Urban dust fecal pollution in Mexico City: Antibiotic resistance and virulence factors of Escherichia coli

Urban dust fecal pollution in Mexico City: Antibiotic resistance and virulence factors of Escherichia coli

ARTICLE IN PRESS Int. J. Hyg. Environ.-Health 209 (2006) 461–470 www.elsevier.de/ijheh Urban dust fecal pollution in Mexico City: Antibiotic resista...

563KB Sizes 0 Downloads 37 Views

ARTICLE IN PRESS

Int. J. Hyg. Environ.-Health 209 (2006) 461–470 www.elsevier.de/ijheh

Urban dust fecal pollution in Mexico City: Antibiotic resistance and virulence factors of Escherichia coli Irma Rosasa,, Eva Salinasa, Leticia Martı´ neza, Edmundo Calvab, Alejandro Craviotoc, Carlos Eslavac, Carlos F. Ama´bile-Cuevasd a

Centro de Ciencias de la Atmo´sfera, Universidad Nacional Auto´noma de Me´xico, Circuito Exterior, Me´xico D. F., 04510, Me´xico Instituto de Biotecnologı´a, Universidad Nacional Auto´noma de Me´xico, Me´xico c Facultad de Medicina, Universidad Nacional Auto´noma de Me´xico, Me´xico d Fundacio´n Lusara, Mexico City, Mexico b

Received 18 March 2005; received in revised form 6 March 2006; accepted 19 March 2006

Abstract Fecal pollution of settled dust samples from indoor and outdoor urban environments, was measured and characterized by the presence of fecal coliforms (FC), and by the characterization of Escherichia coli virulence genes, adherence and antibiotic resistance traits as markers. There were more FC indoors than outdoors (mean values 1089 and 435 MPN/g). Among indoor samples, there were more FC in houses with carpets and/or pets. Using a PCR-based assay for six enteropathogenicity genes (belonging to the EAEC, EHEC and EPEC pathotypes) on randomly selected E. coli isolates, there was no significant difference between isolates from indoors and outdoors (60% and 72% positive to at least one gene). The serotypes commonly associated with pathogenic strains, such as O86 and O28, were found in the indoor isolates; whereas O4, O66 and O9 were found amongst outdoor isolates. However, there were significantly more outdoor isolates resistant to at least one antibiotic (73% vs. 45% from indoors) among the strains positive for virulence factors, and outdoor isolates were more commonly multiresistant. There was no relationship between the presence of virulence genes and resistance traits. These results indicate that outdoor fecal bacteria were more likely from human sources, and those found indoors were related to pets and maintained in carpets. This study illustrates the risk posed by fecal bacteria from human sources, usually bearing virulence and resistance traits. Furthermore, the high prevalence of strains carrying genes associated to EAEC or EHEC pathotypes, in both indoor and outdoor environments, adds to the health risk. r 2006 Elsevier GmbH. All rights reserved. Keywords: Escherichia coli; Environment; Dust; Virulence; Antibiotic resistance

Introduction Fecal pollution is an important health problem mainly in developing countries, where enteric pathogens are Corresponding author. Tel.: +52 55 56225212.

E-mail address: [email protected] (I. Rosas). 1438-4639/$ - see front matter r 2006 Elsevier GmbH. All rights reserved. doi:10.1016/j.ijheh.2006.03.007

disseminated throughout the environment, associated both to the increase in human and animal populations and to the deficient management of fecal waste (Rosas et al., 1997; Sandhu et al., 1999; Wouters et al., 2000). Gastrointestinal pathogens have been identified in agriculture soil and water bodies (Brookes et al., 2004; Gessel et al., 2004). However, microbiological

ARTICLE IN PRESS 462

I. Rosas et al. / Int. J. Hyg. Environ.-Health 209 (2006) 461–470

studies in urban areas to determine the environmental sources of food and water contamination have received minimal attention. House dust has been recognized as a possible reservoir of both toxic and biological components, and inhalation and mouth–hand activities are important direct routes of contamination (Kjaer et al., 2002; Larson and Gomez-Duarte, 2001; von Lindern et al., 2002). Escherichia coli is an important pathogen of diarrheal disease. Pathogenic strains include many different serotypes, categorized into five major groups according to virulence mechanisms: enteropathogenic (EPEC), enterotoxigenic (ETEC), enteroinvasive (EIEC), enterohemorraghic (EHEC) and enteroaggregative (EAEC) (Puente and Finlay, 2001; Nataro et al., 1998; Rich et al., 1999). Other strains, namely diffusely adherent E. coli (DAEC) and cell-detaching E. coli (CDEC), are less well-established as pathogens (Jallat et al., 1994; Okeke et al., 2002). In a previous study from our laboratory, we showed that these bacteria were associated to suspended and settled dust from both indoor and outdoor environments (Rosas et al., 1997). An important risk factor for human health is the survival in the environment of bacteria harboring both virulence and antibiotic resistance genes, which could be transferred to other bacteria horizontally (Escobar-Pa´ramo et al., 2004; Puente and Finlay, 2001; Schwartz et al., 2003). In this study, the fecal coliform (FC) concentrations of urban dust samples and the isolation of E. coli strains were used as indicators of fecal pollution. Also, the presence of different characteristics associated to E. coli pathotypes and antibiotic resistance patterns were determined for both indoor and outdoor isolated strains in order to assess their potential health risk.

