Numerous Antiretroviral Drugs That Control The Disease Biology Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

Thirty years after the discovery of HIV infection, there are numerous antiretroviral drugs that control the disease when administered in a potent combination referred to as Highly Active Antiretroviral Therapy (HAART). This therapy reduces the viral load and improves immune system reconstitution, leading to a significant reduction of HIV-related morbidity and mortality. However, HAART does not completely eliminate HIV, so treatment must continue throughout the patient's life. Prolonged use of HAART has been related to long-term adverse events that can compromise patient health. These deleterious effects have been reported for the majority of antiretroviral drugs and are the most common causes for therapy discontinuation. In most of these adverse events, such as diabetes, cardiovascular diseases, neurological disorders and metabolic alterations, oxidative stress and mitochondrial impairment play important roles. This review covers the implication of antiretroviral drugs in the overproduction of reactive oxygen species and the reduction of antioxidant defences, and in the consequent mitochondrial dysfunction, focusing on the molecular mechanisms involved and the clinical implications for HIV-infected patients.

Current availability of numerous and effective antiretroviral drugs has significantly changed the prognosis of HIV infection. The approved compounds are grouped in six different families depending on their mechanisms of action (Fig.1). According to current guidelines, these drugs should be administered in a potent combined therapy called Highly Active Antiretroviral Therapy (HAART). There are three major families of drugs: nucleoside reverse transcriptase inhibitors (NRTI), non-nucleoside reverse transcriptase inhibitors (NNRTI) and protease inhibitors (PI). Although the most recommended combination is two NRTI combined with either a NNRTI or a boosted PI [1,2], the strategy employed should depend on the status of the infection and/or patient characteristics. The effectiveness of HAART has led to a significant decline in morbidity and mortality among HIV-infected patients due to a reduction of the viral load and an increase in circulating levels of CD4+ T cells, which generates a general improvement in the immune system reconstitution [3,4]. However, this therapy does not completely eliminate HIV but rather controls the infection, which means that patients must continue treatment for life. Long-term exposure to these drugs is associated with several adverse events, including rash, hepatotoxicity, metabolic disturbances, central nervous system (CNS) side effects and cardiovascular diseases, but the contribution of the each of the different families used in HAART to the appearance and development of such toxicities is still unclear (Fig.2).

Mitochondria exert crucial functions in human health and disease and there is a growing awarenes of the importance of mitochondrial dysfunction in a wide variety of pathological situations. The potential mitochondrial sites that can be involved in the onset of this dysfunction are alterations in mitochondrial DNA (mtDNA) replication, membrane structure, inhibition or uncoupling of electron transport and oxidative phosphorylation (OXPHOS) or inhibition of ¢-oxidation. Interestingly, many HAART-associated deleterious effects have been shown to develop as a result of mitochondrial dysfunction and are related to oxidative stress, a condition arising from an imbalance between the production of reactive oxygen species (ROS) and the systems to scavenge them. Clinical data suggest an important role of HAART in the oxidative status of HIV-infected patients. Specifically, increased serum oxidant levels and a reduction of serum total antioxidant status have been reported in HIV patients under antiretroviral treatment [5]. Intracellular generation of ROS occurs in different cellular compartments, but mitochondria are the main source of ROS in the majority of cell types. These molecules develop an important role in many cellular signaling pathways, in which they act as second messengers. However, when overproduced, they are also toxic metabolic byproducts implicated in several human pathologies [6].

