Colonization factors of enterotoxigenic E. coli (ETEC) from residents of Northern Egypt

Microbiol. Res. (1995) 150,429-436

Microbiological Research © Gustav Fischer Verlag Jena

Colonization factors of enterotoxigenic E. coli (ETEC) from residents of Northern Egypt B. A. Oyofol, S. H. EI-Etr\ M. O. Wasfy\ L. Peruski\ B. Kay\ M. Mansour!, J. R. Campbelll, A. M. Svennerholm 2 , A. M. Churilla!, J. R. Murphyh 1 2

U. S. Naval Medical Research Unit No.3, Cairo, Egypt Department of Medical Microbiology and Immunology, University of Goteborg, Sweden

* Present Address: James Ralph Murphy, Center for Infectious Diseases, School of Public Health and Medical School, The University of Texas, Health Science Center at Houston, Houston, TX 77030, USA Accepted: August 6, 1995

Abstract Infection caused by enterotoxigenic Escherichia coli (ETEC) poses a serious health problem to children in developing countries. Colonization of the small intestinal mucosa by ETEC strains is mediated by antigenically specific fimbriae, also known as colonization factor antigens (CFA). The importance of this study arises from reports that active and passive immunization with ETEC strains harboring CFAs induced protective immunity against diarrhea in animal models with preformed antibodies. In humans, ETEC containing CFA/I, II, III and IV have been identified. The aim of this study was to define CF As of ETEC isolated in Alexandria, Egypt. One hundred and seven ETEC isolates from 132 human residents in Alexandria, Egypt were isolated during a birth cohort study. ETEC isolates were screened for heat labile (LT) and heat stable (ST) toxins using a 32p oligonucleotide hybridization probe and a GM 1 ELISA. These isolates were examined using monoclonal antibodies against CF A/I, II, III, IV, and against the putative colonization antigens PCF0159 and PCFOI66, CS 7 and CS 17. CF As were found in 48% of ETEC strains. CFA/I was found in 18% ofthe strains, CFA/II in 10% and CFA/IV in 14%. CF A III was not found. All fifteen strains expressing CFA/IV expressed CS 6 and produced ST. CF A/IV was not found in non-ST producing strains, while CF A/I was absent in ST - only producing strains.

Key words: CF A - ETEC - Diarrhea

Correspondence to: US Naval Medical Res. Unit 3 Research Publication Branch PSC 452, Box 5000 FPO AE 09835-0007 N. York N.Y. USA

Introduction Enterotoxigenic Escherichia coli (ETEC) is the most frequently isolated agent of acute diarrhea among children and travelers in developing countries (Black 1986). These bacteria colonize the small bowel epithelium by means of thread-like appendages (fimbriae) and cause diarrhea. These fimbriae are antigenically diverse and have been called colonization factor antigens (CF As) (Klemm 1985). ETEC strains have been characterized by their capacity to produce a heat-stable enterotoxin (ST), a heat-labile enterotoxin, (LT) or both (ST /LT). Recognized CF As associated with human ETEC isolates are CF A/I, CF A/II and CF A/IY. CF A/I is a single fimbrial antigen (Evans et al. 1975), while CFA/II and CFA/IV have been shown to consist of three distinct fimbrial antigens called coli surface antigens. CF A/II is composed of antigens CS 1, CS 2 and CS 3 (Smith et al. 1983). CF A/IV comprises antigens CS 4, CS 5 and CS 6 (Thomas et al. 1985). CF A/I causes mannose-resistant hemagglutination (MRHA) of human and bovine erythrocytes (Evans et al. 1979), a property which has been shown to be mediated by the same plasmid coding for ST production (Willshaw et al. 1982). CFA/II causes MRHA of bovine, but not human erythrocytes (Evans et al. 1979) and genes coding for its expression have been associated with the plasmids of both L T and ST (Smith et al. 1983). CS 4 and CS 5 strains carrying CF A/IV have been shown to agglutiQate human group A and bovine Microbiol. Res. 150 (1995) 4

