- Email: [email protected]
Characteristics of capsules in enterotoxemic Escherichia coli O139:K12 strains causing swine edema disease
Microbiol. Res. (2002) 157, 191–195 (768) http://www.urbanfischer.de/journals/microbiolres
Characteristics of capsules in enterotoxemic Escherichia coli O139:K12 strains causing swine edema disease Yuko Meno1, Shuji Fujimoto2 1 2
Faculty of Health and Welfare, Seinan-Jogakuin University, 1-3-5 Ibori, Kokura Kita-ku, Kitakyushu, Fukuoka 803-0835, Japan School of Health Science, Kyushu University, 3-1-1 Maidashi, Higashi-ku, Fukuoka, Fukuoka 812–8582, Japan
Accepted: February 8, 2002
Abstract The characteristics of the capsule of the enterotoxemic Escherichia coli (ETEEC) O139:K12 strains that strongly adhere to Hep-2 cells were examined. Electron microscopic studies using the freeze-substitution technique revealed that ETEEC strains had a capsule of approximately 25 nm. These strains show hydrophobic surface properties and strong adherence to human polymorphonuclear leukocytes (PMNs). In contrast, ETEEC strains RK-O139 and ED-1 show weak adherence to HEp-2 cells and fail to express the capsule layer on the cell surface. These ETEEC strains possess hydrophilic surface properties and also adhere to PMNs. The lipopolysaccharide (LPS) analysis by means of sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) showed that ETEEC strains had the same LPS profile and long O-side chains of LPS. Furthermore, all strains were resistant to serum killing activity. These results suggest that the capsule of ETEEC strains does not contribute as an antiphagocytic factor, but as an adherence factor to host cells. Key words: Escherichia coli – capsule – hydrophobicity – surface characteristics
Introduction Enterotoxemic Escherichia coli (ETEEC) is associated with swine edema disease and causes serious problems in piglets at weaning (Imberechts et al. 1992 ; Kausche et al. 1992). ETEEC strains of O139:K12 serogroup were the pathogens most frequently isolated from swine in Japan (Nakazawa et al. 1995). Corresponding author: Y. Meno e-mail: [email protected] 0944-5013/02/157/03-191
$15.00/0
ETEEC strains produce Shiga-like toxin II (SLT-IIv) which has the ability to induce edema disease and plays a critical role in pathogenesis (Gyles et al. 1988; Imberechts et al. 1993). In addition to toxin production, bacterial colonization in the intestine is a crucial step in the pathogenesis of this disease. Imberechts et al. (1993, 1997) reported that colonization of intestinal cells by ETEEC depends on the presence of fimbriae. Nakazawa et al. (1995) have demonstrated that the capsule of the ETEEC strains (O139:K12) was one of the adherence factors to HEp-2 cells. This suggests that the capsule is possibly one of the colonization factors. Generally, a bacterial capsule contains acidic polysaccharides and provides enhanced virulence by impairing ingestion by phagocytic cells, such as polymorphonuclear cells and macrophages (Wicken and Knox 1980). In E. coli K1 and Klebsiella pneumoniae, the capsules play an important role in causing an invasive infection (Schiffer et al. 1976; Simoons-Smit et al. 1986). Recently, we found that the capsule of Vibrio cholerae O139 possesses a hydrophobic character and that the bacteria were well ingested by human PMNs (Meno et al. 1998). Waldor et al. (1994) have shown that the capsule of V. cholerae O139 is important for colonization of the intestinal-tract. Based on these results, we propose that there are two functional roles of the capsule layer of bacteria. One is an antiphagocytic role. The other role is in order to adhere to the intestine. In a previous paper we showed that ETEEC strains, which strongly adhered to HEp-2 cells, had the capsule layer on the cell surfaces but that the strains, which weakly adhered to Hep-2 cells, lacked the capsule (Meno et al. 1996). These findings support the experiMicrobiol. Res. 157 (2002) 3
191
mental evidence presented by Nakazawa et al. (1995), that bacterial adhesion to HEp-2 cells is inhibited by anti-capsular antibody, suggesting that the capsule of ETEEC is one of the adherence factors. In this study, we performed hydrophobicity tests, PMNs association tests, and serum killing tests to determine the role of the capsule of ETEEC.
