Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Distinct Mutations Led to Inactivation of Type 1 Fimbriae Expression in Shigella spp.

Distinct Mutations Led to Inactivation of Type 1 Fimbriae Expression in Shigella spp.

  • Verónica Bravo, 
  • Andrea Puhar, 
  • Philippe Sansonetti, 
  • Claude Parsot, 
  • Cecilia S. Toro
PLOS
x

Abstract

Shigella spp. are responsible for bacillary dysentery in humans. The acquisition or the modification of the virulence plasmid encoding factors promoting entry of bacteria into and dissemination within epithelial cells was a critical step in the evolution of these bacteria from their Escherichia coli ancestor(s). Incorporation of genomic islands (GI) and gene inactivation also shaped interactions between these pathogens and their human host. Sequence analysis of the GI inserted next to the leuX tRNA gene in S. boydii, S. dysenteriae, S. flexneri, S. sonnei and enteroinvasive E. coli (EIEC) suggests that this region initially carried the fec, yjhATS and fim gene clusters. The fim cluster encoding type I fimbriae is systematically inactivated in both reference strains and clinical isolates and distinct mutations are responsible for this inactivation in at least three phylogenetic groups. To investigate consequences of the presence of fimbriae on the outcome of the interaction of Shigella with host cells, we used a S. flexneri strain harboring a plasmid encoding the E. coli fim operon. Production of fimbriae by this recombinant strain increased the ability of bacteria to adhere to and enter into epithelial cells and had no effect on their ability to disseminate from cell to cell. The observations that production of type I fimbriae increases invasion of epithelial cells and that independent mutations abolish fimbriae production in Shigella suggest that these mutations correspond to pathoadaptive events.

Introduction

Members of Shigella spp. and enteroinvasive Escherichia coli (EIEC) are responsible of bacillary dysentery, a major cause of diarrheal diseases in humans [1], [2]. Bacteria invading, multiplying and disseminating in the colonic epithelium induce an acute inflammation of the colon. The Mxi-Spa type III secretion (T3S) system promoting entry of bacteria into epithelial cells and the outer membrane protein IcsA promoting the actin-based motility of intracellular bacteria and their dissemination from cell to cell are encoded by a 220-kb virulence plasmid [3], [4]. The acquisition of the virulence plasmid, or its construction from elements of various origins, was a critical step in the evolution of Shigella spp. from their E. coli ancestor(s). Further, incorporations of genomic islands (GI) encoding such pathogenicity factors as the Shiga toxin of S. dysenteriae [5], the aerobactin transport system [6] and the SigA and PicA proteases [7] also shaped the interactions between Shigella and its host [8], [9].

Among Shigella and EIEC strains, population genetic studies identified six phylogenetic groups (S1, S2, S3, SS, SD1 and A) interspersed within E. coli phylogenetic groups (A, B1, B2, D and E) [10]–[12]. As compared to E. coli, Shigella contains a large number of inactivated genes [13], [14]. These genes might have been inactivated either because they were no longer useful to bacteria following the acquisition of the ability to invade the mucosa, or because their expression was detrimental to the multiplication and survival of bacteria in this new environment [15]. The detection of distinct mutations leading to the inactivation of the same gene (or pathway) in different lineages is suggestive of pathoadaptive mutations [16], [17]. In addition, the analysis of strains in which expression of a gene of interest is experimentally reactivated can provide evidence in support of pathoadaptive mutations. For example, the cad cluster involved in production of cadaverine and the nadA and nadB genes involved in the synthesis of quinolinate are inactivated in Shigella spp. and experimental restoration of the production of these compounds attenuated the virulence of recombinant strains [18]–[20].

Type 1 fimbriae are filamentous surface structures produced by several members of the Enterobacteriaceae family [21]. These fimbriae are encoded by the fimAICDFGH operon containing genes required for their assembly and structure [22]–[24]. FimA is the major structural subunit of fimbriae, FimI is required for fimbriae biosynthesis although its exact role is not known, FimC is the periplasmic chaperone for fimbriae subunits, FimD is the outer-membrane assembly platform, FimF and FimG are adaptor proteins and FimH is the adhesin located at the tip of fimbriae and mediating adhesion of bacteria to mannose containing molecules on host mucosal surfaces [25]–[28]. Phase-variable expression of fimbriae is mediated by the inversion of the 314-bp invertible element fimS containing the promoter of the fim operon; inversion is controlled by the two site-specific tyrosine recombinases FimE and FimB encoded by genes located upstream from fimS [29], [30]. FimB promotes inversion of fimS in both directions, while FimE catalyzes the ON-to-OFF inversion. The fim gene cluster is present within the GI adjacent to the tRNA leuX gene in E. coli.

In the present work, we investigated the role of type I fimbriae in the invasive phenotype of Shigella spp. Sequence analysis of published genomes and whole genome shotgun (WGS) data of representative members of Shigella spp. and EIEC strains and PCR analysis of S. flexneri clinical isolates indicated that the fim cluster is inactivated in all strains and that different mutations are responsible for its inactivation in at least three phylogenetic groups. To analyze the behavior of a fimbriated S. flexneri strain in in vitro models of infection, we used a recombinant S. flexneri strain harboring a plasmid encoding the E. coli fim cluster. Production of fimbriae led to a 50-fold increase in the ability of bacteria to adhere to and invade epithelial cells. The fact that these pathogens lost fimbriae expression in spite of the superior capacity to invade host cells conferred by fimbriae suggests that inactivation of fimbriae production represents a new pathoadaptive event in Shigella spp.

Materials and Methods

Bacterial strains, plasmids and media

The S. flexneri strain M90T (serotype 5a) was used as the wild-type reference strain [31]. The strain BS176 is a noninvasive derivative of M90T cured of the virulence plasmid pWR100 [31], [32]. The plasmid pSH2 harboring the fim operon from the uropathogenic E. coli strain J96 [22] is a derivative of the vector pACYC184. The E. coli fim operon exhibits 97% sequence identity with the fim operon present at the leuX locus of the S. flexneri strain 2457T. Clinical isolates were collected from stool samples of Chilean children aged <14 years with acute diarrhea between 2004 to 2006 [33]. S. flexneri reference and clinical strains, as well as S. sonnei strains, were grown in trypticase soy broth medium (TCS) containing chloramphenicol (25 μg ml-1) when appropriate. To enrich the population of bacteria expressing fimbriae, bacteria were first passaged twice in liquid medium under static conditions at 37°C for 24 h. Then, to screen for bacteria still harboring the virulence plasmid, bacteria were isolated on agar plates containing Congo red. Bacteria from red colonies were passaged again in liquid medium under static conditions for 24 h, sub-cultured at 37°C for 3–4 h with agitation and harvested by centrifugation (3,000 x g for 10 min at 4°C). For adhesion and invasion assays, bacteria were washed with PBS and suspended in Dulbecco Modified Earl’s Medium (DMEM) containing glucose (1 g l-1) and HEPES, pH 7.4, (20 mM). The orientation of the fim promoter in bacteria used for infections was monitored by restriction analysis of a PCR product covering fimS (see below).

