Nutrient Metabolism
Modification of Triacylglycerides and Apolipoprotein in Rats Fed Diets Containing Whole Milk, Skim Milk and Milk Proteins1 GLORIA MARQÃœEZ-RÃœIZ,2B. DIANE Department
of Nutrition,
university
RICHTER AND BARBARA
skim milk and some milk products can exert a hypocholesterolemic effect on humans and experi mental animals (Richardson 1978). Milk constituents such as orotic acid (Richardson 1978), lipoproteins (Ahmed et al. 1979), lactose (Helmes 1977), calcium ¡Keimet al. 1981, Thakur and Jha 1981), 3-hydroxy3-methylglutaric acid (Mann 1977) and a component of the fat globule membrane (Antila et al. 1980, Howard and Marks 1979) have been suggested as hypocholesterolemic factors. Furthermore, products of bacterial fermentation of milk have been reported as hypocholesterolemic (Gilliland 1989, Hepner et al. 1979). Overall, the evidence for the existence of a hypocholesterolemic milk factor is still not definitive, but among all milk products tested, skim milk seems to cause the greatest reduction of serum cholesterol (Ney 1991, Thomson et al. 1982). Most of the studies that have been conducted have focused on the effect of milk products on plasma cholesterol, whereas the influence on plasma and hepatic triacylglyceride concentrations has not been clearly established. Some investigators have reported a plasma triacylglyceride-lowering effect by addition of dairy products to the diet (Kiyosawa et al. 1984, Kritchevsky et al. 1979) whereas others have been unable to document any response (Jenkins et al. 1989, Vrecko et al. 1988). These contrasting results can be attributed to a variety of differences in experimental design, including total fat intake, the length of feeding period and the amount and type of dairy product consumed. Schneeman et al. (1989) showed that whole milk
INDEXING KEY WORDS:
•cholesterol
•casein
O. SCHNEEMAN3
of California, Dauis, CA 95616-8669
ABSTRACT The objective of this study was to de termine the effects of diets containing milk and milk protein fractions on plasma and hepatic lipids, apolipoprotein B mRNA abundance, and plasma apolipoprotein concentrations and lipoprotein composition. Male rats were fed for 6 wk diets that contained (wt/wt) 76% whole milk (WM diet), 55% skim milk (SMFF diet), 22% casein (CAS diet), 22% whey protein isolate (WHY diet) or 55% skim milk-low fat (SMLF diet). The fat concen tration in the SMLF diet was 7%. Butter oil (20%) and corn oil (2%) were added to the SMFF, CAS and WHY diets. Plasma and VLDL triacylglycerides in the WM-fed rats were about half of the level in the groups fed the SMFF and SMLF diets, but not significantly different from those of the WHY-fed group. Hepatic triacyl glycerides generally were lower in the WM-fed group than in the other groups. Plasma cholesterol concentration did not differ among groups. Plasma apolipoprotein B was significantly lower in the WM-fed group than in rats fed the SMLF, SMFF or WHY diets. However, apolipo protein B mRNA abundance in the liver and small intes tinal mucosa did not differ due to dietary treatment. Thus the lipemic response due to whole milk is not associated with milk protein fractions and may be due to the presence of fat globule membrane in the diet con taining whole milk. J. Nutr. 122: 1840-1846, 1992.
•fatty acid synthetase
B
•rats
•triacylglycerides
Bovine milk and some of its products have been considered atherogenic (Ney 1991) because of their cholesterol content and fatty acid profile. The pre dominant fatty acids are saturated; the ratio of polyunsaturated to saturated fatty acids is -0.08. From the literature, however, the relationship between con sumption of dairy products, especially milk, and risk of cardiovascular disease is not clear. Besides im plicating the fat content of milk, investigators have reported that casein, the major milk protein, is hypercholesterolemic in animals (Lovati et al. 1990). In contrast, numerous studies have shown that milk,
Supported by the National Dairy Promotion Board, adminis tered by the National Dairy Council. 2Current address: Instituto de la Grasa y sus Derivados (C.S.I.C.), Avenida Padre Garcia Tejero, 4, Apartado 1078, E-41012 Sevilla, Spain. Gloria Marquez-Ruiz was supported by a Postdoc toral Fellowship from NATO (Spain). 3To whom correspondence should be addressed.
0022-3166/92 $3.00 ©1992 American Institute of Nutrition. Received 7 November 1991. Accepted 16 May 1992.
1840 Downloaded from https://academic.oup.com/jn/article-abstract/122/9/1840/4769454 by University of Glasgow user on 04 April 2018
EFFECT OF MILK ON TRIACYLGLYCERIDES
1841
TABLE 1 Diet composition Diet milkmilklow fat55020506218850 full fat5502020010050
milk760209050 diet2202020029011850
Skim milk powder Whole milk powder CaseinWhey 2020029011850
Corn oilButter oilLactoseSucroseCornstarchCellulose
Mineral mix Vitamin mix1CholesterolCaHP04
60 200.50.214.97Skim
60 60 200.218.06Caseing/kg 2022
0.218.64Whey220
hydroxytolueneEnergy, Butylated
60 2022
60200.217.98
0.218.64Whole
Mf/kgSkim
Composition
was previously published in Schneeman and Gallaher (1980).
powder decreased triacylglyceride levels and the VLDL fraction in rats and that this response was not simply predicted by dietary fat content. Overall, the results suggested that milk is more likely to alter triacylglyceride than cholesterol metabolism in rats. Furthermore, the fraction of whole milk responsible for these effects apparently resides in the portion that contains milk protein or the fat globule membrane, because addition of butter oil alone did not lower either plasma or VLDL triacylglycerides. To investigate whether a protein fraction of whole milk is responsible for the changes in triacylglyceride metabolism, we conducted a controlled study on the effects of milk-based diets and diets containing iso lated milk protein fractions on plasma triacyl glycerides, cholesterol and apolipoprotein level, lipoprotein composition and hepatic lipids in rats.
METHODS Forty male Wistar rats (Hilltop Labs, Scottsdale, PA) were randomly assigned to five groups and fed diets containing whole milk (WM diet),4 skim milk plus full fat (SMFF diet), casein (CAS diet), whey protein isolate (WHY diet) or skim milk with low fat (SMLF diet). The experimental protocol was approved by the Animal Welfare Committee at the University of California, Davis. The compositions of the experi mental diets are shown in Table 1. The SMFF, CAS and WHY diets were adjusted (by the addition of com Downloaded from https://academic.oup.com/jn/article-abstract/122/9/1840/4769454 by University of Glasgow user on 04 April 2018
oil and butter oil) to contain the same fat level and composition as those found in the WM diet. The SMFF, CAS, SMLF and WHY diets were similar in total protein content to the WM diet. Although the casein and whey vary slightly in amino acid content from whole milk, skim milk has the same amino acid pattern. All the diets were balanced for calcium content because calcium may possess a hypocholesterolemic action (Keim et al. 1981, Thakur and Jha 1981). Animals were housed in temperature(21-23°C) and light-controlled rooms with a 12-h light:dark cycle beginning at 0800 h and at ambient humidity (30-60%). After 6 wk of ad libitum feeding, the animals were deprived of food for 16 h. Each rat was anesthetized with a "cocktail" composed of 100 g/L ketamine hydrochloride (0.15 mL), 20 g/L xylazine (0.075 mL) and 10 g/L acepromazine maléate (0.022 mL). Blood was collected from the thoracic aorta or by cardiac puncture into syringes containing a final con centration of 1 g EDTA/L. The intestinal mucosa was removed by scraping the intestine with a glass slide, then weighed and frozen in liquid nitrogen. The liver
Abbreviations used: apo, apolipoprotein; FAS, fatty acid synthetase (EC 2.3.1.85); MFGM, milk fat globular membrane. Diet abbreviations: CAS, diet containing casein; SMFF, diet containing skim milk plus full fat; SMLF, diet containing skim milk with low fat; WHY, diet containing whey protein isolate,- WM, diet con taining whole milk.
