CD22 is required for formation of memory B cell precursors within germinal centers

Abstract

CD22 is a BCR co-receptor that regulates B cell signaling, proliferation and survival and is required for T cell-independent Ab responses. To investigate the role of CD22 during T cell-dependent (TD) Ab responses and memory B cell formation, we analyzed Ag-specific B cell responses generated by wild-type (WT) or CD22-/- B cells following immunization with a TD Ag. CD22-/- B cells mounted normal early Ab responses yet failed to generate either memory B cells or long-lived plasma cells, whereas WT B cells formed both populations. Surprisingly, B cell expansion and germinal center (GC) differentiation were comparable between WT and CD22-/- B cells. CD22-/- B cells, however, were significantly less capable of generating a population of CXCR4hiCD38hi GC B cells, which we propose represent memory B cell precursors within GCs. These results demonstrate a novel role for CD22 during TD humoral responses evident during primary GC formation and underscore that CD22 functions not only during B cell maturation but also during responses to both TD and T cell-independent antigens.

Fig 1. CD22^-/- B cells mount normal early TD Ab responses but do not form memory B cells or long-lived plasma cells.

Splenocytes from WT or CD22^-/- B1-8^hi mice containing 2 x 10⁵ NP-specific B cells were adoptively transferred to B6 recipients 24 h prior to immunization with 50 micrograms NP-CGG in alum. (A) ELISA analysis of sera from WT (black squares) or CD22^-/- (white circles) recipients measuring anti-NP IgG1^a Abs on the indicated days. Data are from 3–5 mice per group per time point and are representative of 2 independent experiments. (B,C) The frequency of AFCs secreting NP-specific IgG1 Ab from spleen (B) or bone marrow (C) as determined by ELISPOT are plotted. Each dot represents a single animal with the mean indicated by a horizontal line. For D-G, splenoctyes containing 2 x 10⁵ NP-specific B cells from both WT and CD22^-/- B1-8^hi mice were adoptively transferred to Ly5.1 recipient animals that were then immunized as in A. (D) Representative flow cytometry analysis identifying NP-specific splenic B cells 125 days p.i. (E) Total number of NP-specific splenic B cells on day 42, 60 and 125 p.i. Each dot represents a single animal with the mean indicated by a horizontal line. 3 animals were used per time point. (F) Surface expression of the indicated marker was determined on total splenic B cells (shaded histogram) or WT B1-8^hi B cells (black line) from recipient mice immunized 125 days previously. Data are representative of 3 independent experiments using 3 mice per group. (G) Mean number of NP-specific B cells +/- SEM are plotted for immune mice (day 125 p.i.) before and 4 days after re-challenge with 20 micrograms soluble NP-CGG. One of two independent experiments using 3–4 mice/group is shown.

Fig 2. CD22-deficient B cells expand normally and undergo germinal center differentiation.

Adoptive transfers and immunizations were performed as in Fig 1A/1B. (A) Representative flow cytometry analysis showing the frequency of WT and CD22^-/- splenic B1-8^hi B cells 7 days p.i. Data from 3 independent experiments are summarized in (B). Each dot represents a single animal with the mean indicated by a horizontal line. (C) Representative flow cytometric plots show the frequency of NP-specific splenic B cells (gated as in A) that co-express GL7 and CD95 7 days p.i. Data from 3 independent experiments are summarized in (D) where each dot represents a single animal with the mean indicated by a horizontal line. (E) NP-specific B cells were assessed for cell death. Representative flow cytometry plots depict Annexin V binding and Mitotracker Red exclusion among live NP-specific GL7⁺ splenic B cells identified as in (A) 7 days p.i. (F) Summarized data from one of 3 independent experiments using 3 mice per group.

Fig 3. CD22-deficient B cells fail to form CD38^hiCXCR4^hi GC B cells.

Adoptive transfers and immunizations were performed as in Fig 1. (A) Representative flow cytometric plots show CXCR4 and CD86 expression among NP-specific GL7⁺ splenic B cells gated as in Fig 2A 7 days p.i. (B) Summary data from 2 independent experiments using 3 mice per group showing the frequency of NP-specific GL7⁺ LZ (CXCR4^loCD86^hi) B cells 7 days p.i. Each dot represents an individual animal. (C,E) Representative flow plots show the frequency of cells expressing CD38 (C) or CD38 and CXCR4 (E) among NP-specific GL7⁺ B cells 7 days p.i. (D,F) Summary data of CD38 (C) or CD38 and CXCR4 expression (E) from 3 independent experiments using 2–3 mice per group. Each dot represents an individual animal. (G) Histograms show PNA, CCR6 or CD95 expression on each population (P1-P4) of NP-specific GL7⁺ splenic B cells depicted in the diagram at left. Total (shaded), WT (solid line) and CD22^-/- (dashed line) B cells are shown. Data are representative of 2 independent experiments using 5–7 mice each. (H) Frequencies of WT (white circles) and CD22-/- (black circles) NP-specific GL7⁺CD38^hiCXCR4^hi splenic B cells are plotted over time. Data are from one experiment utilizing 7 mice/group at each time point. (I) Summary data depicting the frequency of CD38^hiCXCR4^hi GC B cells in WT and CD22^-/- mice (E) from 2 independent experiments using 4–6 mice per group. Each dot represents an individual animal.

Fig 4. CD22-deficient GC B cells are capable of Ag processing and presentation.

Recipient mice were immunized with 50 micrograms NP-CGG and 6 days later injected i.v. with 5 micrograms NP-streptavidin-I-Ealpha or streptavidin-I-Ealpha. 18 h later mice were sacrificed and spleens analyzed by flow cytometry. (A) Representative analysis of peptide:MHCII expression among NP-specific GL7⁺ WT (solid lines) and CD22^-/- (dashed lines) B cells. Total B220⁺ cells are depicted by shaded histograms for reference. (B) Summarized data from experiment in (A). Each dot represents a single animal with the mean indicated by a horizontal line. Data are representative of 2 independent experiments using 3–7 mice each.