Article Eur. J. Clin. Mierobiol. Infect. Dis., August 1991, p. 625-629 0934-9723/91/08 0625-05 $ 3.00/0
Vol. 10. No. 8
625
Analysis of Escherichia coli Isolates from Subjects with Travellers' Diarrhoea Using DNA Probes and Serotyping
C.M.A. Rademakerl,3*, M.R.L. Krul 4, W.H. Jansen 4, N.M. Vos 1, I.M. Hoepelmanl,2, 3, M. R o z e n b e r g _ A r s k a l , 3, j. Verhoefl, 3
Escherichia coli isolated from faeces o f 54 healthy volunteers who visited Tunisia for eight days were examined. These volunteers participated in a randomized double. blind placebo-controlled study to establish whether ciprofloxacin could prevent travellers' diarrhoea. Escherlchia coil strains isolated before travel, during episodes o f travellers' diarrhoea, immediately after return and five weeks after return were serotyped and tested for the presence of virulence genes indicating diarrheogenic properties by hybridization with a set of four non-radioactively labelled D N A probes. Subjects receiving ciprofloxacin prophylactically to prevent travellers' diarrhoea w e r e asymptomatic and no Escherichia coil could be cultured shortly after return home. Sixty-four percent of subjects (18/28) who did not receive antibiotic prophylaxis suffered from travellers' diarrhoea. Hybridization tests detected 8 enterotoxigenic Escherichia colt strains producing heat stable toxin, 13 enterotoxigenic strains producing heat labile toxin and 10 strains which produced both heat labile and stable toxin. Of the 31 probe positive strains, 29 (94 %) were cultured from 11 volunteers with travellers' diarrhoea. A bacterial cause was thus determined in 6 1 % of the volunteers who experienced travellers' diarrhoea.
Eseherichia coli is generally known as a benign COmmensal of the healthy human intestinal tract. The flora usually consists of several serotypes fluctuating with time (1). Little is known about which ecological factors determine the changes in faecal flora, however both travel in foreign countries and intake of antibiotics are known to result in changes in flora (2). At least four types of Escherichia coli can cause gastro-intestinal infections: enterotoxigenic Escherichia coli producing heat labile toxin (ETEC LT), enterotoxigenic Escherichia coli producing heat stable toxin (ETEC ST), enteroinvasive Escherichia coli (EIEC), enterohaemorrhagic Escherichia coli (EHEC) and enteropathogenic Escherichia coli (EPEC)
(3).
In a previous study the efficacy of ciprofloxacin in preventing travellers' diarrhoea was determined (4). Fifty-four healthy volunteers travelled to
Sousse, Tunisia for eight days; 1 of 25 subjects (4 %) receiving ciprofloxacin versus 18 of 28 (64 %) subjects in the placebo group suffered from travellers' diarrhoea (p < 0.0001). This resulted in a 94 % protection level of the antimicrobial agent and indicated that travellers' diarrhoea in our volunteers was of bacterial aetiology. Despite multiple qualitative examination of diarrhoeal faeces samples, only one species of enteropathogenic bacteria was isolated (Campylobacter jejuni). This provided the impetus for closer examination of 750 Escherichia coli isolates obtained during the study period. Thus in the present study five Escherichia coli colonies of the volunteers were serotyped before, during and after travel. In addition, all isolates were tested using a set of four non-radioactively labelled D N A probes to detect the four major categories of diarrheogenic Escherichia col# ETEC LT, ETEC ST, EIEC and EPEC.
D e p a r t m e n t o f C h m c a l M i c r o b i o l o g y a n d L a b o r a t o r y of I n f e c t i o u s D i s e a s e s , 2 D e p a r t m e n t of I n t e r n a l M e d i c i n e ,
and 3U-gene Research, University Hospital, PO Box 85 500 3508 GA Utrecht The Netherlands
Materials and Methods
Protection, Department of Bacteriology, Bilthoven, The Netherlands.
Subjects. Our subjects were 54 healthy volunteers aged 19 to 50 (average age 28) who spent eight days in Tunisia
4 N a h•o n ,a l I n s t i t u t e o f Public , H e a l t h a n d E n•v i r o n m e n t a l
626
in September 1988. They participated in a randomized double-blind study testing ciprofloxacin in the prevention of travellers' diarrhoea (4). Twenty-six subjects received eiprofloxaein and 28 subjects received a placebo. Informed, written consent was obtained from all volunteers.
Laboratory Investigations. Routine stool specimens were obtained in the week prior to the study and 1 to 3 days after return, and follow-up samples 4 to 5 weeks after return from Tunisia. Five Escherichia coli colonies were selected from each faecal culture, placed in a peptone medium containing 25 % glycerol and stored at -70 °C. Stool specimens were also collected during an episode of diarrhoea in Tunisia. All samples were cultured on the day of collection. Five colonies suspected to be Escherichia coli from each faecal sample were inoculated in different tubes containing Iso-Sensitest agar (Oxoid, UK), incubated at 37 °C overnight and kept at 4 °C. The isolates were brought back to The Netherlands and after identification stored in the peptone glycerol medium at -70 °C (4).
