Angiotensin-converting enzyme 2 as a novel target for gene therapy for hypertension


Corresponding author M. J. Katovich: Department of Pharmacodynamics, University of Florida, College of Pharmacy, PO Box 100487, Gainesville, FL 32610-0487, USA. Email:


Less than one-third of patients with hypertension have their blood pressures (BP) controlled with current traditional therapeutic approaches for the treatment and control of hypertension. Pharmacological approaches may have reached a plateau in their effectiveness and thus newer innovative strategies need to be studied not only to increase the number of patients that can achieve BP control, but also to find a way to cure, not just manage, the disease. Continuous advances in gene delivery systems coupled with the completion of the Human Genome Project, now make it possible to investigate genetic means for the treatment and possible cure for hypertension. The renin–angiotensin system (RAS) has long been known to regulate BP, and salt and water metabolism. This system is unique in having both a peripheral circulating system and a tissue-based system. Each of these components have been ascribed a variety of physiological effects that have been associated with not only an increase in BP, but also in a variety of the pathophysiological manifestations associated with hypertension, such as cardiac hypertrophy and kidney dysfunction. We and others have used an antisense gene therapy approach, targeting the classical components of the RAS, to effectively attenuate the development of hypertension and related cardiovascular pathophysiologies in numerous experimental models of hypertension. Recently other components of the RAS have been elucidated and some of these components may be potential targets in a gene therapy approach. This article will focus on angiotensin-converting enzyme 2 (ACE2) as a new, potential target of gene therapy for hypertensive disorders.

Hypertension is defined as a condition of persistently high systolic blood pressure (BP) of at least 140 mmHg and/or a diastolic BP greater than 90 mmHg, or a condition in which BP is controlled to recommended guideline levels of 140/90 mmHg utilizing antihypertensive medication (Deshmukh et al. 1998; Chobanian et al. 2003). This condition of a sustained elevation in BP contributes to serious health complications such as myocardial ischaemia and infarction, renal failure, stroke and retinal damage (Stamler et al. 1993; Whelton, 1994). In addition, hypertension is a very complex disorder with a majority of hypertensive patients being insulin resistant and/or salt sensitive (Sowers, 1991; Ferrannini & Natali, 1991; Reaven, 1991). Recently Fields et al. (2004) has estimated that the number of people in the US with hypertension has increased by approximately 30% over the last decade. The total prevalence rate (percentage) for hypertension for 1999–2000 in the US is 31.3%, which represents at least 65 million adults, a substantial increase from a decade ago. This disorder accounts for 6% of all deaths worldwide and is the most common treatable risk factor for cardiovascular disease (Murray & Lopez, 1997).

The major goal of current therapy for hypertension is to control BP and prevent the complications (end-organ damage) associated with the disease. However, despite the large arsenal of pharmacological antihypertensive agents currently available, successful control of BP (≤ 140/90 mmHg) is only observed in a small percentage of patients. In the US, only 34% of hypertensive patients have their BP controlled to the recommended level of ≤ 140/90 mmHg with pharmacological intervention (Chobanian et al. 2003). This low percentage of individuals controlled with conventional therapy is more troubling when overwhelming evidence from numerous clinical trials clearly elucidates the lifesaving benefits gained from normalizing BP with the use of antihypertensive medication. This lack of control of BP leads to thousands of unnecessary deaths each year. Therefore, proper treatment and management of hypertension is of critical importance to society and should be a matter of extreme urgency worldwide.

Rationale for gene therapy for hypertension

Currently there are five major drug classes used in the treatment of hypertensive patients: diuretics; calcium channel blockers; β-blockers; and two separate drug classes that inhibit portions of the RAS. Despite all of these therapeutic agents, conventional pharmacological therapy does not control the disease in a majority of patients. Possible reasons for poor control of BP include: education; economics; availability of health care; incorrect targets of therapy, poor follow-up care; drug availability; and/or compliance. Compliance has been defined as ‘the extent to which a person's behaviour coincides with medical care or advice’ (Haynes et al. 1979). In the US it has been estimated that compliance in hypertensive patients may be as low as 30–50%, with compliance decreasing over a patient's lifetime (Rudd, 1995). One of the main reasons for non- or partial compliance is that many patients experience unpleasant side effects from their medications (Benson & Britten, 2002). Patients with hypertension tend to have low compliance rates because this disease is generally asymptomatic. Patients actually can ‘feel better’ when they are not on their medication, and they therefore either reduce or stop taking their medication. Thus, we are at a stage in the treatment of hypertension where more focus must be placed on other novel treatment paradigms rather than further development of conventional drugs. Gene therapy offers the possibility of producing long-term therapeutic effects with specificity based on the particular genetic target, which would minimize side effects. Likewise, compliance would no longer be a significant issue, as this type of therapy could be administered a minimal number of times over the patient's lifetime. Gene therapy may be the next frontier for the treatment and possible cure of complex diseases such as hypertension. Because progress is being made so quickly in the area of molecular biology and virology, it is becoming readily apparent that more emphasis needs to be placed on the discovery of appropriate target genes by physiologists.