1842
MARQUEZ-RUIZ
was removed, freeze-clamped, frozen in liquid ni trogen and stored at -70°C. RNA was isolated from liver and intestinal mucosa samples. Tissue samples were homogenized in a guanidine isothiocyanate buffer, and total RNA was isolated by centrifugation on a cesium chloride-EDTA cushion (Chirigwin et al. 1979). Total RNA was esti mated by absorbance at 260 nm. The relative abun dance of mRNA coding for apolipoprotein (apo) B and actin was determined by slot-blot hybridization with specific 32P-labeled cDNA probes. Conditions for hy bridization and cDNA radiolabeling were those described by Maniatis et al. (1982). Autoradiographs of hybridizations were measured by densitometry. Relative abundance of apo B mRNA and of actin mRNA in each sample were expressed per microgram of total RNA and the ratio of apo B to actin mRNA abundance calculated. Plasma VLDL, LDL and HDL fractions were iso lated by sequential ultracentrifugation, as previously described (Ney et al. 1986). The plasma and lipoprotein fractions were analyzed for cholesterol by the method of Allain et al. (1974) and for triacylglycerides using Sigma Procedure no. 336 (Sigma Chemical, St. Louis, MO). Total protein in the lipoprotein fractions was determined by a modified Lowry method (Peterson 1977). Plasma apolipoprotein levels were measured by rocket immunoelectrophoresis, following the method of Laurell (1966) and Dory and Roheim (1981). Antiserum was obtained from Paul Roheim (Louisiana State University, New Orleans, LA). Gradient gel electrophoresis was used to examine the distribution of particle size in the HDL fraction. Liver lipids were extracted by the procedure of Folch et al. (1957) and the extract analyzed for choles terol and triacylglycerides. An aliquot of each liver was also used for determination of fatty acid synthetase (FAS, EC 2.3.1.85) activity. The activity was determined in the 100,000 x g supernatant using [114C]acetyl CoA as substrate (Hsu et al. 1969). The protein concentration of the supernatant was deter
ET AL.
mined by the modified Lowry method (Peterson 1977). One unit of enzyme activity is defined as that required to incorporate 1 umol of [l-14C]acetyl CoA into fatty acid per min at 37°C. The data are expressed as means and pooled SE. Data were analyzed by ANOVA; if the F value was significant, differences between means were deter mined by least significant difference using a proba bility value of 5% as statistically significant.
RESULTS The weights of body, liver and intestinal mucosa of rats at the end of the experiment are shown in Table 2. The initial average body weight of the rats was 133.5 ±2.4 g. The final body weights differed signifi cantly among dietary groups, although the differences were not large. Liver weight, expressed per 100 g of body weight, was significantly lower in the group fed the WHY diet than in the SMLF-fed group, but no significant differences were found among other di etary groups. The relative weight of the intestinal mucosa was significantly higher in the WM-fed group than in the SMFF- and WHY-fed groups. Plasma triacylglyceride, cholesterol and apolipo protein concentrations and hepatic lipids are shown in Table 3. Plasma triacylglycerides concentrations in the WM-fed group were 53% of those in the groups fed the CAS, SMFF and SMLF diets and tended to be lower than those in the WHY-fed group, but this difference was not significant (P > 0.05). Plasma cho lesterol concentrations were similar among all groups. The WM-fed rats had lower plasma apolipoprotein B levels than did the rats fed the SMLF, SMFF or WHY diets. Apolipoprotein A-I levels did not differ among groups. Apolipoprotein A-IV was lowest in rats fed the WM diet, although the difference was signif icant only when compared with the CAS-fed group. Apolipoprotein E concentration was highest in the WM-fed group and differed significantly from the SMFF- and WHY-fed groups.
TABLE 2 Body, liver and small intestinal mucosa weights of rats fed diets containing whole milk, skim milk or milk protein fractions1 Diet milk-full milk-low fat348.6b10.8ab3.10b0.83abSkim fat363.8bc11.0ab3.01 gLiver,weight, Body gLiver, wtIntestinal g/100 gbody mucosa, gì 100 g body wtSkim
4b3.01 ab0.98abWhey350.7b10.1a2.92a0.79aWholemilk319.5a9.3a3.07ab1.07bPooledSE8 ab0.76aCasein378.5C11.
'Values are means for eight rats. Values in a row with different superscript Downloaded from https://academic.oup.com/jn/article-abstract/122/9/1840/4769454 by University of Glasgow user on 04 April 2018
letters are significantly
different (P < 0.05).
EFFECT OF MILK ON TRIACYLGLYCERIDES
1843
TABLE 3 Plasma and liver triacylglyceride,
cholesterol
and apolipoprotein (apo) concentrations skim milk or milk protein fractions^
in rats fed diets containing
whole milk,
Diet milk-
milk-
fat1.681.15C24 full low fat1.751.15C239C157ab539950ab5.57bc13.0ab13.53abSkim
milk1.800.60a117a106a6131232b3.63a2.1a9.38aPooledSE0.0
PlasmaTotal cholesterol,Triacylglycerides,Apo mg/LApo B, mg/LApo A-IV, mg/LApo A-I, mg/LLiverTotal E,
7C242ab536711a6.12e18.9b13.27abCasein1.810.90b151ab254b5871012ab4.32ab5.2a16.06bcWhey1.620.79ab188b180ab5567
wet cholesterol,Percent wtcholesterol, esterifiedTriacylglycerides,mmol/Lmmol/Liunol/g %\>jnol/g wet wtSkim 'Values are means for eight rats. Values in a row with different superscript
No significant differences in apo B mRNA abun dance were found either in liver or in intestinal mucosa among dietary groups. Thus, in the liver the mRNA ratios of apo B to actin were 1.63, 1.61, 1.41, 1.67 and 1.39 (pooled SEM, 0.018) for the groups fed the SMLF, SMFF, CAS, WHY and WM diets, respec tively. Likewise, the values determined in intestinal mucosa were not significantly different among the groups: 1.42, 1.35, 1.80, 1.49 and 1.52 (pooled SEM, 0.27) for the groups fed the SMLF, SMFF, CAS, WHY and WM diets, respectively. Liver cholesterol concentration was significantly lower in the WM-fed group than in the groups fed the WHY, SMFF and SMLF diets (Table 3). The per centage of cholesterol esterified was significantly lower in the groups fed the CAS and WM diets than in those fed the WHY and SMFF diets. Liver triacylglycerides concentrations in the WM-fed group were significantly lower than those in the CAS- and WHYfed groups and tended to be lower than in rats fed the two skim milk diets. The composition of the lipoprotein fractions is pre sented in Table 4. The VLDL triacylglycerides con centrations were lowest in the groups fed the WM and WHY diets, as were the cholesterol and protein con centrations. Hence, the overall VLDL fraction was reduced in the groups consuming WM and WHY diets. In the LDL fraction, triacylglycerides were sig nificantly higher in the WM-fed group than in the groups fed the SMFF, SMLF and WHY diets, and the triacylglycerides were significantly lower in the SMFF-fed group than in the groups fed the SMLF, CAS and WM diets. The percentage of cholesterol that was esterified in the plasma and lipoprotein frac tions did not differ among groups. Table 5 shows the mean diameter of the HDL fraction and the percent distribution of the HDL subDownloaded from https://academic.oup.com/jn/article-abstract/122/9/1840/4769454 by University of Glasgow user on 04 April 2018
letters are significantly
different (P < 0.05|.
fraction that was within the range of 12.5-17.9 nm in diameter. The values reported indicated significantly larger mean peak diameter of HDL particles in the WM-fed group than in the others. Furthermore, a greater proportion of particles corresponding to the HDLi size range was found in the WM-fed group compared with the rest of the dietary treatments. The values obtained for FAS activity in liver, ex pressed as mU/g protein in the supernatant, were 3.14, 1.50, 1.76, 1.50 and 1.29 (pooled SEM, 0.17) in rats fed the SMLF, SMFF, CAS, WHY and WM diets, respectively. The values for activity were generally low, probably due to the fasting state of the rats, which may have resulted in a marked decrease in the activity of lipogenic enzymes. The FAS activity was significantly higher in the SMLF-fed group than in the others, undoubtedly due to the relatively lower fat and higher carbohydrate content of this diet. Interest ingly, in a statistical comparison between the CAS and WM dietary treatments only, FAS activity was significantly lower in the WM-fed group despite the fact that both diets were equally high in fat content (22%).
