Wild-Type and Mutant HCN Channels in a Tandem Biological-Electronic Cardiac Pacemaker
Background— Biological pacemakers (BPM) implanted in canine left bundle branch function competitively with electronic pacemakers (EPM). We hypothesized that BPM engineered with the use of mE324A mutant murine HCN2 (mHCN2) genes would improve function over mHCN2 and that BPM/EPM tandems confer advantage over either approach alone.
Methods and Results— In cultured neonatal rat myocytes, activation midpoint was −46.9 mV in mE324A versus −66.1 mV in mHCN2 (P<0.05). mE324A manifested a positive shift of voltage dependence of gating kinetics of activation and deactivation compared with mHCN2 (P<0.05) in myocytes as well as Xenopus oocytes. In intact dogs in complete atrioventricular block, saline (control), mHCN2, or mE324A virus was injected into left bundle branch, and EPM were implanted (VVI 45 bpm). Twenty-four–hour ECGs were monitored for 14 days. With EPM discontinued, there was no difference in duration of overdrive suppression among groups. However, basal heart rates in controls were less than those in mHCN2, which did not differ from those in E324A (45 versus 57 versus 53 bpm; P<0.05). When spontaneous rate fell below 45 bpm, EPM intervened at that rate, triggering 83% of beats in control, contrasting (P<0.05) with 26% (mHCN2) and 36% (mE324A). On day 14, epinephrine (1 μg/kg per minute IV) induced a 50% heart rate increase in all mE324A, one third of mHCN2, and one fifth of control (P<0.05 mE324A versus control or mHCN2).
Conclusions— mE324A induces faster, more positive pacemaker current activation than mHCN2 and stable, catecholamine-sensitive rhythms in situ that compete with EPM comparably but more catecholamine responsively than mHCN2. BPM/EPM tandems function reliably, reduce the number of EPM beats, and confer sympathetic responsiveness to the tandem.
Received February 6, 2006; revision received June 5, 2006; accepted June 9, 2006.
Despite the utility of electronic pacemakers in treating heart block and/or sinus node dysfunction, there is need of alternatives that more completely reproduce normal function.1 To this end, we and others have developed various approaches to biological pacemaking. These include overexpression of β2-adrenergic receptors to increase endogenous atrial rates,2,3 dominant negative constructs to reduce the hyperpolarizing inward rectifier current Ik1,4,5 and overexpressed HCN2 channels to increase the rate of impulse initiation.6–8 Each of the 3 approaches manipulates 1 of the basic determinants of native pacemaker function in normal hearts, ie, any intervention that increases sympathetic input decreases repolarizing current and/or increases depolarizing current during diastole should increase the rate of impulse initiation.9 Methods used to achieve these ends have involved gene transfer via viral infection or naked plasmid transfection2,3; use of embryonic stem cells incorporating a complement of native genes10; or adult mesenchymal stem cells engineered as platforms to carry pacemaker genes.8 More recent are attempts to reproduce pacemaker action potentials in noncardiac cells11 and/or to induce fusion of noncardiac and cardiac cells.12
Editorial p 986
Clinical Perspective p 999
One of the key issues in advancing the field of biological pacemaking is identification of a gene or genes that (1) optimize heart rhythm such that excessively long pauses do not occur after sudden failure of endogenous rhythms and (2) induce rhythms having physiologically low basal rates while maintaining an appropriate response to catecholamines and acetylcholine. In addition, a practical issue arises from the need during clinical trials (and likely for some significant time thereafter) to implant electronic along with biological pacemakers in a tandem design. We must ensure that the interactions between the biological and electronic pacemakers are both safe and synergistic and confer more benefit to patients than would electronic pacemakers alone.
To this end, we report here a comparison with regard to basal rates, overdrive suppression, and catecholamine responsiveness between 2 pacemaker genes, cloned murine HCN2 (mHCN2) and the mutant mE324A (a single E to A mutation in mHCN2 at residue 324)13, indicating that the mutant confers a subtle yet important advantage in physiological response. We also will demonstrate that biological and electronic approaches used in tandem can synergize to create a superior functional unit than might characterize either approach alone at present.
