- Email: [email protected]
Enterotoxins in Aeromonas-associated gastroenteritis
Microbes and Infection, 1, 1999, 1129−1137 © 1999 Éditions scientifiques et médicales Elsevier SAS. All rights reserved
Review
Enterotoxins in Aeromonas-associated gastroenteritis Ashok K. Chopra*, Clifford W. Houston Department of Microbiology and Immunology, The University of Texas Medical Branch, Galveston, TX 77555–1070, USA
ABSTRACT – Aeromonas species produce an array of virulence factors, and the pathogenesis of Aeromonas infections is therefore complex and multifactorial. Aeromonas-associated gastroenteritis is especially a problem in young children. The potential involvement of enterotoxins in the pathogenesis of Aeromonas infections is discussed. © 1999 Éditions scientifiques et médicales Elsevier SAS Aeromonas / cytotoxic enterotoxin / aerolysin / cytotonic enterotoxin / pathogenic mechanisms
1. Introduction Aeromonas species (spp.), which recently have been placed in a new family – Aeromonadaceae, are predominately pathogenic to poikilotherms [1]. However, Aeromonas has emerged as an important human pathogen because of suspected food-borne outbreaks [2, 3] and the increased incidence of Aeromonas isolation from patients with traveler’s diarrhea [4, 5]. Misidentification of Aeromonas is a continuing problem, and in 1998, the Department of Health Services, Berkeley, CA, reported two unusual cases of Aeromonas infections one associated with bacteremia (A. schubertii), and another in which the organism was recovered from an infected gall bladder (A. veronii bv. veronii) [6]. These strains initially were identified as Vibrio damsela and V. cholerae, respectively. Aeromonas is found in both fresh and salt water and in virtually all foods [7], and causes a wide variety of human infections, including septicemia, wound infections, meningitis, pneumonia and gastroenteritis [1, 8]. In addition, Aeromonas-associated hemolytic uremic syndrome, burnassociated sepsis, and a variety of respiratory infections, including epiglottis, have been reported [8]. The ability of these microorganisms to grow well at refrigeration temperatures and produce exotoxins at these low temperatures [9] could be important in their role as food poisoning agents. Kirov et al. [10] reported that 10% of the Aeromonas strains isolated from food could adhere to HEp-2 cells at 37 and 5 oC, and the adherent strains of Aeromonas were virulent and produce exotoxins [10, 11]. It also has been shown that a species of Aeromonas adheres to human blood cells [12]. Among the 14 species of Aeromonas isolated to date [1], A. hydrophila, A. caviae, and A. veronii bv. * Correspondence and reprints Microbes and Infection 1999, 1129-1137
sobria, have most commonly been isolated from human infections and been shown to produce a variety of biologically active extracellular products. These include hemolysins, cytotoxins, enterotoxins, proteases, leukocidin, phospholipases, endotoxin, outer membrane proteins, and fimbriae or adhesions. Some Aeromonas spp. also possess S-layers and form capsules [13]. Taken together, the ability of these bacteria to invade host cells and disseminate to virtually any organ via blood and their capacity to produce various virulence factors contribute to the pathogenesis of disease mediated by Aeromonas. An epidemiological link has been suggested between the consumption of contaminated foods and human illness, with aeromonas as a predominant flora in fecal and biopsy specimens [2, 3, 8]. In addition, serological evidence of infection was demonstrated by the presence of agglutinating antibodies to the somatic antigen and neutralizing antibodies to the exotoxins [8]. Recent reports of A. hydrophila and A. sobria (now referred to as A. veronii bv. sobria) isolation from humans with sepsis, peritonitis, urinary tract infections, severe muscle degeneration, and bacteremia with myonecrosis and gas-gangrene in a hemodialysis patient demonstrated the importance of this organism in various diseases [14–16]. In a more recent study in Japan, Funada and Matsuda [17] reported Aeromonas bacteremia in patients with hematologic diseases, with no exposure to water or fish. A clinical picture ranging from mild gastroenteritis to severe gastroenteritis was observed in 71% of the patients, with an overall mortality rate of 35%. In another report, the association of A. hydrophila gastroenteritis with hypercalcemia was cited [18]. Deutsch and Wedzina [19] reported a case of left-sided segmental colitis due to A. sobria (A. veronii bv. sobria), and supported the view that Aeromonas spp. needed to be considered in the differential diagnosis of colitis. In Malaysia, bacteremia due to A. hydrophila in a patient recovering from cholera was reported [20]. Like1129
Review
wise, a case of severe acute diarrhea produced by A. sobria (A. veronii bv. sobria) in a patient colectomized for Crohn’s disease was published [21]. Martino et al. [22] reported that patients with acute nonlymphoblastic leukemia developed septic shock due to Aeromonas infection. The latter confirms the potentially aggressive nature of these bacteria in neutropenic cancer patients, as previously reported [23]. Overall, the literature suggests that Aeromonas, once mainly considered an opportunistic pathogen in immunocompromised humans, is now implicated as the etiologic agent in numerous clinical situations involving immunocompetent individuals of all age groups.
2. Epidemiology of Aeromonasassociated gastroenteritis One of the most commonly reported infections by aeromonas is gastroenteritis that ranges from a mild, selflimiting watery diarrhea to a more severe, invasive Shigella-like dysenteric form [24]. Further, chronic diarrhea exceeding one year’s duration due to A. caviae or A. hydrophila also has been reported [8]. In a survey examining the causes of traveler’s diarrhea treated in Tokyo between 1985 and 1995, Aeromonas spp. (A. veronii bv. sobria and A. hydrophila) were isolated from 5.5% of the travelers returning from developing countries [4]. Aeromonas spp. were isolated from 8.7% of Finnish tourists with diarrhea who had traveled to Morocco and from 1.4% of nondiarrheal tourists. Species of Aeromonas were found as the sole pathogen in 5.5% of those patients [5]. In a 27-month prospective study conducted in Chicago, we found Aeromonas species were the only potential bacterial enteropathogens isolated from children suffering from diarrhea and ranging from 1 to 27 months of age [25]. Gastroenteritis due to aeromonas is especially a problem for the pediatric population. These microorganisms have been detected on a worldwide basis as the only pathogen from 2–20% of the children suffering from diarrhea and from only 0–2% of children without diarrhea [26, 27]. In a study by Goodwin et al. [28], A. hydrophila was isolated from the feces of 32 adult patients during a nine-month period in three hospitals in Western Australia. All 32 isolates produced enterotoxin detected by the suckling mouse test, and all patients except one had diarrhea. In another study, 224 cases of Aeromonas gastroenteritis were reported in Iowa, from January to June 1991, thus making this organism the most prevalent enteric pathogen isolated from stools of patients suffering from gastroenteritis [29]. A. hydrophila has been reported to be the most common enteric pathogen in adults with diarrhea in Bangkok, Thailand [30]. Serum samples from adult individuals usually contain antibodies to aeromonas, and Khaitovich [31] reported that 47% of the patients with diarrhea due to other enteropathogens had Aeromonas-specific antibodies. Treatment with prophylactic doxycycline significantly prevented the colonization of Aeromonas sp. in the gastrointestinal tract [32]. The high environmental prevalence of these pathogenic bacteria should be regarded as an important threat 1130
Chopra and Houston
to public health, since Aeromonas infections generally are acquired through consumption of water and food. Borrell et al. [33] reported that most pathogenic Aeromonas spp. were prevalent in environmental samples, with A. veronii bv. sobria being the most common in lakes and reservoirs (42%) and treated drinking water (25%), and A. caviae was the most common in sea water (26%) and milk products (36%). A. hydrophila (18%) was the second most prevalent species isolated in untreated water [33]. Interestingly, Albert et al. [34] reported that some strains of Aeromonas shared common antigens with that of V. cholerae O139. They reported that a polyclonal antiserum to a cross-reacting A. trota strain cross-protected infant mice against cholera on challenge with virulent V. cholerae O139. Based on hybridization groups (HGs), most human isolates have been allocated to DNA HG 4 (A. caviae) and HG 1 (A. hydrophila) [35]. Further, human isolates have been placed in BD-1 and BD-2 types, according to PhenePlate typing, and the HG1/BD-2 type may represent a pathogenic A. hydrophila type that is able to produce diarrhea in humans. Studies of Haque et al. [36] reported the presence of Shiga-like toxin 1 on a plasmid in some strains of A. hydrophila and A. sobria (A. veronii bv. sobria) which could account for Aeromonas-associated gastroenteritis. However, in general the role of plasmids in Aeromonas virulence is questionable [37].
3. Enterotoxins of aeromonas Two categories of enterotoxins – cytotonic and cytotoxic – have been discovered in culture filtrates of Aeromonas isolates. The cytotonic enterotoxin, like cholera toxin (CT), does not cause degeneration of crypts and villi of the small intestine, whereas the cytotoxic enterotoxin results in extensive damage to epithelium. The cytotoxic enterotoxin of aeromonas also has been referred to by other investigators as β-hemolysin, and/or aerolysin [3841]. These toxin molecules have hemolytic and cytotoxic activities in addition to an enterotoxic activity [42]. However, toxin molecules with hemolytic or cytotoxic activity alone also have been isolated from Aeromonas sp. [43].
4. Animal models for detecting enterotoxins and enterotoxigenicity of aeromonas Ligated ileal loops in adult mice, rats, and rabbits and the suckling mouse assay have been successfully used by several investigators to examine enterotoxic activity in culture filtrates of Aeromonas isolates and of purified enterotoxins [44], and likewise, the enterotoxigenicity of Aeromonas spp. However, these methods do not use the natural oral route of infection in humans and are not suitable for quantifying virulence of the organism. More recently, virulence of the organism has been correlated successfully by oral administration of organisms in suckling mice and by determining the LD50 [44]. Sanderson et al. [45] noted that only the streptomycin-treated adult Microbes and Infection 1999, 1129-1137
Enterotoxins in Aeromonas-associated gastroenteritis
mice could colonize Aeromonas species in the intestine when it was given by an intragastric route. However, diarrheal symptoms were not produced in this model. Likewise, protein-malnourished mice could colonize aeromonas in the intestine when it was given in drinking water over four days. Although the animals did not develop diarrhea, 75% of them were shedding aeromonas two days after the challenge period [45]. Therefore, an animal model is needed to demonstrate Aeromonasinduced gastroenteritis using a natural route of infection. Recently, Graf [46] suggested a possible use of medicinal leech (Hirudo medicinalis) as a model for digestive tract association of Aeromonas spp. A. veronii bv. sobria was exclusively found in the digestive tract of the leech. Interestingly, human fecal isolates of A. hydrophila and A. veronii bv. sobria colonized the digestive tract to the same extent as the symbiotic isolate. Further, human isolates proliferated to the same extent in the crop fluid as the symbiotic isolate six hours after blood feeding, indicating a potential for this digestive tract model in studying bacterial-host interaction.
