H. Schellekens et al. / EBioMedicine xxx (xxxx) 103176
Fig. 7. B. longum APC1472 increases Bifidobacterium abundance without impacting the overall composition of the gut microbiota in humans. The gut microbiota was assesed at the begin- ning (pre) and end of the study (12 weeks, past). Alpha (A-C) and beta diversity (D) were investigated, as wel as the bacterial genera present (E-F). Microbial taxa were centre-log-trans- formed (CLR). Significant differences between pre and post were anlysed using the Mann-Whitney U test, whereas treatment differences were analysed using an ANCOVA controlling for sex and pre-intervention Bifidobacterium abundance. Data are depicted as boxplot or scatter dot plot, where the dots depict individual datapoints, with n = 48 for the placebo group and n =74 for the B. longum APC1472 treatment group. * indicates a significant effect (* p <0.05, ** p <0.01).
bacterial strains, including B. longum APC1472, might result in a higher treatment efficacy, as some evidence suggests that multi strain probi- otics may be more effective . Age has also been shown to affect body fat distribution and metabolism increasing both the risk and the severity of obesity development . Therefore, some of the discrepan- cies and lack of translation between the mice study and humans could be explained by the relatively low age of the mice (adolescence to adult- hood) versus the human cohort with an average age at midlife. A low age may facilitate a better response to changes in metabolic and physio- logic responses and therefore a higher capacity to positively respond to therapeutic interventions. Moreover, the administration strategies were differences between both studies. The mouse study followed a preven- tion strategy as B. longum APC1472 was administered before obesity was established, while in the human study the participants were already obese at the time of administration and, therefore, presented a more se- vere condition to ameliorate. Most notably, the B. longum APC1472 supplementation significantly improved glucose tolerance in HFD-induced obese mice. Similarly, B. longum APC1472 decreased fasting blood glucose levels in overweight/ obese individuals ( − 0.266 mmol/L compared to placebo). It is important to note that the participants in this study had average fasting blood glu- cose levels of 5.0 mmol/L, which is considered healthy and non-diabetic ( n = 11 were prediabetic). Above 5.6 mmol/L is considered prediabetic, whereas above 6.9 is considered diabetic [83,84]. These data indicate that B. longum APC1472 may have a bigger effect-size on fasting blood glucose levels in a prediabetic or diabetic population, which warrants further investigations. This is further reinforced by the obese subpopula- tion analysis of the obese individuals, rather than overweight and obese combined, which revealed a fasting blood glucose level ( − 0.295 mmol/ L compared to placebo), which constitutes a bigger effect-size in fast- ing blood glucose levels ( η 2 = 0.092 vs 0.075), indicating a more potent treatment efficacy in obese individuals. This warrants further investiga- tion into the effect of B. longum APC1472 in a cohort of prediabetic or diabetic individuals.
The underlying mechanisms for the decreased fasting blood glucose levels may be associated with the changes in ghrelinergic signalling, as B. longum APC1472 was found to attenuate ghrelinergic signalling in vitro  and ghrelin has been shown to be involved in glucose homeostasis via inhibition of insulin secretion . Moreover, insulin receptor substrate 1 (IRS-1) has been reported to play a key role in glucose homeostasis being involved in glucose transporter 4 (GLUT-4) mobilization [86,87]. Low IRS-1 expression levels have been associ- ated with glucose and insulin sensitivity impairments [86,87]. There- fore, increased IRS-1 expression in epididymal fat tissue of B. longum APC1472 treated mice may have also influenced glucose homeostasis. Nevertheless, glucose metabolism is multifactorial and other mecha- nisms are likely also affected following the supplementation of the B. longum APC1472. However, while the biggest effect-size was observed on plasma glucose levels in both the preclinical and human intervention studies, it is also possible that the other observed effects are secondary to the decrease in plasma glucose levels. Notably, obesity is associated with decreased circulating levels of ghrelin [44,72], which we also observed in the HFD-fed mice and the reason why the ghrelinergic system has been implicated as a promising therapeutic target to combat obesity [45,88]. Indeed, the “ hunger hor- mone ” ghrelin was first described as a growth hormone secretagogue, but its key role in the regulation of appetite, food intake, adiposity and metabolism have directed the main therapeutic focus of ghrelin and its receptor towards obesity research with promising anti-obesity potential [41,45,76,89-92]. Interestingly, B. longum APC1472 supplementation increased levels of active ghrelin, but not total ghrelin levels, in healthy obese individuals. The increase in active ghrelin may indicate an ame- lioration of the deficiencies in ghrelinergic signalling associated with obesity. It is also interesting to note that B. longum APC1472 was se- lected on its ability to modulate the ghrelinergic system in vitro . Future studies are warranted to investigate if administration of other bacterial strains and their metabolites, including SCFAs, which equally
Powered by FlippingBook