Nox1 Downregulators: A New Class of Therapeutics

Abstract

Chronic non-communicable diseases share the pathomechanism of increased reactive oxygen species (ROS) production by nicotinamide adenine dinucleotide phosphate (NADPH) oxidases, known as Nox. The recent discovery that expression of Nox1, a Nox isoform that has been implicated in the pathogenesis of cardiovascular and kidney disease and cancer, is regulated by the expression and activity of G protein-coupled estrogen receptor (GPER) led to the identification of orally active small-molecule GPER blockers as selective Nox1 downregulators (NDRs). Preclinical studies using NDRs have demonstrated beneficial effects in vascular disease, hypertension, and glomerular renal injury. These findings suggest the therapeutic potential of NDRs, which reduce Nox1 protein levels, not only for cardiovascular disease conditions including arterial hypertension, pulmonary hypertension, heart failure with preserved ejection fraction (HFpEF), and chronic renal disease, but also for other non-communicable diseases, such as cerebrovascular disease and vascular dementia, Alzheimer’s disease, autoimmune diseases and cancer, in which elevated Nox1-derived ROS production plays a causal role.

1. Therapies for chronic non-communicable diseases: Unmet needs

Chronic, non-communicable diseases are responsible for more than half of global mortality [1, 2], with half of these deaths caused by only four diseases, predominantly affecting the arterial vasculature: arterial hypertension, diabetes, coronary artery disease, and stroke [3–6]. Despite continued efforts, the prevalence of these diseases continues to rise, largely because of the obesity pandemic and population aging [7], indicating unmet needs that require additional therapeutic options.

2. Reactive oxygen species in chronic diseases

Reactive oxygen species (ROS) are highly reactive, short-lived molecules that are formed under physiological conditions and act as second messengers, modulating rapid signaling and the activity of transcription factors [8]. ROS contribute to the regulation of vascular tone, cell proliferation and, if ROS bioactivity is abnormally increased under pathological conditions, to fibrosis, vasoconstriction, thrombosis, vascular stiffening and calcification. Part of this effect is brought about by decreasing nitric oxide (·NO) bioavailability through inactivation by ·O₂⁻ at a diffusion-limited rate [9], reducing or abrogating the vasoprotective properties of ·NO that include vasodilatation and inhibition of cell proliferation, as well as inhibition of inflammation and thrombosis [9]. Numerous pathologies, such as cerebral ischemia, arterial remodeling and vascular hypertrophy, ischemia-reperfusion injury in stroke and myocardial infarction, heart failure and proteinuric renal disease, have been linked to ROS-dependent mechanisms [10].

3. Dietary supplementation of “antioxidant vitamins” without effect

Over several decades, clinical studies have explored the theoretical benefits of non-specifically reducing ROS bioactivity through supplementation with antioxidant vitamins as “scavengers” of ROS. However, such antioxidants react with superoxide anion at a rate that is one billion times slower than its reaction with ·NO [11]. Not surprisingly therefore, multiple clinical trials using dietary supplementation of antioxidant vitamins, such as vitamin C or vitamin E, found little to no effect on cardiovascular morbidity or mortality [10–14]. The failure of these studies has been ascribed to the fact that many were observational cohort studies and not randomized trials, as well as to the type, dosage, treatment duration, and administration route of the antioxidant vitamins [12–14].

4. NADPH oxidases: key sources of reactive oxygen species

Non-communicable diseases are associated with the upregulation or activation of ROS-producing cellular enzymes [8], which include nicotinamide adenine dinucleotide phosphate (NADPH) oxidase (Nox) [15], “uncoupled” endothelial ·NO synthase (generating ·O₂⁻ instead of ·NO) [9, 16, 17], xanthine oxidase, and mitochondrial respiratory enzymes [8]. Cellular ROS formation is endogenously controlled by anti-oxidant enzymes such as superoxide dismutase, catalase, and glutathione peroxidase, as well as by antioxidant steroid hormones, such as estrogens (that also downregulate the Nox-associated protein Rac1) [18], calcitriol (“vitamin” D) [19], and antioxidant vitamins [10].

