Postprandial lipemia is an independent risk factor for CVD, due to effects on lipoprotein remodelling, oxidative stress, inflammation, haemostasis and endothelial dysfunction. However, it is unknown whether the total, peak or duration of the lipemic response determines risk. The PREDICT I study is the largest study to date to measure postprandial lipemic responses and intermediary acutely changing cardiometabolic risk factors at multiple time points using a standardized test meal model.

A multi-center postprandial study of 1,000 individuals from the UK (unrelated, identical and non-identical twins) and 100 unrelated individuals from the US, assessed postprandial (hourly 0–6h) metabolic responses to sequential mixed-nutrient dietary challenges (50 g fat, 85 g carbohydrate at 0 h; 22 g fat, 71 g carbohydrate at 4h) in a clinic setting. We investigated the relationship of different postprandial triacylglycerol (TG) measures (4 and 6 h TG iAUC, 4 and 6 h TG concentration, 4 and 6 h TG increase from fasting) with lipoprotein remodelling (XXL-VLDL (including chylomicron remnants and VLDL particles) and XL-VLDL particle concentrations (average diameters > 75, 64 nm respectively), HDL-C) and levels of glycosylated acute phase proteins (GlycA; marker of cardiovascular inflammation), all of which have been implicated as independent predictors of CVD risk.

Following adjustment (for use of medication, demographic characteristics, fasting TG, insulin and glucose levels), all six postprandial TG measures (4 and 6 h TG iAUC, 4 and 6 h TG concentration, 4 and 6 h TG increase from fasting) were strongly correlated with markers of atherogenic lipoprotein remodelling and the marker of cardiovascular inflammation (GlycA). The strongest correlation (interim analysis) was observed for the 6 h TG increase from fasting (all P < 0.001, Pearson's coefficient r = 0.94 [95%CI's; 0.93, 0.95] for XXL-VLDL-P; r = 0.95 [95%CI's; 0.95, 0.96] for XL-VLDL-P; r = 0.89 [95%CI's; 0.88, 0.91] for GlycA ; r = -0.61 [95%CI's; -0.66, -0.55] for HDL-C). Inter-individual variability in postprandial lipemic responses was high in the tightly controlled clinic setting (interim analysis, n = 656); IQR (median) was; iAUC (0–6h) 2.39 (2.31) mmol/L.h; Cmax 1.32 (2.06) mmol/L; Tmax 30.0 (300) min; and increase above fasting at 6 h 0.78 (0.62) mmol/L.

This is the most detailed postprandial study performed to date and suggests that identifying predictors of the postprandial 6 h TG rise will have the highest CVD relevance. Ongoing exploration in PREDICT I of the determinants of postprandial lipemic responses considering environmental, genetic and microbiome variables will significantly advance our ability to predict an individual's postprandial response and its links to cardiovascular risk.

Our recent study (1) showed that the amount of dietary carbohydrates in obesity interventions has differential effects on cardiovascular risk markers (CVM) and effects magnitude depends on intervention duration. Very-low carbohydrate high-fat diets (VLCD) were superior in ameliorating lipid markers compared to high-carbohydrate low-fat diets (LFD).

We updated our systematic review and meta-analysis to include long-term effects of VLCD (< 50 g /day) on weight, glucose, total cholesterol, insulin and blood pressure (BP) among overweight/obese adults in comparison to LFD.

Medline, PubMed, Cochrane Central, and CINAHLPlus were searched to identify large (n > 100) randomised controlled trials (RCT) with duration ≥ 6 months. Risk of bias, a random effects model and subgroup analyses based on duration of follow-up were performed using Review Manager. Results were reported according to PRISMA.

Four open label RCT (n = 723; 362 VLCD; 361 LFD) with some form of behavioral intervention and duration 6–24 months were identified. VLCD showed more favorable effects on diastolic BP at 6 months (-1.96; 95%CI, -2.99 to 00.93; P = 0.0002) and 24 months (-2.69; 95%CI, -4.87 to -0.51; P = 0.001), near significant level at 12 months (-1.79; 95%CI, -3.56 to 0.04; P = 0.05) and an overall total favourable effect (-1.98; 95%CI, -2.73 to -1.22). The decrease in systolic BP was greater among VLCD for the whole period and the overall total effect reached the level of significance (-1.76; 95%CI, -3.56 to 0.04; P = 0.05). VLCD showed beneficial effect on total cholesterol level at 6 and 12 months (-0.01 mmol/L; 95%CI, -0.01 to –0.00; P = 0.002 and -0.01 mmol/L; 95%CI, -0.01 to –0.00; P = 0.005, respectively). The mean changes in weight, and fasting glucose and insulin levels revealed non-significant differences between both diets at any measured time, although these parameters decreased within both groups compared to baseline.

VLCD led to significant total weighted mean decrease of diastolic BP and near significant decrease of systolic BP independent of changes in body weight, fasting glucose or insulin levels. The present data on decreased levels of diastolic BP and total cholesterol, combined with our recently published results on increased HDL-cholesterol, decreased triglycerides and no significant effect on LDL-cholesterol (1) provide evidence that VLCD are superior to LFD in improving traditional CVM in longer term.

Semen quality is a key indicator of male fecundity and therefore is closely related to male fertility. In recent decades a significant decrease of semen parameters was observed worldwide. Except for environmental pollution, low physical activity, body weight and comorbidities, unhealthy diet was linked to semen quality decline. Mediterranean Diet (MedDiet) due to its anti-oxidative and anti-inflammatory properties, is considered as beneficial for semen. However, results of existing studies are inconclusive. The aim of this review was to pool associations between adherence to the MedDiet and semen quality parameters in observational studies. Electronic databases including PubMed, Embase and Scopus were searched until April 2019 with no restriction to language or time. Studies were eligible for inclusion if they filled following criteria: (1) observational design (retrospective or prospective), (2) adherence to MedDiet assessed using score or data-driven method and (3) semen quality parameters (total motility, progressive motility, total count, concentration or volume) as outcomes. Mean differences or odds ratios of abnormal semen parameters between the highest and lowest quintiles of MedDiet were converted to standardized mean differences (SMD). Random effects models with inverse variance method were used to pool differences. Heterogeneity was assessed using the Cochrane Q test and I2 statistic with value > 50% indicating substantial heterogeneity. Five cross-sectional studies extracted from 840 articles, met the eligibility criteria and were included in the meta-analysis. Higher adherence to MedDiet compared to lower adherence was associated with increased total (SMD = 0.59 95%CI: 0.12–1.06, I2 = 80%, pheterogenity < 0.01, n = 4) and progressive (SMD = 0.53 95%CI: 0.16–0.91, I2 = 0%, pheterogenity = 0.94, n = 2) sperm motility. No associations were observed between adherence to MedDiet and total sperm count (SMD = 0.14 95%CI: -0.20–0.48, I2 = 70%, pheterogenity < 0.01, n = 5), concentration (SMD = 0.22 95%CI: -0.04–0.47, I2 = 48%, pheterogenity = 0.10, n = 5), volume (SMD = 0.02 95%CI: -0.20–0.25, I2 = 21%, pheterogenity = 0.28, n = 5), as well as morphology (SMD = -0.10 95%CI: -0.36–0.16, I2 = 38%, pheterogenity = 0.18, n = 4). The results suggest that higher adherence to the MedDiet may have a protective effect on semen quality, probably by increased intake of antioxidant-rich products. Limitations of findings include substantial between-study heterogeneity and use of cross-sectional design. Beneficial effects of the MedDiet on semen quality have to be confirmed in further prospective studies.

The gut epithelium is a protective interface between the external environment and the human body. This epithelium interacts with a multitude of internal stimuli from the bloodstream and immune cells, and luminal stimuli from microorganisms and the diet. Disruptions to the epithelium are seen in inflammatory bowel diseases and coeliac disease. The human adenocarcinoma cell line (Caco-2) is an in vitro model used to assess the interactions between nutrients and gut epithelium. Long-chain omega-3 (n-3) polyunsaturated fatty acids (PUFAs), such as eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA), have anti-inflammatory effects via the production of anti-inflammatory eicosanoids, interactions with immune cells and reductions in pro-inflammatory cytokines and chemokines. The aim of this study is to assess the anti-inflammatory properties of DHA and EPA in stimulated Caco-2 monolayers. Caco-2 cells were seeded at 70,000 cells/cm2 and grown to confluence before being allowed to fully differentiate (approx. 21 days total). Cytokines (TNF-α, IFN-γ, and IL-1β) and peptic-tryptic (PT-) gliadin were used as inflammatory stimulants. EPA and DHA incubations occurred 48 hours pre-stimulation. Tight junction function and morphology was determined using trans-epithelial electrical resistance measurements and confocal microscopy. Inflammatory markers, including IL-6, IL-8, and IL-17, were assessed by multiplex. Stimulatory cytokines induced tight junction dysfunction and increased pro-inflammatory mediator production in Caco-2 cells. PT-gliadin, DHA and EPA treatment did not alter paracellular permeability or stimulant-induced production of pro-inflammatory mediators. Further investigation of the inflammatory role of n-3 PUFAs and PT-gliadin in the Caco-2 model is required. Future work will assess the composition of PT-gliadin by electrophoresis and whether co-incubation of n-3 PUFAs and inflammatory cytokines will alter paracellular permeability and mediator output of Caco-2 cells.

Breast cancer treatment and prognosis is informed by biomarker expression. Expression of Oestrogen Receptor-alpha (ERα) for example influences whether the patient receives endocrine- or chemo-therapy. Nutritional status is a modifier of disease free survival and elevated circulating cholesterol associates with increased risk of relapse. Cholesterol hydroxylation produces ‘oxysterols’ which are selective Liver X Receptor alpha (LXRα) modulators and ERα agonists. In ER-positive breast cancer, oxysterols induce proliferation and resistance to endocrine therapy, whilst in ER-negative disease oxysterols are anti-proliferative and pro-metastatic suggesting that there are breast cancer subtype specific differences in the genomic targets of the oxysterol-LXR pathway. This study explored the regulation of LXRα signalling in ER-positive and ER-negative breast cancer, and how ligand, receptor and co-factors combine to regulate LXRα target gene expression in different breast cancer types.

