Vol. 58, No. 7
INFECTION AND IMMUNITY, JUlY 1990, p. 2401-2403 0019-9567/90/072401-03$02.00/0 Copyright C 1990, American Society for Microbiology
Comparative Study of Attachment to and Invasion of Epithelial Cell Lines by Shigella dysenteriae ANUBHA SEN, MYRON A. LEON, AND SUNIL PALCHAUDHURI* Department of Immunology and Microbiology, Wayne State University School of Medicine, Detroit, Michigan 48201 Received 27 December 1989/Accepted 20 April 1990 Henle 407 and HeLa cells were compared as hosts for Shigella dysenteriae at a low multiplicity of infection. Efficiency of attachment and invasion without centrifugation, as well as selectivity for pathogenic over nonpathogenic S. dysenteriae without Congo red, were much greater for Henle 407 cells than for HeLa cells.
portion of the invasion plasmid and lacks all known ipa genes (unpublished data). After every five or six passages on artificial media, bacterial virulence was monitored by the Sereny test as well as by size determination of invasion plasmids (12). Strains which produced keratoconjunctivitis within 48 h were streaked on Congo red-TSB agar (3), to which galactose was added to a final concentration of 0.2%, or L agar plus streptomycin (100 ,ug/ml). Colonies were tested on minimal medium plates, with and without supplements, to confirm the genetic purity of the strains isolated
The ability of enteroinvasive bacteria such as Shigella spp. to invade cell culture monolayers has been used as a measure of their pathogenicity. Indeed, Shigellaflexneri and Shigella sonnei strains which lose the ability to penetrate HeLa cells are uniformly avirulent in animal hosts (7, 14-16). The invasive phenotype has been mapped to a 220-kilobase extrachromosomal element. Spontaneous deletions in the plasmid lead to noninvasive variants (16). The weak binding between Shigella spp. and HeLa cells causes inefficient adherence, and therefore invasion assays have not been highly reproducible. Daskaleros and Payne (3) have improved the efficiency of this interaction for S. flexneri by pretreating HeLa cells or bacterial culture with Congo red dye or hemin. Problems with the invasion assay also arise in assessing the specific location of the bacteria after reaction with the HeLa cells, i.e., whether they are intracellular, residing between the host cells, or attached to the cell monolayers. In order to differentiate adherent bacteria from intracellular organisms, investigators have used a variety of physical, microbiological, and chemical techniques (1, 9, 10, 17). The majority of investigators have used antibiotics to remove or kill adherent noninternalized bacteria (1, 5, 6, 9, 11, 17). Most studies of invasion mechanisms have used S. flexneri, whereas studies with Shigella dysenteriae isolates are sparse (2, 4, 5, 10). However, in recent years S. dysenteriae has been the causative organism of most of the outbreaks of shigellosis in developing countries, with a high percentage of child mortality (7, 12, 18). We have, therefore, established a sensitive in vitro model for studying invasion by S. dysenteriae. Monolayers of the cell line Henle 407 have been found to be susceptible hosts for S. dysenteriae infection in the absence of either centrifugation or addition of Congo red and thus provide a useful model for studying the interaction of S. dysenteriae with normal host cells. Fifty strains of S. dysenteriae were collected from India, Bangladesh, and other sources. Analysis demonstrated that their large invasion plasmids all had similar sizes (220 kilobases). Invasion plasmids from 10 of these 50 isolates were digested with EcoRI. Patterns obtained by agarose gel electrophoresis were identical (12). Accordingly, one isolate, CG097, supplied by J. M. Buysse from the stocks of D. J. Kopecko, was chosen as the representative virulent organism for our studies. Virulent CG097 (arg trp met Sm9) and an avirulent derivative, CG097-2 (arg trp met Smr), were grown on L agar plus streptomycin (100 ,ug/ml), as these strains are resistant to streptomycin. CG097-2 has lost a 37-kilobase *
