The Journal of Nutrition Vol. 128 No. 5 May 1998, pp. 881-885

Dietary Fatty Acid Sources Affect Conjugated Linoleic Acid Concentrations in Milk from Lactating Dairy Cows^1,2,3

Miriam L. Kelly⁴, Julie R. Berry, Debra A. Dwyer, J.M. Griinari^*, P. Yvan Chouinard, Michael E. Van Amburgh, and Dale E. Bauman⁵

Department of Animal Science, Cornell University, Ithaca, NY 14853 and ^* Valio Inc., Research and Development Center, 00101 Helsinki, Finland

	ABSTRACT

Abstract Introduction Methods Results Discussion References

Conjugated linoleic acid (CLA), a naturally occurring anticarcinogen found in dairy products, is an intermediary product of ruminal biohydrogenation of polyunsaturated fatty acids. Our objective was to determine the effect of different dietary oils, which vary in fatty acid composition, on CLA concentrations in milk from lactating dairy cows. Twelve Holstein cows were randomly assigned to a 3 × 3 Latin square design. Dietary treatments were the addition (53 g/kg dietary dry matter) of peanut oil (high oleic acid), sunflower oil (high linoleic acid) and linseed oil (high linolenic acid). Each treatment period was 2 wk, and milk samples were collected on the last 4 d of each period. Milk yield (34.2 ± 1.3 kg/d) and milk fat (2.25 ± 0.06%) were not different among treatments. Milk protein during the sunflower oil treatment (mean, 3.44% protein) was significantly higher (P < 0.01) than during the other treatments. Milk fat concentration of CLA during the sunflower oil treatment was significantly different from other treatments (P < 0.001) and ~500% greater than typically observed when cows consume traditional diets. CLA concentrations (mg/g of milk fat) were 13.3, 24.4 and 16.7 during peanut oil, sunflower oil and linseed oil treatment, respectively. CLA concentration in milk fat can be enhanced by the addition of polyunsaturated fatty acids to the diet, especially oils high in linoleic acid.

	INTRODUCTION

Abstract Introduction Methods Results Discussion References

Conjugated linoleic acid (CLA) refers to a mixture of positional and geometric isomers of octadecadienoic acid with conjugated double bonds. CLA is the only fatty acid shown unequivocally to inhibit carcinogenesis in experimental animals (NRC 1996). Unlike most anticarcinogenic compounds, CLA can both reduce the incidence of tumors in a number of experimental animal models and serve as a cytotoxic agent against existing tumor cells (Belury 1995, Parodi 1997, Scimeca et al. 1994). The anticarcinogenic effects of CLA occur at low dietary concentrations; the current average intake of humans is close to the dietary level that demonstrates anticarcinogenic effects in experimental animal models (Ip et al. 1994). Recent publications have also shown that CLA may have positive effects on cardiovascular risk factors in animal models (Lee et al. 1994, Nicolosi et al. 1993).

CLA is found predominately in food products from ruminant animals; dairy products are the major source in the human diet. CLA is formed as a result of incomplete biohydrogenation of dietary fatty acids in the rumen. In the rumen, dietary lipids are rapidly hydrolyzed, and the resulting free unsaturated fatty acids are subjected to biohydrogenation by the rumen microorganisms (Harfoot and Hazlewood 1988). As a consequence, the ruminant animal absorbs mainly saturated fatty acids, and food products from ruminants tend to have a more saturated fatty acid composition regardless of the fatty acid composition of the diet. However, when biohydrogenation is not complete, CLA can escape the rumen and be absorbed from the gastrointestinal tract, thereby providing the mammary gland with the source of CLA that is found in milk fat.

Typical concentrations of CLA in milk fat are 3-6 mg/g of fat, but the levels of CLA in milk can vary widely among herds (Kelly and Bauman 1996, Reil 1963). However, the specific factors that cause these variances have not been extensively investigated. Previous studies demonstrated that the milk fat concentration of CLA was dependent on the presence of unsaturated fat in the diet (Griinari et al. 1996), and that CLA concentrations increased as dietary level of corn oil increased (McGuire et al. 1996). The objective of this study was to compare the effect of dietary addition of different plant oils on milk fat content of CLA. The plant oils examined were similar in unsaturated fatty acid content but varied in proportions of C18 mono-, di- and trienes.

