Transcriptome sequencing and annotation for the Jamaican fruit bat (Artibeus jamaicensis)

Abstract

The Jamaican fruit bat (Artibeus jamaicensis) is one of the most common bats in the tropical Americas. It is thought to be a potential reservoir host of Tacaribe virus, an arenavirus closely related to the South American hemorrhagic fever viruses. We performed transcriptome sequencing and annotation from lung, kidney and spleen tissues using 454 and Illumina platforms to develop this species as an animal model. More than 100,000 contigs were assembled, with 25,000 genes that were functionally annotated. Of the remaining unannotated contigs, 80% were found within bat genomes or transcriptomes. Annotated genes are involved in a broad range of activities ranging from cellular metabolism to genome regulation through ncRNAs. Reciprocal BLAST best hits yielded 8,785 sequences that are orthologous to mouse, rat, cattle, horse and human. Species tree analysis of sequences from 2,378 loci was used to achieve 95% bootstrap support for the placement of bat as sister to the clade containing horse, dog, and cattle. Through substitution rate estimation between bat and human, 32 genes were identified with evidence for positive selection. We also identified 466 immune-related genes, which may be useful for studying Tacaribe virus infection of this species. The Jamaican fruit bat transcriptome dataset is a resource that should provide additional candidate markers for studying bat evolution and ecology, and tools for analysis of the host response and pathology of disease.

Figure 1. Distribution of mapped contigs.

Histograms displaying the proportion of contigs mapped to particular features of protein coding genes of human and mouse (UTR is the untranslated region, and CDS is the coding sequence). The upper panel (A) displays the raw count and the lower panel (B) normalized values (the proportion discovered relative to how many could be discovered within each category). The raw count of SNPs (C) and Indels (D) mapped to particular features of protein coding genes of human and mouse.

Figure 2. Species with more than 100 top hits from B2G.

Figure 3. B2G annotation for Molecular Function, Biological Process, and Cellular Component Level 2.

Figure 4. Distribution of immune genes at the GO slim level based on CateGOrizer.

Figure 5. Venn Diagram comparison of unannotated transcript mapped to Myotis lucifugus genome, Pteropus vampyrus genome, and Pteropus alecto transcriptome.

Figure 6.KEGG. KEGG Mapped Genes.

A graphical representation for two KEGG pathways: (A) B cell receptor signaling pathway and (B) natural killer cell-mediated cytotoxicity. Because the original KEGG graph is much larger, we only present the center genes and genes neighboring these central nodes. If mapped, genes representing nodes were either highlighted with light green or light blue colors. Genes highlighted with light blue contain SNPs.

Figure 7. Substitution estimation scatter plot.

We calculated the nonsynonymous mutation rate (dN) and synonymous mutation rate (dS) using orthologous genes between bat and human. Two lines were drawn representing the two dN/dS cutoffs of 0.7 (green) and 1.0 (blue).

Figure 8. Unrooted species tree from the orthologous dataset across six Boreoeutheria mammals.

The species tree was generated from 2378 gene loci. There was 95% bootstrap support for placing bats (Chiroptera) sister to Perissodactyla, Cetartiodactyla, and Carnivora.