Ubc 2010 Fall Macri Vincenzo

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

ii

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 voltage-

sensor 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.

iii

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

iv

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

v

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

vi

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

vii

Appendix A A novel KCNA1 mutation associated with global delay and persistent

cerebellar dysfunction .........................................................................................................180

Appendix B Biohazard approval certificate.....................................................................186

viii

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

ix

List of figures

Figure 1.1 Effects of autonomic agonists on spontaneous activity and hyperpolarization-

activated 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

x

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

xi

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.

xii

Dedication

To My Parents

xiii

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.

http://www.ncbi.nlm.nih.gov/pubmed?term=%22Demos%20MK%22%5BAuthor%5D

http://www.ncbi.nlm.nih.gov/pubmed?term=%22Macri%20V%22%5BAuthor%5D

http://www.ncbi.nlm.nih.gov/pubmed?term=%22Farrell%20K%22%5BAuthor%5D

http://www.ncbi.nlm.nih.gov/pubmed?term=%22Nelson%20TN%22%5BAuthor%5D

http://www.ncbi.nlm.nih.gov/pubmed?term=%22Chapman%20K%22%5BAuthor%5D

http://www.ncbi.nlm.nih.gov/pubmed?term=%22Accili%20E%22%5BAuthor%5D

http://www.ncbi.nlm.nih.gov/pubmed?term=%22Armstrong%20L%22%5BAuthor%5D

1

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

2

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).

3

1.1.2 Biophysical properties of If

In SAN cells, If activates at potentials more negative than -30 mV and becomes fully-

activated 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

4

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

5

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

6

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.

7

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; Fernandez-

Velasco 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,

8

Figure 1.1 Effects of autonomic agonists on spontaneous activity and hyperpolarization-

activated 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).

IfIf

9

Na+/Ca

2+ 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

10

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 L-

type 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.

11

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,

12

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

13

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 C-

terminus contains the C-linker and Cyclic Nucleotide Binding Domain (CNBD) shown in

blue. The CNBD is shown binding cAMP (Zagotta et al., 2003).

+ +1 122 33 5 56 6p p

cAMP

Na+, K+

+ +1 122 33 5 56 6p p

cAMP

Na+, K+

14

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

15

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.

KcsA

HCN2

Top View Side View

S5

M1M2

S6

KcsA

HCN2

Top View Side View

S5

M1M2

S6

16

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+

.

17

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

18

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

19

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

20

(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

21

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 C-

linker 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).

22

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 C-

linkers 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

23

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

24

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

25

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 C-

terminus which lacked the CNBD and abolished cAMP modulation, which resulted in

26

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

27

(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

28

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.

29

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.

30

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 voltage-

dependent 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

31

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

32

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 N-

terminal 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

33

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

34

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

35

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

36

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.

KcsA MthK

closed opened

KcsA MthK

closed opened

37

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 Kv

channels.

38

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

39

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

40

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

41

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 S1

(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

42

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 co-

ordinate 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).

C

I

GY

G

C

I

GY

G

KcsA HCN2

1

2

3

4

43

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

44

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,

45

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

46

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.,

47

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

48

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.

49

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

50

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

51

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

52

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.

53

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 cysteine-

scanning substitutions. American Journal of Physiology 294, C136-144.

54

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, 349-

356.

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.

55

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.

56

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 sino-

atrial 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, 809-

813.

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.

57

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


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, 931-

939.

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.

58

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.

59

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, 23-

40.

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.

60

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, 473-

482.

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


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


DiFrancesco, D., and Tromba, C. (1988b). Muscarinic control of the hyperpolarization-

activated current (if) in rabbit sino-atrial node myocytes. The Journal of Physiology 405,

493-510.

61

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.

62

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


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, 9617-

9623.

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.

63

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.

64

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, 500-

544.

Ishii, T.M., Takano, M., and Ohmori, H. (2001). Determinants of activation kinetics in

mammalian hyperpolarization-activated cation channels. The Journal of Physiology 537, 93-

100.

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.

65

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.

66

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


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.

67

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.

68

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.

69

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, 19-

45.

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


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, 27056-

27061.

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.

70

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.

71

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


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, 5101-

5109.

72

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


Proenza, C., and Yellen, G. (2006). Distinct populations of HCN pacemaker channels

produce voltage-dependent and voltage-independent currents. The Journal of General


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 hyperpolarization-

activated cyclic nucleotide-gated channel from sea urchin sperm. Biophysical Journal 83,

1953-1964.

73

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 hyperpolarization-

activated 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.

74

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, 2929-

2942.

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 hyperpolarization-

activated cation channel (HCN) mRNA expression in cardiac tissues. Circ Res 85, e1-6.

75

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


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, 34635-

34643.

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 ion-

permeation pathway of a resting voltage-sensing domain. Nature 445, 546-549.

