Docosahexaenoic acid enrichment in NAFLD is associated with improvements in hepatic metabolism and hepatic insulin sensitivity: a pilot study.

This corrects the article DOI: 10.1038/ejcn.2017.9.

Docosahexaenoic acid enrichment in NAFLD is associated with improvements in hepatic metabolism and hepatic insulin sensitivity: a pilot study.

BACKGROUND/OBJECTIVE: Treatment of subjects with non-alcoholic fatty liver disease (NAFLD) with omega-3 polyunsaturated fatty acids (FAs) suggests high levels of docosahexaenoic acid (DHA) tissue enrichment decrease liver fat content. We assessed whether changes in erythrocyte DHA enrichment (as a surrogate marker of changes in tissue enrichment) were associated with alterations in hepatic de novo lipogenesis (DNL), postprandial FA partitioning and hepatic and peripheral insulin sensitivity in a sub-study of the WELCOME trial (Wessex Evaluation of fatty Liver and Cardiovascular markers in NAFLD (non-alcoholic fatty liver disease) with OMacor thErapy). SUBJECTS/METHODS: Sixteen participants were randomised to 4 g/day EPA+DHA (n=8) or placebo (n=8) for 15-18 months and underwent pre- and post-intervention measurements. Fasting and postprandial hepatic FA metabolism was assessed using metabolic substrates labelled with stable-isotope tracers (2H2O and [U13C]palmitate). Insulin sensitivity was measured by a stepped hyperinsulinaemic-euglycaemic clamp using deuterated glucose. Participants were stratified according to change in DHA erythrocyte enrichment (< or ⩾2% post intervention). RESULTS: Nine participants were stratified to DHA⩾2% (eight randomised to EPA+DHA and one to placebo) and seven to the DHA<2% group (all placebo). Compared with individuals with erythrocyte <2% change in DHA abundance, those with ⩾2% enrichment had significant improvements in hepatic insulin sensitivity, reduced fasting and postprandial plasma triglyceride concentrations, decreased fasting hepatic DNL, as well as greater appearance of 13C from dietary fat into plasma 3-hydroxybutyrate (all P<0.05). CONCLUSIONS: The findings from our pilot study indicate that individuals who achieved a change in erythrocyte DHA enrichment ⩾2% show favourable changes in hepatic FA metabolism and insulin sensitivity, which may contribute to decreasing hepatic fat content.

Metabolically healthy and unhealthy obesity: differential effects on myocardial function according to metabolic syndrome, rather than obesity.

BACKGROUND: The term 'metabolically healthy obese (MHO)' is distinguished using body mass index (BMI), yet BMI is a poor index of adiposity. Some epidemiological data suggest that MHO carries a lower risk of cardiovascular disease (CVD) or mortality than being normal weight yet metabolically unhealthy. OBJECTIVES: We aimed to undertake a detailed phenotyping of individuals with MHO by using imaging techniques to examine ectopic fat (visceral and liver fat deposition) and myocardial function. We hypothesised that metabolically unhealthy individuals (irrespective of BMI) would have adverse levels of ectopic fat and myocardial dysfunction compared with MHO individuals. SUBJECTS: Individuals were categorised as non-obese or obese (BMI ⩾30 kg m(-2)) and as metabolically healthy or unhealthy according to the presence or absence of metabolic syndrome. METHODS: Sixty-seven individuals (mean±s.d.: age 49±11 years) underwent measurement of (i) visceral, subcutaneous and liver fat using magnetic resonance imaging and proton magnetic resonance spectroscopy, (ii) components of metabolic syndrome, (iii) cardiorespiratory fitness and (iv) indices of systolic and diastolic function using tissue Doppler echocardiography. RESULTS: Cardiorespiratory fitness was similar between all groups; abdominal and visceral fat was highest in the obese groups. Compared with age- and BMI-matched metabolically healthy counterparts, the unhealthy (lean or obese) individuals had higher liver fat and decreased early diastolic strain rate, early diastolic tissue velocity and systolic strain indicative of subclinical systolic and diastolic dysfunction. The magnitude of dysfunction correlated with the number of components of metabolic syndrome but not with BMI or with the degree of ectopic (visceral or liver) fat deposition. CONCLUSIONS: Myocardial dysfunction appears to be related to poor metabolic health rather than simply BMI or fat mass. These data may partly explain the epidemiological evidence on CVD risk relating to the different obesity phenotypes.

