Microbiol. Immunol., 42(12), 845-849,
Relationship of Pathogenic Tsuyoshi
Sunabe,
1998
between O-Serogroup and Presence Factor Genes in Escherichia coli and Yasuko
Honma *
Department of Bacteriology, Faculty of Medicine, University of the Ryukyus,Okinawa 903-0215, Japan ReceivedJuly 27, 1998;in revisedform, September22, 1998.AcceptedSeptember24, 1998 Abstract: A total of 383 isolates of serogroup-based enteropathogenic and enteroinvasive Escherichia coli (310 strains of EPEC and 73 strains of EIEC) were examined for the presence of corresponding pathogenic genes. The serogroup-based EPEC consisted of 232 strains isolated from diarrhea patients and of 78 strains from healthy carriers. The gene encoding intimin, eaeA, was detected in 42 of the 232 EPEC strains from patients (18.1%) and 9 of the 78 strains from carriers (11.5%). The difference was not significant. The bfp gene on the EAF plasmid was detected in 7 of the 42 eaeA-positive EPEC strains from patients but was not detected in the 9 strains from carriers. In serogroup-based EIEC, a chromosomal ipaH gene encoding one of the invasive plasmid antigens was detected in 4 of the 60 strains from patients (6%) but not in the 13 strains from carriers. The 4 ipaH-positive strains possessed the invasive plasmid. These results suggested that the serogroup-based diagnosis of EPEC and EIEC is not sufficient for identifying strains carrying the eaeA or ipaH gene. Key words: EPEC, EIEC, O-serogroup
Diarrheal diseases are one of the leading causes of death for young children in tropical countries. Diarrhea-
identification of EIEC. However, screening of EPEC and EIEC by the slide agglutination test using traditionally
genic Escherichia coli is the main causative agent of diarrheal disease. Diarrheagenic E. coli are classified into at least 5 categories; i.e., enteropathogenic E. coli (EPEC), enterotoxigenic E. coli (ETEC), enteroinvasive E. coli
prepared anti-sera for the corresponding O antigens is usually carried out in clinical laboratories, and on many occasions, serogroup-based EPEC is regarded as real EPEC by clinicians. Although it is said that the presence of the EAF plasmid and/or eaeA corresponds well to the presence of the Class-I O antigen, the reverse has not been confirmed. It is said that 70 to 80% of E. coli strains possessing eaeA belong to the Class-I serogroup
(EIEC), enterohemorrhagic E. coli (EHEC), and enteroaggregative E. coli (EAggEC) (7). The identification of ETEC and EHEC is routinely made by detection of the corresponding toxins. EAggEC is not usually handled in clinical laboratories, because a method to identify the organism has not yet been established except for testing the characteristic pattern of adherence to cultured HEp2 cells, which is too laborious to carry out in clinical laboratories. Thus, little is known about the pathogenesis, epidemiology, and serotypes of EAggEC strains (16). Because there is a three-stage model of EPEC patho-
(7), but it is not known what percentage of E. coli with Class-I antigen have the eaeA gene. The reliability of diagnosis with a sero-agglutination test urgently needs to be addressed, because it is the final diagnosis in many clinical laboratories. In this communication, we determined the carrier rate of the pathogenic genes in serogroup-based EPEC and EIEC isolated from diarrhea patients and healthy carriers.
genesis including initial adherence via bundle-forming pili (BFP) and intimate adherence, EPEC can be identified by detecting the EAF plasmid-encoded BFP (14) or the chromosomal eaeA gene encoding a 94 kDa outer membrane protein (intimin) (10). A Sereny test (5) or examination of pathogenic genes, such as the gene for the invasive plasmid antigen (ipa) (13) is required for the
Materials
and Methods
Bacterial strains. A total of 383 E. coli strains agglutinated with a commercially purchased E. coli diagnosAbbreviations:
*
Address
correspondence
to Dr. Yasuko Honma,
Department
of
Bacteriology,Facultyof Medicine, University of the Ryukyus, 207 Uehara, Nishihara, Okinawa 903-0215, Japan. Fax: + 81-98895-2951. E-mail:
[email protected]
pili; EAggEC,
enteroag-
E. coli; EIEC, E. coli; ETEC,
enterotoxigenic
locus;
sive plasmid 845
BFP, bundle-forming
gregative E. coli; EHEC, enterohemorrhagic enteroinvasive E. coli; EPEC, enteropathogenic E. coli; ial, invasion-associated antigen;
PCR,
polymerase
chain reaction.
