Substituting dietary saturated for monounsaturated fat impairs insulin sensitivity in healthy men and women: The KANWU Study.

AIMS/HYPOTHESIS: The amount and quality of fat in the diet could be of importance for development of insulin resistance and related metabolic disorders. Our aim was to determine whether a change in dietary fat quality alone could alter insulin action in humans. METHODS: The KANWU study included 162 healthy subjects chosen at random to receive a controlled, isoenergetic diet for 3 months containing either a high proportion of saturated (SAFA diet) or monounsaturated (MUFA diet) fatty acids. Within each group there was a second assignment at random to supplements with fish oil (3.6 g n-3 fatty acids/d) or placebo. RESULTS: Insulin sensitivity was significantly impaired on the saturated fatty acid diet (-10%, p = 0.03) but did not change on the monounsaturated fatty acid diet (+2%, NS) (p = 0.05 for difference between diets). Insulin secretion was not affected. The addition of n-3 fatty acids influenced neither insulin sensitivity nor insulin secretion. The favourable effects of substituting a monounsaturated fatty acid diet for a saturated fatty acid diet on insulin sensitivity were only seen at a total fat intake below median (37E%). Here, insulin sensitivity was 12.5% lower and 8.8% higher on the saturated fatty acid diet and monounsaturated fatty acid diet respectively (p = 0.03). Low density lipoprotein cholesterol (LDL) increased on the saturated fatty acid diet (+4.1%, p < 0.01) but decreased on the monounsaturated fatty acid diet (MUFA) (-5.2, p < 0.001), whereas lipoprotein (a) [Lp(a)] increased on a monounsaturated fatty acid diet by 12% (p < 0.001). CONCLUSIONS/INTERPRETATION: A change of the proportions of dietary fatty acids, decreasing saturated fatty acid and increasing monounsaturated fatty acid, improves insulin sensitivity but has no effect on insulin secretion. A beneficial impact of the fat quality on insulin sensitivity is not seen in individuals with a high fat intake (> 37E%).

Dietary variables and glucose tolerance in pregnancy.

OBJECTIVE: To investigate relationships between dietary macronutrient intakes and glucose tolerance in pregnancy RESEARCH DESIGN AND METHODS: Nulliparous pregnant Chinese women diagnosed with gestational diabetes mellitus (GDM) (n = 56) were compared to age-, gestational age-, height-, and parity-matched groups with normal glucose tolerance (n = 77) and glucose intolerance (IGT) (n = 38) based on the results of an oral glucose tolerance test (National Diabetes Data Group criteria), performed between 24 and 28 weeks of pregnancy. A 24-h recall dietary assessment was also obtained at the time of screening. RESULTS: Subjects with IGT and GDM were significantly heavier (66.1 +/- 1.4 and 68.6 +/- 1.2 kg, respectively, mean +/- SEM) (P < 0.0001) than the normal group (61.2 +/- 1.8 kg) and had a higher BMI. Overall energy intake was similar between groups, as were the intakes of each macronutrient (%kcal). However, there was a highly significant reduction in polyunsaturated fat intake in the IGT and GDM groups whether expressed as %kcal, % of total fat, or fat kcal. This effect was independent of body weight or BMI whether assessed by ordinal logistic regression or by analysis of a weight- and BMI-matched subgroup of the subjects (P = 0.002 for %kcal; n = 47 normal, 26 IGT, and 43 GDM subjects). In logistic regression analysis of the complete data set, increased body weight (P < 0.0001) and decreased polyunsaturated fat intake (P = 0.0014) were both independent predictors of glucose intolerance (IGT and GDM), as were increased body weight and a low dietary polyunsaturated to saturated fat ratio. CONCLUSIONS: Increased polyunsaturated fat intake is associated with a reduced incidence of glucose intolerance during pregnancy. This finding may have major implications for dietary management of women with or at risk of developing GDM.

Role of dietary factors: macronutrients.