from single colonies. Cultures were stored at room temperature, rather than at 4°C, for no longer than 1 week. These conditions favor maintenance of intact invasion plasmids. To confirm the presence of intact invasion plasmids, plasmid isolation was performed on a sample of each experimental culture and plasmid size was compared with a known intact standard by agarose gel electrophoresis (12). Henle 407 cells (embryonic, intestinal epithelial, showing HeLa markers, human, Shiga toxin resistant; ATCC CCL 6), Hct8 cells (adenocarcinoma, ileocecal, showing no HeLa markers, human, Shiga toxin resistant; ATCC CCL 244), and a toxin-resistant HeLa cell derivative of ATCC CCL 2 were used as host cells. These lines were maintained in RPMI 1640 containing 10% fetal calf serum, gentamicin (50 jig/ml), and 0.01 M HEPES (N-2-hydroxyethylpiperazineN'-2-ethanesulfonic acid) (complete medium), in plastic tissue culture flasks (25 cm2; Corning Glass Works) in 5% CO2. For passage, the cells were washed and dissociated with 0.25% trypsin in Hanks balanced salt solution. After centrifugation, the pellet was suspended in complete medium and seeded into 24-well plates at 5 x 104 cells per well. A semiconfluent monolayer of cells was obtained in 24 h. The wells were washed three times with RPMI 1640 to remove antibiotic and fetal calf serum. Samples of overnight bacterial cultures, grown at 37°C in antibiotic-free L broth, were diluted in RPMI 1640 and added to the wells of tissue culture monolayers at an approximate multiplicity of infection of 5. This multiplicity was chosen, rather than the multiplicities of up to 2,000 used by some workers (5, 14), to minimize nonspecific binding and to increase discrimination between behaviors of different cell lines as bacterial hosts. In some experiments, the organisms were pretreated with Congo red dye (100 ,ug/ml) at 37°C for 40 min (3) prior to dilution in RPMI 1640. After addition of bacteria, the 24-well plates were either directly incubated at 37°C for 1 h in 5% CO2 or centrifuged at 2,700 x g for 5 min and then incubated as above. The wells were then washed four times with phosphate-buffered saline in order to remove
Corresponding author. 2401
2402
NOTES
INFECT. IMMUN.
TABLE 1. Adherence of S. dysenteriae to tissue culture lines in the absence of centrifugation Cell linea
Bacterial inoculum per well
HeLa Henle 407 Hct8
5 x 106 5 x 106 5 x 106
Count of adherent bacteria per well' CG097 CG097-2
3.5 x 102 3.6 x 105 1.3 x 105
1.2 x 102 4.2 x 102 2.8 x 102
a Count for each line, 106 cells per well. b Adherence time was 1 h, followed by four washings with phosphatebuffered saline to remove nonadherent bacteria and treatment of the monolayer with 0.25% trypsin to release adherent bacteria (2). Data represent means of two determinations.
nonadherent extracellular bacteria. Complete RPMI 1640 containing 50 ,ug of gentamicin per ml was added, and the plates were incubated for 2 h at 37°C in 5% CO2. Under these conditions, free CG097 and CG097-2 showed 100% killing in 60 min (data not shown). Monolayer-bacterium complexes were washed twice with antibiotic-free RPMI 1640 and treated with 0.25% trypsin (1) at 37°C for 5 min to dissociate the adherent bacteria, which were then separated from the cells by centrifugation at 1,000 x g for 3 min. Supernatants containing these bacteria were plated for viable counts. Intracellular viable organisms were recovered with 0.5% sodium deoxycholate treatment (9) after adherent bacteria were removed. The incubation period with sodium deoxycholate was always less than 30 min, since 0.5% sodium deoxycholate is bactericidal after 30 min of incubation. In preliminary studies we observed that many S. flexneri M9OT cells invaded HeLa cells, but, with the same ratio of bacteria to HeLa cells, few pathogenic S. dysenteriae cells invaded the HeLa cells. We then compared Henle 407 and Hct8 cell monolayers with HeLa cells as hosts for attachment of S. dysenteriae CG097 and its avirulent derivative CG097-2. To test for the effect of centrifugation on attachment, bacteria were added to monolayers and incubated with or without centrifugation. Table 1 presents the data obtained without centrifugation. Many CG097 cells adhered to the Henle 407 cells, while few CG097 cells adhered to the HeLa cells. Stained cover slip preparations confirmed the differences in binding in a qualitative manner (data not shown). Hct8, a human ileocecal cell line, showed adherence which was much higher than that of HeLa but comparable to that of Henle. Furthermore, the selectivity of Henle 407 and Hct8 for CG097 over CG097-2 was much greater than the selectivity shown by HeLa. Centrifugation greatly improved the binding of CG097 to HeLa cells but did not significantly change the binding of the organisms to Henle 407 cells (data not shown).
