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

Based on the analysis of the Illumina-sequenced 16S amplicons, we found a significant change of bacterial community diversity over the reproductive season in the oral microbiota only, for both bat species (GLM 7 : χ 2 2 = 1562.4, p < 0.001; GLM 8 : χ 2 3 = 5.245, p = 0.004) (electronic supplementary material, table S1). In M. natalensis for example, bacterial diversity in saliva was particularly high in September (electronic supplementary material, figure S1a). In faeces, bacterial community diversity was driven by sex (GLM 6 : χ 2 1 = 36.123, p = 0.002), with females harbouring less diversity compared to males (electronic supplementary material, figure S1b). This may be explained by a confounding effect of reproductive status; as in M. natalensis, all males were in a non-reproductive state (non-scrotal) and most females were reproductively active. This would contrast with the results of Phillips et al. [20] from phyllostomid bats from South America, where reproductive females had the highest microbiota diversity compared to all other bat groups (scrotal males, non-reproductive males and non-reproductive females). A marginal difference between sexes was also observed in the saliva of R. aegyptiacus (GLM 8 : χ 2 1 = 1.651, p = 0.040), but with the opposite pattern (electronic supplementary material, figure S1b). Finally, we did not find any association between infectious agent shedding and the diversity of bacterial communities in the respective body habitats, except for Leptospira bacteria in urine. Indeed, the presence of Leptospira bacteria (detected by RT-PCR) in urine in M. natalensis was associated with lower diversity in the urinary microbiota (GLM 5 : χ 2 1 = 447.320, p = 0.004, figure 3a), and as expected the urinary microbiota diversity decreased with increased Leptospira load (r 2 = 0.52, p < 0.001, figure 3b).