UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The unique pore and selectivity filter of HCN channels Macri, Vincenzo S 2010

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2010_fall_macri_vincenzo.pdf [ 3.42MB ]
Metadata
JSON: 24-1.0071068.json
JSON-LD: 24-1.0071068-ld.json
RDF/XML (Pretty): 24-1.0071068-rdf.xml
RDF/JSON: 24-1.0071068-rdf.json
Turtle: 24-1.0071068-turtle.txt
N-Triples: 24-1.0071068-rdf-ntriples.txt
Original Record: 24-1.0071068-source.json
Full Text
24-1.0071068-fulltext.txt
Citation
24-1.0071068.ris

Full Text

THE UNIQUE PORE AND SELECTIVITY FILTER OF HCN CHANNELS by  Vincenzo S. Macri M.Sc., Simon Fraser University, 2002  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in  The Faculty of Graduate Studies (Physiology)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) July 2010 ©Vincenzo S. Macri, 2010  Abstract Hyperpolarization-activated Cyclic Nucleotide-modulated (HCN) channels are similar in structure and function to potassium channels. In both, changes in membrane voltage produce directionally similar movement of positively charged residues in the voltage sensor to alter the pore structure at the intracellular side and gate ion flow. Both classes of channels also allow mainly potassium ions to flow, are blocked by cesium ions, and are activated by extracellular potassium. However, HCN channels open when hyperpolarized, whereas most potassium channels open when depolarized. Thus, electromechanical coupling between the voltage sensor and gate is opposite. A key determinant of this coupling is the intrinsic stability of the pore. In potassium channels, the closed, and not the open, pore is more stable, however this it not known for HCN channels. HCN channels are also significantly permeable to sodium despite containing the GYG potassium channel signature selectivity filter sequence. In potassium channels, the selectivity filter sequence is „T/S-V/I/L/T-GYG‟, which forms a row of four binding sites through which dehydrated potassium ions flow. In HCN channels, the equivalent residues are „C-I-GYG‟, but whether they form four similarly arrayed cation binding sites is not known. In this thesis, we show using the mammalian HCN2 channel, that the stabilities of the open and closed pore are similar, the voltage sensor must apply force to close the pore, and that the interactions between the pore and voltagesensor are weak. Furthermore, our data suggest that the conserved cysteine of the selectivity filter does not form a fourth binding site for permeating ions, which prevents it from contributing to either permeation or associated gating functions of the selectivity filter.  ii  Table of contents Abstract .................................................................................................................................... ii Table of contents .................................................................................................................... iii List of tables.......................................................................................................................... viii List of figures .......................................................................................................................... ix Acknowledgements ................................................................................................................ xi Dedication .............................................................................................................................. xii Co-authorship statement ..................................................................................................... xiii 1. Introduction ..........................................................................................................................1 1.1 The funny current, If ....................................................................................................... 1 1.1.1 History of If .............................................................................................................. 1 1.1.2 Biophysical properties of If ...................................................................................... 3 1.1.3 The role of If in pacemaking in the heart ................................................................. 7 1.1.4 Autonomic modulation of If and heart rate .............................................................. 9 1.2 HCN channels ............................................................................................................... 11 1.2.1 Cloning and expression .......................................................................................... 11 1.2.2 Predicted transmembrane segments and cytoplasmic termini ............................... 12 1.2.3 Proposed architecture of the HCN channel pore.................................................... 14 1.2.4 Biophysical properties of HCN channels ............................................................... 17 1.2.5 Isoform specific channel opening, modulation by cAMP and the CNBD ............. 19 1.2.6 Mutations in HCN4 are linked to human bradyarrhythmias .................................. 24 1.3 Voltage-dependent gating and pore opening in HCN channels .................................... 26 1.3.1 Isoform differences in activation rates are attributed to S1 and S2 ....................... 26  iii  1.3.2 The S3-S4 linker modifies voltage-dependent gating ............................................ 27 1.3.3 The S4 domain in voltage dependent gating .......................................................... 28 1.3.3.1 S4 primary structure ........................................................................................ 28 1.3.3.2 Functional role of the S4 residues ................................................................... 29 1.3.3.3 S4 movement .................................................................................................. 30 1.3.4 Coupling voltage-sensing to pore opening ............................................................ 33 1.3.5 The activation gate in S6........................................................................................ 34 1.3.6 The proposed glycine hinge in S6 .......................................................................... 35 1.3.7 Energetics of pore opening in HCN channels ........................................................ 38 1.4 The structure and function of the HCN selectivity filter .............................................. 39 1.4.1 Proposed structure of the selectivity filter ............................................................. 39 1.4.2 The GYG residues of the selectivity filter ............................................................. 43 1.4.3 The C-terminal residues located immediately outside the GYG ........................... 44 1.4.4 Extracellular K+ and Na+ may affect conductance at the selectivity filter............. 45 1.4.5 Conductance and gating at the fourth ion binding site of the selectivity filter ...... 47 1.5 Statement of thesis objectives ....................................................................................... 49 1.6 References ..................................................................................................................... 53 2. Alanine scanning of the S6 segment reveals a unique and cyclic AMP-sensitive association between the pore and voltage-dependent opening in HCN channels ............82 2.1 Introduction ................................................................................................................... 82 2.2 Experimental procedures .............................................................................................. 84 2.2.1 Mutagenesis ........................................................................................................... 84 2.2.2 Tissue culture and expression of HCN2 constructs ............................................... 84  iv  2.2.3 Whole-cell patch clamp electrophysiology ............................................................ 85 2.2.4 Data analysis .......................................................................................................... 85 2.2.5 Western blot analysis ............................................................................................. 86 2.3 Results ........................................................................................................................... 87 2.3.1 Alanine/valine scanning of the distal S6 reveals small changes in perturbation energy.............................................................................................................................. 87 2.3.2 Cyclic AMP shifts the balance of perturbation energies of the S6 mutations toward negative values ................................................................................................................ 89 2.3.3 The effects of S6 mutations on Z are consistent with an altered closed to open transition ......................................................................................................................... 97 2.4 Discussion ................................................................................................................... 104 2.5 Acknowledgements ..................................................................................................... 111 2.6 References ................................................................................................................... 112 3. The unique form and function of the HCN channel selectivity filter ..........................118 3.1 Introduction ................................................................................................................. 118 3.2 Methods ...................................................................................................................... 120 3.2.1 Site-directed mutagenesis .................................................................................... 120 3.2.2 Tissue culture and expression of HCN2 constructs ............................................. 121 3.2.3 Whole-cell patch clamp electrophysiology .......................................................... 121 3.2.4 Data analysis ........................................................................................................ 122 3.3 Results ......................................................................................................................... 124 3.3.1The cysteine 400 sulfhydryl side chain does not impact selectivity ..................... 124 3.3.2 The cysteine 400 sulfhydryl side chain does not impact cation flow .................. 126  v  3.3.3 Enhanced block by extracellular cesium supports a contribution to the permeation path by the threonine side chain.................................................................................... 132 3.3.4 Effects of the T400 mutation on HCN2 function are dependent on potassium ions residing within the internal cavity................................................................................. 134 3.3.5 The T400 mutation facilitates channel opening ................................................... 137 3.4 Discussion ................................................................................................................... 140 3.5 Acknowledgements ..................................................................................................... 144 3.6 References ................................................................................................................... 145 4. Concluding chapter ..........................................................................................................153 4.1 Overview ......................................................................................................................153 4.2 A comparison of the energetics of pore opening in HCN and Kv channels .................154 4.3 The majority of S6 mutations alter channel opening ...................................................157 4.4 The input of energy is conserved in HCN and Kv channels ........................................158 4.5 Physiological implications for a naturally opened HCN channel pore ........................158 4.6 The sulfhydryl side chain group of cysteine 400 of the CIGYG selectivity filter does not contribute to K+ and Na+ selectivity and conductance.................................................160 4.7 A role for the selectivity filter in gating in HCN channels ..........................................162 4.8 K+ and Na+ selectivity in HCN channels .....................................................................163 4.9 The selectivity filter motif, CIGYG, sets the reversal potential and conductance response to physiological levels of extracellular K+ ..........................................................165 4.10 Future research directions ..........................................................................................166 4.11 References ..................................................................................................................169  vi  Appendix A A novel KCNA1 mutation associated with global delay and persistent cerebellar dysfunction .........................................................................................................180 Appendix B Biohazard approval certificate.....................................................................186  vii  List of tables Table 2.1 A, B The effects of S6 pore mutations on voltage-dependent gating at basal (A) and saturating (2 mM; B) levels of cAMP ...............................................................................92 Table 2.2 Allosteric model parameters at basal and saturating (2 mM) levels of cAMP .....103  viii  List of figures Figure 1.1 Effects of autonomic agonists on spontaneous activity and hyperpolarizationactivated current (If) in cardiac sinoatrial node myocytes from the rabbit ................................8 Figure 1.2 The HCN channel subunit .....................................................................................13 Figure 1.3 Homology model of the HCN2 channel pore based upon KcsA suggests a similar architecture ...............................................................................................................................15 Figure 1.4 X-ray crystal structure of the C-linker and CNBD of the HCN2 channel .............21 Figure 1.5 Comparison of the closed and opened channel pore in K+ channels .....................36 Figure 1.6 The residues which make up the selectivity filter of HCN2 may form four ion binding sites similar to KcsA ...................................................................................................42 Figure 2.1 HCN2 channels are most stable in the open state ..................................................90 Figure 2.2 Saturating levels of cAMP (2 mM) further stabilize the open state ......................95 Figure 2.3 Glycine 424 is critical for the expression of cell surface HCN2 channels ............98 Figure 2.4 Experimental and model Z values are comparable and change minimally over the range of observed mid-activation voltages ............................................................................102 Figure 2.5 Distribution of amino acids in distal HCN2 S6 segment that are critical for energetic balance of open and closed configurations ............................................................108 Figure 3.1 Mutation of the innermost binding site from cysteine to threonine, but not serine or alanine, shifts the reversal potential to more positive potentials in physiological solutions ................................................................................................................................................127 Figure 3.2 The T400 mutation reduces the maximum potassium conductance ...................129 Figure 3.3 Wild type and T400 channel conductance increases by the same relative amount in response to raising extracellular potassium .......................................................................130 Figure 3.4 Potassium conductance is selectively reduced in individual cells expressing the T400 channel ..........................................................................................................................131 Figure 3.5 Extracellular Cs+ blocks the T400 channel with greater sensitivity and at a site closer to the extracellular side of the selectivity filter ...........................................................133  ix  Figure 3.6 Reduced potassium conductance of the T400 channel reverts to wild type phenotype by lowering and raising intracellular potassium and sodium, respectively ................................................................................................................................................136 Figure 3.7 Block of the T400 channel by Cs+ reverts to wild type phenotype by lowering and raising intracellular potassium and sodium, respectively ......................................................138 Figure 3.8 The T400 mutation facilitates HCN2 channel opening only when intracellular potassium and sodium are high and low, respectively ...........................................................139 Figure 4.1 The input of energy is conserved in HCN and Shaker channels .........................159  x  Acknowledgements Thank you to my senior supervisor, Dr. Eric Accili, for his mentorship and support, and the freedom to develop and pursue my own research path. I would also like to thank the members of my supervisory committee, Dr. Steven Kehl, Dr. David Fedida, Dr. Mark Paetzel and Dr. Ed Moore, for their insightful and valuable feedback on my research. Thank you to all the members of the lab for the scientific discussions and friendship.  Thank you to my parents, Stefano and Caterina, and family for their continuing support, love, and encouragement during my graduate studies. Thank you to my fiancée, Laura, for her unconditional love, support, encouragement, and friendship.  xi  Dedication  To My Parents  xii  Co-authorship statement Chapter 2: Alanine scanning of the S6 segment reveals a unique and cyclic AMP-sensitive association between the pore and voltage-dependent opening in HCN channels Vincenzo Macri designed, collected, and analysed all electrophysiology data. Vincenzo Macri designed the site-directed mutagenesis experiments and Hamed Nazzari performed the site-directed mutagenesis and western blot experiments and Evan McDonald performed the site-directed mutagenesis. Vincenzo Macri performed the modeling and analysed the modeled data. Vincenzo Macri and Eric Accili prepared and edited the manuscript. A version of this chapter has been published. Macri, V, Nazzari, H, McDonald, E, Accili, EA. (2009) Alanine scanning of the S6 segment reveals a unique and cyclic AMP-sensitive association between the pore and voltage-dependent opening in HCN channels. Journal of Biological Chemistry, 284: 15659-67. Chapter 3: The unique form and function of the HCN channel selectivity filter Vincenzo Macri designed, collected, and analysed the majority of the electrophysiology data. Damiano Angoli collected and analysed some of the electrophysiology data. Vincenzo Macri designed and performed all the site directed mutagenesis experiments. Vincenzo Macri and Eric Accili prepared and edited the manuscript. A version of this chapter has been submitted for publication. Macri, V, Angoli, D, Accili, EA. The unique form and function of the HCN channel selectivity filter. Appendix: A novel KCNA1 mutation associated with global delay and persistent cerebellar dysfunction Michelle Demos designed, performed and prepared the case study data, clinical and genetic analysis, and the manuscript. Vincenzo Macri designed, performed and analysed all of the electrophysiology data. Vincenzo Macri and Eric Accili prepared and edited the electrophysiology portion of the manuscript and edited the manuscript. Kevin Farrell designed, performed and prepared the collection of the clinical data and edited the manuscript. Tanya Nelson designed, performed and prepared the clinical report and edited the manuscript. Kristine Chapman collected the neurophysiology data, designed and prepared the clinical report and edited the manuscript. Linlea Armstrong designed, performed and prepared the case study data, clinical information and genetic and functional studies, and edited the manuscript. This work has been published. Demos, MK, Macri, V, Farrell, K, Nelson, TN, Chapman, K, Accili, E, Armstrong, L. (2009) A novel KCNA1 mutation associated with global delay and persistent cerebellar dysfunction. Movement Disorders, 24: 788-82.  xiii  1. Introduction 1.1 The funny current, If 1.1.1 History of If Before the discovery of If, IK2, an outward pure K+ carrying current, was considered to be the pacemaker current in the heart (Hauswirth et al., 1968; Noble and Tsien, 1968). However, IK2 was incorrectly identified and was later found to be the same current as If (DiFrancesco, 1981a). IK2 was initially described in spontaneously active Purkinje fibres. Researchers hypothesized that IK2 contributed to pacemaking since this current was activated during the action potential and was subsequently turned off during the interval between action potentials known as the diastolic depolarization phase (Hauswirth et al., 1968; Noble and Tsien, 1968). Therefore, the turning off of IK2 resulted in depolarization of the membrane which led to threshold firing of the next action potential (Hauswirth et al., 1968; Noble and Tsien, 1968). However, this was considered intuitively difficult to understand since an inward current was needed to depolarize the membrane potential.  In 1976, the first report of an inward current that was activated upon membrane hyperpolarization was described in sino-atrial node (SAN) cells (Noma and Irisawa, 1976). This inward current, like IK2, was also found to be important during the diastolic depolarization phase but in SAN cells (Brown and DiFrancesco, 1980; Brown et al., 1979; DiFrancesco and Ojeda, 1980). Both currents were modulated by adrenaline which resulted in an increase in the spontaneous firing rate of action potentials both in Purkinje fibres and SAN cells (Brown et al., 1979; DiFrancesco and Ojeda, 1980; Hauswirth et al., 1968). However, unlike IK2, this inward current was named If for its funny properties. If was  1  characterized as a slowly developing inward current activated by hyperpolarization. The inward current depolarized the membrane to initiate threshold firing of the next SAN action potential (Brown and DiFrancesco, 1980; Brown et al., 1979). Furthermore, the reversal potential of If was determined to be ~ -20 mV in physiological solutions of K+ and Na+ and was sensitive to changes in both extracellular K+ and Na+ (DiFrancesco and Ojeda, 1980; Yanagihara and Irisawa, 1980). These observations suggested that, unlike IK2, If was not a pure K+ current but was a mixed K+ and Na+ current (DiFrancesco and Ojeda, 1980; Yanagihara and Irisawa, 1980).  Experiments using extracellular Ba2+ helped to reinterpret IK2 and allowed for the correct identification of If as the pacemaker current (DiFrancesco, 1981a). The inwardly rectifying K+ current, IK1, was found to be significantly larger in Purkinje fibres than in SAN cells. Because of this significant size difference, IK1 contaminated the reversal potential measurements of IK2 in Purkinje fibers but allowed for the identification of If in SAN cells. The application of extracellular Ba2+ to Purkinje fibres blocked IK1, revealing the true reversal potential of IK2 which was the same as If in SAN cells (DiFrancesco, 1981a). These experiments revealed that IK2 in Purkinje fibres was the same If current that was described in SAN cells (DiFrancesco, 1981a, b; DiFrancesco and Ojeda, 1980).  Shortly after the description of If in cardiac tissue, an identical current was discovered in neurons, but was named Ih since like If, was activated upon hyperpolarization (Bader et al., 1979; Halliwell and Adams, 1982; Mayer and Westbrook, 1983).  2  1.1.2 Biophysical properties of If In SAN cells, If activates at potentials more negative than -30 mV and becomes fullyactivated at ~-100 mV (Brown and DiFrancesco, 1980; DiFrancesco, 1991; DiFrancesco et al., 1986). The rate of channel opening also increases as the membrane potential becomes more hyperpolarized and at -100 mV reaches steady state at ~ 250 ms. The midpoint of activation (V1/2) was measured to be ~ -52 mV (DiFrancesco et al., 1986). If deactivates at depolarized potentials and is completely closed at potentials more positive than -30 mV. The rates of current activation and deactivation are similar in time course and the onsets of these currents are sigmoid in shape (DiFrancesco, 1984; DiFrancesco et al., 1986). Furthermore, a characteristic delay occurs before the onset of current activation which shortens in length as the membrane potential becomes more hyperpolarized.  The sigmoid shape of current  activation and deactivation and the observed delay before channel opening suggests that If does not obey classic Hodgkin-Huxley current kinetics (DiFrancesco, 1984; Hodgkin and Huxley, 1952). Several years after the identification and from the subsequent cloning of the molecular determinants of If, a cyclic allosteric model with multiple closed and opened states was shown to accurately describe the kinetics and voltage-dependence of If (Altomare et al., 2001; DiFrancesco, 1999). The molecular determinants of If, Hyperpolarization-activated Cyclic Nucleotide-gated (HCN) channels, and the cyclic allosteric model will be discussed in further detail in section 1.2.  In physiological solutions of K+ and Na+, the reversal potential of If was measured to be ~ -20 mV (DiFrancesco, 1984; DiFrancesco et al., 1986; Hestrin, 1987; Maccaferri et al., 1993; McCormick and Pape, 1990; Solomon and Nerbonne, 1993). Based upon this value, the  3  calculated Na+ and K+ permeability ratio (PNa/PK) was ~0.3 using the Goldman Hodgkin Katz equation. This value was much larger compared to K+-selective channels (PNa/PK ~0.01) which suggested that If had a high level of Na+ permeability (Edman and Grampp, 1989; Frace et al., 1992; Hille, ; Ho et al., 1993; Magee, 1998; Wollmuth and Hille, 1992). The reversal potential was found to be sensitive to changes in extracellular Na+ and K+, suggesting that both Na+ and K+ contribute to If (DiFrancesco, 1981b; DiFrancesco et al., 1986; Ho et al., 1993, 1994). The permeability of other monovalent cations, such as Li+ and TI+, were also tested. The permeability ratios for these ions versus K+ were PLi/PK ~ 0.06 and PTI/PK ~ 1.1, respectively (DiFrancesco, 1982; Edman and Grampp, 1989; Wollmuth and Hille, 1992). In mixed solutions of extracellular TI+ and K+, the current amplitudes were observed to be smaller than with TI+ or K+ alone, which was indicative of an anamolous mole fraction effect (Wollmuth, 1995). The anamolous mole fraction effect and the extracellular K+ and Na+ dependent changes in reversal potential suggested that the If channel functions as a single file multi-ion pore (Frace et al., 1992; Wollmuth, 1995; Wollmuth and Hille, 1992). A minimum If channel pore size of < 4 Å was also estimated using the organic cations, ammonium (NH4, 3.7 Å) which was permeable (PNH4/PK ~0.17) and methylammonium (MA, 4.0 Å) which was not permeable (PMA/PK ~ 0.06) (Wollmuth and Hille, 1992).  Extracellular Cs+ and Rb+ block inward but not outward If currents. However, extracellular Cs+ blocks inward If more efficiently and with a steeper voltage-dependence compared to extracellular Rb+ (DiFrancesco, 1982). In SAN cells, the IC50 (0 mV) values were ~ 1.8 mM for Cs+ and ~4.1 mM for Rb+ and the δ values were ~0.7 for Cs+ and 0.05 for Rb+, which were calculated using the Woodhull model (DiFrancesco, 1982). According to the Woodhull  4  model, the difference in δ values suggested that both extracellular Cs+ and Rb+ block at sites located ~70% and ~5% of the electric field, respectively (DiFrancesco, 1982; Woodhull, 1973). The difference in δ values suggested that the Cs+ blocking site was located deep in the channel pore while the Rb+ blocking site was located at a more superficial site near the extracellular entrance of the channel pore.  Extracellular and intracellular K+ were both found to be strong modulators of the If whole cell slope conductance (Gf). Raising extracellular K+, but not raising extracellular Na+, was shown to increase Gf in both cardiac tissue and neurons (DiFrancesco, 1981b, 1982; DiFrancesco et al., 1986; Edman and Grampp, 1989; Frace et al., 1992; Solomon and Nerbonne, 1993). The increase in Gf was most dramatic in the physiological range of extracellular K+ concentrations, 2-10 mM and saturated at ~ 20 mM (Edman and Grampp, 1989; Frace et al., 1992). Intracellular K+ was also shown to be an important modulator of Gf. Replacing intracellular K+ (140 mM) with Cs+ (140 mM) dramatically increased the ability of extracellular Na+ to enhance Gf (Ho et al., 1993). Taken together, these findings suggest that both extracellular and intracellular K+ modulate the flow of K+ and Na+ through the If channel pore.  Extracellular K+ was also shown to be necessary for Na+ to permeate the If channel pore. Lowering extracellular K+ concentration decreased the permeability of Na+ relative to K+ which suggested that the extracellular K+ enhanced Na+ permeation (DiFrancesco, 1981b; Frace et al., 1992). This observation was further supported by experiments showing that replacement of extracellular K+ with the non-permeant N-methyl-D-glucamine in the  5  presence of only extracellular Na+, resulted in a complete loss of inward current (Frace et al., 1992; Wollmuth and Hille, 1992).  These experiments showed that a small amount of  extracelullar potassium was needed to maintain an inward current.  However, outward  currents could be measured with the K+-free, Na+ containing extracellular solutions. The outward currents were carried by both intracellular K+ and Na+ which suggested that the If channel was able to open at hyperpolarized potentials in K+-free, Na+ containing extracellular solutions and that Na+ permeated very slowly in the absence of extracellular K+.  The measurement of single If channels remained elusive for several years after its initial discovery in cardiac tissue and neurons (Bader et al., 1979; Brown and DiFrancesco, 1980; Brown et al., 1979; Halliwell and Adams, 1982; Yanagihara and Irisawa, 1980).  The  inability to detect single If channels suggested that the movement of K+ and Na+ across the membrane may have occurred via a transporter/exchanger mechanism which is much slower (300 ions/sec) than ion flux through a channel (1x108 ions/sec) (DiFrancesco, 1986). Then in 1986, small single channel currents were measured in cell-attached recordings from SAN cells (DiFrancesco, 1986). At a fully-activated potential of -102 mV, the unitary current amplitude was -0.085 pA. Plotting these unitary currents against test voltage gave a linear relationship, with a single channel conductance of ~1 pS. The measurement of If single channels, established that the flux of K+ and Na+ across the cell membrane was indeed through an ion channel.  6  1.1.3 The role of If in pacemaking in the heart In specialized cells of the SAN, If has been suggested to provide an inward current during the diastolic depolarization phase of the SAN action potential which helps to drive spontaneous activity in the heart (Brown and DiFrancesco, 1980; Brown et al., 1979; DiFrancesco, 1991, 1993; DiFrancesco and Ojeda, 1980). At the end of an SAN action potential, when the membrane potential is ~ -55 mV, If channels open and the inward current helps to depolarize the membrane potential to reach threshold to start a new action potential (Fig. 1.1). Depolarization activates the L-type calcium current which produces the upstroke of the action potential (DiFrancesco, 1993). The role of If in contributing to pacemaking was supported by the results of experiments using the specific If blocker ivabradine which reduced heart rate with little or no cardiac side effects (Bois et al., 1996).  Myocytes isolated from the atrial or ventricular tissue lack spontaneous pacemaking activity and have very little or no expression of If (Robinson et al., 1997; Shi et al., 1999; Wu et al., 1991).  However, If has been observed in adult ventricular myocytes after cardiac  hypertrophy and in neonatal ventricular myocytes which both exhibit spontaneous pacemaking activity which suggests that the expression of If is needed to confer spontaneous activity in otherwise quiescent cells (Cerbai et al., 1996; Cerbai et al., 1999; FernandezVelasco et al., 2006).  However, all spontaneous activity cannot be attributed to If alone. Other membrane-bound ion translocation proteins such as T-type calcium channels, RyR Ca2+ release channels,  7  If Figure 1.1 Effects of autonomic agonists on spontaneous activity and hyperpolarizationactivated current (If) in cardiac sinoatrial node myocytes from the rabbit Spontaneous action potentials recorded in control conditions and in the presence of either isoprenaline (Iso) or acetylcholine (ACh) at the concentrations indicated. The rate of acceleration (by Iso) and slowing (by ACh) are due to changes in the degree of activation of If which is reflected in the rate of diastolic depolarization (Accili et al., 2002).  8  Na+/Ca2+ exchangers and a sustained Na+ background current from an unidentified source, also provide inward currents during the diastolic depolarization phase of the SAN action potential (Bers, 2006; Lipsius and Bers, 2003; Vinogradova et al., 2002). Therefore, it is not completely clear to what extent or proportions these other inward currents, in addition to If, contribute to spontaneous activity in the SAN (Bogdanov et al., 2006; Bucchi et al., 2003; Lipsius and Bers, 2003).  1.1.4 Autonomic modulation of If and heart rate The SAN is innervated by both the sympathetic and parasympathetic branches of the autonomic nervous system (DiFrancesco, 1993). The sympathetic nervous system during stress or exercise increases heart rate by releasing adrenaline. The increase in heart rate can be attributed, in part, to adrenaline‟s effect on If (Brown et al., 1979; Zaza et al., 1996). Adrenaline binds to β-adrenergic receptors and raises the intracellular cyclic adenosine mono-phosphate (cAMP) levels via activation of adenylyl cyclase. Using inside-out patches from SAN cells, it was shown that the direct binding of cAMP to the cytoplasmic side of the If channel resulted in ~ +10 mV shift in the V1/2 at a saturating concentration of 2 mM (DiFrancesco and Tortora, 1991). Single channel experiments also showed that cAMP decreased the first latency of If channel opening but had no effect on single channel conductance (DiFrancesco, 1986; DiFrancesco and Mangoni, 1994). Therefore, the positive shift in the V1/2 and the shorter first latency of opening demonstrated that cAMP increased the amount of If available during the diastolic depolarization (ranging from -40 to -65 mV) of action potential in SAN cells. The increase in current availability of inward current at diastolic potentials helps to reach threshold more quickly and shortens the interval between  9  SAN action potentials (Fig. 1.1). The increase in heart rate can be attributed, in part, to adrenaline‟s effect on If (Brown et al., 1979; Zaza et al., 1996). However, both If and the Ltype Ca2+ current are inward currents that depolarize the membrane during diastolic depolarization.  While both currents display a similar dose response to β-adrenergic  stimulation, based upon their I-V relationships, If and the L-type Ca2+ current contribute to the early and late phase of the diastolic depolarization phase, respectively (Zaza et al., 1996).  The parasympathetic nervous system slows heart rate by releasing acetylcholine which acts on muscarinic receptors and inhibits the production of cAMP (DiFrancesco et al., 1989; DiFrancesco and Tromba, 1988b). Acetylcholine shifts the V1/2 of If to more hyperpolarized potentials by ~ -10 mV and has no effect on the open channel If-V relationship in SAN cells (DiFrancesco et al., 1989; DiFrancesco and Tromba, 1988a). This negative shift produced by acetylcholine has the opposite effect of adrenaline. If is activated at more hyperpolarized potentials, thus producing less inward current during the diastolic depolarization phase. This results in delayed firing and increasing the interval between SAN action potentials (Fig. 1.1). In addition to If, activation of the acetylcholine sensitive K+ current (IK,Ach) during the diastolic depolarization phase has also been suggested to be important in contributing to slowing heart rate. However, activation of IK,Ach required acetylcholine concentrations of ~ 20-fold greater than for the inhibition of If (DiFrancesco et al., 1989). A reduction in SAN firing rate was observed at low doses of acetylcholine (0.01-0.03 M) and at these concentrations If was significantly reduced. Therefore, these findings suggest that at low levels such as might occur during mild vagal stimulation, acetylcholine selectively acts on If to reduce heart rate.  10  1.2 HCN channels 1.2.1 Cloning and expression About twenty years after the identification of If in SAN cells, the genes that encode for If were cloned and were called HCN channels (Ludwig et al., 1998; Santoro et al., 1997; Santoro et al., 1998; Seifert et al., 1999). HCN channels, based upon primary amino acid structure, were suggested to be most similar to voltage-gated K+ (Kv) (e.g. HERG, human ether-a-go-go, and KAT1, plant channel from Arabidopsis thaliana) and Cyclic Nucleotide Gated (CNG) channels (Robinson and Siegelbaum, 2003). HCN channels were cloned from both heart and brain tissue from various mammals such as mouse, rabbit, and human (Ishii et al., 1999; Ludwig et al., 1998; Ludwig et al., 1999; Mistrik et al., 2005; Moroni et al., 2001; Santoro et al., 1998; Seifert et al., 1999; Stieber et al., 2005; Vaccari et al., 1999). These cloning efforts identified four mammalian HCN channels: HCN1, HCN2, HCN3 and HCN4. Each of the four HCN channels produced currents that had biophysical properties similar to If/Ih, described in cardiac tissue and neurons. A non-mammalian HCN channel was also cloned from sea urchin sperm, spHCN (Gauss et al., 1998). The cloning of HCN channels has advanced our understanding of their structure and function, tissue expression, and role in cardiac and neurophysiology (Robinson and Siegelbaum, 2003; Wahl-Schott and Biel, 2009).  