Differential effects of fatness, fitness and physical activity energy expenditure on whole-body, liver and fat insulin sensitivity.

AIMS/HYPOTHESIS: The relative contributions of fitness (maximal oxygen uptake), physical activity energy expenditure (PAEE) and fatness to whole-body, liver and fat insulin sensitivity is uncertain. The aim of this study was to determine whether fitness and PAEE are associated with whole-body, liver and fat insulin sensitivity independently of body fat. MATERIALS AND METHODS: We recruited 25 men (mean [SD] age 53 [6] years). Whole-body (M value) and liver (percentage suppression of endogenous glucose output) insulin sensitivity were estimated using a hyperinsulinaemic-euglycaemic clamp. Insulin sensitivity in fat (insulin sensitivity index for NEFA) was estimated during an OGTT. Total and truncal fat were measured by dual-energy X-ray absorptiometry, fitness by treadmill, and PAEE (n = 21) by 3 day heart rate monitoring and Baecke questionnaire. RESULTS: In univariate analyses, fatness was strongly associated with insulin sensitivity (whole-body, liver and fat). Fitness was associated with whole-body (r = 0.53, p < 0.007) and liver (0.42, p = 0.04) insulin sensitivity, while PAEE was associated with liver insulin sensitivity (r = 0.55, p = 0.01). Regression models were established to describe associations between fatness, fitness and physical activity and measures of insulin sensitivity (whole-body, fat and liver) as outcomes. Only fatness was independently associated with whole-body insulin sensitivity (B coefficient -0.01, p = 0.001). Fitness was not associated with any outcome. Only PAEE was independently associated with liver insulin sensitivity (B coefficient 13.5, p = 0.02). CONCLUSIONS/INTERPRETATION: Fatness explains most of the variance in whole-body insulin sensitivity. In contrast, PAEE explains most of the variance in liver insulin sensitivity.

Cortisol clearance and associations with insulin sensitivity, body fat and fatty liver in middle-aged men.

AIMS/HYPOTHESIS: The regulation of cortisol metabolism in vivo is not well understood. We evaluated the relationship between cortisol metabolism and insulin sensitivity, adjusting for total and regional fat content and for non-alcoholic fatty liver disease. MATERIALS AND METHODS: Twenty-nine middle-aged healthy men with a wide range of BMI were recruited. We measured fat content by dual-energy X-ray absorptiometry and magnetic resonance imaging (MRI), liver fat by ultrasound and MRI, the hypothalamic-pituitary-adrenal axis by adrenal response to ACTH(1-24), unconjugated urinary cortisol excretion, corticosteroid-binding globulin, and cortisol clearance by MS. We assessed insulin sensitivity by hyperinsulinaemic-euglycaemic clamp and by OGTT. RESULTS: Cortisol clearance was strongly inversely correlated with insulin sensitivity (M value) (r = -0.61, p = 0.002). Cortisol clearance was increased in people with fatty liver compared with those without (mean+/-SD: 243 +/- 10 vs 158 +/- 36 ml/min; p = 0.014). Multiple regression modelling showed that the relationship between cortisol clearance and insulin sensitivity was independent of body fat. The relationship between fatty liver and insulin sensitivity was significantly influenced by body fat and cortisol clearance. CONCLUSIONS/INTERPRETATION: Cortisol clearance is strongly associated with insulin sensitivity, independently of the amount of body fat. The relationship between fatty liver and insulin sensitivity is mediated in part by both fatness and cortisol clearance.

Oral estrogen antagonizes the metabolic actions of growth hormone in growth hormone-deficient women.

