521

toxin gene. Although our results show a link between these

genotypic changes

and the

epidemiological spread,

we

enhanced the dissemination of the highly virulent clone through human populations. Invasive strains that have the speA gene do not uniformly express SPE A in vitro (unpublished), so invasive potential may not be related to possession of the gene. Our findings are consistent with the possibility that invasive genes are horizontally transmissible between members of the species. The speA gene is carried by a temperate bacteriophage,15 and bacteriophages are the most likely vehicle of transmission. Infection and lysogenisation of other serotypes could account for their association with invasive disease in this outbreak. We thank Mr Dwight Johnson for serotyping S pyogenes strains and Mrs Helen Pederson for preparing the paper. This study was supported by grants from the National Institutes of Health, AI16722(PC), HL36611(PS), and the Emma B. Howe Foundation (E. L. K.). cannot

conclude that those changes

REFERENCES

LA, Woodard DR, Schlievert PM, Tomory GS. Clinical and bacteriologic observations of a toxic shock-like syndrome due to Streptococcus pyogenes. N Engl J Med 1987; 317: 146-49. 2. Stevens DL, Tanner MH, Winship J, et al. Severe group A streptococcal infections associated with a toxic shock-like syndrome and scarlet fever toxin A. N Engl J Med 1989; 321: 1-7. 3. Belani K, Schlievert PM, Kaplan EL, Ferrieri P. Association of exotoxin producing group A streptococci and severe disease in children. Pediatr Infect Dis J 1991; 10: 351-54. 4. Gaworzewska ET, Colman G. Changes in the pattern of infection caused by Streptococcus pyogenes. Epidemiol Infect 1988; 100: 257-69. 5. Hribalova V. Streptococcus pyogenes and the toxic shock syndrome. Ann 1. Cone

Intern Med 1988; 108: 772. 6. Begovac J, Marton E, Lisic M, Beus I, Bozinovic D, Kuzmanovic N. Group A beta-hemolytic streptococcal toxic shock-like syndrome. Pediatr Infect Dis J 1990; 9: 369-70. 7. Martin PR, Hoiby EA. Streptococcal serogroup A epidemic in Norway 1987-1988. Scand J Infect Dis 1990; 22: 421-29. 8. Bartter T, Dascal A, Carroll K, Curley FJ. Toxic strep syndrome: a manifestation of group A streptococcal infection. Arch Intern Med 1988; 148: 1421-24. 9. Musser JM, Hauser AR, Kim MH, Schlievert PM, Nelson K, Selander RK. Streptococcus pyogenes causing toxic shock-like syndrome and other invasive diseases: clonal diversity and pyrogenic exotoxin expression. Proc Natl Acad Sci USA 1991; 88: 2668-72. 10. Hauser AR, Stevens DL, Kaplan EL, Schlievert PM. Molecular analysis of pyrogenic exotoxins from Streptococcus pyogenes isolates associated with toxic shock-like syndrome. J Clin Microbiol 1990; 29: 1562-67. 11. Schwartz B, Facklam RR, Breiman RF. Changing epidemiology of group A streptococcal infection in the USA. Lancet 1990; 336: 1167-71. 12. Lee PK, Schlievert PM. Quantification and toxicity of group A streptococcal pyrogenic exotoxins in an animal model of toxic shock syndrome-like illness. J Clin Microbiol 1989; 27: 1890-92. 13. Schlievert PM, Bettin KM, Watson DW. Production of pyrogenic exotoxin by groups of streptococci: association with group A. J Infect Dis 1979; 140: 676-81. 14. Johnson DR, Wlazlo A, Kaplan EL. Group A streptococci isolated from uncomplicated pharyngitis during the resurgence of streptococcal sequelae in the United States. Proceedings of XIth Lancefield International Symposium, 1990: Siena, Italy (in press). 15. Martin NJ, Kaplan EL, Gerber MA, et al. Comparison of epidemic and endemic group G streptococci by restriction enzyme analysis. J Clin Microbiol 1990; 28: 1881-86. 16. Johnson LP, Tomai MA, Schlievert PM. Bacteriophage involvement in group A streptococcal pyrogenic exotoxin A production. J Bacteriol 1984; 166: 623-27. 17. Marshall GS, Patel CG, Buck G. Meningitis caused by toxigenic group A beta-hemolytic streptococcus in a pediatric patient with acquired immunodeficiency syndrome. Pediatr Infect Dis J 1991; 10: 339-40. 18. Atwater RM. Studies in epidemiology of acute rheumatic disease in the United States, based on mortality statistics. Am J Hyg 1927; 7: 343-69. 19. Cleary PP, Kaplan EL, Livdahl C, Skjold S. DNA fingerprints of Streptococcus pyogenes are M type specific. J Infect Dis 1988; 158: 1317-23. 20. Stollerman GH. The relative rheumatogenicity of strains of group A streptococci. Mod Concepts Cardiovasc Dis 1975; 44: 35-40. 21. Yu CE, Ferretti JJ. Molecular epidemiologic analysis of the type A streptococcal exotoxin (erythrogenic toxin) gene (speA) in clinic Streptococcus pyogenes strains. Infect Immun 1989; 57: 3715-19.

