The giant panda genome codes for all necessary enzymes connected with a carnivorous digestive system but lacks genes for enzymes needed to digest cellulose, the principal component of their bamboo diet. cellulose-digesting enzymes and one hemicellulose-digesting enzyme, cellulase, -glucosidase, and xylan 1,4–xylosidase, in group I. Comparing glycoside hydrolase profiles of pandas with those of herbivores and omnivores, we found a moderate abundance of oligosaccharide-degrading enzymes for pandas (36%), close to that for humans (37%), and the lowest abundance of cellulases and endohemicellulases (2%), which may reflect low digestibility of cellulose and hemicellulose in the panda’s unique bamboo diet. The presence of putative cellulose-digesting microbes, in combination with adaptations related to feeding, physiology, and morphology, show that giant pandas have evolved a number of traits to overcome the anatomical and physiological challenge of digesting a diet high in fibrous matter. Access to dietary resources shapes animal evolution (1). Early on, animals lost the ability to synthesize many key compounds, and instead this function is performed by symbionts (2). For example, microbial symbionts assist with extracting nutrients from food and key compounds from the environment, and also synthesize necessary metabolic compounds (1). Gut microbiota share specialized relationships with their hosts, and advances in genomics are revealing the dynamics of these relationships (3). Recent developments in culture-independent methodologies based on large-scale comparative analyses of microbial small-subunit ribosomal RNA genes (16S ribosomal RNA) and metagenomics have revealed the extent of microbial diversity and metabolic potential in greater detail (2C7). These techniques can now be applied to animals that have acquired a profoundly new 138890-62-7 IC50 diet, presenting an opportunity to investigate host physiological and microbial systems in an evolutionary context. The giant panda (and and group I (10 OTUs, 1,457 sequences, 26.4% of the total sequences) and XIVa (3 OTUs, 185 sequences, 3.3% of the total sequences) (and genus (of group I (((TOP10 (Tiangen). Approximately 192C900 colonies from each sample PCR product were chosen at random. Plasmid inserts were sequenced bidirectionally using BigDye Terminator (Applied Biosystems) and vector-specific primers (M13F: 5-GTAAAACGACGGCCAG-3; M13R: 5-CAGGAAACAGCTATGAC-3). Sequences were analyzed with DNASTAR (DNAStar) and trimmed to remove vector sequences. After trimming and adjusting for quality values, the average single sequence read length was 700 nucleotides. Bidirectional sequence reads of clone inserts provided nearCfull-length 16S rRNA gene sequences of 1 1,400 bp. Chimera Checking. Each sequence was edited 138890-62-7 IC50 manually in conjunction with its chromatogram and secondary structure information. Only unambiguous nucleotide positions were included in the analysis, and primer sequences were excluded. A multiple sequence alignment was generated with the NAST online tool (27), and chimeras were identified with Bellerophon version 3 (28), implemented at the Greengenes Web site (http://greengenes.lbl.gov) 138890-62-7 IC50 with the following (default) parameters: Sequences were compared with others within the same host species and with the Greengenes Core Set, identity to the core set was set to 97%, the match length to sequence threshold was set to 1 1,050 bp and 1,150 bp, respectively, the window size was set to 300, the count of similar sequences to search for each window was 7, the parent-to-fragment ratio 138890-62-7 IC50 was 90%, and Rabbit Polyclonal to RPL39L the divergence ratio threshold was set at 1.1. Determining OTU and Taxonomy Assignments. Clusters. We downloaded sequences from GenBank according to known clusters (15). We then combined these sequences with all of the OTUs in this study and constructed the neighbor-joining phylogenetic tree (1,000 bootstraps) using MEGA4 (29). Based on phylogenetic information (in the same clade), we determined the relationships between all OTUs and known clusters. Estimation of Microbial Diversity. We calculated the Good coverage estimation as [1 ? (may be the amount of singleton sequences and may be the total.