DISCUSSION The results of this study demonstrate that a diet containing whole milk affects plasma and hepatic lipids in rats. The significant reduction of plasma triacylglycerides in the WM- and WHY-fed groups was related to a significant reduction in the VLDL fraction. These results support the observations made in previous experiments by Schneeman et al. (1989) that whole milk lowers triacylglyceride and VLDL
"\ MARQUEZ-RUIZ
1844
ET AL.
TABLE 4
Triacylglycerìde,cholesterol and protein concentrations of plasma ¡ipoproteinfractions in rats fed diets containing whole milk, skim milk or milk protein fractions1 Diet Skim milklow fat VLDL 405C Triacylglycerides, \unol/L 129C Cholesterol, ¡imol/L Percent cholesterol esters, % 48.8 127b Protein, mg/L LDL 35b Triacylglycerides, \imol/L Cholesterol, \unol/L 152 58.5ab Percent cholesterol esters, % Protein, mg/L 73 HDL Triacylglycerides, \imol/L 68 Cholesterol, \anollL 1013 80.6ab Percent cholesterol esters, % 863 Protein, mg/L 'Values are means for eight rats. Values in a row
Skim milkfull fat
Casein
Whey
Whole milk
Pooled
151a 54a
40 16 3.2 8.6
398C 144C
307bc 110bc
229ab 81ab
51.2 127b
53.0 123b
50.7 96a
49.9 85a
18a 141 61.2b
49bc
34ab
182 57.8a
171 59.2ab
57e 166 57.9a
66 56 903 83.7b 794
89 74 1010 81.4ab 827
89 66 917 78.8a 804
84 74 1017 81.3ab 888
SE
5.7 17 1.1 83 7.9 50 1.4 41
with different superscript letters are significantly different (P < 0.05).
concentrations compared with a diet containing casein. Additionally, plasma apo B, the primary apolipoprotein associated with triacylglyceride-rich par ticles, was lower in rats fed a diet that contained whole milk. However, apo B mRNA abundance was not affected by the dietary treatments, indicating that with these dietary manipulations, pretranslational regulation of apo B synthesis does not occur despite the differences in plasma apo B levels. Other investi gators have shown that synthetic rates of apo B do not correlate strongly with secretion of triglyceride-rich lipoprotein (Renner et al. 1986). In contrast to the WHY-fed group, the tendency for lower plasma triacylglycerides in the WM-fed group cannot be explained by an accumulation of lipid in the liver, because hepatic lipid levels were lowest in the WM-fed group. The overall reduction of the VLDL fraction in rats fed the diets containing whole milk may be accounted for by a more rapid turnover of VLDL. Additionally, the hypotriglyceridemic effect of whole milk may be determined, at least in part, by a reduction in hepatic FAS activity. Unfortunately, FAS activities were very low in the current experiment; it has been shown that fasting and fat feeding markedly decrease de novo synthesis of fatty acids from acetate or acetyl-CoA in rat liver (Goodridge 1987, Schillabeer et al. 1990). However, even under these cir cumstances, FAS activity tended to be lower in the WM-fed group than in the CAS-fed group. These ob servations are in agreement with the results reported by Kritchevsky et al. (1983) showing that FAS activity was lowest in rat pups consuming fluid milk com Downloaded from https://academic.oup.com/jn/article-abstract/122/9/1840/4769454 by University of Glasgow user on 04 April 2018
pared with water regardless of the maternal drinking fluid. In a previous study, they had also noted that FAS activity was similar in rats given whole milk and skim milk as their drinking fluid and considerably lower than in a control group given water (Kritch evsky et al. 1979). These results suggest that further research is needed on factors in whole milk that may modify fatty acid synthesis. The mean diameter of the HDL fraction was higher in the WM-fed group than in rats receiving the other dietary treatments. This shift in particle size in rats may reflect an increase in reverse cholesterol
TABLE 5
Plasma HDL size and subfraction distribution in rats fed diets containing whole milk, skim milk or milk protein fractions1
DietSkim
of HDL with diameter mean 12.5-17.0 nm%40a36a38a37a peak diameternm12.0a12.0a12.0a12.0a
fatSkimmilk-low fatCaseinWhey milk-full 12.6b0.1Percentage 48b1.8 Whole milkPooled SEHDL 'Values are means for eight rats. Values in a column with different superscript letters are significantly different (P < 0.05).
EFFECT OF MILK ON TRIACYLGLYCERIDES
transport. Additionally, the liver cholesterol and per centage of cholesterol esterified were lowest in the WM-fed group. The effect of whole milk on plasma and hepatic lipids is not related simply to the level and type of fat in the diet, because rats fed the SMFF, CAS and WHY diets received butter oil and corn oil in proportions that provided lipid levels and composition compa rable to those in the WM diet. Additionally, despite the fact that rats fed the WM diet ingested over threefold more fat (of which -62% was saturated fatty acids) than the group fed the low fat control diet (SMLF diet), the plasma triacylglyceride concentra tions observed in the whole milk-fed group were half of those found in animals fed the low fat skim milk diet. These results confirm the view that some com ponent of whole milk other than fat content influ ences the lipemic response. The whey proteins of milk have been found to lower (Joshida et al. 1988, Norton et al. 1987), to not affect (Stahelin and Ritzel 1979) or to elevate serum cholesterol (Lovati et al. 1990, Sautier et al. 1983, Vrecko et al. 1988). In the present study, plasma cholesterol concentration did not respond to the type of dietary protein, because we found similar concen trations for the CAS- and WHY-fed groups, and these concentrations did not differ from those found in the milk-based dietary groups. The effect of the WM diet on plasma and liver triacylglycerides was not matched by any of the diets containing milk protein fractions; thus, it is unlikely that a single protein component is responsible for the effect. The WM diet differed from the other dietary treatments in the higher concentration of fat globule membrane material in the whole milk; thus the triacylglyceride-lowering effect of whole milk may be associated with the membrane material. Based on the estimation that the milk fat globule membrane (MFGM) is present at a concentration of -900 mg/100 g fat globules (Walstra 1985), the MFGM content of the WM diet fed in the current experiment was 180 mg/100 g diet. In contrast, the SMFF diet contained similar protein and carbohydrate components as whole milk but has a lower MFGM content, because the fat was provided by adding clarified butter and corn oil. Several reports have speculated that the presence of MFGM might account for the hypocholesterolemic effect of milk (Antila et al. 1980, Howard and Maries 1979), but the effect on triacyl glyceride metabolism has not been investigated. The results of the present study suggest that whole milk affects aspects of lipid metabolism in a complex fashion and that changes in triacylglyceride concen tration and triacylglyceride-rich particle composition provide important information regarding the lipemic response in rats to whole milk consumption. More investigation is needed into the potential effects of dairy products on lipogenic enzymes and clearance of triacylglyceride-rich lipoprotein particles. Downloaded from https://academic.oup.com/jn/article-abstract/122/9/1840/4769454 by University of Glasgow user on 04 April 2018
1845
ACKNOWLEDGMENTS We appreciate the assistance of Suzette Middleton in conducting the mRNA analysis and the valuable discussions with Bruce German.
