Voltage-dependent anion channel-2 interaction with nitric oxide synthase enhances pulmonary artery endothelial cell nitric oxide production

Abstract

Increased pulmonary artery endothelial cell (PAEC) endothelium-dependent nitric oxide synthase (eNOS) activity mediates perinatal pulmonary vasodilation. Compromised eNOS activity is central to the pathogenesis of persistent pulmonary hypertension of the newborn (PPHN). Voltage-derived anion channel (VDAC)-1 was recently demonstrated to bind eNOS in the systemic circulation. We hypothesized that VDAC isoforms modulate eNOS activity in the pulmonary circulation, and that decreased VDAC expression contributes to PPHN. In PAECs derived from an ovine model of PPHN: (1) there is eNOS activity, but not expression; and (2) VDAC1 and -2 proteins are decreased. Immunocytochemistry, coimmunoprecipitation, and in situ proximity ligation assays in human PAECs (hPAECs) demonstrate binding between eNOS and both VDAC1 and -2, which increased upon stimulation with NO agonists. The ability of agonists to increase the eNOS/VDAC interaction was significantly blunted in hypertensive, compared with normotensive, ovine PAECs. Depletion of VDAC2, but not VDAC1, blocked the agonist-induced increase in eNOS activity in hPAECs. Overexpression of VDAC2 in hypertensive PAECs increased eNOS activity. Binding of VDAC2 enhances eNOS activity in the pulmonary circulation, and diminished VDAC2 constrains eNOS in PAECs derived from fetal lambs with chronic intrauterine pulmonary hypertension. We speculate that decreases in VDAC2 may contribute to the limited eNOS activity that characterizes pulmonary hypertension.

Figure 1.

Chronic intrauterine hypertension decreases nitric oxide synthase (NOS) activity in pulmonary artery endothelial cells (PAECs) without changing eNOS expression. (A) PAECs isolated from normotensive (Norm) and hypertensive (HTN) fetal sheep were treated with either vehicle or the NO agonist, A23187 (2 μM), and NOS activity was determined by conversion of L-arginine to L-citrulline at 30 minutes (***P < 0.001 in Norm A23187-stimulated versus vehicle Norm, and ^§§§P < 0.001 in A23187-stimulated HTN versus vehicle HTN, and versus A23187-stimulated Norm PAECs; n = 3). Data expressed as percent increase in NOS activity from baseline in Norm and HTN PAECs in response to A23187 treatment. (B) Quantitative RT-PCR (qRT-PCR) was performed and the absolute copy number of eNOS was normalized to 18S rRNA to determine relative expression. There was no difference in eNOS expression between Norm and HTN PAEC. (C) Top panel: Western immunoblot to determine the level of eNOS protein expression in Norm versus HTN fetal sheep. Bottom panel: Summary of densitometry data with eNOS protein expression normalized to α-actinin (n = 3 PAEC isolates from three separate animals).

Figure 2.

Chronic intrauterine hypertension decreases expression of VDAC isoforms 1 and 2. qRT-PCR was performed and the absolute copy number of VDAC1 (A) and VDAC2 (B) was normalized to 18S rRNA to determine relative expression in Norm versus HTN fetal ovine PAECs. Western immunoblot was to determine the level of VDAC1 (C) and VDAC2 (D) protein in Norm versus HTN fetal ovine PAECs, with VDAC1 and -2 protein expression normalized to α-actinin (*P < 0.05, with n = 5 and n = 3 PAEC isolates from Norm and HTN animals, respectively).

Figure 3.

eNOS colocalizes with VDAC1 and -2 in human PAECs (hPAECs). Representative immunocytochemistry on hPAECs to detect VDAC1 (A) or VDAC2 (B) and eNOS. Incubation of hPAECs with anti-VDAC antibody (red) and eNOS (green) demonstrate significant colocalization of both VDAC isoforms with eNOS, particularly in a perinuclear distribution. Insets demonstrate higher magnification images. Calibration mark = 100 μm in all images.