SHORT REPORTS epithelial cell protein phosphorylation in enteropathogenic Escherichia coli Intestinal

diarrhoea

ability of enteropathogenic Escherichia coli (EPEC) to cause diarrhoea in man is associated with the formation of characteristic histopathological The

lesions in small-intestine enterocytes, with gross cytoskeletal damage and loss of brush-border microvilli. Investigation of enterocyte protein phosphorylation in response to EPEC infection showed that the major phosphorylated protein, identified by immunoprecipitation, is myosin lightchain—an important cytoskeletal protein known to affect actin organisation in non-muscle cells. High enterocyte concentrations of actin and myosin were observed at sites of bacterial infection. Our findings indicate that enterocyte cytoskeletal changes in response to EPEC may be directly triggered by bacterial adherence through signal transduction pathways that stimulate protein kinase activity.

Enteropathogenic Escherichia coli (EPEC) are an important cause of acute and persistent infantile diarrhoea world wide, and in some developing countries represent the most common bacterial enteric pathogen.1 EPEC colonise the small intestine to produce "attaching and effacing" (AE) lesions characterised by localised destruction of brushborder microvilli in infected enterocytes, gross cytoskeletal rearrangement, and actin accretion at sites of bacterial attachment.2 Studies in animals and man indicate that the ability of EPEC to cause diarrhoea correlates with the formation of AE lesions.1 The genes responsible for EPEC adherence have been identified but do not account for enterocyte cytoskeletal lesion changes, and the genetic basis of AE lesion formation remains unclear. EPEC do not produce recognised enterotoxins or show shigella-like invasiveness. Could the intracellular changes that culminate in AE lesion formation and diarrhoea result instead from interaction of EPEC with surface receptors in the brushborder membrane? Several hormonal systems (eg, vasoactive intestinal peptide and angiotensin II) regulate intestinal ion fluxes via second messengers such as calcium, cyclic adenosine monophosphate, or phosphatidylinositol metabolites;3 such signal transduction pathways generally involve activation of specific protein kinases. We investigated whether phosphorylation of specific target cell proteins, especially proteins in the epithelial cell cytoskeleton, might mediate enterocyte AE lesion formation and EPEC diarrhoea.

522

A B C D

E

F G

Fig 1-Stimulation of protein phosphorylation in Caco-2

and

small-intestine cells infected with EPEC.

