Prior to the completion of the genomic sequence of L. monocytogenes [2] our knowledge of the PrfA regulon was very limited. At that time, the only known genes that did not belong to the virulence gene cluster but were also regulated by PrfA were inlAB, [111] and inlC [7]. The availability of the complete listerial genome sequence allowed the in silico screening of the sequence for genes preceded by putative PrfA boxes. This screening identified four additional previously unknown genes harboring PrfA boxes. One of the genes, lmo0838, later called hpt, codes for a putative hexose permease, the expression of which was shown to be strictly PrfA-dependent [66]. The other three genes preceded by PrfA boxes are genes of unknown function.

A recent systematic approach aiming to elucidate the complete PrfA regulon used a whole-genome macroarray carrying PCR products of the 2853 ORFs of the L. monocytogenes EGD-e genome [112] and compared the expression profiles of the wild-type strain and a prfA-deletion mutant. With this approach three groups of differently regulated genes were identified. Group I comprises, in addition to the 10 already known PrfA-regulated genes, two new genes, both positively regulated and preceded by a PrfA box. Group II comprises eight negatively regulated genes, one of which is preceded by a PrfA box and the others form an operon. Group III comprises 53 genes, of which only two are preceded by a PrfA box and which are either activated or repressed under different conditions; most of the genes in this group are transcribed from sigma B-dependent promoters. Taken together, the results suggest that PrfA, on the one hand, may positively regulate a core set of 12 genes, preceded by a PrfA box, which are probably expressed from sigma A-depen-dent promoters, and on the other hand, negatively regulates eight genes. A second set of PrfA-regulated genes lacks PrfA boxes and is expressed from sigma B-dependent promoters. These data reveal that PrfA can act either as an activator or as a repressor, and suggest that PrfA may directly or indirectly activate sets of genes in association with different sigma factors [112]. However, several promoters of these additional genes that were affected either positively or negatively by PrfA, when tested in an in vitro transcription system where all truly PrfA-depen-dent virulence gene promoters showed similar PrfA-dependent transcription as in vivo, did not yield PrfA-dependent transcripts in the presence or absence of SigB [57]. It is therefore more likely that the in vivo observed influence of PrfA on the transcription of these genes is indirect.

Besides the above-mentioned sigma factors SigA [RpoD, r43 (lmo1454)] and SigB [RpoF, r37 (lmo0895)], three further sigma factors were identified in the L. monocytogenes genome, namely SigH [RpoH, r30 (lmo0243)], SigL [RpoN, r54 (lmo2461)], and an ECF-type sigma factor (lmo0423) [2]. Whereas sigA is an essential gene, deletion mutants in the other four sigma factor-encoding genes were obtained; disruption of the gene coding for the alternative sigma factor SigH resulted in a mutant that demonstrated reduced growth potential in minimal medium but was without a defect in the infectious process [113]; target genes of SigH are, however, unknown to date.

The role of the alternative r factor, encoded by the rpoN gene, was investigated by comparing the global gene expression of the wild-type EGD-e strain and an rpoN mutant [114]. Gene expression using the whole-genome macroarrays mentioned before [112] identified 77 genes whose expressions were modulated in the rpoN mutant as compared to the wild-type strain. Most of the changes in gene expression were related to carbohydrate metabolism, and in particular to pyruvate metabolism. However, (a) further analyses showed that only the mptACD operon was directly controlled by r and (b) in silico analysis suggested that r may directly control the expression of four different phosphotransferase system (PTS) operons, including mptACD. RpoN (r ) is hence mainly involved in the control of carbohydrate metabolism in L. monocytogenes [114].

SigB-dependent genes in L. monocytogenes were identified in a coupled bioinfor-matics/microarray strategy [115]: first, candidate SigB-dependent promoters were searched for in the EGD-e sequence by biocomputing and the data generated were used to develop a specialized 208-gene microarray which included 166 genes downstream of the predicted SigB-dependent promoters as well as selected virulence and stress response genes. This array was hybridized with RNA from WT and a DsigB--mutant which resulted in the identification of more than 50 clearly SigB-dependent genes including both stress response genes (e.g., gadB, ctc, and the glutathione reductase gene lmo1433) and virulence genes (e.g., inlA, inlB, and bsh). These data demonstrate that SigB not only regulates the expression of genes important for survival under environmental stress conditions but also contributes to regulation of virulence gene expression in L. monocytogenes [115].

An ECF (lmo0423) deletion mutant showed unaltered virulence in a mouse sepsis model and a basically unchanged gene expression pattern compared to the wild-type strain when grown under aerobic conditions in a rich culture medium

(BHI). Growth retardation of the ECF mutant was, however, observed under microaerophilic conditions, suggesting that genes essential for the metabolism under these conditions may be under the control of this ECF sigma factor (M. Rauch and W. Goebel, unpublished results).

0 0

Post a comment