Experiments were performed with the use of protocols approved by the Columbia University Institutional Animal Care and Use Committee and conform to the Guide for the Care and Use of Laboratory Animals (National Institutes of Health publication No. 85-23, revised 1996).
Newborn rat ventricular myocyte cultures were prepared as previously described.14 Briefly, 1- to 2-day-old Wistar rats (Charles River Laboratories, Wilmington, Mass) were euthanized, hearts were removed quickly, and ventricles were dissociated with the use of a standard trypsinization procedure. Myocytes were harvested, plated in 35-mm dishes, and incubated in serum-free medium at 37°C, 5% CO2. For electrophysiology experiments, the cell monolayer was briefly exposed to 0.1% trypsin, and cells were replated onto fibronectin-coated coverslips, then studied 2 to 6 hours later.
Adenoviral Constructs and Expression
An adenoviral construct of mouse mHCN2 driven by the CMV promoter (AdmHCN2) was prepared as previously described.15 The mE324A point mutation was introduced into the mHCN2 sequence with the QuikChange XL Site-Directed Mutagenesis Kit (Stratagene, La Jolla, Calif) and packaged in the pDC515 (AdMax, Microbix Biosystems, Inc, Toronto, Ontario, Canada) shuttle vector to create the pDC515mE324A vector driven by the CMV promoter. PDC515mE324A then was cotransfected with pBHGfrtΔE1,3FLP vector (AdMax) into E1-complementing HEK293 cells. Successful recombination of the 2 vectors resulted in production of the adenovirus mE324A (AdmE324A), which was subsequently plaque-purified, amplified in HEK293 cells, and harvested after CsCl banding to achieve a titer of at least 1011 fluorescent focus units per milliliter. For consistency with earlier studies,6 when samples were prepared for in vivo injection, 3×1010 fluorescent focus units of each adenovirus was mixed with an equal amount of a green fluorescent protein (GFP)–expressing adenovirus in a total volume of 700 μL.
Infection of the newborn ventricular myocytes was performed on cell monolayer cultures 4 days after initial plating. Cells were exposed to a virus-containing mix (multiplicity of infection 20, in 250 μL culture medium) for 2 hours, rinsed twice, and incubated in serum-free medium at 37°C, 5% CO2 for 24 to 48 hours before the cells were resuspended as described above for electrophysiological study. Because we had previously found that >90% of cells exposed to AdmHCN2 in vitro expressed the current,15 we did not coinfect with GFP–expressing adenovirus to aid in the selection of infected cells.
The whole-cell patch clamp technique was used to record mHCN2 current from the myocytes as previously described.15 Experiments were performed on cells superfused at 35°C. The external solution contained the following (expressed in mmol/L): NaCl, 140; NaOH, 2.3; MgCl2, 1; KCl, 10; CaCl2, 1; HEPES, 5; glucose, 10; pH 7.4. The external solution had a pH value of 7.4. MnCl2 (2 mmol/L) and BaCl2 (4 mmol/L) were added to block other currents. The pipette solution contained the following (expressed in mmol/L): aspartic acid, 130; KOH, 146; NaCl, 10; CaCl2, 2; EGTA-KOH, 5; Mg-ATP, 2; HEPES-KOH, 10; pH 7.2.