We have determined whether oral estrogen reduces the biological effects of growth hormone (GH) in GH-deficient (GHD) women compared with transdermal estrogen treatment. In two separate studies, eight GHD women randomly received either oral or transdermal estrogen for 8 wk before crossing over to the alternate route of administration. The first study assessed the effects of incremental doses of GH (0.5, 1.0, 2.0 IU/day for 1 wk each) on insulin-like growth factor I (IGF-I) levels during each estrogen treatment phase. The second study assessed the effects of GH (2 IU/day) on lipid oxidation and on protein metabolism using the whole body leucine turnover technique. Mean IGF-I level was significantly lower during oral estrogen treatment (P < 0.05) and rose dose dependently during GH administration by a lesser magnitude (P < 0.05) compared with transdermal treatment. Postprandial lipid oxidation was significantly lower with oral estrogen treatment, both before (P < 0.05) and during (P < 0.05) GH administration, compared with transdermal treatment. Protein synthesis was lower during oral estrogen both before and during GH administration (P < 0.05). Oral estrogen antagonizes several of the metabolic actions of GH. It may aggravate body composition abnormalities already present in GHD women and attenuate the beneficial effects of GH therapy. Estrogen replacement in GHD women should be administered by a nonoral route.

Renal glucose production compensates for the liver during the anhepatic phase of liver transplantation.

The extent of the renal contribution to postabsorptive endogenous glucose production (EGP) in humans is controversial. We measured EGP in the absence of the liver during the anhepatic phase (AH) of liver transplantation in five patients (aged 46.4+/-10.2 years, two women). Stable labeling of plasma glucose (PG) was achieved for a 2-h period before the AH by primed continuous infusion of di-deuterated 6,6[2H2]glucose (1.7 mg/min) and continued throughout the AH. PG was maintained above the fasting level (6.1+/-2.73 mmol/l) with 5% dextrose labeled with 6,6[2H2]glucose throughout the AH (mean level during the AH 0.98+/-0.45 mg x kg(-1) x min(-1)). Isotopic enrichment remained stable at 0.84+/-0.21% atom percent excess throughout. EGP, calculated by use of a modified Steele equation, decreased from 2.6+/-1.24 at baseline to 0.97+/-0.9 mg x kg(-1) x min(-1) (36% baseline, P = 0.045) but recovered at approximately 30 min to reach 1.38+/-0.83 mg x kg(-1) x min(-1) (54% baseline) by 60 min. Epinephrine, lactate, free fatty acid, and glycerol levels increased significantly (0.79+/-0.74 to 3.65+/-2.1 nmol/l, P = 0.005; 1.88+/-0.43 to 3.46+/-0.9 mmol/l, P = 0.024; 543.9+/-215.5 to 705.5+/-219.2 micromol/l, P = 0.012; 75.6+/-30.2 to 139+/-96.3 micromol/l, P = 0.003, respectively). These data show that postabsorptive nonhepatic glucose production in humans may contribute to greater than one-third of overall EGP, increasing when required, and that it is associated with a stress response and increased gluconeogenic substrate availability. We conclude that extrahepatic tissues, most notably those of the kidney, make a significant contribution to EGP in humans.

Recombinant human insulin-like growth factor-I (rhIGF-I) therapy in adults with type 1 diabetes mellitus: effects on IGFs, IGF-binding proteins, glucose levels and insulin treatment.