Under healthy conditions, Nox represent the main cellular source of ROS in the vasculature and are the only enzymes whose primary function is to produce ROS; Nox consist of a multienzyme complex (Figure 1) of which at least 7 mammalian isoforms have been identified, including five Nox proteins (Nox1-Nox5) and two dual oxidases, Duox1 and Duox2 [20]. All Nox isoforms catalyze the transfer of two electrons from NAPDH via their FAD domain and two iron-heme prosthetic groups to molecular oxygen [21]. Nox1, Nox2, Nox3, Nox 5, and the Duox oxidases are regulated via intracellular signal transduction, whereas Nox4 exhibits constitutive activity [11, 22–24]. Overall ROS production by Nox/Duox enzymes also depends on the expression levels of the enzymatic subunit in particular, as evidenced by decreased ROS production upon GPER inhibition, which decreases Nox1 protein levels [20, 25]. Autocrine regulation of Nox1-derived ROS is also essential for the proliferation of stem cells, indicating a novel “self-regulating” function of Nox1 [26, 27], a mechanism also implicated in the self-renewal of thyroid cancer cells [28]. Nox1 is induced by growth-stimulating and pro-inflammatory mediators such as angiotensin II and endothelin-1 [29], as well as by RNA-binding proteins, such as HuR (human antigen R) [30], which is tightly regulated by hydrogen sulfide (H₂S) generated from cystathionine gamma lyase [31, 32].

Figure 1. Schematic representation of the membrane-bound Nox1 enzyme complex and its regulation by angiotensin receptor blockers (ARBs) and Nox1 downregulators (NDRs).

ARBs inhibit angiotensin II-induced assembly and function of Nox1 and its specific activator (Noxo1, Rac1) and adaptor proteins (p22_phox, Noxa1) through protein kinase C (PKC)-dependent and other rapid signaling pathways. NDRs reduce Nox1-dependent superoxide (O₂⁻) production by downregulating Nox1 protein synthesis. H₂O₂, hydrogen peroxide; TF, transcription factor. Modified from reference [20] and reproduced with permission of Elsevier.

Nox1, Nox2, Nox3, and Nox5 (all considered homologues of Nox2/gp91(phox) [33, 34]), generate superoxide. Species- and tissue-specific expression and function of the different Nox isoforms have been identified, with Nox1 and Nox2 isoforms propagating disease. By contrast, Nox4 has protective effects in the cardiovascular system [24, 35–37] and mediates its effects via formation of H₂O₂, which acts as a vasodilator through endothelium-dependent hyperpolarization, thereby controlling blood pressure [38]. Moreover, Nox4 has a physiological role in maintaining myocardial function during exercise [39]. Nox5 is found only in the human but not in the rodent genome; it is regulated in a calcium-dependent fashion and highly expressed in the human kidney and vasculature [22, 34, 40]. Studies using overexpression of human Nox5 in rodents found remarkable effects on disease pathologies. Transgenic podocyte-specific overexpression of Nox5 in rodents with diabetic nephropathy resulted in podocyte injury, early-onset proteinuria and hypertension [41], suggesting an important role for human diseases as well. Similarly, in rodents, Nox5 overexpression that was limited to either endothelial or to vascular smooth muscle and mesangial cells was associated with glomerular injury, inflammation and fibrosis [42].

5. NADPH oxidases and drug discovery

Over the past decade, drug discovery programs have focused their attention on identifying small molecule compounds that directly interfere with ROS production at their enzymatic source instead of scavenging them [10, 11, 43]. We and others have recently reviewed the current status of the development of these new drugs, as well as their potential shortcomings. Recently, a combined Nox1/4 inhibitor (GKT-831, formerly GKT-137831) entered Phase I and Phase II clinical trials for diabetic nephropathy and primary biliary cholangitis [21, 22]; however, these trials failed to show any benefits. It is likely that the disappointing results were in part related to the fact that “protective” Nox4 was also blocked by the GKT-831 compound, in addition to the “pathogenic” Nox1 isoform. In fact, Guzik and associates recently concluded that there existed an unmet need for identifying new molecules that provide isoform-specific inhibition of Nox homologues and that efforts “must focus on generating small molecular weight inhibitors of NADPH oxidases, allowing the selective inhibition of dysfunctional NADPH oxidase homologs. This appears to be the most reasonable approach, potentially much more efficient than non-selective scavenging of all ROS by the administration of antioxidants.” [11]. So far, only few compounds have been identified that allow selective inhibition of one Nox isoform [20, 25]. Schramm et al. as well as Altenhöfer et al. have recently reviewed some of the Nox inhibitors that have been developed, including VAS2870, VAS3947, GK-136901, GKT-831 and S17834 [11, 21].

