Ubc 2010 Fall Macri Vincenzo
-
Upload
delia-pitariu -
Category
Documents
-
view
20 -
download
1
Transcript of 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.
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
Physiology 530, 21-33.
Chen, J., Mitcheson, J.S., Lin, M., and Sanguinetti, M.C. (2000). Functional roles of charged
residues in the putative voltage sensor of the HCN2 pacemaker channel. The Journal of
Biological Chemistry 275, 36465-36471.
Chen, J., Mitcheson, J.S., Tristani-Firouzi, M., Lin, M., and Sanguinetti, M.C. (2001). The
S4-S5 linker couples voltage sensing and activation of pacemaker channels. Proceedings of
the National Academy of Sciences of the United States of America 98, 11277-11282.
Chen, S., Wang, J., Zhou, L., George, M.S., and Siegelbaum, S.A. (2007). Voltage sensor
movement and cAMP binding allosterically regulate an inherently voltage-independent
closed-open transition in HCN channels. The Journal of General Physiology 129, 175-188.
Cheng, L., Kinard, K., Rajamani, R., and Sanguinetti, M.C. (2007). Molecular mapping of
the binding site for a blocker of hyperpolarization-activated, cyclic nucleotide-modulated
pacemaker channels. The Journal of Pharmacology and Experimental Therapeutics 322, 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
Physiology 308, 353-367.
DiFrancesco, D., and Tortora, P. (1991). Direct activation of cardiac pacemaker channels by
intracellular cyclic AMP. Nature 351, 145-147.
DiFrancesco, D., and Tromba, C. (1988a). Inhibition of the hyperpolarization-activated
current (if) induced by acetylcholine in rabbit sino-atrial node myocytes. The Journal of
Physiology 405, 477-491.
DiFrancesco, D., and Tromba, C. (1988b). Muscarinic control of the 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
Physiology 115, 193-208.
Gonzalez, C., Rosenman, E., Bezanilla, F., Alvarez, O., and Latorre, R. (2001). Periodic
perturbations in Shaker K+ channel gating kinetics by deletions in the S3-S4 linker.
Proceedings of the National Academy of Sciences of the United States of America 98, 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
Biological Chemistry 278, 50724-50731.
Larsson, H.P., Baker, O.S., Dhillon, D.S., and Isacoff, E.Y. (1996). Transmembrane
movement of the shaker K+ channel S4. Neuron 16, 387-397.
Lesso, H., and Li, R.A. (2003). Helical secondary structure of the external S3-S4 linker of
pacemaker (HCN) channels revealed by site-dependent perturbations of activation
phenotype. The Journal of Biological Chemistry 278, 22290-22297.
Lipsius, S.L., and Bers, D.M. (2003). Cardiac pacemaking: I(f) vs. Ca(2+), is it really that
simple? Journal of Molecular and Cellular Cardiology 35, 891-893.
Liu, Y., Holmgren, M., Jurman, M.E., and Yellen, G. (1997). Gated access to the pore of a
voltage-dependent K+ channel. Neuron 19, 175-184.
Long, S.B., Campbell, E.B., and Mackinnon, R. (2005). Crystal structure of a mammalian
voltage-dependent Shaker family K+ channel. Science 309, 897-903.
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
Physiology 431, 291-318.
Milanesi, R., Baruscotti, M., Gnecchi-Ruscone, T., and DiFrancesco, D. (2006). Familial
sinus bradycardia associated with a mutation in the cardiac pacemaker channel. N Engl J
Med 354, 151-157.
Mistrik, P., Mader, R., Michalakis, S., Weidinger, M., Pfeifer, A., and Biel, M. (2005). The
murine HCN3 gene encodes a hyperpolarization-activated cation channel with slow kinetics
and unique response to cyclic nucleotides. The Journal of Biological Chemistry 280, 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
Physiology 113, 17-34.
Pape, H.C. (1996). Queer current and pacemaker: the hyperpolarization-activated cation
current in neurons. Annual Review of Physiology 58, 299-327.
Paterson, D.J. (1996). Role of potassium in the regulation of systemic physiological function
during exercise. Acta physiologica Scandinavica 156, 287-294.
Pathak, M.M., Yarov-Yarovoy, V., Agarwal, G., Roux, B., Barth, P., Kohout, S., Tombola,
F., and Isacoff, E.Y. (2007). Closing in on the resting state of the Shaker K(+) channel.