LITERATURECITED Ahmed, A. A., McCarthy, R. D. &. Porter, G. A. (1979) Effect of milk constituents on hepatic cholesterolgenesis. Atherosclerosis 32: 347-357. Allain, C. C., Poon, L. S., Chan, C.S.G., Richmond, W. & Pu, P. C. (1974) Enzymatic determination of total serum cholesterol. Clin. Chem. 20: 470-475. Antila, M., Ali-Yrkko, S., Antila, V., Antila, P., Ronnemaa, T., Jarvelainen, H. &. Viikari, J. (1980) Is fat globule membrane essential for cholesterol-lowering effect of milk? Lancet 15: 602. Chirigwin, J. M., Przybyla, A. E., MacDonald, R. J. &. Rutter, W. J. (1979) Isolation of biologically active ribonucleic acid from sources enriched in ribonuclease. Biochemistry 18: 5294-5299. Dory, L. & Roheim, P. S. (1981) Rat plasma lipoproteins and apolipoproteins in experimental hypothyroidism. J. Lipid Res. 22: 287-296. Folch, I., Lees, M. & Sloane Stanley, G. H. (1957) A simple method for the isolation and purification of total lipides from animal tissues. J. Biol. Chem. 226: 497-509. Gilliland, S. E. (1989) Acidophilus milk products: a review of potential benefits to consumers. J. Dairy Sci. 72: 2483-2494. Goodridge, A. G. (1987) Dietary regulation of gene expression: enzymes involved in carbohydrate and lipid metabolism. Annu. Rev. Nutr. 7: 157-185. Helmes, P. (1977) Hypocholesterolemic effect of milk. Lancet 2: 556 (letter). Hepner, G., Fried, R., St Jeor, S., Fusetti, L. & Morin, R. (1979) Hypocholesterolemic effect of yogurt and milk. Am. J. Clin. Nutr. 32: 19-24. Howard, A. N. & Marks, J. (1979)Effect of milk products on serumcholesterol. Lancet 3: 957. Hsu, R. Y., Butterworth, P.H.W. & Porter, }. W. (1969) Pigeon liver fatty acid synthase. Methods Enzymol. 14: 33-39. Jenkins, DJ.A., Wolever, T.M.S., Spiller, G., Buckley, G., Lam, Y., Jenkins, A. L. & Josse, R. G. (1989) Hypocholesterolemic effect of vegetable protein in a hypocaloric diet. Atherosclerosis 78: 99-107. Joshida, K., Masatoshi, Y. & Ahiko, K. (1988) Effects of addition of arginine, cystine, and glycine to the bovine milk-simulated amino acid mixture on the level of plasma and liver cholesterol in rats. J. Nutr. Sci. Vitaminol. 34: 567-576. Keim, N. L., Marie«, J. A. & Amundson, C. H. (1981) The cholesterolemic effect of skim milk in young men consuming controlled diets. Nutr. Res. 1: 429-442. Kiyosawa, H., Sugawara, N. C., Sugawara, N. & Yiyake, H. (1984) Effect of skim milk and yogurt on serum lipids and development of sudanophilic lesions in cholesterol-fed rabbits. Am. J. Clin. Nutr. 40: 479-484. Kritchevsky, D., Tepper, S. A., Morrisey, R. B., Czamecki, S. K. & Klurfeld, D. M. (1979) Influence of whole or skim milk on cholesterol metabolism in rats. Am. J. Clin. Nutr. 32: 597-600. Kritchevsky, D., Tepper, S. A., Mueller, M. A., Czamecki, S. K. & Klurfeld, D. M. (1983) Pre- and post-weaning milk consumption effects on lipid metabolism in rats. Atherosclerosis 47: 241-250. Laureti, C. B. (1966)Quantitative estimation of proteins by electrophoresis in agarose gel containing antibodies. Anal. Biochem. 15: 45-52. Lovati, M. R., West, C. E., Sirtori, C. R. & Beynen, A. C. (1990) Dietary animal proteins and cholesterol metabolism in rabbits. Br. J. Nutr. 64: 473-485.
1846
MARQUEZ-RUIZ
Maniatis, T., Fritsch, E. F. & Sambrook, J. (1982) Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. Mann, G. V. (1977)A factor in yogurt which lowers cholesterolemia in man. Atherosclerosis 26: 335-340. Ney, D. M. (1991| Potential for enhancing the nutritional properties of milk fat. J. Dairy Sci. 74: 4002-4012. Ney, D. M., Lefevre, M. & Schneeman, B. O. (1986) Alteration in lipoprotein composition with intravenous compared to intragastric fat-free feeding in the rat. J. Nutr. 116: 2106-2120. Norton, S. A., Béâmes, C. G., Maxwell, C. V. & Morgan, G. L. (1987) Effect of dietary whey upon the serum cholesterol of the pig. Nutr. Rep. Int. 36: 273-279. Peterson, G. R. (1977) Simplification of protein assay method of Lowry et al. which is more generally applicable. Anal. Biochem. 83: 346-356. Renner, F., Samuelson, A., Rogers, M. & Glickman, R. M. (1986) Effect of saturated and unsaturated lipid on the composition of mesenteric triglyceride-rich lipoprotein in rats. J. Lipid Res. 27: 72-81. Richardson, T. (1978) The hypocholesterolemic effect of milk—a review. J. Food Prot. 41: 226-235. Sautier, C, Dieng, K., Flament, C., Doucet, C., Suquet, J. P. & Lemonnier, D. (1983) Effects of whey protein, casein, soya-bean and sunflower proteins on the serum, tissue and faecal steroids
Downloaded from https://academic.oup.com/jn/article-abstract/122/9/1840/4769454 by University of Glasgow user on 04 April 2018
ET AL. in rats. Br. J. Nutr. 49: 313-319. Schillabeer, G., Hornford, J., Förden,f. M., Wong, N.C.W. & Lau, D.C.W. (1990) Hepatic and adipose tissue lipogenic enzyme mRNA levels are suppressed by high fat diets in the rat. J. Lipid Res. 31: 623-631. Schneeman, B. O. & Gallaher, D. (1980)Changes in small intestinal digestive enzyme activity and bile acids with dietary cellulose. J. Nutr. 110: 584-590. Schneeman, B. O., Rice, R. &. Richter, B. D. (1989) Reduction of plasma and hepatic triacylglycerides with whole milkcontaining diets in rats. J. Nutr. 119: 965-970. Stahelin, H. B. & Ritzel, G. (1979) Effect of whey on plasma lipids. Int. J. Vitamin. Nutr. Res. 49: 229-230. Thakur, C. P. & Jha, A. N. (1981) Influence of milk, yoghurt and calcium on cholesterol-induced atherosclerosis in rabbits. Atherosclerosis 39: 211-215. Thomson, L. U., Jenkins, D.J.A., Amer, M.A.V., Reichter, R., Jenkins, A. & Kamulsky, ]. (1982) The effect of fermented and unfermented milks on serum cholesterol. Am. f. Clin. Nutr. 36: 1106-1111. Vrecko, K., Mlekusch, W. & Aloia, R. C. (1988) Effect of skim milk and whey diets on plasma lipid levels of rabbits in a cross-over study. Atherosclerosis 72: 11-17. Walstra, P. (1985) Some comments on the isolation of fat globule membrane material. J. Dairy Res. 52: 309-312.
