Defining Mechanisms of Interaction Between Enterohemorrhagic Escherichia coli O157:H7 and Host Epithelia

Bacteria-Host Communication: The Language of Hormones. Sperandio V, Torres AG, Jarvis B, et al. Proc Natl Acad Sci USA 2003;100:8951–6. Summary: Quorum sensing is a newly recognized mechanism for regulating bacterial growth and gene expression based on the cell density-dependent release and detection of soluble autoinducers (AI). For example, AI-2 is synthesized by an enzyme encoded by the luxS gene, which is expressed by several commensal and pathogenic bacteria. Sperandio et al. previously reported that regulation of enterohemorrhagic Escherichia coli (EHEC) O157:H7 virulence genes, including flagella, Shiga-toxins, and the locus for enterocyte effacement (LEE) pathogenicity island, is dependent on luxS expression (Proc Natl Acad Sci USA 1999;96:15196–201; J Bacteriol 2001;183:5187–97; Mol Microbiol 2002;43:809–21; Infect Immun 2002;70:3085–93). These results confirmed that EHEC virulence gene expression is controlled by a soluble quorum sensing factor and suggested that AI-2 synthesized by gut commensals may act on the quorum-sensing apparatus of EHEC. Indeed, in the current study the authors found that fecal filtrates obtained from healthy human volunteers contained AI-2 activity. An EHEC luxS mutant that displayed attenuated virulence gene expression was tested for its ability to cause characteristic attaching and effacing lesion formation on epithelial cells. Surprisingly, the mutant still formed attaching-effacing lesions. This unexpected finding led the authors to consider that a eukaryotic-derived signaling compound could compensate for the luxS mutation and lack of AI-2 production by the mutant bacterium. To address this novel hypothesis, Sperandio et al. tested eukaryotic cell culture medium preconditioned by incubation with epithelial cells and found that it restored virulence to the luxS mutant. Size fractionation revealed a small compound (<1 kilodalton) present in fetal bovine serum that is normally added to culture media was responsible for the observed effects. This led to the suspicion that a hormone may be involved. With the use of purified catecholamines, epinephrine and norepinephrine, the signaling effects of the luxS mutant were restored. The response to epinephrine was specific because other intestinal hormones, such as gastrin, had no effect. The effect also was independent of catecholamines released by epithelial cells or the EHEC O157:H7-derived catechol, enterobactin. The nonselective β-adrenergic receptor antagonist propranolol and the nonselective α-adrenergic receptor antagonist phentolamine both blocked the stimulatory effects of epinephrine on attaching-effacing lesion formation by the EHEC luxS mutant. Biochemically purified AI-2 from EHEC O157:H7 culture supernatants and AI-2 synthesized in vitro from recombinant LuxS protein both failed to activate LEE reporter gene transcription in vitro. These negative results clearly demonstrated that regulation of these virulence genes in EHEC O157:H7 is not under the control of AI-2. However, a separate fraction of EHEC supernatants lacking AI-2 activity did activate the LEE reporter gene and restored virulence gene expression (e.g., flagella). This fraction also displayed an electrospray-mass spectrometry profile different from that of AI-2 and was named AI-3. Comment: Infection of humans with EHEC O157:H7 causes hemorrhagic colitis and the hemolytic uremic syndrome, the latter representing the most common cause of potentially preventable acute renal failure in children (Curr Opin Infect Dis 2003;16:259–63). Current evidence indicates that manipulation of host cell signal transduction cascades by EHEC allows the organism to establish infection (Can J Gastroenterol 2002;16:771–8). Because antibiotic therapy may prove harmful in EHEC-infected individuals (N Engl J Med 2000;342:1930–6), there is an urgent need to delineate novel therapeutic targets. To this end, research delineating the role of EHEC O157:H7 quorum sensing is significant because quorum sensing is increasingly recognized as an important virulence determinant of pathogenic bacteria (Curr Opin Microbiol 2003;6:191–7). The novel findings by Sperandio et al. reinforce the notion that secretion of AI molecules can activate quorum sensing between different bacterial species (Curr Opin Microbiol 2003;6:191–7). Furthermore, this study provides mechanistic insight regarding how a small number of EHEC organisms might cause disease. For example, activation of EHEC quorum sensing by AI molecules derived from bacterial commensals could enhance the expression of EHEC O157:H7 virulence genes needed to propel the