	MATERIALS AND METHODS

Abstract Introduction Methods Results Discussion References

All procedures involving animals were approved by the Cornell University Institutional Animal Care and Use Committee.

Twelve Holstein cows were separated from the rest of the herd and housed in individual tie stalls at the Cornell University Teaching and Research Center (Harford, NY). The cows (mean, 134 d postpartum) were randomly assigned to a 3 × 3 Latin square design. The four Latin squares were completed simultaneously. Treatments included the addition of peanut oil, sunflower oil or linseed oil to the diet. Peanut oil was acquired from F & T Distributing (Ithaca, NY). Sunflower and linseed oils were provided by Cargill (Liverpool, NY).

Formulation of the diets was completed using the Cornell Net Carbohydrate and Protein System (CNCPS) (Fox et al. 1992, Russell et al. 1992, Sniffen et al. 1992). The CNCPS was used to estimate microbial protein yield and metabolizable protein requirements. On the basis of this analysis, a protein supplement, low in rumen degradability and fat, was formulated (Taylor By-Products, Wyalusing, PA) and included in the diet. A summary of dietary ingredients is presented in Table 1. Oils were added at a level of 5.3% of dietary dry matter, resulting in dietary lipid content of 8.5%. On a daily basis, each oil was blended with the basal diet in a mixer wagon to ensure equal distribution of the oil in the ration. Cows had free access to the balanced diet throughout the day with fresh feed provided each morning. Feed intakes were monitored daily.

Over a 14-d pretrial period, the cows were gradually acclimated to the treatment diet that they would receive in period 1, to allow the rumen microbial population to adjust to the high oil content. After the adjustment period, each of the dietary treatments was fed to the cows for a 2-wk period in a Latin square design. Samples of the diets were taken weekly and used for dry matter and nutrient analysis. Dry matter was determined by drying at 60°C for 48 h. Weekly samples were composited for each period for nutrient analysis by the Northeast Dairy Herd Improvement Association (Ithaca, NY). Chemical composition of the diet is presented in Table 1.

Cows were milked twice daily (1000 and 2200 h) in the milking parlor at the facility. Milk yield was recorded daily and milk samples collected each milking of the last 4 d of each 2-wk treatment period. Milk samples were stored at 4°C until analyzed for fat and protein composition by infrared analysis (Northeast Dairy Herd Improvement Association). Another set of milk samples was combined on an individual cow basis to form a pool for each day and each period, and stored at 20°C until analysis for fatty acid composition.

For fatty acid analysis, the frozen milk samples were rapidly thawed according to the method of Lynch et al. (1992). An aliquot of thawed milk (~30 mL) was centrifuged at 17,800 × g for 30 min at 8°C and then 300-530 mg of fat cake was removed. Lipid extraction of milk fat was performed according to Hara and Radin (1978) by using a hexane/isopropanol (3:2; v/v) solution (18 mL/g of fat cake) followed by a 67 g/L sodium sulfate solution (12 mL/g of fat cake).

Fatty acid methyl esters for milk fat analysis were prepared by the transmethylation procedure of Christie (1982) with modifications. Hexane (2mL, HPLC grade) was added to the 40 mg of lipid followed by 40 µL of methyl acetate. After vortexing, 40 µL of methylation reagent (1.75 mL methanol/0.4 mL of 5.4 mol/L sodium methylate) was added. The mixture was vortexed and allowed to react for 10 min; then 60 µL of termination reagent (1 g oxalic acid/30 mL diethyl ether) was added. The sample was then centrifuged for 5 min at 2400 × g at 5°C and the liquid portion transferred to labeled Wheaton vials and stored at 20°C.

Gas chromatography of fatty acid methyl esters was performed using a Hewlett Packard GCD system (G1800 A; Avondale, PA), equipped with HP G107A GCD software for peak integration. In general, specifications were similar to those given by Shantha and Decker (1993). Conditions were as follows: a fused silica capillary column (2-4082, Suplecowax 10, 60 m × 0.32 mm i.d. × 0.25 µm film thickness; Supleco, Bellefonte, PA); ultra high purity helium carrier gas; an electron ionization detector and a split injection inlet (87.5:1, v/v). The injector and detector were maintained at 250°C, and the column temperature programming was 50°C for 1 min, then increased 20°C/min to 200°C and held for 38 min.

