Potential Role of Metal Chelation to Prevent the Cardiovascular Complications of Diabetes

Abstract

Context

For decades, there has been epidemiologic evidence linking chronic toxic metal exposure with cardiovascular disease, suggesting a therapeutic role for metal chelation. Given the lack of compelling scientific evidence, however, the indications for metal chelation were never clearly defined. To determine the safety and efficacy of chelation therapy, the National Institutes of Health funded the Trial to Assess Chelation Therapy (TACT). TACT was the first double-blind, randomized, controlled trial to demonstrate an improvement in cardiovascular outcomes with edetate disodium therapy in patients with prior myocardial infarction. The therapeutic benefit was striking among the prespecified subgroup of patients with diabetes.

Design

We review the published literature focusing on the atherogenic nature of diabetes, as well as available evidence from clinical trials, complete and in progress, of metal chelation with edetate disodium therapy in patients with diabetes.

Results

The TACT results support the concept that ubiquitous toxic metals such as lead and cadmium may be modifiable risk factors for cardiovascular disease, particularly in patients with diabetes.

Conclusions

The purpose of this review is to discuss the potential mechanisms unifying the pathogenesis of atherogenic factors in diabetes with toxic metal exposure, and the potential role of metal chelation.

This review summarizes the epidemiological and basic evidence linking toxic metal accumulation and diabetes-related cardiovascular disease, supported by the salutary effects of chelation in TACT.

On 4 November 2012, the results of the 10-year National Institutes of Health–funded Trial to Assess Chelation Therapy (TACT) were presented at the American Heart Association (1). Remarkably, infusions of an edetate disodium–based infusion reduced recurrent cardiovascular events in the study cohort composed of nondiabetic patients (63%) and patients with type 1 and 2 diabetes (37%) with prior myocardial infarction (MI). Subsequent analyses identified the largest effects [41% reduction in major cardiovascular events (P = 0.0002), and a 43% reduction in total mortality (P = 0.011)] in patients with diabetes (2). The salutary effect of edetate disodium–based chelation in the prevention of cardiovascular disease appears to be substantially larger than with other interventions in recent cardiovascular outcome studies (EMPA-REG, CANVAS, SUSTAIN-6); however, the TACT findings have received relatively little attention. Although metals may contribute to atherosclerotic cardiovascular disease in all types of diabetes, most data to date, including TACT, have addressed the role of chelation in a population that by far predominates in type 2 diabetes. Therefore, we think that our discussion is most relevant to type 2 diabetes. Despite current advances in cardiovascular disease prevention and care, patients with diabetes still develop atherosclerosis at an accelerated rate despite optimal therapy (3), a phenomenon known as residual risk. Numerous atherogenic risks coexist in the diabetic population, perhaps accounting for the residual risk. First, endothelial dysfunction is potentiated in diabetes. This is a multifactorial phenomenon, with a decline in nitric oxide–mediated vascular reactivity as well as enhanced endothelial expression of adhesion molecules in the initial phases of plaque formation (3–6). Second, diabetic dyslipidemia also plays a role owing to the combination of impaired lipid influx/efflux equilibrium, a distinct diabetes atherogenic lipid phenotype, and a resulting increased lipid core of atherogenic plaque in patients with type 2 diabetes. Part of the diabetic lipid phenotype is incremental formation and uptake of oxidized low-density lipoprotein (LDL). Oxidized LDL is a modified lipoprotein toxic to endothelium and a potent chemoattractant for macrophages. It is normally degraded by paraoxonase (PON), a high-density lipoprotein (HDL)–associated esterase/lactonase found mainly in serum, but the degradation may be diminished in diabetes as a result of a lower and/or modified HDL cholesterol (3, 6, 7). This phenomenon along with the higher macrophage apoptosis may explain the observed larger necrotic cores in lesions from individuals with diabetes vs nondiabetic patients (3, 7, 8). Fourth, to further complicate the situation, thrombogenicity might be enhanced, probably through elevation of plasminogen activator inhibitor levels, prothrombotic substances, and increased platelet aggregation. Increased protease activity has also been reported, provoking instability in the fibrous cap of the lesion and greater intraplaque hemorrhage as a result (3, 4, 7–9). The underlying and unifying pathogenic basis for many of these atherogenic factors in diabetes may be directly related to hyperglycemia, insulin resistance, oxidative stress, inflammation, or a combination of all of these.