Mitochondria and oxidative stress

Mitochondria play an important role in glucose and fatty acid oxidation to synthetize energy, and are the most important source of ROS in the majority of eukaryotic cell lines in both physiological and pathological conditions. ROS are small, short-lived and highly reactive molecules formed by incomplete one-electron reduction of oxygen, and include molecules such as superoxide (O2.-), hydroxyl (HO.) and peroxyl (RO2.) radicals, and hydrogen peroxide (H2O2). They are generated by aerobic cells as by-products of numerous metabolic reactions and are present in many different cellular compartments and situations. Cellular ROS sources include mitochondrial electron transport chain (ETC), NADPH oxidase, peroxisomes, cytochrome p450, glucose oxidase, xanthine oxidase, cyclooxygenase, lipooxygenase and -glutamyl transpeptidase, as well as several non-enzymatic autooxidation processes. In addition, there are several exogenous stimuli capable of inducing ROS generation, such as hyperoxia, smoke, anoxia/reoxygenation, mineral dusts, toxins and ionising radiation [7]. Low to moderate levels of ROS are important players in several cellular and developmental processes, and are involved in intracellular signaling pathways. The maintenance of these levels is mediated by several cellular mechanisms, such as enzymatic and non-enzymatic systems, whose function is to scavenge ROS, prevent ROS formation and repair cellular damage produced by oxidative stress. The main antioxidant enzymes include peroxidases, superoxide dismutase (SOD), thioredoxin and catalase, and the non-enzymatic defences are comprised of different molecules, such as glutathione (GSH), vitamin A, C and E, coenzyme Q, cyt c, NADPH, melatonin, uric acid, ¡-keto acids, bilirubin, carotenoids and metal-binding proteins. An imbalance between ROS generation and these antioxidizing mechanisms produces an accumulation of these molecules and leads to oxidative stress, a condition in which cellular constituents, including proteins, DNA and lipids, are oxidized and damaged [8]. These effects have been correlated with aging and an increasing number of diseases (cancer, diabetes, neurodegenerative diseases, inflammation and heart failure) [9] (Fig.3).

Mitochondrial ROS are produced mainly at the mitochondrial respiratory chain. The incomplete reduction of oxygen during respiration generates superoxide (O2- -), which can be converted to hydrogen peroxide (H2O2) and the highly reactive hydroxyl radical (OH-). Although a huge variety of mitochondrial reactions can produce superoxide, the two major sites for superoxide production are complexes I and III of the respiratory chain [10]. The molecular source of O2-‑ at Complex III is believed to be the ubisemiquinone radical intermediate (QH・), formed during the Q cycle at the Qo site of Complex III. In the case of Complex I, the mechanisms responsible are unclear, with several domains having suggested as the site of ROS generation. It seems that Complex I and III actually generate two different pools of superoxide, with Complex I releasing it in the matrix, and Complex III releasing it in the intermembrane space [11]. However, many stimuli have been described to increment ROS production in the mitochondria, including disruption of the ETC, alterations in O2 concentration and „™m, an increment in the NADH/NAD+ ratio, decompartmentalization of Ca2+, lack of ADP, as well as the presence of some endogenous and exogenous molecules such as cytokines (TNF-¡€© and drugs (rotenone, cyanide) [12].

A direct involvement of ROS and oxidative stress in mitochondrial apoptosis has been demonstrated, and, in addition to HIV infection, many compounds can trigger intrinsic apoptotic signaling through induction of ROS production [6]. ROS and mtDNA damage promote the permeabilization of the mitochondrial outer membrane and the release of apoptogenic factors (cytochrome c, AIF or Smac/Diablo) to the cytosol, triggering caspase-dependent or caspase-independent cytosolic signaling events. Mitochondrial GSH is an essential antioxidant defence for the maintenance of mitochondrial redox status, thus protecting against mitochondrial dysfunction and cell death. Under physiological conditions, mitochondrial GSH is able to cope with the stress derived from many apoptotic stimuli. However, depletion of mitochondrial GSH below a certain threshold compromises ROS detoxification and leads to its accumulation, which can result in the activation of apoptotic pathways [13]. In this regard, mitochondrial GSH depletion has been associated with cellular damage and cell death in a increasing number of pathologies [14-16].

Mitochondrial dysfunction and oxidative stress in human diseases

Mitochondrial toxicities are usually chronic disorders manifested in different ways and of variable severity depending on the number of damaged mitochondria within the affected cells. In this way, cells with a large number of mitochondria only manifest major cellular injuries when a significant number of mitochondria are malfunctional. Tissue specificity is another important factor that influence mitochondrial toxicities, since some tissues show a higher dependence on OXPHOS [17]. Mitochondrial dysfunction and the oxidative stress that follows it have been described in a large number of both congenital and acquired illnesses, including diabetes, cardiovascular diseases, cancer, aging and neurodegenerative disorders. Other pathologies can result from oxidative stress-induced apoptotic signalling produced by ROS increases and/or antioxidant decreases, disruption of intracellular redox homeostasis, and irreversible oxidative modifications of lipids, proteins or DNA. Then, we describe briefly some of the mechanisms by which ROS contribute to the appearance and development of several important human diseases that have also been associated to long-term antiretroviral therapy.

Diabetes and Insulin Resistance.