5. Cytotonic enterotoxins Since the first report of a cytotonic enterotoxin of aeromonas, several investigators have identified an enterotoxic factor in culture filtrates of Aeromonas species that could be responsible for fluid secretion in the small intestine of animals. However, not much effort has been made to purify and characterize the class of this enterotoxin. A. hydrophila was first shown to be enteropathogenic by Annapurna and Sanyal [47], who injected live cells into ligated intestinal loops of rabbits. It later was shown by Wadstrom’s group [48] that a 15-kDa extracellular cytotonic enterotoxin was produced by A. hydrophila. This enterotoxin induced fluid accumulation in rabbit, rat, and mouse intestinal loops [49]. Ljungh et al. [50] demonstrated that A. hydrophila cytotonic enterotoxin, like CT, gave a positive rabbit skin test. These workers [50], as well as Dubey et al. [51], observed increases in the levels of 3’,5’-adenosine monophosphate (cAMP) in rabbit intestinal epithelial cells treated with enterotoxin. Neutralization experiments with rabbit intestinal loops, rabbit skin, and Y1 adrenal cells failed to demonstrate any immunological relationship between Aeromonas cytotonic enterotoxin and CT or Escherichia coli heat-labile (LT-I) enterotoxin [50, 52]. James et al. [53], on the other hand, reported that A. hydrophila produced two types of enterotoxins. One was heat-labile (56 oC for 10 min), caused fluid secretion in suckling mice, and did not cross-react with CT. The second was heat-stable (100 oC for 30 min), caused fluid accumulation in a rat perfusion system, and reacted with cholera antitoxin. James et al. [53] further demonstrated that preincubation with anticholera toxin significantly reduced intestinal secretions induced by culture filtrates of A. hydrophila in jejunal perfusion experiments in rats. Likewise, Bundell et al. [54] reported that immunization of rats with CT not only gave protection against pure CT during intestinal perfusion, but also significantly protected Microbes and Infection 1999, 1129-1137
Review
against the secretory effects of E. coli LT-I and a CT-like toxin produced by A. sobria (A. veronii bv. sobria). Chakraborty et al. [43] reported cloning of a cytotonic enterotoxin gene in 1984 from a diarrheal isolate of A. veronii bv. sobria (now referred to as A. trota). The cell lysate from this clone caused fluid secretion in rabbit ligated ileal loops and in suckling mice. No DNA homology was detected between their cytotonic enterotoxin gene, CT, and LT-I genes, based on Southern blot analysis. This enterotoxin gene [43], which has not been characterized further, appeared to be different from that isolated by Potomski et al. [55] from A. sobria (A. veronii bv. sobria. The latter enterotoxin cross-reacted with CT, and the biological activity associated with this enterotoxin was neutralized by antibodies to CT. Further, the purified enterotoxin from Potomski’s group exhibited bands of 43.5, 29.5, and 26 kDa [55] on sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE), unlike the 15-kDa cytotonic enterotoxin isolated by Wadstrom’s and Sanyal’s groups [50, 51]. Likewise, Shultz and McCardell [56] demonstrated three protein bands of 89, 37, and 11 kDa in cell lysates of an A. hydrophila isolate that reacted with cholera antitoxin on an immunoblot. Despite differing molecular sizes and ability to react with antibodies to CT, all of these toxins had a mechanism of action similar to that of CT; i.e., they elevated cAMP levels in eukaryotic cells [57]. Our laboratory provided evidence for the presence of a cytotoxic enterotoxin and a cytotonic enterotoxin in culture filtrates of Aeromonas isolates [58]. Both of these toxins were biologically heat labile at 56 oC/20 min. Subsequently, the heat-labile cytotonic enterotoxin (referred to as Alt) was purified from a diarrheal isolate, SSU, of A. hydrophila. The purified native Alt exhibited a molecular mass of 44 kDa and was biologically active in in vivo and in vitro models. Alt was not related to CT, based on Western blot analysis, Southern blot hybridization using CT-specific probes, and its inability to bind to GM1ganglioside, a receptor for CT and LT-I [59]. In Chinese hamster ovary (CHO) cells, Alt elevated cAMP and prostaglandin (e.g., PGE2) levels [59, 60]. The gene encoding Alt was cloned [61] and expressed using various E. coli-based expression vectors and subsequently purified [57]. The recombinant-Alt (r-Alt) in E. coli was slightly smaller (35–38 kDa), compared to native Alt (44 kDa) purified from A. hydrophila [59]. Alt exhibited 45–51% identity in a 175-aa overlap with the carboxyl half of lipase and phospholipase C (PLC) of A. hydrophila [57]. An aa sequence VHFLGHSLGA (aa residues 218–227) of Alt was highly homologous to the putative substrate-binding domain found within bacterial, fungal, porcine, and human lipases. Interestingly, the CHO cell elongation and the fluid secretory ability of r-Alt in rat ligated ileal loop was much lower (ten- to 15-fold) compared to the native Alt purified from Aeromonas sp. [57], indicating differential processing of the toxin in E. coli and aeromonas which affected the toxin’s enterotoxicity. Cyclic AMP levels of the intestinal mucosa in r-Alt-treated rat loops were 3–4 fold higher, and the PGE2 levels in the loop fluid were elevated by at least five-fold compared to phosphate-buffered saline (PBS)-treated control loops. Re1131
Review
cent expression of the alt gene back in aeromonas using a tac promoter-based, multi-host-range expression vector pMMB66 indicated the r-Alt was 44 kDa when produced from plasmid pMMB66 in aeromonas and secreted into the medium, like native Alt. Further, the biological activity of recombinant and native Alt produced from aeromonas was identical (unpublished data). Active immunization of mice with purified r-Alt resulted in a significant decrease (39%) in the fluid secretory response when challenged with wild-type A. hydrophila [57]. We recently have developed an alt gene minus mutant of A. hydrophila SSU by double crossover homologous recombination. The effect of this mutation on the enterotoxic activity is under investigation. Studies of Pal et al. [61] identified PLC as the enterotoxic factor of the bifunctional hemolysin-PLC molecule of V. cholerae O139. In a recent study by Granum et al. [62], 75% of the Aeromonas isolates from food and water in Norway contained the alt gene, based on PCR amplification, and one of the A. hydrophila strains was probably involved in an outbreak of food poisoning caused by ingestion of raw, fermented fish. We identified another cytotonic enterotoxin gene (designated as ast) in the genomic library of A. hydrophila SSU. The ast gene encoded a product (Ast) which caused CHO cells to elongate, an enterotoxic activity stable at 56 oC for 20 min [63]. The crude Ast preparations evoked fluid secretion in the rat small intestine, and the cAMP levels in the mucosal cells were elevated, compared to appropriate control (unpublished data). Expression of the DNA fragment containing the ast gene in E. coli exhibited prominent bands of 32 and 67 kDa. The relationship of these two polypeptides to the structure of Ast is under investigation. McCardell et al. [64] purified from A. hydrophila a heatstable enterotoxin (56 oC/20 min) which exhibited a size of 70 kDa. This toxin was non-CT cross-reactive, did not increase cAMP, cGMP, and PGE2 levels on CHO cells, but evoked intestinal fluid accumulation in infant mice. Whether this enterotoxin has any similarity to Ast needs to be determined.
6. Cytotoxic enterotoxin Asao’s group [39] purified a hemolysin molecule from A. hydrophila with various biological activities, which included hemolytic, cytotoxic and enterotoxic activities, as well as lethality in mice. However, this toxin was not characterized further at the molecular level. Subsequently, we reported purification from A. hydrophila SSU of a cytolytic enterotoxin (52 kDa; now designated as Act) with activities similar to that associated with Asao’s hemolysin [65, 66]. The 50% lethal dose of Act for mice was 27.5 ng, and interestingly, Act reacted with antibodies to CT in the enzyme-linked immunosorbant assay and Western blots. However, these antibodies failed to neutralize the biological activities of Act [65]. Subsequent cloning and DNA sequence analysis of the cytolytic enterotoxin gene (act) revealed no homology with the CT encoding gene (ctx) [42], suggesting a conformational epitope on Act that reacted with antibodies to CT at a site(s) distinct from site(s) responsible for biological activities. Potomski 1132
Chopra and Houston
et al. [67] similarly isolated from A. sobria (A. veronii bv. sobria) a cytotoxic enterotoxin of 63 kDa and which possessed biological activities similar to those of Act, but did not cross-react with CT. No further molecular characterization of this 63-kDa protein has been reported. Two different groups reported cloning of an aerolysin gene from A. trota and A. bestiarum (originally classified as A. hydrophila) [40, 41] while our studies were in progress. The structural gene (aerA) coding for an aerolysin was shown to be related to Asao’s hemolysin based on Western blot analysis [40]. At the DNA level, approximately 75% homology was noted in these two aerolysin structural genes; however, the 5’ flanking DNA sequences to the aerolysin genes were quite divergent [42]. Although the first 25 NH2-terminal aa residues of Act purified from A. hydrophila SSU were identical to the aerolysin reported by Buckley’s group from A. bestiarum [38], Act and their aerolysin differed in aa composition [38]. Molecular characterization of the act gene revealed 79–93% homology at the aa level between Act and the two aerolysins [42]. In a 14-aa stretch (from position 449–462), Act exhibited only 21–36% homology with the two aerolysins. The 5’ flanking sequences to the act and aerolysin genes exhibited 48–71% homology [42]. Differential neutralization of Act with Act- and aerolysin-specific monoclonal antibodies ([68] and unpublished data) and the role of selected aa residues, as determined by the site-directed mutagenesis, in aerolysin and Act-associated hemolytic activity was very intriguing [69]. Some aa residues, such as histidine (His) 107 and 132 were crucial for aerolysin’s hemolytic activity, whereas the analogous His residues 130 and 155 did not affect the hemolytic activity of Act. Likewise, tryptophan (Trp) 371 had little or no effect on the hemolytic titers of aerolysin; however, an analogous Trp 394 abrogated the biological activity of Act [69]. Detailed structure-function studies on Act and aerolysin indicated that Act/aerolysin were closely related. However, some heterogeneity at the aa level could lead to a possible differential folding of these molecules, resulting in differential neutralization of these toxins by specific monoclonal antibodies and the observed hemolytic activity associated with various mutated toxins [70]. It also was evident from the literature that Act/aerolysin and α-toxin of Staphylococcus aureus were activated by binding to the target cell membrane, with subsequent oligomerization and pore formation [66, 71–73]. Most studies on aerolysin are targeted toward measuring hemolytic activity. However, site-directed mutagenesis data on Act indicated possibly different loci on a single chain of Act which might be associated with various biological activities [69]. Blomquist and Sjorgen [74] similarly reported that one monoclonal antibody did not neutralize the hemolytic or dermonecrotic effects of α-toxin from S. aureus, but did abrogate lethality. Therefore, different biological activities of the α-toxin molecule may be evoked by different molecular regions which also may be the case with our Act. Interestingly, a peptide (aa residues 245–274) has been identified in Act which competed with native Act to bind to CHO cells. The polyclonal anti-peptide antibodies to this peptide as well as to peptide (aa residues 361–405) Microbes and Infection 1999, 1129-1137
Enterotoxins in Aeromonas-associated gastroenteritis
significantly neutralized the biological activity of Act [69]. Our studies in which we demonstrated that the lack of the act gene in Aeromonas isolate A52 would dramatically increase the LD50 for mice indicated the importance of this enterotoxin in the pathogenesis of Aeromonas-mediated diseases. These findings were substantiated further by the generation of transposon mutants of A. hydrophila SSU with significantly reduced biological activity of Act compared to that seen with wild-type Aeromonas spp. These mutants were nonlethal to mice when injected intraperitoneally at a concentration of 5 × 107 compared to wildtype A. hydrophila, which was lethal at this dose [75]. Likewise, the LD50 of isogenic mutants of A. hydrophila in which the act gene was inactivated by double crossover homologous recombination was 1 × 108 compared to wild-type Aeromonas sp., which was 3 × 105. Reintegration of the native act gene in place of the truncated toxin gene in isogenic mutants resulted in complete restoration of Act’s biological activity and virulence in mice. The animals injected with a sublethal dose of wild-type aeromonas or its revertant, but not the isogenic mutant, had circulating toxin-specific neutralizing antibodies [75]. Our studies substantiated the earlier findings of Chakraborty et al. [76] in which they showed that aerolysin-deficient mutants of A. trota were less virulent in mice than wild-type Aeromonas sp. were. Since the aerolysin gene from A. trota differed significantly from the act gene from A. hydrophila [42], it was important to delete the act gene from an authentic strain of A. hydrophila, particularly as Kuhn et al. [35] reported that A. hydrophila type HG1/BD-2 might cause diarrhea in humans. It was noted that mutants of A. hydrophila SSU with drastically reduced hemolytic and cytotoxic activities did not have transposition within the act gene; however, the transcription of the act gene was affected drastically in the transposon mutants [75]. The production of Act was repressed by adding glucose and iron in the medium. However, the presence of calcium significantly increased Act’s production (unpublished data). We also have detected a fur (ferric-upregulation repressor)-like sequence in the genomic DNA of Aeromonas. Further studies on the regulation of the act gene are currently under study. Among various phospholipids and gangliosides tested, cholesterol abrogated the hemolytic activity of the toxin. Myristylated cholesterol, on the other hand, did not alter the biological activity of Act, indicating that the 3’-OH group of cholesterol could be important for toxin interaction with this membrane constituent [66]. Once Act interacted with cholesterol, aggregates formed, as demonstrated by immunoblot analysis, and these aggregates were heat stable [66]. The association of 14C-labeled cholesterol with Act by gel filtration also was demonstrated. Binding of the toxin to erythrocytes was temperature dependent, with no binding occurring at 4 oC. However, at 37 oC the toxin bound to erythrocytes within 1–2 min [66]. Nelson et al. [77] reported that Thy-1, a major surface glycoprotein of T lymphocytes, is a high-affinity receptor for aerolysin from A. bestiarum. Aerolysin is a bilobal toxin, and very recently, Diep et al. [78] reported that aerolysin contained two receptor-binding sites. The large lobe contained a site for glycosylphosphatidylinositolMicrobes and Infection 1999, 1129-1137
Review
anchored proteins, while the site located in the small lobe bound a surface carbohydrate determinant. Interestingly, Act did not bind to glycophorin, unlike aerolysin, further indicating the uniqueness of Act when compared with aerolysin of A. bestiarum. Act inhibited the phagocytic ability of mouse phagocytes, either in vivo or in vitro, and interferon (IFN)-γ pretreatment overcame this toxic effect. Act’s direct inhibition of phagocytic activity might be a pathological mechanism associated with some Aeromonas-mediated diseases, whereas IFN-γ protected the host against these diseases [79]. Act significantly stimulated the chemotactic activity of human leukocytes, and this stimulatory effect was inhibited by various concentrations of pertussis toxin (PT), suggesting that human leukocytes possessed Act receptors which might be coupled to PT-sensitive G-protein [80]. Studies of Krause et al. [81] indicated that aerolysin, like Act, activated PT-sensitive G-proteins in granulocytes. Our recent data also indicated that Act increased tumor necrosis factor (TNF)-α and IL-1β production in macrophages and intestinal epithelial cells (IEC6), which could be responsible for the pathology seen with Aeromonas- or Act-infected small intestine.
