Enteroaggregative Escherichia coli Virulence Factors Are Found To Be Associated with Infantile Diarrhea in Brazil

Transcription

1 JOURNAL OF CLINICAL MICROBIOLOGY, Mar. 2004, p Vol. 42, No /04/$ DOI: /JCM Copyright 2004, American Society for Microbiology. All Rights Reserved. Enteroaggregative Escherichia coli Virulence Factors Are Found To Be Associated with Infantile Diarrhea in Brazil Andresa Zamboni, 1 Sandra H. Fabbricotti, 1 Ulysses Fagundes-Neto, 2 and Isabel C. A. Scaletsky 1 * Departamento de Microbiologia, Imunologia e Parasitologia and Departamento de Pediatria, 1 and Universidade Federal de São Paulo, 2 Escola Paulista de Medicina, São Paulo, SP, Brazil Received 30 June 2003/Returned for modification 7 November 2003/Accepted 1 December 2003 We have previously shown that enteroaggregative Escherichia coli (EAEC) is an important pathogen among Brazilian infants. Most EAEC strains harbor a plasmid (paa) from which a DNA fragment has been used as a probe (EAEC probe). To better understand the characteristics of EAEC in Brazil, 109 strains carrying and lacking the EAEC probe sequence were tested for the presence of paa plasmid-borne and chromosomal factors. Common virulence factors of probe-positive and probe-negative isolates included the presence of the Pet, EAST-1, Shf, Irp2, ShET1/Pic, and Hly virulence markers. The presence of AggR or one other virulence factor (AAF/I, AAF/II, AAF/III, or Aap) was predominantly identified only in probe-positive strains. In EAEC probepositive strains, the virulence marker Aap was found significantly more frequently (P 0.023) in isolates from children with diarrhea (22%) than in isolates from controls (3%). EAST-1 and Shf were the markers most frequently detected (61%) in EAEC probe-negative strains and were found to be significantly associated with diarrhea (P and P 0.020, respectively). Furthermore, our data suggest that AggR can be used as an important genetic marker for EAEC probe-positive strains. Enteroaggregative Escherichia coli (EAEC) is increasingly recognized as a cause of diarrhea worldwide (18). EAEC is defined by its characteristic stacked brick aggregative adherence (AA) pattern of adherence to HEp-2 cell (17). Most EAEC strains harbor a 60- to 65-MDa virulence plasmid (paa). A 1-kb fragment of paa, referred to as the EAEC probe or CVD432 (1), has been widely used for epidemiological studies (7, 19, 25, 29). The paa plasmid also encodes AA fimbriae (AAF) I, II, and III (2, 4, 15); the transcriptional activator AggR (16); enteroaggregative heat-stable enterotoxin 1 (EAST-1) (23); a 104-kDa cytotoxin designated Pet (8); the cryptic secreted protein Shf (5); and a novel antiaggregation protein (dispersin) encoded by the aap gene (formerly known as aspu) (27). In addition to the paa plasmid, some EAEC strains express putative virulence factors that are encoded on the chromosome, including a 116-kDa secreted mucinase (Pic) (12), yersiniabactin (26), and the E. coli -hemolysin (30). Shigella enterotoxin 1 (ShET1) is encoded by the antisense strand of the pic gene (10, 21). However, none of these factors is consistently found in all EAEC isolates, as determined by hybridization studies. To find a specific virulence marker for the detection of EAEC in epidemiological and clinical studies, we tested 109 strains, isolated from Brazilian children in a previously described study (25), for their abilities to hybridize to eight paa plasmid-derived and three chromosomal gene probes. conducted in different regions of Brazil from 1997 to 1999 (25). The children were admitted to public hospitals in the following cities (states) for treatment: São Paulo (São Paulo), Joinville (Santa Catarina), Natal (Rio Grande do Norte), Goiania (Goiás), and São Luiz (Maranhão). That study used rectal swab specimens from 338 children with acute diarrhea (case patients) and 322 asymptomatic children, matched for age with the case patients, without any gastrointestinal symptoms for at least 30 days prior to inclusion in the study (controls). In the study mentioned above, each fecal specimen was examined by standard methods for the presence of Shigella spp., Salmonella spp., Giardia lamblia, Yersinia enterocolitica, Campylobacter spp., Cryptosporidium spp., and rotavirus. Four separate lactose-fermenting colonies and two non-lactose-fermenting colonies of each distinct morphological type were cultivated in commercial test systems (PROBAC do Brasil, São Paulo, Brazil), for biochemical