A window into lysogeny: revealing temperate phage biology with transcriptomics

Abstract

Prophages are integrated phage elements that are a pervasive feature of bacterial genomes. The fitness of bacteria is enhanced by prophages that confer beneficial functions such as virulence, stress tolerance or phage resistance, and these functions are encoded by 'accessory' or 'moron' loci. Whilst the majority of phage-encoded genes are repressed during lysogeny, accessory loci are often highly expressed. However, it is challenging to identify novel prophage accessory loci from DNA sequence data alone. Here, we use bacterial RNA-seq data to examine the transcriptional landscapes of five Salmonella prophages. We show that transcriptomic data can be used to heuristically enrich for prophage features that are highly expressed within bacterial cells and represent functionally important accessory loci. Using this approach, we identify a novel antisense RNA species in prophage BTP1, STnc6030, which mediates superinfection exclusion of phage BTP1. Bacterial transcriptomic datasets are a powerful tool to explore the molecular biology of temperate phages.

Keywords: transcriptomics; RNA-seq; bacteriophage; lysogeny; prophage.

Fig. 1.

The transcriptomic landscape of the BTP1 prophage of S . enterica serovar Typhimurium D23580 across 22 different RNA-seq experiments. RNA-seq and dRNA-seq data are from Canals et al. [20] and Hammarlöf et al. [21]. Each coloured horizontal track represents a different RNA-seq condition (Table S2). The upper panel shows sequence reads mapped to the positive strand and the lower panel reads mapped to the negative strand. The dRNA-seq data are shown in red, and were used to identify the TSSs, which are indicated by curved black arrows on the main annotation track. Annotated phage genes are grouped into functional clusters. Non-coding RNAs are represented beneath the main annotation track.

Fig. 2.

The transcriptomic landscape of the Gifsy-2 prophage of S . enterica serovar Typhimurium D23580 across 22 different RNA-seq experiments. RNA-seq and dRNA-seq data are from Canals et al. [20] and Hammarlöf et al. [21]. Each coloured horizontal track represents a different RNA-seq condition (Table S2), the upper panel shows sequence reads mapped to the positive strand and the lower panel reads mapped to the negative strand. The dRNA-seq data are shown in red, and were used to identify the TSSs, which are indicated by curved black arrows on the annotation track. Annotated phage genes are grouped into functional clusters. ncRNAs are annotated with red font. Striped arrows indicate pseudogenes and dotted red lines indicate where ORFs have been disrupted. Due to multiple copies of certain genes, some RNA-seq reads could not be mapped uniquely to the chromosome, these reads were ignored, and so transcriptomic signal is absent from parts of the prophage (e.g. the Gifsy-2 transposase STMMW_10641).

Fig. 3.

The transcriptomic landscape of the ST64B prophage of S . enterica serovar Typhimurium D23580 across 22 different RNA-seq experiments. RNA-seq and dRNA-seq data are from Canals et al. [20] and Hammarlöf et al. [21]. Each coloured horizontal track represents a different RNA-seq condition (Table S2), and the upper panel shows sequence reads mapped to the positive strand and the lower panel reads mapped to the negative strand. The dRNA-seq data are shown in red, and were used to identify the TSSs, which are indicated by curved black arrows on the annotation track. Annotated phage genes are grouped into functional clusters. ncRNAs are annotated with red font. Striped arrows indicate pseudogenes and dotted red lines indicate where ORFs have been disrupted.

Fig. 4.

The transcriptomic landscape of the Gifsy-1 prophage of S . enterica serovar Typhimurium D23580 across 22 different RNA-seq experiments. RNA-seq and dRNA-seq data are from Canals et al. [20] and Hammarlöf et al. [21]. Each coloured horizontal track represents a different RNA-seq condition (Table S2), the upper panel shows sequence reads mapped to the positive strand and the lower panel reads mapped to the negative strand. The dRNA-seq data are shown in red, and were used to identify the TSSs, which are indicated by curved black arrows on the annotation track. Annotated phage genes are grouped into functional clusters. ncRNAs are annotated with red font.