E u r . J. Clin. M i c r o b i o l . Infect. Dis.
detected by an enzyme immunoassay using an antibodyconjugate and subsequent enzyme-catalyzed colour reaction with 5-bromo-4-ehloro-3-indolyl phosphate and nitroblue tetrazolium salt (DIG DNA Labelling and Detection Kit, Boehringer, FRG) (14). Experiments were carried out as specified by the manufacturer except that 50 lag of denaturated herring sperm DNA per ml hybridization solution was added and the concentration blocking reagent in buffer 2 was doubled to 1.0 %. The concentration of probe DNA in the hybridization solution was 80 ng/ml. Filters were read blind, i.e. before reviewing results of investigations to detect travellers' diarrhoea and examining serotypes. The vectors of the cloned probes were tested to exclude probe contamination by vector DNA.
Synthetic Oligonucleotide Probe. An alkaline-phosphatase-conjugated synthetic oligonucleotide probe was used to detect ETEC ST (SNAP Hybridization System NEP010, Dupont). The oligonucleotide is 26 bases long and homologous to the STh gene. Hybridization experiments were carried out as specified by the manufacturer.
Serotyping. All 750 Escherichia coli isolates were serotyped for the presence of O and K antigens by means of a mechanized microtechnique using all available O (O1171) antigens and O:K antisera against the K antigens usually associated with each O antigen (5).
DNA Probes. We used the recombinant plasmids pWD299, pHS5753 and pJPN16 encoding for ETEC LT (6), EIEC (7) and EPEC (8) respectively as the source of the probes. These were kindly provided by Mr Sansonetti, Institut Pasteur, Paris (9). All plasmids were maintained in Escherichia coli HB101, a non-toxigenic strain, and were isolated as described bY Birnboim and Doly (10). The ETEC LT probe consisted of a 850-bp HindllI digestion fragment which was cloned into vector pBR322, the EIEC probe was a 4.5-kbHindlII digestion fragment cloned into vector pACYC184 and the EPEC probe was a 1-kb BamH1-SalI fragment cloned into vector pBR322. The EPEC probe, also called the EAF probe (EPEC adherence factor), detects the adhesin which confers both adherence to HEp-2 cells and virulence in humans. The appropriate DNA fragments and vector DNA were separated e[ectrophoretically on 1% agarose gels containing ethidium bromide (500 ng/ml). Bands corresponding to the inserts were cut out and DNA was isolated using the Geneclcan method according to the conditions specified by the manufacturer (Bio 101, USA). Hybridization Assays. Escberichia coli isolates were grown overnight at 37 °C in Evans medium (11). 150 lal (approximately 3 x 108 cfu) were transferred to Gene Screen Plus membranes (Dupont, NEN Research Products, USA) using a Bio-Dot micro-filter apparatus (Bio Rad, USA). The HB101 Escherichia coli containing the target of the DNA probe, three probe positive and two non-toxigenic Escherichia coli strains were used as control strains per 10 x 10 cm blot. All control strains have been described previously with the exception of Shigella sp. BBD 85-1657 which was a field isolate (12). Bacterial DNA was denaturatcd with 0.5 M NaOH and steam as described by Notermans et al. (13). Filters were air dried, baked at 80 °C and stored at room temperature. Probe DNA was nonradioactively labelled by random primed incorporation of digoxigenin-labelled deoxyuridine-triphosphate. After hybridization the hybrids were
Statistical Analysis. Student's t-test for paired samples was used for comparing the mean number of different serotypes per set of five Escherichia coli colonies before, during and after travel.
Results
Clinical Symptoms. E i g h t e e n of the 28 v o l u n t e e r s w h o did n o t r e c e i v e a n t i b i o t i c p r o p h y l a x i s exp e r i e n c e d t r a v e l l e r s ' d i a r r h o e a . I n 11 c a s e s ( 6 1 % ) the d i a r r h o e a was a s s o c i a t e d with e n t e r o t o x i g e n i c Escherichia coli a n d in o n e case Campylobacter jejuni was also isolated. O f t h e s e p a t i e n t s , eight h a d mild clinical illness ( t h r e e or m o r e w a t e r y stools p e r day, f r e q u e n t b o w e l m o v e m e n t s a c c o m p a n i e d by c r a m p s ) a n d three, i n c l u d ing the o n e in w h o m Campylobacter jejuni was also isolated, suffered severe d i a r r h o e a l illness with v o m i t i n g a n d / o r fever > 38.0 °C. I n s e v e n v o l u n t e e r s d i a r r h o e a was of u n k n o w n aetiology. Clinical s y m p t o m s were the s a m e as in the g r o u p in which E T E C was p r e s e n t : five p e o p l e exp e r i e n c e d mild illness a n d two w e r e seriously ill. N o n e of the p e o p l e with t r a v e l l e r s ' d i a r r h o e a had b l o o d y stools. O n e v o l u n t e e r w h o r e c e i v e d antibiotic p r o p h y l a x i s h a d d i a r r h o e a , b u t n o Escherichia coli c o u l d be isolated. Diarrheogenic Escherichia coli. T h e results of h y b r i d i z a t i o n tests are s h o w n in T a b l e 1. O f 750
Escherichia coli isolates tested, 8 strains o n l y h y b r i d i z e d with the E T E C S T p r o b e , 13 o n l y h y b r i d i z e d with the E T E C L T p r o b e a n d 10 h y b r i d i z e d with b o t h the S T a n d the L T p r o b e s . All 31 E T E C p r o b e positive strains w e r e c u l t u r e d from d i a r r h o e a l s a m p l e s o b t a i n e d in T u n i s i a or
Vo1.10,1991
627
Table 1: Diarrheogenic Escherichia coil detected using probes in 11 subjects with diarrhoea (61% ) and two asymptomatic subjects. Numbers in brackets indicate numbers of subjects from whom Escherichia coil strains were isolated. Pre-travel strains a (n = 250)
Diarrhoea b and post-travelc strains
Follow-up strains d (n = 243)
(n :257) ETEC STh
0
8 (3)
0
ETEC LT
0
13 (7)~
0
ETEC STh + LT
0
10 (3)
0
EIEC inv
0
0
5 (1) f
EPEC EAF
4 (1)g
0
0