Since Congo red binding is associated with increased infectivity of S. flexneri in the HeLa cell model (3), we assessed the ability of S. dysenteriae, which had prebound Congo red, to invade Henle 407 cells. Invasive S. dysenteriae CG097 showed neither increased adherence to nor increased penetration of Henle 407 cells in the presence of Congo red (Table 2). Under identical conditions, S. dysenteriae showed significant increases in both adherence to and penetration of HeLa cells, regardless of the pathogenicity of the organisms. In fact, the avirulent isolate, S. dysenteriae CG097-2, pretreated with Congo red, showed a 28-fold increase in adherence to and a 9-fold increase in penetration of HeLa cells. The reduction in discrimination between virulent and avirulent organisms, at least as regards invasiveness of HeLa cells by S. dysenteriae, suggests that the use of Congo red is inappropriate for such studies of S. dysenteriae. These data are not necessarily in conflict with the recent suggestion of Sankaran et al. (13) that three membrane proteins, whose concentrations are significantly increased when S. dysenteriae or S. flexneri is grown in the presence of Congo red, are virulence proteins. Aside from the fact that their incubation period with Congo red was much longer than that used in our protocol, they did not compare counts of adherent or intracellular organisms with or without Congo red treatment. Such data would be required, particularly at low multiplicities of infection, to establish a link between regulation of one or more of the three membrane proteins and virulence. Although invasiveness of S. flexneri (8, 16, 17) and S. dysenteriae (2, 4, 10) has generally been examined with HeLa cells as an in vitro model, both Henle 407 and Hct8 cells proved more useful than HeLa cells as hosts for S. dysenteriae on several counts. Efficiency of attachment and of invasion, as well as selectivity for pathogenic over nonpathogenic S. dysenteriae, for Henle 407 and Hct8 cells was significantly greater than that found for HeLa cells. This increased efficiency, even in the absence of centrifugation, may be due to the presence of high-affinity membrane receptors on these cell lines. Since Henle 407 proved to be a somewhat better host cell line than Hct8, we have chosen to use Henle 407 cells as our in vitro model for study of infection by S. dysenteriae. In the course of these studies, we have observed that the large invasion plasmid of S. dysenteriae CG097 is maintained more stably than that of S. flexneri M9OT (unpublished data). This finding suggests a possible reason for the greater prevalence of S. dysenteriae in epidemics. However, the fact that even the CG097 invasion plasmid suffers spontaneous deletion of invasion genes makes it imperative to verify the size of the plasmid before each experiment, as was done in
TABLE 2. Effect of Congo red on adherence and invasion of S. dysenteriae with Henle 407 or HeLa cells as hosts Bacterial count (CFU) with indicated cell linea Strain
CG097 CG097 CG097-2
None
CG097-2
Congo red
None
Congo red
HeLa
Henle 407
Chromogen Adherent
Intracellular
Adherent
Intracellular
3.8 x 104 1.3 x 104
4.8 8.9 1.1 3.3
104
3.7 x 102 4.0 x 104 82
2.1 x 102 1.0 X 103 50 4.4 x 102
55
No growth
x
x 104 X 102 x 102
2.2 x 103
a Time of incubation at 37°C was 1 h without gentamicin and, after washing, an additional incubation for 2 h with gentamicin (50 ,ug/ml). Adherent bacteria released by trypsinization and internalized (intracellular) bacteria released by sodium deoxycholate were plated for viable colony counts. The data presented are from one of several experiments which generated similar results.
NOTES
VOL. 58, 1990
this work. We speculate that the high multiplicities of infection used by many who work with Shigella spp. may, in fact, represent empirical adjustments to compensate for loss of invasion genes by a variable fraction of the bacteria. We
are
indebted to S. Nikam for introducing
us
to the invasion
assay.
