Deletion of pic results in decreased virulence for a clinical isolate of Shigella flexneri 2a from China

Background Shigella is a major pathogen responsible for bacillary dysentery, a severe form of shigellosis. Severity of the disease depends on the virulence of the infecting strain. Shigella pathogenicity is a multi-gene phenomenon, involving the participation of genes on an unstable large virulence plasmid and chromosomal pathogenicity islands. Results A multiplex PCR (mPCR) assay was developed to detect S. flexneri 2a from rural regions of Zhengding (Hebei Province, China). We isolated and tested 86 strains using our mPCR assay, which targeted the ipaH, ial and set1B genes. A clinical strain of S. flexneri 2a 51 (SF51) containing ipaH and ial, but lacking set1B was found. The virulence of this strain was found to be markedly decreased. Further testing showed that the SF51 strain lacked pic. To investigate the role of pic in S. flexneri 2a infections, a pic knockout mutant (SF301-∆ pic) and two complementation strains, SF301-∆ pic/pPic and SF51/pPic, were created. Differences in virulence for SF51, SF301-∆ pic, SF301-∆ pic/pPic, SF51/pPic and S. flexneri 2a 301 (SF301) were compared. Compared with SF301, both SF51 and SF301-∆ pic exhibited lower levels of Hela cell invasion and resulted in reduced keratoconjunctivitis, with low levels of tissue damage seen in murine eye sections. The virulence of SF301-∆ pic and SF51 was partially recovered in vitro and in vivo through the addition of a complementary pic gene. Conclusions The pic gene appears to be involved in an increase in pathogenicity of S. flexneri 2a. This gene assists with bacterial invasion into host cells and alters inflammatory reactions.


Background
Bacteria of the genus Shigella are fastidious Gram-negative organisms that cause an estimated 164.7 million cases of shigellosis annually worldwide, and are responsible for 1.1 million deaths [1]. Shigellosis is an acute intestinal infectious disease. Its symptoms range from mild watery diarrhea to a life-threatening dysenteric syndrome with blood, mucus and pus in stools [2][3][4]. The severity of the disease depends on the virulence of the infecting strain. Therefore, clinical diagnosis tests for Shigellosis should not only focus on the determination of the strain's biochemical and serological types, but also on the determination of the strain's virulence. Based on biotyping, the Shigella genus contains four species with 48 serotypes (including subgroups). In China, Shigella flexneri 2a (S. flexneri 2a) is the predominant subgroup [2].
To simultaneously, effectively, and rapidly detect the pathogen and determine its virulence, three chromosomeand plasmid-encoded virulence genes (ipaH, ial, and set1B) [3,[5][6][7] were chosen to assist in the development of a multiplex PCR (mPCR) assay. ipaH is present on both the chromosome and on the large Shigella virulence plasmid. Therefore, ipaH is considered a stable PCR target for pathogen identification [8][9][10][11]. The ial gene is located in the cell-entry region of the large virulence plasmid that encodes an important part of the molecular machinery required for bacterial invasion and intracellular survival [4,[12][13][14]. This region is bracketed by insertion-like (IS) elements IS100 and IS600, with a high tendency for automatic deletion [4,13,15,16]. Detection based on ial provides some information pertaining to bacterial virulence but can easily generate false negative results [4,17]. The set1B gene is located on pathogenicity island 1 (PAI-1) of the chromosome and encodes Shigella enterotoxin 1 subunit B. Enterotoxin 1 causes the watery phase of diarrhea in Shigellosis [6,18,19]. Studies have shown that set1B is present exclusively in S. flexneri 2a [6,18,19]. An mPCR system should be able to determine, in a single reaction, whether the genes related to pathogenesis of a particular Shigella strain are encoded on the chromosome or the plasmid, and also to determine the serotype of a particular strain [4,5].
The S. flexneri 2a pic gene, which is located at an unstable chromosomal site of S. flexneri 2a PAI-1, is spontaneously deleted at a low frequency [20]. Previous studies have shown that the pic and set1B loci are overlapping genes encoded on opposite strands, and set1B is within pic [21]. The Pic protein is a 116 kDa autotransporter protein, secreted by the serine protease auto-transporter from members of the Enterobacteriaceae family [21,22]. To date, pic has only been found in enteroaggregative Escherichia coli (EAEC), uropathogenic E. coli (UPEC) and S. flexneri 2a. Pic has been shown to exhibit hemagglutination and mucinolytic activities in vitro [21][22][23][24]. However, it has also been shown that Pic is unable to elicit a cytotoxic effect in the HT29-C1 and HEp-2 epithelial cell lines [24,25].
The major aims of our study were to detect and determine the strain of the Shigella pathogen and determine its virulence. We also investigated whether attenuation of SF51 virulence correlated to the loss of pic, by constructing a pic-deleted mutant and two complementation strains.