Genome and WGS accession numbers

Strains for which genomes sequences and WGS results were used in this study are the following: E. coli MG1655, NC_000913.3 [34]; EIEC 53638 (serotype O144), AAKB02000001.1; S. boydii 4444–74 (serotype 2), NZ_AKNB1000000; S. boydii Sb227 (serotype 4), CP000036 [35]; S. boydii 3594–74 (serotype 4), NZ_AFGC000000.1; S. boydii strains CIP52-54 (serotype 7), ERR200284 (F.-X. Weill, personal communication); S. boydii CIP56-18 (serotype 11), ERR200299 (F.-X. Weill, personal communication); S. boydii CDC3083-94 (serotype 18), NC_010658; S. dysenteriae Sd197 (serotype 1), CP000034 [35]; S. dysenteriae 155–74 (serotype 2), NZ_AFFZ01000000; S. flexneri 2457T (serotype 2a), AE014073 [13]; S. flexneri CP301 (serotype 2a), AE005674 [36]; S. flexneri NCTC1 (serotype 2a), LM651928 [37]; S. flexneri M90T (serotype 5a), CM001474.1 [38]; S. flexneri Sf8401 (serotype 5b), CP000266 [39]; S. flexneri CFSAN027317 (ATCC 12025) (serotype 6), JWSL01000004.1; S. flexneri CCH60 (serotype 6), AKMW01000080.1; S. flexneri CDC 796–83 (serotype 6), AERO01000069.1; S. flexneri collection of 59 strains covering 14 serotypes (1a, 1b, 1d, 2a, 2b, 3a, 3b, 4a, 4av, 4b, X, Xv, Y and Yv), AZOG00000000.1 to AZQM00000000.1 [40]; S. flexneri collection of the 16 type strains covering all the established serotypes (1a, 1b, 1c, 2a, 2b, 3a, 3b, 3c, 4a, 4b, 5a, 5b, 6, X, Y and E1037), ERS088060-ERS088076 [41]; S. sonnei Ss046, CP000038 [35]; S. sonnei strain 53G, HE616528.

Characterization of the leuX GI in clinical isolates

The presence and the organization of the GI at the leuX locus were characterized by tiling-PCR and restriction analysis. The genomic DNA was purified using the Genomic DNA Purification Kit from Promega®. Primers (Table 1) used to amplify 6 fragments covering the whole GI were designed according to the sequence of the S. flexneri 2a strains 2457T. PCR fragment 3 containing fimB was analyzed by using the restriction enzyme BfaI (Fermentas). Sizes of BfaI fragments obtained from most strains were 1862, 1583 and 258 bp; the presence of the 258-bp fragment indicates the presence of a mutation at codon 162 of fimB. In strain 2457T, the 1583-bp fragment was replaced by 1392- and 191-bp fragments, due to the presence of an additional BfaI site at codon 130 of fimA.

Determination of the orientation of the fim promoter region

The orientation of the invertible DNA element fimS carrying the fim promoter was determined as previously described [42]. Briefly, primers Inv-1 and Inv-2 (hybridizing to each side of fimS) were used to amplify a 601-bp DNA fragment that was digested with SnaBI (Fermentas) and restriction fragments were resolved on 2% agarose gels. Due to the asymmetric location of the SnaBI cleavage site within the invertible element, different restriction fragments were obtained depending on the element orientation: the phase ON orientation yielded fragments of 403 and 198 bp whereas the phase OFF orientation yielded fragments of 440 and 161 bp.

Electron Microscopy

To visualize type 1 fimbriae on the bacterial surface, a drop of bacterial suspensions was placed onto a cupper electron microscope grid for 15 minutes at 20°C. The suspension was removed by absorption with a filter paper and the sample was fixed with a solution containing 5% formaldehyde and stained with 0.5% uranyl acetate prior to examination. Samples were analyzed on Jeol 1200 EX set at 80 kV.

Haemagglutination assay

The presence of functional type 1 fimbriae on the bacterial surface was tested using a modified protocol of the mannose-sensitive haemagglutination assay (MSHA) on Guinea pig erythrocytes [43], [44]. Briefly, equal volumes of a suspension containing 2×109 bacteria per ml in PBS and a suspension of 5% Guinea pigs erythrocytes were mixed in the absence or in the presence of 0.2 mM mannose and agglutination was observed after 20 min of gentle shaking at 20°C.

Adhesion, invasion and dissemination assays

HeLa and TC7 cells were grown in a humidified incubator at 37°C with 10% CO2 in DMEM (Hyclone®) supplemented with 10% fetal calf serum (FCS) (Hyclone®) and non-essential amino acids. HeLa cells (2x105 cells) were grown in 12-well culture plates overnight to obtain a semi-confluent monolayer. For adhesion assays, HeLa cells were infected at a multiplicity of infection (MOI) of 100 bacteria per cell and plates were centrifuged at 180 x g for 10 min and incubated at 37°C for 30 min in the presence or in the absence of 0.2 mM mannose. Infected cells were washed five times with PBS and lysed in PBS containing sodium deoxycholate (0.1%, w/v). Cell-associated bacteria, both adherent onto the cell surface and intracellular, were quantified by plating dilutions of lysates onto LB agar plates. Data are the means and SD of three independent experiments performed in duplicate. In parallel, infected cells were fixed with ethanol and stained with Giemsa. The ability of bacteria to enter and disseminate within a differentiated monolayer of intestinal epithelial cells was evaluated using the plaque assay [45]. Briefly, TC7 cells were seeded into six-well plates and incubated at 37°C for 72 hours. Cells were infected at different MOIs (0.1, 0.01, 0.001 and 0.0001) in the absence or in the of presence 0.2 mM mannose and incubated for 2 h at 37°C without any centrifugation step. Infected cells were then washed, overlaid with DMEM containing agarose (0.5%, w/v), calf serum (10%, v/v) and gentamicin (50 μg ml-1) and incubated for 48 or 72 h at 37°C in the presence of CO2. The agarose layer was then removed and cells were fixed and stained with Giemsa.

Statistical analysis

Data was analyzed either by one-way ANOVA or t-test. Statistical significance was assumed at a P-value ≤ 0.05. The data analysis was performed using the Prism 5.01 software (GraphPad).