A butter oil reference standard (CRM 164; Commission of the European Communities, Community Bureau of Reference, Brussels, Belgium) was routinely chromatographed and used to determine rates of recoveries and correction factors for individual fatty acids. A standard mixture of fatty acids (GLC-60; Nu-Chek-Prep, Elysian, MN) was also chromatographed to verify retention times and rates of recoveries. Furthermore, because CLA was of special interest, a CLA standard provided by M. Pariza (University of Wisconsin, Madison, WI) was used to determine correct retention times for the cis-9, trans-11 isomer of CLA.

All cows completed the full Latin square. Statistical analysis was conducted with the use of the following model: where Y_ij represents the observations for the dependent variables, µ is the the least-square mean average, S_i is the systematic effect of square i, C_j(S_i) is the systematic effect of cow j within square i, P_k is the systematic effect of period k, T_l is the systematic effect of treatment l, (ST)_il is the interaction between square i and treatment l, and E_ijkl is the residual error.

The interaction between square and treatment was tested, found to be nonsignificant and then removed from the model. Main effects (oil type) were tested using Duncan's multiple range procedure with the General Linear Models procedure (SAS 1989). Differences were considered significant when P  0.1. Values in the text are presented as means ± SEM.

	RESULTS

Abstract Introduction Methods Results Discussion References

The three plant oils used in the study were chosen because they differed in fatty acid composition (Table 2). The major fatty acid in peanut oil was oleic acid (51.5%); sunflower oil was highest in linoleic acid (69.4%), and linseed oil contained a high proportion of linolenic acid (51.4%). The dry matter intake was relatively constant during each period and did not differ among the three treatments (P > 0.1; data not presented). Across all treatments, dry matter intake averaged 21.1 ± 0.3 kg/d.

Milk production performance and composition data are presented in Table 3. Milk yield did not differ among treatments (P > 0.1), averaging 34.2 ± 1.3 kg/d. The addition of plant oils to the diet did not disturb cow performance or well being on the basis of the consistency of the daily milk yield and dry matter intake. The milk fat test was relatively low for all treatments but neither milk fat test nor milk fat yield differed among treatments (P > 0.1). Milk protein content was significantly (P < 0.01) greater during the sunflower oil treatment compared with the peanut and linseed oil treatments.

The fatty acid composition of milk fat is presented in Table 4. The addition of peanut oil to the diet resulted in a greater concentration of palmitic acid (16:0) (P < 0.001). Similarly, the addition of sunflower oil increased the concentration of octadecenoic acid (18:1) relative to those when peanut or linseed oils were fed (P < 0.001). Although the level of linolenic acid (18:3) in the milk was low, feeding linseed oil doubled the concentration as well as increased the level of the "other" fatty acids in the milk (P < 0.001).

The milk fat concentration of CLA was affected by diet during the sunflower oil treatment, resulting in the highest concentrations of CLA (Table 4). In spite of the fact that cows were at a similar stage of lactation and milk production level, and received the same diet under the same management, there was a substantial variation among individuals regardless of treatment (Fig. 1). This variation was greatest for the sunflower oil diet which had a fivefold range in milk fat concentration of CLA, with CLA representing over 5% of milk fatty acids in the cow with the greatest milk concentration. From the standpoint of the anticarcinogenic effects of CLA, the goal is to increase the CLA concentration of milk fat rather than the absolute amount secreted in milk. However, milk yield of CLA was greatest when the cows received the sunflower oil treatment, 17.8 g CLA/d compared with 10.2 and 13.5 g CLA/d for peanut oil and linseed oil treatments, respectively.

View larger version (9K): [in this window] [in a new window]   Fig 1. Milk content of conjugated linoleic acid (CLA) for individual cows fed supplemental oils. Cows (n = 12) received diets containing 5.3% of the specific plant oil as indicated. Treatment averages for CLA concentrations were 13.3, 24.4 and 16.7 mg/g of milk fat for peanut oil, sunflower oil and linseed oil, respectively (SEM = 1.7 mg/g milk fat). The sunflower oil diet differed from other treatments (P < 0.001).