Material and methods Environmental sampling Samples were collected in the southeastern area of Mexico City: 30 homes of voluntary residents were included. All the homes were located 25 km from the urban-rural area of the dried basin of the Texcoco Lake. The study area had a high population density and insufficient public services. Indoor sampling: two indoor samples from each home were collected during the dry season in 2001 (one in February and another in May). The settled dust samples were obtained in the living room, on an area of 2 m2 during 5 min, using a vacuum cleaner (VK 121, Vorwerk, Spain) with sterile bags. Characteristics such as carpets, pets, visible growth mold, number of windows and doors as well as number of occupants were registered as answered by adult family members.

Outdoor sampling: simultaneously to indoor sampling, soil samples were collected in front of each house (2 m2) on a plastic dustpan and a sterile brush. The samples were thoroughly mixed and sieved through Tyler 30, 50 mesh sieves, to remove fibers and large particles.

Coliform evaluation and E. coli isolation Five-hundred milligrams of dust was suspended in 10 ml of diluent (0.01% Tween 80, 1% bacteriological peptone and 2% inositol) and vortexed for 1 min. To quantify FC, serial dilutions (1:10) were performed and analyzed by the multiple tube fermentation technique (American Public Health Association, 1985). An aliquot of positive tubes was plated on McConkey agar and the colonies considered as presumptive E. coli were isolated. The isolated bacteria were identified using an automated method (BioMe´rieux Vitek, Inc., Hazelwood, MO), with Gram negative identification cards (GNI+). Five confirmed E. coli strains from each sample were selected for the subsequent analysis.

Characterization of E. coli strains: virulence factors Strains were screened by a multiplex PCR assay for the presence of 15 different genes (Table 1) to identify diarrheogenic E. coli of the ETEC, EPEC, EIEC and EHEC groups (Lo´pez-Saucedo et al., 2003). To identify EAEC strains, a multiplex PCR using primers for the aggC, aap genes and for the pCVD432 plasmid, was standardized in this study. Each PCR tube contained 50 ml of reaction mix (10 mM Tris–HCl (pH 8.3), 50 mM KCl, 1.5 mM MgCl2, 500 mM of deoxynucleosides triphosphates, 1.25 U of Taq DNA polymerase (Invitrogen, Life Technologies, Carlsbad, CA), a mixture of the six primers (Table 1) and 10 ml of a 1:100 dilution of DNA in sterile water. The PCR program was 94 1C (3 min, 1 cycle); 50, 72 and 94 1C (1 min each temperature, 35 cycles); and a final extension step (8 min, 72 1C). Individual PCRs for the astA, aggA, aafA, aggR and pet genes were also performed (Gioppo et al., 2000; Eslava et al., 1998). Aliquots (10 ml) of the PCR products were separated by gel electrophoresis through a 2% horizontal agarose gel in TAE buffer for 1 h. Gels were stained with ethidium bromide, visualized under UV transillumination and photographed with a DC290 zoom digital camera (Kodak, Rochester, NY). The figures were produced with a computer graphic program (Kodak ID Image Analysis Software).

HEp-2 cells adherence assay Strains positive for at least one virulence factor were analyzed for adherence by the HEp-2 cell assay as

ARTICLE IN PRESS I. Rosas et al. / Int. J. Hyg. Environ.-Health 209 (2006) 461–470

Table 1.

463

PCR amplification for different gene loci of diarrheogenic Escherichia coli strains

Reference strains

Locus

Primers sequence (50 –30 )

Amplicon size (bp)

Concentration (pMol) in mix

ETEC O78:H11

lt

CGGTCTCTATATTCCCTGTT GGCGACAGATTATACCGTGC ATTTTTCTTTCTGTATTGTCTT CACCCGGTACAAGCAGGATT AATGGTGCTTGCGCTTGCTGC GCCGCTTTATCCAACCTGGTA GACCCGGCACAAGCATAAGC CCACCTGCAGCAACAAGAGG GACCCGGCACAAGCATAAGC CCACCTGCAGCAACAAGAGG CTGGATTTAATGTCGCATAGTG AGAACGCCCACTGAGATCATC GGCACTGTCTGAACCTGCTCC TCGCCAGTTATCTGACATTCT GGTATGATGATGATGAGTCCA GGAGGCCAACAATTATTTCC

450

5.0

190

6.47

324

2.5

385

3.88

385

3.88

150

3.88

255

2.5

650

10.25

111

50

548

1  103

432

1  103

538

10

550

20  103

232

50

308

1  103

630

50

st EPEC O127:H6

bfpA eaeA

EHEC O157:H7

eaeA stx1 stx2

EIEC O124NM

ial

EAEC O42 17-2

astA pet aggA aggC aafA aap aggR pCVD432

CCATCAACACAGTATATCCGA GGTCGCGAGTGACGGCTTTGT GTATCAAAAATGCCCCTCTGGTAA ATTCGCCCCCTTGTGGATAGTCA GCGTTAGAAAGACCTCCAATA GCCGGATCCTTAAAAATTAATTCCGGC TATTAAACCATGGTAGCG GCCAAGATCCGAGATTGA ACATGCATGCAAAAAATCAGAATGTTTGTT CGGGATCCATTTGTCACAAGCTCAGC CTTTTCTGGCATCTTGGGT GTAACAACCCCTTTGGAAGT CTAATTGTACAATCGATGTA ATGAAGTAATTCTTGAAT CTGGCGAAAGACTGTATCAT CAATGTATAGAAATCCGCTGTT

previously described (Cravioto et al., 1991). Positive and negative control adherent strains were: E. coli E2348/69 (localized); E. coli 49766 (diffuse); O42 (aggregative) and E. coli K12 HB101 (non-adherent). The adherence patterns were analyzed according to Scaletsky et al. (1984) and Nataro et al. (1987).