Results

Analysis of the GI-leuX in Shigella spp. strains

In the E. coli K-12 strain MG1655, the fim cluster is located in a 50-kb GI also carrying the fecABCDE operon (encoding an iron transport system), the yjhATS genes (encoding factors involved in uptake and catabolism of sialic acids) and a number of IS elements and phage remnants. This GI is flanked by uxuABR (encoding proteins involved in hexuronate degradation) and gntP (encoding a fructuronate transporter) on one side and by leuX (encoding a Leu tRNA) and yjgB (encoding a predicted alcohol dehydrogenase) on the other side (Fig. 1). Comparison of the sequences of this region in MG1655 and representative strains of S. flexneri, S. boydii, S. dysenteriae, S. sonnei and EIEC from the six phylogenetic groups identified GIs of different sizes carrying various assortments of the fim, yjh and fec genes (Fig. 1); (i) group S3: the ~23-kb GI of S. flexneri strains 2457T and Sf301 (serotype 2a) and M90T and Sf8401 (serotypes 5a and 5b, respectively) carries the fim cluster and yjhATS; (ii) group SS: the 21-kb GI of the S. sonnei strain Ss046 carries yjhATS and fec genes; the 29-kb GI of the S. sonnei strain 53G carries the 3' part of the fim cluster, yjhATS and fec genes; (iii) group SD1: only fec genes are present in the vicinity of leuX and yjgB in S. dysenteriae strain Sd197 (serotype 1); the left boundary of the GI cannot be defined because the flanking gntP and uxuABR genes are missing; (iv) group S1: the 18-kb GI of the S. boydii strain CDC3083-94 (serotype 18) carries only a remnant of the fim cluster; in the S. boydii strain Sb227 (serotype 4), there is an inversion of a region encompassing part of the GI and only the fim cluster is present between gntP and leuX. In the absence of complete genome sequence data, we used WGS results for strains of the other phylogenetic groups; (v) group S2: in the S. dysenteriae strain 155–74 (serotype 2), the fim, yjhATS and fec genes, as well as the uxuABR and gntP genes, are absent; in S. boydii strains CIP52-54 and CIP56-18 of serotypes 7 and 11, respectively (F.-X. Weill, personal communication), only the 3' part of fimD and the fimFGH genes are present next to gntP; (vi) group A: in the EIEC strain 53638 (serotype O144), the 4-kb region located between gntP and leuX carries only the 3’ part of the fim cluster. These observations suggest that the leuX GIs harbored by Shigella and EIEC strains derive from a region carrying the fec, yjhATS and fim clusters and that different segments of this GI were lost or modified during the evolution of different phylogenetic groups.

thumbnail
Fig 1. Genomic organization of the GI leuX in E. coli K12 and representative members of Shigella spp.

A schematic representation (drawn to scale) of the main gene clusters present in the GI leuX is shown for the E. coli K-12 strain MG1655, the S. flexneri strain 2457T, the S. sonnei strain Ss046, the S. dysenteriae strain Sd197, the S. boydii strains CDC 3083–94 and Sb227, the S. dysenteriae strain 155–74, the S. boydii strain CIP52-54 and the EIEC strain 53638; the phylogenetic group to which each strain belongs is indicated in parenthesis. The GI encompasses the region located between uxuABR-gntP and leuX-yjgB. For the sake of clarity, IS (including IS inserted in fim genes) and phage remnants are not indicated.

https://doi.org/10.1371/journal.pone.0121785.g001

Further analysis of the published sequences of the fim cluster revealed that fimD is interrupted by an IS1 at codon 221 in the two S. flexneri strains of serotype 2a (2457T and Sf301) and fimI is interrupted by an IS1 at codon 35 in the S. flexneri strains of serotypes 5a (M90T) and 5b (Sf8401). Furthermore, fimB is inactivated by a nonsense mutation at codon 162 (TAG instead of CAG in E. coli K12) in these four S. flexneri strains. In the S. sonnei strain 53G, the 3' end of fimB (from codon 188) up to the fimC-fimD intergenic region is replaced by an IS1, there is an insertion of IS1 at codon 204 of fimD and of an unknown IS at codon 596 of fimD. In the S. boydii strain CDC3083-94, fimB, fimE and fimA are missing, fimD carries a frameshift mutation due to the deletion of one nucleotide at codon 27 and a nonsense mutation at codon 563 (TAG instead of CAG in E. coli K12) and an IS1 is inserted at codon 98 of fimC. In the S. boydii strain Sb227 (serotype 4), fimD carries the same nonsense mutation at codon 563 and fimH is interrupted by an IS629 at codon 264. There is an insertion of an ISSfl7 at codon 801 of fimD in the S. boydii strain CIP52-54 (serotype 7) and an insertion of an IS600 at codon 509 of fimD in the S. boydii strain CIP56-18 (serotype 11); furthermore, the fimBEAIC genes are absent from these two strains. In the EIEC strain 53638, only the last two genes of the fim operon, fimG and fimH, are present. These observations indicate that the fim cluster is inactivated in Shigella and EIEC representative strains belonging to all phylogenetic groups.

The fim operon is inactivated in S. flexneri clinical isolates

To determine if the presence of the fim cluster is a common feature of S. flexneri strains and whether or not this gene cluster is complete and functional in clinical isolates, we analyzed a collection of 60 strains isolated from patients suffering from shigellosis in Chile between 2004 and 2006 [33]. Tiling-PCR was used to characterize the GI-leuX region in S. flexneri clinical isolates of serotypes 1a, 2a, 2b, 3a and 3b (Table 2), as well as in the reference strains 2457T (serotype 2a) and M90T (serotype 5a). Primers were designed using the genome of 2457T to cover the entire GI with six overlapping PCR fragments (Fig. 2). The six PCR fragments were amplified from all but one of the 60 isolates; in this last strain, only PCR fragment 6 that does not contain any fim gene was amplified. The size of PCR fragment 2 containing fimD was 4.2 kb in 39 strains and 3.4 kb in 20 strains (Table 2); these sizes are consistent with the presence of IS1 in fimD in 39 strains (as in 2457T and Sf301) and its absence in the remaining 20 strains (as in Sf8401). Accordingly, fimD is inactivated by IS1 in most strains of serotype 2a (39 out of 43 strains), but not in strains of serotypes 1b (1 strain), 2b (7 strains), 3a (7 strains) and 3b (1 strain) (Table 2). These results were confirmed by nested PCR using internal primers for IS1 (data not shown). Sizes of PCR fragments amplified from clinical strains indicated that none of these strains carried an IS in any other fim gene. We also analyzed 60 S. sonnei clinical strains; none of the six PCR fragments was amplified from any of these strains, indicating that, as in the sequenced genome of Ss046, the fim cluster is not present in these S. sonnei Chilean clinical isolates.

thumbnail
Table 2. Distribution of inactivation events of the fim cluster among S. flexneri Chilean clinical isolates clinical and reference strains.

https://doi.org/10.1371/journal.pone.0121785.t002

thumbnail
Fig 2. Characterization of the GI leuX in S. flexneri.

