Protection from angiotensin II-induced cardiac hypertrophy and fibrosis by systemic lentiviral delivery of ACE2 in rats


  • M. J. Huentelman and J. L. Grobe contributed equally to this work.

Corresponding author M. K. Raizada: Department of Physiology and Functional Genomics, University of Florida, College of Medicine, PO Box 100274, Gainesville, FL 32610, USA. Email:


Angiotensin converting enzyme 2 (ACE2), a newly discovered member of the renin–angiotensin system (RAS), is a potential therapeutic target for the control of cardiovascular disease owing to its key role in the formation of vasoprotective peptides from angiotensin II. The aim of the present study was to evaluate whether overexpression of ACE2 could protect the heart from angiotensin II-induced hypertrophy and fibrosis. Lentiviral vector encoding mouse ACE2 (lenti-mACE2) or GFP was injected intracardially in 5-day-old Sprague–Dawley rats. This resulted in expression of mACE2 in cardiac tissue for the duration of the study. Infusion of 200 ng kg−1 min−1 angiotensin II for 4 weeks resulted in an 80 mmHg increase in systolic blood pressure, a significant increase in the heart weight to body weight ratio (HW : BW), and marked myocardial fibrosis in control rats. Transduction with lenti-mACE2 resulted in significant attenuation of the increased HW : BW and myocardial fibrosis induced by angiotensin II infusion. These observations demonstrate that ACE2 overexpression results in protective effects on angiotensin II-induced cardiac hypertrophy and fibrosis.

The discovery of angiotensin converting enzyme 2 (ACE2) has revolutionized our understanding of the renin–angiotensin system (RAS). Its discovery has stimulated the investigation of the potential role of this enzyme in the control and treatment of cardiovascular diseases (Donoghue et al. 2000). ACE2 shares ∼40% similarity of its catalytic domain with the somatic form of ACE (Donoghue et al. 2000; Tipnis et al. 2000). Despite this similarity, ACE2 differs from ACE both in its substrate specificity and in its inability to be inhibited by ACE inhibitors (Donoghue et al. 2000; Tipnis et al. 2000). Studies have demonstrated that ACE2 plays a central role in balancing the vasoconstrictor and proliferative actions of angiotensin (Ang) II at the angiotensin II type I receptor (AT1R) by increasing the concentration of angiotensin 1–7 and altering levels of other vasoactive compounds, such as apelin and the connexins (Donoghue et al. 2000; Danilczyk et al. 2003; Oudit et al. 2003; Burrell et al. 2004; Katovich et al. 2005). Alterations of the ACE2 gene or its expression are implicated in multiple cardiovascular diseases. Evidence for this includes the following: (i) disruption of the ACE2 gene in the mouse results in an elevation of Ang II, impaired cardiac contractility, and the induction of hypoxia responsive genes in cardiac tissue (Crackower et al. 2002; Donoghue et al. 2003); in addition, overexpression of Ang-(1–7) in transgenic rats delays the development of cardiac hypertrophy (Santos et al. 2004); (ii) transgenic mice overexpressing ACE2 exhibit lower systolic blood pressure (BP) (Crackower et al. 2002); (iii) ACE2 levels are decreased in several animal models of hypertension (Crackower et al. 2002; Danilczyk et al. 2003; Garcia et al. 2003); (iv) Ang-(1–7), a major product of ACE2, acts as a vasodilator and ACE inhibitor (Iyer et al. 1998; Collister & Hendel, 2003); (v) both Ang-(1–7) and cardiac ACE2 mRNA levels are altered during losartan treatment following myocardial infarction (Ishiyama et al. 2004); and (vi) the ACE2 gene maps to a defined quantitative trait locus associated with hypertension (Crackower et al. 2002). Collectively, these observations led us to propose that ACE2 inhibition could exacerbate cardiovascular pathologies while its overexpression may result in cardiovascular benefits. The objective of our present study was to evaluate this by testing the hypothesis that ACE2 overexpression would protect the heart from angiotensin II-induced cardiac hypertrophy and fibrosis.