a Cultured pre-travel. bCultured during episodes of travellers' diarrhoea. c Cultured 1-3 days after travel. d Cultured on follow-up. e Including one asymptomatic person. f Rough strains. g Asymptomatic person. from routine samples taken immediately after return. Twenty-nine probe positive strains were found in 11 volunteers with diarrhoea. Pathogens were detected in I to 4 (mean 2.6) of the set of 5 Eseherichia colt isolates. T h e probe positive strains of persons with diarrhoea belonged to serotypes O6:K15, O8:K47, O27:K-, O39:K-, O40:K-, O64:K-, O148:K- and O159:K-. Two E T E C LT probe positive strains from an asymptomatic volunteer were serotype O64:K-. Table 2 shows the relation between toxin type and serotype. In one volunteer without symptoms, four E P E C probe positive strains of serotype O73:K18 were isolated in the routine pre-travel stool sample. In another subject, who suffered diarrhoea in Tunisia, five E I E C probe positive strains were isolated in the routine follow-up stool samples. These Were non-typeable, rough strains. Changes in Escherichia colt Serotypes. The composition of the Escherichia colt flora fluctuated Table 2: Serotypes of probe positive Escherichia coil
considerably throughout the study period. A large inter- and intra-individual variation in serotypes was found in the placebo group. T h e diarrhoeal samples and the samples taken shortly after return differed completely from the pre-travel samples, regardless of whether diarrhoea was present. T h e mean number of different serotypes per set of five colonies are shown in Table 3. C o m p a r e d to pretravel samples, the mean n u m b e r of different serotypes doubled from 1.7 pre-travel to 3.4 (p < 0.05) in diarrhoeal stool samples obtained in Tunisia. Shortly after return the mean was 2.7 both in the group with diarrhoea as well as in the group without symptoms. Follow-up samples showed the pre-travel mean number (1.7) but again a profound change of serotypes. In the ciprofloxacin group, the Escherichia colt cultured from pre-travel and follow-up samples showed also completely different serotypes but the mean number was in the same order of magnitude.
strains, isolated from volunteers with travellers' diarrhoea.
Discussion
Toxin type
Serotypes
ETEC LT
O8:K47; O39K-; O40K-; O64K-*
ETEC ST
O27:K-; O 148:K-; O159:K-
ETEC LT+ ST
O6:K15; O159:K-
Enterotoxigenic Escherichia cob play an important role in travellers' diarrhoea in tourists to African countries (15, 16). In 1986 Black (17) reported that enterotoxigenic Escherichia colt were found in a median of 36 % of ill travellers in aetiologic studies in Kenya and Morocco.
*This serotype was also isolated from an asymptomatic volunteer.
In tile present study an incidence of enterotoxigenic Escherichia colt of 6 1 % (11/18) was found
--.____.
628
Eur. J. Clin. Microbiol. Infect. Dis.
Table3: Mean number •f different Escherichia c•li ser•types per set •f five c•••nies is••ated fr•m each faecal sample. Numbers in brackets indicate numbers of subjects from whom Escherichia coli strains were isolated. Treatment
Pre-travelstrains a Diarrhoea strains u Post-travelstrainsc Follow-up strains d
Placebo
1.7 (28)
3.4 (18)
2.7 (28)
1.7 (27)
Ciprofloxaein
1.9 (26)
0
0
1.5 (24)
a Cultured bCultured c Cultured d Cultured
pre-travel. during episodes of travellers' diarrhoea. 1-3 days after travel. on follow-up.
in a group of travellers with diarrhoea in Tunisia. In this study five Escherichia coli colonies were chosen f r o m each diarrhoeal stool sample and examined using a colony hybridization technique with a set of four non-radioactively labelled D N A probes. Subsequently, results of serotyping revealed that one to four different serotypes of Escherichia coli strains were isolated per set of five colonies obtained f r o m each stool sample. This indicates that Escherichia coli isolates taken from diarrhoeal samples are not pure cultures of one serotype, and if m o r e than five strains per sample had b e e n examined m o r e episodes of diar r h o e a might have b e e n associated with a proven bacterial cause (18). O n e person without s y m p t o m s had two p r o b e positive strains in the stool sample taken immediately after return. T h e s e strains were of serotype O64:K- and were also found in a n o t h e r subject with diarrhoea. This serotype is known to be pathogenic in animals but its i m p o r t a n c e in h u m a n s has not b e e n established. Hybridization tests with p r o b e s for identification of enteroinvasive Escherichia coli and enteropathogenic Escherichia coli showed that travellers' diarrhoea in our study could not be associated with the presence of these pathogens. We did not screen for e n t e r o h a e m o r r h a g i c Escherichia coli because no episode of diarrhoea was acc o m p a n i e d by faecal blood loss and all cultures on sorbitol M a c C o n k e y agar were negative. T h e Escherichia coli flora changed considerably in all travellers regardless of whether diarrhoea was present or whether antibiotic prophylaxis was used, C o m p l e t e l y different serotypes to those found in the pre-travel cultures were isolated in the follow-up samples. Similar findings have b e e n described by o t h e r investigators and suggest that the e n v i r o n m e n t , in particular the diet, influences the intestinal flora (2, 19). O u r colony hybridization assay involved time-
consuming procedures. It would be of value to d e v e l o p a less time-consuming m e t h o d to screen faeces samples using a set of different probes. Moreover, a hybridization m e t h o d to identify pathogens directly in stool smears might provide m o r e information on the quantity of p a t h o g e n s present in the intestine. In conclusion, enterotoxigenic Escherichia coli were isolated in 6 1 % (11/18) of a group of volunteers who were affected by travellers' diarrhoea in Tunisia. Colony hybridization with non-radioactively D N A labelled p r o b e s was a useful tool for the detection and epidemiology of these pathogens in diarrhoeal illness. T h e question remains as to how m a n y colonies per stool specimen need to be screened to optimize the detection of diarrheogenic Escherichia coli in cases of travellers' diarrhoea.