Neuron 56, 124-140.
Posson, D.J., Ge, P., Miller, C., Bezanilla, F., and Selvin, P.R. (2005). Small vertical
movement of a K+ channel voltage sensor measured with luminescence energy transfer.
Nature 436, 848-851.
Proenza, C., Angoli, D., Agranovich, E., Macri, V., and Accili, E.A. (2002a). Pacemaker
channels produce an instantaneous current. The Journal of Biological Chemistry 277, 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
Biological Chemistry 277, 29634-29642.
Proenza, C., and Yellen, G. (2006). Distinct populations of HCN pacemaker channels
produce voltage-dependent and voltage-independent currents. The Journal of General
Physiology 127, 183-190.
Prole, D.L., and Yellen, G. (2006). Reversal of HCN channel voltage dependence via
bridging of the S4-S5 linker and Post-S6. The Journal of General Physiology 128, 273-282.
Robinson, R.B., and Siegelbaum, S.A. (2003). Hyperpolarization-activated cation currents:
from molecules to physiological function. Annual Review of Physiology 65, 453-480.
Robinson, R.B., Yu, H., Chang, F., and Cohen, I.S. (1997). Developmental change in the
voltage-dependence of the pacemaker current, if, in rat ventricle cells. Pflugers Arch 433,
533-535.
Roncaglia, P., Mistrik, P., and Torre, V. (2002). Pore topology of the 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
Journal of General Physiology 117, 91-101.
Solomon, J.S., and Nerbonne, J.M. (1993). Hyperpolarization-activated currents in isolated
superior colliculus-projecting neurons from rat visual cortex. The Journal of Physiology 462,
393-420.
Stieber, J., Herrmann, S., Feil, S., Loster, J., Feil, R., Biel, M., Hofmann, F., and Ludwig, A.
(2003a). The hyperpolarization-activated channel HCN4 is required for the generation of
pacemaker action potentials in the embryonic heart. Proceedings of the National Academy of
Sciences of the United States of America 100, 15235-15240.
Stieber, J., Stockl, G., Herrmann, S., Hassfurth, B., and Hofmann, F. (2005). Functional
expression of the human HCN3 channel. The Journal of Biological Chemistry 280, 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
Biological Chemistry 277, 18994-19000.
Tsang, S.Y., Lesso, H., and Li, R.A. (2004). Dissecting the structural and functional roles of
the S3-S4 linker of pacemaker (hyperpolarization-activated cyclic nucleotide-modulated)
channels by systematic length alterations. The Journal of Biological Chemistry 279, 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
hyperpolarization-activated cyclic nucleotide-gated channels. The Journal of Biological
Chemistry 276, 29930-29934.
Wahl-Schott, C., and Biel, M. (2009). HCN channels: structure, cellular regulation and
physiological function. Cell Mol Life Sci 66, 470-494.
Wainger, B.J., DeGennaro, M., Santoro, B., Siegelbaum, S.A., and Tibbs, G.R. (2001).
Molecular mechanism of cAMP modulation of HCN pacemaker channels. Nature 411, 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
Physiology 61, 687-708.
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
Hyperpolarization-activated Cyclic Nucleotide-modulated (HCN) channels are similar in
structure and function to Shaker K+ channels (Gauss et al., 1998; Ludwig et al., 1998;
Santoro et al., 1998). As in Shaker, HCN channels are comprised of 4 subunits which each
consist of six predicted membrane-spanning segments (S1-S6). The S1-S4 segments form the
voltage-sensing domain, and the S5 and S6 segments, the pore-forming domain. The S4
segment in both channels contains positive charges that move similarly in response to
changes in membrane voltage (Bell et al., 2004; Mannikko et al., 2002; Vemana et al., 2004),
to then alter the pore structure at the intracellular side of the S6 segment; this region
functions as a voltage-controlled gate to cation flow (Giorgetti et al., 2005; Macri et al.,
2002; Rothberg et al., 2003; Shin et al., 2001). Despite these similarities, HCN channels are
opened by hyperpolarization of the membrane potential, whereas Shaker channels open in
response to depolarization. Thus, the electromechanical coupling between the voltage sensor
and the gate is reversed in these two channels.
