Heme Oxygenase-1

Abstract

HomeCirculationVol. 114, No. 20Heme Oxygenase-1 Free AccessResearch ArticlePDF/EPUBAboutView PDFView EPUBSections ToolsAdd to favoritesDownload citationsTrack citationsPermissions ShareShare onFacebookTwitterLinked InMendeleyReddit Jump toFree AccessResearch ArticlePDF/EPUBHeme Oxygenase-1A Novel Drug Target for Atherosclerotic Diseases? Roland Stocker, PhD and Mark A. Perrella, MD Roland StockerRoland Stocker From the Centre for Vascular Research, School of Medical Sciences, University of New South Wales, and Department of Haematology, Prince of Wales Hospital, Sydney, Australia (R.S.), and Pulmonary and Critical Care Division, Brigham and Women’s Hospital and Harvard Medical School, Boston, Mass (M.A.P.). Search for more papers by this author and Mark A. PerrellaMark A. Perrella From the Centre for Vascular Research, School of Medical Sciences, University of New South Wales, and Department of Haematology, Prince of Wales Hospital, Sydney, Australia (R.S.), and Pulmonary and Critical Care Division, Brigham and Women’s Hospital and Harvard Medical School, Boston, Mass (M.A.P.). Search for more papers by this author Originally published14 Nov 2006https://doi.org/10.1161/CIRCULATIONAHA.105.598698Circulation. 2006;114:2178–2189Schmid and coworkers1 were the first to report on the presence of heme oxygenase (HO) in liver microsomes capable of degrading heme to bilirubin, and this activity was subsequently dissociated from cytochrome P-450.2,3 HO catalyzes the first and rate-limiting step in the oxidative degradation of heme (Fe-protoporphyrin-IX) to carbon monoxide (CO), ferrous iron (Fe2+), and biliverdin-IX (Figure 1). The enzyme binds heme in a 1:1 molar complex, and HO-bound heme acts as prosthetic group and substrate. The reaction requires 3 molecules of molecular oxygen (O2) per heme molecule oxidized and reducing equivalents derived from nicotinamide adenine dinucleotide phosphate or nicotinamide adenine dinucleotide (reduced form) and transferred to the oxygenase via nicotinamide adenine dinucleotide phosphate:cytochrome P-450 reductase. Regiospecific oxidation of heme is achieved in a stepwise reaction, with α-meso-hydroxyheme and verdoheme as intermediates, and the dissociation of CO followed by that of Fe2+.4 The release of biliverdin from HO is accelerated by biliverdin reductase, which reduces the green pigment to bilirubin-IX,4 which is then excreted into bile as the glucuronic acid conjugate. Download figureDownload PowerPointFigure 1. Oxidative metabolism of heme by HO and biliverdin reductase, giving rise to CO, iron, biliverdin, and bilirubin.Originally, the interest in HO was related to its well-established function in heme catabolism and the turnover of erythrocytes. For many years, CO and bilirubin were regarded as toxic waste byproducts of the HO reaction, but in 1987, a potential beneficial role of bilirubin was proposed5 based on the in vitro antioxidant activities of the pigment. Over the last decade, however, the interest in HO has shifted greatly from a metabolic to a protective function of the enzyme in a variety of conditions associated with cellular stress and pathologies, and this has been the subject of excellent recent reviews.6,7The relationship of HO to atherosclerotic vascular disease was suggested initially in 1994 by an observational study reporting that low serum concentrations of bilirubin are associated with increased risk of coronary artery disease.8 Since then, the implied cardioprotective role of HO has been developed and substantiated significantly in experimental models of atherosclerotic vascular disease, including atherosclerosis,9 intimal hyperplasia,10 and myocardial infarction.11 There is now good evidence that induction of the inducible isozyme HO-1 by a broad spectrum of physical and chemical agents leads to several vascular cell-specific protective activities in the setting of inflammatory atherosclerotic diseases. Despite this recent work, however, the precise relationship between HO-1 and atherosclerosis remains unknown. In the present study, we review recent progress in our understanding of the protective mechanisms of HO-1 in atherosclerotic vascular disease and highlight areas of insufficient knowledge that require additional