The purpose of this review is to discuss the potential mechanisms of the beneficial effects of chelation, and to put these findings in context by focusing on the atherogenic nature of diabetes, as well as the role that oxidant metals, both essential and toxic, may play in the pathobiology of complications of diabetes. We analyze the available evidence from clinical trials, complete and in progress, of metal chelation with edetate disodium therapy in the prevention of diabetic-associated cardiovascular disease.

Design and Results of TACT With Emphasis on the Diabetes Population

A chelating drug is an organic molecule that has a pocket with a negative charge and as such is able to capture metal ions, bind them, and permit renal or biliary excretion, thus removing the metal from the body. Natural chelators, such as the heme molecule, are essential for vital processes, binding iron within the Hb molecule. Edetate disodium, a synthetic amino acid, has a high affinity for ions with +2 to +6 valences. Edetate and its salts (disodium, calcium disodium, and others) may bind toxic and essential metals, including divalent cations such as calcium and magnesium, and enhance their renal excretion. Transition metals are elements with a partially filled electron subshell. Their interchangeable prevalent ionic form, the catalytic metal ion, makes them useful cofactors in biochemical redox reactions. However, these same chemical properties can also be responsible for uncontrolled redox reactions in the cell, leading to cell poisoning. Nonessential metals such as cadmium, lead, and mercury are toxic metals generally at lower levels of exposure. Toxicity depends on chemical species, absorbed dose, route of exposure, and duration of exposure. The edetates were first patented by Munz in 1938 and were used as lead chelators for industrial poisoning and as treatment of hypercalcemia (10). Serendipitously, at a time when there were no effective treatments for atherosclerosis and diffuse coronary calcification was common in autopsy specimens, Clarke et al. (11, 12) reasoned that decalcification of these arteries might be beneficial. The first publication reporting edetate disodium infusions for cardiovascular disease was in 1956 (11). Clarke et al. used edetate disodium in 20 patients with atherosclerotic disease and reported improvement of angina symptomatic relief in 19. Starting in the 1990s, small clinical studies followed with mixed results (13). These enrolled, in aggregate, <400 patients, had surrogate endpoints such as angina or claudication and short follow-up periods (14). In 2002, the Cochrane Collaborative reviewed the literature to date and concluded that there was insufficient evidence to declare whether EDTA chelation was beneficial or harmful for patients with atherosclerotic disease (15).

In the clinical arena, chelation persisted in alternative medicine practice (16) although persistent questions regarding safety and efficacy remained (17). This uneasy equipoise prompted the National Center for Complementary and Alternative Medicine (now the National Center for Complementary and Integrative Health) to release a $30 million request for applications cofunded by the National Heart, Lung, and Blood Institute of the National Institutes of Health.

Design of TACT and overall results

TACT was designed to test the most prevalent chelation solution in use. As published, it included not only up to 3 g of edetate disodium, but also 7 g of ascorbic acid, B vitamins, procaine, 500 U of unfractionated heparin, magnesium, and other electrolytes. Moreover, most patients receiving chelation also received high doses of oral multivitamins and multiminerals (OMVM). Therefore, a factorial design was selected for TACT, in which eligible patients were randomly assigned to one of four groups (1): (i) active IV chelation infusions plus active OMVM; (ii) active IV chelation infusions plus placebo OMVM; (iii) placebo IV chelation infusions plus active OMVM; and (iv) placebo IV chelation infusions plus placebo OMVM.

Based on discussions with experts in the chelation community, TACT included 30 weekly infusions followed by 10 maintenance infusions up to 8 weeks apart. OMVM or placebo was to be taken daily throughout the duration of the trial. Eligible patients were 50 years of age or older, had a prior MI >6 weeks before the enrollment date, and had a serum creatinine level ≤2.0 mg/dL.

The primary endpoint for the study was a composite of all-cause mortality, myocardial infarction, stroke, coronary revascularization, and hospitalization for angina. The major secondary endpoint was cardiovascular mortality, stroke, or recurrent MI.

TACT enrolled 1708 participants (839 randomly assigned to chelation and 869 to placebo) at 134 sites in the United States and Canada, of which 37% had diabetes. Study patients had a median age of 65 years, 18% were female, 9% were nonwhite, and 83% had either a prior coronary bypass or percutaneous coronary intervention. At baseline, 92% were taking antithrombotic therapy (aspirin, clopidogrel, or warfarin). The median LDL was 89 mg/dL, with 73% of patients taking statins at baseline.

