Functional Assessment of High-Risk APOL1 Genetic Variants

Despite intensive investigation, little is known about the fundamental molecular mechanism underlying APOL1-associated kidney disease. Alterations to circulating protein levels are likely to reside along the causal pathway between the presence of the APOL1 G1 and G2 gain-of-function genetic variants and phenotypic variation. Mass spectrometry–based proteomics can be a powerful tool to advance the knowledge of disease pathology in an unbiased, large-scale manner. In addition, proteomics can be used to discover new biomarkers and inform prevention, diagnostics, and therapeutics.

In this issue of CJASN, Chen et al. evaluated the associations between presence of APOL1 high-risk variants and 6790 plasma proteins derived from whole blood, measured by the SOMAscan aptamer-based assay among participants from the African American Study of Kidney Disease and Hypertension and brought forward significant findings for replication in participants with African ancestry from the Atherosclerosis Risk in Communities study (1). The investigators discovered differences in plasma circulating levels of seven proteins among African American Study of Kidney Disease and Hypertension participants with APOL1 2 high-risk variants, compared with those with 0 or 1 high-risk variants. Of these, five were available in Atherosclerosis Risk in Communities and three associations remained significant upon replication: lower apolipoprotein L1 (aptamer 1), lower 72 kDa type IV collagenase, and lower metalloproteinase inhibitor 2 per additional APOL1 risk allele.

The association between APOL1 risk alleles and APOL1 protein levels comes in contrast to previous smaller studies (2,3), but is consistent with a recent study by Katz and colleagues that found an association of variation at rs73885319 (part of G1 haplotype) with circulating APOL1 among 1852 Black adults from the Jackson Heart Study (4). Beyond differences in APOL1, Chen et al. identify new candidate proteins, type IV collagenase and metalloproteinase inhibitor 2, which are associated with APOL1 high-risk variants. These may represent novel biomarkers of APOL1 nephropathy, and these associations should be the starting point of further mechanistic explorations. All 6790 APOL1-high risk genotype–plasma protein associations are available in a supplemental data file to facilitate future work in this area (1).

Subsequent analyses by Chen et al. related the concentrations of the top candidate proteins with prospective risk of kidney disease progression. None of the tested proteins were associated with the adverse outcome after adjustment for baseline eGFR. These findings are consistent with those of several groups that have assessed the implications of circulating APOL1 and showed levels do not correlate with kidney disease risk (3,5).

The utilization of the highly sensitive SOMAscan platform within these well-characterized cohorts of individuals is a notable strength of the study. The cohorts examined offer dense phenotype data from individuals with CKD and from the general population, at high risk of kidney disease progression.

The study by Chen et al. exposes a fundamental limitation of proteomic approaches confined to proteins derived from whole blood. Although circulating plasma protein data are more readily accessible in larger samples, they may not necessarily capture the biologic effects from tissue types that are more relevant to the disease being studied. This issue may underlie the lack of association between the proteins and kidney progression outcomes. For one, protein levels are known to vary depending on the tissue type they are investigated in. We know that APOL1 is expressed in vascular cells, glomerular endothelial cells, podocytes, and kidney tubular cells (6). Studies in mouse models have shown that expression of APOL1 high-risk variants in podocytes is associated with important functional and histomorphological changes that resemble human kidney disease (7). Transcriptomic analyses in kidney biopsy samples among patients with nephrotic syndrome reported associations of high-risk genotype with increased fractional interstitial area, interstitial fibrosis, and tubular atrophy (8). In addition, kidney transplant studies, which provide a unique setting to distinguish between the effects of systemic versus kidney APOL1 expression, suggest kidney APOL1 as the principal driver of APOL1 nephropathy. As a result, it would be expected that data derived from kidney tissue would be the most pertinent for characterizing the function of APOL1 high-risk variants.

Large-scale analyses of kidney-based protein quantitative trait loci have not yet been performed because the necessary data have yet to emerge. The Kidney Precision Medicine Project will provide an important step forward in that direction by developing tools to apply mass spectrometry–based proteomics to proximal tubular and glomerular cells and by recruiting patients of African ancestry over the next few years (9,10).

In summary, the Chen et al. study provides evidence of associations of APOL1 high-risk variants with plasma concentrations of APOL1 protein. However, circulating levels of APOL1 were not associated with kidney disease progression outcomes among well-phenotyped individuals from longitudinal cohorts. This study highlights the need for comprehensive and consistent cell type–specific proteomics data in nephrology. Thankfully, novel techniques and resources have been evolving rapidly in attempts to fill this gap.

Disclosures

C. Robinson-Cohen reports serving as a CJASN Editorial Board Member and as Genetics Section Editor for Clinical Nephrology.

Funding

None.

Author Contributions

C. Robinson-Cohen conceptualized the study, was responsible for investigation, provided supervision, and wrote the original draft.