To measure the HCN activation curve, a standard 2-step protocol was used. Hyperpolarizing steps from −25 to −135 mV for mHCN2 and from −5 or −15 to −135 mV for mE324A were applied from a holding potential of −10 mV, followed by a tail current step (to −125 or −135 mV). The duration of test steps was longer at less hyperpolarized potential for mHCN2 channels to more closely approach steady state activation at all voltages. For each cell, the normalized plot of tail current versus test voltage was fit with a Boltzmann function, and then the voltage of half maximum activation (V1/2) and slope factor (s) were defined from the fitting. Activation kinetics were determined from the same tracings, whereas deactivation kinetics were determined from tracings recorded at each test potential after full activation was achieved by a prepulse to −135 mV. Time constants were then obtained by fitting the early time course of activation or deactivation current tracings with a monoexponential function; the initial delay and any late slow activation or deactivation phase were ignored.15,16 Current densities are expressed as the value of the time-dependent component of current amplitude, measured at the end of the test potential and normalized to cell membrane capacitance. Records were not corrected for liquid junction potential, which we previously determined to be 9.8 mV under these conditions.15
Intact Canine Studies
Adult mongrel dogs (Chestnut Ridge Kennels, Chippensburg, Pa) weighing 22 to 25 kg were anesthetized with propofol 6 mg/kg IV and inhalational isoflurane (1.5% to 2.5%). With the use of a steerable catheter, saline (n=5), AdmHCN2 (n=6), or AdmE324A (n=4) was injected into the left bundle branch as described previously.7 In 2 additional dogs, AdmE324A was injected into the left ventricular septal myocardium as an internal control. Complete atrioventricular (AV) block was induced via radiofrequency ablation, and each site of injection was paced via catheter electrode to distinguish electrocardiographically the origin of the idioventricular rhythm during the follow-up period.
An electronic pacemaker (Guidant, Discovery II, Flextend lead, Guidant Corp, Indianapolis, Ind) was implanted and set at VVI 45 bpm. ECG, 24-hour Holter monitoring, pacemaker log record check, and overdrive pacing at 80 bpm were performed daily for 14 days. For each dog, the percent electronic and percent biologically induced beats were calculated daily and then averaged into 3-day bins for each dog. For statistical calculations, the unit of analysis was the individual dog, with each animal represented once in each 3-day bin. To evaluate β-adrenergic responsiveness, on day 14, epinephrine (1.0, 1.5, and 2.0 μg/kg per minute for up to 10 minutes each) was infused to an end point of a 50% increase in idioventricular rate or ventricular arrhythmia (single ventricular premature beats with a morphology other than that of the dominant idioventricular rhythm or ventricular tachycardia), whichever occurred first. If none of the aforementioned responses was observed within 10 minutes after onset of the maximal dose of 2 μg/kg per minute, the infusion was terminated.
Newborn rat myocytes infected with GFP, mHCN2, or mE324A were scraped from 35-mm dishes in Laemmli sample buffer and gently sonicated. Protein concentrations were determined with a modified Lowrey method, and 40 μg of whole-cell lysates was separated by 7.5% SDS-PAGE (Bio-Rad Laboratories, Hercules, Calif) and then transferred to nitrocellulose membrane. Immunoblotting with anti-HCN2 antibodies (Alomone Labs, Jerusalem, Israel) was performed according to manufacturer’s instructions. Proteins were visualized on x-ray film (Ewen Parker X-ray Corp, Elmsford, NY) with the use of ECL chemiluminescence (Pierce Biotechnology, Inc, Rockford,Ill).
Data are presented as mean±SEM. In cell culture studies, experimental data were compared with a Student t test or χ2 test with Yates correction or 2-way ANOVA, as appropriate. To minimize the effect of culture-to-culture variability, for each culture cells were prepared expressing each construct, and data from at least 3 separate cultures were pooled for each comparison. In experiments in situ, the 5 saline-injected dogs and the 2 injected into the myocardium (rather than left bundle branch) with AdmE324A showed no electrophysiological differences and were combined into 1 control group for subsequent analysis. One-way ANOVA was used to evaluate the effect of an implanted construct on electrophysiological parameters. Subsequent analysis was performed with the Bonferroni test, in which equal variances were assumed, and the Games-Howell test, in which variances were unequal. A 2-way contingency table analysis was conducted to evaluate whether epinephrine had different effects across 3 groups. Data were analyzed with the use of SPSS for Windows software (SPSS, Inc, Chicago, Ill). P<0.05 was considered significant.
