|Home | About | Journals | Submit | Contact Us | Français|
The presence of antibodies to the intimin β-binding region (Int280-β) of attaching and effacing Escherichia coli (AEEC) in serum from 20 goat kids from 2 herds, as well as in goat colostrum, was investigated by enzyme-linked immunosorbent assay. In addition, the onset and subsequent pattern of shedding of AEEC from the same goat kids over a 6-mo period was investigated. All the colostrum and serum samples tested contained antibodies against Int280-β. The association between the antibody titer and the isolation of AEEC suggests that antibodies to intimin β do not prevent colonization of the intestine by AEEC in goat kids. The AEEC were generally shed only transiently. Most AEEC isolated from the kids belonged to serogroup O26. Three isolates belonged to serogroup O157. These data show that goat kids may be a reservoir of AEEC that are potentially pathogenic for humans.
La présence d’anticorps dirigés contre la région d’adhésion β de l’intimine (Int280-β) des Escherichia coli attachant et effaçant (AEEC) a été étudiée à partir d’échantillons de sérum de 20 chevreaux provenant de 2 troupeaux, ainsi que dans le colostrum par épreuve immuno-enzymatique. Également, on étudia le début ainsi que le patron d’excrétion des AEEC par les mêmes chevreaux durant une période de 6 mois. Des anticorps dirigés contre Int280-β étaient présents dans tous les échantillons de colostrum et de sérum éprouvés. L’association entre le titre d’anticorps et l’isolement d’AEEC suggère que les anticorps contre l’intimine β n’empêchent pas la colonisation de l’intestin par les AEEC chez les chevreaux. Les AEEC n’étaient généralement excrétés que de manière transitoire. La plupart des AEEC isolées des chevreaux appartenaient au sérogroupe O26. Trois isolats appartenaient au sérogroupe O157. Les résultats démontrent que les chevreaux pourraient être un réservoir des AEEC qui sont potentiellement pathogènes pour les humains.
(Traduit par Docteur Serge Messier)
Attaching and effacing Escherichia coli (AEEC) have been associated with diarrhea in humans and animals. These bacteria cause attaching and effacing (AE) lesions in the gut mucosa that are characterized by intimate bacterial adhesion to the enterocyte and effacement of brush-border microvilli (1). Enteropathogenic E. coli (EPEC) and enterohemorrhagic E. coli (EHEC) cause AE lesions in the human intestinal mucosa (1), and ruminants can be reservoirs of these strains (2). In contrast to EPEC, EHEC strains produce verotoxins (VTs) (1). The EPEC strains have been classified as typical (possessing the bfpA gene) or atypical (lacking the bfpA gene). In contrast to typical EPEC, atypical EPEC strains frequently possess several virulence genes, such as astA, ehxA, and cnf3 (3,4).
The genes necessary for AE lesion formation are encoded in a pathogenic island called the locus of enterocyte effacement (LEE). The LEE encodes intimin, which mediates intimate bacterial adhesion to the enterocytes and is encoded by the eae gene (1). The overall pattern of the sequences of eae genes from different AEEC strains shows high conservation in the N-terminal region and variability in the first 280 amino acids starting from the C terminus of the intimin protein, which are involved in binding to enterocytes (1). On the basis of antigenic variation, polymerase chain reaction (PCR) analysis, and sequencing, several types of intimin have been identified (2).
Human colostrum and milk collected from women living in areas in which AEEC infection is endemic contain antibodies that react with surface antigens, especially intimin, of AEEC strains (5,6). The data suggest that the development of specific immunity to intimin may play a role in protecting against AEEC infection. In goats, in contrast to humans, there is a lack of information regarding the antibody response to intimin.
Most of the caprine AEEC isolates previously reported have been EPEC (7,8). Although intimin β has been found to be prevalent among EPEC isolated from healthy and diarrheic goats, EPEC isolates with this intimin type have also been associated with neonatal diarrhea in goat kids (2). Most AEEC isolates from healthy goats have been found in goat kids (7,8), but, to our knowledge, no longitudinal studies in goat kids colonized with this type of E. coli have been reported.
This study was designed to determine the presence of antibodies to the intimin β-binding region (Int280-β) of AEEC in serum from goat kids collected during their first 4 wk of life, and in goat colostrum, and to assess the role of such antibodies in preventing intestinal colonization by AEEC in the kids. Other objectives of the study were to investigate the onset and subsequent pattern of shedding of AEEC from goat kids over a 6-mo period and to characterize the AEEC colonies isolated.
