Antigen-induced B cell apoptosis is independent of complement C4

Abstract

Deficiencies in early complement components are associated with the development of systemic lupus erythematosus (SLE) and therefore early complement components have been proposed to influence B lymphocyte activation and tolerance induction. A defect in apoptosis is a potential mechanism for breaking of peripheral B cell tolerance, and we hypothesized that the lack of the early complement component C4 could initiate autoimmunity through a defect in peripheral B lymphocyte apoptosis. Previous studies have shown that injection of a high dose of soluble antigen, during an established primary immune response, induces massive apoptotic death in germinal centre B cells. Here, we tested if the antigen-induced apoptosis within germinal centres is influenced by early complement components by comparing complement C4-deficient mice with C57BL/6 wild-type mice. We demonstrate that after the application of a high dose of soluble antigen in wild-type mice, antibody levels declined temporarily but were restored almost completely after a week. However, after antigen-induced apoptosis, B cell memory was severely limited. Interestingly, no difference was observed between wild-type and complement C4-deficient animals in the number of apoptotic cells, restoration of antibody levels and memory response.

Fig. 1

Primary and secondary immune response in wild-type and C4^–/– mice measured by anti-4-hydroxy-3-nitrophenylacetyl (NP)-enzyme-linked immunosorbent assay (ELISA). Experimental groups (see also Table 1) were treated differently on day 0 and 9: (a) (wild-type) and (e) (C4^–/–): suicide experiment [injection of 50 µg NP-chicken gamma globulin (CGG) in alum intraperitoneally (i.p.) on day 0 and of 4 mg NP-bovine serum albumin (BSA) i.p. on day 9], (b) (wild-type) and (f) (C4^–/–): injection of an immunogenic dose of antigen (50 µg NP-CGG precipitated in alum i.p) on day 0, (c) (wild-type) and (g) (C4^–/–): injection of a high dose of antigen (4 mg NP-BSA i.p) on day 9, (d) (wild-type) and (h) (C4C4^–/–): no treatment before day 180. All mice were injected with 100 µg NP-BSA on day 180. IgG production was significantly reduced in the C4^–/– mice, but during the secondary immune response there was no difference in the dominant subclass IgG1 between C4^–/– and wild-type mice (a, b, e and f). We also show here that the antigen dose used for inducing the secondary immune response did not induce a significant IgG production in untreated mice, besides the mice that had previously undergone a primary immune response. A single injection of a high dose of antigen did not induce an immune response in wild-type and C4^–/– mice. Injected with 100 µg NP-BSA on day 180, the mice reacted as previously untreated mice did, showing no sign of a secondary immune response (c and g).

Fig. 2

Re-increase in 4-hydroxy-3-nitrophenylacetyl (NP)-specific antibody levels after the suicide experiment in wild-type and C4^–/– mice as an indirect marker for the apoptosis of antigen-specific B cells measured by anti-NP-enzyme-linked immunosorbent assay (ELISA). The left graph shows the IgG1 response, while the right graph shows IgG2b. Both groups (wild-type and C4^–/–) were injected on day 0 with 50 µg NP-chicken gamma globulin (CGG) in alum intraperitoneally (i.p.), and on day 9, at the high of the immune response, with 4 mg NP-bovine serum albumin (BSA) i.p. Specific B cells were not eliminated completely, because their IgG1 and IgG2b levels re-increased after 1 week (*p < 0·05).

Fig. 3

(a) Immunohistochemical staining for terminal deoxynucleotidyl transferase-mediated deoxyuridine triphosphate (dUTP) nick end labelling (TUNEL) (blue, detection of apoptotic cells) and peanut agglutinin (red, staining of germinal centre B cells) of the murine spleen (the bar corresponds to 50 µm). Images are labelled according to experimental group (D and H data not shown). In groups A (wild-type) and E (C4^–/–) (suicide experiment) several apoptotic cell cluster occurred within the germinal centres. In groups B (wild-type) and F (C4^–/–) (normal primary immune response) fewer TUNEL positive clusters were observed. In groups C (wild-type) and G (C4^–/–), which had received only a high dose of antigen without any previous immunization, apoptotic cells were found throughout the red and white pulp of their spleens. (b) Immunohistochemical staining for caspase 3 (red, detection of apoptotic cells) and peanut agglutinin (green, staining of germinal centre B cells) of the murine spleen (the bar corresponds to 50 µm). Images are labelled according to experimental group. In groups A (wild-type) and E (C4^–/–) (suicide experiment) several apoptotic cell cluster occurred within the germinal centres. In groups C (wild-type, data not shown) and G (C4^–/–), which had received only a high dose of antigen without any previous immunization, apoptotic cells were found throughout the red and white pulp of their spleens. (c) Number of apoptotic cell clusters in wild-type and C4^–/– mice (number of TUNEL positive cell clusters per GC/per section in immunohistochemistry). After the injection of a high dose of antigen during an established immune response (suicide experiment) many apoptotic cells were present in the splenic germinal centres. In the spleens of mice on day 9 of an ongoing immune response of untreated mice and of mice which received only a high dose of antigen, no significant difference (P < 0·05) could be identified in the number of apoptotic cells in the germinal centres between the C4^–/– and the wild-type mice.