Cloning of murine ACE2 in lentiviral vector and production of lenti-mACE2

Mus musculus ACE2 (mACE2) cDNA (Komatsu et al. 2002) was used as a template in polymerase chain reaction (PCR) amplification with the use of the following primer sequence obtained from GenoMechanix LLC (Gainesville, FL, USA): ACE2 NheI: sense, 5′-AAGCTAGCATAGCCAGGTCCTCCTGGCTCCTTC-3′; ACE2 SalI antisense, 5′-AAGTCGACCTAAAAGGAAGTCTGAGCATCATCACTG-3′. mACE2 amplification product was cloned into PCR-BluntII-TOPO vector (Invitrogen, Carlsbad, CA, USA). ACE2 coding sequence was excised with NheI-Sal I and subcloned into the same site in the lentivirus cloning vector pTY.EF1.IRES.EGFP as previously described (Huentelman et al. 2004b). The presence of an internal ribosome entry site (IRES) element permits the expression of two gene products from a single promoter. Note that the control construct (lenti-GFP) contains all sequence elements except for the therapeutic transgene (mACE2). The constructs are diagramatically depicted in Fig. 1A. Lentiviral particles were prepared as previously described (Huentelman et al. 2002; Coleman et al. 2003). Viral medium was titrated using an HIV-1 p24 antigen ELISA assay (Beckman Coulter, Fullerton, CA, USA) according to manufacturer's instructions. Viral vectors yielded titres of ∼1 × 1010 infectious units ml−1.

Figure 1.

Lentiviral vector constructs used in the present study and ACE2 enzyme activity in lenti-mACE2-infected astroglial cells
A, control vector construct, lenti-GFP (top) and experimental vector construct, lenti-mACE2-GFP (bottom). LTR, long-terminal repeat; SIN LTR, self-inactivating LTR; F, cPPT DNA flap; hEF1; human elongation factor 1 α promoter; IRES; internal ribosome entry site; GFP, enhanced green fluorescent protein; and Ψ, Psi packaging signal. B, astroglial cells in primary culture were established in 12-well cultures dishes essentially as previously described (Lu et al. 1995). One-week-old cultures were incubated with 10 MOI of lenti-mACE (○) or lenti-GFP (•) with 8 μg ml−1 polybrene for 12 h at 37°C (Huentelman et al. 2002; Coleman et al. 2003; Huentelman et al. 2004b). Culture media were removed; cells were fed with fresh media and allowed to grow for an additional 7 days. ACE2 enzyme actively was measured in cell lysates essentially as described in the Methods.

ACE2 activity assay

ACE2 activity in astroglial cells following lenti-mACE2 infection with 10 multiplicites of infection (MOI) was determined as previously described (Huentelman et al. 2004a,b). The assay is based on the use of Flurogenic Peptide Substrate VI (FPS VI, R and D Systems, Minneapolis, MN, USA). ACE2 removes the C-terminal dinitrophenyl moiety that quenches the inherent fluorescence of the 7-methoxycoumain group, resulting in an increase in fluorescence in the presence of ACE2 activity at excitation and emission spectra of 328 and 392 nm, respectively. Briefly, protein was isolated from lenti-GFP- and lenti-mACE2-infected astroglial cells using a buffer comprised of 75 mm Tris pH 7.5, 1 m NaCl, and 0.5 μm ZnCl2. Protein content was then determined using a Bradford assay. Samples containing ACE2 enzyme (up to 50 μl) were incubated with 100 μm FPS VI, 10 μm captopril (to inhibit ACE activity) and reaction buffer (1 m NaCl, 75 mm Tris and 0.5 mm ZnCl, pH 7.5) in a final volume of 100 μl at 37°C. The change in fluorescence was monitored using a BioTek Synergy HT spectrophotometer. Total ACE2 activity was determined in the absence of the peptide-based ACE2 inhibitor DX600 (R and D Systems) while specific ACE2 activity was calculated by subtracting the total activity in the presence of 10 μm captopril and 100 μm DX600. Specific ACE2 activity is expressed as picomoles of substrate converted to product per unit time, and is normalized for protein content. Infection of astroglial cells with 10 MOI of lenti-mACE2 resulted in an ∼6-fold increase in ACE2 enzyme activity (Fig. 1B).