Modification of Triacylglycerides and Apolipoprotein in Rats Fed Diets Containing Whole Milk, Skim Milk and Milk Proteins1 GLORIA MARQÃœEZ-RÃœIZ,2B. DIANE Department
of Nutrition,
university
RICHTER AND BARBARA
skim milk and some milk products can exert a hypocholesterolemic effect on humans and experi mental animals (Richardson 1978). Milk constituents such as orotic acid (Richardson 1978), lipoproteins (Ahmed et al. 1979), lactose (Helmes 1977), calcium ¡Keimet al. 1981, Thakur and Jha 1981), 3-hydroxy3-methylglutaric acid (Mann 1977) and a component of the fat globule membrane (Antila et al. 1980, Howard and Marks 1979) have been suggested as hypocholesterolemic factors. Furthermore, products of bacterial fermentation of milk have been reported as hypocholesterolemic (Gilliland 1989, Hepner et al. 1979). Overall, the evidence for the existence of a hypocholesterolemic milk factor is still not definitive, but among all milk products tested, skim milk seems to cause the greatest reduction of serum cholesterol (Ney 1991, Thomson et al. 1982). Most of the studies that have been conducted have focused on the effect of milk products on plasma cholesterol, whereas the influence on plasma and hepatic triacylglyceride concentrations has not been clearly established. Some investigators have reported a plasma triacylglyceride-lowering effect by addition of dairy products to the diet (Kiyosawa et al. 1984, Kritchevsky et al. 1979) whereas others have been unable to document any response (Jenkins et al. 1989, Vrecko et al. 1988). These contrasting results can be attributed to a variety of differences in experimental design, including total fat intake, the length of feeding period and the amount and type of dairy product consumed. Schneeman et al. (1989) showed that whole milk
INDEXING KEY WORDS:
•cholesterol
•casein
O. SCHNEEMAN3
of California, Dauis, CA 95616-8669
ABSTRACT The objective of this study was to de termine the effects of diets containing milk and milk protein fractions on plasma and hepatic lipids, apolipoprotein B mRNA abundance, and plasma apolipoprotein concentrations and lipoprotein composition. Male rats were fed for 6 wk diets that contained (wt/wt) 76% whole milk (WM diet), 55% skim milk (SMFF diet), 22% casein (CAS diet), 22% whey protein isolate (WHY diet) or 55% skim milk-low fat (SMLF diet). The fat concen tration in the SMLF diet was 7%. Butter oil (20%) and corn oil (2%) were added to the SMFF, CAS and WHY diets. Plasma and VLDL triacylglycerides in the WM-fed rats were about half of the level in the groups fed the SMFF and SMLF diets, but not significantly different from those of the WHY-fed group. Hepatic triacyl glycerides generally were lower in the WM-fed group than in the other groups. Plasma cholesterol concentration did not differ among groups. Plasma apolipoprotein B was significantly lower in the WM-fed group than in rats fed the SMLF, SMFF or WHY diets. However, apolipo protein B mRNA abundance in the liver and small intes tinal mucosa did not differ due to dietary treatment. Thus the lipemic response due to whole milk is not associated with milk protein fractions and may be due to the presence of fat globule membrane in the diet con taining whole milk. J. Nutr. 122: 1840-1846, 1992.
•fatty acid synthetase
B
•rats
•triacylglycerides
Bovine milk and some of its products have been considered atherogenic (Ney 1991) because of their cholesterol content and fatty acid profile. The pre dominant fatty acids are saturated; the ratio of polyunsaturated to saturated fatty acids is -0.08. From the literature, however, the relationship between con sumption of dairy products, especially milk, and risk of cardiovascular disease is not clear. Besides im plicating the fat content of milk, investigators have reported that casein, the major milk protein, is hypercholesterolemic in animals (Lovati et al. 1990). In contrast, numerous studies have shown that milk,
Supported by the National Dairy Promotion Board, adminis tered by the National Dairy Council. 2Current address: Instituto de la Grasa y sus Derivados (C.S.I.C.), Avenida Padre Garcia Tejero, 4, Apartado 1078, E-41012 Sevilla, Spain. Gloria Marquez-Ruiz was supported by a Postdoc toral Fellowship from NATO (Spain). 3To whom correspondence should be addressed.
0022-3166/92 $3.00 ©1992 American Institute of Nutrition. Received 7 November 1991. Accepted 16 May 1992.
1840 Downloaded from https://academic.oup.com/jn/article-abstract/122/9/1840/4769454 by University of Glasgow user on 04 April 2018
EFFECT OF MILK ON TRIACYLGLYCERIDES
1841
TABLE 1 Diet composition Diet milkmilklow fat55020506218850 full fat5502020010050
milk760209050 diet2202020029011850
Skim milk powder Whole milk powder CaseinWhey 2020029011850
Corn oilButter oilLactoseSucroseCornstarchCellulose
Mineral mix Vitamin mix1CholesterolCaHP04
60 200.50.214.97Skim
60 60 200.218.06Caseing/kg 2022
0.218.64Whey220
hydroxytolueneEnergy, Butylated
60 2022
60200.217.98
0.218.64Whole
Mf/kgSkim
Composition
was previously published in Schneeman and Gallaher (1980).
powder decreased triacylglyceride levels and the VLDL fraction in rats and that this response was not simply predicted by dietary fat content. Overall, the results suggested that milk is more likely to alter triacylglyceride than cholesterol metabolism in rats. Furthermore, the fraction of whole milk responsible for these effects apparently resides in the portion that contains milk protein or the fat globule membrane, because addition of butter oil alone did not lower either plasma or VLDL triacylglycerides. To investigate whether a protein fraction of whole milk is responsible for the changes in triacylglyceride metabolism, we conducted a controlled study on the effects of milk-based diets and diets containing iso lated milk protein fractions on plasma triacyl glycerides, cholesterol and apolipoprotein level, lipoprotein composition and hepatic lipids in rats.
METHODS Forty male Wistar rats (Hilltop Labs, Scottsdale, PA) were randomly assigned to five groups and fed diets containing whole milk (WM diet),4 skim milk plus full fat (SMFF diet), casein (CAS diet), whey protein isolate (WHY diet) or skim milk with low fat (SMLF diet). The experimental protocol was approved by the Animal Welfare Committee at the University of California, Davis. The compositions of the experi mental diets are shown in Table 1. The SMFF, CAS and WHY diets were adjusted (by the addition of com Downloaded from https://academic.oup.com/jn/article-abstract/122/9/1840/4769454 by University of Glasgow user on 04 April 2018
oil and butter oil) to contain the same fat level and composition as those found in the WM diet. The SMFF, CAS, SMLF and WHY diets were similar in total protein content to the WM diet. Although the casein and whey vary slightly in amino acid content from whole milk, skim milk has the same amino acid pattern. All the diets were balanced for calcium content because calcium may possess a hypocholesterolemic action (Keim et al. 1981, Thakur and Jha 1981). Animals were housed in temperature(21-23°C) and light-controlled rooms with a 12-h light:dark cycle beginning at 0800 h and at ambient humidity (30-60%). After 6 wk of ad libitum feeding, the animals were deprived of food for 16 h. Each rat was anesthetized with a "cocktail" composed of 100 g/L ketamine hydrochloride (0.15 mL), 20 g/L xylazine (0.075 mL) and 10 g/L acepromazine maléate (0.022 mL). Blood was collected from the thoracic aorta or by cardiac puncture into syringes containing a final con centration of 1 g EDTA/L. The intestinal mucosa was removed by scraping the intestine with a glass slide, then weighed and frozen in liquid nitrogen. The liver
Abbreviations used: apo, apolipoprotein; FAS, fatty acid synthetase (EC 2.3.1.85); MFGM, milk fat globular membrane. Diet abbreviations: CAS, diet containing casein; SMFF, diet containing skim milk plus full fat; SMLF, diet containing skim milk with low fat; WHY, diet containing whey protein isolate,- WM, diet con taining whole milk.