work in the future.HO-1: Critical Contribution to Iron HomeostasisA discussion of the role of HO-1 in atherosclerotic vascular disease should begin with a brief review of the established role of the enzyme in iron homeostasis. HO-1 in macrophages of the reticuloendothelial system plays a key role in the reuse of the iron essential for erythropoiesis. In adult humans, the majority of Fe-protoporphyrin-IX degraded by HO-1 is derived from hemoglobin, resulting in a daily production of ≈28 mg Fe2+ or nearly 1% of the total body iron store.12 This iron is returned almost quantitatively to the circulation where it is bound tightly by the plasma glycoprotein transferring, which transports the iron to the bone marrow where it is used to synthesize hemoglobin in developing erythroid cells. Indeed, mice lacking functional HO-1 develop an anemia that is associated with abnormally low levels of serum iron and the accumulation of iron in liver and kidneys.13 This function of HO-1 in iron homeostasis implies that at least in macrophages, HO-1 activity is associated with the release of cellular iron.In the context of atherosclerotic vascular diseases, it is not clear whether a role of HO-1 in iron reuse extends to macrophages, or indeed to other cells, in the vessel wall and/or to heme derived from sources other than hemoglobin. Ferris et al14 reported decreased iron efflux from fibroblasts deficient in HO-1 compared with control cells and transfection of an epithelial cell line with HO-1 to augment iron release without changes to the cellular iron storage protein ferritin. In apparent contrast, an earlier study reported Ultraviolet A radiation induced HO-1 activity to co-induce ferritin and that ferritin to store released iron.15 Homeostasis of cellular iron is accomplished by coordinated regulation of iron import and export proteins.16 Export of iron to transferrin is generally thought to occur via the plasma membrane transporter ferroportin, and this requires oxidation of Fe2+ to Fe3+ by ceruloplasmin.16 In vitro vascular endothelial cells express ferroportin in response to high glucose17 and proinflammatory cytokines.18 It is not known, however, whether this extends to endothelial and/or vascular smooth muscle cells in vivo, whether ferroportin expression is responsive to increased cellular HO-1 activity, or whether ceruloplasmin is present in atherosclerotic vessels. What is known is that arteries with atherosclerotic lesions express HO-119 and contain more iron than corresponding healthy vessels.20 Because iron can conceivably affect atherosclerotic disease, investigations of the role of HO-1 on iron content in diseased arteries are warranted. In addition, the possibility that non–bone marrow HO-1 supplies iron for growth of nonerythroid stem cells, such as endothelial progenitor cells implicated in the repair of vascular injury relevant to atherosclerotic vascular disease, deserves investigation.HO-1: Functions Beyond Iron HomeostasisThere are 2 genetically distinct isozymes of HO: the inducible HO-1 and a constitutively expressed form, HO-2. As indicated, HO-1 is expressed most strongly in tissues involved in erythrocyte or hemoglobin metabolism, whereas in most other tissues, HO-1 typically occurs at low basal levels but responds rapidly by transcriptional activation to diverse stimuli. In contrast, HO-2 is strongly expressed in testes but also is present ubiquitously in other tissues, including the vasculature, and it does not generally respond to transcriptional activation. HO-1 is classically characterized as a protein associated with the endoplasmic reticulum,1 although recent evidence suggests that the enzyme also is present in caveolae of the plasma membrane of endothelial cells7 and in liver mitochondria of heme-treated rats.21 A function of nonmicrosomal HO-1 currently is not established, although the reported colocalization of biliverdin reductase21 suggests that the enzyme is also metabolically active at these sites and hence may contribute to a localized regulation of heme protein turnover.HO-1: Protection Against Atherosclerotic DiseaseHO-1 is a cytoprotective enzyme, and its induction commonly occurs in the setting of increased cellular stress to help maintain physiological homeostasis. HO-1 is induced