The expression patterns of the four mammalian HCN channels in the heart and brain have been studied at both the protein and mRNA level in various mammals. HCN1 was found to be expressed abundantly in the thalamus, dorsal root ganglion cells, and in the SAN cells (Ludwig et al., 1998; Ludwig et al., 1999; Santoro et al., 2000; Shi et al., 1999). HCN2 was determined to be present in different regions of the brain, such as the cortex and thalamus,  11  and within the ventricles and atria. HCN2 was also found at lower levels in the SAN cells (Ludwig et al., 1998; Santoro et al., 2000; Shi et al., 1999). Low levels of HCN3 have been shown in the olfactory bulb and hypothalamus, and in the heart ventricle (Mistrik et al., 2005; Stieber et al., 2005). HCN4 was found in the thalamus and in the ventricle but was most abundant in SAN cells (Ishii et al., 1999; Ludwig et al., 1998; Santoro et al., 1997; Seifert et al., 1999; Shi et al., 1999).  Furthermore, HCN1, HCN2 and HCN4 are expressed in  atrioventricular nodal cells and Purkinje fibers of the heart, while HCN3 is expressed in the embryonic heart (Han et al., 2002; Ishii et al., 1999; Ludwig et al., 2003; Moosmang et al., 2001; Moroni et al., 2001; Shi et al., 1999).  1.2.2 Predicted transmembrane segments and cytoplasmic termini HCN channels are composed of four subunits (Biel et al., 2009; Robinson and Siegelbaum, 2003). The four subunits can assemble to make tetrameric channels which are expressed on the plasma membrane (Proenza et al., 2002b; Whitaker et al., 2007; Xue et al., 2002). Each subunit contains six-transmembrane (S1-S6) spanning segments with a cytoplasmic amino and carboxy terminus (Fig. 1.2). HCN channels, like CNG channels, also have a cyclic nucleotide binding domain (CNBD) located in the C-terminus. When considering only the six transmembrane spanning segments and the CNBD, the four mammalian HCN channels display >80% amino acid identity (Jackson et al., 2007; Ludwig et al., 1998; Viscomi et al., 2001).  When considering only the cytoplasmic amino and carboxy terminus, the four  mammalian HCN channels show a significantly lower percentage of amino acid identity and are also substantially different in length.  The first four transmembrane spanning segments  (S1-S4) form the voltage sensing domain and the fifth and sixth transmembrane spanning  12  Na+, K+ p  1 2 3 + 5 6  p  6 5 + 3 2 1  cAMP  Figure 1.2 The HCN channel subunit Two of four HCN subunits are shown placed in the plasma membrane denoted by the two horizontal black lines. Each subunit contains six transmembrane spanning helices, numbered 1 to 6 with the fourth helix being represented with a positive sign to denote it as the putative voltage sensor. In red are the p (pore-helices) and S6 helices which form part of the pore and are proposed to come into contact with permeating ions. Each subunit also contains an intracellular N- and C-terminus, where the Cterminus contains the C-linker and Cyclic Nucleotide Binding Domain (CNBD) shown in blue. The CNBD is shown binding cAMP (Zagotta et al., 2003).  13  segments (S5-S6), along with a pore-helix and selectivity filter, form the pore domain. The structure and function of the transmembrane spanning segments, CNBD and selectivity filter will be discussed in further detail in the following sections.  1.2.3 Proposed architecture of the HCN channel pore In the x-ray crystal structures of K+ channels, such as KcsA from Streptomyces lividans, KvAP from the archeabacterium Aeropyrum Pernix, Kv1.2 from rat brain and KirBac1.1 from Burkholderia pseudomallei, each channel pore is composed of four subunits (Doyle et al., 1998; Jiang et al., 2003a; Kuo et al., 2003; Long et al., 2005). The four subunits of each K+ channel pore come together forming an inverted teepee structure with a central ion conduction pathway (Fig. 1.2). The pore domain of each subunit consists of an outer (M1 or S5) and an inner (M2 or S6) helix, a pore helix and the GYG K+ channel signature sequence which forms the selectivity filter.  HCN channels are also composed of four subunits which come together to form a functional channel (Proenza et al., 2002b; Whitaker et al., 2007; Xue et al., 2002). Although there is no x-ray crystal structure of the HCN channel pore, a HCN2 pore homology model based on the x-ray crystal structure of the KcsA K+ channel pore, suggests that the general pore architecture of HCN and K+ channels may be similar (Fig. 1.3) (Giorgetti et al., 2005). Each subunit also consists of an outer (S5) and an inner helix (S6), pore helix and the proposed selectivity filter also contains the GYG K+ channel signature sequence. While the amino acid identity of the residues which form the pore of HCN2 and KcsA is low, ~ 18%, amino acid  14  KcsA M1  Top View  M2  Side View  HCN2 S5 S6  Figure 1.3 Homology model of the HCN2 channel pore based upon KcsA suggests a similar architecture Top left, x-ray crystal structure of the KcsA K+ channel pore showing four subunits together forming a tetrameric channel with a central ion pathway. Top right, two of four subunits are shown to highlight the inverted teepee architecture of KcsA pore with M1 (outer helix) and M2 (inner helix). The GYG residues of the selectivity filter are also shown which highlight the narrowest region of the pore. Bottom left, homology model of HCN2 based upon the KcsA K+ channel pore which also shows four subunits together forming a tetrameric channel with a central ion pathway. Bottom right, two of four subunits are highlighted to show the proposed inverted teepee architecture of HCN2 pore with the S5 (outer helix) and S6 (inner helix). The GYG residues of the HCN2 selectivity filter are also shown which highlight the narrowest region of the pore as in KcsA.  15  identity increases to ~ 30% when including only the residues starting at the pore helix up to the selectivity filter.  Experimental evidence also suggests that the orientation of the HCN channel pore in the plasma membrane may be similar to the K+ channel pore. Amino acid residues predicted to be located extracellularlly or intracellularlly, were confirmed in HCN channels using cysteine accessibility experiments (Au et al., 2008; Roncaglia et al., 2002; Xue and Li, 2002). Specifically, an endogenous conserved cysteine residue was predicted to be located in the extracellular loop between the S5 and pore helix of HCN channels. This cysteine residue in HCN1 could be modified when the cystiene modifying agent, methanethiosulfonate ethyltrimethlammonium (MTSET) was applied only to the extracellular solution which resulted in a reduction in current. Mutation of this cysteine to serine, C318S, abolished the effect of extracellular MTSET (Xue and Li, 2002). In a similar experiment using spHCN channels, two residues located just C-terminal to the GYG of the selectivity filter, K433 and F434, were also are predicted to be located extracellularlly. Mutation of these residues to cysteines also resulted in a reduction in current when Cd2+ was applied only to the extracellular solution (Au et al., 2008). Cd2+ was used as the probe since it also modifies cysteine residues. Using spHCN channels, it was also shown that the conserved cysteine residue, C428, of the selectivity filter sequence, CIGYG, was shown to abolish current when Cd2+ was applied only in the intracellular solution (Roncaglia et al., 2002). Mutation of the cysteine to serine removed the effect of intracellular Cd2+.  16  1.2.4 Biophysical properties of HCN channels The four mammalian HCN channels, HCN1-4, display classic If/Ih biophysical properties (Ishii et al., 1999; Ludwig et al., 1998; Ludwig et al., 1999; Mistrik et al., 2005; Santoro et al., 1998; Seifert et al., 1999). These are: 1) an inward current which is activated upon membrane hyperpolarization (Ishii et al., 1999; Ludwig et al., 1998; Ludwig et al., 1999; Mistrik et al., 2005; Santoro et al., 1998); 2) the activating and deactivating currents are sigmoid in shape and display a characteristic delay before the onset of activation which can be removed by pre-hyperpolarizing pulses (Altomare et al., 2001; Ishii et al., 1999; Ludwig et al., 1998; Stieber et al., 2005); 3) a PNa/PK ~ 0.3-0.4 in physiological solutions of K+ and Na+ (Ludwig et al., 1998; Moroni et al., 2000; Seifert et al., 1999); 4) a reversal potential sensitive to changes in extracellular K+ and Na+ (Gauss et al., 1998; Moroni et al., 2000); 5) whole cell slope conductance (Gf) that is significantly increased when raising extracellular K+; while raising exttracellular Na+ only has a modest effect on Gf (Ludwig et al., 1998; Macri et al., 2002; Moroni et al., 2000); 6) Na+ currents are not supported without the presence of K+ (Lyashchenko and Tibbs, 2008); 7) a low single channel conductance, ~ 1.5 pS (Dekker and Yellen, 2006); and 8) complete blockage by extracellular Cs+ in the millimolar range at fully-activated potentials (-140 mV), giving a valence of block (δ) ~ 0.7 determined from the Woodull model (Ludwig et al., 1999; Macri and Accili, 2004; Moroni et al., 2000; Stieber et al., 2005).  The single channel conductance of If measured from SAN cells and HCN channels in a heterologous expression system were similar. The single channel conductance was directly measured to be ~1.5 pS for HCN2 (Dekker and Yellen, 2006). The single channel  17  conductance was directly measured to be ~1 pS from SAN cells (DiFrancesco, 1986). Slightly larger values for single channel conductance were also determined indirectly using non stationary noise analysis for HCN2 and spHCN which were ~2.5 pS (Dekker and Yellen, 2006; Flynn et al., 2007; Johnson and Zagotta, 2005). The HCN single channel conductance value determined from the direct measurement of single channels is much smaller compared to other related Kv channels, such as Shaker and KAT1 which have a single channel conductance of ~15 pS and ~24 pS, respectively (Heginbotham and MacKinnon, 1993; Schachtman et al., 1992).  However, single channel conductance values are also quite  variable among different types of K+ channels ranging from 3 to 200 pS (Hille, 2001).  HCN channels produce an instantaneous current which is positively correlated with the size of the time-dependent inward current, If. The instantaneous current (Iinst) occurs before the onset of the time dependent inward current, If (Gauss et al., 1998; Macri et al., 2002; Proenza et al., 2002a). The current density of Iinst is significantly larger compared to endogenous instantaneous currents measured from mammalian cell lines not expressing HCN channels (Macri and Accili, 2004; Proenza et al., 2002a). Iinst, when plotted against test voltage shows a linear relationship with a reversal potential similar to If (~-20 mV). The Iinst reversal potential was sensitive to changes in extracellular K+ and Na+ suggesting that Iinst was like If, and not a pure K+ current (Macri and Accili, 2004; Proenza et al., 2002a). Iinst was not blocked by Cs+ but was reduced by the specific HCN pore blocker, ZD7288 which suggested that Iinst may flow through the HCN channel pore (Macri and Accili, 2004; Proenza et al., 2002a; Proenza and Yellen, 2006). Further support that Iinst flows through the HCN channel pore was demonstrated using a mutant spHCN channel with a cysteine engineered in the  18  middle of the S6 which showed a significant reduction in Iinst with the application of intracellular Cd2+ (Proenza and Yellen, 2006). These results suggested Iinst did not originate in another region of the channel such as the voltage sensing domain (S1-S4), as has been shown for Kv channels (Tombola et al., 2007). Further evidence that Iinst was associated with the HCN channel was supported by experiments showing that raising intracellular cAMP concentrations increased Iinst in a similar fashion observed for If (Proenza and Yellen, 2006).  1.2.5 Isoform specific channel opening, modulation by cAMP and the CNBD The rate of channel opening is different between the four mammalian HCN isoforms. The four mammalian HCN channels, open in the following order, from fastest to slowest: HCN1> HCN2> HCN3> HCN4 (Altomare et al., 2001; Ishii et al., 1999; Ludwig et al., 1998; Ludwig et al., 1999; Mistrik et al., 2005; Moroni et al., 2001; Seifert et al., 1999; Stieber et al., 2005). The V1/2 for human HCN2 and HCN4 were similar, -95 and -100 mV, respectively, and human HCN1 and HCN3 were -69 mV and -77 mV, respectively (Stieber et al., 2005). The slope factor, k, which is determined from fitting activation curves with the Boltzmann equation, did not vary significantly between the four human HCN isoforms.  Modulation of HCN channel opening by cAMP is different between the four mammalian HCN channels. The HCN2 and HCN4 isoforms showed the greatest response to cAMP while HCN1 and HCN3 responded minimally (Ishii et al., 1999; Ludwig et al., 1998; Ludwig et al., 1999; Mistrik et al., 2005; Santoro et al., 1998; Seifert et al., 1999). For HCN2 and HCN4, saturating concentrations of intracellular cAMP (2 mM) shift the V1/2 by ~ +10 mV and +20 mV, respectively, and increased the rate of channel opening approximately four-fold  19  (Ludwig et al., 1999; Stieber et al., 2005).  For HCN1 and HCN3 the V1/2 was not  significantly modulated by intracellular cAMP, however for HCN3, cAMP did produce a slight hyperpolarized shift in V1/2 (Santoro et al., 1998; Stieber et al., 2005). The difference in the ability for cAMP to modulate the four HCN isoforms was determined to be the result of the degree of inhibition incurred by the C-linker and CNBD of the C-terminus on the channel transmembrane domains (Viscomi et al., 2001; Wainger et al., 2001).  Using a chimeric mutagenesis approach, it was shown that the binding of cAMP to the CNBD removes an inhibitory effect of the C-linker specifically for HCN2 and HCN4 and not for HCN1 channels. Replacing the C-termini of HCN4 with HCN1 produced a chimeric HCN4-HCN1-C-terminal channel with activation rates similar to HCN1 (Viscomi et al., 2001; Wainger et al., 2001). Furthermore, a truncated HCN2 C-terminal mutant channel showed faster activation rates and shifted the V1/2 to more positive values compared to wild type HCN2 channels (Wainger et al., 2001). Additional experiments revealed that both the C-linker and the CNBD were required to exchange the V1/2 and activation rate phenotypes of the mammalian HCN channels, while the distal C- terminus was not important (Wang et al., 2001). Therefore, the C-linker and CNBD function to set a basal level of inhibition on channel opening which is greater for HCN2 and HCN4 and much less for HCN1.  Figure 1.4 shows the x-ray crystal structure of the C-linker and Cyclic Nucleotide Binding Domain (CNBD) for the HCN2 channel (Zagotta et al., 2003). The C-linker is composed of  20  Figure 1.4 X-ray crystal structure of the C-linker and CNBD of the HCN2 channel Top, one of four subunits is shown to highlight the structure of the C-linker and CNBD. The Clinker is composed of seven alpha helices, A‟ to F‟, and the CNBD is formed by a beta roll consisting of 8 beta sheets and the P helix, and the remaining three alpha helices, A to C. The binding pocket for cAMP is formed by the interface of the beta roll and the C helix. Bottom left, side view of the C-linker and CNBD from each subunit together located below the HCN channel. Bottom right, top view of the C-linker and CNBD from each subunit together shown as a tetramric structure with a central pore. The central pore does not form ion permeation pathway (Zagotta et al., 2003).  21  seven alpha helices, A‟ to F‟, and the CNBD is composed of a beta roll consisting of 8 beta sheets, a P helix, and three alpha helices, A to C. The binding pocket for cAMP is formed by the interface of the beta roll and C-helix of the CNBD. The C-linkers and CNBD are located just below the core transmembrane spanning domains. The C-linker and CNBD of HCN channels are closely related in primary structure to CNG, HERG, and KAT1 channels (Ludwig et al., 1998; Santoro et al., 1998; Zagotta et al., 2003). The C-linker and CNBD of one subunit come together to form a four fold symmetrical structure with a central pore in the HCN2 channel. However, the central pore of the C-linker and CNBD was shown not to form part of the ion permeation pathway since mutagenesis of residues which form the central pore did not change the single channel conductance compared to the wild type HCN2 channel (Johnson and Zagotta, 2005).  The C-linkers of each subunit are connected to their adjacent neighbours subunit. The Clinkers function to couple cAMP binding at the CNBD to the transmembrane domains, thereby modifying HCN channel opening. For HCN2 channels, the direct binding of cAMP results in a positive shift in the V1/2 (~+10 mV) and increases the open channel probability (Craven and Zagotta, 2004).  The binding of cAMP has been suggested to release an  inhibitory effect of the C-linker and CNBD on HCN2 which is transmitted via the C-linker to the S6 of the pore (Craven and Zagotta, 2004; Flynn et al., 2007; Zhou and Siegelbaum, 2007). This notion has also been suggested in the related CNG channels (Craven and Zagotta, 2004; Paoletti et al., 1999). The direct binding of cAMP to the CNBD was therefore suggested to stabilize the open state of HCN2 channels.  Specifically, mutation of a  positively charged residue to a negative residue, K472E, which is located in the B‟ helix of  22  the C-linker resulted in a positive shift in the V1/2 (~+10 mV) in the absence of cAMP. Furthermore, the V1/2 of the K472E mutant channel was unresponsive to cAMP. Based on the x-ray crystal structure of the C-linker and CNBD of HCN2,  the mutation of the  positively charged residue, K472, disrupted two salt bridge interactions with both intersubunit (E502, D‟ helix of adjacent subunit) and intrasubunit (D542, B roll of same subunit) negatively charged residues (Craven and Zagotta, 2004). These findings suggested that these residues stabilize the closed state and that the binding of cAMP to the CNBD breaks the salt bridges, thereby stabilizing the open state.  The opening and closing of If and HCN channels has been shown to be modulated allosterically by both voltage and cAMP (Altomare et al., 2001; DiFrancesco, 1999). The Altomare model employs a ten state cyclic allosteric model which includes 5 closed and 5 open state reactions which are voltage-dependent. The model assumes each HCN subunit has one independent voltage sensor that undergoes gating transitions in response to changes in voltage which contribute to channel opening and closing (Altomare et al., 2001). Therefore, each of the four voltage sensors is suggested to transition from a reluctant to a willing state which occurs through a cooperative allosteric interaction of all four subunits. The model accurately describes most features of HCN channel gating, such as the delay observed with activation, mid point of activation and the differences in the activation/deactivation time constants of the mammalian HCN channels. For example, the faster opening and closing rates for HCN1 compared to HCN2 could be explained by the greater ease with which the HCN1 voltage sensor moves from the reluctant to the willing  23  state. Furthermore, the binding of cAMP enhances these transitions and favors the open state (DiFrancesco, 1999; Wang et al., 2002; Zhou and Siegelbaum, 2007).  The Altomare model assumed that the closed to closed and open to open and closed to open transitions were all voltage dependent. However, recent evidence has shown that the closed to open transitions may be voltage-independent for HCN2 channels (Chen et al., 2007). In HCN2 channels, the activation rates were shown to be rate limiting at extreme hyperpolarized voltages (> -150 mV) and that cAMP increased the maximal amount of current in addition to shifting the V1/2 to positive potentials. Interestingly, HCN1 channels which have much faster opening kinetics and are relatively insensitive to cAMP did not show saturation of the activation kinetics at extreme hyperpolarized voltages. These observations suggested that for HCN1 channels the closed to open transition were voltage dependent. Using a chimeric approach, it was determined that the difference between the closed to open transitions for HCN1 and HCN2 were suggested to reside in the S4-S6 transmembrane segments (Chen et al., 2007).  1.2.6 Mutations in HCN4 are linked to human bradyarrhythmias Sinus bradycardia is classified clinically with patients exhibiting a slower than normal heart rate. Recently, point mutations in the human HCN4 gene have been linked to clinical sinus bradycardia (Milanesi et al., 2006; Nof et al., 2007; Schulze-Bahr et al., 2003; Ueda et al., 2004). Patients identified with sinus bardycardia were found to have point mutations in the pore forming domain and in the C-terminus of the HCN4 channel. These point mutations resulted in either a truncated C-terminus including the CNBD, a non-functional CNBD, reduced channel expression or channels which opened at very negative potentials. All of the  24  HCN4 point mutations resulted in slowing the spontaneous activity or firing rate of the SAN. The slowing of spontaneous activity was the result of less If being available during the diastolic depolarization phase of the action potential since the mutations significantly reduced current density or shifted the V1/2 to more hyperpolarized potentials. These studies highlight the importance of HCN4 channels in contributing to and setting basal heart rate in humans.  However, a temporal HCN4 knock out in the adult mouse did not have a drastic effect on spontaneous activity and did not interfere with β-adrenergic regulation of heart rate (Herrmann et al., 2007).  Based on these findings, HCN4 was suggested to provide a  depolarization reserve, since HCN4 knock out adult mice exhibited recurrent sinus pauses after vagal stimulation. Therefore, the presence of HCN4 was suggested to provide an inward current to counterbalance membrane repolarization after vagal stimulation. Furthermore, global HCN4 knockout mice were found to be embryonic lethal between days 9.5 to 11.5 which suggested an importance of HCN4 in development (Stieber et al., 2003a). The HCN4 knock out studies suggested that If is important for preventing dysrrhythmias and in embryonic development, but was not required for maintaining spontaneous activity in the heart.  However, recent experiments in adult mice, using heart specific expression of the human HCN4 573X mutant gene, was used to further elucidate the role of HCN4 in pacemaking in the mouse (Alig et al., 2009). In humans, HCN4 573X mutation resulted in a truncated Cterminus which lacked the CNBD and abolished cAMP modulation, which resulted in  25  clinical sinus bradycardia (Schulze-Bahr et al., 2003). In adult mice, the HCN4 573X mutation exhibited slower hearts at rest and during exercise but did not display recurrent sinus pauses as in the temporal HCN4 knock out adult mouse. Taken together, these studies provide support for the role of HCN4 channels in setting basal heart rate and contributing to pacemaking in both the mouse and human heart.  1.3 Voltage-dependent gating and pore opening in HCN channels 1.3.1 Isoform differences in activation rates are attributed to S1 and S2 As discussed above in section 1.2.4.4, the four mammalian HCN isoforms open at different rates, from fastest to slowest: HCN1>HCN2>HCN3>HCN4.  The different rates of activation between HCN1 and HCN4 are attributed to differences in S1, S1-S2 linker, and S2. At fully-activated potentials (> -130 mV), HCN4 activates ~10 times slower compared to HCN1 (Ishii et al., 1999). HCN4 and HCN1 are the slowest and fastest of the four mammalian HCN channels (Biel et al., 2009; Robinson and Siegelbaum, 2003). Using a chimeric mutagenesis approach, it was determined that the difference in activation rate between HCN4 and HCN1 could be attributed to S1, S1-S2 linker, and S2 (Ishii et al., 2001).  Swapping the entire region from either HCN1 or HCN4 into the  background of the other, resulted in chimeric HCN4 channels with activation rates as fast as HCN1, and chimeric HCN1 channels with activation rates as slow as HCN4.  At first the above results were supported by experiments showing that S1, S1-S2 linker, and S2 were also important for the differences in the activation rates between HCN2 and HCN4  26  (Stieber et al., 2003b). However, a single amino difference in the N-terminal region of S1 was actually determined to be completely responsible for the difference in the activation rate between HCN2 and HCN4 (Stieber et al., 2003b). Exchanging L272 of HCN4 with the analogous residue F221 of HCN2 conferred the HCN2 activation rate phenotype upon HCN4. The reverse residue exchange conferred the HCN4 activation rate phenotype upon HCN2. The same result was not achieved when replacing the leucine residue of HCN4 with the analogous residue of HCN1 (Stieber et al., 2003b). Taken together, these results may suggest that for HCN2 and HCN4, the S1, S1-S2 linker, and S2 are similar in structure, while HCN1 and HCN4 are not.  1.3.2 The S3-S4 linker modifies voltage-dependent gating In Kv channels, a mutagenesis scan showed that the extracellular S3-S4 linker which connects the S3 and S4 formed an alpha helix and was important in activation gating (Gonzalez et al., 2000, 2001; Mathur et al., 1997). Therefore, an alanine mutagenesis scan of the S3-S4 linker of HCN1 was employed to determine whether the S3-S4 linker also formed an alpha helix and was important for channel gating. The mutagenesis scan revealed that, compared to wild type HCN1 channels, three residues, G231, M232, and E235, resulted in a significant change in the free energy of activation while four residues, D233, S234, V236, and Y237, did not. The residues with the same phenotype clustered into two separate groups when plotted on a alpha helical wheel, suggesting that the S3-S4 linker was an alpha helix, as was determined for Kv channels (Lesso and Li, 2003).  Furthermore, shortening or  lengthening the S3-S4 linker corresponded to depolarizing and hyperpolarizing shifts in the V1/2, respectively (Tsang et al., 2004). These results suggested that S3-S4 linker, which is to  27  the tethered to the S4, influences its position or movement in response to voltage in HCN channels.  1.3.3 The S4 domain in voltage dependent gating 1.3.3.1 S4 primary structure The primary amino acid structure of the S4 domains are both similar and different for Kv and HCN channels. In Kv channels, the S4 contains a string of four to seven basic residues (e.g. lysine or arginine) which are separated by two hydrophobic residues. This sequence of positively charged residues is highly conserved across all Kv channels and is important for sensing changes in membrane potential (Shealy et al., 2003; Yellen, 2002). In mammalian HCN(1-4) channels, the S4 is also highly conserved and consists of the same general arrangement of basic residues (e.g. lysine or arginine), where each basic residue is separated by two hydrophobic residues (Jackson et al., 2007; Robinson and Siegelbaum, 2003; Shealy et al., 2003).  However, mammalian HCN(1-4) channels have nine positively charged  residues instead of the typical four to seven as observed in Kv channels (Jackson et al., 2007; Ludwig et al., 1998; Santoro et al., 1998). In addition, the nine basic residues cluster into two groups which are separated by a serine residue. The first and second groups consist of five and four basic residues, respectively. The similarities and differences in the S4 primary amino acid structure of HCN and Kv channels has triggered several investigations into determining how the positively charged residues of the S4 domain contribute to voltage sensing. This will be discussed in the following sections, 1.3.3.2 and 1.3.3.3.  28  1.3.3.2 Functional role of the S4 residues To determine the role of the nine positively charged S4 residues in voltage-dependent gating in mammalian HCN channels, each basic residue was mutated to the uncharged amino acid glutamine (Q). Mutagenesis experiments with the HCN2 channel showed that neutralization of each of the first four of the nine basic residues, K291Q, R294Q, R297Q, and R300Q, resulted in a negative shift in V1/2 (~-12 mV) with no effect on the slope factor (k), activation kinetics, and current amplitude (Chen et al., 2000; Vaca et al., 2000). However, mutation of all of the first four residues produced a quadruple mutant channel which displayed an additive hyperpolarizing shift in the V1/2 (~-44 mV). These results suggested that the first four residues stabilize one or many closed states, since a greater hyperpolarizing voltage was needed to open the quadruple mutant channel.  Mutation of the fifth basic residue, R303Q, resulted in ionic currents which were detected at very negative potentials or were non measurable. Mutations of the sixth, eighth and ninth basic residues, R309Q, R315Q, and R318Q, respectively, showed a significant reduction in the membrane surface expression of the mutant channels (Chen et al., 2000; Vaca et al., 2000). Specifically, surface expression for R309Q was reduced by 94%, which completely accounted for the loss of measurable current. For R315Q and R318Q, surface expression was reduced by 75% and 54%, respectively. The result for R315Q and R318Q suggested that, in addition to reduced surface expression, inhibition of channel opening could have also contributed to the lack of measurable current. Finally, the seventh residue, R312Q, also reduced the amount of time dependent current, but by ~ 4 times.  29  The serine residue, S306, separates the first five basic residues from the last four basic residues. Mutation of S306 to glutamine was also carried out to determine its role in voltagedependent gating. The S306Q mutant channel produced currents which were reduced by ~9 fold and showed very little time dependence.  Interestingly, mutation of the equivalent  residue in the non-mammalian spHCN channel resulted in a dramatic reduction in gating current. The reduction in gating current observed in the spHCN channel suggested a role for the S306 in voltage-sensing (Mannikko et al., 2002).  1.3.3.3 S4 movement The S4 in HCN channels responds to changes in membrane potential and undergoes conformational changes (Bruening-Wright et al., 2007; Mannikko et al., 2002). For example, in spHCN channels, mutation of the middle S4 residue, S338C, eliminated most of the gating current which was consistent with S4 movement in response to changes in membrane potential (Mannikko et al., 2002). Furthermore, it was also observed in spHCN channels, that fluorescence versus voltage curves, using an N-terminal S4 mutant residue, R332C, overlapped completely with charge versus voltage curves determined from gating currents (Bruening-Wright et al., 2007). These findings suggest S4 movement corresponds to gating charge movement which is indicative of voltage sensing associated with the S4. However, how the S4 moves in HCN channels has not been definitely resolved (Bell et al., 2004; Mannikko et al., 2002; Vemana et al., 2004)  Because the polarity of voltage-dependent opening and closing is reversed in HCN channels compared to Kv channels, it was first hypothesized that the movement of the S4 may also be  30  reversed (Mannikko et al., 2002). To determine whether this was the case, a substituted cysteine accessibility mutagenesis approach using the N-and C-terminal residues of the S4 of spHCN and HCN1 was employed using intracellular and extracellular MTSET (Bell et al., 2004; Mannikko et al., 2002; Vemana et al., 2004). This approach had been previously used successfully to determine the direction of S4 movement for Shaker K+ channels (Larsson et al., 1996).  The results with the substituted cysteine accessibility mutagenesis approach for Shaker K+ channels showed that buried N-terminal S4 residues became accessible to external MTSET only upon membrane depolarization when the channels were open and not during hyperpolarization when the channels were closed (Larsson et al., 1996). Conversely, buried C-terminal S4 residues became accessible to intracellular MTSET only upon membrane hyperpolarization when the channels were closed and not during membrane depolarization when the channels were open (Larsson et al., 1996).  The same experimental approach with HCN1 and spHCN channels gave similar results to those found for Shaker K+ channels. The buried N-terminal S4 residues were accessible to external MTSET only upon membrane depolarization (+50 mV). The buried C-terminal S4 residues were accessible to internal MTSET only upon membrane hyperpolarization (-100 mV) (Mannikko et al., 2002; Vemana et al., 2004). Furthermore, the S4 serine residue which is located in the middle of the S4 and separates the strings of basic residues for both spHCN (S338) and HCN1 (S253 and L254), was found to be accessible during both membrane depolarization and hyperpolarization. The non-specific voltage-dependent accessibility of  31  the serine residue to MTSET suggested that the middle portion of the S4 can be reached from either the outside or the inside of the cell membrane.  The above findings supported the notion that movement of the S4 was conserved in both hyperpolarization-activated HCN channels and depolarization-activated K+ channels. Therefore, the S4 moved upward and downward upon membrane depolarization and membrane hyperpolarization, respectively. It was also hypothesized that the S4 movement in HCN channels happened via a helical screw mechanism similar to what had been proposed for Kv channels. The helical screw mechanism proposed that the S4 helix translates 5-14 Å through the lipid membrane and undergoes some rotation (Baker et al., 1998; Cha et al., 1999; Larsson et al., 1996; Pathak et al., 2007).  However, Bell et al. suggested an alternative to the helical screw mechanism for the S4 voltage sensor movement in HCN channels. Using the substituted cysteine accessibility mutagenesis approach, as above, Bell et al. observed that the HCN1 N-terminal S4 cysteine subsituted residue, T249C, showed no voltage dependent accessibility to external MTSET (Bell et al., 2004). This was different to the Vemana et al. finding using the identical Nterminal S4 residue, T249C. Vemana et al. observed that the T249C residue showed greater accessibility to external MTSET upon membrane depolarization. Nonetheless, Bell et al. hypothesized that the N-terminal residues of the S4 were relatively static and that the movements of neighboring subunits open and collapse around the C-terminal S4 residues. This hypothesis was consistent with the proposed transporter model for Kv channels, which involved limited S4 movement (2-4 Å) through a narrowlly focused electric field created by  32  deformations and aqueous crevices of the lipid membrane (Ahern and Horn, 2005; Cha et al., 1999; Chanda et al., 2005; Posson et al., 2005).  As a further alternative, a paddle model has also been put forward to explain the orientation of the S1 to S4 alpha helices and their potential movements in Kv channels. Based upon the KvAP and Kv1.2 x-ray crystal structures, large movements (12-15 Å) of the S4 and part of the S3 were suggested to occur through the lipid membrane during changes in membrane potential (Jiang et al., 2003a; Jiang et al., 2003b; Long et al., 2007; Ruta et al., 2005). However, it is important to note that voltage sensing is dynamic and that crystal structures represent only a static conformation of the channel protein.  1.3.4 Coupling voltage-sensing to pore opening Even though the opening and closing of Kv and HCN channels are reversed with respect to voltage, the S4-S5 linker is important in coupling S4 movement to pore opening in both channels. The S4-S5 linker couples the movement of the S4 to the activation gate located in the lower end of the S6 in both Kv and HCN channels (Chen et al., 2001; Decher et al., 2004; Macri and Accili, 2004; Tristani-Firouzi et al., 2002).  In HCN2, an alanine mutagenesis scan of the residues which form the S4-S5 linker, produced mutant channels which shifted the V1/2 to more depolarized potentials and in some instances produced constitutively open channels (e.g. Y331 and R339) (Chen et al., 2001). These findings suggested that the S4-S5 linker mutations uncoupled the S4-S5 linker from the activation gate located in the lower end of the S6.  