confirmation of the species or genus. All E. coli isolates were tested with specific DNA probes designed to detect enterotoxigenic E. coli (heat-labile toxin- and heat-stable toxin-specific probes), enteroinvasive E. coli (Inv-specific probe), Shiga-toxinproducing E. coli (Shiga toxin 1- and Shiga toxin 2-specific probes), EAEC (EAEC-specific probe), diffusely adhering E. coli (daac- and AIDA-I-specific probes), and EPEC (eae- and EAF-specific probes). Serotyping. Identification of the somatic (O) antigens of the strains was done by standard agglutination methods (9) with specific O1 to O175 antisera acquired commercially (Universidad de Santiago de Compostela, Santiago de Compostela, Spain). Twelve strains were tested in the Enteric Section of the Instituto Adolfo Lutz (São Paulo, Brazil) by using H antisera prepared with type strains. DNA hybridization. Hybridization assays were performed with specific DNA fragments amplified from prototype strains or fragments derived from cloned DNA probes (Table 1). These fragments were labeled with [ - 32 P]dCTP, and colony hybridization assays were performed as described previously (25). Statistical analysis. Data for children with diarrhea and for the controls were compared by a two-tailed chi-square or Fisher s exact test. MATERIALS AND METHODS Bacterial strains. The strains examined in this work were isolated during a study of the epidemiology of acute diarrhea in children less than 2 years old * Corresponding author. Mailing address: Departamento de Microbiologia, Imunologia e Parasitologia, Universidade Federal de São Paulo, Escola Paulista de Medicina, Rua Botucatu, 862, São Paulo, SP, Brazil. Phone: Fax: E- mail: scaletsky@ecb.epm.br. RESULTS A total of 338 children with diarrhea and 322 matched controls without diarrhea were studied. We identified potential diarrheagenic E. coli strains by HEp-2 cell adherence assays and by their abilities to hybridize with specific DNA probes. Of the 660 fecal specimens analyzed, 109 (67 from patients and 42 from controls) were identified as EAEC strains by the AA pattern. Eight-one (74%) of the 109 EAEC strains hybridized 1058

2 VOL. 42, 2004 EAEC VIRULENCE FACTORS AND INFANTILE DIARRHEA 1059 TABLE 1. Probes hybridizing to colony blots of EAEC isolates Target gene Properties of target Gene location Fragment (reference) agga AAF/I subunit Plasmid 450-bp amplified fragment (24) aafa AAF/II subunit Plasmid 550-bp amplified fragment (4) agg3a AAF/III subunit Plasmid 462-bp amplified fragment (2) asta EAST-1 toxin Plasmid 111-bp amplified fragment (31) pet 104-kDa cytotoxin Plasmid 832-bp fragment of pcefn1 (8) aap 10-kDa secreted protein Plasmid 232-bp amplified fragment (5) aggr Transcription of AAFs Plasmid 308-bp amplified fragment (5) shf Cryptic ORF a Plasmid 613-bp amplified fragment (5) shet1/pic Shigella enterotoxin 1/mucinase Chromosome 1,100-bp amplified fragment (19) irp2 Yersiniabactin Chromosome 264-bp amplified fragment (5) hly -Hemolysin Chromosome 6,400-bp fragment of psf4000 (30) a ORF, open reading frame. with the EAEC-specific probe. Fourteen EAEC strains hybridized with the daac-specific probe for diffusely adherent E. coli and none of the 109 strains reacted with the probes for enteropathogenic E. coli (EPEC). The number of E. coli isolates with these characteristics in each specimen varied from one to three. For the present study, only one isolate was selected from each of these 109 specimens. The isolates in only 14 of the 109 specimens (8 from patients and 6 from controls) were EPEC, diffusely adherent E. coli, Shigella spp., or rotavirus. In our study, although EAEC was not significantly associated with diarrhea, EAEC isolates were recovered more frequently from children with diarrhea than from children without diarrhea. In total, 81 EAEC probe-positive and 28 EAEC probenegative strains were characterized by DNA hybridization to detect the genes for the proposed EAEC virulence factors. The frequencies of strains with different genetic profiles regarding their association with diarrhea were further analyzed. Characteristics of EAEC probe-positive strains. Of the 81 EAEC probe-positive strains, 73 (90%) strains belonged to 34 distinct serogroups, of which O44 and O49 were the most frequent (Tables 2 and 3). The virulence factor aggr was the single most common gene marker identified and was found in 43 strains (88%) from patients and 31 strains (97%) from controls. All strains hybridized with three or more paa-derived