Urine APOL1 Isoforms Reflect Plasma-Derived Liver-Synthesized Proteins

Visual Abstract

APOL1 kidney risk variants (KRVs) cause approximately 30% of ESKD in Blacks.^1,2 Studies support kidney-synthesized APOL1 KRV isoforms, not liver-synthesized circulating isoforms, as causing nephropathy.^3,4 Small molecule inhibitors and APOL1 antisense oligonucleotides are novel treatments for APOL1-associated nephropathy on the horizon.⁵ These therapies provide hope for slowing disease progression in a subset of the 5.4 million U.S. Blacks with two APOL1 KRVs at risk for nephropathy (13% of 42 million Blacks with two KRVs) and millions more in Africa and with recent African ancestry across the globe. Given evolving therapies, it may prove useful to measure relative urine and plasma APOL1 protein isoform concentrations and determine whether urine contains kidney-derived isoforms. This question was addressed in transplant recipients with differing APOL1 genotypes in liver and functioning kidney.

Plasma and urine were collected from 26 adult recipients of a deceased donor kidney >6 months post-transplant. They lacked active infection, rejection, or allograft failure and had stable kidney function. Donor-recipient pairs had different APOL1 genotypes, one with two KRVs and the other zero KRVs. The Wake Forest School of Medicine Institutional Review Board approved the study, and participants provided written informed consent.

Genotyping and sample preparation methods are described in Supplemental Material and Supplemental Tables 1 and 2. Urines were subject to ultrafiltration to concentrate samples before immunoprecipitation using APOL1 mAb-coated magnetic beads. APOL1 protein levels were tracked throughout using a sandwich ELISA method. Samples were subjected to tryptic digestion prior to analysis by liquid chromatography coupled to tandem mass spectrometry (LC-MS/MS) using a series of stable isotope-labeled internal standard protein epitope signature tags unique for G0, G1, and G2.

Table 1 displays recipient characteristics; 24 carried high-risk APOL1 genotypes (11 G1:G1, seven G1:G2, and six G2:G2) and received APOL1 low-risk genotype (G0:G0) kidneys. Two carried APOL1 low-risk (G0:G0) genotypes and received high-risk genotype kidneys (one G1:G1 and one G2:G2). Mean time post-transplant was 1.9 years.

Table 1.

Demographic and clinical characteristics of kidney transplant recipients

Characteristic	Recipient
Age, yr, mean (SD)	57.1 (12.5)
Men, %	69
Years since transplant, mean (SD)	1.9 (1.2)
Urine protein-creatinine ratio, mg/g, mean (SD)	928.9 (2730)
Systolic BP, mm Hg, mean (SD)	134.1 (18.1)
Serum creatinine, mg/dl, mean (SD)	1.56 (0.48)
Recipient genotype
High-risk genotype donor
G0:G0	2
Low-risk genotype donor
G1:G1	11
G2:G2	6
G1:G2	7

Urine APOL1 concentration determined using ELISA ranged from 0.4 to 66 ng/ml (median [interquartile range, 25–75]: 1.04 [0.38, 3.23] ng/ml). Supplemental Figure 1 reveals that urine APOL1 concentrations did not differ significantly on the basis of recipient genotype. Urine APOL1 protein was captured on beads (78% efficiency) and concentrated to determine urine APOL1 isoforms using LC-MS/MS. Upon elution of APOL1 from beads, 69.2% (18 of 26) of samples yielded protein concentrations resulting in detectable APOL1 peptides by mass spectrometry.

Urine isoform determination was possible in 18 recipients, 16 with APOL1 high-risk genotypes and two with low-risk genotypes (Supplemental Table 3). APOL1 G0 protein was not detected in urine samples from high-risk recipients of low-risk kidneys; G1 or G2 variants were. APOL1 G0 was detected in the urine of both low-risk recipients of high-risk kidneys; one had approximately 14 g proteinuria per day and very high urine APOL1 levels with only 5% APOL1 of donor kidney origin. Significant variability was observed in urine APOL1 concentrations; however, urine APOL1 and albuminuria were significantly correlated (r²=0.50, P<0.001).