The authors had full access to the data and take full responsibility for its integrity. All authors have read and agreed to the manuscript as written.
Studies of Myocytes
Biophysical Properties of mHCN2 and mE324A Expressed in Myocyte Cultures
In whole-cell voltage clamp experiments, mHCN2- and mE324A-expressing myocytes both gave rise to an inward current in response to hyperpolarizing voltages. Representative normalized current tracings obtained at test potentials ranging from −25 to −125 mV, from a holding potential of −10 mV, are shown in Figure 1A and 1B. It is apparent from the expanded currents in the insets that the activation threshold of mE324A channels is less negative than that of mHCN2 channels.
The difference in voltage dependence of activation between mHCN2 and mE324A is more evident from the mean activation relationships shown in Figure 1C. The curves were obtained from tail currents, as described in Methods. The individual activation curves were each fit to the Boltzmann equation, and the calculated midpoint (V1/2) and slope factor (s) from all cells were averaged and compared statistically. Mean parameters for mHCN2-expressing (n=14) and mE324A-expressing (n=16) cells, respectively, were as follows: V1/2=−66.1±1.5 mV and −46.9±1.2 mV (P<0.05) and s=10.7±0.5 mV and 9.6±0.4 mV (P>0.05). Thus, in agreement with data obtained previously in oocytes13 and confirmed by us (see online-only Data Supplement), the mE324A mutation resulted in a positive shift of the activation curve relative to that of mHCN2 when both constructs were expressed in newborn myocytes.
The activation kinetics of mE324A channels appeared faster than those of mHCN2 (Figure 1A and 1B insets). To demonstrate this difference, time constants of activation and deactivation were measured at different voltages (see Methods) and averaged (Figure 1D). These data show that the faster activation kinetics observed for mE324A channels were due to a positive shift of the voltage dependence of gating kinetics. Both activation and deactivation voltage dependence shifted positively, so that at the positive voltages at which deactivation was measured, the deactivation was slower for mE324A than for mHCN2. Moreover, this shift is comparable to that in the current-voltage relationship. Indeed, the relative peaks of the kinetic-voltage relations were consistent with the previously determined V1/2 values.
Finally, we investigated whether the mutant channel expressed current as well as the wild-type channel. The percentage of myocytes expressing mE324A current was significantly smaller than the percentage expressing mHCN2 (36.6% of 93 cells versus 74.5% of 47 cells, respectively; P<0.05 by χ2 test) in 6 cell cultures. Moreover, in the cells that expressed current, the mE324A current density (measured at −135 mV) was &2.5 times smaller than that of mHCN2 (21.0±3.5 pA/pF, n=12, versus 53.5±8.7 pA/pF, n=10, respectively; P<0.05). These functional data were consistent with Western blot analysis of protein expression (Figure 1E), which showed markedly less detection in mE324A-infected cultures than in mHCN2 cultures.
The positive shift in the activation relation and kinetics would be expected to result in more current being passed earlier in the cardiac cycle with mE324A in comparison to mHCN2. To be beneficial as a biological pacemaker, however, it also is necessary to preserve autonomic responsiveness. To assess this, mHCN2 and mE324A activation curves were compared in the absence and in the presence of cyclic adenosine monophosphate (cAMP) in the pipette solution (Figure 2). Both channels responded to the presence of saturating intracellular cAMP, as detailed in the Figure 2 legend.
Intact Animal Studies
The electronic pacemaker triggered 83±5% of all beats in the 7 controls, contrasting (P<0.05) with 26±6% in the 6 mHCN2 and 36±7% in the 4 mE324A animals (for the latter 2, P>0.05). A temporal analysis of the electronically paced beats for the HCN2 versus the electronic pacemaker is shown in Figure 3A. Note that a significantly lower number of beats was initiated electronically in the mHCN2 group throughout the study period. Results for mE324A did not differ significantly from mHCN2 and are not shown here.
