Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Structural determinants regulating surface expression and function of Kv-related ion channels Nazzari, Hamed 2010

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

Item Metadata

Download

Media
24-ubc_2010_fall_nazzari_hamed.pdf [ 16.65MB ]
Metadata
JSON: 24-1.0071289.json
JSON-LD: 24-1.0071289-ld.json
RDF/XML (Pretty): 24-1.0071289-rdf.xml
RDF/JSON: 24-1.0071289-rdf.json
Turtle: 24-1.0071289-turtle.txt
N-Triples: 24-1.0071289-rdf-ntriples.txt
Original Record: 24-1.0071289-source.json
Full Text
24-1.0071289-fulltext.txt
Citation
24-1.0071289.ris

Full Text

STRUCTURAL DETERMINANTS REGULATING SURFACE EXPRESSION AND FUNCTION OF Kv-RELATED ION CHANNELS by Hamed Nazzari B.Sc. (Hon), Simon Fraser University, 2005  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) August 2010 © Hamed Nazzari, 2010  ABSTRACT To date, the mechanisms and structural determinants which contribute to the regulation of ion channel trafficking, surface expression and function have only been limitedly explored. Through biophysical and molecular characterization of the hyperpolarization-activated cyclicnucleotide gated channel 2 (HCN2), we have identified a four-amino acid motif (EEYP) in the B-helix of the cyclic-nucleotide binding domain (CNBD) that strongly promotes channel export from the endoplasmic reticulum (ER) and cell surface expression but does not contribute to the inhibition of channel opening. We further demonstrate that this motif augments a step in the trafficking pathway and/or the efficiency of correct folding and assembly. The role of posttranslational modifications, specifically N-linked glycosylation, has also been investigated in two different HCN isoforms. All four mammalian HCN channel isoforms have been shown to undergo N-linked glycosylation in the brain. HCN channels have further been suggested to require N-glycosylation for function, a provocative finding that would make them unique in the voltage-gated potassium channel superfamily. Here, we show that both the HCN1 and HCN2 isoforms are also predominantly N-glycosylated in the embryonic heart, where they are found in significant amounts and where HCN-mediated currents are known to regulate beating frequency. Surprisingly, we find that N-glycosylation is not required for HCN2 function, although its cell surface expression is highly dependent on the presence of N-glycans. Comparatively, disruption of N-glycosylation only modestly impacts cell surface expression of HCN1 and leaves permeation and gating functions almost unchanged. The evolutionary significance of this isoforms specific regulation is also examined. Finally, the role of palmitoylation in the regulation of Kv4 channels is examined. Using acylbiotin exchange (ABE) chemistry we are able demonstrate that Kv4.2 is present as a palmitoylated protein in both rat cortical neurons and COS-7 cells. Through mutational analysis of the twelve intracellular cysteine residues within ii  Kv4.2, we were able to localize the site of palmitoylation to the intracellular COOH-terminus. Palmitoylation of Kv4.2 does not contribute to the regulation of activation and inactivation gating parameters. Rather, inhibition of palmitoylation through either mutation of COOHterminal cysteine residues or the pharmacological agent 2-bromopalmitate results in significant reductions in overall current density measurements.  iii  TABLE OF CONTENTS ABSTRACT.................................................................................................................................................................. ii	
   TABLE OF CONTENTS ........................................................................................................................................... iv	
   LIST OF FIGURES ................................................................................................................................................... vii	
   ABBREVIATIONS................................................................................................................................................... viii	
   ACKNOWLEDGEMENTS ........................................................................................................................................ x	
   CO-AUTHORSHIP STATEMENT ......................................................................................................................... xii	
   CHAPTER 1: INTRODUCTION.............................................................................................................................. 1	
   1.1 OVERVIEW OF HYPERPOLARIZATION-ACTIVATED CYCLIC-NUCLEOTIDE GATED CHANNELS ................................................................................................................................................................................... 2	
   1.1.1 Physiological role of If in the heart............................................................................................................. 6	
   1.1.2 Physiological role of Ih in neurons.............................................................................................................. 9	
   1.1.3 HCN channels and disease........................................................................................................................ 11	
   1.1.4 Structure and function of the cyclic-nucleotide binding domain .............................................................. 13	
   1.1.5 Accessory proteins regulating HCN channels .......................................................................................... 16	
   1.1.6 Post-translational modifications regulating HCN channels..................................................................... 17	
   1.2 POTASSIUM CHANNEL OVERVIEW.......................................................................................................... 18	
   1.2.1 Overview of the Kv4.x potassium channels............................................................................................... 19	
   1.2.2 Physiological role of Ito in the heart ......................................................................................................... 21	
   1.2.3 Physiological role of IA in the central nervous system.............................................................................. 23	
   1.2.4 Post-translational modification of Kv4.x channels ................................................................................... 25	
   1.2.5 Accessory proteins regulating Kv4.x channels ......................................................................................... 26	
   1.2.6 Kv4.x channels and disease....................................................................................................................... 31	
   1.3 BIOGENESIS AND SURFACE TRAFFICKING OF ION CHANNELS....................................................... 33	
   1.3.1 Gene expression and transcriptional regulation....................................................................................... 33	
   1.3.2 Translation through the endoplasmic reticulum ....................................................................................... 34	
   1.3.3 ER retention signals and quality control mechanisms .............................................................................. 35	
   1.3.4 Mechanisms regulating forward trafficking from the ER to the Golgi apparatus .................................... 38	
   1.3.5 Golgi to plasma membrane trafficking ..................................................................................................... 41	
   1.3.6 Integration, distribution and compartmentalization at the cell surface ................................................... 43	
   1.4 ER ASSOCIATED DEGRADATION.............................................................................................................. 46	
   1.5 OVERVIEW OF N-LINKED GLYCOSYLATION ........................................................................................ 47	
   1.5.1 Regulation of ion channel trafficking and function through N-glycosylation........................................... 48	
   1.5.2 Glycosylation of ion channels and disease ............................................................................................... 50	
   1.6 OVERVIEW OF PALMITOYLATION........................................................................................................... 50	
   1.6.1 Regulation of ion channels through palmitoylation.................................................................................. 52	
   1.6.2 Identification of palmitoylated proteins .................................................................................................... 53	
   1.7 SCOPE OF THE THESIS................................................................................................................................. 55	
   1.8 REFERENCES.................................................................................................................................................. 57	
   CHAPTER 2: REGULATION OF CELL SURFACE EXPRESSION OF FUNCTIONAL PACEMAKER CHANNELS BY A MOTIF IN THE B-HELIX OF THE CYCLIC NUCLEOTIDE-BINDING DOMAIN.... 82	
   2.1 INTRODUCTION ............................................................................................................................................ 83	
   2.2 MATERIALS AND METHODS...................................................................................................................... 85	
   2.2.1 Mutagenesis and expression ..................................................................................................................... 85	
   2.2.2 Whole cell patch-clamp electrophysiology ............................................................................................... 86	
   2.2.3 Immunocytochemistry and microscopy ..................................................................................................... 87	
   2.2.4 Western blot analysis ................................................................................................................................ 88	
   2.2.5 Proteinase K treatment ............................................................................................................................. 89	
   2.2.6 Sucrose gradient analysis ......................................................................................................................... 90	
   2.3 RESULTS ......................................................................................................................................................... 91	
    iv  2.3.1 Identification of a motif in the CNBD B-helix potentially important for channel trafficking and function ............................................................................................................................................................................ 91	
   2.3.2 The EEYP motif promotes ER export and cell surface expression of mature HCN2................................ 92	
   2.3.3 The EEYP motif is not required to form functional channels and does not contribute to inhibition of channel opening ................................................................................................................................................. 94	
   2.3.4 The EEYP motif regulates cell surface expression by a mechanism that does not lead to substantive degradation or disruption of subunit assembly.................................................................................................. 96	
   2.3.5 The EE and YP amino-acid couplets have an additive effect on cell surface expression ......................... 99	
   2.4 DISCUSSION ................................................................................................................................................. 100	
   2.5 GRANTS......................................................................................................................................................... 104	
   2.6 ACKNOWLEDGEMENTS ............................................................................................................................ 105	
   2.7 REFERENCES ............................................................................................................................................... 122	