Targets of the renin–angiotensin system for gene therapy

The RAS is one hormonal system in which dysregulated expression and hyperactivity has been associated with the development and maintenance of hypertension. Both the systemic (endocrine) and tissue (paracrine/autocrine) versions of the RAS contribute to hypertension (Dzau, 1988; Bader et al. 2001). This system is also involved in aspects of insulin resistance (Yavuz et al. 2003; Henriksen & Jacob, 2003), nitric oxide metabolism (Liu & Persson, 2004), oxidative stress (Zhou et al. 2004) and vascular smooth muscle and cardiac hypertrophy (Higashi et al. 2003; Yamakawa et al. 2003). The classical understanding of the RAS has led to the use of angiotensin-converting enzyme (ACE) inhibitors and angiotensin II type I receptor blockers (ARBs) as therapeutic agents in the treatment of hypertension. Because these agents are so effective, one could hypothesize that genetic manipulation of this system to inhibit signalling within the RAS may, in principle, be an ideal method to attempt a genetic cure for this disease. In addition, numerous studies have demonstrated links between genes of the RAS (i.e. angiotensinogen, renin, ACE and angiotensin I receptor (AT1R) and hypertension in various populations (Luft, 2002; Zhu & Cooper, 2003). Therefore, we and others (Iyer et al. 1996; Lu et al. 1997; Phillips et al. 1997; Peng et al. 1998; Wang et al. 1999; Reaves et al. 1999, 2003; Pachori et al. 2000, 2002; Katovich et al. 2001; Yamakawa et al. 2003) have used an antisense mRNA technique to successfully down-regulate transcription of the ACE and/or the AT1R and thereby prevent the development of hypertension in both genetic and non-genetic models of experimental hypertension. Not only was BP reduced, but cardiac hypertrophy, fibrosis, perivascular necrosis in cardiac tissues, and endothelial dysfunction were prevented in hypertensive animals treated with antisense to the AT1R (Iyer et al. 1996; Lu et al. 1997; Wang et al. 1999; Pachori et al. 2002). Although most studies have focused on the AT1R subtype, there is an additional receptor subtype that can be activated by angiotensin II, AT2R (Chung et al. 1996; Matsukawa & Ichikawa, 1997). Generally the actions of the AT2R opposes those of the AT1R (Chung et al. 1996; Matsukawa & Ichikawa, 1997; Gallinat et al. 2000). Using gene therapy approaches, we have recently demonstrated that inhibition of expression of the AT2R by antisense mRNA can elevate BP in Sprague-Dawley rats (Wang et al. 2004) and over-expression of the AT2R is cardio-protective in both genetic and non-genetic models of hypertension (Metcalfe et al. 2004; Falcon et al. 2004), without having significant effects on BP. Because expression of the AT2R is significantly elevated in the heart with this technique (Metcalfe et al. 2004), we speculate that this beneficial effect is mediated at the tissue RAS level. Although we have been successful in targeting three components of the RAS (AT1, AT2 and ACE), the RAS is much more complex, and new discoveries regarding angiotensin degradation fragments (Cesari et al. 2002; Roks & Henning, 2003) frequently provide more potential targets for gene therapy in hypertension. Figure 1 identifies most of the components of the RAS, and thus identifies some current and future targets for gene therapy.

Figure 1.

Schematic of the renin–angiotensin system
Possible mechanisms of action by the Ang (1–7) fragment are indicated by dashed lines. Lines ending in bars indicate inhibition, while lines ending in arrowheads indicate stimulation.