6. Dual-acting drugs

The majority of drugs that have resulted from discovery and screening efforts and later entered clinical application were initially designed to target solely one receptor or enzyme. However, several drugs inhibit more than one target and may even act as a combined agonist/antagonist. After the AT₁ antagonist losartan (DuP 793) was licensed for clinical use, the active metabolite of losartan, EXP3174, was found to not only inhibit the AT₁ receptor but also the thromboxane A₂/prostaglandin H₂ receptor [44, 45]. Similarly, the endothelin receptor antagonist darusentan was found to also reduce ACE activity in certain tissues under pathological conditions [46, 47]. In the field of steroid pharmacology, drugs such as selective estrogen receptor modulators (SERMs) or selective estrogen receptor downregulators (SERDs), which inhibit or reduce levels of nuclear estrogen receptors α and β [48], were subsequently found to also act as agonists of GPER [49]. Aside from these accidental discoveries, efforts in drug development have recently been directed toward the generation of dual-acting drugs. These include dual inhibitors of cyclooxygenase / thromboxane A₂ receptor [50, 51], dual 5-HT₆ receptor / D₃ receptor antagonists [52], dual VEGF receptor / PDGF receptor antagonists [53], dual AT₁ / ET_A antagonists [54], dual AT₁ / α₁-receptor antagonists [55], dual muscarinic antagonists / β₂ agonists [56, 57], and dual norepinephrine reuptake inhibitors / 5-HT receptor antagonists [58]. With regard to Nox, dual Nox1/Nox4 and pan-Nox inhibitors, which inhibit all Nox isoforms, have been synthesized. However, pharmacological inhibition of the “protective” Nox4 isoform is not desirable as it may abrogate its beneficial physiological and disease-protective effects [24, 35–37, 39]. Accordingly, Rajaram et al. have recently called for reinforcing safety in clinical drug development: “with the emergence of pharmacological NOX4 inhibitors in clinical trials, caution should be taken in identifying potential side effects in patients prone to acute kidney injury and cardiovascular disease” [37].

7. Nox1: An emerging therapeutic target

The human Nox1 gene is located on the X chromosome, and thus some of its effects, like other X-linked genes, may determine possible sex-dependent risks of disease [59]. Nox1 (Figure 1), highly expressed in organs rich in smooth muscle or smooth muscle-like cells, such as mesangial cells (initially designated «mitogenic oxidase» [60]), has recently emerged as a promising new therapeutic target [15, 20, 25, 61, 62]. Nox1 contributes to the pathogenesis and progression of chronic, non-communicable diseases, including vascular disease, heart failure, diabetes, proteinuric renal disease, cancer / tumor angiogenesis, and fibrosis [14, 15, 61]. We have recently reviewed drug discovery efforts to identify inhibitors that block one or more Nox isoforms [20]. These include non-selective small molecules as well as peptides such as Nox1ds, a synthetic, cell-permeable peptide containing an 11-amino acid sequence of the docking site of the Nox1 activator NoxA1. This peptide prevents assembly and subsequent activation of the Nox1 multienzyme complex, but has only been shown to have activity in vitro [20, 63].

8. Constitutive, ligand-independent activity of receptors

The concept of unliganded, constitutively active G protein-coupled receptors, originally introduced by Costa and Herz in 1989 [64], is now firmly rooted in receptor pharmacology and occurs through the spontaneous population of active receptor states [65, 66]. The paradigm that a steroid receptor strictly requires activation by its ligand ligand (e.g. phosphorylation by MAPK) has also been revised [67]. Another example are certain drugs, such as the AT₁ receptor antagonist olmesartan, that mediate some of their effects via nuclear estrogen receptors even in the absence of their natural ligand estrogen [68, 69]. Ligand-independent activity of ERα has also been reported by Karas and associates, who showed that activation of unliganded ERα induces multiple genes associated with vascular injury, inhibiting endothelial cell proliferation and migration, and triggering VSMC proliferation, as well as inflammatory responses in both cell types, thereby counteracting estrogen-dependent effects mediated by ERα [70].