All the forms of diabetes are complex multifactorial diseases that are highly prevalent all over the world. Diabetic patients exhibit pancreatic ¢-cell dysfunction and insulin resistance, with subsequent chronic hyperglycemia and augmentation of ROS, leading to the aggravation of this disease. Due to their low antioxidant capacity, ¢-cells are extremely sensitive to oxidative stress, which is critically involved in the impairment of these cells during the development and progression of diabetes and also in their associated complications (atherosclerosis) [18]. The metabolic alterations of diabetes cause an enhanced ROS production in endothelial cells in both large and small vessels, as well as in the myocardium, leading to the activation of several pathways involved in the pathogenesis of these complications and to the inactivation of antiatherosclerotic enzymes (endothelial nitric oxide synthase and prostacyclin synthase). It is important to note that insulin resistance itself induces atherosclerosis and cardiomyopathy in absence of hyperglycemia due to an increased release and oxidation of free fatty acids from the adipocytes, generating an overproduction of ROS that activates the same pathways of hyperglycemia-induced ROS [19]. Insulin resistance is associated not only with diabetes and its complications, but also with fatty liver disease and the polycystic ovarian syndrome, and so, it is an essential factor in the prevention of a vast burden of premature mortality and morbidity causes [20]. A restoration of pancreatic ¢-cell function and insulin sensitivity have been demonstrated in diabetic mice after suppresion of ROS [21], which means that therapy with antioxidants is likely to exert beneficial effects on the development of diabetes and its complications. However, clinical trials with antioxidants have shown little effect on diabetes progression, and this has fuelled the search of stronger and more appropiate antioxidants in order to improve this therapy.


The lipodystrophy syndromes are a heterogeneous group of disorders which may be congenital or acquired in origin. They are characterized by changes in body fat composition, dyslipidaemia (changes in plasma concentration of cholesterol, High Density Lipoprotein Cholesterol (HDL-c), Low Density Lipoprotein Cholesterol (LDL-c), triglycerides) and a metabolic syndrome-like condition. The metabolic alterations observed in patients with lipodystrophy, mainly dyslipidaemia and insulin resistance, not only involve disturbances in adipose tissue, but also in other organs such as liver, skeletal muscle and pancreas. Complications affecting nonadipose tissue may be the consequence of a direct stimuli (HIV, drugs) as well as responses to pathogenic signals coming from adipose tissue [22]. A chronic elevation of circulating fatty acids can become cytotoxic, thus generating oxidative stress and alterations in mitochondrial structure and function [23]. The involvement of ROS in body fat changes associated with this disease has been proposed in several studies that have linked ROS-mediated cell death in adipocytes to fat wasting during lipoatrophy [24].

Cardiovascular events.

This term includes all the incidents that can cause damage to blood vessels (arteries and veins) and to the heart, ranging from vascular dysfunction and myocardial infarction to stroke. Increased production of ROS in mitochondria, accumulation of mitochondrial DNA damage and progressive respiratory chain dysfunction are associated with atherosclerosis and cardiomyopathy in humans [25]. There are several mechanisms by which mitochondrial dysfunction and ROS generation may lead to vascular dysfunction and atherosclerosis, including reduction of aerobic capacity, increase of susceptibility to ischemic injury and enhanced oxidative stress-mediated apoptosis in vascular cells. Specifically, ROS generation in vascular endothelial cells induces the oxidation of LDL and the expression of ROS-sensitive inflammatory genes (for instance, vascular cell adhesion molecule-1 and monocyte chemotactic protein-1), thereby inducing changes in endothelial structure and function and contributing to the initiation of atherosclerosis [26]. Another important mechanism by which ROS contribute to the development of vascular diseases is their direct impairment on endothelium-derived nitric oxide bioactivity, which contributes to the initiation and progression of endothelial dysfunction [27]. Endothelial mitochondria maintain the regulatory balance between mitochondrial calcium concentration, nitric oxide and ROS production and so, alterations in such important processes could lead to atherosclerosis and heart disease [28].

Neuro-psychiatric effects.

