Allogeneic CD4 T Cells Sustain Effective BK Polyomavirus-Specific CD8 T Cell Response in Kidney Transplant Recipients

Abstract

Introduction: BK polyomavirus-associated nephropathy (BKPyVAN) is a significant complication in kidney transplant recipients (KTRs), associated with a higher level of plasmatic BK polyomavirus (BKPyV) replication and leading to poor graft survival.

Methods: We prospectively followed-up with 100 KTRs with various degrees of BKPyV reactivation (no BKPyV reactivation, BKPyV-DNAuria, BKPyV-DNAemia, and biopsy-proven BKPyVAN [bp-BKPyVAN], 25 patients per group) and evaluated BKPyV-specific T cell functionality and phenotype.

Results: We demonstrate that bp-BKPyVAN is associated with a loss of BKPyV-specific T cell proliferation, cytokine secretion, and cytotoxic capacities. This severe functional impairment is associated with an overexpression of lymphocyte inhibitory receptors (programmed cell death 1 [PD1], cytotoxic T lymphocyte-associated protein 4, T cell immunoreceptor with Ig and ITIM domains, and T cell immunoglobulin and mucin domain-containing-3), highlighting an exhausted-like phenotype of BKPyV-specific CD4 and CD8 T cells in bp-BKPyVAN. This T cell dysfunction is associated with low class II donor-recipient human leukocyte antigen (HLA) divergence. In contrast, in the context of higher class II donor-recipient HLA (D/R-HLA) divergence, allogeneic CD4 T cells can provide help that sustains BKPyV-specific CD8 T cell responses. In vitro, allogeneic HLA-mismatched CD4 T cells rescue BKPyV-specific CD8 T cell responses.

Conclusion: Our findings suggest that in KTRs, allogeneic CD4 T cells can help to maintain an effective BKPyV-specific CD8 T cell response that better controls BKPyV replication in the kidney allograft and may protect against BKPyVAN.

Keywords: BK polyomavirus reactivation; BK polyomavirus-associated nephropathy; BK polyomavirus-specific T cell; donor-recipient HLA divergence; kidney transplantation.

Graphical abstract

Figure 1

High levels of plasma BKPyV loads and poor graft survival in KTRs with biopsy-proven BKPyVAN. We examined clinical and biological features in 4 groups of KTR according to the stage of BKPyV reactivation (KTRs without BKPyV reactivation, KTRs with BKPyV-DNAuria, KTRs with BKPyV-DNAemia and KTRs with bp-BKPyVAN; 25 KTRs by group). (a) and (b) show plasma and urinary BKPyV loads. (c) shows graft function in KTR groups. Graft function was evaluated by eGFR (Modification of Diet in Renal Disease-4 formula) at 3 time points: 1-month, BKPyVAN diagnosis or 12 months posttransplantation, and at the end of the follow-up. (d) shows graft survival. Statistical analysis was performed by using Kruskal-Wallis tests followed by Dunn’s multiple comparison tests between the BKPyV-DNuria, BKPyV-DNAemia, or bp-BKPyVAN KTRs (a and b) nonparametric repeated-measures 2-way analysis of variance (c), and Kaplan-Meier survival curves with log-rank (d) tests. Boxes represent the median and 25th and 75th percentiles; whiskers represent the 10th and 90th percentiles. eGFRs are represented as the median and interquartile range. ∗P < 0.05, ∗∗P < 0.005, ∗∗∗P < 0.0005, ∗∗∗∗P < 0.0001. BKPyV, BK polyomavirus; BKPyVAN, BK polyomavirus-associated nephropathy; bp-BKPyVAN, biopsy-proven BKPyVAN; eGFR, estimated glomerular filtration rate; KTR, kidney transplant recipient.