Further support for this coupling  33  mechanism was shown in a double mutant channel containing the point mutations Y331S and R318Q (Chen et al., 2001). The S4-S5 linker mutation, Y331S, produced a constitutively open channel when observed in isolation. The S4 mutation, R318Q, allowed the mutant channel to traffic to the cell membrane but did not give measurable currents on its own. However, R318Q in the presence of Y331S resulted in a double mutant channel with measurable currents. The Y331S mutation is therefore credited with uncoupling the effect of the S4 mutation, R318Q, on channel opening.  In addition, experiments have suggested that the S4-S5 linker and the S6 are in close proximity to each other. In the HCN2 channel, a positively charged residue in the S4-S5 linker, R339, and a negatively charged residue of the S6, D443, was suggested to form a salt bridge since disrupting this interaction by neutralizing the positive or negative residue resulted in constitutively open channels (Decher et al., 2004). It was also observed in spHCN channels, that a double cysteine mutant channel located in the S4-S5 linker, F359C, and post S6, K482C, could co-ordinate Cd2+ at hyperpolarized potentials. Co-ordination of Cd2+ between these residues suggested that the S4-S5 linker and post S6 were in close proximity when the channel was open (Prole and Yellen, 2006).  1.3.5 The activation gate in S6 The activation gate of HCN channels is located in the lower S6 region. In spHCN channels, the activation gate was first shown to be located at the cytoplasmic side of the channel using the specific HCN blocker, ZD7288 (Shin et al., 2001). Using excised-out patches, ZD7288 could be trapped in the closed state which suggested that the opening and closing processes  34  occurred at the cytoplasmic side of the channel. Experiments using the T464C mutant channel, a residue which is located near the lower end of the S6, and Cd2+, suggested that the S6 region forms the voltage-controlled constriction point of the pore. In the T464C mutant channel, Cd2+ reduced currents by ~95% at hyperpolarized potentials when the channels were open (Rothberg et al., 2002). However, less than 10% of the current was inhibited at depolarized potentials when the mutant channel was closed. Therefore, Cd2+ accessibility occurred only when the pore was open. Similar observations were also found using an analogous residue with Shaker K+ channels (del Camino and Yellen, 2001; Liu et al., 1997).  1.3.6 The proposed glycine hinge in S6 In K+ channels, the middle portion of the S6 is kinked at a central pivot point which is called the glycine hinge. When the S6 helices open, a low resistance pathway is formed which allows ions to flow through the pore. When these helices close, ion flux is significantly prevented. This structural rearrangement can be observed from the x-ray crystal structures of KcsA  from  Streptomyces  lividans,  and  MthK,  from  Methanobacterium  thermoautotrophicum, which captured the K+ channel pore in the closed and open states, respectively (Fig. 1. 5) (Doyle et al., 1998; Jiang et al., 2002b).  In both Kv and HCN channels, a conserved glycine in the S6 is important for channel biogenesis and function (Cheng et al., 2007; Ding et al., 2005; Jackson et al., 2007; Macri et al., 2009; Shealy et al., 2003). Mutation of the conserved glycine to alanine in Shaker K+ channels resulted in a non-functional channel (Ding et al., 2005). However, function could be restored in a double mutant channel which contained a glycine residue substituted one  35  KcsA  closed  MthK  opened  Figure 1.5 Comparison of the closed and opened channel pore in K+ channels Left, x-ray crystal structure of two of four subunits showing the KcsA pore in the closed state. Note the M2 (inner helices) come into contact at the lower end indicating that this conformation acts as a physical barrier to prevent the flow of ions through the channel pore. Right, x-ray crystal structure of two of four subunits showing the MthK pore in the open state. Note the M2 (inner helices) at the lower end are far apart from each other indicating that in this conformation the flow of ions through the channel pore is permitted.  36  position C-terminal to the alanine mutation. Furthermore, mutation of the glycine gave rise to unglycosylated channels which indicated a lack of surface expression on the plasma membrane.  In the HCN2 channel, mutation of the glycine to an alanine, G424A, similarly resulted in non-measurable currents (Cheng et al., 2007; Macri et al., 2009). These results were due to a trafficking or folding defect since the G424A mutation resulted in unglycosylated channels which indicated a lack of surface expression on the plasma membrane (Macri et al., 2009). It was not determined whether the creation of a double mutant by inserting a glycine residue in another region of the S6 restored channel function. Based upon the location of the T464C and the accessibility to Cd2+, as discussed above, the bending point of the S6 in HCN channels occurs below the conserved glycine residue.  The S6 regions of most Kv channels also have a PXP motif, but HCN channels do not. The PXP motif is located below the conserved glycine hinge residue in the S6 in Kv channels (Jackson et al., 2007; Shealy et al., 2003). The PXP motif has also been suggested to be a bending point during opening and closing in Kv channels.  Mutating the PXP residues  resulted in non-functional channels. However, re-inserting the PXP motif a few residues below or above the mutated PXP residues rescued channel function (Labro et al., 2003). Therefore, the bending points of S6 in HCN channels are similar but not the same as in K v channels.  37  1.3.7 Energetics of pore opening in HCN channels As discussed above, the pore of HCN and K+ channels is proposed to be structurally similar based upon several findings. For example, the orientation and structure of the HCN pore in the plasma membrane is thought to be similar to K+ channels based upon cysteine accessibility mutagenesis studies and homology modeling (Au et al., 2008; Giorgetti et al., 2005; Roncaglia et al., 2002; Xue and Li, 2002). Furthermore, the lower end of the S6 contains the activation gate in both HCN and Kv channels (del Camino et al., 2000; Liu et al., 1997; Rothberg et al., 2002; Shin et al., 2001). In addition, the S4-S5 linker couples the movement of the S4 to the activation gate in both HCN and Kv channels (Chen et al., 2001; Decher et al., 2004; Macri and Accili, 2004; Tristani-Firouzi et al., 2002). The S4 of HCN channels contain a string of positively charged residues that sense changes in voltage in a similar fashion to K+ channels (Bell et al., 2004; Larsson et al., 1996; Mannikko et al., 2002; Vemana et al., 2004).  For the Shaker K+ channel, it has been suggested that the closed state is intrinsically more stable and that depolarization and the voltage sensors must work to open the channel pore. This was concluded since most alanine/valine point mutants of the S6 shifted the activation curve to hyperpolarized potentials favoring the open state (Hackos et al., 2002; Yifrach and MacKinnon, 2002). The point mutations prevented optimal protein packing of the closed pore as observed in the x-ray crystal structure of the KcsA pore (Fig. 1.5) (Doyle et al., 1998). Therefore, it was suggested that the closed pore was the low energy stable state. Furthermore, it was suggested that x-ray crystal structure of the MthK K+ channel, from  38  Methanobacterium thermoautotrophicum, which captures the K+ channel pore in the open state, represented the high energy unstable state.  In HCN channels, it is not known whether the closed pore is the low energy state. In HCN channels the voltage sensor moves in a somewhat similar fashion as in Kv channels: upwards upon depolarization and downwards upon hyperpolarization.  Therefore, to explain the  reverse voltage dependence of pore opening, the coupling of the voltage sensors to the activation gate located in the S6 was thought to be reversed (Bell et al., 2004; Mannikko et al., 2002; Vemana et al., 2004). This would imply that the closed state of the HCN channel pore would also be the low energy conformation as in Kv channels. Chapter 2 of this thesis addresses whether the closed or open pore is the low energy state in HCN channels by employing the same experimental approach used for the Shaker K+ channel.  1.4 The structure and function of the HCN selectivity filter 1.4.1 Proposed structure of the selectivity filter As discussed in section 1.2, all members of the potassium channel family, which include HCN channels, share a common pore structure that forms a central ion permeation path (Biel et al., 2009; Yellen, 2002). X-ray crystallography has revealed a K+ channel pore structure that can be divided into two functional domains: the selectivity filter and the activation gate (Doyle et al., 1998; Jiang et al., 2002b; Jiang et al., 2003a; Long et al., 2005). These two functional domains, located at opposite ends of the ion permeation path, each have a unique function. The selectivity filter is located near the top end of the channel pore and contains the GYG signature sequence residues and physically separates the extracellular environment  39  from the internal pore cavity in both K+ and HCN channels (Au et al., 2008; Doyle et al., 1998; Giorgetti et al., 2005; Jackson et al., 2007; Jiang et al., 2003a; Long et al., 2005; Shealy et al., 2003). The role for this region in regulating ion flow has not been examined directly in HCN channels, the similarity of this region to the selectivity filter of K+ selective channels makes it probable that cation binding sites exist and that the movement of cations through the pore proceeds in a manner that is similar (Doyle et al., 1998; Hille, 2001; Zagotta, 2006). But despite this striking similarity to K+ channels, HCN channels also allow the passage of Na+ (DiFrancesco, 1981b). The passage of Na+ is critical for the depolarization of cells at subthreshold membrane potentials following hyperpolarization (DiFrancesco, 1993; Kaupp and Seifert, 2001; Pape, 1996; Robinson and Siegelbaum, 2003)  The structure of the selectivity filter of HCN channels is proposed to be similar to K+ channels because of a shared primary amino acid identity and a homology of the HCN2 channel selectivity filter based on the x-ray crystal structure of the KcsA K+ channel. In most K+ channels, including KcsA, the amino acid residues TVGYG form the selectivity filter, however in HCN channels the amino acid residues CIGYG form the proposed selectivity filter (Fig. 1.6). As shown in Fig 1.3, the selectivity filters of both the KcsA and HCN2 channel are positioned in place by the pore helices.  The x-ray crystal structures from both bacterial and mammalian K+ channels show that the selectivity filter residues, TVGYG, produce a stack of backbone carbonyl oxygen atoms that form negatively charged rings that co-ordinate dehydrated K+ ions (Doyle et al., 1998; Jiang et al., 2002a; Jiang et al., 2003a; Long et al., 2005; Zhou et al., 2001). The backbone  40  carbonyl oxygen atoms create four cation binding sites, S1 (Y-G), S2 (G-V), S3 (V-T) and S4 (T- and the threonine hydroxyl, - OH, side chain group), which function to mimic the environment of a hydrated K+ ion in solution (Fig 1.6). The S4 is located just above the central pore cavity and S1 is located near the extracellular entrance. Hydrated cations and water are located below and above these sites. An external binding site located just above the selectivity filter at the extracellular entrance, defined as S0, has also been identified and holds a partially dehydrated K+ ion. The negatively charged rings of backbone carbonyl oxygen atoms energetically balance the cost of hydrating and dehydrating K+, however, the energetic cost for dehydrating Na+ would be too high, thus resulting in the low permeability of Na+ relative to K+ (Zhou et al., 2001). Despite the high degree of K+ selectivity, fast conduction rates reaching the limits of diffusion are achieved through a single file multi-ion process where two K+ ions simultaneously occupy two sites, S1 and S3 or S2 and S4 which are separated by water. K+ movement through the selectivity filter occurs via a „knock on‟ mechanism, where electrostatic repulsion shuttles K+ between S1, S3 and S2, S4 (Aqvist and Luzhkov, 2000; Berneche and Roux, 2001; Morais-Cabral et al., 2001; Zhou and MacKinnon, 2003)  In HCN channels, the predicted fourth cation binding site (S4) is the most striking difference when comparing the selectivity filter of HCN channels to other K+ selective channels (Fig. 1.6). The HCN2 homology model of the selectivity filter based upon the KcsA K + channel shows that the first three binding sites are formed by the backbone carbonyl oxygen atoms S 1 (Y-G), S2 (G-V), S3 (V-T) which recapitulate the three binding sites formed by the backbone carbonyl oxygen atoms of the KcsA K+ channel (Giorgetti et al., 2005). However, part of  41  KcsA  HCN2  G 1 2  Y  G I  3 4  C Figure 1.6 The residues that make up the selectivity filter of HCN2 may form four ion binding sites similar to KcsA Left, x-ray crystal structure showing two of four subunits which form the selectivity filter of KcsA. The TVGYG residues contribute negatively charged backbone carbonyl oxygens (in red) which form four cation binding sites which coordinate dehydrated K+ ions (numbered green spheres). Right, homology model of two of four subunits which form the selectivity filter of HCN2 based upon KcsA. The CIGYG residues may also contribute negatively charged backbone carbonyl oxygens (in red) to form four cation binding sites that may also co-ordinate dehydrated K+ and Na+ ions (numbered green spheres). Note that the fourth binding site in KcsA is formed by the backbone carbonyl oxygen of threonine and the hydroxyl group of threonine. However, the proposed fourth binding site in HCN2 is different from KcsA since it is formed by the backbone carbonyl oxygen of threonine and the sulfur group of cysteine (Morais-Cabral et al., 2001; Giorgetti et al., 2005).  42  the fourth binding site in K+ channels is formed by the hydroxyl groups of the four threonine residues, while in HCN channels it is proposed to be formed by the sulfydryl groups of the four cysteine residues. The sulfhydryl side chain groups have been suggested to form a divalent cation binding site and be part of the permeation pathway in HCN channels, since mutation to a threonine or serine reduced current block by intracellular Mg2+ and Cd2+ in HCN2 and spHCN channels, respectively (Roncaglia et al., 2002; Vemana et al., 2008). These intracellular divalent blocking studies have suggested that the opposite α-carbons of the sulfhydryl side chain groups are ~ 11 Å apart. However, the opposite α-carbons of the hydroxyl side chain group are ~ 3 Å apart, based upon the x-ray crystal structure of the KcsA K+ channel. Nevertheless, whether the sulfhydryl side chain groups of this cysteine interact with permeating ions, as in K+ channels, is not known.  1.4.2 The GYG residues of the selectivity filter Compared to K+ channels, a site-directed mutagenesis approach in HCN channels has provided limited information on the function of the GYG residues of the selectivity filter. For example, in Shaker and Kv2.1, mutagenesis has shown that the GYG selectivity filter residues are important for maintaining high selectivity for K+ over Na+ (Chapman et al., 2001; Heginbotham et al., 1992; Heginbotham et al., 1994). For the HCN1 and HCN2 channels, mutation of any of the GYG residues produced mutant channels which could traffic to the cell membrane, but were non-functional. The HCN1 and HCN2 GYG selectivity filter mutant channels did not produce measurable currents (Er et al., 2003; Macri et al., 2002; Xue et al., 2002). These results suggested that the GYG residues of the selectivity filter are important for HCN channel function. However, because of the lack of measurable currents  43  for HCN1 and HCN2, no information could be determined about how these residues might contribute to ion selectivity, as in K+ channels. However, for the HCN4 channel, mutation of the second glycine of the GYG did produce measurable currents. The mutant channels were activated at only very negative potentials (>-120 mV), but sustained wild type ion selectivity (Nof et al., 2007).  1.4.3 The C-terminal residues located immediately outside the GYG Experiments have suggested that the residue that immediately follows the GYG was also not involved in ion selectivity. In HCN channels, either a positive (R, K) or non-charged (Q, A) residue immediately follows the GYG amino acid residues (Gauss et al., 1998; Jackson et al., 2007; Ludwig et al., 1998; Santoro et al., 1998). To determine whether the residue that immediately follows the GYG was involved in ion selectivity, the positive or uncharged residues were replaced with a negative aspartate residue, which immediately follows the GYG in most K+ channels. These GYGD selectivity filter mutant HCN channels did produce measurable currents, but did not confer high selectivity for K+ over Na+ in either spHCN, HCN1 or HCN2 channels (Azene et al., 2003; Roncaglia et al., 2002).  However, the residues located just C-terminal to the GYG were determined to be important in controlling the effects of extracellular K+ on channel gating. In HCN2, increasing the ratio of extracellular K+ to Na+ accelerated the rate of channel closing and shifted the V1/2 to more negative voltages (Azene et al., 2003; Macri et al., 2002). In HCN1, increasing the ratio of extracellular K+ to Na+ accelerated both the rate of channel opening and closing and also shifted the V1/2 to more negative voltages (Azene et al., 2003). In HCN1 and HCN2,  44  mutation of residues located just C-terminal to the GYG motif, A352 and A354, to negative or polar residues abolished the effects of extracellular K+ on channel gating (Azene et al., 2003; Azene et al., 2005). Therefore, it was suggested that the effect of extracellular K+ on channel gating was due to conformational changes associated with the selectivity filter.  1.4.4 Extracellular K+ and Na+ may affect conductance at the selectivity filter Raising extracellular K+ has been observed to significantly increase whole-cell slope conductance in both native tissue and HCN channels expressed in heterologous systems (DiFrancesco, 1981b, 1982; Edman and Grampp, 1989; Frace et al., 1992; Ludwig et al., 1998; Macri et al., 2002; Moroni et al., 2000; Solomon and Nerbonne, 1993). The most dramatic increases on Gf occurred over the physiological range of extracellular K+ (5.4 -10 mM) and were found to saturate at ~20 mM (DiFrancesco, 1981b, 1982; Edman and Grampp, 1989; Frace et al., 1992; Macri et al., 2002; Solomon and Nerbonne, 1993). The effect of extracellular K+ on conductance may by physiologically important. For example, during exercise, extracellular K+ may rise to levels as high as 9 mM which would depolarize the resting membrane potential (Paterson, 1996). Depolarization of the membrane would be detrimental to the activation of If, since less inward current would be available. Therefore, the increase in conductance would counter membrane depolarization with elevated levels of extracellular K+.  It has been suggested that the observed increase in conductance in response to raising extracellular K+ may be due to an allosteric effect where extracellular K+ would bind to an external site to increase the open channel probability or to increase the permeation of K+ and  45  Na+ through the open channel pore (DiFrancesco, 1982; Edman and Grampp, 1989; Maruoka et al., 1994). However, to date, the molecular mechanism which underlies the effect of raising extracellular K+ on conductance remains unknown.  Experiments suggested that the permeation pathway and, specifically, the selectivity filter of HCN channels, may be the target of interest in controlling the effect of extracellular K+ on conductance. In support of an effect of extracellular K+ on permeation, the reversal potential was found to be sensitive to changes in extracellular K+ (DiFrancesco, 1981b; Frace et al., 1992). Furthermore, the complete removal of extracellular K+, with only extracellular Na+ remaining, eliminated current flow in the inward direction whereas outward current remained (Frace et al., 1992; Wollmuth, 1995; Wollmuth and Hille, 1992). These findings suggested that permeation was impaired whereas the ability of the channel to open in response to voltage was spared. In excised patches using HCN2, the effect of K+ to maintain Na+ currents was shown to be bidirectional which suggested that the regions responsible could be accessed from either side of the channel thus implicating the permeation pathway and the selectivity filter (Lyashchenko and Tibbs, 2008).  The ability of both extracellular K+ and extracellular Na+ to modify conductance in a way that reflects their relative ability to permeate implies that their effects are controlled by the permeation pathway. In studies of native tissue, increases in extracellular Na+ were shown to affect conductance very little or not at all (DiFrancesco, 1981b, 1982; Edman and Grampp, 1989; Ho et al., 1993). However, recent experiments on the human HCN2 channel showed for the first time that increases in extracellular Na+ did increase conductance (Moroni et al.,  46  2000). Even though they were not compared directly, the magnitude and sensitivity of the changes in Gf produced by extracellular Na+ appeared to be smaller than those produced by extracellular K+.  1.4.5 Conductance and gating at the fourth ion binding site of the selectivity filter The fourth or innermost binding site, S4 of the selectivity filter alters gating and conductance in K+ and HCN channels. In the bacterial KcsA K+ channel, x-ray crystallography showed that a mutation of the threonine to a cysteine (T75C) decreased the occupancy of K+ at S4 which led to a significant reduction in single channel conductance when raising extracellular K+ concentration compared to wild type (Zhou and MacKinnon, 2004). In the Shaker K+ channel, mutation of the threonine to a serine (T442S) did not alter ion selectivity but did increase the duration of the single channel openings and shifted the voltage dependence of opening to more negative potentials, thus destabilizing the closed state (Heginbotham et al., 1994; Yool and Schwarz, 1991). In HCN4 channels, mutation of the cysteine (C479) to a threonine decreased the relative permeability of K+ over Na+, accelerated channel opening and closing and did not alter the V1/2 of channel opening (D'Avanzo et al., 2009). However, the HCN4 study was limited since only the single point mutation was performed and it did not address how the cysteine contributed to ion selectivity and the effects of extracellular K+ on conductance, which are both fundamental properties of the „funny‟ current.  We therefore asked the question: Does the proposed fourth ion binding site play a role in regulating ion selectivity and the effects of extracellular K+ on conductance? As discussed above, a site-directed mutagenesis approach has provided limited information on the role the  47  GYG residues play in regulating ion selectivity and conductance in HCN channels. In Chapter 3, we mutated the conserved cysteine residue to threonine, serine (which is much smaller in volume than threonine but contains the hydroxyl side chain group) and alanine (which has the same volume as serine but contains a methyl side chain group), to determine the role, if any, that the conserved cysteine residue of the CIGYG selectivity filter plays in regulating ion selectivity and conductance.  48  1.5 Statement of thesis objectives  HCN channels are the molecular determinants of the hyperpolarization-activated cyclic nucleotide-gated, funny current, If (Gauss et al., 1998; Ludwig et al., 1998; Santoro et al., 1998). HCN channels contribute and help to regulate the excitability of spontaneously active cells found in cardiac tissue and neurons (Biel et al., 2009; Robinson and Siegelbaum, 2003). HCN channels are members of the Kv channel superfamily and therefore share four common defining features related to their structure and function (Biel et al., 2009; Robinson and Siegelbaum, 2003). HCN like Kv channels have: 1) an S4 voltage sensor made up of a string of positive residues which moves upwards and downwards upon membrane depolarization and hyperpolarization, respectively (Bell et al., 2004; Mannikko et al., 2002; Vemana et al., 2004), 2) an S6 which forms the inner pore cavity and contains the voltage-controlled activation gate which undergoes conformational changes that open and close the channel pore (Rothberg et al., 2002; Rothberg et al., 2003; Shin et al., 2001), 3) an intracellular S4-S5 linker which couples S4 voltage sensor movement to the activation gate (Chen et al., 2001; Decher et al., 2004; Macri and Accili, 2004; Prole and Yellen, 2006), and 4) a selectivity filter that has the GYG K+ channel signature sequence motif which is important for allowing current flow (Azene et al., 2003; Er et al., 2003; Macri et al., 2002; Xue et al., 2002). Based upon these four defining features, it may be suggested that HCN and Kv channels are in general, related in structure and function.  However, despite a shared structure and function, there are two strikingly apparent functional differences between HCN and Kv channels. These are 1) the HCN channel pore opens and  49  closes upon membrane hyperpolarization and depolarization, respectively, even though the S4 moves in a similar fashion to Kv channels and 2) the HCN current, If is carried by both K+ and Na+ despite having the GYG K+ channel signature sequence motif (Biel et al., 2009; Robinson and Siegelbaum, 2003).  These two functional differences are vital for the  proposed role of HCN channels in contributing and regulating excitability in spontaneously active cells.  During repolarization of the action potential and under physiological  concentrations of K+ and Na+, HCN channels open and provide an inward current carried mostly by Na+ which depolarizes the membrane to help reach threshold firing of the next action potential. To date the mechanisms underlying these two physiologically important processes remain unknown. Working within this context, this thesis sets out to answer two important questions: 1) is the structure of the closed pore of HCN channels similar to Kv channels even though pore opening occurs with a reversed polarity? and 2) how do the residues which form the selectivity filter, CIGYG, regulate K+ and Na+ flow through the HCN channel pore?  In Chapter 2 the main objective was to determine whether the closed pore in HCN channels was the low energy conformation as in Kv channels. In the Shaker K+ channel, it has been suggested that the closed pore is intrinsically more stable and that depolarization and the voltage sensors must work to open the channel since an alanine/valine scan of the S6 disrupted the closed state by shifting the V1/2 to more hyperpolarized potentials (Hackos et al., 2002; Yifrach and MacKinnon, 2002). In HCN channels, the voltage sensor moves in the same direction as in Kv channels, upwards upon depolarization and downwards upon hyperpolarization, however the coupling of the voltage sensors to the activation gate is  50  thought to be reversed (Bell et al., 2004; Mannikko et al., 2002; Vemana et al., 2004). Because the pore structure is thought to be similar between K+ and HCN channels, and that only the coupling of the voltage sensor to the activation gate is different between the two channels, we hypothesize that the closed state of the HCN channel pore would also be the low energy conformation as in Shaker. To determine whether the closed pore was the low energy state in HCN channels, an alanine/valine scan of the S6 using the HCN2 channel was employed as in the Shaker study. Surprisingly, the closed pore was not the low energy state in HCN channels, but the energetic equilibrium between the open and closed states was similar since the mutations resulted in shifts in V1/2 that were mixed.  In Chapter 3, the main objective was to determine the role the conserved cysteine residue of the selectivity filter contributes to HCN2 channel function. The permeation pathway in HCN channels has been suggested to be formed by the residues which make up the selectivity filter, CIGYG (Giorgetti et al., 2005). It has been previously suggested that the cysteine residue of the selectivity filter forms an intracellular binding site for Mg2+ and Cd2+ (Roncaglia et al., 2002; Vemana et al., 2008). However, whether the cysteine residue, which is completely conserved in mammalian HCN channels, also plays a role in controlling ion selectivity and the effects of extracellular K+ on conductance is not known. We found that mutation of the cysteine to a threonine but not alanine or serine, of the selectivity filter in the HCN2 channel, which recapitulates the S4 binding site of the selectivity filter of K+ selective channels, reduced the relative permeability of K+ to Na+. Furthermore, the T400 mutation reduced K+ conductance but had no effect on Na+ conductance. Channel opening was also facilitated by the threonine substitution; strikingly, both channel opening and K+ conduction  51  phenotypes could be reverted to wild type by increasing intracellular sodium concentrations. These data show that, in HCN channels, the sulfhydryl side chain group does not contribute to permeation and gating, and that the backbone carbonyls, in part, control these functions.  52  1.6 References  Accili, E.A., Proenza, C., Baruscotti, M., and DiFrancesco, D. (2002). From funny current to HCN channels: 20 years of excitation. News Physiol Sci 17, 32-37.  Ahern, C.A., and Horn, R. (2005). Focused electric field across the voltage sensor of potassium channels. Neuron 48, 25-29.  Alig, J., Marger, L., Mesirca, P., Ehmke, H., Mangoni, M.E., and Isbrandt, D. (2009). Control of heart rate by cAMP sensitivity of HCN channels. Proceedings of the National Academy of Sciences of the United States of America.  Altomare, C., Bucchi, A., Camatini, E., Baruscotti, M., Viscomi, C., Moroni, A., and DiFrancesco, D. (2001). Integrated allosteric model of voltage gating of HCN channels. The Journal of General Physiology 117, 519-532.  Aqvist, J., and Luzhkov, V. (2000). Ion permeation mechanism of the potassium channel. Nature 404, 881-884.  Au, K.W., Siu, C.W., Lau, C.P., Tse, H.F., and Li, R.A. (2008). Structural and functional determinants in the S5-P region of HCN-encoded pacemaker channels revealed by cysteinescanning substitutions. American Journal of Physiology 294, C136-144.  53  Azene, E., Xue, T., and Li, R.A. (2003). Molecular basis of the effect of potassium on heterologously expressed pacemaker (HCN) channels. The Journal of Physiology 547, 349356.  Azene, E.M., Sang, D., Tsang, S.Y., and Li, R.A. (2005). Pore-to-gate coupling of HCN channels revealed by a pore variant that contributes to gating but not permeation. Biochemical and Biophysical Research Communications 327, 1131-1142.  Bader, C.R., Macleish, P.R., and Schwartz, E.A. (1979). A voltage-clamp study of the light response in solitary rods of the tiger salamander. The Journal of Physiology 296, 1-26.  Baker, O.S., Larsson, H.P., Mannuzzu, L.M., and Isacoff, E.Y. (1998). Three transmembrane conformations and sequence-dependent displacement of the S4 domain in shaker K+ channel gating. Neuron 20, 1283-1294.  Bell, D.C., Yao, H., Saenger, R.C., Riley, J.H., and Siegelbaum, S.A. (2004). Changes in local  S4  environment  provide  a  voltage-sensing  mechanism  for  mammalian  hyperpolarization-activated HCN channels. The Journal of General Physiology 123, 5-19.  Berneche, S., and Roux, B. (2001). Energetics of ion conduction through the K+ channel. Nature 414, 73-77.  54  Bers, D.M. (2006). The beat goes on: diastolic noise that just won't quit. Circulation Research 99, 921-923.  Biel, M., Wahl-Schott, C., Michalakis, S., and Zong, X. (2009). Hyperpolarization-activated cation channels: from genes to function. Physiological Reviews 89, 847-885.  Bogdanov, K.Y., Maltsev, V.A., Vinogradova, T.M., Lyashkov, A.E., Spurgeon, H.A., Stern, M.D., and Lakatta, E.G. (2006). Membrane potential fluctuations resulting from submembrane Ca2+ releases in rabbit sinoatrial nodal cells impart an exponential phase to the late diastolic depolarization that controls their chronotropic state. Circulation Research 99, 979-987.  Bois, P., Bescond, J., Renaudon, B., and Lenfant, J. (1996). Mode of action of bradycardic agent, S 16257, on ionic currents of rabbit sinoatrial node cells. British Journal of Pharmacology 118, 1051-1057.  Brown, H., and Difrancesco, D. (1980). Voltage-clamp investigations of membrane currents underlying pace-maker activity in rabbit sino-atrial node. The Journal of Physiology 308, 331-351.  Brown, H.F., DiFrancesco, D., and Noble, S.J. (1979). How does adrenaline accelerate the heart? Nature 280, 235-236.  55  Bruening-Wright, A., Elinder, F., and Larsson, H.P. (2007). Kinetic relationship between the voltage sensor and the activation gate in spHCN channels. The Journal of General Physiology 130, 71-81.  Bucchi, A., Baruscotti, M., Robinson, R.B., and DiFrancesco, D. (2003). I(f)-dependent modulation of pacemaker rate mediated by cAMP in the presence of ryanodine in rabbit sinoatrial node cells. Journal of Molecular and Cellular Cardiology 35, 905-913.  Cerbai, E., Barbieri, M., and Mugelli, A. (1996). Occurrence and properties of the hyperpolarization-activated current If in ventricular myocytes from normotensive and hypertensive rats during aging. Circulation 94, 1674-1681.  Cerbai, E., Pino, R., Sartiani, L., and Mugelli, A. (1999). Influence of postnatal-development on I(f) occurrence and properties in neonatal rat ventricular myocytes. Cardiovascular Research 42, 416-423.  Cha, A., Snyder, G.E., Selvin, P.R., and Bezanilla, F. (1999). Atomic scale movement of the voltage-sensing region in a potassium channel measured via spectroscopy. Nature 402, 809813.  Chanda, B., Asamoah, O.K., Blunck, R., Roux, B., and Bezanilla, F. (2005). Gating charge displacement in voltage-gated ion channels involves limited transmembrane movement. Nature 436, 852-856.  56  Chapman, M.L., Krovetz, H.S., and VanDongen, A.M. (2001). GYGD pore motifs in neighbouring potassium channel subunits interact to determine ion selectivity. The Journal of Physiology 530, 21-33.  Chen, J., Mitcheson, J.S., Lin, M., and Sanguinetti, M.C. (2000). Functional roles of charged residues in the putative voltage sensor of the HCN2 pacemaker channel. The Journal of Biological Chemistry 275, 36465-36471.  Chen, J., Mitcheson, J.S., Tristani-Firouzi, M., Lin, M., and Sanguinetti, M.C. (2001). The S4-S5 linker couples voltage sensing and activation of pacemaker channels. Proceedings of the National Academy of Sciences of the United States of America 98, 11277-11282.  Chen, S., Wang, J., Zhou, L., George, M.S., and Siegelbaum, S.A. (2007). Voltage sensor movement and cAMP binding allosterically regulate an inherently voltage-independent closed-open transition in HCN channels. The Journal of General Physiology 129, 175-188.  Cheng, L., Kinard, K., Rajamani, R., and Sanguinetti, M.C. (2007). Molecular mapping of the binding site for a blocker of hyperpolarization-activated, cyclic nucleotide-modulated pacemaker channels. The Journal of Pharmacology and Experimental Therapeutics 322, 931939.  Craven, K.B., and Zagotta, W.N. (2004). Salt bridges and gating in the COOH-terminal region of HCN2 and CNGA1 channels. The Journal of General Physiology 124, 663-677.  57  D'Avanzo, N., Pekhletski, R., and Backx, P.H. (2009). P-loop residues critical for selectivity in K channels fail to confer selectivity to rabbit HCN4 channels. PloS One 4, e7712.  Decher, N., Chen, J., and Sanguinetti, M.C. (2004). Voltage-dependent gating of hyperpolarization-activated, cyclic nucleotide-gated pacemaker channels: molecular coupling between the S4-S5 and C-linkers. The Journal of Biological Chemistry 279, 13859-13865.  Dekker, J.P., and Yellen, G. (2006). Cooperative gating between single HCN pacemaker channels. The Journal of General Physiology 128, 561-567.  del Camino, D., Holmgren, M., Liu, Y., and Yellen, G. (2000). Blocker protection in the pore of a voltage-gated K+ channel and its structural implications. Nature 403, 321-325.  del Camino, D., and Yellen, G. (2001). Tight steric closure at the intracellular activation gate of a voltage-gated K(+) channel. Neuron 32, 649-656.  DiFrancesco, D. (1981a). A new interpretation of the pace-maker current in calf Purkinje fibres. The Journal of Physiology 314, 359-376.  DiFrancesco, D. (1981b). A study of the ionic nature of the pace-maker current in calf Purkinje fibres. The Journal of Physiology 314, 377-393.  58  DiFrancesco, D. (1982). Block and activation of the pace-maker channel in calf purkinje fibres: effects of potassium, caesium and rubidium. J Physiol 329, 485-507.  DiFrancesco, D. (1984). Characterization of the pace-maker current kinetics in calf Purkinje fibres. The Journal of Physiology 348, 341-367.  DiFrancesco, D. (1986). Characterization of single pacemaker channels in cardiac sino-atrial node cells. Nature 324, 470-473.  DiFrancesco, D. (1991). The contribution of the 'pacemaker' current (if) to generation of spontaneous activity in rabbit sino-atrial node myocytes. The Journal of Physiology 434, 2340.  DiFrancesco, D. (1993). Pacemaker mechanisms in cardiac tissue. Annu Rev Physiol 55, 455-472.  DiFrancesco, D. (1999). Dual allosteric modulation of pacemaker (f) channels by cAMP and voltage in rabbit SA node. The Journal of Physiology 515, 367-376.  