Dynamics of rumen microbiome in sika deer (Cervus nippon yakushimae) from unique subtropical ecosystem in Yakushima Island, Japan

Abstract

Yaku sika deer (Cervus nippon yakushimae) are endemic to Yakushima Island, whose landscape covered with primary evergreen forest is recognized as a World Heritage Site. In this study, the rumen bacterial microbiota (RBM) of wild Yaku sika was characterized using high throughput sequencing of bacterial 16S rRNA genes combined with targeted cultivation and functional analyses. Comparative analyses of RBM datasets from other ruminant animals revealed distinct community structure among domesticated and wild ruminants. Wild Yaku sika RBM exhibited higher species richness than other sika deer (i.e. wild Ezo sika and domesticated sika deer), likely reflecting their dietary variations associated with unique ecosystem in the island. The Yaku sika RBM of high deer population density samples exhibited higher diversity and contained higher proportion of Firmicutes than those of lower density samples. Moreover, the highest abundance of tannase gene were observed in individuals from the highest population density area, consistent with the previous observation that Yaku sika in the high density areas expanded their feed to include tannin-rich unpalatable plants. This study indicated that RBM of unique wild Yaku sika contribute to the flexibility of dietary shift and thus maintaining nutritional status of Yaku sika under high density conditions.

Figure 1

A contour map of estimate Yaku sika population density in 2013 (A) and 2016 (B). The numbers on the maps indicate the IDs of the ten areas separated by river boundaries (RB). Samples used in this study were obtained from following areas on the map; RB9Y13, RB9Y16, RB1Y12, and RB1Y16. Color keys show the estimated deer density as the number of deer individuals per km². Redrawn from figures in Kagoshima Prefecture Nature Conservation Division under permission of Kagoshima Prefecture.

Figure 2

Yaku sika rumen bacterial community compositions at phylum level (A) and genus level (B). “Other phyla” contains several phyla with < 1% relative abundance. “Other genera” contains several genera among other phyla.

Figure 3

(A) PCoA based on weighted UniFrac distance among ruminants. The weighted UniFrac distance was tested using permutational multivariate analysis of variance. (B) Rarefaction curves of 16S rRNA OTUs among sika deer. The number of OTUs of Yaku sika rumen bacterial microbiota was compared to those of Ezo sika and domestic sika deer. *P ≤ 0.001.

Figure 4

(A) PCoA analysis based on weighted UniFrac distance and (B) rarefaction curves of 16S rRNA OTUs among Yaku sika groups. *P ≤ 0.015.

Figure 5

The ratios of tanB to16S rRNA genes copy numbers (tanB/16S rDNA). Bars indicate the mean values of triplicate qPCR replicates from individual deer (n = 5/group), and error bars representing the standard deviation are shown but absent due to their small scale.