Many studies suggest an important role of mitochondria in aging-related neurodegenerative conditions such as Alzheimer's disease, Parkinson's disease, multiple sclerosis, Huntington's disease and amyotrophic lateral sclerosis. In spite of being a heterogeneous group of disorders, they are all characterized by a progressive and selective loss of structure or function of neurons, including neuronal cell death. The mechanisms involved in neurodegeneration are multiple, and consistent evidence shows that production of ROS is an important process by which mitochondria contribute to neurodegeneration [29]. One of the proposed mechanisms is that mitochondrial dysfunction contributes to the damage of vulnerable genes in the aging brain. The promoters of these genes are both more sensitive to oxidative stress and deficient in repair, effects that could be enhanced by mitochondrial dysfunction, which increases ROS and/or decreases the availability of ATP [30]. The occurrence of oxidative stress during neurodegeneration is also confirmed by the detection of aldehyde adducts in brains with these diseases. These adducts are advanced lipid peroxidation end products that induce protein dysfunctions and alter cellular responses. Although the rate of oxidation and the aldehyde adduct formation is usually low, it increases with ageing, as well as with the decrease in antioxidant defences [31]. Importantly, ROS also play a critical role in the pathophysiology of several neuropsychiatric disorders such as schizofrenia or bipolar disorder. In this regard, recent studies have reported alterations in antioxidant enzymes and augmented products of lipid peroxidation in patients with the latter disorder [32].


Several mechanisms are known to produce liver injury, and oxidative stress is one of them. Mild to moderate degrees of hepatic steatosis are present in patients with antiretroviral-related liver injury and are likely to be a predisposing factor in drug-related toxicity. Liver steatosis enables the formation of ROS and oxidized lipids which, if sustained, can lead to a severe disease called non-alcoholic steatohepatitis (NASH) [33]. In fact, oxidative stress activates a variety of proinflammatory stimuli which promote adhesion and infiltration of polymorphonuclear cells and contribute to the progression of the disease. Altered mitochondria have been found in livers of patients with NASH, which display decreased activities of mitochondrial respiratory chain complexes, oxidative degeneration of mitochondrial DNA and oxidative modified proteins [34]. The importance of oxidative stress in NASH has been ascertained by findings in clinical studies and animal models which have suggested that treatment with antioxidants alleviates lipid accumulation and liver injury [35,36]. The weakening of antioxidant defences has important consequences in hepatocyte survival, which is a key aspect of liver disease. The disruption in mitochondrial GSH levels produces an accumulation of ROS and oxidative stress but also alters cellular redox status and modulates multiple signaling pathways, producing either cell death by apoptosis or necrosis or sensitization to other cell-death stimuli [37]. All these processes have been implicated in the pathogenesis of many liver diseases.

Oxidative stress and antiretroviral therapy-associated adverse events

Many HAART-related complications are thought to be attributable to toxic effects on mitochondria [38], and recent research has suggested an involvement of oxidative stress in some of these adverse events [39-42]. Each family of antiretroviral drugs exerts different side effects. Moreover, most studies point to some compound-specific alterations that are unrelated to pharmacological group (for instance, CNS adverse events have been frequently described in patients undergoing an EFV-based therpay, but not in those taking other NNRTI). This variability, together with the fact that antiretroviral drugs are administered in combination and that drug-induced mitochondrial toxicities are tissue specific, has complicated the assessment of the specific mitochondrial toxicity profile of each of the compounds.

Some of these complications have been widely linked to mitochondrial dysfunction and oxidative stress. This review describes the involvement of antiretroviral drugs in cellular ROS generation and the following oxidative stress, and focuses on the molecular mechanisms responsible and on the clinical manifestations of this damage.

Nucleoside Reverse Transcriptase Inhibitors (NRTI)

NRTI are analogues of native nucleosides and require intracellular phosphorylation to be active. The phosphorylated forms lack the 3'-OH group, and so their incorporation into the nascent chain during DNA replication by reverse transcriptase produces an inhibiton of the viral enzyme and the termination of viral DNA replication. However, the triphosphate forms of these analogues are also potential substrates for the DNA polymerase  (Pol-), the only enzyme responsible for mtDNA replication and maintenance, and can also provoke the termination of the DNA chain during mtDNA replication. NRTI treatment compromises mtDNA replication and interferes with the synthesis of essential proteins of the mitochondrial ETC, generating OXPHOS inhibition and a decrease in cellular bioenergetics. Mitochondrial dysfunction is also manifested by changes in the mitochondrial membrane potential and increased ROS generation [43]. mtDNA depletion leads to an impaired synthesis of mtDNA-encoded respiratory chain polypeptides, which can partially block the flow of electrons in the respiratory chain. As a result, they accumulate in complex I and III, where they react with oxygen to form the superoxide anion radical [44]. These effects have been described with almost all the NRTI, with dideoxy-NRTI (zalcitabine, didanosine and stavudine -ddC, ddI and d4T-) proving to be the most potent inhibitors of Pol- [45].