Acknowledgement This study was supported by a grant from Bayer AG, Wuppertal, FRG.
References 1. Shooter RA, Bettelheim KA, Lennox-King SMJ, O'Farreli S: Escherichia coli scrotypes in the faeces of healthy adults over a period of several months. Journal of Hygiene 1977, 78: 95-98. 2. ~rskov IF,Sack RB, t~rskov 1, Froehlich JL: Changing fecal Escherichia coli flora during travel. European Journal of Clinical Microbiology and Infectious Diseases 1984, 3: 306--309. 3. Levine MM: Escherichia coli that cause diarrhea: cnterotoxigenic, enteropathogenic, enteroinvasive, enterohemorrhagic, and enteroadherent. Journal of Infectious Diseases 1987, 155: 377-389.
Vo1.10,1991
4. Rademaker CMA, Hoepelman IM, Wolfllagen MJHM, Beumer H, Rozenberg-Arska M, Verhoef J: Results of a double-blind placebo-controlled study using ciprofloxacin for prevention of travelers' diarrhea. European Journal of Clinical Microbiology and Infectious Diseases 1989, 6: 690-694. 5. Guinde PAM, Agterberg CM, Jansen WH: Escherichia colt 0 antigen typing by means of a mechanized microtechnique. Applied Microbiology 1972, 24: 127-131. 6. Dallas WS, Gill DM, Falkow S: Cistrons encoding Escherichia colt heat-labile toxin. Journal of Bacteriology 1979, 139: 850-858. 7. Boileau CR, d'Hauteville HM, Sansone|ti PJ: DNA hybridization technique to detect Shigella species and enteroinvasive Escherichia coil Journal of Clinical Microbiology 1984, 20: 959-961. 8. Nataro JP, Scaletsky l e A , Kaper JB, Levine MM, Trahulsi LR: Plasmid-mediated factors conferring diffuse and localized adherence of enteropathogenic Escherichia coll. Infection and Immunity 1985, 48: 378383. 9. Bt~h~xertMG, d'Hauteville ltM, Sansonetti PJ: Detection of enteric pathotypcs of Eschelqchia coli by hybridization using six DNA probes. Annales de l'Institut Pasteur/Microbiologie 1988, 139: 189-202. 10. Birnholm H e , Doly J: A rapid alkaline extraction procedure for screening recombinant plasmid DNA. Nucleic Acids Research 1979, 7: 1513-1523. 11. Evans DG, Evans D J, Gnbach SL: Identification of enterotoxigenic Escherichia colt and serum antitoxin activity by vascular permeability factor assay. Infection and Immunity 1973, 8: 731-735. 12. Guin6e PAM, Jansen WH: Detection of enterotoxigcnicity and attachment factors in Eschcrichia colt strains of human, porcine and bovine origin; a comparative study. Zentralblatt fiir Bakteriologie, Mikrobiologie und Hygiene 1979, 243: 245-257.
629
13. Notermans S, Heuvelman K J, Wernars K: Synthetic enterotoxin B DNA probes for detection of enterotoxigenic Staphylococcus aureus strains. Applied and Environmental Microbiology 1988, 54: 531-533. 14. Sommerfelt H, Grewal HMS, Bhan MK: Simplified and accurate nonradioactive polynucleotide gene probe assay for identification of enterotoxigenie Escherichia coil Journal of Clinical Microbiology 1990, 28: 49-54. 15. Sack DA, Kaminsky DC, Sack RB, ltotia JN, Arthur RR, Kapikian AZ, Orskov IF, Orskov 1: Prophylactic doxycycline for travelers' diarrhea: results of a prospective double-blind study of Peace Corps volunteers in Kenya. New England Journal of Medicine 1978, 298: 758--763. 16. Sack RB, Froehlich JL, Zulieh AW, Hidi DS, Kapikan AZ, Orsknv F, Orskov I, Greenberg HB: Prophylactic doxycycline for travelers' diarrhea: results of a prospective double-blind study of Peace Corps volunteers in Morocco. Gastroenterology 1979, 76: 1368-1378. 17. Black RE: Pathogens that cause travelers' diarrhea in Latin America and Africa. Reviews of Infectious Diseases 1986, 8, Supplement 2: 131-135. 18. Echeverria P, Taylor DN, Seriwatana J, Brown JE, Lexomboon U: Examination of colonies and stool blots for detection of enteropathogens by D N A hybridization with eight probes. Journal of Clinical Microbiology 1989, 27: 331-334. 19. Stenderup J, Orskov h ~rskov F: Changes in serotype and resistance pattern of the intestinal Escherichia colt flora during travel. Scandinavian Journal of Infectious Diseases 1983, 15: 367-373.