probes, with three being the modal value. Strains hybridizing with at least one of the probes for AAF/II, Aap, and Pet were more commonly recovered from children with diarrhea than from controls. The asta, shf, shet1/pic, and hly virulence markers were detected at similar frequencies from patients and controls. The irp2 gene was found more frequently in controls than in patients. When we investigated the strains for the presence of combinations of genes, we found that 48 EAEC strains (59%) possessed aggr and irp2, followed by 36 EAEC strains (44%) possessing the combination aggr and shf; 3 strains were shf positive, and 1 strain was shf and irp2 positive but aggr negative. Characteristics of EAEC probe-negative strains. Twentyone (75%) EAEC probe-negative strains belonged to 16 distinct serogroups (Table 4). The majority of strains in this group carried two or more of the genes for which assays were conducted, with two being the modal value; but two strains isolated from a control did not test positive for any of the factors. The AAF, aggr, and aap genes were present in only a minority of strains. The asta, pet, shf, ShET1/pic, irp2, and hly genes were found more frequently in the patients than in the controls. The combination asta and shf was found in 16 strains (57%), followed by 7 strains (25%) possessing the combination asta, shf, and irp2. Distribution of virulence factors in children with and without diarrhea. The distribution of EAEC probe-positive and probe-negative strains with the different virulence factors in children with and without diarrhea is presented in Table 5. The common virulence factors found in EAEC probe-positive and probe-negative isolates included the Pet, EAST-1, Irp2, Shf, ShET1/Pic, and Hly virulence markers. AAFs, AggR, and Aap were predominantly identified only in probe-positive strains. The AggR virulence factor was the most common gene marker identified in EAEC probe-positive strains and was found in 74 strains (91%), followed by Irp2, which was found in 49 strains (60%). In this group of strains, the Aap virulence marker was found significantly more frequently (P 0.023) in isolates from children with diarrhea (22%) than in controls (3%). EAST-1 and Shf were the most frequently detected markers (61%) in EAEC probe-negative strains and were found to be significantly associated with diarrhea (P and P 0.020, respectively). DISCUSSION EAEC has been associated with endemic pediatric diarrhea worldwide. However, the pathogenic mechanisms of EAEC infection are not fully understood. Moreover, there appears to be a significant heterogeneity of virulence markers among EAEC isolates (19). By using EAEC strains isolated in a casecontrol study, we assessed the prevalence of putative virulence factors, such as AFF/I, AAF/II, AAF/III, AggR, Aap, EAST-1, Pet, Shf, ShET1/Pic, Irp2, and -Hly, in an attempt to identify their roles as enteric virulence factors. In our study, a correlation between the presence of a specific EAEC marker and diarrhea was found for both EAEC probepositive and probe-negative strains. Except for a study conducted in Nigeria (19), in which the presence of genes related to AAF/II was associated with diarrhea in children, none of the paa markers has been statistically associated with diarrhea, until now. Recently, Piva et al. (20) detected pic sequences significantly more frequently in EAEC probe-positive strains from children with diarrhea than from controls. Among the EAEC probe-positive strains, we detected the aap gene much more frequently in strains from patients than

3 1060 ZAMBONI ET AL. J. CLIN. MICROBIOL. TABLE 2. Distributions of plasmid and chromosomal genes among 49 EAEC probe-positive isolates from patients Strain Serotype or serogroup b Probe result Plasmid gene Chromosomal gene AAF/I AAF/II AAF/III aggr aap asta pet shf shet1/pic irp2 hly RN785-1 O1 HSP161-1 O3:H2 MA817-1 O5 MA555-7 O17:NM RN781-7 O17 MA687-3 O18:NM MA563-2 O21 RN463-2 O38 SC725-1 a O40 HDV139-1 a O42:H28 HDV147-6 O42 HSP41-1 O44:H18 MA233-1 O44 RN459-1 a O44:NM SC717-4 O44 SP21-2 O48:H18 HDV73-1 O49:H18 HDV111-1 O49:H33 SC289-1 O49 SC327-5 O49 RN349-5 O49 SC389-1 O49 MA821-3 O54 MA253-1 O55 RN513-3 a O66 MA655-5 O66 HSP59-1 O78:H2 MA565-1 O83 MA697-7 a O86 RN501-7 O90 MA537-1 O90 MA703-1 O90 HDV O92:H2 HSP O117 RN407-1 O117 MA495-2 O117 MA567-4 O125 MA659-2 O125:NM MA823-6 O125 RN735-1 O126:NM HSP713-1 O126 HDV83-1 O153:H26 RN747-9 O155 MA545-1 O170 MA807-5 O170 RN733-3 O174:NM RN611-3 