Clara Salame

Background: Emulsifiers are the most ubiquitous food additives. Experimental studies showed an emulsifier-induced gut microbiota dysbiosis stimulating inflammation. This study is the first to investigate the association between food additive emulsifiers and type 2 diabetes risk in a large population-based cohort. Methods: Repeated 24h dietary records with brand-specific information were administered within the French NutriNet-Santé cohort (2009-2022), allowing qualitative and quantitative assessment of food additive exposures. Multi-adjusted cox models were used for statistical analysis. Results: This study included 104,168 adults (79.2% women, mean age at baseline=42.7y (SD 14.5)). 1,057 incident T2D were diagnosed. Intakes of total emulsifiers (HR = 1.18[1.00-1.39] for tertile 3 vs 1, p-trend=0.049), total carrageenan (HR = 1.20[1.01-1.42], p = 0.04), tripotassium phosphate (E340, HR = 1.45[1.10-1.91], p < 0.001), triphosphates (E451, HR = 1.39[1.16-1.67], p < 0.001), polyphosphates (E452, HR = 1.36[1.14-1.62], p < 0.001), total polyglycerol esters of FAs (HR = 1.34[1.10-1.63], p < 0.001), polyglycerol esters of interesterified ricinoleic acid (E476, HR = 1.26[1.03-1.54], p < 0.001), sodium citrate (E331, HR = 1.19[1.02-1.40], p = 0.03), carob bean gum (E410, HR = 1.19[1.01-1.40], p = 0.03), guar gum (E412, HR = 1.35[1.14-1.59], p < 0.001), xanthan gum (E415, HR = 1.29[1.09-1.52], p < 0.001) and ammonium salts of phosphatidic acid (E442, HR = 1.31[1.03-1.66], p = 0.01) were positively associated with T2D risk. The only inverse association was observed for sodium bicarbonate (E500, HR = 0.84[0.71-1.00], p = 0.047). Conclusions: This study revealed direct associations between T2D risk and exposures to nine individual and two groups of food additive emulsifiers widely used in industrial foods in a large prospective cohort of French adults. If confirmed, these findings could lead to a reevaluation of regulations governing the use of food additive emulsifiers in the food industry for a better consumer protection. Key messages: • Study finds positive associations between certain emulsifiers and type 2 diabetes risk in French adults. • Results highlight the need for re-evaluation of regulations for food additive emulsifiers to protect consumer health.

To address gaps in knowledge, our objectives were to (1) to determine whether there are age-related changes in sweet taste detection thresholds, as has been observed for sweet taste preferences, and (2) determine whether detection thresholds and taste preferences were significantly related to each other from childhood to adulthood. We combined data from studies that used the same validated psychophysical techniques to measure sucrose taste detection threshold and the most preferred sucrose concentration in children (n = 108), adolescents (n = 172), and adults (n = 205). There were significant effects of age group on both sucrose detection thresholds (p < 0.001) and most preferred sucrose concentration (p < 0.001). While children had higher sucrose detection thresholds than adolescents, who in turn tended to have higher detection thresholds than adults, both children and adolescent most preferred sucrose concentrations were higher than that of adults (all p < 0.05). Among each age group, and when combined, the lowest sucrose concentration detected was not significantly correlated with the most preferred sucrose concentration (all p > 0.18). These data provide further evidence that age-related changes in sucrose taste preferences that occur during adolescence cannot be explained by changes in taste sensitivity and that these two dimensions of sweet taste undergo distinct developmental trajectories from childhood to adulthood.

Here, we tested the hypothesis that sucralose differentially affects metabolic responses to labeled oral glucose tolerance tests (OGTTs) in participants with normal weight and obesity. Participants (10 with normal weight and 11 with obesity) without diabetes underwent three dual-tracer OGTTs preceded, in a randomized order, by consuming sucralose or water, or by tasting and expectorating sucralose (e.g., sham-fed; sweetness control). Indices of β-cell function and insulin sensitivity (S_I) were estimated using oral minimal models of glucose, insulin, and C-peptide kinetics. Compared with water, sucralose ingested (but not sham-fed) resulted in a 30 ± 10% increased glucose area under the curve in both weight groups. In contrast, the insulin response to sucralose ingestion differed depending on the presence of obesity: decreased within 20–40 min of the OGTT in normal-weight participants but increased within 90–120 min in participants with obesity. Sham-fed sucralose similarly decreased insulin concentrations within 60 min of the OGTT in both weight groups. Sucralose ingested (but not sham-fed) increased S_I in normal-weight participants by 52 ± 20% but did not affect S_I in participants with obesity. Sucralose did not affect glucose rates of appearance or β-cell function in either weight group. Our data underscore a physiological role for taste perception in postprandial glucose responses, suggesting sweeteners should be consumed in moderation.