Reducing colonization of Salmonella Enteritidis in chicken by targeting outer membrane proteins


M.I. Khan, Department of Pathobiology, Box U-89, University of Connecticut, Storrs, Connecticut 06269, USA (e-mail:


Aims: To evaluate the ability of Salmonella enterica ser. Enteritidis outer membrane proteins (OMPs) of 75·6 and 82·3 kDa to inhibit or reduce in vivo colonization of S. Enteritidis on intestinal mucosa in chickens.

Methods and Results: Nine-week-old specific-pathogen-free chickens were subcutaneously immunized with 75·6 or 82·3 kDa protein, and challenged with a virulent strain of S. Enteritidis. Chickens were killed, and portions of small intestine and caecum were removed at necropsy. The population of S. Enteritidis attached to chicken intestinal mucosa was determined. The population of S. Enteritidis recovered from the small intestine and caecum of chickens immunized with 75·6 or 82·3 kDa protein was significantly (P < 0·05) lower than that recovered from the control birds.

Conclusions:Salmonella Enteritidis OMPs 75·6 kDa and 82·3 kDa were effective in reducing colonization of S. Enteritidis on intestinal mucosa in chickens.

Significance and Impact of the Study:Salmonella Enteritidis OMPs 75·6 or 82·3 kDa could be used as potential vaccines to reduce S. Enteritidis colonization in chickens.


Salmonella enterica ser. Enteritidis has emerged as a significant food-borne pathogen in the United States. Poultry is an important reservoir of S. Enteritidis infection in humans (Allen-Vercoe et al. 1997). Epidemiological studies have revealed that the major source of S. Enteritidis outbreaks in humans include contaminated poultry meat and eggs. Contamination of egg contents may occur by direct transmission from infected ovaries and/or oviducts or by contamination of eggshell with faeces from a laying bird excreting Salmonella (Borland 1975; Timoney et al. 1989). In addition to the public health impact, S. Enteritidis causes tremendous economic losses to the poultry industry, amounting to ca. $64–$114 million annually (Bryan and Doyle 1995).

The ability of S. Enteritidis to establish infection depends on their ability to attach, colonize and invade intestinal epithelial cells (D'Aoust et al. 1991). Attachment is the first step in the process of pathogenesis bringing the bacteria in close contact with the host epithelial cells, initiating the further processes of pathogenesis such as colonization and invasion (Baloda et al. 1988; Finlay and Falkow 1989). Attachment of S. Enteritidis is mediated by proteins collectively known as adhesins (Beachy 1981). Blocking the primary stage of infection, namely bacterial attachment to host receptors may be the most effective strategy to prevent bacterial infection (Wizemann et al. 1999). Previously we identified two outer membrane proteins (OMPs), which are involved in the attachment of S. Enteritidis to intestinal epithelial cell line, Int-407 (Fadl et al. 2002). These proteins include 75·6 and 82·3 kDa OMPs, which were expressed when S. Enteritidis was incubated with Int-407. It was also found that antibodies against the 75·6 and 82·3 kDa proteins significantly reduced S. Enteritidis attachment to Int-407 (Fadl et al. 2002). Further, it was observed that immunization of chickens with the above proteins induced a strong antibody response (Fadl et al. 2002). The objective of this study was to evaluate the ability of the 75·6 and 82·3 kDa proteins to inhibit or reduce in vivo colonization of S. Enteritidis on chicken intestinal mucosa.

Materials and methods

Bacterial culture

Salmonella Enteritidis strain (SE 28) obtained from Dr Mike Opitz, University of Maine (Orono, ME, USA) was used for the study. Salmonella Enteritidis was cultured in 100 ml of sterile Luria Bertani broth (Sambrook et al. 1989) in 250-ml Erlenmeyer flasks at 37°C for 24 h with agitation (150 rev min−1). Following incubation, S. Enteritidis cells were harvested by centrifugation at 7000 × g for 10 min. Bacterial cells were washed three times with sterile phosphate-bufferered saline (PBS, pH 7·00), and subsequently resuspended in PBS.

Chickens and housing. Twelve 8-week-old specific-pathogen-free (SPF), white leghorn chickens were obtained from SPAFAS Inc. (Norwich, CT, USA). All chickens were tagged and assigned randomly to one of the three groups of four birds each. Each group was housed separately in a wire-bottomed battery cage. Birds were acclimated 1 week before inoculation. Feed and water were provided ad libitum.