76

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


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, 43752-

43759.

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.

77

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


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, 805-

810.

78

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


79

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, 35-

42.

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.

80

Zagotta, W.N. (2006). Membrane biology: permutations of permeability. Nature 440, 427-

429.

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.

81

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.

82

2. Alanine scanning of the S6 segment reveals a unique and cyclic AMP-

sensitive association between the pore and voltage-dependent opening in

HCN channels1

2.1 Introduction


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.

83

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

84

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 F-

12 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).

85

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)

86

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) = (2

ZFV½,wt + 2

ZFV½, 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 10g/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

87

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 (I422-

D443) 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

88

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

89

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

90

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).

91

Z

D.

ZF

V1

/2

V1/2

C.HCN2 Q440A

A.

voltage (mV)

no

rma

lized

cu

rren

t

B.

L438A

3 s 3 s

3 s

1 nA 0.5 nA

1 nA

C427A

1 nA

3 s

-150 -130 -110 -90 -70 -50 -30 -10 10

0.0

0.2

0.4

0.6

0.8

1.0

I42

2A

V4

23A

G4

24

AA

42

5V

T4

26A

C4

27A

Y4

28A

A4

29V

M4

30

AF

43

1A

I43

2A

G4

33

AH

43

4A

A4

35V

T4

36A

A4

37V

L43

8A

I43

9A

Q4

40

A

Basal cAMP

S4

41A

L44

2A

D4

43A

I42

2A

V4

23A

G4

24

AA

42

5V

T4

26A

C4

27A

Y4

28A

A4

29V

M4

30

AF

43

1A

I43

2A

G4

33

AH

43

4A

A4

35V

T4

36A

A4

37V

L43

8A

I43

9A

Q4

40

AS

44

1A

L44

2A

D4

43A

-30

-20

-10

0

10

20

30

-2

-1

0

1

2

3

4

-8

-6

-4

-2

0

2

4

6

8

*

*

*

** *

* *

*

*

*

*

HCN2

Q440A

C427A

L438A

Figure 2.1

92

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.

93

Table 1A

basal cAMP

HCN2 channel n V1/2 (mV) Z -ZFV1/2 (kcal/mol) Δ ZFV1/2 (kcal/mol)

wild type 9 -108.9 ± 1.8 2.24 ± 0.18 -5.47 ± 0.36

I422A 6 -111.4 ± 3.4 1.96 ± 0.09 -4.92 ± 0.15 0.54 ± 0.39

V423A 6 -104.7 ± 2.9 2.93 ± 0.40 -6.96 ± 1.03 -1.49 ± 1.09

G424A no expression

A425V no current

T426A no current

C427A 5 -96.4 ± 2.4* 2.85 ± 0.28 -6.20 ± 0.63 -0.73 ± 0.73

Y428A no current

A429V 4 -116.4 ± 5.5 1.93 ± 0.12 -5.05 ± 0.18 0.42 ± 041

M430A 4 -101.0 ± 0.84* 2.76 ± 0.22 -6.34 ± 0.51 -0.87 ± 0.62

F431A 5 -119.7 ± 2.0* 2.15 ± 0.18 -5.81 ± 0.57 -0.34 ± 0.68

I432A 5 -104.5 ± 1.3 2.26 ± 0.39 -5.37 ± 0.96 0.11 ± 1.03

G433A 8 -115.0 ± 2.1* 1.66 ± 0.11* -4.30 ± 0.25* 1.17 ± 0.44

H434A 5 -115.1 ± 2.1* 2.61 ± 0.13 -6.80 ± 0.44* -1.33 ± 0.57

A435V 5 -113.4 ± 4.2 2.29 ± 0.35 -5.77 ± 0.74 -0.29 ± 0.82

T436A 5 -110.7 ± 1.5 2.40 ± 0.24 -6.00 ± 0.56 -0.53 ± 0.67

A437V 6 -120.2 ± 3.5* 2.32 ± 0.21 -6.28 ± 0.54 -0.81 ± 0.65

L438A 8 -120.1 ± 2.1* 2.09 ± 0.07 -5.67 ± 0.22 -0.19 ± 0.42

I439A 5 -90.4 ± 4.9* 2.00 ± 0.17 -4.12 ± 0.43* 1.35 ± 0.41

Q440A 8 -111.0 ± 2.7 2.25 ± 0.34 -5.55 ± 0.76 -0.08 ± 0.85

S441A 6 -110.4 ± 2.2 1.87 ± 0.11 -4.67 ± 0.27 0.79 ± 0.46

L442A 7 -108.5 ± 1.9 1.99 ± 0.17 -4.89 ± 0.47 0.57 ± 0.61

D443A 6 -105.6 ± 2.2 2.20 ± 0.19 -5.27 ± 0.48 0.19 ± 0.60

Table 1B

2 mM cAMP

HCN2 channel n V1/2 (mV) Z -ZFV1/2 (kcal/mol) Δ ZFV1/2 (kcal/mol)

