Total body electrical conductivity (TOBEC) is a simple and non-invasive method for the assessment of body composition in vivo. Information regarding the applicability of TOBEC in the condition of abnormal fluid balance is scarce. In the present paper we give the results of the comparison between TOBEC and total body water (TBW; assessed by the tritium dilution technique) in three groups of animals: (1) healthy (n 17), (2) expanded fluid volume by secondary biliary cirrhosis (SBC; n 9) and (3) Fiirosemide®-treated rats (n 9). The TOBEC score and TBW by tritium dilution were found to be highly correlated in the pooled sample (r 0·90) and in normal (r 0.·87), SBC (r 0·73) and Furosemide-treated (r 0·89) rats. However, the relationship between TOBEC and TBW, described by least-squares regression analysis, was found to be similar for SBC and normal rats but was significantly different for Furosemide-treated and normal rats. These findings suggest that TOBEC is unable to track TBW accurately when the ratio between intracellular and extracellular water is chronically or acutely altered.

In black bears the last 6–8 weeks of gestation and the first 10–12 weeks of lactation occur in winter while the mother is in a dormant state, and reportedly does not eat, drink, urinate or defaecate. Measurements were made of the body composition and organ weights of cubs, of the composition of milk, and of milk intake (by dilution of 2H2O), in the first 3 months after birth. Additional milk samples were collected until 10 months postpartum. Bear cubs were small at birth, only 3·7 g/kg maternal weight, and chemically immature, as indicated by the high concentration of water (840 g/kg) in their bodies. Organ weights at birth were similar to those of puppies. In the first month after birth cubs gained 22 g/d or 0·23 g/g milk consumed; the milk was high in fat (220 g/kg) and low in water (670 g/kg). About 30% of the ingested energy and 51% of the ingested N were retained in the body. Over the entire 12-week period bear cubs required about 11 kg milk, containing (kg) water 7, fat 2·5, protein 0·8 and total sugar 0·25, to achieve a 2·5 kg weight gain. The birth of immature young and the production of high-fat, low-carbohydrate milk seem to be maternal adaptations to limit the utilization of glucogenic substrates during a long fast. Isotope recycling indicates that mothers may also recover most of the water (and perhaps much of the N) exported in milk by ingesting the excreta of the cubs. Lactation represents another aspect of the profound metabolic economy of the fasting bear in its winter den.

A short-term N balance study was conducted in twelve healthy female adults aged 21–32 years to determine their protein requirement. Four dietary protein levels (0·3, 0·4, 0·5 and 0·6 g protein/kg per d) were used. Energy intake of the subjects was kept constant at 0·18 MJ/kg per d. All subjects maintained their normal activity throughout the study period. N excretion was determined from the measurements of N in a total collection of urine, faeces, sweat and menstrual fluid for each dietary period. N balance during the four protein levels were −15·15 (SD 5·95), −5·53 (SD 6·71), + 6·15 (SD 4·76) and +12·05 (SD 8·63) mg N/kg per d for 0·3, 0·4, 0·5 and 0·6 g protein/kg per d respectively. The calculated average N requirements from regression analysis was 76·0 (sd 3·37) mg N/kg per d (0·48 g protein/kg per d). The estimate of allowance for individual variation to cover the 97·5 % population was 95 mg N/kg per d (0·6 g protein/kg per d). The net protein utilization (NPU) of the diet was 0·55. When compared with a similar study with men, there was a significant difference in the protein requirement between sexes. Thus, the unjustifiable sex difference in the protein allowance recommended by the Food and Agriculture Organization/World Health Organization/United Nations University (1985) Expert Consultation group must be reviewed.

