Antisense RNA targeting of primase interferes with bacteriophage replication in Streptococcus thermophilus

Abstract

The putative primase gene and other genes associated with the Sfi21-prototype genome replication module are highly conserved in Streptococcus thermophilus bacteriophages. Expression of antisense RNAs complementary to the putative primase gene (pri3.1) from S. thermophilus phage kappa 3 provided significant protection from kappa 3 and two other Sfi21-type phages. Expression of pri3.10-AS, an antisense RNA that covered the entire primase gene, reduced the efficiency of plaquing (EOP) of kappa 3 to 3 x 10(-3) and reduced its burst size by 20%. Mutant phages capable of overcoming antisense inhibition were not recovered. Thirteen primase-specific antisense cassettes of different lengths (478 to 1,512 bp) were systematically designed to target various regions of the gene. Each cassette conferred some effect, reducing the EOP to between 0.8 and 3 x 10(-3). The largest antisense RNAs (1.5 kb) were generally found to confer the greatest reductions in EOP, but shorter (0.5 kb) antisense RNAs were also effective, especially when directed to the 5' region of the gene. The impacts of primase-targeted antisense RNAs on phage development were examined. The expression of pri3.10-AS resulted in reductions in target RNA abundance and the number of phage genomes synthesized. Targeting a key genome replication function with antisense RNA provided effective phage protection in S. thermophilus.

FIG. 1.

The antisense RNA expression vectors used in this study that contain the P6 promoter cassette cloned in opposite orientations relative to the cat194 gene. (A) pTRK687; (B) pTRK696. Abbreviations: T7, coliphage T7 transcription terminator; P6, L. acidophilus P6 promoter; repBCA, genes encoding plasmid replication factors; cat194, chloramphenicol resistance gene; oriSH71, plasmid origin of DNA replication; pri3.1, phage κ3-derived putative primase. Restriction endonuclease recognition sites: E, EcoRI; S, Sau3AI; P, PstI.

FIG. 2.

Schematic of the pri3.1 gene (arrow) and upstream putative ribosome binding site (small vertical black bar). The boundaries of the subcloned fragments are shown as bars demarcated with dotted lines. Base pair (bp) coordinates are numbered relative to the 5′ position of the primase translation initiation codon.

FIG. 3.

(A) Plaque size of phage κ3 when titered on NCK1125 (left) and NCK1125(pTRK687) (right). (B) Comparison of primase-derived antisense RNA constructs on the EOP and plaque size of phage κ3 relative to that of the plasmid control, pTRK687. Horizontal and vertical dotted lines represent an EOP of 1.0 and a plaque size of 4 mm, respectively. The error bars for EOP were calculated from the results of three independent experiments.

FIG. 4.

The effect of various phages on the growth of NCK1125(pTRK687) (•) and NCK1125 (pTRK790::pri3.10-AS) (▪). (A) Growth in the absence of phage. Growth in the presence of phage κ3 (B); κ5 (C); κ9 (D); and κ6 (E). The multiplicities of infection were 0.1% ± 20% for all phages.

FIG. 5.

RNA-RNA slot blot hybridization of three antisense RNA-expressing constructs and the vector control strain NCK1125(pTRK687) with detect-S, a pri3.12-derived single-stranded RNA probe that was complementary to the sense strand of the phage κ3-encoded primase mRNA (Fig. 2). RNA was isolated from cells in the absence (top panel) or presence (bottom panel) of phage κ3.

FIG. 6.

The impact of antisense RNA expression on the in vivo accumulation of HindIII-digested phage κ3-specific DNA fragments over time. (A) Vector control strain NCK1125(pTRK687); (B) NCK1125(pTRK789::pri3.8-AS); (C) NCK1125(pTRK790::pri3.10-AS). Each plasmid encoded two HindIII sites; plasmid-specific DNA bands are demarked with black triangles. Abbreviations: φ, purified phage κ3 genomic DNA; M, 1-kb molecular size marker (GIBCO-BRL).