Early Everolimus Initiation Fails to Counteract the Cytotoxic Response Mediated by CD8+ T and NK Cells in Heart Transplant Patients

Abstract

The positive long-term effects of conversion to everolimus (EVL) after heart transplantation (HT) have been evaluated in several studies. However, the timing of EVL initiation, the best way to combine it with other immunosuppressive treatments, and the impact of these combinations on the immune response are poorly understood aspects. Here, we analyzed the immune phenotype and function of HT patients (n = 56) at short and long terms (prospective and retrospective cohorts), taking into account the time of EVL initiation: early (3 months post-transplant, EVL-E group) or late (>1 year post-transplant, EVL-L group) compared with mycophenolate mofetil treatment (MMF group). We show that early EVL conversion from MMF allows the increase of cytotoxic (CD56^dim CD16⁺) NK and effector-memory (EM, CD45RA^- CCR7^-) CD8⁺ T cell subsets, which show a significantly higher level of expression of cytotoxic molecules, IFN-γ production and degranulation ability under activation. NK cell expansion is accompanied by an altered balance of receptor expression, increasing the activation state, and lytic activity of those cells. Those changes are detected after as little as 1 month after EVL conversion in association with the expansion of regulatory T cells and the decrease in B cell frequency. However, no changes in the immune cells subsets were observed after late EVL initiation (EVL-L) compared with the MMF group. Our results imply that only early EVL conversion induces key changes in the post-transplant immune response, preserving an efficient anti-viral response, but simultaneously showing a limited ability to counteract the cytotoxic response to the allograft.

Keywords: CD8+ T cells; DNA methylation; NK cells; cytotoxicity; everolimus; heart transplantation; mTOR inhibitors.

Figure 1

Design of the retrospective study. HT (heart transplant) patients enrolled with respect to the immunosuppressant treatment and the timing of EVL initiation. MMF group (n = 20), patients under MMF treatment; EVL-E group (n = 12), patients converted to EVL 3 months after HT; EVL-L (n = 8), patients converted to EVL >1 year after HT. MMF, mycophenolate mofetil; EVL, everolimus.

Figure 2

Immune phenotype in peripheral blood of HT patients, by treatment group. The percentage of the immune cell subsets was determined in 50 μl of peripheral blood obtained from HT patients at >1 year post-transplant who were receiving therapy with MMF (MMF group, n = 20) or who were converted to EVL early (EVL-E, n = 12) or late (EVL-L, n = 8). Each circle represents one HT patient, and the mean and standard deviation are depicted as black bars. Significant differences between groups were determined by the Mann–Whitney U-test. *p < 0.05.

Figure 3

Effect of EVL on the cell subsets of CD4⁺ and CD8⁺ T cells and functional properties of cytotoxic CD8⁺ T cells. (A) Distribution of CD4⁺ and CD8⁺ T lymphocytes into naive (N, CD45RA⁺ CCR7⁺), central memory (CM, CD45RA⁻ CCR7⁺), effector-memory (EM, CD45RA⁻ CCR7⁻) and terminally differentiated effector-memory re-expressing CD45RA (TEMRA, CD45RA⁺ CCR7⁻) subsets, in HT patients from MMF (n = 20), EVL-E (n = 12) and EVL-L (n = 8) groups. (B) Histograms represent the percentage of CD8⁺ T lymphocytes expressing cytolytic molecules (Granzyme B and Perforin), producing IFN-γ cytokine or with the ability to degranulate (show by the cell-surface expression of the CD107a marker) for each patient from the MMF (n = 20) and EVL-E (n = 12) groups. Dot-plots are representative of an independent experiment from each treatment group. *p < 0.05.

Figure 4

Effect of EVL on the subsets and functionality of NK cells. (A) Distribution of NK cell subsets into CD56^dim CD16⁺ (cytotoxic cells), CD56^bright CD16^+/− (cells producing cytokines) and CD56⁻ CD16⁺ (cells expanded in chronic viral infections) in HT patients from the MMF (n = 20), EVL-E (n = 12) and EVL-L (n = 8) groups. Each circle represents one HT patient, and the mean and standard deviation are depicted as black bars. Significant differences between groups were determined by the Mann–Whitney U-test. *p < 0.05. (B) Degranulation (CD107a⁺ cells) of NK cells under activation with IL-2 + IL-15 cytokines for each patient from the MMF (n = 20) and EVL-E (n = 12) groups. Significant differences between groups were determined by Student's t-test or the Mann–Whitney U-test. *p < 0.05. Dot-plots show one representative experiment of each group.

Figure 5

Changes in the immune phenotype of HT patients during the first year post-transplant from the prospective study. Distribution of CD8⁺ T, regulatory T, NK and B cells in peripheral blood from HT patients (n = 16) under EVL treatment from the third month post-transplant and follow-up at different times during the first year; before EVL initiation (Pre-E), and 1, 3, and 9 months after EVL conversion (Pre-E1, Pre-E2, Post-E3). Each circle represents one HT patient, and the mean and standard deviation are depicted as black bars. Differences between Pre-E and each time after EVL conversion (Post-E1, Post-E2 and Post-E3) were analyzed by the Wilcoxon test and *p < 0.05.

Figure 6

DNA methylation of immune genes in HT patients during the first year post-transplant from the prospective study. (A) The DNA methylation levels of IFNG, FASLG, and PRF1 genes were quantified by pyrosequencing in PBMCs isolated from HT patients (n = 16) under early EVL treatment and follow-up at different times during the first year; before EVL initiation (Pre-E), and 1, 3, and 9 months after EVL conversion (Pre-E1, Pre-E2, Post-E3). Each circle represents one HT patient and differences between Pre-E and each time after EVL conversion (Post-E1, Post-E2, and Post-E3) were analyzed by the Wilcoxon test. *p < 0.05. (B) Correlation between the DNA methylation level of each gene and the percentage of CD8⁺ T cell subsets and NK cells at all assayed times during the first year post-transplant. Negative and positive Pearson correlation coefficients indicate that demethylation of the gene is significantly associated with a higher or lower percentage of the cell subset, respectively. N, naive; EM, effector memory. (C) Association between the loss of IFNG DNA methylation and the increase in the percentage of EM CD8⁺ T and NK cells for each patient during the first year post-transplant (Post-E3–Pre-E times).