Antibiotic susceptibility assay The susceptibility of the E. coli strains to 13 antimicrobial agents (Table 4) was determined by an automated procedure, using Vitek-Gram negative susceptibility cards (BioMe´rieux Vitek Inc., Hazelwood, MO).

Serotyping

Integron detection

Serotyping of strains positive for at least one virulence factor was performed using rabbit antisera (SERUNAM; Mexico), prepared against the 175 somatic (O) and 56 flagellar (H) antigens. The microagglutination test was used to determine the bacterial serogroup (Orskov and Orskov, 1984).

The presence of class 1 integrons among the E. coli isolates was assessed by a PCR-based assay (Le´vesque and Roy, 1993); this assay was performed only upon strains resistant to sulfonamides, as determined by disk diffusion tests, performed in addition to the procedures above.

ARTICLE IN PRESS 464

I. Rosas et al. / Int. J. Hyg. Environ.-Health 209 (2006) 461–470

Distribution of virulence factors and antibiotic resistance among E. coli

Fecal coliforms MPN X 10³/g

6 Indoor Outdoor

5 4 3 2 1 0 -1 0

20

40 60 Percentile

80

100

Fig. 1. Accumulative frequency of FC concentrations in outdoor and indoor environments. The detection limit was 3 MPN/g.

Results Correlation between dust FC concentrations from indoor and outdoors FC concentrations for outdoor and indoor samples are shown in Fig. 1. FC concentrations and their distribution according to house characteristics can be found in Table 2. No statistically significant correlation was found between FC concentrations from indoor and outdoor samples (r ¼ 0:39, p40:05), thus R2 ¼ 0:15. Significant differences were observed between indoor and outdoor FC concentrations (Z ¼ 2:9, po0:05).

A multiplex PCR assay was used to identify an array of E. coli diarrheogenic virulence factors. One or two virulence factors could be detected in 60% of the indoor and in 72% of the outdoor E. coli strains (Table 3; Figs. 2 and 3). Among these, amplification from different EAEC primers was the most common feature, and similar for both indoor and outdoor strains (37%, 38%) followed by EHEC (22%, 34%) and EPEC (2%, 0%) (Table 3; Fig. 2). No significant differences were observed among the percentage of virulence genes according to their site of isolation, neither indoor nor outdoor (X2 ¼ 3.7; p40:05), but it was significant when the antibiotic resistance of each group of virulence factors was considered (X 2 ¼ 11:4; po0:05), with higher values for outdoor. The antibiotic susceptibility assay showed a total of 54% and 67% of resistant E. coli strains from indoor and outdoor, respectively (Table 3). Among the resistant outdoor strains, 61 of 83 (73%) harbored a virulence factor associated to either EAEC or EHEC, with 79% of EHEC resistant to at least one antibiotic. A lower proportion (45%) of the indoor strains was antibiotic resistant and positive for the tested primers. Class 1 integrons were found mostly among outdoor isolates: 16 of 116 (14%) vs. 3 of 183 (2%); all amplicons were about the same size (1 kb). Integrons were found in approximately the same ratio amongst outdoor strains harboring or not virulence factors (5/33, 15% vs. 11/83, 13%, respectively).

Antibiotic resistance among E. coli strains with adherence patterns and virulence factors Comparison of dust FC, E. coli isolation and presence of virulence factors according to home characteristics Table 2 shows the percentage of samples with FC values of more than 3 MPN/g (detection limit) and their concentrations. Sixty-one percent of indoor dust samples presented FC 4 3 MPN/g with a geometric mean (GM) of 131, and 53% of the outdoor samples with a GM of 27. E. coli could be isolated from all FC positive indoor samples and from 75% of the outdoor samples. Significant differences (po0:05) were observed between FC concentrations of homes with and without carpets, as well as those with pets. Higher FC values were found in homes with carpets compared to those with pets. However, the percentage of E. coli that harbored virulence genes was higher in homes with pets (Table 2).

The E. coli strains that were positive for some of the virulence factors tested were classified in three adherence patterns: diffuse adherence (DA), chain-like adherence (CLA) and aggregative adherence to cover slip (AAcs). The DA pattern was the most commonly observed among these strains (Table 4). The DA pattern was present in 77% (7/9) of the serotypes that were isolated from indoors and in 50% (6/12) of those from outdoors. No association of the DA pattern was observed in connection with the presence of either aap, astA, pCVD432/astA or stx1. AAcs was observed in 11% (1/9) of the indoor strains, which were positive for aap, and in 33% (4/12) of the outdoor strains, all of which were positive for stx1. The cell-detaching (CD) phenotype was observed in 8% (1/12) strains from outdoors. On the other hand, antibiotic multiresistance was associated with strains positive for some virulence

ARTICLE IN PRESS I. Rosas et al. / Int. J. Hyg. Environ.-Health 209 (2006) 461–470

Table 2.

