CD40-independent help by memory CD4 T cells induces pathogenic alloantibody but does not lead to long-lasting humoral immunity

Abstract

CD40/CD154 interactions are essential for productive antibody responses to T-dependent antigens. Memory CD4 T cells express accelerated helper functions and are less dependent on costimulation when compared with naïve T cells. Here, we report that donor-reactive memory CD4 T cells can deliver help to CD40-deficient B cells and induce high titers of IgG alloantibodies that contribute to heart allograft rejection in CD40-/- heart recipients. While cognate interactions between memory helper T and B cells are crucial for CD40-independent help, this process is not accompanied by germinal center formation and occurs despite inducible costimulatory blockade. Consistent with the extrafollicular nature of T/B cell interactions, CD40-independent help fails to maintain stable levels of serum alloantibody and induce differentiation of long-lived plasma cells and memory B cells. In summary, our data suggest that while CD40-independent help by memory CD4 T cells is sufficient to induce high levels of pathogenic alloantibody, it does not sustain long-lasting anti-donor humoral immunity and B cell memory responses. This information may guide the future use of CD40/CD154 targeting therapies in transplant recipients containing donor-reactive memory T cells.

Keywords: Alloantibody; CD40 costimulation; memory T cells.

Figure 1

Donor-reactive memory CD4 T cells induce alloantibody production in recipients treated with anti-CD154 mAb. Wild type B6 female mice were injected with 5×10⁶ naïve (circles) or in vitro primed (squares) Mar cells. Three weeks later, all animals were treated with anti-CD154 mAb MR1 and transplanted with C3H male heterotopic heart allografts. Control wild type B6 female recipients of C3H male heart allografts did not receive Mar cells or MR1 treatment (triangles). Serum titers of donor or third party-reactive IgG alloAb were determined on d. 14 post transplant. The titers of third party BALB/c-reactive Ab were ≤ 64 for IgG2c, IgG2b and IgG1 and ≤ 512 for IgG3. The experiment was performed three times with similar results.

Figure 2

Memory but not naïve CD4 T cells provide help to CD40−/− B cells. B6 CD40−/− female mice containing naïve Mar cells (open circles) or memory Mar cells (closed circles) were transplanted with C3H male heart allografts. Serum titers of IgG alloantibody isotypes were measured at indicated time points post transplant. In the absence of Mar cell transfer, CD40−/− heart allograft recipients had low serum titers of donor C3H-reactive Ab (≤64 for all IgG isotypes). The titers of third party Balb/c-reactive Ab were ≤ 64 for IgG2c, IgG2b and IgG1 and ≤ 512 for IgG3 in all groups. N=4-8 mice/group. The experiment was performed twice with similar results.

Figure 3

Donor-reactive alloantibody induced by memory CD4 T cells in CD40−/− recipients contribute to heart allograft rejection. A. The kinetics of BALB/c male heart allograft survival in CD40−/− recipients (circles) and in CD40−/− recipients injected with memory Mar cells (squares) with or without CD8 depletion (closed and open symbols, respectively). N = 6-9 animals/group. B. Heart allografts were harvested at the time of rejection. H&E staining and immunohistochemical staining for CD3 and C4d were performed on paraffin-embedded tissue sections. The images are representative of 4-5 rejecting heart allografts analyzed in each group. Original magnification 200×.

Figure 4

CD4 memory T cells provide help for alloantibody response in the absence of spleen germinal center formation. A-C. Immunofluorescence staining of frozen spleen sections from CD40−/− heart allograft recipients containing naïve (A) or memory (B) Mar cells performed on d. 14 post transplant using anti-B220 Ab (red) and PNA (green). Wild type heart transplant recipients previously sensitized with skin allografts were used as a positive control (C). Similar to the findings on d. 14, CD40−/− recipients containing memory Mar cells had no detectable PNA⁺ germinal centers on d. 7 and 21 following transplantation (not shown). Representative images are shown for 4-5 recipients/group. Original magnification 80x. D-E. Spleen and bone marrow cells were isolated from CD40−/− recipients of C3H heart allografts on d. 21 post transplant or from naïve CD40−/− mice as a control and cultured in vitro for 24 h without restimulation. Culture supernatants were collected and examined for C3H-reactive IgG as described in Methods. The data are expressed as channel MFI (D) and percentage of positively stained donor thymocytes (E). For third party BALB/c thymocytes, the values for all samples tested were < 4 and <0.3%, respectively. N=4 mice/group.

Figure 5

Cognate interactions between memory CD4 T cells and B cells are required to initiate alloantibody responses. Bone marrow cells from either wild type or MHC class II −/− mice were mixed at 1:1 ratio with bone marrow cells from μMT mice and injected into lethally irradiated μMT female recipients. A. I-A^b expression on B220⁺ cells at 4 weeks after bone marrow transplantation. B. MHC II−/− → μMT chimeras were intraperitoneally injected with TNP-Ficoll. Serum anti-TNP IgG antibodies were analyzed by ELISA on d. 14 after immunization. N= 4 mice per group. C. MHC II−/− → μMT and WT → μMT bone marrow chimeras and control μMT mice were injected with 5 × 10⁶ primed Mar cells and received male C3H heart allografts three weeks later. Serum samples were collected on d. 21 after transplantation and tested for the presence of IgG antibodies against donor C3H and third party BALB/c alloantigens. The titers of BALB/c-reactive Ab were ≤ 45 for IgG2c, IgG2b and IgG1 and ≤ 135 for IgG3 in all groups. The experiment was performed twice with similar results.

Figure 6

CD40-independent help by memory CD4 T cells does not lead to the formation of long-lived ASCs and memory B cells. BALB/c male heart allografts were placed into CD40−/− recipients containing memory Mar cells (closed circles), CD40−/− recipients without memory cell transfer (open circles) and wild type B6 recipients (triangles). Six weeks after transplantation, all recipients received intraperitoneal injection of 20 × 10⁶ BALB/c male spleen cells. A. Serum titers of donor-reactive IgG alloantibodies were measured at indicated time points post transplant and 2 weeks after challenge (d. 56). B. The frequencies of spleen and bone marrow cells secreting donor D^d-binding IgG were determined by ELISPOT assay. N= 4-6 mice per group at each time point. P values are shown for comparisons between values before and after challenge within the same group.