumen fistulas were randomly assigned in a 3 × 3 Latin square experiment. At the initiation of the trial, cows averaged 591 ± 91 kg of body weight and were 111 ± 12 day postpartum (means ± SE). Cows were fed a total mixed ration formulated using the Cornell Net Carbohydrate and Protein System (13). The diet was formulated to meet or exceed nutrient requirements (23), with chopped alfalfa hay as the major forage component and cracked corn as the primary concentrate (Table 1). Cows were fed ad libitum intake with equal portions of fresh feed offered at 0600 and 1800 daily. Water was available at all times.

Treatments were abomasal infusion of 1) skim milk (control), 2) cis-9,trans-11 CLA supplement, and 3) trans-10,cis-12 CLA supplement. The composition of CLA supplements (Natural Lipids, Hovdebygda, Norway) is presented in Table 2. The CLA supplements were emulsified with skim milk to obtain a volume so infusions would provide a uniform and continuous supply of CLA. Three separate emulsions were prepared using a microfluidizer (model 110T; Microfluidics, Newton, MA) at a pressure of 8,000 lb/in.² as previously described (5). Target emulsion concentrations for 9,11 CLA and 10,12 CLA supplements were 0.34 and 0.27%, respectively. Actual concentrations, as determined by difference in total solids content (method 990.20; Ref. 1) between the emulsion and skim milk, were 0.33 and 0.27% for 9,11 CLA and 10,12 CLA supplements, respectively. Skim milk (control infusate) and CLA emulsions were stored at 4°C until infused.

Infusion periods lasted 4 days, with a 7-day interval between periods. Abomasal infusions involved a 0.5 cm (ID) polyvinyl chloride tubing, which passed through the rumen fistula and sulcus omasi into the abomasum as previously described (32). Emulsions were continuously infused using pumps (Plum Infusion System XL 11555; Abbott Laboratories, North Chicago, IL) programmed to provide 4 liters/day.

Cows were milked at 0600 and 1800 daily. At each milking, yield was determined and milk was sampled. One aliquot was stored at 4°C with a preservative (bronopol tablet; D&F Control System, San Ramon, CA) until analyzed for fat and protein content by infrared analysis (New York DHI, Ithaca, NY). Because milk fat content was severely reduced during some treatments, we verified the accuracy of the infrared analysis (method 989.05; Ref. 1) and results were in agreement (±2.2%, n = 6). A second aliquot of milk was stored at 20°C until analyzed for fatty acid composition. Lipid extraction was performed according to Hara and Radin (17). Milk fatty acids were transesterified according to the method of Christie (7) with modifications (5). The CLA supplements consisted of free fatty acids and were methylated using 1% sulfuric acid in methanol as described by Christie (8).

Fatty acid methyl esters were quantified by a gas chromatograph (Hewlett Packard GCD system HP G1800 A; Avondale, PA) equipped with a Supelcowax-10 fused silica capillary column [60 m × 0.37 mm (ID) with 0.25-µm film thickness] as previously described (5). Each peak was identified and quantified using pure methyl ester standards (Nu Check Prep, Elysian, MN). Additional standards for CLA isomers were obtained from Natural Lipids. A butter oil reference standard (CRM 164; Commission of the European Communities, Community Bureau of Reference, Brussels, Belgium) was used to determine recoveries and correction factors for individual fatty acids. Further analysis by high-resolution nuclear magnetic resonance spectroscopy (¹³C) demonstrated that the 9,11 CLA and 10,12 CLA were almost exclusively the cis-9,trans-11 isomer and trans-10,cis-12 isomer, respectively (M. Aursand and A. Saebø, Natural Lipids; personal communication).

Blood samples were obtained from the coccygeal vein after each milking during the infusion period. Sodium heparin (100 U/ml of blood) was used to prevent coagulation. Plasma was immediately harvested (2,300 g, 15 min at 4°C) and stored at 20°C until analyzed. Plasma glucose concentration was determined by enzymatic colorimetric analysis using a commercial kit (510A; Sigma, St. Louis, MO). Nonesterified fatty acids (NEFA) were determined by enzymatic colorimetric analysis (Wako Pure Chemical Industries, Osaka, Japan) as modified by Sechen et al. (31). Both glucose and NEFA were analyzed within a single assay, and the intra-assay coefficients were 2.0 and 4.4%, respectively.