Figure 4.

eNOS physically interacts with VDAC1 and -2 in hPAECs, and enhanced interaction occurs with stimulation by NO agonists. (A) Representative coimmunoprecipitations performed with either mouse IgG1 or anti-eNOS as the immunoprecipitating antibody, and immunoblotted with anti-eNOS, -VDAC1, or -VDAC2 antibodies. (B) Summary of densitometric data from replicate coimmunoprecipitations. Data are normalized to immunoprecipitations performed with the isotype control (mouse IgG1) (**P < 0.01, n = 4–7). (C) Representative images from in situ proximity ligation assays performed to detect the interaction of eNOS and either VDAC1 or -2 in hPAECs under basal conditions, and after stimulation with either A23187 or histamine. Calibration mark = 50 μm. **P < 0.01 versus vehicle; ***P < 0.001 versus vehicle.

Figure 5.

Chronic intrauterine hypertension blunts the interaction of eNOS and VDAC isoforms upon stimulation with NO agonists. Representative immunocytochemistry on hPAECs to detect VDAC1 (A) or VDAC2 (B) and eNOS. Incubation of hPAECs with anti-VDAC antibody (red) and eNOS (green) demonstrate significant colocalization of both VDAC isoforms with eNOS. Boxes delineate the image field chosen to depict under higher magnification. Calibration mark = 100 μm. (C) Representative images from in situ proximity ligation assays performed to detect the interaction of eNOS and VDAC1 in Norm and HTN ovine PAECs under basal conditions, and after stimulation with either A23187 or histamine (***P < 0.001 versus Norm or HTN vehicle control; ^§P < 0.05 versus HTN vehicle treated and P < 0.01 versus Norm histamine stimulated). Calibration mark = 50 μm. (D) Representative images from in situ proximity ligation assays performed to detect the interaction of eNOS and VDAC2 in Norm and HTN ovine PAECs under basal conditions, and after stimulation with either A23187 or histamine (***P < 0.001 versus vehicle-stimulated Norm cells; ^§P < 0.05 versus HTN vehicle-treated cells and <0.01 versus A23187-stimulated Norm cells; ^##P < 0.01 versus Norm vehicle-treated cells and histamine-stimulated HTN cells). Calibration mark = 50 μm.

Figure 6.

Loss of VDAC2 in hPAECs impairs the histamine-induced increase in NOS activity. (A) qRT-PCR (left) and Western immunoblot (right) to determine VDAC1 gene expression 24 hours after transfection with siRNA. For qPCR, target gene expression was normalized to the levels of 18S rRNA within each sample, and normalized to the average of samples transfected with nontargeting control (NTC) siRNA (***P < 0.001, n = 3). For WB, summarized densitometric data with expression normalized to tubulin levels within each sample and normalized to the average ratio found in hPAECs transfected with NTC siRNA (***P < 0.001, n = 12). (B) qRT-PCR (left) and Western immunoblot (right) to determine VDAC2 gene expression 24 hours after transfection with siRNA. For qPCR, target gene expression was normalized to the levels of 18S rRNA within each sample, and normalized to the average of samples transfected with NTC siRNA (***P < 0.0001, n = 6). For WB, summarized densitometric data are presented with expression normalized to tubulin levels within each sample, and normalized to the average ratio found in hPAECs transfected with NTC siRNA (***P < 0.001, n = 15). NOS activity was measured in VDAC1- (C) or VDAC2 (D) -depleted hPAECs after treatment with either vehicle (white bars) or histamine (200 μM; black bars), and NO activity determined by the conversion of L-arginine to L-citrulline at 30 minutes (*P < 0.05, **P < 0.01, and ***P < 0.001 versus vehicle and ^§§§P < 0.001 versus histamine-stimulated NTC; n = 9).

Figure 7.

Overexpression of VDAC2 increases eNOS activity in HTN PAECs. (A) Representative confocal micrographs after immunostaining to detect green fluorescent probe (GFP) in untransfected, GFP-vector–transfected, and VDAC2-GFP–transfected HTN PAECs. Calibration mark = 100 μm. (B) Western immunoblot on protein lysates obtained from untransfected, GFP-vector–transfected, and VDAC2-GFP–transfected HTN PAECs demonstrated strong expression of the 29-kD GFP in the vector-transfected cells, and the presence of the 60-kD predicted GFP-tagged VDAC2 protein in the VDAC2-GFP–transfected cells. (C) NOS activity was measured in HTN PAECs 24 hours after transfection with either empty vector or VDAC2 overexpression vector, and after treatment with either vehicle (white bars), or A23187 (2 μM; black bars) (***P < 0.001 versus VDAC2 overexpression PAECs treated with vehicle and ^§§P < 0.01 versus empty vector PAECs stimulated with A23187; n = 3).