8; lane 9-13 positive controls (TEM-3, TEM-6, TEM-9, TEM-10, SHV-2). Discussion The Barents Sea subpopulation of polar bears has little contact with human activities [10], and the samples investigated in this study were collected from the subpopulation in their natural environment in Svalbard, Norway. MI-503 mw Fresh faeces were collected from live, sedated bears and immediately frozen. There is a potential loss of bacteria when samples are stored before cultivation of bacteria. The pure faeces samples were stored at -70°C and the rectum swabs were stored in 20% glycerol at the same temperature. Achá

et al [31] found that there was not a great loss of bacterial number and species when pure faeces samples were stored at -70°C compared to faeces samples mixed with a cryoprotectant such as glycerol, as long as the samples were not repeatedly thawed and analysed in shorter intervals. The samples processed in this study were not repeatedly thawed and analysed and we expect little loss of bacterial

number and species compared to if the samples were mixed with glycerol before storing. The 16S rRNA gene libraries were made from DNA extracted from faeces, and the samples were pooled after PCR to ensure that bacterial DNA from all animals was equally represented. The number of PCR cycles were reduced to a minimum, as the frequency of formation of chimeric molecules increases by the number of PCR cycles Staurosporine mw [32]. We used 30 cycles for the amplification of the 16S rRNA genes, and did not detect possible chimeras using the Chimera Detection Program. Seventeen different phylotypes were identified among the 161 sequences analysed (Table 2). The coverage of the combined libraries was 97%, which indicate that we have detected the majority of the present Urocanase microbioma in the faeces. In a study based

on faecal microbial communities of 106 individual mammals representing 60 species from 13 taxonomic orders, including captive bears and pandas, Ley et al [33] observed that host diet and phylogeny both influence bacterial diversity, which increases from carnivorous to omnivorous to herbivorous animals. In captive carnivores between 19 and 75 OTUs were observed using the 96% similarity criteria, while in herbivore animals up to 223 OTUs were detected. Within members of the Ursidae family including carnivorous, herbivorous and omnivorous bears, the number of OTUs ranged from 14 to 34 which is consistent with our findings. Only four of the seventeen phylotypes were < 97% related to any known cultivated species (Table 2). This is in contrast to observations made in other studies that the microbial diversity reflected by cultivation represents only a minor fraction of the microbial diversity. In a study of the microbial diversity in reindeer, 92.5% of the bacterial diversity represented novel taxonomic groupings [7].