Reduced Virulence of an fliC Mutant of Shiga-Toxigenic Escherichia coli O113:H21
http://www.100md.com
感染与免疫杂志 2006年第3期
School of Molecular and Biomedical Science, University of Adelaide, Adelaide, S.A. 5005, Australia
ABSTRACT
The contribution of flagellin to the virulence of the O113:H21 Shiga-toxigenic Escherichia coli (STEC) strain 98NK2 was investigated in the streptomycin-treated mouse model. Groups of mice were challenged with either the wild-type STEC or a fliC deletion derivative thereof. There was no difference in the level of gut colonization by the two strains, but the fliC mutant was significantly less virulent than its parent; the overall survival rates were 43.7% and 81.2%, respectively (P < 0.025). This is the first report of a nontoxic accessory virulence factor contributing to a fatal outcome of STEC infection in this model. Although H21 FliC is known to be a potent inducer of CXC chemokines, including interleukin 8, there was no obvious difference in the recruitment of polymorphonuclear leukocytes to the intestinal epithelium of mice challenged with either strain. However, immunofluorescence microscopy suggested that the fliC mutant was less capable of forming a close association with the colonic epithelium. This may have reduced the uptake of Stx2 by mice infected with the mutant.
TEXT
Shiga-toxigenic Escherichia coli (STEC) strains are a major cause of severe gastrointestinal disease in humans. STEC patients typically present with abdominal cramps and/or watery diarrhea, which may be followed by severe bloody diarrhea and hemorrhagic colitis, and in some cases, hemolytic-uremic syndrome (HUS) (11, 14). STEC strains are generally considered to be noninvasive pathogens, and after ingestion and establishment of intestinal colonization, they release Shiga toxin (Stx) into the gut lumen. Toxin is then absorbed across the gut epithelium into the circulation system, where it targets host cells expressing its specific glycolipid receptor globotriaosylceramide (Gb3). In humans, Gb3 receptors are present at high levels in renal tubular epithelium; significant levels are also expressed in microvascular endothelial cells of the kidney, intestine, pancreas, and brain, particularly after exposure to inflammatory cytokines, and Stx-mediated damage at these sites accounts for the pathological features of STEC disease and HUS (14).
The translocation of Stx from the gut lumen into underlying tissues is a crucial step in pathogenesis, and polymorphonuclear leukocytes (PMNs) are now thought to facilitate this process. Recruitment of PMNs to the gut epithelium is postulated to increase paracellular permeability and break down the tight junction barrier, thereby increasing toxin uptake (5). The presence of low doses of Stx1 can also induce both the respiratory burst and superoxide production in PMNs, resulting in localized tissue damage, increased absorption of Stx, and also the induction of Stx2-encoding bacteriophages (7, 24). PMNs have also been reported to play a role in the transport of Stx from the gut to remote target sites (18, 19). These findings stimulated interest in determining the STEC factors responsible for recruiting PMNs to the site of infection. Treatment of intestinal epithelial cells with purified Stx has been shown to induce expression of interleukin 8 (IL-8), a potent PMN chemoattractant (3), as well as other members of the CXC chemokine family, in a dose-dependent manner, via a ribotoxic stress response pathway (20, 21). However, a more recent study using live STEC strains with defined mutations in various virulence-related genes has shown that flagellin (FliC) rather than Stx was responsible for the bulk of the proinflammatory chemokine responses of Hct-8 cells to STEC infection (17). In the present study, the direct contribution of FliC to the pathogenesis of STEC disease was investigated using a streptomycin-treated mouse model.
Bacterial strains. The O113:H21 STEC strain 98NK2, which was responsible for an outbreak of HUS in Adelaide, South Australia, in 1998, has been described previously (13). A derivative of this strain (98NK2fliC), in which nucleotides 64 to 1281 of the fliC gene have been replaced with a (nonpolar) kanamycin resistance cartridge, has also been described previously (17). Spontaneous streptomycin (Str)-resistant derivatives of these strains were isolated by growth on LB agar with a 50-μg streptomycin disk, as previously described (15), and were designated 98NK2SR and 98NK2fliCSR, respectively. Western blot analysis of whole-cell lysates of these strains, using H21 antiserum, showed a 51-kDa immunoreactive species (the expected size of H21 FliC) in 98NK2SR but no H21 flagellin expression in 98NK2fliCSR (result not shown). Motility of 98NK2SR and 98NK2fliCSR was also assessed by growth on semisolid LB agar. The significant difference in the diameters of spread (47 ± 2 mm and 3 ± 1 mm, respectively; P < 0.0001) confirmed that the latter strain was nonmotile. In order to confirm that neither introduction of the fliC mutation nor selection for streptomycin resistance affected production or secretion of Stx2, supernatants from fresh overnight LB broth cultures of 98NK2, 98NK2fliC, 98NK2SR and 98NK2fliCSR were assayed for cytotoxicity on Vero cells as described previously (12); in each case, the Stx2 titer was identical (40,960 50% tissue culture infective doses per ml). To assess general protein synthesis, the four strains were grown in LB in the presence of [3H]leucine and incorporation of label into trichloroacetic acid-insoluble material was measured over 3 h. There was no significant difference in [3H]leucine incorporation in any of the strains at any time point (1, 2, or 3 h) (result not shown).
