ABSTRACT
Salmonella enterica serovar Typhimurium colonizes the gut of chickens and is cleared from the intestine within about 3 weeks. Infection induces high levels of specific antibody, but B cells do not play an essential role in clearance of primary infection or in the enhanced clearance after secondary challenge.
Consumption of contaminated poultry meat and eggs is a major cause of human salmonellosis, with the vast majority of these cases caused by Salmonella enterica serovar Enteritidis or Typhimurium. In chickens these serovars colonize the gastrointestinal tract, with only low-level systemic involvement (4). High levels of Salmonella-specific immunoglobulin M (IgM), IgG, and IgA have been reported that coincide with clearance of salmonellae from the gut lumen (5, 7, 13, 14).
B-cell-deficient chickens have been produced experimentally by surgical removal of the bursa of Fabricius or by chemical (cyclophosphamide) or hormonal (testosterone) treatment (reviewed in reference 22). The use of chemical or hormonal treatment is effective at B-cell removal but also transiently affects other cell types, including the gut epithelial cells, bone marrow cells, T cells, and thrombocytes (11, 16, 18-20). Chemical and/or hormonal treatments decrease the capacity to clear enteric Salmonella infection in the chicken (1, 9, 10), and these results have been interpreted as evidence for the involvement of B cells in immunity. While these treatments are effective in producing B-cell-deficient chickens, our data indicate that there are differential effects on the biology of serovar Typhimurium infection depending on the method employed.
The course of infection was compared in intact, surgically or chemically bursectomized line 61 chickens at 6 weeks old. Surgical bursectomy was achieved according to the method of Glick and Olah (12) by removal of the bursa at 17 days of embryonic development. Chemical ablation of B cells was achieved by daily intramuscular injection of 3 mg cyclophosphamide during the first 4 days posthatch (19). Chickens were reared as described previously (5), and all groups were challenged orally with ∼2 × 108 CFU of naladixic acid-resistant serovar Typhimurium F98 (24) at 6 weeks of age. Infection was monitored by plating cloacal swabs onto brilliant green agar supplemented with 20 μg/ml naladixic acid and 1 μg/ml novobiocin as described previously (5). Following incubation (24 h, 37°C), plates were scored using a modified version of the system described by Smith and Tucker (24) (Table 1).
The course of infection was identical with intact and surgically bursectomized chickens (Fig. 1). In contrast, chickens rendered B cell deficient by posthatch treatment with cyclophosphamide had significantly greater numbers of salmonellae than both of the other groups between 1 and 41 days postinfection (dpi) (Fig. 1, representative of three separate experiments). Moreover, most intact and surgically bursectomized chickens cleared infection by 20 dpi, whereas the cyclophosphamide-treated chickens continued to excrete serovar Typhimurium until 41 dpi. Neither surgical nor cyclophosphamide treatment affected the numbers of salmonellae detected in the spleen or liver (data not shown).
Since there was a differential effect of surgical bursectomy and cyclophosphamide treatment on the magnitude and course of infection with serovar Typhimurium, it was important to examine the effectiveness of the respective treatments. The status of the B-cell compartment was verified by assessment of circulating anti-Salmonella antibodies in serum (taken at 21 dpi) and by fluorescence-activated cell sorting analysis of splenocytes. Antigen-specific enzyme-linked immunosorbent assay was performed using a soluble Salmonella antigen preparation (STAgP) as described previously (6). Intact chickens responded to infection by production of STAgP-specific serum IgM, IgG, and IgA (as reported previously (5), whereas no antibody could be detected in either surgically bursectomized or cyclophosphamide-treated chickens (Fig. 2A). Phycoerythrin-labeled anti-BU-1 (Cambridge Bioscience, Cambridge, United Kingdom) recognizes chicken B cells (23, 25) and was used for fluorescence-activated cell sorting analysis. As expected, both surgical bursectomy and cylophosphamide treatment effectively removed B cells from the spleen (0.34% and 0.63% Bu1+ cells, respectively, compared with 19.80% with intact animals) (Fig. 2B).