The study was conducted on 2 Murciano–Granadina goat herds in the Murcia region of southeastern Spain. The practice of artificial rearing, in which goat kids are withdrawn after parturition and fed pasteurized colostrum and milk replacer, was implemented in 1 of the herds (herd A). In contrast, in the other herd studied (herd B), the goat kids were fed colostrum and milk from their dams. Herd A was studied from June to November 2004, and herd B was studied from November 2004 to April 2005. Serum samples were collected from a total of 20 goat kids (9 from herd A and 11 from herd B) once weekly during the kids’ first 4 wk of life. Serum samples were also collected from 2 kids per herd just after birth, before they were fed colostrum. One sample of pasteurized colostrum was collected from herd A. In herd B, serum was collected from female goats approximately 1 wk before parturition, and colostrum was collected at the time of parturition. Additionally, fecal samples were taken from the 20 goat kids once weekly during the first 4 wk of life and then once every month for the next 5 mo of life.
Int280-β, expressed as a His-tagged polypeptide, was prepared as previously described (9), with minor modifications. Briefly, a DNA fragment encoding Int280-β was amplified by PCR from the RDEC-1 strain (a rabbit EPEC strain that possesses intimin β) with use of the following primers and standard PCR amplification conditions: Int280βF (5′-TCGCGGATCCATTACTGAGATTAAGGCTGA-3′) and Int280βR (5′-GCGAAGCTTGTTTTACACAAAACAGAAAAAG-3′). The PCR product was cloned into BamHI/HindIII-digested pET-28a (Novagen; Merck Biosciences, Darmstadt, Germany). The recombinant plasmid was transformed into E. coli BL21 cells and His-Int280 expressed under isopropyl-β-D-thiogalactopyranoside induction and affinity-purified with metal resin (Clontech; Clontech-Takara Bio Europe, Saint-Germain-en-Laye, France) with the use of standard protocols.
The enzyme-linked immunosorbent assay (ELISA), used to determine the presence and amount of antibodies to Int280-β in colostrum and serum, was performed as previously described (10), with minor modifications. Briefly, ELISA plates were coated overnight at 4°C with 20 μg/mL of Int280-β, 50 μL/well. The plates were incubated with 2-fold serial dilutions of the colostrum and serum samples and the reactions revealed with peroxidase-conjugated rabbit antibodies against goat antigen (Dako Diagnósticos, Barcelona, Spain) and o-phenylenediamine as substrate. The serum samples taken from 2 goat kids per herd just after birth, before they were fed colostrum, were used as negative controls.
Immunoblot analysis was used to determine the optical density (OD) cutoff associated with a positive ELISA result. Dilutions of 20 serum samples that had OD values in the ELISA between 0.15 and 0.5, as well as the 4 negative-control samples, were tested with this technique. After Tris–Tricine–polyacrylamide gel electrophoresis, Int280-β (100 μg/mL) was transferred electrophoretically onto polyvinylidene difluoride membranes. The membranes were incubated with the serum samples and the reactions revealed with peroxidase-conjugated rabbit antibodies against goat antigen (Dako) and enhanced chemiluminescence.
For isolation of AEEC, the fecal samples were plated onto MacConkey agar. After overnight incubation, 4 colonies with the typical appearance of E. coli were randomly chosen from each sample and identified as E. coli by biochemical tests. All E. coli isolates were analyzed by PCR for possession of the eae gene as described previously (7). All AEEC colonies were tested for the bfpA, astA, ehxA, and cnf3 genes and the types of eae and vt genes by PCR as described previously (2,3,4,8). The determination of O antigens for serogrouping was carried out in the Laboratorio de Referencia de E. coli (Lugo, Spain).
From the immunoblot studies, the cutoff for a positive ELISA result was considered to be an OD of 0.2 or more above that for the negative controls. All colostrum and serum samples from the adult goats contained antibodies against Int280-β. The titer of these antibodies in the pasteurized colostrum obtained from herd A was 1:2048; the titer in colostrum and serum from adult goats in herd B ranged from 1:128 to 1:2048. Moreover, antibodies to Int280-β were detected in all the serum samples obtained from goat kids in both herds (Tables I and andII).II). A relationship was observed in herd B between the titer of antibodies in the colostrum from adult goats and the titer in the serum from their respective kids.
Shedding of AEEC by goat kids began in the 2nd week of life in herd B but in the 3rd month in herd A. In all but 1 of the animals studied, AEEC colonies were detected in at least 1 sample, but in most kids in both herds the shedding was transient. When the AEEC colonies from a fecal sample showed the same characteristics (intimin type, possession of the vt, bfpA, astA, ehxA, and cnf3 genes, and serogroup) it was assumed that they were the same isolate. In total, 29 AEEC isolates were identified. The characteristics of the AEEC colonies isolated from the goat kids are shown in Tables I to III. In all but 1 of the kids colonized with AEEC, only 1 isolate per animal was identified in each sample. However, different AEEC colonies were isolated from 4 kids at different times.