OBJECTIVE: Insulin-like growth factor-I (IGF-I) has both insulin-like and anabolic actions but unlike insulin, IGF-I circulates bound to a number of specific binding proteins that regulate its availability and activity. Patients with type 1 diabetes mellitus have low levels of circulating IGF-I despite increased growth hormone (GH) secretion, and are a group that may benefit from rhIGF-I therapy. Understanding the relationship between IGF-I and its binding proteins is necessary to appreciate the actions of exogenously administered rhIGF-I. Therefore, we examined the effects of 19 days' subcutaneous administration of rhIGF-I (50 micrograms/kg BID) on the levels of IGF-I, IGF-II and the IGF-binding proteins (IGFBPs), as well as the daily dose of insulin necessary to maintain glycaemic control in patients with type 1 diabetes mellitus. DESIGN AND PATIENTS: This was an open study, and the patients were studied initially while resident (days 1-5) in the hospital and thereafter (days 6-24) as outpatients. Serum was collected at baseline and at intervals throughout the study for the measurement of total IGF-I, IGF-II, IGFBP-1, -2, -3, free insulin and growth hormone (GH). Daily insulin doses and glucometer readings were recorded throughout the study. The changes in each of these variables were examined. The subjects were six adults (35.3 +/- 4.0 years, mean +/- SE), with type 1 diabetes, and all had reasonable glycaemic control (HbA1c 7.2 +/- 0.5%). RESULTS: rhIGF-I administration increased circulating total IGF-I over two-fold (15.3 +/- 1.9 vs. 33.7 +/- 5.4 nmol/l, mean +/- SEM, P < 0.01, day 1 vs. day 20) and decreased plasma IGF-II concentration (85.0 +/- 4.7 vs. 50.6 +/- 4.7 nmol/l, P < 0.01, day 1 vs. day 20). The dose of insulin required for adequate glycaemic control decreased significantly during rhIGF-I therapy (46 +/- 7 vs. 31 +/- 8 U/day, P < 0.05, day -1 vs. day 19), as did the fasting free insulin concentration (8.4 +/- 1.5 vs. 5.0 +/- 0.8 mU/l, P < 0.05, baseline vs. day 5). IGFBP-2 concentration increased (388 +/- 115 vs. 758 +/- 219 micrograms/l, P < 0.05, day 1 vs. day 20), but IGFBP-1 and IGFBP-3 were unchanged during rhIGF-I treatment. Mean nocturnal GH concentration decreased (12.7 +/- 3.3 vs. 3.8 +/- 0.9 mU/l, P = 0.05) after 4 days' rhIGF-I therapy. CONCLUSION: Twice daily rhIGF-I therapy in adults with type 1 diabetes resulted in an increase in circulating IGF-I with a reciprocal decrease in IGF-II, and a marked elevation of IGFBP-2 concentration. The levels of IGFBP-1 and -3 were not dramatically changed despite a reduction in the concentration of serum free insulin, and a large decrease in the requirement for insulin. The mechanisms behind these changes remains unclear but alterations in circulating levels of of IGFBPs may alter IGF-I bioactivity. If rhIGF-I is to have an application in the management of adults with type 1 diabetes, further work is necessary to determine the metabolic consequences of the alterations seen in the IGFs and their binding proteins following rhIGF-I administration.

rhIGF-I administration reduces insulin requirements, decreases growth hormone secretion, and improves the lipid profile in adults with IDDM.

IDDM is associated with elevated circulating levels of growth hormone (GH) and reduced insulin-like growth factor I (IGF-I). GH antagonizes the action of insulin-increasing insulin requirements in IDDM. The effects of subcutaneously administered rhIGF-I on glycemic control, insulin requirements, and GH secretion were studied in eight adults with IDDM. Patients received either placebo or rhIGF-I (50 microg/kg b.i.d.) for 19 days in a randomized, double-blind, parallel-design, placebo-controlled trial. Overnight GH, plasma glucose, free insulin, IGF-I, fructosamine, and lipid profiles were assessed during this period. rhIGF-I therapy increased IGF-I concentration from 117.1 +/- 14.2 (mean +/- SE) ng/ml (baseline) to 310.5 +/- 40.6 and 257.1 +/- 41.2 ng/ml on day 5 (P < 0.01 vs. baseline) and day 20 (P < 0.01 vs. baseline), respectively. After 19 days of rhIGF-I treatment, fructosamine concentrations were unchanged compared with baseline (439 +/- 32 vs. 429 +/- 35 micromol/l, day -1 vs. day 20, respectively), yet insulin requirements were decreased by approximately 45% (0.67 +/- 0.08 vs. 0.36 +/- 0.07 U x kg(-1) x day(-1), day -1 vs. day 19, respectively, P < 0.005). After 4 days of rhIGF-I therapy, there was a decrease in free insulin levels (8.38 +/- 1.47 vs. 4.98 +/- 0.84 mU/l, P < 0.05), mean overnight GH concentration (12.6 +/- 3.3 vs. 3.8 +/- 2.1 mU/l, P = 0.05), and total cholesterol and triglycerides (4.68 +/- 0.31 vs. 4.25 +/- 0.35 mmol/l, P < 0.05, 1.27 +/- 0.19 vs. 0.95 +/- 0.21 mmol/l, P < 0.001, respectively). There was no change in any variable in the placebo-treated patients. This study demonstrates that subcutaneous administration of rhIGF-I decreases insulin requirements and improves the plasma lipid profile while maintaining glycemic control in adults with IDDM. The excess nocturnal release of GH, characteristic of IDDM, is also decreased by rhIGF-I therapy.