7. Other virulence factors of aeromonas Many strains of Aeromonas possess a regularly arrayed surface layer (S-layer) consisting of tetragonally arranged subunits tethered to the bacterial cell surface via lipopolysaccharide. Merino et al. [82] reported that Aeromonas spp. belonging to serogroup 0:11 with S-layer resisted complement-mediated killing by impeding complement activation. However, serum resistance of Aeromonas strains lacking the S-layer was due to their inability to form C5b or C5b-9 needed to lyse the cells. In their subsequent studies, Merino et al. [83] documented that a 39-kDa outer membrane protein which bound to C1q was not accessible in Aeromonas strains possessing O antigen which imparted serum resistance to the organism. The ability of some strains of A. hydrophila to invade eukaryotic cells and resist complement-mediated lysis could result in bacteremia and other invasive diseases associated with Aeromonas infections. Aeromonas spp. produce a wide range of proteases which cause tissue damage and aid in establishing an infection by overcoming host defenses and by providing nutrients for cell proliferation [84]. At least three types of proteases have been identified, which include heat-labile serine protease and heat-stable and EDTA-sensitive or -insensitive metalloproteases. In addition, some aminopeptidases may have a specific function such as activation of Act/aerolysin. Aeromonas spp. produce glycerophospholipid:cholesterol acyltransferase (GCAT) which functions as a lipase or phospholipase and could cause erythrocyte lysis by digesting their plasma membranes [84]. Although the role of GCAT in fish disease furunculosis has been suggested, its role as a virulence factor in humans in presently undefined. A microorganism must be able to first colonize before it can cause infection. A. hydrophila produces various lec1133
Review
tins and adhesins which allow bacteria to adhere to specific glycoconjugates on epithelial surfaces, on erythrocytes or in mucin on the gut mucosa. Ascencio et al. [85] recently reported bacterial cell surface extracts containing active mucin-binding components that varied in size from 22–95 kDa in different species of Aeromonas. Barnett et al. [86] reported the presence of two distinct families of type IV pili (bundle-forming pili and Tap) from Aeromonas species associated with gastroenteritis. In addition, the presence of flagellins allows Aeromonas to reach the target cell, where they colonize [86]. During an infection, a microbial pathogen must acquire all of its iron from the host. Because of the array of host iron-withholding defenses, an efficient mechanism to divert some of the metal to microbial metabolism is essential for bacterial virulence. To obtain their supply of iron, organisms synthesize and excrete iron-specific ligands of low molecular mass, collectively known as siderophores. Aeromonas spp. produce either one of the siderophores, enterobactin or amonabactin [84]. While the amonabactin producers have evolved both siderophore-dependent and -independent means for iron acqusition from a vertebrate host, enterobactin producers have to rely exclusively on nonsiderophore heme utilization, because enterobactin is inactive in vertebrate serum. Both siderophores contain 2,3-dihydroxybenzoic acid (DHB), with amonabactin produced in two biologically active forms. Amonabactin T contained 2,3-DHB, lysine, glycine and tryptophan, while amonabactin P contained phenylalanine instead of tryptophan [84]. The ferric siderophore receptor gene (fstA) of A. salmonicida exhibited significant homology with fstA genes of Vibrio anguillarum, Yersinia enterocolitica, Pseudomonas aeruginosa, and Bordetella bronchiseptica indicating that homologs of this protein were widespread in Gram-negative bacterial pathogens [84]. Quorum sensing, which is a mechanism for controlling gene expression in response to an expanding bacterial population, has been reported in Aeromonas spp. and is a subject of intensive investigation in many Gram-negative bacteria [87]. The quorum-sensing signal molecule belongs to the N-acylhomoserine lactone (AHL) family, and the signal generator proteins responsible for the synthesis of AHLs belong to the LuxI family. Accumulation of this molecule above a threshold concentration provides an indication that the minimal bacterial population size has been reached and that the appropriate target gene(s) should be activated via the LuxR family of transcriptional activators. The genes for a quorum-sensing signal generator and a response regulator were cloned from both A. hydrophila and A. salmonicida and termed ahyRI and asaRI, respectively. Downstream of both ahyI and asaI genes was a gene with close homology to iciA, an inhibitor of chromosome replication in E. coli, indicating that, in Aeromonas, cell division might be linked to quorum sensing. The LuxR protein consists of two domains with an AHL-binding site within the N-terminal end and a helixturn-helix DNA-binding motif within the C-terminal domain. It is plausible that the expression of various virulence factors of Aeromonas, in addition to cell division, could be controlled by quorum sensing. 1134
Chopra and Houston
In conclusion, the pathogenesis of Aeromonas infections involves a combination of virulence factors. Although the role of aeromonas in causing nonintestinal infections has been established, only recently has an epidemiological link associating them with gastroenteritis been suggested. However, the lack of an animal model to reproduce Aeromonas-associated diarrhea and no documented Aeromonas-associated large outbreaks are concerns in firmly establishing the role of aeromonas in gastroenteritis. It needs to be emphasized that in case-control studies all over the world, isolation of Aeromonas species as the only enteropathogen from children with diarrhea clearly has been established. However, it still is not totally clear which vital virulence factors are necessary for aeromonas to instigate gastroenteritis. The role of both cytotoxic and cytotonic enterotoxins in causing gastroenteritis has been elucidated in our laboratory. Studies on the structure-function relationship and the mechanism of action of these toxins would provide a means to intervene in Aeromonas-induced intestinal infections. Most Aeromonas studies are restricted to case-control studies with little emphasis on the molecular and biochemical characterization of virulence factors. Our studies are among the first to characterize enterotoxins of aeromonas on a molecular basis and to study bacterial-host interaction.
Acknowledgments This work was supported by grants R01 AI41611 from the National Institutes of Health and the John Sealy Memorial Endowment Fund.
References [1] Carnahan A.M., Altwegg M.,in: Austin B., Altwegg M., Gosling P.J., Joseph S. (Eds.), Taxonomy. The genus Aeromonas, John Wiley & Sons, NY, 1996, pp. 1–38. [2] Altwegg M., Martinetti L.G., Luthy-Hottenstein J., Rohrbach M., Aeromonas-associated gastroenteritis after consumption of contaminated shrimp, Eur. J. Clin. Microbiol. Infect. Dis. 10 (1991) 44–45. [3] Hanninen M.L., Oivanen P., Hirvela-Koski V., Aeromonas species in fish, fish-eggs, shrimp and freshwater, Inter. J. Food Microbiol. 34 (1997) 17–26. [4] Yamada S., Matsushita S., Dejsirilet S., Kudoh Y., Incidence and clinical symptoms of Aeromonas-associated travellers’ diarrhoea in Toyko, Epidemiol. Infect. 119 (1997) 121–126. [5] Hanninen M.L., Salmi S., Mattila L., Taipalinen R., Siitonen A., Association of Aeromonas spp with travellers’ diarrhoea in Finland, J. Med. Microbiol. 42 (1995) 26–31. [6] Abbott S.L., Seli L.S., CatinoJr M., Hartley M.A., Janda J.M., Misidentification of unusual Aeromonas species as members of the genus Vibrio: a continuing problem, J. Clin. Microbiol. 36 (1998) 1103–1104. [7] Buchanan R.L., Palumbo S.A., Aeromonas hydrophila and Aeromonas sobria as potential food poisoning species: a review, J. Food Safety 7 (1985) 15–29. Microbes and Infection 1999, 1129-1137
Enterotoxins in Aeromonas-associated gastroenteritis
[8] Janda J.M., Abbott S.L., Evolving concepts regarding the genus Aeromonas: an expanding panorama of species, disease presentations, and unanswered questions, Clin. Infect. Dis. 27 (1998) 332–344. [9] Majeed K.N., Egan A.F., Macrae I.C., Production of exotoxins by Aeromonas spp. at 5 °C, J. Appl. Bacteriol. 69 (1990) 332–337. [10] Kirov S.M., Ardestani E.K., Hayward L.J., The growth and expression of virulence factors at refrigeration temperature by Aeromonas strains isolated from foods, Inter. J. Food Microbiol. 20 (1993) 159–168. [11] Grey P.A., Kirov S.M., Adherence to Hep-2 cells and enteropathogenic potential of Aeromonas spp., Epidemiol. Infect. 110 (1993) 279–287. [12] Atkinson H.M., Trust T.J., Hemagglutination properties and adherence ability of Aeromonas hydrophila, Infect. Immun. 27 (1980) 938–946. [13] Merino S., Rubires X., Knochel S., Tomas J.M., Emerging pathogens: Aeromonas spp., Inter. J. Food Microbiol. 28 (1995) 157–168. [14] Mani S., Sadigh M., Adriole V.T., Clinical spectrum of Aeromonas hydrophila infections: report of 11 cases in a community hospital and review, Infect. Dis. Clin. Pract. 4 (1995) 79–86. [15] Kohashi T., Sakai H., Marumo F., Sato C., Aeromonas sobria infection with severe muscle degeneration in a patient with alcoholic liver cirrhosis, Amer. J. Gastroenterol. 90 (1995) 2234–2235. [16] Lin S.H., Shieh S.D., Lin Y.F., Debrauwer E., vanLanduyt H.W., Gordts B., Boelaert J.R., Fatal Aeromonas hydrophila bacteremia in a hemodialysis patient treated with deferoxamine, Amer. J. Kidney Dis. 27 (1996) 733–735. [17] Funada H., Matsuda T., Aeromonas bacteremia in patients with hematologic diseases, Int. Med. 36 (1997) 171–174. [18] Murakami T., Inoue T., Kawakami H., Takahashi H., Yasunaga T., Hypercalcemia associated with Aeromonas hydrophila gastroenteritis, Acta Paediatrica Japonica 37 (1995) 192–195. [19] Deutsch S.F., Wedzina W., Aeromonas sobria-associated leftsided segmental colitis [Review], Amer. J. Gastroenterol. 92 (1997) 2104–2106. [20] Riley P.A., Parasakthi N., Liam C.K., Development of Aeromonas hydrophila bacteremia in a patient recovering from cholera, Clin. Infect. Dis. 22 (1996) 867–868. [21] Do Sola Earle C., Quezel-Guerraz N.M., Rojas L.H., Cantos A.S., Garcia A.J., Severe acute gastroenteritis due to Aeromonas in a patient colectomized for Crohns’ disease, Revista Espanola de Enfermedades Digestivas 89 (1997) 48–50. [22] Martino R., Santamaria A., Pericas R., Sureda A., Brunet S., Acute rhabdomyolysis and myonecrosis complicating Aeromonas bacteremia in neutropenic patients with hematologic malignancies: report of two cases, Haematologica 82 (1997) 692–694. [23] Holston K.V.I., Zandvliet S.E., Rodriguez S., Nguyen H.T., Bodey G.P., Spectrum of Aeromonas and Plesiomonas infections in patients with cancer and AIDS, Experientia 47 (1991) 437–439. [24] Janda J.M., Duffey P.S., Mesophilic aeromonads in human disease: current taxonomy, laboratory identification, and infectious disease spectrum, Rev. Infect. Dis. 10 (1988) 980–997. Microbes and Infection 1999, 1129-1137
Review
