H. Schellekens et al. / EBioMedicine xxx (xxxx) 103176
cases, a q-value < 0.1 was considered significant. α -diversity was com- puted using the iNEXT library . 4.2. Faecal SCFA quantification Faecal samples were homogenised with acidified water (HCl pH 3) at a ratio of 1:7.5 w/v and analysed by gas chromatography flame ion- isation detection (GC-FID) using a Varian 3800 GC system, fitted with an Agilent DB-FFAP column (30 mL x 0.32 mm ID x 0.25 μ m df; Agi- lent) and a flame ionisation detector with a CP-8400 auto-sampler. He- lium was employed as the carrier gas at an initial flow rate of 1.3 mL/ min. The initial oven temperature was 50 °C, was maintained for 30 s, raised to 140 °C at 10 °C/min and held for 30 s, before being increased to 240 °C at 20 °C/min, and held for 5 min (total run time 20 min). The temperatures of the detector and the injection port were 300 °C and 240 °C, respectively. A split-less injection of 0.2 µL was carried out for each sample or standard using a 10 µL syringe (Agilent) installed to a CP-8400 auto-sampler (Varian). A 5 m guard column was installed be- tween the injector and analytical column (Restek). Peak integration was performed using Varian Star Chromatography Workstation version 6.0 software. Vials containing 1800 µL of water were run between each sam- ple duplicates as blanks to control for any potential carryover. Standards were included in each run to maintain the calibration. For further details on sample and standards preparation see supplementary information. 4.3. Questionnaires Using self-report scales, participants were assessed for perceived stress using Cohen's Perceived Stress Scale and anxiety and depression using the Hospital Anxiety and Depression Scale (HADS) at baseline, af- ter 6 and after 12 weeks, as previously described, . In addi- tion, satiety/hunger was determined using a visual analogue Hunger/ Satiety scale, physical activity using the International Physical Activity Questionnaire (IPAQ) . Nutrient intake was assessed using a Food Frequency Questionnaire (FFQ), as previously described . 4.4. Statistical analysis Preclinical data were assessed for normality using the Shapiro-Wilk test. Normally distributed data were analysed using a two-way ANOVA, followed by Fisher's least significant difference (LSD) post hoc test. Non-parametric datasets were analysed using the Kruskal-Wallis test,
followed by the Mann-Whitney U test with Bonferroni adjustment of p-values. Body weight changes and glucose levels in glucose toler- ance test were analysed with a two way repeated-measures ANOVA (with Diet and Probiotic as two independent factors and Time as a re- peated-measured factor), followed by LSD post hoc test at each time point. Statistical analysis was performed using SPSS software (IBM SPSS statistics 22). Preclinical data are represented as mean ± SEM. For the human intervention study, differences between the treatment and placebo groups at the last visit (i.e. visit 4) were analysed using an analysis of covariance (ANCOVA), correcting for baseline variance (i.e. visit 2) and sex. Comparisons between baseline measurements (visit 2) and post-intervention measurements (visit 4) were analysed using an unpaired student's T-test. Analyses were performed on the intention to treat populations. Statistical analysis was performed using SPSS soft- ware version 26 (IBM Corp). Data in table are presented as mean ± SEM or 95% CI. P-Values <0.05 were considered statistically signifi- cant. For significant associations, a Benjamini-Hochberg procedure was performed with a threshold of q <0.1. Partial eta-squared ( η 2 )was used to estimate effect size . Effect sizes were interpreted as following: η 2 ≤ 0.06 was considered small, 0.06 > η 2 ≤ 0.14 was considered mod- erate, η 2 ≥ 0.14 was considered large. 5. Results 5.1. B. longum APC1472 decreases body weight gain and fat depots accumulation in obese mice B. longum APC1472 decreased body weight gain after 15 weeks of administration ( F (1, 33) = 4.751, p =0.037) ( Fig. 2A, 2B ). HFD feed- ing increased caloric intake (F (1, 15) = 9.229, p =0.008) ( Figure S1C ), body weight (F (1, 33) = 29.715, p < 0.001) ( Fig. 2A ) and fat depot accumulation (mesenteric (F (1, 33) = 61.328, p < 0.001), retroperitoneal (F (1, 32) = 128.409, p < 0.001), subcutaneous ( F (1, 31) = 124.091, p < 0.001) and epididymal ( F (1, 33) = 81.673, p < 0.001)) ( Fig. 2C, D, E, F ). Pairwise comparisons showed a significant decreased body weight effect of B. longum APC1472 in HFD-fed mice ( p= 0.047) ( Fig. 2B ), which was independent of caloric intake ( Fig- ure S1 ). Furthermore, the administration of B. longum APC1472 signif- icantly reduced fat depot accumulation (mesenteric ( F (1, 33) = 5.908, p = 0.021), and subcutaneous ( F (1, 33) = 4.270, p =0.047)) ( Fig. 2C, D, E, F ). Finally, pairwaise comparisons revealed a significant de- creased fat depot accumation effect of B. longum APC1472 administra- tion in HFD-fed mice (mesenteric p = 0.002, retroperitoneal p =0.05 and subcutaneous p =0.023).
Fig. 2. Effects of Bifidobacterium longum APC1472 on body weight and fat depots accumulation in mice. (A) Weekly body weight gain, (B) total body weight gain and (C) mesenteric, (D) retroperitoneal, (E) subcutaneous and (F) epididymal fat depots accumulation (% of total body weight) in control mice treated with drinking water containing sterile PBS (2% v/v) and glycerol (0.5% v/v) and fed a control low-fat diet (LFD) ( n = 10) or a high-fat diet (HFD) ( n = 9) and in mice treated with B. longum APC1472 in drinking water (2 × 10 8 CFU/mL) and fed a LFD ( n =9inA,B,C,EandF; n =8inD)oraHFD( n =9inA,B,C,D,andF; n = 8 in E) for 15 (A, and B) or 16 weeks (C, D, E and F). Data are shown as mean ± SEM.. Data are significant different ( p <0.05) accordingly to Repeated Measures ANOVA (A) or two-way ANOVA followed by LSD post-hoc test (B, C, D, E and F). * indicates significant diet treatment effect (* p <0.05, ** p <0.01, *** p <0.001) and # indicates significant B. longum APC1472 treatment effect ( # p <0.05, ## p <0.01).
Powered by FlippingBook