The effect of recombinant human growth hormone on glucose and leucine metabolism in Cushing's syndrome.

Cushing's syndrome is characterized by central obesity and muscle wasting. As GH is anabolic, it may be able to counteract the loss of body protein. To evaluate the potential therapeutic use of GH preoperatively, eight patients with Cushing's syndrome received sc injections of recombinant human GH (0.07 U/kg.day) for 7 days. Whole body leucine and glucose turnover were measured after an infusion of [1-13C]leucine and [6,6-2H2]glucose before (day 0) and after 2 and 7 days of GH treatment. Compared with the value on day 0, there was a significant increase on days 2 and 7 in insulin (P < 0.005 and P < 0.001), C peptide (P < 0.01 and P < 0.005), insulin-like growth factor I (P < 0.001), and glucose concentrations (P < 0.01 and P < 0.005) and a decrease in the leucine concentration (P < 0.005). There was no significant change in glucose production rate, glucose MCR, leucine production rate (a measure of protein degradation), or nonoxidative leucine disappearance rate (a measure of protein synthesis). The leucine MCR was increased after 7 days (P < 0.05), and the clearance of leucine into protein (nonoxidative leucine disappearance rate/leucine concentration) was increased (P < 0.05) after 2 and 7 days of GH treatment. This is consistent with GH stimulating the availability of amino acid transporters. GH may, therefore, have a therapeutic role in the preoperative treatment of Cushing's syndrome.

Simvastatin decreases the hepatic secretion of very-low-density lipoprotein apolipoprotein B-100 in heterozygous familial hypercholesterolaemia: pathophysiological and therapeutic implications.

We studied six patients with heterozygous familial hypercholesterolaemia (FH) before and after 8 weeks of treatment with simvastatin (40 mg day-1), an inhibitor of 3-hydroxy-3-methyl-glutaryl-Coenzyme A. Simvastatin decreased plasma low-density lipoprotein (LDL) cholesterol by 43% (P = 0.002), triglycerides by 15% [corrected] (P = 0.05) and mevalonic acid (a measure of in vivo cholesterol synthesis) by 20% (P = 0.002); high-density lipoprotein cholesterol increased by 17% (P = 0.02). The hepatic secretion rate of very-low-density lipoprotein apolipoprotein B-100 (VLDL apoB) was measured directly using a primed, constant intravenous infusion of 1-[13C]-leucine with monitoring of the isotopic enrichment of apoB by gas chromatography-mass spectrometry; fractional secretion rate (FSR) was derived using a monoexponential function. Simvastatin decreased the FSR, ASR and pool size of VLDL apoB by 17% (14.3 (SEM 3.6)) vs. (11.9 (SEM 3.5) pools day-1, P = 0.10), 83% (51.4 (SEM 17.9) vs. (8.6 (SEM 1.4), P = 0.007 mg kg-1 day-1) and 65% (234.2 (SEM 30.4) vs. 82.6 (SEM 24.0) mg, P = 0.02), respectively. The change in the ASR of VLDL apoB was significantly correlated with the change in plasma LDL cholesterol concentration (P = 0.04), but not with the change of triglyceride or mevalonic acid. We conclude that the hepatic secretion of VLDL apoB in FH is decreased by simvastatin, which may partly explain the fall in plasma cholesterol.(ABSTRACT TRUNCATED AT 250 WORDS)

Increased hepatic secretion of very-low-density-lipoprotein apolipoprotein B-100 in heterozygous familial hypercholesterolaemia: a stable isotope study.