by a number of stressors,7 and one may predict that risk factors for the development of coronary heart disease and other cardiovascular disease processes will mediate HO-1 expression. In fact, this is the case because increased blood pressure22 and altered laminar flow in blood vessels,23 advanced glycation end products,24 cigarette smoke,25 oxidized lipids,26 and a multitude of systemic inflammatory processes7 lead to increased cellular HO-1 expression. Moreover, this constellation of processes that lead to HO-1 induction may suggest a role for HO-1 in mediating the cardiovascular disease processes associated with obesity and metabolic syndrome,27 which have become significant public health problems. Pragmatically determining that upregulation of HO-1 plays a protective role in atherosclerotic disease would require investigation into patients who either lack HO-1 or do not robustly express HO-1 in the presence of risk factors for coronary heart disease. With the availability of these patients, investigations into the prevalence of coronary heart disease would provide critical insight into the physiological role of HO-1.HO-1 deficiency is very rare in humans; however, an autopsy report from a 6-year-old boy with HO-1 deficiency revealed hyperlipidemia associated with fatty streaks and fibrous plaques in the aorta.28 The concept that HO-1 may be causally related to cardiovascular diseases in humans also has been suggested by studies assessing polymorphisms in the 5′-flanking sequence of the human HO-1 gene.29 Different polymorphisms have been identified in the HO-1 promoter, with the most studied being (GT)n dinucleotide-length polymorphism. In general, the belief is that compared with longer (GT)n repeats, shorter (GT)n repeats have higher transcriptional activity and thus higher expression levels. For example, in a Chinese population of type II diabetic patients, long (GT)n repeats (≥32 GTs) were associated with increased risk for coronary artery disease.30 Conversely, in a Japanese population of patients with significant risk factors (hyperlipidemia, diabetes, and smoking) for coronary artery disease, shorter (GT)n repeats (<27 GTs) were associated with less disease.31 However, this concept does not appear to hold for white patients with myocardial infarctions or stable coronary artery disease.32 In contrast, studies assessing the association of length of (GT)n repeats with the risk of restenosis after percutaneous transluminal angioplasty have been more consistent across ethnic backgrounds.29,33 In addition to (GT)n dinucleotide-length polymorphism, a single nucleotide polymorphism in the HO-1 promoter, T(-413)A, correlated with a reduced incidence of ischemic heart disease in a Japanese population.34 In this study, the AA genotype (A on each allele) at position −413 was less likely to be associated with myocardial infarction and angina pectoris, and cell culture experiments suggested the AA genotype to have significantly higher basal promoter activity that was independent of the length of (GT)n repeats.34 These studies advocate the potential role of HO-1 gene regulation in atherosclerotic disease processes. Truly understanding HO-1 as a modulator of atherosclerosis, however, requires further in-depth investigation into this specific disease process.Atherosclerosis is an inflammatory disease in which lipid deposition in the arterial wall, resulting from elevated levels of plasma cholesterol, is central to lesion development. This process involves the uptake of modified low-density lipoprotein (LDL) by macrophages and is associated with a state of heightened oxidative stress and damage.35 As atherosclerotic lesions progress, migration and proliferation of smooth muscle cells and deposition of fibrous tissue lead to an advanced, complicated lesion. Wang and colleagues19 previously demonstrated that expression of HO-1 is prominent in endothelial cells, macrophages, and foam cells in human and animal atherosclerotic lesions. Ishikawa and colleagues36,37 showed that inducers of HO-1 reduce lesion size in Watanabe heritable hyperlipidemic rabbits36 and LDL-receptor–deficient mice.37 Using adenovirus-mediated gene transfer of HO-1, Juan and colleagues38 demonstrated that selective overexpression of HO-1 