Arginase Inhibition Reverses Endothelial Dysfunction, Pulmonary Hypertension, and Vascular Stiffness in Transgenic Sickle Cell Mice

Abstract

Background: In sickle cell disease (SCD), hemolysis results in the release and activation of arginase, an enzyme that reciprocally regulates nitric oxide (NO) synthase activity and thus, NO production. Simply supplementing the common substrate L-arginine, however, fails to improve NO bioavailability. In this study, we tested the hypothesis that arginase inhibition would improve NO bioavailability and thereby attenuate systemic and pulmonary vascular endothelial dysfunction in transgenic mice with SCD.

Methods: We studied 5-month-old transgenic sickle cell (SC) mice and age matched wild-type (WT) controls. SC mice were treated with the arginase inhibitor, 2(S)-amino-6-boronohexanoic acid (ABH; approximately 400 μg/d) for 4 weeks or left untreated.

Results: Vascular arginase activity was significantly higher at baseline in untreated SC mice compared to WT controls (SC versus WT, 346 ± 69.3 vs 69 ± 17.3 pmol urea/mg protein/minute; P = 0.0043; n = 4-5 animals per group). Treatment with ABH may significantly decrease arginase activity to levels near WT controls (SC + ABH 125.2 ± 17.3 pmol urea/mg protein/minute; P = 0.0213). Aortic strips from untreated SC mice showed decreased NO and increased reactive oxygen species (ROS) production (NO: fluorescence rate 0.76 ± 0.14 vs 1.34 ± 0.17 RFU/s; P = 0.0005 and ROS: fluorescence rate 3.96 ± 1.70 vs 1.63 ± 1.20 RFU/s, P = 0.0039; n = 3- animals per group). SC animals treated with ABH for 4 weeks demonstrated NO (fluorescence rate: 1.16 ± 0.16) and ROS (fluorescence rate: 2.02 ± 0.45) levels comparable with age-matched WT controls (n = 3- animals per group). The maximal endothelial-dependent vasorelaxation response to acetylcholine was impaired in aortic rings from SC mice compared with WT (57.7% ± 8.4% vs 80.3% ± 11.0%; P = 0.02; n = 6 animals per group). The endothelial-independent response was not different between groups. In SC mice, the right ventricular cardiac output index and end-systolic elastance were similar (4.60 ± 0.51 vs 2.9 ± 0.85 mL/min/100 g and 0.89 ± 0.48 vs 0.58 ± 0.11 mm Hg/μL), whereas the pulmonary vascular resistance index and right ventricular end-systolic pressure were greater (2.9 ± 0.28 vs 5.5 ± 2.0 mm Hg × min/μL/100 g and 18.9 ± 1.1 vs 23.1 ± 4.0 mm Hg; n = 8 animals per group). Pulse wave velocity (a measure of arterial stiffness) was greater in SC mice compared with WT (3.74 ± 0.54 vs 3.25 ± 0.21 m/s; n = 20 animals per group), arginase inhibition for 4 weeks significantly reduced the vascular SC phenotype to one similar to WT animals (P = 0.0009).

Conclusions: Arginase inhibition improves NO bioavailability and thereby attenuates systemic and pulmonary vascular endothelial dysfunction in transgenic mice with SCD. Therefore, arginase is a potential therapeutic target in the treatment of cardiovascular dysfunction in SCD.

Figure 1. Arginase activity in mouse aortic homogenates

Arginase activity is higher in sickle cell mice (SC; n = 5) compared to wild type controls (WT; n = 4; p=0.0009). Arginase inhibition (2(S)-Amino-6-BoronoHexanoic acid [ABH]; n = 5) in SC mice decreases arginase activity, to a level comparable to WTs (p=0.69 compared to WT and p=0.002 compared to SC; (*post hoc p-value < 0.01)).

Figure 2. Nitric oxide levels and reactive oxygen species

A) Nitric oxide production is decreased in sickle cell (SC) aorta compared to wild types (WT; p=0.009). Arginase inhibition with 2(S)-Amino-6-BoronoHexanoic acid (ABH) increases nitric oxide production (n = 3 animals per group; p=0.39 compared to WT and p=0.05 compared to SC). B) Rates of reactive oxygen species (ROS) production are higher in SC mice compared to WT (p=0.03). ABH treatment decreases this in SC mice (n = 5 animals per group; p=0.87 compared to WT and p=0.07 compared to SC); (*post hoc p-value < 0.01).

Figure 3. Vasoreactivity and endothelial function

A) Sickle cell mice (SC) have a reduced maximal endothelial dependent vasodilatory response to acetycholine (ACh) compared to wild types (WT, p=0.18), which becomes apparent at an ACh dose of 10⁻⁵ M. Inhibition of arginase restores endothelial-dependent vasodilation (p>0.99 compared to WT). B) There is no difference in the endothelial-independent response (sodium nitroprusside) between the groups (n = 6 animals per group, p>0.99 in all comparisons).

Figure 4. Right heart catheterization

A) Pulmonary vascular resistance index (PVRI) and B) right ventricular end systolic pressure (RVESP) were elevated in sickle cell (SC) mice compared to wild types (WT). C) Right ventricular cardiac output index (RVCOI) and right ventricular end-systolic elastance (RVEes) were similar in SC mice compared to WT. Treatment with ACh for 4 weeks restored PVRI, RVESP, RVCOI, and RVEes (n = 8 animals per group, post hoc p-values: PVRI: WT vs SC p=0.004, WT vs SC+2(S)-Amino-6-BoronoHexanoic acid (ABH ) p=0.76, SC vs SC+ABH p=0.19; RVESP: WT vs SC p=0.03, WT vs SC+ABH p=0.70, SC vs SC+ABH p=0.13; RVCOI: WT vs SC p=0.10, WT vs SC+ABH p=0.80, SC vs SC+ABH p=0.03; RVEes: WT vs SC p=0.18, WT vs SC+ABH p=0.83, SC vs SC+ABH p=0.44; *p<0.01).

Figure 5. Right ventricular pressure volume loops

Representative traces of right ventricular pressure volume loops in A) wild type, B) sickle cell, and C) sickle cell mice after treatment with 2(S)-Amino-6-BoronoHexanoic acid (ABH).

Figure 6. Pulse wave velocity

Pulse wave velocity is higher in sickle cell (SC) mice than in wild types (WT, p=0.09). SC mice treated with 2(S)-Amino-6-BoronoHexanoic acid (ABH), have significantly reduced PWV to a level that is not statistically different from WT (n = 20 animals per group; p=0.73 compared to WT and p=0.001 compared to SC; *p <0.01).