wild type 8 -98.8 ± 2.6 1.84 ± 0.13 -4.10 ± 0.28

I422A 6 -94.9 ± 2.8 1.63 ± 0.11 -3.52 ± 0.32 0.58 ± 0.43

V423A 5 -93.0 ± 2.5 2.02 ± 0.28 -4.22 ± 0.55 -0.12 ± 0.62

G424A no expression

A425V 3 -117.0 ± 5.2* 2.30 ± 0.64 -6.03 ± 1.52* -1.93 ± 1.55

T426A 4 -117.1 ± 3.2* 2.32 ± 0.46 -6.11 ± 1.15* -2.01 ± 1.19

C427A 5 -85.1 ± 1.5* 2.72 ± 0.55* -5.18 ± 0.97 -1.07 ± 1.01

Y428A 5 -104.9 ± 3.6 2.80 ± 0.64* -6.81 ± 1.75* -2.70 ± 1.77

A429V 4 -113.7 ± 3.7* 2.92 ± 0.42* -7.43 ± 0.92* -3.32 ± 0.96

M430A 6 -90.5 ± 2.8* 2.87 ± 0.22* -5.89 ± 0.42* -1.79 ± 0.51

F431A 6 -116.3 ± 3.1* 2.06 ± 0.31 -5.45 ± 0.66* -1.34 ± 0.72

I432A 6 -107.1 ± 5.2 2.00 ± 0.17 -4.77 ± 0.38 -0.67 ± 0.47

G433A 9 -102.7 ± 2.5 1.63 ± 0.09 -3.77 ± 0.22 0.32 ± 0.36

H434A 4 -99.4 ± 2.8 2.49 ± 0.22* -5.62 ± 0.60* -1.52 ± 0.67

A435V 4 -100.7 ± 4.2 2.91 ± 0.24* -6.61 ± 0.55* -2.51 ± 0.62

T436A 5 -105.2 ± 2.2 1.69 ± 0.17 -4.00 ± 0.36 0.09 ± 0.46

A437V 6 -113.3 ± 3.5* 2.17 ± 0.32 -5.43 ± 0.64* -1.48 ± 0.70

L438A 7 -110.4 ± 4.2* 1.52 ± 0.11* -3.76 ± 0.26 0.33 ± 0.38

I439A 4 -78.9 ± 2.4* 1.60 ± 0.16 -2.83 ± 0.23* 1.27 ± 0.36

Q440A 10 -102.2 ± 2.6 1.73 ± 0.14 -3.95 ± 0.24 0.15 ± 0.37

S441A 7 -87.5 ± 2.9* 1.90 ± 0.11 -3.74 ± 0.19 0.36 ± 0.34

L442A 7 -91.8 ± 2.2* 1.70 ± 0.09 -3.53 ± 0.22 0.56 ± 0.36

D443A 6 -86.4 ± 3.6* 1.79 ± 0.18 -3.44 ± 0.24 0.65 ± 0.37

Table 2.1A

Table 2.1B

94

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

95

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).

96

A.

B.

no

rma

lized

cu

rren

t

voltage (mV)

Z

V1/2

C.

D.

ZF

V1

/2

2 mM cAMP

HCN2 Q440A

3 s

1 nA

3 s

1 nA

3 s

1 nA

A437V

0.5 nA

3 s

T426A

HCN2 Q440A

3 s

1 nA

3 s

1 nA

3 s

1 nA

A437V

0.5 nA

3 s

T426A

-150 -130 -110 -90 -70 -50 -30 -10 10

0.0

0.2

0.4

0.6

0.8

1.0

-150 -130 -110 -90 -70 -50 -30 -10 10

0.0

0.2

0.4

0.6

0.8

1.0

I42

2A

V4

23A

G4

24

AA

42

5V

T4

26A

C4

27A

Y4

28A

A4

29V

M4

30

AF

43

1A

I43

2A

G4

33

AH

43

4A

A4

35V

T4

36A

A4

37V

L43

8A

I43

9A

Q4

40

AS

44

1A

L44

2A

D4

43A

I42

2A

V4

23A

G4

24

AA

42

5V

T4

26A

C4

27A

Y4

28A

A4

29V

M4

30

AF

43

1A

I43

2A

G4

33

AH

43

4A

A4

35V

T4

36A

A4

37V

L43

8A

I43

9A

Q4

40

AS

44

1A

L44

2A

D4

43A

-30

-20

-10

0

10

20

30

-2

-1

0

1

2

3

4

-8

-6

-4

-2

0

2

4

6

8

**

*

*

*

**

*

*

**

*

* * **

**

*

* *

* *

* * **

*

*

HCN2

Q440A

A437V

T426A

Figure 2.2

97

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

98

G42

4A

HCN2

N38

0Q

UT

IM

A. B.HCN2

G424A

1 nA

1 nA

3 s

3 s

-150 mV

-35 mV

114136 kDa

GAPDH

-150 mV

-35 mV

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.

