Trained immunity suppression determines kidney allograft survival

Abstract

The innate immune system plays an essential role in regulating the immune responses to kidney transplantation, but the mechanisms through which innate immune cells influence long-term graft survival are unclear. The current study highlights the vital role of trained immunity in kidney allograft survival. Trained immunity describes the epigenetic and metabolic changes that innate immune cells undergo following an initial stimulus, allowing them have a stronger inflammatory response to subsequent stimuli. We stimulated healthy peripheral blood mononuclear cells with pretransplant and posttransplant serum of kidney transplant patients and immunosuppressive drugs in an in vitro trained immunity assay and measured tumor necrosis factor and interleukin 6 cytokine levels in the supernatant as a readout for trained immunity. We show that the serum of kidney transplant recipients collected 1 week after transplantation can suppress trained immunity. Importantly, we found that kidney transplant recipients whose serum most strongly suppressed trained immunity rarely experienced graft loss. This suppressive effect of posttransplant serum is likely mediated by previously unreported effects of immunosuppressive drugs. Our findings provide mechanistic insights into the role of innate immunity in kidney allograft survival, uncovering trained immunity as a potential therapeutic target for improving graft survival.

Keywords: graft survival; innate immunity; kidney transplantation; trained immunity.

Figure 1.

The effect of human kidney transplant recipient’s serum on trained immunity. (A) Schematic representation of serum collection and the subsequent serum-trained immunity assay. (B, C) PBMCs were incubated for 24 hours with pre- and posttransplant serum of 96 kidney transplant recipients. After a 5-day resting period, cells were restimulated with LPS for 24 hours, and IL-6 and TNF cytokine production was measured in the supernatant by ELISA (n = 3). RPMI culture medium without patient serum was used as a negative control. (D) Correlation between IL-6 and TNF levels measured in the supernatant after serum-trained immunity assay with pre- and posttransplant serum of 96 kidney transplant recipients. (E) Correlation between IL-6 and TNF levels measured in the supernatant after posttransplant serum-trained immunity assay of 96 kidney transplant recipients and their whole blood tacrolimus levels posttransplant. (F) Comparison between IL-6 and TNF response in cells trained with posttransplant serum from patients with DGF compared to patients with no DGF. (G) Comparison of tertiles of IL-6 and TNF responses from cells trained with posttransplant serum and creatinine levels at 3 and 6 months after transplantation. (H) Volcano plot of 76 inflammation-related proteins (Olink inflammation panel) measured in posttransplant serum of the highest tertile compared to the lowest tertile of the IL-6 and TNF values in the trained immunity assay. Data are expressed as mean ± SD (B, C, G) or as mean ± SEM (F). Fold changes were calculated as value/mean RPMI (D, E). P values were calculated using unpaired t test (B, C, F), 1-way analysis of variance (G), or Mann-Whitney U test (H). ****P < .0001. DGF, delayed graft function; ELISA, enzyme-linked immunosorbent assay; IL-6; inter-leukin 6; LPS, lipopolysaccharide; PBMC, peripheral blood mononuclear cell; SD, standard deviation; SEM, standard error of the mean; TNF, tumor necrosis factor; CASP-8, caspase-8.

Figure 2.

Trained immunity is associated with long-term graft survival. (A) Schematic representation of the method of collecting the data for Kaplan-Meier survival analyses. Results of serum-induced trained immunity were divided into tertiles (low = green, middle = blue, and high = red) or by cutoffs determined based on the ROC curves. (B, C) Kaplan-Meier survival analyses of death-censored graft survival of the tertiles of serum-induced trained immunity for the IL-6 and TNF responses to LPS restimulation. Data are expressed as survival probability (%) and death-censored graft survival (y). P values were calculated with a log-rank test. (D) ROC curves for the IL-6 and TNF responses to LPS restimulation. (E, F) Kaplan-Meier survival analyses of death-censored graft survival of the low vs the high posttransplant serum-induced trained immunity for the IL-6 and TNF responses to LPS restimulation, based on the cutoffs determined by the ROC curves. Data are expressed as survival probability (%) and death-censored graft survival (years). P values were calculated with a log-rank test. CI, confidence interval; DCGS, death-censored graft survival; IL-6; interleukin 6; LPS, lipopolysaccharide; ROC, receiver operating characteristic; TNF, tumor necrosis factor.

Figure 3.