ipa, inva-
T. SUNABE
846
ticand-sera kit (Denka Seiken Co., Tokyo) were used. Among the isolates, 292 strains were isolated from diarrhea patients and 91 strains from healthy carriers in the Peoples' Democratic Republic of Lao (Lao PDR) between 1996 and 1997, the Dominican Republic in 1996, Indonesia in 1993, Kenya in 1981, and Japan in 1997. The organisms were stocked in tubes of nutrient agar diluted 1:2 with water until use. As a positive control for the eaeA gene, encoding for the 94 kDa protein intimin, and bfp gene, encoding the bundle-forming pilus protein, E. coli O111 strain RIMD0509829, which was a gift from Prof. T. Honda (Research Institute for Microbial Disease, Osaka University) was used. As a positive control for ipaH, encoding invasive plasmid antigen H, and the 230-kbp invasive plasmid, Shigella flexneri strain YSH6000 (11) was used. Slide agglutination test. The organisms subcultured on a nutrient agar plate were suspended in normal saline solution and autoclaved at 121 C for 30 min. They were then centrifuged and washed with normal saline solution before being analyzed by the agglutination test. They were serogrouped by a slide agglutination test using commercial antisera. Classification of serogroups. In the commercially available diagnostic antisera kit, there were 19 antisera for EPEC and 9 antisera for EIEC. The serogroups were classified into 4 categories. The first 2 categories were those proposed by Levine (7). The first, EPEC Class-I, included O26, O55, O86a, O111, O119, O125, Ol26,
O127a, O128
Class-II,
and O142;
included O18,
third, tentatively
designated
and the second,
044, O112 "EPEC
included O1, O146, O151, O157, O158
and O114. Others"
EPEC The
in this study,
and O166;
and
the fourth, EIEC, included O28ac, O29, O1 12ac, O124, Ol43, O144, O152, O164 and O167. The third catego-
Table
1. Primers
and PCR conditions
used in this study
AND Y. HONMA
ry
is serogroups
ual
of
the
empirically
Shiga-like of kit the
(0.2%
casamino
was
Seiken
Co.)
strains
were acids,
was
like
toxin
ined
120
(VT1
using PCR
the
of
reaction
E.
The col
(PCR).
assay The were
3 sets used
to
amplification Gannon The
eaeA
based
of
for
The with
Ol44,
et ipaH
0
the
was
of
min,
37
C
culture
and
Shiga-
was
exam-
gene
with
virulence
polymerase
all
study
eaeA
sets
are
of
chain
primers
shown
by
The
were
those
and
in Table
Gannon
gene.
examined
while strains.
amplifying
Gunzburg
at
The
30
by
described
eaeA
isolates,
eaeA-positive
used
NaCl,
8.5)
et
al
conditions described
1. (2) for by
(2). gene
EPEC
0.25%
pH
associated
in this
primers
the
manual. medium
supernatant
detected
sequences
amplify of
et al
genes
used
the
kit.
i were The
conditions
extract,
g for
in the
VTEC-RPLA
assays.
factors
VT2)
to
a VTEC-
CA-YE
shaking.
at 7,500 •~ and
toxin
using
trypton,
rounds/min
man-
Seiken.