[25] Challapalli M., Tess B.R., Cunningham D.G., Chopra A.K., Houston C.W., Aeromonas associated diarrhea in children, Pediatr. Infect. Dis. J. 7 (1988) 693–698. [26] Gracey M., Gastroenteritis in Australian children. Studies on the aetiology of acute diarrhea, Ann. Trop. Pediatr. 8 (1988) 68–75. [27] Verenkar, M., Naik V., Rodrigues S., Singh I., Aeromonas species and Plesiomonas shigelloides in diarrhoea in Goa, Ind. J. Path. Microbiol. 38 (1995) 169–171. [28] Goodwin C.S., Harper W.E.S., Stewart J.K., Gracey M., Burke V., Robinson J., Enterotoxigenic Aeromonas hydrophila and diarrhea in adults, Med. J. Australia 1 (1983) 25–26. [29] Quinn J.P., UHL enteric bacterial disease surveillance, Lab. Hotline 28 (1991) 2. [30] Echeverria P., Pitarangsi C., Eampokalap B., Vibulbandhitkit S., Boonthai P., Rowe B.A., Longitudinal study of the prevalence of bacterial enteric pathogens among adults with diarrhea in Bangkok, Thailand, Diagn. Microbiol. Infect. Dis. 1 (1983) 193–204. [31] Khaitovich A.B., G0al’tseva G.V., Serological diagnosis of diarrhea caused by some representatives of the genera Vibrio and Aeromonas, J. Hyg. Epidemiol. Microbiol. Immunol. 27 (1983) 211–218. [32] Echeverria P., Sack R.B., Blacklow N.R., Bodhidatta P., Rowe B., McFarland A., Prophylactic doxycycline for traveler’s diarrhea in Thailand. Further supportive evidence of A. hydrophila as an enteric pathogen, Am. J. Epid. 120 (1984) 912–921. [33] Borrell, N., Figueras M.J., Guarro J., Phenotypic identification of Aeromonas genomospecies from clinical and environmental sources, Can. J. Microbiol. 44 (1998) 103–108. [34] Albert, M.J., Ansaruzzaman M., Shimada T., Rahman N.A., Bhuiyan S., Nahar F., Qadri M., Islam S., Characterization of Aeromonas trota strains that cross-react with Vibrio cholerae O139 Bengal, J. Clin. Microbiol. 33 (1995) 3119–3123. [35] Kuhn, I., Albert M.J., Ansaruzzaman M., Bhuiyan N.A., Alabi S.A., Islam M.S., Neogi P.K., Huys G., Janssen P., Kersters K., Mollby R., Characterization of Aeromonas spp. isolated from humans with diarrhea, from healthy controls, and from surface water in Bangladesh, J. Clin. Microbiol. 35 (1997) 369–373. [36] Haque, Q.M., Sugiyama A., Iwade Y., Midorikawa Y., Yamauchi T., Diarrheal and environmental isolates of Aeromonas spp. produce a toxin similar to shiga-like toxin 1, Curr. Microbiol. 32 (1996) 239–245. [37] Brown, R.L., Sanderson K., Kirov S.M., Plasmids and Aeromonas virulence, FEMS Immun. Med. Microbiol. 17 (1997) 217–223. [38] Rose J.M., Houston C.W., Coppenhaver D.H., Dixon J.D., Kurosky A., Purification and chemical characterization of a cholera toxin-cross-reactive cytolytic enterotoxin produced by a human isolate of Aeromonas hydrophila, Infect. Immun. 57 (1989) 1165–1169. [39] Asao T., Kinoshita Y., Kozaki S., Uemura T., Sakaguchi G., Purification and some properties of Aeromonas hydrophila hemolysin, Infect. Immun. 46 (1984) 122–127. 1135
Review
[40] Chakraborty T., Huhle B., Bergbauer H., Goebel W., Cloning, expression, and mapping of the Aeromonas hydrophila aerolysin gene determinant in Escherichia coli K-12, J. Bacteriol. 167 (1986) 368–374. [41] Howard S.P., Buckley J.T., Molecular cloning and expression in Escherichia coli of the structural gene for the hemolytic toxin aerolysin from Aeromonas hydrophila, Mol. Gen. Genet. 204 (1986) 289–295. [42] Chopra A.K., Houston C.W., Peterson J.W., Jin G.F., Cloning, expression, and sequence analysis of a cytolytic enterotoxin gene from Aeromonas hydrophila, Can. J. Microbiol. 39 (1993) 513–523. [43] Chakraborty T., Montenegro M.A., Sanyal S.C., Helmuth R., Bulling E., Timmis K.N., Cloning of enterotoxin gene from Aeromonas hydrophila provides conclusive evidence of production of a cytotonic enterotoxin, Infect. Immun. 46 (1984) 435–441. [44] Wong C.Y., Mayrhofer G., Heuzenroeder M.W., Atkinson H.M., Quinn D.M., Flower R.L., Measurement of virulence of aeromonads using a suckling mouse model of infection, FEMS Immun. Med. Microbol. 15 (1996) 233–241. [45] Sanderson K., Ghazali F.M., Kirov S.M., Colonization of streptomycin-treated mice by Aeromonas species, J. Diarrhoeal Dis. Res. 14 (1996) 27–32. [46] Graf J., Symbiosis of Aeromonas veronii biovar sobria and Hirudo medicinalis, the medicinal leech: a novel model for digestive tract associations, Infect. Immun. 67 (1999) 1–7. [47] Annapurna E., Sanyal S.C., Enterotoxicity of Aeromonas hydrophila, J. Med. Microbiol. 10 (1977) 317–323. [48] Wadstrom T., Ljungh A., Wretlind B., Enterotoxin, haemolysin and cytotoxic protein in Aeromonas hydrophila from human infections, Acta. Path. Microbiol. Scand. Sect. B 84 (1976) 112–114. [49] Ljungh A., Kronevi T., Aeromonas hydrophila toxinsintestinal fluid accumulation and mucosal injury in animal models, Toxicon 20 (1982) 397–407. [50] Ljungh A., Eneroth P., Wadstrom T., Cytotonic enterotoxin from Aeromonas hydrophila, Toxicon 20 (1982) 787–794. [51] Dubey R.S., Bhattacharya A.K., Sanyal S.C., Elevation of adenosine 3’5’ cyclic monophosphate level by Aeromonas hydrophila enterotoxin, Ind. J. Med. Res. 74 (1981) 668–674. [52] Dubey R.S., Sanyal S.C., Characterization and neutralization of Aeromonas hydrophila enterotoxin in the rabbit ilealloop model, J. Med. Microbiol. 12 (1979) 347–354. [53] James C., Dibley M., Burke V., Robinson J., Gracey M., Immunological cross-reactivity of enterotoxins of Aeromonas hydrophila and cholera toxin, Clin. Exp. Immunol. 47 (1982) 34–42. [54] Bundell C.S., Watson I.M., Burke V., Gracey M., Protection of rats against cholera toxin and cholera-like enterotoxins by immunization with enteric-coated cholera toxin, Ann. Trop. Paediatrics 6 (1986) 199–204. [55] Potomski J., Burk V., Robinson J., Fumarola D., Miragliotta G., Aeromonas cytotonic enterotoxin cross-reactive with cholera toxin, J. Med. Microbiol. 23 (1987) 179–186. [56] Shultz A.J., McCardell B.A., DNA homology and immunological cross-reactivity between Aeromonas hydrophila cytotonic toxin and cholera toxin, J. Clin. Microbiol. 26 (1988) 57–61. 1136
Chopra and Houston
[57] Chopra A.K., Peterson J.W., Xu X.J., Coppenhaver D.H., Houston C.W., Molecular and biochemical characterization of a heat-labile enterotoxin from Aeromonas hydrophila, Microbial Path. 21 (1996) 357–377. [58] Chopra A.K., Houston C.W., Genaux C.T., Dixon J.D., Kurosky A., Evidence for production of an enterotoxin and cholera toxin cross-reactive factor by Aeromonas hydrophila, J. Clin. Microbiol. 24 (1986) 661–664. [59] Chopra A.K., Houston C.W., Purification and partial characterization of a cytotonic enterotoxin produced by Aeromonas hydrophila, Can. J. Microbiol. 35 (1989) 719–727. [60] Chopra A.K., Vo T.N., Houston C.W., Mechanism of action of a cytotonic enterotoxin produced by Aeromonas hydrophila, FEMS Microbiol. Lett. 91 (1992) 15–19. [61] Pal S., Datta A., Nair G.B., Guhathakurta B., Use of monoclonal antibodies to identify phospholipase C as the enterotoxic factor of the bifuntional hemolysinphospholipase C molecule of Vibrio cholerae O139, Infect. Immun. 66 (1998) 3974–3977. [62] Granum P.E., O’Sullivan K., Tomás J.M., Ørmen Ø., Possible virulence factors of Aeromonas spp. from food and water, FEMS Immun. Med. Microbiol. 21 (1998) 131–137. [63] Chopra A.K., Pham R., Houston C.W., Cloning and expression of putative cytotonic enterotoxin-encoding genes from Aeromonas hydrophila, Gene 139 (1994) 87–91. [64] McCardell B.A., Madden J.M., Kothary M.H., Sathyamoorthy V., Purification and characterization of CHO cell-elongating toxin produced by Aeromonas hydrophila, Microbial Path. 19 (1995) 1–9. [65] Rose J.M., Houston C.W., Kurosky A., Bioactivity and immunological characterization of a cholera toxin-crossreactive cytolytic enterotoxin from Aeromonas hydrophila, Infect. Immun. 57 (1989) 1170–1176. [66] Ferguson M.R., Xu X.J., Houston C.W., Peterson J.W., Coppenhaver D.H., Popov V.L., Chopra A.K., Hyperproduction, purification and mechanism of action of the cytotoxic enterotoxin produced by Aeromonas hydrophila, Infect. Immun. 65 (1997) 4299–4308. [67] Potomski J., Burke V., Watson I., Gracey M., Purification of cytotoxic enterotoxin of Aeromonas sobria by use of monoclonal antibodies, J. Med. Microbiol. 23 (1987) 171–177. [68] Chopra A.K., Houston C.W., Kurosky A., Genetic variation in related cytolytic toxins produced by different species of Aeromonas, FEMS Microbiol. Lett. 78 (1991) 231–237. [69] Ferguson M.R., Xu X.J., Houston C.W., Peterson J.W., Chopra A.K., Amino-acid residues involved in biological functions of the cytolytic enterotoxin from Aeromonas hydrophila, Gene 156 (1995) 79–83. [70] Buckley J.T., Howard S.P., The cytotoxic enterotoxin of Aeromonas hydrophila is aerolysin, Infect. Immun. 67 (1999) 466–467. [71] Bhakdi S., Tranum-Jenson J., Alpha-toxin of Staphylococcus aureus, Microbiol. Rev. 55 (1991) 733–751. [72] Chakroborty T., Schmid A., Notermans S., Benz R., Aerolysin of Aeromonas sobria: evidence for formation of ionpermeable channels and comparison with alpha-toxin of Staphylococcus aureus, Infect. Immun. 58 (1990) 2127–2132. [73] Parker M.W., vander Goot F.G., Buckley J.T., Aerolysunthe ins and outs of a channel forming protein, Mol. Microbiol. 19 (1996) 205–212. Microbes and Infection 1999, 1129-1137
Enterotoxins in Aeromonas-associated gastroenteritis
[74] Blomqvist L., Sjogren A., Production and characterization of monoclonal antibodies against S. aureus alpha toxin, Toxicon 26 (1988) 265–273. [75] Xu X.J., Ferguson M.R., Popov V.L., Houston C.W., Peterson J.W., Chopra A.K., Role of a cytotoxic enterotoxin in Aeromonas-mediated infections: development of transposon and isogenic mutants, Infect. Immun. 66 (1998) 3501–3509. [76] Chakraborty T., Huhle B., Hof H., Bergbauer H., Goebel W., Marker exchange mutagenesis of the aerolysin determinant in Aeromonas hydrophila demonstrates the role of aerolysin in A. hydrophila associated systemic infections, Infect. Immun. 55 (1987) 2274–2280. [77] Nelson K.L., Raja S.M., Buckley J.T., The glycosylphosphatidylinositol-anchored surfce glycoprotein Thy-1 is a receptor for the channel-forming toxin aerolysin, J. Biol. Chem. 272 (1997) 12170–12174. [78] Diep D.B., Nelson K.L., Lawrence T.S., Sellman B.R., Tweten R.K., Buckley J.T., Expression and properties of an aerolysin-Clostridium septicum alpha toxin hybrid protein, Mol. Microbiol. 31 (1999) 785–794. [79] Jin G.F., Houston C.W., The effect of Aeromonas hydrophila enterotoxins on the phagocytic function of mouse phagocytes, Dig. Dis. Sci. 37 (1992) 1697–1703. [80] Jin G.F., Chopra A.K., Houston C.W., Stimulation of neutrophil leukocyte chemotaxis by a cloned cytolytic enterotoxin of Aeromonas hydrophila, FEMS Microbiol. Lett. 98 (1992) 285–290. [81] Krause K.H., Fivaz M., Monod A., Van Der Goot F.G.,
Microbes and Infection 1999, 1129-1137
Review
[82]
[83]
[84] [85]
[86]
[87]
Aerolysin induces G-protein activation and Ca2+ release from intracellular stores in human granulocytes, J. Biol. Chem. 273 (1998) 18122–18129. Merino S., Rubires X., Aguilar A., Alberti S., HernandezAlles S., Benedi V.J., Tomas J.M., Mesophilic Aeromonas sp. Serogroup 0:11 resistance to complement-mediated killing, Infect. Immun. 64 (1996) 5302–5309. Merino S., Nogueras M.M., Aguilar A., Rubires X., Alberti S., Benedi V.J., Tomas J.M., Activation of the complement classical pathway (C1q binding) by mesophilic Aeromonas hydrophila outer membrane protein, Infect. Immun. 66 (1998) 3825–3831. Pemberton J.M., Kidd S.P., Schmidt R., Secreted enzymes of Aeromonas, FEMS Microbiol. Lett. 152 (1997) 1–10. Ascencio F., Martinez-Arias W., Romero M.J., Wadstrom T., Analysis of the interaction of Aeromonas caviae, A, hydrophila and A, sobira with mucins, FEMS Immun. Med. Microbiol. 20 (1998) 219–229. Barnett T.C., Kirov S.M., Strom M.S., Sanderson K., Aeromonas spp, possess at least two distinct type IV pilus families, Microbial Path. 23 (1997) 241–247. Swift S., Karlyshev A.V., Fish L., Durant E.L., Winson M.K., Chhabra S.R., William S.P., MacIntyre S., Stewart G.S., Quorum sensing in Aeromonas hydrophila and Aeromonas salmonicida: identification of the LuxRI homologs AhyRI and AsaRI and their cognate N-acylhomoserine lactone signal molecules, J. Bacteriol. 178 (1997) 5271–5281.