99

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).

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

(Equation 2.2)

C C1 C2 C3 C4

O O1 O2 O3 O4

L

K

aK

C C1 C2 C3 C4

O O1 O2 O3 O4

L

K

aK

Po =1

1 + L(V)

1+1/K(V)

1+1/aK(V)

4Po

1

1 + L(V)

1+1/K(V)

1+1/aK(V)

4Po =

1

1 + L(V)

1+1/K(V)

1+1/aK(V)

4Po

1

1 + L(V)

1+1/K(V)

1+1/aK(V)

4

100

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 well-

fitted, 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,

101

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.

102

-160 -140 -120 -100 -80 -60 -40

0

1

2

3

4

5

6

7

8

-160 -140 -120 -100 -80 -60 -400

1

2

3

4

5

6

7

8

Basal cAMP 2 mM cAMPB.A.

Z

V1/2 (mV)

Z

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).

103

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.

Table 2

basal cAMP 2mmcAMP

L=/* L' 0.0001594 L=/* L' 0.0003785 1198 208.4

K=/ * K' 1086 K=/ * K' 13.33 106.4 86.66

z=-z 1.123 z=-z 0.8974

z=-z 0.8437 z=-z 0.9621

a 0.2 a 0.2

r 25.85 r 25.85

L' range 0.01- 50 L' range 0.1- 500

K' range 10̂ -25 - 10̂ 6 K' range 10̂ -15- 10̂ 8

Table 2.2

104

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

105

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

106

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.

107

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.

108

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.

109

Bottom

G433A

G424A

H434A

T426A

F431A

A435VA437V

I439AS6

Y428A

A429VM430A

S5

A425V

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

Top

S5

S6

SF

Figure 2.5

110

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.

111

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.

112

2.6 References


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 hyperpolarization-

activated 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 S4-

S5 linker couples voltage sensing and activation of pacemaker channels. Proc Natl Acad Sci U

S A 98, 11277-11282.

113

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. 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.



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.

114




Gauss, R., Seifert, R., and Kaupp, U.B. (1998). Molecular identification of a

hyperpolarization-activated channel in sea urchin sperm. Nature 393, 583-587.


region of HCN channels. Biophys J 89, 932-944.


gate in the shaker K+ channel. J Gen Physiol 119, 521-532.



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


115




hyperpolarization-activated cyclic nucleotide-gated channels. J Biol Chem 279, 16832-16846.


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.

116

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, 83-

91.


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.

117


hyperpolarization-activated cation channels: evidence for an intracellular activation gate. J Gen

Physiol 117, 91-101.


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, 805-

810.



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.

118

3. The unique form and function of the HCN channel selectivity filter2

3.1 Introduction


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.

119

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 appropriately-

spaced 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

120

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 side-

chain 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).

121

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).

122

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,

123

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 voltage-

independent 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-

124

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(V

1/2-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

125

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).

126

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

127

HCN2

T400

leakage

0.5 nA

0.5 s

0 nA

Instantaneous If-60 mV

-60 mV

3530

-150

5

-150

30-

C.A.

HCN1 CIGYG

HCN2 CIGYG

HCN3 CIGYG

HCN4 CIGYG

Kir1.1 TIGYG

Kir2.1 TIGYG

Kir3.1 TIGYG

Kir3.4 TIGYG

KCNQ1 TIGYG

SK SIGYG

BK TVGYG

Shaker TVGYG

Kv1.2 TVGYG

Kv1.5 TVGYG

Kv2.1 TVGYG

KvAP TVGYG

KcsA TVGYG

Mthk TVGYG

Kv3.1 TLGYG

Kv4.2 TLGYG

Kat1 TTGYG

B.

0 nA

test voltage (mV)

Insta

nta

neous

I (pA

/pF

)f

HCN2

T400

-60 -50 -40 -30 -20 -10 10 20 30

-30

-20

-10

10

20

30

40

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).

128

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

129

Na+ only

-150 mV

-35 mV

HCN2

50 pA/pF

0.5 s

-150 mV

-35 mV

T400

A. K+ only

-150 mV-35 mV

-150 mV

-35 mV

250 pA/pF

0.5 s

HCN2

T400

B.

-600

-500

-400

-300

-200

-100

0

Na+ only

K+ only

HCN2

HCN2

T400

T400

(6) (7)

(6)

(7)

I at

-150 m

V (

pA

/pF

)f

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).

130

A.

-800

-700

-600

-500

-400

-300

-200

-100

0

I

at

-15

0 m

V (

pA

/pF

)

(6)

(6)

(6)

(6)

[5.4

K]o

[135

K]of

HCN2

T400

[5.4

K]o

[135

K]oB.