   CHAPTER 3: EVOLUTIONARY EMERGENCE OF N-GLYCOSYLATION AS A VARIABLE PROMOTER OF HCN CHANNEL SURFACE EXPRESSION ........................................................................ 125	
   3.1 INTRODUCTION .......................................................................................................................................... 126	
   3.2 MATERIALS AND METHODS.................................................................................................................... 128	
   3.2.1 Mutagenesis............................................................................................................................................. 128	
   3.2.2Western Blotting....................................................................................................................................... 128	
   3.2.3Immunocytochemistry and microscopy .................................................................................................... 129	
   3.2.4 ELISA ...................................................................................................................................................... 130	
   3.2.5 Electrophysiology and data analysis ...................................................................................................... 130	
   3.2.6 HCN sequence collection and analysis ................................................................................................... 131	
   3.3 RESULTS ....................................................................................................................................................... 132	
   3.3.1 HCN2 channels undergo N-glycosylation in native cardiac tissue ........................................................ 132	
   3.3.2 N-glycosylation promotes mHCN2 cell surface expression.................................................................... 134	
   3.3.3 The effect of N-glycosylation on cell surface expression correlates only partly with an increase in total protein .............................................................................................................................................................. 135	
   3.3.4 HCN2 channels lacking N-glycosylation form functional channels ....................................................... 136	
   3.3.5 N-glycosylation of HCN channels emerged in an ancestor common to chordates to promote cell surface expression......................................................................................................................................................... 137	
   3.4 DISCUSSION ................................................................................................................................................. 141	
   3.5 ACKNOWLEDGEMENTS ............................................................................................................................ 143	
   3.6 REFERENCES ............................................................................................................................................... 152	
   CHAPTER 4: COOH-TERMINAL PALMITOYLATION OF THE KV4.2 VOLTAGE-GATED POTASSIUM CHANNEL REGULATES CELL SURFACE EXPRESSION................................................... 158	
   4.1 INTRODUCTION .......................................................................................................................................... 159	
   4.2 MATERIALS AND METHODS .................................................................................................................... 162	
   4.2.1 Mutagenesis and expression ................................................................................................................... 162	
   4.2.2 Electrophysiological and data analysis .................................................................................................. 162	
   4.2.3 Cortical neuron preparation ................................................................................................................... 163	
   4.2.4 Cell lysate preparation and immunoprecipitation .................................................................................. 164	
   4.2.5 Acyl-Biotin-Exchange (ABE) chemistry assay ........................................................................................ 164	
   4.3 RESULTS ....................................................................................................................................................... 165	
   4.3.1 Kv4.2 is palmitoylated in E18 rat embryonic cortical neurons .............................................................. 165	
   4.3.2 2-Bromopalmitate treatment of rat cortical neurons reduces the total amount of palmitoylated Kv4.2 166	
   4.3.3 Rat Kv4.2 and Kv4.3 are palmitoylated in COS-7 cells.......................................................................... 166	
   4.3.4 Kv4.2 is palmitoylated within the intracellular COOH-terminus ........................................................... 167	
   4.3.5 Palmitoylation does not modulate gating parameters of Kv4.2.............................................................. 168	
   4.3.6 Palmitoylation of Kv4.2 increase current density measurements ........................................................... 168	
   4.4 DISCUSSION ................................................................................................................................................. 170	
   4.5 ACKNOWLEDGEMENTS ............................................................................................................................ 173	
   4.6 REFERENCES ............................................................................................................................................... 181	
   CHAPTER 5: CONCLUSIONS AND GENERAL DISCUSSION ..................................................................... 185	
   5.1 REFERENCES ............................................................................................................................................... 195	
    v  APPENDICES.......................................................................................................................................................... 197	
   APPENDIX A: CONTRIBUTIONS TO OTHER PUBLISHED MATERIAL ................................................... 198	
   APPENDIX B: ALANINE SCANNING OF THE S6 SEGMENT REVEALS A UNIQUE AND CAMP – SENSITIVE ASSOCIATION BETWEEN THE PORE AND VOLTAGE-DEPENDENT OPENING IN HCN CHANNELS ......................................................................................................................................................... 199	
   APPENDIX C: HCN2 AND HCN4 ISOFORMS SELF-ASSEMBLE AND CO-ASSEMBLE WITH EQUAL PREFERENCE TO FORM FUNCTIONAL PACEMAKER CHANNELS ........................................................ 200	
   APPENDIX D: IN SITU CO-DISTRIBUTION AND FUNCTIONAL INTERACTIONS OF SAP97 WITH SINOATRIAL ISOFORMS OF HCN CHANNELS............................................................................................ 201	
    vi  LIST OF FIGURES Figure 1.1 Membrane topology of HCN channels. ........................................................................ 5	
   Figure 1.2 Rhythmic activity in cells of the sinoatrial node........................................................... 8	
   Figure 1.3 Structure of the mouse HCN2 C-linker and CNBD construct bound to cAMP.......... 15	
   Figure 2.1 The distal B-helix is conserved among ion channels but not protein kinases with similar cyclic nucleotide-binding domains (CNBDs). ............................................................... 106	
   Figure 2.2 Schematic representation of CNBD B-helix mutants. .............................................. 107	
   Figure 2.3 The EEYP motif promotes cell surface expression of the mature form of HCN2.... 108	
   Figure 2.4 The EEYP motif is not required to form functional channels................................... 110	
   Figure 2.5 The EEYP motif does not contribute to inhibition of channel opening. ................... 112	
   Figure 2.6 The EEYP motif does not prevent channel degradation. .......................................... 114	
   Figure 2.7 Enhanced degradation accompanies reduced If density upon NH2-terminal truncation and thus defines an intracellular fate distinct from that seen in the EEYP mutants................... 116	
   Figure 2.8 HCN2-4A and wild-type channels assemble to the same extent. ............................. 118	
   Figure 2.9 Individual elements within the EEYP contribute to its function............................... 120	
   Figure 3.1 Mouse (m) hyperpolarization-activated cyclic nucleotide-modulated (HCN)2 channels undergo N-glycosylation in native cardiac tissue. ...................................................................... 144	
   Figure 3.2 N-glycosylation is required for mHCN2 function in multiple expression systems. . 145	
   Figure 3.3 N-glycosylation promotes mHCN2 cell surface expression. .................................... 147	
   Figure 3.4 N-glycosylation of HCN channels emerged in an ancestor common to chordates. .. 149	
   Figure 3.5 N-glycosylation is not required for efficient mHCN1 cell surface expression. ........ 150	
   Figure 3.6 Disruption of N-glycosylation minimally impacts mHCN1 function. ...................... 151	
   Figure 4.1 Schematic representation of acyl-biotinyl exchange (ABE) assay. .......................... 174	
   Figure 4.2 Kv4.2 is palmitoylated in E18 embryonic rat cortical neurons................................. 175	
   Figure 4.3 2-bromopalmitate reduces Kv4.2 palmitoylation in rat cortical neurons.................. 176	
   Figure 4.4 Kv4.2 and Kv4.3 are palmitoylated in COS-7 cells.................................................. 177	
   Figure 4.5 Kv4.2 is palmitoylated within the intracellular COOH-terminus. ............................ 178	
   Figure 4.6 Palmitoylation does not modulate steady-state activation or inactivation properties. .................................................................................................................................................... 179	
   Figure 4.7 Palmitoylation of Kv4.2 results in enhanced expression of channels at the cell surface. .................................................................................................................................................... 180	