DiFrancesco, D., Ducouret, P., and Robinson, R.B. (1989). Muscarinic modulation of cardiac rate at low acetylcholine concentrations. Science 243, 669-671.  59  DiFrancesco, D., Ferroni, A., Mazzanti, M., and Tromba, C. (1986). Properties of the hyperpolarizing-activated current (if) in cells isolated from the rabbit sino-atrial node. The Journal of Physiology 377, 61-88.  DiFrancesco, D., and Mangoni, M. (1994). Modulation of single hyperpolarization-activated channels (i(f)) by cAMP in the rabbit sino-atrial node. The Journal of Physiology 474, 473482.  DiFrancesco, D., and Ojeda, C. (1980). Properties of the current if in the sino-atrial node of the rabbit compared with those of the current iK, in Purkinje fibres. The Journal of Physiology 308, 353-367.  DiFrancesco, D., and Tortora, P. (1991). Direct activation of cardiac pacemaker channels by intracellular cyclic AMP. Nature 351, 145-147.  DiFrancesco, D., and Tromba, C. (1988a). Inhibition of the hyperpolarization-activated current (if) induced by acetylcholine in rabbit sino-atrial node myocytes. The Journal of Physiology 405, 477-491.  DiFrancesco, D., and Tromba, C. (1988b). Muscarinic control of the hyperpolarizationactivated current (if) in rabbit sino-atrial node myocytes. The Journal of Physiology 405, 493-510.  60  Ding, S., Ingleby, L., Ahern, C.A., and Horn, R. (2005). Investigating the putative glycine hinge in Shaker potassium channel. The Journal of General Physiology 126, 213-226.  Doyle, D.A., Morais Cabral, J., Pfuetzner, R.A., Kuo, A., Gulbis, J.M., Cohen, S.L., Chait, B.T., and MacKinnon, R. (1998). The structure of the potassium channel: molecular basis of K+ conduction and selectivity. Science 280, 69-77.  Edman, A., and Grampp, W. (1989). Ion permeation through hyperpolarization-activated membrane channels (Q-channels) in the lobster stretch receptor neurone. Pflugers Arch 413, 249-255.  Er, F., Larbig, R., Ludwig, A., Biel, M., Hofmann, F., Beuckelmann, D.J., and Hoppe, U.C. (2003). Dominant-negative suppression of HCN channels markedly reduces the native pacemaker current I(f) and undermines spontaneous beating of neonatal cardiomyocytes. Circulation 107, 485-489.  Fernandez-Velasco, M., Ruiz-Hurtado, G., and Delgado, C. (2006). I (K1) and I (f) in ventricular myocytes isolated from control and hypertrophied rat hearts. Pflugers Arch 452, 146-154.  Flynn, G.E., Black, K.D., Islas, L.D., Sankaran, B., and Zagotta, W.N. (2007). Structure and rearrangements in the carboxy-terminal region of SpIH channels. Structure 15, 671-682.  61  Frace, A.M., Maruoka, F., and Noma, A. (1992). External K+ increases Na+ conductance of the hyperpolarization-activated current in rabbit cardiac pacemaker cells. Pflugers Arch 421, 97-99.  Gauss, R., Seifert, R., and Kaupp, U.B. (1998). Molecular identification of a hyperpolarization-activated channel in sea urchin sperm. Nature 393, 583-587.  Giorgetti, A., Carloni, P., Mistrik, P., and Torre, V. (2005). A homology model of the pore region of HCN channels. Biophysical Journal 89, 932-944.  Gonzalez, C., Rosenman, E., Bezanilla, F., Alvarez, O., and Latorre, R. (2000). Modulation of the Shaker K(+) channel gating kinetics by the S3-S4 linker. The Journal of General Physiology 115, 193-208.  Gonzalez, C., Rosenman, E., Bezanilla, F., Alvarez, O., and Latorre, R. (2001). Periodic perturbations in Shaker K+ channel gating kinetics by deletions in the S3-S4 linker. Proceedings of the National Academy of Sciences of the United States of America 98, 96179623.  Hackos, D.H., Chang, T.H., and Swartz, K.J. (2002). Scanning the intracellular S6 activation gate in the shaker K+ channel. The Journal of General Physiology 119, 521-532.  62  Halliwell, J.V., and Adams, P.R. (1982). Voltage-clamp analysis of muscarinic excitation in hippocampal neurons. Brain Res 250, 71-92.  Han, W., Bao, W., Wang, Z., and Nattel, S. (2002). Comparison of ion-channel subunit expression in canine cardiac Purkinje fibers and ventricular muscle. Circulation Research 91, 790-797.  Hauswirth, O., Noble, D., and Tsien, R.W. (1968). Adrenaline: mechanism of action on the pacemaker potential in cardiac Purkinje fibers. Science 162, 916-917.  Heginbotham, L., Abramson, T., and MacKinnon, R. (1992). A functional connection between the pores of distantly related ion channels as revealed by mutant K+ channels. Science 258, 1152-1155.  Heginbotham, L., Lu, Z., Abramson, T., and MacKinnon, R. (1994). Mutations in the K+ channel signature sequence. Biophys J 66, 1061-1067.  Heginbotham, L., and MacKinnon, R. (1993). Conduction properties of the cloned Shaker K+ channel. Biophysical Journal 65, 2089-2096.  Herrmann, S., Stieber, J., Stockl, G., Hofmann, F., and Ludwig, A. (2007). HCN4 provides a 'depolarization reserve' and is not required for heart rate acceleration in mice. The EMBO Journal 26, 4423-4432.  63  Hestrin, S. (1987). The properties and function of inward rectification in rod photoreceptors of the tiger salamander. The Journal of Physiology 390, 319-333.  Hille, B. (2001). Ion channels of excitable membranes.  Ho, W.K., Brown, H.F., and Noble, D. (1993). Internal K ions modulate the action of external cations on hyperpolarization-activated inward current in rabbit isolated sinoatrial node cells. Pflugers Arch 424, 308-314.  Ho, W.K., Brown, H.F., and Noble, D. (1994). High selectivity of the i(f) channel to Na+ and K+ in rabbit isolated sinoatrial node cells. Pflugers Arch 426, 68-74.  Hodgkin, A.L., and Huxley, A.F. (1952). A quantitative description of membrane current and its application to conduction and excitation in nerve. The Journal of Physiology 117, 500544.  Ishii, T.M., Takano, M., and Ohmori, H. (2001). Determinants of activation kinetics in mammalian hyperpolarization-activated cation channels. The Journal of Physiology 537, 93100.  Ishii, T.M., Takano, M., Xie, L.H., Noma, A., and Ohmori, H. (1999). Molecular characterization of the hyperpolarization-activated cation channel in rabbit heart sinoatrial node. The Journal of Biological Chemistry 274, 12835-12839.  64  Jackson, H.A., Marshall, C.R., and Accili, E.A. (2007). Evolution and structural diversification of hyperpolarization-activated cyclic nucleotide-gated channel genes. Physiological Genomics 29, 231-245.  Jiang, Y., Lee, A., Chen, J., Cadene, M., Chait, B.T., and MacKinnon, R. (2002a). Crystal structure and mechanism of a calcium-gated potassium channel. Nature 417, 515-522.  Jiang, Y., Lee, A., Chen, J., Cadene, M., Chait, B.T., and MacKinnon, R. (2002b). The open pore conformation of potassium channels. Nature 417, 523-526.  Jiang, Y., Lee, A., Chen, J., Ruta, V., Cadene, M., Chait, B.T., and MacKinnon, R. (2003a). X-ray structure of a voltage-dependent K+ channel. Nature 423, 33-41.  Jiang, Y., Ruta, V., Chen, J., Lee, A., and MacKinnon, R. (2003b). The principle of gating charge movement in a voltage-dependent K+ channel. Nature 423, 42-48.  Johnson, J.P., Jr., and Zagotta, W.N. (2005). The carboxyl-terminal region of cyclic nucleotide-modulated channels is a gating ring, not a permeation path. Proceedings of the National Academy of Sciences of the United States of America 102, 2742-2747.  Kaupp, U.B., and Seifert, R. (2001). Molecular diversity of pacemaker ion channels. Annu Rev Physiol 63, 235-257.  65  Kuo, A., Gulbis, J.M., Antcliff, J.F., Rahman, T., Lowe, E.D., Zimmer, J., Cuthbertson, J., Ashcroft, F.M., Ezaki, T., and Doyle, D.A. (2003). Crystal structure of the potassium channel KirBac1.1 in the closed state. Science 300, 1922-1926.  Labro, A.J., Raes, A.L., Bellens, I., Ottschytsch, N., and Snyders, D.J. (2003). Gating of shaker-type channels requires the flexibility of S6 caused by prolines. The Journal of Biological Chemistry 278, 50724-50731.  Larsson, H.P., Baker, O.S., Dhillon, D.S., and Isacoff, E.Y. (1996). Transmembrane movement of the shaker K+ channel S4. Neuron 16, 387-397.  Lesso, H., and Li, R.A. (2003). Helical secondary structure of the external S3-S4 linker of pacemaker (HCN) channels revealed by site-dependent perturbations of activation phenotype. The Journal of Biological Chemistry 278, 22290-22297.  Lipsius, S.L., and Bers, D.M. (2003). Cardiac pacemaking: I(f) vs. Ca(2+), is it really that simple? Journal of Molecular and Cellular Cardiology 35, 891-893.  Liu, Y., Holmgren, M., Jurman, M.E., and Yellen, G. (1997). Gated access to the pore of a voltage-dependent K+ channel. Neuron 19, 175-184.  Long, S.B., Campbell, E.B., and Mackinnon, R. (2005). Crystal structure of a mammalian voltage-dependent Shaker family K+ channel. Science 309, 897-903.  66  Long, S.B., Tao, X., Campbell, E.B., and MacKinnon, R. (2007). Atomic structure of a voltage-dependent K+ channel in a lipid membrane-like environment. Nature 450, 376-382.  Ludwig, A., Budde, T., Stieber, J., Moosmang, S., Wahl, C., Holthoff, K., Langebartels, A., Wotjak, C., Munsch, T., Zong, X., et al. (2003). Absence epilepsy and sinus dysrhythmia in mice lacking the pacemaker channel HCN2. The EMBO Journal 22, 216-224.  Ludwig, A., Zong, X., Jeglitsch, M., Hofmann, F., and Biel, M. (1998). A family of hyperpolarization-activated mammalian cation channels. Nature 393, 587-591.  Ludwig, A., Zong, X., Stieber, J., Hullin, R., Hofmann, F., and Biel, M. (1999). Two pacemaker channels from human heart with profoundly different activation kinetics. The EMBO Journal 18, 2323-2329.  Lyashchenko, A.K., and Tibbs, G.R. (2008). Ion binding in the open HCN pacemaker channel pore: fast mechanisms to shape "slow" channels. The Journal of General Physiology 131, 227-243.  Maccaferri, G., Mangoni, M., Lazzari, A., and DiFrancesco, D. (1993). Properties of the hyperpolarization-activated current in rat hippocampal CA1 pyramidal cells. Journal of Neurophysiology 69, 2129-2136.  67  Macri, V., and Accili, E.A. (2004). Structural elements of instantaneous and slow gating in hyperpolarization-activated cyclic nucleotide-gated channels. The Journal of Biological Chemistry 279, 16832-16846.  Macri, V., Nazzari, H., McDonald, E., and Accili, E.A. (2009). Alanine scanning of the S6 segment reveals a unique and cyclic AMP-sensitive association between the pore and voltage-dependent opening in HCN channels. The Journal of Biological Chemistry.  Macri, V., Proenza, C., Agranovich, E., Angoli, D., and Accili, E.A. (2002). Separable gating mechanisms in a Mammalian pacemaker channel. The Journal of Biological Chemistry 277, 35939-35946.  Magee, J.C. (1998). Dendritic hyperpolarization-activated currents modify the integrative properties of hippocampal CA1 pyramidal neurons. J Neurosci 18, 7613-7624.  Mannikko, R., Elinder, F., and Larsson, H.P. (2002). Voltage-sensing mechanism is conserved among ion channels gated by opposite voltages. Nature 419, 837-841.  Maruoka, F., Nakashima, Y., Takano, M., Ono, K., and Noma, A. (1994). Cation-dependent gating of the hyperpolarization-activated cation current in the rabbit sino-atrial node cells. J Physiol 477 ( Pt 3), 423-435.  68  Mathur, R., Zheng, J., Yan, Y., and Sigworth, F.J. (1997). Role of the S3-S4 linker in Shaker potassium channel activation. The Journal of General Physiology 109, 191-199.  Mayer, M.L., and Westbrook, G.L. (1983). A voltage-clamp analysis of inward (anomalous) rectification in mouse spinal sensory ganglion neurones. The Journal of Physiology 340, 1945.  McCormick, D.A., and Pape, H.C. (1990). Properties of a hyperpolarization-activated cation current and its role in rhythmic oscillation in thalamic relay neurones. The Journal of Physiology 431, 291-318.  Milanesi, R., Baruscotti, M., Gnecchi-Ruscone, T., and DiFrancesco, D. (2006). Familial sinus bradycardia associated with a mutation in the cardiac pacemaker channel. N Engl J Med 354, 151-157.  Mistrik, P., Mader, R., Michalakis, S., Weidinger, M., Pfeifer, A., and Biel, M. (2005). The murine HCN3 gene encodes a hyperpolarization-activated cation channel with slow kinetics and unique response to cyclic nucleotides. The Journal of Biological Chemistry 280, 2705627061.  Moosmang, S., Stieber, J., Zong, X., Biel, M., Hofmann, F., and Ludwig, A. (2001). Cellular expression and functional characterization of four hyperpolarization-activated pacemaker channels in cardiac and neuronal tissues. Eur J Biochem 268, 1646-1652.  69  Morais-Cabral, J.H., Zhou, Y., and MacKinnon, R. (2001). Energetic optimization of ion conduction rate by the K+ selectivity filter. Nature 414, 37-42.  Moroni, A., Barbuti, A., Altomare, C., Viscomi, C., Morgan, J., Baruscotti, M., and DiFrancesco, D. (2000). Kinetic and ionic properties of the human HCN2 pacemaker channel. Pflugers Arch 439, 618-626.  Moroni, A., Gorza, L., Beltrame, M., Gravante, B., Vaccari, T., Bianchi, M.E., Altomare, C., Longhi, R., Heurteaux, C., Vitadello, M., et al. (2001). Hyperpolarization-activated cyclic nucleotide-gated channel 1 is a molecular determinant of the cardiac pacemaker current I(f). The Journal of Biological Chemistry 276, 29233-29241.  Noble, D., and Tsien, R.W. (1968). The kinetics and rectifier properties of the slow potassium current in cardiac Purkinje fibres. The Journal of Physiology 195, 185-214.  Nof, E., Luria, D., Brass, D., Marek, D., Lahat, H., Reznik-Wolf, H., Pras, E., Dascal, N., Eldar, M., and Glikson, M. (2007). Point mutation in the HCN4 cardiac ion channel pore affecting synthesis, trafficking, and functional expression is associated with familial asymptomatic sinus bradycardia. Circulation 116, 463-470.  Noma, A., and Irisawa, H. (1976). Membrane currents in the rabbit sinoatrial node cell as studied by the double microelectrode method. Pflugers Arch 364, 45-52.  70  Paoletti, P., Young, E.C., and Siegelbaum, S.A. (1999). C-Linker of cyclic nucleotide-gated channels controls coupling of ligand binding to channel gating. The Journal of General Physiology 113, 17-34.  Pape, H.C. (1996). Queer current and pacemaker: the hyperpolarization-activated cation current in neurons. Annual Review of Physiology 58, 299-327.  Paterson, D.J. (1996). Role of potassium in the regulation of systemic physiological function during exercise. Acta physiologica Scandinavica 156, 287-294.  Pathak, M.M., Yarov-Yarovoy, V., Agarwal, G., Roux, B., Barth, P., Kohout, S., Tombola, F., and Isacoff, E.Y. (2007). Closing in on the resting state of the Shaker K(+) channel. Neuron 56, 124-140.  Posson, D.J., Ge, P., Miller, C., Bezanilla, F., and Selvin, P.R. (2005). Small vertical movement of a K+ channel voltage sensor measured with luminescence energy transfer. Nature 436, 848-851.  Proenza, C., Angoli, D., Agranovich, E., Macri, V., and Accili, E.A. (2002a). Pacemaker channels produce an instantaneous current. The Journal of Biological Chemistry 277, 51015109.  71  Proenza, C., Tran, N., Angoli, D., Zahynacz, K., Balcar, P., and Accili, E.A. (2002b). Different roles for the cyclic nucleotide binding domain and amino terminus in assembly and expression of hyperpolarization-activated, cyclic nucleotide-gated channels. The Journal of Biological Chemistry 277, 29634-29642.  Proenza, C., and Yellen, G. (2006). Distinct populations of HCN pacemaker channels produce voltage-dependent and voltage-independent currents. The Journal of General Physiology 127, 183-190.  Prole, D.L., and Yellen, G. (2006). Reversal of HCN channel voltage dependence via bridging of the S4-S5 linker and Post-S6. The Journal of General Physiology 128, 273-282.  Robinson, R.B., and Siegelbaum, S.A. (2003). Hyperpolarization-activated cation currents: from molecules to physiological function. Annual Review of Physiology 65, 453-480.  Robinson, R.B., Yu, H., Chang, F., and Cohen, I.S. (1997). Developmental change in the voltage-dependence of the pacemaker current, if, in rat ventricle cells. Pflugers Arch 433, 533-535.  Roncaglia, P., Mistrik, P., and Torre, V. (2002). Pore topology of the hyperpolarizationactivated cyclic nucleotide-gated channel from sea urchin sperm. Biophysical Journal 83, 1953-1964.  72  Rothberg, B.S., Shin, K.S., Phale, P.S., and Yellen, G. (2002). Voltage-controlled gating at the intracellular entrance to a hyperpolarization-activated cation channel. The Journal of General Physiology 119, 83-91.  Rothberg, B.S., Shin, K.S., and Yellen, G. (2003). Movements near the gate of a hyperpolarization-activated cation channel. The Journal of General Physiology 122, 501-510.  Ruta, V., Chen, J., and MacKinnon, R. (2005). Calibrated measurement of gating-charge arginine displacement in the KvAP voltage-dependent K+ channel. Cell 123, 463-475.  Santoro, B., Chen, S., Luthi, A., Pavlidis, P., Shumyatsky, G.P., Tibbs, G.R., and Siegelbaum, S.A. (2000). Molecular and functional heterogeneity of hyperpolarizationactivated pacemaker channels in the mouse CNS. J Neurosci 20, 5264-5275.  Santoro, B., Grant, S.G., Bartsch, D., and Kandel, E.R. (1997). Interactive cloning with the SH3 domain of N-src identifies a new brain specific ion channel protein, with homology to eag and cyclic nucleotide-gated channels. Proceedings of the National Academy of Sciences of the United States of America 94, 14815-14820.  Santoro, B., Liu, D.T., Yao, H., Bartsch, D., Kandel, E.R., Siegelbaum, S.A., and Tibbs, G.R. (1998). Identification of a gene encoding a hyperpolarization-activated pacemaker channel of brain. Cell 93, 717-729.  73  Schachtman, D.P., Schroeder, J.I., Lucas, W.J., Anderson, J.A., and Gaber, R.F. (1992). Expression of an inward-rectifying potassium channel by the Arabidopsis KAT1 cDNA. Science 258, 1654-1658.  Schulze-Bahr, E., Neu, A., Friederich, P., Kaupp, U.B., Breithardt, G., Pongs, O., and Isbrandt, D. (2003). Pacemaker channel dysfunction in a patient with sinus node disease. J Clin Invest 111, 1537-1545.  Seifert, R., Scholten, A., Gauss, R., Mincheva, A., Lichter, P., and Kaupp, U.B. (1999). Molecular characterization of a slowly gating human hyperpolarization-activated channel predominantly expressed in thalamus, heart, and testis. Proceedings of the National Academy of Sciences of the United States of America 96, 9391-9396.  Shealy, R.T., Murphy, A.D., Ramarathnam, R., Jakobsson, E., and Subramaniam, S. (2003). Sequence-function analysis of the K+-selective family of ion channels using a comprehensive alignment and the KcsA channel structure. Biophysical Journal 84, 29292942.  Shi, W., Wymore, R., Yu, H., Wu, J., Wymore, R.T., Pan, Z., Robinson, R.B., Dixon, J.E., McKinnon, D., and Cohen, I.S. (1999). Distribution and prevalence of hyperpolarizationactivated cation channel (HCN) mRNA expression in cardiac tissues. Circ Res 85, e1-6.  74  Shin, K.S., Rothberg, B.S., and Yellen, G. (2001). Blocker state dependence and trapping in hyperpolarization-activated cation channels: evidence for an intracellular activation gate. The Journal of General Physiology 117, 91-101.  Solomon, J.S., and Nerbonne, J.M. (1993). Hyperpolarization-activated currents in isolated superior colliculus-projecting neurons from rat visual cortex. The Journal of Physiology 462, 393-420.  Stieber, J., Herrmann, S., Feil, S., Loster, J., Feil, R., Biel, M., Hofmann, F., and Ludwig, A. (2003a). The hyperpolarization-activated channel HCN4 is required for the generation of pacemaker action potentials in the embryonic heart. Proceedings of the National Academy of Sciences of the United States of America 100, 15235-15240.  Stieber, J., Stockl, G., Herrmann, S., Hassfurth, B., and Hofmann, F. (2005). Functional expression of the human HCN3 channel. The Journal of Biological Chemistry 280, 3463534643.  Stieber, J., Thomer, A., Much, B., Schneider, A., Biel, M., and Hofmann, F. (2003b). Molecular basis for the different activation kinetics of the pacemaker channels HCN2 and HCN4. The Journal of Biological Chemistry 278, 33672-33680.  Tombola, F., Pathak, M.M., Gorostiza, P., and Isacoff, E.Y. (2007). The twisted ionpermeation pathway of a resting voltage-sensing domain. Nature 445, 546-549.  75  Tristani-Firouzi, M., Chen, J., and Sanguinetti, M.C. (2002). Interactions between S4-S5 linker and S6 transmembrane domain modulate gating of HERG K+ channels. The Journal of Biological Chemistry 277, 18994-19000.  Tsang, S.Y., Lesso, H., and Li, R.A. (2004). Dissecting the structural and functional roles of the S3-S4 linker of pacemaker (hyperpolarization-activated cyclic nucleotide-modulated) channels by systematic length alterations. The Journal of Biological Chemistry 279, 4375243759.  Ueda, K., Nakamura, K., Hayashi, T., Inagaki, N., Takahashi, M., Arimura, T., Morita, H., Higashiuesato, Y., Hirano, Y., Yasunami, M., et al. (2004). Functional characterization of a trafficking-defective HCN4 mutation, D553N, associated with cardiac arrhythmia. The Journal of Biological Chemistry 279, 27194-27198.  Vaca, L., Stieber, J., Zong, X., Ludwig, A., Hofmann, F., and Biel, M. (2000). Mutations in the S4 domain of a pacemaker channel alter its voltage dependence. FEBS letters 479, 35-40.  Vaccari, T., Moroni, A., Rocchi, M., Gorza, L., Bianchi, M.E., Beltrame, M., and DiFrancesco, D. (1999). The human gene coding for HCN2, a pacemaker channel of the heart. Biochimica et Biophysica Acta 1446, 419-425.  76  Vemana, S., Pandey, S., and Larsson, H.P. (2004). S4 movement in a mammalian HCN channel. The Journal of General Physiology 123, 21-32.  Vemana, S., Pandey, S., and Larsson, H.P. (2008). Intracellular Mg2+ is a voltage-dependent pore blocker of HCN channels. American Journal of Physiology 295, C557-565.  Vinogradova, T.M., Bogdanov, K.Y., and Lakatta, E.G. (2002). beta-Adrenergic stimulation modulates ryanodine receptor Ca(2+) release during diastolic depolarization to accelerate pacemaker activity in rabbit sinoatrial nodal cells. Circulation Research 90, 73-79.  Viscomi, C., Altomare, C., Bucchi, A., Camatini, E., Baruscotti, M., Moroni, A., and DiFrancesco, D. (2001). C terminus-mediated control of voltage and cAMP gating of hyperpolarization-activated cyclic nucleotide-gated channels. The Journal of Biological Chemistry 276, 29930-29934.  Wahl-Schott, C., and Biel, M. (2009). HCN channels: structure, cellular regulation and physiological function. Cell Mol Life Sci 66, 470-494.  Wainger, B.J., DeGennaro, M., Santoro, B., Siegelbaum, S.A., and Tibbs, G.R. (2001). Molecular mechanism of cAMP modulation of HCN pacemaker channels. Nature 411, 805810.  77  Wang, J., Chen, S., Nolan, M.F., and Siegelbaum, S.A. (2002). Activity-dependent regulation of HCN pacemaker channels by cyclic AMP: signaling through dynamic allosteric coupling. Neuron 36, 451-461.  Wang, J., Chen, S., and Siegelbaum, S.A. (2001). Regulation of hyperpolarization-activated HCN channel gating and cAMP modulation due to interactions of COOH terminus and core transmembrane regions. The Journal of General Physiology 118, 237-250.  Whitaker, G.M., Angoli, D., Nazzari, H., Shigemoto, R., and Accili, E.A. (2007). HCN2 and HCN4 isoforms self-assemble and co-assemble with equal preference to form functional pacemaker channels. The Journal of Biological Chemistry 282, 22900-22909.  Wollmuth, L.P. (1995). Multiple ion binding sites in Ih channels of rod photoreceptors from tiger salamanders. Pflugers Arch 430, 34-43.  Wollmuth, L.P., and Hille, B. (1992). Ionic selectivity of Ih channels of rod photoreceptors in tiger salamanders. The Journal of General Physiology 100, 749-765.  Woodhull, A.M. (1973). Ionic blockage of sodium channels in nerve. The Journal of General Physiology 61, 687-708.  78  Wu, J.Y., Vereecke, J., Carmeliet, E., and Lipsius, S.L. (1991). Ionic currents activated during hyperpolarization of single right atrial myocytes from cat heart. Circulation Research 68, 1059-1069.  Xue, T., and Li, R.A. (2002). An external determinant in the S5-P linker of the pacemaker (HCN) channel identified by sulfhydryl modification. The Journal of Biological Chemistry 277, 46233-46242.  Xue, T., Marban, E., and Li, R.A. (2002). Dominant-negative suppression of HCN1- and HCN2-encoded pacemaker currents by an engineered HCN1 construct: insights into structure-function relationships and multimerization. Circulation Research 90, 1267-1273.  Yanagihara, K., and Irisawa, H. (1980). Inward current activated during hyperpolarization in the rabbit sinoatrial node cell. Pflugers Arch 385, 11-19.  Yellen, G. (2002). The voltage-gated potassium channels and their relatives. Nature 419, 3542.  Yifrach, O., and MacKinnon, R. (2002). Energetics of pore opening in a voltage-gated K(+) channel. Cell 111, 231-239.  Yool, A.J., and Schwarz, T.L. (1991). Alteration of ionic selectivity of a K+ channel by mutation of the H5 region. Nature 349, 700-704.  79  Zagotta, W.N. (2006). Membrane biology: permutations of permeability. Nature 440, 427429.  Zagotta, W.N., Olivier, N.B., Black, K.D., Young, E.C., Olson, R., and Gouaux, E. (2003). Structural basis for modulation and agonist specificity of HCN pacemaker channels. Nature 425, 200-205.  Zaza, A., Robinson, R.B., and DiFrancesco, D. (1996). Basal responses of the L-type Ca2+ and hyperpolarization-activated currents to autonomic agonists in the rabbit sino-atrial node. The Journal of Physiology 491 ( Pt 2), 347-355.  Zhou, L., and Siegelbaum, S.A. (2007). Gating of HCN channels by cyclic nucleotides: residue contacts that underlie ligand binding, selectivity, and efficacy. Structure 15, 655-670.  Zhou, M., and MacKinnon, R. (2004). A mutant KcsA K(+) channel with altered conduction properties and selectivity filter ion distribution. J Mol Biol 338, 839-846.  Zhou, Y., and MacKinnon, R. (2003). The occupancy of ions in the K+ selectivity filter: charge balance and coupling of ion binding to a protein conformational change underlie high conduction rates. J Mol Biol 333, 965-975.  80  Zhou, Y., Morais-Cabral, J.H., Kaufman, A., and MacKinnon, R. (2001). Chemistry of ion coordination and hydration revealed by a K+ channel-Fab complex at 2.0 A resolution. Nature 414, 43-48.  81  2. Alanine scanning of the S6 segment reveals a unique and cyclic AMPsensitive association between the pore and voltage-dependent opening in HCN channels1  2.1 Introduction Hyperpolarization-activated Cyclic Nucleotide-modulated (HCN) channels are similar in structure and function to Shaker K+ channels (Gauss et al., 1998; Ludwig et al., 1998; Santoro et al., 1998). As in Shaker, HCN channels are comprised of 4 subunits which each consist of six predicted membrane-spanning segments (S1-S6). The S1-S4 segments form the voltage-sensing domain, and the S5 and S6 segments, the pore-forming domain. The S4 segment in both channels contains positive charges that move similarly in response to changes in membrane voltage (Bell et al., 2004; Mannikko et al., 2002; Vemana et al., 2004), to then alter the pore structure at the intracellular side of the S6 segment; this region functions as a voltage-controlled gate to cation flow (Giorgetti et al., 2005; Macri et al., 2002; Rothberg et al., 2003; Shin et al., 2001). Despite these similarities, HCN channels are opened by hyperpolarization of the membrane potential, whereas Shaker channels open in response to depolarization. Thus, the electromechanical coupling between the voltage sensor and the gate is reversed in these two channels.  A key determinant of this coupling is the intrinsic stability of the closed and open conformations of the pore. In Shaker channels, it has been proposed that the pore is  1  A version of this chapter has been published. Macri, V, Nazzari, H, McDonald, E, Accili, EA. (2009) Alanine scanning of the S6 segment reveals a unique and cyclic AMP-sensitive association between the pore and voltage-dependent opening in HCN channels. Journal of Biological Chemistry, 284: 15659-67.  82  intrinsically most stable when closed and that the voltage sensor works to open the pore during depolarization (Hackos et al., 2002; Yifrach and MacKinnon, 2002). Results from an alanine/valine scan of residues across the entire Shaker pore, by single point mutation, showed that most mutations made the channel easier to open and steepened the channel‟s response to changes in voltage. It was argued that because most mutations likely destabilize protein packing, the closed conformation must be the stable state; this is consistent with the observed crystal structures of Shaker-related channels KcsA and MthK, in the closed and open states respectively, wherein more optimally and tightly packed helices were seen in the closed conformation (Doyle et al., 1998; Jiang et al., 2002a, b).  Because of presumed shared architecture of the gate between HCN and Shaker channels, HCN channels might also be most stable when closed and thus the voltage sensor would work to open the pore upon hyperpolarization. To test this hypothesis, we performed an alanine/valine scan of the C-terminal 22 amino acids of the S6 segment in HCN2, used as a prototype, and examined pore energetics as described previously in Shaker (Yifrach and MacKinnon, 2002). The choice of this region for mutation was based on: 1) in Shaker, the corresponding region harbors one of two clusters of gating-sensitive residues; and 2) it contains the voltage-controlled gate. Surprisingly, the effects of the mutations on channel opening and on the steepness of the channel‟s response to voltage are mixed and smaller than those in Shaker. These findings imply that, in HCN2, the stability of the open and closed pore are similar, the interactions between the pore and voltage-sensor, both structural and functional, are weaker than in Shaker, and that the voltage sensor must apply force to the pore to close it. Thus, Shaker is closed and HCN2 is open in the absence of input from the  83  voltage sensor. Moreover, cyclic AMP binding to the HCN2 channel heightens the effects of the mutations, indicating stronger interactions between the pore and voltage-sensor, and tips the energetic balance towards a more stable open state.  2.2 Experimental procedures 2.2.1 Mutagenesis Single-point alanine/valine mutant HCN2 channels were constructed in one of two ways. First, some mutants were constructed by overlapping PCR mutagenesis using a mouse HCN2 template in pcDNA3.1, as previously described (14). For remaining mutants, base pairs 1172-2216 of the mouse HCN2 template were amplified by PCR primers containing distal EcoRI and BamHI sites and subcloned into pBluescript. Quickchange (Stratagene, La Jolla, CA) was then used to generate mutations in this amplified fragment. Next, BlpI and AgeI digested fragments were inserted into the mouse HCN2 template. All mutations were confirmed via DNA sequencing (NAPS facility, University of British Columbia).  2.2.2 Tissue culture and expression of HCN2 constructs Chinese hamster ovary (CHO-K1) cells (ATCC, Manassas, VA) were maintained in Hams F12 media supplemented with antibiotics and 10% FBS (Gibco, Burlington, Ontario), and maintained at 37oC with 5% CO2. Cells were plated onto glass cover slips. Two days after splitting, mammalian expression vectors encoding wild type or mutant HCN2 channels (2 g per 35 mm dish), and a green fluorescent protein (GFP) reporter plasmid (0.3 g per dish), were transiently co-transfected into the cells using the FuGene6 transfection reagent (Roche Biochemical, Indianapolis, IN).  84  2.2.3 Whole-cell patch clamp electrophysiology Cells expressing GFP were chosen for whole-cell patch clamp recordings 24-48 hours post transfection. The pipette solution contained (in mM): 130 K-Asp, 10 NaCl, 0.5 MgCl2, 1 EGTA, and 5 HEPES with pH adjusted to 7.4 using KOH. For experiments at saturating levels of cAMP, 2 mM cAMP (Na salt) was added to the pipette solution. Extracellular recording solution contained (in mM): 135 KCl, 5 NaCl, 1.8 CaCl2, 0.5 MgCl2, and 5 HEPES with pH adjusted to 7.4 using KOH. Whole-cell currents were recorded using an Axopatch 200B amplifier and Clampex software (Axon Instruments, Union City, CA) at room temperature. Patch clamp pipettes were pulled from borosilicate glass and fire polished before use (pipette R= 2.5-4.5 M).  2.2.4 Data analysis Data were filtered at 2 kHz and were analyzed using Clampfit (Axon Instruments, Union City, CA), Origin (Microcal, Northhampton, MA) and Excel (Microsoft, Seattle, WA) software. If activation curves were determined from tail currents at a 2 s pulse to -35 mV following 3 to 15 s test pulses ranging from -150 mV to -10 mV, in 20 mV steps. Single tail current test pulses were followed by a 500 ms pulse to +5 mV to ensure complete channel deactivation. The resting current was allowed to return to its baseline value before subsequent voltage pulses. If activation curves were determined by plotting normalized tail current amplitudes versus test voltage and fitting these with a single order Boltzmann function,  f(V) = Imax/(1 + e(V½-V)/k)  (Equation 2.1)  85  to determine the midpoint of activation (V½) and slope factor (k). The effective charge (Z) was calculated using the equation Z = RT/kF, where T = 295K and R and F have their usual thermodynamic meanings. Changes in free energy between open and closed states were given by -ZFV½. The perturbation in free energy produced by introduction of the point mutations (∆(ZFV½)) was given by –F(ZmutV½mut – ZwtV½wt). The standard errors for ∆(ZFV½) were calculated using ∆(ZFV½) = (2ZFV½,wt + 2ZFV½, mut)1/2.  Differences in values for V½, Z and ZFV½ between the wild type channel and mutant channels were determined independently using an unpaired t-test (P<0.05 was considered significant).  2.2.5 Western blot analysis Each sample was derived from cells on 35mm plates that had been lysed in 100 L of lysis buffer containing 50mM Tris at pH 8.0, 1% NP40, 150mM NaCl, 1mM EDTA, 1mM PMSF, 2mM each of Na3VO4 and NaF, and 10g/mL each of aprotinin, pepstatin, and leupeptin. Samples were left on ice for 30 minutes, during which time they were vortexed every 5 minutes for ~5 s. After centrifugation to remove cell debris (25,000g, 25 minutes), protein concentration of the supernatant was determined by Bradford assay. 20 µg samples of supernatant  were  fractionated  by  sodium  dodecyl  sulphate-polyacrylamide  gel  electrophoresis (SDS-PAGE, 8%) and electroblotted to polyvinylidene fluoride (PVDF) membrane (Bio-Rad, Mississauga, ON). Blots were washed three times in TBST (50 mM Tris, pH 7.4, 150 mM NaCl, 0.1% Tween 20) and then blocked with 5% non-fat dry milk (Bio-Rad) in TBST for 1 hour at room temperature. Blots were then incubated with a rabbit  86  polyclonal antibody specific to the C-terminus of HCN2 (Affinity Bioreagents, Golden, CO), at a dilution of 1:500 in TBST with 5% non-fat dry milk for 2.5 hours at room temperature. Blots were washed in TBST for 10 minutes, three times, and then incubated with horseradish peroxidase conjugated to goat anti-rabbit 1:3000 dilution in 5% non-fat dry milk with TBST for 1 hour at room temperature; they were subsequently washed 3 times in TBST. Signals were obtained with ECL Western Blotting Detection Reagents (GE Healthcare, Baie d‟Urfe, QC). Protein loading was controlled by probing all Western blots with goat anti-GAPDH antibody (Santa Cruz Biotechnology, Santa Cruz, CA).  2.3 Results 2.3.1 Alanine/valine scanning of the distal S6 reveals small changes in perturbation energy To determine the most stable conformation of the channel, we performed a single-point alanine/valine scan of the C-terminal 22 amino acids of the S6 segment in HCN2 (I422D443) and examined channel opening, as described previously in Shaker (Yifrach and MacKinnon, 2002). We hypothesized that, as for Shaker channels, the values for V½ would be shifted in the positive direction and Z would be larger, due to disruption of a more stable closed state by introduced alanine or valine residues. This assumes that the closed conformation of the channel is at an energetic minimum, and that all of the mutations within the S6 will result in positive perturbation energies. The S6 sites involved in positive perturbations promote a more stable closed conformation whereas those that produce negative perturbations promote a more stable open conformation. The relative numbers that shift in the two directions give an approximation of the relative stability of the open versus the closed conformations e.g. a larger number of negative perturbation energies would  87  suggest a more stable open state, an equal number of positive and negative perturbation energies would suggest that the stabilities of the open and closed conformations are about equal. Finally, this assumes that each residue contributes equally to stability.  Wild type and mutant channels were expressed independently in CHO cells from which If was recorded using the whole-cell patch clamp approach. If activation curves were determined by plotting normalized tail current amplitudes versus test voltage and fitting these with Equation 2.1 (Experimental Procedures). From this fit, values for V½ and Z were determined to thereby allow calculation of perturbation energies (Table 2.1A,B). Gating parameters and perturbation energies of wildtype channels were compared to those of the mutant channels using an unpaired t-test. Eighteen of 22 single-point mutations expressed measurable levels of If from which activation curves could be derived (Fig. 2.1A,B). Levels of If for G424A, A425V, T426A, and Y428A were not detectable. More mutants had a V½ value that were either significantly more negative (5/18) or unchanged (10/18) from that of wild type, than those which were more positive (3/18) (Fig. 2.1C, upper). With one exception, all Z values of mutants were unchanged from that of wild type (Fig. 2.1C, lower). Finally, with the exception of three values, the free energies of mutants were unchanged from that of wild type (Fig. 2.1D). The mix of positive and negative shifts in V½, and lack of change in free energies in the mutant channels suggest that, contrary to our hypothesis, the stabilities of the open and closed conformations are similar. These data are in accordance with recent findings from an alanine/valine scan of the S6 in HCN2 expressed in Xenopus oocytes, which showed that most mutations shifted the opening of the channel to more  88  negative potentials or had no effect; however, the energetic repercussions of these changes on gating were not explored (Cheng et al., 2007).  