Animal procedures and treatments with lenti-mACE2

Sprague–Dawley rats were purchased from Charles River Laboratories (Wilmington, MA, USA). At 5 days of age, rats were lightly anaesthetized with methoxyflurane (inhalation) and a single 40 μl bolus of 3 × 108 particles of either lenti-GFP (n= 6) or lenti-mACE2 (n= 6) was injected into the left cardiac ventricular cavity as previously described (Iyer et al. 1996; Lu et al. 1997; Pachori et al. 2002). After viral injection, the animals were returned to their respective mothers for an additional 17–20 days for weaning. At 15 weeks of age, osmotic minipumps (model 2004, Alzet, Durect, Cupertino, CA, USA) were inserted subcutaneously between the shoulder blades to deliver 200 ng kg−1 min−1 Ang II or 0.9% saline at an infusion rate of 0.25 μl h−1 for 4 weeks. Thus four groups (n= 3 per group) were incorporated into the study: lenti-GFP with saline pump; lenti-GFP with Ang II pump; lenti-mACE2 with saline pump; and lenti-mACE2 with Ang II pump. All animal procedures were conducted under the approval of our Institutional Animal Care and Use Committee. This protocol of lentiviral vector-mediated gene transfer has been demonstrated to cause a long-term and robust transduction of predominantly the liver and the heart, with little transduction of other organs such as kidney, adrenal glands and lung (Coleman et al. 2003). Studies have established that transduction of the heart is predominantly a result of infection of cardiomyocytes by the lentiviral vector (Zhao et al. 2002; Bonci et al. 2003; Fleury et al. 2003; Sakoda et al. 2003).

Indirect blood pressure measurements

Indirect BP measurements were carried out using the tail-cuff method (Iyer et al. 1996; Lu et al. 1997). Briefly, animals were lightly heated for 3–5 min under a 200 W heat lamp before placement into a temperature-controlled Plexiglass restraint cage to which the animals had previously become accustomed. A pneumatic pressure sensor was attached to the tail distal to a pneumatic pressure cuff, both under the control of a Programmed Electro-Sphygmomanometer (Narco Bio Systems, Austin, TX, USA). Voltage outputs from the pressure sensor bulb and inflation cuff were recorded and analysed electronically using a PowerLab signal transduction unit and associated Chart software (ADInstruments, Colorado Springs, CO, USA). Systolic blood pressure values from each animal were determined by averaging a minimum of three separate indirect pressure measurements.

Measurement of cardiac fibrosis

After 4 weeks of Ang II infusion, rats were killed by halothane inhalation followed by decapitation and hearts were removed, blotted free of blood, and weighed to determine heart weight to body weight ratios (HW : BW) as described previously (Pachori et al. 2002; Metcalfe et al. 2004). Hearts were postfixed in ice-cold PLP solution (2% paraformaldhyde, 75 mm lysine, 37 mm sodium phosphate and 10 mm sodium peroxide) and processed for Masson's Trichrome staining to assess the extent of myocardial collagen deposition. Six 10 μm transverse (short-axis) sections of the heart at the level of the papillary muscles were analysed from each experimental animal for bright blue staining (collagen) using the NIH ImageJ analysis program (Rasband, 2005). Blue (collagen) staining was normalized against red (cardiac myocyte) staining for each heart.

mACE2 transgene expression

Total RNA was isolated from the left ventricular free wall of experimental and control virus-infected animals using RNeasy fibrous tissue mini kit (Qiagen, Valencia, CA, USA) and one-step real-time RT-PCR was performed with an Applied Biosystems Prism 7000 HT detection system according to the manufacturer's instructions. Primers and probes used were as follows: (forward primer), 5′-ACCCTTCTTACATCAGCCCTACTG-3′; (reverse primer), 5′-TGTCCAAAACCTACCCCACATAT-3; and (probe), 5′-ATGCCTCCCTGCTCATTTGCTTGGT-3′. Relative expression of mACE2 mRNA was calculated using a comparative method described in the Applied Biosystems User bulletin 2. Controls for no reverse transcription and no template were included to ensure the absence of genomic DNA contamination in the assay.


