High RIPK3 expression is associated with a higher risk of early kidney transplant failure

Abstract

Renal ischemia-reperfusion injury (IRI) is associated with reduced allograft survival, and each additional hour of cold ischemia time increases the risk of graft failure and mortality following renal transplantation. Receptor-interacting protein kinase 3 (RIPK3) is a key effector of necroptosis, a regulated form of cell death. Here, we evaluate the first-in-human RIPK3 expression dataset following IRI in kidney transplantation. The primary analysis included 374 baseline biopsy samples obtained from renal allografts 10 minutes after onset of reperfusion. RIPK3 was primarily detected in proximal tubular cells and distal tubular cells, both of which are affected by IRI. Time-to-event analysis revealed that high RIPK3 expression is associated with a significantly higher risk of one-year transplant failure and prognostic for one-year (death-censored) transplant failure independent of donor and recipient associated risk factors in multivariable analyses. The RIPK3 score also correlated with deceased donation, cold ischemia time and the extent of tubular injury.

Keywords: Molecular biology; Nephrology.

Graphical abstract

Figure 1

Gene expression analysis of baseline biopsies of deceased versus living donors (A–D) Schematic workflow showing how baseline biopsies were sampled 10 min after reperfusion of the donor graft (B) Principal component analysis (PCA) between the living and deceased donors’ gene expression profiles (C) The dendrogram denotes network modular organization of differentially expressed genes comparing the deceased with living donor dataset (D) Module expression profiles of long ischemia based on the module eigengene. (E–H) Heatmap represents the KEGG pathway enrichment analysis of each co-expression module as calculated by WGCNA (F) GSEA of the TNF signaling pathway in long ischemia NES, normalized enrichment score (G) Heatmap representing the members of the TNF signaling pathway genes significantly expressed in deceased donors (H) Schematic overview of TNF signaling with members being denoted in magenta when significantly deregulated in deceased donors compared to living donations.

Figure 2

RIPK3 expression levels in baseline biopsies (A–E) Workflow depicting how baseline biopsies were evaluated for RIPK3 scoring and statistical analysis. From a total of 406 available biopsies, 374 were stained and evaluated within this study. 21 biopsies could not be assessed, and 11 biopsies came from transplants that succumbed to surgical complications, leading to their exclusion (B) Representative images of cortical specimens from baseline biopsies. The exact scores of the illustrated specimens with low and high RIPK3 expression are from left to right as follows: 0; 1.0; 2.34 and 3.0. Scale bars as depicted (C) Representative images of negative controls, specifically, (I) tumor-distant non-inflamed and non-fibrotic renal parenchyma from kidneys after tumor nephrectomy; (II) kidneys from end stage allograft failure with severe interstitial fibrosis and tubular atrophy; (III and IV) kidneys with membranous glomerulonephritis and nephrotic proteinuria. Scale bars as depicted (D) Scatterplot (with median reported in red) depicting the distribution of RIPK3 score across the investigated cohort (E) RIPK3 score is significantly higher in biopsies from deceased donors. Data are presented as scatterplot and in the graph the median is reported. p value from Mann-Whitney test is reported in figure.

Figure 3

RIPK3 expression predicts kidney transplant failure (A) Kaplan-Meier estimates of death-censored transplant failure. Shown are estimates of the probabilities of the primary endpoint (i.e., the permanent need for dialysis after transplantation, which consists of both primary non-function (without surgical complications) and follow up end-stage transplant failure requiring the reinstitution of dialysis) comparing renal allograft baseline biopsies with a RIPK3 score of 0–2.0 (≤2) and greater than 2.0 (>2). Estimates are shown for the first year (left) and for the follow up period from year 2–5 (right). Data were censored for death-censored graft survival at the time of death with a functioning graft, at last day of detected kidney function, and either at 12 months (for one-year transplant failure) or at 60 months (for the follow up period 2–5 years). p-Values were calculated using the log rank test. (B) Kaplan-Meier estimates of non-death-censored transplant failure. Shown are estimates of the probabilities of the secondary endpoint, which was a composite of primary non-function (without surgical complications), follow-up end-stage transplant failure requiring the reinstitution of dialysis, or recipient death with a functioning allograft for renal allograft baseline biopsies, with a RIPK3 score 0 to 2.0 (≤2) and greater than 2.0 (>2). Estimates are shown for first year (left) and for the follow-up period from year 2–5 (right). Data were censored for non-death-censored graft survival at last day of detected kidney function and either at 12 months (for one-year transplant failure) or at 60 months (for the follow-up period 2–5 years. p-values were calculated using the log rank test.

Figure 4

RIPK3 expression and its association with acute tubular injury (A) Representative images of PAS reaction of cortical specimen with corresponding RIPK3 staining. Scale bar as depicted. (B and C) Frequency distribution of acute tubular injury (ATI) in the whole cohort. p-value from chi-square test is reported in figure (C) Frequency distribution of ATI in living and deceased donation cohorts, stratified above and below the RIPK3 score median. p value from chi-square test is reported in figure.