9. Constitutive ligand-independent regulation of Nox1 through GPER

GPER, previously known as G protein-coupled receptor 30 (GPR30) [71], was first cloned in 1991 as an orphan receptor [71] and subsequently shown to mediate rapid signaling in response to estrogen [72–75]. The creation of GPER-deficient mice and pharmacological antagonists [49, 76, 77] has facilitated the study of ligand-dependent and -independent (taken here as constitutive or basal activity in the absence of known or added specific agonists) functions of GPER. Over the last decade, we have observed that some of the effects of GPER do not require estrogens (at circulating levels seen in females) or pharmacological ligands, suggesting the possibility of “constitutive” effects mediated by this receptor [20].

The requirement of GPER expression and activity for the expression of Nox1 was a serendipitous finding, a discovery we have previously summarized in detail [20]. In brief, in view of the numerous studies using GRAs (G protein-coupled estrogen receptor agonists), we had hypothesized that with aging, ROS-related mechanisms would worsen functional responses in isolated arteries. However, contrary to our expectations, this was not the case: vascular function in aged animals lacking Gper was better than in aged wild-type animals and essentially resembled that in young animals [25, 78]. In line with these unexpected findings, we had observed that GPER deficiency reduces activation of the endothelin system in the myocardium, which normally increases with age partly due to ROS [79]. In a series of experiments, we subsequently established that expression of Nox1, whose activity depends on intracellular signaling events [8], is regulated by GPER as a hitherto unknown constitutive activator of Nox1 (but not of other Nox isoforms). Moreover, we established that GPER-dependent expression of Nox1 and the associated ROS production play an obligatory role in pathological processes associated with increased ROS activity [78, 79]. Such pathologies include cardiac fibrosis, diastolic heart failure (heart failure with preserved ejection fraction, HFpEF), arterial hypertension, and chronic renal disease, and likely other chronic non-communicable diseases [20].

In the process of this work, G36 [77] (Figure 2), originally developed as orally active small molecule GPER blocker from its precursor G15 [76], has emerged as the first ever Nox1 downregulator (NDR), capable of reducing Nox1 protein levels, thereby reducing the abundance and activity of this ROS-producing enzyme in the cardiovascular system and the kidney [20]. This mechanism of action of NDRs is different from that obtained with synthetic GPER agonists such as G-1 or its natural ligand 17β-estradiol, which increase the bioactivity of ·NO, thereby indirectly scavenging ROS (such as superoxide anion) [80], while also possibly exerting direct antioxidant effects [81]. By contrast, NDRs such as G36 lack direct antioxidant activity [25]. Thus, ligand-dependent activation of GPER by synthetic GPER agonists [82] or estrogen, as well as inhibition of constitutive/basal activity of GPER (through the reduction of Nox1 abundance) by NDRs [25] - albeit though different mechanisms and pathways - can both yield similar net beneficial effects on pathophysiology and end-organ injury [83]. Whether G36 acts as an inverse agonist to inhibit constitutive activity of GPER, or as a neutral antagonist to inhibit basal activity of GPER (due to low levels of endogenous agonist ligands, as in male mice in vivo or in estrogen-free medium in vitro) remains unclear. Nevertheless , G36 inhibits disease progression in estrogen-dependent pathologies such as endometrial cancer [84], by acting as a “classical” GPER antagonist, interfering with estrogen-stimulated endometrial epithelial cell proliferation [76, 77]. NDRs, as a new class of therapeutics, hold promise to be efficacious in a number of chronic non-communicable diseases, similar to what has been observed with downregulators of certain steroid receptors in hormone-dependent forms of cancer [48], where the abundance of the target protein is reduced by increasing its turnover [85]. The currently known pathological conditions that represent therapeutic targets for Nox1 downregulators are discussed in detail below. We will also briefly discuss other potential disease targets in which Nox1 has been identified to play a role.

Figure 2. Chemical structure of the first selective Nox1 downregulator, G36.

The small molecule compound G36 is the first selective Nox1 downregulator to reduce expression and function of Nox1, which is under constitutive control of G protein-coupled estrogen receptor (GPER). G36 was originally identified as a GPER blocker and thus has dual functions, on the one hand inhibiting estrogen-dependent pathways involving GPER, while on the other inhibiting Nox1 expression and reducing Nox1 protein levels that are activated through constitutive GPER expression independently of estrogen.

