Genetic Variation Influences the B-Cell Response to Immunization with a Pneumococcal Polysaccharide Conjugate Vaccine

Abstract

CBA/J mice immunized with pneumococcal 23F-CRM₁₉₇ vaccine produce significantly lower titers of 23F-specific antibodies and fewer 23F-specific antibody-secreting cells (ASC) than did BALB/c or (CBA/J × BALB/c)F₁ (CCBAF₁) mice. The reduced 23F-specific titers of CBA/J versus BALB/c or CCBAF₁ mice are presumably related to lower frequencies of 23F-specific ASC influenced by genetic variation.

The multivalent pneumococcal (Pn) polysaccharide (PS)-protein conjugate vaccine includes seven different Pn serotypes separately conjugated to the identical carrier protein CRM₁₉₇ (cross-reactive material, a nontoxic diphtheria toxin mutant). In contrast to the unconjugated PnPS vaccines, this vaccine protects infants against infection with vaccine serotypes of Streptococcus pneumoniae (6, 7). Despite the covalent coupling of each PnPS component of the multivalent conjugate vaccine to the same carrier protein, some serotypes elicit relatively low titers of PS-specific antibodies (2, 3, 6). Evidence obtained with mice and humans demonstrates that responsiveness to PnPS in general may be genetically controlled (1, 4, 5).

We previously described low 23F-specific antibody titers after immunization with the 23F-CRM₁₉₇ vaccine in CBA/J mice, despite high CRM₁₉₇-specific antibody titers and the presence of CRM₁₉₇-reactive T cells (3). Immunization of CBA/J mice with 6B-CRM₁₉₇ or 19F-CRM₁₉₇ conjugate vaccines resulted in high levels of both PS- and CRM₁₉₇-specific antibodies. Although immunization with 23F-CRM₁₉₇ yielded responses to different sets of CRM₁₉₇-derived peptides than did immunization with the 6B-CRM₁₉₇ and 19F-CRM₁₉₇ conjugates, the overall levels of T-cell proliferation in response to any of the different conjugates were comparable. Thus, we concluded that the apparent qualitative alterations in the T-cell response to 23F-CRM₁₉₇ in CBA/J mice were unlikely to be responsible for the poor immunogenicity of serotype 23F PS (3).

In the present study, we asked if the weak PS-specific antibody responses of CBA/J mice following immunization with 23F-CRM₁₉₇ is influenced by genetic differences between distinct inbred strains of mice. Therefore, we compared responses to 23F-CRM₁₉₇ conjugate vaccine among CBA/J, BALB/c, and (CBA/J × BALB/c)F₁ (CCBAF₁) mice. Both serum titers of 23F-specific antibody and numbers of 23F-specific antibody-secreting cells (ASC) after immunization were assessed.

Mice of the three genetic backgrounds (CBA/J, BALB/c, and CCBAF₁) were immunized intraperitoneally with either 19F-CRM₁₉₇ or 23F-CRM₁₉₇ and boosted 2 weeks later. Eight-week-old pathogen-free female mice were immunized with 10 μg (PS content) of 19F-CRM₁₉₇ or 23F-CRM₁₉₇ (kindly supplied by Ron Eby, Wyeth-Lederle Vaccines, West Henrietta, N.Y.). The PS/protein mass ratios of the vaccines were as follows: 19F-CRM₁₉₇, 0.66; 23F-CRM₁₉₇, 0.52 (R. Eby, personal communication). CBA/J and BALB/c mice were obtained from The Jackson Laboratory (Bar Harbor, Maine). CCBAF₁ mice were obtained by in-house breeding. Ten micrograms of PnPS was chosen as the immunizing dose because previous experiments had demonstrated that this is the minimal amount of antigen that is able to yield maximal or near-maximal antibody titers in CBA/J mice (3). Mice immunized with sterile phosphate-buffered saline served as negative controls. Sera were screened for anti-PnPS and anti-CRM₁₉₇ antibodies via an enzyme-linked immunosorbent assay (ELISA) in which 96-well plates were coated with PnPS or CRM₁₉₇ as previously described (3). Quantitation of antigen-specific antibody levels was based on standard curves generated with distinct murine monoclonal antibodies specific for, respectively, 19F PS (59-1; immunoglobulin G1 [IgG1]), 23F PS (53-2; IgG1), or CRM₁₉₇ (E7-10; IgG1). The monoclonal antibodies were all generously provided by Phil Fernsten, Wyeth-Lederle Vaccines.

Immunization of CBA/J mice with 23F-CRM₁₉₇ resulted in low-level anti-23F antibody titers. In contrast, immunization of BALB/c and CCBAF₁ mice with 23F-CRM₁₉₇ resulted in 23F-specific antibody titers significantly greater than those of CBA/J mice (P < 0.004) (Fig. 1A). CBA/J and BALB/c mice also produced significantly lower titers of CRM₁₉₇-specific antibodies (Fig. 1C).