A total of 55,222 placebo or active infusions were administered. Seventy-six percent of participants completed at least 30 (of a total of 40 possible) infusions. All 40 infusions were completed by 65% of participants. The median follow-up was 55 months. Active infusions of edetate chelation reduced the primary composite endpoint by 18%: hazard ratio (HR), 0.82; 95% CI, 0.69 to 0.99; P = 0.035; 5-year number needed to treat (NNT) to prevent one event was 18.

Results in patients with diabetes

The results of active infusion of edetate disodium chelation in the 37% of patients with diabetes (n = 633) showed a prespecified analysis of a 41% reduction (HR, 0.59; 95% CI, 0.44 to 0.79; P = 0.0002) in the primary endpoint during 5 years compared with placebo (Fig. 1A). The 5-year NNT to reduce one primary event during 5 years in this prespecified population was 6.5. All individual components of the primary endpoint were consistently lower in the active IV chelation group (Table 1). All-cause mortality was reduced by 43% (10% vs 16%; HR, 0.57; 95% CI, 0.36 to 0.88; P = 0.011; 5-year NNT to prevent one death was 12; Fig. 1B), as was the principal secondary endpoint (cardiovascular death, reinfarction, or stroke; 11% vs 17%; HR, 0.60; 95% CI, 0.39 to 0.91; P = 0.017; Table 1) (2). There was a 52% relative reduction in the risk of recurrent MI among patients with diabetes (HR, 0.48; 95% CI, 0.26 to 0.88; P = 0.015). There was no difference between groups in control of glucose as defined by fasting glycemia during the course of the initial 30 infusions, suggesting that glycemic control was not an explanation for these findings. These findings ultimately led to the funding of the second TACT (TACT2) by the National Center for Complementary and Integrative Health, the National Heart, Lung, and Blood Institute, the National Institute of Diabetes and Digestive and Kidney Diseases, and the National Institute of Environmental Health Sciences.

Figure 1.

TACT Kaplan–Meier estimates of the primary composite endpoint. EDTA chelation therapy vs placebo: (A) subset of patients with diabetes mellitus; (B) mortality in patients with diabetes mellitus by infusion group.

Table 1.

Clinical Endpoints by Infusion Arms for Patients With Diabetes Mellitus

Endpoint	EDTA Chelation (n = 322)	Placebo (n = 311)	HR (95% CI)	P Value
Primary endpoint	80 (25%)	117 (38%)	0.59 (0.44–0.79)	<0.001
Death	32 (10%)	50 (16%)	0.57 (0.36–0.88)	0.011
MI	16 (5%)	30 (10%)	0.48 (0.26–0.88)	0.015
Stroke	4 (1%)	3 (1%)	1.19 (0.27–5.30)	0.829
Coronary revascularization	48 (15%)	62 (20%)	0.68 (0.48–0.99)	0.042
Hospitalization for angina	5 (2%)	6 (2%)	0.72 (0.22–2.36)	0.588
Secondary endpoint	35 (11%)	52 (17%)	0.60 (0.39–0.91)	0.017
Cardiovascular death	19 (6%)	27 (9%)	0.63 (0.35–1.13)	0.118

What is the mechanism underlying the benefit seen?

If edetate can reduce the clinical consequences of atherosclerotic coronary artery disease, questions arise as to what possible mechanisms might be responsible. Additionally, edetate seems to be limited in the ways it might affect pathophysiology by its confinement to the vascular space and its rapid excretion in the urine. The “chemical claw” is thought to do one thing in the body: bind cations and facilitate their excretion. To investigate whether edetate disodium chelated toxic metals, Waters et al. (18) collected 24-hour urine for 2 days prior to and 2 days after an edetate disodium–based infusion and analyzed the samples for various toxic and essential metals. Following the infusion, the excretion of lead during 2 days increased by 3830%, and cadmium increased by 514%. Arenas et al. (19) performed a similar experiment in 26 TACT-eligible patients, 13 (50%) with diabetes. Toxic metals were detected in the baseline urine of >80% of patients. There was no increase in urine metal excretion following a TACT placebo infusion. Following an active infusion, however, levels of urine metals increased (lead by 3835%, P < 0.0001; cadmium by 633%, P < 0.0001; nickel by 373%, P < 0.0001; and aluminum by 275%, P < 0.001). These findings demonstrate that edetate disodium mobilizes lead and cadmium, the most vasculotoxic of the abovementioned metals, from their chronic storage compartments and facilitates their excretion. Whether these effects of chelation are the mechanism for its therapeutic action depends, of course, on whether accumulation of these metals is important in the atherosclerotic disease process. The major findings of TACT, particularly in patients with diabetes, merit a brief review of the atherogenicity of diabetes, with special emphasis of the role that edetate-chelatable metals may play.