This work was supported by Public Health Service grant AI-25104 from the National Institutes of Health. LITERATURE CITED 1. Bhogale, S. R., K. D. Sharma, and R. S. Kamat. 1983. Role of heat labile antigen of Shigella flexneri in Hela cell invasion. J. Med. Microbiol. 16:37-43. 2. Binns, M. M., S. Vaughan, and K. N. Timmis. 1985. 0-antigens are essential virulence factors of Shigella sonnei and Shigella dysenteriae 1. Zentralbl. Bakteriol. Mikrobiol. Hyg. Abt. 1 Orig. A 181:197-205. 3. Daskaleros, P. A., and S. M. Payne. 1987. Congo red binding phenotype is associated with hemin binding and increased infectivity of Shigella flexneri in the HeLa cell model. Infect. Immun. 55:1393-1398. 4. Fontaine, A., J. Arondel, and P. J. Sansonetti. 1988. Role of Shiga toxin in the pathogenesis of bacillary dysentery, studied by using a Tox- mutant of Shigella dysenteriae 1. Infect. Immun. 56:3099-3109. 5. Hale, T. L., P. J. Sansonetti, P. A. Schad, S. Austin, and S. B. Formal. 1983. Characterization of virulence plasmids and plasmid-associated outer membrane proteins in Shigella flexneri, Shigella sonnei, and Escherichia coli. Infect. Immun. 40:340350. 6. Kihlstrom, E. 1977. Infection of HeLa cells with Salmonella typhimurium 395 MS and MR10 bacteria. Infect. Immun. 17: 290-295. 7. Kopecko, D. J., P. J. Sansonetti, L. S. Baron, and S. B. Formal. 1981. Invasive bacterial pathogens of the intestine; Shigella virulence plasmids and potential vaccine approaches, p. 111122. In S. B. Levy, R. C. Clowes, and E. L. Koenig (ed.), Molecular biology, pathogenicity and ecology of bacterial plasmids. Plenum Publishing Corp., New York. 8. Labrec, E. H., H. Schneider, T. J. Magnani, and S. B. Formal.
2403
1964. Epithelial cell penetration as an essential step in the pathogenesis of bacillary dysentery. J. Bacteriol. 88:1503-1518. 9. Lawlor, K. M., P. A. Daskaleros, R. E. Robinson, and S. M. Payne. 1987. Virulence of iron transport mutants of Shigella flexneri and utilization of host iron compounds. Infect. Immun. 55:594-599. 10. Niesel, D. W., C. E. Chambers, and S. L. Stockman. 1985. Quantitation of HeLa cell monolayer invasion by Shigella and Salmonella species. J. Clin. Microbiol. 22:897-902. 11. Okamura, N., T. Nagai, R. Nakaya, S. Kondo, M. Murakami, and K. Hisatsune. 1983. HeLa cell invasiveness and 0 antigen of Shigella flexneri as separate and prerequisite attributes of virulence to evoke keratoconjunctivitis in guinea pigs. Infect. Immun. 39:505-513. 12. Palchaudhuri, S., R. Kumar, D. Sen, R. Pal, S. Ghosh, B. Sarkar, S. K. Bhattacharya, and S. C. Pal. 1985. Molecular epidemiology of plasmid patterns in Shigella dysenteriae type I obtained from an outbreak in West Bengal (India). FEMS Microbiol. Lett. 30:187-191. 13. Sankaran, K., V. Ramachandran, Y. V. B. K. Subrahmanyam, S. Rajarathnam, S. Elango, and R. K. Roy. 1989. Congo red-mediated regulation of levels of Shigella flexneri 2a membrane proteins. Infect. Immun. 57:2364-2371. 14. Sansonetti, P. J., T. L. Hale, G. J. Dammin, C. Kapfer, H. H. Collins, Jr., and S. B. Formal. 1983. Alterations in the pathogenicity of Escherichia coli K-12 after transfer of plasmid and chromosomal genes from Shigella flexneri. Infect. Immun. 39:1392-1402. 15. Sansonetti, P. J., D. J. Kopecko, and S. B. Formal. 1981. Shigella sonnei plasmids: evidence that a large plasmid is necessary for virulence. Infect. Immun. 34:75-83. 16. Sansonetti, P. J., D. J. Kopecko, and S. B. Formal. 1972. Involvement of a plasmid in the invasive ability of Shigella flexneri. Infect. Immun. 35:852-860. 17. Sansonetti, P. J., A. Ryter, P. Clerc, A. T. Maurelli, and J. Mouner. 1986. Multiplication of Shigella flexneri within HeLa cells: lysis of the phagocytic vacuole and plasmid-mediated contact hemolysis. Infect. Immun. 51:461-469. 18. Shmilovitz, M., B. Kretzer, and S. Levi. 1985. A new provisional serovar of Shigella dysenteriae. J. Clin. Microbiol. 21:240-242.
