Shifts in the gut microbiome observed in wildlife faecal samples exposed to natural weather conditions: lessons from time-series analyses using next-generation sequencing for application in field studies

1. Gut microbiome studies in wildlife species are of increasing interest to understand the effects of intrinsic and extrinsic factors on the variation of bacterial communities, which are associated with host health. Current research is mainly restricted to faecal samples obtained during veterinary inspections or to collection of faeces instantly after defecation. For some elusive wildlife species, the collection of faeces in the field can only occur hours or days after defecation and might be the only option to investigate their gut microbiomes. However, to interpret bacterial sequencing results, it is essential to understand how abiotic and biotic factors affect bacteria in faecal material after defecation. 2. We analysed gut microbiomes of two ungulate species, the giraffe (Giraffa camelopardalis) and the springbok (Antidorcas marsupialis), based on field-collected faecal material and further investigated how exposure to the environment influenced gut bacteria over time. We applied high-throughput microbial 16S rRNA gene sequencing to daily faecal subsamples exposed to the environment and assigned taxonomy to sequence reads. We used diversity measures, GAMs and SourceTracker (Knights, Kuczynski & Charlson, Nature Methods, 2011; 8: 761) to assess how community compositions and specifically the ratios of anaerobic to facultatively aerobic and aerobic bacteria were effected during exposure. 3. Our results demonstrate shifts in microbial communities over time in both species with a decrease of anaerobic bacteria and an increase of facultatively aerobic and aerobic bacteria. SourceTracker analyses revealed that the native microbiome still contributed up to 90% to the giraffe microbiome after 1 week, but only up to 31% in springbok. Furthermore, changes in proportions of bacterial taxa appeared earlier and were stronger in springboks than in giraffes. Differences in sample degradation between the species were due to a combination of environmental factors and differences in faecal characteristics. 4. Bacterial data derived from faecal samples without a known history can lead to false description of gut microbiomes. Therefore, we highlight the need for preliminary tests to assess up to which point such samples can still deliver reliable results. Such a strategy will enable the investigation of gut microbiomes in wildlife species where immediate sample collection is not possible.