Molecular control of blood flow and angiogenesis: role of nitric oxide


William C. Sessa, Yale University School of Medicine, Vascular Biology & Therapeutic Program, 10 Amistad St, New Haven, CT 06520, USA.
Tel.: 203 737 2290.


Summary.  In the past decade, the importance of the vascular endothelium as a multifunctional regulator of vascular smooth muscle physiology and pathophysiology has been appreciated. Indeed, the endothelium responds to hemodynamic stimuli (pressure, shear stress and wall strain) and locally manufactured mediators (such as bradykinin, prostaglandins, angiotensin II and nitric oxide) that can influence blood flow, cell trafficking into tissue and angiogenesis. In this chapter, the importance of nitric oxide (NO) as a mediator of blood flow control, vascular permeability and angiogenesis will be discussed.

Nitric oxide (NO), a free radical gas produced by the NO synthase (NOS) family of proteins, is a ubiquitous second messenger in diverse physiological responses in the cardiovascular system including vasodilation, anti-coagulation, vascular remodeling and angiogenesis. In vitro and in vivo studies using NOS inhibitors or more recently, using mice deficient in each of the three mammalian isoforms, namely: (i) endothelial NOS (eNOS); (ii) cytokine-inducible NOS (iNOS) or (iii) neuronal NOS (nNOS), have shown definitive roles of NO and its sources in the cardiovascular system. The purpose of this review is to highlight the evidence supporting various functional roles of NO in control of local blood flow and angiogenesis.

NO and vasodilation

NO is clearly the major endothelium-dependent vasodilator produced by large blood vessels. In the aorta or carotid arteries of mice deficient in eNOS (eNOS −/− mice), endothelium-dependent vasodilation is eliminated, thus supporting eNOS derived NO as the source of the relaxing factor that led to the Nobel Prize in 1998. However, the vasodilatory roles for NO in arterioles and venules appear less unanimous and the importance of NO as a vasodilatory agent (deciphered by the sensitivity of the response to blockage of NOS) depends on the stimulus (flow versus other agonists such as acetylcholine, VEGF, histamine), the vascular bed studied (coronary, cremaster, gracillus, mesentery), the species (human, rat, mouse, hamster) and the number of endothelial-smooth muscle interconnections (i.e. myo-endothelial gaps). In eNOS (−/−) mice, the local vasodilatory actions of acetylcholine or flow induced changes in blood flow are not diminished, due to compensation by upregulation of nNOS, endothelium-derived hyperpolarizing factor (EDHF) and vasodilatory prostaglandins [1,2]. However, the local vasodilatory action of histamine was absent from second order arterioles in eNOS (−/−) mice [3]. These are only a few of the many examples highlighting some of the differential response to NO as aforementioned. Anther theory imparting NO as an important factor in microvascular blood flow control is the concept that NO bound to hemoglobin in red blood cells can serve as a stable NO-adduct for delivery at sites of resistance. Clearly NO binds to the heme moiety of hemoglobin while a reactive thiol, cysteine β 93, and hemoglobin can undergo allosteric changes in structure initiated by a drop in arteriolar pO2, and release NO from these sites [4]. Although this principle can be demonstrated in model systems, the physiological role of NO via its release from hemoglobin in regulating blood flow in the microcirculation remains controversial. Directed experiments where mice were generated containing a genetic mutation of cysteine β 93 in hemoglobin does not result in pO2 dependent changes in blood flow or hypoxic vasodilation, arguing strongly against this theory [5]. An alternative, emerging theory of how the endothelium may control microvascular blood flow is through another endothelium-dependent vasodilator, EDHF. In systems where NOS is blocked or absent, EDHF may subserve a role as a key regulator of microvascular blood flow control.

Preserving a non-thrombogenic surface of the endothelium

In addition to the role of NO as a vasodilator, NO in the microcirculation can prevent platelets from adhering to the endothelium and can assist with the dis-aggregation of activated platelets to the endothelium or underlying basement membrane. This concept was unequivocally demonstrated in mice deficient for eNOS [6,7]. Bleeding times were decreased in eNOS (−/−) mice and platelet adhesion to venules was increased in response to bacterial lipopolysachharide (LPS) when a NOS inhibitor was given. Similarly, mice deficient in eNOS, but not iNOS or nNOS also exhibited increased platelet adhesion. In both instances, eNOS in the endothelium as well as in the platelet, per se contributed to the anti-platelet actions of eNOS-derived NO.

Maintaining an anti-inflammatory state of the vessel wall

Inflammation can be described as the physiological response of the microcirculation to injury or infections. Designed to close off or destroy injured tissue, inflammatory response is hallmarked by increases in leukocyte-endothelium interactions. Recently, NO has emerged as a crucial endogenous anti-inflammatory mediator in a number of pathophysiological states including hypercholesterolemia and ischemia-reperfusion injury. More specifically, NO can act by downregulating cytokines, resulting in the downregulation of endothelial cell adhesion molecules (ECAMs). The mechanism of leukocyte recruitment is mediated by these ECAMs such as the selectin family (P- and E-selectin), which are important modulators of leukocyte-endothelium interaction via leukocyte rolling along the endothelium and adhesion to the endothelium. Once the cells begin to roll, they can then firmly attach to the endothelium via integrin interaction with endothelial intercellular adhesion molecules (ICAMs) to promote leukocyte adhesion.