Plasma concentrations of insulin and insulin-like growth factor-I (IGF-I) were quantified using a double-antibody RIA. Insulin RIA was as described (22) using pancreatic bovine insulin (lot 615-70N-80; Lilly Research Laboratories, Greenfield, IN) for iodination and to create standards. Primary antibody was guinea pig anti-porcine insulin (lot 122845-P; Linco Research, St. Louis, MO), secondary antibody was caprine anti-guinea pig IgG serum (lot GP2022; Linco Research), and the carrier was normal guinea pig IgG serum (lot NGP026; Linco Research). Plasma concentrations of IGF-I were determined after dissociation and inactivation of the IGF binding proteins using a glycyl-glycine HCl extraction procedure as validated by Plaut et al. (29). Recombinant bovine IGF-I (lot GTS-3; Monsanto, St. Louis, MO) was used for iodination and to create standards; primary anti-serum (polyclonal UB2-495) was supplied by the National Hormone and Pituitary Program. Both insulin and IGF-I were analyzed within a single assay, and the intra-assay coefficients of variation were 7.4 and 3.7%, respectively.

Data were statistically analyzed as a 3 × 3 Latin square design using the PROC MIXED procedure of SAS (30) according to the following model
where Y_ijk is observation, µ is overall mean, T_i is treatment (i = 1, 2, and 3), P_j is period (j = 1, 2, and 3), C_k is cow (k = 1, 2, and 3), and E_ijk is residual error.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

We abomasally infused CLA supplements as an experimental method to bypass the rumen fermentation processes, thus avoiding biohydrogenation of the polyunsaturated supplement. Our goal was to infuse 10 g/day of the CLA isomer of interest. Actual rates of infusion were very close to our target, being 1 and 3% higher for the cis-9,trans-11 CLA and trans-10,cis-12 CLA isomers, respectively (Table 2). In addition to the CLA isomer of interest, both supplements contained small amounts of other CLA isomers. As a result, the 9,11 CLA supplement provided ~1.1 g/day of trans-10,cis-12 CLA and the 10,12 CLA supplement provided ~0.7 g/day of cis-9,trans-11 (Table 2).

Milk fat percentage and milk fat yield were substantially reduced when the 10,12 CLA supplement was infused (Table 3). These effects were specific for the trans-10,cis-12 CLA; infusion of a similar amount of the 9,11 CLA supplement had no effect on milk fat. The temporal pattern revealed that milk fat percentage progressively decreased during the 4-day infusion of the 10,12 CLA supplement and returned to normal by day 4 postinfusion (Fig. 1). In contrast to milk fat, yield of milk and milk protein was not altered by any treatment, although infusion of the 10,12 CLA supplement slightly reduced milk protein content (Table 3). Furthermore, feed intake was similar for all treatment groups, with a tendency for a lower intake during infusion of the 10,12 CLA supplement.

View larger version (14K): [in this window] [in a new window]   Fig. 1.   Temporal pattern of milk fat content during abomasal infusion of conjugated linoleic acid (CLA) supplements. Infusions were for 4 days as indicated, and treatments were control (), 9,11 CLA supplement (; 10 g/day), and 10,12 CLA supplement (; 10 g/day). Values represent means from 3 cows; SE for milk fat percentage ranged from 0.003 to 0.300. See Table 2 for composition of CLA supplements.

The fatty acid composition of milk fat is presented in Table 4. As expected, each CLA supplement caused an increase in the corresponding CLA isomer concentration in milk fat. Infusion of the 9,11 CLA supplement increased the milk fat content of cis-9,trans-11 CLA to 8.0 mg/g of fat, whereas infusion of the 10,12 CLA supplement enhanced the milk fat content of trans-10,cis-12 CLA from trace levels to 3.9 mg/g of fat. For most other fatty acids, milk composition was comparable for control and 9,11 CLA supplement treatments (Table 4). In contrast, during infusion of the 10,12 CLA supplement, there was a reduction in the contribution of C_4:0 to C_16:0 fatty acids and a corresponding increase in the percentage of most longer-chain fatty acids. The temporal patterns for the reduction in milk fat yield of C_4:0 to C_16:0 and total fatty acids are shown for the 10,12 CLA supplement in Fig. 2. In addition, the ratios of C_14:0 to C_14:1 and C_18:0 to C_18:1 were substantially increased with the abomasal infusion of the 10,12 CLA supplement (Table 4).

View larger version (13K): [in this window] [in a new window]   Fig. 2.   Temporal pattern for reduction in yield of milk fatty acids during 4-day abomasal infusion of 10,12 CLA supplement. Decrease in yield is presented for total fatty acids (open bars) and de novo synthesized fatty acids (C4:0 to C16:0; solid bars), and SE averaged 220 mmol/day. On day 0, 62% of milk fatty acids (molar basis) were C4:0 to C16:0 and 38% were C18 and longer.