All results are derived from three or four animals in each group and are expressed as means ±s.e.m. Two-way ANOVA was used to determine effects of virus and angiotensin II treatment, and values of P < 0.05 were considered statistically significant. Post hoc analyses were carried out using one-way ANOVA with P < 0.05 considered significant.


Systemic delivery resulted in significant overexpression (P < 0.05) of mACE2 in ventricles of all six lenti-mACE2 treated animals (Fig. 2). Delta CT values (mACE2 cycle number minus 18S cycle number) were approximately 12.15 cycles lower than for the lenti-GFP treated animals. This indicates that mACE2 gene transfer resulted in an expression increase of approximately 4500 copies of mACE2 mRNA.

Figure 2.

Real-time RT-PCR for mouse ACE2 in rat heart ventricles
Measurement of murine ACE2 using real-time RT-PCR indicated a significant increase (P < 0.05) in expression levels in the heart ventricles of lenti-mACE2 treated rats.

Infusion of Ang II for 4 weeks resulted in an approximate 90 mmHg increase in BP over saline-infused controls (P= 0.0005). This increase in BP was unaffected by lenti-mACE2 treatment (P= 0.97, Fig. 3).

Figure 3.

The effect of lenti-mACE2 on Ang II-induced increases in systolic blood pressure measured indirectly
Fifteen weeks following transduction with either lenti-GFP or lenti-mACE2, rats were infused with Ang II for 4 weeks, as described in the Methods. Systolic BP was then determined using an indirect method. Two-way ANOVA revealed a significant difference (†P= 0.0005) between saline and Ang II. No significant effect (P= 0.97) of lenti-mACE2 treatment was observed.

Animals were killed after 4 weeks of Ang II infusion, and HW : BW ratio and cardiac fibrosis were assessed. Angiotensin II infusion resulted in a 22% increase in HW : BW ratio (P < 0.005, Fig. 4). A post hoc analysis comparing the HW : BW ratios between lenti-GFP and lenti-mACE2 virus-transduced animals that received Ang II infusion revealed significant (P < 0.05) mACE2-mediated reduction of HW : BW ratio.

Figure 4.

The effect of lenti-mACE2 transduction on cardiac hypertrophy following Ang II infusion
HW : BW ratios were determined after death as described in the Methods. Four weeks of Ang II infusion caused significant increases in HW : BW ratios (†P < 0.005). Post hoc analysis of GFP–Ang II versus mACE2–Ang II groups found a significant attenuation of Ang II-induced hypertrophy by mACE2 (*P < 0.05).

Chronic Ang II infusion caused a marked increase in mid-myocardial cardiac fibrosis as determined by Masson's Trichrome staining (Fig. 5). Quantification of collagen (Fig. 6) revealed that Ang II infusion led to a significantly increased collagen deposition (P= 0.01), and lenti-mACE2 treatment significantly attenuated this effect (P= 0.02).

Figure 5.

The effect of lenti-mACE2 transduction on Ang II-induced myocardial fibrosis
Following termination of Ang II infusion (after 4 weeks of infusion), hearts were fixed in PLP solution, sectioned and stained with Masson's Trichrome as described in the Methods. Images are representative of the extent of myocardial fibrosis in lenti-GFP with saline infusion, lenti-GFP with Ang II infusion and lenti-mACE2 with Ang II infusion at ×10 and ×20 magnification.

Figure 6.

