HCN Family: The Pacemaking Channels

Introduction

The ability of excitable cells to generate rhythmic, spontaneously firing action potentials is called pacemaking. In heart, pacemaking is accomplished by rhythmic discharge of the sinoatrial node (SA node). The firing rate is determined by a specific current slowly depolarizing a membrane to the threshold level, triggering the action potential. The ionic conductance underlying the cardiac pacemaking was identified over 20 years ago and termed I_f (f for funny)^{1, 2}. At the same time, a similar current was described in neurons and in retina, termed, respectively, I_h (h for hyperpolarization-activated) and I_q (q for queer)². The unique features of I_f/I_h/I_q current are: (1) its activation by hyperpolarization; (2) conductance of Na⁺ and K⁺ ions (K⁺/Na⁺ ~3-4); (3) blocking by low concentration (0.1-5 mM) of extracellular Cs⁺; (4) enhancement by cyclic AMP (cAMP). Surprisingly, the cAMP dependence was not mediated by protein phosphorylation, but involved direct binding of the cyclic nucleotide to the channel².

In the SA node, hyperpolarization at the termination of the cardiac action potential leads to activation of I_f current, which gradually depolarizes the membrane. Activation of β-adrenergic receptors by sympathetic stimulation leads to activation of G_s protein and adenylate cyclase. The synthesized cAMP binds to the channel, shifting its activation curve to more positive voltages, which results in the increase of the diastolic depolarization rate, leading to heart rate acceleration^{1, 2}. The switch from slow-wave sleep to waking and attentiveness of the animal is correlated with cAMP-dependent changes in I_h activity in thalamic neurons. Binding of norepinephrine or serotonin to the appropriate receptors increases the intracellular cAMP concentration. Higher cAMP concentration leads to activation of I_h and facilitates the shift of these neurons from a state of rhythmic oscillation and low excitability during slow-wave sleep to a state of increased excitability and responsiveness during periods of waking, attentiveness and cognition³.

HCN family are pacemaking channels

The molecular counterpart of I_h currents was unknown until 1997. The first channel was cloned via a yeast two-hybrid system by screening for proteins interacting with N-Src⁴. Independently, channels underlying I_h currents were cloned by searching the EST databases for similarities with cyclic nucleotide-binding domain (CNBD) of CNG channels⁵.

In 1998, the common nomenclature was accepted: the new clones were designated HCN for Hyperpolarization-activated Cyclic Nucleotide-gated channels⁶. Structurally, HCN channels belong to the Shaker (6TM) superfamily, with 6 transmembrane domains and pore region between S5 and S6. The CNBD is located in the C-terminus region. CNBD is also present in two other groups of 6TM channels: cyclic nucleotide-gated (CNG) cation channels and EAG (KCNH) family of K⁺ channels. While CNG are non-selective cation channels, permeable to Ca²⁺, and EAG family are highly selective K⁺ channels, HCN channels are permeable to Na⁺ and K⁺, with a slight selectivity to the latter. In agreement with this, the P region of HCN channels contains GYG signature sequence, characteristic to most K⁺ channels. However, the next residue, which is D or E in most K⁺ channels, varies in different HCN channels^2,7.

To date, four different HCN channels (HCN1-4), have been identified in mammals. When functionally expressed, they give rise to I_h-like currents, differing in their activation kinetics and cAMP modulation. Based on their similarity with other 6TM channels, one can suggest a possible tetrameric architecture for the functional unit. When heterologously expressed, HCN1 and HCN2 form functional heteromers with intermediate properties⁸.

The four HCN channels differ in their tissue localization. HCN1, the least sensitive to cAMP, is expressed in cerebral cortex, hippocampus, cerebellum, and facial motor nucleus^11,12. In agreement with this, the cAMP dependent shift of I_h is minimal in hippocampal neurons compared to the SA node. HCN1 is the only HCN channel in retinal photoreceptors and the major isoform in dorsal root ganglia neurons¹¹. HCN2 and HCN3 are widespread in the brain. HCN2 is most highly abundant in mamillary bodies, pontine nucleus, ventral cochlear nucleus, and nucleus of the trapezoid body, while HCN3 expression is most pronounced in supraoptic nucleus of hypothal amus¹. The slowest and the most sensitive to cAMP, HCN4, is the major isoform in the SA node^11,13,14. In brain, HCN4 is mostly expressed in medial habenula and in thalamus¹². I_h/I_f current is a target of a class of agents called “specific bradycardic agents”, like UL-FS 49 (zatebradine)¹⁶, its derivative DK-AH 268¹⁵ or ZD7288⁹. The latter has been shown to block heterologously expressed HCN1¹⁰. However, the detailed data of the specificity of these agents is still unavailable. Like native I_h currents, heterologously expressed HCN channels are blocked by low concentrations of Cs⁺ and are unaffected by classic K⁺ channel blockers, Ba²⁺ or tetraethylammonium (TEA).

Our knowledge of HCN channels is still fragmentary, partly due to their recent discovery and lack of specific molecular tools.

References

1. Ludwig, A. et al. (1999) Cell. Physiol. Biochem. 9, 179 (review).
2. Biel, M. et al. (1999) Rev. Physiol. Biochem. Pharmacol. 136, 165 (review).
3. Pape, H.C. (1996) Annu. Rev. Physiol. 58, 299 (review).
4. Santoro, B. et al. (1997) Proc. Natl. Acad. Sci. USA 94, 14815.
5. Ludwig, A. et al. (1998) Nature 393, 587.
6. Clapham, D.E. (1998) Neuron 21, 5 (review).
7. Santoro, B. and Tibbs, G.R. (1999) Ann. N. Y. Acad. Sci, 868, 741 (review).
8. Ulens, C. and Tytgat, J. (2001) J. Biol. Chem. 276, 6069.
9. BoSmith, R.E. et al. (1993) Br. J. Pharmacol. 110, 343.
10. Shin, K. et al. (2001) J. Gen. Physiol. 117, 91.
11. Moosmang, S. et al. (2001) Eur. J. Biochem. 268, 1646.
12. Monteggia, L.M. et al. (2000) Brain. Res. Mol. Brain. Res. 81, 129.
13. Shi, W. et al. (1999) Circ. Res. 85, e1.
14. Ishii, T.M. et al. (1999) J. Biol. Chem. 274, 12835.
15. Pape, H.C. (1994) Neuroscience 59, 363.
16. Kobinger, W. and Lillie, C. (1984) Eur. J. Pharmacol. 104, 9.