Schellekens_BLongum_Obesity_Stress_EBIOM_103176 (003)

H. Schellekens et al. / EBioMedicine xxx (xxxx) 103176


Fig. 4. B. longum APC1472 supplementation does not impact BMI and W/H ratio in overweight and obese individuals. Body mass index (BMI) (A, B) and waist-to-hip ratio (W/H ratio) (C, D) were measured as the beginning of the study (pre), after 6 weeks (mid) and after 12 weeks (post) of treatment. All BMI and W/H ratio data are depicted of all 3 timepoints (A, C), as well as the change after 12 weeks compared to at the beginning of the study (B, D). 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. 5.9. B. longum APC1472 does not affect satiety, mood, perceived stress and cortisol awakening response in humans Considering that the gut microbiota has been implicated in the mod- ulation of host mood and food intake behaviour [10,64], we investi- gated whether B. longum APC1472 could improve levels of the stress hormone cortisol upon waking (i.e. cortisol awakening response), or self-reported measures of satiety, and self-reported measures of mood (i.e. perceived stress, anxiety and depression) ( Table 3 ). B. longum APC1472 did not impact cortisol awakening response, or self-reported satiety, perceived stress, anxiety and depression measures. 5.10. B. longum APC1472 improves fasting glucose levels, active ghrelin and cortisol awakening response in obese individuals Participants in this study were either overweight ( n =40; 28 ≥ BMI < 30) or obese ( n =82; 30 ≥ BMI < 35). It is possible that B. longum APC1472 may evoke a stronger effect in obese individuals as they have a stronger phenotype compared to overweight individuals. As such, we investigated whether any of the anthropomorphic measures, blood bio- markers and measures of mood were affected by B. longum APC1472 in the obese subpopulation only, compared to placebo ( Fig. 6 and Table S7 – 11 for population characteristics and full statistical results). Similar to the analysis on the entire study population, B. longum APC1472 and placebo reduced fasting glucose levels over the 12-week treatment pe- riod ( Fig. 6A ). However, glucose levels were 0.295 mmol/L (95% CI [ − 0.5, − 0.1]) lower in the B. longum APC1472 group compared to the

placebo group ( F (1,75) = 7.566, p = 0.007), in obese individuals, with a moderate effect size ( η 2 =0.092) ( Fig. 6B ). Furthermore, B. longum APC1472 increased active ghrelin levels ( F (1,74) = 4.903, p =0.030), with a moderate effect size ( η 2 = 0.062). Moreover, B. longum APC1472 also reduced cortisol awakening response ( F (1,51) = 4.415, p = 0.041), with a moderate effect size ( η 2 = 0.080), in the obese sub- population analysis. Overall, these results show beneficial effects of B. longum APC1472 on fasting plasma glucose levels, active ghrelin levels and cortisol awak- ening response in obese individuals. It is also important to note that the effect size in the obese subpopulation ( η 2 = 0.092) was bigger than the effect size in the overall study population ( η 2 = 0.075). This indicates that B. longum APC1472 has a more robust beneficial effect on fasting glucose levels in obese, rather than in overweight, individuals. 5.11. B. longum APC1472 does not induce major rearrangements on the microbiota composition but increases the abundance of Bifidobacterium We subsequently investigated whether the observed changes induced by the B. longum APC1472 strain were mediated in part through mod- ulation of the gut microbiota. Investigations into the caecal micro- biota in the preclinical experiment revealed that there was a signif- icant dissimilarity in beta diversity between LFD- and HFD-fed mice ( p < 0.01) ( Figure S5A ), with a decreased relative abundances of Bacteroidetes phylum and increased relative abundances of Firmicutes class Clostridia , respectively ( Figure S5B ), which is in line with previ- ous studies [65,66]. Different phylotypes were responsible for the cae- cal microbiota differences amongst the treatment groups ( Figure S5C ), showing increments on different Firmicutes members in HFD-fed mice treated with B. longum APC1472. Moreover, B. longum APC1472 par

Powered by