Virulence in the streptomycin-treated mouse model. Animal experimentation was conducted in accordance with the Australian Code of Practice for the Care and Use of Animals for Scientific Purposes and was approved by the Animal Ethics Committee of the University of Adelaide. Groups of 16 BALB/c mice (5- to 6-week-old males) were pretreated for 24 h with Str (5 mg/ml in drinking water) to inhibit the normal gut microflora, thereby enabling streptomycin-resistant STEC to colonize the gastrointestinal tract as previously described (12). Mice were then fed approximately 108 CFU of strain 98NK2SR or 98NK2fliCSR, suspended in 50 μl 10% NaHCO3 and 20% sucrose. Fecal pellets were collected at intervals thereafter and plated onto LB agar containing Str for CFU determination (Fig. 1). One day after challenge, pellets contained approximately (5.9 ± 1.3) x 109 and (6.8 ± 1.5) x 109 CFU/g, for 98NK2SR- and 98NK2fliCSR-treated mice, respectively. Thereafter, colonization levels of 98NK2SR and 98NK2fliCSR gradually declined to approximately (9.3 ± 3.1) x 108 and (7.0 ± 3.0) x 108 CFU/g, respectively, by day 12 (Fig. 1). There were no significant differences in the levels of colonization of mice challenged with either strain at any time point.
All mice challenged with strain 98NK2SR showed signs of sickness 5 days after challenge, including ruffled fur, lethargy, and reduced food and water intake. In mice that subsequently succumbed, this was followed by hind limb paralysis, ataxia, and death. Interestingly, only one mouse challenged with 98NK2fliCSR showed signs of hind limb paralysis and ataxia in the 4 h prior to death; the remaining mice appeared to be healthy. As shown in Fig. 2, 9 of 16 mice (56%) challenged with 98NK2SR died during the course of the experiment (median survival time, 7.6 days), but only 3 of 16 mice (19%) challenged with 98NK2fliCSR died (median survival time, >14 days). This difference in survival rate was statistically significant (P < 0.025 by the 2 test), directly implicating flagellin in the virulence of 98NK2 in the streptomycin-treated mouse model.
Given the clear role of flagellin in eliciting IL-8 and other CXC chemokine responses in human epithelial cell lines infected with STEC (17), it was tempting to speculate that FliC-mediated induction of analogous chemokines in the mouse gut might contribute to pathogenesis of disease by increasing PMN recruitment, thereby facilitating uptake of Stx. To examine this, additional groups of four streptomycin-treated mice were infected with strain 98NK2SR or 98NK2fliCSR. At 1 and 2 days after challenge, mice were anesthetized by inhalation of Halothane, and blood samples taken via cardiac puncture were collected into EDTA tubes. Blood cell smears were freshly made from 5 μl of EDTA blood, air dried, and fixed by dipping in methanol for 2 min. Smears were then Giemsa stained for 4 min and examined by light microscopy. The percentage of PMNs was determined by differential counting of 100 leukocytes. Infection with either 98NK2SR or 98NK2fliCSR resulted in a significant elevation of peripheral blood PMNs after 1 day relative to uninfected control mice, but there was no significant difference between the two infected groups (Fig. 3). After 2 days, peripheral blood PMN counts for the infected groups were not significantly different from those of the controls (Fig. 3).
One-centimeter segments of distal ileum and proximal colon were also collected from euthanized mice (taken 1 cm from the cecum) and fixed overnight in 4% formaldehyde before being embedded in paraffin. Five-micron sections were cut from each tissue block and placed on saline-coated slides. Sections were dewaxed using two 7-min treatments with Safsolvent (Ajax APS Chemicals, Australia), rehydrated gradually through 2-min washes in 100%, 90%, 70%, and 50% ethanol, and then placed in phosphate-buffered saline (PBS). Before immunostaining, sections were encircled with a Liquid Blocker (Super Pap pen; Daido Sangyo Co., Ltd., Tokyo, Japan). The sections were incubated with 50 μl of 50-μg/ml monoclonal rat antibody to mouse Ly-6G (specific for granulocyte-restricted cell surface protein [mainly neutrophils]) (PharMingen, BD Biosciences) followed by 50 μl of a 1:50 dilution of fluorescein isothiocyanate (FITC)-conjugated donkey anti-rat immunoglobulin (Jackson ImmunoResearch). Both antibodies were diluted in 10% heat-inactivated fetal calf serum in PBS, and the incubations were carried out in a humidified atmosphere in the dark for 1 h at ambient temperature. Slides were washed three times with PBS after each incubation. Stained tissue sections were then mounted in 30% glycerol and examined by fluorescence microscopy (AX 70; Olympus), and the digital images were captured with a Precision digital imaging system (V++; Digital Optics Ltd., Auckland, New Zealand). There was evidence of increased PMN-specific staining of ileal tissue of mice infected with either strain 98NK2SR or 98NK2fliCSR relative to uninfected control mice after 1 day (Fig. 4), but not after 2 days (result not shown). These findings are consistent with the peripheral blood PMN counts shown in Fig. 3. However, in colonic tissue, there was no difference in PMN staining between infected and uninfected mice on either day 1 (Fig. 4) or day 2 (result not shown). Furthermore, there was no obvious difference in PMN staining of ileal or colonic tissue between mice infected with 98NK2SR and those infected with 98NK2fliCSR on either day. Thus, it appears that the significant difference in virulence between 98NK2SR and 98NK2fliCSR cannot be attributed to differences in the level or duration of PMN recruitment to the site of infection.