Materials and methods Bacterial strains and culture conditions. ETEEC strains isolated from pigs with edema disease were used in this study. These strains were kindly provided by M. Nakazawa of the National Institute of Animal Health, Tsukuba, Japan. Strains 107/86, IW-2, ED-3, ED-43 and ED-61 belong to serotype O139:K12. Strains RK-O139 and ED-1 express O139 antigen but fail to form the capsule antigen. More precise information regarding these strains was previously described (Meno et al. 1996). The bacteria were cultured in either Luria broth or on L agar plates at 37 °C. Surface hydrophobicity. Bacterial surface hydrophobicity was tested by the same method as reported previously (Rosenberg et al. 1980; Williams et al. 1986). Five ml of bacterial suspension of 8 × 108 cfu/ ml and 1.0 ml of xylene were mixed vigorously in a glass tube with a vortex mixer for 30 sec. The hydrophobicity was expressed as the percentage reduction of the optical density of the aqueous phase after being mixed with xylene. Association to PMNs. Human polymorphonuclear leukocytes were isolated from the blood of healthy adult donors (Boyum 1976). The PMNs counts were then performed using a standard method, and the final leukocyte pellet was adjusted to a concentration of about 4 × 106 PMNs per ml in Hanks balanced salt solution. The bacterial association test using PMNs was carried out as follows: 0.5 ml of PMNs (4 × 106 PMNs/ ml), 0.1 ml of bacterial suspension (2 × 109 cfu/ml) and 0.4 ml of the absorbed normal human serum were mixed in polystyrene tubes. To remove the antibodies for E. coli in normal human serum (NHS), the NHS was previously absorbed with the bacteria used in each experiment (1 × 108 cfu/ml) for 30 min at room temperature. The mixtures were then incubated by shaking for 15 min at 37 °C in a water bath. Samples (0.1 ml) were taken from the mixture and added to 1.0 ml of cold phosphate buffered saline (PBS) containing 1.0% bovine serum albumin. After centrifugation for 10 min at 160 × g, the pelleted cells were resuspended in 0.1 ml of PBS. 1 drop (0.05 ml) of the suspension was pipetted onto a slide and stained with Giemsa solution. 192
Microbiol. Res. 157 (2002) 3
100 PMNs each on duplicate slides were examined to determine the number of associated bacteria. The results were expressed as the percentage of PMNs with associated bacteria out of the total PMNs counted. Killing by normal human serum. The sensitivity of E. coli strains to the bactericidal activity of NHS was tested as follows: 0.35 ml of bacterial suspension (1 × 108 cfu/ml) and 0.65 ml of absorbed NHS in polystyrene tubes were incubated at 37°C in a water bath. At 30 min intervals, samples (0.1 ml) were taken from the mixture, diluted with PBS, and then plated on L agar plates. The viable cells were counted after incubation at 37°C overnight. To remove the antibody for E. coli in NHS, the NHS was mixed with bacteria used in each experiment (1 × 108 cfu/ml) for 30 min at room temperature. After centrifugation at 6,000 × g for 10 min, the supernatant was used for serum killing test. Analysis of LPS. Lipopolysaccharides were extracted from whole bacterial cells by the method of Preston and Penner (Preston and Penner 1987). The extracted LPS was fractionated by SDS-PAGE (Laemmli 1970), and the gel was stained by the silver staining method of Tsai et al. (1982). Electron microscopy. The method of freeze-substitution used in this paper was described previously (Amako et al. 1988). Thin sections were cut with a diamond knife, stained with uranyl acetate and lead citrate, and then examined with a JEM 2000EX electron microscope at 100 kV.
Results Surface structure The surface morphology of the ETEEC strains was examined by electron microscopy. A thin-sectioned electron micrograph shows capsule formation on the strain 107/86 cell surface (Fig. 1). The surface of the bacterial cells was covered with a layer consisting of fine fibers and the thickness of the layer was approximately 25 nm. The layer was also observed in strains IW-2, ED-3 ED-43 and ED-61, but not observed in the strains RK-O139 and ED-1 (data not shown). These results were identical with those of a previous report (Meno et al. 1996). Surface hydrophobicity and association to PMNs The surface hydrophobicity of the ETEEC was measured and expressed as the percentage of bacteria that remained in the aqueous phase after mixing with hydrocarbon (xylene). As shown in Fig. 2A, most cells of encapsulated strains moved to the hydrocarbon phase
Fig. 1. Electron micrograph of E. coli strain 107/86. The bacterial cell (arrows) is covered with a capsule of short fine fibers. Bar: 100 nm
Fig. 3. SDS-PAGE profiles of LPS from E. coli strains. Lanes 1 to 5 are encapsulated strains, 107/86, IW-2, ED-3, ED-43 and ED-61. Lanes 6 and 7 are non-encapsulated strains RKO139 and ED-1.