In herd B, AEEC colonies were isolated from goat kids during their first 4 wk of life (Table II). In the 7 animals in this herd that shed AEEC possessing intimin β during their first 4 wk of life, the titers of antibodies to Int280-β measured when the AEEC isolates with this intimin type were found ranged from 1:32 to 1:512. In the 3 kids of herd B that were colonized with AEEC of other intimin types during the same period the titers of antibodies to Int280-β measured when these AEEC colonies were isolated ranged from 1:32 to 1:256.
To our knowledge, this is the first work in which the presence of antibodies to intimin β in colostrum and serum from adult and juvenile goats has been studied. We measured titers of antibodies only to the extracellular C-terminal portion of intimin β that is 280 amino acids long; this portion is involved with binding to enterocytes (1). Our observations are therefore more pertinent to intimin β-positive AEEC than to AEEC that express other intimin variants, and only AEEC that possess intimin β have been associated with diarrhea in goat kids (2).
In both herds, all colostrum and serum samples from adult females contained antibodies against Int280-β. This finding suggests that the dams had been exposed to AEEC that possessed intimin β and may reflect the prevalence of this type of E. coli in the environment, the high immunogenicity of intimin, or both (10). Antibodies against Int280-β were also detected in all serum samples from the goat kids. In 1 of the herds studied (herd B) the AEEC colonies were isolated during the kids’ first 4 wk of life. Although our data are insufficient for statistical analysis, the association between the titers of antibodies to Int280-β and the isolation of AEEC from the animals in herd B suggests that antibodies to intimin β do not prevent colonization of the intestine of goat kids by AEEC. In agreement with the results of this study, van Diemen et al (9) reported that immunization of calves with the cell-binding domain of intimin type β or γ induced humoral immunity but did not protect against intestinal colonization of EHEC O26:H– (Int280-β) or EHEC O157:H7 (Int280-γ).
The onset of shedding of AEEC was different in the 2 herds that we studied. It is possible that this difference was due to different feeding and management practices. Changes in diet and animal management practices have been shown to affect the shedding of verotoxin-producing E. coli (11,12). Although AEEC colonies were detected in at least 1 sample from all the animals studied except 1, AEEC were generally shed transiently by the colonized goat kids.
The present results are consistent with earlier findings that most AEEC isolated from healthy goats possess intimin β and the ehxA gene and are negative for the vt and astA genes and that goats are not a significant reservoir of typical EPEC (none of the EPEC strains found in this study had the bfpA gene) (2,3,7,8). In this study, the AEEC colonies isolated belonged to 7 serogroups, but almost 60% belonged to serogroup O26. Seven of the 17 O26 colonies possessed the cnf3 gene, which encodes a new type of cytotoxic necrotizing factor recently described by our group in atypical EPEC from sheep and goats (4). In addition, 3 isolates belonged to serogroup O157 and carried the vt2 gene. Since EPEC O26 and EHEC O157 have been implicated as human pathogens (1,13), the results of this study show that goat kids may be a potential source of infection for humans. However, in contrast with the data obtained here, previous studies (2,7,8) have been unable to detect serogroups O26 and O157 among AEEC isolated from healthy goats. The differences between studies in the detection of AEEC serogroups may be at least partially due to the transient nature of the shedding of AEEC by colonized goat kids.
In conclusion, this article describes for the first time the presence of antibodies against intimin β in colostrum and serum obtained from goat kids. Moreover, the study results suggest that these antibodies do not prevent colonization of the intestine by AEEC in goat kids. The data obtained also show that colonization with AEEC in goat kids is transient and suggest that these animals may be a reservoir of AEEC potentially pathogenic for humans.
This study was supported by grants from the Dirección General de Investigación (grant AGL2001-1476), the Universidad Complutense-Comunidad de Madrid (grant CCG07-UCM/AGR-2568), and the Instituto de Salud Carlos III (grants FIS G03-025-COLIRED-O157 and PI051428). We thank Elena Neves and Rocío Sánchez for technical assistance, and Dr. Jorge Blanco (Laboratorio de Referencia de E. coli, Facultade de Veterinaria, Universidade de Santiago de Compostela, Lugo, Spain) for serogrouping the attaching and effacing Escherichia coli strains.