Inflammatory mechanisms that may contribute to atherosclerosis and endothelial dysfunction, especially through superoxide, were reviewed in this lecture. Data suggests that gene transfer of extracellular superoxide dismutate reduces blood pressure in hypertensive animals and protects against endothelial dysfunction in heart failure. Endothelial dysfunction perhaps contributes to the events of atherosclerosis.

Several risk factors may lead to inflammation and the generation of superoxides, leading to endothelial dysfunction and hence atherosclerosis and its complication of thrombosis. Treatment to regress atherosclerosis may have some very beneficial effects.

The standard list of risk factors for cardiovascular disease (CVD) includes hypertension, hypercholesterolemia, diabetes, smoking, age, males, hyperhomocysteinemia, visceral fat, elevated C-reactive protein (CRP) levels), and familial premature CVD. These risk factors have different effects in different organs. However, Heistad contends that to better understand vascular disease, a “superfamily” of risk factors would be useful, and he challenges young scientists to develop such a superfamily of risk factors that considers the mediators of vascular disease, the mechanism, and the target organs.

Impaired endothelial function is present in atherosclerosis. Endothelial dysfunction (ED) is a predictor of CV risk and is probably a cause of CV events. The key mechanism for ED is the interaction of nitric oxide (NO) and superoxide to form peroxynitrite. High levels of superoxide and peroxynitrite can damage vessels. The generation of superoxide is clearly associated with increased risk of CVD.

A common mediator of vascular disease is reactive oxygen species, especially superoxide. A number of enzymes can generate superoxide, including NAD(P)H oxidase, xanthine oxidase, NO synthase in the absence or presence of reduced levels of substrate,  mitochondria, and myeloperoxidase. Superoxide dismutase dismutes the superoxide to hydrogen peroxide.

Therefore, superoxide is considered to be important in vascular disease. Hence its reduction should reduce CVD. However, this has been difficult to demonstrate. Although trials have shown that vitamin E reduces CV events (as a primary or secondary endpoint), it remains unclear whether antioxidant vitamins protect against CVD. A better understanding of the causes of CVD and the role of superoxide is needed.

Work by Miller in Heistad’s lab developed and validated a new laboratory methodology for identifying the presence of superoxide in the blood vessel wall, and demonstrated that the vessel media is an important source of superoxide in atherosclerosis. In the control vessel, some superoxide is present in the endothelium, but in the atherosclerotic rabbit vessel, superoxide is present in the endothelium and surprisingly in the vessel media.

Superoxide is increased in the vessel wall in a number if disease states. Yokoyama and colleagues showed a great increase in superoxide in coronary atherectomy samples from patients with unstable angina. Lund and Heistad showed increased superoxide in the vessel wall, including the media, in a diabetic rabbit. Nakani showed that a 1-week infusion of angiotensin II increases superoxide in the endothelium and the media.

Oxidative stress is thought to contribute to hypertension because of the increased levels of reactive oxygen species, especially superoxide, in the vessel wall in several models of hypertension, and because superoxide inactivates NO and hence increases blood pressure. Antioxidant vitamins and superoxide dismutase (SOD) generally, but not always, reduce arterial pressure.

A recent study from study from Heistad’s lab provided evidence that superoxide may be contributing to hypertension, in their experimental model, and that gene therapy may be a useful treatment for hypertension. The injection of an adenovirus expressing extracellular SOD (AdECSOD) was injected intravenously into spontaneously hypertensive rats (SHR) and normotensive WKY rats. The adenovirus makes extracellular SOD in the liver, which is released into the circulation. The SOD was hypothesized to travel to blood vessels outside the liver, reduce superoxide, increase NO, and reduce blood pressure. Indeed, mean arterial pressure was significantly reduced in the SHR but not the WKY rats.

The effects of heparin on ECSOD were then studied in Heistad’s lab to determine whether the heparin binding site is important for the function of ECSOD.  ECSOD binds to the heparin sulfate proteoglycan on the outside of the cell, through heparin binding sites. Intravenous heparin injection releases ECSOD from the cell.

Their study showed that mean arterial pressure was not lowered by the copper-zinc SOD genome (CuZnSOD) lacking the heparin-binding domain (HBD), whereas it was reduced with the AdECSOD gene transfer. These data suggest that the HBD is critical to the normal function of ECSOD.

Further study showed that deletion of the HBD did not interfere with the function of ECSOD in the blood, but it had no effect on the tissue. They concluded that despite the high circulating levels of ECSOD minus HBD, it was not able to bind to tissues and reduce blood pressure.

Foltz and colleagues described a R213G mutation in the heparin-binding domain of ECSOD in humans, which is associated with a 10-fold increase in circulating levels of ECSOD. Whether this mutation is associated with increased susceptibility to a variety of CVD is of interest.