The genetic organization of the ~23-kb GI inserted at the leuX locus in the S. flexneri reference strain 2457T is shown (drawn to scale). The gntP and leuX-yjgB genes flanking the GI are indicated by hatched arrows, fim genes by white arrows, yjhATS genes by dark-gray arrows and other genes present in the GI by pale-gray arrows. The position of IS1 in fimD is indicated by a black box. The position and sizes of the six PCR fragments generated for the tiling PCR are indicated below the genetic map; the expected sizes of PCR fragment 2 corresponding to fimD with and without IS1 are indicated. For the sake of clarity, other IS elements and phage remnants are not indicated.

https://doi.org/10.1371/journal.pone.0121785.g002

In the sequenced S. flexneri genomes, the mutation introducing a nonsense codon at codon 162 of fimB creates a BfaI restriction site (ATC TAG). To determine whether S. flexneri Chilean clinical isolates carry the same mutation in fimB, PCR fragment 3 containing fimB was digested with BfaI; sizes of BfaI fragments were consistent with the presence of a BfaI site at codon 162 of fimB in all strains (Table 2). Even though the fim cluster is present in almost all S. flexneri Chilean clinical isolates and is presumably intact (i.e. not interrupted by any IS) in 20 strains, fimB is inactivated by the same mutation in all strains (Table 2). Restriction analysis of the fragment encompassing fimS amplified from ten clinical isolates that did not contain an IS in any fim genes indicated that the promoter of the fim operon is in the OFF orientation in all these isolates (data not shown), as it is in all the 129 Shigella genome sequences and WGS results available.

To determine whether the nonsense mutation in fimB is present in all S. flexneri strains, we analyzed results of two WGS projects, one encompassing 59 strains covering 14 serotypes (1a, 1b, 1d, 2a, 2b, 3a, 3b, 4a, 4av, 4b, X, Xv, Y and Yv) isolated from different provinces of China between 1997 and 2006 [40] and the other encompassing the 16 type strains covering all the established serotypes (1a, 1b, 1c, 2a, 2b, 3a, 3b, 3c, 4a, 4b, 5a, 5b, 6, X, Y and E1037) isolated from various regions between 1949 and 1972 and held at Public Health England (PHE) [41], as well as the sequence of the S. flexneri 2a strain NCTC1 isolated from a English soldier on the Western front in 1915 [37]. In all these strains, but the serotype 6 strain of the PHE collection, fimB carries the same mutation in codon 162. Further analysis of WGS results available for three other serotype 6 strains, CFSAN027317, CCH60 and CDC 796–83 confirmed that serotype 6 strains do not contain the mutation in fimB and, instead, all contain both a nonsense codon at codon 563 of fimD and an insertion of IS629 at codon 264 of fimH. The same two mutations are also present in the S. boydii strains Sb227 and 3594–74 (both of serotype 4) and 4444–74 (serotype 2). These later observations are consistent with previous analysis showing that S. flexneri serotype 6 strains are more closely related to S. boydii strains than to any other S. flexneri strains [11], [46].

Construction of a S. flexneri strain producing functional type 1 fimbriae

To analyze a S. flexneri strain producing fimbriae, the S. flexneri strain M90T was transformed with the plasmid pSH2, a derivative of the vector pACYC184 carrying the entire E. coli fim cluster [22]. To enrich the proportion of bacteria in which the fim promoter was in the ON orientation, bacteria were first cultured in static conditions and the orientation of the promoter was monitored by restriction analysis of a PCR product covering fimS (Fig. 3). Results indicated that the fim promoter was in the ON orientation in ~80% of the plasmid population and RT-PCR on fimA, fimH and fimD confirmed the transcription of fim genes in M90T/pSH2 (data not shown). To visualize production of type 1 fimbriae at the surface of bacteria, electron microscopy images of negatively stained M90T bacteria harboring pSH2 or pACYC184 were obtained.

thumbnail
Fig 3. Determination of the orientation of the fimS invertible element.

Schematic representations of the region encompassing fimS in the ON and OFF orientations are shown in panels A and B, respectively; positions of fimS (box), inverted repeats (IR) on both sides of fimS, the promoter of the fim operon (small open arrow), the 5' end of fimA (large grey arrow), primers Inv-1 and Inv-2 (small black arrows), the SnaBI cleavage site and sizes of the SnaBI digestion products of the PCR fragment amplified by using Inv-1 and Inv-2 are indicated for each orientation. (C) The region encompassing fimS was amplified by PCR by using primers Inv-1 and Inv-2 and digested with SnaBI and restriction fragments were resolved on a 2% agarose gel; lane 1, M90T/pSH2 after growth for 24 h in static conditions; lane 2, M90T/pACYC184 after growth for 24 h in static conditions; lane 3, M90T/pSH2 after growth for 48 h in static conditions. The fimS region is carried by both the chromosome and the plasmid pSH2 in M90T/pSH2 and only by the chromosome in M90T/pACYC184; due to the higher copy number of pSH2 as compared to the chromosome, fimS was preferentially amplified from the pSH2 plasmid in M90T/pSH2. Diagrams on panels A and B were adapted from [44].

https://doi.org/10.1371/journal.pone.0121785.g003

Electron microscopy analysis showed the presence of fimbriae at the surface of 74% of bacteria harboring pSH2, but not of bacteria harboring pACYC184 (Fig. 4). Bacteria producing type 1 fimbriae promote agglutination of erythrocytes in the absence, but not in the presence of mannose [43], [44]. Haemagglutination assays performed using Guinea pig erythrocytes confirmed that M90T/pSH2, but not M90T/pACYC184, expressed functional fimbriae (Fig. 4).

thumbnail
Fig 4. Characterization of type 1 fimbriae production in derivatives of the S. flexneri strain M90T.

Representative electron microscopy pictures of derivatives of S. flexneri strain M90T harboring pACYC184 (vector) or pSH2 (carrying the fim operon) are shown in panels A and B, respectively. Scale bar, 1 μm. Results of an haemagglutination assays performed in the absence or in the presence of 0.2 mM mannose with the S. flexneri strain M90T harboring either pSH2 or pACYC184 are shown in panel C.

https://doi.org/10.1371/journal.pone.0121785.g004

Effect of type 1 fimbriae production on the interaction of S. flexneri with epithelial cells

To investigate the potential consequence of type 1 fimbriae production on the interaction of Shigella with epithelial cells, we monitored adhesion of M90T/pSH2 to HeLa cells. Following centrifugation of bacteria onto semi-confluent cell monolayers, infected cells were incubated for 30 min at 37°C, washed, fixed with ethanol and stained with Giemsa. Observation of infected cells revealed a 50-fold increase in the number of cell-associated M90T/pSH2 as compared to M90T/pACYC184 (Fig. 5). The increased adhesion of M90T/pSH2 was abolished in the presence of mannose, confirming that it was dependent upon functional type 1 fimbriae (Fig. 5).

thumbnail
Fig 5. Adhesion of S. flexneri bacteria expressing type 1 fimbriae to HeLa cells.