Increased ROS production has been reported in adipose tissue in HAART-associated lipodystrophy syndrome patients. Interestingly, the thymidine analogues have been implicated in peripheral lipoathropy in several studies in vitro and in vivo that have demonstrated these drugs to induce mitochondrial dysfunction, increased oxidative stress and fat inflammation and loss. The thymidine analogues zidovudine (AZT) and d4T have been described to produce mitochondrial changes and oxidative stress in fibroblasts and adipose cells, and to be partly involved in the premature senescence observed in these cell types after treatment with AZT and d4T. None of these parameters were changed after treatment with abacavir (ABC), lamivudine (3TC), ddI or tenofovir disoproxil fumarate (TDF) [46]. The same NRTI were associated with increased ROS and cytokine production in differentiated adipocytes and macrophages, suggesting a potential role for these compounds in lipoathropy and/or insuline resistance [39].

Hepatotoxicity is a main adverse event in HAART-treated patients, and is associated with the three major families of antiretroviral drugs [47,48]. Liver toxicity caused by NRTI can be inflicted through different mechanisms, including oxidative stress [49]. The majority of NRTI can cause mitochondrial damage, and therefore have the potential to lead to liver injury, though some (AZT, ddI and d4T) have been more strongly linked to these adverse events than others. The use of NRTI (particularly AZT, ddI and d4T) has been associated in different studies with hepatic steatosis and steatohepatitis [50,51], diseases that are directly related to oxidative stress. NRTI can interact directly with cellular bioenergetics by impairing mitochondrial respiration through inhibition of complex I of the ETC. This mechanism of mitochondrial toxicity, independent of Pol  inhibition and mtDNA depletion, has been observed in HepG2 treated with AZT, ddI or ddC. These NRTI were capable of affecting complex I activity and increasing superoxide production in this complex [52]. Other studies to evaluate AZT have demonstrated an increased oxidative stress both in vitro [53] and in vivo [54]. ABC has been associated with a reduction in respiration and ATP levels in Hep3B cells that is apparently not related to complex I inhibition [55]. However this drug did not increase ROS in this human cell line.

ABC and ddI have been widely associated with a higher risk of cardiovascular events in HIV-infected patients [56,57], which has been confirmed by several in vitro studies [58]. However, the implication of mitochondrial dysfunction and oxidative stress in the induction of these effects is still unclear. ABC (and also 3TC and AZT) have been shown to significantly increase superoxide anion both in porcine pulmonary arterial rings and in human pulmonary arterial endothelial cells (HPAECs). Other kinds of cardiac complications have been described after long-term treatment with NRTI, such as cardiomyopathy due to mtDNA depletion. In one study, mice treated with ddC and AZT showed cardiomyopathy with several mitochondrial alterations, including disminished mtDNA copy numbers and enhanced myocardial formation of ROS [59]. This mitochondrial toxicity was antagonized by uridine supplementation, implicating pyrimidine pool depletion in its pathogenesis. Recent evidence obtained through the use of genetically engineered mice points to an important role of H2O2 and oxidative stress in the mitochondrial cardiomyopathy produced by AZT [60].

Protease Inhibitors (PI)

Long-term exposure to antiretroviral therapy containing PI has been linked to an increased incidence of toxicities in which oxidative stress plays a pathogenic role, such as metabolic (lipodystrophy, insulin resistance and diabetes) and cardiovascular diseases (hypertension, heart failure and atherosclerosis).