    vii  ABBREVIATIONS Amino Acid  3 Letter Code  1 Letter Code  Alanine Arginine Asparagine Aspartate Cysteine Glutamine Glutamate Glycine Histidine Isoleucine Leucine Lysine Methionine Phenylalanine Serine Threonine Tryptophan Tyrosine Valine  Ala Arg Asn Asp Cys Gln Glu Gly His Ile Leu Lys Met Phe Ser Thr Try Tyr Val  A R N D C Q E G H I L K M P S T W Y V  ADP: Adenosine diphosphate AF: Atrial fibrillation AMPA: α-amino-3-hydroxyl-5-methyl-4-isoxazole-propionate ATP: Adeonsine triphosphate bAPs: Back-propagating action potentials BKca: Voltage-activated potassium channels CAP: Catabolite gene activator protein cAMP: Cyclic adenosine monophosphate CDGs: Congenital disorders of glycosylation cDNA: Complementary deoxyribonucleic acid cGMP: Cyclic guanosine monophosphate CNBD: Cyclic-nucleotide-binding-domain CNS: Central nervous system CHO: Chinese hamster ovary COP: Coat protein complex CPTV: Catecholaminergic polymorphic ventricular tachycardia DPPX: Dipeptidyl aminopeptidase-like protein DRM: Detergent-resistant membrane DXE: Diacidic motifs E1: Ubiquitin activating enzyme E2: Ubiquitin-carrier enzyme E3: Ubiquitin-protein ligase viii  EAG: Ether-à-go-go ER: Endoplasmic reticulum ERAD: Endoplasmic reticulum associated degradation ERGIC: Endoplasmic reticulum-golgi intermediate compartment GEF: Guanine exchange factor GTPase: Guanosine triphosphatase HCN: Hyperpolarization-activated-cyclic-nucleotide-gated HEK: Human embryonic kidney hERG: Human Ether-à-go-go Related Gene IA:A-type current If: “Funny” current Iq: Queer current Ih: Hyperpolarized current Iins: Instantaneous current Ito: Transient outward current Ito,f: Fast transient outward current Ito,s: Slow transient outward current KChAP: Potassium channel associated protein KChIP: Potasium channel interacting protein KCNA: Potassium voltage-gated channel, shaker-related subfamily KCNH: Potassium voltage-gated channel, subfamily H KCNK: Potassium voltage-gated channel, subfamily K KCNQ: Potassium voltage-gated channel, KQT-like subfamily KCR1: Kupffer cell receptor Kir: Inward rectifying potassium channel Kv: Voltage-gated potassium channel LTP: Long-term potentiation LQTS: Long-QT syndrome M2: Muscarinic acetylcholine receptors MI: Myocardial infarction mRNA: Messenger ribonucleic acid N-X-S/T: N-glycosylation sequon NMDA: N-methyl-D-aspartic acid PKA: Protein kinase A PKC: Protein kinase C RNA: Ribonucleic acid rRNA: Ribosomal ribonucleic acid SA: Sinoatrial node SNAP-25: Synaptosomal-associated protein 25 SNARE: SolubleN-ethylmaleimide-sensitive factor-attachment protein receptors SUR: Sulfonylurea receptor Syn-1A: Syntaxin 1A TEA: Tetraethylammonium tRNA: Transfer ribonucleic acid UPS: Ubiquitin proteasome system UTRs: Untranslated regions VDCC: Voltage-dependent calcium channel ix  ACKNOWLEDGEMENTS I would like to express my sincere gratitude to my supervisor Dr. Eric Accili. Thank you for all of your support, guidance and mentorship throughout my graduate studies. To my supervisory committee members Drs. Kenneth Baimbridge, David Fedida, Christopher Loewen and Filip Van Petegem, thank you for your expertise, input and guidance over the years. Throughout my time here at UBC I have had the opportunity to meet some great people. I have many fond memories of my time spent in the lab with my labmates HJ, GW, EM, AH, CP, SC, DA, you all helped to create a great lab atmosphere. To all of my family and friends thank you for all of your support and encouragement throughout the years. To AJR, thank you for everything. Finally and most importantly, to my parents, thank you for everything you sacrificed to get me to this point. I could not have reached this milestone without you both in my life. I am eternally grateful for all of your love, support, encouragement and belief in me throughout the years. Mom, during the final months of writing this thesis we went through some very difficult times together. The strength and determination you have shown in your battle with cancer is truly courageous and inspiring. I love you and you are forever my hero.  x  DEDICATION  To my parents  xi  CO-AUTHORSHIP STATEMENT CHAPTER 2: Regulation of cell surface expression of functional pacemaker channels by a motif in the B-helix of the cyclic nucleotide-binding domain Nazzari, H., Angoli, D., Chow, S.S., Whitaker, G., Leclair, L., McDonald, E., Macri, V., Zahynacz, K., Walker, V., Accili, E.A. A version of this chapter has been published in the American Journal of Physiology – Cell Physiology Contributions: HN was responsible for experimental design and work, data analysis and writing of the manuscript in conjunction with EAA and completed of revisions required for publication. Figure 1 was completed by EAA. Figure 2 was completed by HN. Mutants used in Figure 3 were generated by HN and EM, western blots by HN and SSC, immunocytochemistry by HN and SSC. Figure 4 panels A and B were completed by HN and DA, Figure 4C was completed by HN. Figure 5 was completed by HN and DA. Figure 6 panels A-C were completed by HN and DA with two electrophysiology traces contributed by VM. Figure 6D was completed by HN and SSC. Figure 7 was completed by SSC, GW and LL. Figures 8 and 9 was completed by HN. CHAPTER 3: Evolutionary emergence of N-glycosylation as a variable promoter of HCN channel surface expression Hegle, A.P*., Nazzari, H*., Roth, A., Angoli, D., Accili, E.A. * Equal Contribution A version of this chapter has been published in the American Journal of Physiology – Cell Physiology Contribution: HN was responsible for experimental design and work in conjuction with APH and EAA. Figure 1 was completed by APH. Figure 2 panels A and C were completed by APH, panel B was completed by HN and DA. Figure 3 panels A-D were completed by HN, panel E was completed by APH. Figure 4 panel A was completed by AR, panel B by APH and panel C by HN. Figure 5 panels A and B were completed by APH, panels C and B were completed by HN. Figure 6 was completed by HN. EAA, APH and HN contributed towards the writing and editing of the manuscript. CHAPTER 4: C-terminal palmitoylation of the Kv4.2 voltage-gated potassium channel regulates cell surface expression Nazzari, H., Kang, R., Cheng, Y., Fedida, D., Accili, E.A. A version of this chapter will be submitted for publication Contribution: HN was responsible for project conception and designing of experiments. HN and RK performed palmitoylation assays. RK was responsible for cortical neuron preparation. HN constructed all mutants used in this study. YC performed electrophysiology experiments. Manuscript was written by HN with input from EAA.  xii  CHAPTER 1: INTRODUCTION  1.1 OVERVIEW OF HYPERPOLARIZATION-ACTIVATED CYCLIC-NUCLEOTIDE GATED CHANNELS Since their initial characterization in sinoatrial (SA) node cells of the heart (64, 65, 295), hyperpolarization activated currents have garnered a great deal of attention due to their inherently unique biophysical properties. The current in heart was originally coined Ih (hyperpolarized current) or If (“funny” current) and later, in neurons, Iq (queer current); throughout this thesis we will refer to it as If or Ih. Several distinguishing characteristics – primarily its unique selectivity and gating properties – make If an exceptional case among ionic currents. If is a cationic current carried by both sodium and potassium ions, and has a reversal potential of approximately -20mV under physiologic ionic concentrations (64). Thus, at resting membrane potentials If is an inwardly conducted current. Additionally, unlike most ionic currents found in the heart and neurons which are activated upon depolarization, If is activated at hyperpolarized potentials and its gating is strongly regulated by intracellular cyclic adenosine monophosphate (cAMP) levels (25, 66, 66). Increases in intracellular cAMP exert their effects directly and independently of protein kinase A (PKA), enhancing the opening of the channels and shifting the steady-state voltage dependence to more depolarized potentials. Interestingly, If does not exhibit the voltage-dependent inactivation which is seen in most other voltage-gated channels, including potassium and sodium channels (257). Kinetically, If can be broken down into two separate components: a minor instantaneous component (Iins) (2, 180, 216, 216) which is fully activated within a few milliseconds and the major more slowly developing component which reaches steady-state levels in a range of milliseconds to seconds (2, 60, 180) . It should be noted that both of these kinetic components vary substantially depending on cell type and microenvironment (e.g., auxiliary proteins, pH, temperature, etc.).  2  The molecular correlates of If were identified and characterized over a decade ago (177, 181, 236). Because of their dual gating properties the underlying channels were referred to as Hyperpolarization-activated cyclic-nucleotide-gated (HCN) channels. Originally cloned from the rat brain, four isoforms of HCN channels have since been identified in mammals (HCN1-4) (178), each of which is encoded by a separate gene (255). Heterologous expression of these complementary deoxyribonucleic acid (cDNAs) has confirmed that HCN channels are the molecular correlate of If. The four isoforms exhibit distinctive gating properties, responsiveness to cAMP, and tissue distribution (170, 179). In terms of channel activation properties the kinetic order follows HCN1>HCN2>HCN3>HCN4, whereas responsiveness to cAMP is reversed and follows HCN4>HCN2>HCN3>HCN1 (180, 276). With respect to tissue expression, HCN2 and HCN4 have been identified primarily as cardiac isoforms; while HCN1 has been deemed to be a primarily neuronal expressing isoform and HCN3 has shown variable tissues expression in both cardiac and neuronal tissue and has not been well characterized to date (30, 49, 79). Structurally, HCN channels have a similar topology to most voltage-gated channels. Each isoform is comprised of four identical subunits which come together in a tetrameric assembly. Each subunit consists of six transmembrane helices which are interconnected through extracellular and intracellular linkers (267, 308) (Figure 1.1). The N- and C-termini are intracellular domains, with the COOH-terminus containing the cyclic-nucleotide-binding-domain (CNBD) (48). Similar to other voltage-gated channels the S4 transmembrane segment harbors several positively charged arginine and lysine resides regularly spaced at every third position and serves as the voltage sensor of these channels. The pore loop between the S5 and S6 transmembrane segment forms the selectivity filter and contains the potassium selective consensus sequence GYG (158). Interestingly, despite their preference for potassium ions, HCN channels are also able to conduct sodium at physiological ionic concentrations. In general, HCN 3  channels conduct potassium and sodium ions at a 4:1 ratio. Several reports have shown that HCN channels exhibit a small permeability for calcium ions as well (159, 304). The relevance of calcium entry through these channels is, however, not well understood. In addition to the GYG motif the selectivity filter of HCN channels is closely related to those of potassium channels in terms of amino acid sequence. Efforts to resolve the inherent differences in ion selectivity between HCN and voltage-gated potassium (Kv) channels, despite this high degree of sequence homology within the selectivity filter have proved unsuccessful to date. The unique characteristics of If play an important role in regulating cellular excitability and electrical responsiveness in many different cell types. Since its initial characterization in the SA node, If has been shown to play a significant role in the excitability of many other cell types including, ventricular myocytes, hippocampal neurons, and thalamic neurons.  4  Figure 1.1 Membrane topology of HCN channels. Two of the four subunits of the HCN channel are shown. The transmembrane segments (S1–S6) are shown in black, except for the pore region (S6 and pore loop), which is shown in red. The S4 contains positively charged arginine residues and serves as the voltage sensor of the channel. The COOH-terminal region contains the cyclic nucleotide–binding domain (CNBD, blue, shown with cGMP bound) and the C-linker (green), which connects the CNBD to the pore region. Figure adapted from Craven KB and Zagotta, 2006.  5  1.1.1 Physiological role of If in the heart In the heart, the SA node serves as the primary pacemaking region, with a number of different ionic currents contributing to the generation of the spontaneous rhythmic firing of action potentials required for normal heart function (61, 167) (Figure 1.2). HCN channels, activated during hyperpolarization of the membrane after an action potential has fired, conduct the inward cation current If necessary for membrane depolarization back to the threshold levels for the firing of the subsequent action potential. Not surprisingly, because of their role in regulating the diastolic depolarization phase of the action potential, modulation of HCN channels and If has been shown to result in significant changes in action potential firing rates (32, 60, 62, 64). The inherent properties of HCN channels make them ideal candidates for modulation through the autonomic nervous system. Physiological adaptations to changes in sympathetic/parasympathetic stimulation need to be rapid and highly regulated. The exerted effects of cAMP on HCN channels through direct binding of the CNBD in the channels COOHterminus contributes to this rapid and effective modulation. This is a direct advantage to other cAMP induced cellular responses, which involve cAMP-mediated phosphorylation events and therefore are inherently slower and less responsive to changes in cAMP concentrations. In the heart, this modulation through autonomic stimuli in turn modulates the steepness of the slow diastolic depolarization phase of the action potential. Sympathetic stimulation through βadrenergic (β1 and β2) receptors triggers an increase in intracellular cAMP concentrations that, in turn, directly enhance HCN channel opening by shifting the voltage dependence of activation to more positive potentials (4, 276). Activation of the parasympathetic system through specific muscarinic acetylcholine receptors (M2) conversely reduces cytosolic cAMP levels and shifts the voltage dependence to more negative potentials (2, 63). Because of this direct modulation, 6  local concentrations of cAMP surrounding HCN channels are critical. In the SA node, it has been demonstrated that β2-adrenergic receptors both found at a much higher density than in surrounding tissue and are co-localized with the HCN4 channels. The dominant presence of HCN4 in the SA node is relevant as well since it is the most responsive of the HCN channels to changes in intracellular cAMP and therefore is able to respond most readily to changes in autonomic stimulation (97, 261).  