429

erythrocytes, whereas CS 6 did not cause MRHA (Thomas et al. 1985). A number of other putative colonization factors (PCFs) have been described and characterized. These include those referred to as CF A/III, CS 7, CS17, PCF0159, PCF0166, PCF2230, PCF0148, PCF09 and PCF8786 (Tacket et al. 1987; Viboud et al. 1993). Previous studies have associated ETEC CF As or PCFs with particular 0 serogroups and enterotoxin types (Tacket et al. 1987; Viboud et al. 1993). For. example, PCF0159 was found only in ST and LT producing strains of serotypes 0159: H 4 or 0159:H20 (Tacket et al. 1987). CFA/III has been found only in LT-producing strains of serotype 025: H 16 (Hibberd et al. 1990). The role of CFAs in the pathogenesis of ETECassociated diarrhea has been well described in human volunteer studies (Evans et al. 1984; Levine et a/. 1984; Satterwhite et al. 1978). Further, it has been shown in animal models that active or passive immunization with ETEC strains, harboring CF As, induced protective immunity against diarrhea (Ahren and Svennerholm 1986; Svennerholm et al. 1988). Human volunteers immunized with purified CFA/I or CFA/II produced specific serum and intestinal antibody responses (Evans et al. 1984). This observation argues for the inclusion of CF As into a multivalent vaccine formula (Svennerholm et al. 1989). Infection by CF A-bearing ETEC strains from different countries differ from 25 and 75% (Cravioto et al. 1988; McConnell et al. 1991). The goal of this study was to define the relationship between the expression of LT, ST, and CF As by ETEC strains isolated in Alexandria, Egypt due to the upcoming ETEC vaccine trial in this region. In addition, the relationship between types of CF As, enterotoxin production, mannose resistant hemagglutination (MRHA) and hydrophobicity was evaluated.

Materials and methods Study population. ETEC were isolated from members of households where at least one child was enrolled in a birth cohort study of etiology of diarrhea. The study was made in the Abees village complex, Alexandria Governorate, Egypt. Epidemiological surveillance/definition of events. All births occurring in households under surveillance were eligible for enrollment. Households were visited twice per week. During each visit a questionnaire was administered which was, in part, targeted to detection of cases of diarrhea occurring in enrolled births. A rectal swab was collected from each enrolled child

430

Microbiol. Res. 150 (1995) 4

once per week. Additionally, a care giver/head of household report of diarrhea caused collection of a rectal swab at either the visit when the report was made or at the following visit. The same sequence of enhanced sampling was triggered by a care giver report of a child passing four or more stools in the 24 hour period prior to a household visit. A care giver report of diarrhea in an older member of a household resulted in collection of stool from that individual. Either a care giver report of diarrhea or a care giver report of the passage of > three stools in the 24 hours preceding a visit resulted in a declaration of a case. ETEC isolated from a rectal swab was considered the probable cause of a case when ETEC was isolated at the time of or within seven days of declaration of a case. Volunteers had to be diarrhea free and culture negative for diarrheogenic bacteria for 14 days before they were eligible for classification as a new case. Bacterial strains and cultivation. Specimens collected from patients were cultured on MacConkey agar plates for selection of E. coli isolates. After overnight culture, five E. coli lactose fermenting colonies from each patient were picked and stored on nutrient agar stab cultures until analyzed for heat-labile (LT) and heat-stable (ST) enterotoxins. A total of 107 ETEC strains from 132 human residents in Alexandria governorate, Egypt was examined for LT and ST toxin production and CFA expression. Of the 132 subjects, 58 were symptomatic, while 74 patients were asymptomatic for diarrhea. For analysis, lactose fermenting colonies were selected. Bacteria were then inoculated into CFA media, Luria-Bertani broth in 96 well micro titer plates for GM 1 ELISA and lifted on Whatman 541 filters for use in colony hybridization with 32p labelled probes. Oligonucleotide probes for ETEC. 32p probing for ST and L T genes was done according to the method of Murray et al. (1987). The oligonucleotides used consisted of the following sequences: STH, (5') GCT GTG AAT TGT GTT GTA ATC C (3'); STp , (5') GCT GTG AAC TTT GTT GTA ATC C (3'); and LT (5') GCG AGA GGA ACA CAA ACC GG (3'). For hybridization, five individual colonies from each patient were spotted and lysed on Whatman 541 filters. Oligonucleotides were labeled by using T 4 polynucleotide kinase and [32 p] ATP. Colony hybridization was performed as previously described (Murray et al. 1987). ELISA. The GM 1 ELISA for LT and ST genes was performed as described by Svennerholm et al. (1986). Microtiter plates (Dynatech Immunoclone I) were