Protease inhibitor-dependent increases in atherosclerosis are thought to be a consequence of metabolic disturbances, including dyslipidemia and insulin resistance, and also of a direct action on endothelial cells and function [61]. Endothelial dysfunction is considered an initial step in the development of cardiovascular complications, and is characterized by high levels of ROS. An association between PI-containing therapies and oxidative stress have been stated in different studies with laboratory animals and endothelial cells in culture. Saquinavir (SQV) has been reported to produce toxicological effects in human endothelial cells, increasing apoptotic cell death via ROS production [62]. In one study, human aortic endothelial cells (HAECs) treated with either indinavir (IDV) or nelfinavir (NFV) for 3 days were shown to exert higher levels of ROS production following oxidative stress induced by pro-inflammatory stimuli [63]. This increased formation of ROS promoted leukocyte recruitment in this cell line, leading to an augmentation in mononuclear cell adhesion in both Jurkat and in U-937 cells. Recent studies have suggested that long-term treatment (30 days) with ritonavir (RTV) or lopinavir (LPV) boosted with RTV triggers premature senescence in human coronary artery endothelial cells by a mechanism involving accumulation of prelamin A and oxidative stress [64]. In other studies, RTV, SQV and amprenavir (APV) directly cause endothelial dysfunction in porcine coronary arteries, partially through oxidative stress, and similar results were obtained when HPAECs were treated with RTV and IDV for 24h, suggesting a contribution of these drugs to the high incidence of pulmonary artery hypertension in HIV-infected patients [65,66]. These studies also demonstrated that the endothelial dysfunction generated as a result of these treatments can be reversed by several antioxidants, including ginsenosides. An increase in mitochondria-derived ROS has also been observed following short-term incubation (6-24h) of human umbilical vein endothelial cells with IDV, both alone or in combination with AZT [67]. The same combination was evaluated in an atherogenic mouse model (C57BL/6) fed with a high-fat diet in order to analyze their involvement in premature atherosclerosis [68]. Results showed high levels of ROS and disminished GSH levels in aortic rings from the three groups evaluated (AZT, IDV and AZT+IDV) after 2 and 20 weeks of treatment. Interestingly, the combined treatment did not potentiate oxidative stress in the aortic samples more than the single drug administrations. In cardiac myocytes, RTV and LPV, but not APV and nelfinavir (NFV), acutely induced several mitochondrial changes, including mitochondrial ROS generation [69].

ROS may mediate activation of the 3 major classes of MAPKs (ERK1/2 and BMK1, c-Jun N-terminal protein kinases -JNKs-, and p38) in a variety of cell types, leading to changes in gene expression. In fact, RTV was shown to induce a ROS-mediated activation of ERK1/2 in human microvascular endothelial cells, pointing to an important role of MAPK signaling pathways in the development of PI-endothelial dysfunction [70].

Several studies have reported associations between PI and metabolic alterations, but PI-contribution to HAART-related lipodystrophy is still unclear. Lipodystrophy is a syndrome characterized by changes in body fat distribution and systemic effects similar to those observed in metabolic syndrome (hypertriglyceridemia, hypercholesterolemia, insulin resistance and type II diabetes mellitus), and its incidence is high among HIV-infected patients (approximately 25-50%) [71,72]. Mitochondrial dysfunction and the consequent increase in ROS and oxidative stress have been associated with different features of the development of lipohypertrophy and/or lipoatrophy, which vary according to the severity of the mitochondrial damage generated. IDV, NFV, LPV and RTV, but not atazanavir (ATV) or APV, increase ROS and chemokine/cytokine production, generating alterations in adipokine secretion and lipid content in human adipocytes. Only LPV and NFV have been shown to produce similar results in human macrophages [39]. Murine 3T3-L1 adipocytes chronically exposed to RTV exhibit changes in the expression of genes related to the development of dyslipidemia and lipodystrophy, such as inflammatory cytokines, genes involved in endoplasmic reticulum and oxidative stress, and apoptosis-related genes [73]. The implication of NFV in lipodystrophic syndrome was also assessed by evaluating its effect on mature adipocyte cell death. Evidence pointed to a specific NFV-mediated induction of necrosis in 3T3-F442A in an oxidative stress-dependent manner, as confirmed by the almost total reversion of this process after ascorbate administration [74]. These results may partially explain the mechanisms underlying lipoatrophy.

Patients under PI-containing HAART have high incidences of insulin resistance syndrome (IRS) and type 2 diabetes, but the molecular mechanisms involved have not been fully elucidated [75]. It is likely that several mechanisms contribute to insulin resistance during PI-therapy [76], as the inhibition of peripheral glucose uptake and changes in hepatic glucose production. Several studies have attempted to characterize the direct influence of individual PI on insulin resistance using in vitro, animal and healthy human volunteers models [77]. In vitro data with the adipocyte cell line 3T3-L1 have shown that NFV-mediated induction of oxidative stress plays an important role in adipocyte insulin resistance [78]. Short-term treatment with NFV, but not with SQV or ATV, has been shown to increase ROS generation in the pancreatic insulinoma cell line INS-1, leading to a reduction in GSH levels and ATP production, a suppressed expression of cytosolic SOD and an augmentation of mitochondrial uncoupling protein-2 levels [40].