decreases lesion size in apolipoprotein E–deficient (ApoE−/−) mice.There are additional lines of evidence in support of HO-1 playing a protective role in atherosclerotic lesion formation (the Table). Long-term inhibition of the HO activity using metalloporphyrins promotes lesion formation in LDL-receptor–deficient mice37 and rabbits.36 A potentially serious limitation of these studies, however, is that metalloporphyrins are not selective for HO-1 or even the HO system,39 particularly when used at high doses. Thus, the use of HO-1–deficient (HO-1−/−) mice was important to specifically establish a protective role of endogenous HO-1 in atherosclerosis. For this, Yet et al9 subjected ApoE−/− mice and mice deficient in ApoE and HO-1 (ApoE−/−HO-1−/−) to a Western diet for 8 weeks and then analyzed them for the development of atherosclerosis. Despite similarly elevated total plasma cholesterol, ApoE−/−HO-1−/− mice had larger and more advanced lesions than ApoE−/− mice. The lesions in ApoE−/−HO-1−/− mice were complicated with fibrous caps, comparable to plaques seen in ApoE−/− mice on a Western diet for longer periods of time (12 weeks). These results, in conjunction with the HO-1 gene transfer studies, provide strong evidence for a beneficial effect of HO-1 on experimental atherosclerosis. Support for a Protective Role of HO-1 Against Atherosclerosis and Related DiseasesEvidenceReferencePresence of HO-1 in human and experimental atherosclerotic lesions19High serum bilirubin levels associated with decreased risk of coronary artery disease8Low serum bilirubin levels associated with increased risk of coronary artery disease7HO-1 induction/upregulation decreases experimental atherosclerosis36–38HO-1 inhibition/downregulation increases experimental atherosclerosis36, 37Increased atherosclerosis in ApoE−/−HO-1−/− vs ApoE−/−HO-1+/+ mice9Increased vein graft stenosis in ApoE−/−HO-1−/− vs ApoE−/−HO-1+/+ mice9HO-1 overexpression decreases intimal thickening in experimental restenosis10, 40, 46HO-1 deficiency/downregulation increases intimal thickening in experimental restenosis10, 120Cardiac-specific HO-1 overexpression protects against experimental ischemia/reperfusion injury11, 85Lung-specific overexpression of HO-1 ameliorates pulmonary vascular hypertension in a model of chronic hypoxia86When evaluating the potential for HO-1 as a therapeutic target, we should also assess the ability of HO-1 to ameliorate vascular complications after coronary artery bypass surgery or percutaneous transluminal angioplasty, namely vein graft failure and restenosis. Although coronary artery stenting is being used increasingly to treat patients with obstructive atherosclerotic lesions, coronary artery bypass graft surgery remains an important treatment for multivessel disease. Autologous vein grafts provide a convenient conduit for bypass graft surgery, and although early grafts occlude as a result of thrombotic events, late-onset graft occlusion is the result of intimal thickening and superimposed atherosclerosis. In experimental vein graft stenosis, HO-1−/− mice develop more robust neointima consisting of smooth muscle cells 10 days after surgery than wild-type mice,9 suggesting that HO-1 plays a protective role in the pathophysiology of not only atherosclerosis but also vein graft failure.As indicated, studies assessing human HO-1 polymorphisms in disease suggest that higher levels of HO-1 expression are associated with a reduced risk for restenosis after percutaneous transluminal angioplasty.29 This concept has been confirmed in animal models of restenosis. Thus, adenovirus-mediated transfer of the HO-1 gene reduces intimal thickening in balloon-injured pig femoral10 and rat carotid arteries.40 Conversely, arterial lesions induced by wire injury are more severe in HO-1−/− than wild-type mice,10 again establishing a protective role for HO-1 in models of injury-induced vascular disease. Together, the above results strongly support the notion that HO-1 plays a protective role in experimental atherosclerotic vascular disease. Although the relevance of these findings to the human disease has yet to be established, the overall outcome of the preclinical studies carried out to date clearly points to HO-1 as a potential novel target for therapy.Induction