1842
MARQUEZ-RUIZ
was removed, freeze-clamped, frozen in liquid ni trogen and stored at -70°C. RNA was isolated from liver and intestinal mucosa samples. Tissue samples were homogenized in a guanidine isothiocyanate buffer, and total RNA was isolated by centrifugation on a cesium chloride-EDTA cushion (Chirigwin et al. 1979). Total RNA was esti mated by absorbance at 260 nm. The relative abun dance of mRNA coding for apolipoprotein (apo) B and actin was determined by slot-blot hybridization with specific 32P-labeled cDNA probes. Conditions for hy bridization and cDNA radiolabeling were those described by Maniatis et al. (1982). Autoradiographs of hybridizations were measured by densitometry. Relative abundance of apo B mRNA and of actin mRNA in each sample were expressed per microgram of total RNA and the ratio of apo B to actin mRNA abundance calculated. Plasma VLDL, LDL and HDL fractions were iso lated by sequential ultracentrifugation, as previously described (Ney et al. 1986). The plasma and lipoprotein fractions were analyzed for cholesterol by the method of Allain et al. (1974) and for triacylglycerides using Sigma Procedure no. 336 (Sigma Chemical, St. Louis, MO). Total protein in the lipoprotein fractions was determined by a modified Lowry method (Peterson 1977). Plasma apolipoprotein levels were measured by rocket immunoelectrophoresis, following the method of Laurell (1966) and Dory and Roheim (1981). Antiserum was obtained from Paul Roheim (Louisiana State University, New Orleans, LA). Gradient gel electrophoresis was used to examine the distribution of particle size in the HDL fraction. Liver lipids were extracted by the procedure of Folch et al. (1957) and the extract analyzed for choles terol and triacylglycerides. An aliquot of each liver was also used for determination of fatty acid synthetase (FAS, EC 2.3.1.85) activity. The activity was determined in the 100,000 x g supernatant using [114C]acetyl CoA as substrate (Hsu et al. 1969). The protein concentration of the supernatant was deter
ET AL.
mined by the modified Lowry method (Peterson 1977). One unit of enzyme activity is defined as that required to incorporate 1 umol of [l-14C]acetyl CoA into fatty acid per min at 37°C. The data are expressed as means and pooled SE. Data were analyzed by ANOVA; if the F value was significant, differences between means were deter mined by least significant difference using a proba bility value of 5% as statistically significant.
RESULTS The weights of body, liver and intestinal mucosa of rats at the end of the experiment are shown in Table 2. The initial average body weight of the rats was 133.5 ±2.4 g. The final body weights differed signifi cantly among dietary groups, although the differences were not large. Liver weight, expressed per 100 g of body weight, was significantly lower in the group fed the WHY diet than in the SMLF-fed group, but no significant differences were found among other di etary groups. The relative weight of the intestinal mucosa was significantly higher in the WM-fed group than in the SMFF- and WHY-fed groups. Plasma triacylglyceride, cholesterol and apolipo protein concentrations and hepatic lipids are shown in Table 3. Plasma triacylglycerides concentrations in the WM-fed group were 53% of those in the groups fed the CAS, SMFF and SMLF diets and tended to be lower than those in the WHY-fed group, but this difference was not significant (P > 0.05). Plasma cho lesterol concentrations were similar among all groups. The WM-fed rats had lower plasma apolipoprotein B levels than did the rats fed the SMLF, SMFF or WHY diets. Apolipoprotein A-I levels did not differ among groups. Apolipoprotein A-IV was lowest in rats fed the WM diet, although the difference was signif icant only when compared with the CAS-fed group. Apolipoprotein E concentration was highest in the WM-fed group and differed significantly from the SMFF- and WHY-fed groups.
TABLE 2 Body, liver and small intestinal mucosa weights of rats fed diets containing whole milk, skim milk or milk protein fractions1 Diet milk-full milk-low fat348.6b10.8ab3.10b0.83abSkim fat363.8bc11.0ab3.01 gLiver,weight, Body gLiver, wtIntestinal g/100 gbody mucosa, gì 100 g body wtSkim
4b3.01 ab0.98abWhey350.7b10.1a2.92a0.79aWholemilk319.5a9.3a3.07ab1.07bPooledSE8 ab0.76aCasein378.5C11.
'Values are means for eight rats. Values in a row with different superscript Downloaded from https://academic.oup.com/jn/article-abstract/122/9/1840/4769454 by University of Glasgow user on 04 April 2018
letters are significantly
different (P < 0.05).
EFFECT OF MILK ON TRIACYLGLYCERIDES
1843
TABLE 3 Plasma and liver triacylglyceride,
cholesterol
and apolipoprotein (apo) concentrations skim milk or milk protein fractions^
in rats fed diets containing
whole milk,
Diet milk-
milk-
fat1.681.15C24 full low fat1.751.15C239C157ab539950ab5.57bc13.0ab13.53abSkim
milk1.800.60a117a106a6131232b3.63a2.1a9.38aPooledSE0.0
PlasmaTotal cholesterol,Triacylglycerides,Apo mg/LApo B, mg/LApo A-IV, mg/LApo A-I, mg/LLiverTotal E,
7C242ab536711a6.12e18.9b13.27abCasein1.810.90b151ab254b5871012ab4.32ab5.2a16.06bcWhey1.620.79ab188b180ab5567
wet cholesterol,Percent wtcholesterol, esterifiedTriacylglycerides,mmol/Lmmol/Liunol/g %\>jnol/g wet wtSkim 'Values are means for eight rats. Values in a row with different superscript
No significant differences in apo B mRNA abun dance were found either in liver or in intestinal mucosa among dietary groups. Thus, in the liver the mRNA ratios of apo B to actin were 1.63, 1.61, 1.41, 1.67 and 1.39 (pooled SEM, 0.018) for the groups fed the SMLF, SMFF, CAS, WHY and WM diets, respec tively. Likewise, the values determined in intestinal mucosa were not significantly different among the groups: 1.42, 1.35, 1.80, 1.49 and 1.52 (pooled SEM, 0.27) for the groups fed the SMLF, SMFF, CAS, WHY and WM diets, respectively. Liver cholesterol concentration was significantly lower in the WM-fed group than in the groups fed the WHY, SMFF and SMLF diets (Table 3). The per centage of cholesterol esterified was significantly lower in the groups fed the CAS and WM diets than in those fed the WHY and SMFF diets. Liver triacylglycerides concentrations in the WM-fed group were significantly lower than those in the CAS- and WHYfed groups and tended to be lower than in rats fed the two skim milk diets. The composition of the lipoprotein fractions is pre sented in Table 4. The VLDL triacylglycerides con centrations were lowest in the groups fed the WM and WHY diets, as were the cholesterol and protein con centrations. Hence, the overall VLDL fraction was reduced in the groups consuming WM and WHY diets. In the LDL fraction, triacylglycerides were sig nificantly higher in the WM-fed group than in the groups fed the SMFF, SMLF and WHY diets, and the triacylglycerides were significantly lower in the SMFF-fed group than in the groups fed the SMLF, CAS and WM diets. The percentage of cholesterol that was esterified in the plasma and lipoprotein frac tions did not differ among groups. Table 5 shows the mean diameter of the HDL fraction and the percent distribution of the HDL subDownloaded from https://academic.oup.com/jn/article-abstract/122/9/1840/4769454 by University of Glasgow user on 04 April 2018
letters are significantly
different (P < 0.05|.
fraction that was within the range of 12.5-17.9 nm in diameter. The values reported indicated significantly larger mean peak diameter of HDL particles in the WM-fed group than in the others. Furthermore, a greater proportion of particles corresponding to the HDLi size range was found in the WM-fed group compared with the rest of the dietary treatments. The values obtained for FAS activity in liver, ex pressed as mU/g protein in the supernatant, were 3.14, 1.50, 1.76, 1.50 and 1.29 (pooled SEM, 0.17) in rats fed the SMLF, SMFF, CAS, WHY and WM diets, respectively. The values for activity were generally low, probably due to the fasting state of the rats, which may have resulted in a marked decrease in the activity of lipogenic enzymes. The FAS activity was significantly higher in the SMLF-fed group than in the others, undoubtedly due to the relatively lower fat and higher carbohydrate content of this diet. Interest ingly, in a statistical comparison between the CAS and WM dietary treatments only, FAS activity was significantly lower in the WM-fed group despite the fact that both diets were equally high in fat content (22%).