We evaluated escape time daily by performing three 30-second periods of ventricular overdrive pacing at 80 bpm followed by an abrupt cessation of pacing. The average time between the final electronically paced beat and the first intrinsic beat was then determined. Escape times ranged from 1 to 5 seconds across all 3 groups and incorporated a wide variability, such that no significant differences were seen. Hence, no advantage accrued to any group with regard to escape intervals. There was a different result with regard to basal heart rates throughout the 14-day period, however. As shown in Figure 3B, average heart rate in saline controls was that determined by the rate of the electronic pacemaker (45 bpm). This was significantly slower throughout the study than that of mHCN2- or mE324A-injected dogs, which groups did not differ from one another.
An example of the interrelationship between the biological and the electronic components of the tandem is shown in Figure 4. Note that as the biological component slows, the electronic takes over, and that as the biological component speeds in rate, the electronic ceases to fire.
Figure 5 demonstrates the response to epinephrine in terminal experiments. Figure 5A shows representative ECGs for all 3 groups before and during infusion of epinephrine 1 μg/kg per minute. Control rates were 42, 44, and 52 bpm for the saline, mHCN2, and mE324A animals, respectively. With epinephrine, rates increased to 44, 60, and 81 bpm. Figure 5B summarizes the rate changes occurring at all doses of epinephrine. The mE324A group manifested a >50% increase in heart rate in all dogs at the lowest dose of epinephrine. No animals showed ectopy. Only one half of the mHCN2 group generated a ≥50% increase in heart rate, and the remainder had either a <50% increase in heart rate or the occurrence of ventricular premature depolarizations. Finally, in the saline group, all dogs showed a <50% increase in rate and/or ventricular premature depolarizations throughout the range of epinephrine concentrations. Hence, there was greater epinephrine sensitivity in the mE324A group than in either of the other groups.
Our data provide information of interest at 3 levels: (1) with regard to the utility of studying ion currents as predictors of biological pacemaker function in situ; (2) with regard to the direct comparison between wild-type mHCN2 and a mutant channel mE324A; and (3) with regard to the potential for tandem therapy as an alternative to either electronic pacemaking or biological pacemaking as a unique therapy.
Biophysical Properties of Ion Currents as Predictors of Biological Pacemaker Function
Our studies in neonatal rat myocytes (Figures 1 and 2⇑) and in Xenopus oocytes (online-only Data Supplement, Figures I to IV) gave concordant results with regard to the function of mHCN2 and mE324A. That the mE324A mutation induced faster, more positive pacemaker current activation in these in vitro settings than did mHCN2 might be interpreted as suggesting that the mutant channel would result in a faster pacemaker rate and/or a shorter escape interval after overdrive pacing than occurred in saline-injected or mHCN2-injected hearts. However, this was not the case: Both the saline- and mHCN2-injected hearts showed escape times equivalent to those of the mE324A-injected hearts. As for automatic rates per se, these were equivalent for mHCN2- and mE324A-injected hearts, and both were significantly faster than those injected with saline. In other words, for 2 important descriptors, rate attained and overdrive suppression, there was no clear discrimination between the effects of mHCN2 and mE324A in situ. One explanation for this may be that the percentage of myocytes expressing mE324A current was significantly less than that expressing mHCN2. Moreover, there was a lesser current density in the E324A group, and this was associated with reduced protein detection on Western blot. Thus, although a greater fraction of channels activates faster at a given voltage with mE324A compared with mHCN2, the total number of channels available or net current flow may be approximately equivalent at physiologically relevant voltages such as −55 mV (see insets in Figure 1).
