A freshwater cyanophage whose genome indicates close relationships to photosynthetic marine cyanomyophages

Abstract

Bacteriophage S-CRM01 has been isolated from a freshwater strain of Synechococcus and shown to be present in the upper Klamath River valley in northern California and Oregon. The genome of this lytic T4-like phage has a 178,563 bp circular genetic map with 297 predicted protein-coding genes and 33 tRNA genes that represent all 20-amino-acid specificities. Analyses based on gene sequence and gene content indicate a close phylogenetic relationship to the 'photosynthetic' marine cyanomyophages infecting Synechococcus and Prochlorococcus. Such relatedness suggests that freshwater and marine phages can draw on a common gene pool. The genome can be considered as being comprised of three regions. Region 1 is populated predominantly with structural genes, recognized as such by homology to other T4-like phages and by identification in a proteomic analysis of purified virions. Region 2 contains most of the genes with roles in replication, recombination, nucleotide metabolism and regulation of gene expression, as well as 5 of the 6 signature genes of the photosynthetic cyanomyophages (hli03, hsp20, mazG, phoH and psbA; cobS is present in Region 3). Much of Regions 1 and 2 are syntenic with marine cyanomyophage genomes, except that a segment encompassing Region 2 is inverted. Region 3 contains a high proportion (85%) of genes that are unique to S-CRM01, as well as most of the tRNA genes. Regions 1 and 2 contain many predicted late promoters, with a combination of CTAAATA and ATAAATA core sequences. Two predicted genes that are unusual in phage genomes are homologues of cellular spoT and nusG.

Fig. 1. Distribution of S-CRM01 in the Klamath River valley, 2009

Positive identifications were made in samples from the Williamson River delta area of Upper Klamath Lake (WRD; 15 October), Copco Reservoir (near dam; 13 October), Iron Gate Reservoir (IG, near dam, 18 August & 15 September), and at two sites on the lower Klamath River: Brown Bear sampling site at Horse Creek (BB, 6 August & 15 September) and Seiad Valley (SV, 15 September).

Fig. 2

Electron micrograph of S-CRM01 phage particles negatively stained with phosphotungstic acid. An intact particle is shown at left and a contracted particle at right. Note the icosahedral head, prominent neck, two-ringed baseplate and injection tube.

Fig. 3. Circular genetic map showing the gene organization of phage S-CRM01

The genome can be considered as comprised of Regions 1-3 (outer ring) that are dominated by structural genes, replication-related genes, and unique genes, respectively. Subsequent rings represent: positive strand ORFs, negative strand ORFs, tRNAs (maroon), and positive mass spectrometry identification (mauve). The inner rings show GC content around a 50% mid-line (black), and GC skew (G+ = cyan, C+ = blue). Putative ORF function is described by color as listed below the circular map.

Fig. 4. Extensive synteny between S-CRM01 and marine cyanomyophages

A. Dot plots comparing gene order between S-CRM01 and phage T4 and indicated cyanomyophage genomes. The S-CRM01 genome regions 1-3 (see Fig. 3) are indicated by colored shading and are labeled at right. Syntenous segments I-IV are gp13–gp46 (SCRM01 g29–g55), gp5-td (g59–g80), nrdB-gp45 (g116–g156), and g166–167 (cobS), respectively. B. Diagram indicating the location of the inversion between the S-CRM01 and S-PM2 genomes. Note that the P-SSM2 and S-PM2 genomes have insertions between syntenous segments II and III, and that P-SSM2 and Syn9 genome numbering convention places the regions syntenous with Region 1C at the start of the genomes.

Fig. 5. Phylogenetic relationship of S-CRM01 to other myophages

A. Consensus tree for the most highly supported topology across all tested values of included non-universal genes (missing data). Values at nodes indicate the average bootstrap value for all trees with that node/the number of trees for which that node occurred out of a total of 27 trees (see Materials and Methods). The marine cyanomyoviruses have been grouped into the clades MC1-4 as indicated. B and C. Consensus trees showing alternative topologies among clades MC1-3 when including 0–10% and 15–40% included non-universal genes, respectively. D. The number of orthologous clusters analyzed for each phage as a function of % included non-universal genes are indicated and color-matched. The genes used for each level of analysis are indicated in Table S2.

Fig. 6. Late promoter consensus sequence

Weblogo representation of the sequences of 81 predicted late promoters (Table S3), indicating the prevalence of both CTAAATA and ATAAATA core sequences. Nucleotide 0 marks the predicted start of transcription based on T4 late promoters (Miller et al., 2003a).