Sarcoplasmic reticulum Ca²⁺ ATPase as a therapeutic target for heart failure

Abstract

Background

The cardiac isoform of the sarco/endoplasmic reticulum Ca²⁺ATPase (SERCA2a) plays a major role in controlling excitation/contraction coupling. In both experimental and clinical heart failure (HF), SERCA2a expression is significantly reduced which leads to abnormal Ca²⁺ handling and deficient contractility. A large number of studies in isolated cardiac myocytes, in small and large animal models of HF showed that restoring SERCA2a expression by gene transfer corrects the contractile abnormalities and improves energetics and electrical remodeling. Following a long line of investigation, a clinical trial is underway to restore SERCA2a expression in patients with HF using adeno-associated virus type 1.

Objective

This review addresses the following issues regarding HF gene therapy: 1) new insights on calcium regulation by SERCA2a; 2) SERCA2a as a gene therapy target in animal models of HF; 3) advances in the development of viral vectors and gene delivery; 4) clinical trials of HF using SERCA2a

Methods

This review focuses on the new advances in SERCA2a- targeted gene therapy made in the last three years.

Conclusion

SERCA2a is an important therapeutic target in various cardiovascular disorders. Ongoing clinical gene therapy trials will provide answers on its safety and applicability.

Introduction

Heart failure (HF) remains a major cause of morbidity and mortality in the developed world. In the population under the age of 65 HF prevalence approaches 1%¹. Despite the fact that pharmacologic and device-based treatments have improved survival, the 5-year mortality rate remains at an alarming 50% in this population². Faced with these challenges, researchers have vigorously explored novel therapeutic options, including gene-based treatments. Over the last decade, important progress has been made in understanding the various intracellular and molecular mechanisms that become abnormal during HF^3–5. One of the key abnormalities in both human HF and experimental models of HF is abnormal intracellular calcium ion (Ca²⁺) handling caused by a defect in sacoplasmic reticulum (SR) function^{4, 6}. Deficient SR Ca²⁺ uptake in myocytes is associated with a decrease in the expression and activity of the sarco/ endoplasmic reticulum calcium ATPase cardiac isoform 2a, (SERCA2a)^3–5. Unloaded ([Ca²⁺])_SR in turn regulates the effectors of SR Ca²⁺ release and Ca²⁺ influx channels, further worsening the abnormal Ca²⁺ distribution. SERCAs play a critical role in the control of spatio-temporal patterns of intracellular calcium signaling – a pattern, that controls virtually every cellular function, including contraction, proliferation/hyperthrophic growth and apoptosis^{6, 7}. Normalization of SERCA2a function has been shown to increase contractility in failing human cardiomyocytes and to improve hemodynamics along with survival in rodent and large animal models of HF^3–5. The overexpression of SERCA2a also has been found to restore energetic supply and to decrease ventricular arrhythmias in a model of ishemia/reperfusion^8–10. Furthermore, a number of novel beneficial mechanisms have been uncovered in vascular smooth muscle cells (VSMCs) and endothelial cells (EC)^{11, 12}. Therefore, SERCA2a is one of the most promising targets for the treatment of HF.

Advances in the understanding of the molecular basis of myocardial dysfunction together with the evolution of gene transfer technology, has finally placed CHF within reach of gene-based therapy^{5, 13, 14}. Myocardial dysfunction results in myocardium that has a mixture of diseased, dysfunctional and healthy myocytes. The goal of gene therapy is to treat dysfunctional myocytes and to prevent disease or loss of healthy myocytes.

This review focuses on gene therapy using SERCA2a or small molecules regulating SERCA2a activity to treat HF. We present new insight in the mechanism of Ca²⁺ regulation by SERCA2a, examine the complex beneficial effects as well as the risk associated with SERCA2a overexpression. We describe the new advances in the development of viral vectors and gene delivery and discuss the potential of SERCA2a as a therapeutic target in cardiovascular disease.

1. Calcium cycling abnormalities in heart failure

Calcium is crucial to the very process that enables the chambers of the heart to contract and relax, a process called exitation-contraction coupling^4–6. During systole, the action potential induces a minor Ca²⁺-influx through sarcolemmal L-type calcium channel (LTCC) (Figure 1). This triggers further calcium release from the SR via the major SR Ca²⁺-release channel ryanodine receptor (RyR2, the cardiac isoform of RyR). The combination of Ca²⁺ influx and release raises free intracellular Ca²⁺ concentration ([Ca²⁺]) from 0.1–0.2 μM to 2.0 – 10.0 μM, thereby allowing Ca²⁺ to bind to the myofilament protein troponin C resulting in sarcomere shortening and muscle contraction. Muscle relaxation is initiated by RyR2 closure accompanied by Ca²⁺ dissociation from the troponin complex and its reuptake into the SR by SERCA2a (75% of Ca²⁺ removal in human) its removal by the sarcolemmal Na²⁺/ Ca²⁺ exchanger (NCX) (25% of Ca²⁺ removal in human). SERCA pump serves a dual function: 1) to cause muscle relaxation by lowering cytosolic [Ca²⁺], and 2) to restore SR Ca²⁺ load necessary for muscle contraction. SERCA2a enzymatic activity is controlled by the inhibitory peptide phospholamban (PLN). In its dephosphorylated form, PLN inhibits SERCA2a activity by lowering its affinity for Ca²⁺.