We also examined two metabolites and two hormones associated with lipid metabolism and energy homeostasis (Table 5). Neither CLA supplement had an effect on circulating levels of glucose (means = 75.2 mg/dl) or insulin (means = 3.7 ng/ml). There was a small increase (13%) in plasma NEFA levels when cows received the 10,12 CLA supplement, whereas infusion of the 9,11 CLA supplement resulted in slightly greater (17%) plasma concentrations of IGF-I.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Recent studies have demonstrated that abomasal infusion of CLA supplements containing a mixture of CLA isomers considerably reduces the percentage and yield of milk fat in lactating dairy cows (5, 6, 20). A number of different dietary conditions, such as plant oil supplements, high-concentrate/low-fiber diets, and diets with small fiber particle size, also cause a dramatic reduction in milk fat yield. Several theories have been proposed and subsequently shown to be inadequate to explain the mechanisms for dietary-induced milk fat depression (9, 12). Recently, Griinari et al. (15) established that dietary-induced milk fat depression corresponds to an increase in milk fat content of trans-10 C_18:1 and suggested that this specific trans isomer or related metabolites might be responsible. The trans-10 C_18:1 found in milk fat originates in the rumen via reduction of trans-10,cis-12 CLA formed in the biohydrogenation of linoleic acid (14). We also found that rumen concentrations of trans-10,cis-12 CLA are negatively correlated with milk fat percentage and yield for diets that cause milk fat depression (16).

On the basis of the above observations, we hypothesized that the trans-10,cis-12 CLA isomer would decrease milk fat synthesis and, results of the present study confirm our hypothesis. After 4 days of abomasal infusion, the 10,12 CLA supplement resulted in a 42% reduction in milk fat percentage and a 44% decrease in milk fat yield (Fig. 1). In contrast, infusion of similar amounts of the cis-9,trans-11 CLA isomer had no effect on milk fat (Table 3). Effects of the 10,12 CLA supplement on milk fat are comparable to the reductions observed in other studies with lactating cows that used a mixture of CLA isomers (5, 6, 20). Furthermore, effects of CLA on milk synthesis are specific for fat as yields of milk and milk protein were not altered by abomasal infusion of the trans-10,cis-12 CLA isomer (Table 3) or by infusion of other CLA supplements containing a mixture of isomers (5, 6). The CLA isomers containing a double bond at carbon-10 appear to be key in reducing milk fat synthesis. We recently demonstrated that a CLA supplement containing an equal mixture of cis/trans 8,10 and 9,11 isomers of CLA caused a reduction in milk fat synthesis (6). This effect was likely due to the 8,10 CLA isomer, given results of the present study showing the 9,11 CLA supplement had no effect.

Mechanisms by which CLA inhibits milk fat synthesis are unknown. The temporal pattern for the reduction in milk fat content in dairy cows was progressive over the 4-day infusion and returned to normal levels over a similar time interval after 10,12 CLA supplementation ceased (Fig. 1). Fatty acids in milk fat arise from both de novo synthesis by the mammary gland (C_4:0 to C_14:0 plus part of C_16:0), and mammary uptake of preformed fatty acids (C₁₈ plus part of C_16:0). The 10,12 CLA supplement resulted in a more dramatic reduction in fatty acids originating from de novo synthesis compared with preformed fatty acids (Table 4). Similar results were obtained when Chouinard et al. (6) abomasally infused CLA supplements comprised of various CLA isomers, all of which contained cis/trans 10,12 CLA or cis/trans 8,10 CLA . When data for milk fat composition (Table 4) and milk fat yield (Table 3) are combined, 78% of the reduction (mmol basis) in milk fat that occurred with infusion of 10,12 CLA supplement was accounted for by fatty acids of chain length C₄ to C₁₆ (Fig. 2). Thus CLA may be inhibiting the activity or synthesis of key enzymes involved in de novo fatty acid synthesis such as acetyl CoA carboxylase and fatty acid synthetase.

We also observed an increase in the ratios of C_14:0 to C_14:1 and C_18:0 to C_18:1 in milk fat of cows receiving the 10,12 CLA supplement (Table 4); similar shifts occurred when a mixture of CLA isomers was used (5, 6). These ratios are related to the activity of ⁹-desaturase in the mammary gland, and the increase indicates that trans-10,cis-12 CLA caused a reduction in the activity or the amount of this enzyme. The net effect was a reduction in yields of myristoleic and oleic acids, two fatty acids that are preferentially acylated to the sn-3 position of the triglyceride. This is important because milk fat triacylglycerides must contain a range of fatty acids that have characteristics allowing for proper fluidity (28). Lee et al. (19) reported that a mixture of CLA isomers decreased the mRNA abundance of ⁹-desaturase in the liver of in vivo-treated mice and in vitro studies with a mouse hepatocyte cell line. They further showed that the inhibitory effect on mRNA expression in the H2.35 hepatocytes was by CLA isomers other than cis-9,trans-11 isomer. Using ratios of specific fatty acid pairs as a proxy for desaturase activity, our results are consistent with the lack of an effect by cis-9,trans-11 CLA and further demonstrate that it is the trans-10,cis-12 CLA isomer that alters desaturase activity.