	DISCUSSION

Abstract Introduction Methods Results Discussion References

The major objective of this study was to examine the effect of feeding different oils on CLA concentration of milk fat. All of the plant oil diets resulted in milk fat concentrations of CLA that were several fold greater than the 3-6 mg/g of milk fat typically observed with traditional forage/concentrate diets. CLA concentrations were affected by dietary treatment; concentrations during the sunflower oil treatment were significantly greater than during the other treatments (Table 4).

The biohydrogenation pathway of linoleic acid involves an initial isomerization step resulting in the formation of cis-9, trans-11 CLA, which undergoes sequential reduction steps yielding trans-11-octadecenoic acid and then stearic acid (Fig. 2). Investigations of the pathway have involved identifying some of the rumen bacteria that catalyze the biohydrogenation sequence, as well as isolating and characterizing the enzymes that catalyze the specific reactions (Harfoot and Hazlewood 1988, Vivani 1970). Although linoleate isomerase and CLA reductase have been purified from the microorganism Butyrivibrio fibrisolvens (Hughes et al. 1982, Kepler and Tove 1967), in general, no one species of microorganisms carries out the full sequence of biohydrogenation (Harfoot and Hazlewood 1988). In a review, Vivani (1970) proposed that CLA was also formed as an intermediate in the biohydrogenation pathway of linolenic acid. However, in the biohydrogenation studies with rumen microorganisms, linolenic acid (cis-9, cis-12, cis-15-octadecatrienoic acid) has been shown to be converted to cis-9, trans-11, cis-15 conjugated triene, then to trans-11, cis-15 18:2, and finally to an octadecenoic acid which is either trans-11, trans-15, or cis-15 (Harfoot and Hazlewood 1988). Therefore, linolenic pathways do not involve CLA as an intermediate. However, it has been proposed that cis-9, trans-11 CLA could be produced endogenously from trans-11 octadenoic acid in the body tissues by -9 desaturase (Griinari et al. 1997). In this study, the feeding of linseed oil, which contains high concentrations of linolenic acid, increased CLA levels but not to the extent observed when cows consumed sunflower oil.

View larger version (21K): [in this window] [in a new window]   Fig 2. Pathway of biohydrogenation of linoleic acid to stearic acid by rumen microorganisms. Adapted from Harfoot and Hazlewood (1988).

Previous work has suggested that the biohydrogenation sequence of linoleic acid can lead to an increase in CLA levels in milk fat. McGuire et al. (1996) showed that increasing dietary unsaturated fatty acids by adding corn oil, which contains ~50% linoleic acid, resulted in an increase in milk CLA levels with the magnitude of the increase directly related to the dietary level of corn oil. Griinari et al. (1996) also showed that changing forage to concentrate ratios and the addition of dietary unsaturated fatty acids such as corn oil also enhanced milk fat concentrations of CLA. In this study, all plant oils contained linoleic acid; sunflower oil had the greatest content and resulted in the highest concentration of CLA in milk fat (Fig. 1).

Reil (1993) and Kelly and Bauman (1996) reported a wide variation in milk fat content of CLA among individual herds. However, variation among individual cows has been less extensively examined. Recently, Kelly et al. (1998) examined this variation for cows fed either a pasture diet or a totally mixed diet of forage and concentrate. Although they observed that individuals maintained relatively constant milk fat concentrations of CLA across time, there was a threefold variation among individuals fed the same diet. Results in this study also show that a substantial individual variation exists (Fig. 1). This is especially impressive because all cows were at a similar stage of lactation, consumed the same diet at a similar level of intake and produced a similar amount and composition of milk. For example, when cows consumed the sunflower oil diet, the CLA concentration averaged 24.4 mg/g of milk fat, but individuals ranged from 9.9 to 51.7 mg/g of milk fat. Obtaining milk with a high CLA content would be of interest to examine the anticarcinogenic effects with natural food products in studies with experimental animal models. Results from this study show that it is possible to obtain high concentrations of CLA in milk fat; during the sunflower oil treatment, one third of the cows had CLA concentrations >33 mg/g of milk fat, with CLA averaging >5% of the milk fatty acids in the cow with the highest concentration (Fig. 1).