Insulin resistance is an important early marker of the metabolic syndrome disease cluster. Our understanding of the role of dietary macronutrients in the etiology of insulin resistance is currently limited by a paucity of credible intervention studies in humans. In contemplating such studies there are many issues that need consideration from actual study design (e.g., duration of intervention, study population, cross-over or not, nutrient formulation) to practical issues such as palatability and compliance (i.e., that terribly important issue of achievability because realistically individuals must be "free range" in order to complete studies of sufficient duration). Initiatives to support well-designed multicenter studies on diet and insulin resistance would have a major impact on our ability to treat, but more importantly to prevent, the metabolic syndrome diseases.

Interrelationships between muscle fibre type, substrate oxidation and body fat.

OBJECTIVE: To investigate interrelationships between muscle fibre type, respiratory exchange ratio (RER) during exercise at a fixed workload and adiposity. DESIGN: Cross-sectional study. SUBJECTS: 21 untrained, healthy male subjects. MEASUREMENTS: Body fat composition by dual-energy X-ray absorptiometry (DEXA). Exercise test at 55% of VO2max, muscle fibre type composition, muscle NADH and citrate synthase enzyme activity levels; serum insulin, glucose and cortisol concentrations. RESULTS: Percent body fat was inversely correlated to the proportion of type I muscle fibres (r=-0.55, P<0.02). In addition percent trunk fat was negatively correlated with percent type I fibres (r=-0.58, P<0.01) while this relationship was not present for percent leg fat. There was no relation between RER at rest or during exercise and muscle fibre type composition or percent body fat. CONCLUSION: Body fat and percent type I muscle fibres were correlated, supporting skeletal muscle fibre type as a potential etiological factor in obesity. No correlation was observed between percent body fat and substrate oxidation at rest or during moderate exercise, indicating that muscle fuel substrate mix does not appear to provide a mechanism for this relation under either condition.

Does dietary fat influence insulin action?

What is clear from the research thus far is that dietary fat intake does influence insulin action. However, whether the effect is good, bad, or indifferent is strongly related to the fatty acid profile of that dietary fat. The evidence has taken many forms, including in vitro evidence of differences in insulin binding and glucose transport in cells grown with different types of fat in the incubation medium, in vivo results in animals fed different fats, relationships demonstrated between the membrane structural lipid fatty acid profile and insulin resistance in humans, and finally epidemiological evidence linking particularly high saturated fat intake with hyperinsulinemia and increased risk of diabetes. This contrasts with the lack of relationship, or even possible protective effect, of polyunsaturated fats. In particular, habitual increased n-3 polyunsaturated dietary fat intake (as fish fats) would appear to be protective against the development of glucose intolerance. It is reassuring that the patterns of dietary fatty acids that appear beneficial for insulin action and energy balance are also the patterns that would seem appropriate in the fight against thrombosis and cardiovascular disease. Mechanisms, though, still need to be defined. However, there are strong indicators that defining the ways in which changes in the fatty acid profile of membrane structural lipids are achieved, and in turn influence relevant transport events, plus understanding the processes that control accumulation and availability of storage lipid in muscle may be fruitful avenues for future research. One of the problems of moving the knowledge gained from research at the cellular level through to the individual and on to populations is the need for more accommodating research designs. In vitro studies may provide in-depth insights into intricate mechanisms, but they do not give the "big picture" for practical recommendations. On the other hand, correlational studies tend to be fairly blunt instruments, requiring large numbers that are very often not feasible if a greater depth of understanding of the biological processes is to be incorporated. There may be benefit in turning to the clinical case study as a framework for a more comprehensive analysis of the links between dietary fats and insulin action. The real challenge is to keep the depth of analysis rigorous enough to be able to explain and accommodate individual variation (i.e., the diversity of both environmental and genetic backgrounds) while at the same time satisfying the cultural need to provide appropriate overall dietary guidelines. Finally, David Kritchevsky brought to our attention a delightful quote from Mark Twain: "There is something fascinating about science. One gets such a wholesale return of conjecture for such a trifling investment of fact." In the field of dietary fats and the Metabolic Syndrome, this quotation is, unfortunately, apt. Much more research is necessary to define how dietary fats really work to affect insulin action. Well designed, long-term studies in "free range" humans must be undertaken if dietary guidelines for the Metabolic Syndrome are to be based on anything more than a "trifling" amount of "fact."

