Chitin-glucan supplementation improved postprandial metabolism and altered gut microbiota in subjects at cardiometabolic risk in a randomized trial

The beneficial effects of dietary fibers (DF) are now well recognized and the current recommended daily intake is 25–30 g/day¹.When assessing DF contribution to health, DF interaction with gut microbiota is also widely reported as of interest. DF may actually affect gut microbiota composition and function and induce the production of gut-derived metabolites such as short-chain fatty acids (SCFA), which in turn can improve glucose and lipid parameters². DF can have different origins and can be classified according to their physicochemical properties (including solubility, viscosity or fermentability)³. These parameters are used to classify DF which are thus subdivided into resistant starch (RS), non-starch polysaccharides (NSPs) and resistant oligosaccharides (ROs)^4,5. DF also have different physiological effects which are not always associated with their classification, particularly in terms of their solubility^5,6. Actually, solubility is often correlated with fermentability so that soluble DF are fermented more quickly than insoluble DF, with the exception of RS which are highly fermentable although the majority of them are insoluble DF^6,7. Of note, the health effects and the underlying mechanisms of action of insoluble DF remain unclear. In this context, the project FiberTAG (Joint Programming Initiative “A Healthy Diet for a Healthy Life” 2017–2020 https://www.fibertag.eu/↗) aimed at establishing a set of biomarkers of gut barrier function and bacterial co-metabolites beyond SCFA, such as long-chain fatty acids (LCFA), bile acids (BA), intestinal permeability and endotoxemia, which link DF intake and gut-microbiota related health effects. Moreover, we also focused on noninvasive biomarkers such as exhaled gases⁸. In particular, lactulose hydrogen (H₂) breath test is primarily used as a marker for small intestinal bacterial overgrowth (SIBO). Since several reports compared the fermentation of DF using the measurement of end-expiratory H₂ concentrations and suggested that they depend on the type of DF⁹, exhaled H₂ during lactulose H₂ breath test could be of interest as it would allow to assess the fate of DF and its cardiometabolic effects.

Here, we were particularly interested in chitin-glucan (CG), a novel insoluble DF considered as safe food ingredient by the European Food Safety Authority¹⁰. It is the major component of the cell walls of the fungi Aspergillus niger and is mainly composed of a branched β-1, 3/1, 6 glucan that is linked to chitin via a β-1, 4 linkage. Preclinical studies in high fat diet (HFD) rodent models showed potential beneficial effects of CG supplementation, namely on gut microbiota composition and function but also on the cardiometabolic profile^11,12. The first attempt to study the effect of this CG on gut microbiota in humans was carried out as part of an in vitro model using a Simulator of the Human Intestinal Microbial Ecosystem (SHIME)¹³. This study showed that CG was fermented in the colon and induced the production of gut-derived metabolites such as short chain fatty acids (SCFA). Consistently, CG-derived effects on gut microbiota and bacterial metabolites in humans have been studied recently, revealing the interaction of CG with the gut microbiota in healthy subjects¹⁴. Together, these results suggest that it would be of interest to study the effects of CG in subjects at cardiometabolic risk who may benefit more distinctly than healthy individuals from such intervention.

In this double-blind, randomized, cross-over exploratory study, we aimed at assessing whether a 3-week daily supplementation of 4.5 g of CG had different effects from maltodextrin (as control) on gut microbiota composition and functions as well as cardiometabolic profile (including postprandial phenotyping) in subjects at cardiometabolic risk (CMR) by assessing a comprehensive set of biomarkers. We also investigated the relevance of targeted biomarkers to identify responders to the fiber intervention, allowing us to further characterize those participants who were able to metabolize CG.

In this exploratory study performed in a population of subjects at CMR of whom 90% were dyslipidemic, we showed that compared to maltodextrin as control, 3-week supplementation with CG modified the gut microbiota composition by changing the relative abundance of several gut bacterial taxa. This effect was accompanied by a significant improvement of postprandial glucose and lipid profiles.