tively. The surface of the non-encapsulated strains showed hydrophilic properties. The association of the ETEEC strains with human PMNs was also examined. After the ETEEC strains were incubated with PMNs for 15 min, all strains with or without the presence of the capsule on the cell surface were observed to associate with PMNs (Fig. 2B). The association rate ranged between 40% and 56%, and there was almost no difference in the association rate among the ETEEC strains. Serum killing and LPS profile
Fig. 2. Hydrophobicity of the surface in E. coli strains (A) and association of E. coli strains with PMNs (B)
and only a small portion remained in the aqueous phase, thus indicating their hydrophobic surface characteristics. The hydrophobicity rate of encapsulated strains ranged from 19% to 37%. In contrast, the hydrophobicity rate of the strains RK-O139 and ED-1 which did not express a capsule layer were 0% and 5%, respec-
All ETEEC strains under examination were resistant to serum killing (data not shown). The viable count of ETEEC strains was hardly reduced after 60 min in serum. This killing activity was lost when the serum was heated at 56°C for 30 min. Because LPS was one of important factors in activation of complement, we performed an LPS analysis of the strains. LPS extracted from E. coli strains was electrophoresed and detected by the silver-staining. LPS profiles of all strains were identical (Fig. 3). All strains had high-molecular-weight ladders representing O-side chains. These results show that complement-mediated bacterial killing is not inhibited by the presence of the capsule. Microbiol. Res. 157 (2002) 3
193
Discussion
Acknowledgements
The ETEEC strains had a capsule measuring approximately 25 nm in thickness. This was less than 1/2 to 1/6 as thick as that of invasive bacteria such as E. coli K1, K. pneumoniae (Amako et al. 1988) and V. vulnificus (Amako et al. 1984). The capsule of invasive bacteria consists of long filaments and showed high electron density. On the other hand, that of ETEEC consists of short fibers and was less dense. There seems to be a difference in the volume of capsular polysaccharides expressed among these bacteria. The volume of capsular polysaccharides in ETEEC may be less than that of E. coli K1, K. pneumoniae and V. vulnificus. We also examined the characteristics of capsules of ETEEC strains. Whereas encapsulated strains had a hydrophobic surface character, the non-encapsulated strains were hydrophilic. Most bacterial capsules contain acidic polysaccharides with various chemical compositions and thus confer a negative charge to the cell surface. The capsule of ETEEC strains contains 2-keto3 deoxy-D-manno octonic acid (Schmidt and Jann 1983) as acidic sugar, but it did not provide hydrophilic properties to the cell surface. The reason why the surface of the ETEEC strains is hydrophobic is not known. The small capsules of the ETEEC may contain only few acidic sugar molecules in their small capsules and therefore may be charged negatively only weakly. The ETEEC strains adhered well to PMNs with or without the presence of capsules. These results suggest that the capsule of the ETEEC may not have an antiphagocytic function like that of invasive bacteria. In a previous paper we reported that there were two types of capsule layers in bacteria. One is hydrophilic and plays an antiphagocytic role as the capsule in K. pneumoniae. The other type is hydrophobic without an antiphagocytic function, as in the case of V. cholerae O139 (Meno et al. 1998). Therefore, it is possible that the capsules of the ETEEC strains belong to the latter type. Fimbriae have been described as one of colonization factors in the pig’s intestine (Imberechts et al. 1997). The hydrophobic capsule of ETEEC may assist the fimbriae in attaching to host cells. Encapsulated and non-encapsulated strains had long LPS O-side chains and were resistant to complementmediated killing. In serum resistance, LPS and the length of LPS O-side chains play an important role for complement activation in the alternative pathway (Tomas et al. 1986). Although the capsule in ETEEC strains masks LPS on the cell, it seems not to prevent complement from interacting with LPS. This evidence suggests that the capsule is not a complement resistance factor.
We thank Dr. K. Amako for the valuable advice and criticism in this work.
