Genetic Modifiers Influencing the Development of Albuminuria in Children with Sickle Cell Anemia

Background:  Nephropathy in sickle cell anemia (SCA) begins in childhood and portends chronic kidney disease, renal failure, and early mortality among affected adults. Individuals of African descent have disproportionately higher rates of developing non-diabetic renal disease. Several candidate genetic variants have been identified, including some specific to African Americans, which are associated with the development of albuminuria and renal disease. The influence of genetic polymorphisms on albuminuria and elevated glomerular filtration rate (GFR) in children with SCA, both early signs of sickle nephropathy, has not been investigated.  

Objectives:  To determine the influence of selected single nucleotide polymorphisms (SNPs) on the development of albuminuria and elevated GFR in children with SCA; to identify novel genetic variants influencing albuminuria and GFR by whole exome sequencing (WES).

Design/Methods:  Genomic DNA was collected on children with SCA enrolled in two prospective studies with pre-hydroxyurea renal assessments (n=185):  (1) Hydroxyurea Study of Long-Term Effects (HUSTLE, NCT00305175, n=79) with no prior disease-modifying therapy; and (2) Transcranial Doppler (TCD) With Transfusions Changing to Hydroxyurea (TWiTCH, NCT 01425307, n=106) on chronic transfusions for abnormal TCD velocities. Albuminuria was defined as ≥30mg albumin/gm creatinine on the pre-hydroxyurea urine specimen. GFR was measured in HUSTLE using plasma DTPA (technetium 99m-labeled diethylenetriaminepentaacetic acid) clearance, and estimated GFR (eGFR) in TWiTCH based on serum creatinine. DNA samples were genotyped for 8 candidate SNPs previously associated with renal disease, using PCR-based allelic discrimination, bidirectional Sanger sequencing, and analysis of variable number tandem repeats (VNTR). Associations between albuminuria and genetic polymorphisms were tested using an additive model and correlation trend test. Linked WES data from the same patients were analyzed to identify other variants associated with albuminuria and GFR.

Results: Albuminuria was present in 13.1% of patients, including 16.3% in HUSTLE and 11.0% in TWiTCH. APOL1 genetic variants were common (G1 allele frequency = 21.9%, G2 allele = 16.0%, Table) and similar to published cohorts. Children with two APOL1 G1 alleles had an increased risk of albuminuria that approached statistical significance (p=0.053). Conversely, the presence of the DARC SNP that confers Duffy antigen expression had a protective effect (p=.038). WES analysis did not identify additional non-synonymous APOL1 variants linked with albuminuria. However, 93 non-synonymous variants were associated with DTPA GFR (p<0.001). Using patients with eGFR as a validation cohort, 7 variants in FUBP1, ZFAND4, CD163, GMFG and HLA-E maintained their association with kidney filtrative function (p<0.05). In particular, two variants in CD163, which is a macrophage scavenger receptor for hemoglobin-haptoglobin complexes, were strongly associated with increased GFR in both patient cohorts.

 

Gene	SNP	Location	MAF	Albuminuria (Additive)	Variant frequency
					Cases	Controls
APOL1  G1	rs73885319	22q12	21.9%	0.053	32.0%	20.3%
APOL1  G2	rs71785313	22q12	16.0%	0.445	10.0%	16.0%
DARC	rs2814778	1q23	13.6%	0.038	4.2%	15.7%
eNOS    4a	VNTR	7q35	31.2%	0.333*	31.5%	22.7%
eNOS	rs1799983	7q35	12.5%	0.663	14.6%	12.2%
eNOS	rs2070744	7q35	15.6%	0.299	10.4%	16.3%
CUBN	rs7918972	10p12	16.2%	0.457	12.5%	16.8%
CUBN	rs1801239	10p12	2.7%	0.199	0.0%	3.1%

Table. Candidate genes associated with microalbuminuria. All SNPs were tested with either an additive or recessive genetic model. *The eNOS VNTR was analyzed by the chi-square method.

Conclusion:  Genetic polymorphisms associated with chronic kidney disease in African American adults may influence the development of early-onset albuminuria among children with SCA, including an increased risk among children with ≥1 APOL1 G1 alleles and a decreased risk associated with the DARC SNP. Previously published eNOS and CUBN variants had no measureable effects. WES analysis suggests novel genetic variants including CD163 SNPs may influence the development of elevated GFR in children with SCA, and provide candidate genes for future research.