Ethics
All procedures performed on mice were conducted according to national ( Beijing. The affected subject exhibited a severe acute clinical manifestation of Shigellosis. The complete genome of SF301 was sequenced and has since been used as a reference strain for S. flexneri 2a in China. E. coli ATCC 25922 was provided by Dr. Bijie Hu from Zhongshan Hospital (Shanghai, China). E. coli SM10 λpir and plasmid pSB890 were provided by Dr. Daoguo Zhou from Purdue University (West Lafayette, IN, USA). The pSC plasmid was modified from pREP4 (Qiagen, Hilden, Germany), which contains a p15A origin of replication and a kanamycin resistance gene. E. coli DH5α was purchased from Invitrogen (Carlsbad, CA, USA). S. flexneri and E. coli were grown at 37°C in Luria-Bertani (LB) medium (Oxoid, Wesel, Germany). All bacterial strains were grown on Salmonella-Shigella (SS) agar (Oxoid) before being transferred to an LB agar plate. Strains were then incubated overnight at 37°C, then stored at −20°C in LB broth containing 15% glycerol.

Plaque formation tests on HeLa cells
Twelve strains containing ipaH, ial and set1B were further tested to determine their virulence in HeLa cells (ATCC CCL-2) using a plaque formation test [26]. HeLa cells were grown in 24-well tissue culture plates until they formed semi-confluent monolayers. The culture medium used was RPMI1640 supplemented with 10% fetal calf serum (FCS), and 1% penicillin-streptomycin; and cultures were incubated at 37°C/5% CO 2 . Cells were washed three times with phosphate-buffered saline (PBS), and bacteria added to the semi-confluent HeLa cultures at a multiplicity of infection (MOI) of 100. After incubating at 37°C for 90 min, growth medium containing 5% (w/v) agar and 20 μg/mL gentamicin was poured into the 24well plates, then incubated at 37°C/5% CO 2 for 72 h. HeLa cells were inoculated with SF301 as a positive control, and with E. coli ATCC 25922 as a negative control.
Sequence and analysis of virulence genes on PAI-1 of SF51 SF51 genomic DNA was extracted using a QIAamp DNA Mini Kit (Qiagen). PCR primers for amplification of pic, sigA, int and orf30 from PAI-1 of the SF51 clinical isolate were designed according to the SF301 sequence. Amplicons were cloned into a pCR-XL-TOPO vector using a TOPO W XL PCR Cloning Kit (Invitrogen), and the inserts were sequenced by Sangon Biotech (Shanghai, China) Co. Ltd, then identified using the standard nucleotide basic local alignment search tool (BLASTn; NCBI).