465

Fecal coliform concentrations and percentage of E. coli isolates from indoor and outdoor environments

House characteristics

E. coli (MPN)

Fecal coliform (MPN/g) Houses

Samples 43a (%)

Mean7SD

Samples (%)

Positive for virulence genes (%)

227671862 743 4557618 53 po0.05c

16/16 (100)

45/80 (56)

21/21(100)

65/103 (63)

148271576 456 6587119 52c po0.05c 4357660 27 108971457 131 po0.05c

20/20 (100)

68/98 (69)

17/17 (100)

42/85 (49)

24/32 (75)

83/116 (72)

37/37 (100)

110/183 (60)

b

GM Carpet Yes

10

16/20 (80)

No

20

21/40 (53)

Pets Yes

12

20/24 (83)

No

18

17/36 (47)

Outdoors

30

32/60 (53)

Indoors

30

37/60 (61)

a

Detection limit. GM: geometric mean, MPN/g. c Mann–Whitney test. b

Table 3.

Virulence factors present in urban dust strains and their relation to antibiotic resistance to at least one antibiotic

E. coli pathotype

Indoor

Outdoor

Virulence factors (%)

Resistant (%)

Virulence factors (%)

Resistant (%)

EAEC: aap astA aggC pCVD432 pCVD432/astA aggC/aap pCVD432/aap aap/astA

67/183 (37) 51/67 (76) 11/67 (16) 0/67 (0) 1/67 (2) 0/67 (0) 1/67 (2) 1/67 (2) 3/67 (4)

27/67 (40) 14/27 (58) 9/27 (38) 0/27 (0) 1/27 (4) 0/27 (0) 1/27 (4) 0/27 (0) 2/27 (8)

44/116 (38) 16/44 (36) 16/44(36) 7/44 (16) 2/44 (5) 1/44 (2) 0/44 (0) 1/44 (2) 1/44 (2)

30/44 (68) 8/30 (24) 9/33 13/30 (39) 7/30 (21) 2/30 (6) 1/30 (3) 1/30 (3) 0/30 (0) 1/33 1/30 (3)

EHEC: stx1

40/183 (22)

19/40 (48)

39/116 (34)

31/39 (79)

3/183 (2)

3/3 (100)

0/116 (0)

0/116 (0)

110/183 (60) 100/183 (54)

49/110 (45)

83/116 (72) 78/116 (67)

61/83 (73)

EPEC: eaeA a,b

Total positive Total resistants a

Strains positive for virulence factors and those resistant to an antibiotic among them. Comparison between percentage of positive strains for both virulence factors and antibiotic resistance from indoor and outdoor samples (X2 test, contingency table, po0.05). b

factors related to EAEC, mainly isolated from outdoors (Table 4). Among these E. coli strains, seven serogroups OR, O170, O6, O159, O86, O28 and O8 were found in

indoor dust; and, in similar numbers, the outdoor isolates had the serogroups O87, O4, O66, O9, O70, OR and O8.

ARTICLE IN PRESS 466

I. Rosas et al. / Int. J. Hyg. Environ.-Health 209 (2006) 461–470

1

2

3

4

5

6

7

8

9

10

11

pCVD432 632 pb aggC 538 pb

aaf/A 550 pb pet 548 pb

aggA 432 pb aggR 308 pb aap 232 pb astA 111 pb

Fig. 2. PCR amplification of the EAEC genes. Lane 1: 123 bp ladder marker; lane 2: positive control from EAEC strains 17-2 of the aaf/A gene; lane 3: EAEC strains O42 from pet gene; lanes 4 and 5: outdoor strains; lanes 6–9: indoor strains; lane 10: negative control; and lane 11: positive controls EAEC strains O42 and O17-2 of the aggC, aggA, aggR, aap, astA and pCVD432 plasmid from EAEC.

pb

861 — 615 — 492 —

1

2

3

4

5

6

7

8

ial 650

369 —

lt 450 eaeA 384

246 —

bfpA 324 stx2 255

123 —

st 190 stx1150

Fig. 3. Multiplex PCR primers to determine the presence of ial, lt, eaeA, bfpA, stx2, st and stx1 in Escherichia coli strains isolated from indoor and outdoor urban dust (Table 1). Lane 1: molecular weight marker (123 bp DNA); lanes 2–7: E. coli samples of positive strains; lane 8: positive controls.

Antibiotic resistance patterns and virulence factors Statistically significant differences were observed between the percentage of E. coli strains resistant to important clinical antibiotics, from indoor and outdoor

isolates. This was observed among all strains whether negative or positive for virulence factors tested (X2 ¼ 48.4; po0:05) (Table 4). This was similar for strains positive for virulence genes. Resistance, especially towards tetracycline, was more prevalent among the strains positive to any virulence factor (Table 5). The exception was resistance to ticarcillin-clavulanic acid, which was only detected among indoor strains, and ciprofloxacin resistance at a relatively high prevalence in outdoor samples.

