A tripartite complex of suPAR, APOL1 risk variants and αvβ3 integrin on podocytes mediates chronic kidney disease

Abstract

Soluble urokinase plasminogen activator receptor (suPAR) independently predicts chronic kidney disease (CKD) incidence and progression. Apolipoprotein L1 (APOL1) gene variants G1 and G2, but not the reference allele (G0), are associated with an increased risk of CKD in individuals of recent African ancestry. Here we show in two large, unrelated cohorts that decline in kidney function associated with APOL1 risk variants was dependent on plasma suPAR levels: APOL1-related risk was attenuated in patients with lower suPAR, and strengthened in those with higher suPAR levels. Mechanistically, surface plasmon resonance studies identified high-affinity interactions between suPAR, APOL1 and α_vβ₃ integrin, whereby APOL1 protein variants G1 and G2 exhibited higher affinity for suPAR-activated avb3 integrin than APOL1 G0. APOL1 G1 or G2 augments α_vβ₃ integrin activation and causes proteinuria in mice in a suPAR-dependent manner. The synergy of circulating factor suPAR and APOL1 G1 or G2 on α_vβ₃ integrin activation is a mechanism for CKD.

Figure 1

Synergy of an APOL1 high-risk genotype and suPAR levels determines yearly loss of kidney function. (a,b) Estimated yearly change in estimated glomerular filtration rate (eGFR) according to APOL1 status (green, low risk; red, high risk) and continuous levels of suPAR at cohort enrollment for the Emory Cardiovascular Biobank (EmCAB; n = 487) (a) and the African American Study of Kidney Disease and Hypertension cohort (AASK; n = 607) (b). Yearly change in eGFR was estimated from linear mixed-effects models of eGFR adjusted for age, sex, body-mass index, hypertension, diabetes mellitus (in EmCAB), smoking status, randomization group (in AASK) and suPAR level in both baseline levels and slope over time. In EmCAB, suPAR was log-transformed. In AASK, suPAR was log-transformed and modeled using restricted cubic splines to allow for deviations from log linearity.

Figure 2

APOL1 forms high-affinity interactions with suPAR and α_vβ₃ integrin. All sensorgrams were generated using SPR assays. Rate constants (k_a and k_d) were determined by kinetic fitting of the sensorgrams using a one-to-one binding equation, and equilibrium dissociation constant (K_D) values were determined by calculating k_d/k_a. (a) Sensorgrams of suPAR (0–300 nM) binding to immobilized APOL1 protein variants (G0, G1 and G2), and calculated K_D values. (b) Sensorgrams of α_vβ₃ integrin (0–10 nM) binding to immobilized suPAR in the absence (inactive form) and presence (active form) of Mn₂₊ (0.5 mM), and calculated K_D values. (c,d) Steady-state affinity fitting curves of α_vβ₃ (c) or α₃β₁ (d) integrin (0–40 nM) binding to immobilized APOL1 (G0, G1 and G2) in the absence (inactive) and presence (active) of Mn₂₊ (0.5 mM). (e) Sensorgrams of α_vβ₃ integrin (0–20 nM) binding to immobilized APOL1 (G0, G1 and G2) in the presence of Mn₂₊ (0.5 mM), and calculated K_D values. (f) Bar graphs showing the determined K_D values for suPAR in the presence of α_vβ₃ (5 nM), increasing concentrations of suPAR (0–300 nM) and immobilized APOL1 proteins (G0, G1 and G2) in the absence (inactive) or presence (active) of Mn₂₊ (0.5 mM). The average K_D values were determined from two independent experiments, and error bars for K_D measurements indicate s.e.m. (n = 2) (a–f). A, analyte; IM, immobilization.