Vol. 10. No. 8
625
Analysis of Escherichia coli Isolates from Subjects with Travellers' Diarrhoea Using DNA Probes and Serotyping
C.M.A. Rademakerl,3*, M.R.L. Krul 4, W.H. Jansen 4, N.M. Vos 1, I.M. Hoepelmanl,2, 3, M. R o z e n b e r g _ A r s k a l , 3, j. Verhoefl, 3
Escherichia coli isolated from faeces o f 54 healthy volunteers who visited Tunisia for eight days were examined. These volunteers participated in a randomized double. blind placebo-controlled study to establish whether ciprofloxacin could prevent travellers' diarrhoea. Escherlchia coil strains isolated before travel, during episodes o f travellers' diarrhoea, immediately after return and five weeks after return were serotyped and tested for the presence of virulence genes indicating diarrheogenic properties by hybridization with a set of four non-radioactively labelled D N A probes. Subjects receiving ciprofloxacin prophylactically to prevent travellers' diarrhoea w e r e asymptomatic and no Escherichia coil could be cultured shortly after return home. Sixty-four percent of subjects (18/28) who did not receive antibiotic prophylaxis suffered from travellers' diarrhoea. Hybridization tests detected 8 enterotoxigenic Escherichia colt strains producing heat stable toxin, 13 enterotoxigenic strains producing heat labile toxin and 10 strains which produced both heat labile and stable toxin. Of the 31 probe positive strains, 29 (94 %) were cultured from 11 volunteers with travellers' diarrhoea. A bacterial cause was thus determined in 6 1 % of the volunteers who experienced travellers' diarrhoea.
Eseherichia coli is generally known as a benign COmmensal of the healthy human intestinal tract. The flora usually consists of several serotypes fluctuating with time (1). Little is known about which ecological factors determine the changes in faecal flora, however both travel in foreign countries and intake of antibiotics are known to result in changes in flora (2). At least four types of Escherichia coli can cause gastro-intestinal infections: enterotoxigenic Escherichia coli producing heat labile toxin (ETEC LT), enterotoxigenic Escherichia coli producing heat stable toxin (ETEC ST), enteroinvasive Escherichia coli (EIEC), enterohaemorrhagic Escherichia coli (EHEC) and enteropathogenic Escherichia coli (EPEC)
(3).
In a previous study the efficacy of ciprofloxacin in preventing travellers' diarrhoea was determined (4). Fifty-four healthy volunteers travelled to
Sousse, Tunisia for eight days; 1 of 25 subjects (4 %) receiving ciprofloxacin versus 18 of 28 (64 %) subjects in the placebo group suffered from travellers' diarrhoea (p < 0.0001). This resulted in a 94 % protection level of the antimicrobial agent and indicated that travellers' diarrhoea in our volunteers was of bacterial aetiology. Despite multiple qualitative examination of diarrhoeal faeces samples, only one species of enteropathogenic bacteria was isolated (Campylobacter jejuni). This provided the impetus for closer examination of 750 Escherichia coli isolates obtained during the study period. Thus in the present study five Escherichia coli colonies of the volunteers were serotyped before, during and after travel. In addition, all isolates were tested using a set of four non-radioactively labelled D N A probes to detect the four major categories of diarrheogenic Escherichia col# ETEC LT, ETEC ST, EIEC and EPEC.
D e p a r t m e n t o f C h m c a l M i c r o b i o l o g y a n d L a b o r a t o r y of I n f e c t i o u s D i s e a s e s , 2 D e p a r t m e n t of I n t e r n a l M e d i c i n e ,
and 3U-gene Research, University Hospital, PO Box 85 500 3508 GA Utrecht The Netherlands
Materials and Methods
Protection, Department of Bacteriology, Bilthoven, The Netherlands.
Subjects. Our subjects were 54 healthy volunteers aged 19 to 50 (average age 28) who spent eight days in Tunisia
4 N a h•o n ,a l I n s t i t u t e o f Public , H e a l t h a n d E n•v i r o n m e n t a l
626
in September 1988. They participated in a randomized double-blind study testing ciprofloxacin in the prevention of travellers' diarrhoea (4). Twenty-six subjects received eiprofloxaein and 28 subjects received a placebo. Informed, written consent was obtained from all volunteers.
Laboratory Investigations. Routine stool specimens were obtained in the week prior to the study and 1 to 3 days after return, and follow-up samples 4 to 5 weeks after return from Tunisia. Five Escherichia coli colonies were selected from each faecal culture, placed in a peptone medium containing 25 % glycerol and stored at -70 °C. Stool specimens were also collected during an episode of diarrhoea in Tunisia. All samples were cultured on the day of collection. Five colonies suspected to be Escherichia coli from each faecal sample were inoculated in different tubes containing Iso-Sensitest agar (Oxoid, UK), incubated at 37 °C overnight and kept at 4 °C. The isolates were brought back to The Netherlands and after identification stored in the peptone glycerol medium at -70 °C (4).
E u r . J. Clin. M i c r o b i o l . Infect. Dis.
detected by an enzyme immunoassay using an antibodyconjugate and subsequent enzyme-catalyzed colour reaction with 5-bromo-4-ehloro-3-indolyl phosphate and nitroblue tetrazolium salt (DIG DNA Labelling and Detection Kit, Boehringer, FRG) (14). Experiments were carried out as specified by the manufacturer except that 50 lag of denaturated herring sperm DNA per ml hybridization solution was added and the concentration blocking reagent in buffer 2 was doubled to 1.0 %. The concentration of probe DNA in the hybridization solution was 80 ng/ml. Filters were read blind, i.e. before reviewing results of investigations to detect travellers' diarrhoea and examining serotypes. The vectors of the cloned probes were tested to exclude probe contamination by vector DNA.
