Memory B cells, but not long-lived plasma cells, possess antigen specificities for viral escape mutants

Abstract

Memory B cells (MBCs) and long-lived plasma cells (LLPCs) persist after clearance of infection, yet the specific and nonredundant role MBCs play in subsequent protection is unclear. After resolution of West Nile virus infection in mice, we demonstrate that LLPCs were specific for a single dominant neutralizing epitope, such that immune serum poorly inhibited a variant virus that encoded a mutation at this critical epitope. In contrast, a large fraction of MBC produced antibody that recognized both wild-type (WT) and mutant viral epitopes. Accordingly, antibody produced by the polyclonal pool of MBC neutralized WT and variant viruses equivalently. Remarkably, we also identified MBC clones that recognized the mutant epitope better than the WT protein, despite never having been exposed to the variant virus. The ability of MBCs to respond to variant viruses in vivo was confirmed by experiments in which MBCs were adoptively transferred or depleted before secondary challenge. Our data demonstrate that class-switched MBC can respond to variants of the original pathogen that escape neutralization of antibody produced by LLPC without a requirement for accumulating additional somatic mutations.

Figure 1.

LLPCs recognize the K307/T330, but not a mutant K307E/T330I epitope. (A) Mice were infected with WNV, and serum was collected at the indicated time points after infection. Levels of DIII or DIII-K307E/T330I–specific IgG were measured by ELISA. Endpoint titers are expressed as the reciprocal serum dilution that was 3 SD above background. The data reflect 6–10 mice per time point from four experiments. (B) Bone marrow cells were collected at the indicated time points after WNV infection and DIII or DIII-K307E/T330I–specific plasma cells were enumerated by ELISPOT assay. The data reflect six to nine mice per time point from four experiments. (C) Bone marrow cells were collected from mice infected with WNV or Chikungunya virus (CHIKV) and antigen-specific plasma cells were enumerated by intracellular staining. Data are shown as percentage of total B220⁻CD138⁺ cells. The percentage of cells staining as DIII or DIII-K307E/T330I–specific LLPC in CHIKV-infected mice was considered background and subtracted (0.1%). The data reflect five mice from three experiments. Horizontal bars in A–C indicate the mean. (D) Representative flow cytometry plots showing DIII or DIII-K307E/T330I–specific plasma cells from mice at 60 d after infection with WNV or CHIKV. Statistical significance (***, P < 0.001; **, P < 0.01; *, P < 0.05) was determined by an unpaired, two-tailed Student’s t test.

Figure 2.

LLPCs weakly neutralize a variant WNV. (A) Mice were infected with WNV or WNV-K307E (left) or immunized with an inactivated WNV vaccine (right), and sera were collected at day 60. Neutralizing antibody titers were measured by plaque reduction assay. Means and SD are shown of six experiments. (B) EC50 titers were calculated from neutralization curves of WNV-WT or WNV-K307E incubated with serum from WNV-infected or vaccinated mice. Means and SD are shown from three or six experiments. The dashed line represents the limit of detection of assay. (C) Mice were injected with 1 µl WNV immune serum or PBS before infection with WNV or WNV-K307E or PBS. Data reflect 20–26 mice per group in three experiments. Survival differences were determined by the log-rank test. NS, not significant. (D) Mice were injected with 1 µl of immune or PBS immediately before infection with WNV (left) or WNV-K307E (right) mice. Viral RNA in serum was measured by qRT-PCR. Means and SD are shown of two experiments with five mice per group. Statistical significance (***, P < 0.001; **, P < 0.01; *, P < 0.05) was determined by an unpaired, two-tailed Student’s t test.

Figure 3.

MBCs recognize mutant DIII-K307E/T330I epitopes. (A) Mice were infected with WNV, splenocytes were collected at the indicated time points, and CD19⁺ B cells were isolated by positive selection. The frequency of DIII- or DIII-K307E/T330I-specific IgG⁺ MBC was assessed by LDA. Virus-specific IgG in supernatants from LDA cultures were measured by ELISA and frequencies of virus-specific MBC determined by linear regression analysis. The limit of detection was ∼2 MBCs per 10⁶ B cells (dashed line), and data reflect 5–14 mice per group from four experiments. P-values were determined using an unpaired, two-tailed Student’s t test. (B) The frequency of splenic DIII- or DIII-K307E/T330I-specific MBC was assessed by flow cytometry. The percentage of cells staining as DIII⁺ or DIII-K307E/T330I⁺ MBC in CHIKV-infected mice was considered background and subtracted (6.1 and 5.4 MBCs per 10⁶ CD19⁺ cells). Data reflect 7–11 mice per group from five experiments and are displayed as frequency of antigen-specific MBCs per 10⁶ CD19⁺ cells. P-values were determined using an unpaired, two-tailed Student’s t test. (C) Flow cytometry plots were from B and show MBC recognizing DIII or DIII-K307E/T330I from WNV- or CHIKV-infected mice. (D) Ratio of frequency of DIII-K307E/T330I- to DIII-specific MBC or LLPC in individual mice as measured by flow cytometry, LDA, and ELISPOT. MBC frequencies were shown in A and B, and LLPC frequencies were from Fig 1 (B and C). P-values were determined by the Mann-Whitney test (***, P < 0.001; **, P < 0.01; *, P < 0.05). Horizontal bars indicate the mean.

Figure 4.

MBC clones can recognize variant epitope better than WT DIII. (A) Percentage of MBC clones that bind DIII, both DIII and DIII-K307E/T330I, or DIII-K307E/T330I by ELISA. Positive-scoring wells of the lowest dilution from LDA in A were assigned specificities based on binding to DIII and DIII-K307E/T330I. Data represent ∼9.8 clones per mouse from 8–14 mice (low frequency of MBC at day 300 precluded analysis). (B and C) Individual WNV-specific MBCs were sorted and cultured. Supernatant was tested for binding to DIII or DIII-K307E/T330I by ELISA to identify DIII-LR–specific (B) or DIII–cross-reactive (C) MBC. Data represent four independent experiments. (D) Clone 2F2 was expressed ectopically in 293T cells and supernatant was tested for binding to DIII or DIII-K307E/T330I. Representative results are shown, and statistical significance was determined using a paired, two-tailed Student’s t test using data from four independent experiments performed in duplicate (*, P < 0.01).

Figure 5.

MBCs can respond to and neutralize variant virus. (A) Neutralization curves of WNV-WT or WNV-K307E with supernatant from stimulated MBCs. Data reflects four independent experiments performed in duplicate. Error bars indicate SD. (B) IgH^a MBCs (CD19⁺IgM⁻IgD⁻lin⁻CCR6⁺CD80⁺) from WNV-vaccinated mice were sorted and transferred into allotypic IgH^b recipients and challenged with WNV-WT or WNV-K307E 1 d later. Specificity of the MBC-derived antibody at different time points was determined by ELISA using an IgG2a^a-specific detection antibody. Data reflects eight mice per group and is the average of five experiments (C) Mice were immunized with JEV-DIII and treated with B cell–depleting (CD20 mAb) or an isotype control antibody. Serum was tested before WNV infection (left) and at days 3 and 8 (right) after infection for IgG binding to WNV-DIII by ELISA. The data reflect a total of five to nine mice per group from three experiments. P-values were determined using an unpaired, two-tailed Student’s t test (*, P < 0.01). NS, indicates not significant.