To understand the enhanced responsiveness of mE324A to catecholamines, we considered the following: The inward current contributed by If in the diastolic range of potentials is determined by the number of f channels opened during diastolic depolarization. This is determined by 3 variables, the maximal f conductance, the steady state open probability at a given potential, and the rate at which the f gating variable approaches the steady state value. Given the bell-shaped relation of the kinetics-voltage relation of HCN channels (Figure 1D), the position of the control HCN2 activation relation within the diastolic potential range is such that a positive shift with cAMP might actually slow kinetics of activation at some diastolic potentials, thereby dampening the effects of β stimulation. In contrast, the much more positive position of the mE324A activation relation guarantees that the activation kinetics are much faster throughout the diastolic voltage range, and a further positive shift by cAMP will therefore accelerate activation at all diastolic potentials. Therefore, it is reasonable to suggest that this more rapid rate of activation at all diastolic potentials underlies the enhanced response to catecholamines.
Potential for Tandem Therapy as an Alternative to Either Electronic or Biological Pacemaking
In this section we will first discuss the stability of biological pacemaking in its own right and then the relationship of biological to electrical activity in a tandem mode. The stability of biological pacemaking can be considered in light of the data in Figure 3. Note that in the first 24 to 72 hours, there is expression of spontaneous pacemaker activity 80% of the time. As previously described by us,7 this reflects both biological pacemaker activity and ectopic activity induced by injury secondary to injection during the first 24 to 72 hours after injection. Specifically, both saline- and HCN2-injected animals manifest the same pacemaker activity on days 1 to 3.7 Because both injury-induced and pacemaker-induced rhythms originate at the same site at this time, it is impossible to distinguish them via pace mapping. In contrast, in Figure 3, between days 3 to 12 biological activity attributable to the HCN2 injection is stable at 70% of the beats, whereas in saline-injected dogs, the electronic pacemaker is responsible for 80% to 90% of beats. After this time, biological activity begins to wane, as is seen on days 13 to 14. Indeed, in animals that we have carried to 2.5 to 3 weeks, little biological activity remains (data not shown).
Note as well that the electronic pacemaker was set at a rate of 45 bpm, in a demand mode. As a result, any time the rate of the biological pacemaker fell below 45 bpm the electronic pacemaker intervened. Dogs of the size we studied generate a wide range of sinus rates over a 24-hour period.17 These rates are conditioned by their own sinus arrhythmia influenced by autonomic control while awake and asleep. With regard to autonomic control, we have previously demonstrated the vagal control of the HCN2-based biological pacemaker in situ6 and show in the present report that there is β-adrenergic modulation as well. Moreover, a biological pacemaker placed in ventricular muscle will be in contact with a greater number of cells that express IK1 than in sinus node, and this will have a damping effect on pacemaker rate. At any time that biological pacemaker rate falls below 45 bpm, as during sleep or rest and/or in association with enhanced vagal or decreased sympathetic input, the electronic pacemaker would be expected to take over. That this happened does not reflect a failure of the biological pacemaker but its normal variability. In future studies, we will document heart rate variability with the biological pacemaker and will study autonomically blocked animals as well to further test the extent of autonomic control.