During all treatments, milk fat content was relatively low (mean 2.25%), compared with a pretreatment average of 3.38%. This low milk fat concentration is characteristic of milk fat depression that commonly occurs when diets high in plant oils are fed (Davis and Brown 1970). High intakes of dietary fat can also cause a decrease in milk protein percentage and yield. In this case, the dietary fat adversely affects microbial fermentation and microbial protein yield, thereby decreasing the supply of amino acids available for absorption by the cow (Palmquist and Jenkins 1980). In this study, the formulation of the diet was completed with the use of the CNCPS with special attention to metabolizable protein supply (Fox et al. 1992, Russell et al. 1992, Sniffen et al. 1992). The diet was designed to maximize microbial protein yield and allow for adequate rumen escape protein by the inclusion of a supplement formulated with less rapidly degraded proteins. The absence of any apparent negative effects on milk protein content, even with the relatively high fat diets, demonstrated the success of the formulations.

Overall, this study indicates that CLA content in milk fat of dairy cows can be increased via dietary manipulation. The addition of plant oils high in unsaturated fatty acids, particularly linoleic acid, markedly enhanced the milk fat content of CLA. The greatest increase was observed when sunflower oil was added to the diet, resulting in CLA concentrations that averaged four- to fivefold greater than typically observed. A substantial individual variation was also observed, indicating that many additional factors must be affecting ruminal production and subsequent milk concentrations of CLA.

	ACKNOWLEDGMENTS

The skilled assistance of Dottie Ceurter, Alla Zarifyan and Kristin Perkins is greatly appreciated. The authors also gratefully acknowledge the donation of the sunflower oil and linseed oil by Cargill and the protein supplement by Taylor By-Products.

	FOOTNOTES

Manuscript received 3 November 1997. Initial reviews completed 25 November 1997. Revision accepted 14 January 1998.