Although there was no significant difference between strains 98NK2SR and 98NK2fliCSR in terms of their capacity to colonize the gut (Fig. 1), it was possible that the fliC mutation might affect the nature of any interaction between the STEC bacteria and the mucosa. This was examined by subjecting ileal and colonic tissue sections to immunofluorescence microscopy using 1:200 polyclonal rabbit anti-E. coli O113 serum (Institute of Medical and Veterinary Science, Adelaide, Australia) and 1:100 Alexa 488-conjugated donkey anti-rabbit immunoglobulin (Molecular Probes). There was very little detectable staining of bacteria in any of the ileal tissue sections, regardless of challenge organism (result not shown). However, for colonic sections, there was evidence of a close interaction between 98NK2SR and the epithelial cell surface, including penetration of occasional bacteria into the cells (Fig. 5). In contrast, 98NK2fliCSR was detected principally in the lumen, with little evidence of direct interaction with the epithelial surface (Fig. 5). This difference in host cell interaction was not, however, reflected in in vitro adherence assays. We found no significant difference between the adherence of wild-type 98NK2 and the fliC mutant to either Hct-8 (human colonic epithelial) or HEp-2 cells (result not shown).
Conclusions. In this study we have shown a clear role for flagellin in the virulence of the O113:H21 STEC strain 98NK2 in the streptomycin-treated mouse model. The overall survival rate for mice challenged with the wild-type STEC was 43.7% compared with 81.2% for those challenged with the fliC mutant. The streptomycin-treated mouse model of STEC disease was established by Wadolkowski et al. (22, 23), and in this model virulence is manifested as fatal Stx2-dependent renal tubular necrosis. Not all STEC strains are virulent in this model, but E. coli K-12 strains expressing a range of stx2-related genes cloned in high- rather than low-copy-number plasmids are lethal (8, 12, 23). This suggests that other STEC virulence factors are unimportant if Stx2 production is sufficiently high. The present study is the first report of a nontoxic accessory virulence factor contributing to a fatal outcome of STEC infection in this model.
We have previously shown that H21 and other common STEC flagellin types (e.g., H7) are potent inducers of IL-8 and other CXC chemokine responses in human colonic epithelial cells in vitro (17). In cases of human STEC disease, elevated IL-8 levels in plasma have been shown to correlate with both leukocytosis and poor prognosis (2, 6, 9, 10, 25). We postulated that chemokine induction by STEC flagellin would translate into increased PMN recruitment to the intestinal epithelium in vivo and that this would increase penetration of Stx2 from the gut lumen into underlying tissues, as proposed by Hurley et al. (5). However, although there was a clear difference in virulence, we could find no evidence for a difference in PMN recruitment to the intestinal mucosa between mice challenged with the wild-type or fliC mutant STEC strains. Furthermore, although STEC infection resulted in elevated PMN counts in peripheral blood relative to uninfected mice, there was no difference in PMN counts between groups infected with strains 98NK2SR and 98NK2fliCSR. On the other hand, although there was no difference in gut colonization as judged by fecal CFU counts, we did see a marked difference in the intimacy of interaction between the STEC strains themselves and the colonic, but not the ileal, mucosa. Wild-type 98NK2SR was closely associated with the epithelial surface, while 98NK2fliCSR largely remained in the lumen. These findings are consistent with the known role of flagella as mediators of adherence to, and invasion of, epithelial cells by a range of pathogenic bacteria (16). Flagella are known to contribute to adherence and invasion by facilitating penetration of the mucous layer covering the gastrointestinal mucosa, thereby providing the pathogen with access to receptors on epithelial cells (16). However, there is evidence that, at least in some pathogens, flagella can also contribute to colonization by acting directly as adhesins. Giron et al. (4) have shown that fliC mutants of enteropathogenic E. coli are defective in adherence to epithelial cell lines in vitro and that H2 and H6 flagella purified from enteropathogenic E. coli, but not H7 flagella from STEC, are capable of directly binding to HeLa cells. Our own findings did not support a role for H21 FliC in adherence of 98NK2 to Hct-8 or HEp-2 cells in vitro. Best et al. (1) have recently reported that H7 flagella contribute significantly to gastrointestinal colonization of chickens by nontoxigenic E. coli O157:H7. However, it is not clear from the latter study whether the beneficial contribution of H7 flagella to colonization is due to enhanced penetration of mucus or direct FliC-mediated adherence.
Notwithstanding the above uncertainties regarding the manner in which flagellin contributes to gut colonization, our finding that FliC contributes directly to STEC virulence in streptomycin-treated mice provides a model system in which to explore the protective efficacy of STEC FliC-based vaccines. Vaccines based on STEC adherence factors, such as intimin, are already being tested for capacity to prevent colonization of livestock, and supplementation with common STEC FliC types may further minimize carriage and hence entry of STEC into the human food chain.
ACKNOWLEDGMENTS
This work was supported by program grant 284214 from the National Health and Medical Research Council of Australia.
REFERENCES
1. Best, A., R. M. La Ragione, A. R. Sayers, and M. J. Woodward. 2005. Role for flagella but not intimin in the persistent infection of the gastrointestinal tissues of specific-pathogen-free chicks by Shiga toxin-negative Escherichia coli O157:H7. Infect. Immun. 73:1836-1846.
2. Fitzpatrick, M. M., V. Shah, R. S. Trompeter, M. J. Dillon, and T. M. Barratt. 1992. Interleukin-8 and polymorphoneutrophil leucocyte activation in hemolytic uremic syndrome of childhood. Kidney Int. 42:951-956.
3. Gale, L. M., and S. R. McColl. 1999. Chemokines: extracellular messengers for all occasions BioEssays 21:17-28.
4. Giron, J. A., A. G. Torres, E. Freer, and J. B. Kaper. 2002. The flagella of enteropathogenic Escherichia coli mediate adherence to epithelial cells. Mol. Microbiol. 44:361-379.
5. Hurley, B. P., C. M. Thorpe, and D. W. K. Acheson. 2001. Shiga toxin translocation across intestinal epithelial cells is enhanced by neutrophil transmigration. Infect. Immun. 69:6148-6155.