Quantitative analysis of collagen deposition in heart sections
Quantitative analysis of slides at ×10 magnification was carried out using ImageJ (NIH) (Rasband, 2005) Two-way ANOVA revealed significant effects of Ang II infusion (†P= 0.01), lenti-mACE2 treatment (*P= 0.02), and a significant interaction (P= 0.01).


This study demonstrates, for the first time, that non-targeted overexpression of ACE2 protects against the development of angiotensin-induced myocardial fibrosis. The beneficial effects on cardiac hypertrophy and fibrosis are observed in the absence of any significant normalization of systolic BP. Thus, our observation is supportive of previous findings showing that beneficial cardiovascular outcomes are possible even in the absence of the pharmacological lowering of BP (Regan et al. 1997; Sleight, 2000).

Previous studies from our laboratory have established that lentivector-mediated gene delivery effectively targets cardiac tissue in addition to other organ systems, including liver, kidney and lung. Delivery of the lentivector into the systemic circulation via the left ventricular cavity targets between 20 and 40% of the cardiac mass (including both ventricles and atria) with upwards of 90–95% of transduced cells exhibiting myocyte morphology (Coleman et al. 2003). The reason for such specific myocyte targeting within cardiac tissue is under investigation, but similar cell type specific transduction using the lentivector has been demonstrated by others (Zhao et al. 2002; Bonci et al. 2003; Fleury et al. 2003; Sakoda et al. 2003).

Although systemic delivery of the lentiviral vector is capable of transducing other tissues to a much lesser extent, the results reported in this study are consistent with the local therapeutic transgene expression in the cardiac tissue. This is supported by the fact that the mACE2 construct used in this study encoded for the full-length protein product and therefore is predicted to be membrane localized (Huentelman et al. 2004b). Additionally, our experimental observations illustrate that BP was not altered in lenti-mACE2 treated animals, thereby suggesting a local-acting, not systemic, therapeutic effect. In spite of these views, one cannot completely exclude the role of limited transduction of other cardiovascular tissues by ACE2 at the present time. An immunocytochemical mapping of the heart and other cardiovascular tissues would be critical in resolving this issue. At the present time, we cannot identify whether local, cardiac expression of mACE2 is providing the cardioprotective effects observed, or whether expression at other, non-cardiac locations (i.e. certain cardiovascular control centres of the brain) are responsible for the observed changes in hypertrophy and fibrosis. Lack of ACE2 antibodies which selectively differentiate transgenic, mouse ACE2 from endogenous, rat ACE2, however, prevents us from conducting such an experiment at the present time.

In the lenti-mACE2 treated animals there was no demonstrated fibrosis in the heart. This is surprising since this protocol does not transduce 100% of cardiac tissue. The mechanism by which this occurs remains speculative, but we suggest that intra- or intercellular communications directly or though some yet unknown paracrine/autocrine factor may enable the propagation of signals from ACE2 overexpressing cells to the entire heart. This proposal is consistent with the observation from whole cell therapy experiments, where significant improvements in cardiac function are seen following the implantation of only a few thousand altered stem cells in the heart (Mangi et al. 2003). The beneficial effects on hypertension-induced cardiac remodelling by ACE2 were without any observed adverse effect on the heart. Although we did not perform rigorous evaluations of cardiac function in this preliminary study, we could not ascertain any differences in heart weight, heart rate, or morphological differences in echocardiograms (data not shown) between mACE2-overexpressing and control hearts. These observations are in stark contrast to the studies of Donoghue et al. (2003), where overexpression of ACE2 in the mouse resulted in ventricular tachycardia and fibrillation and sometimes sudden death. We believe that overexpression of this enzyme in the heart after normal cardiac development may be critical in providing these cardioprotective effects. This conclusion is further supported by our preliminary data with the spontaneously hypertensive rat (SHR) model of hypertension, in which the postnatal delivery of lenti-mACE2 caused similar cardioprotective effects in the SHR (J. Vazquez, unpublished observations).