10. Potential therapeutic applications of Nox1 downregulators

10.1. Arterial hypertension and vascular disease

The first hint suggesting that GPER-dependent regulation of Nox1 could play a role in arterial hypertension came from experiments demonstrating that Gper deficiency largely abrogates ROS/superoxide production in response to angiotensin II, the prototypic inducer of Nox1 [33, 34, 60] and one of the main drivers of arterial hypertension and renal disease in humans (Figure 3). Angiotensin II directly enhances AT₁-Nox1 binding to stimulate vascular smooth muscle cell growth [86]. Nox1-derived ROS are also one of the mechanisms underlying enhanced vasoconstriction in response to angiotensin II [87]. Consistent with the genetic ablation of GPER, its pharmacological inhibition with the Nox1 downregulator G36 was similarly effective [25] (Figure 3). As it was initially unclear whether this effect was specific to Nox1 or might also involve other isoforms of Nox, we performed experiments to determine the expression of different Nox proteins in vascular smooth muscle cells incubated with G36. As shown in Figure 3C, G36 reduced the abundance of only the Nox1 protein, but not that of Nox2 or Nox4, indicating its specificity and selectivity [25]. The effects of Nox inhibition on ·O₂⁻ production were confirmed by employing the peptide inhibitor of Nox1/Nox2, gp91ds-tat [88]. The results were similar to those obtained with either Gper-deficient mice or wild-type mice treated with G36, while angiotensin II had no effect at all in animals lacking the Gper gene (Figure 3). Finally, as a proof of principle to confirm that the reduced ROS production was only due to the lack of Nox1 expression, the human Nox1 gene was introduced into Gper-deficient cells using adenovirus-mediated gene transfer, a procedure that restored the cells’ ability to produce superoxide [Fig 3].

Figure 3. Regulation of Nox1-dependent superoxide production through constitutive expression/activation of GPER in murine and human vascular smooth muscle cells in vitro.

Treatment with siRNA targeting Gper abrogated Ang II-induced superoxide production (A). Comparable effects were observed using the Nox1 downregulator G36 (B). Selectivity of G36 for the Nox1 isoform was demonstrated by a reduction of Nox1 protein, but not Nox2 or Nox4 protein (C). In contrast to wild-type cells (D), in vascular smooth muscle cells derived from mice lacking Gper, Ang II was ineffective at stimulating superoxide production (E). Conversely, adenoviral introduction of human Nox1 into cells from animals genetically lacking murine Gper restored the cells’ ability to produce superoxide anion in response to Ang II (F). Ang II, angiotensin II; siCTL, scrambled siRNA, siGPER, siRNA against Gper; NDR, Nox1 downregulator. Gper+/+, wild-type; Gper−/−, gper-deficient; tat, Nox1/2 inhibitor gp91ds-tat; AdGFP, adenoviral vector carrying sequence for control protein GFP; AdNox1, adenoviral vector carrying Nox1 sequence. *p<0.05 vs. siCTL; *† and **p<0.05 vs CTL; ***p<0.05 vs solvent CTL; ****p<0.05 vs AdGFP control protein; Modified from reference [25] and reproduced with permission of the AAAS.

To this point, we had established that the activity of one of the principal drivers of Nox1, angiotensin II, depends on the basal activity of GPER, a concept that we deemed worthy of testing in vivo as well. In a well-established model of angiotensin II-induced arterial hypertension, we recapitulated the effects of angiotensin on arterial superoxide production in vivo. Interestingly, in animals lacking the Gper gene, angiotensin II failed to elicit any changes in blood pressure, reinforcing the obligatory role of this receptor in activating Nox1 and the associated superoxide production [20, 25]. However, in animals expressing the Gper gene, angiotensin II caused robust superoxide production and increased arterial blood pressure; the GPER antagonist G36 completely prevented this increase in ROS, and attenuated the arterial hypertension induced by angiotensin II [25] (Figure 4). Importantly, G36 treatment also prevented the impaired endothelium-dependent relaxation that develops after chronic angiotensin II infusion [25], a vascular dysfunction that is an independent prognostic marker of survival in patients with coronary artery disease [89].