Fatty acids, triglycerides and syndromes of insulin resistance.

Muscle plays a major role in insulin-stimulated glucose disposal. There is now a range of evidence in humans and experimental animals demonstrating strong relationships between the fatty acid composition of structural membrane lipids and insulin action. The in vivo work is correlative but the in vitro studies suggest a causal relationship exists. Good insulin action is associated with an increased proportion of n-3 fatty acids, low saturates, a low n-6/n-3 ratio and possibly increased monounsaturates. What is reassuring is that there is a pleasing symmetry with the fatty acid pattern that might lead to decreased thrombosis. There is little argument about saturated fats with a reduction having a range of beneficial effects. However, the n-3 fatty acids might also be a key to amelioration of both insulin resistance and thrombosis. The sites of action of n-3s are multiple: decreased triglyceride and VLDL production; inhibition of thromboxane A2 production, increased thromboxane A3 and decreased platelet aggregation; reduction of triglyceride and VLDL concentration; improved blood rheology and membrane transport; action on the endothelium and proliferation of the intimal cells, and improvement of vascular tone. The data here are now strong and reasonably consistent. Similarly, after initial controversy, the evidence for n-3s playing a beneficial role in insulin action is now accumulating. The n-6 PUFAs are a bit of a worry: while arachidonic acid levels in muscle phospholipid has linked positively to insulin action in our studies, linoleic is negative. Linoleic acid, in high amounts, is known to inhibit the delta6 fatty acid desaturase enzyme and with the competition between n-6 and n-3 fatty acids for the enzymes of desaturation and elongation it does focus on a high n-6/n-3 ratio as a critical factor in both insulin resistance and atherosclerosis.

A review of observational studies on the relationship between cholesterol and coronary heart disease.

The likelihood that an association observed in epidemiological studies is one of cause and effect is often evaluated using criteria first put forward by Bradford Hill. The evidence on whether abnormal concentrations of cholesterol and lipoproteins in the blood plasma cause coronary heart disease (CHD) is considered using Bradford Hill's criteria. Evidence from observational studies, backed by evidence from clinical, animal and laboratory studies, leaves no doubt that a high plasma cholesterol concentration is a cause of CHD.

Diet and coronary heart disease. The National Heart Foundation of Australia.

Over the last four decades there has been extensive research into the links between diet and coronary heart disease. The most recent literature is reviewed in this position statement. The clinical and public health aspects of the National Heart Foundation's nutrition policy are based on this review. The key points are as follows: 1. Saturated fatty acids A high intake of saturated fatty acids is strongly associated with elevated serum cholesterol and LDL-cholesterol levels and increased risk of coronary heart disease. 2. The n-6 polyunsaturated fatty acids The n-6 polyunsaturated fatty acids (principally linoleic acid) lower serum cholesterol levels when substituted for saturated fats and probably have an independent cholesterol-lowering effect. 3. The n-3 polyunsaturated fatty acids (fish oils) The n-3 polyunsaturated fatty acids reduce serum triglyceride levels, decrease the tendency to thrombosis and may further reduce coronary risk through other mechanisms. 4. Monounsaturated fatty acids Monounsaturated fatty acids reduce serum cholesterol levels when substituted for saturated fatty acids. It is not clear whether this is an independent effect or simply the result of displacement of saturates. 5. Trans fatty acids Trans fatty acids may increase serum cholesterol levels and can be reckoned to be equivalent to saturated fatty acids. 6. Total fat Total fat intake, independent of fatty acid type, is not strongly associated with coronary heart disease but may contribute to obesity. Associations between total fat intake and coronary heart disease are primarily mediated through the saturated fatty acid component. 7. Dietary cholesterol Dietary cholesterol increases serum cholesterol levels in some people and may increase risk of coronary heart disease. 8. Alcohol A high intake of alcohol increases blood pressure and serum triglyceride levels and increases mortality from cardiovascular disease. Light alcohol consumption reduces the risk of coronary heart disease. 9. Sugar The consumption of sugar is not associated with coronary heart disease. 10. Sodium and potassium High salt intake is related to hypertension especially in the subset of "salt-sensitive" people. Potassium intake may be inversely related to hypertension. 11. Overweight and obesity Abdominal obesity increases the risk of coronary heart disease probably by adversely influencing conventional risk factors. 12. Vegetarianism A high intake of plant foods reduces the risk of coronary heart disease through several mechanisms, including lowering serum cholesterol and blood pressure levels.