coated with GM 1 ganglioside (0.3 nml/ml) (Sigma) in PBS and stored at 4°C. Plates were washed with PBS and blocked with 0.1 % BSA-PBS for 30 min at 37°C. LB medium containing 45 Jlg/lincomycin/ml and 2.5 mg glucose/ml was added. Bacterial isolates were then inoculated as one colony per well. Plates were incubated overnight at 37°C with gentle shaking. A second set of coated plates was washed, blocked and incubated with ST-cholera toxin "B" subunit (ST-CTB) conjugate diluted 1: 300 (100 Ill/ well) for 60 min at room temperature. Fifty III of bacteria culture fluid was added from the first set of plates, to corresponding wells in the second set. An equal volume of monoclonal antibody ST (1: 3) diluted 1: 400 was then added, and plates were incubated for 90 min at room temperature. The first set of plates was washed and monoclonal antibody LT -39, diluted 1: 100, was added (100 Ill/well). The plates were incubated for 90 min at room temperature. The presence of bound antibody was demonstrated by addition of an antimouse immunoglobulin-horseradish peroxidase conjugate with orthophenylenediamine (OPD)-H 2 02 as enzyme substrate. Plates were read using an ELISA reader at 450 nm after 10 - 20 min. Dot blot CFA test. Two III of bacterial suspension diluted in PBS was added to previously soaked and marked nitrocellulose membrane strips. Membranes were blocked in 1 % BSA at room temperature for 60 min with rocking. Monoclonal antibodies against CFA/I, CS1, CS2, CS3, CS4, CS5, CS6, CS7, CS 17, PCF0159, PCF0166 and CFA/III were added to different membrane strips containing the same bacterial culture diluted 1: 20 and 1: 30 in PBS, and strips were incubated for 2 hr at room temperature with rocking. Membrane strips were washed with PBS. Bound antibody was then demonstrated by the addition of anti-mouse immunoglobulin-horseradish peroxidase conjugate with 4-chloro-naphthol-H 2 0 2 as enzyme substrate. Positive isolates were determined visually at a scale of + 1 to + 4 depending on the relative intensity of blue colour (Lopez-Vidal and Svennerholm 1990; Viboud et al. 1993). Hemagglutination. Fresh erythrocytes of human type A, bovine, guinea pig, chicken, goat, sheep, rabbit, and horse were washed twice in 0.85% NaCI and suspended to 3% in saline with or without 1 % D-mannose. Ten III each suspension was mixed with 10 III of the bacterial suspensions and observed for 2 min at room temperature, and for another 5 min at 4°C when bovine erythrocytes were used. The hemagglutination (HA) was considered to be MRHA if the same degree of HA occurred with and without Dmannose (Evans et al. 1979).

Salting out (SO) test. The SO test was performed as described by Lindahl et af. (1981) with a modification to determine the surface hydrophobicity of bacterial cells. Ten III of the bacterial suspension (10 10 organism per ml) was mixed in a 96 well round-bottom micro titer plate with an equal volume of selected concentrations of ammonium sulfate. The original procedure calls for individual mixing of the mixture on a glass slide. The plate was rotated for 2 min at room temperature, and aggregation of bacteria was observed. The bacterial surface hydrophobicity was evaluated by determining the lowest concentration of ammonium sulfate in which bacteria aggregated. Statistical analysis. Age and stools of patients per 24 hours were evaluated using the two sample t-test, other variables by Chi square (Yate's corrected).

Results Relationship between production of CFAs and toxin production. One hundred and seven ETEC strains were detected by 32p probe as compared to 98 ETEC strains, detected by GM 1 ELISA procedure. Both techniques showed an agreement of 92% between results. Fifty (47%) isolates produced ST only, 33 (31%) produced LT only, and 24 (22%) produced ST/LT (Table 1). Of the 107 ETEC isolates, 19 (18%) expressed CFA/I, 11 (10%) expressed CFA/II, 15 (14%) expressed CFA/IV and 6 (6%) expressed PCF0159. CFA/II and/IV isolates produced ST only, while 18 of the 19 CFA/I strains produced ST /L T. PCF 0159 was associated with the production of ST and ST /L T.

Table 1. Presence of CF A/I, CFA/II, CF A/IV and PCF0159 in relation to toxin production in ETEC strains isolated from Egypt Marker

No. (%) of strains producing toxin ST LT ST and LT Total

CFA/I 0 CFA/II CS1, CS2, CS3 5 CS3 6 CFA/IV CS6 15 PCF0159 4 None 20 TOTAL 50 (47)

1

0 0

0

18 0 0

0

0 0 0 2 32 4 33 (31) 24 (22)

19 (18) 11 (10) 15 (14) 6 (6) 56 (52) 107

* All strains tested were negative for CFA/III, PCF0166, CS 7 and CS 17. Microbiol.