of HO-1Signaling PathwaysThe gene coding for HO-1 is highly regulated, and in most cell types, HO-1 is expressed in response to numerous stimuli.7 Interestingly, the overwhelming majority of stimuli cause a rapid and temporary induction of the HO-1 gene; only a few mediators that suppress HO-1 are known.41 Regulation of the HO-1 gene is predominantly at the transcriptional level.7 Multiple enhancer regions have been identified in the 5′-flanking sequence of the HO-1 gene, and depending on the specific stimulus and the cell type involved, various transcription factors will interact with their cognate DNA binding domains in the HO-1 promoter to regulate gene transcription. One of the first pathways to link extracellular stimuli to activation of HO-1 is the mitogen-activated protein kinase (MAPK) pathway.7 The MAPKs are a family of serine-threonine protein kinases that regulate many cellular events, including responses to environmental stimuli. The MAPK pathway encompasses 3 signaling cascades (ie, extracellular signal regulated kinases 1/2, c-Jun-N-terminal kinase, and p38-MAPK) that phosphorylate downstream targets. In the case of HO-1 transcriptional events, downstream targets of these kinase cascades are transcription factors involved in HO-1 gene regulation. Beyond the MAPKs, other kinases that have emerged as mediators of HO-1 gene regulation include phosphatidylinositol 3-kinase and protein kinases A, C, and G.7We have already mentioned that genetic polymorphisms in the human HO-1 gene promoter modulate the level of transcriptional activity and the magnitude by which HO-1 responds to a pathophysiological stimulus. These polymorphisms are associated with an altered risk profile for cardiovascular disease.29 To date, little is known about the interaction of nuclear proteins within these polymorphic regions. However, a number of excellent reviews have covered the complex topic of HO-1 gene regulation by transcriptional events involving classic DNA binding domains and their associated transcription factors, as well as cell type and species specificity (see Reference 7 and references therein). Therefore, for the purposes of the present review, we focus on selected transcriptional events activated by proatherogenic stimuli.There are several critical regulatory domains present in a 10-kb region of the 5′-flanking sequence of the HO-1 gene. Two of the most highly studied enhancer regions, called E1 and E2, contain stress-response elements (StRE) that are conserved between human, mouse, and rat genes and that structurally resemble the antioxidant response element (ARE) and the AP-1/TPA, Maf, and cyclic adenosine monophosphate (cAMP) response elements.7 Nuclear proteins that bind to these elements belong to the basic-leucine zipper (bZIP) superfamily of transcription factors and include AP-1 (Fos/Jun), CREB/ATF, Maf, Cap’n’collar-bZIP (Nrf, Bach), and biliverdin reductase.42 These nuclear proteins bind to StRE as homodimers or heterodimers (within or between families). Because of the similarity of StRE with consensus AP-1 binding sites, complexes such as c-Fos/c-Jun heterodimers were the focus of initial HO-1 gene regulation studies.43 However, more recent studies have emphasized additional bZIP family members as important mediators of the StRE response. For example, during exposure to heme or heavy metals, cytosolic Nrf2 is stabilized and then translocates to the nucleus, where it binds to consensus StRE/ARE binding sites and forms heterodimers with Maf proteins. Because Maf proteins do not have transactivation domains, Nrf2 drives transcriptional activity. The heme binding protein Bach1 is another binding partner for Maf proteins at StRE/ARE sites. Because Bach1 also lacks a transactivation domain, its heterodimerization with Maf proteins results in repression of HO-1 expression. Heme, the substrate for HO-1 enzyme activity, abrogates the repression of Bach1 by inhibiting its binding to DNA.44 This inhibition of Bach1 binding allows activators of the HO-1 gene such as Nrf2 to bind with Maf proteins at StRE/ARE sites45 and thus provides a feedback loop for the regulation of HO-1 expression. The potential importance of Bach1 in cardiovascular disease is supported by the recent finding that mice deficient in