HeLa cells infected for 30 min with M90T/pACYC184 or M90T/pSH2 in the absence or in the presence of 0.2 mM mannose were fixed with ethanol and stained with Giemsa. Numbers of bacteria per cell are shown in (A). Data are the means of three independent experiments performed in duplicate; *** denotes P-value < 0.001. Representative pictures of Giemsa stained cells are shown in (B); scale bar, 40 μm.

https://doi.org/10.1371/journal.pone.0121785.g005

To evaluate invasion and dissemination of bacteria in a confluent monolayer of TC7 cells, we used the plaque assay that does not involve any centrifugation of bacteria onto the cells; in this assay, the number of plaques is indicative of the ability of bacteria to enter epithelial cells and the size of plaques is indicative of the ability of bacteria to spread from cell to cell [45]. M90T/pSH2 induced the formation of approximately 50 times more plaques than M90T/pACYC184, an increase that was abolished when invasion was performed in the presence of mannose (Fig. 6). Plaques of similar sizes were produced by the two strains, suggesting that expression of fimbriae on the bacterial surface had no effect on the capacity of bacteria to disseminate from cell to cell. These results demonstrated the increased invasive capacity conferred upon Shigella by production of fimbriae.

thumbnail
Fig 6. Invasion and dissemination of S. flexneri bacteria expressing type 1 fimbriae.

(A) Confluent monolayers of TC7 cells were infected with M90T/pACYC184 or M90/pSH2 at multiplicities of infection (MOI) of 0.01 (upper row) or 0.001 (lower row), overlaid with gentamicin-containing agar, incubated at 37°C for 72 h, fixed and stained with Giemsa. (B) The graph indicates the number of plaques formed in the monolayer infected with either M90T harboring pACYC184 (vector) or pSH2 (carrying the fim operon), as calculated using different MOIs. Values are the means and SD of three independent experiments, each performed in triplicate at a MOI of 0.001; ** indicates P-value between 0.001 and 0.01. (C) Confluent monolayers of TC7 cells were infected with M90T/pACYC184 or M90/pSH2 at a MOI of 0.001 in the absence or in the presence of 0.2 mM mannose during the first 2 h (entry step).

https://doi.org/10.1371/journal.pone.0121785.g006

Discussion

This study analyzed the presence, integrity and functionality of the fim cluster carried by the GI inserted next to leuX in the genome of members of Shigella spp. and EIEC. The presence of all or part of the fim, yjfATS and fec clusters in these strains suggests that the leuX GI initially carried these three gene clusters. Analysis of published genome sequences indicated that the fim cluster is either deleted or inactivated in members of all Shigella and EIEC phylogenetic groups: (i) in group A, there is a deletion encompassing fimBEAICDF (in EIEC 53638); (ii) in group S1, there is a mutation introducing a nonsense codon at codon 563 of fimD, an insertion of IS629 after codon 264 of fimH and either a frameshift mutation at codon 25 of fimD (in S. boydii CDC3083-94) or a deletion of fimEBA (in S. boydii Sb227); (iii) in group S2, the fim, yjhATS and fec clusters and the uxuABR and gntP neighboring genes are absent in S. dysenteriae 155–74 and only the 3' part of fimD and the fimFGH genes are present in S. boydii strains CIP52-54 and CIP56-18; (iv) in group S3, there is a mutation introducing a nonsense codon at codon 162 of fimB in all S. flexneri strains except strains of serotype 6, as well as an insertion of IS1 in fimD in some serotype 2a strains and in fimI in some serotype 5b strains; (v) in group SD1, there is a deletion encompassing the whole fim cluster and the adjacent uxuABR and gntP genes (in S. dysenteriae Sd197); (vi) in group SS, there is a deletion of the whole fim cluster in S. sonnei Ss046, all 60 Chilean clinical isolates tested and most strains for which WGS data are available and only remnants of the fim cluster are present in the S. sonnei strain 53G and a few other strains. PCR and restriction analysis of fimB from 60 S. flexneri Chilean clinical isolates and the survey of the genome sequences and WGS results of over 80 S. flexneri strains from various origins [37], [40]- [41] indicated that the very same mutation in fimB is present in all S. flexneri strains, except strains of serotype 6. The four serotype 6 strains we analyzed contain both the same mutation in fimD and the same insertion of IS629 in fimH; the very same two mutations are present in some S. boydii strains, consistent with the close relationship between S. flexneri serotype 6 and S. boydii strains [11], [46]. The presence of the same mutation in fimB in S. flexneri strains of all serotypes (except strains of serotype 6 discussed above) isolated over a period of almost a century from three continents, i.e. in 1915 in France (NCTC1), in 1954 in Tokyo (2457T), in 1955 in Mexico City (M90T), in 1984 in Beijing (Sf301), from 1949 to 1972 in various places (PHE collection), in the early 2000s in Santiago (clinical isolates used in this study) and from 1997 to 2006 in various provinces of China, suggests that this mutation occurred prior to serotype diversification [11]. Even though the nature of the initial mutation that led to inactivation of a gene cluster cannot be demonstrated in the case of deletions encompassing several genes, there is evidence that inactivation of the fim cluster is due to different point mutations in at least three cases.

The identification of mutations inactivating the fim cluster in Shigella spp. shed light on previous observations by Snellings et al. reporting that bacteria expressing type 1 fimbriae were recovered from a subset of Shigella strains, albeit at a low frequency, after five to ten serial transfers in static culture conditions and that fimbriae expression was accompanied by inversion of fimS in the ON orientation in S. flexneri Fim+ derivatives [47]. It is likely that fimB was inactivated in the S. flexneri strains used by Snellings et al., as it is in all S. flexneri genomes and strains examined in the present study; accordingly, the reported low frequency of switching to the fimbriated phase was probably due to the necessity of reverting the non-sense mutation in fimB (or expressing a suppressor tRNA) prior to obtaining inversion of fimS and expression of the fim operon. No fimbriated derivatives were obtained from M90T [47] in which, as shown here, fimI is also inactivated by an IS1. Recovery of fimbriated derivatives from one S. boydii strain and one S. dysenteriae strain, the serotypes of which were not reported, is more difficult to explain from the analysis of S. boydii and S. dysenteriae available genome sequences. Indeed, there is both a nonsense mutation at codon 564 of fimD and an insertion of IS629 at codon 264 of fimH in S. boydii strains Sb27 (serotype 4), CDC3083-94 (serotype 18), 4444–74 (serotype 2) and 3594–74 (serotype 4); fimBEAIC and the 5' part of fimD are missing in S. boydii strains CIP52-54 and CIP56-18 (serotypes 7 and 11) and the whole fim cluster is absent from the nine other S. boydii strains for which WGS results are available in GenBank. The fim cluster is absent from the complete genome of the S. dysenteriae strain Sd197, as well as from WGS results of seven other S. dysenteriae strains, and only a truncated fim cluster interrupted by 3 IS is present in WGS results of the S. dysenteriae strain 222–75. Future analysis of larger datasets for S. boydii and S. dysenteriae genomes will help to define further inactivation events of the fim cluster in these Shigella spp.