Inoculation. The OMPs, 75·6 kDa and 82·3 kDa of S. Enteritidis incubated with intestinal epithelial cell line Int-407, were separated on a 4–20% gradient polyacrylamide gel (Fadl et al. 2002). The 82·3 and 75·6 kDa proteins were individually cut from sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) gels, and harvested by electroelution (Avakian and Kleven 1990). The concentration of the purified proteins was determined by Bradford assay (Bio-Rad Laboratories, Richmond, CA, USA). The two proteins were individually harvested and pooled from a total of 10 gels. Before pooling, each electroelution batch was checked for homogeneity in Coomassie blue-stained SDS-PAGE gels. The proteins were dissolved in SDS-PAGE sample buffer and stored at −80°C until used.

Two groups of chickens each containing four birds were inoculated subcutaneously in the thigh with 10 μg of each of the proteins. Group I was inoculated with the 82·3 kDa protein, whereas group II was inoculated with the 75·6-kDa protein. Group III, which was not inoculated with any proteins served as the control. The inoculated chickens were boosted twice with the same amount of proteins with a time interval of 2 weeks.

Bird challenge and sample collection. One week after the last boost, chickens were challenged with S. Enteritidis, by oral inoculation of 1 ml of S. Enteritidis culture containing ca. 8 × 108 CFU, using a syringe and a 10 cm rubber catheter. Following the challenge, chickens were observed for typical symptoms of Salmonellosis.

Duplicate faecal swabs were collected from all the birds 24 h postchallenge. The swabs were washed in 10 ml tryptic soya broth (TSB, Difco, Detroit, MI, USA), and incubated at 37°C for 24 h. After 48 h of challenge, all the chickens were killed. Triplicate portions of caecum and ileum (10 g each) were aseptically collected, washed gently in sterile PBS, and transferred into sterile TSB and PBS.

Bacteriological examination. All samples transferred into TSB were incubated at 37°C for 24 h. Following incubation, the culture was streaked on xylose lylose desoxycholate agar plates (XLD; Genetrack, MA, USA), and incubated at 37°C for 24 h.

Tissue samples collected in PBS were homogenized using a tissue tearor (Biospec products, Bartlesville, OK, USA). The tissue homogenates were serially diluted (1 : 10) in sterile PBS, and 0·1-ml aliquots from all the dilutions were plated on XLD agar. The plates were incubated at 37°C for 24 h. Salmonella colonies were identified by morphology, and representative colonies were confirmed with a Salmonella-specific latex agglutination kit (Oxoid, Ogdenberg, NY, USA).

Western blot.Salmonella Enteritidis OMPs separated on a 4–20% gradient SDS-PAGE were electrophoretically transferred to a nitrocellulose membrane (Sigma, St Louis, MO, USA) by the method of Towbin et al. (1979). The separated proteins were tested with preimmunization serum and by serum collected from the inoculated birds 1 week after the last boost.

Statistical analysis. The data obtained from each bird from the three groups were pooled and analysed for statistical significance by t-test (Steele and Torrie, 1980).


Western blots for Salmonella Enteritidis OMPs tested by preimmunization serum and the serum from inoculated birds showed that preimmunization serum did not recognize the S. Enteritidis proteins (Fadl et al. 2002). On the other hand, serum from inoculated birds reacted with the 75·6 and 82·3 kDa proteins.

