Colobines are a unique group of Old World monkeys that principally eat leaves and seeds rather than fruits and insects. We report the sequencing at 146× coverage, de novo assembly and analyses of the genome of a male golden snub-nosed monkey (Rhinopithecus roxellana) and resequencing at 30× coverage of three related species (Rhinopithecus bieti, Rhinopithecus brelichi and Rhinopithecus strykeri). Comparative analyses showed that Asian colobines have an enhanced ability to derive energy from fatty acids and to degrade xenobiotics. We found evidence for functional evolution in the colobine RNASE1 gene, encoding a key secretory RNase that digests the high concentrations of bacterial RNA derived from symbiotic microflora. Demographic reconstructions indicated that the profile of ancient effective population sizes for R. roxellana more closely resembles that of giant panda rather than its congeners. These findings offer new insights into the dietary adaptations and evolutionary history of colobine primates.Knowledge of the patterns and processes underlying the evolution of alternative dietary strategies in nonhuman primates is critical to understanding hominin evolution, nutritional ecology and applications in biomedicine 1 . Colobines, a group of Old World monkeys, serve as an important model organism for studying the evolution of the primate diet because of their adaptation to folivory: they primarily eat leaves and seeds rather than fruits and insects as their major food source. In their specialized and compartmentalized stomachs, colobines allow symbiotic bacteria in the foregut to ferment structural carbohydrates and then recover nutrients by digesting the bacteria 2 . This strategy is similar to that used by other foregut fermenters found in an evolutionarily distantly related group of mammals (for example, artiodactyls). Although a number of primate genomes have been sequenced thus far, high-quality genome sequence information is absent for Asian and African colobines, a key group for elucidating the evolution and adaptation of primates as a whole. Snub-nosed monkeys (Rhinopithecus species) are a group of endangered colobines, which were once widely distributed in Asia but are now limited to mountain forests in China and Vietnam 3 (Supplementary Fig. 1).The golden snub-nosed monkey (GSM, R. roxellana) is recognized as an iconic endangered species in China for its golden coat, blue facial coloration, snub nose and specialized life history. Among its congeners, the black-white snub-nosed monkey (R. bieti), endemic to the Tibetan plateau, has the highest altitudinal distribution (>4,000 m above sea level) of any nonhuman primate. Given the above features and the fact that Rhinopithecus species consume difficult-to-digest foods that contain tannins (for example, leaves and pine seeds), we expected to identify genetic adaptations that enhance the breakdown of toxins, improve the regulation of energy metabolism and facilitate the digestion of symbiotic microbacteria. RESULTS Genomic sequences and the accumulation of...

Snub-nosed monkeys (genus Rhinopithecus) are a group of endangered colobines 2 endemic to South Asia. Here, we re-sequenced the whole genomes of 38 snub-nosed monkeys representing four species within this genus. By conducting population 4 genomic analyses, we observed an similar load of deleterious variation in snub-nosed monkeys living in both smaller and larger populations and found that genomic 6 diversity was lower than that reported in other primates. Reconstruction of Rhinopithecus evolutionary history suggested that episodes of climatic variation over 8 the past 2 million years, associated with glacial advances and retreats and population isolation, have shaped snub-nosed monkey demography and evolution. We further 10 identified several hypoxia-related genes under selection in R. bieti (black snub-nosed monkey), a species that exploits habitats higher than any other nonhuman primate. 12These results provide the first detailed and comprehensive genomic insights into genetic diversity, demography, genetic burden and adaptation in this radiation of 14 endangered primates. 16