INFECTION AND IMMUNITY, JUlY 1990, p. 2401-2403 0019-9567/90/072401-03$02.00/0 Copyright C 1990, American Society for Microbiology
Comparative Study of Attachment to and Invasion of Epithelial Cell Lines by Shigella dysenteriae ANUBHA SEN, MYRON A. LEON, AND SUNIL PALCHAUDHURI* Department of Immunology and Microbiology, Wayne State University School of Medicine, Detroit, Michigan 48201 Received 27 December 1989/Accepted 20 April 1990 Henle 407 and HeLa cells were compared as hosts for Shigella dysenteriae at a low multiplicity of infection. Efficiency of attachment and invasion without centrifugation, as well as selectivity for pathogenic over nonpathogenic S. dysenteriae without Congo red, were much greater for Henle 407 cells than for HeLa cells.
portion of the invasion plasmid and lacks all known ipa genes (unpublished data). After every five or six passages on artificial media, bacterial virulence was monitored by the Sereny test as well as by size determination of invasion plasmids (12). Strains which produced keratoconjunctivitis within 48 h were streaked on Congo red-TSB agar (3), to which galactose was added to a final concentration of 0.2%, or L agar plus streptomycin (100 ,ug/ml). Colonies were tested on minimal medium plates, with and without supplements, to confirm the genetic purity of the strains isolated
The ability of enteroinvasive bacteria such as Shigella spp. to invade cell culture monolayers has been used as a measure of their pathogenicity. Indeed, Shigellaflexneri and Shigella sonnei strains which lose the ability to penetrate HeLa cells are uniformly avirulent in animal hosts (7, 14-16). The invasive phenotype has been mapped to a 220-kilobase extrachromosomal element. Spontaneous deletions in the plasmid lead to noninvasive variants (16). The weak binding between Shigella spp. and HeLa cells causes inefficient adherence, and therefore invasion assays have not been highly reproducible. Daskaleros and Payne (3) have improved the efficiency of this interaction for S. flexneri by pretreating HeLa cells or bacterial culture with Congo red dye or hemin. Problems with the invasion assay also arise in assessing the specific location of the bacteria after reaction with the HeLa cells, i.e., whether they are intracellular, residing between the host cells, or attached to the cell monolayers. In order to differentiate adherent bacteria from intracellular organisms, investigators have used a variety of physical, microbiological, and chemical techniques (1, 9, 10, 17). The majority of investigators have used antibiotics to remove or kill adherent noninternalized bacteria (1, 5, 6, 9, 11, 17). Most studies of invasion mechanisms have used S. flexneri, whereas studies with Shigella dysenteriae isolates are sparse (2, 4, 5, 10). However, in recent years S. dysenteriae has been the causative organism of most of the outbreaks of shigellosis in developing countries, with a high percentage of child mortality (7, 12, 18). We have, therefore, established a sensitive in vitro model for studying invasion by S. dysenteriae. Monolayers of the cell line Henle 407 have been found to be susceptible hosts for S. dysenteriae infection in the absence of either centrifugation or addition of Congo red and thus provide a useful model for studying the interaction of S. dysenteriae with normal host cells. Fifty strains of S. dysenteriae were collected from India, Bangladesh, and other sources. Analysis demonstrated that their large invasion plasmids all had similar sizes (220 kilobases). Invasion plasmids from 10 of these 50 isolates were digested with EcoRI. Patterns obtained by agarose gel electrophoresis were identical (12). Accordingly, one isolate, CG097, supplied by J. M. Buysse from the stocks of D. J. Kopecko, was chosen as the representative virulent organism for our studies. Virulent CG097 (arg trp met Sm9) and an avirulent derivative, CG097-2 (arg trp met Smr), were grown on L agar plus streptomycin (100 ,ug/ml), as these strains are resistant to streptomycin. CG097-2 has lost a 37-kilobase *