194
Microbiol. Res. 157 (2002) 3
References Amako, K, Meno, Y., Takade, A. (1988): Fine structures of the capsules of Klebsiella pneumoniae and Escherichia coli K1. J. Bacteriol. 170, 4960–4962. Amako, K., Okada, K., Miake, S. (1984): Evidence for the presence of a capsule in Vibrio vulnificus. J. Gen. Microbiol. 130, 2741–2743. Boyum, A. (1976): Isolation of lymphocytes, granulocytes, and macrophages. Scand. J. Immunol. 5, 9–15. Gyles, C. L., De Grandis, S. A., MacKenzie, C., Brunton, J. L. (1988): Cloning and nucleotide sequence analysis of the genes determining verocytotoxin production in a porcine edema disease isolate of Escherichia coli. Microb. Pathog. 5, 419–426. Imberechts, H., Bertschinger, H. U., Nagy, B., Deprez, P., Pohl, P. (1997): Fimbrial colonization factors F18ab and F18ac of Escherichia coli isolated from pigs with postweaning diarrhea and edema disease. Adv. Exp. Med. Biol. 412, 175–183. Imberechts, H., De Greve, H., Lintermans, P. (1992): The pathogenesis of edema disease in pigs. A review. Vet. Microbiol. 31, 221–233. Imberechts, H., De Greve, H., Hernalsteens, J. P., Schlicker, C., Bouchet, H., Pohl, P., Charlier, G., Bertschinger, H. U., Wild, P., Vandekerckhove, J., Van Damme, J., Montagu, M. V. (1993): The role of adhesive F107 fimbriae and of SLT-IIv toxin in the pathogenesis of edema disease in pigs. Zentralbl. Bakteriol. 278, 445–450. Kausche, F. M., Dean, E. A., Arp, L. H., Samuel, J. E., and Moon, H. W. (1992): An experimental model for subclinical edema disease (Escherichia coli enterotoxemia) manifest as vascular necrosis in pigs. Am. J. Vet. Res. 53, 281–287. Laemmli, U. K. (1970): Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680–685. Meno, Y., Fujimoto, S., Amako, K. (1996): Fine surface structure of enterotoxemic Escherichia coli O139:K12 strains associated with swine edema disease. Arch. Microbiol. 166, 357–360. Meno, Y., Waldor, M. K., Mekalanos, J. J., Amako, K. (1998): Morphological and physical characterization of the capsular layer of Vibrio cholerae O139. Arch. Microbiol. 170, 339–344. Nakazawa, M., Kataoka, Y., Ohya, T. (1995): Adherence to HEp-2 cells of enterotoxemic Escherichia coli O-group 139 from pigs with edema disease. Jpn. J. Bacteriol. 50, 551–555. Preston, M., Penner, J. (1987): Structural and antigenic properties of lipopolysaccharides from serotype reference strains of Campylobacter jejuni. Infect. Immun. 55, 1806–1812.
Rosenberg, M., Gutnick, D., Rosenberg, E. (1980): Adherence of bacteria to hydrocarbons: a simple method for measuring cell-surface hydrophobicity. FEMS Microbiol. Lett. 9, 29–33. Schiffer, M. S., Oliveira, E., Glode, M. P., McCracken, G. M. Jr., Sarff, L. M., Robbins, J. B. (1976): A review: relation between invasiveness and the K1 capsular polysaccharide of Escherichia coli. Pediatr. Res. 10, 82–89. Schmidt, M. A., Jann, K. (1983): Structure of the 2-keto-3deoxy-D-manno-octonic-acid-containing capsular polysaccharide (K12 antigen) of the urinary-tract-infective Escherichia coli O4:K12:H-. Eur. J. Biochem. 131, 509–517. Simoons-Smit, A. M., Verweij-van Vught, A. M., Maclaren, D. M. (1986): The role of K antigens as virulence factors in Klebsiella. J. Med. Microbiol. 21, 133–137. Tomas, J. M., Benedi, V. J., Ciurana, B., Jofre, J. (1986): Role
of capsule and O antigen in resistance of Klebsiella pneumoniae to serum bactericidal activity. Infect. Immun. 54, 85–89. Tsai, C. M., Boykins, R., Gordon, C. (1982): Heterogeneity and variation among Neisseria meningitides lipopolysaccharides. J. Bacteriol. 155, 498–504. Waldor, M. K., Colwell, R., Mekalanos, J. (1994): The Vibrio cholerae O139 serogroup antigen includes an O-antigen capsule and lipopolysaccharide virulence determinants. Proc. Natl. Acad. Sci. USA. 91, 11388–11392. Wicken, A. J., Knox, K. W. (1980): Bacterial cell surface amphiphiles. Biochim. Biophys. Acta 604, 1–26. Williams, P., Lambert, P. A., Haigh, C. G., Brown, M. R. (1986): The influence of the O and K antigens of Klebsiella aerogenes on surface hydrophobicity and susceptibility to phagocytosis and antimicrobial agents. J. Med. Microbiol. 21, 125–132.