Immunostaining showed high levels of AdECSOD and AdECSOD minus HBD in the liver. But, in the artery there was minimal AdECSOD minus HBD, although there were high levels of AdECSOD. Endothelial function was greatly improved with AdECSOD, but not with AdECSOD minus HBD. Protein levels in the glomeruli were increased with AdECSOD, but not with AdECSOD minus HBD. One mechanism whereby ECSOD reduces blood pressure may be through alteration of the sodium balance, based on further study by Heistad in collaboration with DiBona.

The current concept is that reactive oxygen species, especially superoxide, is the common denominator in CVD, although low levels of superoxide and hydrogen peroxide are important in signaling. Although there have been some excellent studies of SOD in patients, one prior to angioplasty to reduce reperfusion injury and another in head injury, they were negative. However, would ECSOD without the HBD be beneficial in patients?

Work by Takeshita and colleagues in the canine model suggested that superoxide is increased in heart failure. One concept is that activation of NAD(P)H in blood vessels by angiotensin in the setting of heart failure generates superoxide, thereby contributing to vasomotor dysfunction, a significant problem in heart failure.

A study by other investigators showed that endothelium-bound ECSOD activity is reduced in patients with heart failure compared to control subjects. So if the renin-angiotensin system is activated, the reduction in ECSOD may contribute to endothelial dysfunction.

Work by Iida and Heistad looked at improving endothelial function in rats. Seven weeks after ligation of the left anterior descending artery, 3 groups of rats were given either AdECSOD or AdECSOD minus HBD. The control group was not ligated and was given AdECSOD. In heart failure, the ejection fraction was greatly reduced, lung weight increased, and superoxide levels in the aorta markedly increased. AdECSOD normalized the levels of superoxide in the aorta, whereas AdECSOD minus HBD did not. Endothelial function was restored by AdECSOD in this study.

In the future, with adenoviruses that do not cause an inflammatory response, it may be feasible to use gene therapy in heart failure to reduce levels of superoxide and thereby reduce peripheral resistance and improve vascular function. This is an area for future research.

Inducible nitric oxide synthase (iNOS) is expressed in a variety of CVD. It is perhaps part of the inflammatory response induced by risk factors that leads to CVD.

Work in Heistad’s lab showed that iNOS impairs endothelial function. In their experiment, a recombinant virus expressing iNOS was injected into the rabbit carotid artery. Endothelial function was impaired with iNOS and restored to normal with aminoguanadine. This finding has been consistent in a number of settings in vivo and in vitro in further work by this group.

In human intracranial blood vessels, obtained at the time of surgery for seizures, Heistad’s group showed after gene transfer that bradykinin causes endothelium-dependent relaxation and iNOs impairs relaxation. Interestingly, iNOS also impaired responses to nitroprusside. So, iNOS can impair responses both by an endothelial effect and by NO generated by nitroprusside. Thus, the inflammatory response and expression of iNOS are sufficient to produce endothelial dysfunction.

Interleukin (IL)-10 is an important anti-inflammatory cytokine. Heistad and colleagues showed that endothelial function was significantly impaired in IL-10 deficient mice, but not wild-type mice, with the same low dose of an endotoxin. Work by another group work showed that IL-10 modulates the development of atherosclerosis. Overexpression of IL-10 in wild-type mice attenuated the development of atherosclerosis, while the IL-10 deficient mice had increased atherosclerosis. 

Endothelial function improves with regression of atherosclerosis, as shown by experimental and human studies. Work in the carotid artery of an atherosclerotic monkey by Heistad and colleagues showed reduction in superoxide across the vessel wall when atherosclerotic lesions regress, along with reduction of macrophages and subunits of NAD(P)H.

The protein C anticoagulant pathway plays a role in atherosclerosis. At the site of injury, thrombin not only contributes to the thrombus, but binds to thrombomodulin on the endothelium, and thereby activates circulating protein C.  Activated protein C (APC) is an extremely potent anticoagulant. APC is used in severe sepsis and has been shown to reduce mortality in patients with sepsis.

The activity of thrombomodulin on the endothelium is reduced in atherosclerosis.
Heistad’s group showed that activated PTT is greatly attenuated and APC slightly attenuated in atherosclerosis after injection of thrombin in normal and atherosclerotic monkeys. This may be one mechanism that contributes to the susceptibility to thrombosis in atherosclerotic arteries. After regression of atherosclerosis, thrombomodulin increases. Thrombin causes a greater increase in ACP and a pretty good recovery in activated PTT in the atherosclerotic monkeys. So, the anticoagulant mechanism is improved with regression of atherosclerosis, and this may be contributing to the marked reduction in CV events after regression of atherosclerosis.