Electrophoretic separation of phosphoproteins induced in Caco-2 cells by incubation with !!ve(!aneA)orki!!ed(B) EPEC strain E2348-69 (0127 H6); or after treatment with calmodulin inhibitor 48/80 (C) or okadaicacid (D) ;and In small- intestinemucosaaherinfectionwith EPEC strain E2348-69 (E) or laboratory strain HB101(pJPN11) (F). Phosphoprotein precipitated by antimyosin light-chain antibodies shown in lane G Stimulation of protein phosphorylation in response to EPEC infection was examined in small-intestine mucosal biopsies obtained, with informed consent, from healthy adults. In parallel experiments, we also used the cultured human enterocyte cell-line Caco-2, which differentiates after 15 days in culture to form polarised microvillated cell monolayers.2 Target material was incubated in phosphate-free medium that contained 50 u.Ci (1-85 MBq) carrier-free 32P for 4 h at 37°C before infection with EPEC ( 10g colony-forming units) for a further 2 h. Total cell proteins were solubilised, separated by electrophoresis in 12-5% polyacrylamide gels, and analysed by autoradiography.4

Infection of Caco-2 monolayers with EPEC strain E2348-695 (0127) increased the phosphorylation state of

several cell proteins, the most prominent of which, at 20 kDa (fig 1, lane A), was apparent as early as 30 min after infection. Stimulated phosphorylation of various proteins, including one at 20 kDa, was also observed in EPEC-infected human small-intestine mucosa (fig 1, lane E). 6 other EPEC strains that induced AE lesion formation within the same period in the brush borders of Caco-2 cells and cultured mucosas gave identical results (data not shown). The effect was not seen with heat-killed EPEC (fig 1, lane B) or sonication, or with a laboratory E coli strain HB 101 (p JPN 11) that carried the genetic determinants for adherence but not for AE lesion formation4,’ (fig 1, lane F). Lack of stimulated phosphorylation in Caco-2 cells incubated with the cahnodulin inhibitor 48/80 before EPEC infection (fig 1, lane C) indicates an involvement of Ca2+ /ca1modulindependent protein kinases. We attempted to mimic the action of EPEC with various agonists, including phorbol ester, ionomycin, epidermal growth factor, and bacterial toxins. The only compound which elicited a similar phosphorylation pattern, and then only after prolonged incubation for 8 h, was okadaic acid (fig 1, lane D), a potent inhibitor of protein phosphatases 1 and 2a and believed to be the cause of diarrhoeal shellfish poisoning.6 The result, a net increase in protein phosphorylation, is similar for EPEC and okadaic acid but the striking difference in 32P incorporation kinetics indicates that the pattern of protein phosphorylation induced by EPEC infection is unlikely to result from phosphatase inhibition alone. The major 20 kDa phosphoprotein was identified as myosin light-chain by immunoprecipitation with purified antibodies specific for the phosphorylated form of the regulatory light-chain of smooth-muscle myosin.7 A single 32P_labelled 20 kDa protein was precipitated from extracts of EPEC-infected intestinal mucosa and Caco-2 cells (fig I, lane G). The same protein was also precipitated by antiserum against whole smooth-muscle myosin (Sigma, Poole, Dorset) but in this case as a part of an actinomyosin complex (data not shown). Immunofluorescence microscopy with this antiserum showed accumulation of myosin in the AE lesion coincident with that previously found for actin2 (fig 2). Our data clearly indicate that EPEC stimulate protein kinase activity in enterocytes to which they adhere. The major phosphorylated species is myosin light-chain, an

Fig 2-Fluorescence micrographs showing corresponding accumulations of actin (A) and myosin (B) in AE lesions of Caco-2 cells infected with EPEC strain E2348-69.

523

integral cytoskeletal component, phosphorylation of which regulates changes in cell shape and actin organisation in non-muscle cells.8 We believe this to be the first report of membrane signal transduction stimulated directly by bacterial interaction with target cell surfaces, rather than mediation by toxins. However, bacterial attachment alone is insufficient to cause such effects, which implies the existence of genes for AE lesion formation distinct from those