Although there have been a number of studies of effects of diet and hormones on lipoprotein lipase (EC 3.1.1.34; LPL) activity and levels of LPL mRNA (Raynolds et al. 1990), there have been no studies which have investigated effects of different dietary fatty acids on LPL gene expression. In the present study male Wistar Albino rats were pair-fed diets containing 50 g fat/kg of different fatty acid composition for 2 weeks. The diets fed were (1) a mixed oil (450 g saturated fatty acids, 420 g monounsaturated fatty acids, 130 g polyunsaturated fatty acids/kg; n 8), (2) maize oil (n 8), or (3) fish oil (n 8). Animals were killed, RNA was extracted from liver and perirenal and epididymal fat pads, and analysed by ‘Northern methodology’. Samples were hybridized to a human cDNA probe for LPL (Gotoda et al. 1989). Two transcripts were identified in epididymai and perirenal adipose tissue which were approximately 3·7 and 1·7 kb in size. The results suggested that (1) fish oil-fed animals had significantly greater production of LPL mRNA in epididymai adipose tissue compared with maize oil-fed animals (P < 0·05), (2) maize oil-fed animals had significantly greater production of LPL mRNA in perirenal fat compared with the other dietary groups (P < 0·05), (3) expression in the liver was not significant. Rats fed on a fish oil diet had significantly reduced plasma triacylglycerol concentrations compared with the mixed-oil group (P < 0·05), but there were no significant differences in plasma cholesterol. The differences in LPL could not be explained directly by the changes in plasma immunoreactive-insulin and glucose-dependent insulinotrophic polypeptide levels in the three groups.

A long-term N balance study was carried out to determine the adequacy of an estimated protein requirement level recommended for young healthy Nigerian women and the effect of the menstrual cycle on the requirement. Eleven healthy young women, 25 (SD 2·6) years, were fed on a diet providing 0·6 g protein (N × 6.25)/kg per d and an average energy intake of 0·17 (SD 0·012) MJ/kg per d. Urine, faeces, sweat and menstrual fluids were collected for estimation of N balance. Menstrual N loss varied among individuals ranging from 46 to 124 mg N/d with an average of 89 (SD 21.8) mg N/d. Individual N balance was found to vary according to the day of the menstrual cycle. Positive N balances were recorded at about ovulation while negative balances were observed just before the onset of menstruation. The average N balance ranged from + 8·49 (SD 5·64) to − 430 (SD 7·84) mg N/kg per d. Nevertheless, an overall cumulative positive N balance of + 5·7 (SD 6·98) mg N/kg per d which did not change significantly with time was observed for the last 5 d of two consecutive 20 d diet periods, although three subjects were in negative N balance. Blood biochemical measurements were stable except for one subject who had elevated serum aspartate aminotransferase (EC 2.6.1.1) levels. These findings suggest that our estimate of protein requirements was sufficient to achieve N balance equilibrium in a majority (70 %) of young women. However, to satisfy 97·5 % of the population, slight adjustments might be necessary in the energy intake since subjects who were in cumulative negative N balance also lost weight.

Adult laca mice were dosed orally with 150 μl whole milk containing 2·5 μCi of either labelled stearic acid ([1-14C]18: 0; n 20) or labelled linoleic acid ([1 − 14C]18: 2; n 20). The mice were killed in groups of four at 6, 12, 24, 48 and 96 h following dosing and samples of perirenal and epididymal fat pads were taken from both inner and outer sites at each location. Significant differences in the rate of loss of label between sites were found. No differences (P = 0·018) between fat pad locations (epididymal v. perirenal) were found. A significant interaction between rate of loss of labelled fatty acid and site (P = 0·019) reflected the fact that between-site variations in this context were confined to labelled linoleic acid. In a second study weanling adult (n 10) and adult (n 10) mice were killed and their epididymal and perirenal fat pads prepared for histological examination. Both transverse and longitudinal sections were taken at the inner and outer sites of each fat pad location. Following staining, both the size and number of blood vessels were measured using computer-linked microscopy. In all instances there were significant differences between sites with the inner site consistently showing greater numbers and areas of blood vessels. In general the number of blood vessels in the inner site tended to be greater in older mice, while the reverse was seen for younger mice. The results lend support to the concept of multiple pools of triacylglycerol-fatty acids in adipose tissue such that the main determinant of shortterm supply of essential fatty acids is the quantity recently ingested.

Hepatocytes were isolated from sixteen fasting neonatal pigs and used in two experiments: (1) to determine the effect of various factors on the ability for hepatic oxidation of fatty acids and (2) to clarify the relationship between fatty acid oxidation and glucose synthesis. In Expt 1, newborn pigs were either fasted from birth for 24 h or allowed to suck ad lib. for 3 d followed by a 24 h fast. In the presence of pyruvate, oxidation of octanoate (2 mM) was about 30-fold greater than oleate (1 mM) regardless of age, but glucose synthesis was not enhanced beyond that observed for pyruvate alone. Inclusion of carnitine (1 mM), glucagon (100 nM) or dibutryl cAMP (50 μM) in the incubation media did not stimulate either fatty acid oxidation (octanoate or oleate) or glucose synthesis. Extending the period of fasting to 48 h (Expt 2) failed to enhance the fatty acid oxidative capacity or glucose synthesis rate. Likewise, the redox potential of the giuconeogenic substrate (lactate v. pyruvate) did not influence glucose synthesis regardless of the oxidative capacity exhibited for fatty acids. These data indicate that fatty acid oxidative capacity is not the first limiting factor to full expression of gluconeogenesis in hepatocytes isolated from fasted newborn pigs.