A key determinant of this coupling is the intrinsic stability of the closed and open
conformations of the pore. In Shaker channels, it has been proposed that the pore is
1 A version of this chapter has been published. Macri, V, Nazzari, H, McDonald, E, Accili, EA. (2009) Alanine
scanning of the S6 segment reveals a unique and cyclic AMP-sensitive association between the pore and
voltage-dependent opening in HCN channels. Journal of Biological Chemistry, 284: 15659-67.
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
Altomare, C., Bucchi, A., Camatini, E., Baruscotti, M., Viscomi, C., Moroni, A., and
DiFrancesco, D. (2001). Integrated allosteric model of voltage gating of HCN channels. J Gen
Physiol 117, 519-532.
Barbuti, A., Baruscotti, M., Altomare, C., Moroni, A., and DiFrancesco, D. (1999). Action of
internal pronase on the f-channel kinetics in the rabbit SA node. J Physiol 520 Pt 3, 737-744.
Bell, D.C., Yao, H., Saenger, R.C., Riley, J.H., and Siegelbaum, S.A. (2004). Changes in local
S4 environment provide a voltage-sensing mechanism for mammalian 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.
Decher, N., Chen, J., and Sanguinetti, M.C. (2004). Voltage-dependent gating of
hyperpolarization-activated, cyclic nucleotide-gated pacemaker channels: molecular coupling
between the S4-S5 and C-linkers. J Biol Chem 279, 13859-13865.
DiFrancesco, D. (1999). Dual allosteric modulation of pacemaker (f) channels by cAMP and
voltage in rabbit SA node. J Physiol 515 ( Pt 2), 367-376.
DiFrancesco, D., and Tortora, P. (1991). Direct activation of cardiac pacemaker channels by
intracellular cyclic AMP. Nature 351, 145-147.
114
Doyle, D.A., Morais Cabral, J., Pfuetzner, R.A., Kuo, A., Gulbis, J.M., Cohen, S.L., Chait,
B.T., and MacKinnon, R. (1998). The structure of the potassium channel: molecular basis of
K+ conduction and selectivity. Science 280, 69-77.
Gauss, R., Seifert, R., and Kaupp, U.B. (1998). Molecular identification of a
hyperpolarization-activated channel in sea urchin sperm. Nature 393, 583-587.
Giorgetti, A., Carloni, P., Mistrik, P., and Torre, V. (2005). A homology model of the pore
region of HCN channels. Biophys J 89, 932-944.
Hackos, D.H., Chang, T.H., and Swartz, K.J. (2002). Scanning the intracellular S6 activation
gate in the shaker K+ channel. J Gen Physiol 119, 521-532.
Ishii, T.M., Takano, M., Xie, L.H., Noma, A., and Ohmori, H. (1999). Molecular
characterization of the hyperpolarization-activated cation channel in rabbit heart sinoatrial
node. J Biol Chem 274, 12835-12839.
Jiang, Y., Lee, A., Chen, J., Cadene, M., Chait, B.T., and MacKinnon, R. (2002a). Crystal
structure and mechanism of a calcium-gated potassium channel. Nature 417, 515-522.
Jiang, Y., Lee, A., Chen, J., Cadene, M., Chait, B.T., and MacKinnon, R. (2002b). The open
pore conformation of potassium channels. Nature 417, 523-526.
115
Ludwig, A., Zong, X., Jeglitsch, M., Hofmann, F., and Biel, M. (1998). A family of
hyperpolarization-activated mammalian cation channels. Nature 393, 587-591.
Macri, V., and Accili, E.A. (2004). Structural elements of instantaneous and slow gating in
hyperpolarization-activated cyclic nucleotide-gated channels. J Biol Chem 279, 16832-16846.
Macri, V., Proenza, C., Agranovich, E., Angoli, D., and Accili, E.A. (2002). Separable gating
mechanisms in a Mammalian pacemaker channel. J Biol Chem 277, 35939-35946.
Mannikko, R., Elinder, F., and Larsson, H.P. (2002). Voltage-sensing mechanism is conserved
among ion channels gated by opposite voltages. Nature 419, 837-841.
Much, B., Wahl-Schott, C., Zong, X., Schneider, A., Baumann, L., Moosmang, S., Ludwig, A.,
and Biel, M. (2003). Role of subunit heteromerization and N-linked glycosylation in the
formation of functional hyperpolarization-activated cyclic nucleotide-gated channels. The
Journal of Biological Chemistry 278, 43781-43786.