Construction of SF301-Δ pic
The upstream and downstream portions of pic were amplified by PCR. Primers uppic-F-NotI and uppic-R-XbaI (Table 1) were used to amplify the upstream fragment of pic, with primers downpic-F-XbaI and downpic-R-BamHI (Table 1) used to amplify the downstream fragment. The amplified downstream fragment of pic was digested with XbaI and BamHI and ligated into pSB890 which had been cut with the same restriction endonucleases [27]. We designated the resulting plasmid pSB890-pic downstream. The amplified upstream pic fragment was digested with NotI and XbaI and ligated into pSB890-pic downstream that had been digested with NotI and XbaI. The resulting vector was designated pSB890-Δ pic and transformed into E. coli SM10 λpir cells, then introduced into SF301 through a bacterial conjugation test. After culturing on a sucrose LB agar plate at 22°C, sucrose-tolerant colonies were screened using Shigella-specific minimal medium [7] and a PCR employing primers Upuppic-F and Downdownpic-R ( Table 1). The mutant strain with the pic deletion was identified by sequencing and named SF301-Δ pic. Construction of complementation strains SF301-Δ pic/pPic and SF51/pPic A plasmid containing pic was constructed using pSC modified from pREP4. The pic gene was amplified from SF301 genomic DNA using PCR. The PCR primers used were pic-pSC-F-PfMlI and pic-pSC-R-AclI (Table 1). Amplicons were inserted into pSC, creating pSC-pic, which was verified by restriction enzyme digestion and nucleic acid sequencing. Verified pSC-pic was transformed into SF301-Δ pic and SF51, resulting in SF301-Δ pic/pPic and SF51/pPic, respectively.

S. flexneri growth curves
The growth curves of S. flexneri 2a strains were determined by measuring the optical density at 600 nm (OD 600 ) as described previously [28]. Briefly, overnight cultures were diluted 1:200 and incubated at 37°C with shaking (220 rpm). Samples (1 mL) of the bacterial cultures were taken every 30 min over 16 h and OD measured. Growth curves were created by plotting OD 600 against incubation time (h).

S. flexneri HeLa cell invasion assays
S. flexneri cell invasion assays were used to test the virulence of a SF51 clinical strain without set1B, SF301-Δ pic, wild-type SF301, SF301-Δ pic/pPic and SF51/pPic. The ability of bacteria to invade HeLa cells was determined using a gentamicin protection assays [29]. HeLa cells were grown in 6-well tissue culture plates in DMEM supplemented with 10% FCS and incubated at 37°C/5% CO 2 until they formed semi-confluent monolayers. SF51, SF301-Δ pic, SF301-Δ pic /pPic, SF51 /pPic and SF301 were individually added to semi-confluent HeLa cells at an MOI of 100. Bacteria were diluted and plated on LB agar plates. Colony-forming units (CFUs) were counted and added to HeLa cells. Plates were centrifuged at 900 × g for 5 min. After incubating at 37°C for 90 min, cells were washed three times with PBS, and gentamicin added to the medium at a final concentration of 10 μg/mL. The mixture was then incubated for 20 min at 37°C. HeLa cells in each well were lysed with 1 mL of PBS containing 0.1% Triton X-100 for 10 min at room temperature. Lysates were diluted and plated onto LB agar plates in triplicate.
Colonies that grew on LB plates were counted. Results were expressed as the number of bacteria recovered from gentamicin-treated cells divided by the number of inoculated bacteria added to the cell. Cells inoculated with E. coli ATCC 25922, an avirulent strain, were the negative controls. Cell invasion assays were performed in triplicate for each strain, and the assay repeated twice.