ACE2 as a novel target for gene therapy for hypertension

Genomic-based studies by two independent groups recently resulted in the discovery of ACE2 (Tipnis et al. 2000; Donoghue et al. 2000). Originally this enzyme was found in the testis, kidney and heart (Tipnis et al. 2000; Donoghue et al. 2000) but now it has also been identified in a wide variety of tissues, and is probably localized in much the same places as ACE (Harmer et al. 2002). It shares 40% homology with ACE, but differs greatly in substrate specificity, and its activity is not altered by ACE inhibitors (Tipnis et al. 2000). ACE2 is one of several enzymes that catalyse the formation of degradation fragments angiotensin 1–9 (Ang (1–9)) and angiotensin 1–7 (Ang (1–7)) from both angiotensin I and angiotensin II, respectively (see Fig. 1).

It has been suggested that Ang (1–7) antagonizes the action of angiotensin II directly at the AT1R as well as indirectly via other pathways, such as antagonizing ACE, and its plasma and tissue concentrations are increased during both ACE inhibition and ARB treatment (Donoghue et al. 2000; Zhu et al. 2002; Stegbauer et al. 2003; Yagil & Yagil, 2003; Ishiyama et al. 2004). It has been speculated that Ang (1–7) may contribute to the antihypertensive effects of therapeutic agents used to block the RAS (Iyer et al. 1998). It also has been demonstrated that Ang (1–7) in the heart has beneficial actions on cardiac contractility, coronary perfusion and endothelial dysfunction (Tallant et al. 1999; Strawn et al. 1999; Ueda et al. 2000; Clark et al. 2001; Ferreira et al. 2001; Loot et al. 2002). Liomar et al. (2004) has demonstrated effective vasodilatory actions of Ang (1–7) in the mesenteric artery and we have demonstrated similar action in in vitro utilizing the aorta (Fig. 2). Collectively, these findings suggest that ACE2, and its enzymatic product, Ang (1–7), are unique in the RAS cascade, as they appear to set the balance of pressor/depressor tone of the RAS and have the potential to be cardio- and reno-protective.

Figure 2.

Aortic vascular smooth muscle relaxation response to Ang (1–7)
Intact aortic rings isolated from ovariectomized rats were preconstricted with 10−7m phenylephrine, then exposed to increasing doses of angiotensin (1–7), and maximal relaxation responses were recorded. Data are significantly different (P < 0.01) from matched time controls by repeated measures ANOVA.

It has been demonstrated that in mice lacking the ACE2 gene, there is an enhanced pressor response to angiotensin II (Allred et al. 2002) and an elevation in plasma angiotensin II level (Crackower et al. 2002). ACE2 mRNA has been inversely correlated to BP in experimental models of hypertension (Crackower et al. 2002; Donoghue et al. 2003). There is also considerable evidence suggesting that ACE2 may also be critically involved in cardiac function. ACE2-deficient mice exhibit severe reductions in cardiac contractility (Crackower et al. 2002), which is completely reversed by deletion of ACE (Crackower et al. 2002). The involvement of ACE2 in cardiac function is further supported by the increase in cardiac Ang (1–7) formation in patients with cardiomyopathy (Zisman et al. 2003). Recent studies have also hypothesized a reno-protective role of ACE2 in diabetes (Tikellis et al. 2003). Thus, it is conceivable that the balance between ACE and ACE2 may be a pivotal mechanism in not only the regulation of BP but also in tissue pathophysiology that has been associated with the tissue RAS.

The specificity of ACE2 is not limited to angiotensin I and II, as it can also act with high catalytic efficiency on several other peptides, such as apelin-13 and apelin-32 and some of the kinin metabolites (with the exception of bradykinin), neurotensin and related peptides as well as opioid peptides, such as dynorphin (Vickers et al. 2002). These products have a variety of functions and thus ACE2 may play a key role in inflammation, neurotransmission and cardiovascular functions. Although ACE2 acts on several substrates, the production of Ang (1–7) appears to have the most relevant role in cardiovascular control (Roks et al. 1999; Allred et al. 2000). This new knowledge of the RAS suggests that over-expression of ACE2, or over-expression of the Ang (1–7) receptor are potential targets for gene therapy in the treatment of hypertension.