To investigate the potential role of fimbriae production on the interaction of Shigella with host cells, we used a S. flexneri strain expressing the fim operon from E. coli. Production of fimbriae increased the ability of bacteria to adhere to epithelial cells in vitro, consistent with the reported role of type 1 fimbriae in promoting adhesion of E. coli to epithelial cells [48], [49]. Increasing the capacity of S. flexneri to adhere to epithelial cells by experimental expression of the adhesin AfaE or coating bacteria with poly-lysine was reported to increase entry of bacteria into cells in vitro [50], [51]. However, it was conceivable that exposure of long fimbriae on the surface of bacteria might interfere with either the delivery of T3S effectors or the accessibility of IcsA that are involved in entry and intracellular mobility of bacteria, respectively. In the plaque assay, in which the interaction between bacteria and cells is not forced by a centrifugation step, the invasive capacity of the fimbriated strain was 50 folds higher than that of the non-fimbriated strain. Plaques of similar size were produced by the Fim+ and Fim- strains, indicating that production of fimbriae did not affect the ability of S. flexneri to disseminate from cell to cell. These results indicated that production of fimbriae does not interfere with delivery of T3S effectors and, by increasing the ability of bacteria to interact with epithelial cells, increases the ability of these bacteria to invade these cells in vitro.

In deciphering the interactions pathogenic bacteria establish with their host(s), much attention has been paid to the identification of their specific virulence determinants, most often acquired by lateral transfer. However, both the loss and the gain of genetic material have contributed to the adaptation of bacteria to new environments. The importance of pathoadaptive mutations is increasingly recognized [16], [52], [53]. The presence of antigenic surface structures that do not have an overall favorable role in infection are counter-selected, possibly because they increase detection of bacteria by the innate and adaptive immune systems. Indeed, genes encoding the flagellum and curli are inactivated in Shigella spp. [54]–[56]. Moreover, expression of other immunogenic surface structures, such as the LPS, is modified at different stages of infection; in an initial phase of infection of epithelial cells, lipid A acylation is decreased, likely allowing Shigella to evade the immune detection [57]. Even though expression of type 1 fimbriae in Shigella produced a hyper-invasive phenotype in vitro, it may be detrimental to bacterial survival in the gut, given that expression of a highly immunogenic structure at the onset of infection might trigger a strong immune response. Moreover, it was shown that a fimbriated Shigella strain was more sensitive to killing by human granulocytes than a non-fimbriated strain [58] and that expression of FimH in uropathogenic E. coli (UPEC) induced rapid neutrophil recruitment in mice [59]. There is accumulating evidence indicating that Shigella inhibits the inflammatory response, including during invasion of epithelial cells [4], [60]–[69]. For instance, the T3S effector IpgD was shown to prevent the hemichannel-dependent secretion of the endogenous danger signal ATP, an early alert response to infection in intestinal epithelial cells [69]. The finding that Shigella actively dampens the immune response highlights the necessity of these invasive bacteria to avoid immune detection in the first place. The observation that independent mutations led to inactivation of fimbriae production in spite of the increased invasion capacity conferred upon Shigella by fimbriae expression suggests that these mutations might correspond to pathoadaptive events.

Acknowledgments

We thank Joelle Mounier for technical assistance, Marie-Christine Prevost and Perrine Bomme (Plate-Forme de Microscopie Ultrastructurale, Imagopole, Institut Pasteur) for expert assistance in electron microscopy analysis, Erik Denamur for discussions on Shigella phylogeny and François-Xavier Weill and Kate Baker (Wellcome Trust Sanger Institute) for their help in the analysis of WGS data from the PHE strain collection.

Author Contributions

Conceived and designed the experiments: VB CP CT PS. Performed the experiments: VB AP. Analyzed the data: VB CP CT PS. Contributed reagents/materials/analysis tools: CP CT. Wrote the paper: VB CP. Editing and proof reading the manuscript: VB AP CP PS CT.

