IntroductionWith the failure of high-density lipoprotein (HDL) cholesterol raising therapies to reduce cardiovascular risk, attention has turned towards HDL composition and function. There are several techniques for the isolation of lipoproteins from plasma, with density ultracentrifugation considered the gold-standard method. It is not known which form of density ultracentrifugation is optimal for HDL composition and function analyses.MethodsThis study compared HDL composition by nLC-MS/MS and ELISA, subclass distribution by gel electrophoresis, and in vitro vascular anti-inflammatory function in HDL fractions isolated by sodium bromide sequential density ultracentrifugation (SDU) and iodixanol density gradient ultracentrifugation (DGU).ResultsHDL composition differed between the two isolation techniques, with DGU-isolated HDL fractions containing a higher total protein content than SDU (10.14 ± 1.49 mg/mL compared to 3.18 ± 1.10 mg/mL respectively, p < 0.001, mean ± SD) but lower proteomic detection of key HDL proteins such as apolipoprotein A-I, apolipoprotein A-II and paraoxonase-1 in DGU-isolated HDL. HDL subclass distribution could not be determined in DGU HDL fractions due to contaminating plasma proteins. Vascular anti-inflammatory function was higher in DGU HDL (77.5 ± 4.4 % compared to 58.3 ± 4.8%, mean ± SD, p = 0.014) possibly due to the presence of confounding plasma proteins.ConclusionFurther steps may need to be added to DGU methods to obtain a ‘cleaner’ HDL fraction and therefore currently SDU is better suited to studies of HDL composition and function.

Purpose Adipocyte glucocorticoid (GC) signalling influences lipid storage and insulin sensitivity, and South Asians develop insulin resistance at lower BMI than White Europeans. We tested whether early modest weight gain alters adipocyte GC-related transcripts differently by ancestry and whether within-person transcript changes track dynamic insulin responses. Methods White European (n = 21) and South Asian (n = 14) men underwent ~ 6% diet-induced weight gain. Abdominal subcutaneous adipocytes were sampled at baseline and post–weight gain for RT-qPCR assessment of GC-responsive transcripts ( FKBP5, TSC22D3/GILZ ) and related targets ( HSD11B1, HSD11B2, IL2, IL6 ). Metabolic responses were characterised using a standardised mixed-meal test with 5-hour profiles of glucose, insulin, C-peptide and triglycerides; hepatic triglyceride content was quantified by MRI. Results Weight gain reduced FKBP5 (− 23.65 ± 8.82% in White Europeans; −17.68 ± 11.62% in South Asians; P = 0.007) and GILZ (− 11.40 ± 2.83%; −5.95 ± 4.01%; P = 0.001 for change with weight gain), with no ethnicity×intervention interaction (P ≥ 0.26). HSD11B1/HSD11B2 and IL2/IL6 did not change. At baseline, FKBP5 and GILZ were associated with adiposity, liver fat and adipocyte size. Within-person ΔFKBP5 (post–weight gain minus baseline) correlated with Δpostprandial insulin (r = 0.46, P = 0.006) and ΔC-peptide (r = 0.34, P = 0.049). ΔGILZ correlated with Δfasting glucose (r = − 0.40, P = 0.017) and Δpostprandial insulin (r = 0.34, P = 0.049). In mixed-effects models, change in postprandial insulin remained an independent correlate of both transcripts. Conclusion Early modest weight gain downregulates adipocyte GC-responsive transcripts similarly across ancestries, and dynamic transcript changes track insulin exposure. These findings implicate insulin dynamics as a potential driver of adipocyte GC signalling adaptations during early weight gain, linking adipocyte transcriptional responses to clinically relevant postprandial insulin physiology.