Figure 1. Excitation-contraction coupling in cardiomyocytes.

A small amount of Ca²⁺ uptake via LTCC causes a large amount of Ca²⁺ release from the SR via RyR – Ca²⁺-induced Ca²⁺ release. Rise of intracellular Ca²⁺ causes contraction of myofilaments. Ca²⁺ uptake into the SR by SERCA or extrusion of Ca²⁺ trough NCX channels results in relaxation of myofilaments. Abbreveations: Na-K ATPase: plasma membrane sodium-potassium ATPase; Ca²⁺: calcium ion; Ica: L-type calcium current; LTCC: L-type calcium channel; K⁺: potassium ion; PLN: Phospholamban; PMCA: Plasma Membrane Calcium ATPase; RyR: Ryanodine Receptor calcium channel; Na⁺: sodium ion; NCX: - sodium-calcium exchanger; SR/ER: Sarco/Endoplasmic Reticulum; SERCA: Sarco/Endoplasmic Reticulum Ca²⁺ ATPase 2a.

The system of Ca²⁺ cycling is finely regulated by the β-adrenergic signaling system through phosphorylation and dephosphorylation (Figure 2). In response to stress, binding of agonists to β-Adrenergic receptors (βARs) results in adenylyl cyclase (AC) stimulation, cAMP production and activation of protein kinase A (PKA). PKA phosphorylates LTCC and RyR, thus increasing cytosolic Ca²⁺ influx, whereas phosphorylation of troponin-I increases the overall contractility, mediating the inotropic and chronotropic effect of βAR. Phosphorylation of PLN on serine-16 by PKA relieves its inhibition of SERCA2a and contributes to a frequency-dependent acceleration of SR calcium reuptake, mediating the lusitropic effect of βARs. An additional critical target in calcium cycling is inhibitor-1 (I-1). I-1 becomes active upon PKA phosphorylation and inhibits the type 1 serine/threonine protein phosphatase (PP1) resulting in amplification of βAR responses in the heart¹⁵ (Figure 2).

Figure 2. The regulation of calcium cycling via adrenergic pathway and its alterations in failing heart.

Schematic showing the positive control of cardiac function through classic β-adrenergic pathway and a negative control of cardiac function through the α-adrenergic pathway. In response to stress, binding of agonist to the βARs results in inotrope (increased contractility), chronotrope (increased heart rate) and lusiotrope (enhanced relaxation) effects. These effects are mediated by PKA phosphorylation of several targets, resulting in enhanced Ca²⁺ cycling. Negative control of heart function through αARs is mediated by PP1 dephosphorylation of several PKA targets. Abbreveations: αAR: α-adrenergic receptor; AC: Adenylyl cyclase; βAR: β-adrenergic receptor; βARK: β-adrenergic receptor kinase; CIRC: Calcium induced calcium release; Gαq: Gq protein α-subunit; Gαs: Stimulatory G-protein α-subunit; I-1: Inhibitor 1; Ica: L-type calcium current; LTCC: L-type calcium channel; P: Phosphorylation; PKA: Protein kinase A; PKC: Protein kinase C; PLN: Phospholamban; PLC: Phospholipase C; PP1: Serine/threonine phosphatase 1; RyR: Ryanodine receptor calcium channel; SERCA: Sarco/endoplasmic reticulum Ca²⁺ ATPase.

Chronic HF is associated with increased sympathetic outflow, which may be compensatory early in the disease state; however long-term neurohormonal activation induces significant damage to cardiomyocytes. Long-term chronic stimulation of βARs results in multiple alterations in the βAR signaling cascade including βAR down regulation, upregulation of βAR kinase (βARK) and increased inhibitory G-protein α-subunit function^{3, 13}. In failing cardiomyocytes the intracellular [Ca²⁺] is elevated, the amplitude of the [Ca²⁺] transient is decreased, and its duration is prolonged. These significant alterations in Ca²⁺ handling along with SR Ca²⁺ leak trough the RyR, decreased SR Ca²⁺ uptake and decreased SR Ca²⁺ content, result in abnormal excitation-contraction coupling^{3, 5}.

The importance of SERCA2a in HF has been studied extensively in both animal models and in humans. These studies have shown that SERCA expression and activity is reduced in failing myocardium^{3, 5}. Increased inhibitory function of PLN¹⁶ and increased activity of PP1 resulting in dephosphorylation and inactivation of PKA targets¹⁵ also contribute to impaired SR Ca²⁺ uptake. The common finding is a substantial decrease in SR Ca²⁺ uptake which is a crucial component of the impaired contractile performance of the failing heart.