Potential health effects of greenhouse effect and ozone layer depletion in Australia.

OBJECTIVE: To identify potential health effects of the greenhouse effect and ozone layer depletion in Australia. DATA SOURCES: Data were derived from a number of sources: (i) published articles accessed from relevant databases in the disciplines of health, public health and climatology over the past 20 years; (ii) published conference proceedings, review monographs and government reports covering the topic; (iii) a survey of experts in public health and climatology/geography (150 individuals were surveyed in the first phase with a 63% response rate); and (iv) a consensus conference in which 22 invited experts reviewed the results of the literature review and survey and a second conference in which 18 senior members of the health bureaucracy and public health profession considered the implications of the findings. STUDY SELECTION: Over 200 published articles or monographs were reviewed. Criteria for selection were whether the papers contributed information to the objectives of the review. DATA EXTRACTION: Because of the nature of the problem under investigation, predictions based on reasonable scientific assumptions were the major content of the review rather than conclusions based on scientific research. DATA SYNTHESIS: The major predicted health effects of long-term climatic change in Australia are skin and eye damage from increased ultraviolet radiation exposure, increased incidence of some respiratory diseases, vector-borne and water-borne diseases, and the social and physical effects of natural hazards and social and economic restructuring. The most vulnerable groups include the aged, the very young, the chronically ill, those living in poorly designed neighbourhoods and those working in outdoor occupations or heavy industry. CONCLUSIONS: The potential effects on health of long-term climatic change cover the broad spectrum of public health concerns. Detailed predictions of likely problems in specific geographic areas are not yet possible, but progressive development of such predictive capability is a high priority. Doctors will have an increasingly important role in monitoring local health status and participating in disease prevention and surveillance programmes.

Retroendocytosis of high density lipoproteins by the human hepatoma cell line, HepG2.

When human HepG2 hepatoma cells were pulsed with 125I-labeled high density lipoproteins (HDL) and chased in fresh medium, up to 65% of the radioactivity released was precipitable with trichloroacetic acid. Cell-internalized 125I-HDL contributed to the release of acid-precipitable material; when cells were treated with trypsin before the chase to remove 125I-HDL bound to the outer cell membrane, 50% of the released material was still acid-precipitable. Characterization of the radioactive material resecreted by trypsinized cells revealed the presence of particles that were similar in size and density to mature HDL and contained intact apolipoproteins (apo) A-I and A-II. The release of internalized label occurred at 37 degrees C but not at 4 degrees C. Monensin, which inhibits endosomal recycling of receptors, decreased the binding of 125I-HDL to cells by 75%, inhibited the release of internalized radioactivity as acid-precipitable material by 80%, and increased the release of acid-soluble material by 90%. In contrast, the lysosomal inhibitor chloroquine increased the association of 125I-HDL to cells by 25%, inhibited the release of precipitable material by 10%, and inhibited the release of acid-soluble radioactivity by 80%. Pre-incubation with cholesterol caused a 50% increase in the specific binding, internalization, and resecretion of HDL label. Cholesterol affected the release of acid-precipitable label much more (+90%) than that of acid-soluble material (+20%). Taken together, these findings suggest that HepG2 cells can bind, internalize, and resecrete HDL by a retroendocytotic process. Furthermore, the results with cholesterol and monensin indicate that a regulated, recycling, receptor-like molecule is involved in the binding and intracellular routing of HDL.

Effects of blocking plasma lipid transfer protein activity in the rabbit.