To gain an understanding of the genomic structure and evolutionary history of the giant panda major histocompatibility complex (MHC) genes, we determined a 636,503-bp nucleotide sequence spanning the MHC class II region. Analysis revealed that the MHC class II region from this rare species contained 26 loci (17 predicted to be expressed), of which 10 are classical class II genes (1 DRA, 2 DRB, 2 DQA, 3 DQB, 1 DYB, 1 DPA, and 2 DPB) and 4 are non-classical class II genes (1 DOA, 1 DOB, 1 DMA, and 1 DMB). The presence of DYB, a gene specific to ruminants, prompted a comparison of the giant panda class II sequence with those of humans, cats, dogs, cattle, pigs, and mice. The results indicated that birth and death events within the DQ and DRB-DY regions led to major lineage differences, with absence of these regions in the cat and in humans and mice respectively. The phylogenetic trees constructed using all expressed alpha and beta genes from marsupials and placental mammals showed that: (1) because marsupials carry loci corresponding to DR, DP, DO and DM genes, those subregions most likely developed before the divergence of marsupials and placental mammals, approximately 150 million years ago (MYA); (2) conversely, the DQ and DY regions must have evolved later, but before the radiation of placental mammals (100 MYA). As a result, the typical genomic structure of MHC class II genes for the giant panda is similar to that of the other placental mammals and corresponds to BTNL2∼DR1∼DQ∼DR2∼DY∼DO_box∼DP∼COL11A2. Over the past 100 million years, there has been birth and death of mammalian DR, DQ, DY, and DP genes, an evolutionary process that has brought about the current species-specific genomic structure of the MHC class II region. Furthermore, facing certain similar pathogens, mammals have adopted intra-subregion (DR and DQ) and inter-subregion (between DQ and DP) convergent evolutionary strategies for their alpha and beta genes, respectively.

Current understanding of historic climate oscillations that have occurred over the past few million years has modified scientific views on evolution. Major climatic events have caused local and global extinction of plants and animals and have impacted the spatial distribution of many species. The endangered golden snub-nosed monkey (Rhinopithecus roxellana) currently inhabits three isolated regions of China: the Sichuan and Gansu provinces (SG), the Qinling Mountains in Shaanxi province (QL), and the Shennongjia Forestry District in Hubei province (SNJ). However, considerable uncertainty still exists about their historical dispersal routes under the influence of environment change. To date, two dispersal routes have been proposed: (1) the QL and SNJ populations originated from the SG population; and (2) the SG population recolonized from the QL and SNJ populations. We used the mitochondrial DNA complete control region to perform statistical assessments of the relative probability of alternative migration scenarios and the role of environmental change on the geographic dispersal of Rhinopithecus roxellana. Thirty haplotypes were identified from the three geographic regions and a high degree of genetic structure was observed. The most recent common ancestor among the mitochondrial DNA haplotypes was estimated to live around 0.47-1.88 million years ago and five notable haplotype clusters were found. Phylogenetic analysis and historical gene flow estimates suggested that the QL and SNJ populations originated from the SG population, with at least two dispersal events from the SG population occurring during the Pleistocene (1.17±0.70 and 0.53±0.30 Ma). Composite dispersal history of the golden snub-nosed monkey can be explained by both environmental change inducing global climate change and the influence of the Tibetan Plateau uplift. Such range shifts involved considerable demographic changes, as revealed in the dramatic decreases in population size during the last 25,000 years.

The major histocompatibility complex class I genes play crucial roles in the adaptive immune system of vertebrates against intracellular pathogens. To date, no class I genes from the giant panda (Ailuropoda melanoleuca) has been reported, even none from species of Ursidae. In this study, we successfully identified three class I genes from a giant panda bacterial artificial chromosome library and designated them as Aime-128, 152, and 1906, respectively. Pairwise sequence alignments revealed that (1) the Aime-1906 always possessed the lowest identities (52-86%) in different regions compared with the Aime-128 and 152 and (2) the Aime-128 also varied from the Aime-152 in the regions of 5' untranslated region (UTR), 3' UTR, and exon1, whose similarities were 83%, 87%, and 91%, respectively. Comparison of structure characteristics indicated that the Aime-128 possessed all conserved amino acids important to the function of antigen presentation while the Aime-152 and 1906 presented two and five mutated residues. Analysis of phylogenetic trees demonstrated that the Aime-128, 152, and 1906 were clustered into three different branches with 99% or 100% bootstrap values. As a result, these three kinds of evidence supported that the Aime-1906, 152, and 128 should be derived from different loci. Furthermore, in view of a prestop codon in the exon 7 and patterns of amino acid replacement within alleles, the Aime-1906 gene is predicted to be a nonclassical locus, which is most closely related to dog leukocyte antigen 79 in the phylogenetic tree constructed with various mammalian class I loci.