	LITERATURE CITED

Abstract Introduction Methods Results Discussion References

• Belury M. A. Conjugated dienoic linoleate: a polyunsaturated fatty acids with unique chemical properties. Nutr. Rev. 1995; 53:83-89 [Medline]
• Christie W. W. A simple procedure of rapid transmethylation of glycerolipids and cholesteryl esters. J. Lipid Res. 1982; 23:1072-1075 [Abstract]
• Davis, C. L. & Brown, R. E. (1970) Low-fat milk syndrome. In: Digestion and Metabolism in the Ruminant (Phillipson, A. T., ed.), pp. 545-565. Oriel Press, Newcastle upon Tyne, England.
• Fox D. G., Sniffen C. J., O'Connor J. D., Russell J. B., Van Soest P. J. A net carbohydrate and protein system for evaluating cattle diets: III. Cattle requirements and diet adequacy. J. Anim. Sci. 1992; 70:3578-3596 [Abstract]
• Griinari, J. M., Chouinard, P. Y. & Bauman, D. E. (1997) Trans fatty acid hypothesis of milk fat depression revised. In: Proc. Cornell Nutr. Conf. Feed Manuf., pp. 208-216. Cornell University, Ithaca, NY.
• Griinari, J. M., Dwyer, D. A., McGuire, M. A. & Bauman, D. E. (1996) Partially hydrogenated fatty acids and milk fat depression. J. Dairy Sci. 79 (suppl. 1): 177 (abs.).
• Hara A., Radin N. S. Lipid extraction of tissues with a low-toxicity solvent. Anal. Biochem. 1978; 90:420-426 [Medline]
• Harfoot, C.G. & G.P. Hazlewood. 1988. Lipid metabolism in the rumen. In: The Rumen Microbial Ecosystem (Hobson, P.N., ed.), pp. 285-322. Elsevier Science Publishers B.V., Amsterdam, The Netherlands.
• Hughes P. E., Hunter W. J., Tove S. B. Biohydrogenation of unsaturated fatty acids: purification and properties of cis-9, trans-11-octadecadienoic reductase. J. Biol. Chem. 1982; 257:3643-3649 [Abstract/Free Full Text]
• Ip C., Scimeca J. A., Thompson H. J. Conjugated linoleic acid: a powerful anticarcinogen from animal fat sources. Cancer 1994; 74:1050-1054 [Medline]
• Kelly, M. L. & Bauman, D. E. (1996) Conjugated linoleic acid: a potent anticarcinogen found in milk fat. In: Proc. Cornell Nutr. Conf. Feed Manuf., pp. 68-74. Cornell University, Ithaca, NY.
• Kelly, M. L., Kolver, E. S., Bauman, D. E., VanAmburgh, M. E. & Muller, L. D. (1998) Effect of dietary intake of pasture on conjugated linoleic acid (CLA) concentrations in milk from lactating cows. J. Dairy Sci. (in press).
• Kepler C. R., Tove S. B. Biohydrogenation of unsaturated fatty acids: III. Purification and properties of a linoleate ¹²-cis, ¹¹-trans-isomerase from Butyrivibrio fibrisolvens. J. Biol. Chem. 1967; 242:5686-5692 ^{[Abstract/Free Full Text]}
• Lee K. N., Kritchevsky D., Pariza M. W. Conjugated linoleic acid and atherosclerosis in rabbits. Atherosclerosis 1994; 108:19-25 [Medline]
• Lynch J. M., Barbano D. M., Bauman D. E., Hartnell G.F., Nemeth M. A. Effect of a prolonged-release formulation of n-methionyl bovine somatotropin (Sometribove) on milk fat. J. Dairy Sci. 1992; 75:1794-1809 [Abstract]
• McGuire, M.A., McGuire, M. K., Guy, M. A., Sanchez, W. K., Shultz, T. D., Harrison, L. Y., Bauman, D. E. & Griinari, J. M. (1996) Effect of dietary lipid concentration on content of conjugated linoleic acid (CLA) in milk from dairy cattle. J. Anim. Sci. 74 (suppl. 1): 266 (abs.).
• National Research Council (1996) Carcinogens and Anticarcinogens in the Human Diet. National Academy of Science, Washington, DC.
• Nicolosi, R.J., Courtemanche, K. V., Laitinen, L., Scimeca, J. A. & Huth, P. J. (1993) Effect of feeding diets enriched in conjugated linoleic acid on lipoproteins and aortic atherogenesis in hamsters. Circulation 88 (suppl.): I-457 (abs.).
• Palmquist D. L., Jenkins T. C. Fat in lactation rations: review. J. Dairy Sci. 1980; 63:1-14
• Parodi P. W. Cow's milk fat components as potential anticarcinogenic agents. J. Nutr. 1997; 127:1055-1060 [Abstract/Free Full Text]
• Reil R. R. Physico-chemical characteristics of Canadian milk fat: unsaturated fatty acids. J. Dairy Sci. 1963; 46:102-106 ^{[Abstract/Free Full Text]}
• Russell J. B., O'Connor J. D., Fox D. G., Van Soest P. J., Sniffen C. J. A Net Carbohydrate and Protein System for evaluating cattle diets: I. Ruminal fermentation. J. Anim. Sci. 1992; 70:3551-3561 [Abstract]
• SAS Institute Inc. (1989) SAS/STAT User's Guide, version 6, 4th ed. SAS Institute, Cary, NC.
• Scimeca, J. A., Thompson, H. J. & Ip, C. (1994) Effect of conjugated linoleic acid on carcinogenesis. In: Diet and Breast Cancer, pp. 59-65. Amer. Inst. Breast Cancer, Plenum Press, New York, NY.
• Shantha N. C., Decker E. A. Conjugated linoleic acid concentrations in processed cheese containing hydrogen donors, iron and dairy-based additives. Food Chem. 1993; 47:257-261
• Sniffen C. J., O'Connor J. D., Van Soest P. J., Fox D. G., Russell J. B. A Net Carbohydrate and Protein System for evaluating cattle diets: II. Carbohydrate and protein availability. J. Anim. Sci. 1992; 70:3562-3577 [Abstract]
• Viviani R. Metabolism of long-chain fatty acids in the rumen. Adv. Lipid Res. 1970; 8:267-346 [Medline]

0022-3166/98 $3.00 ©1998 American Society for Nutritional Sciences