The differential outcome of infection with cyclophosphamide-treated or surgically bursectomized birds may have been due to chemical disturbance of T-cell responses to serovar Typhimurium antigens. This was examined using a standard [3H]thymidine proliferation assay with splenocytes taken at 48 dpi exposed to antigen (STAgP, 8.1 μg/ml), mitogen (phytohemagglutinin, 20 μg/ml), or unsupplemented medium as described elsewhere (5). Splenocytes from intact or B-cell-deficient chickens proliferated in response to STAgP (proliferation was significantly higher than that with medium alone; P < 0.05) (Fig. 2C) or mitogen (data not shown) with no significant differences according to treatment group. Uninfected birds do not respond to STAgP (data not shown) (5). Supplementation of proliferation assay cultures with irradiated splenocytes (to provide B cells for ex vivo antigen presentation) from infected birds did not alter the capacity of splenocytes from B-cell-deficient chickens to respond to STAgP (data not shown). Although cyclophosphamide treatment transiently affects the T-cell response, these are reported to recover completely by 4 weeks posttreatment (11, 19, 20), and our data confirm this within a Salmonella system.
In the mouse, B cells are not required for the clearance of primary infection with serovar Typhimurium but are involved in immunity to secondary infection (21). Although the systemic nature of serovar Typhimurium infection in mice is quite different from the enteric localization in the chicken, it was appropriate to rechallenge the B-cell-deficient chickens. Following clearance of the primary infection (67 dpi), the intact and B-cell-deficient chickens were rechallenged with spectinomycin-resistant serovar Typhimurium F98 (bacteriologically monitored by cloacal swab using brilliant green agar supplemented with 50 μg/ml spectinomycin). Following rechallenge, all groups cleared the infection more rapidly than they had cleared the primary challenge and at a similar rate irrespective of treatment group. The rechallenged chickens also cleared infection more rapidly than parallel primary (PP) infections in age-matched intact or cyclophosphamide-treated chickens (Fig. 3).
Although substantial antibody responses have been consistently reported in chickens infected with serovar Typhimurium (2, 3, 5, 7, 17), the results presented here indicate that antibodies and B cells are not required to clear either a primary or secondary infection. Our data support the findings of previous studies with cyclophosphamide (1, 10), but the lack of effect of embryonic surgical bursectomy indicates that the effect lies within a non-B-cell compartment. Brownwell et al. (8) used surgical bursectomy to assess the role of the bursa in infection with serovar Typhimurium. Unfortunately, the bursectomy was performed at 8 to 9 days posthatch and would not result in complete B-cell deficiency; therefore, the requirement for B cells could not be determined (15). In our studies, surgical removal of the bursa at 17 days of embryonic development resulted in complete ablation of the B-cell compartment. At present, it is not clear which non-B-cell mechanisms are involved in clearance of primary infection. Moreover, the lack of effect of cyclophosphamide treatment at secondary infection suggests that the mechanism of clearance differs between primary and secondary infection. In the chicken, serovar Typhimurium infection is largely restricted to the gut lumen, and identifying the non-B-cell immune mechanisms that mediate bacterial clearance is important for specific anti-Salmonella vaccine development for chickens and may have implications in the broader context of control of enteric bacterial infection.
Effect of surgical bursectomy or neonatal cyclophosphamide treatment on fecal excretion of serovar Typhimurium in chickens infected at 6 weeks of age. Results are expressed as means of scored swabs and are representative of three experiments. Error bars represent the standard error of the mean, two asterisks indicate a significant difference between the cyclophosphamide-treated group and both other groups, and a single asterisk indicates a significant difference between only the cyclophosphamide-treated and surgically bursectomized chickens using the Mann-Whitney test (P < 0.05).
Characterization of B- and T-cell responses for intact, surgically bursectomized, or cyclophosphamide-treated chickens after infection with serovar Typhimurium. Enzyme-linked immunosorbent assay for STAgP-specific IgM, IgG, and IgA (A), percentage of BU-1+ cells in the spleen (B), proliferation of splenocytes (48 dpi) in response to STAgP (C). Each error bar represents the standard error of the mean, and an asterisk indicates a significant difference from the intact controls (A and B) and between antigen-stimulated and unstimulated controls (C) using Student's t test (P < 0.05).
Effect of bursectomy on fecal excretion of serovar Typhimurium following a secondary infection. Results are expressed as means of scored swabs. Each error bar represents the standard error; an asterisk represents a significant difference between cyclophosphamide PP and intact PP using the Mann-Whitney test (P < 0.05). There were no significant differences between any of the groups following rechallenge.
Scoring of plates for serovar Typhimurium infection
ACKNOWLEDGMENTS
We thank the staff of the production and experimental units of the IAH and the Biotechnology and Biological Sciences Research Council, United Kingdom, and DEFRA-HEFCI for funding this research (grant no. 8/BFP11365 and VT-0104).
FOOTNOTES
- Received 11 August 2005.
- Returned for modification 20 October 2005.
- Accepted 22 November 2005.
- Copyright © 2006 American Society for Microbiology