required solely for bacterial adherence. We have previously observed raised intracellular calcium concentrations in cells infected with EPEC,9 and suggested that the characteristic cytoskeletal damage (vesiculation of microvilli and formation of AE lesions) might partly be explained by calcium activation of actin-severing proteins. From our present findings, we propose that extensive phosphorylation of myosin light-chain, activated by sustained high concentrations of calcium at sites of EPEC contact, irreversibly destroys microvillus function and stimulates actinomyosin accretion within AE lesions. Such brush-border damage, if extensive, would cause diarrhoea by reduction of the absorptive capacity of the intestinal mucosa. Nevertheless, this mechanism of action alone is unlikely to account for the diarrhoea associated with EPEC small-bowel colonisation. In animals, stimulators of protein kinase activity produce a similar degree of hypersecretion to that caused by cholera toxin.lO Could EPEC-induced phosphorylation of other cell proteins-for example those involved in ion transport across the membrane-also be important in the mechanism of secretory EPEC diarrhoea? This work was funded by the Medical Research Council and the Wellcome Trust. H. A. M.-H. was supported by the National University of Mexico. We thank Dr J. Bennet (Department of Anatomy and Cell Biology, St Mary’s Hospital Medical School, London, UK) for myosin light-chain antibodies.

REFERENCES 1. Robins-Browne RM. Traditional enteropathogenic Escherichia coli of infantile diarrhea. Rev Infect Dis 1987; 9: 28-53. 2. Knutton S, Baldwin T, Williams PH, McNeish AS. Actin accumulation at sites of bacterial adhesion to tissue culture cells: basis of a new diagnostic test for enteropathogenic and enterohemorrhagic Escherichia coli. Infect Immun 1989; 57: 1290-98. 3.Williamson JR, Monck JR. Hormone effects on cellular Ca2+ fluxes. Annu Rev Physiol 1989; 51: 107-24. 4. Baldwin TJ, Brooks SF, Knutton S, Manjarrez-Hemandez HA, Aitken A,Williams PH. Protein phosphorylation by protein kinase C in HEp-2 cells infected with enteropathogenic Escherichia coli. Infect Immun

1990; 58: 761-65. S, Lloyd DR, McNeish AS. Adhesion of enteropathogenic

5. Knutton

Escherichia coli to human intestinal enterocytes and cultured human intestinal mucosa. Infect Immun 1987; 55: 69-77. 6. Cohen P, Holmes CFB, Tsukitani Y. Okadaic acid: a new probe for the study of cellular regulation. Trends Biochem Sci 1990; 15: 98-102. 7. Bennet JP, Cross RA, Kendrick-Jones J, Weeds AG. Spatial pattern of myosin phosphorylation in contracting smooth muscle cells: evidence for contractile zones. J Cell Biol 1988; 107: 2623-29. 8. Keller TCS III, Mooseker MS. Ca++-calmodulin-dependent phosphorylation of myosin and its role in brush border contraction in vitro. J Cell Biol 1982; 95: 943-59. 9. Baldwin TJ, Ward W, Aitken A, Knutton S, Williams PH. Elevation of intracellular free calcium levels in HEp-2 cells infected with enteropathogenic Escherichia coli. Infect Immun 1991; 59: 1599-604. 10. Fondacaro JD, Henderson LS. Evidence for protein kinase C as a regulator of intestinal electrolyte transport. Am J Physiol 1985; 249: G422-26.

ADDRESSES: Laboratory of Protein Structure, National Institute for Medical Research, Mill Hill, London NW7 (H. A. Manjarrez-Hernandez, BSc, A Aitken, PhD); Department of Genetics, University of Leicester, Leicester (T. J. Baldwin, PhD, P. H. Williams, PhD); and Institute of Child Health, University of Birmingham, Birmingham B16, UK (S. Knutton, PhD). Correspondence to Dr P H. Williams, Department of Genetics, University of Leicester, Leicester LE1 7RH, UK.

Reversed cerebral asymmetry in women with breast cancer

Altered intrauterine hormonal environment might predispose to both atypical cerebral asymmetry and breast cancer. We therefore investigated computed tomographic scans of 79 right-handed, white patients with breast cancer and 97 controls to assess the pattern of cerebral asymmetry. Women with breast cancer had a reversed pattern of cerebral asymmetry significantly more often than did controls (p