P450 in the Angiogenesis Affair

Angiogenesis is a fundamental process of growth and differentiation of new blood vessels in the body. Angiogenesis is new vessel growth from pre-existing vessels by budding, sprouting, and subsequent formation of patent capillaries with lumens, whereas vasculogenesis is new vessel growth from endothelial cell precursors, or stem cells.1,2 Both processes occur at different times in different tissues and are likely regulated by complex sets of effectors in physiological and pathological conditions. In recent years, many potent peptide growth factors and cytokines were identified that regulate the angiogenesis process. They include basic fibroblast growth factor (FGF-2), vascular endothelial (VEGF), insulin-like (IGF-I), hepatocyte (HGF), platelet-derived (PDGF), placenta (PlGF), pigment epithelium-derived (PEDF), transforming (TGF-β) growth factors, stromal-derived factor (STF), interleukin (IL)-6, -8, and -10, tumor necrosis factor (TNF)-α, and angiopoetins (for review, see3). Some of these factors are proangiogenic, whereas the others are antiangiogenic. Angiogenic factors act on endothelial cells through cell surface receptors that mediate downstream signaling eliciting mitogenic and motogenic cell responses. In recent years, some signaling pathways of angiogenic growth factor/cytokine receptors have been unraveled. For instance, the importance of protein kinase Cβ (PKCβ) in VEGF signaling has been emphasized and exploited for therapeutic purposes.4 At the same time, many aspects of receptor signaling that mediate angiogenic responses remain unclear.3

One of the gaps in our understanding of downstream cellular events leading to angiogenesis seems to be cleared in a publication appearing in the current issue of American Journal of Pathology (Chen et al, Am J Pathol, 2005, 165:615-624). In this paper from Scicli’s group,5 it is shown that cytochrome P450 4A (CYP4A) plays an important role in angiogenesis. Pharmacological inhibition of CYP4A activity with N-hydroxy-N′-(4-butyl-2-methylphenol) formamidine (HET0016) blocked VEGF-induced endothelial cell proliferation in vitro. Furthermore, HET0016 inhibited growth factor-induced angiogenesis in the corneal micropocket model and in the model of corneal angiogenesis induced by glioblastoma implantation. The authors identified 20-hydroxyeicosatetraenoic acid (20-HETE) as a CYP4A metabolite that likely stimulated angiogenesis. They went on to show that a 20-HETE agonist was able to induce endothelial cell mitogenesis in vitro and angiogenesis in the rat cornea model. The data clearly implicate CYP4A in angiogenesis and suggest a possible novel way to block this process by pharmacological inhibition of the enzyme. This may be very important for developing drugs aimed at counteracting pathological neovascularization as seen in solid tumors and in various proliferative retinopathies.6

The cytochrome P450 (CYP) family of enzymes has been long known to mediate detoxification of steroid hormones, vitamins, xenobiotics, and various drugs.7 Many CYPs are expressed in liver, although there are isoforms specific for other tissues, such as heart, vasculature, kidney, lung, and smooth muscle.7,8 In addition to the above-mentioned substrates, these microsomal enzymes can also metabolize arachidonic acid. This capacity makes them important players in the regulation of vascular tone, blood flow, and angiogenesis.9–13

Arachidonic acid is the precursor of pro-inflammatory eicosanoids. It is produced from membrane phospholipids by phospholipase A₂ in response to different stimuli and may be metabolized by several enzymes to yield a variety of important biological mediators. Cyclooxygenases produce prostaglandins and lipoxygenases generate leukotrienes14 from arachidonic acid. Additionally, arachidonic acid can be metabolized by CYP4A to 20-HETE or by CYP2C and CYP2W to epoxyeicosatrienoic acids (EET). These metabolites have varying effects on blood flow. Whereas EETs are vasorelaxants, 20-HETE is generally a potent vasoconstrictor.13 However, in pulmonary arteries 20-HETE may act as vasorelaxant.15 20-HETE mediates vascular tone increase through inhibition of K⁺_Ca channels, membrane depolarization and further increase in Ca²⁺.7,10,16 It is also a vascular oxygen sensor.11 Formation of 20-HETE can be regulated by nitric oxide that inactivates CYP4A.17,18 20-HETE is an important signaling molecule of hormonal systems, such as endothelin-1 and angiotensin II.7

What is the evidence for the participation of CYP4A and its product, 20-HETE, in angiogenesis? Sa et al19 obtained the first data supporting the role of CYP4A and 20-HETE in FGF-2-mediated angiogenesis. They showed that FGF-2 was able to activate phospholipase A₂ in endothelial cells. This could be due to FGF-2-induced increase in the production of arachidonic acid, in turn stimulating CYP4A. Further evidence was obtained by Amaral et al10 who showed that angiogenesis induced by electrical stimulation in skeletal muscle was dependent on 20-HETE. They made an interesting observation that a neutralizing antibody to a potent angiogenic factor, VEGF, blocked 20-HETE increases induced by stimulation. These results suggested that 20-HETE was important for induction of angiogenesis and that its formation was VEGF-dependent. Presumably, VEGF could increase 20-HETE by stimulating phospholipase A₂20 and arachidonic acid production. Most recently, Jiang et al12 used adenovirus-mediated transfer to transfect smooth muscle cells and microdissected renal arteries with CYP4A gene. The expression of transfected CYP4A elicited angiogenic activity in the arteries manifested by markedly increased endothelial cell sprouting.

