Complete Genome Sequence of Algoriphagus sp. PR1, Bacterial Prey of a Colony-Forming Choanoflagellate

Abstract

Bacteria are the primary food source of choanoflagellates, the closest known relatives of animals. Studying signaling interactions between the Gram-negative Bacteroidetes bacterium Algoriphagus sp. PR1 and its predator, the choanoflagellate Salpingoeca rosetta, provides a promising avenue for testing hypotheses regarding the involvement of bacteria in animal evolution. Here we announce the complete genome sequence of Algoriphagus sp. PR1 and initial findings from its annotation.

The marine Bacteroidetes species Algoriphagus sp. PR1 was coisolated with the choanoflagellate Salpingoeca rosetta from mud core samples near Hog Island, VA (13). Bacteroidetes species make up 6 to 30% of the total bacteria in the oceans (4, 11). Furthermore, they play an important role in the global carbon cycle because of their ability to degrade polysaccharides and other macromolecules (6, 8, 9, 22). Of the three clades that constitute the Bacteroidetes phylum (Cytophaga, Flavobacteria, and Bacteroides), the Cytophaga clade, of which Algoriphagus is a member, has been the least studied.

The complete genome sequence of Algoriphagus sp. PR1 was determined using shotgun sequencing, 454 (16), and Illumina technologies (2). Initial assembly of a draft whole-genome shotgun sequence into 12 contigs was generated at the J. Craig Venter Institute (JCVI) based upon 50,413 Sanger sequencing reads from genomic libraries harboring 4-kb and 40-kb fragments. Resequencing of Algoriphagus sp. PR1 was performed at the Broad Institute, and a 30× assembly containing a single gap was generated using the 454 Newbler assembler for 454 data (21) and the Velvet assembler (25) for Illumina data. The remaining gap is small and appears to be contained within a single gene.

The Algoriphagus sp. PR1 genome was found to be a single circular 4.89-Mbp chromosome that is 38.69% GC rich, contains 3,954 predicted genes, and is similar in size to previously sequenced genomes from other marine Bacteroidetes (1, 18-20). Ab initio gene models were generated using GeneMark (3), Glimmer3 (5), and Metagene (17). Predicted genes were generated from BLAST hits to the UniRef90 database, and a synteny-based approach was used to transfer open reading frames (ORFs) from the draft PR1 genome. The final ORF set was derived by comparison of in silico ORFs, ORFs from BLAST hits and mapped ORFs with hits to Pfam (10), and the top BLAST hits against UniRef90. ORFs with overlap relative to noncoding RNA features were removed when appropriate. Discrepancies in the final ORFs were resolved manually. Noncoding features were identified with RNAmmer (14), tRNAScan (15), and RFAM (12). There are 39 tRNAs and 9 rRNA operons. The genome contains genes required for a complete tricarboxylic acid cycle and complete glycolysis and pentose phosphate pathways. Algoriphagus sp. PR1 forms pink-pigmented colonies, and the genome encodes numerous carotenoid biosynthetic enzymes.

Given the capacity of Bacteroidetes bacteria to degrade macromolecules, we catalogued the diversity of carbohydrate-active enzymes in Algoriphagus sp. PR1. We found Algoriphagus sp. PR1 to have 62 glycoside hydrolases, 71 glycosyltransferases, 2 polysaccharide lyases, and 10 carbohydrate esterases, constituting a high capacity for polysaccharide degradation. While the expansion of these groups of enzymes is a characteristic of the Bacteroidetes phylum (1, 7, 23, 24), Algoriphagus sp. PR1 possesses a repertoire more similar to that of gut commensal Bacteroidetes than marine Bacteroidetes, which may in part be related to its interactions with choanoflagellates. The sequencing and annotation of the Algoriphagus sp. PR1 genome provide a foundation for comparative studies of microbe-eukaryote interactions.

Nucleotide sequence accession numbers.

The JCVI genome sequence of Algoriphagus sp. PR1 is available in GenBank under accession number AAXU01000000, and the accession number for the Broad genome sequence is AAXU02000000.