APOL1 variants increase risk for FSGS and HIVAN but not IgA nephropathy

Abstract

A chromosome 22q13 locus strongly associates with increased risk for idiopathic focal segmental glomerulosclerosis (FSGS), HIV-1-associated nephropathy (HIVAN), and hypertensive ESRD among individuals of African descent. Although initial studies implicated MYH9, more recent analyses localized the strongest association within the neighboring APOL1 gene. In this replication study, we examined the six top-most associated variants in APOL1 and MYH9 in an independent cohort of African Americans with various nephropathies (44 with FSGS, 21 with HIVAN, 32 with IgA nephropathy, and 74 healthy controls). All six variants associated with FSGS and HIVAN (additive ORs, 1.8 to 3.0; P values 3 × 10(-2) to 5 × 10(-5)) but not with IgA nephropathy. In conditional and haplotype analyses, two APOL1 haplotypes accounted for virtually all of the association with FSGS and HIVAN on chromosome 22q13 (haplotype P value = 5.6 × 10(-8)). To assess the role of MYH9 deficiency in nephropathy, we crossbred Myh9-haploinsufficient mice (Myh9(+/-)) with HIV-1 transgenic mice. Myh9(+/-) mice were healthy and did not demonstrate overt proteinuria or nephropathy, irrespective of the presence of the HIV-1 transgene. These data further support the strong association of genetic variants in APOL1 with susceptibility to FSGS and HIVAN among African Americans.

Figure 1.

Haplotype analysis localized the most significant signal within the APOL1 locus. (A) Two-SNP and three-SNP haplotype sliding-window analysis localized the strongest signal within the APOL1 locus. The y axis shows the −log (P value) for the association statistic. The x axis shows the tested SNP and their location within the APOL1 and MYH9 genes. (B) Haplotype analysis of the APOL1 SNPs demonstrated that G1 and G2 reside on distinct haplotypes and each confers an independent risk of disease compared with the reference haplotype. G1 is defined by the rs73885319-G and rs60910145-G alleles, and G2 is defined by the rs71785313 deletion allele (D). The odds ratios (ORs) were calculated in reference to the most common haplotype, which does not contain any risk variants (ATI). *P value <0.05 versus the reference haplotype. The P value for the global haplotype test is also indicated.

Figure 2.

Myh9 haploinsufficiency does not induce nephropathy in HIV-1 transgenic mice. (A-C) Renal histology (Periodic acid-Schiff staining) was normal in B6 (A), B6-Myh9^+/− (B), and TgB6-Myh9^+/− (C) mice. (D) Proteinuria (quantified as μg of albumin/mg of creatinine ratio in urine) was increased in the Myh9 haploinsufficient, HIV transgenic mice (TgB6-Myh9^+/−) compared with the B6 and B6-Myh9^+/− parental strains but was still within the range of normal for inbred mice (e.g. DBA/2J or FVB/NJ). n = four to five mice per group.