The ACE2 gene is the first member of the RAS whose overexpression holds potential for its use in gene therapy for cardiovascular diseases and hypertension. It is a multifunctional enzyme that not only regulates the production of vasodilators, such as Ang-(1–7), but also influences other peptides that may exert effects on the cardiovascular system, such as apelin and the connexins (Donoghue et al. 2000; Danilczyk et al. 2003; Oudit et al. 2003; Burrell et al. 2004; Katovich et al. 2005). Targeted overexpression with the use of cell/tissue specific promoters and the availability of a secreted form of this enzyme would permit investigation of the effects of both local and systemic ACE2 on tissue pathophysiology and BP regulation. A secreted form of ACE2 lacking the transmembrane anchor has been cloned (Huentelman et al. 2004b) and is currently being investigated for its ability to regulate BP through the alteration of plasma ACE2 levels. Finally, the mechanism by which cardiac overexpression of ACE2 prevents hypertrophy and fibrosis remains to be investigated. Based on the available data, it is reasonable to suggest that overexpression of this enzyme may shift the RAS away from vasoconstrictor, proliferative and hypertrophic responses towards vasodilatory and antihypertrophic responses (Oudit et al. 2003; Burrell et al. 2004). This view is supported by observations that increased cardiac ACE2 mRNA and plasma levels of Ang-(1–7) are seen in the reversal of cardiac pathophysiology after myocardial infarct (Loot et al. 2002; Averill et al. 2003). Our observation that ACE2 is involved in several aspects of cardioprotection is nonetheless an important milestone in recognizing the future therapeutic potential of this enzyme.

The RAS has been implicated in the accumulation of collagen and the resultant fibrosis in the heart (Brilla, 2000; Linjen et al. 2004). The effects of the RAS on collagen synthesis and/or degradation have also been demonstrated using cell culture techniques to avoid the confounding in vivo haemodynamics and other humoral factors that can influence myocardial collagen formation (Lijnen & Petrov, 1999). Angiotensin II has been demonstrated to stimulate collagen secretion and production in cardiac fibroblasts via AT1R (Lijnen et al. 2001). Angiotensin II type 2 receptors (AT2R) are also re-expressed in cardiac fibroblasts in failing hearts and this re-expression may exert anti-AT1R actions on the progression of interstitial fibrosis during cardiac remodelling (Ohkubo et al. 1997). We have demonstrated that overexpression of cardiac AT2R produces similar cardioprotective effects on hypertrophy and fibrosis in two models of hypertension (Falcon et al. 2004; Metcalfe et al. 2004), without altering blood pressure. Similar effects may be observed with overexpression of ACE2. Thus, the present study further emphasizes a role for components of the RAS in cardiac hypertrophy and fibrosis.

Current therapeutic strategies for the control and treatment of cardiovascular diseases (CVD), including hypertension, are primarily limited to the use of pharmacological agents, many of which inhibit components of the RAS. Despite their demonstrated success, the prevalence of CVD has increased significantly in the last decade. Clearly, there is an urgent need to develop new strategies (i.e. to identify new drug targets, develop novel therapeutic molecule delivery methods, explore the hope of cell-based therapies, etc.) for the successful control of CVD. The discovery of ACE2, with its potential to shift the adverse effects of RAS hyperactivity toward beneficial outcomes in the cardiovascular system, holds this promise. Our study is timely in that it presents evidence that overexpression of ACE2 prevents cardiac hypertrophy and fibrosis induced by angiotensin II. This provides conceptual in vivo support for ACE2 as a viable target for the future development of pharmacological and genetic upregulating strategies for the treatment of CVD.



This work was supported by the NIH grant HL56921. Justin Grobe is a predoctoral fellow of the American Heart Association; Florida/Puerto Rico affiliate. Observation describing the effect of ACE2 on fibrosis was presented at the 2004 Vascular Biology Working Group Annual meeting in New Orleans and was included as unpublished data in a review that followed (Katovich et al. 2005).

Author's present address

M.J.Huentelman: The Translational Genomics Research Institute, 445 N. Fifth Street, 5th Floor, Phoenix, AZ 85004, USA.