Diabetes, Oxidative Stress, and Metals

Glycoxidation

There is substantial evidence suggesting that diabetic complications may arise, at least partially, as a result of uncompensated oxidative stress. Hyperglycemia leads to an increased production of reactive oxygen species (ROS) and to nonenzymatic glycoxidation of proteins, which alters their structure and function (4, 5). This process is initiated by the oxidation of an aldose or ketose to a more reactive dicarbonyl sugar (glucosone), which will then react with a protein to form a ketoimine adduct. The resulting reduced oxygen products, such as superoxide and hydrogen peroxide, may cause oxidative damage to neighboring molecules (7, 20). Therefore, autoxidative glycosylation requires, or at least is enhanced by, metal catalyzed oxygen chemistry to produce oxygen free radicals. The products of these reactions have been termed advanced glycation end-products (AGEs) owing to the interplay between glycation and oxidation in their formation (21, 22).

The effects of AGEs can be direct or receptor-dependent. Given that advanced glycation occurs during a prolonged period of time, structural components of connective tissue matrix and basement membrane components (i.e., collagen IV) as well as other long-lived proteins such as myelin, tubulin, plasminogen activator 1, and fibrinogen may be targeted (21, 23). Alternatively, the receptor-dependent effects of AGEs can be mediated by different proteins or receptors that bind these components, with receptor of the advanced glycation end-products (RAGE) being the best characterized. RAGE, once activated, triggers secondary messenger pathways such as protein kinase C. Nuclear factor κB is a key target of the RAGE pathway and results in the increased transcription of intercellular adhesion molecule-1, E-selectin, endothelin-1, tissue factor, vascular endothelial growth factor, and proinflammatory cytokines as well as upregulation of RAGE itself. In endothelial cells, RAGE activation may activate additional damaging pathways involving nicotinamide dinucleotide phosphate oxidase p21RAS and MAPKs, further enhancing the production of ROS (21).

The impact of AGEs in diabetic complications was studied in type 1 diabetes in the DCCT Skin Collagen Ancillary Study Group, which compared intensive vs conventional glucose management and performed skin biopsies in a subset of the DCCT cohort. There was an association of AGEs with microvascular complications of diabetes. The association of microvascular complications with AGE measures persisted after adjustment for chronic HbA1c levels, and glycated collagen (furosine) had the strongest correlation with complications. This suggested that AGEs were more proximal to the cause of diabetic complications. This implicated tissue catalytic and inhibitory factors such as metals and oxidative stress that determine their formation rate (24). In another study, serum levels of AGEs were elevated in patients with type 2 diabetes when compared with nondiabetic controls and patients with diabetes and coronary heart disease demonstrated higher levels of AGEs in general (25).

Glycation products formed on intact proteins might bind transition metals, possibly allowing them to participate in reduction/oxidation reactions (1, 23). Of note, redox-active metals are sequestered in nonreactive forms (chelates) within proteins in vivo, being transported this way through compartments. However, metal homeostasis may become dysregulated during hyperglycemia, as glycated proteins may have lower capacity to form strong complexes with metals, potentially increasing their activation tendency in redox processes (7, 23, 26). This “glycochelates” hypothesis was tested in vitro by Qian et al. (23) by comparing the iron and copper binding capacity of three proteins (albumin, gelatin, and elastin) in their natural vs glycated state. Binding capacity was enhanced in the glycated forms, as was the redox activity of the bound metals. If this mechanism were active in vivo, it might produce redox-active glycochelates in areas such as the elastic laminae of arteries and elsewhere, potentially causing long-term consequences and contributing to the genesis of many diabetes complications. Thus, there is some evidence suggesting that metals may play an important role in the complications of diabetes, and the results of TACT suggest that macrovascular complications are responsive to metal chelation. Qian et al. (23) concluded that perhaps these oxidation reactions that take place in diabetes are not just limited to copper and iron but to transition metals as well. The chelation of metals in TACT2 may be providing a direct means of testing this hypothesis.