DISCUSSION The results of this study demonstrate that a diet containing whole milk affects plasma and hepatic lipids in rats. The significant reduction of plasma triacylglycerides in the WM- and WHY-fed groups was related to a significant reduction in the VLDL fraction. These results support the observations made in previous experiments by Schneeman et al. (1989) that whole milk lowers triacylglyceride and VLDL
"\ MARQUEZ-RUIZ
1844
ET AL.
TABLE 4
Triacylglycerìde,cholesterol and protein concentrations of plasma ¡ipoproteinfractions in rats fed diets containing whole milk, skim milk or milk protein fractions1 Diet Skim milklow fat VLDL 405C Triacylglycerides, \unol/L 129C Cholesterol, ¡imol/L Percent cholesterol esters, % 48.8 127b Protein, mg/L LDL 35b Triacylglycerides, \imol/L Cholesterol, \unol/L 152 58.5ab Percent cholesterol esters, % Protein, mg/L 73 HDL Triacylglycerides, \imol/L 68 Cholesterol, \anollL 1013 80.6ab Percent cholesterol esters, % 863 Protein, mg/L 'Values are means for eight rats. Values in a row
Skim milkfull fat
Casein
Whey
Whole milk
Pooled
151a 54a
40 16 3.2 8.6
398C 144C
307bc 110bc
229ab 81ab
51.2 127b
53.0 123b
50.7 96a
49.9 85a
18a 141 61.2b
49bc
34ab
182 57.8a
171 59.2ab
57e 166 57.9a
66 56 903 83.7b 794
89 74 1010 81.4ab 827
89 66 917 78.8a 804
84 74 1017 81.3ab 888
SE
5.7 17 1.1 83 7.9 50 1.4 41
with different superscript letters are significantly different (P < 0.05).
concentrations compared with a diet containing casein. Additionally, plasma apo B, the primary apolipoprotein associated with triacylglyceride-rich par ticles, was lower in rats fed a diet that contained whole milk. However, apo B mRNA abundance was not affected by the dietary treatments, indicating that with these dietary manipulations, pretranslational regulation of apo B synthesis does not occur despite the differences in plasma apo B levels. Other investi gators have shown that synthetic rates of apo B do not correlate strongly with secretion of triglyceride-rich lipoprotein (Renner et al. 1986). In contrast to the WHY-fed group, the tendency for lower plasma triacylglycerides in the WM-fed group cannot be explained by an accumulation of lipid in the liver, because hepatic lipid levels were lowest in the WM-fed group. The overall reduction of the VLDL fraction in rats fed the diets containing whole milk may be accounted for by a more rapid turnover of VLDL. Additionally, the hypotriglyceridemic effect of whole milk may be determined, at least in part, by a reduction in hepatic FAS activity. Unfortunately, FAS activities were very low in the current experiment; it has been shown that fasting and fat feeding markedly decrease de novo synthesis of fatty acids from acetate or acetyl-CoA in rat liver (Goodridge 1987, Schillabeer et al. 1990). However, even under these cir cumstances, FAS activity tended to be lower in the WM-fed group than in the CAS-fed group. These ob servations are in agreement with the results reported by Kritchevsky et al. (1983) showing that FAS activity was lowest in rat pups consuming fluid milk com Downloaded from https://academic.oup.com/jn/article-abstract/122/9/1840/4769454 by University of Glasgow user on 04 April 2018
pared with water regardless of the maternal drinking fluid. In a previous study, they had also noted that FAS activity was similar in rats given whole milk and skim milk as their drinking fluid and considerably lower than in a control group given water (Kritch evsky et al. 1979). These results suggest that further research is needed on factors in whole milk that may modify fatty acid synthesis. The mean diameter of the HDL fraction was higher in the WM-fed group than in rats receiving the other dietary treatments. This shift in particle size in rats may reflect an increase in reverse cholesterol
TABLE 5
Plasma HDL size and subfraction distribution in rats fed diets containing whole milk, skim milk or milk protein fractions1
DietSkim
of HDL with diameter mean 12.5-17.0 nm%40a36a38a37a peak diameternm12.0a12.0a12.0a12.0a
fatSkimmilk-low fatCaseinWhey milk-full 12.6b0.1Percentage 48b1.8 Whole milkPooled SEHDL 'Values are means for eight rats. Values in a column with different superscript letters are significantly different (P < 0.05).
EFFECT OF MILK ON TRIACYLGLYCERIDES
transport. Additionally, the liver cholesterol and per centage of cholesterol esterified were lowest in the WM-fed group. The effect of whole milk on plasma and hepatic lipids is not related simply to the level and type of fat in the diet, because rats fed the SMFF, CAS and WHY diets received butter oil and corn oil in proportions that provided lipid levels and composition compa rable to those in the WM diet. Additionally, despite the fact that rats fed the WM diet ingested over threefold more fat (of which -62% was saturated fatty acids) than the group fed the low fat control diet (SMLF diet), the plasma triacylglyceride concentra tions observed in the whole milk-fed group were half of those found in animals fed the low fat skim milk diet. These results confirm the view that some com ponent of whole milk other than fat content influ ences the lipemic response. The whey proteins of milk have been found to lower (Joshida et al. 1988, Norton et al. 1987), to not affect (Stahelin and Ritzel 1979) or to elevate serum cholesterol (Lovati et al. 1990, Sautier et al. 1983, Vrecko et al. 1988). In the present study, plasma cholesterol concentration did not respond to the type of dietary protein, because we found similar concen trations for the CAS- and WHY-fed groups, and these concentrations did not differ from those found in the milk-based dietary groups. The effect of the WM diet on plasma and liver triacylglycerides was not matched by any of the diets containing milk protein fractions; thus, it is unlikely that a single protein component is responsible for the effect. The WM diet differed from the other dietary treatments in the higher concentration of fat globule membrane material in the whole milk; thus the triacylglyceride-lowering effect of whole milk may be associated with the membrane material. Based on the estimation that the milk fat globule membrane (MFGM) is present at a concentration of -900 mg/100 g fat globules (Walstra 1985), the MFGM content of the WM diet fed in the current experiment was 180 mg/100 g diet. In contrast, the SMFF diet contained similar protein and carbohydrate components as whole milk but has a lower MFGM content, because the fat was provided by adding clarified butter and corn oil. Several reports have speculated that the presence of MFGM might account for the hypocholesterolemic effect of milk (Antila et al. 1980, Howard and Maries 1979), but the effect on triacyl glyceride metabolism has not been investigated. The results of the present study suggest that whole milk affects aspects of lipid metabolism in a complex fashion and that changes in triacylglyceride concen tration and triacylglyceride-rich particle composition provide important information regarding the lipemic response in rats to whole milk consumption. More investigation is needed into the potential effects of dairy products on lipogenic enzymes and clearance of triacylglyceride-rich lipoprotein particles. Downloaded from https://academic.oup.com/jn/article-abstract/122/9/1840/4769454 by University of Glasgow user on 04 April 2018
1845
ACKNOWLEDGMENTS We appreciate the assistance of Suzette Middleton in conducting the mRNA analysis and the valuable discussions with Bruce German.