ONT MA693-5 ONT MA831-5 ONT a daac positive. b NM, nonmotile; NT, nontypeable. from controls (P 0.023). Recently, Sheikh et al. (27) showed that this gene encodes an antiaggregation protein named dispersin that is expressed in vivo, highly immunogenic, and present in most EAEC probe-positive strains. Those investigators also proposed that this protein might be representative of a functional class of colonization factors. Among the EAEC probe-negative strains, asta and shf were found to be significantly associated with diarrhea (P and P 0.020, respectively). In the present study, 61% of EAEC probe-negative strains from patients harbored the asta gene. A similar asta gene frequency has previously been reported among EAEC probe-negative strains (7), but in that study, EAST-1 was detected more frequently in strains from controls than in strains from patients. Okeke et al. (19) and Rich et al. (22) detected EAST-1 in 23 and 45% of EAEC strains, respectively, but no correlation between the presence

4 VOL. 42, 2004 EAEC VIRULENCE FACTORS AND INFANTILE DIARRHEA 1061 TABLE 3. Distributions of plasmid and chromosomal genes among 32 EAEC probe-positive isolates from controls Strain Serotype or serogroup b Probe result Plasmid gene Chromosomal gene AAF/I AAF/II AAF/III aggr aap asta pet shf shet1/pic irp2 hly SC272-1 O15:NM RN630-1 O15 MA524-1 O17 RN416-3 O17:NM MA496-2 O36 RN572-1 O41 RN612-1 O48 MA670-3 O48:NM MA228-4 O49 MA232-2 O49:NM MA330-4 O49 RN620-1 O49 RN782-1 a O62 RN514-3 O66 RN776-8 O73 MA498-1 O90 RN766-2 O90 RN220-1 O92 MA252-5 O104 RN208-9 a O117 RN190-2 O125:H21 RN196-3 a O125:H21 MA660-5 O125 HDV262-8 O126 MA566-1 O150 RN770-1 O150:NM MA478-2 O153 HSP166-5 ONT:H18 RN460-3 ONT:NM MA486-1 ONT:NM RN632-3 ONT RN784-3 ONT a daac positive. b NM, nonmotile; NT, nontypeable. of EAST-1 and diarrhea was found. Recently, Piva et al. (20) detected EAST-1 in 73% of EAEC probe-positive strains. The asta gene has been detected not only in EAEC strains but also in EPEC, enterotoxigenic E. coli, Shiga-toxin producing E. coli, and enteroinvasive E. coli strains (6, 28, 32). However, its significance in the pathogenesis of E. coli infection remains unclear. We also detected shf in 61% of EAEC probe-negative strains. This gene encodes a cryptic open reading frame that shares 25% amino acid identity with the Staphylococcus epidermidis IcaB protein, which is implicated in intercellular adhesion (11). This gene has been found in many EAEC probe-positive strains and a few EAEC probe-negative strains (5, 7). The majority of the EAEC isolates in this study reacted with one of the antisera used. Other workers have shown that many EAEC strains are O nontypeable (13, 28). The diversity of serotypes seen in this study and the apparent lack of a correlation between a particular serotype and the presence of a particular virulence marker suggest that several clones of EAEC from children with diarrhea may be responsible for diarrhea in different cities in Brazil. It was also interesting that the strains belonging to serogroups O42 and O44 were detected only in children with diarrhea. EAEC isolates of diverse origins have been found to be of these serogroups, and thus, these serogroups could serve as markers for pathogenic EAEC strains. Few studies have evaluated the prevalence of EAEC markers in EAEC probe-positive and probe-negative strains isolated from subjects in case-control studies. The present study provides evidence that some of the recognized paa-encoded factors of EAEC (AAF/I, AAF/II, AAF/III, Aap, and AggR) are prevalent in EAEC probe-positive strains and that others (Pet, EAST-1, Irp2, ShET1/Pic and Hly) are found in both groups of EAEC strains. In a recent examination of strains in a case-control study, Elias et al. (7) found that EAEC probepositive strains isolated in São Paulo exhibited a wide range of virulence characteristics compared to the number exhibited by EAEC probe-negative strains. In that study, they suggested that the EAEC probe-positive strains might have a higher pathogenic potential or, alternatively, that EAEC probenegative strains may harbor virulence factors that have not yet been described. In contrast, Bouzari et al. (3) found that EAEC probe-negative strains in Iran share virulence factors with EAEC probe-positive isolates. Our findings, along with those of Elias et al. (7), suggest the possibility that EAEC probe-positive strains comprise a distinct subcategory of EAEC that could be called typical EAEC.