Fig. 5.

The transcriptomic landscape of the BTP5 prophage of S . enterica serovar Typhimurium D23580 across 22 different RNA-seq experiments. RNA-seq and dRNA-seq data are from Canals et al. [20] and Hammarlöf et al. [21]. Each coloured horizontal track represents a different RNA-seq condition (Table S2), and the upper panel shows sequence reads mapped to the positive strand and the lower panel reads mapped to the negative strand. The dRNA-seq data are shown in red, and were used to identify the TSSs, which are indicated by curved black arrows on the annotation track. Annotated phage genes are grouped into functional clusters.

Fig. 6.

Prophage regulatory or accessory genes show unique transcriptional signatures. These findings suggest that transcriptomic data can be used to heuristically enrich for genes likely to be associated with novel regulatory or accessory functions. (a) Genomic map of S . enterica serovar Typhimurium strain D23580 indicating the location of the five prophage elements. Known accessory loci associated with each prophage element are annotated in the exterior grey ring. (b) Functional categorization of all prophage genes with expression values of <100 TPM (not highly expressed in lysogeny) and >100 TPM (highly expressed in lysogeny) in at least one RNA-seq condition. The majority of highly expressed prophage genes have a known regulatory or accessory function, or have no known function. (c) The 40 prophage genes of S . enterica serovar Typhimurium strain D23580 classified as highly expressed during lysogeny.

Fig. 7.

Examples of prophage genes exhibiting unique transcriptional signatures consistent with regulatory or accessory functions. (a) The bstA locus of prophage BTP1. (b) The STMMW_20121 locus of prophage ST64B. (c) The STMMW_26411 locus of prophage Gifsy-1.

Fig. 8.

Transcriptomic-based identification of a novel prophage-encoded asRNA, STnc6030. (a) The transcriptional context of the STnc6030 ncRNA. The transcriptomic data are the same as shown in Fig. 1. (b) Detection of the STnc6030 transcript by Northern blot using an anti-STnc6030 DIG-labelled riboprobe. The two most abundant transcripts detected by the anti-STnc6030 riboprobe are indicated, with the approximate size estimated from the molecular marker.

Fig. 9.

The BTP1 prophage-encoded asRNA STnc6030 functions as a phage-specific superinfection-exclusion factor. (a) Over-expression of the STnc6030 RNA in D23580 WT background does not affect BTP1 spontaneous induction. Plaque assay of overnight culture supernatants of D23580 WT, D23580 pP_L-STnc6030 and D23580 pP_L (empty vector) on host strain D23580 ΔBTP1. Error bars represent the sd of three biological replicates. (b) Heterologous expression of STnc6030 in D23580 ΔBTP1 completely protects against BTP1 phage, but not P22 infection. Plaque assay of BTP1 and P22 phage on D23580 ΔBTP1 strains containing the pP_L-STnc6030 expression plasmid or the negative control plasmid. (c) Isolation of STnc6030-escape mutants of phage BTP1 suggests the STnc6030 functional ‘seed’ region is located at the 3’ end of the transcript. A high titre BTP1 phage stock was used to identify naturally occurring BTP1 phage mutants that were immune to inhibition by STnc6030. Escape phages were estimated to occur at a frequency of approximately 4×10⁻⁸. Five escape phages of varying plaque morphologies were selected for sequencing. The sequence of the STnc6030 region of the escape phages identified SNPs, and the position and substitution are shown. The SNPs that conferred immunity to STnc6030 interference were clustered within a 36 bp region (shown) corresponding to the 3’ end of the STnc6030 transcript and the 3’ end of the STMMW_03891 gene (located between nt 404 041 and 404 078 on the D23580 chromosome; GenBank accession no. FN424405).