LITERATURECITED Ahmed, A. A., McCarthy, R. D. &. Porter, G. A. (1979) Effect of milk constituents on hepatic cholesterolgenesis. Atherosclerosis 32: 347-357. Allain, C. C., Poon, L. S., Chan, C.S.G., Richmond, W. & Pu, P. C. (1974) Enzymatic determination of total serum cholesterol. Clin. Chem. 20: 470-475. Antila, M., Ali-Yrkko, S., Antila, V., Antila, P., Ronnemaa, T., Jarvelainen, H. &. Viikari, J. (1980) Is fat globule membrane essential for cholesterol-lowering effect of milk? Lancet 15: 602. Chirigwin, J. M., Przybyla, A. E., MacDonald, R. J. &. Rutter, W. J. (1979) Isolation of biologically active ribonucleic acid from sources enriched in ribonuclease. Biochemistry 18: 5294-5299. Dory, L. & Roheim, P. S. (1981) Rat plasma lipoproteins and apolipoproteins in experimental hypothyroidism. J. Lipid Res. 22: 287-296. Folch, I., Lees, M. & Sloane Stanley, G. H. (1957) A simple method for the isolation and purification of total lipides from animal tissues. J. Biol. Chem. 226: 497-509. Gilliland, S. E. (1989) Acidophilus milk products: a review of potential benefits to consumers. J. Dairy Sci. 72: 2483-2494. Goodridge, A. G. (1987) Dietary regulation of gene expression: enzymes involved in carbohydrate and lipid metabolism. Annu. Rev. Nutr. 7: 157-185. Helmes, P. (1977) Hypocholesterolemic effect of milk. Lancet 2: 556 (letter). Hepner, G., Fried, R., St Jeor, S., Fusetti, L. & Morin, R. (1979) Hypocholesterolemic effect of yogurt and milk. Am. J. Clin. Nutr. 32: 19-24. Howard, A. N. & Marks, J. (1979)Effect of milk products on serumcholesterol. Lancet 3: 957. Hsu, R. Y., Butterworth, P.H.W. & Porter, }. W. (1969) Pigeon liver fatty acid synthase. Methods Enzymol. 14: 33-39. Jenkins, DJ.A., Wolever, T.M.S., Spiller, G., Buckley, G., Lam, Y., Jenkins, A. L. & Josse, R. G. (1989) Hypocholesterolemic effect of vegetable protein in a hypocaloric diet. Atherosclerosis 78: 99-107. Joshida, K., Masatoshi, Y. & Ahiko, K. (1988) Effects of addition of arginine, cystine, and glycine to the bovine milk-simulated amino acid mixture on the level of plasma and liver cholesterol in rats. J. Nutr. Sci. Vitaminol. 34: 567-576. Keim, N. L., Marie«, J. A. & Amundson, C. H. (1981) The cholesterolemic effect of skim milk in young men consuming controlled diets. Nutr. Res. 1: 429-442. Kiyosawa, H., Sugawara, N. C., Sugawara, N. & Yiyake, H. (1984) Effect of skim milk and yogurt on serum lipids and development of sudanophilic lesions in cholesterol-fed rabbits. Am. J. Clin. Nutr. 40: 479-484. Kritchevsky, D., Tepper, S. A., Morrisey, R. B., Czamecki, S. K. & Klurfeld, D. M. (1979) Influence of whole or skim milk on cholesterol metabolism in rats. Am. J. Clin. Nutr. 32: 597-600. Kritchevsky, D., Tepper, S. A., Mueller, M. A., Czamecki, S. K. & Klurfeld, D. M. (1983) Pre- and post-weaning milk consumption effects on lipid metabolism in rats. Atherosclerosis 47: 241-250. Laureti, C. B. (1966)Quantitative estimation of proteins by electrophoresis in agarose gel containing antibodies. Anal. Biochem. 15: 45-52. Lovati, M. R., West, C. E., Sirtori, C. R. & Beynen, A. C. (1990) Dietary animal proteins and cholesterol metabolism in rabbits. Br. J. Nutr. 64: 473-485.
1846
MARQUEZ-RUIZ
Maniatis, T., Fritsch, E. F. & Sambrook, J. (1982) Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. Mann, G. V. (1977)A factor in yogurt which lowers cholesterolemia in man. Atherosclerosis 26: 335-340. Ney, D. M. (1991| Potential for enhancing the nutritional properties of milk fat. J. Dairy Sci. 74: 4002-4012. Ney, D. M., Lefevre, M. & Schneeman, B. O. (1986) Alteration in lipoprotein composition with intravenous compared to intragastric fat-free feeding in the rat. J. Nutr. 116: 2106-2120. Norton, S. A., Béâmes, C. G., Maxwell, C. V. & Morgan, G. L. (1987) Effect of dietary whey upon the serum cholesterol of the pig. Nutr. Rep. Int. 36: 273-279. Peterson, G. R. (1977) Simplification of protein assay method of Lowry et al. which is more generally applicable. Anal. Biochem. 83: 346-356. Renner, F., Samuelson, A., Rogers, M. & Glickman, R. M. (1986) Effect of saturated and unsaturated lipid on the composition of mesenteric triglyceride-rich lipoprotein in rats. J. Lipid Res. 27: 72-81. Richardson, T. (1978) The hypocholesterolemic effect of milk—a review. J. Food Prot. 41: 226-235. Sautier, C, Dieng, K., Flament, C., Doucet, C., Suquet, J. P. & Lemonnier, D. (1983) Effects of whey protein, casein, soya-bean and sunflower proteins on the serum, tissue and faecal steroids
Downloaded from https://academic.oup.com/jn/article-abstract/122/9/1840/4769454 by University of Glasgow user on 04 April 2018
ET AL. in rats. Br. J. Nutr. 49: 313-319. Schillabeer, G., Hornford, J., Förden,f. M., Wong, N.C.W. & Lau, D.C.W. (1990) Hepatic and adipose tissue lipogenic enzyme mRNA levels are suppressed by high fat diets in the rat. J. Lipid Res. 31: 623-631. Schneeman, B. O. & Gallaher, D. (1980)Changes in small intestinal digestive enzyme activity and bile acids with dietary cellulose. J. Nutr. 110: 584-590. Schneeman, B. O., Rice, R. &. Richter, B. D. (1989) Reduction of plasma and hepatic triacylglycerides with whole milkcontaining diets in rats. J. Nutr. 119: 965-970. Stahelin, H. B. & Ritzel, G. (1979) Effect of whey on plasma lipids. Int. J. Vitamin. Nutr. Res. 49: 229-230. Thakur, C. P. & Jha, A. N. (1981) Influence of milk, yoghurt and calcium on cholesterol-induced atherosclerosis in rabbits. Atherosclerosis 39: 211-215. Thomson, L. U., Jenkins, D.J.A., Amer, M.A.V., Reichter, R., Jenkins, A. & Kamulsky, ]. (1982) The effect of fermented and unfermented milks on serum cholesterol. Am. f. Clin. Nutr. 36: 1106-1111. Vrecko, K., Mlekusch, W. & Aloia, R. C. (1988) Effect of skim milk and whey diets on plasma lipid levels of rabbits in a cross-over study. Atherosclerosis 72: 11-17. Walstra, P. (1985) Some comments on the isolation of fat globule membrane material. J. Dairy Res. 52: 309-312.