Discussion Fecal pollution is one of the major factors associated with urbanization that contribute to the degrading quality of the environment. In this study, with samples collected in Mexico City in 2001, FC including E. coli were found in both indoor and outdoor environments. Curiously enough, FC were more abundant in indoor samples, particularly in those taken at houses having either carpets or pets. Furthermore, E. coli was found in all FC-positive indoor samples. There is evidence that the home environment could be an important source of infectious diseases (Krause et al., 2005; Kuhnert et al., 2000; Larson and Gomez-Duarte, 2001). Sixty-five percent (193/299) of randomly selected E. coli strains were positive for at least one of the virulence factors tested (Table 3), but there was no significant difference between the prevalence of virulence determinants among indoor and outdoor E. coli. However, in a low percentage of isolates, the virulence factor for EPEC (eae) was only observed indoors (Table 3): this virulence

ARTICLE IN PRESS I. Rosas et al. / Int. J. Hyg. Environ.-Health 209 (2006) 461–470

Table 4.

467

Serotypes, virulence factors and antibiotic resistance of strains with different adherence patterns

Source

Serotype

Pattern of adhesion

Virulence marker

Antibiotic resistance

OR:H30 O170:H O6:H10 O6:H10 O159:H21 O86:H34 O86:H10 O28:H O8:H51

DA DA DA DA DA DA DA CLA AAcs

astA astA aap aap aap stx1 stx1 aap aap

PIP, SXT, AM TE, CL, AM SXT AM, PIP, CTX TE, GM, AN SXT SXT TE TE

O87:H21 O4:H42 O66:H32 O9:H31 O70:H11 OR:H25 O4:H42 O66:H32 O70:H11 O8:H11 O66:H45 O4:H42

DA DA DA DA DA DA CD CLA AAcs AAcs AAcs AAcs

aap aap astA pCVD432/astA stx1 stx1 aap astA stx1 stx1 stx1 stx1

TE, CL TE, CL, CIP, TM TE, CL, AM, AZ, CF, CIP, GM, FT, TM, SXT, AN TE, CL, AM, CF, PIP, TM, SXT TE TE TE, AM, CTX, GM, TM, CL, SXT TE, CIP, GM, FT, CL, SXT, AN TE CF, AM CIP, SXT TM

Indoor

Outdoor

Adherence: diffuse adherence (DA), chain-like adherence (CLA), cell-detaching (CD), adherence to cover slip (AAcs).Antibiotics: ampicillin (AM); amikacin (AN); aztreonam (AZ); cefotaxime (CTX); cephalotin (CF); chloramphenicol (CL); ciprofloxacin (CIP); gentamicin (GM); nitrofurantoin (FT); piperacillin (PIP); tetracyclin (TE); tobramycin (TM); trimethoprim/sulfamethoxazol (SXT).

Table 5.

Antibiotic resistance in E. coli and virulence factors Indoor

Outdoor

Number of resistant strains Absence VF

Presence VF

Absence VF

Presence VF

Antibiotics:

51/73

49/110

17/33

61/83

Tetracyclin Piperacillin Ticarcillin/clavulanic acid Trimethoprim/sulfamethoxazol Ciprofloxacin Chloramphenicol Ampicillin Strains resistant and positive for virulence factors: in/outa

38% 16% 15% 12% 0% 30% 16%

60% 26% 22% 22% 0% 8% 21%

52% 21% 0% 39% 27% 33% 21%

90% 20% 0% 53% 36% 53% 25%

VF: virulence factors. a Comparison between resistant strains positive for virulence factors from indoors and outdoors (X2 test, contingency table, po0.05).

factor has been reported in E. coli isolated from pets (Krause et al., 2005; Nakazato et al., 2004). An entirely different picture was obtained when analyzing antibiotic resistance among the strains posi-

tive for virulence factors. In this respect, 73% of the outdoor isolates were resistant to at least one drug, while only 45% of the isolates from indoor samples were so (Table 3). Furthermore, multiresistance was also

ARTICLE IN PRESS 468

I. Rosas et al. / Int. J. Hyg. Environ.-Health 209 (2006) 461–470

more common among outdoor strains (Table 4). Resistance to tetracycline, chloramphenicol, ciprofloxacin and trimethoprim-sulfamethoxazole was much more common among outdoor isolates (Table 5). It is interesting to comment on some individual resistance traits: the prevalence of ampicillin resistance was similar among indoor and outdoor strains, but all outdoor strains were clavulanate-susceptible, while most indoor isolates that were resistant to ampicillin were also resistant to clavulanate (Table 5). As resistance to ampicillin and clavulanate is likely to be the result of a de-repressed chromosomal ampC gene, whereas ampicillin resistance but not clavulanate resistance is often mediated by a plasmid-borne beta-lactamase, it is possible that the outdoor environment is somehow selecting plasmid-carrying strains for this antibiotic resistance. Integrons, mobile genetic elements that allow the rearrangement of antibiotic resistance genes, which are often found in plasmids (Hall et al., 1996), were also more often found in outdoor isolates. In contrast, ciprofloxacin resistance, mainly determined by chromosomal mutations on target genes (Koutsolioutsou et al., 2001), was only found amongst outdoor strains. In any event, the high incidence of antibiotic resistance of E. coli strains isolated from outdoor samples suggests that they are more likely originating from humans (Webster et al., 2004; Guan et al., 2002; Kelsey et al., 2003; Parveen et al., 2001). None of the E. coli isolates harbored a complete array of the tested virulence factors. This could be related in part to the size of the sample. Also, it is important to consider that the isolated strains reported here were not tested for all known diarrheogenic virulence markers (Czeczulin et al., 1999). Interestingly, neither the genes associated to ETEC nor EIEC pathotypes were found. It has been shown in a report for Mexico City that 46% of childhood diarrhea was associated with ETEC, 8% with EIEC and 5% with EPEC. Nevertheless, neither EAEC nor EHEC probes were included in such a study (Secretarı´ a de Salud, 2001). Recently, the presence of some virulence genes for EAEC has been reported in a small group of Mexican children with diarrhea, as well as in healthy children (Cerna et al., 2003; EstradaGarcı´ a et al., 2005). In another study, among E. coli harboring virulence factors for EAEC, different adherence patterns have been observed: no typical aggregative adherence was found but rather variant aggregative adherence, such as DA (Piva et al., 2003), CLA (Gioppo et al., 2000), CD (Okeke et al., 2002) or AAcs (Kahali et al., 2004). Interestingly, in the study reported in this paper, also no typical aggregative adherence was found among the strains positive for an EAEC virulence factor. It is important to consider that the EAEC strains have only rarely been cultured from non-human sources: from infant feeding bottles in Brazil (Morais and Gomes,1997), from cattle (Sandhu et al., 1999) and

