Candida-elicited murine Th17 cells express high Ctla-4 compared with Th1 cells and are resistant to costimulation blockade

Abstract

Effector and memory T cells may cross-react with allogeneic Ags to mediate graft rejection. Whereas the costimulation properties of Th1 cells are well studied, relatively little is known about the costimulation requirements of microbe-elicited Th17 cells. The costimulation blocker CTLA-4 Ig has been ineffective in the treatment of several Th17-driven autoimmune diseases and is associated with severe acute rejection following renal transplantation, leading us to investigate whether Th17 cells play a role in CD28/CTLA-4 blockade-resistant alloreactivity. We established an Ag-specific model in which Th1 and Th17 cells were elicited via Mycobacterium tuberculosis and Candida albicans immunization, respectively. C. albicans immunization elicited a higher frequency of Th17 cells and conferred resistance to costimulation blockade following transplantation. Compared with the M. tuberculosis group, C. albicans-elicited Th17 cells contained a higher frequency of IL-17(+)IFN-γ(+) producers and a lower frequency of IL-10(+) and IL-10(+)IL-17(+) cells. Importantly, Th17 cells differentially regulated the CD28/CTLA-4 pathway, expressing similarly high CD28 but significantly greater amounts of CTLA-4 compared with Th1 cells. Ex vivo blockade experiments demonstrated that Th17 cells are more sensitive to CTLA-4 coinhibition and therefore less susceptible to CTLA-4 Ig. These novel insights into the differential regulation of CTLA-4 coinhibition on CD4(+) T cells have implications for the immunomodulation of pathologic T cell responses during transplantation and autoimmunity.

Figure 1. Candida elicited CD4⁺ T cells can mediate costimulation blockade resistant graft rejection

(A) Mice were adoptively transferred with 10⁶ OT-I and OT-II cells 24 h before immunization with M.Tb or Candida and OVA peptide. (B–C) Draining popliteal LNs were isolated on (B) day 9 or (C) day 14 post-immunization and the frequency and absolute number of of Thy1.1⁺ OT-II cells were determined (p = n.s., n = 10–12/group, 2 experiments). (D–E) At 14 d post-immunization, mice were transplanted with mOVA skin and monitored for graft survival (n = 13–14/group, 3 experiments). (D) Mice were left untreated (p = n.s.) or (E) treated with 500 µg each CTLA-4 Ig and anti-CD154 (p = 0.0002). Statistical comparisons performed using unpaired two-tailed Student’s t-test (B) and log rank test (D–E), *p < 0.05, ***p < 0.0005. No Immun., no immunization control.

Figure 2. Candida elicits a higher frequency of Th17 cells than M.Tb

Draining popliteal LNs from M.Tb OVA and Candida OVA mice were collected on day 9 post-immunization and restimulated for 4 h with PMA/Iono (n = 10–13/group, 2 experiments). The frequencies of (A) IFN-γ⁺ Th1 and (B) IL-17⁺ Th17 cells among antigen-specific CD4⁺ T cells (p = 0.0005). (C) Expression of RORγT expression on Th1 and Th17 cells. Statistical comparisons performed using unpaired two-tailed Student’s t-test, ***p<0.0005.

Figure 3. M.Tb and Candida elicit a similar CD4⁺ T_EM phenotype

(A–B) Draining popliteal LNs from M.Tb OVA and Candida OVA were collected on day 14 post immunization (n = 7–8/group, 2 experiments). (A) The frequency of antigen-specific OT-II CD44^high cells in both groups. (B) The frequency of OT-II T_EM and T_CM cells by CD44 and CCR7 expression. (C–D) Draining popliteal LNs from M.Tb OVA and Candida OVA were collected on day 9 post immunization and restimulated for 4 h with PMA/Iono. Frequency of (C) IL-2 and (D) TNF expression among OT-II CD4⁺ T cells (n = 10–13/group, 2 experiments). Statistical comparisons performed using unpaired two-tailed Student’s t-test, *p<0.05.

Figure 4. Candida elicits more pathogenic Th17 cells than M.Tb

Draining popliteal LNs from M.Tb OVA and Candida OVA mice were collected on day 9 post-immunization and restimulated for 4 h with PMA/Iono (n = 7–12/group, 2 experiments. (A) The frequency of IFN-γ⁺IL-17⁺ cells among OT-II cells (p = 0.0042). (B) CCR6 expression was assessed on OT-II Th1 and Th17 cells (p = 0.0422). (C) The frequency of IL-10⁺ cells among OT-II cells (p = 0.0040). (D) The frequency of IL-10⁺IL-17⁺ cells among OT-II IL-17⁺ cells was assessed (p = 0.0035) Bar graphs show average ± SEM. Statistical comparisons performed using unpaired two-tailed Student’s t-test, *p<0.05, **p<0.005.