There may be species and physiological state differences in the effects of CLA on fat synthesis. Similar to our results with lactating cows, Masters et al. (21) observed that dietary supplementation of CLA to nursing women reduced the milk fat content and fat secretion. However, Chin et al. (4) reported that addition of CLA (mixture of isomers) to the diet of lactating rats had no effect on growth of the nursing pups, thereby suggesting that milk fat content was not altered. Dietary CLA also causes a reduction in body fat accretion during the growth phase in a number of species. Mice, rats, chicks, and pigs have substantial reductions in body fat when fed diets containing a mixture of CLA isomers (10, 11, 26, 27, 33). Comparisons are complicated by differences in the amount and type of CLA isomers used in various studies, but the level of dietary CLA (0.5-2.0% of diet) that decreases body fat in growing animals is substantially greater than the CLA level we used to obtain a reduction in milk fat synthesis (~0.05% of diet). Recently, Park et al. (27) provided evidence that the trans-10,cis-12 CLA was the isomer causing the reduction in body fat accretion in growing mice. Thus it seems likely the trans-10,cis-12 CLA isomer that we find causes a reduction in milk fat synthesis may also be the CLA isomer causing reductions in body fat in different species of growing animals.

The specific mechanisms whereby CLA alters lipid metabolism are not clear but could involve a number of processes. One mechanism may involve increases in rates of lipolysis and fatty acid oxidation. Support for this mechanism includes observations of increased hormone-sensitive lipase activity and enhanced catecholamine-induced lipolysis in adipocytes isolated from rats fed CLA (25), greater glycerol release in 3T3-L1 adipocytes cultured with CLA (27), and enhanced carnitine palmitoyltransferase activity in several tissues of mice fed CLA (26). In lactating cows, circulating concentrations of NEFA are highly correlated with rates of lipolysis (3), and the relatively minor changes we observed with CLA supplement (Table 5) suggest that CLA has little or no effect on lipolysis. Another possible mechanism for CLA to alter lipid metabolism would be to reduce tissue uptake of fatty acids. This involves lipoprotein lipase, and the activity of this enzyme was decreased when 3T3-L1 adipocyte cultures were incubated with CLA (26, 27). If lipoprotein lipase in the mammary glands was affected in our study, a reduction in the use of preformed fatty acids for milk fat synthesis would be expected. A reduction was observed, but effects on de novo synthesized fatty acids were more extensive (Fig. 2). In growth studies, the extent to which the CLA-induced reduction in body fat reflects changes in de novo synthesis of fatty acids versus uptake of preformed fatty acids has not been examined. However, both processes appear to be altered on the basis of studies by West et al. (33), which reported dietary CLA reduced body fat in mice fed either a high-carbohydrate or a high-fat diet.

In the lactating cow, the 10,12 CLA supplement had no effect on circulating concentrations of glucose, insulin, or IGF-I. However, studies with laboratory animals have indicated that dietary CLA can affect energy metabolism and glucose homeostasis. In the AKR/J mice, CLA caused an increase in plasma insulin concentrations (10) and a shift in the nutrient mixture oxidized for energy (33). In addition, CLA has been shown to improve glucose tolerance and glucose homeostasis in the Zucker obese fatty rat. In particular, CLA enhanced insulin sensitivity and this effect may have been mediated by peroxisome proliferator-activated receptors (PPAR)  and other members of the PPAR family (18).

In summary, abomasal infusion of 10 g/day trans-10,cis-12 CLA (~0.05% of intake) dramatically reduced milk fat content and yield. Effects were specific for the fat component of milk and specific for the 10,12 CLA supplement, because abomasal infusion of cis-9,trans-11 CLA had no effect on milk fat. Furthermore, the fat reduction was evident within 24 h after initiating 10,12 CLA infusion and persisted throughout the infusion period. The mechanisms by which the trans-10,cis-12 CLA isomer causes a reduction in milk fat synthesis is unknown, but changes in the milk fatty acid composition suggest an inhibition of the pathways of de novo lipogenesis and a reduction in ⁹-desaturase activity. Overall, data support our hypothesis that the trans-10 double bonds of octadecadienoic acids are a major factor in the inhibition of milk fat synthesis. When our results are combined with other recent data (27), it appears that CLA isomers with trans-10 double bonds may be responsible for CLA-induced reductions in milk fat synthesis during lactation and body fat accretion during growth in a wide range of species.

Perspectives