from typical food in Mexico (Adachi et al., 2002). Moreover, it is possible that EAEC strains could be milder pathogens, eliciting inflammation without diarrhea or else could be pathogenic to animals (Okeke et al., 2000). In this study, among the E. coli isolates, 22% of the indoor strains and 34% from outdoors were positive for stx1 (Table 3). This could be related to their natural hosts such as farm ruminants (Beutin et al., 1998); also, EHEC strains have been considered an important pathogen associated to diarrheal symptoms with the potential of being transmitted to humans and animals (Vernozy-Rozand et al., 2004). Some of the serogroups of the urban dust E. coli isolates described here have been associated to clinical cases in other studies: mainly O86 with DA and positive to either aap or stx1, virulence factors which have been reported among the EPEC, EAEC and EHEC groups (Zhang et al., 2000; Gonza´lez et al., 1997). Among the outdoor soil strains, we found that the O9 serogroup was associated with DA, with multiresistance to antibiotics and with EAEC virulence factors (Table 4). Interestingly, the O9 serogroup has also been found associated to EAEC virulence factors (Rodrı´ guez, 2002). The O4 serotype, presenting CD with multiple resistance (Table 4), has also been found among E. coli isolated from Nigerian children (Okeke et al., 2002). The association between virulence factors and antibiotic resistance observed in this study could reflect the importance of protecting the virulence gene reservoir from potential vertical and horizontal transfer. Finally, the high prevalence of virulence factors associated to either EAEC or EHEC in E. coli strains indicates that these pathotypes need to be included in the national surveillance of infectious diseases.

Acknowledgments We thank to Dr. Iruka Okeke for the review of the manuscript Ariadnna Cruz for the technical assistance in the PCR assays, and Armando Navarro for the serotyping analysis.

References Adachi, J.A., Ericsson, C.D., Jiang, Z.-D., DuPont, M.W., Pallegar, S.R., DuPont, H.L., 2002. Natural history of enteroaggregative and enterotoxigenic Escherichia coli infection among US travelers to Guadalajara. Me´x. J. Infect. Dis. 185, 1681–1683. American Public Health Association, 1985. Standard Methods for the Examination of Water and Wastewater, 16a ed. American Public Health Association, Washington, DC. Beutin, L., Zimmermann, S., Gleier, K., 1998. Human infections with Shiga toxin-producing Escherichia coli other

ARTICLE IN PRESS I. Rosas et al. / Int. J. Hyg. Environ.-Health 209 (2006) 461–470