Lipoxidation

Oxidative modification of LDL cholesterol is considered a crucial step in the pathogenesis of atherosclerosis and might be enhanced in patients with diabetes, suggesting a relationship between hyperglycemia and accelerated oxidation (27). Cell-mediated oxidation of LDL depends on the balance between cellular pro-oxidants [i.e., reduced form of NAD phosphate (NADPH) oxidase, lipoxygenase, myeloperoxidase] and antioxidants (glutathione system, superoxide dismutase) as well as the balance between pro-oxidants and antioxidants in lipoproteins themselves (7). Traces of transition metal ions in free form or in redox-active complexes are generally agreed to be essential for the production of oxidized LDL. LDL oxidation starts with the consumption of its antioxidants, the role of which is to scavenge lipid peroxyl radicals. After this depletion occurs, an initiating free radical extracts a hydrogen atom from one of the polyunsaturated fatty acids contained in the LDL lipids. Then, metal ions catalyze a series of propagation reactions, which include the breakdown of lipid hydroperoxides and the formation of aldehydes. As a result, the apolipoprotein B100 molecule undergoes oxidative modification, affecting its charge and dimensional configuration, which renders it unable to bind to the LDL receptor, binding to the CD36, scavenger receptor A, and lectin-like receptor family instead (7, 28–30).

LDL oxidation by macrophages has a major role during the early stage of atherosclerosis. Macrophages have the ability to reduce iron or copper ions, which then rapidly react with lipid hydroperoxides, leading to the formation of reactive lipid radicals and the conversion of the metal back to its oxidized form. Superoxide ions are required for the LDL oxidation initiation, and in macrophages their main source is the NADPH oxidase system (7, 31).

In diabetes, LDL particles are characteristically small and dense and are considered more atherogenic owing to their capacity to penetrate the intima, poor affinity to the LDL receptor, and increased susceptibility to oxidation and glycation (32). At the same time, glycated LDL or AGE-LDL also has proinflammatory activity, mediated by its interactions with MCP-1, RAGE and the scavenger receptors, and Toll-like 4 receptor, triggering signaling pathways to produce proinflammatory cytokines (32, 33). In this context, in vivo studies have observed an increased level of plasma lipid hydroperoxides in patients with diabetes, with a decrease in antioxidant capacity (measured by the total radical-trapping antioxidant potential) not associated with HbA1c levels (25). Furthermore, glycation of HDL in diabetes may inversely impact its vasoprotective functions such as reverse cholesterol transport, antioxidant, anti-inflammatory, and vasodilatory effects (7, 29–33). For example, and as mentioned previously, PON protective action against lipid peroxidation in lipoproteins, macrophages, and atherosclerotic lesions might be impaired in diabetes (7, 34). It has been shown in vitro that glucose inactivates PON-1 and glycated PON-1 has reduced ability to hydrolyze membrane hydroperoxides (7). Also, several studies have found an inverse association between chronic exposure to different metals (such as cadmium, lead, and mercury) and PON-1 activity (7, 35, 36). Therefore diabetes can be considered as a state of increased free radical production and compromised total plasma antioxidant capacity, resulting in enhanced oxidative stress. Metal-catalyzed oxygen chemistry, therefore, may be important in establishing and enhancing oxidative state.

Effect of Metals on Atherosclerotic Disease and Diabetes Mellitus

Lead

Lead has also been extensively linked to cardiovascular outcomes of atherosclerotic origin, including coronary disease, hypertension, stroke, and peripheral vascular disease.

Animal studies have shown fatty degeneration and sclerotic changes on arterial walls of rats exposed to lead (37). In humans, there is a direct relationship of blood lead levels with cardiovascular events and cardiovascular mortality (37, 38). The National Health and Nutrition Examination Survey (NHANES), a large prospective, representative cohort of the US population, showed a significant association between blood lead level and mortality from both myocardial infarction and stroke at lead levels <10 µg/dL (39). Several epidemiological studies have also shown a significant association between blood lead level and systemic blood pressure (40, 41). In the Normative Aging Study, longitudinal analysis of bone lead levels was associated with increased systolic blood pressure and increased rate of all-cause mortality driven largely by cardiovascular causes, particularly ischemic heart disease (41, 42). A recent analysis of NHANES data estimated ∼250,000 deaths from cardiovascular disease attributable to low level lead exposure with concentration <5 µg/dL (43). In this study, participants who had a high blood lead level were also more likely to have elevated levels of serum cholesterol and higher rates of hypertension and diabetes. Although not a transition metal, lead shares the role of transition metals in potentiating oxidative stress by favoring the formation of ROS and increasing lipid peroxidation. As previously explained, experimental studies have shown an association between lead exposure and decrease in major antioxidant enzymes such as glutathione peroxidase, glutathione reductase, superoxide dismutase, and catalase (44, 45). It has been observed that lead exposure in lead-acid battery factory workers decreased serum PON-1 activity, further contributing to the atherosclerosis process (38). Chen et al. (46) found that patients with type 2 diabetes and higher lead body burden treated with EDTA chelation therapy showed a significant delay in the progression of diabetic nephropathy. However, no specific study has analyzed the relationship of lead exposure and diabetic cardiovascular disease and atherosclerosis. The design of TACT2, with metal levels assayed several times during the trial, should clarify this relationship.