Synthetic Oligonucleotide Probe. An alkaline-phosphatase-conjugated synthetic oligonucleotide probe was used to detect ETEC ST (SNAP Hybridization System NEP010, Dupont). The oligonucleotide is 26 bases long and homologous to the STh gene. Hybridization experiments were carried out as specified by the manufacturer.
Serotyping. All 750 Escherichia coli isolates were serotyped for the presence of O and K antigens by means of a mechanized microtechnique using all available O (O1171) antigens and O:K antisera against the K antigens usually associated with each O antigen (5).
DNA Probes. We used the recombinant plasmids pWD299, pHS5753 and pJPN16 encoding for ETEC LT (6), EIEC (7) and EPEC (8) respectively as the source of the probes. These were kindly provided by Mr Sansonetti, Institut Pasteur, Paris (9). All plasmids were maintained in Escherichia coli HB101, a non-toxigenic strain, and were isolated as described bY Birnboim and Doly (10). The ETEC LT probe consisted of a 850-bp HindllI digestion fragment which was cloned into vector pBR322, the EIEC probe was a 4.5-kbHindlII digestion fragment cloned into vector pACYC184 and the EPEC probe was a 1-kb BamH1-SalI fragment cloned into vector pBR322. The EPEC probe, also called the EAF probe (EPEC adherence factor), detects the adhesin which confers both adherence to HEp-2 cells and virulence in humans. The appropriate DNA fragments and vector DNA were separated e[ectrophoretically on 1% agarose gels containing ethidium bromide (500 ng/ml). Bands corresponding to the inserts were cut out and DNA was isolated using the Geneclcan method according to the conditions specified by the manufacturer (Bio 101, USA). Hybridization Assays. Escberichia coli isolates were grown overnight at 37 °C in Evans medium (11). 150 lal (approximately 3 x 108 cfu) were transferred to Gene Screen Plus membranes (Dupont, NEN Research Products, USA) using a Bio-Dot micro-filter apparatus (Bio Rad, USA). The HB101 Escherichia coli containing the target of the DNA probe, three probe positive and two non-toxigenic Escherichia coli strains were used as control strains per 10 x 10 cm blot. All control strains have been described previously with the exception of Shigella sp. BBD 85-1657 which was a field isolate (12). Bacterial DNA was denaturatcd with 0.5 M NaOH and steam as described by Notermans et al. (13). Filters were air dried, baked at 80 °C and stored at room temperature. Probe DNA was nonradioactively labelled by random primed incorporation of digoxigenin-labelled deoxyuridine-triphosphate. After hybridization the hybrids were
Statistical Analysis. Student's t-test for paired samples was used for comparing the mean number of different serotypes per set of five Escherichia coli colonies before, during and after travel.
Results
Clinical Symptoms. E i g h t e e n of the 28 v o l u n t e e r s w h o did n o t r e c e i v e a n t i b i o t i c p r o p h y l a x i s exp e r i e n c e d t r a v e l l e r s ' d i a r r h o e a . I n 11 c a s e s ( 6 1 % ) the d i a r r h o e a was a s s o c i a t e d with e n t e r o t o x i g e n i c Escherichia coli a n d in o n e case Campylobacter jejuni was also isolated. O f t h e s e p a t i e n t s , eight h a d mild clinical illness ( t h r e e or m o r e w a t e r y stools p e r day, f r e q u e n t b o w e l m o v e m e n t s a c c o m p a n i e d by c r a m p s ) a n d three, i n c l u d ing the o n e in w h o m Campylobacter jejuni was also isolated, suffered severe d i a r r h o e a l illness with v o m i t i n g a n d / o r fever > 38.0 °C. I n s e v e n v o l u n t e e r s d i a r r h o e a was of u n k n o w n aetiology. Clinical s y m p t o m s were the s a m e as in the g r o u p in which E T E C was p r e s e n t : five p e o p l e exp e r i e n c e d mild illness a n d two w e r e seriously ill. N o n e of the p e o p l e with t r a v e l l e r s ' d i a r r h o e a had b l o o d y stools. O n e v o l u n t e e r w h o r e c e i v e d antibiotic p r o p h y l a x i s h a d d i a r r h o e a , b u t n o Escherichia coli c o u l d be isolated. Diarrheogenic Escherichia coli. T h e results of h y b r i d i z a t i o n tests are s h o w n in T a b l e 1. O f 750
Escherichia coli isolates tested, 8 strains o n l y h y b r i d i z e d with the E T E C S T p r o b e , 13 o n l y h y b r i d i z e d with the E T E C L T p r o b e a n d 10 h y b r i d i z e d with b o t h the S T a n d the L T p r o b e s . All 31 E T E C p r o b e positive strains w e r e c u l t u r e d from d i a r r h o e a l s a m p l e s o b t a i n e d in T u n i s i a or
Vo1.10,1991
627
Table 1: Diarrheogenic Escherichia coil detected using probes in 11 subjects with diarrhoea (61% ) and two asymptomatic subjects. Numbers in brackets indicate numbers of subjects from whom Escherichia coil strains were isolated. Pre-travel strains a (n = 250)
Diarrhoea b and post-travelc strains
Follow-up strains d (n = 243)
(n :257) ETEC STh
0
8 (3)
0
ETEC LT
0
13 (7)~
0
ETEC STh + LT
0
10 (3)
0
EIEC inv
0
0
5 (1) f
EPEC EAF
4 (1)g
0
0
a Cultured pre-travel. bCultured during episodes of travellers' diarrhoea. c Cultured 1-3 days after travel. d Cultured on follow-up. e Including one asymptomatic person. f Rough strains. g Asymptomatic person. from routine samples taken immediately after return. Twenty-nine probe positive strains were found in 11 volunteers with diarrhoea. Pathogens were detected in I to 4 (mean 2.6) of the set of 5 Eseherichia colt isolates. T h e probe positive strains of persons with diarrhoea belonged to serotypes O6:K15, O8:K47, O27:K-, O39:K-, O40:K-, O64:K-, O148:K- and O159:K-. Two E T E C LT probe positive strains from an asymptomatic volunteer were serotype O64:K-. Table 2 shows the relation between toxin type and serotype. In one volunteer without symptoms, four E P E C probe positive strains of serotype O73:K18 were isolated in the routine pre-travel stool sample. In another subject, who suffered diarrhoea in Tunisia, five E I E C probe positive strains were isolated in the routine follow-up stool samples. These Were non-typeable, rough strains. Changes in Escherichia colt Serotypes. The composition of the Escherichia colt flora fluctuated Table 2: Serotypes of probe positive Escherichia coil