Figure 2

Loss of BKPyV-specific CD8 functionality in KTRs with biopsy-proven BKPyVAN. We examined BKPyV-specific T cell responses in KTR groups according to BKPyV reactivation levels. (a and b) show proliferative capacities of BKPyV-specific CD4 T cells and BKPyV-specific CD8 T cells, respectively. (c) shows representative plots of CD4 and CD8 T cell gating and CFSE dilution following activation with BKPyV peptide pools for 5 days. (d) shows IFNγ and TNFα production capacities of BKPyV-specific CD8 T cells. (e) shows representative plots of coproduction of IFNγ and TNFα in CD8 T cells after overnight activation with BKPyV peptide pools. (f) shows the proportion of KTRs able to perform coproduction of IFNγ and TNFα. (g) shows cytotoxic capacities of BKPyV-specific CD8 T cells. (h) shows representative plots of 7AAD expression on autologous target cells (CD8-depleted PBMCs loaded with BKPyV peptides) in contact with CD8 T cells activated overnight with BKPyV peptide pools. BKPyV-specific CD4 or CD8 T cells were expressed as the number of cells per 10,000 CD4 or CD8 T cells and then subjected to natural logarithm or square root transformation to normalize the data distributions. n represents the number of patients. Two BKPyV-specific responses were analyzed for each patient, after activation with 2 BKPyV-specific peptide pools (LT-Ag and VP1-peptides). Statistical analysis was performed (a, b, d, and g) using Kruskal-Wallis tests followed by Dunn’s multiple comparison tests or (f) by χ² for trend test. Scatter dot plots are shown, with the median and interquartile range. ∗P < 0.05, ∗∗P < 0.005, ∗∗∗P < 0.0005, ∗∗∗∗P < 0.0001. 7AAD, 7-Aminoactinomycin D; BKPyV, BK polyomavirus; BKPyVAN, BK polyomavirus-associated nephropathy; IFNγ, interferon-γ; KTR, kidney transplant recipient; LT-Ag, large tumor antigen; PBMC, peripheral blood mononuclear cell; TNFα, tumor necrosis factor-α; VP1, viral protein-1.

Figure 3

BKPyV-specific CD8 T cell exhaustion in KTRs with biopsy-proven BKPyVAN. We examined the expression of inhibitory receptors on BKPyV-specific CD4 and CD8 T cells in KTR groups as defined above. (a) shows a representative plot of CFSE dilution on CD4 T cells following activation with 2 BKPyV peptide pools (LT-Ag and VP1 peptide pools) for 5 days. (b) shows the correlation between proliferative BKPyV-specific CD4 T cells and plasma BKPyV loads. (c) shows a representative plot of CFSE dilution in CD8 T cells following activation with 2 BKPyV peptide pools for 5 days. (d) shows the correlation between proliferative BKPyV-specific CD8 T cells and plasma BKPyV loads. (e) shows a representative plot of PD1 and CTLA4 expression on IFNγ and/or TNFα producing CD4 T cells following overnight activation with 2 BKPyV peptide pools (Boolean gate analysis). (f) shows a representative plot of PD1 and CTLA4 expression on IFNγ and/or TNFα producing CD8 T cells following overnight activation with 2 BKPyV peptide pools (Boolean gate analysis). (g) and (h) show the expression of PD1 and CTLA4 on BKPyV-specific CD4 T cells, respectively. (i) and (j) show the expression of PD1 and CTLA4 on BKPyV-specific CD8 T cells, respectively. (k) shows a representative plot of PD1, TIGIT, and TIM3 coexpression on proliferative BKPyV-specific CD4 T cells (CFSE low cells following activation for 5 days with 2 BKPyV peptide pools). (l) shows a representative plot of PD1, TIGIT and TIM3 coexpression on proliferative BKPyV-specific CD8 T cells (CFSE low cells following activation for 5 days with 2 BKPyV peptide pools). (m) shows the coexpression of PD1, TIM3, and TIGIT on BKPyV-specific CD4 T cells. (n) shows the coexpression of PD1, TIM3, and TIGIT on BKPyV-specific CD8 T cells. CFSE low BKPyV-specific CD4 or CD8 T cells were expressed as the number of cells per 10,000 CD4 or CD8 T cells and then subjected to natural logarithm transformation to normalize the data distributions (b and d). The proportion of patients expressing the inhibitory receptors PD1, TIM3, and TIGIT, were classified as those expressing none (PD1- TIGIT- TIM3-), 1 (PD1⁺ or TIGIT⁺ or TIM3⁺), 2 (PD1⁺ & TIGIT⁺ or PD1⁺ & TIM3⁺ or TIGIT⁺ & TIM3⁺) or 3 (PD1⁺ TIGIT⁺ TIM3⁺) of the inhibitory receptors (m and n). Statistical analysis was performed by using nonparametric Spearman’s correlation (b and d), Mann-Whitney U tests (g–j), and χ² tests for trend (m and n). Spearman's rank correlation coefficient and the 95% confidence interval are represented. Box and whiskers plots show the median and (10th–90th percentile). ∗P < 0.05, ∗∗P < 0.005, ∗∗∗∗P < 0.0001. BKPyV, BK polyomavirus; BKPyVAN, BK polyomavirus-associated nephropathy; CTLA4, cytotoxic T lymphocyte-associated protein 4; IFNγ, interferon-γ; KTR, kidney transplant recipient; LT-Ag, large tumor antigen; PBMC, peripheral blood mononuclear cell; PD1, programmed cell death 1; TIGIT, T cell immunoreceptor with Ig and ITIM domains; TIM3, T cell immunoglobulin and mucin domain-containing-3; TNFα, tumor necrosis factor-α; VP1, viral protein-1.

