Exploring the link between gender, GI distress, and the gut microbiome

Methods

Summary

24-30 resistance-trained men (n = 12-15) and women (n = 12-15) will complete a resistance exercise protocol (RE) consisting of the squat, shoulder press, deadlift, bent-over row, and leg press exercises or a resting control session (CON) in a randomized, counterbalanced design. The exercise protocol will utilize a load of 70% 1-RM for 4 sets of 10 repetitions and a 90-second rest period length between sets and exercises. Fecal samples will be collected in the 24 hours prior and following the exercise protocol. Blood samples will be collected before exercise (PRE), as well as immediately- (IP), 15, 30, and 60 minutes post-exercise. GI symptom questionnaires will be completed by participants to assess subjective GI issues. Blood and fecal samples will be assayed to quantify small intestine damage (I-FABP), GI permeability (Lactulose/rhamnose ratio), endotoxin leakage, as well as intestinal (calprotectin) and systemic (TNF-α) inflammation. We will also analyze fecal samples for short-chain fatty acids (SCFA) utilize new technology termed shallow shotgun metagenomics to classify the gut microbes of these resistance-trained participants down to the species level and provide insight as to the functional role of their microbiome.

Pre Analysis Plan

Prior to analysis all data will be assessed to ensure normal distribution, homogeneity of variance and sphericity. Non-normally distributed data will be transformed using the natural log (LN) and if sphericity is violated, a Greenhouse Geisser correction will be applied. General Linear Model (GLM) repeated measures analyses [time x trial x sex] will be used to compare trials (RE, CON) for all variables. Following a significant F ratio, separate one-way repeated-measures ANOVA will be performed to assess the effect of time during each trial, and separate paired-samples t-tests will be used to compare trials at each time point. Separate independent t-tests will also be used to evaluate specific differences between sexes when applicable. An alpha level of p < 0.05 will be considered statistically significant for all comparisons. Based on a standard deviation of 511 pg·ml⁻¹ and 347 pg·ml⁻¹ for post-exertional I-FABP for males and females, respectively (22) and using a standard alpha (0.05) and beta value (0.8), a sample size of n= 4 per group was calculated to have adequate statistical precision to detect a > 110% increase in I-FABP post-exercise. Such increases in I-FABP after exercise have also been correlated with the magnitude of intestinal permeability (24). Current participant numbers are also in accordance with sufficient statistical precision to detect significant differences in gastrointestinal symptoms (3). From the metagenome sequencing the group comparisons of gut microbiota will be analyzed with non-parametric Wilcoxon matched-pair signed rank test and corrected for multiple comparison with Benjamini-Hochberg procedure. Regarding the taxonomic data, all analyses will be made with Quantitative Insights Into Microbial Ecology (QIIME) open-sourced software from the randomly subsampled Operational Taxonomic Unit (OUT) table with rarefaction level matching the sample with the lowest total OTU count. The bacterial diversity of the samples (α-diversity metrics) and statistically significant differences in the OTU abundances will be computed with QIIME. Any correlations between non-normally distributed gut microbiota and other variables will be determined using Spearman's rank correlation coefficient in SPSS. The GLM in SPSS will be used to determine whether differences in taxa occur due to biological sex or whether they are dependent on age, weight, body fat %, total energy intake, macronutrient composition, or fiber.