Twenty-four hours after challenge, the inoculated chickens showed mild to moderate diarrhoea, and S. Enteritidis was recovered from the faecal swabs of all the birds. Salmonella Enteritidis was also recovered from the caecal contents 48 h postchallenge, as well as from the homogenized caecal and small intestine samples from all the birds. Enumeration of S. Enteritidis from the homogenized tissue samples of caecum revealed a significant decrease (P < 0·05) in the pathogen population in the samples from the chickens inoculated with 75·6 and 82·3 kDa proteins, compared with the S. Enteritidis population in the non-inoculated control group (Fig. 1). The mean log S. Enteritidis population recovered from the caecal samples of birds inoculated with 82·3 and 75·6 kDa were 3·7 ± 0·50 and 5·0 ± 0·50 g−1, respectively, compared with the control birds (7·1 ± 0·70 g). Similar to the results obtained with caecum, S. Enteritidis counts recovered from the small intestine of birds inoculated with the 82·3 and 75·6 kDa OMPs were significantly (P < 0·05) lower than those recovered from the control birds (P < 0·05). The mean log S. Enteritidis population recovered from the small intestine of birds inoculated with 82·3 and 75·6 kDa were 4·2 ± 0·78 g−1, 3·8 ± 0·80 g−1, and 5·8 ± 1·1 g−1, respectively (Fig. 2). Further, among the birds inoculated with the 75·6-kDa OMP, a significantly (P < 0·05) lower population of S. Enteritidis was recovered from the small intestine than from the caecum. However, there was no significant difference (P > 0·05) between the S. Enteritidis counts recovered from of the caecum and small intestine of chickens inoculated with the 82·3 kDa OMP.

Figure 1.

Effect of 75·6 and 82·3 kDa outer membrane proteins on the population of Salmonella Enteritidis in the caecum of chicken

Figure 2.

Effect of 75·6 and 82·3 kDa outer membrane proteins on the population of Salmonella Enteritidis in the small intestine of chicken


Reduction of carriage of S. Enteritidis by chicken would lead to reduced contamination of poultry carcasses and eggs by the pathogen, thereby potentially reducing food-borne outbreaks of S. Enteritidis. We previously observed that antibodies against S. enteritidis OMPs 75·6 and 82·3 kDa significantly reduced S. Enteritidis attachment to intestinal epithelial cell line Int-407 (Fadl et al. 2002). In order to evaluate the in vivo effect of these proteins to reduce colonization of S. Enteritidis on chicken intestinal mucosa, we immunized SPF chickens with 75·6 and 82·3 kDa proteins.

Vaccination of chickens using inactivated and live attenuated vaccines has been used to protect them from S. Enteritidis infection, and to reduce faecal shedding of the pathogen (Gast et al. 1992; Gast et al. 1993). These vaccines were able to provide protection against virulent strains of S. Enteritidis. However, they failed to stop shedding of the organism in faeces (Gast et al. 1993). Further, inactivated vaccines, despite eliciting acceptable levels of humoral immunity, were found to be poor inducers of cell-mediated immunity (Collins 1974; Arnon et al. 1989). OMPs from Salmonella spp. have been reported to protect poultry against experimentally induced Salmonella infection, and used as potential vaccine candidates (Charles et al. 1994; Meenakshi et al. 1999).

In the present study, chickens were injected with the 75·6 and 82·3 kDa OMPs subcutaneously without any adjuvant. It has been reported that OMPs from S. Enteritidis administered with an adjuvant elicited a stronger antibody response than OMP preparations containing no adjuvant (Meenakshi et al. 1999). Moreover, immunization of chickens with Salmonella-OMPs using different adjuvants, and routes of inoculation has been shown to induce various degrees of antibody response (Charles et al. 1994; Hopkins et al. 1995). For example, Charles et al. (1994) vaccinated turkeys with S Enteritidis OMPs using liposomes, mineral oil and immunostimulatory complexes (ISCOM) as adjuvants. The antibody response to positively charged liposomal-OMP vaccine was greater and the vaccine was able to decrease shedding of S. Enteritidis, compared with the mineral oil- and the ISCOM-adjuvanted vaccines.

Salmonella Enteritidis being an enteric pathogen, mucosal immunity (secretory IgA) plays a fundamental role in the protection against Salmonella infection (Hackett 1993). In this study, as the birds were inoculated with the S. enteritidis OMPs subcutaneously, we did not determine the mucosal immune response. Immunization of chickens with the 75·6 and 82·3 kDa OMPs by routes that induce a strong mucosal immune response may potentially increase the inhibitory effect of these proteins on S. Enteritidis colonization in chicken. Although the present study indicated that OMPs 75·6 and 82·3 kDa were effective in reducing in vivo colonization of S. Enteritidis in chickens, the results should be considered preliminary because of potential variations among the tested parameters, and the lack of a large sample size. Therefore, future experiments will involve a larger group of birds and replicates, especially after determining the suitable adjuvant and route for administering the OMPs in chickens.