Expression and Function of a Biological Pacemaker in Canine Heart
Background— We hypothesized that localized overexpression of the hyperpolarization-activated, cyclic nucleotide-gated (HCN2) pacemaker current isoform in canine left atrium (LA) would constitute a novel biological pacemaker.
Methods and Results— Adenoviral constructs of mouse HCN2 and green fluorescent protein (GFP) or GFP alone were injected into LA, terminal studies performed 3 to 4 days later, hearts removed, and myocytes examined for native and expressed pacemaker current (If). Spontaneous LA rhythms occurred after vagal stimulation-induced sinus arrest in 4 of 4 HCN2+GFP dogs and 0 of 3 GFP dogs (P<0.05). Native If in nonexpressed atrial myocytes was 7±4 pA at −130 mV (n=5), whereas HCN2+GFP LA had expressed pacemaker current (IHCN2) of 3823±713 pA at −125 mV (n=10) and 768±365 pA at −85 mV.
Conclusions— HCN2 overexpression provides an If-based pacemaker sufficient to drive the heart when injected into a localized region of atrium, offering a promising gene therapy for pacemaker disease.
Received December 6, 2002; revision received January 21, 2003; accepted January 21, 2003.
Implantable electronic devices have represented state-of-the-art therapy for high degrees of heart block since the 1960s. Such devices save lives, and refinements in design have made them far more palatable to patients than they had been originally. Nonetheless, the ideal pacemaker, in terms of both physiological function of the heart and adaptability to the human body, would be biological.1–5 The search for such a pacemaker has centered on 3 gene therapy strategies: (1) upregulation of β2-adrenergic receptors by transfecting cloned receptors that increase heart rate responses to adrenergic input1,2; (2) viral infection producing dominant negative inhibition of inwardly rectifying potassium current (IK1), such that the balance of inward currents suffices to depolarize ventricular myocardial cells5; and (3) adenoviral transfer of α (hyperpolarization-activated cyclic nucleotide-gated [HCN2])3 and/or β (minK-related peptide 1 [MiRP1])6 subunits of the endogenous human pacemaker current to induce autonomically responsive pacemaker function in ventricular myocytes. The first two approaches have seen proof of concept demonstrated in animal models.1,5 The third approach might be less problematic and proarrhythmic in that it incorporates the endogenous pacemaker channel gene, which selectively activates only during diastole. The present study provides proof of concept that HCN2 overexpression locally in left atrium (LA) induces both current and in situ pacemaker function.
Protocols were approved by the Columbia University Animal Care and Use Committee.
We prepared an adenoviral construct of mouse HCN2 (mHCN2, GenBank AJ225122) driven by the cytomegalovirus promoter, as previously described.3 The construct AdHCN2 was purified through plaque assay, amplified to a large stock, and harvested and titrated after CsCl banding. The same procedure was used to construct an adenoviral vector of enhanced green fluorescent protein (AdGFP), the sequence of which was taken from its original vector pIRES2-EGFP (Clontech) at BamHI and NotI sites and subcloned into the shuttle vector pDC516. The final titer for AdHCN2 was 3.4×1011 ffu/mL and AdGFP 1.4×1012 ffu/mL. In each experiment, 2 to 3×1010 ffu of each virus was injected.
Intact Animal Studies
Under sterile conditions and after sodium thiopental induction (17 mg/kg IV) and inhalational isoflurane (1.5% to 2.5%) anesthesia, 23- to 27-kg male or female mongrel dogs were subjected to a pericardiectomy. We injected AdGFP+AdHCN2 or AdGFP alone subepicardially in 0.6 mL of solution into the root of the LA appendage and sewed a reference electrode to the right atrium.
Animals recovered for 3 to 4 days and then were anesthetized. Both cervical vagal trunks were isolated and the chest opened. During continuous ECG monitoring, graded right and/or left vagal stimulation was performed via bipolar platinum iridium electrodes7 to suppress sinus rhythm such that escape pacemaker function might occur. Then, 1 cm of atrium surrounding the injection site was excised for study of ion currents.