2.3.2 Cyclic AMP shifts the balance of perturbation energies of the S6 mutations toward negative values Cyclic AMP stabilizes the open conformation of HCN channels by removing a tonic inhibitory action of the cyclic nucleotide-binding domain (CNBD), located in the C-terminus, on pore opening (Barbuti et al., 1999; Chen et al., 2007; Craven and Zagotta, 2004; DiFrancesco, 1999; DiFrancesco and Tortora, 1991; Wainger et al., 2001). Inhibition by the CNBD occurs by a coupled interaction with the C-linker, a structure that connects the CNBD to the S6 helices, which is thought to apply a force on these helices to inhibit pore opening (Craven and Zagotta, 2004; Zhou and Siegelbaum, 2007). Cyclic AMP binding reverses the coupled interaction which then alleviates inhibition of pore opening thereby promoting a more stable open state. Given a more stable open conformation upon cAMP binding, we hypothesized that, in saturating levels of this cyclic nucleotide, the S6 mutations would produce more dramatic effects on V½ and Z, and a shift in perturbation energies towards more negative values.  To test this hypothesis, identical experiments were conducted with all 22 mutant channels and the wild type channel at saturating levels of cAMP (2 mM). All but one mutant (G424A) expressed measurable levels of If from which activation curves could be determined (Fig. 2.2A, B). For the wild type HCN2 channel, V½ was shifted +10.1 mV and Z was decreased  89  Figure 2.1 HCN2 channels are most stable in the open state A. Current traces recorded from CHO cells expressing wild type and three representative S6 alanine mutant HCN2 channels. Currents were elicited by test voltage pulses ranging from 150 mV to -10 mV, in 20 mV steps from a holding potential of -35 mV. The tail currents were elicited at -35 mV. B. Representative If activation curves determined by plotting tail current amplitudes which were normalized to their maximum value (I/Imax), versus test voltages (HCN2, squares; Q440A, circles; C427A, upright triangles; L438A, inverted triangles). The curved lines represent fitting by Equation 2.1 (see Experimental Procedures). C. Bar graphs depicting the changes in V½ (upper) and Z (lower) values for each mutant channel relative to wild type. D. Bar graph depicting change in perturbation of free energy, ∆(ZFV½), for each mutant channel relative to the wild type channel. Four mutant channels did not yield measurable levels of If (solid line through numbered residue, X axis).  90  1 nA 3s  C427A  0.8  0.6  L438A  1 nA 3s  voltage (mV) I422A V423A G424A A425V T426A C427A Y428A A429V M430A F431A I432A G433A H434A A435V T436A A437V L438A I439A Q440A S441A L442A D443A  HCN2  1 nA  Q440A  C427A  0.4 L438A  ZFV1/2  A. C. 20  10  V1/2  0.5 nA 3s  Z  3s  -1  B.  1.0  D.  HCN2  2  -2  0.2 -4  0.0  -6  I422A V423A G424A A425V T426A C427A Y428A A429V M430A F431A I432A G433A H434A A435V T436A A437V L438A I439A Q440A S441A L442A D443A  normalized current  Basal cAMP  Q440A 30 *  * *  0  -10 * **  -20  * *  -30  4  3  2  1  0  *  -2  8  6  4  * *  0  *  -150 -130 -110 -90 -70 -50 -30 -10 10  -8  Figure 2.1  91  Table 2.1 A, B The effects of S6 pore mutations on voltage-dependent gating at basal (A) and saturating (2 mM; B) levels of cAMP The V½ and Z values are from fits of activation curves with Equation 2.1 for wild type and mutant channels (Table 2.1A, basal cAMP; Table 2.lB, 2 mM cAMP). The free energy of the open or closed state is shown as -ZFV½. The difference in free energy between each mutant channel relative to wild type is indicated by Δ(ZFV½). Data are presented as the mean ± sem. Asterisks represent significant differences from wild type.  92  Table Table2.1A 1A basal cAMP Z  -ZFV1/2 (kcal/mol)  Δ ZFV1/2 (kcal/mol)  2.24 ± 0.18 1.96 ± 0.09 2.93 ± 0.40  -5.47 ± 0.36 -4.92 ± 0.15 -6.96 ± 1.03  0.54 ± 0.39 -1.49 ± 1.09  2.85 ± 0.28  -6.20 ± 0.63  -0.73 ± 0.73  1.93 ± 0.12 2.76 ± 0.22 2.15 ± 0.18 2.26 ± 0.39 1.66 ± 0.11* 2.61 ± 0.13 2.29 ± 0.35 2.40 ± 0.24 2.32 ± 0.21 2.09 ± 0.07 2.00 ± 0.17 2.25 ± 0.34 1.87 ± 0.11 1.99 ± 0.17  -5.05 ± 0.18 -6.34 ± 0.51 -5.81 ± 0.57 -5.37 ± 0.96 -4.30 ± 0.25* -6.80 ± 0.44* -5.77 ± 0.74 -6.00 ± 0.56 -6.28 ± 0.54 -5.67 ± 0.22 -4.12 ± 0.43* -5.55 ± 0.76 -4.67 ± 0.27 -4.89 ± 0.47  0.42 ± 041 -0.87 ± 0.62 -0.34 ± 0.68 0.11 ± 1.03 1.17 ± 0.44 -1.33 ± 0.57 -0.29 ± 0.82 -0.53 ± 0.67 -0.81 ± 0.65 -0.19 ± 0.42 1.35 ± 0.41 -0.08 ± 0.85 0.79 ± 0.46 0.57 ± 0.61  -105.6 ± 2.2  2.20 ± 0.19  -5.27 ± 0.48  0.19 ± 0.60  n  V1/2 (mV)  2 mM cAMP Z  -ZFV1/2 (kcal/mol)  Δ ZFV1/2 (kcal/mol)  8 6 5  -98.8 ± 2.6 -94.9 ± 2.8 -93.0 ± 2.5 no expression -117.0 ± 5.2* -117.1 ± 3.2* -85.1 ± 1.5* -104.9 ± 3.6 -113.7 ± 3.7* -90.5 ± 2.8* -116.3 ± 3.1* -107.1 ± 5.2 -102.7 ± 2.5 -99.4 ± 2.8 -100.7 ± 4.2 -105.2 ± 2.2 -113.3 ± 3.5* -110.4 ± 4.2* -78.9 ± 2.4* -102.2 ± 2.6 -87.5 ± 2.9* -91.8 ± 2.2* -86.4 ± 3.6*  1.84 ± 0.13 1.63 ± 0.11 2.02 ± 0.28  -4.10 ± 0.28 -3.52 ± 0.32 -4.22 ± 0.55  0.58 ± 0.43 -0.12 ± 0.62  2.30 ± 0.64 2.32 ± 0.46 2.72 ± 0.55* 2.80 ± 0.64* 2.92 ± 0.42* 2.87 ± 0.22* 2.06 ± 0.31 2.00 ± 0.17 1.63 ± 0.09 2.49 ± 0.22* 2.91 ± 0.24* 1.69 ± 0.17 2.17 ± 0.32 1.52 ± 0.11* 1.60 ± 0.16 1.73 ± 0.14 1.90 ± 0.11 1.70 ± 0.09 1.79 ± 0.18  -6.03 ± 1.52* -6.11 ± 1.15* -5.18 ± 0.97 -6.81 ± 1.75* -7.43 ± 0.92* -5.89 ± 0.42* -5.45 ± 0.66* -4.77 ± 0.38 -3.77 ± 0.22 -5.62 ± 0.60* -6.61 ± 0.55* -4.00 ± 0.36 -5.43 ± 0.64* -3.76 ± 0.26 -2.83 ± 0.23* -3.95 ± 0.24 -3.74 ± 0.19 -3.53 ± 0.22 -3.44 ± 0.24  -1.93 ± 1.55 -2.01 ± 1.19 -1.07 ± 1.01 -2.70 ± 1.77 -3.32 ± 0.96 -1.79 ± 0.51 -1.34 ± 0.72 -0.67 ± 0.47 0.32 ± 0.36 -1.52 ± 0.67 -2.51 ± 0.62 0.09 ± 0.46 -1.48 ± 0.70 0.33 ± 0.38 1.27 ± 0.36 0.15 ± 0.37 0.36 ± 0.34 0.56 ± 0.36 0.65 ± 0.37  HCN2 channel  n  V1/2 (mV)  wild type I422A V423A G424A A425V T426A C427A Y428A A429V M430A F431A I432A G433A H434A A435V T436A A437V L438A I439A Q440A S441A L442A  9 6 6  4 4 5 5 8 5 5 5 6 8 5 8 6 7  -108.9 ± 1.8 -111.4 ± 3.4 -104.7 ± 2.9 no expression no current no current -96.4 ± 2.4* no current -116.4 ± 5.5 -101.0 ± 0.84* -119.7 ± 2.0* -104.5 ± 1.3 -115.0 ± 2.1* -115.1 ± 2.1* -113.4 ± 4.2 -110.7 ± 1.5 -120.2 ± 3.5* -120.1 ± 2.1* -90.4 ± 4.9* -111.0 ± 2.7 -110.4 ± 2.2 -108.5 ± 1.9  D443A  6  HCN2 channel wild type I422A V423A G424A A425V T426A C427A Y428A A429V M430A F431A I432A G433A H434A A435V T436A A437V L438A I439A Q440A S441A L442A D443A  5  Table 2.1B 1B Table  3 4 5 5 4 6 6 6 9 4 4 5 6 7 4 10 7 7 6  93  0.4 compared to the values determined at basal cAMP (Table 2.1A,B). The majority of V½ values in the mutant channels were more negative (6/21) or unchanged (9/21) compared to wild type, whereas fewer values were more positive (6/21) (Fig. 2.2C, upper). The majority of Z values were larger (6/21) or unchanged (14/21) compared to wild type, whereas only one value was smaller (Fig. 2.2C, lower). A majority of free energies were more negative (9/21) or unchanged (11/21) compared to wild type, but only one value was more positive (Fig. 2.2D).  Comparing free energies in saturating cAMP with those in basal cAMP (Fig. 2.1D and Fig. 2.2D), there was a lower proportion of more positive free energies (1/21 versus 2/18), a lower proportion of unchanged free energies (11/21 versus 15/18) and a higher proportion more negative free energies (9/21 versus 1/18). For one site (G433A), free energy was significantly positive in basal cAMP but, in saturating concentrations of cAMP, it was not altered significantly. The shift of perturbation energies towards the negative, when assayed at saturating levels of cAMP, suggest that the open conformation becomes more stable as a result of cAMP binding.  Three of the mutants that were not functional in basal cAMP recovered function in saturating levels cAMP (A425V, T426A and Y428A), which may have been due to one or both of the following reasons. First, in basal cAMP levels, the mutations may have shifted the range of current activation to very negative voltages at which function cannot be reliably ascertained (i.e. more negative than -150 mV). In elevated cAMP, the activation range would have  94  Figure 2.2 Saturating levels of cAMP (2 mM) further stabilize the open state A. Current traces recorded from CHO cells expressing wild type and three representative S6 alanine mutant HCN2 channels at saturating levels of cAMP. Currents were elicited by test voltage pulses ranging from -150 mV to -10 mV, in 20 mV steps from a holding potential of -35 mV. The tail currents were elicited at -35 mV. B. Representative If activation curves determined by plotting tail current amplitudes which were normalized to their maximum value (I/Imax), versus test voltages (HCN2, squares; Q440A, circles; A437V, upright triangles; T426A, inverted triangles). The curved lines represent fitting by Equation 2.1 (see Experimental Procedures). C. Bar graphs depicting the changes in V½ (upper) and Z (lower) values for each mutant channel relative to wild type. D. Bar graph depicting change in perturbation of free energy, ∆(ZFV½), in mutant channels relative to wild type. One mutant channel did not yield measurable levels of If (solid line through numbered residue, X axis).  95  1 nA  3s  T426A  0.8  0.6  0.4  A437V  0.5 nA 3s  voltage (mV) I422A V423A G424A A425V T426A C427A Y428A A429V M430A F431A I432A G433A H434A A435V T436A A437V L438A I439A Q440A S441A L442A D443A  HCN2  3s  A437V  T426A  -150 -130 -110 -90 -70 -50 -30 -10 10  ZFV1/2  A. C.  Q440A  V1/2  3s -30  1 nA  Z  B.  D.  HCN2  0.2 -2  0.0  -4  -6  I422A V423A G424A A425V T426A C427A Y428A A429V M430A F431A I432A G433A H434A A435V T436A A437V L438A I439A Q440A S441A L442A D443A  normalized current  2 mM cAMP 30  20 * *  10  *  -20 **  2  * *  *  ** *  1  *  *  *  **  *  -1  2  *  *  *  *  0  1 nA  -10 *  4  3  * *  0  *  1.0 -2  8  Q440A  6  4  *  0  *  **  -8  Figure 2.2  96  moved to less negative voltages where the likelihood of detecting channel activity is increased using our protocols. Second, the number of functional channels at the cell surface or single channel conductance may have been reduced by the mutations. For HCN2 channels, cAMP has been suggested to increase open probability in addition to shifting the activation curve to more positive voltages (Craven and Zagotta, 2004), which could have overcome reductions in number of functional channels or single channel conductance. A reduction in the number of functional channels or single channel conductance by these three mutations is supported by the significantly lower levels of current they produce compared to the wild type channel (wt HCN2, -421 ± 98 pA/pF, n= 8; A425V, -71 ± 8 pA/pF n = 3; T426A, -116 ± 22 pA/pF, n = 4; Y428A, -100 ± 16 pA/pF, n = 5; all of the mutants are significantly different from wild type HCN2, p<0.05).  The G424A mutant did not yield current in either basal or elevated cAMP. A lack of function has also been reported for the identical mutant when expressed in Xenopus oocytes (Cheng et al., 2007). Western blotting showed that this mutant did not undergo complex glycosylation, unlike the wild type channel but like a channel in which the N-glycosylation site has been mutated (N380Q) (Fig. 2.3). These data suggest that G424 is important for plasma membrane localization of functional channels.  2.3.3 The effects of S6 mutations on Z are consistent with an altered closed to open transition In Shaker, an alanine/valine scan of the pore showed that Z values increased as V½ values became more negative (Yifrach and MacKinnon, 2002). This relationship is consistent with effects on the final closed to open step in a linear gating scheme in which each of the four  97  M I  1 nA  38 0Q G 42 4A  N  C  U T  -35 mV  N2  B.  HCN2  H  A.  136 kDa 114  3s -150 mV  G424A  GAPDH  -35 mV -150 mV  1 nA 3s  Figure 2.3 Glycine 424 is critical for the expression of cell surface HCN2 channels A. Current traces elicited from cells expressing wild type HCN2 (upper trace) or HCN2 G424A (lower trace) in response to hyperpolarizing voltage pulses to -150 mV from a holding potential of -35 mV. B. Western blot probed with a rabbit polyclonal antibody directed against the C-terminus of HCN2. Lane 1, untransfected cells (UT), Lane 2, wt HCN2, Lane 3, HCN2 N380Q (N-glycosylation mutant), Lane 4, HCN2-G242A. The arrows indicate the presence of mature (M, ~136 kDa), immature (I, ~114 kDa) protein forms. These data are representative of 3 independent experiments. Note the absence of a mature form of HCN2 in lanes containing HCN2 N380Q (as demonstrated previously (Much et al., 2003; Nazzari et al., 2008)) and HCN2 G424A.  98  voltage sensors moves independently and, once all sensors reach the permissive state, the pore opens by a voltage-independent concerted transition (Schoppa and Sigworth, 1998; Zagotta et al., 1994).  For HCN2, we were struck by the mutation-induced changes in Z because they were very small compared to those in Shaker. To determine whether the comparatively small changes in Z are still consistent with an altered closed to open step in HCN2, we applied an allosteric model that captures most aspects of HCN channel gating behavior (Altomare et al., 2001). K  C  C1  C2  C3  C4    O  O1  O2   L  O4  O3 aK  In this model, the voltage sensor in each of the four monomeric subunits moves from reluctant to willing states (C to C4) independently to then allosterically trigger closed to open transitions. Successive engagement of each subunit enhances the probability of channel opening (Po) given by  1  Po =  1+1/K(V) 1 + L(V)  1+1/aK(V)  4  (Equation 2.2)  99  where K(V) and L(V) are the equilibrium constants for voltage sensor movement and the closed to open step, respectively. One important way in which this model differs from the scheme used to describe Shaker is that the closed to open step is dependent upon voltage. Using this model, Altomare et al (2001) showed that HCN-mediated currents were wellfitted, and that isoform-specific positions of the activation curves and delays in both current activation and deactivation could be predicted.  We used this allosteric model to generate hypothetical values of Z and V½ by varying the rate of either the closed to open step (L(V)) or voltage-sensor movement (K(V)) to assess which change could best predict the effects of the S6 mutations on Z. Because the HCN2 S6 mutations are in a region of the pore that contains the gate, an effect on the closed to open transition, and thus on L(V), would be expected. Z values derived from model Po curves by varying L(V), but not by varying K(V), should then approximate our experimental Z values.  To test this, Po curves were generated using Equation 2.2 with a range of L(V) and K(V) values and model parameters specific for either basal or 2 mM cAMP. Model parameters were determined by best fitting and are shown in Table 2.2. Select Po curves that spanned a similar range of voltages as those determined experimentally were then fitted with Equation 2.1 to yield theoretical values for Z and V½, which were then plotted in Fig. 2.4A and 2.4B. Both the Z values obtained by varying L(V) and those observed experimentally do not vary greatly with V½; this held true at basal and at saturating levels of cAMP (in Fig. 2.4, compare the experimentally determined Z values with those determined from the model using a range of L(V) values, represented by the individual symbols and the black lines,  100  respectively). In contrast, the Z values obtained by varying K(V) in the model increase at more negative voltages and plateau in range of voltages separate from that in which most of the experimentally determined Z values are found, in both basal and saturating levels of cAMP (in Fig. 2.4, compare the experimentally determined Z values with those determined from the model using a range of K(V) values, represented by the individual symbols and the gray lines, respectively). Furthermore, when K(V) was decreased in the model, the activation curves reached a point at which Z and V½ values changed very little, even with very small values for K(V). Consequently, there are no model Z values at voltages less negative than ~95 mV in Fig. 2.4 (note that the gray lines do not continue to less negative voltages in this Figure). These data are consistent with an impact of the S6 mutations primarily on L(V) and thus on the closed to open transition.  However, some Z values were affected significantly by the mutations, especially when cAMP was elevated (note the colored points in Fig. 2.4). This is not predicted by the model when varying either L(V) or K(V), suggesting that combined effects of the mutations on both voltage sensor movement and the closed to open step, and/or on other transitions prior to the final steps, contribute significantly to the observed changes in Z.  101  A.  B.  Basal cAMP 8  8  7  7  6  6  5  Z  2 mM cAMP  5  Z  4  4  3  3  2  2  1  1  0  0 -160  -140  -120  -100  V1/2 (mV)  -80  -60  -40  -160  -140  -120  -100  -80  -60  -40  V1/2 (mV)  Figure 2.4 Experimental and model Z values are comparable and change minimally over the range of observed mid-activation voltages Plots of Z values versus V½ values for wild type HCN2 channels and each mutant channel examined, at basal (left) and 2 mM cAMP (right). Each line is derived from paired Z and V½ values determined from model Po curves at varying L(V) (black) and K(V) (gray) (see Results). Also shown are individual values for Z and V½ obtained experimentally for wild type (filled black diamonds), mutants that are significantly different from wild type (filled red or blue diamonds, which are smaller or larger than wild type, respectively) and mutants that are not significantly different from wild type (open squares).  102  Table Table2.2 2 basal cAMP L = / * L'  K= / * K'  L' range K' range      z = -z z = -z a r  0.0001594 1198 1086 106.4 1.123 0.8437 0.2 25.85 0.01 - 50 10^-25 - 10^6  2 mmcAMP L = / * L' K= / * K'  L' range K' range      z = -z z = -z a r  0.0003785 208.4 13.33 86.66 0.8974 0.9621 0.2 25.85 0.1 - 500 10^-15 - 10^8  Table 2.2 Allosteric model parameters at basal and saturating (2 mM) levels of cAMP Parameters were obtained by statistical fitting in Matlab, using those from Altomare et al (2001) as initial values, which were determined for the wild type human HCN2 channel.  103  2.4 Discussion The mixed effects on the voltage-dependence of channel opening and very small perturbation energies produced by the majority of S6 mutations in basal levels of cAMP, and an abundance of mutations with negative perturbation energies in saturating levels of cAMP, suggest that the stability of the open and closed states are similar, and that cAMP binding shifts the energetic balance toward a more stable open state. This implies that the voltage sensors must apply force upon the HCN2 pore to close. This is unlike Shaker channels, which are most stable in the closed conformation and in which voltage sensor works to open the pore (Yifrach and MacKinnon, 2002). Thus, voltage-dependent channel gating in both HCN and Shaker channels is constrained such that the force exerted by the voltage sensor on the gate occurs during depolarization of the membrane potential.  Our findings explain the presence of an “instantaneous” current at all voltages in wild type HCN channels (Chen et al., 2001; Gauss et al., 1998; Ishii et al., 1999; Proenza et al., 2002; Proenza and Yellen, 2006), and the frequent observation that artificial perturbations to HCN lead to even larger constitutively-active currents. A resting conductance of ~2% has been estimated for HCN2 channels, whereas a value between 4-8% has been estimated for sea urchin HCN channels, without and with cAMP, respectively (Proenza and Yellen, 2006). Our data imply that the channel open probability does not reach zero, yielding a significant resting conductance, and that the voltage sensor is unable to exert sufficient force to realize this end. The production of greater constitutive current seen with a number of single-point mutations in the S4-S5 and C- linkers (Chen et al., 2000; Chen et al., 2001; Decher et al., 2004; Macri and Accili, 2004), and upon cadmium binding to cysteine substitutions near the  104  intracellular side of the pore (Rothberg et al., 2003), when understood in the context of a naturally open pore, suggests that these perturbations weaken the link between the voltage sensor and pore. Alternatively, residual current through a channel in the closed state may contribute to a resting conductance but this would not depend upon the energetic balance between the open and closed states. Nevertheless, a constitutively open channel may not necessarily be an inevitable consequence of a pore that is more stable when open. At more positive voltages, the voltage sensor could actively keep the channel shut. This is the opposite of what happens in a channel with a pore that is more stable when closed, like Shaker, in which the voltage sensors work to keep the channel open.  Perturbation energies induced by the S6 mutations in HCN2 were smaller than those in Shaker (Yifrach and MacKinnon, 2002) which suggest weaker interactions between the voltage-sensing elements and the pore. Loose coupling between the voltage sensor and pore, as might be expected from a weak structural interaction, has been proposed recently for HCN channels (Bruening-Wright et al., 2008). These authors showed that the energetics of voltage sensor movement is little affected in sea urchin HCN channels that have been “locked open”, as opposed to the energetics of voltage sensor movement in locked open Shaker channels which are significantly affected. The lack of apparent coupling in a locked open HCN channel is completely consistent with the notion that the pore is naturally open without input from the voltage sensing elements.  A difference in gating dynamics of HCN2 from Shaker is also suggested by our finding that the effective charge Z, determined from the slope of the activation curve, was changed only  105  minimally by the single-point S6 mutations. In contrast, single-point mutations in the S6 of Shaker altered Z and perturbation energy to a much greater extent, and the Z values increased as V½ values became more negative (Yifrach and MacKinnon, 2002). This difference in observed Z between these 2 channels may arise from the fact that, in HCN2, the closed to open transition as well as the movement of the voltage sensor may be voltage dependent (Altomare et al., 2001; Yifrach and MacKinnon, 2002). Thus, the slope of the HCN2 activation curve would reflect contributions from both processes, whereas that of Shaker would reflect a contribution primarily from voltage sensor movement. It should be noted that in 2007 a study on HCN2 channels suggested that the closed to open transition may instead be voltage independent (Chen et al., 2007). It will be interesting to determine whether the gating model developed in that study predicts the small changes in Z seen in our study.  Cyclic AMP has been proposed to stabilize the HCN open state by removing an inhibitory action of the CNBD on pore opening. In the absence of cAMP, inhibition by the CNBD occurs by a coupled interaction with the C-linker region that is thought to apply a force on the S6 helices to actively inhibit pore opening (Craven and Zagotta, 2004; Zhou and Siegelbaum, 2007). Our data showing a significant shift of perturbation energies to more negative values by mutations in the S6 are consistent with this proposed action of cAMP and identify a cluster of residues around the proposed activation gate (Rothberg et al., 2002) that are modified by the inhibitory action of the CNBD (Fig. 2.5). Our data are also consistent with previous work in sea urchin HCN wherein mutation of a single residue in S6 (F459L) produced an equivalent effect to cAMP on gating (Shin et al., 2004). The corresponding site in mouse HCN2 (F431) is one of the ten cAMP-sensitive sites identified in our study.  106  Our data suggest that the primary effect of the S6 mutations is on the closed to open step, the final step of the activation process, which seems reasonable for several reasons. First, the mutations that are energetically sensitive cluster in a region of the S6 that likely forms the activation gate (Rothberg et al., 2002; Rothberg et al., 2003; Shin et al., 2001). Second, the small effects of the mutations on effective charge can be mostly, although not completely, explained by effects on the pore opening step. Third, cAMP, which releases the inhibitory influences on pore opening, significantly shifts perturbation energies towards the negative, suggesting that both the mutations and the CNBD target the same region. Nevertheless, an allosteric effect of the mutations on voltage sensor movement could have contributed to the observed alterations in gating. We found that the significant effects on the effective charge (Z) produced by some of the mutations could not be explained by an allosteric model in which only the pore opening step, or only the voltage-sensor movement, was altered. Other strategies are required to determine whether the voltage-sensing elements of HCN channels contribute to the observed effects of the S6 mutations on gating. It is important to note that the perturbation energies of the S6 mutations in HCN2 are small relative to those in the prototypical Shaker channel, especially at basal levels of cAMP; therefore, neither the pore or voltage sensor are apparently affected despite mutations in and around the activation gate. These small perturbation energies, along with their shift toward the negative by cAMP, are strong support for both a weak interaction between the pore and voltage sensor, compared to Shaker, and a pore that is not at its energetic minimum when closed. The evidence demonstrating that the effects of the mutations on perturbation energy in saturating cAMP levels are larger, and shifted towards negative, greatly strengthens this conclusion.  107  Figure 2.5 Distribution of amino acids in distal HCN2 S6 segment that are critical for energetic balance of open and closed configurations S6 residues with significant perturbation energies (see Table 2.1) are categorized and mapped according to color on to homology model of the HCN2 pore in the closed state (Giorgetti et al., 2005). A color key for each residue mutated is shown below. Ten sites, including 2 sites at the N-terminal end and 4 sites at the C-terminal end, were unaffected by the mutations and G424A did not produce current with or without cAMP. A tetramer is shown on the left, whereas the one subunit alone is shown on the right.  108  Top SF S5 G424A  A425V  S5  T426A A429V F431A H434A  Y428A M430A G433A A437V  S6  S6  A435V I439A  Bottom  No change in energy with both basal and 2 mM cAMP Change in energy with basal cAMP Change in energy with 2 mM cAMP Change in energy with both basal and 2 mM cAMP No expression  Figure 2.5  109  A naturally open pore in HCN2 has important implications for the structural orchestration of gating. The direction of charge and voltage sensor movement is similar between HCN and Shaker-related channels, despite the inverted dependence of HCN channel opening to voltage, which implies that the coupling of voltage sensor movement to channel opening is inverted (Bell et al., 2004; Mannikko et al., 2002; Vemana et al., 2004). We suggest that positive force is applied by the voltage sensor to the C-terminal region of the S6 helices during depolarization to cause the gate to close in HCN2, rather than to open as in Shaker. The structural details of this action will have to await more sophisticated analyses such as the determination of HCN crystal structure, but we believe our present findings provide a glimpse into a fundamentally different way of cycling between open and closed states in the Kv superfamily of voltage-gated channels.  110  2.5 Acknowledgements  VM is the recipient of doctoral scholarships from the Michael Smith Health Research Foundation and the Canadian Institutes for Health Research. HN is the recipient of doctoral scholarships from the Michael Smith Health Research Foundation and the Natural Sciences and Engineering Research Council of Canada. EAA is the recipient of a Tier II Canada Research Chair. Supported by grants from the Heart and Stroke Foundation of British Columbia & the Yukon (EAA). We would also like to thank Patrick Fletcher for help with Matlab and Martin Biel (Munich) for mouse HCN2 cDNA.  111  2.6 References  Altomare, C., Bucchi, A., Camatini, E., Baruscotti, M., Viscomi, C., Moroni, A., and DiFrancesco, D. (2001). Integrated allosteric model of voltage gating of HCN channels. J Gen Physiol 117, 519-532.  Barbuti, A., Baruscotti, M., Altomare, C., Moroni, A., and DiFrancesco, D. (1999). Action of internal pronase on the f-channel kinetics in the rabbit SA node. J Physiol 520 Pt 3, 737-744.  Bell, D.C., Yao, H., Saenger, R.C., Riley, J.H., and Siegelbaum, S.A. (2004). Changes in local S4 environment provide a voltage-sensing mechanism for mammalian hyperpolarizationactivated HCN channels. J Gen Physiol 123, 5-19.  Bruening-Wright, A., Pandey, S., and Larsson, P. (2008). Loose Coupling Between The Voltage Sensor And The Activation Gate In HCN Channels Suggests A Molecular Mechanism For Voltage Gating. Biophysical Journal 94, 119.  Chen, J., Mitcheson, J.S., Lin, M., and Sanguinetti, M.C. (2000). Functional roles of charged residues in the putative voltage sensor of the HCN2 pacemaker channel. J Biol Chem 275, 36465-36471.  Chen, J., Mitcheson, J.S., Tristani-Firouzi, M., Lin, M., and Sanguinetti, M.C. (2001). The S4S5 linker couples voltage sensing and activation of pacemaker channels. Proc Natl Acad Sci U S A 98, 11277-11282. 112  Chen, S., Wang, J., Zhou, L., George, M.S., and Siegelbaum, S.A. (2007). Voltage sensor movement and cAMP binding allosterically regulate an inherently voltage-independent closedopen transition in HCN channels. J Gen Physiol 129, 175-188.  Cheng, L., Kinard, K., Rajamani, R., and Sanguinetti, M.C. (2007). Molecular mapping of the binding site for a blocker of hyperpolarization-activated, cyclic nucleotide-modulated pacemaker channels. J Pharmacol Exp Ther 322, 931-939.  Craven, K.B., and Zagotta, W.N. (2004). Salt bridges and gating in the COOH-terminal region of HCN2 and CNGA1 channels. J Gen Physiol 124, 663-677.  Decher, N., Chen, J., and Sanguinetti, M.C. (2004). Voltage-dependent gating of hyperpolarization-activated, cyclic nucleotide-gated pacemaker channels: molecular coupling between the S4-S5 and C-linkers. J Biol Chem 279, 13859-13865.  DiFrancesco, D. (1999). Dual allosteric modulation of pacemaker (f) channels by cAMP and voltage in rabbit SA node. J Physiol 515 ( Pt 2), 367-376.  DiFrancesco, D., and Tortora, P. (1991). Direct activation of cardiac pacemaker channels by intracellular cyclic AMP. Nature 351, 145-147.  113  Doyle, D.A., Morais Cabral, J., Pfuetzner, R.A., Kuo, A., Gulbis, J.M., Cohen, S.L., Chait, B.T., and MacKinnon, R. (1998). The structure of the potassium channel: molecular basis of K+ conduction and selectivity. Science 280, 69-77.  Gauss, R., Seifert, R., and Kaupp, U.B. (1998). Molecular identification of a hyperpolarization-activated channel in sea urchin sperm. Nature 393, 583-587.  Giorgetti, A., Carloni, P., Mistrik, P., and Torre, V. (2005). A homology model of the pore region of HCN channels. Biophys J 89, 932-944.  Hackos, D.H., Chang, T.H., and Swartz, K.J. (2002). Scanning the intracellular S6 activation gate in the shaker K+ channel. J Gen Physiol 119, 521-532.  Ishii, T.M., Takano, M., Xie, L.H., Noma, A., and Ohmori, H. (1999). Molecular characterization of the hyperpolarization-activated cation channel in rabbit heart sinoatrial node. J Biol Chem 274, 12835-12839.  Jiang, Y., Lee, A., Chen, J., Cadene, M., Chait, B.T., and MacKinnon, R. (2002a). Crystal structure and mechanism of a calcium-gated potassium channel. Nature 417, 515-522.  Jiang, Y., Lee, A., Chen, J., Cadene, M., Chait, B.T., and MacKinnon, R. (2002b). The open pore conformation of potassium channels. Nature 417, 523-526.  114  Ludwig, A., Zong, X., Jeglitsch, M., Hofmann, F., and Biel, M. (1998). A family of hyperpolarization-activated mammalian cation channels. Nature 393, 587-591.  Macri, V., and Accili, E.A. (2004). Structural elements of instantaneous and slow gating in hyperpolarization-activated cyclic nucleotide-gated channels. J Biol Chem 279, 16832-16846.  Macri, V., Proenza, C., Agranovich, E., Angoli, D., and Accili, E.A. (2002). Separable gating mechanisms in a Mammalian pacemaker channel. J Biol Chem 277, 35939-35946.  Mannikko, R., Elinder, F., and Larsson, H.P. (2002). Voltage-sensing mechanism is conserved among ion channels gated by opposite voltages. Nature 419, 837-841.  Much, B., Wahl-Schott, C., Zong, X., Schneider, A., Baumann, L., Moosmang, S., Ludwig, A., and Biel, M. (2003). Role of subunit heteromerization and N-linked glycosylation in the formation of functional hyperpolarization-activated cyclic nucleotide-gated channels. The Journal of Biological Chemistry 278, 43781-43786.  Nazzari, H., Angoli, D., Chow, S.S., Whitaker, G., Leclair, L., McDonald, E., Macri, V., Zahynacz, K., Walker, V., and Accili, E.A. (2008). Regulation of cell surface expression of functional pacemaker channels by a motif in the B-helix of the cyclic nucleotide-binding domain. American Journal of Physiology 295, C642-652.  115  Proenza, C., Angoli, D., Agranovich, E., Macri, V., and Accili, E.A. (2002). Pacemaker channels produce an instantaneous current. J Biol Chem 277, 5101-5109.  Proenza, C., and Yellen, G. (2006). Distinct populations of HCN pacemaker channels produce voltage-dependent and voltage-independent currents. J Gen Physiol 127, 183-190.  Rothberg, B.S., Shin, K.S., Phale, P.S., and Yellen, G. (2002). Voltage-controlled gating at the intracellular entrance to a hyperpolarization-activated cation channel. J Gen Physiol 119, 8391.  Rothberg, B.S., Shin, K.S., and Yellen, G. (2003). Movements near the gate of a hyperpolarization-activated cation channel. J Gen Physiol 122, 501-510.  Santoro, B., Liu, D.T., Yao, H., Bartsch, D., Kandel, E.R., Siegelbaum, S.A., and Tibbs, G.R. (1998). Identification of a gene encoding a hyperpolarization-activated pacemaker channel of brain. Cell 93, 717-729.  Schoppa, N.E., and Sigworth, F.J. (1998). Activation of Shaker potassium channels. III. An activation gating model for wild-type and V2 mutant channels. J Gen Physiol 111, 313-342.  Shin, K.S., Maertens, C., Proenza, C., Rothberg, B.S., and Yellen, G. (2004). Inactivation in HCN channels results from reclosure of the activation gate: desensitization to voltage. Neuron 41, 737-744.  116  Shin, K.S., Rothberg, B.S., and Yellen, G. (2001). Blocker state dependence and trapping in hyperpolarization-activated cation channels: evidence for an intracellular activation gate. J Gen Physiol 117, 91-101.  Vemana, S., Pandey, S., and Larsson, H.P. (2004). S4 movement in a mammalian HCN channel. J Gen Physiol 123, 21-32.  Wainger, B.J., DeGennaro, M., Santoro, B., Siegelbaum, S.A., and Tibbs, G.R. (2001). Molecular mechanism of cAMP modulation of HCN pacemaker channels. Nature 411, 805810.  Yifrach, O., and MacKinnon, R. (2002). Energetics of pore opening in a voltage-gated K(+) channel. Cell 111, 231-239.  Zagotta, W.N., Hoshi, T., and Aldrich, R.W. (1994). Shaker potassium channel gating. III: Evaluation of kinetic models for activation. J Gen Physiol 103, 321-362.  Zhou, L., and Siegelbaum, S.A. (2007). Gating of HCN channels by cyclic nucleotides: residue contacts that underlie ligand binding, selectivity, and efficacy. Structure 15, 655-670.  117  3. The unique form and function of the HCN channel selectivity filter2  3.1 Introduction Hyperpolarization-activated Cyclic Nucleotide-modulated (HCN) channels are similar in structure and function to potassium-selective channels (Biel et al., 2009; Robinson and Siegelbaum, 2003). HCN channels pass predominantly potassium, are blocked by millimolar levels of cesium ions (Hille, 2001; Ludwig et al., 1999; Mistrik et al., 2005; Moroni et al., 2000) and activated by extracellular potassium (Heginbotham and MacKinnon, 1993; Ludwig et al., 1998; Macri and Accili, 2004; Macri et al., 2002; Moroni et al., 2000; Sakmann and Trube, 1984; Stampe et al., 1998; Yang and Sigworth, 1998). Differences in permeation also exist between HCN and potassium channels. HCN channels are only minimally inhibited by barium or TEA (DiFrancesco, 1981a, b; Ludwig et al., 1998; Wollmuth and Hille, 1992), both of which are strong blockers of potassium channels (Hille, 2001). Sodium ordinarily passes in significant amounts in HCN channels, although it is less permeable than potassium, and passes only when potassium is present on the same side of the plasma membrane (DiFrancesco, 1981b; Ludwig et al., 1998; Moroni et al., 2000; Pape, 1996). Because larger organic cations permeate (D'Avanzo et al., 2009; Wollmuth and Hille, 1992), the minimum diameter of the HCN pore may be wider than that of potassium channels (Doyle et al., 1998; Hille, 2001). Finally, the single channel conductance of HCN channels is very small, less than 2 pS when measured in very high concentrations of potassium (Dekker and Yellen, 2006; DiFrancesco, 1986), as compared to 5-50 pS for potassium channels measured at physiological potassium  2  A version of this chapter has been submitted for publication. Macri, V, Angoli, D, Accili, EA. The unique form and function of the HCN channel selectivity filter.  