Nazzari, H., Angoli, D., Chow, S.S., Whitaker, G., Leclair, L., McDonald, E., Macri, V.,
Zahynacz, K., Walker, V., and Accili, E.A. (2008). Regulation of cell surface expression of
functional pacemaker channels by a motif in the B-helix of the cyclic nucleotide-binding
domain. American Journal of Physiology 295, C642-652.
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.
Rothberg, B.S., Shin, K.S., and Yellen, G. (2003). Movements near the gate of a
hyperpolarization-activated cation channel. J Gen Physiol 122, 501-510.
Santoro, B., Liu, D.T., Yao, H., Bartsch, D., Kandel, E.R., Siegelbaum, S.A., and Tibbs, G.R.
(1998). Identification of a gene encoding a hyperpolarization-activated pacemaker channel of
brain. Cell 93, 717-729.
Schoppa, N.E., and Sigworth, F.J. (1998). Activation of Shaker potassium channels. III. An
activation gating model for wild-type and V2 mutant channels. J Gen Physiol 111, 313-342.
Shin, K.S., Maertens, C., Proenza, C., Rothberg, B.S., and Yellen, G. (2004). Inactivation in
HCN channels results from reclosure of the activation gate: desensitization to voltage. Neuron
41, 737-744.
117
Shin, K.S., Rothberg, B.S., and Yellen, G. (2001). Blocker state dependence and trapping in
hyperpolarization-activated cation channels: evidence for an intracellular activation gate. J Gen
Physiol 117, 91-101.
Vemana, S., Pandey, S., and Larsson, H.P. (2004). S4 movement in a mammalian HCN
channel. J Gen Physiol 123, 21-32.
Wainger, B.J., DeGennaro, M., Santoro, B., Siegelbaum, S.A., and Tibbs, G.R. (2001).
Molecular mechanism of cAMP modulation of HCN pacemaker channels. Nature 411, 805-
810.
Yifrach, O., and MacKinnon, R. (2002). Energetics of pore opening in a voltage-gated K(+)
channel. Cell 111, 231-239.
Zagotta, W.N., Hoshi, T., and Aldrich, R.W. (1994). Shaker potassium channel gating. III:
Evaluation of kinetic models for activation. J Gen Physiol 103, 321-362.
Zhou, L., and Siegelbaum, S.A. (2007). Gating of HCN channels by cyclic nucleotides: residue
contacts that underlie ligand binding, selectivity, and efficacy. Structure 15, 655-670.
118
3. The unique form and function of the HCN channel selectivity filter2
3.1 Introduction
Hyperpolarization-activated Cyclic Nucleotide-modulated (HCN) channels are similar in
structure and function to potassium-selective channels (Biel et al., 2009; Robinson and
Siegelbaum, 2003). HCN channels pass predominantly potassium, are blocked by millimolar
levels of cesium ions (Hille, 2001; Ludwig et al., 1999; Mistrik et al., 2005; Moroni et al.,
2000) and activated by extracellular potassium (Heginbotham and MacKinnon, 1993; Ludwig
et al., 1998; Macri and Accili, 2004; Macri et al., 2002; Moroni et al., 2000; Sakmann and
Trube, 1984; Stampe et al., 1998; Yang and Sigworth, 1998). Differences in permeation also
exist between HCN and potassium channels. HCN channels are only minimally inhibited by
barium or TEA (DiFrancesco, 1981a, b; Ludwig et al., 1998; Wollmuth and Hille, 1992), both
of which are strong blockers of potassium channels (Hille, 2001). Sodium ordinarily passes in
significant amounts in HCN channels, although it is less permeable than potassium, and passes
only when potassium is present on the same side of the plasma membrane (DiFrancesco,
1981b; Ludwig et al., 1998; Moroni et al., 2000; Pape, 1996). Because larger organic cations
permeate (D'Avanzo et al., 2009; Wollmuth and Hille, 1992), the minimum diameter of the
HCN pore may be wider than that of potassium channels (Doyle et al., 1998; Hille, 2001).
Finally, the single channel conductance of HCN channels is very small, less than 2 pS when
measured in very high concentrations of potassium (Dekker and Yellen, 2006; DiFrancesco,
1986), as compared to 5-50 pS for potassium channels measured at physiological potassium
2 A version of this chapter has been submitted for publication. Macri, V, Angoli, D, Accili, EA. The unique form and
function of the HCN channel selectivity filter.