Sereny tests and pathohistological examination
A mouse Sereny test was used to evaluate the virulence of all strains we examined in this study, as described by Murayama [30]. A single red colony of S. flexneri on Congo red agar [Tryptic soy broth (Oxoid), 1.5% (w/v) agar and 0.01% (w/v) Congo red] was inoculated into LB broth at 37°C for 8 h with constant shaking. Female BALB/c mice (4-5-weeks-old) were infected with 1 × 10 8 CFUs per eye (n = 4 eyes, two mice in each group). Symptoms and signs of keratoconjunctivitis in mice infected with bacteria were observed at 24, 48, 72, and 96 h post-inoculation [28,30]. Eyes inoculated with E. coli ATCC 25922 and normal saline (NS) served as the negative controls. The invasiveness of bacteria was scored according to the following system: '−' indicates no inflammation, and an infection level score of 0; '±' is indicative of low levels of keratoconjunctivitis, and an infection level score of 0.5; '+' indicates slight conjunctival inflammation with eyelid edema, and an infection level score of 1; '++' indicates mild keratoconjunctivitis with eyelid edema, increased tear film evaporation and periocular hair loss, and an infection level score of 2; and '+++' indicating fully developed keratoconjunctivitis with eyelid swelling, periocular hair loss, blepharophimosis, conjunctival follicles and purulent discharge, and an infection level score of 3. At 24, 48, 72, and 96 h post-inoculation, mice were euthanized and the eyes removed and fixed in 4% formalin in PBS (pH 7.2). After hemotoxylin and eosin (H&E) staining, eye sections were examined using a light microscope.

Statistical analysis
Experimental data were analyzed with SPSS and comparisons made using Student's t-test. Differences with a P-value less than 0.05 were considered statistically significant.

Results
Detection of ipaH, ial, and set1B in S. flexneri clinical isolates The ipaH gene was detected in all 86 S. flexneri clinical isolates, whereas ial was detected in 45 isolates (52.3%), and set1B detected in 69 isolates (80.2%). Amplicons for ipaH, ial and set1B were not seen from E. coli ATCC 25922 samples. All three genes were detected in the SF301 positive control.

HeLa cell invasion of S. flexneri clinical isolates
Nine isolates in our study contained all three genes, one SF68 isolate contained ipaH and set1B, one SF51 isolate had ipaH and ial, and one SF36 isolate contained only ipaH ( Table 2). The nine isolates that contained all three virulence-associated genes demonstrated high invasive ability in HeLa cells (>200 plaques/well). In HeLa cells, SF68 and SF36 were less invasive resulting in 4 and 2 plaques/well, respectively. SF51 lacked set1B but retained ial, and showed a decrease in invasiveness (20 plaques/ well; Table 2). Using PCR and nucleotide sequencing, it was shown that SF51 lacked the entire pic gene on PAI-1, but harbored sigA and other significant open reading frames ( Figure 1).

HeLa cell invasion of SF301 and mutants
The growth curves for SF51, SF301-Δ pic, SF301, SF301-Δ pic/pPic and SF51/pPic were similar under aerobic growth conditions ( Figure 2). Gentamicin protection tests on HeLa cells revealed that the cell invasion ratio for clinical isolate SF51 with standard strain SF301 decreased by 34%, while the invasion ratio for SF301-Δ pic compared with SF301 decreased by 61% ( Figure 3A). The invasion abilities were partially recovered by the introduction of pic into deleted mutant SF301-Δ pic, which increased the ratio by 31% (to a final cell invasion ratio of 51%, Figure 3A). The invasion abilities of SF51/pPic increased by 59% compared with SF51, with cell invasion ratios of 35% and 22%, respectively ( Figure 3B). The E. coli ATCC 25922 strain was not found to invade HeLa cells.