6. Karpman, D., A. Andreasson, H. Thysell, B. S. Kaplan, and C. Svanborg. 1995. Cytokines in childhood hemolytic uremic syndrome and thrombotic thrombocytopenic purpura. Pediatr. Nephrol. 9:694-699.
7. King, A. J., S. Sundaram, M. Cendoroglo, D. W. K. Acheson, and G. T. Keusch. 1999. Shiga toxin induces superoxide production in polymorphonuclear cells with subsequent impairment of phagocytosis and responsiveness to phorbol esters. J. Infect. Dis. 179:503-507.
8. Lindgren, S. W., J. E. Samuel, C. K. Schmitt, and A. D. O'Brien. 1994. The specific activities of Shiga-like toxin type II (SLT-II) and SLT-II-related toxins of enterohemorrhagic Escherichia coli differ when measured by Vero cell cytotoxicity but not by mouse lethality. Infect. Immun. 62:623-631.
9. Litalien, C., F. Proulx, M. M. Mariscalco, P. Robitaille, J. P. Turgeon, E. Orrbine, P. C. Rowe, P. N. McLaine, and E. Seidman. 1999. Circulating inflammatory cytokine levels in hemolytic uremic syndrome. Pediatr. Nephrol. 13:840-845.
10. Murata, A., T. Shimazu, T. Yamamoto, N. Taenaka, K.-I. Nagayama, T. Honda, H. Sugimoto, M. Monden, N. Matsuura, and S. Okada. 1998. Profiles of circulating inflammatory- and anti-inflammatory cytokines in patients with hemolytic uremic syndrome due to E. coli O157 infection. Cytokine 10:544-548.
11. Nataro, J. P., and J. B. Kaper. 1998. Diarrheagenic Escherichia coli. Clin. Microbiol. Rev. 11:142-201.
12. Paton, A. W., A. J. Bourne, P. A. Manning, and J. C. Paton. 1995. Comparative toxicity and virulence of Escherichia coli clones expressing variant and chimeric Shiga-like toxin type II operons. Infect. Immun. 63:2450-2458.
13. Paton, A. W., M. C. Woodrow, R. M. Doyle, J. A. Lanser, and J. C. Paton. 1999. Molecular characterization of a Shiga toxigenic Escherichia coli O113:H21 strain lacking eae responsible for a cluster of cases of hemolytic-uremic syndrome. J. Clin. Microbiol. 37:3357-3361.
14. Paton, J. C., and A. W. Paton. 1998. Pathogenesis and diagnosis of Shiga toxin-producing Escherichia coli infections. Clin. Microbiol. Rev. 11:450-479.
15. Paton, J. C., T. J. Rogers, R. Morona, and A. W. Paton. 2001. Oral administration of formaldehyde-killed recombinant bacteria expressing a mimic of the Shiga toxin receptor protects mice from fatal challenge with Shiga-toxigenic Escherichia coli. Infect. Immun. 69:1389-1393.
16. Ramos, H. C., M. Rumbo, and J. C. Sirard. 2004. Bacterial flagellins: mediators of pathogenicity and host immune responses in mucosa. Trends Microbiol. 12:509-517.
17. Rogers, T. J., A. W. Paton, S. R. McColl, and J. C. Paton. 2003. Enhanced CXC chemokine responses of human colonic epithelial cells to locus of enterocyte effacement- negative Shiga-toxigenic Escherichia coli. Infect. Immun. 71:5623-5632.
18. te Loo, D. M. W. M., L. A. H. Monnens, T. J. A. M. van der Velden, M. A. Vermeer, F. Preyers, P. N. M. Demacker, L. P. W. J. van den Heuvel, and V. W. M. van Hinsberg. 2000. Binding and transfer of verocytotoxin by polymorphonuclear leukocytes in haemolytic uremic syndrome. Blood 95:3396-3402.
19. te Loo, D. M. W. M., V. W. M. Van Hinsbergh, L. P. W. J. Van Den Heuvel, and L. A. H. Monnens. 2000. Detection of verocytotoxin bound to circulating polymorphonuclear leukocytes of patients with hemolytic uremic syndrome. J. Am. Soc. Nephrol. 12:800-806.
20. Thorpe, C. M., B. P. Hurley, L. L. Lincicome, M. S. Jacewicz, G. T. Keusch, and D. W. K. Acheson. 1999. Shiga toxins stimulate secretion of interleukin-8 from intestinal epithelial cells. Infect. Immun. 67:5985-5993.
21. Thorpe, C. M., W. E. Smith, B. P. Hurley, and D. W. K. Acheson. 2001. Shiga toxins induce, superinduce, and stabilize a variety of C-X-C chemokine mRNAs in intestinal epithelial cells, resulting in increased chemokine expression. Infect. Immun. 69:6140-6147.
22. Wadolkowski, E. A., J. A. Burris, and A. D. O'Brien. 1990. Mouse model for colonization and disease caused by enterohemorrhagic Escherichia coli O157:H7. Infect. Immun. 58:2438-2445.
23. Wadolkowski, E. A., L. M. Sung, J. A. Burris, J. E. Samuel, and A. D. O'Brien. 1990. Acute renal tubular necrosis and death of mice orally infected with Escherichia coli strains that produce Shiga-like toxin type II. Infect. Immun. 58:3959-3965.
24. Wagner, P. L., D. W. K. Acheson, and M. K. Waldor. 2001. Human neutrophils and their products induce Shiga toxin production by enterohemorrhagic Escherichia coli. Infect. Immun. 69:1934-1937.