South Asians (SAs) in the UK are at an increased risk of cardiovascular disease (CVD), develop type 2 diabetes mellitus at a lower age and body mass index, and have a lower high-density lipoprotein cholesterol (HDL-C) concentration than their white European (EU) counterparts. The failure of HDL-C raising therapies for CVD risk reduction has turned attention to its composition and function. A previous study comparing the effect of moderate weight gain on SA and EU men found baseline and weight gain-induced ethnic differences in body composition, adipocyte function and insulin resistance (ClinicalTrials.gov registration: NCT02399423). This study investigated differences in HDL protein composition, subclass distribution and in vitro vascular functions at baseline and after weight gain in the same cohort of men. HDL protein composition was determined by nano liquid chromatography tandem mass spectrometry using label-free quantification. HDL subclass distribution was measured by native gel electrophoresis. HDL in vitro paraoxonase-1 (PON-1) activity was measured by monitoring the PON-1 mediated hydrolysis of phenylacetate. In vitro HDL anti-inflammatory function was assessed in an endothelial cell assay of adhesion molecule inhibition. SAs had higher levels of immunity- and inflammation-related proteins and a detrimental profile of lipid metabolism-related proteins at baseline and with weight gain (including lower apolipoprotein (apo) A-IV and apoF and higher apoC-III) compared with EU. HDL subclass distribution and in vitro vascular function were not different between EUs and SAs. HDL protein composition reflects systemic physiology and acts as a mechanistic marker of impaired lipid metabolism in SAs.

Obesity is a worldwide health crisis and poses significant challenges in critical care. Many studies suggest an 'obesity paradox', in which obesity, defined by body mass index (BMI), is associated with better outcomes. However, the inability of BMI to discriminate between fat and muscle or between visceral adipose tissue and subcutaneous adipose tissue, limits its prediction of metabolic ill health. We suggest that the 'obesity paradox' may be more reflective of the limitations of BMI than the protective effect of obesity. We explore the biological processes leading to visceral fat accumulation, and the evidence linking it to outcomes in critical illness. In the 'spillover' hypothesis of adipose tissue expansion, caloric excess and impaired expansion of storage capacity in the subcutaneous adipose tissue lead to accumulation of visceral adipose tissue. This is associated with a chronic inflammatory state, which is integral to the link between visceral adiposity, type 2 diabetes mellitus, and ischaemic heart disease. We review the current evidence on visceral adiposity and critical illness outcomes. In COVID-19, increased visceral adipose tissue, irrespective of BMI, is associated with more severe disease. This is mirrored in acute pancreatitis, suggesting visceral adiposity is linked to poorer outcomes in some hyperinflammatory conditions. We suggest that visceral adiposity's chronic inflammatory state may potentiate acute inflammation in conditions such as COVID-19 and acute pancreatitis. Further work is required to investigate other critical illnesses, especially sepsis and acute respiratory distress syndrome, in which current evidence is scarce. This may give further insights into pathophysiology and inform tailored treatment and nutrition strategies based on body fat distribution.

Background/Objectives: A balanced nutritious diet is vital during pregnancy for both the mother and the baby. The aims of this longitudinal study were to (1) determine any differences in macro- and micronutrient intakes in a group of UK women during pregnancy (and in the post-partum period) who were overweight or obese (BMI mean (SD) 31.1 (2.9)) at antenatal booking appointment compared with women who were within the ideal BMI range (BMI mean (SD) 22.1 (1.9)) and (2) determine the proportion of women who met the Harmonized Average Requirements (H-AR) during pregnancy. Methods: Forty-two participants attended four clinic visits: three during pregnancy, one in each trimester (V1, V2, and V3), and one 12 weeks post-partum (V4). Dietary intake was assessed by 24 h diet recall and analysed using DietPlan6. Results: There were no differences in energy and macronutrient intakes between overweight/obese and lean women. During pregnancy, the overweight/obese women consumed a mean (SD) of 3238 (941) sodium (mg per day), which was approximately 10% higher compared to 2934 (732) sodium (mg per day) in the lean group (p = 0.015). Dietary and supplemental intakes of the sodium to potassium ratio was 21% higher in overweight/obese women compared to the lean women, p = 0.0031 (mean (SD) of 1.17 (0.35) versus 0.93 (0.28), respectively). Virtually all women did not meet the H-AR for niacin, folate, and vitamin D through dietary intake alone. Conclusions: The ‘eat better and not more’ message during pregnancy is supported.