In many cases, altered Ca²⁺ cycling precedes the observed depression of mechanical performance; consequently, improving of Ca²⁺ cycling has potential as an attractive strategy against HF^{3, 5}.

2. Effect of SERCA2a gene transfer on calcium regulation and cell functions

Calcium is an ubiquitous and pleiotropic second messenger, thus, modulation of calcium signaling by SERCA2a gene transfer can affect multiple functions in different cardiovascular cell types. At least 11 isoforms of SERCA encoded by three separate homologous genes (ATP2A1, ATP2A2 and ATP2A3) have been discovered today^{7, 17, 18}. Given the functional differences between isoforms, only cardiac specific isoform SERCA2a is therapeutically useful, others are or may be harmful or ineffective^19–21.

Contractility

Adenovirus-mediated gene transfer of SERCA2a to failing human cardyomyocytes restored the calcium transient and improved contraction and relaxation velocity to the level of non-failing myocytes²². The enhanced contractility induced by SERCA2a overexpression was due to an enhanced SR Ca²⁺ uptake during diastole, increased SR Ca²⁺ content, and more effective Ca²⁺ efflux during systole (Figure 3).

Figure 3. Effect of SERCA2a overexpression on contractile function in diseased heart.

The overexpession of SERCA2a in failing hearts has been shown to result in the recovery of contractility. SERCA2a overexpression improves calcium cycling and myocyte shortening. Abbreveations: CIRC: Calcium induced calcium release; L-type calcium current; LTCC: L-type calcium channel; PLN: Phospholamban; RyR: Ryanodine receptor calcium channel; SERCA: Sarco/endoplasmic reticulum Ca²⁺ ATPase.

Mechanoenergetics

Calcium overload leading to increased calcium uptake by mitochondria, resulted in mitochondrial dysfunction and decreased ATP production^{23, 24}. Normalization of intracellular calcium by SERCA2a overexpression should prevent mitochondrial dysfunction and decrease energy cost for Ca²⁺ regulation. Sakata et al. investigated the oxygen cost of left ventricular (LV) contractility, obtained by plotting myocardial oxygen consumption per beat against equivalent maximal elastance^{8, 9}. In diseased hearts, the oxygen cost of LV contractility was increased; SERCA2a gene transfer restored the increased oxygen cost to the normal level. In addition, restoring SERCA2a was shown to correct the creatine kinase activity by normalizing the creatine to ATP ratio in animal model of HF²⁵. Therefore SERCA2a gene transfer restores cardiac energetics both by improving mechanoenergetic efficiency and by enhancing energy supply.

Arrhythmias

Diastolic Ca²⁺ overload occurring in ischemia and chronic HF is responsible for cellular electric instability resulting in tachyarrythmias⁵. The increase in Ca²⁺ ions in the cell and SR leads to generation of asynchronous Ca²⁺ waves during diastole via abnormal activity of RyRs. These waves activate the electrogenic NCX and thereby produce arrhythmogenic spontaneous aftercontractions and afterdepolarizations²⁶. Recent data demonstrated that SR Ca²⁺ load via SERCA2a is responsible to sensitization of RyR, thereby creating a traveling wave of high Ca²⁺ sensitivity initiating Ca²⁺ release²⁷. SERCA2a overexpression in rabbit cardiomyocytes, favoring sequestration of Ca²⁺ in the SR, result in rapid inactivation of subsequent Ca²⁺ currents and reduced Ca²⁺ entry through LTCC²⁸. In this model, reduction of trans-sarcolemmal Ca²⁺ flux by SERCA2a overexpression without changes of normal systolic transient also reduced the occurrence of aftercontractions and afterdepolarization²⁸. As such, reduction of cytosolic Ca²⁺ overload might constitute a viable strategy for combating arrhythmia (Figure 4). In support of this hypothesis, Del Monte et al. showed that SERCA2a gene transfer in a model of ischemia/reperfusion injury effectively suppressed arrhythmias and reduced myocardial infarct size¹⁰.

Figure 4. Model showing the effect of SERCA2a overexpression on arrhythmias.

The development of an unstable equilibrium is the mechanistic basis of increased arrhythmogenic risk. At the basal level (left) Ca²⁺ homeostasis is maintained by the balanced activities of LTCC and NCX, and between RyR2 and SERCA2a. Normal cardiac activation (middle) proportionally increases LTCC, NCX, RyR2 and SERCA function to maintain synchronized Ca²⁺ fluxes resulting in a stable state. Acute disruption of a single component relative to others (e.g. RyR) results in the ablation of synchronized Ca²⁺ fluxes and directly increases the arrhythmogenic propensity of the system (right). This state, illustrated using the experimentally observed findings of increased RyR2 leak and NCX activity and diminished SERCA activity, is inherently more prone to destabilization. SERCA2a gene transfer removes Ca²⁺ overload, stabilizes RyR2 and prevents arrhythmias. Abbreveations: Ca²⁺: calcium ion; LTCC: L-type calcium channel; PLN: Phospholamban; RyR: Ryanodine Receptor calcium channel; NCX: - sodium-calcium exchanger; SR: Sarcoplasmic Reticulum; SERCA: Sarco/Endoplasmic Reticulum Ca²⁺ ATPase 2a.