Figure 4

Mismatched allogeneic CD4 T cells restore BKPyV-specific CD8 T cell functionality by allogeneic CD4 help. (a) shows class I and class II D/R-HLA divergence (calculated using the Grantham score) in KTRs with BKPyV-DNAemia and biopsy-proven BKPyVAN. (b) corresponds to the schematic representation of allogeneic CD4 T cell rescue to BKPyV-specific CD8 T cells. (c) shows CD8 proliferation (CFSE dilution) with and without BKPyV peptide pools in the presence of autologous or allogeneic CD4 T cells. Briefly, PBMCs from bp-BKPyVAN KTRs were depleted from CD4 T cells and cultured for 5 days with allogeneic or autologous CD4 T cells and BKPyV peptide pools (Methods). A control group corresponds to PBMCs from BKPyV-DNAemia KTRs treated in autologous condition. (d) shows BKPyV-specific CD8 proliferation in the presence of autologous or allogeneic CD4 T cells in bp-BKPyVAN KTRs and the control group. Results are expressed as delta percentages of CFSE low CD8 T cells between the conditions with and without BKPyV peptide pools. Statistical analysis was performed by using (a) Mann-Whitney U tests, (c) Wilcoxon matched-pairs signed-rank tests between autologous and allogeneic comparisons, and (d) Kruskal-Wallis tests followed by Dunn’s multiple comparison tests. ∗∗P < 0.005, ∗∗∗P < 0.0005, ∗∗∗∗P < 0.000. BKPyV, BK polyomavirus; BKPyVAN, BK polyomavirus-associated nephropathy; CD4 BK-v, BKPyV-specific CD4 T cell; CD8 BK-v, BKPyV-specific CD8 T cell; donor APC, donor antigen-presenting cell; D/R-HLA divergence, donor-recipient human leukocyte antigen divergence; KTR, kidney transplant recipient; ns, nonsignificant.

Figure 5

A schematic representation of how high-class II D/R-HLA divergence may improve BKPyV-specific CD8 T cell responses in KTRs. BKPyV, BK polyomavirus; CD4 allo, allogeneic CD4 T cell; CD4 BK-v, BKPyV-specific CD4 T cell; CD8 BK-v, BKPyV-specific CD8 T cell; donor APC, donor antigen-presenting cell; D/R-HLA divergence, donor-recipient human leukocyte antigen divergence; MHC, major histocompatibility complex.