7  Figure 1.2 Rhythmic activity in cells of the sinoatrial node. Heart sinoatrial node pacemaker current. Sinoatrial node (black oval) of the heart (right) and the ion channels which contribute towards generating pacemaker activity (left). Each of the currents generated contributes to a specific phase of the sinoatrial action potential, indicated by the dotted arrows. Figure adapted from Craven KB and Zagotta, 2006.  8  1.1.2 Physiological role of Ih in neurons The unique properties of HCN channels as already discussed allows these channels to serve as distinct regulators of cellular excitability. In our discussion of Ih in neurons two particular properties of HCN channel are necessary to mention. Firstly, HCN channels are constitutively open at neuronal resting membrane potentials, passing an inward depolarizing current which contributes to establishing and maintaining the resting membrane potential (105). Secondly, HCN channels are able to counteract both depolarizing and hyperpolarizing currents by undergoing activation (inward current) at hyperpolarizing potentials and deactivation (outward current) at depolarized potentials. It is these two specific properties of Ih that provide the molecular basis for the diverse set of physiological roles which these channels play in neurons (20, 44, 178, 183). Specifically, HCN channels have been well documented to play a role in physiological processes including: dendritic integration, working memory, synaptic transmission, motor learning and the regulation of resonance and oscillations in specific neurons (300). For the purposes of this thesis each of these processes will not be examined in great detail, rather selected features will be discussed further. Dendritic integration is a process which is critical for signal processing in neurons. Several ion channels have been shown to contribute this process, including AMPA glutamate receptors, Kv4 channels and HCN channels (53, 130, 165). Dendritic integration has been well characterized in CA1 pyramidal neurons of the hippocampus, here HCN channels have been shown to be expressed in their highest levels at the distal dendrites, with expression decreasing in a gradient-like manner as distance from the soma decreases (162, 197). It is believed that the Ih, specifically from the HCN1 isoform in these cell types play an important role in contributing to the dampening of distally originating excitatory post-synaptic potentials (EPSPs) arriving at the level of the soma (205). Closing of HCN channels at positive membrane potentials decreases 9  the flow of positively charged inward current flow and allows the repolarizing currents to dominate and therefore increasing the rate of EPSP decay (205, 206). Therefore as an EPSP travel from a distal to more proximal location their signal is attenuated to a greater degree than proximally originating EPSPs. As such, the temporal summation of these distally and proximally arriving EPSPs is approximately equal at the level of the soma. In addition to attenuating EPSPs in select neurons, HCN channels also play a crucial role in controlling the oscillations and rhythmic firing in neurons (132, 300). Thalamocortical neurons have been shown to undergo two types of firing: burst and continuous (14, 15). These two modes of firing have been attributed to rapid eye movement (REM) and non-REM sleep patterns. During REM sleep thalamocortical neurons are depolarized by afferent inputs and transition to what is referred to as the transmission firing mode (117, 118). In this mode sodium spikes result in constant depolarization of the membrane, which results in continued inactivation of T-type Ca2+ channels. As a result of this constant firing the output frequency of these neurons is increased which results in increased synaptic transmission. Burst firing is most often a characteristic of non-REM sleep and epileptic events (260). This firing is tightly regulated through the orchestrated effects of the low threshold T-type Ca2+ current, IT, along with Ih. Hyperpolarization of the membrane activates the HCN channels, resulting in the inward flow of Ih which depolarizes the membrane potential enough to activate the T-type channels which causes a further depolarization and burst firing action potentials carried out by Na+/K+ currents (260, 300). The inactivation of T-type channels decreases the amount of inward current flow, additionally during this spike firing HCN channels are also deactivated. Overall this leads to an overshoot hyperpolarization of the membrane and subsequent activation of HCN channels which will initiate another cycle of burst firing (126, 140, 141). The roles of Ih in neurons which have  10  been discussed only highlight the complex role this current plays in the regulation of physiological processes involving neurons. 1.1.3 HCN channels and disease Advances in genomic screening technologies have revolutionized and dramatically enhanced the search for genetically determined channelopathies. Mutations in ion channel genes have been associated with a number of inheritable diseases. In terms of cardiac specific channelopathies the list includes, but is not limited to, long and short QT syndromes, Brugada syndrome, and catecholaminergic polymorphic ventricular tachycardia (CPVT). Most of these disorders have been associated with mutations found in voltage-gated potassium (Kv) channels. Many other ion channels have been associated with various disease phenotypes, HCN channels however remain a relatively uncharacterized protein in terms of its contribution to different disease states (106). Four mutations in the HCN4 isoform have been found to be associated with disease states (176, 195, 246, 271). Each of these mutations has been linked to some form of cardiac dysfunction. In one study, a single base pair deletion (1631delC) in exon 3 of the human HCN4 gene results in a premature stop codon and an HCN4 channel with no CNBD (246). Heterologous expression of this truncated version of HCN4 demonstrated that these channels are able to traffic to the cell surface, but are no longer sensitive to changes in intracellular cAMP. Interestingly, this insensitivity to cAMP was seen in both homomeric and hetermeric channels, indicating a dominant negative phenotype of this mutation in humans. This mutation was found in a patient suffering from an idopathic sinus node dysfunction characterized by marked sinus bradycardia. The inheritance pattern of this mutation could not be further delineated and therefore the association of 1631delC with cardiac disease is only speculative. Another study found that a missense mutation in human HCN4 (D553N) was associated with SA node 11  dysfunction and bradycardia in a Japanese family (271). In this family, SA node dysfunction and bradycardia were only two of the symptoms displayed by individuals carrying this mutation; the others include episodes of syncope, atrial fibrillation and chronotropic incompetence. Heterologous expression of D553N channels revealed decreased trafficking of channels to the cell surface in a dominant negative fashion. This decrease in channel surface trafficking correlates with reduced levels of If and is a likely explanation for the bradycardia observed in these patients. However, because of the complexity in symptoms demonstrated by these individuals a direct correlation between D553N and the disease state cannot be made with certainty. Perhaps the best evidence for the association of HCN channels with disease states come from two studies examining two families suffering from asymptomatic bradycardia. In the first study, 27 members of a third generation Italian family were found to carry a point mutation (S672R) in HCN4 (176). This mutation was found to be associated with the bradycardia phenotype in an autosomal dominant pattern. Of the 27 individuals whose genotypes were identified, those who were heterozygous for this mutation exhibited bradycardia (defined in this study as a HR below 60 bpm), while those that did not carry the mutation had heart rates of approximately 60 bpm. The mutation was shown to lie close to the cAMP binding site within the CNBD; nevertheless, modulation by cAMP was unaffected. These channels instead exhibited hyperpolarizing shifts in their voltage dependence of activation and a slowing of their deactivation kinetics. Both of these biophysical changes suggest that the decreased opening of channels at membrane potentials associated with diastole contribute to the bradycardia phenotype observed in these individuals. The second study again involving HCN4 found that a point mutation (G480R) in the pore forming domain was present in a family exhibiting asymptomatic bradycardia (196). Individuals heterozygous for this mutation had heart rates 12  below 55 bpm while homozygous wild type individuals exhibited heart rates greater than 63 bpm. Heterologous expression of this mutation showed that G480R resulted in a hyperpolarizing shift in the activation profile and slowing of channel deactivation kinetics, similar to what was observed with the S672R mutation. Despite the limited number of studies associating HCN channels with disease states the current evidence strongly suggests that HCN channels may be playing a prominent role in the manifestation of several known disease states, and therefore should be continued to be examined as a potential therapeutic target in these diseases. 1.1.4 Structure and function of the cyclic-nucleotide binding domain Modulation of HCN channels by cyclic-nucleotides (cAMP and cyclic guanosine monophosphate (cGMP)) occurs through binding of these nucleotides to the CNBD that comprises approximately 120 amino acid residues in the distal COOH-terminus. The C-linker domain, which is comprised of approximately 80 amino acid residues, connects the CNBD to the bottom of the S6 transmembrane helix (308). Insight into the structural features of the CNBD has been greatly aided by the crystallization of the HCN2 COOH-terminus (Figure 1.3). This crystal structure begins at the end of the S6 helix and includes the entire C-linker and CNBD domains, the region extending past the CNBD is not included. The crystal structure reveals a tetrameric four-fold symmetry with a hole down the central axis of the structure. The C-linker consists of six alpha helices A’-F’, while the CNBD is made up of alpha helices and a beta barrel with an orientation and structure that closely resembles the CNBD’s from other cyclic-nucleotide binding proteins including the cAMP-dependent PKA and the catabolite gene activator protein (CAP) (285). Each CNBD in the HCN2 channel contains an alpha helix (A) followed by a beta roll made up of 8 beta strands followed by another two alpha helices (B and C). Within the beta barrel between strands 6 and 7 there is an additional short alpha helix called the P-helix. The  13  majority of the inter-subunit interactions are mediated by amino acid residues with the C-linker region (50, 308). The crystal structure was solved in the presence of the two cyclic nucleotides known to modulate HCN channels: cAMP and cGMP. These structures demonstrate that the structural architecture of the two ligand-bound channels is very similar with the only differences found in subtle conformational differences of the ligand itself (308). For both ligands the binding pocket involves amino acid residues lying within the beta barrel and C-linker domains. The importance of these residues in HCN2 has been highlighted through electrophysiological studies, which have demonstrated that mutation of these residues abolishes channel modulation by either ligand. A total of seven residues have been shown to interact with these ligands, three of which reside in the beta roll, with the other four residing in the C-linker. Of these seven residues only one, R632 of the C-helix, has been shown to control the efficacy by which cyclic nucleotides enhance channel opening. Additionally, four residues within the C-helix (R632, R635, I636, and K638) have been shown to play a role in the favorable selectivity of cAMP over cGMP in HCN2 (48, 50, 308). In addition to its role in binding cyclic-nucleotides and enhancing channel opening, the CNBD has been shown to play a critical role in trafficking of channels to the cell surface. Removal of the CNBD from HCN2 has been shown to eliminate such trafficking of that channel (3, 217). Aspects of this thesis will highlight specific residues within the CNBD of HCN2 which play critical roles in regulating the forward trafficking through the endoplasmic reticulum (ER) and contribute to enhancing trafficking to the plasma membrane (191).  