25. Westerholt, S., T. Hartung, M. Tollens, A. Gustrau, M. Oberhoffer, H. Karch, B. Klare, K. Pfeffer, P. Emmrich, and R. Oberhoffer. 2000. Inflammatory and immunological parameters in children with haemolytic uremic syndrome (HUS) and gastroenteritis: pathophysiological and diagnostic clues. Cytokine 12:822-827.(Trisha J. Rogers, James C)
ABSTRACT
The contribution of flagellin to the virulence of the O113:H21 Shiga-toxigenic Escherichia coli (STEC) strain 98NK2 was investigated in the streptomycin-treated mouse model. Groups of mice were challenged with either the wild-type STEC or a fliC deletion derivative thereof. There was no difference in the level of gut colonization by the two strains, but the fliC mutant was significantly less virulent than its parent; the overall survival rates were 43.7% and 81.2%, respectively (P < 0.025). This is the first report of a nontoxic accessory virulence factor contributing to a fatal outcome of STEC infection in this model. Although H21 FliC is known to be a potent inducer of CXC chemokines, including interleukin 8, there was no obvious difference in the recruitment of polymorphonuclear leukocytes to the intestinal epithelium of mice challenged with either strain. However, immunofluorescence microscopy suggested that the fliC mutant was less capable of forming a close association with the colonic epithelium. This may have reduced the uptake of Stx2 by mice infected with the mutant.
TEXT
Shiga-toxigenic Escherichia coli (STEC) strains are a major cause of severe gastrointestinal disease in humans. STEC patients typically present with abdominal cramps and/or watery diarrhea, which may be followed by severe bloody diarrhea and hemorrhagic colitis, and in some cases, hemolytic-uremic syndrome (HUS) (11, 14). STEC strains are generally considered to be noninvasive pathogens, and after ingestion and establishment of intestinal colonization, they release Shiga toxin (Stx) into the gut lumen. Toxin is then absorbed across the gut epithelium into the circulation system, where it targets host cells expressing its specific glycolipid receptor globotriaosylceramide (Gb3). In humans, Gb3 receptors are present at high levels in renal tubular epithelium; significant levels are also expressed in microvascular endothelial cells of the kidney, intestine, pancreas, and brain, particularly after exposure to inflammatory cytokines, and Stx-mediated damage at these sites accounts for the pathological features of STEC disease and HUS (14).
The translocation of Stx from the gut lumen into underlying tissues is a crucial step in pathogenesis, and polymorphonuclear leukocytes (PMNs) are now thought to facilitate this process. Recruitment of PMNs to the gut epithelium is postulated to increase paracellular permeability and break down the tight junction barrier, thereby increasing toxin uptake (5). The presence of low doses of Stx1 can also induce both the respiratory burst and superoxide production in PMNs, resulting in localized tissue damage, increased absorption of Stx, and also the induction of Stx2-encoding bacteriophages (7, 24). PMNs have also been reported to play a role in the transport of Stx from the gut to remote target sites (18, 19). These findings stimulated interest in determining the STEC factors responsible for recruiting PMNs to the site of infection. Treatment of intestinal epithelial cells with purified Stx has been shown to induce expression of interleukin 8 (IL-8), a potent PMN chemoattractant (3), as well as other members of the CXC chemokine family, in a dose-dependent manner, via a ribotoxic stress response pathway (20, 21). However, a more recent study using live STEC strains with defined mutations in various virulence-related genes has shown that flagellin (FliC) rather than Stx was responsible for the bulk of the proinflammatory chemokine responses of Hct-8 cells to STEC infection (17). In the present study, the direct contribution of FliC to the pathogenesis of STEC disease was investigated using a streptomycin-treated mouse model.
Bacterial strains. The O113:H21 STEC strain 98NK2, which was responsible for an outbreak of HUS in Adelaide, South Australia, in 1998, has been described previously (13). A derivative of this strain (98NK2fliC), in which nucleotides 64 to 1281 of the fliC gene have been replaced with a (nonpolar) kanamycin resistance cartridge, has also been described previously (17). Spontaneous streptomycin (Str)-resistant derivatives of these strains were isolated by growth on LB agar with a 50-μg streptomycin disk, as previously described (15), and were designated 98NK2SR and 98NK2fliCSR, respectively. Western blot analysis of whole-cell lysates of these strains, using H21 antiserum, showed a 51-kDa immunoreactive species (the expected size of H21 FliC) in 98NK2SR but no H21 flagellin expression in 98NK2fliCSR (result not shown). Motility of 98NK2SR and 98NK2fliCSR was also assessed by growth on semisolid LB agar. The significant difference in the diameters of spread (47 ± 2 mm and 3 ± 1 mm, respectively; P < 0.0001) confirmed that the latter strain was nonmotile. In order to confirm that neither introduction of the fliC mutation nor selection for streptomycin resistance affected production or secretion of Stx2, supernatants from fresh overnight LB broth cultures of 98NK2, 98NK2fliC, 98NK2SR and 98NK2fliCSR were assayed for cytotoxicity on Vero cells as described previously (12); in each case, the Stx2 titer was identical (40,960 50% tissue culture infective doses per ml). To assess general protein synthesis, the four strains were grown in LB in the presence of [3H]leucine and incorporation of label into trichloroacetic acid-insoluble material was measured over 3 h. There was no significant difference in [3H]leucine incorporation in any of the strains at any time point (1, 2, or 3 h) (result not shown).