The effects of CG supplementation on gut microbiota and bacterial co-metabolites in humans has been recently studied in healthy subjects, revealing that CG increased the relative abundance of the butyrate-producing bacteria Roseburia sp, and coherently increase SCFA production including butyric acid¹⁵. In our study, Roseburia sp abundance and fecal SCFA concentrations were not differently modified by the two treatments, even if butyric acid tended to be higher after CG supplementation. These results suggest that the production of metabolites could differ according to the metabolic status of participants or to their gut microbiota characteristics. It is also well known that the pathways of metabolites’ production are diverse and redundant among the different microbiota species. It is therefore possible that the impacted microbial features may be involved in opposite pathways with the resultant corresponding to the “no effect” we observed when measuring the metabolites.

In this study, we also showed that exhaled H₂ following the intake of a fiber-enriched breakfast was increased by CG compared to control, suggesting the ability of CG to induce a higher fermentation in response to the dietary fiber intake.

Interestingly, still compared to control, CG improved postprandial glucose and lipid metabolism as illustrated by the decrease in iAUC of glycemia and triglyceridemia, but also the peak of triglyceridemia in response to a standardized test meal challenge, the FlexMeal. In a 6-week, randomized, double-blind and placebo-controlled study conducted in healthy overweighed volunteers, the same dose of CG also exhibited beneficial effects by reducing blood levels of oxidized low-density lipoproteins (ox-LDL)¹⁶. We did not observe such effect here, possibly due to the difference in terms of study duration. In accordance with the effects observed on postprandial glycemic and lipemic profiles in this study, it has been previously shown that in mice fed with a high-fat diet (HFD), a 4-week administration of CG significantly improved metabolic parameters, including glucose tolerance¹². Another in vivo study also demonstrated beneficial effects of CG in lowering plasma triglycerides in hamsters fed with an atherogenic diet but at fasting state¹¹. Presently, the observation of effects at postprandial and not at fasting state, bolsters the interest of postprandial phase when assessing the effects of nutritional interventions and more generally human health¹⁷. Furthermore, postprandial hyperglycemia and hyperlipemia are now recognized as independent cardiometabolic risk factors¹⁸. An exacerbated and prolonged postprandial lipemia is typically observed in subjects with metabolic disorders¹⁹. Also, epidemiological data reported that post-challenge glycemia is associated with cardiovascular risk, regardless of fasting biomarkers such as fasting glycemia and glycated hemoglobin (HbA1c) and even has a greater predictive power of cardiovascular events²⁰. Indeed, abnormally increased levels of glucose and lipids generates excessive amounts of free radicals (or reactive oxygen species triggering a biochemical cascade throughout the circulation, inducing inflammation, endothelial dysfunction, hypercoagulability, sympathetic hyperactivity and a more atherogenic lipoprotein profile^21–23). Therefore, a reduction of postprandial hyperglycemia and hyperlipemia excursions, as demonstrated in the present study with 3-week supplementation of CG, is of interest to improve cardiometabolic profiles.

H₂ has been reported to have antioxidant, anti-inflammatory, and anti-apoptotic properties²⁴. In patients with type 2 diabetes or impaired glucose tolerance, supplementation of H₂-rich water improved lipid and glucose metabolism²⁵. Similarly, in db/db mice, drinking H₂-water has been shown to decrease levels of plasma glucose, insulin, and triglyceride, the effect of which on hyperglycemia was similar to diet restriction²⁶. The suggested underlying mechanism is an observed increase of the expression of a hepatic hormone, fibroblast growth factor 21 (FGF21), which is a regulator of energy expenditure. Here, the energy metabolism as measured by resting metabolic rate did not show significant differences between the two treatments. Subgroup analyses revealed that the improvement in postprandial glucose metabolism, especially the decrease in the iAUC of postprandial blood glucose, was only observed in the “High-H₂” subgroup composed of subjects with a higher concentration of exhaled H₂ in response to the lactulose H₂ breath test. This study is the first of its kind to identify the production of H₂ as an indicator of a fiber-derived impact on glucose metabolism. “High H₂” subjects had a lower abundance of methanogenic bacteria including Methanobrevibacter smithii, which is the most dominant methanogen in the gastrointestinal tract in humans²⁷. It has been reported that methanogens increase in ob/ob mice²⁸ and that in obese humans CH₄ production is associated with a higher BMI²⁹. The suggested mechanism is that methanogens use the H₂ produced by syntrophic organisms for their own anaerobic metabolism, after which CH₄ is produced as a by-product. This H₂ scavenging allows syntrophic organisms to be more productive, increasing SCFA production and availability of calories for the host^30,31. Subjects included in our study were all H₂ producers and even if the comparison between the two subgroups showed that they differed by their abundance in methanogenic bacteria, no significant difference in terms of exhaled CH₄ was found. Interestingly, at baseline, when firstly challenged with the standardized test meal, subjects belonging to the “High-H₂^” subgroup were the ones who presented a higher fasting concentration of NEFA and glycemia peak after the fiber-enriched breakfast, tAUC, iAUC and peak of glycemia after the standardized test meal challenge. Thus, the subjects who have benefited the most from the intervention in terms of glycemic profile improvement would actually be those who already have impaired postprandial metabolic profiles.