from single colonies. Cultures were stored at room temperature, rather than at 4°C, for no longer than 1 week. These conditions favor maintenance of intact invasion plasmids. To confirm the presence of intact invasion plasmids, plasmid isolation was performed on a sample of each experimental culture and plasmid size was compared with a known intact standard by agarose gel electrophoresis (12). Henle 407 cells (embryonic, intestinal epithelial, showing HeLa markers, human, Shiga toxin resistant; ATCC CCL 6), Hct8 cells (adenocarcinoma, ileocecal, showing no HeLa markers, human, Shiga toxin resistant; ATCC CCL 244), and a toxin-resistant HeLa cell derivative of ATCC CCL 2 were used as host cells. These lines were maintained in RPMI 1640 containing 10% fetal calf serum, gentamicin (50 jig/ml), and 0.01 M HEPES (N-2-hydroxyethylpiperazineN'-2-ethanesulfonic acid) (complete medium), in plastic tissue culture flasks (25 cm2; Corning Glass Works) in 5% CO2. For passage, the cells were washed and dissociated with 0.25% trypsin in Hanks balanced salt solution. After centrifugation, the pellet was suspended in complete medium and seeded into 24-well plates at 5 x 104 cells per well. A semiconfluent monolayer of cells was obtained in 24 h. The wells were washed three times with RPMI 1640 to remove antibiotic and fetal calf serum. Samples of overnight bacterial cultures, grown at 37°C in antibiotic-free L broth, were diluted in RPMI 1640 and added to the wells of tissue culture monolayers at an approximate multiplicity of infection of 5. This multiplicity was chosen, rather than the multiplicities of up to 2,000 used by some workers (5, 14), to minimize nonspecific binding and to increase discrimination between behaviors of different cell lines as bacterial hosts. In some experiments, the organisms were pretreated with Congo red dye (100 ,ug/ml) at 37°C for 40 min (3) prior to dilution in RPMI 1640. After addition of bacteria, the 24-well plates were either directly incubated at 37°C for 1 h in 5% CO2 or centrifuged at 2,700 x g for 5 min and then incubated as above. The wells were then washed four times with phosphate-buffered saline in order to remove
Corresponding author. 2401
2402
NOTES
INFECT. IMMUN.
TABLE 1. Adherence of S. dysenteriae to tissue culture lines in the absence of centrifugation Cell linea
Bacterial inoculum per well
HeLa Henle 407 Hct8
5 x 106 5 x 106 5 x 106
Count of adherent bacteria per well' CG097 CG097-2
3.5 x 102 3.6 x 105 1.3 x 105
1.2 x 102 4.2 x 102 2.8 x 102
a Count for each line, 106 cells per well. b Adherence time was 1 h, followed by four washings with phosphatebuffered saline to remove nonadherent bacteria and treatment of the monolayer with 0.25% trypsin to release adherent bacteria (2). Data represent means of two determinations.
nonadherent extracellular bacteria. Complete RPMI 1640 containing 50 ,ug of gentamicin per ml was added, and the plates were incubated for 2 h at 37°C in 5% CO2. Under these conditions, free CG097 and CG097-2 showed 100% killing in 60 min (data not shown). Monolayer-bacterium complexes were washed twice with antibiotic-free RPMI 1640 and treated with 0.25% trypsin (1) at 37°C for 5 min to dissociate the adherent bacteria, which were then separated from the cells by centrifugation at 1,000 x g for 3 min. Supernatants containing these bacteria were plated for viable counts. Intracellular viable organisms were recovered with 0.5% sodium deoxycholate treatment (9) after adherent bacteria were removed. The incubation period with sodium deoxycholate was always less than 30 min, since 0.5% sodium deoxycholate is bactericidal after 30 min of incubation. In preliminary studies we observed that many S. flexneri M9OT cells invaded HeLa cells, but, with the same ratio of bacteria to HeLa cells, few pathogenic S. dysenteriae cells invaded the HeLa cells. We then compared Henle 407 and Hct8 cell monolayers with HeLa cells as hosts for attachment of S. dysenteriae CG097 and its avirulent derivative CG097-2. To test for the effect of centrifugation on attachment, bacteria were added to monolayers and incubated with or without centrifugation. Table 1 presents the data obtained without centrifugation. Many CG097 cells adhered to the Henle 407 cells, while few CG097 cells adhered to the HeLa cells. Stained cover slip preparations confirmed the differences in binding in a qualitative manner (data not shown). Hct8, a human ileocecal cell line, showed adherence which was much higher than that of HeLa but comparable to that of Henle. Furthermore, the selectivity of Henle 407 and Hct8 for CG097 over CG097-2 was much greater than the selectivity shown by HeLa. Centrifugation greatly improved the binding of CG097 to HeLa cells but did not significantly change the binding of the organisms to Henle 407 cells (data not shown).