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.
Aqvist, J., and Luzhkov, V. (2000). Ion permeation mechanism of the potassium channel.
Nature 404, 881-884.
Bichet, D., Grabe, M., Jan, Y.N., and Jan, L.Y. (2006). Electrostatic interactions in the
channel cavity as an important determinant of potassium channel selectivity. Proceedings of
the National Academy of Sciences of the United States of America 103, 14355-14360.
Bichet, D., Lin, Y.F., Ibarra, C.A., Huang, C.S., Yi, B.A., Jan, Y.N., and Jan, L.Y. (2004).
Evolving potassium channels by means of yeast selection reveals structural elements
important for selectivity. Proceedings of the National Academy of Sciences of the United
States of America 101, 4441-4446.
Biel, M., Wahl-Schott, C., Michalakis, S., and Zong, X. (2009). Hyperpolarization-activated
cation channels: from genes to function. Physiol Rev 89, 847-885.
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.
D'Avanzo, N., Pekhletski, R., and Backx, P.H. (2009). P-loop residues critical for selectivity
in K channels fail to confer selectivity to rabbit HCN4 channels. PLoS One 4, e7712.
Dekker, J.P., and Yellen, G. (2006). Cooperative gating between single HCN pacemaker
channels. The Journal of General Physiology 128, 561-567.
DiFrancesco, D. (1981a). A new interpretation of the pace-maker current in calf Purkinje
fibres. J Physiol 314, 359-376.
DiFrancesco, D. (1981b). A study of the ionic nature of the pace-maker current in calf
Purkinje fibres. J Physiol 314, 377-393.
DiFrancesco, D. (1982). Block and activation of the pace-maker channel in calf purkinje
fibres: effects of potassium, caesium and rubidium. The Journal of Physiology 329, 485-507.
147
DiFrancesco, D. (1986). Characterization of single pacemaker channels in cardiac sino-atrial
node cells. Nature 324, 470-473.
Doyle, D.A., Morais Cabral, J., Pfuetzner, R.A., Kuo, A., Gulbis, J.M., Cohen, S.L., Chait,
B.T., and MacKinnon, R. (1998). The structure of the potassium channel: molecular basis of
K+ conduction and selectivity. Science 280, 69-77.
Furini, S., Zerbetto, F., and Cavalcanti, S. (2007). Role of the intracellular cavity in
potassium channel conductivity. The Journal of Physical Chemistry 111, 13993-14000.
Giorgetti, A., Carloni, P., Mistrik, P., and Torre, V. (2005). A homology model of the pore
region of HCN channels. Biophys J 89, 932-944.
Grabe, M., Bichet, D., Qian, X., Jan, Y.N., and Jan, L.Y. (2006). K+ channel selectivity
depends on kinetic as well as thermodynamic factors. Proceedings of the National Academy
of Sciences of the United States of America 103, 14361-14366.
Heginbotham, L., Lu, Z., Abramson, T., and MacKinnon, R. (1994). Mutations in the K+
channel signature sequence. Biophys J 66, 1061-1067.
Heginbotham, L., and MacKinnon, R. (1993). Conduction properties of the cloned Shaker
K+ channel. Biophys J 65, 2089-2096.
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).
X-ray structure of a voltage-dependent K+ channel. Nature 423, 33-41.
Ludwig, A., Zong, X., Jeglitsch, M., Hofmann, F., and Biel, M. (1998). A family of
hyperpolarization-activated mammalian cation channels. Nature 393, 587-591.
Ludwig, A., Zong, X., Stieber, J., Hullin, R., Hofmann, F., and Biel, M. (1999). Two
pacemaker channels from human heart with profoundly different activation kinetics. The
EMBO journal 18, 2323-2329.
Lyashchenko, A.K., and Tibbs, G.R. (2008). Ion binding in the open HCN pacemaker
channel pore: fast mechanisms to shape "slow" channels. The Journal of General Physiology
131, 227-243.
MacKinnon, R. (2003). Potassium channels. FEBS letters 555, 62-65.
Macri, V., and Accili, E.A. (2004). Structural elements of instantaneous and slow gating in
hyperpolarization-activated cyclic nucleotide-gated channels. The Journal of Biological
Chemistry 279, 16832-16846.