Mouse Sereny tests and pathohistological examination
Mouse Sereny tests confirmed the results of the cell invasion tests. Mild presentation of keratoconjunctivitis was observed 24 h after mice were infected with SF301. Symptoms included eyelid edema, increased tear film evaporation and periocular hair-loss that we scored as either + or ++, with an average infection level score of 1.5. This developed into severe keratoconjunctivitis with maximal blepharophimosis at 48 h that we rated +++, and an average infection level score of 2.8. Keratoconjunctival inflammation continued for 96 h post-inoculation with SF301 ( Figure 4). Both the isolated and constructed picdeletion mutants induced lower levels of inflammation in the eyes of mice than for SF301 (Figure 4). At 48 h postinoculation, the pathogenicity of SF301-Δ pic in mouse * Values represent the mean ± standard deviation of three wells. and sigA were amplified from SF51 DNA but pic was not. All four target genes were amplified from the SF301 positive control. The sequence of all PCR products was confirmed by nucleic acid sequencing.   eyes were assessed as + or ++ with an average infection level scores up to 1.2; for SF51, pathogenicity was rated ± or + with an average infection level score less than 0.6. Virulence was partially recovered by introducing the complementary pSC-pic into the deletion mutants. At 48 h post-inoculation the pathogenicity of SF301-Δ pic/ pPic was rated at + or ++ with an average infection level score 1.9; SF51/pPic pathogenicity was + or ++ with average infection level scores of 1. Light microscopy assessment at 48 h post-infection revealed typical symptoms of SF301 infection. These included limited invasion, corneal epithelial thickening and loss, along with mild, moderate, or severe ulcers. Both pic-deletion mutants showed fewer pathologic changes following H&E staining compared with SF301 ( Figure 5).
Several studies have been conducted to detect virulence genes in Shigella by mPCR, targeting ipaH, ial, and rfc or stx1 for serotype identification [3,5,7,39]. In 2005, Thong [5] first described a new mPCR system to detect S. flexneri 2a by targeting four virulence genes (ipaH, ial, set1A and set1B). This mPCR system was able to determine, in a single reaction, whether genes related to pathogenesis of a particular Shigella strain are associated with the chromosome or plasmid, and whether the serotype of the particular strain can be grouped under S. flexneri 2a [4,5]. In our present study, Thong's mPCR system was modified to identify S. flexneri 2a strains and their virulence using only three virulent genes (ipaH, ial, and set1B). We omitted set1A from the mPCR system, as both set1B and set1A genes have been shown to exist in tandem on PAI-1 of the bacterial chromosome, and they share the same promoter [5,21]. The low prevalence of ial (45/86, 52.3%) verifies that the cell-entry region on the large virulence plasmid of S. flexneri is prone to loss or deletion. The high prevalence of the set1B gene (69/86, 80.2%) verifies that in the rural regions of Zhengding, the isolated epidemic strain of Shigella was S. flexneri 2a. All of our mPCR results were confirmed by serological tests. We confirmed that comparable decreases in virulence occur following the deletion of essential elements in the large virulence plasmid (ipaH and set1B for SF68; and ipaH for SF36) [35][36][37][38]. A clinical SF51 isolate was found to retain ial but had lost set1B, and demonstrated an obvious decrease in HeLa cell invasion. This indicated to us that the chromosome locus around set1B may influence virulence. The location of set1B is known to be in Shigella PAI-1 [7,20], which exists exclusively in S. flexneri 2a. At least four major virulence genes are present in PAI-1 (pic, set1A, set1B, and sigA). The autotransporter SigA exhibits cytopathic effects on HEp-2 cells [40], and the autotransporter Pic exhibits hemagglutination and mucinolytic activities in vitro [20][21][22][23][41][42][43]. Upstream from pic are two IS elements, IS911 and IS629, followed by pic itself, and then a perD IS element [21]. This implies that pic can be spontaneously deleted.
The upstream element int, downstream element orf30, cytopathic factor gene sigA, and the hemagglutinin gene pic on PAI-1 of SF51 were sequenced to verify whether SF51 lost the whole PAI-1 or only part of the genetic locus around set1B. Our results revealed that the entire pic gene on PAI-1 was deleted in this case, whereas other genes (sigA, int, and orf30) were unaffected (Figure 1). This result also suggests that a decrease in virulence of SF51 is not related to sigA, but may be associated with pic deletion.