The effect of feeding grass silage or a forage–concentrate (dried grass–barley) diet ad lib. to young cattle on growth rate, plasma growth hormone (GH) and insulin-like growth factor-1 (IGF-1) concentrations was examined. The effect of including the β-adrenergic agonist cimaterol in the forage–concentrate diet was also investigated. Significantly higher growth rates were observed in animals fed on the forage–concentrate diet than in those fed on the silage diet (P < 0·001), and these were further enhanced by dietary inclusion of cimaterol (P < 0·05). Plasma GH levels were higher in the silage-fed animals (P < 0·001) but IGF-1 levels were significantly lower (P < 0·001) than in the forage–concentrate-fed animals. Cimaterol had no effect on mean plasma GH and IGF-1 concentrations. Nevertheless, plasma IGF-1 levels correlated positively with growth rate across all three treatments (r 0·84, P < 0·001, n 17). Spectral analysis of the GH data obtained for each animal was used to determine whether regular periodicities could be detected in the 24 h profile and whether they differed between the three treatment groups. Regular periodicities of four-five cycles/d were detected in the averaged GH profiles of the silage-fed and the control forage–concentrate-fed animals. In contrast, in animals given cimaterol regular GH cycles were not detectable at any of the frequencies tested. This suggests that cimaterol disrupts the rhythm of GH secretion without altering the overall mean concentrations. The data also suggest that due to an inadequate nutrient supply, the GH–IGF-1 regulatory mechanism was uncoupled in the cattle fed on silage, which may have contributed to the poor growth response of these animals.

The hypothesis was tested that dietary lactose v. glucose stimulates Mg absorption in rats because lactose lowers pH of the ileal lumen, which improves Mg solubility which in turn enhances Mg availability for transport across the ileal epithelium. For comparison, the effects of lactulose were studied because it shares with lactose the characteristic of being poorly digestible. Replacement of glucose by lactose (100 g/Kg) significantly stimulated apparent absorption of Mg. Apart from Mg absorption, lactulose also significantly enhanced absorption of Ca and phosphate. Lactose v. glucose lowered the pH of the ileal lumen from 7·5 to 7·2, whereas lactulose significantly reduced it to 7·0. In in vitro incubations a decrease in pH within the range of fluctuation in vivo was found to cause an improved solubility of Mg, and to a lesser extent also of Ca and phosphate. The smaller fall of ileal pH induced by feeding lactose instead of lactulose may explain why lactose improved Mg absorption only. For all individual rats combined there were negative relationships between ileal pH and apparent absorption of minerals, the relationship being strongest for Mg. Neither lactose nor lactulose was found to raise ileal solubility of minerals, which could relate to the possibility that the time of sampling was not appropriate. It is suggested that lactose-induced stimulation of Mg absorption in rats is caused by a lowering of ileal pH.

The ADP:O values in both cardiac and hepatic mitochondria have significantly decreased with an increase in protein level after 7, 14 and 21 d of feeding (Toyomizu et al. 1992). The present studies were undertaken to clarify tissue-specific effects of dietary protein levels on oxidative phosphorylation in the liver, kidney, skeletal muscles and small intestine and to characterize oxidative metabolism with diverse substrates in the liver. Chicks were fed on semi-purified diets of different protein levels (7, 25, 43 and 61% of metabolizable energy content) for 21 d. The responses of protein levels to oxidative phosphorylation showed tissue-dependency; although liver mitochondria of chickens fed on higher- protein diets exhibited reduced ADP:O values and state 3, neither changes in ADP:O value nor state 3 and state 4 rates were observed in the isolated mitochondria from kidney and skeletal muscles. Small intestinal mucosal mitochondria from chickens fed on a high (61%)-protein-energy diet showed significantly reduced ADP:O value and respiratory control ratio when compared with medium-protein- energy diets (25 and 43%). In liver mitochondria showing the most sensitive dependency to the levels of dietary protein, the ADP:O value decreased with increasing protein levels when pyruvate + malate- or glutamate-requiring complexes I, III and IV of the electron transport chain were used as substrates, but it did not change when succinate-requiring complexes II, III and IV or ascorbate + tetramethyl-p- phenylenediamine requiring complex IV was used. These results imply that impaired oxidative phosphorylation capacities with increasing dietary protein levels may be associated with functional damage to the respiratory chain for electron flow from NAD-linked substrates to the ubiquinone pool.