Plasma lipid transfer protein activity was completely blocked in rabbits for up to 48 h by infusion with goat antibody to rabbit lipid transfer protein. Lipid transfer protein activity in plasma of control animals, infused with antibody from a non-immune goat, decreased during the experiment but was never less than 50% of pre-infusion levels. During the period that lipid transfer protein activity was completely blocked, there were changes in high-density lipoprotein composition (expressed as % by weight) with a reduction in triacylglycerol from 8.4 +/- 2.4% to 1.0 +/- 0.2% (P less than 0.05) and an increase in esterified cholesterol from 10.7 +/- 1.7% to 14.5 +/- 0.3% (P less than 0.1). In conjunction with the observed changes in high-density lipoprotein composition, there was an increase in high-density lipoprotein particle size from a mean radius of 4.7 to 5.4 nm. The change in composition and particle size was not observed in high-density lipoproteins from control animals. There was a change in the distribution of plasma cholesterol in control animals, with a fall in the proportion of cholesterol in high-density lipoproteins (P less than 0.02) and consequently an increase in the proportion of cholesterol in low-density lipoproteins (P less than 0.02). However, the distribution of plasma cholesterol in animals in which lipid transfer protein activity was inhibited was maintained at original levels during the period of inhibition. Consequently, in these animals, there was a less atherogenic distribution of cholesterol during the period of lipid transfer protein inhibition when compared with control animals. The changes observed in lipoproteins, in the absence of lipid transfer protein activity, demonstrate that lipid transfer protein modifies lipoproteins in vivo and appears to contribute to a more atherogenic lipid profile.

Effects of MK-733 on plasma lipid and lipoprotein levels in subjects with hypercholesterolaemia.

MK-733, which is a competitive inhibitor of the rate-limiting step in cholesterol biosynthesis, or a matching placebo was administered to 30 subjects with primary hypercholesterolaemia (who were already receiving dietary treatment) over a period of four weeks in a double-blind trial. Twenty-one subjects manifested heterozygous familial hypercholesterolaemia and nine subjects had polygenic hypercholesterolaemia. Five subjects received placebo, 15 subjects a low dose of MK-733 (2.5-10 mg/day) and 10 subjects received a high dose of MK-733 (20-80 mg/day). Plasma cholesterol levels in subjects who were receiving MK-733 declined significantly and in a dose-dependent fashion (31% reduction in plasma cholesterol levels with a high dose, 19% reduction with a low dose). Eight of 10 subjects who were receiving a high dose of MK-733 achieved better than a 30% reduction in plasma cholesterol levels after four weeks of treatment. The response was independent of the presence or absence of familial hypercholesterolaemia. High-density lipoprotein cholesterol levels did not change significantly, but there was a suggestive, dose-dependent reduction in plasma triglyceride levels after four weeks of treatment. MK-733 was well-tolerated, appeared to be safe, and may ultimately become an important drug in the management of more severe grades of hypercholesterolaemia.

Purification of human plasma lipid transfer protein using fast protein liquid chromatography.

A system for the isolation of human plasma lipid transfer protein (LTP) has been devised using a combination of conventional and high-performance ion-exchange chromatography. Following initial purification by ammonium sulphate precipitation, ultracentrifugation, hydrophobic interaction and cation-exchange chromatography, appropriate fractions were further purified using the Pharmacia fast protein liquid chromatography system. Using this method of purification, human plasma LTP has been purified more rapidly and with greater recovery than with conventional column chromatography. Whereas two forms of LTP were previously reported from the authors' laboratory [LTP-I, molecular mass (Mr) 69,000 and LTP-II, Mr 55,000], with an improved chromatographic system only one form of LTP (LTP-I) has been isolated. This suggests that LTP-II may have been a fragment of LTP-I, produced during the previously used lengthy purification process.

D-galactosamine induced hepatitis in the rabbit: effect on lecithin:cholesterol acyltransferase activity, plasma lipid transfer protein activity and high density lipoproteins.