Transcription

Calcium also acts on Ca²⁺ -dependent transcription pathways via the activation of various kinases and phosphatases⁶ (Figure 5). The Ca²⁺-dependent serine/threonine phosphatase calcineurin was identified as a central pro-hypertrophic signaling molecule in the myocardium²⁹. Its target, nuclear factor of activated T lymphocytes (NFAT), was shown to be necessary and sufficient for mediating pathological hypertrophic remodeling³⁰ apoptosis³¹ and inflammation³² in cardiomyocytes and proliferation of VSMC¹². During hypertrophic /proliferative remodeling, reduced SR Ca²⁺ store and increased expression of transient receptor potential channels (TRPC)⁷ favor store-operated Ca²⁺ entry and sustained activation of calcineurin. In VSMC overexpression of SERCA2a inhibits calcineurin activity and NFAT-driven gene transcription¹².

Figure 5. Model showing Ca²⁺ dependent hyperthrophy/proliferation transcription pathways.

Excitation/contraction calcium cycling is not involved in the control of hyperthrophic/proliferation pathways. Cardiac myocytes also contain lipid rafts rich in caveolin in the sarcolemma and T-tubules that are thought to generate local microdomains of Ca²⁺. Activation of phosphoinositol-3-phosphate-coupled membrane receptors results in calcium release from the SR via IP3R. Depletion of SR calcium close to the sarcolemma induces the formation of Store Operated Calcium complex resulted in calcium influx in local submembrane area. This calcium activates calcineurin (PP2B), which dephosphorylates NFAT, inducing its nuclear translocation and transcriptional activation.

Hypertrophy Proliferation

Myocardial hypertrophy is an adaptive response to hormonal and mechanical stimuli. Several reports suggest that SERCA2a overexpression rescues or prevents cardiac hypertrophy^{33, 34}. Although the mechanism of this benefit was not completely elucidated in the heart, it is speculated that they may result from inhibition of calcineurin, as shown in VSMC¹². In VSMC SERCA2b (ubiquitous isoform) and SERCA2a are co-expressed in adulthood. During proliferation VSMC undergo de-differentiation and acquisition of a synthetic phenotype, associated with loss of SERCA2a. Restoring SERCA2a expression in VSMC by gene transfer inhibits VSMC proliferation of and post-injury restenosis in a rat carotid injury model¹².

NO synthesis

NO synthesis is controlled by intracellular Ca²⁺ oscillations, and increase in SERCA activity by overexpression of S100A1 (see below) also increases NO production in endothelial cells (EC)³⁵. Sakata et al showed that transcoronary gene transfer of SERCA2a increase coronary blood flow³⁴. One of the reasons for this improvement is the increased production of NO caused by SERCA2a overexpression in EC¹¹. This additional effect of SERCA2a overexpression might have a synergic benefit on cardiovascular disease.

3) SERCA2a as a target for heart failure gene therapy in animal models

Several studies have been designed to improve cardiac function using SERCA as a target. The SERCA isoform used is an important consideration. Data from transgenic animals demonstrated functional differences between the isoforms SERCA2a and SERCA2b in the heart. Long-term expression of SERCA2a in transgenic animals improves contractility without adverse effects³⁶, but replacement of SERCA2a by SERCA2b isoform resulted in cardiac dysfunction and hypertrophy^{19, 20}. Adenovirus-mediated gene transfer of SERCA2a improved cardiac hemodynamics and increased survival in a rat model of HF²⁵, whereas gene transfer of the faster skeletal muscle isoform SERCA1 had limited functional and metabolic improvement in the same model and compromised function in hypertrophic hearts²¹.

SERCA2a has been the subject of most gene therapy approaches targeting altered calcium physiology⁵. The increase in contractility concomitant with an increase in SERCA2a expression has been shown by gene transfer and transgenic methods in healthy, HF or diabetes mellitus animals and by gene transfer in a large animal model of HF^{5, 37}. Furthermore, SERCA2a gene transfer restored the oxygen cost of LV contractility to the normal level, indicating an improvement in energy utilization^{8, 9, 25}.

LV dysfunction is a major predictor of electrical instability and arrhythmias. Hence, improvement of myocardial contractility by gene transfer could have a major beneficial electrophysiological effect. Adenovirus-directed gene transfer of SERCA2a in a rat model of ischemia/reperfusion injury effectively suppressed arrhythmias and reduced myocardial infarct size¹⁰. In contrast with this finding, overexpression of SERCA2a in transgenic rats increased the occurrence of fatal ventricular tachyarrhythmia following infarct³⁸. These latter finding could result from adverse electrical remodeling caused by reduced NCX expression and function triggered by enhanced SR Ca²⁺ content. Overall, the overexpression of SERCA2a by gene transfer in the setting of ischemia reperfusion both in small and large animals has been shown to reduce ventricular arrhythmias^{10, 39}.