South Asians (SAs) develop type 2 diabetes at lower body mass index values than white Europeans (WEs). This basic human experimental study aimed to compare the metabolic consequences of weight gain in SA and WE men without overweight or obesity. Fourteen SAs and 21 WEs had assessments of body composition, metabolic responses to mixed-meal ingestion, cardiorespiratory fitness and physical activity, and a subcutaneous abdominal adipose tissue biopsy, before and after 4-6 weeks of overfeeding to induce 5-7% weight gain. Here we show that body mass index and whole-body adipose tissue volume increases similarly between ethnic groups, but SAs gain less lean tissue. SAs experience a substantially greater decrease in insulin sensitivity compared with WEs (38% versus 7% decrease, P = 0.009), have fewer small (37.1% versus 60.0%, P = 0.003) and more large (26.2% versus 9.1%, P = 0.005) adipocytes at baseline and have a smaller decrease in very small adipocytes with weight gain (-0.1% versus -1.9%, P < 0.0001). Ethnic differences in adipocyte morphology are associated with SA's greater adverse metabolic changes with weight gain. ClinicalTrials.gov registration: NCT02399423 .

In obesity, failure of appropriate adipocyte expansion leads to ectopic fat deposition and systemic insulin resistance. Maternal obesity leads to higher incidence of gestational diabetes mellitus (GDM) and pre-eclampsia (PE), however adipocyte expansion is poorly characterised in pregnancy. This study aimed to compare adipocyte size, adipose tissue (AT) fibrosis, mRNA expression and insulin signaling in third trimester subcutaneous AT (SAT) and visceral AT (VAT) biopsies collected at Caesarean section from non-labouring GDM, PE and BMI- and age-matched healthy normotensive pregnant women (NT). SAT adipocyte diameter did not differ in complicated pregnancies compared to NT, however VAT diameter was higher in GDM (NT (n=30): 66.1±12.4 µm, GDM (n=30): 77.6±11.7 µm, p=0.001) but not PE (NT (n=7): 69.8±11.2 µm, PE (n=7) 72.4±8.9 µm, p=0.65). There was VAT hypertrophy in GDM (NT: 8.7±14.1% vs. GDM: 22.0±21.9% very large adipocytes, p=0.002) but not PE (NT: 13.7±21.3% vs. PE: 16.0±15.4%, p=0.67). Total fibrosis, assessed by picrosirius red staining of paraffin-embedded AT, was lower in GDM (NT (n=19): 45±13% area, GDM (n=12): 22±8%, PE (n=18): 45±12%, p<0.001); whereas pericellular fibrosis was higher in GDM (NT (n=18): 16±7%, GDM (n=12): 23±7%, PE (n=16): 15±4%, p=0.02). Higher HIF1A mRNA expression relative to PPIA was observed in GDM VAT (NT (n=11): 18.1±3.9%, GDM (n=11): 46.4±29.8%, PE (n=10): 25.3±3.6%, p=0.03). Lower insulin-stimulated phosphorylation of Akt was observed in GDM and PE SAT compared to NT by immunoblotting, yet did not reach significance (fold change in phospho-Akt:Akt NT (n=9): 43.1±70.1, GDM (n=7): 10.5±35.4, PE (n=5): 5.8±26.9 (p=0.19). These data suggest AT dysfunction in GDM with a phenotype typical of T2DM, whereas AT dysfunction in PE requires further elucidation. Disclosure Z. Lees: None. I. Salt: None. D.J. Freeman: None. Funding The University of Glasgow MVLS Doctoral Training Programme.