Hepatitis was induced in rabbits by a single intraperitoneal injection of D(+)-galactosamine-HCl (750 mg/kg body wt). Plasma lecithin:cholesterol acyltransferase activity fell to 5% and lipid transfer protein activity to 50% of control values 48 hr after injection. Discoid high density lipoprotein began to appear in plasma of treated rabbits 36 hr after injection, along with populations of high density lipoprotein (HDL) which were both smaller (radius 3.7 nm) and larger (radius 5.9 nm) than the original HDL population (radius 4.8 nm).

Immunoprecipitation of lipid transfer protein activity by an antibody against human plasma lipid transfer protein-I.

Two lipid transfer proteins, designated lipid transfer protein-I (Mr 69 000) and lipid transfer protein-II (Mr 55 000), each of which facilitates the transfer of radiolabelled cholesteryl ester, triacylglycerol and phosphatidylcholine between plasma lipoproteins, were purified from human plasma. Immunoglobulin G was prepared from goat antiserum to human lipid transfer protein-I (i.e., anti-human LTP-I IgG). The progressive addition of anti-human LTP-I IgG to buffered solutions containing either a highly purified mixture of human lipid transfer protein-I and lipid transfer protein-II, or highly purified rabbit lipid transfer protein (Abbey, M., Calvert, G.D. and Barter, P.J. (1984) Biochim. Biophys. Acta 793, 471-480) resulted in specific immunoprecipitation and the removal of increasing amounts, up to 100%, of cholesteryl ester, triacylglycerol and phosphatidylcholine transfer activities. However, similar precipitation studies on human and rabbit lipoprotein-free plasma resulted in the progressive removal of all cholesteryl ester and triacylglycerol transfer activities but only 30% (human) or 20% (rabbit) of phosphatidylcholine transfer activity. In all cases more anti-human LTP-I IgG was required to precipitate rabbit lipid transfer activity than human lipid transfer activity. These results suggest that lipid transfer protein-I and lipid transfer protein-II have antigenic sites in common, allowing precipitation of both proteins by specific antibody to lipid transfer protein-I. Most plasma phosphatidylcholine transfer activity is mediated by a protein (or proteins) other than lipid transfer protein-I and lipid transfer protein-II. In lipoprotein-free plasma all cholesteryl ester and triacylglycerol transfer activity, and some phosphatidylcholine transfer activity, is mediated by lipid transfer protein-I (or lipid transfer protein-I and an antigenically similar protein, lipid transfer protein-II.

Influence of various therapies on hyperlipidemia in diabetes mellitus.

The relationship between diabetes mellitus and the factors involved in its treatment, and plasma lipoprotein concentrations, is far from clear. The prognostic value of plasma HDL cholesterol measurements is also unclear, as none of the published data relating to this are concerned with diabetes mellitus. The plasma concentration of HDL at any particular time is the result of a combination of factors. The production of nascent HDL particles by the liver and intestine, and small HDL particles from unknown sources (which may include triglyceride-rich particles undergoing catabolism in the circulation), and the production of HDL esterified cholesterol by LCAT, are balanced by the still poorly understood mechanisms of HDL catabolism outlined above. Until these mechanisms are better understood, we are unlikely to understand the implications of plasma HDL and HDL cholesterol concentrations in diabetes mellitus, and the effects of treatment of diabetes.

Plasma lipids and insulin in gall stone disease: a case-control study.

Fasting plasma lipid and insulin concentrations were measured in 173 patients with gall stones and 284 hospital controls to investigate their relationship to this disease. Multivariate methods of analysis were used to estimate the net associations between individual plasma variables and the risk of developing gall stones. In both sexes increased plasma insulin values were associated with an increased risk of gall stones independently of plasma triglyceride values; increased plasma triglyceride concentrations were associated with an increased risk of gall stones in young subjects only; increased plasma total cholesterol concentrations were associated with a decreased risk of gall stones only after controlling for plasma insulin and triglyceride concentrations; while increased plasma high density lipoprotein cholesterol concentrations were associated with a decreased risk of gall stones, but were confounded by plasma insulin and triglyceride values. These associations were independent of obesity and dietary intake.