PI-induced liver toxicity has been reported in different studies [79,80], and it seems to be inflicted by several mechanisms, as dose-dependent direct liver cell stress (tripanavir and RTV) or disturbances in lipid and sugar metabolism which can contribute to a steatohepatitis syndrome [49]. Although the cellular and molecular mechanisms of such processes have not been well established, it is plausible to speculate that ROS play an important role in them, since ROS increases and oxidative stress have been associated with both cellular damage and alterations in lipid and sugar metabolism in other tissues.

Non-Nucleoside Reverse Transcriptase Inhibitors (NNRTI)

These compounds directly inhibit HIV reverse transcriptase by inducing conformational changes in their catalytic site. They have different chemical structures and do not depend on intracellular phosphorylation to be active. Among the 4 NNRTI approved for anti-HIV therapy, nevirapine (NVP) and efavirenz (EFV) are the current cornerstones of combined antiretroviral therapy. Although first considered as safe drugs, several mild-to-severe side effects are now arising due to the long-term exposure to these compounds. The most common adverse events associated with NNRTI-based regimens are cutaneous reactions, lipid disturbances, neuro-psychiatric effects and hepatotoxicity [81-84].

HIV infection is associated with a high prevalence of disorders related to the central nervous system, such as dementia, motor impairments and behavioral dysfunctions, and similar effects has been observed with EFV [85]. Oxidative stress seems to play an important role in the pathogenesis of neurodegenerative diseases [30] and also in HIV dementia. Oxidized lipids and proteins, formed by ROS-mediated oxidation, have been detected in brain and cerebrospinal fluid of HIV patients with dementia. These patients also exhibit an accumulation of toxic substances in the cerebrospinal fluid that are capable of producing mitochondrial dysfunction in neurons (changes in „™m, induction of oxidative stress and apoptosis) [86]. A cross-sectional study demonstrated that the presence of oxidative stress in HIV patients was associated with antiretroviral therapy, and particularly with the use of EFV [87].

Treatment of human coronary artery endothelial cells (HCAEC) with EFV leads to increased oxidative stress, evident in the induction of superoxide production and decrease of GSH levels, thus significantly increasing the in vitro monolayer permeability of this cells [88]. In this study, antioxidant administration demonstrated that EFV-induced ROS also activates several cellular pathways mediated by JNK and NF«B, and pointed to an involvement of this drug in inflammatory processes. In the light of these data, it is plausible to speculate that this compound contribute to HAART-associated cardiovascular complications in HIV-infected patients.

Numerous studies have analyzed the incidence of liver toxicity in NNRTI-treated patients, as well as the associated risk factors, but the molecular mechanisms involved have been the focus of little attention. We have recently reported evidence of an acute inhibition of mitochondrial function in human hepatic cells, characterized by a reduction of mitochondrial oxygen consumption and mitochondrial membrane potential and an increase in ROS production [55,89]. EFV produces an inhibition of Complex I, leading to a reduction in intracellular ATP levels and accumulation of ROS. The decrease in cellular proliferation and viability observed in these cells may be due to the activation of the intrinsic (mitochondrial) pathway of apoptosis, and seems to be partly due to oxidative stress as some of these effects were reversed by treatment with the antioxidant Trolox. On the contrary, NVP had no effect on respiration, ROS generation, intracellular ATP levels, expression of phosphorylated-AMPK or levels of neutral lipids in Hep3B cells. These data, together with mounting clinical evidence, suggest that the mechanisms of hepatotoxicity induced by NVP and EFV are drug-specific and unrelated to NNRTI as a drug family.

NNRTI have been associated with metabolic disturbances involving lipid metabolism, such as dyslipidemia and lipodystrophy. These reactions and the molecular mechanisms involved are not fully understood, but data suggest they are drug-specific [90-92]. EFV-induced lipodystrophy may be a result of effects on adipocytes, including inhibition of lipogenesis and differentiation, whereas some in vitro studies have suggested that NVP does not inhibit lipogenesis. The increased levels of inflammatory cytokines found in EFV-treated adipocytes suggest a role for oxidative stress in these reactions, as these metabolites are enhancers of ROS production and can, in turn, mediate several signalling pathways, thus leading to recruitment of inflammation [93].