FIG. 1.

23F- and CRM₁₉₇-specific responses of BALB/c, CBA/J, and CCBAF₁ mice after two immunizations with 23F-CRM₁₉₇. (A) Total 23F-specific serum immunoglobulin κ chain titers (micrograms per milliliter) detected by PnPS solid-phase ELISA. (B) Frequency of 23F-specific κ ASC determined by PnPS solid-phase ELISpot assay. The data shown are a compilation of two experiments with the mean ± the standard error of the mean of a total of 24 mice per group. *, P < 0.004. (C)Total CRM₁₉₇-specific serum immunoglobulin κ chain titers (micrograms per milliliter) detected by CRM₁₉₇ solid-phase ELISA. (D) Frequency of CRM₁₉₇-specific κ ASC determined by PnPS solid-phase ELISpot assay. The data shown are a compilation of two experiments with the mean ± the standard error of the mean of a total of 24 mice per group. *, P < 0.003 for comparisons of selected groups with CCBAF₁.

We also determined the frequencies of PS- and CRM₁₉₇-specific ASC by ELISpot assay after immunization with 23F-CRM₁₉₇ as previously described (2, 8). Multiscreen plates (Millipore) were coated with 100 μl of phosphate-buffered saline containing 10 μg of 19F or 23F PS per ml or 1 μg of CRM₁₉₇ per ml. Spleens were harvested 6 days after a secondary immunization, a single-cell suspension was prepared, and 10⁶ cells were added to each well after washing. ASC were detected with goat anti-mouse κ-alkaline phosphatase and 5-bromo-4-chloro-3-indolylphosphate (BCIP)-nitroblue tetrazolium in nitroblue tetrazolium buffer (Sigma). The spots were enumerated with an ImmunoSpot series 1.0 analyzer (Optimas, Bothell, Wash.). Spots could not be detected in unimmunized mice or in mice immunized once in our assay system (data not shown).

The frequencies of both 23F- and CRM₁₉₇-specific ASC were substantially decreased (P < 0.004 and P < 0.03, respectively) in CBA/J mice in comparison to the 23F- and CRM₁₉₇-specific ASC frequencies of BALB/c and CCBAF₁ mice (Fig. 1B and D), consistent with the serum antibody measurements. In contrast to these results, immunization of CBA/J, BALB/c, and CCBAF₁ mice with 19F-CRM₁₉₇ yielded roughly comparable serum antibody titers and ASC with specificities for both 19F PS and CRM₁₉₇ (Fig. 2A to D). Thus, CBA/J mice have the ability to respond, similarly to BALB/c mice, following immunization with CRM_197. The low numbers of 23F-specific ASC in CBA/J mice and the presence of significantly higher numbers in BALB/c and CCBAF₁ mice suggest a genetic influence over the antibody response to 23F, such that a stronger response is expressed in a dominant fashion.

FIG. 2.

19F- and CRM₁₉₇-specific responses of BALB/c, CBA/J, and CCBAF₁ mice after two immunizations with 19F-CRM₁₉₇. (A) Total 19F-specific serum immunoglobulin κ chain titers (micrograms per milliliter) detected by PnPS solid-phase ELISA. (B) Frequency of 19F-specific κ ASC determined by PnPS solid-phase ELISpot assay. The data shown are a compilation of two experiments with the mean ± the standard error of the mean of a total of 16 mice per group. (C) Total CRM₁₉₇-specific serum immunoglobulin κ chain titers (micrograms per milliliter) detected by CRM₁₉₇ solid-phase ELISA. (D) Frequency of CRM₁₉₇-specific κ ASC determined by CRM₁₉₇ solid-phase ELISpot assay. The data shown are a compilation of two experiments with the mean ± the standard error of the mean of a total of 16 mice per group.

We next sought to determine the ability of CBA/J mice to produce anti-23F antibodies after priming with the immunogenic conjugate 19F-CRM₁₉₇, thereby potentially eliciting carrier-specific memory T cells in the absence of 23F-specific memory B cells. Mice were immunized with 19F-CRM₁₉₇ and then immunized 2 weeks later with either 23F-CRM₁₉₇ or 19F-CRM₁₉₇. Additional mice were primed with 23F-CRM₁₉₇ and then immunized with 19F-CRM₁₉₇ or 23F-CRM₁₉₇. As expected, mice primed with 19F-CRM₁₉₇ and boosted with 19F-CRM₁₉₇ produced anti-19F antibodies primarily of the IgG1 isotype and at titers (Fig. 3A) similar to those produced by BALB/c and CCBAF₁ mice (Fig. 2A). Mice primed with 23F-CRM₁₉₇ and boosted with 23F-CRM₁₉₇ produced minimal to undetectable levels of 23F-specific antibodies (Fig. 3B). Mice primed with 19F-CRM₁₉₇, followed by 23F-CRM₁₉₇, produced levels of 19F-specific antibodies consistent with a single immunization (Fig. 3A), but the level of 23F-specific antibodies remained undetectable. Priming with 23F-CRM₁₉₇, followed by immunization with 19F-CRM₁₉₇, also did not yield detectable levels of 23F-specific antibodies (Fig. 3B). These data suggested that lack of CRM₁₉₇-specific T-cell help due to alterations from the conjugation to the 23F PS was not the cause of the poor 23F-specific antibody response since priming with an immunogenic CRM₁₉₇ conjugate did not improve the 23F-specific antibody response.