14  Figure 1.3 Structure of the mouse HCN2 C-linker and CNBD construct bound to cAMP. Ribbon diagram of the COOH-terminal region of HCN2 bound with cAMP. The HCN2 COOHterminal region is composed of two domains. The C-linker domain consists of six α-helices, designated A′ to F′, which are separated by short loops. The CNBD follows the C-linker domain and includes four α-helices (A, P, B, C) with a β-roll between the A- and B-helices. The β-roll comprises eight β-strands in a jelly-roll-like topology. Figure adapted from Zagotta et al, 2003.  15  1.1.5 Accessory proteins regulating HCN channels To date several studies have convincingly demonstrated that HCN channels exist as multimeric protein complexes, with auxiliary subunits playing important regulatory roles. Specifically, these auxiliary proteins have been shown to regulate trafficking, subcellular localization and several biophysical parameters. The MinK-related protein, MiRP1 (encoded by the gene KCNE2) was the first accessory protein reported to be an auxiliary subunit of HCN channels (28, 220, 290, 303). MiRP1, a member of the single-transmembrane family of proteins (1), has also been established as an auxiliary subunit to several other Kv channels (123, 173, 222, 223, 229). It has been demonstrated that MiRP1 interacts and co-immunoprecipitates with HCN2 channels in both rat neonatal cardiomyocytes and canine SA node tissue (220). However, the exact role played by MiRP1 in the regulation of HCN channels remains debatable. One study found that when MiRP1 is co-expressed with either HCN1 or HCN2 in Xenopus oocytes there is an enhancement in channel current density and an increase in activation kinetics, but no change in channel voltage dependence. In contrast, a subsequent study found that co-expression of MiRP1 with HCN4 in either Xenopus oocytes or Chinese hamster ovary (CHO) cells, also resulted in an enhancement of current amplitude, but conversely caused a slowing of activation kinetics and a hyperpolarizing shift in voltage dependence. A later study also examining the role of MiRP1 on HCN4 provided conflicting results, suggesting that MiRP1 did not have a modulatory role on HCN4 when co-expressed in human embryonic kidney (HEK) 293 cells. The strongest evidence supporting the role of MiRP1 as the beta subunit of HCN channels come from co-immunoprecipation experiments of MiRP1 with HCN2 from ventricular myocytes, but a more detailed analysis of this association needs to take place to fully support this idea. In addition to MiRP1, another transmembrane protein (Kupffer cell receptor (KCR) 1) has been identified as a potential interacting partner with HCN (175). KCR1 is a plasma 16  membrane protein with 12 putative transmembrane domains and, similar to MiRP1, has also been shown to be an interacting partner with other Kv channels. In the case of KCR1 it was shown that it interacts with HCN2 in both CHO cells and rat cardiomyocytes and this interaction results in reduced current densities and alters single channel current parameters. Several neuronal scaffolding proteins have been shown to interact with the COOHterminal domain of HCN channels. Of these proteins, the most well characterized binding partner has been the tetratricopeptide-repeat containing Rab8b interacting protein (TRIP8b) (144, 237, 315). TRIP8b has been shown to interact with a conserved tripeptide sequence in the COOH-terminus of both HCN1 and HCN2. TRIP8b is most well known for its role in the subcellular trafficking of vesicles. Similar to MiRP1 the data on the interaction of TRIP8b with HCN channels is conflicting and will need to be reconciled before the exact role of TRIP8b in HCN channel regulation can be elucidated. Finally, two less studied proteins Mint2 and SSCAM have been identified as binding partners of HCN channels in neurons; however these interactions have been, to date, explored only to a limited extent (131). 1.1.6 Post-translational modifications regulating HCN channels A few post-translational modifications have been shown to play important regulatory roles in HCN channels. HCN channels have been shown to be phosphorylated by two different protein kinases. Firstly, a tyrosine kinase of the Src family has been shown to interact with the C-linker-CNBD of HCN2 through its SH3 domain, resulting in phosphorylation within this region of the channel (112, 145, 316). Secondly, HCN channels have been shown to be phosphorylated by the p38-mitogen-activated (MAP) kinase, which belongs to the serine/threonine kinase family (215). In hippocampal neurons activation of p38 resulted in a positive shift in voltage-dependence. However, it has not been determined whether this  17  modulation is the result of direct phosphorylation of the channel or through the phosphorylation of another interacting protein which in turn regulates channel activity (215, 258). N-linked glycosylation is perhaps the most well studied and characterized posttranslational modification of HCN channels (102, 182, 236, 310). Each of the four HCN isoforms contain within their S5-P-loop linker a conserved sequon for N-linked glycosylation, and each of the isoforms have been shown to undergo N-glycosylation in either heterologous expression systems or native tissue. In the case of the HCN2 isoform, it has been demonstrated that elimination of N-glycosylation either through pharmacological inhibition or mutation of the putative glycosylated asparagine residue, results in a drastic reduction of channel expression at the cell surface (182). The functional role of this modification in other HCN isoforms has yet to be properly assayed. Parts of this thesis will provide novel insight into the regulatory roles Nglycosylation plays in the trafficking and functioning of HCN channel isoforms. 1.2 POTASSIUM CHANNEL OVERVIEW Classically, potassium channels can be classified into three distinct groups. The largest group consists of the voltage-activated and Ca2+-activated (Kv and BKca) channels which are composed of four alpha subunits that come together to form a tetrameric assembly (similar to already described HCN channels, see section 1.1) (46, 232). Each of these pore forming subunits consist of six transmembrane helices along with intracellular NH2- and COOH-termini. These potassium channels play significant roles in the repolarizing phase of the action potential. The second group consists of the potassium “leak” channels. These channels consist of four subunits with four transmembrane helices each which come together to form a tetrameric assembly. These channels are not gated and allow the flow of potassium ions down its concentration gradient. The final group is made up of the inward rectifying potassium (Kir) channels, which structurally are the simplest type of potassium channels (46). They consist of four, two transmembrane domain 18  subunits that once again form a tetrameric assembly. The Kir channels allow the flow of current in the inward direction and are blocked by endogenous polyamines, namely spermine, as well as magnesium ions which plug the pore forming region of these channels at positive potentials. The categorization of channels into each of these three groups or families is primarily based on nucleotide sequence homology. Further classification of these three families into sub-families (e.g. Kv1, Kv2, Kv3, etc) is accomplished through distinct nucleotide homology and biophysical properties. 1.2.1 Overview of the Kv4.x potassium channels Aspects of this thesis will provide novel insight into the regulation of the Kv4.x family of potassium channels. The Kv4.x family is comprised of three known isoforms, Kv4.1, Kv4.2 and Kv4.3, which are encoded by three separate genes (23). These three channels share a high degree of homology within their transmembrane regions, but differ with respect to their NH2- and COOH-termini (13, 35, 96). Kv4.x channels have been shown to be expressed in a variety of tissues, including the heart and brain (199, 214, 221). The biophysical features of these channels have been examined in these tissues as well as in variety of heterologous expression systems. The channels demonstrate some distinguishing biophysical properties which make them unique among the superfamily of Kv channels. These include: activation at subthreshold membrane potentials, rapid inactivation kinetics and rapid recovery from inactivation (23). Because of these characteristics they are referred to as “transient” outward currents. Studies involving animal models in which Kv4.x genes have been knocked out have confirmed that these channels represent the molecular correlate responsible for the transient outward current (Ito) in the heart and A-type current (IA) in the brain (16, 121, 194, 214). It should also be noted that Kv1.x channels, specifically Kv1.4, have also been shown to contribute to the generation of these current types (8, 199), but will not be discussed in detail in this thesis. 19  Similar to other Kv channels, the activation of Kv4.x channels is dependent on the ability of individual alpha subunits to sense changes in membrane potential through their voltage sensing domain, which includes a positively charged S4 transmembrane helix. A number of studies have attempted to elucidate the conformational changes this voltage sensor undergoes in response to changes in membrane potentials (83, 164, 264). Crystal structures of this voltage sensing domain have provided insight into the conformational orientation of these transmembrane helices within the membrane (68, 138, 156). Additionally, sensitive fluorescent based techniques have been used to measure movements of the voltage sensor at the atomic level (24, 38, 168). All of these studies have aided in advancing our knowledge of the possible structural rearrangements which may occur during changes in membrane potentials. However, despite the large number of studies devoted to addressing the specific movements of the voltage sensor the issue remains debatable. For the purposes of this thesis, it will simply be taken as a fact that the S4 transmembrane segment translates changes in membrane potential which contributes to channel gating. Inactivation properties and kinetics of Kv channels have been extensively examined. Classically, Kv channels undergo two primary types of inactivation: N and C-type. N-type inactivation also referred to as “ball and chain” inactivation, involves the NH2-terminus, which is thought to contain the “ball” domain and serves to occlude the intracellular face of the pore during its open state (114, 226). This type of inactivation is fast and can be eliminated upon removal of the NH2-terminus. C-type inactivation involves a collapse of the pore and is much slower than N-type inactivation (153, 202). C-type inactivation remains even with removal of the N-terminus, however, its kinetics are somewhat slowed which suggests that there may be some commonality to the mechanisms which regulate these two modes of inactivation.  