HIV-induced oxidative stress and other risk factors

There are many accompanying factors in HIV patients that can worsen antiretroviral- mediated oxidative damage. HIV itself has been shown to alter the redox balance in different tissues and, furthermore, some studies have reported that its envelope glycoprotein (gp120) and transregulatory protein (Tat) induce oxidative stress in brain endothelial cells, leading to a disruption of the blood-brain barrier, whose essential function is preserving the brain from toxic substances present in the blood and which is likely to be involved in HIV-associated dementia [86,94,95]. In this way, HIV infected individuals are continuously exposed to chronic oxidative stress. This alteration in the intracellular redox balance has been found in other viral infections, and can affect the progression of viral-induced diseases. In some cases this impairment is essential for the replication and maintenance of the virus; for instance, HIV has been reported to increase the levels of inflammatory cytokines, which can induce both a depletion of GSH and oxidative stress. This phenomenon may activate NF-«B, leading to a series of downstream signal transduction events that allow HIV expression [96].

Due to the high prevalence of HIV-HCV co-infected patients (about 30% of HIV-infected patients), the fact that oxidative stress has been described in chronic hepatitis C infection is of great relevance. Several studies have stated a HCV-induced oxidative stress in both liver samples from patients and infected hepatic cells in vitro, and have pointed to a synergistic action when a concomitant HIV infection exists [97,98]. Specifically, depleted GSH levels were observed in hepatic and plasma fractions of patients with HCV, and these changes were more pronounced in patients who also had HIV infection [99]. Recent reports have also shown that HIV and HCV independently regulate hepatic fibrosis progression through the generation of ROS, a regulation that occurs in a NF-«B-dependent fashion [100]. Collectively, these data support that HCV induction of ROS is involved in the progression of liver disease and demonstrates that coinfected patients are more likely to experience oxidative stress-induced adverse events.

Several xenobiotics have been described to induce oxidative stress. Thus, it is plausible that they exacerbate the effects of antiretroviral therapy, particularly in patients with concomitant infections. Recent data have demonstrated that cocaine can potentiate gp120 toxicity in astrocytes through a mechanism involving regulation of oxidative stress, mitochondrial membrane potential and MAPK signaling pathways [101]. A similar response has been described with other addictive drugs like methamphetamine, which potentiates gp120 and Tat proteins-induced oxidative stress by decreasing the levels of the antioxidants GSH and glutathione peroxidase in the brain [95].

Data obtained in a murine model of HIV encephalitis (a neuroinflammatory disorder) suggests that alcohol abuse is also a co-factor in the progression of HIV infection in the brain. In this model, alcohol-exposed mice exhibited enhanced oxidative stress and neuroinflammation and an impaired immune response, suggesting that alcohol abuse in HIV-infected patients exacerbates HIV CNS infection and toxicities [102]. Moreover, ethanol decreases GSH levels via both the generation of oxidants and the inhibition of the mitochondrial GSH transporter, generating an imbalance between ROS generation and antioxidants that can produce oxidative stress. This mechanism links alcohol intake with the development of hepatic inflammation and fibrosis [103].

In this regard, it is plausible to speculate that other drugs and compounds inducing oxidative stress may potentiate the susceptibility and progression to ROS-mediated detrimental effects in HIV-infected patients under antiretroviral therapy, aggravating their clinical manifestations and challenging the design of efficient therapeutic strategies.

Concluding remarks

Although the adverse reactions associated with the long-term use of anti-HIV drugs have become a major issue and now constitute the basis of numerous clinical trials, the molecular mechanisms of these toxicities have been the focus of little attention. Besides their well established involvement in the development and progression of many human diseases, increasing evidence suggests that mitochondrial dysfunction and oxidative stress play a critical role in the toxicity of many anti-HIV drugs. ROS are generated in HAART-treated patients due to the HIV infection itself and exposure to antiretroviral drugs. Antiretroviral therapy also undermines antioxidant defences, leading to oxidative damage in DNA, proteins and lipids, and even to cell death. However, therapeutic approaches using currently available antioxidants have not proved to be particularly effective against many of these adverse events and conflicting data have been obtained from clinical and in vitro studies. Unravelling the ROS-mediated pathways involved in the side effects of these drugs could be of great importance in developing safer drug combinations and improving anti-HIV therapy in general.