Contribution of pacemaker channels to autonomous electrical activity of differentiated embryonic stem cells

The heart functions like a pump derived by electrical impulses known as action potentials. A remarkable feature of cardiac action potentials is that they initiate autonomously by specialized cells in the right atrium located in the sinoatrial (SA) node. The action potentials generated by the SA node are then conveyed to atrial myocytes and subsequently to the atrio-ventricular node where the electrical activity is delayed, allowing the atrium to pump the blood into the ventricles. Thereafter, the electrical excitation of ventricles ensues through fibres known as bundle branches. During embryonic development and organogenesis of heart, intrinsic pacemaker activity is one of the many requirements for generation of autonomous electrical impulse and cardiac function. An equally critical aspect of cardiac development is the generation of non-pacing contracting myocytes such as the atrail and ventricular myogenic lineage.

Propagation of action potentials in heart, and hence the ability to pump blood into circulation, relies on flow of ions in and out of cardiac myocytes. This intricate balance of ionic flow is generated by proteinaceous gates referred to as ion channels. The so called pacemaker channels, also known as funny channels, I_f or I_h, control the autonomous electrical activity of the heart as well as the pacemaker activity in certain regions of the brain. Electrophysiological studies have identified distinct functional hallmarks of I_f channels which make them good candidate for pacemaker activity (reviewed in Accili et al. 2002). Unlike most voltage-gated channels, I_f is activated upon membrane hyperpolarization. This property allows inward flow of sodium ions during the repolarization phase of the action potential which then depolarizes the membrane potential back towards a threshold for generation of the next action potential. In addition, I_f current is directly modulated by cyclic nucleotides, independent of kinases, thus allowing quick changes in the duration of action potential and the rapid control of heart beat by the autonomic nervous system. The molecular identity of I_f channels remained largely unknown until three independent groups simultaneously cloned its gene approximately 20 years after it was first discovered by electrophysiological measurements (Accili et al. 2002 and therein). When expressed in heterologous systems, isolated cDNA clones produce currents that are activated by voltage and modulated by cyclic nucleotides. With regard to this dual mode of activation, these channels are now referred to as hyperpolarization-activated cyclic nucleotide-modulated (HCN). In vertebrates, four such genes (HCN1–4) have been identified.

Autonomous pacemaker activity of the heart starts remarkably early during embryogenesis and it is considered one of the first signs of a healthy human embryo. In order to investigate the molecular bases of early cardiac activity, Qu et al. (2008) have examined the contribution of the pacemaker channels to spontaneous beating of cells derived from mouse embryonic stem cells (mESCs) during development in culture, chosen arbitrarily after plating differentiated cells for 2–4 days (early), 5–8 days (intermediate) or 9–15 days (late). Briefly, their study examines the current density of I_f, its correlation to rhythmicity and HCN isoform expression as well as its modulation by automonic stimulation in early, intermediate and late developmental stages. The authors have drawn several important conclusions which significantly advance our understanding of the mechanisms involved in early rhythmic activity of cardiac stem cells and which suggest that stem cells can be utilized as a faithful model recapitulating the features of heart development. Qu et al. (2008) start by demonstrating a developmental increase in the proportion of cell expressing I_f and the density of I_f per individual cell concomitant with an increase in action potential frequency and beating rate. The authors then examine the effect of a specific I_f blocker (ZD7288) on rhythmicity of single cells and report that micromolar concentrations of ZD7288 reduce beating frequency and inhibit I_f current in mESCs. These experimental observations make a solid case for the involvement of the pacemaker current in the maintenance of fast and regular beating rate of mESCs. In their study, Qu et al. (2008) also correlate functional expression of the native pacemaker current to expression of specific HCN isoforms. Of the four isoforms tested, HCN2 and HCN3 are the only two that were detectable by Western blots which were performed using vigilantly selected antibodies. Whereas the expression levels of HCN2 remain constant throughout the development, the expression of HCN3 is dramatically decreased in late developmental stages. Qu et al. (2008) then go on to investigate the role of autonomic regulation in mESCs and they test whether modulation of I_f current contributes to regulation of rhythmicity. They find that treatment of cells with the β-adrenergic agonist isoproterenol (isoprenaline) significantly increased beating rate in early and late developmental stages. Accompanying the increase in beating rate is a positive shift in activation range of I_f current by isoproterenol. On the other hand, micromolar concentrations of the muscarinic agonist acetylcholine shifted the voltage dependence of I_f to more negative voltages and decreased the beating rate of differentiated stem cells. Therefore, the authors conclude that the cellular machinery required for autonomic regulation is present and functional at early developmental stages in mESCs. In addition, they propose that modulation of I_f is a central component of autonomic regulation in mESCs. It is noteworthy to point out that HCN isoform expression and spontaneous beating rate accompanying maturation reported by Qu et al. (2008) are in contrast to previous studies. These variations have been extensively referenced and discussed by Qu et al. (2008) as being either species specific or due to cell line differences obtained from the same species.

Qu et al. (2008) demonstrate a steady expression of HCN2 and a decreased expression of HCN3 during development of cells in culture and they argue that the dynamic change in expression level of HCN3 is consistent with the change in activation rate of I_f during development in culture. However, at face value, the decrease in expression of HCN3 is inconsistent with the increase in I_f current density during development. Several possibilities can explain this seemingly counterintuitive observation. Firstly, it is possible that high levels of HCN3 in early developmental stages inhibit the functional expression of pacemaker channels. This possibility can easily be addressed experimentally in heterologous systems expressing HCN2 and variable amounts of HCN3. Assuming that HCN3 exerts inhibitory effects, increasing amounts of HCN3 expression is expected to suppress generation of functional channels. Interestingly, one group has reported that HCN3 does not form functional homomers when expressed in heterologous systems (Chen et al. 2001). Secondly, it is possible that robust functional I_f current is present in early developmental stages but masked by a current flowing in the opposite direction. Although this explanation is formally possible, the authors have carefully chosen an external solution to suppress potassium and calcium channels during the measurements of I_f. Lastly, it is possible that pacemaker current in early developmental stages is activated at extreme hyperpolarization test pulses outside the range of those used by Qu et al. (2008). Indeed, one proposed mechanism of differentiation of a non-pacing region of the heart is a shift in activation range of I_f to potentials that are outside the physiological range (Yu et al. 1993). Another intriguing observation by Qu et al. (2008) is the heterogeneity of cells with respect to I_f and beating rhythmicity in early developmental stages. This heterogeneity can be explained by the presence of a mixture of pacing and quiescent cells, reminiscent of nodal and contracting myocytes in cardiac tissue. This assumption implies that the cells in later stages of development in tissue culture are less differentiated than those in earlier stages. Consistent with this assumption, several studies (extensively referenced in Maltsev et al. 1993) have reported changes in differentiation stage of cells in culture and reversion of differentiated ventricular cells back to embryonic stages. In summary, analogous to all well-conducted scientific inquiries, the studies of Qu et al. (2008) have raised several important questions which will certainly lead to other exciting studies in the future.