Results in transplantation and transgenic mice support kidney expression of APOL1 risk variants as causing nephropathy.^3,4 Given the potential for novel modifiers of APOL1-associated forms of kidney disease, it was important to quantify urine APOL1 protein and determine its source. If urine contained kidney-synthesized APOL1 protein, it might provide a window into podocyte and kidney cell APOL1 synthesis and serve as a potential marker of kidney-specific APOL1 knockdown with antisense oligonucleotide therapy. We used a unique human model where participants’ liver and kidney expressed different APOL1 genotypes. Urine APOL1 protein was typically derived from liver-synthesized circulating protein.⁶ Urine APOL1 protein and urine albumin concentrations were highly correlated, supporting leakage of both proteins from the circulation across a damaged glomerular filtration barrier.

This is the first direct assessment of urine APOL1 levels and source in humans with differential APOL1 expression in kidney and liver. Although participants had functioning kidney transplants, results appear relevant to the healthy state. Recipients received stable doses of calcineurin inhibitors and mycophenolate mofetil to prevent rejection. Dramatic effects of these medicines on kidney APOL1 protein expression are unlikely; they do not markedly alter the course of APOL1 FSGS.⁷ It is impossible to maintain different APOL1 genotypes in liver and kidney without immunosuppression. Participant mean serum creatinine was 1.56 mg/dl, revealing abundant nephron mass for kidney synthesis of APOL1. APOL1 is produced by podocytes, which can be shed in urine. This analysis measured extracellular (cellfree) urine APOL1. APOL1 concentrations in urine were variable, and samples required processing through a concentration protocol to permit detection by LC-MS/MS. Because of detection limitations, if the contribution of kidney-derived APOL1 to urine APOL1 was small, it might not be detectable.

We conclude that the source of urine APOL1 protein is primarily the circulation and reflects liver synthesis. APOL1 protein isoforms synthesized by the kidney are generally not present in urine after kidney transplantation. Monitoring kidney APOL1 protein expression will require other techniques for measurement after therapeutic gene knockdown.

Disclosures

M. Althage, A. Bogstedt, P.J. Greasley, J. Hartleib-Geschwindner, and I. MacPhee are employed by and owns shares in AstraZeneca. N. Palmer reports patents and inventions with Wake Forest Innovations. T. Miliotis reports current employment with AstraZeneca R&D. I. MacPhee reports Consultancy Agreements with Chiesi; and Honoraria from Astellas. Wake Forest University Health Sciences and B.I. Freedman have rights to an issued United States patent related to APOL1 genetic testing. B.I. Freedman is a consultant for and receives research support from AstraZeneca and RenalytixAI Pharmaceuticals; reports scientific advisor or membership via Editorial Boards of JASN and Kidney International, American Journal of Nephrology; and other interests/relationships as Chief Medical Officer of Health Systems Management, Inc. M. Gautreaux reports honoraria from American Foundation for Donation and Transplantation. All remaining authors have nothing to disclose.

Funding

This work was supported by AstraZeneca.

Supplemental Material

This article contains the following supplemental material online at http://jasn.asnjournals.org/lookup/suppl/doi:10.1681/ASN.2021030411/-/DCSupplemental.

Supplemental Material. Detailed methods.

Supplemental Table 1. Amino acid sequences of the stable isotope-labelled recombinant protein fragments (SIS PrEST standards) where the bold sequences represent proteotypic peptides for the APOL1 isoform variants.

Supplemental Table 2. SRM transitions monitored for determination of APOL1 isoform variants by mass spectrometry.

Supplemental Table 3. APOL1 isoform detection in kidney transplant recipient urine samples.

Supplementary Figure 1. Urine APOL1 concentration by recipient APOL1 genotype.