149
Macri, V., Proenza, C., Agranovich, E., Angoli, D., and Accili, E.A. (2002). Separable gating
mechanisms in a Mammalian pacemaker channel. The Journal of Biological Chemistry 277,
35939-35946.
Mistrik, P., Mader, R., Michalakis, S., Weidinger, M., Pfeifer, A., and Biel, M. (2005). The
murine HCN3 gene encodes a hyperpolarization-activated cation channel with slow kinetics
and unique response to cyclic nucleotides. The Journal of Biological Chemistry 280, 27056-
27061.
Morais-Cabral, J.H., Zhou, Y., and MacKinnon, R. (2001). Energetic optimization of ion
conduction rate by the K+ selectivity filter. Nature 414, 37-42.
Moroni, A., Barbuti, A., Altomare, C., Viscomi, C., Morgan, J., Baruscotti, M., and
DiFrancesco, D. (2000). Kinetic and ionic properties of the human HCN2 pacemaker
channel. Pflugers Arch 439, 618-626.
Nimigean, C.M., Chappie, J.S., and Miller, C. (2003). Electrostatic tuning of ion conductance
in potassium channels. Biochemistry 42, 9263-9268.
Pape, H.C. (1996). Queer current and pacemaker: the hyperpolarization-activated cation
current in neurons. Annual Review of Physiology 58, 299-327.
150
Proenza, C., Angoli, D., Agranovich, E., Macri, V., and Accili, E.A. (2002). Pacemaker
channels produce an instantaneous current. The Journal of Biological Chemistry 277, 5101-
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.
Robinson, R.B., and Siegelbaum, S.A. (2003). Hyperpolarization-activated cation currents:
from molecules to physiological function. Annu Rev Physiol 65, 453-480.
Roncaglia, P., Mistrik, P., and Torre, V. (2002). Pore topology of the hyperpolarization-
activated cyclic nucleotide-gated channel from sea urchin sperm. Biophys J 83, 1953-1964.
Rothberg, B.S., Shin, K.S., Phale, P.S., and Yellen, G. (2002). Voltage-controlled gating at
the intracellular entrance to a hyperpolarization-activated cation channel. The Journal of
General Physiology 119, 83-91.
Sakmann, B., and Trube, G. (1984). Conductance properties of single inwardly rectifying
potassium channels in ventricular cells from guinea-pig heart. The Journal of Physiology
347, 641-657.
151
Shealy, R.T., Murphy, A.D., Ramarathnam, R., Jakobsson, E., and Subramaniam, S. (2003).
Sequence-function analysis of the K+-selective family of ion channels using a
comprehensive alignment and the KcsA channel structure. Biophys J 84, 2929-2942.
Shin, K.S., Rothberg, B.S., and Yellen, G. (2001). Blocker state dependence and trapping in
hyperpolarization-activated cation channels: evidence for an intracellular activation gate. The
Journal of General Physiology 117, 91-101.
Stampe, P., Arreola, J., Perez-Cornejo, P., and Begenisich, T. (1998). Nonindependent K+
movement through the pore in IRK1 potassium channels. The Journal of General Physiology
112, 475-484.
Tao, X., Avalos, J.L., Chen, J., and MacKinnon, R. (2009). Crystal structure of the
eukaryotic strong inward-rectifier K+ channel Kir2.2 at 3.1 A resolution. Science 326, 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.
Zhou, M., and MacKinnon, R. (2004). A mutant KcsA K(+) channel with altered conduction
properties and selectivity filter ion distribution. J Mol Biol 338, 839-846.
Zhou, Y., Morais-Cabral, J.H., Kaufman, A., and MacKinnon, R. (2001). Chemistry of ion
coordination and hydration revealed by a K+ channel-Fab complex at 2.0 A resolution.
Nature 414, 43-48.
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
Altomare, C., Bucchi, A., Camatini, E., Baruscotti, M., Viscomi, C., Moroni, A., and
DiFrancesco, D. (2001). Integrated allosteric model of voltage gating of HCN channels. The
Journal of General Physiology 117, 519-532.
Aqvist, J., and Luzhkov, V. (2000). Ion permeation mechanism of the potassium channel.
Nature 404, 881-884.
Azene, E., Xue, T., and Li, R.A. (2003). Molecular basis of the effect of potassium on
heterologously expressed pacemaker (HCN) channels. The Journal of Physiology 547, 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
the National Academy of Sciences of the United States of America 103, 14355-14360.