5 1062 ZAMBONI ET AL. J. CLIN. MICROBIOL. TABLE 4. Distributions of plasmid and chromosomal genes among 28 EAEC probe-negative isolates from patients and controls Strain source and strain Serotype or serogroup b Probe result Plasmid gene Chromosomal gene AAF/I AAF/II AAF/III aggr aap asta pet shf shet1/pic irp2 hly Patients RN731-5 a O3 SC385-5 O42 HSP53-1 O42 SC319-3 O49:NM RN779-4 O54 SC289-3 a O86:NM SC323-4 O89:NM MA701-1 O98 RN453-1 O117 RN741-5 O126:NM RN153-1 O126 RN771-5 O162:NM RN749-9 O162 HDV ONT RN767-7 a ONT RN751-9 ONT SC779-1 ONT SC373-4 a ONT Controls SC358-1 O6:NM SC378-3 a O28 SC730-1 O49 RN406-5 O52 MA558-7 O66:NM RN772-3 O89 RN502-1 a O141 RN582-3 O174:NM SC796-1 ONT SC798-4 ONT a daac positive. b NM, nonmotile; NT, nontypeable. Recently, Jiang et al. (14) suggested that the agga and aggr genes alone or in combination with other virulence factors (aafa or aap) might be used to identify pathogenic EAEC strains. In the present study, both aggr and agga were found at similar frequencies in patients and controls. TABLE 5. Distributions of virulence-related markers among EAEC isolates In conclusion, our data reinforce the heterogeneity of EAEC strains with regard to both paa plasmid-borne and chromosomal factors. However, we have apparently identified two virulent subpopulations of EAEC: EAEC probe-positive strains carrying the aap gene and EAEC probe-negative strains No. (%) of isolates Virulence factor Cases (n 49) EAEC probe positive Controls (n 32) Total (n 81) Cases (n 18) EAEC probe negative Controls (n 10) Total (n 28) AAF/I 9 (18) 6 (18) 15 (18) AAF/II 6 (12) 3 (9) 9 (11) 1 (5) 0 1 (3.6) AAF/III 18 (37) 12 (37) 30 (37) AggR 43 (88) 31 (97) 74 (91) 3 (17) 1 (10) 4 (14) Aap 11 (22) 1 (3) a 12 (15) 1 (5) 0 1 (3.6) EAST-1 13 (26) 8 (25) 21 (26) 15 (83) 2 (20) b 17 (61) Pet 20 (41) 11 (34) 31 (38) 7 (39) 3 (30) 10 (36) Shf 24 (49) 16 (50) 40 (49) 14 (78) 3 (30) c 17 (61) ShET1/Pic 8 (16) 5 (16) 13 (16) 5 (28) 1 (10) 6 (21) Trp2 28 (57) 21 (66) 49 (60) 8 (44) 2 (20) 10 (36) Hly 9 (18) 5 (16) 14 (17) 2 (11) 1 (10) 3 (11) a Significant at P 0.023, as determined by 2 test. b Significant at P 0.003, as determined by 2 test. c Significant at P 0.020, as determined by 2 test.