20  In the case of Kv4.x channels the exact type(s) of inactivation that is (are) operative has not yet been determined. One study, in which the NH2-terminal domain of Kv4.2 was removed, found that this deletion did not eliminate the fast inactivation process in these channels, but it did result in a slowing of this process (12). This is in stark contrast to channels like Kv1.4 where removal of the NH2-terminal domain results in a complete loss of fast inactivation. In the case of Kv4.3 removal of the NH2-terminal domain did eliminate the fast inactivation component in these channels; therefore there may be slightly different structural arrangements even within the Kv4.x family which account for these differences. Studies examining C-type inactivation in both Kv4.1 and Kv4.2 have shown that these channels do not exhibit the criteria which are typically seen in other channels undergoing this mode of inactivation (120, 122). Namely, inactivation is not disrupted by tetraethylammonium (TEA) or high external potassium concentrations (18, 42). Further examination of the inactivation process in Kv4.x channels will need to be conducted in order to determine the exact mechanisms regulating this process. 1.2.2 Physiological role of Ito in the heart In the heart Kv4.x channels contribute to the generation of the transient outward current (Ito) which is an important electrical component to the repolarization phase of the cardiac action potential. Ito can be separated into separate slow (Ito,s) and fast kinetic components (Ito,f) (194). Kv1.4 has been identified as the molecular correlate to Ito,s (210), while Kv4.1-3 have been associated with the Ito,f component (193). The expression of these channels and the overall levels of Ito have been shown to be variable throughout the myocardium (234). Additionally, the degree to which Ito is responsible for membrane repolarization can vary among different species. In rodents Ito plays a large part in the repolarization of the membrane potential, while in humans Ito only plays a role in the rapid repolarization which takes place immediately after the upstroke phase of the action potential (194, 287). The differing physiological roles and relative 21  importance of Ito between species is most likely due to varying levels of tissue expression and distribution of Kv4.x channels in the myocardium of these species (280, 287). Several studies have provided compelling evidence that the expression and distribution of Kv4.x can vary between species. In rats and lower mammals Kv4.2 has been identified as the primary molecular correlate to Ito,f (291, 292). Knockout studies using antisense oligonucleotides against Kv4.2 have shown that Ito,f is eliminated in these animals, while antisense oligonucleotides against Kv4.3 had no effect (27). However, this is only the case for the atrium of these animals; in the ventricles it has been shown that both Kv4.2 and Kv4.3 are expressed and that they undergo heteromulterization and interact with a common accessory subunit, potassium channel interacting protein 2 (KChIP2) to produce Ito,f in these cells (91). The use of antisense oligonucleotides to either Kv4.2 or Kv4.3 resulted in a reduction, but not elimination of Itof (76). Additionally, the kinetics of Ito,f vary across the rodent ventricle and this variation closely follows the gradient distribution of Kv4.2 expression across it (91). Interestingly, Kv4.3 expression is evenly distributed across the ventricle. Studies examining the molecular correlate of Ito,f in the human ventricle have demonstrated that Kv4.3 acts as the primary pore-forming subunit (21, 67). As with the rodent heart, KChIP2 plays an important role in regulating Ito in the human heart (230). There have been conflicting studies regarding the expression of KChIP2 across the ventricular wall. One study showed that KChIP2 expression is variable throughout the ventricle and its expression closely follows the gradient of Ito observed (230). Another report found differing levels of KChIP2 messenger ribonucleic acid (mRNA) expression, but no difference in KChIP2 protein produced (55). Thus the exact role of KChIP2 in the modulation of Ito in the human heart remains to be elucidated.  22  1.2.3 Physiological role of IA in the central nervous system In the central nervous system (CNS), IA has been shown to be instrumental in regulating neuronal excitability and processing (i.e. long term potentiation and depression) (251, 252). The distribution of Kv4.x channels has been examined throughout the CNS. Studies examining the mRNA expression profile of Kv4.x channels in the adult rat brain have shown that Kv4.1 is the least expressed isoform (252). The expression profile of Kv4.2 and Kv4.3 is complex and varied throughout the nervous system. In general certain cell types have been shown to selectively express one isoform over another while others have shown overlapping expression patterns (252). Briefly, Kv4.2 has been found to be expressed exclusively in cells of the caudate-putamen, pontine nucleus and some nuclei in the medulla. In the case of Kv4.3 it is the predominant isoform in the substantia nigra pas compacta, the retrosplenial cortex, the superior colliculus and the amygdala. The hippocampus has been found to express both Kv4.2 and Kv4.3 (252). Hippocampal CA 1 pyramidal cells express Kv4.2 as the predominant form, with a subset of neurons also expressing Kv4.3. Pyramidal interneurons have been shown to exclusively express Kv4.3, while both Kv4.2 and Kv4.3 are present in CA2 and CA3 pyramidal cells. In the thalamus both Kv4.2 and Kv4.3 are present, with Kv4.3 primarily expressed in the lateral nuclei and Kv4.2 in the medial nuclei. Finally, in the cerebellum Kv4.3 is found primarily expressed in the Purkinje cells and interneurons, while both Kv4.2 and Kv4.2 are found in cerebellar granule cells (252). Although expression profiles for both mRNA and protein levels have been carefully determined for Kv4.x channels in the rat brain, the physiological relevance in these different regions has only been superficially explored. The hippocampus is one area which has received a significant amount attention, and will be discussed here further. Studies examining the expression of Kv4.x channels in the hippocampus have shown that these channels are localized 23  to the somato-dentritic regions of these neurons (135, 165). Studies examining the somatodentritic region of hippocampal CA1 pyramidal cells have shown that A-type currents increase along a gradient, which follows from the soma to the distal dendrites (135). The increase in Atype current amplitudes along the distal dendrite has been attributed to regulating the levels of back-propagating action potentials (bAPs) in these neurons (305, 312). Action potentials which are triggered at the level of the soma propagate along the axon to the axon terminal where they result in neurotransmitter release. In addition they also travel from the soma to the proximal and distal dendrites and signal that the soma has been sufficiently stimulated to fire an action potential. These bAPs have also been identified as a potential mechanism for regulating neuronal processes such as synaptic plasticity long-term potentiation (121, 312). To better understand the roles of Kv4.x channels in processes such as synaptic plasticity, an understanding of the synaptic architecture of these neurons is necessary. In general synaptic plasticity is mediated through two subtypes of glutamate-gated ion channels which are concentrated at post-synaptic membranes: α-amino-3-hydroxyl-5-methyl-4-isoxazole-propionate (AMPA) and N-methyl-D-aspartic acid (NMDA) type receptors. The AMPA receptors are responsible for regulating the majority of fast excitatory synaptic transmission by binding the synaptically released glutamate. NMDA receptors are blocked by intracellular magnesium at resting membrane potentials and are only activated upon glutamate binding and sufficient membrane depolarization, which is achieved by way of sodium influx through the AMPA receptors (299). Once opened, calcium influx through NMDA receptors further activates various signaling cascades that, in turn, trigger increased trafficking of AMPA receptors to the postsynaptic membrane (80). Important to note is that certain synapses lack AMPA receptors and contain only NMDA receptors. The inability of these post-synaptic membranes to be depolarized through the activation of AMPA receptors renders these synapses “silent”, meaning 24  they are unable to be excited through glutamate binding alone. Therefore, any depolarization that the NMDA receptors sense in these synapses must come from a distal origin along the dendrite (23). The A-type currents play a significant role in preventing sufficient depolarizing potentials from reaching these “silent” synapses, thus, preventing their activation. As such, mechanisms which regulate the trafficking and biophysical properties of Kv4.x channels along the dendritic spines of these neurons can play significant roles in modulating membrane depolarization and consequently, NMDA activation and excitability of these “silent” synapses (23). Studies have demonstrated that during processes such as “theta-type” long-term potentiation (LTP) action potentials are induced which then back-propagate along the dendrites and depolarize synapses (281, 289). This depolarization is promoted through mechanisms which modulate Kv4.x channel opening. Activation of the ERK/MAPK pathway during β-adrenergic stimulation of these neurons results in phosphorylation of Kv4.x channels and a depolarizing shift in their activation potentials which decreases channel open probability (23, 289). The reduced repolarizing currents through Kv4.x channels under these conditions allows for sufficient depolarization of the postsynaptic membrane, activation of NMDA receptors, calcium influx and enhanced excitability, both of which promote LTP in these neurons (23). 1.2.4 Post-translational modification of Kv4.x channels To date there have been only a limited number of examples involving either direct or indirect modulation of Kv4.x channels through post-translational modifications. The most well studied case involves changes in Kv4.x channel activity and expression by channel phosphorylation. Studies examining phosphorylation of Kv4.x channels have often yielded complicated and contradictory results. In ventricular myocytes, activation of protein kinase C (PKC) results in suppression of Ito (10, 187). In hippocampal pyramidal neurons on the other hand, activation of PKA or PKC results in depolarizing shift in the activation curve (108, 306), 25  which has been shown to modulate bAPs (see 1.2.2 for details). When examined in Xenopus oocytes, PKC activation suppressed current expression through Kv4.2 and Kv4.3 (187). Interestingly, PKA activation in this same expression system reportedly has no effect on Kv4.2 currents (243), whereas in hippocampal neurons, PKA activation resulted in a depolarizing shift in activation properties (108). The authors of this same study further demonstrated that coexpression of Kv4.2 with KChIP3 (interacting ancillary subunit – see section 1.2.4) restored PKA modulation of Kv4.2 currents in neurons. They convincingly demonstrated that activation of PKA in oocytes leads to phosphorylation of Kv4.2 at serine 552, which, in the presence of KChIP3, led to a shift in the activation curve (242). These data taken together demonstrates that Kv4.x channel modulation through phosphorylation is highly dynamic and complex and also suggests that ancillary subunits associated with Kv4.x channels in vivo may play an important role in phosphorylation mediated changes observed. Although most voltage-gated potassium channels have been shown to undergo some degree of N-glycosylation, Kv4.x channels remain one of the few exceptions. None of Kv4.x channel contains an N-glycosylation sequon (N-X-S/T) within any of their extracellular linkers. Previous studies have shown however that treatment of ventricular myocytes with neuramidase, which selectively inhibits N-type glycosylation, results in decreased Ito and prolongation of action potentials (272). Similarly, expression of Kv4.3 in a sialic acid deficient cell line also reduces overall current density values (272). Taken together, it appears that N-glycosylation is playing an important, but likely indirect, role in regulating Kv4.x channel activity, possibly through the regulation of an interacting partner. 