Since Congo red binding is associated with increased infectivity of S. flexneri in the HeLa cell model (3), we assessed the ability of S. dysenteriae, which had prebound Congo red, to invade Henle 407 cells. Invasive S. dysenteriae CG097 showed neither increased adherence to nor increased penetration of Henle 407 cells in the presence of Congo red (Table 2). Under identical conditions, S. dysenteriae showed significant increases in both adherence to and penetration of HeLa cells, regardless of the pathogenicity of the organisms. In fact, the avirulent isolate, S. dysenteriae CG097-2, pretreated with Congo red, showed a 28-fold increase in adherence to and a 9-fold increase in penetration of HeLa cells. The reduction in discrimination between virulent and avirulent organisms, at least as regards invasiveness of HeLa cells by S. dysenteriae, suggests that the use of Congo red is inappropriate for such studies of S. dysenteriae. These data are not necessarily in conflict with the recent suggestion of Sankaran et al. (13) that three membrane proteins, whose concentrations are significantly increased when S. dysenteriae or S. flexneri is grown in the presence of Congo red, are virulence proteins. Aside from the fact that their incubation period with Congo red was much longer than that used in our protocol, they did not compare counts of adherent or intracellular organisms with or without Congo red treatment. Such data would be required, particularly at low multiplicities of infection, to establish a link between regulation of one or more of the three membrane proteins and virulence. Although invasiveness of S. flexneri (8, 16, 17) and S. dysenteriae (2, 4, 10) has generally been examined with HeLa cells as an in vitro model, both Henle 407 and Hct8 cells proved more useful than HeLa cells as hosts for S. dysenteriae on several counts. Efficiency of attachment and of invasion, as well as selectivity for pathogenic over nonpathogenic S. dysenteriae, for Henle 407 and Hct8 cells was significantly greater than that found for HeLa cells. This increased efficiency, even in the absence of centrifugation, may be due to the presence of high-affinity membrane receptors on these cell lines. Since Henle 407 proved to be a somewhat better host cell line than Hct8, we have chosen to use Henle 407 cells as our in vitro model for study of infection by S. dysenteriae. In the course of these studies, we have observed that the large invasion plasmid of S. dysenteriae CG097 is maintained more stably than that of S. flexneri M9OT (unpublished data). This finding suggests a possible reason for the greater prevalence of S. dysenteriae in epidemics. However, the fact that even the CG097 invasion plasmid suffers spontaneous deletion of invasion genes makes it imperative to verify the size of the plasmid before each experiment, as was done in
TABLE 2. Effect of Congo red on adherence and invasion of S. dysenteriae with Henle 407 or HeLa cells as hosts Bacterial count (CFU) with indicated cell linea Strain
CG097 CG097 CG097-2
None
CG097-2
Congo red
None
Congo red
HeLa
Henle 407
Chromogen Adherent
Intracellular
Adherent
Intracellular
3.8 x 104 1.3 x 104
4.8 8.9 1.1 3.3
104
3.7 x 102 4.0 x 104 82
2.1 x 102 1.0 X 103 50 4.4 x 102
55
No growth
x
x 104 X 102 x 102
2.2 x 103
a Time of incubation at 37°C was 1 h without gentamicin and, after washing, an additional incubation for 2 h with gentamicin (50 ,ug/ml). Adherent bacteria released by trypsinization and internalized (intracellular) bacteria released by sodium deoxycholate were plated for viable colony counts. The data presented are from one of several experiments which generated similar results.
NOTES
VOL. 58, 1990
this work. We speculate that the high multiplicities of infection used by many who work with Shigella spp. may, in fact, represent empirical adjustments to compensate for loss of invasion genes by a variable fraction of the bacteria. We
are
indebted to S. Nikam for introducing
us
to the invasion
assay.