ACE2 has generated considerable interest and its potential tissue effects on the generation of Ang (1–7) would make it an interesting therapeutic target for hypertension. Omapatrilat is a first-generation mixed vasopeptidase inhibitor that displays hypotensive actions in the spontaneously hypertensive rat (SHR), stimulates ACE2 expression and activity, and leads to an increase in Ang (1–7) levels (Ferrario et al. 2002). Unfortunately, serious side effects of omapatrilat have ruled out its application for therapy in human hypertension. However, data generated from studies with this compound have been sufficient to interest the pharmaceutical industry to conclude that ACE2 is indeed promising enough to warrant research aiming at modulating its expression and activity (Dales et al. 2002). Availability of potent and selective inhibitors and activators of ACE2 can provide the pharmacological tools to help elucidate the physiological role of ACE2 and to investigate the therapeutic potential of ACE2 in treating diseases of the cardiovascular system. Thus, ACE2 may emerge as an important player in hypertension research and cardiovascular therapeutics in the next few years. As several clinical trials (McKelvie et al. 1999; Pahor et al. 2000; Yusuf et al. 2000; Sica, 2004) have implicated greater therapeutic effects of tissue alterations of the RAS (when compared to the circulating RAS), one would speculate that over-expression of tissue ACE2 could represent a highly effective strategy for cardiovascular disease management. This over-expression may be difficult to achieve using standard pharmacological agents that have short half-lives, and thus a gene therapy approach may be the only way to achieve successful therapeutic outcomes. Further, over-expressing this new gene target while simultaneously applying current therapeutic agents, such as ACE inhibitors and ARBs, could maximize therapy and significantly affect the healthcare of individuals with cardiovascular (and possibly metabolic) disorders.

Recently we have cloned both a secreted (sh) and membrane-bound form of the ACE2 gene into a lentiviral vector for gene over-expression studies (Huentelman et al. 2002; Coleman et al. 2002). Preliminary studies have shown > 95% transduction in several cell lines (unpublished data). We have also demonstrated that a secreted human ACE2 (Lenti-shACE2) transduced into human coronary artery endothelial cells results in an increase in ACE2 activity in the media, and a shACE2 gene transfer in vivo results in an increase in ACE2 secretion in blood (Huentelman et al. 2004). In addition, preliminary data from our laboratory have recently demonstrated that we can over-express mouse-derived ACE2 in the heart of Sprague-Dawley rats (Fig. 3). Animals over-expressing ACE2 did not develop cardiac hypertrophy or fibrosis (Fig. 4) when made hypertensive with chronic angiotensin II treatement; however, this tissue over-expression did not prevent the elevation in BP (authors' unpublished results). We anticipate that by having both a secreted and membrane-bound ACE2, we may now be able to deliver this ACE2 gene where the RAS is not recognized as important for cardiovascular physiology (such as a leg muscle), and thereby differentiate the effects of tissue and systemic over-expression of ACE2 on both BP control and related cardiovascular disease management. We plan to develop specific promoters and appropriate regulatory gene switches to regulate the expression of the transgenic ACE2. Thus, we believe that ACE2 over-expression could represent a potentially important treatment opportunity for hypertension and related cardiovascular diseases, and that a cell-selective, regulatable lenti vector system utilizing such a promising novel target transgene could bring gene therapy for hypertension one step closer to application in man.

Figure 3.

Mouse ACE2 expression in cardiac tissue of rats
Cardiac tissues from adult rats previously treated at 5 days of age with either green fluorescent protein (GFP)- or mouse-ACE2-lenti virus were used for real-time RT-PCR to determine transgene expression. ACE2 expression was normalized to matched sample ribosomal subunit 18S expression (smaller arbitrary unit indicates greater expression). *P < 0.05 versus control GFP.

Figure 4.

Representative ventricular fibrosis in angiotensin II-induced hypertensive rats treated at 5 days of age with either GFP- or mouse-ACE2-lenti virus
Left ventricles were isolated from the animals, postfixed, and stained with Maisson's Trichrome. Pictures represent one animal from each group. Similar results were observed from other animals in the groups (n= 3–4/group). Analysis indicates that ACE2 completely prevents the development of myocardial fibrosis in angiotensin-induced hypertensive rats.



This work was supported by grants from the National Institutes of Health (HL68085 to Colin Sumners and M.K.R.; HL56921 to M.K.R. and M.J.K.) and the American Heart Association, Florida/Puerto Rico Affiliate (M.J.K.), and an American Heart Pre-Doctoral Fellowship (J.G.). We thank Adam Mecca for technical support.