Res. 150 (1995) 4

431

Analysis of isolates. (i)

CFA/I. All of the 17 strains that reacted with the monoclonal antibody for CFA/I agglutinated human and bovine erythrocytes. Fifteen of these also agglutinated sheep erythrocytes, and 2 agglutinated chicken erythrocytes (Table 2). All 19 produced aggregates in ammonium sulfate concentrations between 0.06 M and 0.14 M, indicating a high degree of hydrophobicity (Table 3). (ii) CFA/II. AlIH strains that reacted with CFA/II monoclonal antibodies, five strains reacted with CS1, CS2, CS3, while six strains reacted with CS 3 only. All agglutinated bovine and chicken erythrocytes, exhibited hydrophobicity in the range of 0.2 to 0.4 M and were only associated with ST (Tables 2, 3).

(iii) CFA/ill. None of the isolates reacted with antibody to CF A/III. (iv) CFA/IV. Fifteen ETEC strains reacted with CF A/IV monoclonal antibodies: six reacted with bovine, chick and sheep erythrocytes; three reacted with human, bovine, chicken, and sheep erythrocytes tested; two agglutinated bovine and chicken; one agglutinated human and bovine erythrocytes, and the rest agglutinated only sheep erythrocytes (Table 2). Seven of the 15 strains showed a hydrophobicity range of 0.1- 8.0 M while the remaining eight showed no precipitation reaction with ammonium sulfate (Table 3). There was no reactivity when CF A strain was tested with monoclonal antibodies against CS 4 and CS 5. Only CS 6 was reactive.

Table 2. MRHA of CF A-positive and CF A-negative E. coli strains CFA property

No. of strains 15 2 2

CFA/I

CFA/II CSl, CS2, CS3 CS3 CFA/IV CS6 CS6 CS6 CS6 CS6 CS6 PCF0159 CFA Negative

MRHA of Erythrocytes ~--------~~--------~~------~~-------------

Human

Bovine

+

+ + +

+

+ +

+ +

+ +

5 6 6 3 2 2 1 1 6 2 1 3

+ +

Chicken

+

+ +

19

+

+

+

Total

5 6

+

+

Sheep

+

+

+

+

15 6

+

+

+

+

+

+

6

Table 3. MRHA and SO properties of strains CFA Property

CFA/I CFA/II CFA/IV PCF0159 Negative strains*

SO (concentration)

No. of strains producing toxin ST

LT

ST/LT

Total

(0.06-0.14 M)* (0.2-0.4 M)* (0.1-0.8 M)* (No aggregation)t (0.2 M - 0.6 M) (No aggregation)t (0.6 M)* (No aggregation)t

0 11

1 0 0 0 0 0 0 0

18 0 0 0 2 0 0 0

19 11 7 8 5

* MRHA positive, SO positive; t MRHA positive, SO negative;

432

Microbiol. Res. ISO (1995) 4

7 7 3 1 2 4

* MRHA negative, SO positive; * MRHA negative, SO negative

1

2 4

Table 4. Relationship among ETEC toxin, CFA, erythrocyte agglutination types, presence and severity of diarrhea

Diarrhea Status

Mean; range; Sex standard deviation No. of Age Stool/ Female LT Volun24h teers

None

52

Any

14

Signiticant

5

6.9; 1-12; 3.3 9.3; 3-19; 5.9 12.2; 5-19; 6.1N

2.2; 1-3; 0.7 3.4; 1-5; 1.1D 4.2; 4-5; o.sN,a

Toxin ST

Percent Positive for Characteristic LS

Colonization factor

RBC agglutination

PPT

I I I IV I PCF

H

SO

I II III

I B I cis

60

33

42

27

27 0

0

23

0

40

54

37

33 48

57

64 D 14

21

14 0

0

7

0

21

36

43

43

0

o

0

0

0

20

20

20

20 20

60

lOON

0

0

29

Notes: Reported are: age in months; number of stools passed in the 24 hour period immediately preceding the household visit where the sample yielding ETEC was obtained; six; ETEC toxin type (as LT = heat labile toxin, ST = heat stable toxin, LS = LT and ST both present together in each colony; there was one instance (for the any group) where LT and ST were both present but on different colonies; this inidividual is classified positive for both LT and ST); colonization factor antigens (as I = CFA/I, II = CFA/II, III = CFA/III, IV = CFA/IV and PCF = PCF0159); hemagglutination (as H = human, B = bovine, C = chicken and S = Sheep erythrocytes); and, SO presents results of salting out tests. Statistically significant between group differences are denoted by superscripts where an uppercase letter denotes p ~ 0.01, a lowercase letter p ~ 0.05 and the lack of a superscript letter p > 0.05. The letter 'n' signifies that the marked group differs from the 'none' diarrhea classification group; the letter 'a' means that there is a difference from the 'any' diarrhea classification group. Comparisons between 'any' and 'significant' groups comprised the 5 individuals in the 'significant' group compared to the 9 (of 14 total) individuals in the 'any' group who were not also members of the 'significant' group.