Shiga-like
in
yeast
the
Denka
according
cultured
Bacto
centrifuged
on
of
examined
0.6%
1%
with
fluid
kit The
strains
K2HPO.4,
overnight
the
the
(Denka
Briefly,
0.871%
depending
antisera
toxinproductivity.
productivity RPLA
used
commercialized
the The
the
hfp
in h/b
all
gene
primers gene
310
serogroup-
was and
examined the
were
in
conditions
described
by
al (4). gene
antigens O28, O29,
was
examined
in
the
E.
coli
isolates
O112ac, O124, O136, O143,
O152 and O164 (12). The presence of the 230kbp invasive plasmid was examined in the ipaH-positive strains. The primers and PCR conditions for amplification of the ipaH gene were the same as those described by Sethabutr et al (13). For detection of the 230-kbp invasive plasmid, 2 sets of primers, KL1 and KL8, described by Lampel et al (6), and primer I and primer II, amplifying the invasion-associated locus (ial), described by Sethabutr et al (13), were used. The template DNAs
O- SEROGROUP
AND
PATHOGENIC
were prepared by suspending the bacterial colonies in sterile, distilled water and boiling for 10 min. Reactions were conducted by using Taq DNA polymerase and a DNA amplification kit from GIBCO BRL (Grand Island, N.Y., U.S.A.) as recommended by the manufacturer. Statistical analysis. The carrier rates of the eaeA gene in E. coli from diarrhea patients and healthy carriers (Table 3) were compared by a two-tailed x2 test with Yates' correction and by Fisher's exact test. Results Serogroups and Categories of the Strains A total of 383 strains were classified into 4 categories, with 134 strains of EPEC Class-I, 63 strains of EPEC Class-II, 113 strains of EPEC Others, and 73 strains of EIEC. One-hundred-and-thirty-four strains of EPEC were classified into 10 serogroups belonging to category EPEC Class-I, 63 strains of EPEC into 3 serogroups of EPEC Class-II, 113 strains into 6 serogroups of EPEC Others, and 73 strains of EIEC into 9
Table
2. Number
of eaeA-positive
strains
GENES
IN E.COLI
847
serogroups of EIEC (Table 2). No strains used in this study produced Shiga-like toxin as examined using the VTEC-RPLA kit. Frequency of Pathogenic Gene-Positive Strains The presence of the chromosomal eaeA gene was examined by PCR using 3 sets of primers (Table 1). A strain was regarded as positive for the eaeA gene if it produced at least 1 of the 3 kinds of PCR products. The eaeA gene was detected in 42 of the 232 serogroupbased EPEC strains isolated from diarrhea patients (18.1%), and in 9 of the 78 strains isolated from healthy carriers (11.5%) (Tables 2 and 3). The eaeA gene was detected in 9 of the 73 strains (12.3%) of the EIEC serogroup (Tables 2 and 3). The positive rate of the eaeA gene in each category is shown in Table 2. There was no significant difference between the carrier rates of the eaeA gene in the serogroup-based EPEC isolated from the patients and healthy carriers (P=0.22, >0.05), nor between the rates in EIEC isolatedfrom patients and carriers (P=0.67, >0.05). There was also
in examined
samples
and their
serogroups
848
T. SUNABE
Table 3. The eaeA gene-positive
Table
4. Frequency
of the eaeA gene-positive
AND Y. HONMA
rate
strains