bacterium through the mucus layer onto the apical epithelial cell surface, and to promote intimate bacterial adherence to this plasma membrane, thereby modulating host signal transduction responses. In 1999, the luxS gene responsible for AI-2 synthesis was characterized in E. coli, Salmonella typhimurium, and Vibrio harveyi (Proc Natl Acad Sci USA 1999;96:1639–44). Soon thereafter the structure characterization of AI-2 was defined (Mol Microbiol 2001;41:463–76; Nature 2002;415:545–9) and luxS-dependent observations in other pathogenic bacteria, such as toxin production by Clostridium perfringens (Mol Microbiol 2002;44:171–9), were established. However, the current study provides compelling data indicating that luxS-dependent production of AI-3, rather than AI-2, is actually responsible for the induction of EHEC virulence gene expression by quorum sensing. Thus, the role of AI-3 in other luxS-dependent systems needs to be clarified. In addition, because it is not activated by AI-2, LEE reporter gene transcription can now be used as a model bioassay system to study AI-3 production by other bacteria. A novel mechanism for cross-talk between eukaryotic and prokaryotic organisms was shown in this study. Indeed, such communication provides a potential mechanism for previous observations that epinephrine increases EHEC O157:H7 Shiga-like toxin production (J Lab Clin Med 1996;128:392–8). In addition, Sperandio et al. (J Bacteriol 2001;183:5187–97) noted previously that an EHEC luxS mutant displayed increased expression of genes important for cell division, whereas the addition of bacterial culture supernatants containing AI-2 (and now we realize also AI-3) slowed growth of the mutant. These data suggested that exposure to AI-2 elicits metabolic processes to slow the growth of EHEC. However, Lyte et al. (Adv Exp Med Biol 1997;412:331–9) found that norepinephrine had the opposite effect and increased the growth of EHEC. Thus, the possibility remains that AI-2 and AI-3 each regulate distinct sets of genes. Furthermore, it is now evident that epinephrine and norepinephrine can act on prokaryotic cells; however, it remains to be determined if AI-3 exerts any functional effects on eukaryotic cell signaling. In addition, delineation of the exact biochemical pathways involved in the synthesis of AI-3 can now be dissected. The identification of the prokaryotic receptor to which AI-3 binds also is of interest. Can the function of AI-3 be manipulated to ameliorate outcomes of EHEC infection in addition to preventing attaching-effacing lesion formation? Using both the EHEC luxS mutant and a pharmacological approach (e.g., propranolol), it now will be of interest to determine if other consequences of E. coli O157:H7 infection, including disruption of epithelial barrier function (Infect Immun 1998;66:1680–7) and inhibition of cytokine signaling (Infect Immun 2003;71:1396–404) are preventable. Indeed, a recent publication suggests that such an approach may well prove effective. Addition of epinephrine or dopamine to murine cecal mucosal tissue mounted into Ussing chambers increased adherence of EHEC, which can be antagonized by either propranolol or phentolamine (Shock 2003;20:183–8). Whether these drugs acted on the bacteria or cecal tissues to elicit the observed effects remains to be defined. An important step forward will be to determine the biologic role of AI-3 in vivo using animal models of EHEC infection. For example, Citrobacter rodentium infection of mice is a model of EHEC O157:H7 infection (Int J Med Microbiol 2003;293:87–93). Thus, determining the virulence of a luxS mutant of C. rodentium in the mouse could prove valuable. In addition, the administration of AI-3 antagonists, such as propranolol and phentolamine, during murine infection will provide information relevant to designing future therapeutic interventions in humans. In summary, the work by Sperandio et al. provides an important advance in the understanding of quorum-sensing mechanisms in general. In addition, the findings of the study establish a role for quorum sensing as an EHEC O157:H7 virulence strategy. The newly identified AI-3 signaling pathway can now be studied in EHEC of multiple serotypes, fecal commensals, and in other pathogenic bacteria. Moreover, the AI-3 signaling pathway represents a site for developing therapeutic interventions, thereby bringing us closer to the primary long-term objective for studying the cellular microbiology of EHEC infection: to find an effective treatment for individuals with infection. Peter J. M. Ceponis Philip M. Sherman The Hospital for Sick Children University of Toronto Toronto, Canada

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.