Adherence to the Mediterranean dietary pattern (MeDiet) and adiposity, respectively, decreases and increases the risk of multiple common age-related diseases through several mechanisms including inflammation, reactive oxygen species (ROS) production in the mitochondria, and DNA methylation. For example, adverse changes in platelets from obese and overweight adults include hyperaggregability and increased ROS. Since platelets are anuclear, their prothrombotic function is fully orchestrated by the mitochondria and the only DNA present is the mitochondrial DNA (mtDNA). In this study, we tested the hypothesis that MeDiet influences patterns of mtDNA methylation in platelets from older adults with greater adiposity.

We selected 134 participants with overweight or obesity (mean BMI = 35.5 ± 5.1 and age = 62 ± 10 years) from the “Susceptibility to particle health effects, miRNA and exosomes”(SPEHRE) Study. Dietary intake was assessed using a food frequency questionnaire and MeDiet adherence was calculated using the MeDiet Score described by Martínez-González et al.(2012). MtDNA was extracted from platelets, linearized, bisulfite converted and DNA methylation was quantified by pyrosequencing at 13 CpG in seven genes that encode for tRNAs (MT-TF and MT-TL1), regulatory regions (D-Loop and MT-OLR), and subunits of the electron-transport-chain (MT-CO1, MT-CO2, and MT-CO3).

In these participants, MeDiet score ranged from 3 to 12 (mean = 6.5), with higher scores reflecting greater MeDiet adherence. Regression analysis showed that higher MeDiet score was associated with lower D-loop (β = -0.031, P = 0.019) and higher MT-CO2 CpG1 (β = 0.040, P = 0.023) methylation. No associations were found between MeDiet Score and methylation level at MT-CO1(2 CpGs), MT-CO2(CpG2), MT-CO3(2 CpGs), MT-TL1(2 CpGs), MT-TF(CpG1), MT-OLR(3 CpGs).In addition, there was no association between mtDNA methylation and BMI.

The D-loop is critical for mitochondrial function since it initiates mtDNA replication and transcription. Increased D-loop methylation has been associated with reduced mitochondrial functionality, and insulin resistance. Our results suggest that higher adherence to MeDiet lowers D-loop methylation which may protect against obesity-related comorbidities (e.g. insulin resistance).

Higher MeDiet scores are associated with MT-CO2 CpG1 hypermethylation. MT-CO2 encodes for a subunit of the Cytochrome-C-oxidase, a highly regulated enzyme involved in the oxidative metabolism. MT-CO2 demethylation, induced by Valproic-Acid administration, has been reported to be associated with increased ROS production. Our results suggest a possible role of MeDiet in mitochondrial ROS regulation via methylation of MT-CO2.

For the first time, we observed associations between MeDiet adherence and mtDNA methylation. Validation of these findings in independent cohorts is required.

Dietary components are contributing factors in the development of Nonalcoholic fatty liver disease (NAFLD). The glycaemic index (GI), glycaemic load (GL) and total antioxidant capacity (TAC) have been considered potential dietary tools influencing diet–disease relationships. The aim of this study was to evaluate associations of the dietary GI, GL, TAC and insulin resistance (IR) condition with hepatic fat in NAFLD adults.

Material and methods: 112 overweight/obese adults with NAFLD (age: 50.8 ± 9 years old) were included in the trial. Dietary intake was assessed by a validated 137-item food frequency questionnaire (FFQ). Anthropometric, glycemic and lipid profiles, fatty liver quantification by magnetic resonance imaging (MRI) and IR measured by the Homeostatic Model Assessment of IR (HOMA-IR) were assessed at baseline. This study was registered as FLiO: Fatty Liver in Obesity study; NCT03183193.

Results: The median of liver fat content by MRI was 6.4 (3.8–10.9) in the recruited population. Participants with higher liver fat content showed significantly increased values of glucose, insulin, HbA1c and HOMA-IR than those with lower liver fat content (p < 0.05). Correlation analyses revealed relevant positive associations of hepatic fat with GI (r = 0.17; p = 0.077) and GL (r = 0.19; p = 0.047). Also, a negative association between liver fat content and TAC (r = -0.22; p = 0.023) was found. Linear regression analyses were used to examine the associations of hepatic fat and dietary quality indicators as well as IR adjusted for potential confounders (sex, age and physical activity). The final models showed that HOMA-IR, GI, GL and TAC were able to explain between 22.4 and 22.8 % (p < 0.001) of the variability of liver fat content.

The pathophysiology of NAFLD is thought to be associated with dietary determinants that contribute to metabolic dysregulation such as IR, ectopic liver fat deposition and hepatic damage. In accordance with other authors, we suggest that monitoring GI, GL and TAC may be useful approaches for the dietary treatment of NAFLD since they are related to hepatic fat. Additionally, it is important to highlight the essential role of IR in NAFLD as a key mediator in the development of NAFLD. Certainly, findings of the present study revealed a significant association of hepatic fat accumulation and IR.

In summary, GI, GL and TAC are potential markers of diet quality, with an impact on susceptible population at hepatic risk.