Virulence in the streptomycin-treated mouse model. Animal experimentation was conducted in accordance with the Australian Code of Practice for the Care and Use of Animals for Scientific Purposes and was approved by the Animal Ethics Committee of the University of Adelaide. Groups of 16 BALB/c mice (5- to 6-week-old males) were pretreated for 24 h with Str (5 mg/ml in drinking water) to inhibit the normal gut microflora, thereby enabling streptomycin-resistant STEC to colonize the gastrointestinal tract as previously described (12). Mice were then fed approximately 108 CFU of strain 98NK2SR or 98NK2fliCSR, suspended in 50 μl 10% NaHCO3 and 20% sucrose. Fecal pellets were collected at intervals thereafter and plated onto LB agar containing Str for CFU determination (Fig. 1). One day after challenge, pellets contained approximately (5.9 ± 1.3) x 109 and (6.8 ± 1.5) x 109 CFU/g, for 98NK2SR- and 98NK2fliCSR-treated mice, respectively. Thereafter, colonization levels of 98NK2SR and 98NK2fliCSR gradually declined to approximately (9.3 ± 3.1) x 108 and (7.0 ± 3.0) x 108 CFU/g, respectively, by day 12 (Fig. 1). There were no significant differences in the levels of colonization of mice challenged with either strain at any time point.
All mice challenged with strain 98NK2SR showed signs of sickness 5 days after challenge, including ruffled fur, lethargy, and reduced food and water intake. In mice that subsequently succumbed, this was followed by hind limb paralysis, ataxia, and death. Interestingly, only one mouse challenged with 98NK2fliCSR showed signs of hind limb paralysis and ataxia in the 4 h prior to death; the remaining mice appeared to be healthy. As shown in Fig. 2, 9 of 16 mice (56%) challenged with 98NK2SR died during the course of the experiment (median survival time, 7.6 days), but only 3 of 16 mice (19%) challenged with 98NK2fliCSR died (median survival time, >14 days). This difference in survival rate was statistically significant (P < 0.025 by the 2 test), directly implicating flagellin in the virulence of 98NK2 in the streptomycin-treated mouse model.
Given the clear role of flagellin in eliciting IL-8 and other CXC chemokine responses in human epithelial cell lines infected with STEC (17), it was tempting to speculate that FliC-mediated induction of analogous chemokines in the mouse gut might contribute to pathogenesis of disease by increasing PMN recruitment, thereby facilitating uptake of Stx. To examine this, additional groups of four streptomycin-treated mice were infected with strain 98NK2SR or 98NK2fliCSR. At 1 and 2 days after challenge, mice were anesthetized by inhalation of Halothane, and blood samples taken via cardiac puncture were collected into EDTA tubes. Blood cell smears were freshly made from 5 μl of EDTA blood, air dried, and fixed by dipping in methanol for 2 min. Smears were then Giemsa stained for 4 min and examined by light microscopy. The percentage of PMNs was determined by differential counting of 100 leukocytes. Infection with either 98NK2SR or 98NK2fliCSR resulted in a significant elevation of peripheral blood PMNs after 1 day relative to uninfected control mice, but there was no significant difference between the two infected groups (Fig. 3). After 2 days, peripheral blood PMN counts for the infected groups were not significantly different from those of the controls (Fig. 3).
One-centimeter segments of distal ileum and proximal colon were also collected from euthanized mice (taken 1 cm from the cecum) and fixed overnight in 4% formaldehyde before being embedded in paraffin. Five-micron sections were cut from each tissue block and placed on saline-coated slides. Sections were dewaxed using two 7-min treatments with Safsolvent (Ajax APS Chemicals, Australia), rehydrated gradually through 2-min washes in 100%, 90%, 70%, and 50% ethanol, and then placed in phosphate-buffered saline (PBS). Before immunostaining, sections were encircled with a Liquid Blocker (Super Pap pen; Daido Sangyo Co., Ltd., Tokyo, Japan). The sections were incubated with 50 μl of 50-μg/ml monoclonal rat antibody to mouse Ly-6G (specific for granulocyte-restricted cell surface protein [mainly neutrophils]) (PharMingen, BD Biosciences) followed by 50 μl of a 1:50 dilution of fluorescein isothiocyanate (FITC)-conjugated donkey anti-rat immunoglobulin (Jackson ImmunoResearch). Both antibodies were diluted in 10% heat-inactivated fetal calf serum in PBS, and the incubations were carried out in a humidified atmosphere in the dark for 1 h at ambient temperature. Slides were washed three times with PBS after each incubation. Stained tissue sections were then mounted in 30% glycerol and examined by fluorescence microscopy (AX 70; Olympus), and the digital images were captured with a Precision digital imaging system (V++; Digital Optics Ltd., Auckland, New Zealand). There was evidence of increased PMN-specific staining of ileal tissue of mice infected with either strain 98NK2SR or 98NK2fliCSR relative to uninfected control mice after 1 day (Fig. 4), but not after 2 days (result not shown). These findings are consistent with the peripheral blood PMN counts shown in Fig. 3. However, in colonic tissue, there was no difference in PMN staining between infected and uninfected mice on either day 1 (Fig. 4) or day 2 (result not shown). Furthermore, there was no obvious difference in PMN staining of ileal or colonic tissue between mice infected with 98NK2SR and those infected with 98NK2fliCSR on either day. Thus, it appears that the significant difference in virulence between 98NK2SR and 98NK2fliCSR cannot be attributed to differences in the level or duration of PMN recruitment to the site of infection.
Although there was no significant difference between strains 98NK2SR and 98NK2fliCSR in terms of their capacity to colonize the gut (Fig. 1), it was possible that the fliC mutation might affect the nature of any interaction between the STEC bacteria and the mucosa. This was examined by subjecting ileal and colonic tissue sections to immunofluorescence microscopy using 1:200 polyclonal rabbit anti-E. coli O113 serum (Institute of Medical and Veterinary Science, Adelaide, Australia) and 1:100 Alexa 488-conjugated donkey anti-rabbit immunoglobulin (Molecular Probes). There was very little detectable staining of bacteria in any of the ileal tissue sections, regardless of challenge organism (result not shown). However, for colonic sections, there was evidence of a close interaction between 98NK2SR and the epithelial cell surface, including penetration of occasional bacteria into the cells (Fig. 5). In contrast, 98NK2fliCSR was detected principally in the lumen, with little evidence of direct interaction with the epithelial surface (Fig. 5). This difference in host cell interaction was not, however, reflected in in vitro adherence assays. We found no significant difference between the adherence of wild-type 98NK2 and the fliC mutant to either Hct-8 (human colonic epithelial) or HEp-2 cells (result not shown).