0

2

4

6

8

10

12

14

Fold

incre

ase in I

at -1

50 m

V

HCN2 T400

f

(6)

(6)

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.

131

HCN2 T400

[5.4K/135Na]o

[135K/5.4Na]o

[5.4K/135Na]o

[135K/5.4Na]o

200 pA/pF

0.5 s

200 pA/pF

0.5 s

A.

I at

-150 m

V (

pA

/pF

)f

(6)(6)

(6)

(6)

-150 mV

-35 mV

-150 mV

[5.4

K/135

Na]

o

[135

K/5.4

Na]

o

[5.4

K/135

Na]

o

[135

K/5.4

Na]

o

HCN2

T400

-35 mV

-150 mV

-600

-500

-400

-300

-200

-100

0

0

1

2

3

4

5

6

7

B. C.

Fold

incre

ase in I

at -1

50 m

Vf

HCN2 T400

(6)

(6)

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.

132

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

133

Ins

tan

tan

eo

us

I (pA

/pF

)f

Ins

tan

tan

eo

us

I (pA

/pF

)f

0.03 mM Cs+0.03 mM Cs+

-150 -120 -90 -60 -30 30

-1000

-800

-600

-400

-200

200

-150 -120 -90 -60 -30 30

-500

-400

-300

-200

-100

100test voltage (mV) test voltage (mV)

B. C.

A.

IC50

[Cs+] (mM)

I /I (-6

0 m

V)

HCN2

T400

HCN2

T400

test voltage (mV)

0.1 1

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

-150 -120 -90 -60 -30 0 30

0.0

0.4

0.8

1.2

1.6

2.0

T400HCN2

Cs+

(mM)

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 If-

V 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.

134

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

135

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+

136

A.[5.4K/135Na]o

[135K/5.4Na]o

[10K/130Na]i

-150 mV

-150 mV

200 pA/pF

0.5 s

-35 mV

[5.4K/135Na]o

[135K/5.4Na]o

[10K/130Na]i

-150 mV

-35 mV

-150 mV

200 pA/pF

0.5 s

B.

-600

-500

-400

-300

-200

-100

0

HCN2 T400

I

at

-150

mV

(p

A/p

F)

f

(6) (6)

(6)

(6)

[5.4

K/135

Na]

o

[135

K/5.4

Na]

o

[5.4

K/135

Na]

o

[135

K/5.4

Na]

o

HCN2

T400

C. D.

0

2

4

6

8

10

12

14

Fo

ld in

cre

ase

in

I

at

-15

0 m

Vf

(6)

(6)

HCN2 T400

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.

137

(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.

138

0.2

0.4

0.6

0.8

1.0

0.0

I /I (-1

50 m

V)

Cs+

T400

0.03 mM Cs+

0.03 mM Cs+

-35 mV

-150 mV

100 pA/pF

0.2 s

-150 mV

[130K/10Na]i(6)

(4)

[130

K/10N

a]i

[10K

/130

Na]

i

-35 mV

A. B.

[10K/130Na]i

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).

139

HCN2

T400

T400

200 pA/pF

1 s

200 pA/pF

1 s

200 pA/pF

1 s

-150 -130 -110 -90 -70 -50 -30 -10

0.0

0.2

0.4

0.6

0.8

1.0

no

rma

lize

d I

f

test voltage (mV)

A.

HCN2

T400

T400

[130K/10Na]i

[130K/10Na]i

[10K/130Na]i

B.

[130K/10Na]i

[130K/10Na]i

[10K/130Na]i

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).

140

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 GYG-

containing 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.

141

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 potassium-

only 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.

142

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 outward-

143

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.

144

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.

145

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.


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


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.


cation channels: from genes to function. Physiol Rev 89, 847-885.

146

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 hGH-

receptor interactions by alanine-scanning mutagenesis. Science 244, 1081-1085.


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.

147

DiFrancesco, D. (1986). Characterization of single pacemaker channels in cardiac sino-atrial

node cells. Nature 324, 470-473.




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.


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., and MacKinnon, R. (1993). Conduction properties of the cloned Shaker

K+ channel. Biophys J 65, 2089-2096.

148

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).




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.



Chemistry 279, 16832-16846.

149


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, 27056-

27061.



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.

150



5109.

Robertson, J.L., Palmer, L.G., and Roux, B. (2008). Long-pore electrostatics in inward-

rectifier potassium channels. The Journal of General Physiology 132, 613-632.


from molecules to physiological function. Annu Rev Physiol 65, 453-480.

Roncaglia, P., Mistrik, P., and Torre, V. (2002). Pore topology of the hyperpolarization-

activated cyclic nucleotide-gated channel from sea urchin sperm. Biophys J 83, 1953-1964.




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.