Recently several preclinical HF studies were conducted on large animal models. A porcine model of HF was created by mitral regurgitation⁴⁰. AAV1/SERCA2a was administrated using single intracoronary perfusion. Two months after treatment SERCA2a-injected animals demonstrated significant improvements in contractile function and restoration of ventricular volumes compared with untreated animals. There was an absolute increase of 16% in median ejection fraction in AAV1/SERCA2a treated animals⁴⁰. A sheep model of HF was created by rapid pacing³⁷. Delivery of AAV1/SERCA2a by cardiac recirculating elicited a dose-dependent improvement in LV function 1–2 month after virus administration 37. In both studies, safety-related parameters including histopathologic analysis, hematology, and clinical chemistry indicate that the treatment did not cause any organ damage or inflammatory response^{37, 40}. Furthermore, in addition to favorable haemodynamic effects, brain natriuretic peptide (BNP) expression was reduced consistent with reversal of the HF molecular phenotype^{37, 40}. To evaluate the potential toxicological effect of AAV1/SERCA2 gene therapy the virus was administrated using transcoronary perfusion in normal Göttingen minipigs⁴¹. The animals received the dose at least 3-fold higher that the highest dose intended for administration in clinical trials. AAV1/SERCA2a DNA was found in the majority of all the tissues sampled 5 days after injection. The highest concentrations were found in the lung and liver, low to non detectable levels of AAV1/SERCA2a were found in blood and brain tissue. Given the wide spread level of “junk” DNA in the human genome, the safety consequence of the presence of vector DNA in non cardiac cells such as liver is expected to be minimal or nonexistent⁴¹. On Days 30 to 90, the tissue concentrations of AAV1/SERCA2a decreased significantly, with many tissues having low to undetectable concentrations of AAV1/SERCA2a. On 90 days after gene transfer, there was an increase in SERCA2a protein expression in the left and right ventricular wall; in addition, coronary arteries, pulmonary arteries, and the diaphragm and lung tissues showed increased expression of SERCA2a, without apparent toxicologic effect⁴¹. However, there was no change in SERCA2 expression in liver, kidney, spleen, retina, biceps, sartorius, and gastrocnemius muscles⁴¹. AAV1/SERCA2a was well tolerated in minipigs; no signs of toxicity, clinical pathology or histology were observed, and values for echocardiography and electrocardiography parameters were found to be within normal range⁴¹.

The overall conclusion from animal studies is that AAV-directed SERCA2a gene transfer appears to be a safe, well-tolerated, and effective treatment for HF.

4. Small molecules targeting SERCA activity

Phospholamban

Inhibition of PLN is another therapeutic approach to target the Ca²⁺ handling pathway in the failing heart^{3, 13}. Gene transfer of a dominant-negative PLN mutant⁴², PLN antisense RNAs⁴³ and intracellular inhibitory PLN antibodies⁴⁴ have been used to increase SERCA2a activity and to rescue HF in animal models. Furthermore, normalization of the calcium transient and restoration of cell contractility has been reported in cardiomyocytes isolated from failing human hearts after adenovirus-mediated PLN antisense gene transfer^{45, 46}. Recently, successful treatment of HF was demonstrated in a rat model of aortic banding by RNAi targeting of PLN⁴⁷. Simple intravenous injection of PLN RNAi carried by rAAV9 provided long-term stable cardiac shPLN expression and restores cardiac function, reduces dilatation, hyperthrophy and fibrosis after 3 months⁴⁷. The excitement generated by these studies has been tempered by the discovery in human two mutations on PLN, leading to PLN-null genotype in homozygous individuals with a phenotype of dilated cardiomyopathy^48–50. These latter developments raise doubts about using PLN ablation as a universal approach to treat all forms of HF.

Inhibitor 1

Targeting PP1 by overexpression of its inhibitory peptide (I-1) to increase SERCA2a activity may constitute a promising therapeutic strategy in HF^{13, 15}. Indeed, silencing PP1 activity in transgenic mice by the expression of a constitutively active I-1 protein (I-1_T35D) results in enhanced PLN phosphorylation, augmented cardiac contractility, rescues from HF development upon aortic constriction and improves mechanical recovery and survival in an ischemia/reperfusion model^{51, 52}. Similarly, adenovirus gene transfer of I-1_T35D results in marked restoration of contractility in animal models and failing human myocytes⁵¹.

