Roles for Nox4 in the contractile response of bovine pulmonary arteries to hypoxia

Abstract

Hypoxia appears to promote contraction [hypoxic pulmonary vasoconstriction (HPV)] of bovine pulmonary arteries (BPA) through removal of a peroxide-mediated relaxation. This study examines the roles of BPA Nox oxidases and mitochondria in the HPV response. Inhibitors of Nox2 (0.1 mM apocynin and 50 muM gp91-dstat) and mitochondrial electron transport (10 muM antimycin and rotenone) decreased superoxide generation in BPA without affecting contraction to 25 mM KCl or the HPV response. Transfection of BPA with small inhibitory RNA (siRNA) for Nox2 and Nox4 decreased Nox2 and Nox4 protein expression, respectively, associated with an attenuation of superoxide detection, without affecting 25 mM KCl contraction. However, Nox4 siRNA, but not Nox2, attenuated HPV in BPA. A Nox4 inhibitor plumbagin (10 muM) increased basal force, decreased superoxide detection and peroxide release, and caused BPA to relax under hypoxia. Although acute removal of peroxide with 0.1 mM ebselen increased 25 mM KCl contraction and decreased hypoxic contraction, prolonged treatment with ebselen only decreased hypoxic contraction without affecting 25 mM KCl contraction, suggesting basal peroxide levels also maintain a contractile mechanism not removed by acute hypoxia. Organ culture of BPA with transforming growth factor (TGF)-beta1 (4 nM) increased Nox4 expression, superoxide, peroxide, and the HPV response. Thus Nox2 and mitochondria are sources for superoxide generation in BPA, which do not appear to influence the HPV response. However, peroxide derived from superoxide generated by Nox4 appears to maintain a basal relaxation in BPA under normoxic conditions, which is removed under hypoxia leading to HPV. Peroxide generated by Nox4 may also function to maintain a contractile mechanism, which is not reversed by acute hypoxia.

Fig. 1.

Effects of inhibition of Nox2 activation on the levels of bovine pulmonary arteries (BPA) superoxide. Superoxide generation detected by 5 μM lucigenin chemiluminescence is decreased in the presence of 100 μM apocynin and 50 μM gp91-dstat (n = 10–12).

Fig. 2.

Effects of inhibition of Nox2 activation on the contraction of BPA to 25 mM KCl and hypoxia. Force generation to 25 mM KCl is not affected in the presence of 100 μM apocynin (n = 6; A) and 50 μM gp91-dstat (n = 7; B). Hypoxic contraction is also not affected in the presence of apocynin (n = 6; C) and gp91-dstat (n = 7; D).

Fig. 3.

Effects of small inhibitory RNA (siRNA) for Nox2 and Nox4 on the expression of these Nox oxidases in BPA. A: Western blot analysis showing 180 nM Nox2 siRNA decreases Nox2 expression, whereas 180 nM Nox4 and 180 nM scrambled siRNA (Scramb) do not affect Nox2 expression (n = 10). B: Western blot analysis showing Nox4 siRNA decreases Nox4 expression, whereas Nox2 and scrambled siRNA do not affect Nox4 expression (n = 7). Ab, antibody.

Fig. 4.

Effects of siRNA for Nox2 and Nox4 on the levels of superoxide in BPA and on the contraction of BPA to 25 mM KCl and hypoxia. A: superoxide levels detected 5 μM lucigenin are decreased in both Nox2 and Nox4 siRNA (180 nM)-treated BPA, whereas scrambled siRNA (Scramb; 180 nM) did not have any effect (n = 12). B: 25 mM KCl contraction is not affected by Nox4, Nox2, or scrambled siRNA (n = 15). C: hypoxic contraction is only decreased in the presence of Nox4 siRNA, whereas Nox2 and scrambled siRNA have no effect (n = 15).

Fig. 5.

Effects of the Nox4 inhibitor plumbagin on BPA contractile function and the response to hypoxia. A: original tracing of force development from BPA rings in the presence and absence of 10 μM plumbagin. B: basal force increases in the presence of plumbagin (n = 10). C: total force developed in the presence of 20 mM KCl (from basal tone before addition of plumbagin) is not changed in the presence of plumbagin (n = 10). D: hypoxic contraction is reversed to relaxation in the presence of 10 μM plumbagin (n = 10).