With regard to tandem pacemaking, we previously have considered the strengths and weaknesses of electronic pacemakers.1,18,19 Clearly, they are the state of the art as life-saving devices for treating a number of cardiac arrhythmias and are being used increasingly for cardiac failure. These advantages more than outweigh their disadvantages, which include (1) their inability to respond to the demands of emotion or exercise (although software has been developed to facilitate variations in heart rate while exercising); (2) the requirement for monitoring and maintenance, including periodic pulse generator changes and, at times, lead replacement; (3) the need to adapt units and leads to the demands of growth and development in children; (4) limitations in sites where leads can be stably implanted, which may compromise cardiac output to variable extents; (5) problems with infection, which although infrequent can be catastrophic; and (6) the potential for interference from other devices.1,18,19
Because electronic pacemakers represent a highly successful form of medical palliation, they will not easily be replaced, but the fact that they are not completely physiological makes them a target for improvement/replacement. However, the only therapy that should replace them is one that is more long-lasting, has less potential for inflicting damage, and is more physiological. It is with this in mind that biological pacemakers are being developed. We have suggested that they should have the potential to (1) create a lifelong, stable physiological rhythm without need of replacement; (2) compete effectively with electronic pacemakers in satisfying the demand for a safe baseline rhythm coupled with autonomic responsiveness to facilitate responsiveness to the demands of exercise and emotion; (3) be implanted at sites adjusted from one patient to another such that propagation through an optimal pathway of activation occurs and efficiency of contraction is optimized; (4) confer no risk of inflammation, neoplasia, or rejection; and (5) have no arrhythmogenic potential. In other words, they should represent not palliation, but cure.1,18
Given this assignment of properties of yin and yang to electronic and to biological pacemakers, why should we consider tandem therapy? There are 2 reasons, one associated with clinical trials and the other associated with more widespread clinical use. After the appropriate safety and efficacy preclinical testing is completed, a study of tandem pacemaking in patients in complete heart block and atrial fibrillation would be a reasonable starting point for a combined phase 1/phase 2 trial. Such a population has need of pacemaker therapy and is not a candidate for AV sequential electronic pacing. The state-of-the-art therapy for such patients—a demand form of electronic ventricular pacing—would be indicated, and a biological implant could be made as well. Moreover, the electronic component set at a sufficiently low rate would ensure a “safety net” in case the biological component failed. Even if phase 1 and phase 2 trials provide evidence of safety and efficacy of the biological pacemaker, however, there is a need to understand how long a biological pacemaker will last. In the first generation of patients to receive them, this should likely be a lifelong question, during which there must be continued electronic backup.
With respect to broader clinical application of the tandem pacemaker concept, there are several issues to consider. First, the system is redundant by design and would have 2 completely unrelated failure modes. Two independent implant sites and independent energy sources would provide a safety mechanism in the event of a loss of capture (eg, due to myocardial infarction). Second, the electronic pacemaker would provide not only a baseline safety net but an ongoing log of all heartbeats for review by clinicians, thus providing insight into a patient’s evolving physiology and the performance of the tandem pacemaker system. Third, because the biological pacemaker will be designed to perform the majority of cardiac pacing, the longevity of the electronic pacemaker could be dramatically improved. Alternatively, longevity could be maintained while the electronic pacemaker could be further reduced in size. Finally, the biological component of a tandem system would provide true autonomic responsiveness, a goal that has eluded ≥40 years of electronic pacemaker research and development.
This study demonstrates the feasibility of engineering a biological pacemaker to meet the demands placed on modern electronic pacemakers and specifically to provide a physio- logical basal heart rate and a means to elevate heart rate during times of increased demand. mHCN2 and mE324A provide biological pacemakers with different characteristics, yet they demonstrate the principle that biological pacemakers, like their electronic counterparts, can be tuned for basal heart rate and catecholamine responsiveness. In exploring the development of a biological pacemaker, we have considered the possibility that the electronic pacemaker in concert with a biological cure may provide an essential bridge to the future of biological therapeutics. Although the bridge may lead us to a future of pure biological therapies, it may itself be an interesting destination providing greater benefit to patients and clinicians.
The authors express their gratitude to Laureen Pagan for her careful attention to the preparation of the manuscript.
Sources of Funding
This work was supported by US Public Health Service–National Heart, Lung, and Blood Institute grants HL-28958 and HL-67101 and by Guidant Corporation.
Drs KenKnight, Girouard, and Qu were employees of Guidant Corp at the time the research was performed. The other authors report no conflicts.
Rosen MR, Brink PR, Cohen IS, Robinson RB. Genes, stem cells and biological pacemakers. Cardiovasc Res. 2004; 64: 12–23.
Edelberg JM, Huang DT, Josephson ME, Rosenberg RD. Molecular enhancement of porcine cardiac chronotropy. Heart. 2001; 86: 559–562.
Qu J, Plotnikov AN, Danilo P Jr, Shlapakova I, Cohen IS, Robinson RB, Rosen MR. Expression and function of a biological pacemaker in canine heart. Circulation. 2003; 107: 1106–1109.