Conclusions. In this study we have shown a clear role for flagellin in the virulence of the O113:H21 STEC strain 98NK2 in the streptomycin-treated mouse model. The overall survival rate for mice challenged with the wild-type STEC was 43.7% compared with 81.2% for those challenged with the fliC mutant. The streptomycin-treated mouse model of STEC disease was established by Wadolkowski et al. (22, 23), and in this model virulence is manifested as fatal Stx2-dependent renal tubular necrosis. Not all STEC strains are virulent in this model, but E. coli K-12 strains expressing a range of stx2-related genes cloned in high- rather than low-copy-number plasmids are lethal (8, 12, 23). This suggests that other STEC virulence factors are unimportant if Stx2 production is sufficiently high. The present study is the first report of a nontoxic accessory virulence factor contributing to a fatal outcome of STEC infection in this model.
We have previously shown that H21 and other common STEC flagellin types (e.g., H7) are potent inducers of IL-8 and other CXC chemokine responses in human colonic epithelial cells in vitro (17). In cases of human STEC disease, elevated IL-8 levels in plasma have been shown to correlate with both leukocytosis and poor prognosis (2, 6, 9, 10, 25). We postulated that chemokine induction by STEC flagellin would translate into increased PMN recruitment to the intestinal epithelium in vivo and that this would increase penetration of Stx2 from the gut lumen into underlying tissues, as proposed by Hurley et al. (5). However, although there was a clear difference in virulence, we could find no evidence for a difference in PMN recruitment to the intestinal mucosa between mice challenged with the wild-type or fliC mutant STEC strains. Furthermore, although STEC infection resulted in elevated PMN counts in peripheral blood relative to uninfected mice, there was no difference in PMN counts between groups infected with strains 98NK2SR and 98NK2fliCSR. On the other hand, although there was no difference in gut colonization as judged by fecal CFU counts, we did see a marked difference in the intimacy of interaction between the STEC strains themselves and the colonic, but not the ileal, mucosa. Wild-type 98NK2SR was closely associated with the epithelial surface, while 98NK2fliCSR largely remained in the lumen. These findings are consistent with the known role of flagella as mediators of adherence to, and invasion of, epithelial cells by a range of pathogenic bacteria (16). Flagella are known to contribute to adherence and invasion by facilitating penetration of the mucous layer covering the gastrointestinal mucosa, thereby providing the pathogen with access to receptors on epithelial cells (16). However, there is evidence that, at least in some pathogens, flagella can also contribute to colonization by acting directly as adhesins. Giron et al. (4) have shown that fliC mutants of enteropathogenic E. coli are defective in adherence to epithelial cell lines in vitro and that H2 and H6 flagella purified from enteropathogenic E. coli, but not H7 flagella from STEC, are capable of directly binding to HeLa cells. Our own findings did not support a role for H21 FliC in adherence of 98NK2 to Hct-8 or HEp-2 cells in vitro. Best et al. (1) have recently reported that H7 flagella contribute significantly to gastrointestinal colonization of chickens by nontoxigenic E. coli O157:H7. However, it is not clear from the latter study whether the beneficial contribution of H7 flagella to colonization is due to enhanced penetration of mucus or direct FliC-mediated adherence.
Notwithstanding the above uncertainties regarding the manner in which flagellin contributes to gut colonization, our finding that FliC contributes directly to STEC virulence in streptomycin-treated mice provides a model system in which to explore the protective efficacy of STEC FliC-based vaccines. Vaccines based on STEC adherence factors, such as intimin, are already being tested for capacity to prevent colonization of livestock, and supplementation with common STEC FliC types may further minimize carriage and hence entry of STEC into the human food chain.
ACKNOWLEDGMENTS
This work was supported by program grant 284214 from the National Health and Medical Research Council of Australia.
REFERENCES
1. Best, A., R. M. La Ragione, A. R. Sayers, and M. J. Woodward. 2005. Role for flagella but not intimin in the persistent infection of the gastrointestinal tissues of specific-pathogen-free chicks by Shiga toxin-negative Escherichia coli O157:H7. Infect. Immun. 73:1836-1846.
2. Fitzpatrick, M. M., V. Shah, R. S. Trompeter, M. J. Dillon, and T. M. Barratt. 1992. Interleukin-8 and polymorphoneutrophil leucocyte activation in hemolytic uremic syndrome of childhood. Kidney Int. 42:951-956.
3. Gale, L. M., and S. R. McColl. 1999. Chemokines: extracellular messengers for all occasions BioEssays 21:17-28.
4. Giron, J. A., A. G. Torres, E. Freer, and J. B. Kaper. 2002. The flagella of enteropathogenic Escherichia coli mediate adherence to epithelial cells. Mol. Microbiol. 44:361-379.
5. Hurley, B. P., C. M. Thorpe, and D. W. K. Acheson. 2001. Shiga toxin translocation across intestinal epithelial cells is enhanced by neutrophil transmigration. Infect. Immun. 69:6148-6155.
6. Karpman, D., A. Andreasson, H. Thysell, B. S. Kaplan, and C. Svanborg. 1995. Cytokines in childhood hemolytic uremic syndrome and thrombotic thrombocytopenic purpura. Pediatr. Nephrol. 9:694-699.