This work was supported by Public Health Service grant AI-25104 from the National Institutes of Health. LITERATURE CITED 1. Bhogale, S. R., K. D. Sharma, and R. S. Kamat. 1983. Role of heat labile antigen of Shigella flexneri in Hela cell invasion. J. Med. Microbiol. 16:37-43. 2. Binns, M. M., S. Vaughan, and K. N. Timmis. 1985. 0-antigens are essential virulence factors of Shigella sonnei and Shigella dysenteriae 1. Zentralbl. Bakteriol. Mikrobiol. Hyg. Abt. 1 Orig. A 181:197-205. 3. Daskaleros, P. A., and S. M. Payne. 1987. Congo red binding phenotype is associated with hemin binding and increased infectivity of Shigella flexneri in the HeLa cell model. Infect. Immun. 55:1393-1398. 4. Fontaine, A., J. Arondel, and P. J. Sansonetti. 1988. Role of Shiga toxin in the pathogenesis of bacillary dysentery, studied by using a Tox- mutant of Shigella dysenteriae 1. Infect. Immun. 56:3099-3109. 5. Hale, T. L., P. J. Sansonetti, P. A. Schad, S. Austin, and S. B. Formal. 1983. Characterization of virulence plasmids and plasmid-associated outer membrane proteins in Shigella flexneri, Shigella sonnei, and Escherichia coli. Infect. Immun. 40:340350. 6. Kihlstrom, E. 1977. Infection of HeLa cells with Salmonella typhimurium 395 MS and MR10 bacteria. Infect. Immun. 17: 290-295. 7. Kopecko, D. J., P. J. Sansonetti, L. S. Baron, and S. B. Formal. 1981. Invasive bacterial pathogens of the intestine; Shigella virulence plasmids and potential vaccine approaches, p. 111122. In S. B. Levy, R. C. Clowes, and E. L. Koenig (ed.), Molecular biology, pathogenicity and ecology of bacterial plasmids. Plenum Publishing Corp., New York. 8. Labrec, E. H., H. Schneider, T. J. Magnani, and S. B. Formal.
2403
1964. Epithelial cell penetration as an essential step in the pathogenesis of bacillary dysentery. J. Bacteriol. 88:1503-1518. 9. Lawlor, K. M., P. A. Daskaleros, R. E. Robinson, and S. M. Payne. 1987. Virulence of iron transport mutants of Shigella flexneri and utilization of host iron compounds. Infect. Immun. 55:594-599. 10. Niesel, D. W., C. E. Chambers, and S. L. Stockman. 1985. Quantitation of HeLa cell monolayer invasion by Shigella and Salmonella species. J. Clin. Microbiol. 22:897-902. 11. Okamura, N., T. Nagai, R. Nakaya, S. Kondo, M. Murakami, and K. Hisatsune. 1983. HeLa cell invasiveness and 0 antigen of Shigella flexneri as separate and prerequisite attributes of virulence to evoke keratoconjunctivitis in guinea pigs. Infect. Immun. 39:505-513. 12. Palchaudhuri, S., R. Kumar, D. Sen, R. Pal, S. Ghosh, B. Sarkar, S. K. Bhattacharya, and S. C. Pal. 1985. Molecular epidemiology of plasmid patterns in Shigella dysenteriae type I obtained from an outbreak in West Bengal (India). FEMS Microbiol. Lett. 30:187-191. 13. Sankaran, K., V. Ramachandran, Y. V. B. K. Subrahmanyam, S. Rajarathnam, S. Elango, and R. K. Roy. 1989. Congo red-mediated regulation of levels of Shigella flexneri 2a membrane proteins. Infect. Immun. 57:2364-2371. 14. Sansonetti, P. J., T. L. Hale, G. J. Dammin, C. Kapfer, H. H. Collins, Jr., and S. B. Formal. 1983. Alterations in the pathogenicity of Escherichia coli K-12 after transfer of plasmid and chromosomal genes from Shigella flexneri. Infect. Immun. 39:1392-1402. 15. Sansonetti, P. J., D. J. Kopecko, and S. B. Formal. 1981. Shigella sonnei plasmids: evidence that a large plasmid is necessary for virulence. Infect. Immun. 34:75-83. 16. Sansonetti, P. J., D. J. Kopecko, and S. B. Formal. 1972. Involvement of a plasmid in the invasive ability of Shigella flexneri. Infect. Immun. 35:852-860. 17. Sansonetti, P. J., A. Ryter, P. Clerc, A. T. Maurelli, and J. Mouner. 1986. Multiplication of Shigella flexneri within HeLa cells: lysis of the phagocytic vacuole and plasmid-mediated contact hemolysis. Infect. Immun. 51:461-469. 18. Shmilovitz, M., B. Kretzer, and S. Levi. 1985. A new provisional serovar of Shigella dysenteriae. J. Clin. Microbiol. 21:240-242.