References

  1. 1. Kotloff KL, Winickoff JP, Ivanoff B, Clemens JD, Swerdlow DL, Sansonetti PJ, et al. Global burden of Shigella infections: implications for vaccine development and implementation of control strategies. Bull World Health Organ. 1999;77:651–666. pmid:10516787
  2. 2. Johnson TJ, Nolan LK. Pathogenomics of the virulence plasmids of Escherichia coli. Microbiol Mol Biol Rev MMBR. 2009;73:750–774. pmid:19946140
  3. 3. Buchrieser C, Glaser P, Rusniok C, Nedjari H, D’Hauteville H, Kunst F, et al. The virulence plasmid pWR100 and the repertoire of proteins secreted by the type III secretion apparatus of Shigella flexneri. Mol Microbiol. 2000;38:760–771. pmid:11115111
  4. 4. Parsot C. Shigella spp. and enteroinvasive Escherichia coli pathogenicity factors. FEMS Microbiol Lett. 2005;252:11–18. pmid:16182469
  5. 5. Newland JW, Strockbine NA, Miller SF, O’Brien AD, Holmes RK. Cloning of Shiga-like toxin structural genes from a toxin converting phage of Escherichia coli. Science. 1985;230:179–181. pmid:2994228
  6. 6. Vokes SA, Reeves SA, Torres AG, Payne SM. The aerobactin iron transport system genes in Shigella flexneri are present within a pathogenicity island. Mol Microbiol. 1999;33:63–73. pmid:10411724
  7. 7. Al-Hasani K, Henderson IR, Sakellaris H, Rajakumar K, Grant T, Nataro JP, et al. The sigA gene which is borne on the she pathogenicity island of Shigella flexneri 2a encodes an exported cytopathic protease involved in intestinal fluid accumulation. Infect Immun. 2000;68:2457–2463. pmid:10768931
  8. 8. Hacker J, Kaper JB. Pathogenicity islands and the evolution of microbes. Annu Rev Microbiol. 2000;54:641–679. pmid:11018140
  9. 9. Schmidt H, Hensel M. Pathogenicity islands in bacterial pathogenesis. Clin Microbiol Rev. 2004;17:14–56. pmid:14726454
  10. 10. Escobar-Páramo P, Giudicelli C, Parsot C, Denamur E. The evolutionary history of Shigella and enteroinvasive Escherichia coli revised. J Mol Evol. 2003;57:140–148. pmid:14562958
  11. 11. Pupo GM, Lan R, Reeves PR. Multiple independent origins of Shigella clones of Escherichia coli and convergent evolution of many of their characteristics. Proc Natl Acad Sci U S A. 2000;97:10567–10572. pmid:10954745
  12. 12. Touchon M, Hoede C, Tenaillon O, Barbe V, Baeriswyl S, Bidet P, et al. Organised genome dynamics in the Escherichia coli species results in highly diverse adaptive paths. PLoS Genet. 2009;5:e1000344. pmid:19165319
  13. 13. Jin Q, Yuan Z, Xu J, Wang Y, Shen Y, Lu W, et al. Genome sequence of Shigella flexneri 2a: insights into pathogenicity through comparison with genomes of Escherichia coli K12 and O157. Nucleic Acids Res. 2002;30:4432–4441. pmid:12384590
  14. 14. Moran NA, Plague GR. Genomic changes following host restriction in bacteria. Curr Opin Genet Dev. 2004;14:627–633. pmid:15531157
  15. 15. Cooper VS, Lenski RE. The population genetics of ecological specialization in evolving Escherichia coli populations. Nature. 2000;407:736–739. pmid:11048718
  16. 16. Maurelli AT. Black holes, antivirulence genes, and gene inactivation in the evolution of bacterial pathogens. FEMS Microbiol Lett. 2007;267:1–8. pmid:17233672
  17. 17. Sokurenko EV, Hasty DL, Dykhuizen DE. Pathoadaptive mutations: gene loss and variation in bacterial pathogens. Trends Microbiol. 1999;7:191–195. pmid:10354593
  18. 18. Maurelli AT, Fernández RE, Bloch CA, Rode CK, Fasano A. “Black holes” and bacterial pathogenicity: a large genomic deletion that enhances the virulence of Shigella spp. and enteroinvasive Escherichia coli. Proc Natl Acad Sci U S A. 1998;95:3943–3948. pmid:9520472
  19. 19. McCormick BA, Fernandez MI, Siber AM, Maurelli AT. Inhibition of Shigella flexneri-induced transepithelial migration of polymorphonuclear leucocytes by cadaverine. Cell Microbiol. 1999;1:143–155. pmid:11207548
  20. 20. Prunier A-L, Schuch R, Fernández RE, Mumy KL, Kohler H, McCormick BA, et al. nadA and nadB of Shigella flexneri 5a are antivirulence loci responsible for the synthesis of quinolinate, a small molecule inhibitor of Shigella pathogenicity. Microbiol Read Engl. 2007;153:2363–2372. pmid:17600080
  21. 21. Klemm P, Schembri MA. Bacterial adhesins: function and structure. Int J Med Microbiol IJMM. 2000;290:27–35.
  22. 22. Orndorff PE, Falkow S. Identification and characterization of a gene product that regulates type 1 piliation in Escherichia coli. J Bacteriol. 1984;160:61–66. pmid:6148338
  23. 23. Kline KA, Dodson KW, Caparon MG, Hultgren SJ. A tale of two pili: assembly and function of pili in bacteria. Trends Microbiol. 2010;18:224–232. pmid:20378353
  24. 24. Capitani G, Eidam O, Glockshuber R, Grütter MG. Structural and functional insights into the assembly of type 1 pili from Escherichia coli. Microbes Infect Inst Pasteur. 2006;8:2284–2290. pmid:16793308
  25. 25. Valenski ML, Harris SL, Spears PA, Horton JR, Orndorff PE. The Product of the fimI gene is necessary for Escherichia coli type 1 pilus biosynthesis. J Bacteriol. 2003;185:5007–5011. pmid:12897022
  26. 26. Connell I, Agace W, Klemm P, Schembri M, Mărild S, Svanborg C. Type 1 fimbrial expression enhances Escherichia coli virulence for the urinary tract. Proc Natl Acad Sci U S A. 1996;93:9827–9832. pmid:8790416
  27. 27. Krogfelt KA, Bergmans H, Klemm P. Direct evidence that the FimH protein is the mannose-specific adhesin of Escherichia coli type 1 fimbriae. Infect Immun. 1990;58:1995–1998. pmid:1971261
  28. 28. Langermann S, Palaszynski S, Barnhart M, Auguste G, Pinkner JS, Burlein J, et al. Prevention of mucosal Escherichia coli infection by FimH-adhesin-based systemic vaccination. Science. 1997;276:607–611. pmid:9110982
  29. 29. McClain MS, Blomfield IC, Eisenstein BI. Roles of fimB and fimE in site-specific DNA inversion associated with phase variation of type 1 fimbriae in Escherichia coli. J Bacteriol. 1991;173:5308–5314. pmid:1679430
  30. 30. Gally DL, Leathart J, Blomfield IC. Interaction of FimB and FimE with the fim switch that controls the phase variation of type 1 fimbriae in Escherichia coli K-12. Mol Microbiol. 1996;21:725–738. pmid:8878036
  31. 31. Sansonetti PJ, Kopecko DJ, Formal SB. Involvement of a plasmid in the invasive ability of Shigella flexneri. Infect Immun. 1982;35:852–860. pmid:6279518
  32. 32. Sansonetti PJ, Ryter A, Clerc P, Maurelli AT, Mounier J. Multiplication of Shigella flexneri within HeLa cells: lysis of the phagocytic vacuole and plasmid-mediated contact hemolysis. Infect Immun. 1986;51:461–469. pmid:3510976
  33. 33. Farfán MJ, Garay TA, Prado CA, Filliol I, Ulloa MT, Toro CS. A new multiplex PCR for differential identification of Shigella flexneri and Shigella sonnei and detection of Shigella virulence determinants. Epidemiol Infect. 2010;138:525–533. pmid:19765332
  34. 34. Blattner FR, Plunkett G, Bloch CA, Perna NT, Burland V, Riley M, et al. The complete genome sequence of Escherichia coli K-12. Science. 1997;277:1453–1462. pmid:9278503
  35. 35. Yang F. Genome dynamics and diversity of Shigella species, the etiologic agents of bacillary dysentery. Nucleic Acids Res. 2005;33:6445–6458. pmid:16275786