Of note, betaglucan (BG) is one component of CG. It has been shown to decrease postprandial glycemia after an oral glucose test tolerance (OGTT) or when added to food products (pasta, bread, soup, cereals, etc.) in healthy or diabetic subjects^32–36. Their postprandial glucose lowering effect has long been reported to be essentially due to a delayed and somewhat reduced carbohydrate absorption from the gut^37,38. However, since different studies have shown BG impacts on gut microbiota without assessing in parallel the impacts on glycemic response, other mechanisms may exist but need further investigations to be elucidated³⁹.

When assessing the link between impacted metabolic and bacterial parameters, Slackia was the only bacterial feature which showed direct correlation with postprandial glycemia. It has been reported that the abundance of this genus was higher in prediabetic subjects and reduced after supplementation in xylooligosaccharide⁴⁰. Here, we showed that this genus was differently impacted by the two treatments and particularly decreased by CG even if not statistically significant and only in « Low-H₂». The improvement of postprandial triglycerides metabolism with CG supplementation was correlated with the decrease in rather pro inflammatory microbial features. The correlation analyses also suggested that the increase of exhaled H₂ after the fiber-enriched breakfast may be explained by abundance’s modification of a cluster of microbial features including butyrate-producers. Consistently, a link between bacterial substrate metabolism and H₂ gas formation have already been proposed in butyrate-producing colon bacteria^41,42.

Our study presents some strengths and limitations. We assessed a cross-over design study with a comprehensive set of biomarkers from gut microbiota composition and functions to fasting and postprandial phenotyping. Even though 4.5 g chitin-glucan were not sufficient to increase SCFA or barrier function biomarkers, it allowed to improve postprandial metabolism, which has been reported to be associated to cardiometabolic risk profile. For the first time, we identified an interesting non-invasive biomarker: exhaled H₂ following a lactulose breath test, which could be of interest to evaluate dietary fiber impact on postprandial glucose metabolism. We also showed that CG supplementation increased exhaled H₂ following a fiber-enriched breakfast. Although several beneficial metabolic effects of H₂ enrichment have been reported (antioxidant, anti-inflammatory, and anti-apoptotic properties, lipid and glucose metabolism improvement)^24–26, we cannot say if the present H₂ could have the same effects. The limitations of this study is that it was a short-term exploratory study performed in a small number of participants, with a comprehensive set of biomarkers so that after adjustment for multiple comparisons, none of the p values remain significant. Due to the explorative character of the study, we thus presented and considered non-adjusted p values < 0.05 as significant.

To conclude, this study showed the potential effects of a 3-week supplementation of CG compared to maltodextrin in at cardiometabolic subjects. No specific fecal biomarkers of CG interactions with the gut microbiota (as shown in the study in healthy volunteers) has been demonstrated in the present study. However, presently the improvement in the postprandial glycemic and lipemic profiles, early determining markers of cardiometabolic risk, underlies the interest of such dietary intervention in this specific at-risk population. The profile response to lactulose H₂ breath test seems to be a parameter of interest to predict the effectiveness of the intervention especially in terms of glycemic response. Further investigations are needed to confirm these observations and to elucidate the underlying mechanisms of CG on cardiometabolic profile in subjects at cardiometabolic risk.