Figure 4. Therapeutic effects of the Nox1 downregulator G36 in vivo in a model of chronic Nox1-dependent, AngII-induced hypertension.

Chronic treatment of wild-type mice with the NDR G36 completely prevented the Nox1-dependent increase of superoxide production (A, quantified in B), and had profound protective antihypertensive effects (C). Treatment was also associated with normalization of endothelium-dependent vascular function (not shown). CTL, control, Ang II, angiotensin II. *p<0.05 vs. CTL; †<0.05 vs. Ang II. Modified from reference [25] and reproduced with permission of the AAAS.

These findings indicated not only a role for GPER in the constitutive expression of Nox1, but also introduced a new therapeutic concept, namely downregulating the abundance of the Nox1 enzyme and concomitantly reducing the ROS it produces [20]. These findings were consistent with and corroborate previous observations by Gavazzi et al., who observed that animals lacking Nox1 had slightly lower blood pressures than wild-type control animals and that Nox1 knockout animals completely lack the increase in blood pressure in response to angiotensin II infusion [90], effects identical to those of Gper deletion in this model of hypertension [25]. It is likely that Nox1 downregulators would not only have potent antihypertensive effects in patients with arterial hypertension but that they would also reduce clinical sequelae of hypertension, such as atherosclerotic vascular disease, chronic renal disease, cerebrovascular disease, stroke and others.

Angiotensin II-dependent mechanisms, which include activation of Nox1 (Figure 1), play a disease-propagating role in atherosclerosis, coronary artery disease, and coronary plaque rupture, the main cause of myocardial infarction [91], whereas the Nox4 isoform has been demonstrated to protect from this disease, causing vasodilation, lowering blood pressure and protecting from atherosclerosis [24, 35]. In this context, it is worth noting that concomitant Nox4 inhibition appears to be undesirable from the therapeutic angle. Selective Nox1 downregulators provide a novel approach to selectively interfere with Nox1 without affecting other isoforms. Selective Nox1 downregulation could also have a role in the treatment of reperfusion injury (associated with substantial ROS production) following myocardial infarction or stroke [92].

10.2. Heart Failure

Heart failure is a clinical syndrome reflecting abnormalities in myocardial contraction or relaxation resulting in failure of the heart during systole (heart failure with reduced ejection fraction, HFrEF) or diastole (heart failure with preserved ejection fraction, HFpEF). Both, HFrEF [93] and HFpEF [25, 94], are associated with activation of ROS production; however, the individual molecular pathways involved have been identified only in part [15]. We have previously observed that activation of GPER conveys protective effects in the cardiovascular system, including inhibition of atherosclerosis in ovariectomized female mice [82], largely recapitulating the protective effects of endogenous estrogens. This is not surprising as GPER mediates part of its effects with estrogens as its ligands [49], and therefore the therapeutic effects of GPER agonists may be largely confined to females [82, 95]. In order to reduce ligand-dependent effects due to the cyclic release of circulating estrogens, we decided to use male animals (where plasma estrogen levels are only a fraction of that of cycling females) and estrogen-free in vitro conditions in order to study whether and how this receptor may exert “ligand-independent” effects in cardiovascular pathologies. Circulating serum levels of estrogen, as measured by high-sensitive gas chromatography-tandem mass spectrometry, are higher in female rodents than in males, and their regulation depend critically on the presence of ERα [96]. It is currently not known how intracrine extragonadal production of sex steroids, such as estrogens or testosterone (the substrate of estrogen-forming aromatase [97]), plays a role in physiology or disease; however, preclinical and clinical studies have found correlations between circulating estrogens levels and the absence or presence of diseases such as coronary atherosclerosis [98].