Given the failure of high-density lipoprotein (HDL) raising therapies to reduce cardiovascular disease risk, attention has turned towards HDL composition and vascular protective functions. In individuals with insulin resistance, exercise interventions recover HDL function. However, the effect of exercise on HDL in otherwise healthy individuals is unknown. This cross-sectional study aimed to measure HDL composition and antioxidant/endothelial anti-inflammatory function in insulin sensitive endurance athlete and healthy control men. HDL was isolated using density gradient ultracentrifugation. HDL composition was measured using microplate assays for apolipoprotein A-I, total cholesterol content and apolipoprotein M. HDL protein composition was measured using nano-liquid chromatography tandem mass spectrometry. HDL subclass distribution was measured by native gel electrophoresis. HDL in vitro antioxidant function was measured by paraoxonase-1 activity assay and anti-inflammatory function assessed in endothelial cells. Compared with controls, endurance athlete HDL had higher apolipoprotein A-1 (1.65 ± 0.62 mg/ml vs 1.21 ± 0.34 mg/ml, P=0.028) and higher total cholesterol content (2.09 ± 0.44 mmol/L vs 1.54 ± 0.33 mmol/L, P<0.001). Proteomics revealed higher apolipoprotein A-II, A-IV and D and transthyretin in endurance athlete HDL versus controls. There was no difference observed in in vitro HDL antioxidant or anti-inflammatory functions between controls and endurance athletes. Despite a more favourable composition, endurance athlete HDL did not have higher in vitro antioxidant or anti-inflammatory function. It is possible that HDL has a ceiling of function, i.e. that healthy HDL function cannot be enhanced by endurance exercise.

Pregnant women who develop pre-eclampsia (PE) and/or intra-uterine growth restriction (IUGR) have reduced polyunsaturated fatty acid (PUFA) status compared to healthy pregnancy(1). It is unknown if pregnant women diagnosed with Gestational Diabetes Mellitus (GDM), and their offspring, also have compromised PUFA status. To determine if women with GDM, and their offspring, have altered PUFA status compared to healthy pregnancy. Pregnant women were recruited from Glasgow Scotland, and Brisbane, Australia from antenatal clinics for this cross sectional study. Third trimester maternal blood samples were collected after an overnight fast and cord blood samples were collected at delivery. Plasma fatty acids were analysed using gas chromatography from women with GDM (n = 37) and healthy pregnancies (n = 27) and their respective offspring (n = 31, from women with GDM, and n = 27 from healthy women). T-tests were used to determine significant differences between maternal with GDM and healthy pregnancy, as well as for their offspring and significance was set at p<0.05. Previously, erythrocyte fatty acids were analysed from women with PE (n = 21), IUGR (n = 13) and healthy pregnancies (n = 86)(1). All results were expressed as mol percent of total fatty acids. There were no differences in maternal plasma arachidonic acid (4.51 ± 1.23 vs. 4.72 ± 0.64, p = 0.39) and plasma EPA & DHA (2.33 ± 0.74 vs 2.69 ± 1.04, p = 0.14) in women with GDM and healthy pregnancies, respectively. There were no differences in fetal plasma arachidonic acid (11.58 ± 2.26 vs. 12.63 ± 1.69, p = 0.08) and plasma EPA & DHA (4.44 ± 1.17 vs. 4.44 ± 1.00, p = 0.89) in offspring from women with GDM and healthy pregnancies, respectively. Women with PE and IUGR had approximately 25% lower erythrocyte EPA & DHA and 35% lower erythrocyte arachidonic acid compared to healthy pregnant(1). Offspring from women with PE and IUGR had approximately 25% lower erythrocyte EPA & DHA and 22% lower erythrocyte arachidonic acid compared to healthy pregnancy(1).Women with PE and IUGR had lower PUFA status likely due to reduced PUFA synthesis(1) and offspring from women with PE and IUGR had reduced PUFA status likely due to ectopic fat in placenta tissue(2). Women with GDM do not have compromised PUFA status suggesting there is no reduced synthesis and transport of PUFA. Offspring from women with GDM do not have reduced PUFA status suggesting there is no problem with PUFA transport across the placenta, unlike offspring from women with PE and/or IUGR. Women with GDM, and their offspring, do not have compromised plasma PUFA status compared to healthy pregnancy.

Women with gestational diabetes mellitus controlled for their plasma glucose levels, exhibit dyslipidaemia that may contribute to offspring obesity and the risk of future gestational diabetes mellitus - Volume 82 Issue OCE2

AIMS:To compare maternal and fetal cord plasma and lipoprotein triglyceride (TG) concentrations in women with Gestational Diabetes Mellitus (GDM), with hyperglycaemia and hypertriglyceridaemia, and healthy women. METHODS:Fasted maternal blood at 28.6 ± 3.4 (T1) and 36.2 ± 1.0 (T2) [mean ± S.D] weeks of gestation, and cord blood were collected. Plasma lipoprotein fractions underwent compositional analysis. RESULTS:Plasma glucose did not differ between GDM and healthy women. T1 maternal plasma TG (2.60 ± 0.89 mmol/l versus 1.71 ± 0.69 mmol/l) and plasma apolipoprotein B (1.30 ± 0.48 g/l versus 0.75 ± 0.40 g/l) were higher in GDM compared to healthy. Maternal plasma TG increased over gestation in both groups. T1 plasma VLDL total protein (38 ± 15 mg/dl versus 25 ± 11 mg/dl), total cholesterol (TC) (30 ± 14 mg/dl versus 16 ± 13 mg/dl) and phospholipid (PL) (43 ± 17 mg/dl versus 26 ± 16 mg/dl) were higher in GDM than healthy, and similarly for IDL, suggesting increased lipoprotein particle number. T1 VLDL-TG enrichment was higher in healthy and increased over gestation in GDM women but decreased in healthy. IDL-TG enrichment (TG/TC) increased over gestation in women with GDM and decreased in healthy. Cord blood VLDL, IDL and LDL from GDM had a two-fold higher TG enrichment than healthy pregnancy. CONCLUSION:Increased maternal lipoprotein number, but not TG enrichment, in GDM mothers may explain TG enrichment of cord lipoproteins.

The roles of DGAT1 and DGAT2 in lipid metabolism and insulin responsiveness of human skeletal muscle were studied using cryosections and myotubes prepared from muscle biopsies from control, athlete, and impaired glucose regulation (IGR) cohorts of men. The previously observed increases in intramuscular triacylglycerol (IMTG) in athletes and IGR were shown to be related to an increase in lipid droplet (LD) area in type I fibers in athletes but, conversely, in type II fibers in IGR subjects. Specific inhibition of both diacylglycerol acyltransferase (DGAT) 1 and 2 decreased fatty acid (FA) uptake by myotubes, whereas only DGAT2 inhibition also decreased fatty acid oxidation. Fatty acid uptake in myotubes was negatively correlated with the lactate thresholds of the respective donors. DGAT2 inhibition lowered acetate uptake and oxidation in myotubes from all cohorts whereas DGAT1 inhibition had no effect. A positive correlation between acetate oxidation in myotubes and resting metabolic rate (RMR) from fatty acid oxidation in vivo was observed. Myotubes from athletes and IGR had higher rates of de novo lipogenesis from acetate that were normalized by DGAT2 inhibition. Moreover, DGAT2 inhibition in myotubes also resulted in increased insulin-induced Akt phosphorylation. The differential effects of DGAT1 and DGAT2 inhibition suggest that the specialized role of DGAT2 in esterifying nascent diacylglycerols and de novo synthesized FA is associated with synthesis of a pool of triacylglycerol, which upon hydrolysis results in effectors that promote mitochondrial fatty acid oxidation but decrease insulin signaling in skeletal muscle cells.

Gestational diabetes mellitus (GDM) is a common disorder of pregnancy with short- and long-term consequences for mother and baby. Pre-eclampsia is of major concern to obstetricians due to its sudden onset and increased morbidity and mortality for mother and baby. The incidence of these conditions continues to increase due to widespread maternal obesity. Maternal obesity is a risk factor for GDM and pre-eclampsia, yet our understanding of the role of adipose tissue and adipocyte biology in their aetiology is very limited. In this article, available data on adipose tissue and adipocyte function in healthy and obese pregnancy and how these are altered in GDM and pre-eclampsia are reviewed. Using our understanding of adipose tissue and adipocyte biology in non-pregnant populations, a role for underlying adipocyte dysfunction in the pathological pathways of these conditions is discussed.

Objective: High-density lipoprotein (HDL) has well-established anti-atherosclerotic properties in the non-pregnant population. HDL may have a potential role in protecting maternal vascular function during pregnancy that fails to occur in preeclampsia. HDL function is determined by its protein and lipid composition. This study aimed to examine protein composition of HDL at different gestations throughout healthy pregnancy and to compare HDL proteome in the third trimester of healthy pregnancy and preeclampsia. Design and method: HDL was isolated from two cohorts of n = 10 healthy pregnant women at 8 different timepoints throughout pregnancy; pre-pregnancy, week 4.6, week 6.1, week 8.4, (cohort 1) and week 15, week 25, week 35 and 3 months post-natal (cohort 2), and from healthy pregnant (n = 10) and preeclampsia (n = 6) women at the third trimester. The proteome of HDL was measured by nano liquid chromatography coupled to tandem mass spectrometry (nLC-MS/MS). Results: There were 16 proteins in HDL that showed significant differences throughout healthy pregnancy which were apolipoprotein A-IV, apolipoprotein C-II, apolipoprotein C-III, apolipoprotein C-IV, apolipoprotein F, apolipoprotein L-I, angiotensinogen, alpha-1-acid glycoprotein 2, alpha-1-antitrypsin, cathelicidin antimicrobial peptide, fibrinogen alpha chain, fibrinogen beta chain, prenylcysteine oxidase 1, serum amyloid a-I, serum paraoxonase/lactonase 3 and vitronectin. In preeclampsia, HDL apolipoprotein A-I was lower than healthy pregnancy, while HDL alpha-1-antitrypsin, pigment epithelium-derived factor and vitamin D-binding protein were higher. Conclusions: There were significant differences in the protein composition of HDL during pregnancy and between healthy pregnancy and preeclampsia in the third trimester. These compositional differences may be involved in HDL function and may indicate HDL's role in vascular protection in pregnancy.

Plasma high density lipoprotein (HDL) exhibits many functions that render it an effective endothelial protective agent and may underlie its potential role in protecting the maternal vascular endothelium during pregnancy. In non-pregnant individuals, the HDL lipidome is altered in metabolic disease compared to healthy individuals and is linked to reduced cholesterol efflux, an effect that can be reversed by lifestyle management. Specific sphingolipids such as sphingosine-1-phosphate (S1P) have been shown to mediate the vaso-dilatory effects of plasma HDL via interaction with the endothelial nitric oxide synthase pathway. This review describes the relationship between plasma HDL and vascular function during healthy pregnancy and details how this is lost in pre-eclampsia, a disorder of pregnancy associated with widespread endothelial dysfunction. Evidence of a role for HDL sphingolipids, in particular S1P and ceramide, in cardiovascular disease and in healthy pregnancy and pre-eclampsia is discussed. Available data suggest that HDL-S1P and HDL-ceramide can mediate vascular protection in healthy pregnancy but not in preeclampsia. HDL sphingolipids thus are of potential importance in the healthy maternal adaptation to pregnancy.

High-density lipoprotein (HDL) is a circulating complex of lipids and proteins known primarily for its role in reverse cholesterol transport and consequent protection from atheroma. In spite of this, therapies aimed at increasing HDL concentration do not reduce the risk of cardiovascular disease (CVD), and as such focus has shifted towards other HDL functions protective of vascular health - including vasodilatory, anti-inflammatory, antioxidant and anti-thrombotic actions. It has been demonstrated that in disease states such as CVD and conditions of insulin resistance such as Type 2 diabetes mellitus (T2DM), HDL function is impaired owing to changes in the abundance and function of HDL-associated lipids and proteins, resulting in reduced vascular protection. However, the gold standard density ultracentrifugation technique used in the isolation of HDL also co-isolates extracellular vesicles (EVs). EVs are ubiquitous cell-derived particles with lipid bilayers that carry a number of lipids, proteins and DNA/RNA/miRNAs involved in cell-to-cell communication. EVs transfer their bioactive load through interaction with cell surface receptors, membrane fusion and endocytic pathways, and have been implicated in both cardiovascular and metabolic diseases - both as protective and pathogenic mediators. Given that studies using density ultracentrifugation to isolate HDL also co-isolate EVs, biological effects attributed to HDL may be confounded by EVs. We hypothesise that some of HDL's vascular protective functions in cardiovascular and metabolic disease may be mediated by EVs. Elucidating the contribution of EVs to HDL functions will provide better understanding of vascular protection and function in conditions of insulin resistance and potentially provide novel therapeutic targets for such diseases.