Hydration of feeds and bacterial attachment to feed particles are thought to play major roles in rumen digestion of fibrous feedstuffs. The objective of the present study was to integrate these phenomena in a mechanistic model that could be used for data analysis. The proposed model was based on the conversion of biomass, where digestion end-products can be used for the synthesis of bacterial mass. Digestion of the potentially digestible fraction and subsequent accumulation of bacterial mass was based on a sequential, three-compartment model. These compartments represented substrate undergoing hydration, digestion, and bacterial mass accumulation. A fraction of the substrate was used for synthesis of bacterial mass. It was assumed that these bacteria associate either temporarily or permanently with the remaining substrate. Dacron bags containing either dry or fully-hydrated lucerne (Medicago sativa), maize (Zea mays) cobs, orchard grass (Dactylis glomeratd), and wheat straw were incubated in the rumen of a steer that was infused continuously with (15NH4)2SO4. The 15N-enrichments of isolated particle-associated bacteria and residue remaining in the bags were used to estimate bacterial attachment. Substrate remaining and microbial mass accumulation were analysed simultaneously. Hydration did not appear to limit digestion. Fractional rate of digestion and appearance of attached bacterial mass was fastest for lucerne. For lucerne, 5 % of the digestion end-products were used for synthesis of bacteria that associated with the substrate, whereas for maize cobs, orchard grass, and wheat straw this was 16, 14, and 19% respectively. Less than 2% of digestion end-products were used for synthesis of bacteria that permanently remained associated with the substrate. Permanent association can occur only with the indigestible fraction, and probably represents bacterial debris. Lysis and/or detachment of bacterial cells was highest for lucerne, and was indicative of the rapid dynamics of lucerne digestion.

The pattern of food intake can affect the regulation of body weight and lipogenesis. We studied the effect of meal frequency on human energy expenditure (EE) and its components. During 1 week ten male adults (age 25–61 years, body mass index 20·7–30·4 kg/m2) were fed to energy balance at two meals/d (gorging pattern) and during another week at seven meals/d (nibbling pattern). For the first 6 d of each week the food was provided at home, followed by a 36 h stay in a respiration chamber. O2 consumption and CO2 production (and hence EE) were calculated over 24 h. EE in free-living conditions was measured over the 2 weeks with doubly-labelled water (average daily metabolic rate, ADMR). The three major components of ADMR are basal metabolic rate (BMR), diet-induced thermogenesis (DIT) and EE for physical activity (ACT). There was no significant effect of meal frequency on 24 h EE or ADMR. Furthermore, BMR and ACT did not differ between the two patterns. DIT was significantly elevated in the gorging pattern, but this effect was neutralized by correction for the relevant time interval. With the method used for determination of DIT no significant effect of meal frequency on the contribution of DIT to ADMR could be demonstrated.

The effect of addition of sugar-beet fibre to the diet on sterol excretion from the small intestine was studied in nine ileostomy subjects. A constant low-fibre diet was given in two 3 d periods with and without 32 g sugar-beet fibre/d in random order. Care was taken to minimize bacterial alteration of the ileostomy contents. The addition of sugar-beet fibre increased net cholesterol excretion by 52 (SE 9)%(P < 0·01), from 294 (SE 99) to 451 (SE 124) mg/d, and decreased bile acid excretion by 26 (SE 15)% (P < 0·01), from 764 (SE 118) to 567 (SE 96) mg/d. The increased cholesterol and decreased bile acid excretion found with sugar-beet fibre addition is different from the pattern associated with fibre sources such as pectin and oat fibre. The interaction between dietary fibre and sterol metabolism may be mediated, therefore, by different mechanisms depending on the fibre source.