Microbiol. Res. 157 (2002) 3
195
Characteristics of capsules in enterotoxemic Escherichia coli O139:K12 strains causing swine edema disease Yuko Meno1, Shuji Fujimoto2 1 2
Faculty of Health and Welfare, Seinan-Jogakuin University, 1-3-5 Ibori, Kokura Kita-ku, Kitakyushu, Fukuoka 803-0835, Japan School of Health Science, Kyushu University, 3-1-1 Maidashi, Higashi-ku, Fukuoka, Fukuoka 812–8582, Japan
Accepted: February 8, 2002
Abstract The characteristics of the capsule of the enterotoxemic Escherichia coli (ETEEC) O139:K12 strains that strongly adhere to Hep-2 cells were examined. Electron microscopic studies using the freeze-substitution technique revealed that ETEEC strains had a capsule of approximately 25 nm. These strains show hydrophobic surface properties and strong adherence to human polymorphonuclear leukocytes (PMNs). In contrast, ETEEC strains RK-O139 and ED-1 show weak adherence to HEp-2 cells and fail to express the capsule layer on the cell surface. These ETEEC strains possess hydrophilic surface properties and also adhere to PMNs. The lipopolysaccharide (LPS) analysis by means of sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) showed that ETEEC strains had the same LPS profile and long O-side chains of LPS. Furthermore, all strains were resistant to serum killing activity. These results suggest that the capsule of ETEEC strains does not contribute as an antiphagocytic factor, but as an adherence factor to host cells. Key words: Escherichia coli – capsule – hydrophobicity – surface characteristics
Introduction Enterotoxemic Escherichia coli (ETEEC) is associated with swine edema disease and causes serious problems in piglets at weaning (Imberechts et al. 1992 ; Kausche et al. 1992). ETEEC strains of O139:K12 serogroup were the pathogens most frequently isolated from swine in Japan (Nakazawa et al. 1995). Corresponding author: Y. Meno e-mail: [email protected] 0944-5013/02/157/03-191
$15.00/0
ETEEC strains produce Shiga-like toxin II (SLT-IIv) which has the ability to induce edema disease and plays a critical role in pathogenesis (Gyles et al. 1988; Imberechts et al. 1993). In addition to toxin production, bacterial colonization in the intestine is a crucial step in the pathogenesis of this disease. Imberechts et al. (1993, 1997) reported that colonization of intestinal cells by ETEEC depends on the presence of fimbriae. Nakazawa et al. (1995) have demonstrated that the capsule of the ETEEC strains (O139:K12) was one of the adherence factors to HEp-2 cells. This suggests that the capsule is possibly one of the colonization factors. Generally, a bacterial capsule contains acidic polysaccharides and provides enhanced virulence by impairing ingestion by phagocytic cells, such as polymorphonuclear cells and macrophages (Wicken and Knox 1980). In E. coli K1 and Klebsiella pneumoniae, the capsules play an important role in causing an invasive infection (Schiffer et al. 1976; Simoons-Smit et al. 1986). Recently, we found that the capsule of Vibrio cholerae O139 possesses a hydrophobic character and that the bacteria were well ingested by human PMNs (Meno et al. 1998). Waldor et al. (1994) have shown that the capsule of V. cholerae O139 is important for colonization of the intestinal-tract. Based on these results, we propose that there are two functional roles of the capsule layer of bacteria. One is an antiphagocytic role. The other role is in order to adhere to the intestine. In a previous paper we showed that ETEEC strains, which strongly adhered to HEp-2 cells, had the capsule layer on the cell surfaces but that the strains, which weakly adhered to Hep-2 cells, lacked the capsule (Meno et al. 1996). These findings support the experiMicrobiol. Res. 157 (2002) 3
191
mental evidence presented by Nakazawa et al. (1995), that bacterial adhesion to HEp-2 cells is inhibited by anti-capsular antibody, suggesting that the capsule of ETEEC is one of the adherence factors. In this study, we performed hydrophobicity tests, PMNs association tests, and serum killing tests to determine the role of the capsule of ETEEC.