Bichet, D., Haass, F.A., and Jan, L.Y. (2003). Merging functional studies with structures of
inward-rectifier K(+) channels. Nature Reviews 4, 957-967.
Bichet, D., Lin, Y.F., Ibarra, C.A., Huang, C.S., Yi, B.A., Jan, Y.N., and Jan, L.Y. (2004).
Evolving potassium channels by means of yeast selection reveals structural elements
important for selectivity. Proceedings of the National Academy of Sciences of the United
States of America 101, 4441-4446.
Biel, M., Wahl-Schott, C., Michalakis, S., and Zong, X. (2009). Hyperpolarization-activated
cation channels: from genes to function. Physiological Reviews 89, 847-885.
Blunck, R., Cordero-Morales, J.F., Cuello, L.G., Perozo, E., and Bezanilla, F. (2006).
Detection of the opening of the bundle crossing in KcsA with fluorescence lifetime
spectroscopy reveals the existence of two gates for ion conduction. The Journal of General
Physiology 128, 569-581.
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
the National Academy of Sciences of the United States of America 98, 11277-11282.
Choate, J.K., Nandhabalan, M., and Paterson, D.J. (2001). Raised extracellular potassium
attenuates the sympathetic modulation of sino-atrial node pacemaking in the isolated guinea-
pig atria. Exp Physiol 86, 19-25.
D'Avanzo, N., Pekhletski, R., and Backx, P.H. (2009). P-loop residues critical for selectivity
in K channels fail to confer selectivity to rabbit HCN4 channels. PloS One 4, e7712.
172
Decher, N., Chen, J., and Sanguinetti, M.C. (2004). Voltage-dependent gating of
hyperpolarization-activated, cyclic nucleotide-gated pacemaker channels: molecular coupling
between the S4-S5 and C-linkers. The Journal of Biological Chemistry 279, 13859-13865.
Dietzel, I., Heinemann, U., Hofmeier, G., and Lux, H.D. (1982). Stimulus-induced changes
in extracellular Na+ and Cl- concentration in relation to changes in the size of the
extracellular space. Experimental brain research Experimentelle Hirnforschung 46, 73-84.
DiFrancesco, D., Ferroni, A., Mazzanti, M., and Tromba, C. (1986). Properties of the
hyperpolarizing-activated current (if) in cells isolated from the rabbit sino-atrial node. The
Journal of Physiology 377, 61-88.
Doyle, D.A., Morais Cabral, J., Pfuetzner, R.A., Kuo, A., Gulbis, J.M., Cohen, S.L., Chait,
B.T., and MacKinnon, R. (1998). The structure of the potassium channel: molecular basis of
K+ conduction and selectivity. Science 280, 69-77.
Flynn, G.E., Black, K.D., Islas, L.D., Sankaran, B., and Zagotta, W.N. (2007). Structure and
rearrangements in the carboxy-terminal region of SpIH channels. Structure 15, 671-682.
Giorgetti, A., Carloni, P., Mistrik, P., and Torre, V. (2005). A homology model of the pore
region of HCN channels. Biophysical Journal 89, 932-944.
173
Hackos, D.H., Chang, T.H., and Swartz, K.J. (2002). Scanning the intracellular S6 activation
gate in the shaker K+ channel. The Journal of General Physiology 119, 521-532.
Heginbotham, L., Lu, Z., Abramson, T., and MacKinnon, R. (1994). Mutations in the K+
channel signature sequence. Biophys J 66, 1061-1067.
Holmgren, M., Shin, K.S., and Yellen, G. (1998). The activation gate of a voltage-gated K+
channel can be trapped in the open state by an intersubunit metal bridge. Neuron 21, 617-
621.
Ishii, T.M., Takano, M., Xie, L.H., Noma, A., and Ohmori, H. (1999). Molecular
characterization of the hyperpolarization-activated cation channel in rabbit heart sinoatrial
node. The Journal of Biological Chemistry 274, 12835-12839.
Jackson, H.A., Marshall, C.R., and Accili, E.A. (2007). Evolution and structural
diversification of hyperpolarization-activated cyclic nucleotide-gated channel genes.
Physiological Genomics 29, 231-245.
Jiang, Y., Lee, A., Chen, J., Cadene, M., Chait, B.T., and MacKinnon, R. (2002). The open
pore conformation of potassium channels. Nature 417, 523-526.
