Validation of Townes As Mice As a Model of Chronic Kidney Disease and Venous Thrombosis Associated with Sickle Cell Trait

Abstract

Individuals with sickle cell trait (SCT) are heterozygous carriers of the sickle beta-globin gene. Systematic analyses of large cohort studies over the past decade have established that SCT is a risk factor for a limited number of complications, most notably chronic kidney disease (CKD) and venous thrombosis (VT). SCT red blood cells (RBCs) can undergo sickling, although the degree and duration of hypoxia required to produce sickling are much greater than in sickle cell disease due to the lower intra-erythrocyte sickle hemoglobin beta (HbS) concentration of 25-45%. Hypoxia-mediated HbS polymerization can be enhanced by cellular dehydration, hyperosmolarity and/or acidosis that may be encountered in the renal inner medulla. Furthermore, prolonged exposure of SCT RBCs to profound hypoxia, such as in the nidus of a venous clot, will also result in sickling. The objective of this study was to determine if Townes AS mice manifest kidney dysfunction analogous to their SCT human counterparts. Furthermore, we evaluated whether VT was more pronounced in a well-described model of inferior vena cava (IVC) stasis. We used humanized male sickle cell trait (AS) and age-matched genetic control (AA) mice (n=7-16). Analysis of urinary renal biomarkers (albuminuria, proteinuria, kidney injury marker 1 (KIM-1), osmolality) and kidney function (transcutaneous measurement of glomerular filtration rate (GFR)) were conducted at 4-week intervals from 8 to 40 weeks of age. VT was induced by complete occlusion of the IVC in AA and AS mice. Sickling of RBCs was determined by histological and electron microscopy analysis of the IVC clot. Senicapoc, an oral Gardos channel inhibitor, was administered twice daily (20mg/kg, bid) for 2 weeks to prevent AS RBC dehydration and subsequent sickling. Human HbS accounts for 30.1% of total Hb expression in AS mice. Consistent with human SCT, peripheral blood counts did not differ from AA controls. Basal plasma thrombin-antithrombin complex levels were also not different between AA and AS mice. We performed a longitudinal study of renal structure and function in AA and AS mice over 40 weeks. At 8 weeks of age, AS mice exhibited no differences in any parameter compared to AA controls. However, AS mice subsequently developed albuminuria and a moderate but statistically significant increase in GFR (Figure 1). At 36 and 40 weeks of age, we observed progressive loss of kidney function demonstrated by statistically significant decreased GFR accompanied by a significant increase in albuminuria in AS mice (Figure 1). Over time, proteinuria gradually increased in AS mice, and had doubled by 40 weeks of age compared to AA controls (4.6 ±1.1 vs. 2.3 ±0.4 mg/24h, p<0.05). Increased levels of urinary KIM-1, a marker of tubular damage, were observed in AS mice throughout the duration of the study, with the highest levels at 16 weeks (228.3 ± 32.6 pg/24h vs. 53.8 ± 8.5 pg/24h, respectively, p˂0.05). Urine osmolality was not different at 8 weeks of age, but progressively declined in AS mice with statistically significant changes vs. controls starting at 24 weeks until the end of study (1946 ±63 vs. 2672 ±135 mOsm/kg at 40 weeks, p=0.002). Histologic analysis confirmed glomerular injury (congestion, glomerulosclerosis) and tubular injury (medullary congestion, interstitial fibrosis) in 40 week old AS mice. Following 24 hours of IVC stasis, thrombus weight was significantly increased (p<0.01) in AS mice (mean±SEM; 29.0 mg +/- 1.7; n=8) compared to AA controls (20.0 mg ± 1.7; n=7). Histologic and electron microscopic evaluation demonstrated that severe hypoxia (pO 2 ≈ 10 mm Hg) present in the nidus of venous thrombi was associated with extensive sickling of RBCs in AS mice. To determine if the enhancement of VT observed in AS mice was mediated by RBC sickling, mice were pretreated with Senicapoc or vehicle. Compared to vehicle-treated AS mice (27.3 mg ± 2.2 n=9) pre-treatment with Senicapoc significantly reduced (p<0.05) clot weight (18.1mg ± 1.1; n=7) to that observed in untreated AA controls. In aggregate, our data demonstrate that Townes AS murine nephropathy closely mimics the renal abnormalities (impaired urinary concentration and albuminuria with later onset loss of GFR) described in humans with SCT. Furthermore, we have established a mouse model of the venous thrombotic complications associated with SCT and have demonstrated that hypoxia-induced sickling of RBCs may contribute to enhanced VT observed in AS mice.