151



comprehensive alignment and the KcsA channel structure. Biophys J 84, 2929-2942.


hyperpolarization-activated cation channels: evidence for an intracellular activation gate. The


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, 1668-

1674.

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.

152

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, 687-

708.

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.





Nature 414, 43-48.

153

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.

154

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

155

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 V1/2

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

156

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.

157

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.

158

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).

159

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.

ShakerPore

O

OC

C

stable

unstable

HCNPore

hyperpolarized

depolarized

Energy

ShakerPore

O

OC

C

stable

unstable

HCNPore

hyperpolarized

depolarized

Energy

160

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

161

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

162

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

163

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; Morais-

Cabral 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

164

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

165

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

166

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 voltage-

sensor 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

167

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

168

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.

169

4.11 References


DiFrancesco, D. (2001). Integrated allosteric model of voltage gating of HCN channels. The



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, 349-

356.

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.

170

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


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.


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


171

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


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 guinea-

pig atria. Exp Physiol 86, 19-25.


in K channels fail to confer selectivity to rabbit HCN4 channels. PloS One 4, e7712.

172



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.




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.


region of HCN channels. Biophysical Journal 89, 932-944.

173


gate in the shaker K+ channel. The Journal of General Physiology 119, 521-532.



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, 617-

621.



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


Jiang, Y., Lee, A., Chen, J., Ruta, V., Cadene, M., Chait, B.T., and MacKinnon, R. (2003).


174

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.



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.

175

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.



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.



5109.

Proenza, C., and Yellen, G. (2006). Distinct populations of HCN pacemaker channels

produce voltage-dependent and voltage-independent currents. The Journal of General


176

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.


from molecules to physiological function. Annual Review of Physiology 65, 453-480.





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.



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.

177

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 hyperpolarization-

activated 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, 539-

550.

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



channel. The Journal of General Physiology 123, 21-32.

178

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, 35-

42.



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, 427-

429.

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.



179

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.



Nature 414, 43-48.

180

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.

http://www.ncbi.nlm.nih.gov/pubmed?term=%22Demos%20MK%22%5BAuthor%5D

http://www.ncbi.nlm.nih.gov/pubmed?term=%22Macri%20V%22%5BAuthor%5D

http://www.ncbi.nlm.nih.gov/pubmed?term=%22Farrell%20K%22%5BAuthor%5D

http://www.ncbi.nlm.nih.gov/pubmed?term=%22Nelson%20TN%22%5BAuthor%5D

http://www.ncbi.nlm.nih.gov/pubmed?term=%22Chapman%20K%22%5BAuthor%5D

http://www.ncbi.nlm.nih.gov/pubmed?term=%22Accili%20E%22%5BAuthor%5D

http://www.ncbi.nlm.nih.gov/pubmed?term=%22Armstrong%20L%22%5BAuthor%5D

A Novel KCNA1 MutationAssociated with Global Delay andPersistent 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

1Department of Pediatric Neurology, British Columbia’sChildren’s Hospital, Vancouver, British Columbia, Canada;

2Department of Cellular and Physiological Sciences,University of British Columbia, Vancouver, BritishColumbia, Canada; 3Department of Pathology

and Laboratory Medicine, Children’s and Women’s HealthCenter of British Columbia, Vancouver, British Columbia,Canada; 4Division of Neurology, Neuromuscular DiseaseUnit, Vancouver Hospital, Vancouver, British Columbia,

Canada; 5Department of Medical Genetics, Children’s andWomen’s Health Center of British Columbia, Vancouver,

British Columbia, Canada

Abstract: Episodic Ataxia Type 1 is an autosomal domi-nant disorder characterized by episodes of ataxia andmyokymia. It is associated with mutations in the KCNA1voltage-gated potassium channel gene. In the presentstudy, we describe a family with novel clinical featuresincluding persistent cerebellar dysfunction, cerebellar at-rophy, and cognitive delay. All affected family membershave myokymia and epilepsy, but only one individual hasepisodes of vertigo. Additional features include posturalabnormalities, episodic stiffness and weakness. A novelKCNA1 mutation (c.1222G>T) which replaces a highlyconserved valine with leucine at position 408(p.Val408Leu) was identified in affected family members,and was found to augment the ability of the channel toinactivate. Together, our data suggests that KCNA1 muta-tions are associated with a broader clinical phenotype,which may include persistent cerebellar dysfunction andcognitive delay. � 2009 Movement Disorder Society

Key words: KCNA1; EA1; cerebellar atrophy; cognitivedysfunction

Episodic Ataxia type 1 (EA1) is a rare autosomal

dominant disorder associated with KCNA1 mutations

that presents in childhood with brief episodes of ataxia

and continuous myokymia.1,2 The clinical spectrum of

EA1 has expanded to include epilepsy, episodes of mus-

cle stiffness, postural abnormalities and weakness.2–8

Persistent cerebellar dysfunction with cerebellar atrophy

is typically absent in patients with EA19 but is a charac-

teristic feature of Episodic Ataxia Type 2 (EA2), which

is associated with mutations in the P/Q-type voltage-

gated 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 epi-

sodes of muscle stiffness triggered by illnesses. The

seizures started in infancy and are controlled on carba-

mazepine. 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. Ex-

amination 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 demon-

strated bilateral epileptiform activity.Patient III-2 (Fig. 1A) is a 14-mo old boy with seiz-

ures, myokymia and mild global developmental delay.