Oat bran concentrated with respect to both soluble and insoluble dietary fibre was fed to adult rats, and its effects on serum cholesterol and liver fat infiltration were studied. The feeds contained 15, 30 or 45 g β-glucan/kg, except in control groups where the fibre of the feed was cellulose. Half the experimental groups received the feeds with an addition of 10 g cholesterol and 2 g cholic acid/kg in order to create an hypercholesterolaemic condition. In normocholesterolaemic rats the concentrated oat bran did not cause any significant changes in serum cholesterol concentrations, but reduced liver weight compared with control group rats fed on the cellulose-containing diet. In hypercholesterolaemic rats the concentrated oat bran reduced serum total cholesterol and increased high-density-lipoprotein cholesterol concentration. The effect was seen already at the 15 g/kg concentration of β-glucan, the higher doses tested did not significantly improve the effect. Infiltration of fat into liver cells, especially in the periportal areas, was observed only in hypercholesterolaemic animals. The fat infiltration was accentuated in proportion to the amount of oat-bran concentrate fed.

Effects of acute (meal) and chronic (diet) level of protein supply on metabolic leucine utilization were investigated in growing (10 weeks) and mature (> 1 year) rats. Rats were conditioned on a high-protein (HP) diet (210 g casein/kg feed) or a low-protein(LP) diet (75 g casein/kg feed) from 7 weeks of age. Overnight-fasted rats were offered a HP or LP meal during a 8 h 14CO2 breath test with a constant infusion of either L-[l-14C]leucine (carboxyl, CL) or L-[U-14C]leucine (universal, UL). Before the meal 14CO2 output was lower for overnight-fasted rats fed on LP than on HP (P < 0·001), and also lower for growing than for mature rats (P < 0·001). Meal ingestion resulted in a rapid increase in 14CO2 output. From 2 h after the start of the meal the effect of acute protein supply on 14CO2 output was significant (P < 0·001), while the effect of chronic protein supply disappeared for CL. After the meal 14CO2 output was transiently lower for growing than for mature rats (P < 0·05), especially after the LP meal. The difference in 14CO 2 output between CL and UL increased transiently after the meal, indicating an increase in decarboxylation relative to total oxidation of leucine. In conclusion: (1) metabolic leucine utilization after overnight fasting depends on the level of chronic protein supply and stage of development of the animal, (2) metabolic leucine utilization after feeding depends primarily on the level of acute protein supply, (3) the transient increase in non-protein label retention suggests a temporal oversupply ofleucine relative to the actual metabolic state.

Information on diet composition, daily energy expenditure, energy storage and the utilization of energy in the prey are important factors when evaluating the food consumption of minke whales (Balaenoptera acutorostrata) during their summer stay in northern waters. The purpose of the present study was in this context to obtain information on the digestible energy (DE) of different prey selected by minke whales. An in vitro three-stage digestion technique, simulating the different compartments of the digestive system, has been developed. The initial step simulated the anaerobic microbial fermentation of substrate in the fortestomach. The next stage included the addition of pepsin (EC 3.4.23.1)–HCI, simulating ventricle enzymic decomposition, and finally, in the third step, fresh extract from duodenal contents was used to simulate enzymic intestinal degradation of the remaining components of the food. The inoculum was normally obtained from animals which had recently eaten the prey to be tested. In such tests we obtained a dry matter disappearance (DMD) and a DE for herring (Clupea harengus) of 80·4 (SD 5·0)% (n 18) and 92·1 (SD 3·7) % (n 16) respectively, and a DMD of krill (Thysanoessa sp.) of 83·4 (SD 4·9)% (n 6). The DMD of krill was reduced to 73·8 (SD 7·3)% (n 8) while the DE was 70·6 (SD 10·4) % (n 7) when inoculum from whales which had recently eaten cod (Gadus morhua) and haddock (Melanogrammus aeglefinus) was used. These results indicate a high digestibility of the most common species of prey in these animals, and also that the whales have little difficulty in changing from one prey species to another.