1137
Review
Enterotoxins in Aeromonas-associated gastroenteritis Ashok K. Chopra*, Clifford W. Houston Department of Microbiology and Immunology, The University of Texas Medical Branch, Galveston, TX 77555–1070, USA
ABSTRACT – Aeromonas species produce an array of virulence factors, and the pathogenesis of Aeromonas infections is therefore complex and multifactorial. Aeromonas-associated gastroenteritis is especially a problem in young children. The potential involvement of enterotoxins in the pathogenesis of Aeromonas infections is discussed. © 1999 Éditions scientifiques et médicales Elsevier SAS Aeromonas / cytotoxic enterotoxin / aerolysin / cytotonic enterotoxin / pathogenic mechanisms
1. Introduction Aeromonas species (spp.), which recently have been placed in a new family – Aeromonadaceae, are predominately pathogenic to poikilotherms [1]. However, Aeromonas has emerged as an important human pathogen because of suspected food-borne outbreaks [2, 3] and the increased incidence of Aeromonas isolation from patients with traveler’s diarrhea [4, 5]. Misidentification of Aeromonas is a continuing problem, and in 1998, the Department of Health Services, Berkeley, CA, reported two unusual cases of Aeromonas infections one associated with bacteremia (A. schubertii), and another in which the organism was recovered from an infected gall bladder (A. veronii bv. veronii) [6]. These strains initially were identified as Vibrio damsela and V. cholerae, respectively. Aeromonas is found in both fresh and salt water and in virtually all foods [7], and causes a wide variety of human infections, including septicemia, wound infections, meningitis, pneumonia and gastroenteritis [1, 8]. In addition, Aeromonas-associated hemolytic uremic syndrome, burnassociated sepsis, and a variety of respiratory infections, including epiglottis, have been reported [8]. The ability of these microorganisms to grow well at refrigeration temperatures and produce exotoxins at these low temperatures [9] could be important in their role as food poisoning agents. Kirov et al. [10] reported that 10% of the Aeromonas strains isolated from food could adhere to HEp-2 cells at 37 and 5 oC, and the adherent strains of Aeromonas were virulent and produce exotoxins [10, 11]. It also has been shown that a species of Aeromonas adheres to human blood cells [12]. Among the 14 species of Aeromonas isolated to date [1], A. hydrophila, A. caviae, and A. veronii bv. * Correspondence and reprints Microbes and Infection 1999, 1129-1137
sobria, have most commonly been isolated from human infections and been shown to produce a variety of biologically active extracellular products. These include hemolysins, cytotoxins, enterotoxins, proteases, leukocidin, phospholipases, endotoxin, outer membrane proteins, and fimbriae or adhesions. Some Aeromonas spp. also possess S-layers and form capsules [13]. Taken together, the ability of these bacteria to invade host cells and disseminate to virtually any organ via blood and their capacity to produce various virulence factors contribute to the pathogenesis of disease mediated by Aeromonas. An epidemiological link has been suggested between the consumption of contaminated foods and human illness, with aeromonas as a predominant flora in fecal and biopsy specimens [2, 3, 8]. In addition, serological evidence of infection was demonstrated by the presence of agglutinating antibodies to the somatic antigen and neutralizing antibodies to the exotoxins [8]. Recent reports of A. hydrophila and A. sobria (now referred to as A. veronii bv. sobria) isolation from humans with sepsis, peritonitis, urinary tract infections, severe muscle degeneration, and bacteremia with myonecrosis and gas-gangrene in a hemodialysis patient demonstrated the importance of this organism in various diseases [14–16]. In a more recent study in Japan, Funada and Matsuda [17] reported Aeromonas bacteremia in patients with hematologic diseases, with no exposure to water or fish. A clinical picture ranging from mild gastroenteritis to severe gastroenteritis was observed in 71% of the patients, with an overall mortality rate of 35%. In another report, the association of A. hydrophila gastroenteritis with hypercalcemia was cited [18]. Deutsch and Wedzina [19] reported a case of left-sided segmental colitis due to A. sobria (A. veronii bv. sobria), and supported the view that Aeromonas spp. needed to be considered in the differential diagnosis of colitis. In Malaysia, bacteremia due to A. hydrophila in a patient recovering from cholera was reported [20]. Like1129
Review
wise, a case of severe acute diarrhea produced by A. sobria (A. veronii bv. sobria) in a patient colectomized for Crohn’s disease was published [21]. Martino et al. [22] reported that patients with acute nonlymphoblastic leukemia developed septic shock due to Aeromonas infection. The latter confirms the potentially aggressive nature of these bacteria in neutropenic cancer patients, as previously reported [23]. Overall, the literature suggests that Aeromonas, once mainly considered an opportunistic pathogen in immunocompromised humans, is now implicated as the etiologic agent in numerous clinical situations involving immunocompetent individuals of all age groups.
2. Epidemiology of Aeromonasassociated gastroenteritis One of the most commonly reported infections by aeromonas is gastroenteritis that ranges from a mild, selflimiting watery diarrhea to a more severe, invasive Shigella-like dysenteric form [24]. Further, chronic diarrhea exceeding one year’s duration due to A. caviae or A. hydrophila also has been reported [8]. In a survey examining the causes of traveler’s diarrhea treated in Tokyo between 1985 and 1995, Aeromonas spp. (A. veronii bv. sobria and A. hydrophila) were isolated from 5.5% of the travelers returning from developing countries [4]. Aeromonas spp. were isolated from 8.7% of Finnish tourists with diarrhea who had traveled to Morocco and from 1.4% of nondiarrheal tourists. Species of Aeromonas were found as the sole pathogen in 5.5% of those patients [5]. In a 27-month prospective study conducted in Chicago, we found Aeromonas species were the only potential bacterial enteropathogens isolated from children suffering from diarrhea and ranging from 1 to 27 months of age [25]. Gastroenteritis due to aeromonas is especially a problem for the pediatric population. These microorganisms have been detected on a worldwide basis as the only pathogen from 2–20% of the children suffering from diarrhea and from only 0–2% of children without diarrhea [26, 27]. In a study by Goodwin et al. [28], A. hydrophila was isolated from the feces of 32 adult patients during a nine-month period in three hospitals in Western Australia. All 32 isolates produced enterotoxin detected by the suckling mouse test, and all patients except one had diarrhea. In another study, 224 cases of Aeromonas gastroenteritis were reported in Iowa, from January to June 1991, thus making this organism the most prevalent enteric pathogen isolated from stools of patients suffering from gastroenteritis [29]. A. hydrophila has been reported to be the most common enteric pathogen in adults with diarrhea in Bangkok, Thailand [30]. Serum samples from adult individuals usually contain antibodies to aeromonas, and Khaitovich [31] reported that 47% of the patients with diarrhea due to other enteropathogens had Aeromonas-specific antibodies. Treatment with prophylactic doxycycline significantly prevented the colonization of Aeromonas sp. in the gastrointestinal tract [32]. The high environmental prevalence of these pathogenic bacteria should be regarded as an important threat 1130
Chopra and Houston
to public health, since Aeromonas infections generally are acquired through consumption of water and food. Borrell et al. [33] reported that most pathogenic Aeromonas spp. were prevalent in environmental samples, with A. veronii bv. sobria being the most common in lakes and reservoirs (42%) and treated drinking water (25%), and A. caviae was the most common in sea water (26%) and milk products (36%). A. hydrophila (18%) was the second most prevalent species isolated in untreated water [33]. Interestingly, Albert et al. [34] reported that some strains of Aeromonas shared common antigens with that of V. cholerae O139. They reported that a polyclonal antiserum to a cross-reacting A. trota strain cross-protected infant mice against cholera on challenge with virulent V. cholerae O139. Based on hybridization groups (HGs), most human isolates have been allocated to DNA HG 4 (A. caviae) and HG 1 (A. hydrophila) [35]. Further, human isolates have been placed in BD-1 and BD-2 types, according to PhenePlate typing, and the HG1/BD-2 type may represent a pathogenic A. hydrophila type that is able to produce diarrhea in humans. Studies of Haque et al. [36] reported the presence of Shiga-like toxin 1 on a plasmid in some strains of A. hydrophila and A. sobria (A. veronii bv. sobria) which could account for Aeromonas-associated gastroenteritis. However, in general the role of plasmids in Aeromonas virulence is questionable [37].
3. Enterotoxins of aeromonas Two categories of enterotoxins – cytotonic and cytotoxic – have been discovered in culture filtrates of Aeromonas isolates. The cytotonic enterotoxin, like cholera toxin (CT), does not cause degeneration of crypts and villi of the small intestine, whereas the cytotoxic enterotoxin results in extensive damage to epithelium. The cytotoxic enterotoxin of aeromonas also has been referred to by other investigators as β-hemolysin, and/or aerolysin [3841]. These toxin molecules have hemolytic and cytotoxic activities in addition to an enterotoxic activity [42]. However, toxin molecules with hemolytic or cytotoxic activity alone also have been isolated from Aeromonas sp. [43].
4. Animal models for detecting enterotoxins and enterotoxigenicity of aeromonas Ligated ileal loops in adult mice, rats, and rabbits and the suckling mouse assay have been successfully used by several investigators to examine enterotoxic activity in culture filtrates of Aeromonas isolates and of purified enterotoxins [44], and likewise, the enterotoxigenicity of Aeromonas spp. However, these methods do not use the natural oral route of infection in humans and are not suitable for quantifying virulence of the organism. More recently, virulence of the organism has been correlated successfully by oral administration of organisms in suckling mice and by determining the LD50 [44]. Sanderson et al. [45] noted that only the streptomycin-treated adult Microbes and Infection 1999, 1129-1137
Enterotoxins in Aeromonas-associated gastroenteritis
mice could colonize Aeromonas species in the intestine when it was given by an intragastric route. However, diarrheal symptoms were not produced in this model. Likewise, protein-malnourished mice could colonize aeromonas in the intestine when it was given in drinking water over four days. Although the animals did not develop diarrhea, 75% of them were shedding aeromonas two days after the challenge period [45]. Therefore, an animal model is needed to demonstrate Aeromonasinduced gastroenteritis using a natural route of infection. Recently, Graf [46] suggested a possible use of medicinal leech (Hirudo medicinalis) as a model for digestive tract association of Aeromonas spp. A. veronii bv. sobria was exclusively found in the digestive tract of the leech. Interestingly, human fecal isolates of A. hydrophila and A. veronii bv. sobria colonized the digestive tract to the same extent as the symbiotic isolate. Further, human isolates proliferated to the same extent in the crop fluid as the symbiotic isolate six hours after blood feeding, indicating a potential for this digestive tract model in studying bacterial-host interaction.