Bach1 have increased expression of HO-1 and less intimal proliferation after vascular (cuff) injury.46In addition, enhancer regions outside E1 and E2 also have been described in the transcriptional regulation of HO-1.7 For example, a balance of Ets protein family members, including transcriptional activators such as Ets-1 and Ets-2, and the repressor Elk-3 regulates the overall transcriptional response of HO-1 during an inflammatory stimulus by binding to domains far downstream from the E1 and E2 enhancer regions.47 This demonstrates the complexity of the system and underscores the importance of differences in HO-1 gene regulation by different stimuli and in different cell types.Response to Physical StressIncreased synthesis of HO-1 protein in response to physical and chemical stress occurs commonly and in most tissues examined.7 This is not surprising when we consider that HO-1 belongs to a larger family of stress proteins in which transcriptional regulation responds to altered environmental conditions. In fact, HO-1 was initially referred to as 32-kDa heat shock protein because of its transcriptional responsiveness to hyperthermia,48 although the protein shares little amino acid homology with heat shock proteins, nor does it display protein chaperone activity. More recently, it has been reported that in cultured vascular smooth muscle cells, endoplasmic reticulum stress increased HO-1 mRNA and protein via the ARE, and this was associated with cell survival.49 The “endoplasmic reticulum stress response” constitutes a general response to endoplasmic reticulum–associated stress such as unfolded proteins, glucose starvation, and disruption of intracellular calcium homeostasis. Endoplasmic reticulum–initiated cell death pathways are increasingly recognized as playing important roles in several diseases, including ischemia/reperfusion injury, heart disease, and diabetes.In addition to these general types of physical stress, HO-1 responds directly to vascular injury such as that which occurs during angioplasty.10 This response appears to be biphasic, with an initial decrease in endogenous HO-1 followed by an increase to above baseline and then a return to background levels of HO-1 expression.10,50 Furthermore, the temporal and spatial pattern of HO-1 expression appears to be similar to that of the G1 cyclin-dependent kinase inhibitors p27Kip1 and p21Cip,10, consistent with the notion that HO-1 functions upstream of and participates in the regulation of smooth muscle cell proliferation in injured vessels.Response to Chemical StressAs reviewed previously,7 HO-1 regulation responds to a large and broad spectrum of chemical stresses, including agents that cause oxidative stress or diminish oxygen, thiol-reactive agents, heavy metals, electrophilic polyphenolic compounds, inflammatory mediators, growth factors, hormones, and environmental pollutants. The following discussion is limited to agents directly relevant to atherosclerotic vascular disease.As we have learned already, atherosclerotic vascular disease is characterized by a state of heightened oxidative stress that includes the presence of oxidized lipids,35 and HO-1 is expressed in atherosclerotic lesions.19 In vitro studies have demonstrated that hydrogen peroxide,51 linoleic acid hydroperoxides,52 oxidized phospholipids53 and LDL54 induce HO-1 in different cell types. When investigated, this induction was shown to be mediated via StRE/ARE in murine and human cells.53 However, whether increased levels of hydrogen peroxide and/or oxidized lipids directly induce HO-1 in atherosclerotic lesions is unclear at present. Hepatic HO activity is unaltered in mice deficient in glutathione peroxidase-1 or selenoprotein P (ie, enzymes involved in the cellular metabolism of hydrogen peroxide and lipid hydroperoxides) but increased in mice deficient in thioredoxin reductase,55 an enzyme that affects several redox-related cellular processes.35Response to Therapeutic AgentsAn increasing number of therapeutic agents have been reported to induce HO-1, in direct support of the notion that HO-1 may represent a novel drug target for atherosclerotic disease. These therapeutic agents include statins, rapamycin, paclitaxel, nitric oxide (NO), aspirin, and probucol. At micromolar concentrations, several statins dose dependently induce HO-1 in the human epithelial cell line ECV30456 and vascular smooth muscle cells.57 In addition, oral administration of statins at 100 mg/kg body weight resulted in a statin- and tissue-specific increase in HO-1 mRNA and protein in mice.58 Because this induction was associated with beneficial cellular effects such as increased resistance to oxidative stress and inhibition of smooth muscle cell proliferation,57 it was speculated that this novel activity may contribute to the pleiotropic and antiatherogenic actions of statins. While attractive, additional studies are required to establish whether statins at concentrations pharmacologically relevant to humans induce HO-1 and, if so, whether this indeed results in significant protection.Rapamycin (sirolimus) is a macrolide antibiotic with potent immunosuppressive properties that inhibits cell proliferation by blocking the progression of cells from the G1 to the S phase of the cell cycle. In experimental models, rapamycin has antiproliferative properties against vascular endothelial and smooth muscle cells,59 and it reduces the fibroproliferative response to vascular injury in vivo.60 This has led to its successful clinical application in the form of sirolimus-eluting stents for the treatment of in-stent restenosis. Rapamycin induces HO-1 expression in primary human endothelial and smooth muscle cells,61 an activity not shared by other immunosuppressive agents such as cyclosporin A. In addition, inhibition of HO activity by tin protoporphyrin resulted in a loss of the antiproliferative activity of rapamycin,61 and smooth muscle cells from HO-1−/− mice were refractory to growth inhibition by rapamycin.62 Similarly, rapamycin was recently reported to induce HO-1 in the lungs of rats and to inhibit the development of monocrotaline-induced pulmonary hypertension, with the protective effect being blocked by co-treatment of the animals with tin protoporphyrin.62 Collectively, these findings suggest that the antiproliferative action of rapamycin may be modulated, at least in part, by its actions on HO-1. Similar to rapamycin, paclitaxel, which induces apoptosis, interferes with normal function of microtubule growth, and is used to treat in-stent restenosis, induces HO-1 in vascular smooth muscle cells.63Nitroglycerin and long-acting nitrates are used commonly in cardiovascular medicine for various anginal syndromes and congestive heart failure and in patients with left ventricular dysfunction. The mechanisms for relief of myocardial ischemia by nitrates are intimately linked to the formation of NO within vascular smooth muscle cells, where it stimulates the enzyme guanylate cyclase, resulting in increases in guanosine 3′,5′-cyclic monophosphate (cGMP) and vasodilation. In addition to regulating vascular tone, it is now increasingly recognized that NO modulates a variety of important physiological activities related to vascular homeostasis, including platelet aggregation, leukocyte trafficking, cell signaling, and the migration and growth of endothelial and smooth muscle cells. NO donors64 and pure, gaseous NO65 induce HO-1. This induction is due, in part, to the increased stability of HO-1 mRNA65 and extends to many different forms of NO (eg, NONOates, S-nitrosothiols, sodium nitroprusside, and pentaerythritol tetranitrate) and vascular cells.7 Activation of soluble guanylate cyclase and enhanced formation of cGMP have commonly been associated with HO-1 induction by NO and related compounds.7 However, this is not the case in all situations, just as not all NO-related compounds induce HO-1. Perhaps most notably, induction of HO-1 in human embryonic lung fibroblasts by gaseous NO was reported to be independent of the guanylate cyclase/cGMP pathway,65 and isosorbide dinitrate, unlike pentaerythritol tetranitrate, was unable to induce HO-1 in endothelial cells.66 The cellular generation of NO also may be responsible for, or contribute to, the ability of therapeutically used drugs such as aspirin67 to induce HO-1.Probucol, a rarely used cholesterol-lowering drug, has been reported to inhibit atherosclerosis in carotid artery68 and neointimal proliferation after coronary angioplasty with69 and without70 stent deployment in