Figure 3

High levels of suPAR are needed to synergize with APOL1 G1 and G2 to induce α_vβ₃ integrin activation on human podocytes. (a,b) Representative western blot images for immunoprecipitation assay using HEK293T cells transfected with plasmids expressing APOL1 G0, Flag-tagged suPAR and Myc-tagged β₃ integrin in different combinations, as indicated. Data are representative of at least three independent experiments. The empty pFlag-CMV3 vector was used as a negative control. (a) Cell lysates were immunoprecipitated using a mouse monoclonal anti-Flag (M2) or monoclonal anti-Myc antibody. (b) Cell lysates were immunoprecipitated using a rabbit polyclonal anti-APOL1. The cell lysates and immunoprecipitates were analyzed by immunoblotting with antibodies, as indicated. Analysis of the input (1%) indicates that the levels of APOL1, β₃ integrin and suPAR are overexpressed in the transfected cells. IP, immunoprecipitation. (c) Bar graph representing level of activated β₃ integrin detected in human podocytes cultured in healthy serum without (Con) or with added proteins: suPAR (2,000, 5,000 and 10,000 pg/ml), APOL1 G0, G1 and G2 (15 µg/ml) or a combination of both proteins. Data represent measurements of >50 cells and are plotted as means ± s.d. (n = 3). ***P < 0.001. Statistical significance was determined by unpaired two-tailed Student’s t-test. (d) Immunofluorescence analysis of human podocytes grown in human serum in the absence (Con) or presence of suPAR (5,000 pg/ml) and three APOL1 variant proteins (15 µg/ml). Cells were stained with AP5 antibody that specifically recognizes the active form of β₃ integrin, and with anti-paxillin antibody to mark focal adhesions. Scale bar, 25 µm. (e) Bar graph representing the level of β₃-integrin activation detected in human podocytes cultured in healthy serum without (Con) or with three serum samples from individuals with FSGS carrying the nonrisk APOL1 genotype, and three different serum samples from individuals with FSGS carrying two APOL1 risk-allele genotypes. FSGS-APOL1 serum #2, and to a substantial degree, serum #3, detached the cells such that the quantification of β₃-integrin activation was either not determined (n.d.) or not accurate (striped bar graph). Data represent measurements of >50 cells and are plotted as means ± s.d. (n = 3). Statistical significance was determined by unpaired two-tailed Student’s t-test. (f,g) Bar graph representing the level of β₃-integrin activation detected in human podocytes cultured in healthy serum without (Con) or with the addition of the indicated concentrations of the proteins. In g, the concentration of suPAR was 5,000 pg/ml. Data represent measurements of >50 cells and are plotted as means ± s.d. (n = 3). *P < 0.01; ***P < 0.001. Statistical significance was determined by unpaired two-tailed Student’s t-test.

Figure 4

In vivo gene delivery of APOL1 G1 and G2, but not of G0, gene constructs caused proteinuria and podocyte foot-process effacement in wild-type mice, but not in Plaur^−/− mice. To express APOL1 in C57BL/6 mice or Plaur^−/− mice, in vivo gene delivery was performed and the kidney phenotype evaluated 16 h after APOL1 gene (G0, G1 and G2) injection. (a) APOL1 protein could be detected in both the glomerulus and liver of injected animals. Double immunostaining of APOL1 (green) and synaptopodin (synpo, red), a podocyte marker. VC, pcDNA 3.1 vector control. Scale bars, 10 µm. (b) Transmission electron microscopy (TEM) analysis of the kidney glomeruli in APOL1 gene (G0, G1 and G2) engineered mice. TEM images display podocyte foot processes. Scale bars, 1 µm. (c,d) Examination of proteinuria (c) and serum APOL1 levels (d) in animals into which APOL1 genes were injected (n = 7 for VC; n = 5 for APOL1 G0; n = 3 for APOL1 G1; n = 7 for APOL1 G2; n = 3 for all Plaur^−/− mice). Analysis of urinary protein is shown as albumin (mg)/creatinine (g) ratio before (T0) and 16 h after (T16) gene delivery. Data are shown as means ± s.e.m. One-way ANOVA, followed by Tukey′s multiple-comparison test in which values at T16 were compared to those at T0; *P < 0.05, **P < 0.01. Welch’s ANOVA was used when unequal variances were noted between groups (P = 0.014). (e) APOL1 G2 protein was detected in the glomerulus and liver of Plaur^−/− mice 16 h after the gene was injected into a tail vein. Double immunostaining of Plaur^−/− (green) and synaptopodin (synpo, red). VC, pcDNA 3.1 vector control. Scale bars, 10 µm.