Analysis of rpoS mRNA in Salmonella dublin: identification of multiple transcripts with growth-phase-dependent variation in transcript stability

Abstract

In Salmonella dublin, rpoS encodes an alternative sigma factor of the RNA polymerase that activates a variety of stationary-phase-induced genes, including some virulence-associated genes. In this work, we studied the regulation and transcriptional organization of rpoS during growth. We found two transcripts, 2.3 and 1.6 kb in length, that represent the complete rpoS sequence. The 2.3-kb transcript is a polycistronic message that also includes the upstream nlpD gene. It is driven by a weak promoter with increasing activity when cells enter early stationary growth. The 1.6-kb message includes 566 bp upstream of the rpoS start codon. It is transcribed from a strong sigma70 RNA polymerase-dependent promoter which is independent of growth. The decay of this transcript decreases substantially in early stationary growth, resulting in a significant net increase in rpoS mRNA levels. These levels are approximately 10-fold higher than the levels of the 2.3-kb mRNA, indicating that the 1.6-kb message is mainly responsible for RpoS upregulation. In addition to the 2.3- and 1.6-kb transcripts, two smaller 1.0- and 0.4-kb RNA species are produced from the nlpD-rpoS locus. They do not allow translation of full-length RpoS; hence their significance for rpoS regulation remains unclear. We conclude that of four transcripts arising from the nlpD-rpoS locus, only one plays a significant role in rpoS expression in S. dublin. Its upregulation when cells enter stationary growth is due primarily to an increase in transcript stability.

FIG. 1

Comparison of the levels of RpoS protein (A) and rpoS mRNA (B) along the growth curve. Arrows mark the protein and mRNA main band, respectively. The samples were standardized to equal amounts of total cellular protein and total RNA, respectively.

FIG. 2

(A) Comparison of the nlpD promoter regions of S. dublin and E. coli (21). The two transcriptional start sites in E. coli and the translational start site are shaded; the −10 and −35 regions of the two promoters determined in E. coli are underlined; the corresponding sequences in S. dublin are marked. The −10 region of the promoter P1 is absent in S. dublin. (B) Determination of the 5′ end of the 1.6-kb main transcript by primer extension. The arrow shows the transcriptional start site. The −10 and −35 regions of the putative ς⁷⁰ RNA polymerase-dependent promoter are also marked. Lanes A, C, G, and T represent the sequencing ladder using the same primer as used for the primer extension reaction.

FIG. 3

Mapping of the different rpoS mRNA transcripts. (A) The probes (A to E) used for the mapping. (B) Band patterns with the different probes. Lanes A to E correspond to probes A to E; bands 1 to 4 corresponds to the 2.3-, 1.6-, 1.0-, and 0.4-kb transcripts. (After longer exposure, the 2.3-kb band was also visualized in lanes D and E.) (C) Alignment of the identified transcripts to the DNA sequence. The arrowhead defines the 3′ end of the 1.6-kb main transcript.

FIG. 4

Transcriptional activities of the nlpD and rpoS promoters. The activities were monitored by measuring the β-galactosidase activities from the transcriptional lacZ fusion vectors pRSNL (■) and pRSP70 (⧫), respectively, and are presented as percentages of the basal transcription level during early exponential phase. The dashed line represents the OD₆₀₀ during growth. Each measurement was performed at least three times.

FIG. 5

Stability of the rpoS mRNA main transcript along the growth curve. The half-life of the 1.6-kb rpoS mRNA was measured at OD₆₀₀ 0.3, 0.8, and 2.3. (A) Relative amounts of rpoS mRNA after addition of rifampin depicted on a semilogarithmic graph. (B) Calculated half-lives (in minutes) and a typical Northern blot. Each experiment was carried out two to four times.