5. Cytotonic enterotoxins Since the first report of a cytotonic enterotoxin of aeromonas, several investigators have identified an enterotoxic factor in culture filtrates of Aeromonas species that could be responsible for fluid secretion in the small intestine of animals. However, not much effort has been made to purify and characterize the class of this enterotoxin. A. hydrophila was first shown to be enteropathogenic by Annapurna and Sanyal [47], who injected live cells into ligated intestinal loops of rabbits. It later was shown by Wadstrom’s group [48] that a 15-kDa extracellular cytotonic enterotoxin was produced by A. hydrophila. This enterotoxin induced fluid accumulation in rabbit, rat, and mouse intestinal loops [49]. Ljungh et al. [50] demonstrated that A. hydrophila cytotonic enterotoxin, like CT, gave a positive rabbit skin test. These workers [50], as well as Dubey et al. [51], observed increases in the levels of 3’,5’-adenosine monophosphate (cAMP) in rabbit intestinal epithelial cells treated with enterotoxin. Neutralization experiments with rabbit intestinal loops, rabbit skin, and Y1 adrenal cells failed to demonstrate any immunological relationship between Aeromonas cytotonic enterotoxin and CT or Escherichia coli heat-labile (LT-I) enterotoxin [50, 52]. James et al. [53], on the other hand, reported that A. hydrophila produced two types of enterotoxins. One was heat-labile (56 oC for 10 min), caused fluid secretion in suckling mice, and did not cross-react with CT. The second was heat-stable (100 oC for 30 min), caused fluid accumulation in a rat perfusion system, and reacted with cholera antitoxin. James et al. [53] further demonstrated that preincubation with anticholera toxin significantly reduced intestinal secretions induced by culture filtrates of A. hydrophila in jejunal perfusion experiments in rats. Likewise, Bundell et al. [54] reported that immunization of rats with CT not only gave protection against pure CT during intestinal perfusion, but also significantly protected Microbes and Infection 1999, 1129-1137
Review
against the secretory effects of E. coli LT-I and a CT-like toxin produced by A. sobria (A. veronii bv. sobria). Chakraborty et al. [43] reported cloning of a cytotonic enterotoxin gene in 1984 from a diarrheal isolate of A. veronii bv. sobria (now referred to as A. trota). The cell lysate from this clone caused fluid secretion in rabbit ligated ileal loops and in suckling mice. No DNA homology was detected between their cytotonic enterotoxin gene, CT, and LT-I genes, based on Southern blot analysis. This enterotoxin gene [43], which has not been characterized further, appeared to be different from that isolated by Potomski et al. [55] from A. sobria (A. veronii bv. sobria. The latter enterotoxin cross-reacted with CT, and the biological activity associated with this enterotoxin was neutralized by antibodies to CT. Further, the purified enterotoxin from Potomski’s group exhibited bands of 43.5, 29.5, and 26 kDa [55] on sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE), unlike the 15-kDa cytotonic enterotoxin isolated by Wadstrom’s and Sanyal’s groups [50, 51]. Likewise, Shultz and McCardell [56] demonstrated three protein bands of 89, 37, and 11 kDa in cell lysates of an A. hydrophila isolate that reacted with cholera antitoxin on an immunoblot. Despite differing molecular sizes and ability to react with antibodies to CT, all of these toxins had a mechanism of action similar to that of CT; i.e., they elevated cAMP levels in eukaryotic cells [57]. Our laboratory provided evidence for the presence of a cytotoxic enterotoxin and a cytotonic enterotoxin in culture filtrates of Aeromonas isolates [58]. Both of these toxins were biologically heat labile at 56 oC/20 min. Subsequently, the heat-labile cytotonic enterotoxin (referred to as Alt) was purified from a diarrheal isolate, SSU, of A. hydrophila. The purified native Alt exhibited a molecular mass of 44 kDa and was biologically active in in vivo and in vitro models. Alt was not related to CT, based on Western blot analysis, Southern blot hybridization using CT-specific probes, and its inability to bind to GM1ganglioside, a receptor for CT and LT-I [59]. In Chinese hamster ovary (CHO) cells, Alt elevated cAMP and prostaglandin (e.g., PGE2) levels [59, 60]. The gene encoding Alt was cloned [61] and expressed using various E. coli-based expression vectors and subsequently purified [57]. The recombinant-Alt (r-Alt) in E. coli was slightly smaller (35–38 kDa), compared to native Alt (44 kDa) purified from A. hydrophila [59]. Alt exhibited 45–51% identity in a 175-aa overlap with the carboxyl half of lipase and phospholipase C (PLC) of A. hydrophila [57]. An aa sequence VHFLGHSLGA (aa residues 218–227) of Alt was highly homologous to the putative substrate-binding domain found within bacterial, fungal, porcine, and human lipases. Interestingly, the CHO cell elongation and the fluid secretory ability of r-Alt in rat ligated ileal loop was much lower (ten- to 15-fold) compared to the native Alt purified from Aeromonas sp. [57], indicating differential processing of the toxin in E. coli and aeromonas which affected the toxin’s enterotoxicity. Cyclic AMP levels of the intestinal mucosa in r-Alt-treated rat loops were 3–4 fold higher, and the PGE2 levels in the loop fluid were elevated by at least five-fold compared to phosphate-buffered saline (PBS)-treated control loops. Re1131
Review
cent expression of the alt gene back in aeromonas using a tac promoter-based, multi-host-range expression vector pMMB66 indicated the r-Alt was 44 kDa when produced from plasmid pMMB66 in aeromonas and secreted into the medium, like native Alt. Further, the biological activity of recombinant and native Alt produced from aeromonas was identical (unpublished data). Active immunization of mice with purified r-Alt resulted in a significant decrease (39%) in the fluid secretory response when challenged with wild-type A. hydrophila [57]. We recently have developed an alt gene minus mutant of A. hydrophila SSU by double crossover homologous recombination. The effect of this mutation on the enterotoxic activity is under investigation. Studies of Pal et al. [61] identified PLC as the enterotoxic factor of the bifunctional hemolysin-PLC molecule of V. cholerae O139. In a recent study by Granum et al. [62], 75% of the Aeromonas isolates from food and water in Norway contained the alt gene, based on PCR amplification, and one of the A. hydrophila strains was probably involved in an outbreak of food poisoning caused by ingestion of raw, fermented fish. We identified another cytotonic enterotoxin gene (designated as ast) in the genomic library of A. hydrophila SSU. The ast gene encoded a product (Ast) which caused CHO cells to elongate, an enterotoxic activity stable at 56 oC for 20 min [63]. The crude Ast preparations evoked fluid secretion in the rat small intestine, and the cAMP levels in the mucosal cells were elevated, compared to appropriate control (unpublished data). Expression of the DNA fragment containing the ast gene in E. coli exhibited prominent bands of 32 and 67 kDa. The relationship of these two polypeptides to the structure of Ast is under investigation. McCardell et al. [64] purified from A. hydrophila a heatstable enterotoxin (56 oC/20 min) which exhibited a size of 70 kDa. This toxin was non-CT cross-reactive, did not increase cAMP, cGMP, and PGE2 levels on CHO cells, but evoked intestinal fluid accumulation in infant mice. Whether this enterotoxin has any similarity to Ast needs to be determined.
6. Cytotoxic enterotoxin Asao’s group [39] purified a hemolysin molecule from A. hydrophila with various biological activities, which included hemolytic, cytotoxic and enterotoxic activities, as well as lethality in mice. However, this toxin was not characterized further at the molecular level. Subsequently, we reported purification from A. hydrophila SSU of a cytolytic enterotoxin (52 kDa; now designated as Act) with activities similar to that associated with Asao’s hemolysin [65, 66]. The 50% lethal dose of Act for mice was 27.5 ng, and interestingly, Act reacted with antibodies to CT in the enzyme-linked immunosorbant assay and Western blots. However, these antibodies failed to neutralize the biological activities of Act [65]. Subsequent cloning and DNA sequence analysis of the cytolytic enterotoxin gene (act) revealed no homology with the CT encoding gene (ctx) [42], suggesting a conformational epitope on Act that reacted with antibodies to CT at a site(s) distinct from site(s) responsible for biological activities. Potomski 1132
Chopra and Houston
et al. [67] similarly isolated from A. sobria (A. veronii bv. sobria) a cytotoxic enterotoxin of 63 kDa and which possessed biological activities similar to those of Act, but did not cross-react with CT. No further molecular characterization of this 63-kDa protein has been reported. Two different groups reported cloning of an aerolysin gene from A. trota and A. bestiarum (originally classified as A. hydrophila) [40, 41] while our studies were in progress. The structural gene (aerA) coding for an aerolysin was shown to be related to Asao’s hemolysin based on Western blot analysis [40]. At the DNA level, approximately 75% homology was noted in these two aerolysin structural genes; however, the 5’ flanking DNA sequences to the aerolysin genes were quite divergent [42]. Although the first 25 NH2-terminal aa residues of Act purified from A. hydrophila SSU were identical to the aerolysin reported by Buckley’s group from A. bestiarum [38], Act and their aerolysin differed in aa composition [38]. Molecular characterization of the act gene revealed 79–93% homology at the aa level between Act and the two aerolysins [42]. In a 14-aa stretch (from position 449–462), Act exhibited only 21–36% homology with the two aerolysins. The 5’ flanking sequences to the act and aerolysin genes exhibited 48–71% homology [42]. Differential neutralization of Act with Act- and aerolysin-specific monoclonal antibodies ([68] and unpublished data) and the role of selected aa residues, as determined by the site-directed mutagenesis, in aerolysin and Act-associated hemolytic activity was very intriguing [69]. Some aa residues, such as histidine (His) 107 and 132 were crucial for aerolysin’s hemolytic activity, whereas the analogous His residues 130 and 155 did not affect the hemolytic activity of Act. Likewise, tryptophan (Trp) 371 had little or no effect on the hemolytic titers of aerolysin; however, an analogous Trp 394 abrogated the biological activity of Act [69]. Detailed structure-function studies on Act and aerolysin indicated that Act/aerolysin were closely related. However, some heterogeneity at the aa level could lead to a possible differential folding of these molecules, resulting in differential neutralization of these toxins by specific monoclonal antibodies and the observed hemolytic activity associated with various mutated toxins [70]. It also was evident from the literature that Act/aerolysin and α-toxin of Staphylococcus aureus were activated by binding to the target cell membrane, with subsequent oligomerization and pore formation [66, 71–73]. Most studies on aerolysin are targeted toward measuring hemolytic activity. However, site-directed mutagenesis data on Act indicated possibly different loci on a single chain of Act which might be associated with various biological activities [69]. Blomquist and Sjorgen [74] similarly reported that one monoclonal antibody did not neutralize the hemolytic or dermonecrotic effects of α-toxin from S. aureus, but did abrogate lethality. Therefore, different biological activities of the α-toxin molecule may be evoked by different molecular regions which also may be the case with our Act. Interestingly, a peptide (aa residues 245–274) has been identified in Act which competed with native Act to bind to CHO cells. The polyclonal anti-peptide antibodies to this peptide as well as to peptide (aa residues 361–405) Microbes and Infection 1999, 1129-1137
Enterotoxins in Aeromonas-associated gastroenteritis
significantly neutralized the biological activity of Act [69]. Our studies in which we demonstrated that the lack of the act gene in Aeromonas isolate A52 would dramatically increase the LD50 for mice indicated the importance of this enterotoxin in the pathogenesis of Aeromonas-mediated diseases. These findings were substantiated further by the generation of transposon mutants of A. hydrophila SSU with significantly reduced biological activity of Act compared to that seen with wild-type Aeromonas spp. These mutants were nonlethal to mice when injected intraperitoneally at a concentration of 5 × 107 compared to wildtype A. hydrophila, which was lethal at this dose [75]. Likewise, the LD50 of isogenic mutants of A. hydrophila in which the act gene was inactivated by double crossover homologous recombination was 1 × 108 compared to wild-type Aeromonas sp., which was 3 × 105. Reintegration of the native act gene in place of the truncated toxin gene in isogenic mutants resulted in complete restoration of Act’s biological activity and virulence in mice. The animals injected with a sublethal dose of wild-type aeromonas or its revertant, but not the isogenic mutant, had circulating toxin-specific neutralizing antibodies [75]. Our studies substantiated the earlier findings of Chakraborty et al. [76] in which they showed that aerolysin-deficient mutants of A. trota were less virulent in mice than wild-type Aeromonas sp. were. Since the aerolysin gene from A. trota differed significantly from the act gene from A. hydrophila [42], it was important to delete the act gene from an authentic strain of A. hydrophila, particularly as Kuhn et al. [35] reported that A. hydrophila type HG1/BD-2 might cause diarrhea in humans. It was noted that mutants of A. hydrophila SSU with drastically reduced hemolytic and cytotoxic activities did not have transposition within the act gene; however, the transcription of the act gene was affected drastically in the transposon mutants [75]. The production of Act was repressed by adding glucose and iron in the medium. However, the presence of calcium significantly increased Act’s production (unpublished data). We also have detected a fur (ferric-upregulation repressor)-like sequence in the genomic DNA of Aeromonas. Further studies on the regulation of the act gene are currently under study. Among various phospholipids and gangliosides tested, cholesterol abrogated the hemolytic activity of the toxin. Myristylated cholesterol, on the other hand, did not alter the biological activity of Act, indicating that the 3’-OH group of cholesterol could be important for toxin interaction with this membrane constituent [66]. Once Act interacted with cholesterol, aggregates formed, as demonstrated by immunoblot analysis, and these aggregates were heat stable [66]. The association of 14C-labeled cholesterol with Act by gel filtration also was demonstrated. Binding of the toxin to erythrocytes was temperature dependent, with no binding occurring at 4 oC. However, at 37 oC the toxin bound to erythrocytes within 1–2 min [66]. Nelson et al. [77] reported that Thy-1, a major surface glycoprotein of T lymphocytes, is a high-affinity receptor for aerolysin from A. bestiarum. Aerolysin is a bilobal toxin, and very recently, Diep et al. [78] reported that aerolysin contained two receptor-binding sites. The large lobe contained a site for glycosylphosphatidylinositolMicrobes and Infection 1999, 1129-1137
Review
anchored proteins, while the site located in the small lobe bound a surface carbohydrate determinant. Interestingly, Act did not bind to glycophorin, unlike aerolysin, further indicating the uniqueness of Act when compared with aerolysin of A. bestiarum. Act inhibited the phagocytic ability of mouse phagocytes, either in vivo or in vitro, and interferon (IFN)-γ pretreatment overcame this toxic effect. Act’s direct inhibition of phagocytic activity might be a pathological mechanism associated with some Aeromonas-mediated diseases, whereas IFN-γ protected the host against these diseases [79]. Act significantly stimulated the chemotactic activity of human leukocytes, and this stimulatory effect was inhibited by various concentrations of pertussis toxin (PT), suggesting that human leukocytes possessed Act receptors which might be coupled to PT-sensitive G-protein [80]. Studies of Krause et al. [81] indicated that aerolysin, like Act, activated PT-sensitive G-proteins in granulocytes. Our recent data also indicated that Act increased tumor necrosis factor (TNF)-α and IL-1β production in macrophages and intestinal epithelial cells (IEC6), which could be responsible for the pathology seen with Aeromonas- or Act-infected small intestine.
