A new piece of the Shigella Pathogenicity puzzle: spermidine accumulation by silencing of the speG gene [corrected]

Abstract

The genome of Shigella, a gram negative bacterium which is the causative agent of bacillary dysentery, shares strong homologies with that of its commensal ancestor, Escherichia coli. The acquisition, by lateral gene transfer, of a large plasmid carrying virulence determinants has been a crucial event in the evolution towards the pathogenic lifestyle and has been paralleled by the occurrence of mutations affecting genes, which negatively interfere with the expression of virulence factors. In this context, we have analysed to what extent the presence of the plasmid-encoded virF gene, the major activator of the Shigella regulon for invasive phenotype, has modified the transcriptional profile of E. coli. Combining results from transcriptome assays and comparative genome analyses we show that in E. coli VirF, besides being able to up-regulate several chromosomal genes, which potentially influence bacterial fitness within the host, also activates genes which have been lost by Shigella. We have focused our attention on the speG gene, which encodes spermidine acetyltransferase, an enzyme catalysing the conversion of spermidine into the physiologically inert acetylspermidine, since recent evidence stresses the involvement of polyamines in microbial pathogenesis. Through identification of diverse mutations, which prevent expression of a functional SpeG protein, we show that the speG gene has been silenced by convergent evolution and that its inactivation causes the marked increase of intracellular spermidine in all Shigella spp. This enhances the survival of Shigella under oxidative stress and allows it to better face the adverse conditions it encounters inside macrophage. This is supported by the outcome of infection assays performed in mouse peritoneal macrophages and of a competitive-infection assay on J774 macrophage cell culture. Our observations fully support the pathoadaptive nature of speG inactivation in Shigella and reveal that the accumulation of spermidine is a key determinant in the pathogenicity strategy adopted by this microrganism.

Figure 1. Superpathway of polyamine biosynthesis I in E. coli and Shigella spp.

Schematic diagram depicting the pathway of polyamine biosynthesis I in E. coli. Steps bounded by the dashed lines are conserved in Shigella spp. The step enclosed by dotted lines is absent in S. boydii. Data were drawn according to http://ecocyc.org.

Figure 2. VirF positively controls the ynfB-speG operon.

A. β-galactosidase activity of the SpeG-LacZ fusion carried by plasmid pULS7 was determined in E. coli ULS153 in the presence of pMYSH6504, a plasmid containing a functional S. flexneri virF gene, or of its virF-depleted variant pMY6504R. Cells were grown at 30°C or 37°C in LB medium and assayed for β-galactosidase at OD₆₀₀ 0.5–0.6. The values reported are expressed in Miller Units and represent the average ± standard deviation of at least 3 independent experiments. B. The in vivo ynfB-speG transcription was monitored by real-time PCR in S. flexneri 2457T and its virF defective derivative 2457TFd. Strains were grown at 37°C in LB medium. At least three wells were run for each sample and the error bars display the calculated maximum (RQMax) and minimum (RQMin) expression levels that represent standard error of the mean expression level (RQ value).

Figure 3. Inactivation of the ynfB-speG locus in Shigella spp. has been attained by convergent evolution.

The operon on the top is based on the E. coli K12 MG1655 sequenced (http://www.ncbi.nlm.nih.gov/genome). Arrows indicate the orientation of ynfB and speG genes. Point mutation, leading to the Il4P substitution, within the ynfB gene found in most S. flexneri and S. dysenteriae strains analysed is indicated by a white cross. Point mutation, leading to the S56R substitution, detected in some S. dysenteriae strains is indicated by a black cross. All the S. flexneri strains show the interruption of SpeG coding sequence due to a stop codon (7^th) resulting from a dinucleotide (GT) deletion indicated by two full stops. The comparative analysis of speG sequences from S. boydii strains shows three different structures that may result from at least two deletion steps, which are schematized by dashed lines. Finally, the ynfB-speG locus of S. sonnei is not reported since it has been completely lost.

Figure 4. The absence of a functional speG gene in Shigella increases survival to the oxidative stress.

Effect of hydrogen peroxide on S. flexneri wild type strain M90T (top left), on its speE-defective derivative M90TEd (top right) and on M90T complemented with the entire ynfB-speG operon (plasmid pULS37, bottom right) or only the speG gene (plasmid pULS55, bottom left). A clear difference can be appreciated in the halo of inhibition around the paper disk soaked with 5 µl of H₂O₂ 30 w.t. % sol. in water (Sigma-Aldrich); sectors of four agar plates are shown at the same enlargement.

Figure 5. Spermidine involvement in response to oxidative stress in a S. flexneri background.

S. flexneri M90T and its derivatives unable to synthesize spermidine (M90TEd) or carrying either the entire ynfB-speG operon (M90T pULS37) or only the promoter proximal ynfB gene (M90T pULS13) were grown in LB (panel A) or in M9 minimal medium (panel B) in the presence of H₂O₂. Survival is expressed as the percentage relative to the S. flexneri M90T wild type strain. The polyamine content of S. flexneri strains, obtained by HPLC analysis, is reported in the tables below panels A and B. Values are expressed as nmol/mg of protein. Panel C: S. flexneri M90TEd was grown in polyamine-free medium (M9) supplemented with increasing amounts of exogenous spermidine. Survival is expressed as the percentage relative to the M90TEd strain grown in spermidine-free medium (set to 100%). Error bars display the standard deviations relative to at least three independent experiments.

Figure 6. Loss of speG confers Shigella an increased fitness within murine peritoneal macrophages.

Time course of intracellular survival within murine peritoneal macrophages of S. flexneri M90T and its derivatives complemented either with the entire ynfB-speG operon (plasmid pULS37) or with the ynfB gene (plasmid pULS55). The data are the average ± standard deviation of the number of viable intracellular bacteria per 10⁵ macrophages from three independent experiments each in triplicate. ▴, S. flexneri M90T; ▪ M90T pULS37; • M90T pULS55.