than serogroup O157 in Germany. Emerg. Infect. Dis. 4, 635–639. Brookes, J.D., Antenucci, J., Hipsey, M., Bruch, M.D., Ashbolt, N.J., Ferguson, C., 2004. Fate and transport of pathogens in lakes and reservoirs. Environ. Int. 30, 741–759. Cerna, J.F., Nataro, J.P., Estrada-Garcı´ a, T., 2003. Multiplex PCR for detection of three plasmid-borne genes of enteroaggregative Escherichia coli strains. J. Clin. Microbiol. 41, 2138–2140. Cravioto, A., Tello, A., Navarro, A., Ruı´ z, J., Villafa´n, H., Uribe, F., Eslava, C., 1991. Association of Escherichia coli HEp-2 adherence patterns with type and duration of diarrhoea. Lancet 337, 262–264. Czeczulin, J.R., Whittam, T.S., Henderson, I.R., NavarroGarcia, F., Nataro, J.P., 1999. Phylogenetic analysis of enteroaggregative and diffusely adherent Escherichia coli. Infect. Immun. 67, 2692–2699. Escobar-Pa´ramo, P., Clermont, O., Blanc-Potard, A., Biu, H., Le Bouguenec, C., Denamur, E., 2004. A specific background is required for acquisition and expression of virulence factors in Escherichia coli. Mol. Biol. Evol. 21, 1085–1094. Eslava, C.A., Navarro, A., Henderson, I.R., Cravioto, A., Nataro, J.P., 1998. Pet autotransporter enterotoxin from enteroaggregative Escherichia coli. Infect. Immun. 66, 3155–3163. Estrada-Garcı´ a, T., Cerna, J.F., Pacheco-Gil, L., Vela´zquez, R.F., Ochoa, T.J., Torres, J., DuPont, H.L., 2005. Drugresistant diarrheogenic Escherichia coli, Mexico. Emerg. Infect. Dis. 11, 1306–1308. Gessel, P.D., Hansen, N.C., Goyal, S.M., Johnston, L.J., Webb, J., 2004. Persistence of zoonotic pathogens in surface soil treated with different rates of liquid pig manure. Appl. Soil. Ecol. 25, 237–243. Gioppo, N.M.R., Elias Jr., W.P., Vidotto, M.C., Linhares, R.E., Saridakis, H.O., Gomes, T.A.T., Trabulsi, L.R., Pelayo, J.C., 2000. Prevalence of HEp-2 cell-adherent Escherichia coli and characterization of enteroaggregative E. coli and chain-like adherent E. coli isolated from children with and without diarrhoea, in Londrina, Brazil. FEMS Microbiol. Lett. 190, 293–298. Gonza´lez, R., Diaz, C., Marin˜o, M., Cloralt, R., Pequeneze, M., Perez-Schael, I., 1997. Age-specific prevalence of Escherichia coli with localized and aggregative adherence in Venezuelan infants with acute diarrhea. J. Clin. Microbiol. 35, 1103–1107. Guan, S., Xu, R., Chen, S., Odumeru, J., Gyles, C., 2002. Development of a procedure for discriminating among Escherichia coli isolates from animal and human sources. Appl. Environ. Microbiol. 68, 2690–2698. Hall, R.M., Recchia, G.D., Collis, C.M., Brown, H.J., Stokes, H.W., 1996. Gene cassettes and integrons: moving antibiotic resistance genes in gram-negative bacteria. In: Ama´bile-Cuevas, C.F. (Ed.), Antibiotic Resistance: From Molecular Basics to Therapeutic Options. Chapman & Hall, New York, pp. 19–34. Jallat, C., Darfeuille-Michaud, A., Rich, C., Joly, B., 1994. Survey of clinical isolates of diarrhoeogenic Escherichia

469

coli: diffusely adhering E. coli strains with multiple adhesive factor. Res. Microbiol. 145, 621–632. Kahali, S., Sarkar, B., Rajendran, K., Khanam, J., Yamasaki, S., Nandy, R.K., Bhattacharya, S.K., Ramamurthy, T., 2004. Virulence characteristics and molecular epidemiology of enteroaggregative Escherichia coli isolates from hospitalized diarrheal patients in Kolkata, India. J. Clin. Microbiol. 42, 4111–4120. Kelsey, R.H., Scott, G.I., Porter, D.E., Thompson, B., Webster, L., 2003. Using multiple antibiotic resistance and land use characteristics to determine sources of fecal coliform bacterial pollution. Environ. Monit. Assess. 81, 337–348. Kjaer, P.J., Ensink, J.H., Jayasinghe, G., Van der Hoek, W., Cairncross, S., Dalsgaard, A., 2002. Domestic transmission routes of pathogens: the problem of in-house contamination of drinking water during storage in developing countries. Trop. Med. Int. Health 7, 604–609. Koutsolioutsou, A., Martins, E.A., White, D.G., Levy, S.B., Demple, B., 2001. A soxRS-constitutive mutation contributing to antibiotic resistance in a clinical isolate of Salmonella enterica (Serovar Typhimurium). Antimicrob. Agents Chemother. 45, 38–43. Krause, G., Zimmermann, S., Beutin, L., 2005. Investigation of domestic animals and pets as a reservoir for intimin(eae) gene positive Escherichia coli types. Vet. Microbiol. 106, 87–95. Kuhnert, P., Boerlin, P., Joachim, F., 2000. Target genes for virulence assessment of Escherichia coli isolates from water, food and environment. FEMS Microbiol. Rev. 24, 107117. Larson, E., Gomez-Duarte, C., 2001. Home hygiene practices and infectious disease symptoms among household members. Public Health Nurs. 18, 116–127. Le´vesque, C., Roy, P.H., 1993. PCR analysis of integrons. In: Persing, D.H., Smith, T.F., Tenover, F.C., White, T.J. (Eds.), Diagnostic Molecular Microbiology, Principles and Applications. American Society for Microbiology, Washington, DC, pp. 590–602. Lo´pez-Saucedo, C., Cerna, J.F., Villegas-Sepulveda, N., Thompson, R., Vela´zquez, F.R., Torres, J., Tarr, P.I., Estrada-Garcı´ a, T., 2003. Single multiplex polymerase chain reaction to detect diverse loci associated with diarrheagenic Escherichia coli. Emerg. Infect. Dis. 9, 127–131. Morais, T.B., Gomes, T.A.T., 1997. Enteroaggregative Escherichia coli in infant feeding bottles. Lancet 349, 1448–1449. Nakazato, G., Glyles, C., Ziebell, K., Keller, R., Trabulsi, L.R., Gomes, T.A.T., Irino, K., Dias Da Silveira, W., Pestana De Castro, A.F., 2004. Attaching and effacing Escherichia coli isolated from dogs in Brazil: characteristics and serotypic relationship to human enteropathogenic E. coli (EPEC). Vet. Microbiol. 101, 269–277. Nataro, J.P., Kaper, J.B., Robins-Browne, R., Prado, V., Vial, P., Levine, M.M., 1987. Patterns of adherence of diarrheagenic Escherichia coli to HEp-2 cells. Pediatr. Infect. Dis. J. 6, 829–831. Nataro, J.P., Steiner, T., Guerrant, R.L., 1998. Enteroaggregative Escherichia coli. Emerg. Infect. Dis. 4, 251–261.