CFA negative strains.

Of the 107 ETEC isolates tested, 56 (52%) were negative for all CF As and PCFs tested. Of these 56 strains, 8 produced MRHA with all of the erythrocytes tested. Strains that gave MRHA of human and bovine erythrocytes predominated (Table 2). The hydrophobicity range of these isolates was less than 0.6 M (Table 3).

PCF0159. None of six strains characterized as PCF0159 agglutinated the erythrocytes tested (Table 2). Five strains had a hydrophobicity range of 0.2 - 0.6 M (Table 3).

Relationship among ETEC toxin, CFA, erythrocyte agglutination types, presence and severity of diarrhea. Review of the demographic/case records found 66 unique cases where ETEC was isolated from children < 24 months of age (Table 4); 52 diarrhea free and 14 diarrhea cases. For analysis diarrheic individuals were stratified to groups 'any' and 'significant' diarrhea where: 'any' diarrhea comprised either a report of the occurrence of diarrhea in a child and/ or the passage of > 3 stools in the 24 hour period prior to the sample collection. 'Significant' diarrhea comprised both report of diarrhea occurrence and passage of > 3 stools prior to sample collection.

The data do show significant relationships between the presence of 'any' or 'significant' diarrhea and the presence of L T toxin.

Discussion The purpose of this study was to define CF As and their association with toxin production in ETEC strains isolated from humans in Alexandria, Egypt due to the upcoming ETEC vaccine trial in this region. Reported surveys of ETEC strains from various parts of the world (Table 5) have shown a

Table 5. Pattern of enterotoxigenic Escherichia coli (ETEC) strains carrying colonization factor antigens (CF A) from different geographic areas Site reference Asia, [24] Thailand, [25] Bangladesh, [23] Rural Mexico, [17] Mexico City, [34] Argentina, [36] Egypt

% of ETEC strains with fimbriae

CFA/I

CFA/II CF A/IV TOTAL

14 21 57 27 18 23 18

13 8 18 14 5 12 10

5 0 ND 34 23 17 14

32 29· 75 75 46 52 42

Microbiol. Res. 150 (1995) 4

433

wide variation in the number of E. coli strains with adhesion factors. In a clinical study conducted in Bangladesh (Gothefors et af. 1985), 75% of the ETEC strains expressed CFA/lor CFA/II; compared to 32.5% of the ETEC strains isolated from various categories of patients in South East Asia possessing CF A/I, CF A/II or CF A/IV (Thomas and Rowe 1982). On the other hand, ETEC isolates from Thai patients showed an overall 29% rate for CF As (Changchawalit et af. 1984). In Mexico City, 46% of the ETEC strains expressed CFAs, with 18% CFA/I, 5% CFA/II and 23% CFA/IV as compared to ETEC strains from Argentina expressing 23 % CF A/I, 12% CFA/II and 17% CFA/IV (Binsztein et a1.1991) (Table 5). In the present study 48 % of the strains expressed CFAs: 18% expressed CFA/I, 10% expressed CFA/II, 14% CFA/IV and 6% expressed PCF 0159. The difference between our results and those mentioned above may in fact be due to differences in the expression of CF As on ETEC in various geographic areas. However, another reason could be the laboratory techniques employed in the identification of CF A/toxin. It has been reported that the low percentage of strains carrying CF As and enterotoxins in certain areas may also be due to repeated subculturing of strains and subsequent loss of plasmids (Ahren et af. 1986; McConnell et af. 1989). Our results agree with the previous observation that ST-only and ST/LT-ETEC strains are more frequently associated with CFA expression, in contrast to LT-only ETEC strains (Ahren et al. 1986). Fifty (52%) ETEC strains tested in this collection had no associated CFA or PCF. It is possible that these strains may be expressing other fimbriae types that have not yet been characterized. The relatively low proportion of PCF -expressing strains (6%), found in this study, is in agreement with the report ofViboud et af. (1993) and McConell et af. (1990) who found 9.5% and 17% ofthe ETEC strains tested expressed PCFs. The low prevalence of PCFs could be attributed to many factors. A loss of plasmids that code for the PCFs can occur as a result of repeated subculturing or long term storage (Mullany et af. 1983). However, it has been reported that the same plasmid that codes for the PCFs also codes for toxin production (Gothefors et af. 1985; McConnell et af. 1990). Though the strains used in this study were subcultured once, they were 100% enterotoxin positive when tested using a 32p probe and 92% positive with ELISA utilizing monoclonal antibodies. Therefore, it is unlikely that the reported low percent PCF found is attributable to plasmid loss alone. Among the 56(52%) strains for which nocolonization factor could be identified, 20 (36%) produced 434

