Synchronized shift of oral, faecal and urinary microbiotas in bats and natural infection dynamics during seasonal reproduction

Dietrich, Muriel; Kearney, Teresa; Seamark, Ernest C. J.; Paweska, Janusz T.; Markotter, Wanda

All statistical analyses were conducted in R [32] , primarily with the vegan, Rcmdr, MuMIn and ggplot2 packages [33] [34] [35] . We examined the temporal dynamics of bacterial and viral shedding by using generalized linear models (GLMs) with a binomial distribution and Pearson Chi-squared tests, separately for each infectious agent. Sampling period (month), sex, age class (adult/juvenile) and reproductive condition (active or not) were included as explanatory variables in the GLMs. We selected the most parsimonious model based on Akaike's information criterion. The effect of variables included in the most parsimonious model was tested using a Chi-square test (χ 2 ) (electronic supplementary material, tables S1 and S2). Analysis of microbiota was performed in parallel for four datasets, produced after gradual exogenous bacterial phylotype removal (electronic supplementary material, text S2). After checking for consistency among results from the different datasets, we present results for dataset D2, for which exogenous phylotypes with a relative abundance greater than 10% in the controls were removed. We analysed αand β-diversity metrics from the rarefied phylotype tables. We calculated microbial richness using the inverse Simpson diversity index and used ANOVA to compare bacterial diversity among body habitats for M. natalensis (as only saliva samples were collected for R. aegyptiacus). Analyses of bacterial diversity were then performed for saliva, urine and faeces separately (electronic supplementary material, text S2), and the effect of the sampling period, sex, age class, reproductive condition and infection status (with HVs, AdVs and Leptospira bacteria) was tested using GLMs assuming a Gaussian distribution. Moreover, for urine, a Pearson's correlation test was performed to analyse the microbiota diversity in relation to the Leptospira load (cycle threshold value measured by RT-PCR).