Materials and methods Bacterial strains and culture conditions. ETEEC strains isolated from pigs with edema disease were used in this study. These strains were kindly provided by M. Nakazawa of the National Institute of Animal Health, Tsukuba, Japan. Strains 107/86, IW-2, ED-3, ED-43 and ED-61 belong to serotype O139:K12. Strains RK-O139 and ED-1 express O139 antigen but fail to form the capsule antigen. More precise information regarding these strains was previously described (Meno et al. 1996). The bacteria were cultured in either Luria broth or on L agar plates at 37 °C. Surface hydrophobicity. Bacterial surface hydrophobicity was tested by the same method as reported previously (Rosenberg et al. 1980; Williams et al. 1986). Five ml of bacterial suspension of 8 × 108 cfu/ ml and 1.0 ml of xylene were mixed vigorously in a glass tube with a vortex mixer for 30 sec. The hydrophobicity was expressed as the percentage reduction of the optical density of the aqueous phase after being mixed with xylene. Association to PMNs. Human polymorphonuclear leukocytes were isolated from the blood of healthy adult donors (Boyum 1976). The PMNs counts were then performed using a standard method, and the final leukocyte pellet was adjusted to a concentration of about 4 × 106 PMNs per ml in Hanks balanced salt solution. The bacterial association test using PMNs was carried out as follows: 0.5 ml of PMNs (4 × 106 PMNs/ ml), 0.1 ml of bacterial suspension (2 × 109 cfu/ml) and 0.4 ml of the absorbed normal human serum were mixed in polystyrene tubes. To remove the antibodies for E. coli in normal human serum (NHS), the NHS was previously absorbed with the bacteria used in each experiment (1 × 108 cfu/ml) for 30 min at room temperature. The mixtures were then incubated by shaking for 15 min at 37 °C in a water bath. Samples (0.1 ml) were taken from the mixture and added to 1.0 ml of cold phosphate buffered saline (PBS) containing 1.0% bovine serum albumin. After centrifugation for 10 min at 160 × g, the pelleted cells were resuspended in 0.1 ml of PBS. 1 drop (0.05 ml) of the suspension was pipetted onto a slide and stained with Giemsa solution. 192
Microbiol. Res. 157 (2002) 3
100 PMNs each on duplicate slides were examined to determine the number of associated bacteria. The results were expressed as the percentage of PMNs with associated bacteria out of the total PMNs counted. Killing by normal human serum. The sensitivity of E. coli strains to the bactericidal activity of NHS was tested as follows: 0.35 ml of bacterial suspension (1 × 108 cfu/ml) and 0.65 ml of absorbed NHS in polystyrene tubes were incubated at 37°C in a water bath. At 30 min intervals, samples (0.1 ml) were taken from the mixture, diluted with PBS, and then plated on L agar plates. The viable cells were counted after incubation at 37°C overnight. To remove the antibody for E. coli in NHS, the NHS was mixed with bacteria used in each experiment (1 × 108 cfu/ml) for 30 min at room temperature. After centrifugation at 6,000 × g for 10 min, the supernatant was used for serum killing test. Analysis of LPS. Lipopolysaccharides were extracted from whole bacterial cells by the method of Preston and Penner (Preston and Penner 1987). The extracted LPS was fractionated by SDS-PAGE (Laemmli 1970), and the gel was stained by the silver staining method of Tsai et al. (1982). Electron microscopy. The method of freeze-substitution used in this paper was described previously (Amako et al. 1988). Thin sections were cut with a diamond knife, stained with uranyl acetate and lead citrate, and then examined with a JEM 2000EX electron microscope at 100 kV.
Results Surface structure The surface morphology of the ETEEC strains was examined by electron microscopy. A thin-sectioned electron micrograph shows capsule formation on the strain 107/86 cell surface (Fig. 1). The surface of the bacterial cells was covered with a layer consisting of fine fibers and the thickness of the layer was approximately 25 nm. The layer was also observed in strains IW-2, ED-3 ED-43 and ED-61, but not observed in the strains RK-O139 and ED-1 (data not shown). These results were identical with those of a previous report (Meno et al. 1996). Surface hydrophobicity and association to PMNs The surface hydrophobicity of the ETEEC was measured and expressed as the percentage of bacteria that remained in the aqueous phase after mixing with hydrocarbon (xylene). As shown in Fig. 2A, most cells of encapsulated strains moved to the hydrocarbon phase
Fig. 1. Electron micrograph of E. coli strain 107/86. The bacterial cell (arrows) is covered with a capsule of short fine fibers. Bar: 100 nm
Fig. 3. SDS-PAGE profiles of LPS from E. coli strains. Lanes 1 to 5 are encapsulated strains, 107/86, IW-2, ED-3, ED-43 and ED-61. Lanes 6 and 7 are non-encapsulated strains RKO139 and ED-1.