no significant difference between the eaeA-positive rates in the serogroup-based EPEC and EIEC(P = 0.48, >0.05). The bfp gene on the EAF plasmid was detected in only 7 (4 strains were from patients in Lao, 3 were from patients in Kenya) of the 60 eaeA-positive strains. The chromosomal ipaH gene was detected in 4 of the 73 strains (5.5%) of the EIEC serogroup, and all ipaH-positive strains also possessed the invasive plasmid. The strains positive for ipaH and invasive plasmid were all isolated from diarrhea patients in Lao. There were no strains which possessed both the eaeA gene and ipaH gene. There were some geographical differences in the frequency of detection of the pathogenic genes (Table 4). Relationship between Pathogenic Genes and Serogroup Although the rate of the eaeA gene in the strains of EPEC Class-I was 20.1%, serogroups O111 and 0119 had the eaeA gene at a higher frequency (Table 2). In the serogroup-based EIEC, the 4 strains positive for ipaH and the invasive plasmid belonged to serogroups 028ac (3 strains) and 0143 (1 strain). Discussion In this study, the positive rates of pathogenic genes in classical serogroup-based EPEC and EIEC were examined. In 1987, Levine reported that the EAF probe (plasmid encoding bfp) hybridized with approximately 75% of
the E. coli strains which were isolated from diarrhea patients and identified as suspect EPEC by multi-step 0 serogrouping (7). The results of the present study, in which 16.5% of the E. coli strains identified as serogroupbased EPEC and 20.1% of the E. coli strains belonging to the Class-I serogroup were shown to carry the eaeA gene, were far different from those reported by Levine. There are some other studies describing the relation between serogroup and pathogenic genes. Nagayama et al (8) showed that 76% (31/41) of eaeA-positive E. coli belonged to traditional EPEC serogroups. Tsukamoto and Kawai (15) reported that the rate of EPEC in the E. coli strains carrying eaeA was about 55% (79/144), and Albert et al (1) reported that 57% (40/70) of EAF-positive and/or eaeA-positive EPEC belonged to a traditional serogroup. The rate of eaeA-positive and/or EAF-positive strains in EPEC identified by serogrouping was recently reported by Giammanco et al (3). They found that 60% (33/55) of Class-I EPEC determined by seroagglutination possessed the eaeA gene, which is far higher than our results of 20.1%. To understand this difference, a geographical study should be carried out. Also, the benefits of screening only by serogrouping to detect EPEC should be reviewed from a variety of viewpoints. There is agreement among the phenotypic characteristics of localized adherence on HEp-2 cells, the presence on the bfpA sequence (9). But EAF-negative strains
O- SEROGROUP
AND
PATHOGENIC
that produce the attaching-and-effacing (A/E) lesion can result from loss of the EAF plasmid. In this study, the positive rate of the EAF plasmid (as examined by PCR to detect bfp) was very low, at 13.7% of eaeA-positive EPEC(7/51). It might be suspected the EAF plasmid was lost during storage. But the EAF-positive rate was high in the eaeA-positive strains isolated in Kenya (3/7, 42.9%) in 1981, which were stored longer than the other strains. Thus, the different EAF-positive rates might be due to some regional difference or due to the properties of each strain. Other points to be noted from the present study are: 1) there was no significant difference in the carrier rates of the eaeA gene in serogroup-based EPEC or EIEC between diarrhea patients and healthy carriers, 2) there was no significant difference between the eaeA-positive rates of serogroup-based EPEC and EIEC, 3) the presence of eaeA was concentrated in strains with serogroups O111 and O119, and 4) ipaH was detected in 6% (4/60) of serogroup-based EIEC isolated from patients. These findings should be confirmed in a largescale study. This study also suggests that there are some factors other than the eaeA gene causing diarrhea in EPEC solely detected by serogrouping, and the screening of EPEC and EIEC using the sero-agglutination test is not as efficient as it has been believed. We are gratefulto Prof. MasaakiIwanaga (Departmentof Bacteriology, Facultyof Medicine,Universityof the Ryukyus)for his adviceon the designof the study and for his supervision. References 1) Albert, M.J., Faruque, S.M., Faruque, A.S.G., Neogi, P.K.B., Ansaruzzaman, M., Bhuiyan, N.A., Alam, K., and Akbar, M.S. 1995. Controlled study of Escherichia coli diarrheal infections in Bangladeshi children. J. Clin. Microbiol. 