118  concentrations (Hille, 2001). These observations suggest that HCN pore structure and function cannot be inferred from existing studies of voltage-gated potassium channels.  The primary sequence of HCNs predicts a pore consisting of the selectivity filter at the outer end and the voltage controlled gate at the inner side; the latter has been supported by functionally analyzing the accessibility of the pore to metals or drugs applied when the channels are open or closed (Giorgetti et al., 2005; Roncaglia et al., 2002; Rothberg et al., 2002; Shin et al., 2001). In GYG-containing potassium channels, the selectivity filter sequence is T/S-V/I/L/T-GYG, (Shealy et al., 2003; Yu and Catterall, 2004) which form a row of four binding sites through which dehydrated potassium ions move (Aqvist and Luzhkov, 2000; Doyle et al., 1998; Jiang et al., 2003). In HCNs, the equivalent residues are C-IGYG, but whether these similarly form four cation binding sites is not known. It has been proposed that the cysteine residues form a ring around the internal opening of the selectivity filter, with their respective alpha carbons lying within 11 Å of each other (Giorgetti et al., 2005; Roncaglia et al., 2002). This orientation and distance comes from experiments showing irreversible reduction of conductance of HCN2 and sea urchin HCN channels, but not of corresponding cysteine-substituted channels, by application of cadmium from the cytoplasmic side, implying that binding of this metal in the permeation path was coordinated by the four appropriatelyspaced cysteine residues.  Even if the selectivity filter cysteines are close to the permeation path, they may not make strong contact with permeating cations because this residue lacks the negatively charged hydroxyl group contained in the side chains of threonine and serine, which contribute to the fourth and most internal ion binding site (S4) of the potassium channel selectivity filter (Doyle 119  et al., 1998; Yu and Catterall, 2004). Indeed, crystallographic studies of KcsA showed that that substitution of the S4 threonine with cysteine removes the hydroxyl group, with the sulfur sidechain replacing the γ-carbon of the threonine side-chain, and dramatically reduces potassium binding at this site (Zhou and MacKinnon, 2004); the KcsA structure was otherwise unaltered and the backbone carbonyl groups forming the first three sites of the selectivity filter remain at 3-4 Å apart. Using the HCN2 isoform as the prototypical HCN channel, we indeed show that the selectivity filter cysteine has little impact on permeation or associated gating functions of the selectivity filter. These functions are likely controlled, at least in part, by sites which are formed by the backbone carbonyl groups of „CIGYG‟ in HCNs.  3.2 Methods 3.2.1 Site-directed mutagenesis Three selectivity filter mutant channels, HCN2 C400T-IGYG (T400), HCN2 C400S-IGYG (S400) and HCN2 C400A-IGYG (A400), were constructed by overlapping PCR mutagenesis from a mouse HCN2 template as previously described (Macri et al., 2002). C-I401V-GYG (V401) and C400T-I-401V-GYG (T400/V401) channels were also constructed but they did not form functional channels when expressed in CHO cells. The amplified mutagenized products were subsequently digested with NheI and BlpI and ligated within the complementary wild type HCN2 vector. The mutations were confirmed by restriction enzyme analysis and automated sequencing (carried out at The Centre for Molecular Medicine and Therapeutics, DNA Sequencing Core Facility, BC Children's and Women's Hospital, University of British Columbia, Vancouver Canada).  120  3.2.2 Tissue culture and expression of HCN2 constructs Chinese Hamster Ovary (CHO) cells (ATCC, Manassas, VA) were maintained in Hams F-12 media supplemented with antibiotics and 10% FBS (Gibco, Burlington, Ontario), and incubated at 37oC with 5% CO2. Cells were plated onto glass coverslips. Two days after splitting, mammalian expression vectors encoding wild type or mutant HCN2 channels (2 g per 35 mm dish), and a green fluorescent protein (GFP) reporter plasmid (0.6 g per dish) were transiently co-transfected into the cells using the FuGene6 transfection reagent (Roche Biochemical, Indianapolis, IN).  3.2.3 Whole-cell patch clamp electrophysiology CHO Cells expressing GFP were chosen for whole-cell patch clamp recordings 24-48 hours post transfection. The pipette solution contained varying concentrations of K aspartate, NaCl or N-methyl D-glucamine (NMG) (see figure legends for each experimental condition) with each solution containing, 0.5 mM MgCl2, 1 mM EGTA, 5 mM HEPES, pH adjusted to 7.4 with KOH or NaOH depending upon the experimental condition. The extracellular solution contained varying concentrations of NaCl, KCl, and NMG (see figure legends for each experimental condition) with each solution containing, 1.8 mM CaCl2, 0.5 mM MgCl2, 5 mM HEPES, pH adjusted to 7.4 with KOH or NaOH depending upon experimental condition. Whole-cell currents were recorded using an Axopatch 200B amplifier and Clampex software (Axon Instruments, Union City, CA) at room temperature (20-22°C). Patch clamp pipettes were pulled from borosilicate glass and fire polished before use (pipette R= 2.5-4.5 M).  121  3.2.4 Data analysis Data were filtered at 2 kHz and were analyzed using Clampfit (Axon Instruments, Union City, CA), Origin (Microcal, Northhampton, MA) and Excel (Microsoft, Seattle, WA) software. Instantaneous If-V relations were generated as described in our previous studies of HCN channels e.g. (Macri and Accili, 2004; Proenza et al., 2002) which were used to determine the reversal potential (Ef). Briefly, a two part protocol was utilized. First, 500 ms test pulses ranging from +30 mV to -150 mV, from a holding potential of -35 mV, were used to determine the amplitude of voltage-independent/leakage currents at each test voltage. Second, a 500 ms pre-pulse to -150 mV from a holding potential of -35 mV, to open the channels, was followed by test potentials ranging from +30 mV to -150 mV to determine total instantaneous currents at each test voltage. The pre-pulse length was kept to 500 ms to minimize ionic fluxes that could occur over the course of the experiment in a single cell. The voltage-independent/leakage currents, (Iinst) were subtracted from the total instantaneous current at each test voltage to yield values of instantaneous If, which were plotted against test voltage to determine Ef which is the point that crosses the x-axis. Ef values were used to determine permeability ratios for Na+ and K+ (PNa/PK) using the following Goldman Hodgkin Katz equation as described previously for HCN channels (Moroni et al., 2000),  Equation 3.1  Ef = (RT/F)ln([Ko+(PNa/PK)Nao]/[Ki+(PNa/PK)Nai])  In order to determine the affinity and voltage dependence of If block by extracellular Cs+ in CHO cells expressing HCN2 or HCN2 C400T, the Hill and Woodhull equations were used. Cs+ dose-response curves (Fig. 3.3C) were measured using various concentrations of extracellular Cs+ and voltages and fitted with the Hill equation, 122  Equation 3.2  ICs+/I = 1/1+ ([Cs+]/IC50)n  where the IC50 is the concentration at which half of the channels are blocked and “n” is the cooperativity factor between Cs+ and the blocking site. Both the wild type and mutant channels had n values near 1. The IC50 values were then plotted against test voltage and fitted with the Woodhull equation as described previously for HCN channels (Woodhull, 1973),  Equation 3.3  IC50(V) = IC50 (0mV)*exp(zFV/RT)  where the IC50 (0 mV) is the concentration required to block 50% of the total current at 0 mV and is the electrical distance of the Cs+ blocking site within the voltage field, in reference to the extracellular surface, and R,T, and F have their usual thermodynamic meaning. The instantaneous If-V relations before and during Cs+ perfusion for the various concentrations used, were determined using the same two part protocol as described above. The voltageindependent currents measured before and during Cs+ perfusion were subtracted from the total instantaneous currents to determine instantaneous If before and during Cs+ perfusion as a function of voltage. 2 tests were used to determine the goodness of fit, which was considered significant at p<0.05.  The voltage-dependence of activation was determined from tail currents at -65 mV following 2s test pulses ranging from -10 mV to –150 mV, in 20 mV steps, using an extracellular solution containing 135 mM K+ and 5.4 mM Na+ and a pipette solution containing 130 mM K-  123  aspartate and 10 mM NaCl. Normalized tail current amplitudes were plotted as a function of test potential and values were fitted with a Boltzmann equation,  Equation 3.4  f(V) = Imax/(1 + e(V1/2-V)/k)  to determine the midpoint of activation (V1/2) and slope factor (k). Single test pulses were often followed by a 200-500 ms pulse to +5 mV to ensure complete channel deactivation, and the resting current was always allowed to return to its baseline value before subsequent voltage pulses.  3.3 Results 3.3.1The cysteine 400 sulfhydryl side chain does not impact selectivity To examine selectivity, we characterized three substitutions of cysteine 400 in the HCN2 channel. Serine and threonine were chosen, which are found naturally at this site in known potassium-selective channels (Fig. 3.1A). Each adds a hydroxyl group to a putative inner binding site of HCN2, although threonine has a larger volume (116 Å3) as compared to serine (89 Å3) because of the additional CH3 group of its side chain. Alanine, with the same volume as serine, was also chosen as it effectively removes a charged side group yet does not likely alter the main-chain conformation or impose strong electrostatic or steric effects (Cunningham and Wells, 1989).  HCN2 channels, like HCNs in native tissue, are permeable to both sodium and potassium ions (Biel et al., 2009). For HCN2 channels expressed in Chinese hamster ovary (CHO) cells, this can be appreciated from the point at which current reverses direction in instantaneous If -V 124  plots, determined using solutions that contain physiological levels of sodium and potassium. To generate these plots, a pre-pulse to -150 mV was given to maximally activate the channels followed by test pulses to a series of less negative test voltages (Fig. 3.1B). The voltage protocol included a prior set of hyperpolarizing pulses to each test voltage from a holding potential of -35 mV, to quantify the voltage-independent current existing at each test voltage prior to channel activation. The subtraction of voltage-independent current from instantaneous current measured after hyperpolarizing pre-pulse yields a measurement of instantaneous If, which was then plotted against test voltage (Fig. 3.1C). As expected for HCN2 (Ludwig et al., 1998; Moroni et al., 2000), this plot crosses the x-axis, or reverses, at ~-24 mV under these conditions, in between the expected reversal potentials for K+ and Na+ calculated from the Nernst equation using physiological cation concentrations.  Reversal potentials for HCN2 channels containing substitutions of cysteine 400 were also determined using solutions with physiological levels of sodium and potassium; this places the theoretical values of ENa and EK far apart to better reveal any differences from the wild type channel. Serine and alanine substitutions of C400 did not significantly impact reversal potential whereas the bulkier threonine significantly shifted Ef to less negative values by ~12 mV (Fig. 3.1C). A similar shift was found when voltage-steps of 5 mV, rather than 30 mV, were used for both wild type and T400 channels to increase accuracy (data not shown). Permeability ratios (PNa/PK) for the wild type and T400 channels were determined using the GHK equation (Equation 3.1) and were 0.35 ± 0.02 (n=12 cells) and 0.58 ± 0.02 (n = 12 cells), respectively, and were significantly different (t-test; p<0.05).  125  3.3.2 The cysteine 400 sulfhydryl side chain does not impact cation flow The ~2 fold increase in PNa/PK ratio after substitution by threonine suggests that its bulkier side group impinges upon the permeation path to modify cation flow, unlike cysteine, serine or alanine. However, it is not clear if this alteration in selectivity is due to an action on potassium or sodium permeation, or on both cations. If the effect of threonine is related to a steric influence of its larger side chain, then the larger potassium ion might be preferentially affected in the T400 channel.  To examine this, we measured whole-cell conductance using solutions that contained either potassium or sodium (at 135 mM, for both intracellular and extracellular solutions). To measure the current density upon full activation, we applied one 2 second hyperpolarizing pulse to -150 mV to CHO cells expressing either the wild type HCN2 or T400 channel (Fig. 3.2A). In potassium-only solutions, the current density was significantly larger for the wild type channel by ~2 fold compared to the T400 channel (Fig. 3.2C).  We also examined conductance with intracellular and extracellular solutions containing only sodium because a threonine-induced increase in the permeation of this cation might have contributed to the greater PNa/PK value. We were surprised to find that both the wild type and T400 channels displayed robust hyperpolarization-activated current (Fig 3.2B). Previous studies of cloned and native HCNs have uniformly suggested that current disappears in the absence of potassium (Andalib et al., 2002; Biel et al., 2009), suggesting that sodium is unable to permeate on its own. Current density in sodium-only solutions calculated for the HCN2 and T400 channels were not significantly different and considerably smaller than densities determined using potassium-only solutions (Fig. 3.2C). Together, the data suggest that the 126  A.  B. CIGYG CIGYG CIGYG CIGYG TIGYG TIGYG TIGYG TIGYG TIGYG SIGYG TVGYG TVGYG TVGYG TVGYG TVGYG TVGYG TVGYG TVGYG TLGYG TLGYG TTGYG  30  5  - 15 0  C.  30  40  - 15 0  30  HCN2  HCN2 0 nA  20  T400  leakage -60  Instantaneous I f  mV  -60  0.5 s  0 nA  mV  -60  0.5 nA  T400  10  test voltage (mV) -50  -40  -30  -20  -10 -10 -20 -30  10  20  Instantaneous I f (pA/pF)  HCN1 HCN2 HCN3 HCN4 Kir1.1 Kir2.1 Kir3.1 Kir3.4 KCNQ1 SK BK Shaker Kv1.2 Kv1.5 Kv2.1 KvAP KcsA Mthk Kv3.1 Kv4.2 Kat1  -3 5  Figure 3.1 Mutation of the innermost binding site from cysteine to threonine, but not serine or alanine, shifts the reversal potential to more positive potentials in physiological solutions A. An alignment of the five amino acids forming the four cation binding sites of the selectivity filter of K+ channels with those residues of the proposed selectivity filter of the four mammalian HCN channels. Amino acids highlighted in black represent complete identities, whereas those highlighted in gray represent conserved identities. Note the conservation of the glycine-tyrosine-glycine „GYG‟ motif and isoleucine/valine among the HCN and K+ channels, and the conservation of the threonine in all of the K+ channels except SK. The amino acid sequences were aligned using ClustalW 1.8. B. Current traces from two representative cells expressing HCN2 (upper) and T400 (lower) in response to an instantaneous If-V voltage protocol in a physiologic solution containing low potassium (5.4 mM) and high sodium (135 mM). If is the slowly increasing component of current elicited in response to test voltage pulses, immediately following leakage current. A double arrow highlights the leakage current at a test voltage of -60 mV. The dashed line represents zero current. A double arrow highlights the instantaneous If at a test voltage of -60 mV, which follows a pre-pulse to -150 mV. Instantaneous If at each test voltage was calculated as the total instantaneous current at each test voltage, following a prepulse to -150 mV, subtracted from the leakage current at that test voltage. The voltage protocol used is shown in the inset above the current traces. C. Plots of instantaneous If versus test voltage determined from „B‟, fitted with straight lines. The measured Ef values were –24.6 ± 1.8 mV for HCN2 (n=12 cells, closed squares) and –12.7 ± 1.1 mV for T400 (n=12 cells, open circles), and were significantly different (t-test, p<0.05). The same procedures were carried out using S400 and A400 mutant channels, which yielded Ef values of –20.0 ± 0.5 mV (n=8 cells) and –20.1 ± 0.6 mV (n=6 cells), respectively; these values were not significantly different from wild type (t-test, p>0.05).  127  30  threonine side chain preferentially inhibits potassium movement.  We also wanted to know if the T400 channel conductance would increase to the same extent as the wild type channel when extracellular potassium is raised, as shown previously for the wild type HCN2 channel (Ludwig et al., 1998; Macri et al., 2002; Moroni et al., 2000). We found that raising extracellular potassium from a low (5.4 mM) to a high concentration (135 mM) caused current density to similarly increase, by ~9 fold, for wild type and T400 channels, even though the absolute current density was significantly lower for the mutant channel at both low and high potassium concentrations (Fig. 3.3). Thus, T400 channel conductance is sensitive to extracellular potassium but the extent to which it responds to this cation is reduced.  In Fig. 3.2, experiments relied on comparisons of currents measured in separate cells using sodium-only or potassium-only solutions. To reduce variability and observe the effect of threonine on potassium movement within the same cell, we took advantage of the known positive effect of exchanging sodium for potassium on HCN2 conductance. Previously, we found that exchanging the low level of potassium and high level of sodium for each other in the extracellular solution, without altering their combined total concentration, produced an increase in current density and slope conductance (Macri and Accili, 2004; Macri et al., 2002). These data can be explained by a difference in the positive effect of permeating cations on conductance, which is larger for the better-permeating potassium ion than for the sodium ion (Moroni et al., 2000). This effect can be appreciated in the current traces shown in Fig. 3.4A, when wild type If at -150 mV was measured first in an extracellular solution containing 5.4 mM potassium and 135 mM sodium and then in a solution containing the reversed concentrations of these cations. For the wild type channel, the exchange of sodium for 128  A.  K+ only -35 mV  HCN2 -35 mV  -150 mV  B.  T400 -150 mV  0.5 s  Na+ only -35 mV  HCN2  I f at -150 mV (pA/pF)  250 pA/pF  0  N2 00 C H T4 (6)  -100  (7)  Na+ only -200 -300  (6)  -400 -500  -150 mV -600 -35 mV  N2 00 C H T4  (7)  K+ only  T400 -150 mV 50 pA/pF 0.5 s  Figure 3.2 The T400 mutation reduces the maximum potassium conductance A. HCN2 (black) and T400 (gray) current traces elicited at –150 mV for 2 s, from a holding potential of -35 mV measured in symmetrical potassium-only (top) or sodium-only (bottom) solutions. B. Bar graph comparing current densities (pA/pF) of the HCN2 (black bar) and T400 (white bar) channels measured in potassium-only or sodium-only solutions. The numbers in parentheses represent the number of cells and the asterisk denotes a significant difference between HCN2 and T400 (t-test, p<0.05).  129  I f at -150 mV (pA/pF)  0 -100  B. 14  (6)  (6)  12  (6)  (6)  -200 -300  T400  -400 -500 -600 -700 -800  ]o ]o K K .4 35 [5 [1  (6)  HCN2  Fold increase in I f at -150 mV  A.  ]o K]o K .4 35 [5 [1  10  (6)  8 6 4 2 0  HCN2 T400  Figure 3.3 Wild type and T400 channel conductance increases by the same relative amount in response to raising extracellular potassium A. Bar graph comparing wild type and T400 steady-state current density, in low (5.4 mM) and high (135 mM) concentrations of extracellular potassium, measured in the same cells in response to test pulses at -150 mV, elicited from a holding potential of -35 mV. Asterisks denote significant difference between current density in low versus high extracellular potassium solutions (t-test, p<0.05).B. Bar graph comparing the relative increase in current density of wild type and T400 when raising extracellular potassium from a low (5.4 mM) to high (135 mM) concentrations, from “A”. There was no significant difference between in the fold-increase between wild type and mutant channels (t-test, p>0.05).For both “A” and “B”, the numbers in parentheses represent the number of cells measured.  130  A.  T400  HCN2  -35 mV  -35 mV  [5.4K/135Na]o  [5.4K/135Na]o  [135K/5.4Na]o  -150 mV  -150 mV  [135K/5.4Na]o 200 pA/pF  200 pA/pF  -150 mV  0.5 s  0.5 s  I f at -150 mV (pA/pF)  0  o o a] a] N N 4 35 5. /1 K/ K 5 3 .4 [1 [5  (6) (6)  -200  (6) -300  T400  -400  -600  7 6  -100  -500  C.  (6)  HCN2  Fold increase in I f at -150 mV  B.  o o a] a] N N 4 35 5. /1 K/ K 5 3 .4 [5 [1  5  (6)  4 3 2  (6)  1 0  HCN2 T400  Figure 3.4 Potassium conductance is selectively reduced in individual cells expressing the T400 channel A. HCN2 (right) and T400 (left) current traces elicited at –150 mV, for 2 s using two extracellular solutions from a holding potential of -35 mV, measured in extracellular solutions containing the indicated potassium and sodium concentrations. B. Bar graph comparing current density measured as shown in „A‟, when switching between the solutions indicated in the same cell expressing HCN2 (black bars) or T400 (white bars). The asterisk denotes a significant difference between the two solutions used (t-test; wild type, p<0.05; T400, p>0.05). C. Bar graph comparing the relative increase in current density of wild type and T400 channels when changing between the indicated extracellular solutions in the same cell. Asterisk denotes a significant difference in the fold-increase between wild type and mutant channels (t-test, p<0.05). For both “B” and “C”, the numbers in parentheses represent the number of cells measured.  131  potassium produced an increase in If, but for the T400 channel the change was small and not significant (Fig. 3.4B,C). Thus, the data are again consistent with an effect of threonine specifically on potassium permeation.  3.3.3 Enhanced block by extracellular cesium supports a contribution to the permeation path by the threonine side chain To further investigate the structural change in the permeation pathway of the T400 channel, we examined the inhibition of wild type and T400 channel function by extracellular cesium. Cesium, which is larger than either sodium or potassium, is thought to bind within the ion conduction pathway of HCNs and obstruct cation flow. In both cloned and native HCNs, the fraction of block increases at more negative voltages (DiFrancesco, 1982; Macri and Accili, 2004; Moroni et al., 2000). The data obtained in these HCN studies follow the classic explanation of voltage-dependent inhibition by Woodhull, in which the charged cation enters the pore and binds to a site located within the electric field (Woodhull, 1973). For the mouse HCN2 channel, we have shown that Cs+ binds with an apparent dissociation constant of about 4 mM at a site located ~80% across the electric field from the outside (Macri and Accili, 2004); this places the Cs+ blocking site very near to the inner aspect of the HCN2 selectivity filter. We thought that the bulkier side chains of threonine might interact more strongly with Cs+, which would then block the channel more efficiently.  The effects of a wide range of Cs+ concentrations on If were determined from CHO cells expressing either HCN2 or T400 channels. Figure 3.5A shows that the mutant channel is blocked more strongly than the wild type channel by low concentrations of cesium (0.03 mM). To quantify the inhibition, the ratio of blocked and unblocked If was calculated for each test 132  A.  T400  HCN2 test voltage (mV) -150  -120  -90  test voltage (mV)  200  -60  -30  -600 -800  -150  -120  -90  100  -60  -30  30 -100  0.03 mM Cs+  -200 -300 -400  -1000  Instantaneous I f (pA/pF)  -400  Instantaneous I f (pA/pF)  -200  0.03 mM Cs+  30  -500  B.  C. 1.4  2.0  I  /I (-60 mV) Cs+  1.2 1.6  1.0  IC50 1.2 (mM)  HCN2  0.8 HCN2  0.6  0.8  0.4 0.2  T400  0.4  T400  0.0  0.0  0.1  1  [Cs+] (mM)  -150 -120 -90  -60  -30  0  30  test voltage (mV)  Figure 3.5 Extracellular Cs+ blocks the T400 channel with greater sensitivity and at a site closer to the extracellular side of the selectivity filter A. Plots of instantaneous If versus voltage in cells expressing HCN2 or T400, determined using the voltage protocol and analysis described in Fig. 3.1, before and during perfusion with 0.03 mM Cs+ with 135 mM K+ and 5.4 mM Na+ in the extracellular solution (HCN2, filled squares, HCN2 + Cs+, filled circles, n = 5 cells; T400 open squares, T400 + Cs+, open circles, n = 6 cells). B. Plot of the ratio of blocked current at -60 mV, obtained from instantaneous IfV curves as shown in “A”, versus Cs+ concentration, for HCN2 (filled squares) and T400 (open circles). The values for the ratio of blocked current represent means ± s.e.m. Solid lines represent fits of the data with the Hill equation (Equation 3.2), which gave values for IC50 and Hill factor (n). C. Plot of IC50 values, obtained from Hill plots as shown in “B”, versus test voltage for HCN2 (filled squares) and T400 (open circles) channels. The values for the IC50 values represent means ± s.e.m. Solid lines represent fits of the data with the Woodhull equation (Equation 3.3). Fitting yielded values for IC50 (at 0 mV) and which were 3.14 ± 0.18 mM and 0.66 ± 0.01, respectively for HCN2, and 0.14 ± 0.05 mM and 0.27 ± 0.06, respectively, for T400. 2 values indicated goodness of fits for both the Hill and Woodhull equations at p < 0.05. 133  voltage, plotted against Cs+ concentration and fit with the Hill equation (Equation 3.2). In Fig. 3.5B, the plot for data collected at -60 mV shows that the mutant channel is blocked to a greater extent than the wild type channel over the same range of Cs+ concentrations; this was true at all voltages examined and the Hill coefficient was approximately one for all cases (data not shown). To examine the voltage dependence of block of If by cesium, values for IC50 were determined from the Hill equation, plotted against test voltage and fitted with the Woodhull equation (Equation 3.4; Fig. 3.5C). The difference in these values between HCN2 and T400 is striking. The value for IC50 at 0 mV (from the Woodhull Equation) was significantly reduced from ~3.14 mM for the wild type to 0.14 mM for the mutant channel. This suggests that Cs+ is able to access and attach more tightly to its binding site, suggesting a stronger interaction of this cation with the threonine side chain. The value for electrical distance ( from the Woodhull equation) was also significantly reduced from ~0.66 in the wild type channel to ~0.27 in the mutant channel. This low value was not necessarily expected and suggests that Cs+ binds predominantly at a more superficial site in the pore and/or that the electric field has expanded; this is reflected in the shallow voltage dependence of cesium inhibition of the mutant channel apparent in the individual If-V curves (Fig. 3.5A) and in the plot of IC50 versus voltage (Fig. 3.5C). This more superficial site could be explained by a structural change in the permeation path or by an outward movement of the Cs+ blocking site because of compromised conduction of potassium.  3.3.4 Effects of the T400 mutation on HCN2 function are dependent on potassium ions residing within the internal cavity In potassium channels, a water-filled cavity is found on the intracellular side of the selectivity filter that normally contains one fully hydrated potassium ion (Zhou et al., 2001). This cavity 134  helps to overcome the dielectric barrier provided by the plasma membrane and determines the movement of potassium between the cavity and the selectivity filter (Bichet et al., 2006; Furini et al., 2007; Grabe et al., 2006; MacKinnon, 2003; Nimigean et al., 2003). The structure and ion-attracting ability of the cavity vary among potassium channels (Robertson et al., 2008; Tao et al., 2009). For KCa channels, it has been shown that potassium ions may be concentrated in the cavity, which promotes their entry into the selectivity filter and increases outward conductance (Brelidze et al., 2003; Furini et al., 2007). Using the same reasoning, we thought that the high concentration of intracellular potassium ions might inhibit inward movement of potassium from the selectivity filter to the cavity and that threonine might provide a bigger barrier for movement into the cavity through a strong interaction with potassium.  To test this, we altered the internal cationic environment and measured the increase in inward current produced by extracellular potassium. We used intracellular solutions in which the levels of potassium ions were reduced and those for sodium were raised, and applied one test voltage pulse to -150 mV. For the T400 channel, raising extracellular potassium now produced an increase in current to a level similar to that seen in the wild type channel (Fig. 3.6). For the wild type channel, the altered intracellular solution did modify current density measured at either low or high concentrations of extracellular potassium, but not to the same extent as the T400 channel (compare Fig. 3.4B,C and Fig. 3.6B,C). Together, these data suggest that potassium inhibits its own movement into the cavity to a greater extent when threonine is present at the internal side of the selectivity filter.  We also tested the inhibitory effect of extracellular Cs+ on the T400 channel, using the raised sodium and lowered potassium intracellular solution. Using a low level of extracellular Cs+ 135  A.  B.  HCN2 [10K/130Na]i -35 mV  T400 [10K/130Na]i  -35 mV  [5.4K/135Na]o  [5.4K/135Na]o  -150 mV  -150 mV  [135K/5.4Na]o -150 mV  200 pA/pF  [135K/5.4Na]o  200 pA/pF  -150 mV 0.5 s  0.5 s  I f at -150 mV (pA/pF)  0 -100  ]o ]o Na Na 4 5 . 3 /5 /1 5K 4K 3 . [1 [5  (6)  ]o ]o Na Na 4 5 . 3 /5 /1 5K 4K 3 . [1 [5  -200  -300 -400  -600  (6)  (6)  14 12  (6)  -500  D.  T400  Fold increase in I f at -150 mV  C.  10  8  (6) (6)  6 4 2 0  HCN2 T400  HCN2  Figure 3.6 Reduced potassium conductance of the T400 channel reverts to wild type phenotype by lowering and raising intracellular potassium and sodium, respectively A. Current traces elicited at –150 mV for 2 s, from a holding potential of -35 mV, from cells expressing the wild type (left) or T400 (right) channel, using a modified intracellular solution and two extracellular solutions as indicated. C. Bar graph comparing the change in current density when switching between the indicated solutions in the same cell expressing HCN2 (black bars) or T400 (white bars). The asterisk denotes a significant difference between current densities measured in the two extracellular solutions (p<0.05). D. Bar graph comparing the relative increase in current density of wild type and T400 channels when switching between the indicated extracellular solutions in the same cell. There was no significant difference in the fold-increase between wild type and mutant channels (t-test, p>0.05). For both “C” and “D”, the numbers in parentheses represent the number of cells measured. 136  (0.03 mM), we found that the block of T400 channel in the altered intracellular solution was reduced (Fig. 3.7) to a level comparable to that of the wild type channel (see Fig. 3.5A). This data suggests that the block by Cs+ is influenced by the movement of potassium out of the selectivity filter into the cavity, as was suggested above.  3.3.5 The T400 mutation facilitates channel opening We noted that the rate of channel activation and deactivation were faster and slower, respectively, in T400 than in the wild type channel (upper and middle traces, Fig. 3.8A). These altered rates are consistent with a shift of the voltage dependence of channel opening to less negative voltages. To determine whether this had occurred, we examined the relationship of channel opening with voltage, by plotting normalized tail current amplitudes versus test voltages and fitting these plotted values with the Boltzmann Equation (Fig. 3.8B; Equation 3.4). We found that the T400 mutation significantly shifted the V1/2 of the activation curve to more positive voltages by about ~+12 mV compared to wild type HCN2. We also plotted the rates of activation versus voltage, and found that those for T400 channel were also shifted in the positive direction along the voltage axis (data not shown).  Importantly, we found that the positive shift in the activation curve produced by the T400 substitution was eliminated when using the intracellular solution with raised sodium and lowered potassium (Fig. 3.8A,B). The reversion of conductance, Cs+ inhibition and activation gating of the T400 channel back to the wild type phenotype is very strong evidence that permeation and gating functions are tightly coupled at the selectivity filter.  137  A.  B.  T400 -35 mV  1.0  (6)  [130K/10Na]i 0.8  -150 mV  I  -35 mV  /I (-150 mV) Cs+  0.03 mM Cs+  0.6  (4)  0.4  0.2  [10K/130Na]i 0.03 mM Cs+ 100 pA/pF -150 mV 0.2 s  0.0  i ]i a] a N N 10 30 / 1 K K/ 30 0 1 1 [ [  Figure 3.7 Block of the T400 channel by Cs+ reverts to wild type phenotype by lowering and raising intracellular potassium and sodium, respectively A. Current traces elicited at –150 mV for 0.5 s, from a holding potential of -35 mV, before and during perfusion with 0.03 mM extracellular Cs+ in the same cell expressing the T400 channel. The top trace was measured with an intracellular solution that contained 130 mM K+ and 10 mM Na+ and the bottom trace was measured with intracellular solution that contained 10 mM K+ and 130 mM Na+. B. Bar graph comparing the ratio of blocked current by 0.03 mM Cs+ as shown in “A”. The numbers in parentheses represent the number of cells and the asterisk denotes a significant difference in the amount of blocked current measured using the indicated intracellular solutions (t-test, p<0.05).  138  A.  B.  HCN2  1.0  [130K/10Na]i 1s  T400 [130K/10Na]i 200 pA/pF  normalized I f  200 pA/pF 0.8  HCN2 [130K/10Na]i T400 [130K/10Na]i  0.6  T400 [10K/130Na]i  0.4  0.2  1s  T400  0.0 -150  -130  -110  -90  -70  -50  -30  -10  test voltage (mV) [10K/130Na]i  200 pA/pF 1s  Figure 3.8 The T400 mutation facilitates HCN2 channel opening only when intracellular potassium and sodium are high and low, respectively A. Current traces elicited by test voltage pulses ranging from -150 mV to -10 mV, in 20 mV steps, from a holding potential of -35 mV, followed by a subsequent pulse to -65 mV. B. Activation curves determined by plotting tail current amplitudes, which were normalized to their maximum value versus test voltage. The curved lines represent fitting of the data with a Boltzmann equation (Equation 3.4) which gave V1/2 and k values. The V1/2 and k values were 107.7 ± 4.1 mV and 9.6 ± 1.1 mV (n=7 cells) for HCN2 (filled squares) and -93.4 ± 3.3 mV and 12.6 ± 1.7 mV (n=7 cells) for T400 (filled circles) measured with 130 mM K+ and 10 mM Na+ intracellular solution. The V1/2 and k values for T400 (open circles) measured with 10 mM K+ and 130 mM Na+ were -115.2 ±2.2 mV and 8.2 ± 0.7 mV (n=6 cells). The values determined for the T400 channel using high potassium, low sodium intracellular solution were significantly different from those of the wild type channel (t-test, p<0.05) using the same intracellular solution and from those of the T400 channel using the low potassium, high sodium, intracellular solution (t-test, p<0.05).  139  3.4 Discussion To help develop an understanding of the HCN selectivity filter structure and function, we examined the anomalous cysteine residue, which is found in place of serine or threonine that contribute to the innermost of four binding sites in the potassium channel selectivity filter. Using HCN2 channels, we show that this cysteine has little impact on permeation, implying that it does not make significant contact with permeating ions or impact the environment near the cytoplasmic entrance to the filter. This contrasts with the selectivity filters of GYGcontaining potassium channels in which threonine has been shown to directly cradle a dehydrated or partially-hydrated potassium ion (Doyle et al., 1998; Morais-Cabral et al., 2001). Specifically, we show that substitution of C400 of HCN2 with alanine or serine has no effect on selectivity, whereas its substitution with the bulkier threonine reduces potassium selectivity and conductance, and enhances blockade by Cs+. Importantly, conductance of the smaller sodium ion is unaltered by threonine substitution, consistent with the notion that the other effects of this residue are related to its bulkier side chain.  