In the present study, we have provided evidence that hypoxia stimulates oxidant production by PA SMCs in vitro
, and increases expression of oxidant-generating enzymes and oxidant levels throughout the lung in vivo
. Elevated lung oxidant levels were associated with loss of CREB in medial PA SMCs in vivo
with concomitant remodeling of the PA wall. Exogenous H2
treatment of PA SMCs in culture also elicited CREB depletion and increased cell proliferation, migration and altered production of ECM components. Oxidant scavengers effectively inhibited loss of CREB in response to H2
O2 in vitro
and chronic hypoxia in vivo
, while simultaneously attenuating SMC proliferation, migration and PA remodeling. Given our previous results14,19
demonstrating that CREB directly controls SMC growth, migration, differentiation and ECM production, the current data suggest that the detrimental impact of ROS on PA remodeling and SMC phenotype are largely mediated by oxidant-induced CREB loss. We are currently generating a SMC-targeted CREB loss-of-function mouse model to directly test the involvement of CREB in hypoxia/oxidant-induced PA remodeling and SMC function.
In the present study we show that oxidants elicit CREB loss in PA SMCs, and that superoxide scavengers completely inhibit this phenomenon. We previously reported that PDGF-BB, a mitogen produced in the arterial wall in response to hypoxia, also stimulates CREB depletion in SMCs via the PI3K/Akt signaling pathway14
. In this study, we demonstrated that activation of this signaling system promotes phosphorylation of CREB by casein kinase 2, which targets CREB for polyubiquitination and proteasomal degradation. Since inhibition of PDGF/PI3K/Akt signaling and oxidant scavenging both prevent CREB loss in SMCs, a common signaling mechanism may be involved in both processes. One potential candidate is the serine/threonine protein kinase Akt, which is activated by both oxidant stress22
in vascular SMCs. Another candidate is the PDGF receptor itself. Oxidants stimulate PDGF receptor phosphorylation and activity13,23
, and non-PDGF growth factors have been shown to stimulate PDGF receptor activity via a mechanism involving ROS and src-family kinases24
. Likewise, antioxidants reduce PDGF receptor signaling in cultured SMCs and in restenotic lesions25
. In our hands H2
alone induces the entire signaling cascade from PDGF receptor phosphorylation to activation of PI3K/Akt to increased casein kinase 2 catalytic subunit expression that we previously showed was responsible for CREB degradation in response to PDGF-BB14
(unpublished data). We propose that ROS and PDGF, both produced in the hypoxic arterial wall, may induce CREB loss via PDGF receptor activation, and are exploring this concept in greater detail.
Studies from our laboratory have demonstrated that CREB depletion in vascular SMCs elicited changes consistent with those observed in SMCs from pathologically remodeled arteries in vivo
. Such changes include decreased expression of SMC markers and contractile factors like SM-myosin, calponin, and fibronectin, increases in proliferation and proliferation-related factors like cyclin D1, and increases in ECM production14
. Leonard et al26
recently reported in vivo
lung-selective phosphorylation of CREB on serine-133 and activation of CREB in the absence of any change in total CREB levels in response to hypoxia. These results likely differ from ours because they were obtained in studies in which mice were exposed for shorter periods of time to modest levels of hypoxia, rather than our studies with rats exposed for 21 days to ~11% O2
. Furthermore, rodents respond moderately to altitude exposure27
and therefore require longer exposure to higher levels of hypoxia to induce PH and vascular remodeling. This may explain why we observe a decrease of total CREB levels in our models.
A somewhat surprising aspect of our experiments was that Tempol did not have any effect on the PA pressure elevation in hypoxic rats. Tempol is a stable nitroxide radical, which has also a low molecular weight and permeates biological membranes. Tempol scavenges both intra-and extracellular deleterious ROS and therefore mimics the benefits of the enzyme superoxide dismutase (SOD)28
. In addition tempol has been shown to effectively suppress pathological conditions associated with marked oxidative stress in vivo29,30,31,32
. Many studies have evaluated the capability of Tempol to prevent the increase in blood pressure in various rat models of hypertension. It has been reported that Tempol normalized right ventricular systolic pressure and reduced right ventricular hypertrophy in chronic hypoxic rats6
. In the majority of animal studies Tempol was administered in the drinking water with doses ranging from 1 to 3 mM. In our study the hypoxic group received 86 mg.kg−1
corresponding to 1 mM. Thus, the inability of Tempol to lower PA pressure and RV hypertrophy is probably not related to the dosage used in our study nor to an accumulation of ROS because immunohistochemical analysis of lung sections obtained from rats treated with Tempol and exposed to hypoxia revealed no positive staining for nitrotyrosine compared to hypoxic untreated rats. There are other studies in which Tempol failed to decrease arterial pressure. For example, Elmarakby et al30
have reported that Tempol did not have any effect of the blood pressure elevation in angiotensin-II induced systemic hypertension in rats. It is not clear whether there are other factors that limit Tempol’s activity.
Finally, our data suggest that pulmonary arterial wall remodeling or thickening may not contribute to the development of the hypoxic hypertension itself, since blockade of remodeling with Tempol had no beneficial affect on PA pressure. This concept is supported by studies employing methods to prevent or compensate for fixation method-dependent changes in lumen area. These studies showed that when the pulmonary vascular bed was maximally vasodilated during lung fixation, there was no reduction in vessel luminal area directly due to the medial and adventitial thickening33,34
. Other experiments have demonstrated that angiotensin-converting enzyme inhibitors and PPARg agonists prevent PA remodeling in rats exposed to chronic hypoxia but do not attenuate the development of PH or RV hypertrophy35,36
. Likewise, Nagaoka et al37
have reported that acute inhibition of RhoA/Rhokinase signaling almost completely reverses PH in rats exposed to chronic hypoxia, although the brief exposure to the inhibitors would not be expected to have any effect on the structural thickening of the PA wall. However, hypoxia-induced PH in rodents fails to mirror the progressive intimal remodeling observed in PH in cows and humans27
. Thus, antioxidant therapies that block remodeling may still prove useful in treating human PH, particularly in combination with the many vasodilators currently in use.
In conclusion, oxidative stress contributes to PA remodeling by promoting SMC growth and ECM deposition; characteristic features of the SMC phenotype in PH. We have demonstrated that chronic hypoxia induced ROS production is associated with an increase of NADPH oxidase expression. H2O2 leads to depletion of CREB and increased SMC proliferation. The development of chronic hypoxic pulmonary vascular remodeling was attenuated in rats treated with the superoxide dismutase mimetic, Tempol. While CREB is not the sole mediator of SMC growth, our study supports a preeminent role for this factor in controlling SMC functions in response to ROS.