Figure 5. Th17 cells express higher levels of CTLA-4 and are less susceptible to CTLA-4 Ig than Th1 cells

Draining popliteal LNs from M.Tb OVA and Candida OVA mice were collected on day 9 post-immunization and (A–D) restimulated for 4 h with PMA/Iono or (E–F) co-cultured with OVA_323–339 and DCs in the presence of CTLA-4 Ig or IgG-Fc for 4 d before brief PMA/Iono restimulation. Data shown depict Th1 and Th7 OT-II cells from both immunization groups (n = 16–21/group, 3–4 experiments). (A) The frequency of CD28⁺ and (B) CD28 expression level on antigen-specific Th1 and Th17 cells (p = n.s.). (C) The frequency of CTLA-4⁺ (p = 0.0006) and (D) CTLA-4 expression level (p = 0.0044) on OT-II Th1 and Th17 cells. (E) CD11c⁺ DCs were purified from naïve splenocytes and analyzed for MHC Class II I-A^b, CD80, and CD86 expression. (F) M.Tb OVA and Candida OVA CD4⁺ T cells were co-cultured with DCs for 4 d in the presence of CTLA-4 Ig IgG-Fc control followed by brief PMA/Iono restimulation. (F) Representative frequencies (top number) and absolute numbers (bottom number) of OT-II Th1 and Th17 cells without PMA/Iono restimulation (first row) or with restimulation (middle/bottom rows, p = 0.0015). Bar graphs include average ± SEM. Statistical comparison performed using (A–D) unpaired or (F) paired two-tailed Student’s t-test. **p<0.005. Endog. CD4⁺, endogenous CD4⁺ T cells. DCs, CD11c⁺ splenic dendritic cells. Isotype, IgG isotype control antibody.

Figure 6. Th17 Cells are more susceptible to CTLA-4 coinhibition than Th1 cells

Draining popliteal LNs from M.Tb OVA and Candida OVA mice were collected on day 9 post-immunization and co-cultured with OVA_323–339 and DCs in the presence of anti-CTLA-4 or IgG for 4 d before brief PMA/Iono restimulation. Data shown depict OT-II Th1 and Th7 cells from both immunization groups (n = 13–15/group, 3 experiments). (A) Representative frequencies (top number) and absolute numbers (bottom number) of OT-II Th1 and Th17 cells without PMA/Iono restimulation (first row) or with restimulation (middle/bottom rows) (B) Th1 and Th17 cell numbers from M.Tb OVA and Candida OVA groups following anti-CTLA-4 blockade (p = 0.0387). Bar graphs depict average ± SEM. Statistical comparison performed using unpaired two-tailed Student’s t-test *p<0.05. Endog. CD4⁺, endogenous CD4⁺ T cells. DCs, CD11c⁺ splenic dendritic cells. Isotype, IgG isotype control antibody.

Figure 7. Th1 and Th17 cells express similar levels of CD154 and are inhibited by CD154 blockade

Draining popliteal LNs from M.Tb OVA and Candida OVA mice were collected on day 9 or day 14 post-immunization and restimulated for 4 h with PMA/Iono (n = 5–7/group, 2 experiments) or (D–E) co-cultured with OVA_323–339 and DCs in the presence of anti-CD154 or IgG for 4 d before brief PMA/Iono restimulation (n = 8–9/group, 2 experiments). (A) Representative histograms and (B) CD154 expression of naïve OT-II, M.Tb OVA, or Candida OVA LN cells on day 9 post immunization. (C) CD154 expression of naïve OT-II, M.Tb OVA, or Candida OVA LN cells on day 14 post immunization. (D) Representative frequencies (top number) and absolute numbers (bottom number) of OT-II Th1 and Th17 cells without PMA/Iono restimulation (first row) or with restimulation (middle/bottom rows). (E) Th1 and Th17 cell numbers from M.Tb OVA and Candida OVA groups following anti-CD154 blockade (Candida Th1/Th17 p = 0.0008). Bar graphs depict average ± SEM. Statistical comparison performed using unpaired two-tailed Student’s t-test *p<0.05, **p<0.005. IgG, isotype control antibody.

Figure 8. Candida polarized mice have greater neutrophil recruitment to skin grafts in the presence of CTLA-4 Ig

M.Tb OVA and Candida OVA mice received an OVA skin transplant on day 14 post-immunization. Mice were left untreated or were treated with 500 µg each CTLA-4 Ig and anti-CD154. On day 6 post-graft skin grafts were harvested and snap frozen for histological analysis of (A) macrophage or (B) neutrophil infiltration. Cellular infiltration was quantified using Aperio ImageScope software (n = 20–22 image fields/group). Bar graphs show average±SEM. Statistical analysis was performed using one-way ANOVA with Bonferroni post test. *p<0.05, **p<0.005, ***p<0.0005.