Cadmium

Cadmium is a toxic metal with widespread population exposure and a long intracorporeal half-life (47–49). Besides its known toxic effects, cadmium has been associated with numerous diseases such as hypertension, kidney disease, diabetes, and atherosclerosis, particularly peripheral artery disease (47). For example, cadmium was associated with increased vascular endothelial permeability by inhibition of endothelial cell proliferation and induction of cell death in vitro (50). Also, Ferramola et al. (36) demonstrated that rats exposed for different intervals of time to cadmium showed an increased level of lipid peroxides, a product of oxidative stress (measured by thiobarbituric acid reactive substances as byproducts of lipid peroxidation), vs controls, which increased even further with longer exposure. At the same time, an initial increase in PON-1 activity was observed in early stages, followed by decreased levels with longer exposure time. In NHANES, urine cadmium concentration, a marker of cumulative exposure, was associated with cardiovascular mortality in men, but not in women (51). Messner et al. (50) measured the carotid intima media thickness of young healthy women and serum metal concentrations. The study showed an independent association between high cadmium levels and increased intima media thickness, and thus a marker of early atherosclerotic vessel wall thickening. Epidemiological studies have also shown an association between cadmium and hypertension similar to the effect of lead (52, 53). Navas-Acien et al. (54) also showed that blood lead and cadmium levels were strongly associated with increased prevalence of peripheral vascular disease in the NHANES population (55). In the Strong Heart Study, a positive correlation between cadmium, cardiovascular mortality, and coronary heart disease was observed, even after adjusting for traditional cardiovascular risks factors (56). Although the association was nominally stronger in the diabetes subpopulation, albeit not statistically significant presumably owing to sample size, the diabetic population might be more susceptible to the effect of toxic metals. In this context, the NHANES III cross-sectional study found a strong positive correlation between urinary cadmium levels, prediabetes, and diabetes, although cigarette smoking, a confounder, also showed a strong correlation with impaired glucose metabolism (57).

Copper and iron

The edetates enhance excretion of copper, another active participant in oxidative reactions that may be particularly damaging in diabetes. Copper is an essential metal. Free copper ions are highly redox active and might contribute to tissue damage by the generation of ROS (7). Copper exists in biological systems primarily as Cu⁺, which is mostly found intracellularly, and Cu²⁺, the main copper form in extracellular matrix. The main copper-binding proteins in plasma are ceruloplasmin (95%) and albumin. However, regulation of copper metabolism might be abnormal in diabetes (58, 59). In vitro studies have shown that chronic hyperglycemia can damage the copper-binding properties of ceruloplasmin and albumin. When incubated with glucose, ceruloplasmin is fragmented and there is a time-dependent release of its bound copper, which then might participate in a Fenton reaction to produce hydroxyl radicals (60). This and the fact that AGEs can bind and localize catalytically active Cu²⁺ might explain some of the increased oxidative stress seen in diabetes and its ensuing cardiovascular complications (7, 58–60). Thus, another mechanism for edetate benefit may be chelation of free copper.

Iron is also intimately linked to oxidative stress, despite its other important functions. Similar to other transition metals, it can also participate in the Fenton reaction on its ferrous form (Fe²⁺) to create highly toxic free radicals and produce lipid peroxides. In diabetes, glycation of transferrin decreases its ability to bind ferrous iron and, owing to the now increased pool of free iron, indirectly stimulates ferritin synthesis (61). Glycated holotransferrin is additionally known to facilitate the production of free oxygen radicals that further amplify iron’s oxidative effects. The toxic iron fraction is usually stored as ferritin. It can be released by the action of reducing agents that convert Fe³⁺ into Fe²⁺, which in turn increases ferritin translation. This is consistent with the observation of increased ferritin levels in patients with diabetes, although, as an acute phase reactant, ferritin levels may also reflect underlying inflammation. Additionally, its protein apoferritin exerts ferroxidase activity, catalyzing the oxidation of Fe²⁺ to Fe³⁺, allowing iron storage and preventing oxidative stress propagation. However, when antioxidant concentrations are low, there is a rapid release of iron from ferritin and a downregulation of ferroxidase activity in this setting, resulting in an enhancement in free radical generation. In this context, animal and human studies in diabetes showed that selective iron chelators ameliorate endothelial dysfunction (7, 61–63). Edetate disodium increases iron excretion. Whether edetate disodium reduces catalytically active iron (Fe³⁺) or copper, however, has yet to be determined.