To confirm that the decreased virulence phenotype in SF51 was associated with deletion of pic, we knocked out pic from the SF301 strain to produce SF301-Δ pic. Additionally, complementation strains SF301-Δ pic/pPic and SF51pic/pPic were constructed to demonstrate that the decreased virulence of SF51 was associated with the deletion of pic. Using gentamicin protection assays, we showed that the Hela cell invasion potential of the pic knockout strains, SF51 and SF301-Δ pic, was decreased compared with the wild-type SF301 strain. This decreased virulence was partially recovered by introducing pSC-pic.
Previous studies have demonstrated that purified recombinant protein Pic (prepared from E.coli HB101 (pPic1)) is not involved in cytotoxic effects on HT29-C1 and HEp-2 cells [24,25]. However, the findings from our current study show that both the clinical and constructed pic-deleted mutants possessed a decreased tendency for cell invasion compared with SF301. Virulence was partially recovered through the insertion of a complementary pic gene into these deletion mutants. Because Pic did not elicit cytopathic effects on epithelial cells, it may be associated with a less efficient interaction process with host cells, lacking any assistance from Figure 5 Histological features of mouse corneas following Sereny tests with SF301 and pic mutants. (A, B) Following inoculation with normal saline, normal corneal epithelium with many layers arranged in an orderly manner can be seen (A: ×50 magnification; B ×400 magnification). (C) After infection with SF301, the corneal epithelium was thinner than that of the control, and vesicular changes (arrowheads) were observed (×100 magnification). (D) Corneal epithelial edema was observed (arrowheads; ×200 magnification). (E) Polymorphic nuclear neutrophilic activity was observed (arrowheads; ×200 magnification). (F) Corneal epithelial derangement and detachment were observed (arrowheads; ×200 magnificaiton). (G) After infection with SF301-Δ pic little damage was observed, but corneal epithelial hyperplasia was noted (arrowheads; ×200 magnification). (H) After infection with SF51, little damage was observed (×200 magnification). bacterial effectors. This phenomenon has also been observed by Vidal et al. [44], who examined the EPEC autotransporter EspC. Purified EspC requires a higher concentration (300 μg/ml vs. 50 μg/ml for other autotransporter cytotoxins) and a longer incubation time (8 h vs. 1 h for EPEC host cells) to produce the same cytotoxic effects as other EPEC isolates. Further studies have confirmed that EspC translocation into epithelial cells results in cytopathic effects in HeLa cells, but require participation of types III and V secretion systems. The mechanism by which Pic is interacted with epithelial cells remains unknown and warrants further study. Further, differences in results observed with Pic regarding decreases in cytopathic effects are likely also associated with other cell lines. Differences in invasion efficiency between Hela cells and HEp-2 cells have been observed for Streptococcus pyrogenes, Campylobacter jejuni and Salmonella typhimurium [45][46][47]; however, the reasons for these differences remain unclear, and further study is required to clarify this.
The mouse Sereny test is commonly used to the test the invasiveness of Shigella [30]. In our work, the virulence of SF51 and SF301-Δ pic was obviously decreased. This was partially recovered by the introduction of pSCpic into deletion mutants. Our findings support the conclusion that pic is associated with the invasion potential of S. flexneri 2a.
Harrington et al. [42] used a mouse model treated with streptomycin to show that Pic promotes intestinal colonization by comparing intestinal colonization abilities of wild-type E. coli 042 and pic mutants (E. coli 042 pic:: aph3 and E. coli 042PicS258A). They demonstrated that the constructed mutants (E. coli 042 pic::aph3 and E. coli 042PicS258A) contained significant defects that adversely affected colonization of mice gastrointestinal tracts compared with E. coli 042. Further work by Harrington et al. suggested that a possible mechanism of promoting intestinal colonization depended on the mucinase activity of Pic. They also showed that this effect is associated with the serine protease catalytic residue in Pic. The research of Harrington et al. supports our findings that Pic is involved in bacterial invasion ability. Whether a decrease in virulence is associated with the mucinase activity of Pic, or other biological activities, should be investigated further.

Conclusions
Our findings suggest that pic, located on PAI-1 of S. flexneri 2a, plays a role in cell invasion during Shigella infections. Further work is necessary to elucidate how Pic affects host-pathogen interactions, and how Pic assists S. flexneri 2a to invade intestinal epithelial cells and cause cytopathic effects.