The paper by Chen et al5 published in this issue of the American Journal of Pathology marks another important step in our understanding of the role of CYP4A and 20-HETE as mediators of growth factor-mediated angiogenesis. The authors have convincingly shown that CYP4A inhibitors abrogated angiogenic response to VEGF, FGF-2, and EGF in an in vivo corneal neovascularization model. This effect may of course be explained in different ways. One may postulate a direct role for CYP4A-produced 20-HETE as a signaling molecule downstream of the angiogenic growth factor receptors. As discussed above, there is experimental evidence in support of this mechanism. An alternative possibility is that the growth factors triggered the production of arachidonic acid metabolites that are pro-inflammatory. These products including 20-HETE could mediate recruitment of polymorphonuclear leukocytes and later, macrophages. These cells produce and secrete VEGF, and are thought to be the main mediators of corneal neovascularization.21 This would explain why CYP4A inhibitors could abrogate the effects of different growth factors with similar efficiencies.5 The latter possibility does not exclude a direct effect of 20-HETE, which may be induced by macrophage-produced VEGF. It would be interesting to sort out these issues in future experiments.

The paper by Chen et al5 also describes the use of glioblastoma cells implanted in the cornea. Tumor cells elicit a strong angiogenic response that is inhibited by CYP4A blockers. Because glioblastoma cells secrete angiogenic growth factors, there is reason to believe that they may induce corneal angiogenesis through the action of these factors, eg, VEGF. Since the corneal model has some limitations related to its inflammatory nature,21 it may be important to further examine the effects of CYP4A inhibitors using tumors grown in other sites, eg, intracranially or subcutaneously.

The development of pathological neovascularization is often associated with hypoxia/ischemia, as observed in malignant tumors and proliferative retinopathies. Hypoxia is known to stimulate important angiogenic mediators including VEGF. This occurs through activation of transcription factor HIF-1α that increases VEGF expression.22,23 The work by Chen et al5 now has established a link between CYP4A and VEGF, and it would be interesting to examine the effects of hypoxia/ischemia on this cytochrome and on its angiogenic product, 20-HETE, in conditions promoting angiogenesis. The available data in this area remain controversial. Some members of the CYP family are upregulated by hypoxia. CYP3A6 and CYP4B1, for instance, are directly increased by hypoxia through HIF-1α mechanism.24,25 Moreover, corneal CYP4B1 is increased by hypoxia in parallel with VEGF.26 Interestingly, blockade of phospholipase A₂ during hypoxia in retinal endothelial cells could inhibit stimulation of VEGF production.27 However, in the rat renal artery and vein ischemia-reperfusion injury model, CYP4A and 20-HETE are both decreased.28 These data pertain to 20-HETE measurements not during ischemia but after 3 hours or more of reperfusion, which may have influenced the results. In contrast, recent data in the heart ischemia-reperfusion model show that levels of both CYP4A and 20-HETE increase in the ischemic phase and shortly thereafter. Moreover, exogenous 20-HETE significantly increases infarct size, possibly by exerting its vasoconstricting effect.29 Taken together, there is a need for expanding these studies using other hypoxic/ischemic models with developing neovascularization, for example, models of proliferative retinopathy or experimental tumor growth.

Critical questions to answer in future experiments concern the mechanism by which CYP4A participates in angiogenesis through production of 20-HETE, and the interaction of CYP system with growth factors in the angiogenic process. In other words, what are the downstream signaling events where growth factors, eg, VEGF, converge with the CYP system? Angiogenic growth factors signal through their surface receptors most of which belong to the receptor-type tyrosine kinase class.3,4,22 Some growth factors including VEGF have more than one receptor. Signaling pathways of these receptors may be different, which may ensure fine-tuning of the system depending on the tissue needs.22 Although there is some convergence in signaling pathways of various angiogenic growth factor receptors, they generally use different intermediates, which forms the basis for growth factor synergy.30 Such a synergy makes it difficult to counteract the action of several growth factors together for therapeutic purposes against pathological neovascularization. Therefore, inhibitors of key downstream signaling molecules may prove useful for developing more efficient drugs against unwanted neovascularization.

It was shown recently that contraction of small coronary arteries by 20-HETE depends on the activation of Rho family of small GTPases.31 At the same time, Rho is needed for activation of serum response factor, a transcription factor that plays a critical role in VEGF signaling.32 Rho system also mediates induction of endothelial and tumor cell migration by FGF-2, IGF-I, and PDGF.33–35 This is the first point of convergence between CYP4A system and angiogenic growth factors. Moreover, both 20-HETE and growth factors are involved in signaling via mitogen-activated protein (MAP) kinase-Ras pathway that is important for growth factor-mediated angiogenesis.3,22,36–38 Finally, 20-HETE can activate PKC,39 which is an important signaling mediator of angiogenic growth factors.4 In the paper by Chen et al5 in this issue of American Journal of Pathology, it was shown that CYP4A inhibitors abrogated the mitogenic action of VEGF on the endothelial cells. The analysis of the literature presented above allows one to assume that such inhibitors would also block the action of other angiogenic growth factors. CYP4A and its product, 20-HETE, are, therefore, emerging as novel mediators of angiogenesis that are closely linked to signaling pathways of major angiogenic growth factors. The ability of CYP4A to activate important angiogenic signaling pathways through 20-HETE makes it a very attractive target for future antiangiogenic therapies. An important advantage of CYP4A inhibitors would be their ability to counteract the action of several growth factors together for a more complete blocking of the angiogenic cascade.