Fig. 6.

The Nox4 inhibitor plumbagin decreases BPA superoxide and peroxide, even in the presence of Nox2 inhibition. A: superoxide detected by 5 μM lucigenin is decreased in the presence of plumbagin (n = 11 to 12). B: peroxide release from BPA detected by 10 μM luminol plus horseradish peroxidase is also decreased in the presence of plumbagin (n = 11 to 12). Superoxide is lowered even further when plumbagin is added to apocynin (Apo; C) and gp91-dstat-treated BPA (gp91; n = 11 to 12; D).

Fig. 7.

Effects of short- and long-term removal of endogenous peroxide by ebselen on the contraction of BPA to KCl and hypoxia. A: long-term treatment (45 min before passive tension, Ebselen-Prol) of BPA with 0.1 mM ebselen does not affect 30 mM KCl contraction. The force generated by 30 mM KCl before DMSO vehicle (control) and acute ebselen (control-Eb) treatments is also shown (n = 11). B: summary data showing that contraction to 25 mM KCl is significantly increased on short-term (30 min) treatment with ebselen, whereas long-term treatment does not affect force generation elicited by 25 mM KCl (n = 11). C: summary data showing that hypoxic contraction is decreased by both short- and long-term treatment of BPA with ebselen (n = 11).

Fig. 8.

Evidence that transforming growth factor (TGF)-β1 increases superoxide and peroxide associated with elevated expression of Nox4. Typical Western blot (A) and summary data (B) showing that 4 nM TGF-β1 treatment increased Nox4 expression significantly in BPA (n = 11) are shown. C: summary data showing that superoxide generation is significantly increased on treatment with TGF-β1 for 48 h. Plumbagin (10 μM) reversed the superoxide increase in TGF-β1-treated vessels, whereas 100 μM apocynin had no effect (n = 8). D: summary data showing that peroxide release is also increased by TGF-β1 and plumbagin significantly decreased peroxide in TGF-β1-treated vessels (n = 4).

Fig. 9.

Evidence that peroxide generation associated with increases in Nox4 expression elicited by TGF-β1 is associated with alterations in the response of BPA to hypoxia. A: peroxide release from BPA organ cultured in the absence and presence of 4 nM TGF-β1 is attenuated by scavenging intracellular peroxide with ebselen and extracellular peroxide with 1 μM catalase (n = 4). Summary data show that treatment of BPA with 4 nM TGF-β1 decreases force generated by 25 mM KCl (B) and increases the magnitude of the contraction to hypoxia (C). The presence of ebselen increases force generated by 25 mM KCl in the absence or presence of pretreatment with TGF-β1, and ebselen markedly attenuates further contraction to hypoxia in both groups of organ-cultured arteries in a manner consistent with roles for changes in peroxide levels in the effects of TGF-β1 and hypoxia (n = 6).

Fig. 10.

Effects of inhibitors of mitochondrial electron transport on the levels of BPA superoxide and the response to hypoxia. Summary data show that 10 μM antimycin and 10 μM rotenone cause a significant decrease in superoxide generation (A) but do not affect force generated by 20 mM KCl (B) or the magnitude of the contraction to hypoxia in BPA (n = 6–8; C).

Fig. 11.

Summary model showing where data generated in the present study provide evidence for a hypothesized role for hypoxia modulating the generation of peroxide by Nox4 in a manner that controls a relaxing mechanism removed by acute exposure to hypoxia leading to a hypoxic pulmonary vasoconstriction (HPV) response. The model includes a hypothesized role for basal levels of peroxide generated by Nox4 controlling a force enhancing process needed for the HPV response, which is not reversed by acute hypoxia. This model also illustrates how hypothesized mechanisms such as peroxide-elicited relaxation through activation of soluble guanylate cyclase (sGC) and increased cGMP protein kinase (PKG)-mediated signaling and peroxide-elicited force enhancement through increased Rho kinase activity have properties of systems that could participate in the BPA response to hypoxia.