FIG. 3.

PS- and CRM₁₉₇- specific antibody titers (micrograms per milliliter) of CBA/J mice primed with either 19F-CRM₁₉₇ or 23F-CRM₁₉₇ and boosted with either conjugate. Total serum immunoglobulin κ chain titers (micrograms per milliliter) of 19F-specific (A), 23F-specific (B), and CRM₁₉₇-specific (C) antibodies in CBA/J mice detected by PnPS (A and B) and CRM₁₉₇ (C and D) solid-phase ELISAs. 19F-specific (A) and CRM₁₉₇-specific (D) IgG1 levels detected in the same sera are also shown. The mean titers (micrograms per milliliter) ± the standard error of the mean of groups of five mice are shown.

Despite the poor 23F-specific immunogenicity of the 23F-CRM₁₉₇ conjugate vaccine, CRM₁₉₇-specific IgG1 antibodies were produced in CBA/J mice after immunization (Fig. 3D). In addition, mice primed with 19F-CRM₁₉₇ and boosted with 23F-CRM₁₉₇ and vice versa produced equivalent amounts of CRM₁₉₇-specific antibodies (Fig. 3C). The CRM₁₉₇-specific antibody response after the second immunization increased in magnitude and consisted primarily of CRM₁₉₇-specific IgG1 (Fig. 3D), typical of a T cell-dependent response. These data also suggest that both 19F-CRM₁₉₇ and 23F-CRM₁₉₇ are capable of eliciting T-cell help for CRM₁₉₇-specific B cells, since T-cell help is required for class switching to IgG1 and increased antibody production after reimmunization. However, the presence of this CRM₁₉₇-specific T-cell help did not result in 23F-specific antibody production. Thus, it is likely that even with the availability of CRM₁₉₇-derived T-cell help, CBA/J mice are unable to produce a substantial 23F-specific antibody response (3).

In order to determine if the differences in 23F PS- or CRM₁₉₇-specific immunogenicity between mouse strains after immunization with 23F-CRM₁₉₇ might involve differences in T-cell responsiveness, we compared both proliferative responses of lymph node cells to whole CRM₁₉₇ and the specificities of T-cell proliferative responses for CRM₁₉₇-derived peptides. While the sets of peptides for the CBA/J and BALB/c cells were not identical (although overlapping), the overall proliferative responses to whole CRM₁₉₇ were comparable (data not shown). These results, in conjunction with the results from sequential immunization with 23F-CRM₁₉₇ and 19F-CRM₁₉₇, suggest that the lack of response to 23F PS (and CRM₁₉₇) by CBA/J mice is not primarily attributable to the T-cell compartment.

The most likely explanation for these findings is a genetically influenced paucity of 23F-specific B-lymphocyte precursors. If in mice, as Zhou et al. (9) found in humans, antibody responses to 23F PS involve a restricted number of light- or heavy-chain variable-region genes, polymorphisms in the relevant variable-region genes might influence the production of 23F-specific precursors. How a relative deficiency of 23F PS-specific precursors in CBA/J mice would account for the relatively poor CRM₁₉₇-specific antibody responses following immunization of these mice with 23F-CRM₁₉₇ remains unclear. Perhaps PS-specific B cells represent a significant proportion of the antigen-presenting cell pool for carrier-specific T cells. Although we cannot absolutely rule out the possibility that conjugation of 23F PS to CRM₁₉₇ generates a uniquely nonstimulatory or inhibitory epitope(s), the ability of 23F-CRM₁₉₇ to prime effectively for an antibody response to CRM₁₉₇ upon a challenge with 19F-CRM₁₉₇ (Fig. 3C and D) argues against these possibilities. Furthermore, we have previously shown that simultaneous immunization with both 23F- and 19F-CRM₁₉₇ conjugates does not preclude a strong response to 19F PS, suggesting that CRM₁₉₇-specific T-cell responses were not inhibited by the 23F-CRM₁₉₇ conjugate (3).

In conclusion, CBA/J mice were poorly responsive after immunization with Pn serotype 23F-CRM₁₉₇, with antibody titers and 23F-specific ASC frequencies markedly lower than those of BALB/c mice. These results suggest that genetic differences between CBA/J and BALB/c mice contribute to the quantitative differences in the 23F PS-specific antibody responses elicited by 23F-CRM₁₉₇. Further experiments are necessary to identify murine loci at which allelic variation influences the magnitude of the antibody response to protein-conjugated Pn capsular PS.