The effects of guar gum, sugar-beet fibre (SBF) and wheat bran supplementation of a high-fat test meal were compared with an NSP-free control meal and a meal containing an equivalent amount of the ion- exchange resin cholestyramine in healthy non-obese human volunteers. Their effects on gastric emptying, postprandial circulating bile acids, triacylglycerols and gastrointestinal hormone levels were studied. The in vitro binding of NSP and cholestyramine to [l-14C]glycocholic acid was measured and compared with their in vivo effect. Guar gum and cholestyramine supplementation significantly lowered circulating postprandial bile acid, triacylglycerol and gastric inhibitory polypeptide concentrations, but sugar-beet fibre and wheat bran were without effect. Liquid gastric emptying, as assessed by circulating paracetamol levels, was slightly accelerated in the guar gum-supplemented meal. Glycocholic acid bound strongly to the insoluble fraction of cholestyramine and the soluble fraction of guar gum. The insoluble fractions of SBF and wheat bran bound only small quantities of glycocholate; no bile acid binding was detected in the soluble fractions of these NSP. The study demonstrates that measurement of postprandial bile acids enables an indirect measurement to be made of bile acid binding to NSP in vivo. The results support the hypothesis that the hypocholesterolaemic action of guar gum is largely mediated via interruption of the enterohepatic bile acid circulation, but indicate that the hypocholesterolaemic action of SBF is mediated by another mechanism.

Broiler chickens were fed on a control diet based on maize or on diets including conventionally-dried (standard pulp) or vacuum-dried (Fipec pulp) sugar-beet pulp at inclusion levels of 23, 46 and 92 g/kg. Diets were fed ad lib. except for half of the chickens fed on the control diet which were fed twice daily at a restricted level. Although not statistically significant, chickens given the 23 g sugar-beet pulp/kg diets generally consumed more feed, had increased body weights and converted feed more efficiently at both day 14 and 21 than those fed on the control diet or diets including 46 or 92 g sugar-beet pulp/kg. Ileal digestibilities of organic matter, crude fat and crude protein generally decreased with increasing levels of sugar-beet pulp in the diet. Birds fed on diets including sugar-beet pulp had reduced total serum cholesterol concentrations. There were only minor differences obtained in production responses, serum cholesterol concentrations, digestibilities and carcass composition between chickens fed on the two types of beet pulp, indicating that the different drying procedures had very little influence on the product. High triacylglycerol and total serum cholesterol concentrations obtained for the restricted level-fed chickens demonstrated a meal frequency factor.

Severe vitamin A deficiency in rats is known to cause anaemia associated with growth retardation and impaired water retention. However, study of the effect of marginal vitamin A intake is of more interest because such intake may mirror the situation in humans in many developing countries. Therefore, in two experiments, the effect of marginal vitamin A deficiency on Fe status was investigated in male rats. After 28 d of feeding either low- or high-vitamin A diets (0 or 120 v. 1200 retinol equivalents/kg feed), body weight and feed intake were not influenced by the level of vitamin A in the diet. Liver weight was lowered by vitamin A deficiency. Water intake was not influenced in rats fed on a low-vitamin A diet. Plasma retinol concentrations were decreased in rats fed on diets low in vitamin A. Marginal vitamin A deficiency produced slightly lower blood haemoglobin concentrations; it did not systematically affect packed cell volume. The concentration of Fe in liver was significantly higher when diets low in vitamin A were fed, but hepatic Fe mass was not affected. Significantly lower Fe levels were observed in femurs of rats with vitamin A deficiency. The effects on liver and femur Fe concentrations were seen with diets adequate in Fe but not with diets deficient in Fe. The efficiency of apparent Fe absorption was significantly increased by low intakes of vitamin A, provided that the dietary Fe concentration was adequate. It is speculated that depressed uptake of Fe by bone marrow is the primary feature of altered Fe status in rats with marginal vitamin A deficiency.

The effect of dietary whey protein v. casein on plasma and liver cholesterol concentrations was investigated in female, weanling rats. Balanced, purified diets containing either whey protein or casein, or the amino acid mixtures simulating these proteins, were used. The high-cholesterol diets (10 g cholesterol/kg feed) had either 150 or 300 g protein or amino acids/kg feed. The diets were fed for 3 weeks. At the low dietary protein level, whey protein v. casein did not affect plasma total cholesterol, but lowered the concentration of liver cholesterol. At the high dietary protein level, whey protein significantly lowered plasma and liver cholesterol and also plasma triacylglycerols. The hypocholesterolaemic effect of whey protein was associated with a decrease in very-low-density-lipoprotein cholesterol. At the high dietary protein concentration, whey protein reduced the faecal excretion of bile acids when compared with casein. The effects of intact whey protein v. casein were not reproduced by the amino acid mixtures simulating these proteins. It is suggested tentatively that the cholesterol-lowering effect of whey protein in rats is caused by inhibition of hepatic cholesterol synthesis.