Dissociation of Myocytes and Studies of HCN Current
Atrial myocytes were dissociated by modifying a previously published procedure.8 Excised atrial tissue was cut into strips, triturated in collagenase-protease solution, and exposed to 2 to 4 digestion cycles in enzyme. Enzyme concentrations were reduced after the initial cycle.
Isolated cells were transferred to a stage-mounted chamber of an inverted epifluorescence microscope to identify green fluorescent protein (GFP)-expressing cells. To measure pacemaker currents, cells were superfused with 35°C Tyrode solution containing (in mmol/L): NaCl, 140; NaOH, 2.3; MgCl2, 1; KCl, 10; CaCl2, 1.0; MnCl2, 2; BaCl2, 4; HEPES, 5; and glucose, 10 (pH 7.4). Pipette solution contained (in mmol/L): aspartic acid, 130; KOH, 146; NaCl, 10; CaCl2, 2; EGTA-KOH, 5; Mg-ATP, 2; and HEPES-KOH, 10 (pH 7.2). To record pacemaker current, cells were held at −50 or −55 mV and stepped to −65 to −145 mV for 6 seconds, followed by an 8-second step to −125 mV to measure tail current.
Fisher’s exact test or Student’s t test was used as appropriate. Data are expressed as mean±SEM. P<0.05 was considered significant.
Intact Animal Studies
Three animals received GFP alone. None showed spontaneous atrial rhythms during vagal stimulation, whereas all four dogs receiving GFP+HCN2 showed spontaneous rhythms during vagal stimulation (P<0.05 versus GFP alone). Moreover, mapping with a hand-held electrode demonstrated early LA activation above the injection site (representative experiment, Figure 1). Increased vagal stimulation terminated atrial activity (data not shown).
There was no measurable native pacemaker current (If) in 5 nonexpressing cells from injected animals, but we identified an If of 7±4 pA at −130 mV in 4 of 5 cells from control animals studied with cAMP in the pipette.9 Membrane capacitance was 97.4±29.6 pF and 90.9±12.9 pF, respectively (P>0.05). In contrast, 10 atrial cells receiving GFP+HCN2 manifested expressed pacemaker current (IHCN2) 500-fold greater, 3823±713 pA (45.3±12.5 pA/pF) at −125 mV and 768±365 pA (8.6±3.8 pA/pF) at −85 mV (representative experiments, Figure 2, A and B). Membrane capacitance here was 105±15.2 pF, not different from the others (P>0.05). The IHCN2 activation threshold was ≈−75 mV. The activation-voltage relation generated by the HCN2 tail current at −125 mV is shown in Figure 2C. Mean voltage of half-maximum activation was −95.2±0.4 mV, and slope factor was 7.9±0.4 mV.
Complete heart block and sinus node dysfunction are major indications for pacemaker implantation.10 Although such electronic devices have had excellent success and minimal morbidity, it would be optimal to offer a therapy incorporating the same plasticity as the normal sinus node and atrioventricular conducting system. Several approaches to provide biological pacemaker function have been tested. After injection of plasmids to overexpress human β2-adrenergic receptors in porcine right atrium, heart rates were 50% faster than those of controls.1 Although clearly demonstrating the value of gene transfer to modify pacemaker function, this study1 focused on modulating native pacemaker cells rather than the pacemaker itself.
A dominant negative strategy to reduce IK1, which normally maintains ventricular myocytes at negative membrane potentials, induced spontaneous impulse initiation in guinea pig heart.5 This provided proof of concept for biological pacemaking, although the site of pacemaker initiation was not identified in vivo, and there was no attempt to achieve localized expression of pacemaking. There is little likelihood that the source of inward current providing pacemaker function derived from If because If magnitude is below the limit of measurement at physiological voltages in normal guinea pig ventricle.11
The present approach to biological pacemaking, using the molecular correlate of If, recruits the pacemaker current endogenous to the heart. The unique voltage dependence of the pacemaker channel means that additional current will flow during diastole and not during the action potential. We anticipated that impulses might be initiated in atrium, given its low IK1 and relatively positive activation of native pacemaker current.12,13 Similarly, we expect success on administration of this construct into Purkinje fibers, inasmuch as these, too, have low IK114 and relatively positive native If.9 In many ways the proximal bundle-branch system would be an optimal site for administering pacemaker constructs because it provides organized propagation to the ventricles. This possibility is presently being tested.