With threonine at the inner side of the selectivity filter, potassium limited its own movement into the cavity and minimized the increase in conductance produced by raising extracellular potassium. In contrast, in the wild type channel, potassium does not limit its own movement to the same extent, which ensures strong modulation of conductance by raising extracellular potassium and maintains an appropriate balance of potassium and sodium permeation. These wild type functions, which are profoundly important under physiological conditions, are likely controlled, at least in part, by sites formed by the backbone carbonyl groups of „CIGYG‟ in HCNs.  140  We were surpised to find robust expression of If in CHO cells containing wild type HCN2 channels with only sodium in the intracellular and extracellular solutions. All previous studies have suggested that potassium is required in order for sodium to permeate HCN channels (Biel et al., 2009; Pape, 1996). The reason we were able to observe this may have been because of our selection of very large CHO cells for measurement, from which even very low current densities can be measured with reasonable resolution. For both wild type and T400 channels, the current density in sodium-only solutions was ~25 pA/pF, much smaller than the potassiumonly currents we observed which were >250 pA/pF. At such a low density, sodium-only currents would be difficult to resolve and separate from other currents in smaller transfected or native cells. Both the relative conductance of sodium and potassium, and their permeability ratio, were altered by two fold in the T400 channel and were consistent with an effect specifically on potassium flux. Together, these data suggest that cation flow through HCN channels may be simply the sum of the individual abilities of sodium and potassium to permeate.  Even though substitution of cysteine 400 with threonine recapitulates a potassium channel selectivity filter, it did not confer high selectivity for potassium. This is not surprising since, in potassium channels, mutation of this threonine to several other amino acids does not render them less selective for potassium (Heginbotham et al., 1994; Hille, 2001). Moreover, inwardly rectifying channels with an intact selectivity filter, but with a pore helix mutation that faces the internal cavity, lack potassium selectivity; further addition of charged residues in the cavity restore potassium selectivity (Bichet et al., 2006; Bichet et al., 2004; Grabe et al., 2006). Our data suggest that the environment of the internal cavity may also help to maintain an appropriate balance of potassium and sodium permeation in HCN channels. 141  A recent study showed that threonine substituted at the same site in HCN4 channels conferred an increase in the relative permeability of large organic cations as compared to potassium permeability (D'Avanzo et al., 2009). Based on Excluded Field Theory, which assumes that permeability is dictated primarily by sieving mechanisms rather than ion binding properties, it was suggested that the threonine residue enlarged pore diameter. Interestingly, this study also found that the relative permeability of sodium and cesium, when compared to potassium permeability, were also larger in the threonine mutant channel. Thus, a selective reduction in potassium permeability such as we found for HCN2 could also explain the HCN4 data found in that study. A role for ion binding properties in the altered permeation of the T400 channel is further supported by the greater sensitivity of potassium movement to the internal cationic environment.  In our our study, channel opening was reversibly facilitated in concert with lowered potassium conductance and altered block by Cs+. These data are further evidence that permeation and channel opening are tightly linked at the selectivity filter in HCN channels (Macri et al., 2002) as they are in potassium channels (VanDongen, 2004).  If cysteine 400 of HCN2 does not form a critical fourth binding site for permeating cations in the selectivity filter, then what is the role for the strongly conserved ring of these residues at the intracellular entrance of the selectivity filter of HCN channels? Previous studies have suggested that these cysteines may provide for regulation of conductance by intracellular oxididation (Giorgetti et al., 2005; Roncaglia et al., 2002) and/or they may contribute to binding of magnesium (Vemana et al., 2008), which induces some rectification of outward142  flowing current (Lyashchenko and Tibbs, 2008; Vemana et al., 2008). Solving the structure of the HCN pore will be an important step toward understanding the role for this cysteine residue and obtaining a complete picture of selectivity filter function for these unusual channels.  143  3.5 Acknowledgements  VM was supported by a Doctoral Research Awards from the Canadian Institutes of Health Research and the Michael Smith Foundation for Health Research. This study was also supported by a Grant-in-Aid from the Heart and Stroke Foundation of British Columbia and Yukon (EAA). EAA is also the recipient of a Tier II Canada Research Chair.  144  3.6 References  Andalib, P., Wood, M.J., and Korn, S.J. (2002). Control of outer vestibule dynamics and current magnitude in the Kv2.1 potassium channel. The Journal of General Physiology 120, 739-755.  Aqvist, J., and Luzhkov, V. (2000). Ion permeation mechanism of the potassium channel. Nature 404, 881-884.  Bichet, D., Grabe, M., Jan, Y.N., and Jan, L.Y. (2006). Electrostatic interactions in the channel cavity as an important determinant of potassium channel selectivity. Proceedings of the National Academy of Sciences of the United States of America 103, 14355-14360.  Bichet, D., Lin, Y.F., Ibarra, C.A., Huang, C.S., Yi, B.A., Jan, Y.N., and Jan, L.Y. (2004). Evolving potassium channels by means of yeast selection reveals structural elements important for selectivity. Proceedings of the National Academy of Sciences of the United States of America 101, 4441-4446.  Biel, M., Wahl-Schott, C., Michalakis, S., and Zong, X. (2009). Hyperpolarization-activated cation channels: from genes to function. Physiol Rev 89, 847-885.  145  Brelidze, T.I., Niu, X., and Magleby, K.L. (2003). A ring of eight conserved negatively charged amino acids doubles the conductance of BK channels and prevents inward rectification. Proceedings of the National Academy of Sciences of the United States of America 100, 9017-9022.  Cunningham, B.C., and Wells, J.A. (1989). High-resolution epitope mapping of hGHreceptor interactions by alanine-scanning mutagenesis. Science 244, 1081-1085.  D'Avanzo, N., Pekhletski, R., and Backx, P.H. (2009). P-loop residues critical for selectivity in K channels fail to confer selectivity to rabbit HCN4 channels. PLoS One 4, e7712.  Dekker, J.P., and Yellen, G. (2006). Cooperative gating between single HCN pacemaker channels. The Journal of General Physiology 128, 561-567.  DiFrancesco, D. (1981a). A new interpretation of the pace-maker current in calf Purkinje fibres. J Physiol 314, 359-376.  DiFrancesco, D. (1981b). A study of the ionic nature of the pace-maker current in calf Purkinje fibres. J Physiol 314, 377-393.  DiFrancesco, D. (1982). Block and activation of the pace-maker channel in calf purkinje fibres: effects of potassium, caesium and rubidium. The Journal of Physiology 329, 485-507.  146  DiFrancesco, D. (1986). Characterization of single pacemaker channels in cardiac sino-atrial node cells. Nature 324, 470-473.  Doyle, D.A., Morais Cabral, J., Pfuetzner, R.A., Kuo, A., Gulbis, J.M., Cohen, S.L., Chait, B.T., and MacKinnon, R. (1998). The structure of the potassium channel: molecular basis of K+ conduction and selectivity. Science 280, 69-77.  Furini, S., Zerbetto, F., and Cavalcanti, S. (2007). Role of the intracellular cavity in potassium channel conductivity. The Journal of Physical Chemistry 111, 13993-14000.  Giorgetti, A., Carloni, P., Mistrik, P., and Torre, V. (2005). A homology model of the pore region of HCN channels. Biophys J 89, 932-944.  Grabe, M., Bichet, D., Qian, X., Jan, Y.N., and Jan, L.Y. (2006). K+ channel selectivity depends on kinetic as well as thermodynamic factors. Proceedings of the National Academy of Sciences of the United States of America 103, 14361-14366.  Heginbotham, L., Lu, Z., Abramson, T., and MacKinnon, R. (1994). Mutations in the K+ channel signature sequence. Biophys J 66, 1061-1067.  Heginbotham, L., and MacKinnon, R. (1993). Conduction properties of the cloned Shaker K+ channel. Biophys J 65, 2089-2096.  147  Hille, B. Ion channels of excitable membranes (2001).  Jiang, Y., Lee, A., Chen, J., Ruta, V., Cadene, M., Chait, B.T., and MacKinnon, R. (2003). X-ray structure of a voltage-dependent K+ channel. Nature 423, 33-41.  Ludwig, A., Zong, X., Jeglitsch, M., Hofmann, F., and Biel, M. (1998). A family of hyperpolarization-activated mammalian cation channels. Nature 393, 587-591.  Ludwig, A., Zong, X., Stieber, J., Hullin, R., Hofmann, F., and Biel, M. (1999). Two pacemaker channels from human heart with profoundly different activation kinetics. The EMBO journal 18, 2323-2329.  Lyashchenko, A.K., and Tibbs, G.R. (2008). Ion binding in the open HCN pacemaker channel pore: fast mechanisms to shape "slow" channels. The Journal of General Physiology 131, 227-243.  MacKinnon, R. (2003). Potassium channels. FEBS letters 555, 62-65.  Macri, V., and Accili, E.A. (2004). Structural elements of instantaneous and slow gating in hyperpolarization-activated cyclic nucleotide-gated channels. The Journal of Biological Chemistry 279, 16832-16846.  148  Macri, V., Proenza, C., Agranovich, E., Angoli, D., and Accili, E.A. (2002). Separable gating mechanisms in a Mammalian pacemaker channel. The Journal of Biological Chemistry 277, 35939-35946.  Mistrik, P., Mader, R., Michalakis, S., Weidinger, M., Pfeifer, A., and Biel, M. (2005). The murine HCN3 gene encodes a hyperpolarization-activated cation channel with slow kinetics and unique response to cyclic nucleotides. The Journal of Biological Chemistry 280, 2705627061.  Morais-Cabral, J.H., Zhou, Y., and MacKinnon, R. (2001). Energetic optimization of ion conduction rate by the K+ selectivity filter. Nature 414, 37-42.  Moroni, A., Barbuti, A., Altomare, C., Viscomi, C., Morgan, J., Baruscotti, M., and DiFrancesco, D. (2000). Kinetic and ionic properties of the human HCN2 pacemaker channel. Pflugers Arch 439, 618-626.  Nimigean, C.M., Chappie, J.S., and Miller, C. (2003). Electrostatic tuning of ion conductance in potassium channels. Biochemistry 42, 9263-9268.  Pape, H.C. (1996). Queer current and pacemaker: the hyperpolarization-activated cation current in neurons. Annual Review of Physiology 58, 299-327.  149  Proenza, C., Angoli, D., Agranovich, E., Macri, V., and Accili, E.A. (2002). Pacemaker channels produce an instantaneous current. The Journal of Biological Chemistry 277, 51015109.  Robertson, J.L., Palmer, L.G., and Roux, B. (2008). Long-pore electrostatics in inwardrectifier potassium channels. The Journal of General Physiology 132, 613-632.  Robinson, R.B., and Siegelbaum, S.A. (2003). Hyperpolarization-activated cation currents: from molecules to physiological function. Annu Rev Physiol 65, 453-480.  Roncaglia, P., Mistrik, P., and Torre, V. (2002). Pore topology of the hyperpolarizationactivated cyclic nucleotide-gated channel from sea urchin sperm. Biophys J 83, 1953-1964.  Rothberg, B.S., Shin, K.S., Phale, P.S., and Yellen, G. (2002). Voltage-controlled gating at the intracellular entrance to a hyperpolarization-activated cation channel. The Journal of General Physiology 119, 83-91.  Sakmann, B., and Trube, G. (1984). Conductance properties of single inwardly rectifying potassium channels in ventricular cells from guinea-pig heart. The Journal of Physiology 347, 641-657.  150  Shealy, R.T., Murphy, A.D., Ramarathnam, R., Jakobsson, E., and Subramaniam, S. (2003). Sequence-function analysis of the K+-selective family of ion channels using a comprehensive alignment and the KcsA channel structure. Biophys J 84, 2929-2942.  Shin, K.S., Rothberg, B.S., and Yellen, G. (2001). Blocker state dependence and trapping in hyperpolarization-activated cation channels: evidence for an intracellular activation gate. The Journal of General Physiology 117, 91-101.  Stampe, P., Arreola, J., Perez-Cornejo, P., and Begenisich, T. (1998). Nonindependent K+ movement through the pore in IRK1 potassium channels. The Journal of General Physiology 112, 475-484.  Tao, X., Avalos, J.L., Chen, J., and MacKinnon, R. (2009). Crystal structure of the eukaryotic strong inward-rectifier K+ channel Kir2.2 at 3.1 A resolution. Science 326, 16681674.  VanDongen, A.M. (2004). K channel gating by an affinity-switching selectivity filter. Proceedings of the National Academy of Sciences of the United States of America 101, 3248-3252.  Vemana, S., Pandey, S., and Larsson, H.P. (2008). Intracellular Mg2+ is a voltage-dependent pore blocker of HCN channels. Am J Physiol Cell Physiol 295, C557-565.  151  Wollmuth, L.P., and Hille, B. (1992). Ionic selectivity of Ih channels of rod photoreceptors in tiger salamanders. The Journal of General Physiology 100, 749-765.  Woodhull, A.M. (1973). Ionic blockage of sodium channels in nerve. J Gen Physiol 61, 687708.  Yang, Y., and Sigworth, F.J. (1998). Single-channel properties of IKs potassium channels. The Journal of General Physiology 112, 665-678.  Yu, F.H., and Catterall, W.A. (2004). The VGL-chanome: a protein superfamily specialized for electrical signaling and ionic homeostasis. Sci STKE 2004, re15.  Zhou, M., and MacKinnon, R. (2004). A mutant KcsA K(+) channel with altered conduction properties and selectivity filter ion distribution. J Mol Biol 338, 839-846.  Zhou, Y., Morais-Cabral, J.H., Kaufman, A., and MacKinnon, R. (2001). Chemistry of ion coordination and hydration revealed by a K+ channel-Fab complex at 2.0 A resolution. Nature 414, 43-48.  152  4. Concluding chapter 4.1 Overview The overall goal of this thesis was to further our understanding of how the HCN pore regulates ion flow. As stated previously in the introduction, HCN channels are members of the potassium channel superfamily and are similar in structure and function (Biel et al., 2009; Robinson and Siegelbaum, 2003). Both HCN and potassium channels have an S4 voltage sensor which moves in the same direction in response to changes in membrane voltage, a voltage-controlled activation gate located in the S6, an S4-S5 linker which couples voltage sensor movement to the activation gate, and a selectivity filter that has the GYG potassium channel signature sequence motif.  Despite these similarities in structure and function, this thesis set out to answer two questions that have been addressed for potassium channels but remained unknown for HCN channels.  Question 1: Is the HCN channel pore energetically more stable in the closed or open state? In potassium channels, the pore is energetically more stable in the closed state. Chapter 2 revealed that for HCN channels the energetic stability of the closed and the opened channel pore were similar with basal levels of cAMP and that saturating levels of cAMP shifted the energetic stability towards the open pore (Macri et al., 2009). Therefore, the pore structures of HCN and potassium channels are energetically different, which may explain the reversed polarity in voltage-dependent pore opening.  153  Question 2: Is the proposed fourth site of the selectivity filter in HCN channels a binding site for permeating ions? In potassium channels the hydroxyl side chain group of threonine forms part of the fourth binding site of the selectivity filter motif, T/S-V/I/L/T-GYG, and contributes to permeation and gating. Chapter 3 revealed that the sulhydryl side chain group of the conserved cysteine which forms part of the proposed fourth binding site of the selectivity filter motif, CIGYG, does not contribute to permeation or to the effect of permeating ions on gating.  The novel findings presented in chapters 2 and 3 provide insight into the unique structure of the HCN channel pore and selectivity filter. This thesis has exposed how together the HCN channel pore and selectivity filter regulate ion flow to produce a current that is indeed „funny‟.  4.2 A comparison of the energetics of pore opening in HCN and Kv channels For Kv channels, the input of energy in the form of a depolarizing voltage pulse is needed to open the channel pore (Yellen, 2002). The depolarizing voltage pulse is sensed by the S4 which is then transmitted to the lower end of the S6, via the S4-S5 linker, which holds the voltage-controlled gate (Larsson et al., 1996; Tristani-Firouzi et al., 2002). This input of energy results in a conformational change in the lower end of the S6 resulting in pore opening (Holmgren et al., 1998; Liu et al., 1997). The x-ray crystal structures of KcsA and MthK represent the pore in the closed and opened state, respectively (Doyle et al., 1998; Jiang et al., 2002). In MthK, the lower end of the S6 is situated approximately 30 degrees from the central axis of the pore (Jiang et al., 2002). In Kv channels, because the input of  154  energy is required to open the channel pore, it was hypothesized that the closed and not the opened pore would be the low energy state (Yifrach and MacKinnon, 2002). Therefore, the KcsA and MthK pores would represent the low and high energy state, respectively.  An alanine/valine mutagenesis scan of the pore forming domain of Shaker, a prototypical Kv channel, revealed that the closed pore was the low energy state, since the majority of the mutations shifted the V1/2 to more hyperpolarized potentials (Hackos et al., 2002; Yifrach and MacKinnon, 2002). The hyperpolarized shift in V1/2 was indicative of the channel being able to open easier, thus the input of less energy was needed to open the channel pore. Therefore, the point mutations functionally destabilized the closed state of the channel pore. Interestingly, the point mutations which resulted in the largest hyperpolarized shifts in V 1/2 clustered in two regions of the Shaker pore, the pore helix and lower end of the S6 which contains the bundle crossing and the voltage-controlled activation gate. Based upon the x-ray crystal structure of KcsA, which represents the low energy closed state, the bundle crossing and pore helix are the two regions which correspond to the most tightly packed amino acids (Doyle et al., 1998; Jiang et al., 2002). Therefore, it was concluded that the point mutations in these regions disrupted this tight packing and destabilized the low energy closed state.  The pore of HCN and K+ channels are proposed to be structurally similar based upon cysteine accessibility mutagenesis studies and homology modeling (Giorgetti et al., 2005; Rothberg et al., 2002; Rothberg et al., 2003). Furthermore, both HCN and Kv channels contain an activation gate near the lower end of the S6 (Holmgren et al., 1998; Liu et al., 1997; Rothberg et al., 2002; Rothberg et al., 2003). In Chapter 2, to determine whether the  155  HCN2 channel pore was also most stable in the closed state, the same alanine/valine mutagenesis approach was employed as in the Shaker study.  We investigated 22  alanine/valine point mutations along the S6 which covered the residues which formed the pore cavity and voltage-controlled gate. HCN and Kv channels open and close with reversed polarity despite the S4 voltage sensor moving in a similar fashion (Bell et al., 2004; Larsson et al., 1996; Mannikko et al., 2002; Vemana et al., 2004). Therefore, the HCN channel pore opens and closes upon membrane hyperpolarization and depolarization, respectively, and the coupling between the S4 and the pore is thought to be different from Kv channels. The difference in the coupling is not known but is hypothesized to occur at the S4-S5 linker, which links the S4 to the activation gate of the pore (Chen et al., 2001; Decher et al., 2004; Prole and Yellen, 2006).  We therefore hypothesized that the alanine/valine point mutations in the pore would predominately shift the V1/2 to more depolarized potentials indicative of destabilizing the low energy closed state which would thus require the input of less energy to open the pore. However, we found that the effects of the S6 point mutations on the voltage-dependence of channel opening were mixed and that the change in energy was very small in basal levels of cAMP (Macri et al., 2009). Therefore, the mutations did not favor either a hyperpolarized or depolarized shift in V1/2, which was different than that observed for Shaker, where the majority of mutations shifted the V1/2 to hyperpolarized potentials (Hackos et al., 2002; Yifrach and MacKinnon, 2002).  The results in Chapter 2 suggest that the energetic  equilibrium between the closed and open states were similar and that the closed pore of HCN channels may not be as tightly packed as compared to KcsA and Shaker K+ channels.  156  Furthermore, using cAMP which stabilizes the open state of HCN channels (Flynn et al., 2007), we found that a majority of the S6 mutations resulted in hyperpolarizing shifts in V1/2, essentially destabilizing the open state and making the HCN channel pore harder to open.  4.3 The majority of S6 mutations alter channel opening For both Kv and HCN channels, the majority of point mutations along the pore forming domain alter the closed to open step. A linear gating model has been previously shown to describe Shaker currents (Zagotta et al., 1994). The four S4 voltage sensors move from a resting to an activated state, and once all four S4 voltage sensors have become active, the pore undergoes a concerted closed to opening step which is voltage-independent. Based on this model, the majority of the Shaker pore point mutations affected the voltage-independent closed to open step, or „late‟ opening transition process. This conclusion was reached since experimentally it was observed that the point mutations which resulted in hyperpolarized shifts in V1/2 also increased Z which was predicted by the Shaker gating model when altering only the rate constant involved in the voltage-independent or „late‟ opening transition step (Yifrach and MacKinnon, 2002). However, for HCN2, the majority of the S6 mutations did not change Z with either hyperpolarized or depolarized shifts in V1/2 with basal or saturating levels of cAMP (Macri et al., 2009). These experimental observations were also predicted by an HCN channel cyclic allosteric gating model when changing only the rate constants involved in the voltage-dependent closed to open step (Altomare et al., 2001; Macri et al., 2009). However, we found that some of the mutations significantly affected Z which could not be explained by an allosteric model in which only the pore opening step was altered.  157  Therefore, an allosteric effect of the mutations on voltage sensor movement could have contributed to the some of observed alterations in gating.  4.4 The input of energy is conserved in HCN and Kv channels The calculated perturbation energies of the S6 mutations in HCN2 were small relative to the Shaker channel, especially at basal levels of cAMP. These small perturbation energies, and a shift toward negative values by cAMP, are strong support for both a weak interaction between the pore and voltage sensor, compared to Shaker, and a pore that is not at its energetic minimum when closed. Taken together, the S4 voltage sensors must apply force upon the HCN2 pore to close. This is unlike Shaker channels, which are most stable in the closed conformation and in which the voltage sensor works to open the pore (Yifrach and MacKinnon, 2002).  Thus, voltage-dependent channel gating in both HCN and Shaker  channels is constrained such that the force exerted by the voltage sensor on the gate occurs during depolarization of the membrane potential (Fig. 4.1).  4.5 Physiological implications for a naturally opened HCN channel pore A naturally opened HCN channel pore may be important for the role these channels play in excitability. A naturally opened pore may be the result of the significant instantaneous current, Iinst that is observed with the expression of HCN channels (Macri and Accili, 2004; Proenza et al., 2002; Proenza and Yellen, 2006). A resting conductance of ~2% has been estimated for HCN2 channels, whereas a value between 4-8% has been estimated for sea urchin HCN channels, without and with cAMP, respectively (Proenza and Yellen, 2006).  158  HCN Pore  Shaker Pore depolarized  O  unstable  C  stable  O  Energy hyperpolarized C  Figure 4.1 The input of energy is conserved in HCN and Shaker channels Diagram showing that the direction of energy moves from hyperpolarized to depolarized potentials in both Shaker and HCN channels to either open or close the channel pore, respectively. The input of energy in the form of membrane depolarization puts both channels in an unstable state (red letters) which is the open pore for Shaker and the closed pore for HCN channels. Therefore, without the input of energy, the pore of Shaker and HCN channels naturally rest in closed and opened state, respectively.  159  These results suggest that the open channel probability does not reach zero, yielding a significant resting conductance which for example could contribute a substantial amount of inward current during the diastolic depolarization phase of SAN action potential. This resting conductance may be important since HCN4 channels which are the most abundantly expressed in SAN cells open and close relatively slow with respect to the time course of the diastolic depolarization phase, seconds versus 100 milliseconds (DiFrancesco et al., 1986; Ishii et al., 1999; Shi et al., 1999). Therefore, a naturally open pore at hyperpolarized potentials may provide an energetically efficient mechanism to supply inward current to depolarize the membrane during the diastolic depolarization phase of the SAN action potential.  4.6 The sulfhydryl side chain group of cysteine 400 of the CIGYG selectivity filter does not contribute to K+ and Na+ selectivity and conductance For all vertebrate HCN channels, the proposed fourth binding site (S4) of the selectivity filter is formed by the conserved cysteine residue which contributes a backbone carbonyl and sulfhydryl side chain group. However, for most K+ channels, S4 is formed by a conserved threonine or to a lesser extent a serine which contributes a backbone carbonyl group and hydroxyl side chain group (Doyle et al., 1998; Giorgetti et al., 2005; Jackson et al., 2007; Shealy et al., 2003; Zhou et al., 2001). In chapter 3, we showed using the HCN2 isoform that mutation of the conserved cysteine, C400, to serine or alanine did not significantly change the relative permeability for Na+ over K+ (PNa/PK) compared to wild type. Similarly, mutation of the equivalent residue in Shaker K+ channels, threonine 442 to alanine or serine also did not significantly change PNa/PK compared to wild type (Heginbotham et al., 1994; Zheng and  160  Sigworth, 1997).  These findings show that the S4 binding site does not contribute  significantly to ion selectivity in HCN and K+ channels.  However, mutation of the C400 to threonine, which is highly conserved in K+ channels, significantly decreased the relative permeability of K+ over Na+ (C400, PNa/PK ~ 0.35 and T400, PNa/PK ~ 0.58). This was also observed in HCN4 channels (D'Avanzo et al., 2009). Interestingly, the reverse mutation in Shaker K+ channels, threonine 442 to cysteine was not tolerated and abolished all ionic and gating current (Zheng and Sigworth, 1997). Because the presence of cysteine at S4 was lethal in the Shaker K+ channel and the presence of threonine was tolerated in HCN channels, this suggests that the structure of S4 and the selectivity filter are different between HCN and K+ channels.  For the T400 channel, the decrease in the relative permeability of K+ over Na+ coincided with a significant decrease in current density measured at -150 mV with symmetrical K+ (135 mM) only solutions compared to wild type HCN2. However, the current density measured at -150 mV with symmetrical Na+ (135 mM) only solutions were not significantly different between the wild type and T400 channel. The T400 mutation reduced both K+ permeability and current density by ~ 2 fold suggesting that the bulkier hydroxyl side chain group inhibits the ability of K+ to traverse the open channel but not Na+, which has a smaller dehydrated ionic radius compared to K+ (Na+ = 0.95 Å and K+ = 1.33 Å). An x-ray crystal structure of the selectivity filter of the KcsA K+ channel showed that mutation of the conserved threonine to cysteine significantly reduced the occupancy of K+ at S4 which suggested that the presence of the sulfhydryl side chain group had a limited interaction with dehydrated K+ (Zhou and  161  MacKinnon, 2004). Taken together, the data suggests that the sulfhydryl side chain group of C400 which forms part of the proposed S4 binding site of the wild type HCN2 channel does not interact with permeating ions. Furthermore, the backbone carbonyls of the CIGYG selectivity filter contribute, in part, to ion selectivity and the effects of extracellular K+ on conductance.  4.7 A role for the selectivity filter in gating in HCN channels The T400 mutation shifted the mid point of voltage dependent opening (V1/2) to more positive values compared to the wild type channel. In Shaker K+ channels, mutation of the equivalent residue T442 to serine shifted the V1/2 to more negative potentials compared to the wild type channel (Yifrach and MacKinnon, 2002; Yool and Schwarz, 1991; Zheng and Sigworth, 1997). Since HCN and Shaker K+ channels are hyperpolarization-activated and depolarization-activated, respectively, the net result was similar: the T400 and S442 mutant channels required less voltage to open the channel pore. The need for less voltage to open the mutant HCN and Shaker channel pores may have been the result of disrupting an energetically favorable interaction with a nearby residue(s) within the selectivity filter or neighboring pore segments, such as the pore-helix or the S6.  Furthermore, we showed the striking result that the functional effects of the T400 mutation on gating and conductance could be restored back to wild type by raising and lowering the intracellular Na+ (130 mM) and K+ (10 mM) concentrations. These findings suggest that gating and permeation influence each other at the selectivity filter. Previous studies of HCN channels have inferred such a relationship by making mutations in and around the selectivity  162  filter and observing changes in gating and conductance (Azene et al., 2003; Azene et al., 2005; D'Avanzo et al., 2009). However, because we were able to restore both gating and conductance in the T400 channel by simply modifying intracellular Na+ and K+ concentration, this strongly suggests that both processes occur at the selectivity filter.  Therefore, the selectivity filter may act as second gate in HCN channels as in other related channels. For example, in KcsA, gating at the selectivity filter has been shown by using life time flouresence spectroscopy (Blunck et al., 2006). Also, the x-ray crystal structures of the KcsA selectivity filter revealed that the backbone carbonyls can adopt a collapsed or opened configuration, in low or high extracellular K+ concentration, respectively (Zhou et al., 2001). This suggests that the collapsed configuration limits ion flow as observed during C-type inactivation in Kv channels (Zhou and MacKinnon, 2003; Zhou et al., 2001). Furthermore, in Kv2.1 channels, increases in both mean open time and in single channel conductance are conferred by increases in the concentration of extracellular K+ (Chapman et al., 2006).  4.8 K+ and Na+ selectivity in HCN channels Although considerable evidence has suggested that the “T/S-V/I/L/T-GYG” motif is critical for the maintenance of high K+ selectivity over Na+ in K+ channels (Aqvist and Luzhkov, 2000; Berneche and Roux, 2001; Doyle et al., 1998; Heginbotham et al., 1994; MoraisCabral et al., 2001; Shi et al., 2006; Zagotta, 2006; Zhou et al., 2001), roles for structures outside of the selectivity filter, such as the pore helix and in the internal pore cavity, have also been shown to be important for maintaining high K+ selectivity over Na+ (Bichet et al., 2006; Bichet et al., 2003; Bichet et al., 2004). We therefore were not completely surprised  163  that replacement of cysteine 400 with a threonine, which recapitulates the inner selectivity filter binding sites of certain K+-selective channels, did not increase the ability of the channel to select K+ over Na+. The opposite result supports that other parts of the channel pore contribute to ion selectivity in both HCN and K+ channels.  A crystal structure of a related non-selective channel, NaK, from Bacillus cereus showed a K+ channel-like selectivity filter motif (TVGYD) (Shi et al., 2006; Zagotta, 2006). The tertiary structure is similar, but not identical, to other known crystal structures of K+ selective channel pore (Doyle et al., 1998; Giorgetti et al., 2005; Jiang et al., 2003). However, the primary structure is also similar to those of HCN and CNG channels, which demonstrate lesser or no preference for K+. Together with our data, these findings suggest that the requirements for obtaining K+ selectivity, and for keeping Na+ from passing, must be very subtle.  The subtleness of variation in structure has been supported experimentally. For example, in the K+ selective Shaker and Kv1.5 channels, the appearance of a significant sodium conductance during and recovery from C-type inactivation was observed (Starkus et al., 1997; Wang et al., 2000). In the KcsA K+ selective channel, molecular dynamic simulations of the second site (S2) have also suggested that even very slight changes in the flexibility of the backbone or the distances between the carbonyls that form the ion binding sites was sufficient to disrupt K+ selectivity (Noskov et al., 2004; Roux, 2005). The HCN selectivity filter of HCN channels may also be potentially more flexible, thereby contributing to the greater permeability for Na+ relative to K+ compared to K+ channels, based on the HCN2  164  pore homology model since the pore helix was predicted to have less of a hydrogen bonding network compared to the pore helix of KcsA (Giorgetti et al., 2005). The greater flexibility of the selectivity filter could therefore better accommodate both dehydrated K+ and Na+ which differ in size by ~ 0.38 Å. Free energy perturbation calculations using the x-ray crystal structure of the non-selective NaK channel also showed that flexibility, and not rigidity or precise geometry of the backbone carbonyls of the selectivity filter, were important in order to maintain a greater selectivity for K+ over Na+ (Noskov and Roux, 2007). Furthermore, a reduction in the number of backbone carbonyls and partial hydration within the selectivity filter were also implicated for contributing to the non selective K+ and Na+ nature of the NaK channel. In the future, the arrival of an x-ray crystal structure of the HCN channel pore will further enhance our understanding of the architecture of the pore and the nature of ion flow through the selectivity filter.  4.9 The selectivity filter motif, CIGYG, sets the reversal potential and conductance response to physiological levels of extracellular K+ In HCN channels, some selectivity for K+ over Na+ is maintained which is critical for setting the reversal potential which allows inward current to flow during diastolic depolarization (Biel et al., 2009; Robinson and Siegelbaum, 2003). Therefore, under normal physiological concentrations of K+ and Na+, the passage of Na+ into cells is important for producing depolarizing inward current in tissues such as the SAN. Moreover, the ability of HCN channels to increase conductance in response to changes in extracellular K+ is important since increasing extracellular K+ would depolarize the resting membrane potential which would limit the amount of available HCN current. Under different physiological and  165  pathophysiological conditions in the heart and brain, extracellular K+ concentration may vary between 3 and 12 mM (Choate et al., 2001; Dietzel et al., 1982; Kleber, 1983; Paterson, 1996; Sykova, 1983).  Based on the data in Chapter 3, the proposed binding sites of the  backbone carbonyls, and not the sulfhydryl side chain group, of the CIGYG selectivity filter, in part, set both the range of voltages over which depolarizing current is available as well as the response of HCN channel conductance to changes in extracellular K+.  4.10 Future research directions Here, I propose two future research directions which naturally extend from the data presented in Chapters 2 and 3.  1) In Shaker K+ channels, an alanine scan of the S5 and S6 showed that the lower end of the S6 near the bundle crossing, and not the S5, significantly altered the energy of pore opening (Yifrach and MacKinnon, 2002). These findings suggested that the lower end of the S6, and not the S5, was a tightly packed structure and that these mutations altered mainly pore opening and not S4 voltage-sensor movement. However, a later study showed that several point mutations in both S5 and S6 also significantly altered gating charge or S4 voltagesensor movement in Shaker K+ channels (Soler-Llavina et al., 2006). Whether mutations in the S5 and S6 alter gating charge or S4 voltage-sensor movement is not known in HCN channels. The measurement of gating currents and use of voltage clamp fluorimetry would be useful to determine whether gating charge or S4 voltage-sensor movement, in addition to pore opening, was also being affected by point mutations in S5 and S6.  However, gating  currents and voltage clamp fluorimetry have only been determined with the sea urchin HCN  166  channel and not from mammalian HCN channels (Bruening-Wright and Larsson, 2007; Mannikko et al., 2002; Mannikko et al., 2005). Therefore, these experiments would be technically challenging using a mammalian HCN channel as in Chapter 2.  