Aging in humans [99] and in rodents [25] is associated with myocardial fibrosis, inflammation and left ventricular hypertrophy, all contributing to diastolic dysfunction and diastolic heart failure. Given that in our previous studies we had only observed beneficial effects using GPER agonists (GRAs), we hypothesized that genetic deficiency of GPER would result in the aggravation of any pathologies that had already developed with aging at the functional, morphological, or molecular level. To exclude the effects of ovarian steroids as ligands, the hypothesis was tested in aged male animals expressing or lacking GPER. Much to our surprise, the results were contrary to what we had expected: aged animals deficient in GPER almost completely lacked the abnormalities that developed with age in wild-type mice, and in fact organs studied in the Gper^−/− mice resembled those of 3-month old control mice used as controls, lacking left ventricular hypertrophy, myocardial fibrosis, and diastolic heart failure (Figure 5). Gper deficiency was associated with reductions in O₂⁻ production in the myocardium. Interestingly, similar to what we had observed in hypertensive animals infused with angiotensin II, abnormal endothelium-dependent vascular function that developed over two years in wild-type mice was not observed in Gper-deficient mice. Diastolic heart failure (HFpEF) is commonly found in elderly patients [99] and there is currently no drug treatment available to interfere with its progression. Interfering with the constitutive activator of Nox1, GPER, by using Nox1 downregulators, could prove to be beneficial for heart failure in patients. A direct role for Nox1 in the changes associated with aging has been corroborated by Meiljes et al., who recently reported that cellular aging in human and murine vascular cells is mediated by Nox1, and that genetic ablation of Nox1 protects against endothelial cell senescence through mechanisms involving thrombospondin-CD47 signaling [62]. More preclinical studies are required to test this concept experimentally, which is however complicated by the limited experimental models of HFpEF, particularly those including the aging heart in otherwise healthy rodents [25], and involving increased oxidative stress [94, 100, 101]. Nevertheless, the therapeutic principle offered by this new class of drugs holds promise for effective treatment of severe disease conditions, such as heart failure.

Figure 5. Constitutive GPER-dependent Nox1 expression is essential for myocardial fibrosis and heart failure with preserved ejection fraction (HFpEF).

Shown are data obtained from aged mice, resembling cardiac changes in humans. Aging was associated with increased myocardial superoxide production (A and B), cardiomyocyte hypertrophy (C) and fibrosis (C and D), left ventricular hypertrophy (E and F), and diastolic dysfunction consistent with heart failure with preserved ejection fraction (HFpEF) as determined by the E/é ratio using transthoracic echocardiography (G). Genetic ablation of Gper essentially prevented myocardial aging at the structural and functional level, completely abrogating ventricular hypertrophy, fibrosis, and HFpEF. Scale bar in panel D: 100 μm. *p<0.05 vs. Gper^+/+ Modified from reference [25] and reproduced with permission of the AAAS.

10.3. Chronic renal disease

Chronic kidney disease is becoming one of the major factors contributing to morbidity and mortality due to the increasing prevalence of diabetes and hypertension, but also due to population aging. While certain drugs have been firmly established in renal medicine to slow the progression of proteinuria and chronic renal disease, they are predominantly inhibitors of the renin-angiotensin-system, its main effector angiotensin II being a key activator of Nox1 [102]. Angiotensin II leads to activation of proliferation-and inflammation-propagating pathways, such as the endothelin system, not only in the vasculature but also in the kidney [103]. Importantly, Nox1 has been recently shown to directly contribute to angiotensin II-induced ROS-mediated aldosterone synthesis in human and rat adrenal cells, an effect that can be abrogated by silencing the Nox1 gene [103]. There is still an unmet need for new therapies that could slow down or even reverse the process of glomerulosclerosis, as shown by pharmacological inhibition of the effects mediated by endothelin-1 [104, 105] or angiotensin II [106, 107]. Similar to the heart, aging in the kidney is associated with the development of pathologic anomalies, such as focal-segmental glomerulosclerosis (FSGS). FSGS is associated with upregulation of the cyclin-dependent kinase inhibitor p21waf1/cip1, which is regulated in an endothelin-dependent fashion [104]. This appears to be an important element in the disease process since endothelin-1, like angiotensin II, is one of the known activators of Nox1 [29]. Consistent with these previous observations, Zhu et al. have recently found that in experimental diabetic chronic kidney disease, Nox1 in the renal cortex promotes renal senescence through p21^waf1/cip1[108]. Cha et al., using a “pan-Nox inhibitor” inhibiting all Nox isoforms (including Nox4), demonstrated a reduction in proteinuria, suggesting therapeutic potential of Nox inhibition in chronic renal disease [109]. We have begun to study the role of Nox1 and Nox1 downregulators in chronic renal disease. Similar to vascular smooth muscle cells, G36 reduces Nox1 abundance in cultured mesangial cells (manuscript in preparation) and protects cultured human podocytes, the gatekeepers of the glomerular filtration barrier [110], from injury in response to exogenous stimuli [111]. Moreover, similar to the heart, mice develop advanced focal segmental glomerulosclerosis with aging [111], while aged mice lacking the Gper gene are largely protected and their kidneys, for the most part, resemble those of young control animals [111]. Also, age-dependent functional abnormalities of the main renal artery are absent in Gper-deficient animals lacking the constitutive activation of Nox1 [112]. Further studies are underway to determine the mechanisms of how Nox1 downregulators can alleviate or even reverse acute or chronic renal injury in different forms of renal disease, including diabetic nephropathy, FSGS, and renovascular disease.