1.2.5 Accessory proteins regulating Kv4.x channels Although, our current knowledge of post-translational modifications regulating Kv4.x channels is limited our knowledge of accessory subunit modulation of Kv4.x channels is 26  significantly more extensive. Numerous accessory subunits have been shown to modulate the biophysical and or trafficking properties of these channels both in vivo and in heterologous expression systems. The first accessory subunits to be discovered were the Kvβ subunits, which interact with most members of the Kv channel superfamily (185). Kvβ subunits are cytoplasmic proteins that do not have transmembrane domains and do not contain any glycosylation or signal peptide sequences (90). Three different Kvβ genes have been identified to date, Kvβ1-3 (103, 139). Each of these subunits modulate channel trafficking and function in differing manners. Primarily they have been shown to modulate voltage dependent properties of channels (73) and enhance cell surface expression (253). One of the less obvious roles of Kvβ subunits is their ability to act as cytoplasmic redox sensors. These proteins have been shown to have a high degree of similarity to several other oxido-reductase enzymes (90). With regard to Kv4.x channels, it has been demonstrated that Kvβ1.2 confers sensitivity to redox modulation and hypoxia in Kv4.2 channels. Upon association with Kvβ1.2, Kv4.2 has been shown to be modifiable by several reducing agents (209). Unlike its effects on other potassium channels, the association of Kv4.2 with Kvβ1.2 did not enhance channel surface expression (209). Kv4.3 has been shown to interact with each of the three Kvβ subunit isoforms (57, 297). Its interaction with Kvβ1 and Kvβ2 results in an increase in current density with no effect on channel kinetics (297). Conversely, Kvβ3 shifts the steady-state inactivation curve and slows recovery from inactivation, but has no effect on channel surface expression (297). Although there is strong evidence supporting the role of Kvβ subunits in Kv4.x modulation, a much better defined set of proteins linked to the regulation of Kv4.x channels are the potassium channel interacting proteins (KChIPs) To date there have been four KChIP isoforms identified (KChIP1-4) (277). All four of these proteins are highly expressed in the brain (5, 227), while KChIP2 is the only isoform found 27  abundantly in the heart (56, 208, 230). The KChIPs belong to the neuronal calcium sensor and EF-hand protein families (277). The NH2-terminus varies between the different KChIPs, while the COOH-terminus is well conserved and contains the four EF-hand-like calcium binding motifs (277). The exact role of these EF hand domains in the regulation of KChIP or Kv4.x channels is not well understood. Interestingly, calcium dependent modulation of IA has been previously reported (278), which suggests that the binding of calcium to these EF-hand domains may be playing a significant role in channel modulation. A recent study has demonstrated that a functional cross-talk may exist between the voltage-gated T-type calcium channel (Cav3.2) and Kv4.2 in certain neurons (7). This study suggested that calcium entry through Cav3.2 binds the EF-hand of KChIP3 which serves as the bridge to modulate Kv4.2 function. Specifically, by modulating the voltage-dependence of these channels it causes a hyperpolarizing shift in the window current which promotes the opening of channels at subthreshold potentials (7). In general KChIP association with Kv4.x channels causes a hyperpolarizing shift in the activation curve, slowing of inactivation kinetics, and an increase in the rate of recovery from inactivation (5). In addition to its role in gating, KChIPs play a significant role in promoting Kv4.x surface expression. There have been numerous studies demonstrating this regulation both in vivo and in heterologous expression systems (256). One particularly study examining a splice variant of KChIP4 (KChIP4a) demonstrated the importance of this ancillary subunit in Kv4.2 trafficking (148). Although the interaction between the two proteins was shown not to not be dependent on channel phosphorylation, the modulatory effects of KChIP4a on Kv4.2, which include enhanced stabilization and membrane expression, do depend on phosyphorylation of Kv4.2 at position S552 (148). In the heart, targeted knockout of the gene encoding KChIP2 results in a complete loss of Ito,f in ventricular myocytes, this is presumably through the elimination of Kv4.2 28  trafficking to the cell surface (265). Overall, the evidence supporting KChIPs as a strong promoter of Kv4.x cell surface expression is clear and convincing. In addition to KChIPs another calcium binding protein, frequenin (also known as neuronal calcium sensor protein-1 or NCS-1) has also been shown to be a Kv4.x auxiliary subunit (92, 188). In heterologous expression studies, frequenin has been shown to promote Kv4.2 expression at the plasma membrane. Frequenin has also been shown to coimmunoprecipitate and interact with Kv4.2 in both COS cells and isolated mouse brain membranes (188). The physiological relevance of the frequenin-Kv4.2 interaction is supported by co-localization experiments which demonstrate that these two proteins show overlapping expression patterns in certain neurons within the mouse brain. Also, in the heart frequenin is expressed in adult mouse ventricles and co-immunoprecipitates with Kv4.3 from adult mouse ventricular extracts (93, 188). Interestingly the expression of frequenin varies depending on developmental state. Frequenin levels have been shown to be the highest in fetal and neonatal mouse hearts compared with the adult heart (189). Thus, it is possible that Kv4.2 modulation via frequenin may only occur during specific developmental stages. Potassium channel associated proteins (KChAPs) are another class of proteins which have been shown to serve as important chaperone partners for several Kv channel members (133, 286). Specific to Kv4.x channels, KChAP has been demonstrated to play a prominent role in the trafficking of Kv4.3 channels to the cell surface (134). It has been demonstrated that in both Xenopus oocytes and mammalian L-cells, co-expressing Kv4.3 with KChAP, the current density values were significantly larger than in cells expressing Kv4.3 alone (134). Additionally, in both systems KChAP did not alter either activation or inactivation processes. The resultant increase in Kv4.3 expression at the plasma membrane was attributed to an overall increase in Kv4.3 protein  29  production. A direct association between the two proteins was demonstrated through coimmunoprecipitation experiments involving KChAP and Kv4.3 from the adult rat heart (134). In the above discussion of KChIPs, studies were presented which demonstrate the ability of these proteins to restore native Kv4.x properties. However, certain Kv4.x gating parameters were found to be altered as a result of KChIP association, including inactivation kinetics. The interaction of KChIP with Kv4.x channels resulted in a slowing in the rate and recovery from inactivation. This suggests that another accessory subunit may be regulating the inactivation properties of Kv4.x channels in native tissue. This discrepancy in inactivation kinetics between heterologously expressed Kv4.2 and native current properties was somewhat resolved through the identification of the dipeptidyl aminopeptidase-like protein (DPPX). DPPX was shown to associate with Kv4.x channels and to facilitate their trafficking and membrane targeting as well as restoring the gating properties of native channels in heterologous expression systems (45, 184). Specifically, co-expression with Kv4.2 was able to reconstitute native inactivation kinetics, a property that is not restored by KChIP expression alone (184, 249). Taken together, DPPX has been shown to have similar modulatory modifications on Kv4.x channels to those of KChIP, with the exception that it is also able restore native inactivation kinetics and therefore appears to be an indispensible accessory subunit in vivo. The majority of accessory subunits presented thus far have been shown to play a regulatory role in trafficking and/or channel function. What role do accessory subunits play in stabilization of Kv4.x channels at the cell surface? The localization and stabilization of Kv4.x channels at the cell surface have been shown to be under the influence of several cytoskeletal proteins. An example of this is the actin-binding protein filamin. Using a yeast two-hybrid approach, Kv4.2 was found to interact with filamin (212). A member of the actin-binding proteins, filamin cross-links actin filaments into a specific orientation and facilitates actin 30  organization. Kv4.2 has been shown to co-immunoprecipitate with filamin from both heterologous cells and brain extracts (212). Functionally, the interaction of Kv4.2 with filamin has been shown to enhance current density by nearly three fold when comparing the expression of Kv4.2 in filamin(+) and filamin(-) cells. Also, Kv4.2 was shown to co-localize with filamin at the filopodial roots in filamin(+) cells, while in filamin(-) cells Kv4.2 demonstrated a nonspecific expression pattern with no localization to filopodial roots (213). The studies presented thus far have attempted to elucidate the physiological relevance of accessory proteins in the regulation and modulation of Kv4.x channels in native tissue. The difficulty faced in each of these studies is the lack of the entire complement of accessory proteins necessary to produce the native Ito or IA currents. As accessory proteins continue to be identified we will gain a fuller understanding of the complete macromolecular complex contributing to the regulation of Kv4.x channels in vivo. 1.2.6 Kv4.x channels and disease Kv4.x channels have been shown to be important contributors in the manifestation of a number of disease states. In particular Kv4.x channels have been shown to play an important role in cardiac pathophysiology (84). Many studies have shown that the deregulation of Kv channels can lead to numerous forms of cardiac disease. Because of their role in the repolarization of cardiac myocytes, disruptions in potassium channel function has been shown to be a leading cause of Long-QT syndrome (101). Initially identified through studies examining defects in the gene coding for the human Ether-à-go-go Related Gene (hERG) channel, it has now been shown that mutations in genes coding for a number of other potassium channels can also lead to LongQT syndrome. Specifically in the case of Kv4.x channels, it has been demonstrated that reductions in Ito can lead to action potential prolongation. In one study, transgenic mice expressing only an N-terminal fragment of Kv4.2 exhibited a marked reduction of Ito and a 31  significant prolongation of action potential duration (288). This study also demonstrated that several of these mice developed age related dilated cardiomyopathy which was associated with cardiac dysfunction in these animals. These mice exhibited characteristics of congestive heart failure, which included chamber dilation and marked hypertrophy (288). In a similar study, mice expressing a dominant negative form of Kv4.2 (a point mutation in the pore-forming domain) resulted in complete elimination of Ito (17). These mice, too, experienced a prolonged action potential; but they exhibited normal behavior and no detectable arrhythmias. In several studies examining myocardial infarctions (MI) in a rat model (110, 111, 219, 298), left ventricular myocytes were shown to exhibit marked reductions in levels of Ito. This decrease in Ito occurs prior to cardiac hypertrophy and persists for 16 weeks post MI. Important to note is that reductions in Ito levels were correlated with reduced levels of Kv4.2 and Kv4.3 mRNA and protein levels (110, 111). This evidence suggests that Ito may be playing an important role in the electrical remodeling of cardiac tissue that occurs post MI. This is of importance due to the fact that electrical remodeling has been shown to lead to various types of arrhythmias. In fact, several studies have directly demonstrated that Ito plays an important role in atrial fibrillation (AF). Individuals exhibiting AF experience prolonged atrial action potentials and refractory periods and Ito is significantly reduced in atrial myocytes from patients with AF compared with those that exhibit normal sinus rhythm (273). The results were corroborated in canine models with either chronic or brief episodes of AF (71). In rabbits, induction of rapid atrial pacing resulted in a marked decrease in Kv4.3 mRNA expression within 24 hours (26). While the roles of Kv4.x channels in the generation of cardiac pathophysiology need to be further explored, the studies outlined here provide substantial evidence that Kv4.x channels play important roles in various forms of cardiac disease and should be considered as potential candidates for the development of novel therapeutics. 