Plotnikov AN, Sosunov EA, Qu J, Shlapakova IN, Anyukhovsky EP, Liu L, Janse MJ, Brink PR, Cohen IS, Robinson RB, Danilo P Jr, Rosen MR. A biological pacemaker implanted in the canine left bundle branch provides ventricular escape rhythms having physiologically acceptable rates. Circulation. 2004; 109: 506–512.
Potapova I, Plotnikov A, Lu Z, Danilo P Jr, Valiunas V, Qu J, Doronin S, Zuckerman J, Shlapakova IN, Gao J, Pan Z, Herron AJ, Robinson RB, Brink PR, Rosen MR, Cohen IS. Human mesenchymal stem cell as a gene delivery system to create cardiac pacemakers. Circ Res. 2004; 94: 841–959.
Biel M, Schneider A, Wahl C. Cardiac HCN channels: structure, function, and modulation. Trends Cardiovasc Med. 2002; 12: 202–216.
Cho HC, Kashiwakura Y, Azene E, Marban E. Conversion of non-excitable cells to self-contained biological pacemakers. Circulation. 2005; 112: II-307. Abstract.
Cho HC, Kashiwakura Y, Marban E. Creation of a biological pacemaker by cell fusion. Circulation. 2005; 112: II-307. Abstract.
Chen J, Mitcheson JS, Tristani-Firouzi M, Lin M, Sanguinetti MC. The S4-S5 linker couples voltage sensing and activation of pacemaker channels. Proc Natl Acad Sci U S A. 2001; 98: 11277–11282.
Altomare C, Bucchi A, Camatini E, Baruscotti M, Viscomi C, Moroni A, DiFrancesco D. Integrated allosteric model of voltage gating of HCN channels. J Gen Physiol. 2001; 117: 519–532.
Detweiler DK. The dog electrocardiogram: a critical review. In: MacFarlane PW, Veitch Lawrie TD, eds. Comprehensive Electrocardiology: Theory and Practice in Health and Disease, Vol. 2. Elmsford, NY: Pergamon Press; 1989: 1267–1329.
We have previously shown that biological pacemakers based on the HCN2 pacemaker gene and implanted as adenoviral constructs into the canine left bundle branch function well during high-degree atrioventricular block. We now hypothesized that biological pacemakers engineered with the use of a specific mutant (E324A) HCN2 gene would improve function over HCN2 and that a biological-electronic pacemaker tandem would confer advantage over either approach alone. We tested the mutant HCN2 gene in cultured neonatal rat myocytes. The mutant gene manifested biophysical differences from the parent HCN2 gene: There were more rapid, more catecholamine-sensitive rhythms with the mutant. In intact dogs in complete atrioventricular block, saline (control), HCN2, or E324A constructs were injected into the left bundle branch, and an electronic pacemaker was implanted (demand mode at 45 bpm). Twenty-four–hour ECGs were monitored for 14 days. Basal heart rates in controls were less than those in mHCN2, which did not differ from those in E324A. When spontaneous rate fell below 45 bpm, electronic pacing intervened at that rate, triggering 83% of beats in controls, contrasting significantly with both HCN2 and E324A (26% and 36% electronic beats, respectively). On day 14, epinephrine (1 μg/kg per minute IV) induced a 50% heart rate increase in all E324A, one third of HCN2, and only one fifth of controls. We concluded that E324A induces faster, more positive pacemaker current activation than HCN2 and stable, catecholamine-sensitive rhythms in situ that compete with electronic pacemakers comparably but more catecholamine responsively than HCN2. Tandem biological-electronic pacemakers function reliably, reduce the number of electronic beats, and confer sympathetic responsiveness to the tandem.
↵*The first 3 authors contributed equally to this work.
The online-only Data Supplement is available with this article at http://circ.ahajournals.org/cgi/content/full/CIRCULATIONAHA.106.617613/DC1.