Jiang, Y., Lee, A., Chen, J., Ruta, V., Cadene, M., Chait, B.T., and MacKinnon, R. (2003).
X-ray structure of a voltage-dependent K+ channel. Nature 423, 33-41.
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.
Macri, V., and Accili, E.A. (2004). Structural elements of instantaneous and slow gating in
hyperpolarization-activated cyclic nucleotide-gated channels. The Journal of Biological
Chemistry 279, 16832-16846.
Macri, V., Nazzari, H., McDonald, E., and Accili, E.A. (2009). Alanine scanning of the S6
segment reveals a unique and cyclic AMP-sensitive association between the pore and
voltage-dependent opening in HCN channels. The Journal of Biological Chemistry.
Mannikko, R., Elinder, F., and Larsson, H.P. (2002). Voltage-sensing mechanism is
conserved among ion channels gated by opposite voltages. Nature 419, 837-841.
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.
Morais-Cabral, J.H., Zhou, Y., and MacKinnon, R. (2001). Energetic optimization of ion
conduction rate by the K+ selectivity filter. Nature 414, 37-42.
Noskov, S.Y., Berneche, S., and Roux, B. (2004). Control of ion selectivity in potassium
channels by electrostatic and dynamic properties of carbonyl ligands. Nature 431, 830-834.
Noskov, S.Y., and Roux, B. (2007). Importance of hydration and dynamics on the selectivity
of the KcsA and NaK channels. The Journal of General Physiology 129, 135-143.
Paterson, D.J. (1996). Role of potassium in the regulation of systemic physiological function
during exercise. Acta physiologica Scandinavica 156, 287-294.
Proenza, C., Angoli, D., Agranovich, E., Macri, V., and Accili, E.A. (2002). Pacemaker
channels produce an instantaneous current. The Journal of Biological Chemistry 277, 5101-
5109.
Proenza, C., and Yellen, G. (2006). Distinct populations of HCN pacemaker channels
produce voltage-dependent and voltage-independent currents. The Journal of General
Physiology 127, 183-190.
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.
Robinson, R.B., and Siegelbaum, S.A. (2003). Hyperpolarization-activated cation currents:
from molecules to physiological function. Annual Review of Physiology 65, 453-480.
Rothberg, B.S., Shin, K.S., Phale, P.S., and Yellen, G. (2002). Voltage-controlled gating at
the intracellular entrance to a hyperpolarization-activated cation channel. The Journal of
General Physiology 119, 83-91.
Rothberg, B.S., Shin, K.S., and Yellen, G. (2003). Movements near the gate of a
hyperpolarization-activated cation channel. The Journal of General Physiology 122, 501-510.
Roux, B. (2005). Ion conduction and selectivity in K(+) channels. Annu Rev Biophys Biomol
Struct 34, 153-171.
Shealy, R.T., Murphy, A.D., Ramarathnam, R., Jakobsson, E., and Subramaniam, S. (2003).
Sequence-function analysis of the K+-selective family of ion channels using a
comprehensive alignment and the KcsA channel structure. Biophys J 84, 2929-2942.
Shi, N., Ye, S., Alam, A., Chen, L., and Jiang, Y. (2006). Atomic structure of a Na+- and
K+-conducting channel. Nature 440, 570-574.
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
Biological Chemistry 277, 18994-19000.
Vemana, S., Pandey, S., and Larsson, H.P. (2004). S4 movement in a mammalian HCN
channel. The Journal of General Physiology 123, 21-32.
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.
Yifrach, O., and MacKinnon, R. (2002). Energetics of pore opening in a voltage-gated K(+)
channel. Cell 111, 231-239.
Yool, A.J., and Schwarz, T.L. (1991). Alteration of ionic selectivity of a K+ channel by
mutation of the H5 region. Nature 349, 700-704.
Zagotta, W.N. (2006). Membrane biology: permutations of permeability. Nature 440, 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.
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.
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.
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.
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.
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
Movement Disorders, Vol. 24, No. 5, 2009
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.
Movement Disorders, Vol. 24, No. 5, 2009
780 M.K. DEMOS ET AL.
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
Movement Disorders, Vol. 24, No. 5, 2009
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.
782 M.K. DEMOS ET AL.
Movement Disorders, Vol. 24, No. 5, 2009
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