Arsenic and mercury

Although the impact of other transition metals and metalloids on atherosclerosis has been studied, evidence in the population with diabetes is limited. For example, arsenic, a metalloid, has been shown to have dose- and time-dependent inhibitory effects on proliferation and viability of endothelial cells and may cause endothelial dysfunction, leading to apoptosis via induction of p21 (47–49, 64). Also, arsenic induces oxidative stress by increased production of ROS (especially superoxide), depletion of antioxidant factors such as glutathione, and indirectly on its methylated form by releasing redox-active iron from ferritin. Therefore, there is an increase in lipid peroxides, enhancing further arsenic-induced endothelial dysfunction and atherosclerosis (64). Arsenic, however, is not effectively chelated by the edetates.

There have been numerous cohort studies on the association of mercury with cardiovascular disease, given this metal’s ample distribution and exposure, especially in marine products. These observations are confounded. Mercury is found in many forms of seafood, but the consumption of seafood is associated with a lower cardiovascular event rate. Moreover, mercury is also not effectively removed from the body by the edetates, so we will not cover the extensive literature on this very interesting metal, other than to point out that the potential atherosclerotic mechanism of mercury, besides the shared pathobiology of transition metals, might involve its effect on selenium and formation of mercury selenide, impairing selenium function as a cofactor for glutathione peroxidase. Also, methylmercury may inactivate catalase, glutathione, and superoxide dismutase, causing further oxidative burden (65).

Chelating Properties of Common Drugs Used in Diabetes

Several medications that are routinely used in patients with diabetes, such as angiotensin-converting enzyme inhibitors (ACEIs), angiotensin II receptor blockers (ARBs), metformin, and carvedilol, among others, have chelating properties (22, 66–68) [Table 2 (53, 57, 59, 62, 67–69)]. These chelating properties should be viewed anew in the post-TACT era. As previously mentioned, chelators may inhibit the metal catalyzed oxidation reactions that form AGEs. In 1986, Brownlee et al. (69) introduced aminoguanidine as the first AGE inhibitor that trapped chemical intermediates in the formation of AGEs. It was a weak chelator that required high doses to achieve efficacy in vivo, which led to toxic effects related to vitamin B6 deficiency, thus hampering its clinical use (70). Subsequently, in search for other drugs with chelating characteristics, Miyata et al. (71) found that olmesartan, an ARB, and temocaprilat, an ACEI, significantly reduced the in vitro production of two AGEs, pentosidine and N-carboxymethyllysine, acting primarily by inhibiting their oxidative formation. They analyzed the effect on metal ions and reported that olmesartan chelated copper, inhibiting the catalyzed auto-oxidation of ascorbic acid. A total of six additional ARBs and four ACEIs were tested in their study with similar AGE-lowering potential and similar mechanisms.

Table 2.

Oral Medications and Their Chelating Properties

Drug	Target	Effects
ACEIs	Inhibit in vitro the metal-catalyzed oxidation of ascorbic acid, increased activities of superoxide dismutase	Decrease AGE formation (57, 62)
ARBs	Inhibit in vitro the metal-catalyzed oxidation of ascorbic acid	Decrease AGE formation (62)
Metformin	Interaction with mitochondrial copper to exert its effects in AMP-activated protein kinase, traps α-dicarbonyl	Decrease in gluconeogenesis (67, 68)
Carvedilol	Inhibition of electron adduction, inhibiting lipid peroxidation in myocardial cell membranes, inhibiting neutrophil release of O2, scavenging peroxychlorous and hypochlorous radicals	Possible role in cardioprotection (69)
Deferoxamine	Chelates iron; intercepts ROS	Helps decrease diabetes-induced endothelial dysfunction (53)
Vitamin C	Intercepts ROS	Decrease in fasting insulin levels (59)