Seizures began at 3 wk and are controlled on carbama-

zepine. 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, myoky-

mia, 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, trig-

gered 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

Potential conflict of interest: None reported.Received 24 October 2008; Accepted 24 December 2009Published online 9 February 2009 in Wiley InterScience (www.

interscience.wiley.com). DOI: 10.1002/mds.22467

*Correspondence to: Dr. Michelle K. Demos, Department of Pedi-atric Neurology, British Columbia’s Children’s Hospital, K3-1764480 Oak St., Vancouver, British Columbia, Canada, V6H 3V4.E-mail: [email protected]

778 M.K. DEMOS ET AL.

Movement Disorders, Vol. 24, No. 5, 2009

181

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 myo-

tonic 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 soft-

ware (Applied Biosystems, Foster City, CA) was used

for comparative analysis of resulting sequence to

KCNA1 consensus sequence (NM_000217). Genotyp-

ing of familial samples was performed using AmpfIstrIdentifiler 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-

FIG. 1. Pedigree and clinical features. (A) Pedigree of family. Blackened symbols represent affected individuals. DNA available from numberedindividuals. (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 clenchedfists. (D) Patient II-1 head MRI at age 17 yr demonstrating cerebellar atrophy. (E) Sequencing of KCNA1 revealed heterozygosity for a nucleotidetransversion (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.]

779CEREBELLAR ATROPHY IN EPISODIC ATAXIA TYPE 1


182

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). Pip-

ettes 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 withKOH. 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 am-

plifier 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-

FIG. 2. Inactivation of the human KCNA1 channel is enhanced by the V408L mutation. (A) Representative current traces from CHO cells trans-fected with wildtype (black line) and mutant (gray line) channels elicited by 8 voltage pulses to 110 mV, 140 mV, and 170 mV from a holdingpotential of 280 mV. Traces are normalized to their maximum (peak) values. (B) Plot of time constants of inactivation (s) determined from a sin-gle exponential fitting procedure of current traces obtained from cells expressing the wildtype (filled bars) or mutant (unfilled bars) channels atthe three test potentials shown in A. s values were significantly faster for the mutant compared with wildtype channels (t-test, P < 0.05). Thenumbers 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) channelsat 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 paren-theses 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 compensatedand currents were not leak-subtracted.



183

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 trans-

mitted to her offspring (III-1 and III-2).Because a mutation of valine 408 to alanine was

previously found to enhance KCNA1 channel inactiva-

tion,14 this behavior was analyzed in CHO cells trans-

fected 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 volt-

age 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 pro-

band’s mother (II-1) has persistent cerebellar dysfunc-

tion associated with cerebellar atrophy on neuroimag-

ing. The proband (III-1) also has mild gait ataxia. Past

reports of patients with EA1 have described mild cere-

bellar dysfunction in some affected family members.

Findings included intention tremor and mild difficulties

with tandem gait and/or arm coordination.3,15,16 In con-

trast 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 at-

rophy, a feature which has not been reported previ-

ously 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 chil-

dren 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-

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 dys-

function 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 dis-

rupts 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 pheno-

type 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 2Canada Research Chair. V. Macri is the recipient of doctoralfellowships from the Canadian Institutes of Health Researchand the Michael Smith Foundation for Health Research. Wethank the patients and their families for their participation inthis study, Dr. J. Jen for her assistance, and Sarah Chow forher technical assistance with preparation of KCNA1 DNA fortransfection.

781CEREBELLAR ATROPHY IN EPISODIC ATAXIA TYPE 1


184

Author Roles: Michelle Demos: This author (first authorand corresponding author) was involved in conception, orga-nization, and execution of this case study, both in terms ofclinical information and genetic studies. She also wrote thefirst draft excluding small portions of genetic and functionalstudies sections; Vincenzo Macri: This author was involvedin conception, organization, and execution of functional stud-ies. He also provided statistical expertise related to the func-tional studies. He also participated in review and critique ofthe manuscript; Kevin Farrell: This author supervised andwas involved in the collection of clinical data and conceptionand organization of information for presentation as a casestudy. He also reviewed and critiqued multiple drafts of themanuscript; Tanya Nelson: This author supervised and wasinvolved in conception, organization, and execution ofgenetic studies. She also wrote the genetics section andreviewed and critiqued manuscript; Kristine Chapman: Thisauthor was involved in collection of clinical data, specificallyneurophysiology data and conception and design of clinicalreport. She also participated in review and critique of themanuscript; Eric Accili: This author supervised and wasinvolved in conception, organization, and execution of func-tional studies. He wrote and provided figures for the func-tional studies section. He also reviewed and critiqued themanuscript; Linlea Armstrong: This author supervised andwas involved in conception, organization, and execution ofthis case study, both in terms of clinical information andgenetic and functional studies. She also reviewed and cri-tiqued the manuscript.