Microbiol. Res. 150 (1995) 4

ST, 32 (57%) produced LT, and 4 (7%) produced both. The detection levels of radioactive probe and the ELISA agreed by 92%. The higher sensitivity of the probe is explained by the fact that it detects the presence of the gene while the ELISA only detects the toxin produced. These results parallel a previous study by Blomen and co-workers (1993), which compared the utility of PCR, DNA hybridization, ELISA and bioassays in the identification of ETEC. CF A/I, the most prevalent adhesion antigen in this study, hemagglutinated human, bovine and sheep erythrocytes. Two reports (Thomas et al. 1982; Wolf et af. 1989) have shown that CF A/IV -positive strains hemagglutinate human, bovine and/or guinea pig erythrocytes. In this study, we found that some of the strains producing the CFA/IV antigen CS 6 hemagglutinated human, bovine, chicken and sheep erythrocytes. These results may suggest the presence of different binding structures on the erythrocytes within the same animal species (McConnell et af. 1990). The mechanisms by which bacteria bind to erythrocytes are variable. They include lectinlike interactions, electrostatic and hydrophobic interactions, hydrogen bonds and Van der Waals interactions (Lindahl et af. 1981). Hydrophobic interactions are thought to be of major importance in the attachment of the CFA/I and CFA/II to host cells (Lindahl et af. 1981); therefore, methods determining hydrophobic binding are useful for the indirect measurement of CFA properties. Results from this study show that not all of the CFA/IV-positive strains tested show hydrophobic properties with the SO test. This may mean that other mechanisms of attachment to erythrocytes exist in CFA/IV or PCF 0159 positive strains. Of the 132 humans examined in this study, 58 (44%) subjects exhibited symptoms of diarrhea, while 74 (56%) were asymptomatic. Fourteen (14%) of the symptomatic patients produced ST, 14% LT, and 7% produced ST/LT. Chi square analysis between these two groups show a significant difference of p < 0.05. ETEC toxin LT was significantly associated with the occurrence of 'any' or 'significant' diarrhea (Table 4). The group with 'significant' diarrhea was older than the asymptomatic group but not the 'any' diarrhea group. The number of stools passed in the 24 hours prior to the household visit where a case was declared for the 'significant' and 'any' diarrhea groups as compared to the 'none' group. The 'significant' diarrhea group had statistically significantly more stool per 24 hours than the 'any' diarrhea group. To our knowledge, this is the first reported study showing the relationships between the expression of LT, ST, and CFAs in ETEC strains isolated from Alexandria, Egypt. These data may help current

research efforts on development of a CFA-based vaccine for humans against ETEC (Evans et al. 1984; Levine et al. 1984; Murray et al. 1987) and provide significant information for the upcoming ETEC vaccine trial in this region. Moreover, for the first time, we report the association of CS 1 together with CS 2 in nature to CF A/II strains. This finding may have some bearing to the efficacy of the ETEC vaccine constituents in this area. Acknowledgments We thank Kerstin Anderson for expert technical assistance. We are also thankful to Dr. S. Lewis for reviewing the manuscript. We also thank Mrs. Salwa A. Mohanna for typing and editorial review of the manuscript. This study was supported by the Naval Medical Research and Development Command, Bethesda, MD, Work Unit No. 00101.EIX.3412. The opinions and assertions contained herein are the private ones of the authors and are not be construed as official or as reflecting the view of the Navy Department, Department of Defence, or the U.S. Government.