We have measured the hepatic secretion of very-low-density-lipoprotein apolipoprotein B-100 (VLDL apo B) in 6 patients with heterozygous familial hypercholesterolaemia (FH) (4 males, 2 females, age 47.0 +/- 2.7 years (mean +/- S.E.M.), weight 71.0 +/- 5.3 kg) and 6 normocholesterolaemic subjects matched for age, weight and sex (4 males, 2 females, age 47.5 +/- 3.1 years, weight 70.0 +/- 4.4 kg) using a stable isotope method. Each subject received a primed, constant infusion of [I-13C]leucine and isotopic enrichment of VLDL apo B was determined using gas-chromatography mass-spectrometry (GCMS). Mean plasma low-density-lipoprotein (LDL) cholesterol and apo B concentrations in the FH group were more than twice that in the control group (FH, 8.5 +/- 0.5 mmol/l vs. controls, 3.3 +/- 0.2 mmol/l, P < 0.001; and FH, 2.0 +/- 0.1 g/l vs. controls, 1.0 +/- 0.04 g/l, P < 0.0001, respectively). Plasma triglyceride (TG) and high-density-lipoprotein (HDL) cholesterol concentrations were not significantly different between the two groups. Although the fractional secretion rates of VLDL apo B were similar (FH, 14.3 +/- 3.6 pools/day vs. controls, 11.6 +/- 1.7 pools/day, P = 0.53), VLDL apo B pool size and VLDL apo B absolute secretion rates (ASR) were significantly higher in the FH group (FH, 234.2 +/- 27.8 mg vs. controls, 66.3 +/- 13.5 mg, P < 0.001; and FH, 51.4 +/- 17.9 mg/kg per day vs. controls, 9.4 +/- 1.1 mg/kg per day, P < 0.02, respectively). We conclude that FH is associated with increased hepatic secretion of VLDL apo B and that this may contribute to the elevated concentration of LDL-cholesterol. The findings are also consistent with the hypothesis that in FH increased hepatic cholesterol availability (due to increased uptake of LDL-cholesterol via the receptor-independent pathway) stimulates hepatic secretion of VLDL apo B.

Increased hepatic secretion of very-low-density lipoprotein apolipoprotein B-100 in obesity: a stable isotope study.

1. We measured the hepatic secretion of very-low-density lipoprotein apolipoprotein B-100 (VLDL apoB) using a stable-isotope gas chromatography-mass spectrometry method in six obese subjects [three males, three females, age 41.5 +/- 3.4 years (mean +/- SEM), weight 105.0 +/- 4.8 kg, plasma total cholesterol concentration 6.2 +/- 0.4 mmol/l, triacylglycerol 2.8 +/- 0.8 mmol/l, high-density lipoprotein cholesterol 1.0 +/- 0.2 mmol/l] and six lean control subjects (three males, three females, age 41.8 +/- 3.7 years, weight 68.2 +/- 4.9 kg, total cholesterol concentration 4.5 +/- 0.3 mmol/l, triacylglycerol 0.8 +/- 0.2 mmol/l, high-density lipoprotein cholesterol 1.3 +/- 0.1 mmol/l). 2. Plasma total cholesterol, triacylglycerol and mevalonic acid (an index of cholesterol synthesis in vivo) concentrations were significantly higher in the obese subjects than in control subjects (P = 0.02, P = 0.03, P = 0.04, respectively). VLDL apoB pool size and absolute secretion rate were significantly higher in the obese subjects than in control subjects (323.4 +/- 99.8 mg versus 53.6 +/- 17.1 mg, P = 0.004; and 42.3 +/- 13.8 mg kg fat-free mass-1 day-1 versus 10.7 +/- 0.4 mg kg fat-free mass-1 day-1, P = 0.01), but there was no significant difference in the fractional catabolic rate of VLDL apoB.(ABSTRACT TRUNCATED AT 250 WORDS)