10.4. Nox1 as a therapeutic target in other chronic non-communicable diseases

Nox1 has been implicated in a number of non-communicable diseases that account for a considerable portion of morbidity and mortality in the general population. Possible clinical applications of Nox1 downregulators include pulmonary arterial hypertension or diseases associated with tissue fibrosis, including liver fibrosis [113] or radiation-induced pulmonary fibrosis, for which Choi et al. have recently shown protective effects of Nox1 inhibition [114]. Selective Nox1 downregulators may also have therapeutic efficacy in chronic inflammatory and autoimmune diseases such autoimmune (type 1) diabetes [115], autoimmune encephalomyelitis/multiple sclerosis [116], or forms of arthritis involving immune reactions [117], diseases all associated with an activation of Nox1. Important new areas of therapeutic application for Nox1 downregulators are neurodegenerative diseases that cause dementia, the most important ones being vascular dementia due to arterial hypertension or Alzheimer’s disease. A possible therapeutic role in the latter has been recently suggested by Nortley et al. who found that in experimental and clinical Alzheimer’s disease β-amyloid induces ROS in a Nox1/Nox2-dependent fashion [118, 119], resulting in pericyte vasoconstriction causing cerebral hypoperfusion, a pathomechanism that suggests therapeutic potential for NDRs [20, 21, 119]. Recent work by Fan et al. has implicated Nox-derived ROS in neuronal damage due to aging in mice, observations further corroborated in postmortem brain tissues of elderly adults [120]. These investigators observed that intrinsic production of angiotensin II within the brain increases with aging and that similar to deletion of the endogenous Nox1-activator GPER [25], deletion of Nox2 largely protected from aging-dependent pathologies [120]. Nox1 also contributes to islet cell dysfunction, and its inhibition reduces islet beta cell dysfunction following exposure to inflammatory cytokines in vitro, suggesting therapeutic potential for type 2 diabetes [121]. Other potential areas of clinical application of Nox1 downregulators include arterial thrombosis, thrombus formation and platelet aggregation, which is sensitive to Nox1 inhibition, possibly through direct interactions between Nox1 and the collagen receptor, GPVI [122], as well as different forms of cancer as either direct or adjuvant therapy [123–128].

11. Conclusion and perspectives

NDRs represent a new class of drugs that regulate the expression of a specific ROS-producing enzyme, thereby reducing its abundance and activity [20]. Nox1 is one of the key isoforms of NADPH oxidases involved in numerous non-communicable disease conditions. Its expression is controlled by the constitutive/basal expression and activity of GPER, a G protein-coupled receptor that previously had only been implicated in rapid and sustained effects mediated by its natural ligand estrogen [49]. NDRs provide several advantages over currently existing compounds targeting Nox pathways: inhibitors of Nox enzymes only interfere with their activity without affecting their expression, and therefore aspects such as dosage, pharmacokinetics and half-life of the drugs become important considerations, particularly with regard to clinical drug development. Moreover, several of the currently available Nox-targeting compounds that are or have been in development block more than one Nox isoform, particularly the “protective” isoform Nox4 [24, 35–37]. Aside from cardiovascular and renal disease conditions, for which preclinical studies suggest therapeutic efficacy of NDRs, future research should also be directed towards exploring the therapeutic efficacy of NDRs in other forms of non-communicable diseases, including pulmonary hypertension, cerebrovascular disease and dementia, chronic inflammatory and autoimmune diseases, and cancer where activation of Nox plays a central role [11, 23, 43, 120].