6 VOL. 42, 2004 EAEC VIRULENCE FACTORS AND INFANTILE DIARRHEA 1063 harboring the asta and the shf genes. Furthermore, our data suggest that aggr can be used as an important genetic marker for EAEC probe-positive strains. ACKNOWLEDGMENTS We thank Chantal Le Bouguénec for the gift of the AAF/III reference strain. This work was funded by the Conselho Nacional de Desenvolvimento Científico e Tecnológico and the Fundação de Amparo à Pesquisa do Estado de São Paulo. REFERENCES 1. Baudry, B., S. J. Savarino, P. Vial, J. B. Kaper, and M. M. Levine A sensitive and specific DNA probe to identify enteroaggregative Escherichia coli, a recently discovered diarrheal pathogen. J. Infect. Dis. 161: Bernier, C., P. Gounon, and C. Le Bouguénec Identification of an aggregative adhesion fimbria (AAF) type III-encoding operon in enteroaggregative Escherichia coli as a sensitive probe for detecting the AAF-encoding operon family. Infect. Immun. 70: Bouzari, S., A. Jafari, A. Azizi, M. Oloomi, and J. P. Nataro Characterization of enteroaggregative Escherichia coli isolates from Iranian children. Am. J. Trop. Med. Hyg. 65: Czeczulin, J. R., S. Balepur, S. Hicks, A. Philips, R. Hall, M. H. Kothary, F. Navarro-Garcia, and J. P. Nataro Aggregative adherence fimbria II, a second fimbrial antigen mediating aggregative adherence in enteroaggregative Escherichia coli. Infect. Immun. 65: Czeczulin, J. R., T. S. Whittam, I. R. Henderson, and J. P. Nataro Phylogenetic analysis of virulence genes in enteroaggregative and diffusely adherent Escherichia coli. Infect. Immun. 67: De Sousa, C. P., and J. D. Dubreuil Distribution and expression of the asta gene (EAST1 toxin) in Escherichia coli and Salmonella. Int. J. Med. Microbiol. 291: Elias, W. P., A. P. Uber, S. K. Tomita, L. R. Trabulsi, and T. A. T. Gomes Combinations of putative virulence markers in typical and variant enteroaggregative Escherichia coli strains from children with and without diarrhea. Epidemiol. Infect. 129: Eslava, C., F. Navarro-Garcia, J. R. Czeczulin, I. R. Henderson, A. Cravioto, and J. P. Nataro Pet, an autotransporter enterotoxin from enteroaggregative Escherichia coli. Infect. Immun. 66: Ewing, W. H Genus Escherichia coli, p In W. H. Ewing (ed.), Edwards and Ewing s identification of Enterobacteriaceae, 4th ed. Elsevier Science, New York, N.Y. 10. Fasano, A., F. Noriega, D. R. Mandeval, S. Chanasongcram, R. Russel, S. Guandalini, and M. M. Levine Shigella enterotoxin 1: an enterotoxin of Shigella flexneri 2a active in rabbit small intestine in vivo and in vitro. J. Clin. Investig. 95: Heilmann, C., O. Schweitzer, C. Gerke, N. Vanittanakom, D. Mack, and F. Goetz Molecular basis of intercellular adhesion in the biofilm-forming Staphylococcus epidermidis. Mol. Microbiol. 20: Henderson, I. R., J. Czeczulin, C. Eslava, F. Noriega, and J. P. Nataro Characterization of Pic, a secreted protease of Shigella flexneri and enteroaggregative Escherichia coli. Infect. Immun. 67: Itoh, Y., I. Nagano, M. Kunishima, and T. Ezaki Laboratory investigation of enteroaggregative Escherichia coli O untypeable:h10 associated with a massive outbreak of gastrointestinal illness. J. Clin. Microbiol. 35: Jiang, Z, D. Greenberg, J. P. Nataro, R. Steffen, and H. L. DuPont Rate of occurrence and pathogenic effect of enteroaggregative Escherichia coli virulence factors in international travelers. J. Clin. Microbiol. 40: Nataro, J. P., Y. Deng, D. R. Maneval, A. L. German, W. C. Martin, and M. M. Levine Aggregative adherence fimbriae I of enteroaggregative Escherichia coli mediate adherence to HEp-2 cells and hemagglutination of human erythrocytes. Infect. Immun. 