In vitro, MDA.MB.468 (ER-negative) cells were more responsive than MCF-7 (ER-positive) cells to synthetic LXRα agonists (T0901317, GW3965) and six oxysterols (22-hydroxycholesterol [22-OHC], 24-OHC, 25-OHC, 27-OHC, 7-ketocholesterol and 24,25-epoxycholesterol), as measured by MTT, LXR-luciferase reporter, and qPCR of canonical targets ABCA1 and APOE (Students t-tests: p < 0.01). Responses to the antagonist GSK2033 was comparable across cell lines. In vivo, LXRa expression correlated with 48/146 target genes in ER-negative (n = 81), but with just 9/146 in ER-positive tumours (n = 234) (Fischer exact test: p < 0.0001) indicating greater LXRa-mediated transcription of target genes in the aggressive subtype. This was not explained by ligand concentration, as we developed a novel fast oxysterol detection system and found no difference in concentration of 22-OHC, 24-OHC, 25-OHC or 27-OHC between ER-negative (n = 11) and ER-positive (n = 11) primary tumours obtained from the Leeds Breast Tissue Bank. However, we did observe that expression of LXRa and 2/7 of its co-activators (SRC, TRRAP) were higher in ER-negative relative to ER-positive disease (using TCGA data from cBioPortal) (Mann-Whitney U test: p < 0.001), and that expression of all LXRa co-repressors were lowest in ER-negative disease (NCOR1, NCOR2, LCOR: Mann-Whitney U test: p < 0.001 for all). siRNA knock-down of NCOR1 and NCOR2 resulted in MCF-7 cells that mimicked the response of MDA.MB.468 cells to oxysterols (as measured by LXR-luciferase and qPCR assay).

These data indicate that despite the anti-proliferative actions of oxysterol-LXRa signalling, there is a, yet to be identified, selective advantage for retention and enhancement of this pathway in ER-negative breast cancer. Dietary routes to selective LXRa modulation (such as plant sterols) may provide patient-led routes to improving ER-negative survival rates.

Glycemic, insulinemic and lipemic postprandial responses are multi-factorial and contribute to diabetes, obesity and CVD. The aim of the PREDICT I study is to assess the genetic, metabolic, metagenomic, and meal-context contribution to postprandial responses, integrating the metabolic burden and gut microbiome to predict individual responses to food using a machine learning algorithm.

A multi-center postprandial study of 1,000 individuals from the UK (unrelated, identical and non-identical twins) and 100 unrelated individuals from the US, assessed postprandial (0–6h) metabolic responses to sequential mixed-nutrient dietary challenges (50 g fat and 85 g carbohydrate at 0 h; 22 g fat and 71 g carbohydrate at 4h) in a clinic setting. Glycemic responses to 5 duplicate isocaloric meals of different macronutrient content and self-selected meals (> 100,000), were tested at home using a continuous glucose monitor (CGM). Baseline factors included metabolomics, genomics, gut metagenomics and body composition. Genetic contributions to postprandial responses were determined by classical twin methods.

Inter-individual variability in postprandial responses (glucose, insulin and triacylglycerol (TG)) was high in the clinic setting: iAUC IQR (median) was (n = 644); glucose (0–2h) 1.97 (1.89) mmol/L.h, insulin (0–2h) 45.6 (67.7) mIU/L.h and TG (0–6h) 2.37 (2.42) mmol/L.h. The unadjusted genetic contribution for glucose, insulin and TG responses were 54%, 29% and 27% respectively. Within-individual concordance (ICC) in glucose responses (iAUC 0–2h) for at home duplicate isocaloric meals was moderate-to-high, depending on the test meal: ICC (95%CI) was; high carbohydrate 0.62 (0.58,0.66), (carbohydrate = 95g/76% energy; n = 764), average lunch 0.57 (0.53, 0.62) (carbohydrate = 68g/54% energy; n = 763), OGTT 0.65 (0.61,0.70) (carbohydrate = 75 g; n = 754), high fat 0.35 (0.28, 0.41) (fat = 40g/71% energy; n = 576) and high protein 0.56, (0.48,0.62) (protein = 41g/32% energy; n = 364). An interim machine learning algorithm predicted 46% of the variation in glycemic responses based on meal content, meal context and participant's baseline characteristics, excluding genetic and microbiome features. Only 29% of variation could be explained by the macronutrient content of the meal.

This is the most comprehensive postprandial study performed to date. The large and modifiable variation in metabolic responses to identical meals in healthy people explains why ‘one size fits all’ nutritional guidelines are problematic. The genetic component to these responses is moderate, leaving the majority of the variation potentially modifiable. By collecting information on glucose responses to > 100,000 meals, alongside environmental, genetic and microbiome variables, we will have excellent power to use machine learning to optimise and predict individual responses to foods.