  36. 36. Wei J, Goldberg MB, Burland V, Venkatesan MM, Deng W, Fournier G, et al. Complete genome sequence and comparative genomics of Shigella flexneri serotype 2a strain 2457T. Infect Immun. 2003;71:2775–2786. pmid:12704152
  37. 37. Baker KS, Mather AE, McGregor H, Coupland P, Langridge GC, Day M, et al. The extant World War 1 dysentery bacillus NCTC1: a genomic analysis. Lancet. 2014;384:1691–1697. pmid:25441199
  38. 38. Onodera NT, Ryu J, Durbic T, Nislow C, Archibald JM, Rohde JR. Genome sequence of Shigella flexneri serotype 5a strain M90T. J Bacteriol. 2012;194:3022. pmid:22582379
  39. 39. Nie H, Yang F, Zhang X, Yang J, Chen L, Wang J, et al. Complete genome sequence of Shigella flexneri 5b and comparison with Shigella flexneri 2a. BMC Genomics. 2006;7:173. pmid:16822325
  40. 40. Zhang N, Lan R, Sun Q, Wang J, Wang Y, Zhang J, et al. Genomic portrait of the evolution and epidemic spread of a recently emerged multidrug-resistant Shigella flexneri clone in China. J Clin Microbiol. 2014;52:1119–1126. pmid:24452172
  41. 41. Ashton PM, Baker KS, Gentle A, Wooldridge DJ, Thomson NR, Dallman TJ, et al. Draft genome sequences of the type strains of Shigella flexneri held at Public Health England: comparison of classical phenotypic and novel molecular assays with whole genome sequence. Gut Pathog. 2014;6:7. pmid:24684748
  42. 42. Bryan A, Roesch P, Davis L, Moritz R, Pellett S, Welch RA. Regulation of type 1 fimbriae by unlinked FimB- and FimE-like recombinases in uropathogenic Escherichia coli strain CFT073. Infect Immun. 2006;74:1072–1083. pmid:16428754
  43. 43. Fein JE. Screening of uropathogenic Escherichia coli for expression of mannose-selective adhesins: importance of culture conditions. J Clin Microbiol. 1981;13:1088–1095. pmid:6114111
  44. 44. Struve C, Bojer M, Krogfelt KA. Characterization of Klebsiella pneumoniae type 1 fimbriae by detection of phase variation during colonization and infection and impact on virulence. Infect Immun. 2008;76:4055–4065. pmid:18559432
  45. 45. Oaks EV, Wingfield ME, Formal SB. Plaque formation by virulent Shigella flexneri. Infect Immun. 1985;48:124–129. pmid:3884506
  46. 46. Yang J, Nie H, Chen L, Zhang X, Yang F, Xu X, et al. Revisiting the molecular evolutionary history of Shigella spp. J Mol Evol. 2007;64:71–79. pmid:17160643
  47. 47. Snellings NJ, Tall BD, Venkatesan MM. Characterization of Shigella type 1 fimbriae: expression, FimA sequence, and phase variation. Infect Immun. 1997;65:2462–2467. pmid:9169792
  48. 48. Godaly G, Frendéus B, Proudfoot A, Svensson M, Klemm P, Svanborg C. Role of fimbriae-mediated adherence for neutrophil migration across Escherichia coli-infected epithelial cell layers. Mol Microbiol. 1998;30:725–735. pmid:10094621
  49. 49. Croxen MA, Finlay BB. Molecular mechanisms of Escherichia coli pathogenicity. Nat Rev Microbiol. 2010;8:26–38. pmid:19966814
  50. 50. Clerc P, Sansonetti PJ. Entry of Shigella flexneri into HeLa cells: evidence for directed phagocytosis involving actin polymerization and myosin accumulation. Infect Immun. 1987;55:2681–2688. pmid:3312007
  51. 51. Mounier J, Popoff MR, Enninga J, Frame MC, Sansonetti PJ, Van Nhieu GT. The IpaC carboxyterminal effector domain mediates Src-dependent actin polymerization during Shigella invasion of epithelial cells. PLoS Pathog. 2009;5:e1000271. pmid:19165331
  52. 52. Bliven KA, Maurelli AT. Antivirulence genes: insights into pathogen evolution through gene loss. Infect Immun. 2012;80:4061–4070. pmid:23045475
  53. 53. Lawrence JG. Common themes in the genome strategies of pathogens. Curr Opin Genet Dev. 2005;15:584–588. pmid:16188434
  54. 54. Al Mamun AA, Tominaga A, Enomoto M. Detection and characterization of the flagellar master operon in the four Shigella subgroups. J Bacteriol. 1996;178:3722–3726. pmid:8682772
  55. 55. Sakellaris H, Hannink NK, Rajakumar K, Bulach D, Hunt M, Sasakawa C, et al. Curli loci of Shigella spp. Infect Immun. 2000;68:3780–3783. pmid:10816548
  56. 56. Tominaga A, Lan R, Reeves PR. Evolutionary changes of the flhDC flagellar master operon in Shigella strains. J Bacteriol. 2005;187:4295–4302. pmid:15937193
  57. 57. Paciello I, Silipo A, Lembo-Fazio L, Curcurù L, Zumsteg A, Noël G, et al. Intracellular Shigella remodels its LPS to dampen the innate immune recognition and evade inflammasome activation. Proc Natl Acad Sci U S A. 2013;110:E4345–4354. pmid:24167293
  58. 58. Gbarah A, Mirelman D, Sansonetti PJ, Verdon R, Bernhard W, Sharon N. Shigella flexneri transformants expressing type 1 (mannose-specific) fimbriae bind to, activate, and are killed by phagocytic cells. Infect Immun. 1993;61:1687–1693. pmid:8097492
  59. 59. Ashkar AA, Mossman KL, Coombes BK, Gyles CL, Mackenzie R. FimH adhesin of type 1 fimbriae is a potent inducer of innate antimicrobial responses which requires TLR4 and type 1 interferon signalling. PLoS Pathog. 2008;4:e1000233. pmid:19057665
  60. 60. Arbibe L, Kim DW, Batsche E, Pedron T, Mateescu B, Muchardt C, et al. An injected bacterial effector targets chromatin access for transcription factor NF-kappaB to alter transcription of host genes involved in immune responses. Nat Immunol. 2007;8:47–56. pmid:17159983
  61. 61. Ashida H, Kim M, Schmidt-Supprian M, Ma A, Ogawa M, Sasakawa C. A bacterial E3 ubiquitin ligase IpaH9.8 targets NEMO/IKKgamma to dampen the host NF-kappaB-mediated inflammatory response. Nat Cell Biol. 2010;12:66–73; sup pp 1–9. pmid:20010814
  62. 62. Ashida H, Toyotome T, Nagai T, Sasakawa C. Shigella chromosomal IpaH proteins are secreted via the type III secretion system and act as effectors. Mol Microbiol. 2007;63:680–693. pmid:17214743
  63. 63. Kim DW, Lenzen G, Page A-L, Legrain P, Sansonetti PJ, Parsot C. The Shigella flexneri effector OspG interferes with innate immune responses by targeting ubiquitin-conjugating enzymes. Proc Natl Acad Sci U S A. 2005;102:14046–14051. pmid:16162672
  64. 64. Li H, Xu H, Zhou Y, Zhang J, Long C, Li S, et al. The phosphothreonine lyase activity of a bacterial type III effector family. Science. 2007;315:1000–1003. pmid:17303758
  65. 65. Newton HJ, Pearson JS, Badea L, Kelly M, Lucas M, Holloway G, et al. The type III effectors NleE and NleB from enteropathogenic E. coli and OspZ from Shigella block nuclear translocation of NF-kappaB p65. PLoS Pathog. 2010;6:e1000898. pmid:20485572
  66. 66. Rohde JR, Breitkreutz A, Chenal A, Sansonetti PJ, Parsot C. Type III secretion effectors of the IpaH family are E3 ubiquitin ligases. Cell Host Microbe. 2007;1:77–83. pmid:18005683
  67. 67. Sanada T, Kim M, Mimuro H, Suzuki M, Ogawa M, Oyama A, et al. The Shigella flexneri effector OspI deamidates UBC13 to dampen the inflammatory response. Nature. 2012;483:623–626. pmid:22407319
  68. 68. Sansonetti PJ, Di Santo JP. Debugging how bacteria manipulate the immune response. Immunity. 2007;26:149–161. pmid:17307704
  69. 69. Puhar A, Tronchère H, Payrastre B, Nhieu GTV, Sansonetti PJ. A Shigella effector dampens inflammation by regulating epithelial release of danger signal ATP through production of the lipid mediator PtdIns5P. Immunity. 2013;39:1121–1131. pmid:24332032