Determinants of health and diseases in humans involve complex interactions between diet, gut microbiota and host metabolism. The liver is a major organ that coordinates host adaptations to environmental factors. Mitochondria and endoplasmic reticulum tightly regulate liver nutrient sensing and metabolic adaptations, shaping its metabolic flexibility. Both organelles interact through contact points (named MAMs) in order to exchange calcium and lipids, and MAMs’ integrity settles metabolic flexibility of the liver. Disrupted MAMs’ integrity in obesity is linked to liver steatosis and insulin resistance. This occurs in association with alterations circulating amino acid profiles, among which glycine. Meta-analyses showed that the low plasma glycine concentration observed in obesity is a predictive factor for developing type 2 diabetes. We thus aimed at exploring whether glycine could participate to the regulation of the hepatic response to insulin and whether mechanisms may involve regulation of mitochondrial-ER (MAMs) interactions.

The study was carried out in 12 week-old male C57B16J mice fed a standard chow (n = 12/group, 36 total) in accordance with the French guidelines for the care and use of animals. Glycine (1.2g/kg) was provided for 3 days in drinking water vs. water vs. isonitrogenous placebo amino acids. Liver was collected after an overnight fast, 15 min after ip injection of saline (n = 6/group) or insulin (0.75U/kg, n = 6/group). Liver samples were fixed in glutaraldehyde/cacodylate for quantifying MAMs by transmission electronic microscopy. Fractions enriched in MAMs were isolated by differential ultracentrifugation on fresh tissue for Western Blot explorations. Insulin response was assessed through Akt phosphorylation at Ser473.

Glycine supplementation increased the percentage of interaction between ER and mitochondria (compare to mitochondrial surface) for all spacing from 10 to 50 nm. Thus, glycine supplementation induced a 45% increase in mitochondrial-ER (MAMs) interactions compared to other groups (p < 0.001). In agreement to the improved MAMs’ integrity, insulin response was enhanced by 50% after glycine supplementation compared to other groups (p < 0.05). Finally, protein analysis of the MAMs' fraction corroborates the importance of PP2A as a key enzyme regulating MAMs’ integrity.

Glycine is therefore an interesting nutritional actor that can play a crucial role in regulating the integrity of MAMs in the liver, thus contributing to the control of the insulin response.

In a randomized crossover study, we challenged 19 NW and 18 OB men with 500, 1’000, and 1’500 kcal HFM. The blood cell transcriptome was measured before and 2, 4, and 6 h after meal ingestion. The untargeted serum metabolome was assessed at the same time points. The genes and metabolites responding postprandially in a dose-dependant manner were identified.

Among the 1’385 features with a postprandial response, 178 showed a significant dose-response between meals. A majority of them increased (amino acids, bile salt conjugates, metabolites in urea cycle), whereas circa one fifth, indicative of an endogenous response, decreased (e.g. carnitine esters). Overall, the postprandial metabolome showed caloric saturation, the metabolome after 1’500 kcal HFM being quantitatively only marginally different than after 1’000 kcal. Although a subset of metabolites responded differently to the caloric increase, the postprandial metabolomes of NW and OB subjects were, overall, mostly characterized by their similarity. This finding contrasted with the blood cell transcriptome, which identified a subset of OB subjects whose expression of genes of the oxidative phosphorylation pathway was, compared to the NW subjects, more increased after ingestion of 1’500 kcal HFM.

We report, for the first time, the response of the human serum metabolome to increasing caloric doses of a meal. The saturation observed between 1’000 and 1’500 kcal reflects on the control mechanisms regulating the metabolites entering the blood circulation. Interestingly, combining the analysis of the serum metabolome with the blood cell transcriptome indicated that the elevated cellular oxidative response to 1’500 kcal HFM in a subset of OB subjects results from a dysregulation of systemic cellular mechanisms in response to food intake, rather than from changes in gastrointestinal processing.

Data modelling studies confirm a role for food fortification to improve vitamin D intakes and status. However, there is very little research reporting the bioavailability of vitamin D from foods to inform food fortification strategies. There is some evidence to suggest that the lipid component of foods impacts vitamin D uptake in cells; however, no studies have examined the absorption kinetics of vitamin D from different lipid delivery systems in humans. The aim of this research is to compare changes in serum vitamin D concentrations using different vitamin D emulsions as delivery systems.

A single blind, controlled, 4-way crossover postprandial study was conducted. Participants consumed 4 different 20μg vitamin D enriched emulsion drinks (olive oil emulsion, fish oil emulsion, mixed micelle emulsion and a non-fat control drink) with a low-fat breakfast on 4 occasions separated by a 1-week washout. Serum 25-hydroxyvitamin D (25(OH)D3) concentrations and cardiometabolic risk markers were measured at baseline and 2, 4, 6 and 8 hours after consuming the drink. Statistical analysis was performed in IBM SPSS Statistics, Version 24 (SPSS Inc. Chicago, IL, USA). Repeated measures ANOVA was performed to assess the effects of the 4 treatments.

Six males and six females aged between 22 and 65 years completed the study. All participants were healthy with a mean BMI of 24.7 ± 4.1kg/m2 for males and 25.1 ± 4.5kg/m2 for females. Each 100 ml drink contained 106 kcal, 4 g fat, 11.9 g carbohydrate and 5.5 g protein, apart from the control which contained 70kcal and 0 g fat. The breakfast contained 321 kcal, 0.7 g fat, 66 g carbohydrate and 10.3 g protein. There was an increase in 25(OH)D3 concentrations over the 8-hour assessment period following consumption of all 4 drinks. Two-way repeated measures ANOVA showed a significant drink*time interaction. No participants experienced any negative side effects after consuming any of the drinks.

The absorption kinetics of vitamin D is different depending on the lipid delivery system. This is the first human study demonstrating the bioavailability of vitamin D3 from various emulsions including a novel mixed micelle emulsion. All emulsions were palatable, effective and safe to consume.