7. Other virulence factors of aeromonas Many strains of Aeromonas possess a regularly arrayed surface layer (S-layer) consisting of tetragonally arranged subunits tethered to the bacterial cell surface via lipopolysaccharide. Merino et al. [82] reported that Aeromonas spp. belonging to serogroup 0:11 with S-layer resisted complement-mediated killing by impeding complement activation. However, serum resistance of Aeromonas strains lacking the S-layer was due to their inability to form C5b or C5b-9 needed to lyse the cells. In their subsequent studies, Merino et al. [83] documented that a 39-kDa outer membrane protein which bound to C1q was not accessible in Aeromonas strains possessing O antigen which imparted serum resistance to the organism. The ability of some strains of A. hydrophila to invade eukaryotic cells and resist complement-mediated lysis could result in bacteremia and other invasive diseases associated with Aeromonas infections. Aeromonas spp. produce a wide range of proteases which cause tissue damage and aid in establishing an infection by overcoming host defenses and by providing nutrients for cell proliferation [84]. At least three types of proteases have been identified, which include heat-labile serine protease and heat-stable and EDTA-sensitive or -insensitive metalloproteases. In addition, some aminopeptidases may have a specific function such as activation of Act/aerolysin. Aeromonas spp. produce glycerophospholipid:cholesterol acyltransferase (GCAT) which functions as a lipase or phospholipase and could cause erythrocyte lysis by digesting their plasma membranes [84]. Although the role of GCAT in fish disease furunculosis has been suggested, its role as a virulence factor in humans in presently undefined. A microorganism must be able to first colonize before it can cause infection. A. hydrophila produces various lec1133
Review
tins and adhesins which allow bacteria to adhere to specific glycoconjugates on epithelial surfaces, on erythrocytes or in mucin on the gut mucosa. Ascencio et al. [85] recently reported bacterial cell surface extracts containing active mucin-binding components that varied in size from 22–95 kDa in different species of Aeromonas. Barnett et al. [86] reported the presence of two distinct families of type IV pili (bundle-forming pili and Tap) from Aeromonas species associated with gastroenteritis. In addition, the presence of flagellins allows Aeromonas to reach the target cell, where they colonize [86]. During an infection, a microbial pathogen must acquire all of its iron from the host. Because of the array of host iron-withholding defenses, an efficient mechanism to divert some of the metal to microbial metabolism is essential for bacterial virulence. To obtain their supply of iron, organisms synthesize and excrete iron-specific ligands of low molecular mass, collectively known as siderophores. Aeromonas spp. produce either one of the siderophores, enterobactin or amonabactin [84]. While the amonabactin producers have evolved both siderophore-dependent and -independent means for iron acqusition from a vertebrate host, enterobactin producers have to rely exclusively on nonsiderophore heme utilization, because enterobactin is inactive in vertebrate serum. Both siderophores contain 2,3-dihydroxybenzoic acid (DHB), with amonabactin produced in two biologically active forms. Amonabactin T contained 2,3-DHB, lysine, glycine and tryptophan, while amonabactin P contained phenylalanine instead of tryptophan [84]. The ferric siderophore receptor gene (fstA) of A. salmonicida exhibited significant homology with fstA genes of Vibrio anguillarum, Yersinia enterocolitica, Pseudomonas aeruginosa, and Bordetella bronchiseptica indicating that homologs of this protein were widespread in Gram-negative bacterial pathogens [84]. Quorum sensing, which is a mechanism for controlling gene expression in response to an expanding bacterial population, has been reported in Aeromonas spp. and is a subject of intensive investigation in many Gram-negative bacteria [87]. The quorum-sensing signal molecule belongs to the N-acylhomoserine lactone (AHL) family, and the signal generator proteins responsible for the synthesis of AHLs belong to the LuxI family. Accumulation of this molecule above a threshold concentration provides an indication that the minimal bacterial population size has been reached and that the appropriate target gene(s) should be activated via the LuxR family of transcriptional activators. The genes for a quorum-sensing signal generator and a response regulator were cloned from both A. hydrophila and A. salmonicida and termed ahyRI and asaRI, respectively. Downstream of both ahyI and asaI genes was a gene with close homology to iciA, an inhibitor of chromosome replication in E. coli, indicating that, in Aeromonas, cell division might be linked to quorum sensing. The LuxR protein consists of two domains with an AHL-binding site within the N-terminal end and a helixturn-helix DNA-binding motif within the C-terminal domain. It is plausible that the expression of various virulence factors of Aeromonas, in addition to cell division, could be controlled by quorum sensing. 1134
Chopra and Houston
In conclusion, the pathogenesis of Aeromonas infections involves a combination of virulence factors. Although the role of aeromonas in causing nonintestinal infections has been established, only recently has an epidemiological link associating them with gastroenteritis been suggested. However, the lack of an animal model to reproduce Aeromonas-associated diarrhea and no documented Aeromonas-associated large outbreaks are concerns in firmly establishing the role of aeromonas in gastroenteritis. It needs to be emphasized that in case-control studies all over the world, isolation of Aeromonas species as the only enteropathogen from children with diarrhea clearly has been established. However, it still is not totally clear which vital virulence factors are necessary for aeromonas to instigate gastroenteritis. The role of both cytotoxic and cytotonic enterotoxins in causing gastroenteritis has been elucidated in our laboratory. Studies on the structure-function relationship and the mechanism of action of these toxins would provide a means to intervene in Aeromonas-induced intestinal infections. Most Aeromonas studies are restricted to case-control studies with little emphasis on the molecular and biochemical characterization of virulence factors. Our studies are among the first to characterize enterotoxins of aeromonas on a molecular basis and to study bacterial-host interaction.
Acknowledgments This work was supported by grants R01 AI41611 from the National Institutes of Health and the John Sealy Memorial Endowment Fund.
References [1] Carnahan A.M., Altwegg M.,in: Austin B., Altwegg M., Gosling P.J., Joseph S. (Eds.), Taxonomy. The genus Aeromonas, John Wiley & Sons, NY, 1996, pp. 1–38. [2] Altwegg M., Martinetti L.G., Luthy-Hottenstein J., Rohrbach M., Aeromonas-associated gastroenteritis after consumption of contaminated shrimp, Eur. J. Clin. Microbiol. Infect. Dis. 10 (1991) 44–45. [3] Hanninen M.L., Oivanen P., Hirvela-Koski V., Aeromonas species in fish, fish-eggs, shrimp and freshwater, Inter. J. Food Microbiol. 34 (1997) 17–26. [4] Yamada S., Matsushita S., Dejsirilet S., Kudoh Y., Incidence and clinical symptoms of Aeromonas-associated travellers’ diarrhoea in Toyko, Epidemiol. Infect. 119 (1997) 121–126. [5] Hanninen M.L., Salmi S., Mattila L., Taipalinen R., Siitonen A., Association of Aeromonas spp with travellers’ diarrhoea in Finland, J. Med. Microbiol. 42 (1995) 26–31. [6] Abbott S.L., Seli L.S., CatinoJr M., Hartley M.A., Janda J.M., Misidentification of unusual Aeromonas species as members of the genus Vibrio: a continuing problem, J. Clin. Microbiol. 36 (1998) 1103–1104. [7] Buchanan R.L., Palumbo S.A., Aeromonas hydrophila and Aeromonas sobria as potential food poisoning species: a review, J. Food Safety 7 (1985) 15–29. Microbes and Infection 1999, 1129-1137
Enterotoxins in Aeromonas-associated gastroenteritis
[8] Janda J.M., Abbott S.L., Evolving concepts regarding the genus Aeromonas: an expanding panorama of species, disease presentations, and unanswered questions, Clin. Infect. Dis. 27 (1998) 332–344. [9] Majeed K.N., Egan A.F., Macrae I.C., Production of exotoxins by Aeromonas spp. at 5 °C, J. Appl. Bacteriol. 69 (1990) 332–337. [10] Kirov S.M., Ardestani E.K., Hayward L.J., The growth and expression of virulence factors at refrigeration temperature by Aeromonas strains isolated from foods, Inter. J. Food Microbiol. 20 (1993) 159–168. [11] Grey P.A., Kirov S.M., Adherence to Hep-2 cells and enteropathogenic potential of Aeromonas spp., Epidemiol. Infect. 110 (1993) 279–287. [12] Atkinson H.M., Trust T.J., Hemagglutination properties and adherence ability of Aeromonas hydrophila, Infect. Immun. 27 (1980) 938–946. [13] Merino S., Rubires X., Knochel S., Tomas J.M., Emerging pathogens: Aeromonas spp., Inter. J. Food Microbiol. 28 (1995) 157–168. [14] Mani S., Sadigh M., Adriole V.T., Clinical spectrum of Aeromonas hydrophila infections: report of 11 cases in a community hospital and review, Infect. Dis. Clin. Pract. 4 (1995) 79–86. [15] Kohashi T., Sakai H., Marumo F., Sato C., Aeromonas sobria infection with severe muscle degeneration in a patient with alcoholic liver cirrhosis, Amer. J. Gastroenterol. 90 (1995) 2234–2235. [16] Lin S.H., Shieh S.D., Lin Y.F., Debrauwer E., vanLanduyt H.W., Gordts B., Boelaert J.R., Fatal Aeromonas hydrophila bacteremia in a hemodialysis patient treated with deferoxamine, Amer. J. Kidney Dis. 27 (1996) 733–735. [17] Funada H., Matsuda T., Aeromonas bacteremia in patients with hematologic diseases, Int. Med. 36 (1997) 171–174. [18] Murakami T., Inoue T., Kawakami H., Takahashi H., Yasunaga T., Hypercalcemia associated with Aeromonas hydrophila gastroenteritis, Acta Paediatrica Japonica 37 (1995) 192–195. [19] Deutsch S.F., Wedzina W., Aeromonas sobria-associated leftsided segmental colitis [Review], Amer. J. Gastroenterol. 92 (1997) 2104–2106. [20] Riley P.A., Parasakthi N., Liam C.K., Development of Aeromonas hydrophila bacteremia in a patient recovering from cholera, Clin. Infect. Dis. 22 (1996) 867–868. [21] Do Sola Earle C., Quezel-Guerraz N.M., Rojas L.H., Cantos A.S., Garcia A.J., Severe acute gastroenteritis due to Aeromonas in a patient colectomized for Crohns’ disease, Revista Espanola de Enfermedades Digestivas 89 (1997) 48–50. [22] Martino R., Santamaria A., Pericas R., Sureda A., Brunet S., Acute rhabdomyolysis and myonecrosis complicating Aeromonas bacteremia in neutropenic patients with hematologic malignancies: report of two cases, Haematologica 82 (1997) 692–694. [23] Holston K.V.I., Zandvliet S.E., Rodriguez S., Nguyen H.T., Bodey G.P., Spectrum of Aeromonas and Plesiomonas infections in patients with cancer and AIDS, Experientia 47 (1991) 437–439. [24] Janda J.M., Duffey P.S., Mesophilic aeromonads in human disease: current taxonomy, laboratory identification, and infectious disease spectrum, Rev. Infect. Dis. 10 (1988) 980–997. Microbes and Infection 1999, 1129-1137
Review
