Transcriptome of the Southern Muriqui Brachyteles arachnoides (Primates:Platyrrhini), a Critically Endangered New World Monkey: Evidence of Adaptive Evolution

Abstract

The southern muriqui (Brachyteles arachnoides) is the largest neotropical primate. This species is endemic to Brazil and is currently critically endangered due to its habitat destruction. The genetic basis underlying adaptive traits of New World monkeys has been a subject of interest to several investigators, with significant concern about genes related to the immune system. In the absence of a reference genome, RNA-seq and de novo transcriptome assembly have proved to be valuable genetic procedures for accessing gene sequences and testing evolutionary hypotheses. We present here a first report on the sequencing, assembly, annotation and adaptive selection analysis for thousands of transcripts of B. arachnoides from two different samples, corresponding to 13 different blood cells and fibroblasts. We assembled 284,283 transcripts with N50 of 2,940 bp, with a high rate of complete transcripts, with a median high scoring pair coverage of 88.2%, including low expressed transcripts, accounting for 72.3% of complete BUSCOs. We could predict and extract 81,400 coding sequences with 79.8% of significant BLAST hit against the Euarchontoglires SwissProt dataset. Of these 64,929 sequences, 34,084 were considered homologous to Supraprimate proteins, and of the remaining sequences (30,845), 94% were associated with a protein domain or a KEGG Orthology group, indicating potentially novel or specific protein-coding genes of B. arachnoides. We use the predicted protein sequences to perform a comparative analysis with 10 other primates. This analysis revealed, for the first time in an Atelid species, an expansion of APOBEC3G, extending this knowledge to all NWM families. Using a branch-site model, we searched for evidence of positive selection in 4,533 orthologous sets. This evolutionary analysis revealed 132 amino acid sites in 30 genes potentially evolving under positive selection, shedding light on primate genome evolution. These genes belonged to a wide variety of categories, including those encoding the innate immune system proteins (APOBEC3G, OAS2, and CEACAM1) among others related to the immune response. This work generated a set of thousands of complete sequences that can be used in other studies on molecular evolution and may help to unveil the evolution of primate genes. Still, further functional studies are required to provide an understanding of the underlying evolutionary forces modeling the primate genome.

Keywords: RNA-seq; de novo transcriptome assembly; immune system; positive selection; primate.

FIGURE 1

Expression profiles of the Brachyteles arachnoides transcriptome from blood cells and fibroblasts. (A) Venn diagram comparing all CDS with BLAST hit expressed (TPM ≥ 1) by blood cells and fibroblasts. (B) Venn diagram comparing the number of CDS homologous to proteins of the innate immune system. (C) HSP coverage plotted against TPM values with marginal density plots considering all CDS. (D) HSP coverage plotted against TPM values with marginal density plots considering CDS homologous to proteins of the innate immune system.

FIGURE 2

Orthogroups intersections among 11 primate species. The top bar plot represents the number of orthogroups intersections. The bottom-left bar plot represents the number of orthogroups per species. The dots indicate the intersections among the species with at least 50 shared orthogroups.

FIGURE 3

Maximum likelihood phylogeny of primate APOBEC3Gs. A PhyML tree of aligned amino acid sequences. Human and NWM APOBEC3As were used to root the tree. A cluster with NWM APOBEC3Gs and retrocopies are shown in pink. The sister clade of OWM, Great Apes and Human APOBEC3Gs are shown in light pink. The scale bar represents the amino acid substitution rate, using the Blosum62 + GAMMA model. NWM APOBEC3Gs sequences associated with RefSeq accessions are marked with hash sign (#).

FIGURE 4

Alignment of CEACAM1 proteins from nine primate species. Human and B. arachnoides sequences are displayed in the first lines to allow a direct comparison of the altered amino acids. Nine primate sequences were compared to the predicted peptide sequence of B. arachnoides. Numbering corresponds to amino acid (aa) positions considered in the FEL analysis for positively selected sites. Boxes correspond to protein domains and lines point to positively selected aa substitutions.

FIGURE 5

3D Model of APOBEC3G homodimer. Electrostatic profile of B. arachnoides (A) and H. sapiens (B) APOBEC3G: the bluest zones on protein surface represent values +74.419 KT/e_c; the reddest zones represent values −74.419 KT/e_c of the electrostatic properties; white regions mean zero values of the electrostatic potential. (C) The structural alignment between APOBEC3G Homo sapiens models (gold cartoon) and B. arachnoides (gray cartoon). (D) Template used for structural modeling, showing the homodimeric interaction between two APOBEC3Gs. Insets show residues (in sticks) under positive selection in muriqui sequence, border color of insets corresponds to each specific highlighted region in C.