Other studies have shown that in addition to renal protection by ACEIs or ARBs such as ramipril, benzapril, and valsartan in hypertensive and diabetic rat models there was also a decrease in AGE accumulation and expression of RAGE in the kidney, as well as decreased expression of NADPH oxidase and other biomarkers of oxidative stress (72–74). Forbes et al. (72) randomized diabetic rats to ramipril (3 mg/L), aminoguanidine (1 g/L), and placebo and followed them for 12 weeks. In rats treated with ramipril and aminoguanidine, renal AGE accumulation was significantly reduced along with renal nitrotyrosine, a marker of protein oxidation, suggesting that both compounds inhibit AGE through effects on ROS production. Thus, these two series of experiments may be interpreted to suggest that the renoprotective effects of ACEIs and ARBs might be secondary to their inhibition of metal-mediated oxidative stress. In another study in diabetic-induced rats, captopril and enalapril were shown to increase activities of the antioxidant enzyme, superoxide dismutase, another mechanism for reducing the cytotoxic effects of enhanced oxidative stress (75).

A prime example of this hypothesis is metformin’s chelating properties, which are possibly key to its metabolic effects in diabetes (76, 77). Several studies have also shown that metformin reduced diabetes-associated vascular risk, demonstrating benefits beyond the expected glucose-lowering effect. Metformin acts as a ligand for several divalent transition metals but has greater affinity for copper (76). This drug might interact with mitochondrial copper to exert its effects in AMP-activated protein kinase and other pathways to inhibit glucose production (76). Additional evidence suggests that metformin also reacts with α-dicarbonyls, reducing toxic dicarbonyls and AGEs (77). Carvedilol, widely used for its cardioprotective properties, also contains an antioxidant (carbazol) compound in its molecular structure. Adding to its potential role as a chelator, this includes: (i) inhibiting electron adduction, (ii) inhibiting lipid peroxidation in myocardial cell membranes, and (iii) inhibiting neutrophil release of O₂ and other antioxidant properties (78). Additionally, iron chelators such as deferoxamine have both shown beneficial effects on endothelial dysfunction in diabetes animal models by interrupting the ROS generating cycle (62). Mild chelating activity might be the basis for the health effects in diabetes of citrus foods and some nutraceuticals (79). Nagai et al. (80) examined the effect of citrate, a natural chelator found in citrus fruits, on diabetic mice observing an inhibition of nephropathy (albuminuria), accumulation of AGEs in lens, and delayed cataract development. This evidence showed that even a common natural dietary chelator can inhibit AGE formation and protect against diabetic complications. Thus, the underlying role of catalytic metal ions in oxidative stress may serve as a target for AGE inhibition and potential mechanism of many of these drugs.

TACT2

Based on the results of TACT, the National Institutes of Health has funded TACT2, a replicative placebo-controlled, factorial clinical trial that will assess the effects of edetate disodium–based chelation therapy and high-dose oral multivitamins and multiminerals on cardiovascular outcomes in patients with diabetes (type 1 or 2) with a prior myocardial infarction. Based on conservative estimations of event rates and effect size, TACT2 will have excellent power to detect a clinically important difference between treatment groups. Additionally, and different from TACT, TACT2 has a strong mechanistic component focusing on toxic metals, and it is funded to maintain a biorepository. Thus, the questions of mechanisms that have arisen following TACT should be answered. The trial is in its early phases of enrollment, and additional clinical sites are being recruited.

Summary

The results of TACT strongly suggest that heavy metals, either as catalysts of oxidative stress or some other mechanism, are an emerging therapeutic target in cardiovascular disease, especially in patients with diabetes. Professionals specializing in the care of patients with diabetes, atherosclerosis, and cardiovascular diseases should be familiar with the potential role that metal chelation with edetate disodium may have as a strategy to reduce residual atherosclerotic risk in optimally treated patients.

Glossary

Abbreviations:

ACEI

angiotensin-converting enzyme inhibitor

AGE

advanced glycation end-product

ARB

angiotensin II receptor blocker

HDL

high-density lipoprotein

HR

hazard ratio

LDL

low-density lipoprotein

MI

myocardial infarction

NADPH

reduced form of NAD phosphate

NHANES

National Health and Nutrition Examination Survey

NNT

number needed to treat

OMVM

oral multivitamin and minerals

PON

paraoxonase

RAGE

receptor for advanced glycation end-products

ROS

reactive oxygen species

TACT

Trial to Assess Chelation Therapy

TACT2

second TACT