There are still many obstacles to overcome before such pacemaker constructs become feasible for clinical testing. Our current practice is to induce transient expression, the duration of which is only as long as virus and resulting protein construct survive in the host. Needed to ensure long-term function of such constructs are identification and utilization of an appropriate delivery system in which the construct is effectively immortalized. Although other candidates for these functions are currently available, uncertainties remain with regard to safety (eg, retrovirus) and permanence of expression (eg, adeno-associated virus). Regulating the level of expression to achieve optimal pacemaker rate is also critical.
Autonomic responsiveness of this pacemaker is another issue. We have demonstrated that HCN2 overexpressed in cultured myocytes responds to alterations in intracellular cAMP3 as well as to the β-adrenergic and muscarinic agonists that physiologically regulate cellular cAMP (data not shown). This suggests that HCN-based biological pacemakers would be subject to modulation by autonomic neurohumors. Moreover, as stated with regard to Figure 1, the LA pacemaker site was suppressed by vagal stimulation, which suggests this limb of autonomic responsiveness is intact.
In summary, we have developed a biological pacemaker using an HCN family isoform that is responsible for pacemaker current in mammalian heart. Although significant issues require solution before testing such constructs in human heart, there are now enough data to suggest that application of such a pacemaker can be effectively explored in animal models. Additional questions to be answered relate to ideal sites for implantation (atrium, Purkinje system), the extent to which diseased sinus nodes can be returned to normal function, and the ideal constructs to be used. Among the candidates are not only HCN2, but also the accessory subunit, MiRP1, which coexpresses in cell membranes with HCN2 and increases pacemaker current.6 Also to be considered are various mutations to HCN channels, whose properties of activation might render them more favorable to initiating impulses in the physiological milieu.
This study was supported in part by a United States Public Health Service-National Heart, Lung and Blood Institute grant (HL-28958), the Guidant Corporation, and the Heritage Affiliate of the American Heart Association. We are grateful for Dr Hamish Young’s contributions to the construction of viruses used in this study, to Drs Maria Obreztchikova, Alessandra Besana, and Lev Protas for assisting with experiments, and Eileen Franey for her careful attention to the preparation of the manuscript.
↵*Drs Qu and Plotnikov contributed equally as first authors.
Edelberg JM, Huang DT, Josephson ME, et al. Molecular enhancement of porcine cardiac chronotropy. Heart. 2001; 86: 559–562.
Qu J, Barbuti A, Protas L, et al. HCN2 over-expression in newborn and adult ventricular myocytes: distinct effects on gating and excitability. Circ Res. 2001; 89: e8–e14.
Rosen MR, Robinson RB. Heart rate: a simple yet complex concept. Dialogues Cardiovasc Med. 2001; 6: 3–16.
Rosenshtraukh L, Danilo P Jr, Anyukhovsky EP, et al. Mechanisms for vagal modulation of ventricular repolarization and of coronary occlusion-induced lethal arrhythmias in cats. Circ Res. 1994; 75: 722–732.
Protas L, DiFrancesco D, Robinson RB. L-type, but not T-type calcium current changes during post-natal development in rabbit sinoatrial node. Am J Physiol. 2001; 281: H1252–H1259.
Yu H, Chang F, Cohen IS. Pacemaker current exists in ventricular myocytes. Circ Res. 1993; 72: 232–236.
Hoppe UC, Beuckelmann DJ. Characterization of the hyperpolarization-activated inward current in isolated human atrial myocytes. Cardiovasc Res. 1998; 38: 788–801.
Melnyk P, Zhang L, Shrier A, et al. Differential distribution of Kir2.1 and Kir2.3 subunits in canine atrium and ventricle. Am J Physiol (Heart Circ Physiol). 2002; 283: H1123–H1133.