60: Nataro, J. P., Y. Deng, Y. Deng, and K. Walker AggR, a transcriptional activator of aggregative adherence factor I expression. J. Bacteriol. 176: Nataro, J. P., J. B. Kaper, R. Robins-Browne, V. Prado, P. Vial, and M. M. Levine Patterns of adherence of diarrheagenic Escherichia coli to HEp-2 cells. Pediatr. Infect. Dis. J. 6: Nataro, J. P., and J. B. Kaper Diarrheagenic Escherichia coli. Clin. Microbiol. Rev. 11: Okeke, I. N., A. Lamikanra, J. Czeczulin, F. Dubovsky, J. B. Kaper, and J. P. Nataro Heterogeneous virulence of enteroaggregative Escherichia coli strains isolated from children in southwest Nigeria. J. Infect. Dis. 181: Piva, I. C., A. L. Pereira, L. R. Ferraz, R. S. N. Silva, A. C. Vieira, J. E. Blanco, M. Blanco, J. Blanco, and L. G. Giugliano Virulence markers of enteroaggregative Escherichia coli isolated from children and adults with diarrhea in Brasilia, Brazil. Infect. Immun. 41: Rajakumar, K., C. Sasakawa, and B. Adler Use of a novel approach, termed island probing, identifies the Shigella flexneri she pathogenicity island which encodes a homology of the immunoglobulin A protease-like family of proteins. Infect. Immun. 65: Rich, C., S. Favre-Bonte, F. Sapena, B. Joly, and C. Forestier Characterization of enteroaggregative Escherichia coli isolates. FEMS Microbiol. Lett. 173: Savarino, S. J., A. Fasano, J. Watson, B. M. Martin, M. M. Levine, S. Guandalini, and P. Guerry Enteroaggregative Escherichia coli heatstable enterotoxin 1 represents another family of E. coli heat-stable toxin. Proc. Natl. Acad. Sci. USA 90: Savarino, S. J., P. Fox, Y. Deng, and J. P. Nataro Identification and characterization of a gene cluster mediating enteroaggregative Escherichia coli aggregative adherence fimbria I. J. Bacteriol. 176: Scaletsky, I. C. A., S. H. Fabbricotti, K. R. Aranda, M. B. Morais, and U. Fagundes-Neto Comparison of DNA hybridization and PCR assays for detection of putative pathogenic enteroadherent Escherichia coli. J. Clin. Microbiol. 40: Schubert, S., A. Rakin, H. Karch, E. Carniel, and J. Heeseman Prevalence of the high-pathogenicity island of Yersinia species among Escherichia coli strains that are pathogenic to humans. Infect. Immun. 66: Sheikh, J., J. R. Czezulin, S. Harrington, S. Hicks, I. R. Henderson, C. Le Bouguenec, P. Gounon, A. Phillips, and J. P. Nataro A novel dispersin protein in enteroaggregative Escherichia coli. J. Clin. Investig. 110: Vial, P. A., R. Robins-Browse, H. Lior, V. Prado, J. B. Kaper, J. P. Nataro, D. Maneval, A. Elsayed, and M. M. Levine Characterization of enteroadherent-aggregative Escherichia coli, a putative agent of diarrheal disease. J. Infect. Dis. 158: Wanke, C. A., J. B. Schorling, L. J. Barret, M. A. De Souza, and R. L. Guerrant Potential role of adherence traits of Escherichia coli in persistent diarrhea in an urban Brazilian slum. Pediatr. Infect. Dis. J. 10: Welch, R. A., R. Hell, and S. Falkow Molecular cloning and physical characterization of a chromosomal hemolysin from Escherichia coli. Infect. Immun. 42: Yamamoto, T., N. Wakisaka, F. Sato, and A. Kato Comparison of the nucleotide sequence of enteroaggregative Escherichia coli heat-stable enterotoxin 1 genes among diarrhea-associated Escherichia coli. FEMS Microbiol. Lett. 147: Yatsuyanagi, J., S. Saito, Y. Miyajima, K. Amano, and K. Enomoto Characterization of atypical enteropathogenic Escherichia coli strains harboring the asta gene that were associated with a waterborne outbreak of diarrhea in Japan. J. Clin. Microbiol. 41: