Oxidative Stress and Obesity, Diabetes, Smoking, and Pollution: Part 3 of a 3-Part Series

Abstract

Oxidative stress occurs whenever the release of reactive oxygen species (ROS) exceeds endogenous antioxidant capacity. In this paper, we review the specific role of several cardiovascular risk factors in promoting oxidative stress, namely diabetes, obesity, smoking, and excessive pollution. Specifically, the risk of developing heart failure is higher in patients with diabetes or obesity, even with optimal medical treatment, and the increased release of ROS from cardiac mitochondria and other sources likely contributes to the development of cardiac dysfunction in this setting. Here, we explore the role of different ROS sources arising in obesity and diabetes, and the impact of excessive ROS production on the development of cardiac lipotoxicity. In parallel, contaminants in the air that we breathe pose a significant threat to human health. This paper provides an overview of cigarette smoke and urban air pollution, considering how their composition and biological effects have detrimental effects on cardiovascular health.

Introduction

Oxidative stress is a pervasive aspect of cardiovascular disease (CVD) and occurs whenever the release of reactive oxygen species (ROS) exceeds endogenous antioxidant capacity. Although physiological levels of ROS are important signaling molecules, prolonged exposure or inappropriate subcellular localization of ROS can have detrimental effects. Previously in this review series, we covered the basic biology of oxidative stress, telomeres, and telomere dysfunction in Part 1, and the role of oxidative stress in both heart failure and vascular disease in Part 2. Here in Part 3, we address the role of several specific and important risk factors, and how they increase cardiovascular risk via increased oxidative stress, namely diabetes, obesity, smoking, and excessive pollution.

Obesity, Diabetes, and Oxidative Stress

Metabolism of the Obese and the Diabetic Heart

The metabolic phenotypes of the diabetic and the obese heart have many similarities: fatty acid uptake and fatty acid oxidation (FAO) are increased, as are levels of intramyocardial and circulating triacylglycerol (TAG) and circulating free fatty acids (FFAs), and glucose uptake and glucose oxidation are reduced (1). Despite reduced glucose uptake, there is increased flux through accessory pathways of glucose metabolism, such as the polyol pathway or the hexosamine biosynthetic pathway (2). In the healthy heart, utilization of FFAs and glucose is well-balanced, and enables the heart to switch between energy sources according to their availability and in response to environmental stimuli. This provides the heart with a high degree of flexibility in substrate utilization. The inability of the obese or diabetic heart to appropriately use glucose results in a reliance on FAO and reduced metabolic flexibility. The concurrent cardiac inefficiency is related to increased mitochondrial uncoupling induced by fatty acids, as well as the low oxygen (O₂) utilization efficiency of FAO (3), resulting in decreased adenosine triphosphate (ATP) production despite fuel oxidation. Metabolic alterations observed in the obese or diabetic heart, such as increased FAO, mitochondrial dysfunction, glucose autoxidation, impaired polyol metabolism, or increased hexosamine metabolism, can cause increased ROS release. Accordingly, metabolite-generated ROS play a major role in the development of various diabetes-related cardiovascular complications. The major metabolic changes in the obese and diabetic heart are summarized in Figure 1.

Figure 1. Metabolic Changes in the Obese and Diabetic Heart.

Overnutrition results in increased levels of circulating fatty acids and glucose. Fatty acids induce an activation of cardiac peroxisome proliferator-activated receptor alpha (PPARα), which enhances the expression of genes that control fatty acid uptake and oxidation as well as glucose oxidation (PDK4). The consequent switch from glucose to free fatty acid oxidation results in metabolic inflexibility and a decrease in cardiac efficiency. The accompanying mitochondrial dysfunction contributes to increased ROS production, uncoupling of the ETC and reduced ATP production. Accumulation of toxic lipids and lipid derivatives such as ceramides can directly alter cellular structures and induce cardiomyocyte dysfunction and cell death, collectively called lipotoxicity. Fatty acids also contribute to the activation of serine/threonine kinases which inhibit insulin signaling via insulin receptor substrate (IRS), finally resulting in reduced glucose uptake. Toxic effects of glucose (glucotoxicity) include formation of advanced glycation end-products (AGEs), a group of modified proteins and/or lipids with damaging potential, as well as increased activity of the hexosamine and polyol pathways. ATP = adenosine triphosphate; ETC = electron transport chain; PDK4 = pyruvate dehydrogenase kinase 4; ROS = reactive oxygen species.

Role of Mitochondrial ROS in Obesity and Diabetes

Across various organs, including the heart (reviewed in (4)), impaired mitochondrial respiration and changes in mitochondrial morphology have been consistently observed in insulin resistance and type 2 diabetes mellitus (DM). Patients with type 2 diabetes have significantly lower cardiac phosphocreatine (PCr)/ATP ratios (5,6), decreased cardiac oxidative capacity, and increased mitochondrial ROS emission (7). Obesity results in disturbed mitochondrial biogenesis and function (respiratory chain complex I), which occurs prematurely in younger patients with obesity (8). Preclinical studies suggest early ROS up-regulation during diabetes-induced cardiac remodeling, but analogous prospective studies in patients with diabetes are lacking. Mitochondrial dysfunction, increased mitochondrial ROS release, and mitochondria-dependent cell death have been reported in the diabetic human heart (7,9–12). Mitochondrial dysfunction appears to be present mainly in cardiac subsarcolemmal, but not in interfibrillar mitochondria of patients with type 2 diabetes (11). Interestingly, impaired mitochondrial function and contractile dysfunction were observed in patients with diabetes, but not in patients with obesity (12). Anderson et al. (7) also demonstrated increased levels of 4-hydroxynonenal– and 3-nitrotyrosine–modified proteins, together with a reduction in the ratio of reduced to oxidized glutathione in diabetic human hearts, indicating persistent oxidative stress in these samples.

Under physiological conditions, electron transport to O₂ is tightly coupled to ATP synthesis. ROS generation by mitochondria occurs through electron leakage from the electron transport chain (ETC) due to a decreased rate of mitochondrial phosphorylation. Normally, <1% of total oxygen consumption leaks from the ETC to generate ROS. Nevertheless, mitochondria are the primary source of cellular ROS production in cardiomyocytes, and increased mitochondrial ROS are a major cause of oxidative stress associated with DM. The hyperglycemia-induced overproduction of superoxide (O₂⁻) by the ETC is recognized as a major cause of the clinical complications associated with diabetes and obesity (13). Indeed, attenuation of mitochondrial ROS release results in completely preserved insulin sensitivity despite a high-fat diet (14,15). Accordingly, mice with overexpression of a mitochondrially-targeted catalase show reduced ROS release and do not develop insulin resistance despite a high-fat diet (14). Recently, it was also shown that a mitochondria-targeted antioxidant prevents insulin resistance and diastolic dysfunction, suggesting that mitochondrial oxidative stress may be involved in both conditions (16).

Hyperglycemia increases ETC flux, resulting in mitochondrial hyperpolarization and O₂⁻ generation (17). Hyperglycemia-induced mitochondrial overproduction of ROS activates 4 major pathways involved in the pathogenesis of cardiovascular complications, including the increased production of advanced glycation end-products (AGEs), the polyol pathway, the hexosamine pathway, and protein kinase C-dependent signal transduction (13). Mitochondrial ROS production can be blocked by certain ETC inhibitors or uncoupling agents (17). Uncoupling proteins (UCPs) such as UCP2 and UCP3, the major human cardiac UCP isoforms (18), can dissipate the electrochemical gradient and thus attenuate ROS production (19) at the expense of decreased cardiac efficiency. Up-regulation of cardiac UCP2 and/or UCP3 has been reported in some, but not in all studies using preclinical models of obesity and DM (20).

FFAs may increase ROS generation in mitochondria functioning in the forward electron transport mode by slowing down the rate of electron flow through complexes I and III of the ETC due to interactions within the complex subunit structure (21). Furthermore, their impact on the electron flow between complexes III and IV can increase ROS generation due to the release of cytochrome c from the inner mitochondrial membrane (21,22). However, a minor depolarization of the inner mitochondrial membrane was suggested to abolish mitochondrial ROS generation, called the mild uncoupling concept (23). FFAs can also decrease ROS generation due to their uncoupling action, albeit only under conditions of reverse electron transport (succinate as substrate) (21). Initially, enhanced fatty acid-induced mitochondrial uncoupling of the obese and diabetic heart may therefore represent an adaptation to increased FA-induced ROS production. However, this does not completely compensate for the increased ROS production, as evidenced by the increased ROS-related damage seen in obesity or DM (20). Over the long term, uncoupling significantly contributes to the energetic deficit, with a decreased PCr/ATP ratio in diabetic hearts (6).

Mitochondria are not only a source, but also a target of ROS. In particular, mitochondrial DNA (mtDNA) appears at increased susceptibility to oxidative damage, which may be attributable to low mtDNA repair capacity and lack of histones, the close proximity to the ETC, or oxidative damage to mitochondrial proteins and lipids. This increased susceptibility has been observed in patients or animals with diabetes or obesity and in vitro models (8,20,24,25). Mitochondrial ROS also impair mitochondrial respiration via oxidative post-translational modifications of complex I and complex II of the ETC, whereas scavenging of mitochondrial ROS inhibited cardiac hypertrophy and improved diastolic function in a preclinical model of obesity (26). The molecular mechanisms linking ROS with cardiac hypertrophy and diastolic dysfunction include oxidative modification and inhibition of the sarcoplasmic reticulum calcium ATPase (SERCA) and impairment of the function of complex II through oxidative post-translational modification (26–28).

In addition to allosteric interactions that modulate enzyme activities, metabolic intermediates, such as acetyl coenzyme A or ROS, are involved in controlling the balance between glucose and fatty acid oxidation. In vitro acetylation of cardiac mitochondria increases ROS production and inhibits pyruvate oxidation (29), suggesting that acetylation of mitochondrial proteins is also involved in the modulation of metabolic flexibility.

Finally, the p66^Shc protein, a redox enzyme that uses reducing equivalents of the ETC to generate mitochondrial ROS (hydrogen peroxide [H₂O₂]) (30), is involved in obesity and DM. It is activated by hyperglycemia (31), and H₂O₂ generated by p66^Shc was shown to be involved in DM-related complications in various organs (32). In addition, p66^Shc is involved in the regulation of glucose transport into cells (33) and is critical in maintaining insulin-dependent signaling (34). The phosphorylation of p66^Shc on Ser36, which is induced via mitochondrial ROS and results in a further increase in ROS production, is known to be activated in various pathologies associated with oxidative stress, including DM and obesity (32). This p66^Shc-mediated ROS production can cause the oxidation of specific phosphatases involved in insulin signal transduction, such as phosphatase and tensin homolog (PTEN) or protein tyrosine phosphatase 1B, resulting in their inactivation (35,36). Deletion of p66^Shc prevents cardiac stem cell aging and development of heart failure in diabetic animals (37), suggesting that targeting p66^Shc may lead to beneficial therapies for diabetic cardiomyopathy and other ROS-related cardiac pathologies. The major sources of ROS generation in cardiomyocytes are summarized in Figure 2.

Figure 2. Major ROS Sources in Cardiomyocytes.

Most ROS in cardiomyocytes are produced at the mitochondrial ETC and by NADPH oxidase 2 (NOX2). In addition, other cytosolic sources, such as cyclooxygenase (COX) or xanthine oxidase (XO), contribute to ROS production in cardiomyocytes from patients with obesity or diabetes. Other mitochondrial proteins, such as p66^shc, monoamine oxidases A (MAO A), and NADPH oxidase 4 (NOX4) are also emerging as major ROS producers. Nitric oxide synthase (NOS) can become a powerful ROS generator, when uncoupled. BH₄ = tetrahydrobiopterin; e⁻ = electron; H⁺ = proton; H₂O₂ = hydrogen peroxide; O₂⁻ = superoxide; ONOO⁻ = peroxynitrite. Other abbreviations as in Figure 1.

Role of Mitochondrial Antioxidant Defense in Obesity and Diabetes

Due to the susceptibility of mitochondria to oxidative damage, ROS detoxifying systems including manganese-dependent superoxide dismutatase (MnSOD), glutathione peroxidases GPX-1 and GPX4, thioredoxin reductases (TrxR2), thioredoxin 2, glutaredoxin (Grx2), and peroxiredoxins (Prdx3 and Prdx5) are located directly within the mitochondria. In addition to increased ROS production by mitochondria or other sources, cardiac expression or activity of many antioxidant enzymes is reduced in obese or diabetic hearts (38), along with a concomitant reduction in circulating antioxidant enzymes (reviewed in (39)). Epidemiological studies reported low plasma vitamin E concentrations to be associated with an increased risk of developing type 2 diabetes (40). The transcription factor Nrf2, which regulates the expression of key antioxidant enzymes, including glutathione peroxidase, superoxide dismutase, peroxiredoxin, thioredoxin or thioredoxin reductase (41), was reported to be induced by lipid peroxidation products. This suggests that ROS products (lipid peroxide) contribute to the induction of antioxidant systems via Nrf2 in cardiomyocytes by a lipid peroxide–induced “pre-conditioning” cardioprotection, but with a mitochondrial hormetic response (42). Accordingly, some studies reported increased antioxidant defenses in obese or diabetic hearts (43).

Although numerous studies have documented signs of increased oxidative stress or an altered antioxidant defense in obese or diabetic hearts, the relationship between the degree of obesity and antioxidant defenses or systemic oxidative stress in humans is still an open question. No correlation at all, a positive correlation, or a link with obesity-related diseases have been described (44–46). Furthermore, the causality of the relationship between mitochondrial function and insulin sensitivity has been challenged (47). If oxidative stress is a major contributor to the progression of diabetes, strategies to reduce ROS, such as antioxidant supplementation, should be protective against diabetes-induced cardiac remodeling. Although the use of antioxidants in animal models of diabetes provided promising results, the clinical translation of this approach has not been straightforward, with randomized controlled trials of vitamin E or vitamin C supplementation failing to demonstrate a clinically significant benefit (48–50) (see also Part 2 of this review series).

Role of Other ROS Sources

Nonmitochondrial sources of ROS include: nicotinamide adenine dinucleotide phosphate (NADPH) oxidase, xanthine oxidase (XO), uncoupled endothelial nitric oxide synthase (eNOS), mitochondrial monoamine oxidase-A (MAO-A), lipoxygenase, cyclooxygenase, and other hemoproteins (51). Among these, the NADPH oxidase (NOX) family, which comprises 7 family members with distinct catalytic subunits that generate ROS through electron transfer from NADPH to molecular oxygen (52), play a pivotal role in ROS production in the diabetic and obese heart. Induction of NOX subunits and ROS production have been reported in patients with metabolic syndrome, and hyperinsulinemia was suggested to contribute to oxidative stress in these patients through activation of NOX (53). The activity of NOX is enhanced in hearts of obese animals (38). NOX inhibition abolishes cardiac O₂⁻ production in obese animals, and may even improve left ventricle function (38,54). High glucose exposure activates Rac1GTP and induces p47^phox translocation to the plasma membrane, resulting in NOX2 activation and increased ROS production in cardiomyocytes, which can be prevented by activation of adenosine monophosphate (AMP)-activated protein kinase (55,56). Chronic hyperglycemia increases AGE formation (57). Accordingly, high levels of AGEs have been found in tissue of patients with diabetes (58) and the AGE signaling pathway may act as a common upstream stimulus for ROS generation. Furthermore, NOX can function cooperatively with other ROS-producing pathways, thereby promoting mitochondrial “ROS-induced ROS release” and exacerbating overall oxidative stress in cardiomyocytes (59). Indeed, AGE binding to the receptor for advanced glycation end-products (RAGE) leads to activation of NOX and increased cytosolic ROS, which facilitates mitochondrial O₂⁻ production in hyperglycemic environments (60,61). The AGE–RAGE interaction also activates nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB), leading to up-regulation of RAGE itself and further ROS generation (62,63). In line with the diverse interactions between NOX and mitochondria, the mitochondria-targeted antioxidant agent mito-TEMPO attenuated myocardial dysfunction in diabetic mice and reduced messenger RNA expression of components of NOX (64), whereas the superoxide dismutase mimetic agent tempol induced a significant reduction in cardiac fibrosis and ROS production while increasing antioxidant enzyme capacity in diabetic rats (65). Sustained activation of nicotinamide adenine dinucleotide (NADH) oxidase in diabetes may also diminish intracellular NADPH, an essential cofactor for eNOS and several antioxidant systems (66). However, ROS formation by high glucose-stimulated NADPH oxidase may up-regulate antioxidant enzymes (66).

The saturated fatty acid palmitate, which circulates in higher concentrations in the blood of patients with diabetes or obesity, induces mitochondrial ROS, which is amplified by NOX2, causing greater mitochondrial ROS generation and mitochondrial dysfunction (67). Deficiency of the fatty acid transporter CD36 prevents cardiac steatosis, and increases insulin sensitivity and glucose utilization, but reduces FA uptake and oxidation, NOX activity, and palmitate-induced ROS production in a genetic mouse model of obesity (ob/ob mouse) (68).

Other important sources of cardiac ROS include XO and MAO-A. XO catalyzes the oxidation of its substrates, hypoxanthine and xanthine, to uric acid using oxygen as an electron receptor, and produces O₂⁻ and H₂O₂. DM is characterized by increased myocardial and serum XO activity, which can be attenuated with the XO inhibitor allopurinol, resulting in reduced cardiac fibrosis and improved systolic and diastolic cardiac performance of treated diabetic animals (69). Accordingly, the Western diet, with its excess fat and fructose, increases uric acid production and promotes cardiomyocyte hypertrophy, oxidative stress, myocardial fibrosis, and impaired diastolic relaxation, which can all be improved with allopurinol-induced reduction in cardiac XO and serum uric acid levels (70). The outer mitochondrial membrane serotonin-degrading enzyme MAO-A is another important source of H₂O₂ in the heart. MAO-A shows greater activity in diabetic cardiomyocytes, and MAO-A inhibition results in improved contractile function in preclinical DM models, despite persistent hyperglycemia and hyperlipidemia (71). However, no difference in MAO expression or MAO-related oxidative stress was recently observed in right atrial appendages from cardiac surgery patients with or without DM (72).

Excessive ROS Production and Cardiac Lipotoxicity

In the obese or diabetic heart, the supply of substrates exceeds the need for ATP synthesis. However, the ability of cardiomyocytes to respond to an increased fatty acid load is significantly reduced with age and obesity, resulting in accumulation of lipids and ceramide, mitochondrial dysfunction, increased ROS production, and possibly reduced cell viability (73–75). Specifically, under these conditions, excess lipids are shunted into nonoxidative pathways, resulting in the generation of toxic lipid intermediates, such as ceramides (73,76). These toxic lipid intermediates promote mitochondrial dysfunction, induce changes in signal transduction, and increase apoptosis; a phenomenon called lipotoxicity (76). Markers of oxidative stress, such as protein carbonyl content and 8-hydroxy-2’-deoxyguanosine, are significantly elevated in cardiomyocytes isolated from young patients with obesity (8). In addition, telomere length, a sensitive indicator of cumulative oxidative stress in post-mitotic cells such as cardiomyocytes (see Part 1 of this review series), is significantly reduced in cardiomyocytes from young subjects with obesity (8). Ectopic lipid accumulation in the heart is associated with cardiac hypertrophy, cardiac dysfunction, and apoptosis (75), and a strong association between increased cardiac fatty acid uptake and cardiomyopathy has been demonstrated in various preclinical models (77,78). Interestingly, cardiac lipid accumulation appears to be reversible in humans, with mechanical unloading by ventricular assist device implantation being shown to correct metabolic derangements and myocardial lipotoxicity in advanced heart failure (79).

ROS-Induced Changes in Insulin Signaling

Insulin-stimulated glucose uptake is impaired in obese and insulin-resistant animals and humans. Furthermore, the metabolic syndrome and insulin resistance are associated with abnormal left ventricular diastolic function and structure, which occurs independently of age, sex, blood pressure, and fasting plasma glucose, and is mainly associated with the state of insulin resistance (80,81).

Cellular insulin signaling occurs through 2 key pathways: the phosphatidylinositol-3-kinase (PI3K)/Akt and mitogen-associated protein kinase (MAPK) pathways (82). It also includes tyrosine phosphorylation of insulin receptor substrate (IRS) proteins (82). The metabolic responses, including glucose uptake by translocation of glucose transporter type 4 (GLUT4) to the cell membrane of cardiomyocytes, are mainly elicited via PI3K-mediated Akt activation. Among the defects that have been suggested as underlying mechanisms for insulin resistance are increased serine phosphorylation of IRS proteins by kinases such as inhibitor of nuclear factor kappa-B kinase subunit beta (IKKβ), c-Jun N-terminal kinase (JNK), p38MAPK, extracellular signal-related kinase (ERK), mechanistic target of rapamycin (mTOR), and S6K, resulting in an attenuation of engagement of IRS and PI3K (Figure 3) (83,84). ROS play a major role in altered activation of many of the aforementioned kinases, and have therefore been proposed as an important link between impaired mitochondrial function and insulin resistance (85,86). Indeed, ROS are known activators of stress-activated proteins p38 MAPK and JNK, which inhibit insulin signal transduction by phosphorylating IRS proteins (87). In addition, mitochondrial ROS stimulate proinflammatory signaling by activation of IKKβ and other kinases that phosphorylate IRS-1 at serine residues (88). Fatty acids (lipid infusion) can also lead to the accumulation of diacylglycerol and other lipid derivatives that activate protein kinase Cs (PKCs), which, in turn, increase serine phosphorylation of IRS proteins and lead to inhibition of insulin signaling (89).

Figure 3. ROS-Induced Changes in Insulin Signaling.

Insulin signaling pathways are initiated by insulin binding to the extracellular alpha subunit of the insulin receptor (IR), followed by a conformational change in the beta subunit of the IR, which has intrinsic tyrosine kinase activity. Receptor activation results in tyrosine phosphorylation of the IRS and subsequent activation of phosphatidylinositol 3-kinase (PI3K). Activation of PI3K leads to stimulation of various downstream serine kinases, including protein kinase B (Akt), which is involved in the translocation of the major glucose transporter GLUT4. Among the mechanisms underlying insulin resistance in cardiomyocytes from patients with obesity or diabetes is increased serine phosphorylation of IRS proteins. Phosphorylation of IRS proteins at particular serine residues inhibits the interaction of IRS proteins with the insulin receptors, resulting in a reduction in activity of the abovementioned signaling pathway and impaired cardiomyocyte glucose uptake. Lipid derivatives stimulate c-Jun N-terminal kinase (JNK), protein kinase C (PKC), and IκB kinase (IKK)-mediated phosphorylation of IRS-1 at serine residues. In addition, mitochondrial dysfunction increases ROS production, which causes activation of serine/threonine kinases, including p38MAPK, JNK, IKKβ, and extracellular signal–regulated kinase (ERK), which increase serine phosphorylation of IRS proteins. Overnutrition, resulting in increased free fatty acids, also contributes to impaired insulin signaling through mechanistic target of rapamycin (mTOR)–S6 kinase 1 (S6K1)-mediated serine phosphorylation of IRS. MAPK = mitogen-activated protein kinase; Other abbreviations as in Figure 1.

Excessive ROS release also impairs protein folding and post-translational modifications that occur in the endoplasmic reticulum (ER). Compromised ER function and ER stress contribute to altered insulin signaling by activation of serine/threonine kinases (90). Insulin signaling is also involved in the regulation of myocardial autophagy (91). Hyperinsulinemia was shown to suppress myocardial autophagy via Akt/mTOR signaling, whereas fasting-induced low insulin levels induce autophagy (91). Recently, it was shown that NOX2-derived O₂⁻, but not mitochondrial O₂⁻ production, induces impaired autophagic flux in response to palmitate in cardiomyocytes (92). This may contribute to impaired cardiac autophagy in response to lipid overload in insulin resistant states (92).

Involvement of ROS in Cardiac Metabolic Memory

Prolonged hyperglycemia induces metabolic changes that alter tissue homeostasis, even after glucose normalization, a phenomenon called metabolic memory (93). Epigenetic mechanisms contribute to the development and maintenance of cardiac metabolic memory. Specific enzymes modify discrete residues on histone tails (writers), other enzymes can remove these marks (erasers), and still others recognize these histone marks (readers), enabling gene expression to proceed (94). Chromatin marks consist of epigenetic post-replicative methylation of DNA at cytosine residues and various post-translational modifications mainly at arginine and lysine residues. Long-term epigenetic effects, such as histone and DNA methylations, are relatively stable and can be transferred as memory to offspring cells. Furthermore, maternal diet and in utero environment can induce epigenetic changes and can be inherited by future generations, triggering diseases such as obesity or DM (95). This chromatin remodeling enables cardiomyocytes to respond to different stimuli by controlling DNA accessibility and thus gene expression.

Hyperglycemia induces long-lasting activation of inflammatory and oxidative stress pathways, resulting in long-lasting or even irreversible epigenetic changes that contribute to the abnormal character of blood cells, endothelial cells, vascular smooth muscle cells, or cardiomyocytes in patients and animals with obesity or diabetes (96). Experimental evidence also suggests hyperglycemia-mediated ROS is a major driver of glycemic memory in endothelial cells (97). So far, only sparse data on potential molecular mechanisms for a metabolic memory in cardiomyocytes have been reported. Recently, it was shown that high glucose levels induce increased levels of the inflammatory cytokine interleukin (IL)-6 and decrease histone-3 methylation at the IL-6 promoter in cardiomyocytes, which was irreversible after removal of high glucose (98). Thus, the high glucose-induced increased inflammatory gene expression in cardiomyocytes was due to a loss of repressive epigenetic histone modifications (98). High glucose-induced mitochondrial dysfunction and apoptosis did not appear to be responsible for the metabolic memory in cardiomyocytes (98). Furthermore, high glucose induces epigenetic regulation of the insulin-like growth factor 1 receptor in cardiomyocytes (99).

In addition to in vitro analyses, epigenome-wide association studies have investigated the impact of obesity or DM on epigenetic changes in patients. In a recent, large study, increased obesity in adults was associated with increased methylation at the hypoxia inducible factor 3A (HIF3A) locus in blood cells and adipose tissue, but not in skin (100). Apparently, epigenetic markers show strong tissue and cell-type specificity (96,101,102). Therefore, the transferability of epigenetic signatures from easily obtainable blood cells of patients with diabetes or obesity to cardiomyocytes needs to be investigated. Finally, metabolic intermediates, such as those of the tricarboxylic acid cycle, glycolysis, FAO, or the hexosamine biosynthetic pathway, are cofactors for chromatin-modifying enzymes (102). Appreciating how alterations in metabolism and nutrition modify epigenetic gene regulation of eukaryotic cells through metabolic intermediates might significantly contribute to therapeutic innovation in obesity and DM.

Conclusions Regarding Obesity, Diabetes and Oxidative Stress

Increased oxidative stress in the heart and cardiomyocytes arises via multiple mechanisms, including mitochondrial dysfunction and uncoupling, increased FAO, enhanced NOX activity, and reduced antioxidant capacity (Central Illustration). Recent data suggest that metabolic memory is an important contributor to CVD, as cells remember exposure to hyperglycemia and oxidative stress. Pharmacological ROS scavenging has been used to improve myocardial energy metabolism and insulin responsiveness, and to reduce cardiac dysfunction in preclinical models of obesity or DM, and in some patient studies (103). However, despite overwhelming evidence of damaging consequences of oxidative stress in obesity and DM, large-scale clinical trials using antioxidant therapies for the treatment of CVD have failed to demonstrate benefit. Vitamin E treatment, for example, had no apparent effect on cardiovascular outcomes in the HOPE (Heart Outcomes Prevention Evaluation) study, a randomized, placebo-controlled, double-blind clinical trial involving more than 8,000 patients with CVD or DM (48). ROS differ in terms of their reaction kinetics, diffusion parameters, site of production, and degradation kinetics. Rather than merely scavenging reactive radicals, a more comprehensive approach may be required that prevents ROS-generation while also promoting ROS scavenging in particular cellular compartments. Importantly, the complex interactions among the various ROS sources within the cell and mitochondria, and the mechanisms responsible for the increase in ROS formation in patients with diabetes or obesity, remain to be elucidated. Future work will also need to address whether the epigenetic signatures associated with obesity and DM can be reverted by drugs that can modify the epigenome.

Central Illustration. Oxidative Stress and Cardiovascular Risk Factors.

Obesity, diabetes, smoking, and pollution are prominent causes of oxidative stress in the cardiovascular system, with these mechanisms increasingly appreciated as playing a major role in disease pathogenesis. AGE = advanced glycation end-product; ETC = electron transport chain; FA = fatty acid; NADPH = nicotinamide adenine dinucleotide phosphate; NOX2 = NADPH oxidase 2; PKC = protein kinase C; ROS = reactive oxygen species.

Role of Oxidative Stress in the Cardiovascular Effects of Air Pollution and Smoking

Introduction

Few would argue that our entitlement to breathe clean air is a basic and essential human right. Yet, the air we breathe is far from clean. Even in the most isolated environments, the atmosphere is a diverse array of substances from aerosolization of the ground, dusts carried in the wind, natural gases, biological material such as pollens and spores, as well as viruses and bacteria. We have evolved complex defense mechanisms to protect our bodies against natural pollutants, but the same cannot be assumed for man-made sources of air pollution, such as those from industry, households, and traffic. In most modern societies, these pollutants are ubiquitous and, for many, exposure is unavoidable. The dramatic mortality that accompanied episodes of high air pollution, such as that of the Meuse Valley fog (Belgium, 1930; 60 deaths from the persistence of industrial emissions over 3 days), the Donora air inversion (United States, 1948; 3 days of high air pollution led to ill-health in over one-third of the town’s population) and the 1952 London smog (United Kingdom, 1952; 4- to 5-day period believed to cause between 4,000 and 10,000 deaths), has cemented the need for regulatory control. The introduction of legislation (e.g., the 1956 Clean Air Act or national regulations such as the 1977 and 1990 U.S. Air Pollution Control Act, 2008 European Union directives) has effectively reduced levels of many pollutants. Yet present-day air pollution continues to be a serious public health issue, with increasing industrialization and the rapid expansion of urban environments. Recent estimates suggest air pollution is responsible for between 3 and 7 million deaths worldwide per year, accompanied by staggering levels of morbidity (3.1% of global disability-adjusted life-years) and associated economic risks (1 to 3 trillion U.S. dollars/year worldwide) (104–106). Indeed, a recent report placed both indoor and outdoor air pollution within the top 10 risk factors for all-cause disease, greater than that caused by risk factors such as sedentary life style or high cholesterol (106).

Given these disturbing figures, it is perhaps surprising that we still continue to expose ourselves to air pollutants for pleasure, by the smoking of cigarettes or other tobacco products. Smoking is believed to kill 6 million people each year worldwide, with 480,000 and 96,000 deaths each year in the United States and United Kingdom alone, respectively (107–109). More than 16 million people in the United States are believed to be living with a disease caused by smoking, with estimated annual costs in the region of $300 billion in the United States and €380 billion in Europe (108,110). Although the ban of smoking in public places has been one of the major successful health interventions of recent years (reduced rates of myocardial infarction by 17%) (111), smoking remains prevalent (currently estimated to be over 1 billion persons worldwide). Thus, although current trends suggest the prevalence of smoking is declining overall, the decline is slow, use is increasing in many low- and middle-income countries, and the global health costs are expected to rise over the next decade (112,113).

There are clear differences in the chemical composition of environmental pollution and smoking. However, the counter is also true; both derive from the combustion of complex carbon-rich materials, with notable similarities in the compositions of these fumes (Figure 4). More importantly, the consequences to health are also comparable, and particularly well-exemplified in relation to CVD. Notably, in terms of mortality, the cardiovascular effects of both smoking and air pollution outweigh that of death from pulmonary conditions (106). The biological pathways that link pulmonary exposure to their cardiovascular actions remain the subject of ongoing research, although oxidative stress is a re-emerging mechanism, and one that undoubtedly contributes to the progression of disease.

Figure 4. Similarities Between the Chemical Composition of Tobacco Smoke and Traffic-Derived Air Pollution.

As shown, there are many similarities between the chemical composition of tobacco smoke and traffic-derived air pollution. Although certain differences exist, the consequences of these pollutants on cardiovascular health are broadly similar. CO = carbon monoxide; CO₂ = carbon dioxide; HCN = hydrogen cyanide; NO₂ = nitrogen dioxide; O₃ = ozone; PAH = polyaromatic hydrocarbon; SO₂ = sulfur dioxide; SVOCs = semi-volatile organic compounds.

We provide an overview of cigarette-related products (C-RP) and combustion-derived environmental air pollution (C-DEAP; principally urban air pollution), to highlight where the similarities exist between their physicochemical properties and the means through which they impair cardiovascular health. We describe the potential biological pathways underlying these actions, with a focus on the role of oxidative stress and how this cellular event could have implications for interventional strategies (Central Illustration).

Sources, Composition, Particles, and Exposure

Smoking

Cigarettes are the most commonly used tobacco products and, despite falling use, cigarette manufacture is substantial (6.35 trillion cigarettes smoked worldwide in 2012) (113). Other tobacco products, such as cigars and tobacco pipes, follow a similar trend, albeit with markedly lower numbers. Water pipes (hookah, shisha) remain a popular means to inhale tobacco smoke in Middle Eastern countries. There is overwhelming evidence that all these forms of smoking have major detrimental health effects (114,115). Electronic (e)-cigarettes have surged in popularity recently, largely attributed to the unconfirmed assumption that they have lesser adverse health effects, a discussion of which can be found elsewhere (e.g., (116–118)).

Tobacco smoke comprises a mixture of gases, aerosolized liquids, and small particles. The composition is highly complex, with more than 4,000 chemicals in C-RP smoke, of which 250 are known to be harmful and 50 more are known to cause cancer (107). The gas/vapor phase of C-RP smoke contains nitrogen, carbon dioxide (CO₂), and carbon monoxide (CO), with smaller, but significant amounts of hydrogen cyanide, nitric acid, methane, benzene, acrolein, acetone, hydrogen sulfide, and a wide selection of hydrocarbons, aldehydes, carbonyls, and nitrosamines (119). The particulates within C-RP fumes are predominantly carbon-based, composed of a mixture of semi-combusted material that contains tar and nicotine. Elemental carbon within the soot particle provides a scaffold for adsorbed organic carbon species and liquids, including phenols, carboxylic acid, paraffin waxes, and nitrosamines among many others. The polyaromatic hydrocarbons (PAHs), quinone species, and transition metals within C-RP particulates have a high capacity to instigate redox reactions in cells.

Air pollution

Environmental air pollution derives from a large number of sources that include both man-made (e.g., industry, power plants, traffic, household combustion and cooking, construction, mechanical wear, agriculture) and environmental sources (e.g., forest fires, volcanic eruptions, aerosolization of soil and dusts, pollen and molds). Indoor air pollution (e.g., from cooking, heating, dust, candles, and accumulation of outdoor pollutants) is a pressing concern, but one that has received less attention than its outdoor counterpart. The majority of attention has fallen on outdoor urban air pollution due to the high incidence of exposure in dense urban populations. Pollution of urban air will depend on the extent of the sources listed previously, as well as geographical and meteorological conditions. Traffic-derived emissions (e.g., diesel exhaust [DE]) are a prominent source of urban pollution, and one that is under particular scrutiny due to their ubiquity and increasing prevalence worldwide.

Urban pollution is a complex cocktail of chemicals that can also be broadly characterized into gases, semi-volatile liquids, and particles (120). Numerous gases are found within C-DEAP, such as sulfur dioxide (SO₂), CO₂, and CO, with recent attention focused on ozone (O₃) and nitrogen dioxide (NO₂) in particular. Ozone readily oxidizes other air pollutants and biological/cellular material, both directly and through free radical generation. NO₂ also acts as an oxidizing agent, as well as modulating cell function via nitric acid formation, nitrosative reactions, and free radical reactions (121). The effects of O₃ and NO₂ are likely to be additive; however, shared sources and the dynamic relationship with other pollutants makes disentangling their actions challenging.

A plethora of semi-volatile organic compounds form the “liquid” phase of air pollution. These include methane, benzene, naphthalene, formaldehyde, and alkanes, as well as a range of PAHs, polychlorinated biphenyls, and polybrominated diphenyl ethers. The close interaction of semi-volatile chemicals with gaseous (e.g., interplay between methane and ozone) and particulate (absorption to the carbonaceous surface) components often results in the ‘liquid’ phase of C-DEAP being grouped within the gases or, more usually, as “particulate matter” for toxicological purposes.

The particulate matter (PM) in air pollution is a mix or organic and inorganic particulates. Environmental PM is categorized into 3 groups according to sampling conventions: coarse particles (PM₁₀; particles with a diameter of 10 µm or less), fine particles (PM_2.5; diameter of 2.5 µm or less), or ultrafine particles (PM_0.1; diameter of <100 nm, also referred to as nanoparticles). Airborne PM is regulated on the basis of PM₁₀ and PM_2.5, and at present, PM_0.1 cannot be measured through existing environmental monitoring networks. Different sources of PM have various size ranges, compositions, and reactivities. C-DEAP particulate is rich in carbon, with the mixture of elemental and organic carbon depending on the fuel source and efficiency of combustion. They carry a cocktail of harmful chemical species on their surfaces that include unburned hydrocarbons, carcinogenic organic carbon species (PAHs, alkanes, quinones), and redox-active transition metals. The biological toxicity of PM is greatly dependent on the composition of the particulate, however, in general, the small size fractions exert greater effects due to their large reactive surface area for a given mass, and their ability to penetrate deep into the alveoli of the lungs, and potentially into the bloodstream.

Mechanisms of Induction of CVD

Epidemiological studies have shown clear associations between various CVDs and both smoking and air pollution, including for coronary artery disease (122–125), peripheral arterial disease (126,127), heart failure (128,129), cerebrovascular disease (124,130–132), cardiac arrhythmia/arrest (133,134), and venous thromboembolism (135) (for overviews see (118) for C-RP and (136,137) for C-DEAP). There is now a substantial body of experimental work demonstrating that these pollutants impair cardiovascular function (138,139). In particular, controlled exposure studies in humans and animals have demonstrated that pollutants have the capacity to impinge on almost all aspects of cardiovascular function (Figure 5A).

Figure 5. Biological Pathways Through Which Inhaled Pollutants Could Induce Cardiovascular Morbidity and Mortality.

(A) Overview of the 3 main hypotheses by which inhaled pollutants could cause cardiovascular effects and their multiple detrimental actions on the blood, vasculature, and heart that lead to cardiovascular morbidity and mortality. Oxidative stress plays a role in exacerbating numerous aspects of each pathway. (B) Section of mouse lung following instillation of ultrafine particles. Note the presence of macrophages densely-loaded with particles (arrows), and the thin cellular barrier between the alveolar space and the pulmonary arterioles. (C) The 3 main hypotheses for how inhaled pollutants cause cardiovascular effects overlaid on a transmission electron micrograph of a rat lung (. Micrograph reprinted, with permission, from Lehnert (257)]: 1) Pollutants induce an inflammatory response in the lungs, leading to release of cytokines and other mediators that spill over into the systemic circulation. 2) Constituents within pollutants may translocate across the alveolar wall and directly interact with the cardiovascular system. 3) Pollutants may activate the autonomic nervous system through sensory receptors on the alveolar surface. Note the thin barrier of alveoli with pulmonary capillaries (shaded in red) and the presence of an alveolar macrophage (green outline) on the pulmonary surface overlying a capillary. Adapted, with permission, from Miller et al. (205). aas = alveolar air space; pc = pulmonary capillary.

Vascular disease and pollution

Air pollution is associated with elevated blood pressure and, accordingly, many sources of C-DEAP increase blood pressure in humans and animals (140,141). Elevated blood pressure is primarily accounted for by altered vascular function, with C-DEAP exposure generally promoting vasoconstriction and decreasing vasodilator responses. Controlled inhalation of dilute DE in humans (at levels representative of busy roadside exposure) impairs the skin microvasculature response (142) and produces a marked attenuation of vasodilatation of forearm blood flow (143). Interestingly, the magnitude of the attenuation of forearm vasodilatation is similar to that in a life-long smoker compared with a nonsmoker (144,145). Vascular impairments are apparent at 2 h after exposure and are still evident 24 h later (146). Increases in arterial stiffness are also found after acute exposure to C-DEAP (147,148). Altered baroreceptor sensitivity may also contribute to the vascular impairment following C-DEAP exposure (149,150). Epidemiological studies have clearly demonstrated that exposure to urban air pollution is associated with atherosclerosis (137,151,152), and data from animal studies supports that C-DEAP can increase the size and, potentially, the vulnerability of plaques (153–155). Finally, there are indications that DE has the capacity to alter ischemia-induced angiogenesis (156).

Blood, circulating factors, and pollution

Exposure to air pollution is frequently, although inconsistently, accompanied by increases in circulating inflammatory markers (157–159). Animal models have shown that inhalation of C-DEAP promotes the adherence of leucocytes to the vascular wall, an early event in atherogenesis (160). There have been mixed findings in regard to whether C-DEAP increases coagulation pathways (161–165). However, inhalation of DE in healthy volunteers potentiated the formation of thrombus in an ex vivo model of arterial injury, predominantly through platelet hyper-reactivity (166). Similar platelet activating effects are seen in cigarette smokers (167). Fibrinolytic pathways are also affected through impaired release of tissue plasminogen activator from endothelial cells (143–145,168). The particulate components of DE alone have the capacity to exacerbate thrombosis and inhibit fibrinolysis (169). C-DEAP also induces alterations in circulating stem cell populations (170,171), potentially disrupting vascular repair (172).

Cardiac disease and pollution

In addition to its effects on the pulmonary and peripheral vasculature, C-DEAP has also been shown to have detrimental effects on the coronary circulation in animal models (173–176). Pulmonary exposure to DE particles increases the susceptibility of the myocardium to ischemia-reperfusion damage in rats (177), and inhalation of dilute DE is associated with greater myocardial ischemia (ST-segment depression during exercise) in patients with ischemic heart disease (168). The PM of C-DEAP has been shown to increase the incidence and duration of arrhythmia (177–179), and there are extensive (but somewhat inconsistent) published reports showing a trend between air pollution exposure and measures of heart rate variability (180,181). Finally, epidemiological studies have found consistent associations between air pollution and heart failure (129), with animal models demonstrating that long-term exposure to PM promotes myocardial hypertrophy and loss of cardiac function (182).

Pollution and other systems

Briefly, it is worth highlighting that C-DEAP has recently been shown to have other systemic effects that adversely impact the cardiovascular system. These include impaired renal blood flow and an increased incidence of renal disease (183,184), exacerbation of metabolic syndrome/diabetes (185–188), changes to the placental circulation (189–191) and epigenetic alterations that may affect the cardiorespiratory health of offspring (192,193).

Biological Mechanisms

Lung to cardiovascular system

The exact mechanisms by which inhaled pollutants lead to effects on the cardiovascular system remain to be determined. However, 3 theories currently predominate (Figure 5) (see also (194)). The classical hypothesis is that inhaled pollutants activate inflammatory cells in the lung, leading to the release of inflammatory mediators that pass into the circulation to influence cardiovascular function (195). A number of constituents of air pollution and tobacco smoke have the capacity to induce pulmonary inflammation and oxidative stress (see later discussion), and parallel pathways can be up-regulated in the systemic circulation and organ systems. At present though, inconsistencies in systemic biomarkers of inflammation between studies and dissociation across the time course of response suggest that this pathway alone cannot fully explain the multiple cardiovascular effects of either exposure. The second theory is that inhaled pollutants activate alveolar receptors, stimulating sensory afferents that alter cardiovascular function via changes in autonomic balance or neuroendocrine regulation (125,196). This pathway is very plausible in terms of the effects of C-DEAP on cardiac electrophysiology, and potentially for some of the rapid (<2 h) cardiovascular effects after exposure. Finally, pollutants themselves (or the constituents released from pollutants) may cross the alveolar wall into the pulmonary circulation, and directly interact with the cardiovascular system (197). Ultrafine particles and/or their constituents have been shown to translocate from the lung to the blood in animal models; however, the fate of these constituents and the mechanisms by which very low levels of translocated particles/constituents can produce widespread cardiovascular impairment requires further research. Overall, evidence exists for all 3 theories, and it is likely that all 3 act in concert to produce the widespread cardiovascular effects of inhaled pollutants (194). The cellular signaling at each stage is subject to ongoing research; however, oxidative stress is a recurring observation and likely driver of all 3 pathways.

Induction Of Oxidative Stress

Cohort studies have found that exposure to high levels of air pollution is associated with a number of biomarkers of oxidative stress, including oxidation of plasma proteins and lipids (e.g., malonaldehyde and protein 2-aminoadipic semialdehyde) (198,199), urinary isoprostanes (200), and oxidative DNA adducts (8-hydroxy-2’-deoxyguanosine) (201–204). In many cases, indications of oxidative stress coincide with that of markers of inflammation, although there is considerable variability in both (205).

Controlled exposure studies in humans have suggested a role for oxidative stress after acute exposure to C-DEAP. DE inhalation attenuates vasodilation in response to both endothelium-dependent and nitric oxide donor drugs, but not in response to vasodilators acting via smooth muscle cell receptors; a profile that is suggestive of NO scavenging by O₂⁻ free radicals (143,206). Exposure to C-DEAP is known to cause a compensatory increase and to eventually deplete antioxidant concentrations in respiratory tract lining fluids and pulmonary cells (207). Likewise, inhalation of DE alters the expression of several antioxidant pathways in peripheral blood monocytes (208). Furthermore, DE particles induce a greater inflammatory response in individuals with genetic deficiencies in various antioxidant systems (209–211).

There are extensive published data from cellular and animal models showing that different types of C-DEAP induce oxidative stress (see (205)). Animal models are especially useful in assessing the mechanisms of action of air pollution in chronic disease processes such as atherosclerosis. Mouse models have shown that the proatherosclerotic effects of C-DEAP are closely associated with oxidation of circulating lipids and arachidonic acid metabolites (212–214), increased urinary isoprotanes (215), nitrotyrosine staining (an indirect marker of oxidative stress) (216), compensations in antioxidant expression (154,216–218), and markers of oxidative stress in plaques/vascular wall (219,220). Furthermore, a diverse array of oxidative pathways are emerging to account for these observations, including an inability of high-density lipoprotein (HDL) cholesterol to protect against low-density lipoprotein (LDL) cholesterol oxidation (217), up-regulation of NOX expression/activity (221,222), eNOS uncoupling (223), and signaling through lectin-like oxidized LDL receptors (212).

Both gaseous and particulate pollutants have the capacity to initiate oxidative reactions in the body. Ozone and NO₂ are both oxidizing agents, and are known to induce oxidative stress in pulmonary cells (121). The lung lining fluids surrounding pulmonary epithelial cells in vivo provide an early defense against acute exposure to inhaled pollutants. However, prolonged or repeated exposure leads to depletion of antioxidants in cell surfactants and changes in pulmonary cell function (207). The gases within C-DEAP can modify blood constituents in the pulmonary circulation, although the precise mechanism by which the oxidative “signal” is carried or transferred to peripheral organs requires further investigation (159). Particulates in C-DEAP can also diminish pulmonary defenses and induce oxidative stress in most cell types of the lung (224). Particles themselves generate free radicals in the absence of biological tissue (194,225) (Figure 6), and directly oxidize isolated DNA, lipids, and enzymes in vitro (see (205)). There is some debate as to whether the in vitro oxidative potential of particulates alone is a useful predictor of toxicity in vivo (207), given that it will underestimate the contribution of oxygen free radicals generated from particle-induced activation of cellular enzymes. However, should particles translocate into the circulation, their oxidative capacity is likely to be an important determinant of their systemic effects. Levels of metals and organic carbon in C-DEAP play a crucial role in the oxidizing and free radical–generating capacity (120,210). Finally, both gaseous and particulate pollutants can induce inflammatory responses in the lung and blood, which provide a means of amplification of the oxidative stress induced by C-DEAP. It is notable that, almost without exception, all the mechanisms described earlier for C-DEAP have been also proposed as potential mediators for the detrimental effects of smoking on the cardiovascular system, especially for oxidative stress pathways (138).

Figure 6. Mechanisms Through Which Inhaled Pollutants Can Induce Cellular Oxidative Stress.

1) Innate generation of free radicals by particles in the absence of biological tissue. Free radicals may be produced by redox reactions between different chemicals on the surface of the particulate, as well as interactions with other constituents within air pollution. 2) Release of cytokines and oxidative mediators from pollutant-induced activation of inflammatory cells. 3) Free radical generation from interaction between pollutants and cells (e.g., from stimulation of enzymes such as nicotinamide adenine dinucleotide phosphate (NADPH) oxidase, xanthine oxidase (XO), uncoupling of endothelial nitric oxide synthase (eNOS), induction of inducible nitric oxide synthase (iNOS), exacerbation of free radicals from mitochondrial inefficiency, or depletion of antioxidant defenses). These pathways may be stimulated by particles themselves, through release of soluble constituents on the particle surface, or intermediate chemical reactions of different constituents within biological fluids. Adapted from Miller et al. (205).

Interventions to Reduce Exposure, Oxidative Stress, and CVD

Reducing sources of air pollution will undoubtedly remain the principal means of preventing its adverse health effects (111,136). However, although the levels of many air pollutants have fallen dramatically in many countries over the last few decades, there are concerns as to whether we are regulating the correct pollutants. For example PM₁₀ and PM_2.5 are currently used for measuring particulate air pollution in the environment, but these measures are greatly skewed by larger particles and are not reliable indicators of ultrafine particles (e.g., those from vehicle exhaust), which have much higher potential to affect health (226). Furthermore, current epidemiological evidence suggests that even air pollution at levels below those recommended by current regulatory guidelines is associated with detrimental health effects (220,227).

Awareness of the adverse health effects of air pollution is growing around the world and there are an increasing number of avenues to reduce it (see (136,228)). Industrial emissions are tightly regulated, and relocation to nonresidential areas and stringent health and safety measures for workers have reduced human exposure. Modern vehicle engines and fuels in general produce lower emissions than their predecessors, and there is a slow, but persistent movement towards alternative energy fuels, such as biodiesel, hydrogen cells, and electric vehicles. Emerging scientific evidence has shown that alternative fuels (229), fuel additives to improve the efficiency of combustion (230), and exhaust filters to remove particles from emissions (231) have the potential to greatly reduce the cardiovascular actions of vehicle exhaust.

Prevention at the level of the individual also merits attention, particularly for those believed to be especially susceptible to air pollution (e.g., young, elderly, and those with pre-existing cardiopulmonary disease). The availability of regularly updated air quality data and low-cost personal measuring devices is providing individuals with the means to avoid prolonged exposure on high air pollution days (120). Furthermore, facemasks may be a simple and cost-effective means to reduce exposure to some pollutants that may result in benefits to cardiovascular health (232).

One subject of debate in the scientific community is whether or not pharmacological agents or supplements can limit the harmful effects of air pollution (recently reviewed in (233)). Although pharmacological agents are unlikely to ever represent a practical means of primary prevention of the health effects of air pollution, their use in scientific investigations provides insight into the biological mechanisms involved. Beta-blockers or antagonists of alveolar sensory receptors can attenuate the cardiac effects of C-DEAP in rodent models (177,234). Additionally, endothelin receptor antagonists (175,235), inhibitors of the renin/angiotensin system (236), and statins (237,238) can reduce the vascular and atherosclerotic effects of air pollutants.

Given the clear involvement of oxidative stress in the cardiovascular actions of air pollution, it is possible that antioxidant agents could represent a useful preventative strategy. Cell culture studies have demonstrated that vitamin C, N-acetylcysteine (NAC), trolox, and inhibitors of iron-mediated free radical generation can prevent the actions of C-DEAP particles in vascular cells (239,240) and macrophages (241). Supplementation of the endogenous antioxidant superoxide dismutase attenuates the vascular effects of C-DEAP particles in isolated blood vessels (205). Furthermore, emerging data from animal models suggests that several antioxidant(-rich) compounds, including curcumin (242), NAC (243), NAD(P)H oxidase inhibitors (244), vanillic acid (245), emodin (246), tempol (235), and even dark chocolate (247), can protect against some cardiovascular actions of C-DEAP. Regrettably, these findings have not been replicated in human studies, with antioxidant-rich foods or supplements having inconsistent effects on PM_2.5-induced heart rate variability (248,249) and vascular dysfunction (250–252). Of concern, and despite the fact that in chronic smokers vitamin C has been shown to improve endothelium-dependent forearm blood flow (252), in another study, the vasoconstrictor effects of DE were actually greater in volunteers receiving vitamin C or NAC (253), These findings mirror those of the largely negative results of large-scale antioxidant trials for CVD in general (254), and may reflect the inability of current compounds to reach and reduce oxidative stress at key biological areas. Nevertheless, the identification of individuals with polymorphisms of antioxidant genes as being particularly susceptible to the effects of air pollution (210) suggests that this avenue of research still deserves further attention in hypothesis-driven studies with long-term intervention.

Finally, although targeting human toxins that are potentially even broader in their range than those in smoking and air pollution, including the full gamut of lifetime environmental exposures, comment is required regarding chelation therapy and TACT (Trial to Assess Chelation Therapy) (255). TACT was an NIH-sponsored, double-blind, placebo-controlled, 2 × 2 factorial randomized trial enrolling 1,708 patients 50 years of age or older who had experienced a myocardial infarction (MI) at least 6 weeks prior and had serum creatinine levels of 2.0 mg/dl or less. Participants were recruited across 134 U.S. and Canadian sites. Patients were randomized to receive 40 infusions of a chelation solution (comprising disodium ethylenediaminetetraacetic acid [EDTA], ascorbate, B vitamins, electrolytes, procaine, and heparin; n = 839) versus placebo (n = 869), and an oral vitamin-mineral regimen versus an oral placebo. The putative active chelating agents (the infusions of vitamins and disodium EDTA), are believed to act by binding divalent and some trivalent cations, including calcium, magnesium, lead, cadmium, zinc, iron, aluminum, and copper, facilitating their urinary excretion (255). The pre-specified primary endpoint was a composite of total mortality, recurrent MI, stroke, coronary revascularization, or hospitalization for angina. Importantly, the primary endpoint occurred in 222 (26%) of the chelation group versus 261 (30%) of the placebo group (hazard ratio, 0.82 [95% CI: 0.69 to 0.99]; p = 0.035). This somewhat unexpected result, published in 2013, was met with a degree of skepticism (256) and vigorous debate ensued. Regardless of these controversies, the efficacy of chelation therapy should soon be definitively shown, as the TACT2 trial (NCT02733185), with a similarly rigorous design as its predecessor, began recruitment in October 2016, with a target enrollment of 1,200 subjects. TACT2 should prove or refute the efficacy and clinical role of this approach.

Conclusions Regarding Pollution, Smoking, and Oxidative Stress

There is now overwhelming evidence that air pollution is associated with CVD, with expert opinion suggesting it should be formally recognized as a risk factor in the same way that smoking of tobacco products has been for many years. There are striking similarities between the physicochemical composition of cigarette smoke and combustion-derived air pollution, and both are associated with multiple detrimental cardiovascular effects.

A number of unresolved issues in the field require attention. First, although it is assumed that certain individuals are more susceptible to air pollution than others (e.g., children, the elderly, and those with pre-existing cardiorespiratory disease) this has not been definitively shown, and requires further elaboration so that practical guidance can be provided to those at risk (e.g., altered activity, dietary supplementation/medication, interventions such as facemasks or air purifiers). Additionally, although the particulates and some gases (e.g., NO₂) are highly likely to be detrimental constituents of urban air pollution, more precise identification of the components that drive the cardiovascular action of air pollution are required (e.g., particle sizes and sources, and the degree of chemical interaction between gases and particulates). Smoking, although declining in prevalence, is also not without its outstanding health issues. The current popularity of “smoking alternatives,” such as e-cigarettes, should continue to be a topic of research, especially to establish the constituents of the fumes that could be associated with health effects. Although the biological mechanisms underlying the cardiovascular actions of smoking and air pollution are still to be fully established, there is a clear role for oxidative stress as a key mediator that will exacerbate and potentially instigate the disease process. In particular, the biological pathways that link the initial lung response to air pollutants to that of the subsequent cardiovascular actions remain an important undetermined area for future research. Precise identification of the mechanisms at play will be extremely useful for identifying which constituents of air pollution are especially harmful and who is particularly susceptible to their effects. At present, though, reducing the prevalence and exposure to both these environmental risk factors remain the key means to preventing the significant burden they inflict on health.

ABBREVIATIONS AND ACRONYMS

AGE

advanced glycation end-product

C-DEAP

combustion-derived environmental air pollution

CVD

cardiovascular disease

DE

diesel exhaust

DM

diabetes mellitus

ETC

electron transfer chain

IRS

insulin receptor substrate

NADPH

nicotinamide adenine dinucleotide phosphate

NOX

nicotinamide adenine dinucleotide phosphate oxidase (NADPH oxidase)

ROS

reactive oxygen species