33: 973-977. 2) Gannon, V.P.J., Rashed, M., King, R.K., and Thomas, E.J.G. 1993. Detection and characterization of the eae gene of Shiga-like toxin-producing Escherichia coli using polymerase chain reaction. J. Clin. Microbiol. 31: 1268-1274. 3) Giammanco, A., Maggio, M., Giammanco, G., Morelli, R., Minelli, F., Scheutz, F., and Caprioli, A. 1996. Characteristics of Escherichia coli strains belonging to enteropathogenic E. coli serogroups isolated in Italy from children with
GENES
IN E. COLI
849
diarrhea. J. Clin. Microbiol. 34: 689-694. 4) Gunzburg, S.T., Tornieporth, N.G., and Riley, L.W. 1995. Identification of enteropathogenic Escherichia coli by PCRbased detection of the bundle-forming pilus gene. J. Clin. Microbiol. 33: 1375-1377. 5) Kopecko, D.J. 1994. Experimental keratoconjunctivitis (Sereny) assay. Methods Enzymol. 235: 39-47. 6) Lampel, K.A., Jagow, J.A., Trucksess, M., and Hill, W.E. 1990. Polymerase chain reaction for detection of invasive Shigella flexneri in food. Appl. Environ. Microbiol. 56: 1536-1540. 7) Levine, M.M. 1987. Escherichia coli that cause diarrhea: enterotoxigenic, enteropathogenic, enteroinvasive, enterohemorrhagic, and enteroadherent. J. Infect. Dis. 155: 377389. 8) Nagayama, K., Yamada, K., Yamamoto, K., and Honda, T. 1996. Detection of bundle-forming pili (BFP) gene and attaching and effacing (eae) gene in E. coli isolated from patients in Kenya. Kansenshogaku Zasshi 70: 142. 9) Nataro, J.P., and Kaper, J.B. 1998. Diarrheagenic Escherichia coli. Clin. Microbiol. Rev. 11: 142-201. 10) Rodrigues, J., Scaletsky, I.C.A., Campos, L.C., Gomes, T.A.T., Whittam, T.S., and Trabulsi, L.R. 1996. Clonal structure and virulence factors in strains of Escherichia coli of the classic serogroup O55. Infect. Immun. 64: 2680-2686. 11) Sasakawa, C., Kamata, K., Sakai, T., Murayama, S.Y., Makino, S., and Yoshikawa, M. 1986. Molecular alteration of the 140-megadalton plasmid associated with the loss of virulence and Congo red binding activity in Shigella flexneri. Infect. Immun. 51: 470-475. 12) Sethabutr, O., Echeverria, P., Hoge, C.W., Bodhidatta, L., and Pitarangsi, C. 1994. Detection of Shigella and enteroinvasive Escherichia coli by PCR in the stools of patients with dysentery in Thailand. J. Diarrheal Dis. Res. 12: 265-269. 13) Sethabutr, O., Venkatesan, M., Murphy, G.S., Eampokalap, B., Hoge, C.W., and Echeverria, P. 1993. Detection of Shigellae and enteroinvasive Escherichia coli by amplification of the invasion plasmid antigen H DNA sequence in patients with dysentery. J. Infect. Dis. 167: 458-461. 14) Sohel, I., Puente, J.L., Ramer, S.W., Bieber, D., Wu, C.Y.,and Schoolnik, G.K. 1996. Enteropathogenic Escherichia coli: identification of a gene cluster coding for bundle-forming pilus morphogenesis. J. Bacteriol. 178: 2613-2628. 15) Tsukamoto, T., and Kawai, T. 1995. The eae gene, adherence to HeLa cells and serotypes of Escherichia coli isolated from diarrhea. Kansenshogaku Zasshi 69: 85-90. 16) Vial, P.A., Robins, B.R., Lior, H., Prado, V., Kaper, J.B., Nataro, J.P., Maneval, D., Elsayed, A., and Levine, M.M. 1988. Characterization of enteroadherent-aggregative Escherichia coli, a putative agent of diarrheal disease. J. Infect. Dis. 158: 70-79.