S100A1

Another important protein that has been recently targeted is S100A1, a low-molecular weight Ca²⁺ binding protein that stabilizes RyR and increases SERCA ATPase activity, resulting in a balanced enhancement of SR Ca²⁺ release and uptake^{13, 53}. In HF the expression of S100A1 is reduced. Gene transfer of S100A1 in rats following cryo-infarction induced an increase in SERCA2a and RyR activities, resulting in improved intracellular calcium handling and overall enhanced systolic and diastolic ventricular function^{54, 55}.

Pharmacotherapy

Recently, a novel inotropic agent that enhances SERCA2a activity was discovered and tested in HF models. Istaroxime, (E, Z)-3-((2-aminoethoxy)imino) androstane-6,17-dione hydrochloride) enhances SR Ca²⁺ content and Ca²⁺ uptake in failing cardiomyocytes and improves heart function in a guinea pig aortic binding model^{56, 57}. This finding opens new perspective for the pharmacotherapy of HF.

5. Advances in the development of viral vectors

The molecular vectors for cardiac gene therapy generally employ expressing cassettes containing strong virus promoters that are constitutively active in a wide spectrum of cells (e.g. cytomegalovirus promoter [CMV]), the gene of interest (SERCA2a) and an appropriate mRNA stabilizing polyadenylation signal. The most commonly used virus vectors are recombinant adeno- and adeno-associated viruses (Table 1).

Table 1.

Comparison of major viral vector systems for cardiovascular gene transfer.

Vectors	Adenovirus	Adeno-Associated Virus
Functional titer (per ml)	Up to 1012	Up to 1010
Genome	dsDNA	ssDNA (can also be dsDNA)
Insert Capacity	7–30 kb	4.8 kb
Integration	No	Yes:chromosome 19 for wild-type AAV2 Low for recombinant vectors
Pattern of gene expression	Transient	Long-term
Cell cycle-dependent transduction	No	No
Immunoreactions	Cytotoxic and immunogenic	Minimally immunogenic
Clinical trial approved	Yes	Yes

Adenoviruses

Recombinant human adenovirus (Ad) vectors are used essentially in preclinical small animal models. Adenoviruses are non-enveloped vectors that are very efficient at transducing target cells in vitro and in vivo, and can be prepared at high titers. The wild type Ad genome is approximately 35 kb of which up to 30 kb can be replaced with foreign DNA. These vectors have demonstrated efficient transduction of adult myocardium in vivo resulting in alteration of global heart function. In animal models, Ads mediate high-level expression for only short term (1 week). However, Ad cause intense local inflammation and stimulate potent cellular and humoral immune responses¹⁴.

Adeno-associated viruses

Recombinant adeno-associated viruses vectors (rAAV), derived from nonpathogenic parvoviruses, have many characteristics that make them particularly well suited for cardiac gene transfer⁵⁸. First, AAV effectively transfect dividing and non-dividing cells. Second, AAV provide stable long term (up to a year) expression in most systems⁵⁹. Third, AAV evoke minimal immune response in humans and are not known to cause human disease. AAV vectors have been studied in hundreds of patients, and have demonstrated an excellent safety profile^{14, 60}.

Major limitation of rAAV vector systems is the limited packaging capacity of 4.8 kb. AAV vectors package single- stranded genomes and require host cell synthesis of complementary strand for transduction. Recently, a new AAV vector (self-complementary [sc]AAV) which can package either two copies of DNA or dimeric-inverted repeat DNA molecules, thereby alleviating the requirement for host-cell DNA synthesis, was developed^{61, 62}. However, the cassette sizes have to be half the size of a normal sized single-stranded vector (i.e., 2.3 kb), which further limits the number of genes that can be used with scAAV.

Today, 12 different AAV serotypes are known, each with different tissue tropisms⁶³. AAV1, 2, 6, 8 and 9 have higher tropism for myocardium^64–66. AAV8 and 9 efficiently cross the endothelial barrier and can be injected intravenously or intraperitoneously^{65, 67}. New mutant AAV pseudotypes have been constructed recently, to improve tissue specificity and safety of gene transfer^{58, 68}. The pseudotype used in human Phase I clinical trial, AAV2/1, contains the capsid sequence of AAV1 and the inverted terminal repeats (ITR) sequences of AAV2. The AAV2 ITRs are incorporated because they have been used in many previous clinical trials and their safety profile is established⁶⁸. The AAV2/1 vector was selected because AAV1 has been shown to be superior to the other AAV serotypes in transducing cardiomyocytes⁶⁴. Newly constructed pseudotypes rAAV2/8 and rAAV2/9 are able to transduce myocardium of neonatal mice at approximately 20- and 200-fold (respectively) higher levels than rAAV2/1^{47, 65}.

Thus, recombinant rAAV vectors hold a great promise for delivering genes to treat heart diseases.

6. Techniques of Gene Delivery

A number of mechanical approaches have been developed for cardiac gene delivery including direct intramyocardial injection, endocardial delivery, ventricular cavity infusion during aortic cross-clamping, pericardial delivery and percutaneous coronary catheterization^{3, 5}.