Pharmacological studies have shown that NOS inhibitors could increase leukocyte adhesion, an effect that can be reversed by large amounts of exogenous l-arginine. Similar studies using eNOS (−/−) mice confirmed that upon activation of an inflammatory response, eNOS derived NO is critical for reducing leukocytes adhesion and the extent of tissue injury. eNOS and nNOS (−/−) mice exhibited increased expression of P-selectin in their mesenteric venules as well as increased leukocyte rolling and adhesion that can be blocked by both the P-selectin neutralizing antibody and a high-affinity P-selectin ligand [8]. This supports previous findings that NOS inhibition by NOS inhibitors caused a significant increase in P-selectin expression in the post-capillary venules. Using another inflammatory model, namely the myocardial ischemia–reperfusion injury, NO was also shown to provide a protective role in the injury cascade leading to inflammation. Taken together, these findings are in agreement with previous observations of the cardioprotective role by NO in the myocardial microcirculation using exogenous NO donors and the exacerbated injury when animals were treated with NOS inhibitor. However, how does NO influence P-selectin expression? Work pioneered by Lowenstein et al., has shown that NO mediated nitrosylation regulates the flux of proteins via the secretory pathway[9]. Indeed, S-nitrosylation of NSF1 reduces agonist stimulated increases in P-selectin expresson, whereas NOS inhibiton or transduction of cells with peptidomimetic of NSF, reduces NSF nitrosylation and promotes P-selectin expression on the cell surface [10]. These data are supported by work from our group showing that NO regulates the flux of protein trafficking from the endoplasmic reticulum to the cell surface [11].

Role of NO in regulating vascular permeability and angiogenesis

One of the initial events in an acute inflammatory response and during angiogenesis is increased extravasation of fluid and protein from post-capillary venules at sites of tissue injury. Under physiological conditions, microvascular fluid exchange is primarily determined by blood flow, Starling forces and the perfused surface area. The perm-selectivity barrier to solute flux is determined by multiple factors including the identity of the endothelium (tissue bed), the glycocalyx, composition of adherens junctions and extracellular matrix.

Using cannulated, venular microvessels in the mesentery, a variety of inflammatory agents such as ATP, ionomycin, bradykinin, histamine and VEGF stimulated an increases in [Ca2+]i slightly preceding the initial peak in vascular leakage. The delay of the onset of changes in permeability suggests that in addition to the initial increase of [Ca2+]i, calcium-dependent signaling cascades (i.e. eNOS) may contribute to vascular permeability. Furthermore, NOS inhibitors can reduce ATP- induced increases in single vessel permeability, while 8-Br-cGMP, a membrane permeable cGMP analog, can potentiate ATP driven increases in permeability. These agents, however, did not affect the ATP induced rise in [Ca2+]i, which further suggests that the increase in [Ca2+]i dependent mechanisms are critical in promoting vascular leak.

Over the years, studies on the action of NO in regulating vascular permeability have yielded conflicting observations. Early studies using NO donors and NOS inhibitors demonstrated the negative regulatory effect of NO in microvascular permeability. For example, administration of a NOS inhibitor, Nω-nitro-l-arginine-methyl ester (l-NAME) increased transvascular fluid and protein flux, whereas subsequent NO donors such as sodium nitroprusside (SNP) reversed this effect in the intestinal circulation. The increase in permeability was not due to hemodynamic alteration, a phenomenon in which l-NAME has been suggested to exert. Many other studies using NOS inhibitors or NO donors also demonstrated that NO negatively regulates permeability in response to a host of inflammatory agents including bradykinin, carrageenan, substance P and mustard oil. In addition, increases in intracellular levels of cGMP by NO donors or activators of guanylate cyclase (sGC) have been associated with a decrease in endothelial permeability.

On the other hand, the NO-sGC-cGMP pathway can positively regulate microvascular permeability. Upon receptor activation and subsequent stimulation of phospholipase C (PLC), increase in cytosolic calcium can lead to activation of eNOS and NO production. NO activates guanylate cyclase (GC) and increases cGMP production, resulting in protein kinase G (PKG) activation and phosphorylation of proteins that regulate the contractile apparatus or cytoskeleton. In cannulated porcine coronary venules, VEGF and histamine can induce increases in permeability, as measured by the rate of FITC-albumin into the extravascular space in a cGMP-PKG dependent manner. Administration of the NO donor SNP, increased permeability similar to that induced by histamine, and this increase can be blocked completely by an inhibitor of guanylate cyclase.

Recent studies have examined the effect of eNOS in modulating vascular permeability using eNOS (−/−) mice or using a synthetic peptide that inhibits eNOS activity. A pivotal study by Jain’s laboratory [12] demonstrated that eNOS (−/−) mice, but not iNOS (−/−) mice showed impairment in VEGF-induced increases in permeability, further confirming the importance of eNOS as the predominant NOS isoforms in the regulation of VEGF induced vascular permeability. To examine further the role of eNOS in regulating vascular permeability, our lab has generated a synthetic fusion protein containing the eNOS inhibitory domain of caveolin-1 and a cell permeable peptide that facilitates the uptake of cargo proteins/oligonucleotides [13]. This peptide, which we termed cavtratin, can inhibit acetylcholine induced endothelium dependent relaxation. Administration of this peptide in vivo blocks inflammation (in response to PAF, carrageenan and mustard oil) and edema formation. Combined with our observations that eNOS (−/−) mice exhibited decreased tumor permeability and progression [14], these studies may provide novel insights into development of new anti-permeability drugs to target tumor growth.

In summary, the discovery of NO has led to the identification of multiple physiological roles for this gaseous second messenger in the microcirculation. We surmise that the local autocrine actions of NO that govern inflammation, vascular permeability and angiogenesis is likely as important as its vasodilatory role in conduit vessels. Future mechanistic studies and the development of novel reagents to manipulate faithfully the NO-sGC-PKG pathway will help dissect the physiologic importance of NO in the microcirculation.

Disclosure of Conflict of Interests

The author states that he has no conflicts of interest.