2) In the Kir3.2 K+ channel, which is highly selective for K+ over Na+ (PNa/PK ~ 0.06), multiple substitutions of a pore helix residue near the S4 binding site of the selectivity filter dramatically reduced K+ selectivity over Na+, ~ 100 fold (PNa/PK ~ 0.6). However, high K+ selectivity over Na+ could be restored to wild type levels by introducing a negatively charged residue, aspartate, at sites along the S6 which face the internal pore cavity (Bichet et al., 2006; Bichet et al., 2004). Furthermore, the large single channel conductance of the BK K+ channel has been attributed to a ring of eight negatively charged glutamate residues located at the cytoplasmic entrance of the internal pore cavity and K+ channels lacking this negatively charged configuration typically have a small single channel conductance (Brelidze et al., 2003). Visual inspection of the residues which form the S6 of HCN channels and from the HCN2 pore homology model based upon the x-ray crystal structure of the KcsA K+ channel pore, reveals that the S6 has no negatively charged residues which face the internal pore cavity; however, there is a single aspartate at the cytoplamic entrance of the internal pore cavity (Giorgetti et al., 2005). Therefore, introducing negatively charged residues at sites along the S6 which face the internal pore cavity (e.g. Q440, T436, G433, A429 and A425), may also significantly increase K+ selectivity over Na+ ~ 100 fold as in Kir3.2. For example, the PNa/PK would decrease from ~ 0.3 for the wild type to ~ 0.03 for the aspartate facing internal pore cavity mutants. Furthermore, increasing the number of negatively charged residues may also increase single channel conductance which is very small, ~ 1 pS, for wild  167  type HCN channels. Therefore, targeting sites along the S6 which are exposed to the internal pore cavity could provide insight on the origin of the significant permeability of Na+ relative to K+ and the low single channel conductance in HCN channels.  168  4.11 References  Altomare, C., Bucchi, A., Camatini, E., Baruscotti, M., Viscomi, C., Moroni, A., and DiFrancesco, D. (2001). Integrated allosteric model of voltage gating of HCN channels. The Journal of General Physiology 117, 519-532.  Aqvist, J., and Luzhkov, V. (2000). Ion permeation mechanism of the potassium channel. Nature 404, 881-884.  Azene, E., Xue, T., and Li, R.A. (2003). Molecular basis of the effect of potassium on heterologously expressed pacemaker (HCN) channels. The Journal of Physiology 547, 349356.  Azene, E.M., Sang, D., Tsang, S.Y., and Li, R.A. (2005). Pore-to-gate coupling of HCN channels revealed by a pore variant that contributes to gating but not permeation. Biochemical and Biophysical Research Communications 327, 1131-1142.  Bell, D.C., Yao, H., Saenger, R.C., Riley, J.H., and Siegelbaum, S.A. (2004). Changes in local  S4  environment  provide  a  voltage-sensing  mechanism  for  mammalian  hyperpolarization-activated HCN channels. The Journal of General Physiology 123, 5-19.  Berneche, S., and Roux, B. (2001). Energetics of ion conduction through the K+ channel. Nature 414, 73-77.  169  Bichet, D., Grabe, M., Jan, Y.N., and Jan, L.Y. (2006). Electrostatic interactions in the channel cavity as an important determinant of potassium channel selectivity. Proceedings of the National Academy of Sciences of the United States of America 103, 14355-14360.  Bichet, D., Haass, F.A., and Jan, L.Y. (2003). Merging functional studies with structures of inward-rectifier K(+) channels. Nature Reviews 4, 957-967.  Bichet, D., Lin, Y.F., Ibarra, C.A., Huang, C.S., Yi, B.A., Jan, Y.N., and Jan, L.Y. (2004). Evolving potassium channels by means of yeast selection reveals structural elements important for selectivity. Proceedings of the National Academy of Sciences of the United States of America 101, 4441-4446.  Biel, M., Wahl-Schott, C., Michalakis, S., and Zong, X. (2009). Hyperpolarization-activated cation channels: from genes to function. Physiological Reviews 89, 847-885.  Blunck, R., Cordero-Morales, J.F., Cuello, L.G., Perozo, E., and Bezanilla, F. (2006). Detection of the opening of the bundle crossing in KcsA with fluorescence lifetime spectroscopy reveals the existence of two gates for ion conduction. The Journal of General Physiology 128, 569-581.  170  Brelidze, T.I., Niu, X., and Magleby, K.L. (2003). A ring of eight conserved negatively charged amino acids doubles the conductance of BK channels and prevents inward rectification. Proceedings of the National Academy of Sciences of the United States of America 100, 9017-9022.  Bruening-Wright, A., and Larsson, H.P. (2007). Slow conformational changes of the voltage sensor during the mode shift in hyperpolarization-activated cyclic-nucleotide-gated channels. J Neurosci 27, 270-278.  Chapman, M.L., Blanke, M.L., Krovetz, H.S., and VanDongen, A.M. (2006). Allosteric effects of external K+ ions mediated by the aspartate of the GYGD signature sequence in the Kv2.1 K+ channel. Pflugers Arch 451, 776-792.  Chen, J., Mitcheson, J.S., Tristani-Firouzi, M., Lin, M., and Sanguinetti, M.C. (2001). The S4-S5 linker couples voltage sensing and activation of pacemaker channels. Proceedings of the National Academy of Sciences of the United States of America 98, 11277-11282.  Choate, J.K., Nandhabalan, M., and Paterson, D.J. (2001). Raised extracellular potassium attenuates the sympathetic modulation of sino-atrial node pacemaking in the isolated guineapig atria. Exp Physiol 86, 19-25.  D'Avanzo, N., Pekhletski, R., and Backx, P.H. (2009). P-loop residues critical for selectivity in K channels fail to confer selectivity to rabbit HCN4 channels. PloS One 4, e7712.  171  Decher, N., Chen, J., and Sanguinetti, M.C. (2004). Voltage-dependent gating of hyperpolarization-activated, cyclic nucleotide-gated pacemaker channels: molecular coupling between the S4-S5 and C-linkers. The Journal of Biological Chemistry 279, 13859-13865.  Dietzel, I., Heinemann, U., Hofmeier, G., and Lux, H.D. (1982). Stimulus-induced changes in extracellular Na+ and Cl- concentration in relation to changes in the size of the extracellular space. Experimental brain research Experimentelle Hirnforschung 46, 73-84.  DiFrancesco, D., Ferroni, A., Mazzanti, M., and Tromba, C. (1986). Properties of the hyperpolarizing-activated current (if) in cells isolated from the rabbit sino-atrial node. The Journal of Physiology 377, 61-88.  Doyle, D.A., Morais Cabral, J., Pfuetzner, R.A., Kuo, A., Gulbis, J.M., Cohen, S.L., Chait, B.T., and MacKinnon, R. (1998). The structure of the potassium channel: molecular basis of K+ conduction and selectivity. Science 280, 69-77.  Flynn, G.E., Black, K.D., Islas, L.D., Sankaran, B., and Zagotta, W.N. (2007). Structure and rearrangements in the carboxy-terminal region of SpIH channels. Structure 15, 671-682.  Giorgetti, A., Carloni, P., Mistrik, P., and Torre, V. (2005). A homology model of the pore region of HCN channels. Biophysical Journal 89, 932-944.  172  Hackos, D.H., Chang, T.H., and Swartz, K.J. (2002). Scanning the intracellular S6 activation gate in the shaker K+ channel. The Journal of General Physiology 119, 521-532.  Heginbotham, L., Lu, Z., Abramson, T., and MacKinnon, R. (1994). Mutations in the K+ channel signature sequence. Biophys J 66, 1061-1067.  Holmgren, M., Shin, K.S., and Yellen, G. (1998). The activation gate of a voltage-gated K+ channel can be trapped in the open state by an intersubunit metal bridge. Neuron 21, 617621.  Ishii, T.M., Takano, M., Xie, L.H., Noma, A., and Ohmori, H. (1999). Molecular characterization of the hyperpolarization-activated cation channel in rabbit heart sinoatrial node. The Journal of Biological Chemistry 274, 12835-12839.  Jackson, H.A., Marshall, C.R., and Accili, E.A. (2007). Evolution and structural diversification of hyperpolarization-activated cyclic nucleotide-gated channel genes. Physiological Genomics 29, 231-245.  Jiang, Y., Lee, A., Chen, J., Cadene, M., Chait, B.T., and MacKinnon, R. (2002). The open pore conformation of potassium channels. Nature 417, 523-526.  Jiang, Y., Lee, A., Chen, J., Ruta, V., Cadene, M., Chait, B.T., and MacKinnon, R. (2003). X-ray structure of a voltage-dependent K+ channel. Nature 423, 33-41.  173  Kleber, A.G. (1983). Resting membrane potential, extracellular potassium activity, and intracellular sodium activity during acute global ischemia in isolated perfused guinea pig hearts. Circulation Research 52, 442-450.  Larsson, H.P., Baker, O.S., Dhillon, D.S., and Isacoff, E.Y. (1996). Transmembrane movement of the shaker K+ channel S4. Neuron 16, 387-397.  Liu, Y., Holmgren, M., Jurman, M.E., and Yellen, G. (1997). Gated access to the pore of a voltage-dependent K+ channel. Neuron 19, 175-184.  Macri, V., and Accili, E.A. (2004). Structural elements of instantaneous and slow gating in hyperpolarization-activated cyclic nucleotide-gated channels. The Journal of Biological Chemistry 279, 16832-16846.  Macri, V., Nazzari, H., McDonald, E., and Accili, E.A. (2009). Alanine scanning of the S6 segment reveals a unique and cyclic AMP-sensitive association between the pore and voltage-dependent opening in HCN channels. The Journal of Biological Chemistry.  Mannikko, R., Elinder, F., and Larsson, H.P. (2002). Voltage-sensing mechanism is conserved among ion channels gated by opposite voltages. Nature 419, 837-841.  174  Mannikko, R., Pandey, S., Larsson, H.P., and Elinder, F. (2005). Hysteresis in the voltage dependence of HCN channels: conversion between two modes affects pacemaker properties. The Journal of General Physiology 125, 305-326.  Morais-Cabral, J.H., Zhou, Y., and MacKinnon, R. (2001). Energetic optimization of ion conduction rate by the K+ selectivity filter. Nature 414, 37-42.  Noskov, S.Y., Berneche, S., and Roux, B. (2004). Control of ion selectivity in potassium channels by electrostatic and dynamic properties of carbonyl ligands. Nature 431, 830-834.  Noskov, S.Y., and Roux, B. (2007). Importance of hydration and dynamics on the selectivity of the KcsA and NaK channels. The Journal of General Physiology 129, 135-143.  Paterson, D.J. (1996). Role of potassium in the regulation of systemic physiological function during exercise. Acta physiologica Scandinavica 156, 287-294.  Proenza, C., Angoli, D., Agranovich, E., Macri, V., and Accili, E.A. (2002). Pacemaker channels produce an instantaneous current. The Journal of Biological Chemistry 277, 51015109.  Proenza, C., and Yellen, G. (2006). Distinct populations of HCN pacemaker channels produce voltage-dependent and voltage-independent currents. The Journal of General Physiology 127, 183-190.  175  Prole, D.L., and Yellen, G. (2006). Reversal of HCN channel voltage dependence via bridging of the S4-S5 linker and Post-S6. The Journal of General Physiology 128, 273-282.  Robinson, R.B., and Siegelbaum, S.A. (2003). Hyperpolarization-activated cation currents: from molecules to physiological function. Annual Review of Physiology 65, 453-480.  Rothberg, B.S., Shin, K.S., Phale, P.S., and Yellen, G. (2002). Voltage-controlled gating at the intracellular entrance to a hyperpolarization-activated cation channel. The Journal of General Physiology 119, 83-91.  Rothberg, B.S., Shin, K.S., and Yellen, G. (2003). Movements near the gate of a hyperpolarization-activated cation channel. The Journal of General Physiology 122, 501-510.  Roux, B. (2005). Ion conduction and selectivity in K(+) channels. Annu Rev Biophys Biomol Struct 34, 153-171.  Shealy, R.T., Murphy, A.D., Ramarathnam, R., Jakobsson, E., and Subramaniam, S. (2003). Sequence-function analysis of the K+-selective family of ion channels using a comprehensive alignment and the KcsA channel structure. Biophys J 84, 2929-2942.  Shi, N., Ye, S., Alam, A., Chen, L., and Jiang, Y. (2006). Atomic structure of a Na+- and K+-conducting channel. Nature 440, 570-574.  176  Shi, W., Wymore, R., Yu, H., Wu, J., Wymore, R.T., Pan, Z., Robinson, R.B., Dixon, J.E., McKinnon, D., and Cohen, I.S. (1999). Distribution and prevalence of hyperpolarizationactivated cation channel (HCN) mRNA expression in cardiac tissues. Circulation Research 85, e1-6.  Soler-Llavina, G.J., Chang, T.H., and Swartz, K.J. (2006). Functional interactions at the interface between voltage-sensing and pore domains in the Shaker K(v) channel. Neuron 52, 623-634.  Starkus, J.G., Kuschel, L., Rayner, M.D., and Heinemann, S.H. (1997). Ion conduction through C-type inactivated Shaker channels. The Journal of General Physiology 110, 539550.  Sykova, E. (1983). Extracellular K+ accumulation in the central nervous system. Progress in Biophysics and Molecular Biology 42, 135-189.  Tristani-Firouzi, M., Chen, J., and Sanguinetti, M.C. (2002). Interactions between S4-S5 linker and S6 transmembrane domain modulate gating of HERG K+ channels. The Journal of Biological Chemistry 277, 18994-19000.  Vemana, S., Pandey, S., and Larsson, H.P. (2004). S4 movement in a mammalian HCN channel. The Journal of General Physiology 123, 21-32.  177  Wang, Z., Hesketh, J.C., and Fedida, D. (2000). A high-Na(+) conduction state during recovery from inactivation in the K(+) channel Kv1.5. Biophysical Journal 79, 2416-2433.  Yellen, G. (2002). The voltage-gated potassium channels and their relatives. Nature 419, 3542.  Yifrach, O., and MacKinnon, R. (2002). Energetics of pore opening in a voltage-gated K(+) channel. Cell 111, 231-239.  Yool, A.J., and Schwarz, T.L. (1991). Alteration of ionic selectivity of a K+ channel by mutation of the H5 region. Nature 349, 700-704.  Zagotta, W.N. (2006). Membrane biology: permutations of permeability. Nature 440, 427429.  Zagotta, W.N., Hoshi, T., and Aldrich, R.W. (1994). Shaker potassium channel gating. III: Evaluation of kinetic models for activation. The Journal of General Physiology 103, 321-362.  Zheng, J., and Sigworth, F.J. (1997). Selectivity changes during activation of mutant Shaker potassium channels. J Gen Physiol 110, 101-117.  Zhou, M., and MacKinnon, R. (2004). A mutant KcsA K(+) channel with altered conduction properties and selectivity filter ion distribution. J Mol Biol 338, 839-846.  178  Zhou, Y., and MacKinnon, R. (2003). The occupancy of ions in the K+ selectivity filter: charge balance and coupling of ion binding to a protein conformational change underlie high conduction rates. J Mol Biol 333, 965-975.  Zhou, Y., Morais-Cabral, J.H., Kaufman, A., and MacKinnon, R. (2001). Chemistry of ion coordination and hydration revealed by a K+ channel-Fab complex at 2.0 A resolution. Nature 414, 43-48.  179  Appendix A A novel KCNA1 mutation associated with global delay and persistent cerebellar dysfunction3  The published work in the Appendix characterized the functional effects of a point mutation located in the pore of the voltage-gated potassium channel, Kv1.1 that was discovered in a family with global delay and persistent cerbellar dysfunction. This pore point mutation is located at the lower end of the inner helix near the activation gate. Kv and HCN channels are closely related by structure and contain an activation gate located at the lower end of the inner helix of the pore. Because of my interest in the lower S6 region of HCN channels, I was curious to know how the point mutation might affect Kv1.1 channel function, especially in light of the fact that this substitution was clinically relevant. Thus, my role was to determine how the point mutation affected Kv1.1 channel function. We found that the substitution increased the rate at which the Kv1.1 channels inactivate, or close, during prolonged stimulation by voltage. This is consistent with other mutations in the Kv1.1 channel that are linked to cerebellar ataxias. It remains to be seen whether a similar inactivation process exists in mammalian HCN channels.  3  This work has been published. Demos, MK, Macri, V, Farrell, K, Nelson, TN, Chapman, K, Accili, E, Armstrong, L.(2009) A novel KCNA1 mutation associated with global delay and persistent cerebellar dysfunction. Movement Disorders, 24: 788-82.  180  778  M.K. DEMOS ET AL.  A Novel KCNA1 Mutation Associated with Global Delay and Persistent Cerebellar Dysfunction Michelle K. Demos, MD,1* Vincenzo Macri, MS,2 Kevin Farrell, MB ChB,1 Tanya N. Nelson, PhD,3 Kristine Chapman, MS,4 Eric Accili, PhD,2 and Linlea Armstrong, MD5 1  Department of Pediatric Neurology, British Columbia’s Children’s Hospital, Vancouver, British Columbia, Canada; 2 Department of Cellular and Physiological Sciences, University of British Columbia, Vancouver, British Columbia, Canada; 3Department of Pathology and Laboratory Medicine, Children’s and Women’s Health Center of British Columbia, Vancouver, British Columbia, Canada; 4Division of Neurology, Neuromuscular Disease Unit, Vancouver Hospital, Vancouver, British Columbia, Canada; 5Department of Medical Genetics, Children’s and Women’s Health Center of British Columbia, Vancouver, British Columbia, Canada Abstract: Episodic Ataxia Type 1 is an autosomal dominant disorder characterized by episodes of ataxia and myokymia. It is associated with mutations in the KCNA1 voltage-gated potassium channel gene. In the present study, we describe a family with novel clinical features including persistent cerebellar dysfunction, cerebellar atrophy, and cognitive delay. All affected family members have myokymia and epilepsy, but only one individual has episodes of vertigo. Additional features include postural abnormalities, episodic stiffness and weakness. A novel KCNA1 mutation (c.1222G>T) which replaces a highly conserved valine with leucine at position 408 (p.Val408Leu) was identified in affected family members, and was found to augment the ability of the channel to inactivate. Together, our data suggests that KCNA1 mutations are associated with a broader clinical phenotype, which may include persistent cerebellar dysfunction and cognitive delay. Ó 2009 Movement Disorder Society Key words: KCNA1; EA1; cerebellar atrophy; cognitive dysfunction  Episodic Ataxia type 1 (EA1) is a rare autosomal dominant disorder associated with KCNA1 mutations that presents in childhood with brief episodes of ataxia  *Correspondence to: Dr. Michelle K. Demos, Department of Pediatric Neurology, British Columbia’s Children’s Hospital, K3-176 4480 Oak St., Vancouver, British Columbia, Canada, V6H 3V4. E-mail: mdemos@cw.bc.ca Potential conflict of interest: None reported. Received 24 October 2008; Accepted 24 December 2009 Published online 9 February 2009 in Wiley InterScience (www. interscience.wiley.com). DOI: 10.1002/mds.22467  and continuous myokymia.1,2 The clinical spectrum of EA1 has expanded to include epilepsy, episodes of muscle stiffness, postural abnormalities and weakness.2–8 Persistent cerebellar dysfunction with cerebellar atrophy is typically absent in patients with EA19 but is a characteristic feature of Episodic Ataxia Type 2 (EA2), which is associated with mutations in the P/Q-type voltagegated calcium channel gene CACLNA4.10,11 We describe and present functional studies of a novel KCNA1 mutation in a family with EA1 in whom there are clinical features not previously described, including persistent cerebellar dysfunction, cerebellar atrophy and delayed cognitive development. PATIENTS AND METHODS Subjects The proband (Patient III-1) (see Fig. 1A,B) is a 4 yr 9-mo old boy with seizures, global developmental delay, myokymia with postural abnormalities, and episodes of muscle stiffness triggered by illnesses. The seizures started in infancy and are controlled on carbamazepine. He walked at 3 yr and his first word was at 4 yr. At 4 yr 9 mo, he functions at a cognitive level of 24 mo. His receptive and expressive language skills are at a 14-mo level and his motor skills are at an 18 mo level. He has chronic swallowing difficulties and gastroesophageal reflux disease requiring a G-tube. Examination in infancy revealed postural abnormalities. Current examination reveals increased tone, myokymia and mild gait ataxia. Head MRI was normal at 4 mo. Electroencephalograms (EEGs) were normal or demonstrated bilateral epileptiform activity. Patient III-2 (Fig. 1A) is a 14-mo old boy with seizures, myokymia and mild global developmental delay. Seizures began at 3 wk and are controlled on carbamazepine. His examination revealed periocular myokymia and increased tone. EEGs were normal or demonstrated rhythmic spikes in the right temporal region. Patient II-1 (Fig. 1A,C) is a 29-yr old woman with mild cognitive difficulties, episodic vertigo, myokymia, and persistent cerebellar dysfunction. She has had infrequent episodes of muscle stiffness triggered by heat. She describes mild generalized weakness exacerbated by temperature extremes, and difficulty swallowing cold substances. Episodes of vertigo, triggered by activity and heat, began at 2 yr. Seizures began in the neonatal period and were controlled on phenytoin which was discontinued at 4 yr. Persistent dysarthria and ataxia was first recognized at 3 yr. She received learning assistance, was placed in a practical skills class and did not formally graduate. A recent  Movement Disorders, Vol. 24, No. 5, 2009  181  CEREBELLAR ATROPHY IN EPISODIC ATAXIA TYPE 1  779  FIG. 1. Pedigree and clinical features. (A) Pedigree of family. Blackened symbols represent affected individuals. DNA available from numbered individuals. (B) Patient III-1 at 4 mo with tightly clenched fists and persistent flexion of hips and knees. (C) Patient II-1 at 2 mo: tightly clenched fists. (D) Patient II-1 head MRI at age 17 yr demonstrating cerebellar atrophy. (E) Sequencing of KCNA1 revealed heterozygosity for a nucleotide transversion (G>T) in affected family members (III-1, III-2, and II-1), (F) but not in the unaffected family members (I-1, I-2) or normal control. [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.]  examination revealed dysarthric speech, mild facial weakness and myokymia of facial muscles and hands. There was also bilateral calf hypertrophy and mild generalized weakness. An intention tremor; difficulty with fine finger and rapid alternating movements; and ataxic gait were also present. Electromyography (EMG) studies demonstrated myokymic discharges, and after muscle cooling to 208C there was electrical silence following dense fibrillation potentials. With this, she was unable to abduct her fingers. No myotonic discharges were present. A head CT scan at 4 mo was normal. A head MRI at 17 yr revealed mild generalized atrophy of cerebellar hemispheres (Fig. 1D), which was unchanged on repeat scan at age 27 yr.  Genetic and Functional Studies DNA was extracted from relevant family members (GentraSystems, Minneapolis, MN). PCR amplification and direct sequencing of the coding and flanking regions of KCNA1 was performed.12 SeqScape software (Applied Biosystems, Foster City, CA) was used for comparative analysis of resulting sequence to KCNA1 consensus sequence (NM_000217). Genotyping of familial samples was performed using AmpfIstr Identifiler chemistry (Applied Biosystems, Foster City, CA) to verify identity and stated relationships. As described previously, Chinese hamster ovary-K1 (CHO) cells (ATCC, Manassas, VA), were transiently co-transfected with pcDNA3.1 vectors encoding wildtype or mutant KCNA1 channels and green fluorescent pro-  Movement Disorders, Vol. 24, No. 5, 2009  182  780  M.K. DEMOS ET AL.  FIG. 2. Inactivation of the human KCNA1 channel is enhanced by the V408L mutation. (A) Representative current traces from CHO cells transfected with wildtype (black line) and mutant (gray line) channels elicited by 8 voltage pulses to 110 mV, 140 mV, and 170 mV from a holding potential of 280 mV. Traces are normalized to their maximum (peak) values. (B) Plot of time constants of inactivation (s) determined from a single exponential fitting procedure of current traces obtained from cells expressing the wildtype (filled bars) or mutant (unfilled bars) channels at the three test potentials shown in A. s values were significantly faster for the mutant compared with wildtype channels (t-test, P < 0.05). The numbers in parentheses represent the number of cells used for each condition and the asterisk above the numbers signifies a significant difference (t-test, P < 0.05). (C) Plot of the fraction of peak current remaining after 8 sec for the wildtype (filled bars) and mutant (unfilled bars) channels at the three test potentials. The fraction of peak current remaining after 8 sec was significantly less for mutant compared with wildtype channels. For either the wildtype or mutant channel, the fraction of current remaining after 8 sec was the same at each test potential. The numbers in parentheses represent the number of cells used for each condition and the asterisk above the numbers signifies a significant difference (t-test, P < 0.05). Data are reported as mean 6 S.E.M. Experiments were conducted at room temperature (20–228C). Series resistance was not compensated and currents were not leak-subtracted.  tein.13 After the appearance of green fluorescence (24–48 hr later), cells were transferred to a recording chamber (200-lL volume) and continually perfused (0.5–1.0 mL/min) with an extracellular solution (5.48 mM KCl, 1358 mM NaCl, 0.58 mM MgCl2, 1.98 mM CaCl2, 58 mM HEPES, adjusted to pH 7.48 with NaOH). Pipettes were filled with a solution of 1308 mM potassium aspartate, 108 mM NaCl, 0.58 mM MgCl2, 58 mM HEPES, and 18 mM EGTA and adjusted to pH 7.48 with KOH. Currents were measured using borosilicate glass electrodes, which had a resistance of 2.0–4.0 mohms when filled, and recorded using an Axopatch 200B amplifier and Clampex software (Axon Instruments). Data  were filtered at 28 kHz and analyzed using Clampfit (Axon Instruments) and Origin (Microcal) software.  RESULTS Sequencing of KCNA1 revealed heterozygosity for a nucleotide transversion (G>T) in all affected family members, but not in unaffected grandparents or normal control (see Fig. 1E,F). This transversion results in the substitution of leucine (L) for valine (V) at amino acid position 408, a highly conserved residue located in the distal pore region of the KCNA1 channel, which was previously implicated in episodic ataxia when con-  Movement Disorders, Vol. 24, No. 5, 2009  183  CEREBELLAR ATROPHY IN EPISODIC ATAXIA TYPE 1 verted to alanine (A).1 Genotyping confirmed identity and stated relationships indicating that the V408L mutation arose de novo in patient II-1 and was transmitted to her offspring (III-1 and III-2). Because a mutation of valine 408 to alanine was previously found to enhance KCNA1 channel inactivation,14 this behavior was analyzed in CHO cells transfected with either wildtype or mutant (V408L) human KCNA1 channels (see Fig. 2). Both the rate and extent of inactivation were greater in the mutant channel compared with the wildtype channel. Neither the voltage range over which channel opening occurred nor current amplitude was significantly altered by the mutation (data not shown). DISCUSSION We report a family whose clinical features further expand the wide clinical spectrum of EA1. The proband’s mother (II-1) has persistent cerebellar dysfunction associated with cerebellar atrophy on neuroimaging. The proband (III-1) also has mild gait ataxia. Past reports of patients with EA1 have described mild cerebellar dysfunction in some affected family members. Findings included intention tremor and mild difficulties with tandem gait and/or arm coordination.3,15,16 In contrast to these earlier reports, the cerebellar dysfunction in the proband’s mother (II-1) appears to be more severe with an earlier onset and greater functional impact. Her head MRI also demonstrated cerebellar atrophy, a feature which has not been reported previously in EA1. It is possible that treatment in infancy with phenytoin may have contributed to the severity of the cerebellar dysfunction and atrophy present in our patient. Given the reports indicating that phenytoin treatment may be associated with permanent cerebellar dysfunction and atrophy,17,18 this case suggests that phenytoin should be used with caution in young children with EA1. This family demonstrates that cognitive dysfunction may also be a feature of EA1. The mother (II-1) has learning difficulties and was educated in a life skills program. In addition, the proband has marked global delay with severe receptive and expressive language delay. Patient III-2 is also globally delayed. We are aware of only one other report of cognitive dysfunction described as mild-to- moderate learning difficulties in one individual with EA1.4 Exposure to warm temperature is recognized as a potential provoking factor for symptoms of EA1.5,7 In our family, the proband’s mothers’ symptoms and EMG results were exacerbated by cold temperatures, suggesting that symptoms of EA1 are provoked by tem-  781  perature extremes. Sensitivity to cold temperatures is not well recognized for EA1; however, mild cramping and worsening of myokymia with cold exposure has been described in two individuals with EA1.2,16 Mice lacking KCNA1 also demonstrated cooling-induced hyperexcitability in synaptic transmission.19 Therefore, KCNA1 may inhibit involuntary muscle contractions during decreases and increases in external temperature by stabilization of central synaptic transmission. The mutation identified in this family is located at the same position as a previously reported mutation (V408A) causing EA1 in an unrelated family.1 Like the V408A mutation, V408L causes the channel to inactivate faster than the wildtype channel.14 This would be expected to reduce the contribution of KCNA1 channels to repolarization of the membrane potentially after neuronal firing resulting in the increased excitability of neurons. A correlation between the degree of KCNA1 dysfunction and EA1 phenotype has been suggested. Mutations associated with relatively severe disease, poorly responsive to medications or associated with seizures, tend to show profound reductions in KCNA1 current amplitude, whereas milder or typical EA1 cases are associated with mutations altering voltage channel activation which more subtly alters potassium flow.20 The more severe phenotype found here suggests that the altered KCNA1 inactivation more profoundly disrupts potassium flow. However, the V408A mutation found previously, which augments channel inactivation in the same way as V408L, is associated with a much less severe phenotype1,9,14 than that found in this study, suggesting that other factors must contribute to the disease. The determination of these contributing factors and more strongly linking genotype to phenotype may help to develop gene and mutation specific therapies for patients with EA1. In conclusion, patients with KCNA1 mutations may also develop persistent cerebellar dysfunction, have cognitive impairment, and exacerbation of symptoms on exposure to cold temperatures. Functional studies demonstrate channel dysfunction but do not fully explain the interfamilial or intrafamilial phenotypic variability of Episodic Ataxia Type 1. Acknowledgments: E. Accili is the recipient of a Tier 2 Canada Research Chair. V. Macri is the recipient of doctoral fellowships from the Canadian Institutes of Health Research and the Michael Smith Foundation for Health Research. We thank the patients and their families for their participation in this study, Dr. J. Jen for her assistance, and Sarah Chow for her technical assistance with preparation of KCNA1 DNA for transfection.  Movement Disorders, Vol. 24, No. 5, 2009  184  782  M.K. DEMOS ET AL.  Author Roles: Michelle Demos: This author (first author and corresponding author) was involved in conception, organization, and execution of this case study, both in terms of clinical information and genetic studies. She also wrote the first draft excluding small portions of genetic and functional studies sections; Vincenzo Macri: This author was involved in conception, organization, and execution of functional studies. He also provided statistical expertise related to the functional studies. He also participated in review and critique of the manuscript; Kevin Farrell: This author supervised and was involved in the collection of clinical data and conception and organization of information for presentation as a case study. He also reviewed and critiqued multiple drafts of the manuscript; Tanya Nelson: This author supervised and was involved in conception, organization, and execution of genetic studies. She also wrote the genetics section and reviewed and critiqued manuscript; Kristine Chapman: This author was involved in collection of clinical data, specifically neurophysiology data and conception and design of clinical report. She also participated in review and critique of the manuscript; Eric Accili: This author supervised and was involved in conception, organization, and execution of functional studies. He wrote and provided figures for the functional studies section. He also reviewed and critiqued the manuscript; Linlea Armstrong: This author supervised and was involved in conception, organization, and execution of this case study, both in terms of clinical information and genetic and functional studies. She also reviewed and critiqued the manuscript.  REFERENCES 1. Browne DL, Gancher ST, Nutt JG, et al. Episodic ataxia/myokymia syndrome is associated with point mutations in the human potassium channel gene, KCNA1. Nat Genet 1994;8:136–140. 2. Van Dyke DH, Griggs RC, Murphy MJ, Goldstein MN. Hereditary myokymia and periodic ataxia. J Neurol Sci 1975;25:109– 118. 3. Brunt ERP, van Weerden TW. Familial paroxysmal kinesigenic ataxia and continuous myokymia. Brain 1990;113:1361–1382. 4. Zuberi SM, Eunson LH, Spauschus A, et al. A novel mutation in the human voltage-gated potassium channel gene (Kv1.1) associates with episodic ataxia type 1 and sometimes with partial epilepsy. Brain 1999;122:817–825. 5. Eunson LH, Rea R, Zuberi SM, et al. Clinical, genetic, and expression studies of mutations in the potassium channel gene  6.  7. 8.  9. 10. 11.  12. 13. 14. 15. 16. 17. 18. 19.  20.  KCNA1 reveal new phenotypic variability. Ann Neurol 2000; 48:647–656. Chen H, von Hehn C, Kaczmarek LK, Ment LR, Pober BR, Hisama FM. Functional analysis of a novel potassium channel (KCNA1) mutation in hereditary myokymia. Neurogenetics 2007; 8:131–135. Klein A, Boltshauser E, Jen J, Baloh RW. Episodic ataxia type 1 with distal weakness: a novel manifestation of a potassium channelopathy. Neuropediatrics 2004;35:147–149. Kinali M, Jungbluth H, Eunson LH, et al. Expanding the phenotype of potassium channelopathy: severe neuromyotonia and skeletal deformities without prominent Episodic Ataxia. Neuromuscul Disord 2004;14:689–693. Rajakulendran S, Schorge S, Kullman DM, Hanna MG. Episodic ataxia type 1: a neuronal potassium channelopathy. Neurotherapeutics 2007;4:258–266. Vighetto A, Froment JC, Trillet M, Aimard G. Magnetic resonance imaging in familial paroxysmal ataxia. Arch Neurol 1988; 45:547–549. Ophoff RA, Terwindt GM, Vergouwe MN, et al. Familial hemiplegic migraine and episodic ataxia type-2 are caused by mutations in the Ca21 channel gene CACNL1A4. Cell 1996;87:543– 552. Lee H, Wang H, Jen JC, Sabatti C, Baloh RW, Nelson SF. A novel mutation in KCNA1 causes episodic ataxia without myokymia. Hum Mutat 2004;24:536. Macri V, Accili EA. Structural elements of instantaneous and slow gating in hyperpolarization-activated cyclic nucleotide-gated channels. J Biol Chem 2004;279:16832–16846. Adelman JP, Bond CT, Pessia M, Maylie J. Episodic ataxia results from voltage-dependent potassium channels with altered functions. Neuron 1995;15:1449–1454. Hand PJ, Gardner RJM, Knight MA, Forrest SM, Storey E. Clinical features of a large Australian pedigree with episodic ataxia type 1. Mov Disord 2001;16:938–939. Hanson PA, Martinez LB, Cassidy R. Contractures, continuous muscle discharges, and titubation. Ann Neurol 1977;1:120–124. Ney GC, Lantos G, Barr WB, Schaul N. Cerebellar atrophy in patients with long-term phenytoin exposure and epilepsy. Arch Neurol 1994;51:767–771. De Marco FA, Ghizoni E, Kobayashi E, Li LM, Cendes F. Cerebellar volume and long-term use of phenytoin. Seizure 2003;12: 312–315. Zhou L, Zhang CL, Messing A, Chiu SY. Temperature-sensitive neurmuscular transmission in Kv1.1 Null Mice: Role of potassium channels under the myelin sheath in young nerves. J Neurosci 1998;18:7200–7215. Jen JC, Hess EJ, Hanna MG, Griggs RC, Baloh RW, CINCH investigators. Primary episodic ataxias: diagnosis, pathogenesis and treatment. Brain 2007;130:2484–2493.  Movement Disorders, Vol. 24, No. 5, 2009  185  Appendix B Biohazard Approval Certificate The University of British Columbia  186  Page I of I  The University of British Columbia  Biohazard Approval Certificate PROTOCOL NUMBER: B09-0277 "  INVESTIGATOR OR COURSE DlRECTOR: Eric Accili DEPARTMENT: Cellular & Physiologi'cal Sc. PROJECT OR COURSE TITLE: Pacemaker Lab APPROVAL DATE: February 18,2010  START DATE: November 18,2009  APPROVED CONTAINMENT LEVEL: 2 FUNDING TITLE: Molecular regulation of pacemaker channel function FUNDlNG AGENCY: Heart and Stroke Foundation of British Columbia and Yukon "  FUNDING TITLE: Comparative studies of pacemaker channels  FUNDING AGENCY: Natural Sciences and Engineering Research Council of Canada (NSERC)   UNFUNDED TITLE: N/A  The Principal Investigator/Course Director is responsible for ensuring that all research or course work involving biological hazards is conducted in accordance with the University of British Columbia Policies and Procedures, Biosafety Practices and Public Health Agency of Canada guidelines.  This certificate is valid for one year from the above start or approval date (whichever is later) provided there are no changes. Annual review is required. A copy of this certificate must be displayed in your facility.  Office of Research Services  102,6190 Agronomy Road, Vancouver, V6T lZ3  Phone: 604-827-5111 FAX: 604-822-5093   https://rise.ubc.ca/rise/Doc/0/HODF2622Q3M4NAM6TS4HNAGJDB/fromString.html  7/22/2010  187  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0071068/manifest

Comment

Related Items