REFERENCES

1. Browne DL, Gancher ST, Nutt JG, et al. Episodic ataxia/myoky-mia syndrome is associated with point mutations in the humanpotassium channel gene, KCNA1. Nat Genet 1994;8:136–140.

2. Van Dyke DH, Griggs RC, Murphy MJ, Goldstein MN. Heredi-tary myokymia and periodic ataxia. J Neurol Sci 1975;25:109–118.

3. Brunt ERP, van Weerden TW. Familial paroxysmal kinesigenicataxia and continuous myokymia. Brain 1990;113:1361–1382.

4. Zuberi SM, Eunson LH, Spauschus A, et al. A novel mutation inthe human voltage-gated potassium channel gene (Kv1.1) associ-ates with episodic ataxia type 1 and sometimes with partial epi-lepsy. Brain 1999;122:817–825.

5. Eunson LH, Rea R, Zuberi SM, et al. Clinical, genetic, andexpression studies of mutations in the potassium channel gene

KCNA1 reveal new phenotypic variability. Ann Neurol 2000;48:647–656.

6. 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.

7. Klein A, Boltshauser E, Jen J, Baloh RW. Episodic ataxia type 1with distal weakness: a novel manifestation of a potassium chan-nelopathy. Neuropediatrics 2004;35:147–149.

8. Kinali M, Jungbluth H, Eunson LH, et al. Expanding the pheno-type of potassium channelopathy: severe neuromyotonia andskeletal deformities without prominent Episodic Ataxia. Neuro-muscul Disord 2004;14:689–693.

9. Rajakulendran S, Schorge S, Kullman DM, Hanna MG. Episodicataxia type 1: a neuronal potassium channelopathy. Neurothera-peutics 2007;4:258–266.

10. Vighetto A, Froment JC, Trillet M, Aimard G. Magnetic reso-nance imaging in familial paroxysmal ataxia. Arch Neurol 1988;45:547–549.

11. Ophoff RA, Terwindt GM, Vergouwe MN, et al. Familial hemi-plegic migraine and episodic ataxia type-2 are caused by muta-tions in the Ca21 channel gene CACNL1A4. Cell 1996;87:543–552.

12. Lee H, Wang H, Jen JC, Sabatti C, Baloh RW, Nelson SF. Anovel mutation in KCNA1 causes episodic ataxia without myo-kymia. Hum Mutat 2004;24:536.

13. Macri V, Accili EA. Structural elements of instantaneous andslow gating in hyperpolarization-activated cyclic nucleotide-gatedchannels. J Biol Chem 2004;279:16832–16846.

14. Adelman JP, Bond CT, Pessia M, Maylie J. Episodic ataxiaresults from voltage-dependent potassium channels with alteredfunctions. Neuron 1995;15:1449–1454.

15. Hand PJ, Gardner RJM, Knight MA, Forrest SM, Storey E. Clin-ical features of a large Australian pedigree with episodic ataxiatype 1. Mov Disord 2001;16:938–939.

16. Hanson PA, Martinez LB, Cassidy R. Contractures, continuousmuscle discharges, and titubation. Ann Neurol 1977;1:120–124.

17. Ney GC, Lantos G, Barr WB, Schaul N. Cerebellar atrophy inpatients with long-term phenytoin exposure and epilepsy. ArchNeurol 1994;51:767–771.

18. De Marco FA, Ghizoni E, Kobayashi E, Li LM, Cendes F. Cere-bellar volume and long-term use of phenytoin. Seizure 2003;12:312–315.

19. Zhou L, Zhang CL, Messing A, Chiu SY. Temperature-sensitiveneurmuscular transmission in Kv1.1 Null Mice: Role of potas-sium channels under the myelin sheath in young nerves. J Neuro-sci 1998;18:7200–7215.

20. Jen JC, Hess EJ, Hanna MG, Griggs RC, Baloh RW, CINCHinvestigators. Primary episodic ataxias: diagnosis, pathogenesisand treatment. Brain 2007;130:2484–2493.



185

186

Appendix B

Biohazard Approval Certificate

The University of British Columbia

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

Ubc 2010 Fall Macri Vincenzo

Documents

Transcript of Ubc 2010 Fall Macri Vincenzo