Direct intramyocardial injection let to efficient local gene delivery, but was often limited by a patchy pattern of vector transfer and procedure related risks^{69, 70}. Intravenous delivery requires injection of at least 10¹¹ AAV genomes to initiate transduction in small animals and is not applicable to human diseases. The best method developed for the rat model involves injecting the virus into the aortic root just above the aortic valve, while the aorta and pulmonary artery are transiently cross-clamped, achieving homogeneous transduction of the myocardium⁷¹.

Endocardial delivery is limited to focal gene transfer to the myocardium. This method of vector delivery for therapeutic angiogenesis has been using in Phase I clinical trials of patients with medically refractory severe angina⁷².

Intracoronary delivery or retrograde infusion of the virus via the coronary veins results are the attractive solutions since all of the myocardial territory supplied by a coronary artery is accessible. Percutaneous catheter-based gene delivery to the myocardium in vivo can be achieved by the endocardial or the intracoronary veins delivery by several approaches:

1. Antegrade Epicardial Coronary Artery Infusion (AECAI). The vector is slowly infused into the coronary arteries⁴².

2. Modified percutaneous antegrade myocardial gene transfer (PAMGT). Catheter-based antegrade intracoronary viral gene delivery with coronary venous blockade⁷³.

3. V-Focus Cardiac Delivery System. The V-Focus system is an investigational device that isolates the coronary circulation from the remainder of the systemic circulation while maintaining perfusion of the myocardium⁷⁴. The V-Focus system enables a closed circuit to be percutaneously established between the coronary arteries and coronary sinus in which the virus vector circulates through the myocardium for approximately 10 min.

The catheter-based techniques were developed for cardiac gene delivery in large animals and may be amenable to translation into human.

7. Clinical trials of heart failure using SERCA2a

The central role of SERCA2a in human HF has been established in patients treated with heart transplant, drugs and devices. Diastolic dysfunction in heart transplant recipients has been shown associated with decreased SERCA2a expression⁷⁵. Improved LV ejection fraction in DCM (dilated cardiomyopathy) patients treated with β-blockers was correlated with increases in SERCA2a mRNA concentrations in myocardial biopsies⁷⁶. Patients with DCM and end-stage HF in whom combined LV assist device and pharmacological therapy induced sufficient clinical recovery to allow removal of the device without recurrence of HF, had increased SR Ca²⁺ content compared with patients who did not recover⁷⁷. In patients (n=11) with advanced CHF, functional improvement related to CRT (cardiac resynchronization therapy) was associated with favorable changes in molecular markers of HF, including increase in SERCA2a mRNA; no significant changes were observed in non responders patients⁷⁸. Improved symptoms and exercise tolerance in HF patients after 3 months treatment with Cardiac Contractility Modulator (CCM) has been associated with increase in SERCA2a mRNA^{79, 80}.

Following a long line of investigation on animal models, gene transfer of SERCA2a has shown tremendous promise and two clinical trials using SERCA2a as a target have been initiated: a Phase I, randomized, double-blinded, placebo-controlled, dose escalation trial of intracoronary administration of AAV1. SERCA2a (MYDICAR™) in patients with CHF (Celladon Corp., CA, USA) and a Phase I study evaluating the safety and biological effects of AAV6. SERCA2a in non-ischemic patients undergoing left ventricular assist device placement.

The investigatioal agent (MYDICAR™) used in the first clinical trial⁴¹ is a rAAV1 vector and consists of a single-stranded 4486 nucleotide DNA containing the human SERCA2a expression cassette (CMV promoter- SERCA2a-bovine growth hormone pA signal) flanked by ITRs derived from AAV2. The MEDRAD Mark V ProVis angiographic injection system is used to produce a slow, diffuse, and homogenous myocardial exposure to AAV1/SERCA2a via AECAI over 10 min⁷⁴.

The objectives of phase I clinical trial are the following: 1) to evaluate the safety and feasibility of a 1 time AECAI administration of MYDICAR™ to subjects with ischemic or nonishemic DCM and Class III/IV symptoms of HF; 2) to explore the activity/efficacy of MYDICAR™ and identify appropriate doses for future studies⁴¹.

The method of cardio-specific vector administration chosen for human testing is AECAI⁴². Vector is slowly infused into the coronary arteries after percutaneous access obtained using a femoral approach. Laboratory assessment is monitored for 12 months after administration of AAV1/SERCA2a and includes following parameters: assessment of cardiac enzymes (creatine kinase-MB and cardiac troponin; BNP and N-terminal prohormone brain natriuretic peptide (NT-proBNP); ELISPOT assay to assess possible anti-AAV1 capsid T-cell response; anti-AAV1 neutralizing antibody assay. Sequential echocardiograms are used to assess response to therapy. Global systolic and diastolic functions are analyzed. Parameters measured include left ventricular sizes and volumes. The primary safety end-point is based on the Stage 1 12-months follow-up data, the primary activity/efficacy analysis is based on the Stage 2 6 –month postdose data.