tively. The surface of the non-encapsulated strains showed hydrophilic properties. The association of the ETEEC strains with human PMNs was also examined. After the ETEEC strains were incubated with PMNs for 15 min, all strains with or without the presence of the capsule on the cell surface were observed to associate with PMNs (Fig. 2B). The association rate ranged between 40% and 56%, and there was almost no difference in the association rate among the ETEEC strains. Serum killing and LPS profile
Fig. 2. Hydrophobicity of the surface in E. coli strains (A) and association of E. coli strains with PMNs (B)
and only a small portion remained in the aqueous phase, thus indicating their hydrophobic surface characteristics. The hydrophobicity rate of encapsulated strains ranged from 19% to 37%. In contrast, the hydrophobicity rate of the strains RK-O139 and ED-1 which did not express a capsule layer were 0% and 5%, respec-
All ETEEC strains under examination were resistant to serum killing (data not shown). The viable count of ETEEC strains was hardly reduced after 60 min in serum. This killing activity was lost when the serum was heated at 56°C for 30 min. Because LPS was one of important factors in activation of complement, we performed an LPS analysis of the strains. LPS extracted from E. coli strains was electrophoresed and detected by the silver-staining. LPS profiles of all strains were identical (Fig. 3). All strains had high-molecular-weight ladders representing O-side chains. These results show that complement-mediated bacterial killing is not inhibited by the presence of the capsule. Microbiol. Res. 157 (2002) 3
193
Discussion
Acknowledgements
The ETEEC strains had a capsule measuring approximately 25 nm in thickness. This was less than 1/2 to 1/6 as thick as that of invasive bacteria such as E. coli K1, K. pneumoniae (Amako et al. 1988) and V. vulnificus (Amako et al. 1984). The capsule of invasive bacteria consists of long filaments and showed high electron density. On the other hand, that of ETEEC consists of short fibers and was less dense. There seems to be a difference in the volume of capsular polysaccharides expressed among these bacteria. The volume of capsular polysaccharides in ETEEC may be less than that of E. coli K1, K. pneumoniae and V. vulnificus. We also examined the characteristics of capsules of ETEEC strains. Whereas encapsulated strains had a hydrophobic surface character, the non-encapsulated strains were hydrophilic. Most bacterial capsules contain acidic polysaccharides with various chemical compositions and thus confer a negative charge to the cell surface. The capsule of ETEEC strains contains 2-keto3 deoxy-D-manno octonic acid (Schmidt and Jann 1983) as acidic sugar, but it did not provide hydrophilic properties to the cell surface. The reason why the surface of the ETEEC strains is hydrophobic is not known. The small capsules of the ETEEC may contain only few acidic sugar molecules in their small capsules and therefore may be charged negatively only weakly. The ETEEC strains adhered well to PMNs with or without the presence of capsules. These results suggest that the capsule of the ETEEC may not have an antiphagocytic function like that of invasive bacteria. In a previous paper we reported that there were two types of capsule layers in bacteria. One is hydrophilic and plays an antiphagocytic role as the capsule in K. pneumoniae. The other type is hydrophobic without an antiphagocytic function, as in the case of V. cholerae O139 (Meno et al. 1998). Therefore, it is possible that the capsules of the ETEEC strains belong to the latter type. Fimbriae have been described as one of colonization factors in the pig’s intestine (Imberechts et al. 1997). The hydrophobic capsule of ETEEC may assist the fimbriae in attaching to host cells. Encapsulated and non-encapsulated strains had long LPS O-side chains and were resistant to complementmediated killing. In serum resistance, LPS and the length of LPS O-side chains play an important role for complement activation in the alternative pathway (Tomas et al. 1986). Although the capsule in ETEEC strains masks LPS on the cell, it seems not to prevent complement from interacting with LPS. This evidence suggests that the capsule is not a complement resistance factor.
We thank Dr. K. Amako for the valuable advice and criticism in this work.