References Ahren, C. M., Gothefors, L. , Stoll, B. J., Salek, M. A., Svennerholm A.-M. (1986) : Comparison of methods for detection of colonization factor antigens of enterotoxigenic Escherichia coli. J. Clin. Microbiol. 23, 586- 591. Ahren, C. M. , Svennerholm, A.-M. (1986) : Experimental enterotoxin-induced Escherichia coli diarrhea and protection induced by previous infection with bacteria of the same adhesin or enterotoxin type. Infect. Immun. SO, 255-261. Binsztein, N., Jouve, M. J., Viboud, G. I., Lopez-Moral, L., Rivas, M., Orskov, I., Ahren, c., Svennerholm, A.-M. (1991): Colonization factors of enterotoxigenic Escherichia coli isolated from children with diarrhea in Argentina. J. Clin. Microbiol. 29, 1893 -1898. Black, R. E. (1986) : The epidemiology of cholera and enterotoxigenic Escherichia coli diarrheal disease. In : Development of vaccines and drugs against diarrhea (Eds : Holmgren, J., Lindberg, A. , Mollby, R.). Studentlitteratur, Lund, Sweden, 23 - 32. Blomen, I., Lofdahl, S., Stenstrom, T. A., Norberg, R. (1993): Identification of enterotoxigenic Escherichia coli isolates: a comparison of PCR, DNA hybridization, ELISAa and bioassays. J. Microbiol. Methods 17, 181-191. Changchawalit, S., Echeverria, P., Taylor, D. N., Leksomboon, U ., Tirapat, C., Eampokalap, B., Rowe, B. (1984) : Colonization factors associated with enterotoxigenic Escherichia coli isolated in Thailand. Infect. Immun. 45, 525-527. Cravioto, A., Reyes, R. E. , Ortega, R ., Fernandez, G ., Hernandez, R., Lopez, D. (1988) : Prospective study of diarrhoeal disease in a cohort of rural Mexican children :

incidence and isolated pathogens during the first two years of life. Epidemiol. Infect. 101, 123 -134. Evans, D. D ., Silver, R. P., Evans, D. J., Jr., Chase, D. G., Gorbach, S. L. (1975). Plasmid-controlled colonization factor associated with virulence in Escherichia coli enterotoxigenic for humans. Infect. Immun. 12, 656-667. Evans, D. J., Jr., Evans, D. G., DuPont, H. L. (1979) : Hemagglutination patterns of enterotoxigenic and enteropathogenic Escherichia coli determined with human, bovine, chicken, and guinea pig erythrocytes in the presence and absence of mannose. Infect. Immun. 23, 336-346. Evans, D. G., Graham, D . Y , Evans, D. J., Jr. (1984) : Administration of purified colonization factor antigens (CF A/I, CF A/II) of enterotoxigenic Escherichia coli to volunteers. Response to challenge with virulent enterotoxigenic Escherichia coli. Gastroenterology 87, 934-940. Gothefors, L., Ahren, C., Stoll, B., Barua, D. K ., Orskov, F., Salek, M. A., Svennerholm, A.-M. (1985): Presence of colonization factor antigens on fresh isolates of fecal Escherichia coli : a prospective study. J. Infect. Dis. 152, 1128-1133. Hibberd, M. L., McConnell, M . M ., Field, A. M., Rowe, B. (1990): The fimbriae of human enterotoxigenic Escherichia coli strain 334 are related to CS 5 fimbriae. J. Gen. Microbiol. 136, 2449 - 2456. Klemm, P. (1985): Fimbrial adhesins of Escherichia coli. Rev. Infect. Dis. 7, 321- 340. Levine, M. M. , Black, R. E., Clements, M. L., Young, C. R. , Cheney, C. P., Schad, P. , Collins, H ., Boedeker, E. C. (1984) : Prevention of enterotoxigenic Escherichia coli diarrheal infection in man by vaccines that stimulate anti-adhesion (antipili) immunity. In : Attachment of organisms to the gut mucosa (Ed: Boedecker, E. C.). CRC Press, Vol. 2. Boca Raton, F. L., 223 - 224. Lindahl, M., Faris, A., Wadstrom, T., Hjerten, S. (1981): A new test based on "salting out" to measure relative surface hydrophobicity of bacterial cells. Biochem. Biophys. Acta. 677,471-476. Lopez-Vidal, Y, Svennerbolm, A.-M. (1990): Monoclonal antibodies against the different subcomponents of CF AIl of enterotoxigenic Escherichia coli and their use in diagnostic tests. J. Clin. Microbiol. 28, 1906-1912. McConnell, M. M. , Chart, H ., Field, A. M ., Hibberd, M. , Rowe, B. (1989): Characterization of a putative colonization factor (pCFO 166) of enterotoxigenic Escherichia coli of serogroup 0166. J. Gen. Microbiol. 135, 1135-1144. McConnell, M. M., Hibberd, M ., Field, A. M., Chart, R. , Rowe, B. (1990): Characterization of a new colonization factor (CS 17) from a human entertoxigenic Escherichia coli of serotype 0114 :H21 which produces only heatlabile enterotoxin. J. Infect. Dis. 161, 343 - 347. McConnell, M . M. , Hibberd, M . L. , Penny, M . E., Scotland, S. M. , Cheasty, T. , Rowe, B. (1991): Surveys of human enterotoxigenic Escherichia coli from three different geographical areas for possible colonization factors . Epidemiol. Infect. 106,477 -484. Microbiol. Res. 150 (1995) 4