In the phase 1 trial a total of 12 patients with advanced HF have received intracoronary infusion of AAV1.SERCA2a⁸¹. Of the 9 reported patients treated, several demonstrated improvements from baseline to month 6 of a number of HF parameters⁸¹. Of note, 2 patients who failed to improve had preexisting anti-AAV-1 antibodies.

Conclusion

Advances in the understanding of the molecular basis of HF together with the evolution of increasingly efficient gene transfer technology, has finally placed congestive HF within reach of gene-based therapy. The intracellular handling of calcium is central to the process of excitation-contraction coupling and a number of calcium cycling proteins have been involved in the pathogenesis of HF. Of these, increasing SERCA2a function have been the subject of a large number of gene therapy approaches targeting calcium physiology. Beyond its beneficial effects on contractile function, SERCA2a had profound effect on hypertrophic remodeling, mechanoenergetics, arrhythmogenesis, vascular smooth muscle and endothelial function. These additional effects have heightened the interest in this versatile Ca²⁺ transporter. However the majority of SERCA2a studies have used animal models, which do not precisely mirror human HF. The patients with end-stage HF may have a greater degree of fibrosis or chronic oxidative stress, reducing efficacy of gene therapy. Early results in the Phase I of a first in-human study of AAV1/SERCA2a in advanced HF showed improvement of a number of HF parameters. Further human studies are greatly needed. Phase 2 of this trial which is placebo controlled randomized and blinded with three therapeutic arms and one control arm is currently under way with more than 30 patients having been enrolled. The results of this important phase 2 trial will be in available in 2010.

Expert opinion

The clinical cardiovascular gene therapy experience is very limited. The available data, however, emphasize the importance of randomized placebo-controlled trials in assessing the efficacy of novel treatments. Larger trials of this type are in progress and the results of these are eagerly awaited. Meanwhile, the large volume of preclinical data is likely to continue to expand as knowledge of the molecular basis of cardiovascular disease increases. Importantly, the accumulation of this data provides the scientific basis from which the human evaluation of novel gene-based therapies can be pursued.

With regard to gene transfer technology, recent serious adverse reactions in young human subjects have been attributed to the vector employed. As a result, efforts to improve vector biosafety have been revitalized; vectors will undergo modifications to further improve biosafety and gene transfer efficiency. Conventional treatment modalities have made significant advances over the last decade in the areas of pharmacotherapy, vascular intervention, surgery and implantable devices. The modest success of gene therapy in cardiovascular disease to date has been in filling a therapeutic need. Future success will depend on either improving conventional therapies of finding effective gene-based therapies. In the next few years, a number of Phase I clinical trials using AAV.SERCA2a based gene therapy for heart failure will provide much needed new information in this field.

Essential to the success of the clinical gene therapy trials in heart failure is the ability to safely use AAV vectors targeting the heart. Even though AAV vectors have been proven to be safe in our trials in patients with heart failure, in our recent experience with the clinical gene therapy trials, we have found that their use for gene delivery has the following limitations: 1) they are not specific for the heart and 2) pre-existing neutralizing antibodies to any individual serotype result in 40% of exclusion of these patients from clinical trials. To identify hybrid viruses that can selectively transduce cardiomyocytes in vivo, and which can also escape the antecedent netralizing antibodies, a number of groups have been employing a combinatorial approach to generate a library of diverse AAV variants, obtained by DNA shuffling, with an enrichment of cardiotropic AAV variants, by directed evolution^{82, 83}

This has allowed the generation of cardiotropic chimerics of AAVs that more specifically targets the heart and escapes the inherent immunity in patients. In the next ten years, this approach will be used to generate novel cardiotropic AAV derived vectors that can also escape patients neutralizing antibodies.

Abbreveations

αAR

α-adrenergic receptor

Ca²⁺

calcium ion

CaM

calmodulin

CaMK

Ca²⁺/CaM – dependent protein kinase

DAG

Diacylglycerol

ET-1

Endothelin-1 receptor

Gαq

Gq protein α-subunit

HDAC

Histone deacetylase

IP3

Inositol-1,4,5-triphosphate

IP3R

IP3 receptor

Ica

L-type calcium current

LTCC

L-type calcium channel

MEF

Myocyte-enhancing factor

NFAT

Nuclear Factor of Activated T lymphocytes

Orai

SOC sub-unite

P

Phosphorylation

PLC

Phospholipase C

PLN

Phospholamban

PP2B

serine/threonin phosphatase 2B, calcineurin

RyR

Ryanodine Receptor calcium channel

SR

Sarcoplasmic Reticulum

SERCA

Sarco/Endoplasmic Reticulum Ca²⁺ ATPase 2a

SRF

Serum Response Factor

TRPC

Transient Receptor Potential Ca²⁺ channel, SOC sub-unite

STIM 1

stromal interacting molecule 1, SR calcium sensor., SOC sub-unite