194
Microbiol. Res. 157 (2002) 3
References Amako, K, Meno, Y., Takade, A. (1988): Fine structures of the capsules of Klebsiella pneumoniae and Escherichia coli K1. J. Bacteriol. 170, 4960–4962. Amako, K., Okada, K., Miake, S. (1984): Evidence for the presence of a capsule in Vibrio vulnificus. J. Gen. Microbiol. 130, 2741–2743. Boyum, A. (1976): Isolation of lymphocytes, granulocytes, and macrophages. Scand. J. Immunol. 5, 9–15. Gyles, C. L., De Grandis, S. A., MacKenzie, C., Brunton, J. L. (1988): Cloning and nucleotide sequence analysis of the genes determining verocytotoxin production in a porcine edema disease isolate of Escherichia coli. Microb. Pathog. 5, 419–426. Imberechts, H., Bertschinger, H. U., Nagy, B., Deprez, P., Pohl, P. (1997): Fimbrial colonization factors F18ab and F18ac of Escherichia coli isolated from pigs with postweaning diarrhea and edema disease. Adv. Exp. Med. Biol. 412, 175–183. Imberechts, H., De Greve, H., Lintermans, P. (1992): The pathogenesis of edema disease in pigs. A review. Vet. Microbiol. 31, 221–233. Imberechts, H., De Greve, H., Hernalsteens, J. P., Schlicker, C., Bouchet, H., Pohl, P., Charlier, G., Bertschinger, H. U., Wild, P., Vandekerckhove, J., Van Damme, J., Montagu, M. V. (1993): The role of adhesive F107 fimbriae and of SLT-IIv toxin in the pathogenesis of edema disease in pigs. Zentralbl. Bakteriol. 278, 445–450. Kausche, F. M., Dean, E. A., Arp, L. H., Samuel, J. E., and Moon, H. W. (1992): An experimental model for subclinical edema disease (Escherichia coli enterotoxemia) manifest as vascular necrosis in pigs. Am. J. Vet. Res. 53, 281–287. Laemmli, U. K. (1970): Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680–685. Meno, Y., Fujimoto, S., Amako, K. (1996): Fine surface structure of enterotoxemic Escherichia coli O139:K12 strains associated with swine edema disease. Arch. Microbiol. 166, 357–360. Meno, Y., Waldor, M. K., Mekalanos, J. J., Amako, K. (1998): Morphological and physical characterization of the capsular layer of Vibrio cholerae O139. Arch. Microbiol. 170, 339–344. Nakazawa, M., Kataoka, Y., Ohya, T. (1995): Adherence to HEp-2 cells of enterotoxemic Escherichia coli O-group 139 from pigs with edema disease. Jpn. J. Bacteriol. 50, 551–555. Preston, M., Penner, J. (1987): Structural and antigenic properties of lipopolysaccharides from serotype reference strains of Campylobacter jejuni. Infect. Immun. 55, 1806–1812.
Rosenberg, M., Gutnick, D., Rosenberg, E. (1980): Adherence of bacteria to hydrocarbons: a simple method for measuring cell-surface hydrophobicity. FEMS Microbiol. Lett. 9, 29–33. Schiffer, M. S., Oliveira, E., Glode, M. P., McCracken, G. M. Jr., Sarff, L. M., Robbins, J. B. (1976): A review: relation between invasiveness and the K1 capsular polysaccharide of Escherichia coli. Pediatr. Res. 10, 82–89. Schmidt, M. A., Jann, K. (1983): Structure of the 2-keto-3deoxy-D-manno-octonic-acid-containing capsular polysaccharide (K12 antigen) of the urinary-tract-infective Escherichia coli O4:K12:H-. Eur. J. Biochem. 131, 509–517. Simoons-Smit, A. M., Verweij-van Vught, A. M., Maclaren, D. M. (1986): The role of K antigens as virulence factors in Klebsiella. J. Med. Microbiol. 21, 133–137. Tomas, J. M., Benedi, V. J., Ciurana, B., Jofre, J. (1986): Role
of capsule and O antigen in resistance of Klebsiella pneumoniae to serum bactericidal activity. Infect. Immun. 54, 85–89. Tsai, C. M., Boykins, R., Gordon, C. (1982): Heterogeneity and variation among Neisseria meningitides lipopolysaccharides. J. Bacteriol. 155, 498–504. Waldor, M. K., Colwell, R., Mekalanos, J. (1994): The Vibrio cholerae O139 serogroup antigen includes an O-antigen capsule and lipopolysaccharide virulence determinants. Proc. Natl. Acad. Sci. USA. 91, 11388–11392. Wicken, A. J., Knox, K. W. (1980): Bacterial cell surface amphiphiles. Biochim. Biophys. Acta 604, 1–26. Williams, P., Lambert, P. A., Haigh, C. G., Brown, M. R. (1986): The influence of the O and K antigens of Klebsiella aerogenes on surface hydrophobicity and susceptibility to phagocytosis and antimicrobial agents. J. Med. Microbiol. 21, 125–132.
Microbiol. Res. 157 (2002) 3
195