435

Mullany, P., Field, A. M., McConnell, M. M., Scotland, S. M., Smith, H. R., Rowe, B. (1983) : Expression of plasmids coding for colonization factor antigen II (CF A/II) and enterotoxin production in Escherichia coli. J. Gen. Microbiol. 129, 3591- 3601. Murray, B. A., Mathewson, J. J ., DuPont, H. 1., Hill, W. E. (1987): Utility of Oligodeoxy-ribonucleotide probes for detecting enterotoxigenic Escherichia coli. J. lnfec. Dis. 155,809-81l. Satterwhite, T. K., DuPont, H. L., Evans, D. G., Evans, D. J., Jr. (1978): Role of Escherichia coli colonization factor antigen in acute diarrhoea. Lancet 2, 181-184. Smith, H. R., Scotland, S. M ., Rowe, B. (1983): Plasmids that code for production of colonization factor antigen II and enterotoxin production in strains of Escherichia coli. Infect. Irnmun. 40, 1236-1239. Svennerholm, A.-M., Wikstrom, M., Lindblad, M., Holmgren, J. (1986): Monoclonal antibodies against E. coli heat-stable toxin (STa) and their use in a diagnostic ST ganglioside GM 1-enzyme-linked immunosorbent assay. J. Clin. Microbiol. 24, 585 - 590. Svennerholm, A.-M., Lopez-Vidal, Y. , Holmgren, J., McConnell, M. M ., Rowe, B. (1988): Role ofPCF8775 antigen and its coli surface subcomponents for colonization, disease, and protective immunogenicity of enterotoxigenic Escherichia coli in rabbits. Infect. Immun. 56, 523-528. Svennerholm, A.-M., Holmgren, J., Sack, D. A. (1989): Development of oral vaccines against enterotoxigenic Escherichia coli of diarrhoea. Vaccine 7, 196-198. Tacket, C. 0., Maneval, D . R., Levine, M. M. (1987): Purification, morphology, and genetics of a new fimbrial putative colonization factor of enterotoxigenic

436

Microbiol. Res. 15() (1995) 4

Escherichia

coli 0159: H 4. Infect. Immun. 55, 1063-1069. Thomas, L. V., Cravioto, A., Scotland, S. M., Rowe, B. (1982): New funbrial antigenic type (E 8775) that may represent a colonization factor in enterotoxigenic Escherichia coli in humans. Infect. Irnmun. 35, 1119 -1124. Thomas, L. v., Rowe, B. (1982): The occurrence of colonization factors (CF A/I, CFA/II and E 8775) in enterotoxigenic Escherichia coli from various countries in South East Asia. Med. Microbiol. Irnmunol. 171, 85-90. Thomas, L. V., McConnell, M . M ., Rowe, B., Field, A. M . (1985) : The possession of three novel coli surface antigens by enterotoxigenic Escherichia coli strains positive for the putative colonization factor PCF8775. J. Gen. Microbiol. 131, 2319 - 2326. Viboud, G. 1., Binsztein, N., Svennerholm, A.-M. (1993): Characterization of monoclonal antibodies against putative colonization factors of enterotoxigenic Escherichia coli and their use in an epidemiological study. J. Clin. Microbiol. 31, 558 - 564. Willshaw, G. A., Smith, H. R., McConnell, M. M., Barclay, E. A., Krnjulac, J., Rowe, B. (1982) : Genetic and molecular studies of plasmids coding for colonization factor antigen I and heat-stable enterotoxin in several Escherichia coli serotypes. Infect. Immun. 37, 858-868. Wolf, M. K., Andrews, G. P., Tall, B. D., McConnell, M. M., Levine, M. M., Boedeker, E. C. (1989): Characterization of CS 4 and CS 6 antigenic components of PCF 8775, a putative colonization factor complex from enterotoxigenic Escherichia coli E 8775. Infect. Irnmun. 57,164-173.