The in vitro and ex vivo effects of sterile inflammation on trained immunity. (A) Schematic representation indicating trained immunity-related DAMPs and their respective receptors. (B) PBMCs were stimulated for 24 hours with DAMPs in 3 concentrations. After a 5-day resting period, cells were restimulated with LPS for 24 hours, and IL-6 and TNF were quantified in the supernatant by ELISA (n = 6). (C) Schematic representation of the mouse ischemia model, subsequent flow cytometry, and ex vivo stimulation assay. (D) Unilateral renal ischemia was induced in 8- to 12-week-old C57BL/6J mice for 30, 35, or 40 minutes. Sham mice underwent the same procedure without clamping the renal artery. Mice were sacrificed on day 3, and BMDMs and splenocytes were collected and stimulated for 24 hours with LPS or culture medium (DMEM:HAMF12) as a control before collecting the supernatant. IL-6 and TNF were quantified by ELISA (n = 3). Data are expressed as log2 fold change compared to untrained (RPMI) PBMCs (A) or as mean ± SEM (B, D). P values were calculated using an unpaired t test (A) or a paired t test (B, D). BMDM, bone marrow-derived macrophage; DAMP, danger-associated molecular pattern; ELISA, enzyme-linked immunosorbent assay; HMGB1, high mobility group box 1 protein; IFN, interferon; IL; interleukin; LPS, lipopolysaccharide; PBMC, peripheral blood mononuclear cell; SEM, standard error of the mean; TNF, tumor necrosis factor; CRP, C-reactive protein; SAP130, Sin3A associated protein; ATP, adenosine triphosphate; RAGE, Receptor for Advanced Glycation Endproducts; TLR, Toll-like receptor; HA, hyaluronic acid; UA, uric acid; HS, heparan sulphate; ns, not significant.

Figure 4.

The in vitro effect of immunosuppressive drugs on the induction of trained immunity. (A, B) PBMCs were stimulated for 24 hours with HKCA (A) or IL-1β (B) alone or together with immunosuppressive drugs in 3 different concentrations for 24 hours. After 24 hours, both the stimulus and the immunosuppressive drugs were washed away. After a 5-day resting period, cells were restimulated with LPS for 24 hours, and IL-6 and TNF cytokine production was measured in the supernatant by ELISA (n = 6). Data are expressed as log2 fold change compared to PBMCs trained with either HKCA or IL-1β. P values were calculated using an unpaired t test. ELISA, enzyme-linked immunosorbent assay; HKCA, heat-killed Candida albicans; IL; interleukin; LPS, lipopolysaccharide; PBMC, peripheral blood mononuclear cell; TNF, tumor necrosis factor; ns, not significant.

Figure 5.

Transcriptional and epigenetic profiles of kidney transplant recipient’s circulating leukocytes. (A) Schematic representation of the method of selection of PBMCs of kidney transplant patients for RNA- and ChIP-sequencing. (B) Genomic annotations of H3K4me3 and H3K27ac peaks in the mean of 6 patients from the lowest tertile and 6 patients from the highest tertile compared to the differential peaks (FDR < 0.05, Diff Peaks panel). (C) Top 10 Gene Ontology (GO) Biological Processes associated with genomic regions showing altered H3K4me3 and H3K27ac in PBMCs of 6 patients from the lowest tertile and 6 patients from the highest tertile of posttransplant serum-induced trained immunity, determined using the Genomic Regions Enrichment of Annotations Tool (FC > 2, FDR < 0.05). (D) H3K4me3 and H3K27ac signal at IL15 gene as visualized in the UCSC genome browser. High represents patients from the highest tertile (red), Low represents patients from the lowest tertile (green). In gray, the expression of IL15 is displayed in monocytes, CD4+ T cells, CD8+ T cells, and naïve B cells. (E) Significantly altered gene sets of the Hallmark database in PBMCs of 6 patients from the lowest tertile and 6 patients from the highest tertile of posttransplant serum-induced trained immunity. ChIP, chromatin immunoprecipitation; FC, fold change; FDR, false discovery rate; H3K27ac, histone 3 lysine 27 acetylation; H3K4me3, histone 3 lysine 4 trimethylation; IL15, interleukin 15; MYC, myelocytomatosis oncogene; NES, normalized enrichment score; PBMC, peripheral blood mononuclear cell; TNF, tumor necrosis factor; TSS, transcription start site; ns, not significant; UCSC, University of California, Santa Cruz.