considerably throughout the study period. A large inter- and intra-individual variation in serotypes was found in the placebo group. T h e diarrhoeal samples and the samples taken shortly after return differed completely from the pre-travel samples, regardless of whether diarrhoea was present. T h e mean number of different serotypes per set of five colonies are shown in Table 3. C o m p a r e d to pretravel samples, the mean n u m b e r of different serotypes doubled from 1.7 pre-travel to 3.4 (p < 0.05) in diarrhoeal stool samples obtained in Tunisia. Shortly after return the mean was 2.7 both in the group with diarrhoea as well as in the group without symptoms. Follow-up samples showed the pre-travel mean number (1.7) but again a profound change of serotypes. In the ciprofloxacin group, the Escherichia colt cultured from pre-travel and follow-up samples showed also completely different serotypes but the mean number was in the same order of magnitude.
strains, isolated from volunteers with travellers' diarrhoea.
Discussion
Toxin type
Serotypes
ETEC LT
O8:K47; O39K-; O40K-; O64K-*
ETEC ST
O27:K-; O 148:K-; O159:K-
ETEC LT+ ST
O6:K15; O159:K-
Enterotoxigenic Escherichia cob play an important role in travellers' diarrhoea in tourists to African countries (15, 16). In 1986 Black (17) reported that enterotoxigenic Escherichia colt were found in a median of 36 % of ill travellers in aetiologic studies in Kenya and Morocco.
*This serotype was also isolated from an asymptomatic volunteer.
In tile present study an incidence of enterotoxigenic Escherichia colt of 6 1 % (11/18) was found
--.____.
628
Eur. J. Clin. Microbiol. Infect. Dis.
Table3: Mean number •f different Escherichia c•li ser•types per set •f five c•••nies is••ated fr•m each faecal sample. Numbers in brackets indicate numbers of subjects from whom Escherichia coli strains were isolated. Treatment
Pre-travelstrains a Diarrhoea strains u Post-travelstrainsc Follow-up strains d
Placebo
1.7 (28)
3.4 (18)
2.7 (28)
1.7 (27)
Ciprofloxaein
1.9 (26)
0
0
1.5 (24)
a Cultured bCultured c Cultured d Cultured
pre-travel. during episodes of travellers' diarrhoea. 1-3 days after travel. on follow-up.
in a group of travellers with diarrhoea in Tunisia. In this study five Escherichia coli colonies were chosen f r o m each diarrhoeal stool sample and examined using a colony hybridization technique with a set of four non-radioactively labelled D N A probes. Subsequently, results of serotyping revealed that one to four different serotypes of Escherichia coli strains were isolated per set of five colonies obtained f r o m each stool sample. This indicates that Escherichia coli isolates taken from diarrhoeal samples are not pure cultures of one serotype, and if m o r e than five strains per sample had b e e n examined m o r e episodes of diar r h o e a might have b e e n associated with a proven bacterial cause (18). O n e person without s y m p t o m s had two p r o b e positive strains in the stool sample taken immediately after return. T h e s e strains were of serotype O64:K- and were also found in a n o t h e r subject with diarrhoea. This serotype is known to be pathogenic in animals but its i m p o r t a n c e in h u m a n s has not b e e n established. Hybridization tests with p r o b e s for identification of enteroinvasive Escherichia coli and enteropathogenic Escherichia coli showed that travellers' diarrhoea in our study could not be associated with the presence of these pathogens. We did not screen for e n t e r o h a e m o r r h a g i c Escherichia coli because no episode of diarrhoea was acc o m p a n i e d by faecal blood loss and all cultures on sorbitol M a c C o n k e y agar were negative. T h e Escherichia coli flora changed considerably in all travellers regardless of whether diarrhoea was present or whether antibiotic prophylaxis was used, C o m p l e t e l y different serotypes to those found in the pre-travel cultures were isolated in the follow-up samples. Similar findings have b e e n described by o t h e r investigators and suggest that the e n v i r o n m e n t , in particular the diet, influences the intestinal flora (2, 19). O u r colony hybridization assay involved time-
consuming procedures. It would be of value to d e v e l o p a less time-consuming m e t h o d to screen faeces samples using a set of different probes. Moreover, a hybridization m e t h o d to identify pathogens directly in stool smears might provide m o r e information on the quantity of p a t h o g e n s present in the intestine. In conclusion, enterotoxigenic Escherichia coli were isolated in 6 1 % (11/18) of a group of volunteers who were affected by travellers' diarrhoea in Tunisia. Colony hybridization with non-radioactively D N A labelled p r o b e s was a useful tool for the detection and epidemiology of these pathogens in diarrhoeal illness. T h e question remains as to how m a n y colonies per stool specimen need to be screened to optimize the detection of diarrheogenic Escherichia coli in cases of travellers' diarrhoea.