7. King, A. J., S. Sundaram, M. Cendoroglo, D. W. K. Acheson, and G. T. Keusch. 1999. Shiga toxin induces superoxide production in polymorphonuclear cells with subsequent impairment of phagocytosis and responsiveness to phorbol esters. J. Infect. Dis. 179:503-507.
8. Lindgren, S. W., J. E. Samuel, C. K. Schmitt, and A. D. O'Brien. 1994. The specific activities of Shiga-like toxin type II (SLT-II) and SLT-II-related toxins of enterohemorrhagic Escherichia coli differ when measured by Vero cell cytotoxicity but not by mouse lethality. Infect. Immun. 62:623-631.
9. Litalien, C., F. Proulx, M. M. Mariscalco, P. Robitaille, J. P. Turgeon, E. Orrbine, P. C. Rowe, P. N. McLaine, and E. Seidman. 1999. Circulating inflammatory cytokine levels in hemolytic uremic syndrome. Pediatr. Nephrol. 13:840-845.
10. Murata, A., T. Shimazu, T. Yamamoto, N. Taenaka, K.-I. Nagayama, T. Honda, H. Sugimoto, M. Monden, N. Matsuura, and S. Okada. 1998. Profiles of circulating inflammatory- and anti-inflammatory cytokines in patients with hemolytic uremic syndrome due to E. coli O157 infection. Cytokine 10:544-548.
11. Nataro, J. P., and J. B. Kaper. 1998. Diarrheagenic Escherichia coli. Clin. Microbiol. Rev. 11:142-201.
12. Paton, A. W., A. J. Bourne, P. A. Manning, and J. C. Paton. 1995. Comparative toxicity and virulence of Escherichia coli clones expressing variant and chimeric Shiga-like toxin type II operons. Infect. Immun. 63:2450-2458.
13. Paton, A. W., M. C. Woodrow, R. M. Doyle, J. A. Lanser, and J. C. Paton. 1999. Molecular characterization of a Shiga toxigenic Escherichia coli O113:H21 strain lacking eae responsible for a cluster of cases of hemolytic-uremic syndrome. J. Clin. Microbiol. 37:3357-3361.
14. Paton, J. C., and A. W. Paton. 1998. Pathogenesis and diagnosis of Shiga toxin-producing Escherichia coli infections. Clin. Microbiol. Rev. 11:450-479.
15. Paton, J. C., T. J. Rogers, R. Morona, and A. W. Paton. 2001. Oral administration of formaldehyde-killed recombinant bacteria expressing a mimic of the Shiga toxin receptor protects mice from fatal challenge with Shiga-toxigenic Escherichia coli. Infect. Immun. 69:1389-1393.
16. Ramos, H. C., M. Rumbo, and J. C. Sirard. 2004. Bacterial flagellins: mediators of pathogenicity and host immune responses in mucosa. Trends Microbiol. 12:509-517.
17. Rogers, T. J., A. W. Paton, S. R. McColl, and J. C. Paton. 2003. Enhanced CXC chemokine responses of human colonic epithelial cells to locus of enterocyte effacement- negative Shiga-toxigenic Escherichia coli. Infect. Immun. 71:5623-5632.
18. te Loo, D. M. W. M., L. A. H. Monnens, T. J. A. M. van der Velden, M. A. Vermeer, F. Preyers, P. N. M. Demacker, L. P. W. J. van den Heuvel, and V. W. M. van Hinsberg. 2000. Binding and transfer of verocytotoxin by polymorphonuclear leukocytes in haemolytic uremic syndrome. Blood 95:3396-3402.
19. te Loo, D. M. W. M., V. W. M. Van Hinsbergh, L. P. W. J. Van Den Heuvel, and L. A. H. Monnens. 2000. Detection of verocytotoxin bound to circulating polymorphonuclear leukocytes of patients with hemolytic uremic syndrome. J. Am. Soc. Nephrol. 12:800-806.
20. Thorpe, C. M., B. P. Hurley, L. L. Lincicome, M. S. Jacewicz, G. T. Keusch, and D. W. K. Acheson. 1999. Shiga toxins stimulate secretion of interleukin-8 from intestinal epithelial cells. Infect. Immun. 67:5985-5993.
21. Thorpe, C. M., W. E. Smith, B. P. Hurley, and D. W. K. Acheson. 2001. Shiga toxins induce, superinduce, and stabilize a variety of C-X-C chemokine mRNAs in intestinal epithelial cells, resulting in increased chemokine expression. Infect. Immun. 69:6140-6147.
22. Wadolkowski, E. A., J. A. Burris, and A. D. O'Brien. 1990. Mouse model for colonization and disease caused by enterohemorrhagic Escherichia coli O157:H7. Infect. Immun. 58:2438-2445.
23. Wadolkowski, E. A., L. M. Sung, J. A. Burris, J. E. Samuel, and A. D. O'Brien. 1990. Acute renal tubular necrosis and death of mice orally infected with Escherichia coli strains that produce Shiga-like toxin type II. Infect. Immun. 58:3959-3965.
24. Wagner, P. L., D. W. K. Acheson, and M. K. Waldor. 2001. Human neutrophils and their products induce Shiga toxin production by enterohemorrhagic Escherichia coli. Infect. Immun. 69:1934-1937.
25. Westerholt, S., T. Hartung, M. Tollens, A. Gustrau, M. Oberhoffer, H. Karch, B. Klare, K. Pfeffer, P. Emmrich, and R. Oberhoffer. 2000. Inflammatory and immunological parameters in children with haemolytic uremic syndrome (HUS) and gastroenteritis: pathophysiological and diagnostic clues. Cytokine 12:822-827.(Trisha J. Rogers, James C)