[25] Challapalli M., Tess B.R., Cunningham D.G., Chopra A.K., Houston C.W., Aeromonas associated diarrhea in children, Pediatr. Infect. Dis. J. 7 (1988) 693–698. [26] Gracey M., Gastroenteritis in Australian children. Studies on the aetiology of acute diarrhea, Ann. Trop. Pediatr. 8 (1988) 68–75. [27] Verenkar, M., Naik V., Rodrigues S., Singh I., Aeromonas species and Plesiomonas shigelloides in diarrhoea in Goa, Ind. J. Path. Microbiol. 38 (1995) 169–171. [28] Goodwin C.S., Harper W.E.S., Stewart J.K., Gracey M., Burke V., Robinson J., Enterotoxigenic Aeromonas hydrophila and diarrhea in adults, Med. J. Australia 1 (1983) 25–26. [29] Quinn J.P., UHL enteric bacterial disease surveillance, Lab. Hotline 28 (1991) 2. [30] Echeverria P., Pitarangsi C., Eampokalap B., Vibulbandhitkit S., Boonthai P., Rowe B.A., Longitudinal study of the prevalence of bacterial enteric pathogens among adults with diarrhea in Bangkok, Thailand, Diagn. Microbiol. Infect. Dis. 1 (1983) 193–204. [31] Khaitovich A.B., G0al’tseva G.V., Serological diagnosis of diarrhea caused by some representatives of the genera Vibrio and Aeromonas, J. Hyg. Epidemiol. Microbiol. Immunol. 27 (1983) 211–218. [32] Echeverria P., Sack R.B., Blacklow N.R., Bodhidatta P., Rowe B., McFarland A., Prophylactic doxycycline for traveler’s diarrhea in Thailand. Further supportive evidence of A. hydrophila as an enteric pathogen, Am. J. Epid. 120 (1984) 912–921. [33] Borrell, N., Figueras M.J., Guarro J., Phenotypic identification of Aeromonas genomospecies from clinical and environmental sources, Can. J. Microbiol. 44 (1998) 103–108. [34] Albert, M.J., Ansaruzzaman M., Shimada T., Rahman N.A., Bhuiyan S., Nahar F., Qadri M., Islam S., Characterization of Aeromonas trota strains that cross-react with Vibrio cholerae O139 Bengal, J. Clin. Microbiol. 33 (1995) 3119–3123. [35] Kuhn, I., Albert M.J., Ansaruzzaman M., Bhuiyan N.A., Alabi S.A., Islam M.S., Neogi P.K., Huys G., Janssen P., Kersters K., Mollby R., Characterization of Aeromonas spp. isolated from humans with diarrhea, from healthy controls, and from surface water in Bangladesh, J. Clin. Microbiol. 35 (1997) 369–373. [36] Haque, Q.M., Sugiyama A., Iwade Y., Midorikawa Y., Yamauchi T., Diarrheal and environmental isolates of Aeromonas spp. produce a toxin similar to shiga-like toxin 1, Curr. Microbiol. 32 (1996) 239–245. [37] Brown, R.L., Sanderson K., Kirov S.M., Plasmids and Aeromonas virulence, FEMS Immun. Med. Microbiol. 17 (1997) 217–223. [38] Rose J.M., Houston C.W., Coppenhaver D.H., Dixon J.D., Kurosky A., Purification and chemical characterization of a cholera toxin-cross-reactive cytolytic enterotoxin produced by a human isolate of Aeromonas hydrophila, Infect. Immun. 57 (1989) 1165–1169. [39] Asao T., Kinoshita Y., Kozaki S., Uemura T., Sakaguchi G., Purification and some properties of Aeromonas hydrophila hemolysin, Infect. Immun. 46 (1984) 122–127. 1135
Review
[40] Chakraborty T., Huhle B., Bergbauer H., Goebel W., Cloning, expression, and mapping of the Aeromonas hydrophila aerolysin gene determinant in Escherichia coli K-12, J. Bacteriol. 167 (1986) 368–374. [41] Howard S.P., Buckley J.T., Molecular cloning and expression in Escherichia coli of the structural gene for the hemolytic toxin aerolysin from Aeromonas hydrophila, Mol. Gen. Genet. 204 (1986) 289–295. [42] Chopra A.K., Houston C.W., Peterson J.W., Jin G.F., Cloning, expression, and sequence analysis of a cytolytic enterotoxin gene from Aeromonas hydrophila, Can. J. Microbiol. 39 (1993) 513–523. [43] Chakraborty T., Montenegro M.A., Sanyal S.C., Helmuth R., Bulling E., Timmis K.N., Cloning of enterotoxin gene from Aeromonas hydrophila provides conclusive evidence of production of a cytotonic enterotoxin, Infect. Immun. 46 (1984) 435–441. [44] Wong C.Y., Mayrhofer G., Heuzenroeder M.W., Atkinson H.M., Quinn D.M., Flower R.L., Measurement of virulence of aeromonads using a suckling mouse model of infection, FEMS Immun. Med. Microbol. 15 (1996) 233–241. [45] Sanderson K., Ghazali F.M., Kirov S.M., Colonization of streptomycin-treated mice by Aeromonas species, J. Diarrhoeal Dis. Res. 14 (1996) 27–32. [46] Graf J., Symbiosis of Aeromonas veronii biovar sobria and Hirudo medicinalis, the medicinal leech: a novel model for digestive tract associations, Infect. Immun. 67 (1999) 1–7. [47] Annapurna E., Sanyal S.C., Enterotoxicity of Aeromonas hydrophila, J. Med. Microbiol. 10 (1977) 317–323. [48] Wadstrom T., Ljungh A., Wretlind B., Enterotoxin, haemolysin and cytotoxic protein in Aeromonas hydrophila from human infections, Acta. Path. Microbiol. Scand. Sect. B 84 (1976) 112–114. [49] Ljungh A., Kronevi T., Aeromonas hydrophila toxinsintestinal fluid accumulation and mucosal injury in animal models, Toxicon 20 (1982) 397–407. [50] Ljungh A., Eneroth P., Wadstrom T., Cytotonic enterotoxin from Aeromonas hydrophila, Toxicon 20 (1982) 787–794. [51] Dubey R.S., Bhattacharya A.K., Sanyal S.C., Elevation of adenosine 3’5’ cyclic monophosphate level by Aeromonas hydrophila enterotoxin, Ind. J. Med. Res. 74 (1981) 668–674. [52] Dubey R.S., Sanyal S.C., Characterization and neutralization of Aeromonas hydrophila enterotoxin in the rabbit ilealloop model, J. Med. Microbiol. 12 (1979) 347–354. [53] James C., Dibley M., Burke V., Robinson J., Gracey M., Immunological cross-reactivity of enterotoxins of Aeromonas hydrophila and cholera toxin, Clin. Exp. Immunol. 47 (1982) 34–42. [54] Bundell C.S., Watson I.M., Burke V., Gracey M., Protection of rats against cholera toxin and cholera-like enterotoxins by immunization with enteric-coated cholera toxin, Ann. Trop. Paediatrics 6 (1986) 199–204. [55] Potomski J., Burk V., Robinson J., Fumarola D., Miragliotta G., Aeromonas cytotonic enterotoxin cross-reactive with cholera toxin, J. Med. Microbiol. 23 (1987) 179–186. [56] Shultz A.J., McCardell B.A., DNA homology and immunological cross-reactivity between Aeromonas hydrophila cytotonic toxin and cholera toxin, J. Clin. Microbiol. 26 (1988) 57–61. 1136
Chopra and Houston
[57] Chopra A.K., Peterson J.W., Xu X.J., Coppenhaver D.H., Houston C.W., Molecular and biochemical characterization of a heat-labile enterotoxin from Aeromonas hydrophila, Microbial Path. 21 (1996) 357–377. [58] Chopra A.K., Houston C.W., Genaux C.T., Dixon J.D., Kurosky A., Evidence for production of an enterotoxin and cholera toxin cross-reactive factor by Aeromonas hydrophila, J. Clin. Microbiol. 24 (1986) 661–664. [59] Chopra A.K., Houston C.W., Purification and partial characterization of a cytotonic enterotoxin produced by Aeromonas hydrophila, Can. J. Microbiol. 35 (1989) 719–727. [60] Chopra A.K., Vo T.N., Houston C.W., Mechanism of action of a cytotonic enterotoxin produced by Aeromonas hydrophila, FEMS Microbiol. Lett. 91 (1992) 15–19. [61] Pal S., Datta A., Nair G.B., Guhathakurta B., Use of monoclonal antibodies to identify phospholipase C as the enterotoxic factor of the bifuntional hemolysinphospholipase C molecule of Vibrio cholerae O139, Infect. Immun. 66 (1998) 3974–3977. [62] Granum P.E., O’Sullivan K., Tomás J.M., Ørmen Ø., Possible virulence factors of Aeromonas spp. from food and water, FEMS Immun. Med. Microbiol. 21 (1998) 131–137. [63] Chopra A.K., Pham R., Houston C.W., Cloning and expression of putative cytotonic enterotoxin-encoding genes from Aeromonas hydrophila, Gene 139 (1994) 87–91. [64] McCardell B.A., Madden J.M., Kothary M.H., Sathyamoorthy V., Purification and characterization of CHO cell-elongating toxin produced by Aeromonas hydrophila, Microbial Path. 19 (1995) 1–9. [65] Rose J.M., Houston C.W., Kurosky A., Bioactivity and immunological characterization of a cholera toxin-crossreactive cytolytic enterotoxin from Aeromonas hydrophila, Infect. Immun. 57 (1989) 1170–1176. [66] Ferguson M.R., Xu X.J., Houston C.W., Peterson J.W., Coppenhaver D.H., Popov V.L., Chopra A.K., Hyperproduction, purification and mechanism of action of the cytotoxic enterotoxin produced by Aeromonas hydrophila, Infect. Immun. 65 (1997) 4299–4308. [67] Potomski J., Burke V., Watson I., Gracey M., Purification of cytotoxic enterotoxin of Aeromonas sobria by use of monoclonal antibodies, J. Med. Microbiol. 23 (1987) 171–177. [68] Chopra A.K., Houston C.W., Kurosky A., Genetic variation in related cytolytic toxins produced by different species of Aeromonas, FEMS Microbiol. Lett. 78 (1991) 231–237. [69] Ferguson M.R., Xu X.J., Houston C.W., Peterson J.W., Chopra A.K., Amino-acid residues involved in biological functions of the cytolytic enterotoxin from Aeromonas hydrophila, Gene 156 (1995) 79–83. [70] Buckley J.T., Howard S.P., The cytotoxic enterotoxin of Aeromonas hydrophila is aerolysin, Infect. Immun. 67 (1999) 466–467. [71] Bhakdi S., Tranum-Jenson J., Alpha-toxin of Staphylococcus aureus, Microbiol. Rev. 55 (1991) 733–751. [72] Chakroborty T., Schmid A., Notermans S., Benz R., Aerolysin of Aeromonas sobria: evidence for formation of ionpermeable channels and comparison with alpha-toxin of Staphylococcus aureus, Infect. Immun. 58 (1990) 2127–2132. [73] Parker M.W., vander Goot F.G., Buckley J.T., Aerolysunthe ins and outs of a channel forming protein, Mol. Microbiol. 19 (1996) 205–212. Microbes and Infection 1999, 1129-1137
Enterotoxins in Aeromonas-associated gastroenteritis
[74] Blomqvist L., Sjogren A., Production and characterization of monoclonal antibodies against S. aureus alpha toxin, Toxicon 26 (1988) 265–273. [75] Xu X.J., Ferguson M.R., Popov V.L., Houston C.W., Peterson J.W., Chopra A.K., Role of a cytotoxic enterotoxin in Aeromonas-mediated infections: development of transposon and isogenic mutants, Infect. Immun. 66 (1998) 3501–3509. [76] Chakraborty T., Huhle B., Hof H., Bergbauer H., Goebel W., Marker exchange mutagenesis of the aerolysin determinant in Aeromonas hydrophila demonstrates the role of aerolysin in A. hydrophila associated systemic infections, Infect. Immun. 55 (1987) 2274–2280. [77] Nelson K.L., Raja S.M., Buckley J.T., The glycosylphosphatidylinositol-anchored surfce glycoprotein Thy-1 is a receptor for the channel-forming toxin aerolysin, J. Biol. Chem. 272 (1997) 12170–12174. [78] Diep D.B., Nelson K.L., Lawrence T.S., Sellman B.R., Tweten R.K., Buckley J.T., Expression and properties of an aerolysin-Clostridium septicum alpha toxin hybrid protein, Mol. Microbiol. 31 (1999) 785–794. [79] Jin G.F., Houston C.W., The effect of Aeromonas hydrophila enterotoxins on the phagocytic function of mouse phagocytes, Dig. Dis. Sci. 37 (1992) 1697–1703. [80] Jin G.F., Chopra A.K., Houston C.W., Stimulation of neutrophil leukocyte chemotaxis by a cloned cytolytic enterotoxin of Aeromonas hydrophila, FEMS Microbiol. Lett. 98 (1992) 285–290. [81] Krause K.H., Fivaz M., Monod A., Van Der Goot F.G.,
Microbes and Infection 1999, 1129-1137
Review
[82]
[83]
[84] [85]
[86]
[87]
Aerolysin induces G-protein activation and Ca2+ release from intracellular stores in human granulocytes, J. Biol. Chem. 273 (1998) 18122–18129. Merino S., Rubires X., Aguilar A., Alberti S., HernandezAlles S., Benedi V.J., Tomas J.M., Mesophilic Aeromonas sp. Serogroup 0:11 resistance to complement-mediated killing, Infect. Immun. 64 (1996) 5302–5309. Merino S., Nogueras M.M., Aguilar A., Rubires X., Alberti S., Benedi V.J., Tomas J.M., Activation of the complement classical pathway (C1q binding) by mesophilic Aeromonas hydrophila outer membrane protein, Infect. Immun. 66 (1998) 3825–3831. Pemberton J.M., Kidd S.P., Schmidt R., Secreted enzymes of Aeromonas, FEMS Microbiol. Lett. 152 (1997) 1–10. Ascencio F., Martinez-Arias W., Romero M.J., Wadstrom T., Analysis of the interaction of Aeromonas caviae, A, hydrophila and A, sobira with mucins, FEMS Immun. Med. Microbiol. 20 (1998) 219–229. Barnett T.C., Kirov S.M., Strom M.S., Sanderson K., Aeromonas spp, possess at least two distinct type IV pilus families, Microbial Path. 23 (1997) 241–247. Swift S., Karlyshev A.V., Fish L., Durant E.L., Winson M.K., Chhabra S.R., William S.P., MacIntyre S., Stewart G.S., Quorum sensing in Aeromonas hydrophila and Aeromonas salmonicida: identification of the LuxRI homologs AhyRI and AsaRI and their cognate N-acylhomoserine lactone signal molecules, J. Bacteriol. 178 (1997) 5271–5281.
1137