Acknowledgement This study was supported by a grant from Bayer AG, Wuppertal, FRG.
References 1. Shooter RA, Bettelheim KA, Lennox-King SMJ, O'Farreli S: Escherichia coli scrotypes in the faeces of healthy adults over a period of several months. Journal of Hygiene 1977, 78: 95-98. 2. ~rskov IF,Sack RB, t~rskov 1, Froehlich JL: Changing fecal Escherichia coli flora during travel. European Journal of Clinical Microbiology and Infectious Diseases 1984, 3: 306--309. 3. Levine MM: Escherichia coli that cause diarrhea: cnterotoxigenic, enteropathogenic, enteroinvasive, enterohemorrhagic, and enteroadherent. Journal of Infectious Diseases 1987, 155: 377-389.
Vo1.10,1991
4. Rademaker CMA, Hoepelman IM, Wolfllagen MJHM, Beumer H, Rozenberg-Arska M, Verhoef J: Results of a double-blind placebo-controlled study using ciprofloxacin for prevention of travelers' diarrhea. European Journal of Clinical Microbiology and Infectious Diseases 1989, 6: 690-694. 5. Guinde PAM, Agterberg CM, Jansen WH: Escherichia colt 0 antigen typing by means of a mechanized microtechnique. Applied Microbiology 1972, 24: 127-131. 6. Dallas WS, Gill DM, Falkow S: Cistrons encoding Escherichia colt heat-labile toxin. Journal of Bacteriology 1979, 139: 850-858. 7. Boileau CR, d'Hauteville HM, Sansone|ti PJ: DNA hybridization technique to detect Shigella species and enteroinvasive Escherichia coil Journal of Clinical Microbiology 1984, 20: 959-961. 8. Nataro JP, Scaletsky l e A , Kaper JB, Levine MM, Trahulsi LR: Plasmid-mediated factors conferring diffuse and localized adherence of enteropathogenic Escherichia coll. Infection and Immunity 1985, 48: 378383. 9. Bt~h~xertMG, d'Hauteville ltM, Sansonetti PJ: Detection of enteric pathotypcs of Eschelqchia coli by hybridization using six DNA probes. Annales de l'Institut Pasteur/Microbiologie 1988, 139: 189-202. 10. Birnholm H e , Doly J: A rapid alkaline extraction procedure for screening recombinant plasmid DNA. Nucleic Acids Research 1979, 7: 1513-1523. 11. Evans DG, Evans D J, Gnbach SL: Identification of enterotoxigenic Escherichia colt and serum antitoxin activity by vascular permeability factor assay. Infection and Immunity 1973, 8: 731-735. 12. Guin6e PAM, Jansen WH: Detection of enterotoxigcnicity and attachment factors in Eschcrichia colt strains of human, porcine and bovine origin; a comparative study. Zentralblatt fiir Bakteriologie, Mikrobiologie und Hygiene 1979, 243: 245-257.
629
13. Notermans S, Heuvelman K J, Wernars K: Synthetic enterotoxin B DNA probes for detection of enterotoxigenic Staphylococcus aureus strains. Applied and Environmental Microbiology 1988, 54: 531-533. 14. Sommerfelt H, Grewal HMS, Bhan MK: Simplified and accurate nonradioactive polynucleotide gene probe assay for identification of enterotoxigenie Escherichia coil Journal of Clinical Microbiology 1990, 28: 49-54. 15. Sack DA, Kaminsky DC, Sack RB, ltotia JN, Arthur RR, Kapikian AZ, Orskov IF, Orskov 1: Prophylactic doxycycline for travelers' diarrhea: results of a prospective double-blind study of Peace Corps volunteers in Kenya. New England Journal of Medicine 1978, 298: 758--763. 16. Sack RB, Froehlich JL, Zulieh AW, Hidi DS, Kapikan AZ, Orsknv F, Orskov I, Greenberg HB: Prophylactic doxycycline for travelers' diarrhea: results of a prospective double-blind study of Peace Corps volunteers in Morocco. Gastroenterology 1979, 76: 1368-1378. 17. Black RE: Pathogens that cause travelers' diarrhea in Latin America and Africa. Reviews of Infectious Diseases 1986, 8, Supplement 2: 131-135. 18. Echeverria P, Taylor DN, Seriwatana J, Brown JE, Lexomboon U: Examination of colonies and stool blots for detection of enteropathogens by D N A hybridization with eight probes. Journal of Clinical Microbiology 1989, 27: 331-334. 19. Stenderup J, Orskov h ~rskov F: Changes in serotype and resistance pattern of the intestinal Escherichia colt flora during travel. Scandinavian Journal of Infectious Diseases 1983, 15: 367-373.