ARTICLE IN PRESS 470

I. Rosas et al. / Int. J. Hyg. Environ.-Health 209 (2006) 461–470

Okeke, I.N., Lamikanra, A., Czeczulin, J., Dubovsky, F., Kaper, J.B., Nataro, J.P., 2000. Heterogeneous virulence of enteroaggregative Escherichia coli strains isolated from children in southwest Nigeria. J. Infect. Dis. 181, 252–260. Okeke, I.N., Steinru¨ck, H., Kanack, K.J., Elliot, S.J., Sundstro¨m, L., Kaper, J.B., Lamikanra, A., 2002. Antibiotic-Resistant cell-detaching Escherichia coli strains from Nigerian Children. J. Clin. Microbiol. 40, 301–305. Orskov, F., Orskov, Y., 1984. Serotyping of Escherichia coli. In: Bergan, T. (Ed.), Methods in Microbiology, vol. 14. Academic Press Ltd., London, pp. 43–112. Parveen, S., Hodge, N.C., Stall, R.E., Farrah, S.R., Tamplin, M.L., 2001. Phenotypic and genotypic characterization of human and nonhuman Escherichia coli. Water Res. 35, 379–386. Piva, I.C., Pereira, A.L., Ferraz, L.R., Silva, R.S.N., Vieira, A.C., Blanco, J.E., Blanco, M., Blanco, J., Giugliano, L.G., 2003. Virulence markers of enteroaggregative Escherichia coli isolated from children and adults with diarrhea in Brasilia, Brazil. J. Clin. Microbiol. 41, 1827–1832. Puente, J.L., Finlay, B.B., 2001. Pathogenic Escherichia coli. In: Groisman, E.A. (Ed.), Principles of Bacterial Pathogenesis. Academic Press, New York, pp. 387–456. Rich, C., Favre-Bonte, S., Sapena, F., Joly, B., Forestier, C., 1999. Characterization of enteroaggregative Escherichia coli isolates. FEMS Microbiol. Lett. 173, 5561. Rodrı´ guez, G., 2002. Principales caracterı´ sticas y diagno´stico de los grupos pato´genos de Escherichia coli. Salud Pu´blica Me´x. 44, 464–475. Rosas, I., Salinas, E., Yela, A., Calva, E., Eslava, C., Cravioto, A., 1997. Escherichia coli in settled-dust and air samples collected in residential environments in Mexico City. Appl. Environ. Microbiol. 63, 4093–4095. Sandhu, K.S., Clarke, R.C., Gyles, C.L., 1999. Virulence markers in Shiga toxin-producing Escherichia coli isolated from cattle. Can. J. Vet. Res. 63, 177–184.

Scaletsky, I.C., Silva, M.L., Trabulsi, L.R., 1984. Distinctive pattern of adherence of enteropathogenic Escherichia coli to HeLa cells. Infect. Immun. 45, 534–536. Schwartz, T., Kohnen, W., Jansen, B., Obst, U., 2003. Detection of antibiotic-resistant bacteria and their resistance genes in wastewater, surface water, and drinking water biofilm. FEMS Microbiol. Ecol. 43, 325–335. Secretarı´ a de Salud, 2001. Informe anual de los paı´ ses participantes en la red de monitoreo/vigilancia de la resistencia a los antibio´ticos. Me´xico, pp. 62–67. Vernozy-Rozand, C., Montet, M.P., Bertin, Y., Trably, F., Girardeau, J.P., Martin, C., Livrelli, V., Beutin, L., 2004. Serotyping, stx2 subtyping, and characterization of the locus of enterocyte effacement island of shiga toxinproducing Escherichia coli and E. coli 0157:H7 strains isolated from the environment in France. Appl. Environ. Microbiol. 70, 2556–2559. von Lindern, I.H., Spalinger, S.M., Bero, B.N., Petrosyan, V., von Braun, M.C., 2002. The influence of soil remediation on lead in house dust. Sci. Total Environ. 303, 59–78. Webster, L.F., Thompson, B.C., Fulton, M.H., Chestnut, D.E., Van Dolah, R.F., Leight, A.K., Scott, G.I., 2004. Identification of sources of Escherichia coli in South Carolina estuaries using antibiotic resistance analysis. J. Exp. Mar. Biol. Ecol. 298, 179–195. Wouters, I.M., Douwes, J., Doekes, G., Thorne, P.S., Brunekreef, B., Heederick, D.J., 2000. Increased levels of markers of microbial exposure in homes with indoor storage of organic household waste. Appl. Environ. Microbiol. 66, 627–631. Zhang, W.-L., Bielaszewska, M., Liesegang, A., Tscha¨pe, H., Schmidt, H., Bitzan, M., Karch, H., 2000. Molecular characteristics and epidemiological significance of Shiga toxin-producing Escherichia coli O26 strains J. Clin. Microbiol. 38, 2134–2140.