32  The presence of Kv4.x channels in high densities at the somatodendritic regions of various neurons throughout the CNS also make them prime candidates as contributors to neurological disorders (135). For example, epilepsy has been linked to Kv4.x channel deregulation. In several studies involving either animal models or human patients with epilepsy, A-type currents have been shown to be significantly downregulated (19, 78, 269). 1.3 BIOGENESIS AND SURFACE TRAFFICKING OF ION CHANNELS 1.3.1 Gene expression and transcriptional regulation The mechanisms responsible for regulating the expression of ion channels at the cell surface begin at the level of the nucleus, and are controlled through gene transcription and mRNA processing. Disruption or changes in either of these processes can result in significant changes in the expression of channels at the cell surface, and in certain instances change channel function which may result in pathophysiological conditions. Gene transcription is controlled through regulatory elements, which include both activators and repressors which bind to genomic DNA sequences upstream of the gene start site and can either upregulate or downregulate expression of the gene (47). Ribonucleic acid (RNA) polymerases are responsible for generating the long stretches of heteronuclear RNA during transcription (250). This immature form of mRNA contains, 5’ and 3’ regulatory stretches referred to as untranslated regions (UTRs). Both 5’ and 3’ UTRs have been shown to play a role in the expression of ion channels. In addition to the UTRs this immature mRNA consists of both exonic and intronic RNA. Eventually, the mature form of the mRNA will contain only exonic RNA, which comprises the coding sequence for the protein. During the removal of intronic RNA, these mRNAs can undergo differential splicing, generating differing coding sequences (284). In the case of ion channels this differential splicing has been shown to play a role in the regulation 33  of several ion channels. In the case of the large calcium activated potassium (BK) channel has been shown to produce multiple splice variants, each of which exhibits unique expression and biophysical profiles (41). Nav1.5 channels have been found to undergo abnormal COOHterminal splicing during heart failure, which results in prematurely truncated channels which are non-functional and have dominant negative effects on the native channel mRNA (157, 245, 245). Finally, once mRNA processing is completed the mature form exits the nucleus and travels to the cytosol where it will bind with a single ribosome and the process of translation begins (154, 186). In the case of ion channels, which are intrinsic transmembrane membrane proteins the translation must occur at the rough endoplasmic reticulum (240, 268). 1.3.2 Translation through the endoplasmic reticulum Ribosomes are composed of two separate subunits made up of ribosomal RNA (rRNA) and protein. The small subunit is responsible decoding the mRNA and the large subunit is responsible for the peptide bond formation through the peptidyl transferase center located at the interface between the small and large subunits. As transfer RNAs (tRNAs) bring amino acids to the site of mRNA-ribosome complex the nascent polypeptide begins to form (186, 240). As peptide synthesis progresses, the peptide chains are often able to begin a self regulating processes which can control their own translation, including elongation, pausing and termination steps. Once a series of specific amino acids are translated the nascent channel peptide-ribosome complex is exported to the endoplasmic reticulum to complete translation. Typically, the nascent peptide of membrane proteins is transported to the ER through binding of the signal recognition particle which recognizes a specific sequence of amino acids. In the case of ion channels, however, a signal peptide is not employed and the process is much more complicated. Multiple transmembrane segments must be sequentially inserted into the plasma membrane through a  34  series of target sequences which direct transport to the ER and insertion into the ER membrane (58). 1.3.3 ER retention signals and quality control mechanisms The process of translocation and translation through the ER is followed by an intricate series of quality control mechanisms which ensure only properly folded and assembled channels undergo forward trafficking to the Golgi apparatus. Incorrectly folded proteins are bound by chaperones where they are retained in the ER (74). If proper folding is not successfully achieved, these proteins undergo retranslocation to the cytosol where they are degraded through the ubiquitin proteasome system (UPS) (see section 1.4). Correctly folded proteins are compartmentalized into coat protein complex (COP) II vesicles and are trafficked to the ERgolgi intermediate compartment (ERGIC) (11, 94). In the case of most proteins undergoing translation through the ER, errors which result during protein folding have been shown to be an important determinant in causing ER retention. In the case of ion channels they are believed to be assembled into oligomeric protein complexes prior to their exit from the ER. It has been proposed by several groups that failures in oligomeric assembly result in the unmasking of ER retention signals. These signals have been shown to consist of di-arginine motifs RR or RXR, where X can be any amino acid and they have been well characterized in a number of potassium channels, the best example of which is the inwared rectifying Kir6.2 potassium channel (309). Exposure of an RXR motif located in the COOHterminus of the channel resulted in decreased surface trafficking and ER retention. This motif has also been shown to play an important role in a number of other channels including Kir2.1 and hERG (160). RXR motifs are also present in many other Kv channels; however, their role in trafficking has yet to be fully elucidated. The ability of these motifs to cause ER retention is  35  thought to be partially due to their ability to bind and interact with a family of proteins known as 14-3-3 proteins (200). 14-3-3 proteins are important regulators in processes including signal transduction, cell cycle control and apoptosis. Additionally, 14-3-3 proteins have been shown to serve as adaptor proteins which can promote cell surface expression of membrane proteins (200). It has been demonstrated that dibasic signal motifs in ion channels can bind these 14-3-3 proteins. One particular study showed that 14-3-3 motifs were able to interact and bind to the dimer and tetrameric form of Kir6.2, but not in its monomeric state (200). Also, only Kir6.2 channels in an oligomeric state were found to ultimately traffic to the cell surface. Interestingly, this same group was able to show that COPI, which is a protein coat involved in the recycling of proteins to the ER from the Golgi, competes with 14-3-3 proteins for binding to the RXR motif in this channel (203). The dibasic motif found in the twin-pore potassium channels TASK-1 and TASK-3 also binds COPI and again this action is competitively inhibited by 14-3-3 protein binding (224). Taken together, it is suggestive that unfolded proteins which are able to escape the ER retention machinery are returned to the ER through binding of COPI to ER retention motifs, whereas interaction of these same motifs with 14-3-3 proteins promotes their trafficking to the cell surface (9). In addition to diarginine motifs already discussed, dilysine repeats, which also serve as ER retention signals, also bind to COPI. Finally, a four amino acid motif KDEL has also been characterized as an ER retention signal. KDEL binds to KDEL receptors found on transport vesicles and regulates Golgi-ER recycling (34, 34, 309). Mutation of residues residing within ER retention motifs have also been associated with several disease phenotypes. Specifically, type-1 long-QT syndrome (LQTS) has been attributed to mutations within the potassium voltage-gated channel, KQT-like subfamily (KCNQ) channels which result in a decrease in Iks current and prolonged action potentials. Many of the mutations 36  in this channels which cause LQTS have been localized to the S2-S3 linker. Interestingly, this region of the channel also contains an RXR retention motif. Specifically, KCNQ1 possesses an RLR motif in the S2-S3 linker region of the channel (152). Mutation of L191P in the KCNQ1 channel in a Chinese pedigree with LQT1 has been shown to drastically reduce channel surface expression. Another study further demonstrated that this mutation decreases the hydrophobicity of the X amino acid within the RXR motif and increases its likelihood of being exposed and thus undergoing ER retention (238). Although much of the work on ER retention signals has been carried out on potassium channels, several other ion channels have been shown to contain these retention motifs as well. N-Methyl-d-aspartate (NMDA) receptors are glutamate-activated ion channels which are found as two sub-types, NR1 and NR2. These channels contain RXR motifs in their COOH-terminus. In the case of a specific splice variant of NR1, its COOH-terminal retention motif can be masked through interaction with a PDZ protein and promote channel surface expression (259). A later study, also examining NR1, showed that intracellular signaling events which trigger the activation of specific kinases and phosphorylation of NR1 are also able to mask retention signals and promote surface expression (248). NR2 subunits also possess a retention motif within their COOH-terminus which prevents them from trafficking to the cell surface in a homomeric form. Heteromeric assembly with NR1 subunits masks this retention motif within NR2 and allows forward trafficking to occur. Splice variants of NR2 which lack this COOH-terminal region containing this motif are able to traffic in a homomeric form. Retention signals have also been identified in the Cav channels. The I-II linker of skeletal muscle Cav1.1 L-type channel contains a retention motif which significantly impairs membrane trafficking; this motif is masked in the presence of the calcium channel β subunit which allows for the rescue of channel surface trafficking (22). 37  In addition to specific retention signals that prevent misfolded channels from trafficking onward to the Golgi apparatus, assembly with accessory proteins has also been shown to play an important role in ER quality control. Again studies involving Kir6.2 have shown that interaction with its accessory subunit sulfonylurea receptor (SUR) 1 is critical in ensuring successful traffic from the ER. It is clear that there exist numerous mechanisms which ensure that only properly folded/assembled channels traffic to the cell surface. However, there also exist regulatory mechanisms which promote forward trafficking to the Golgi apparatus. These mechanisms have also received a significant amount of attention with respect to ion channels and will be discussed further. 1.3.4 Mechanisms regulating forward trafficking from the ER to the Golgi apparatus Although a number of mechanisms contributing to the regulation of forward trafficking of membrane proteins have been shown, with regards to ion channels, the most closely studied of these have been specific forward trafficking motifs. Again, most of the studies examining forward trafficking signals have involved potassium channels. To date there has yet to be an established widely employed consensus sequence of amino acids that facilitates forward trafficking of ion channels. In potassium channels the forward trafficking signals which have been identified have been quite diverse. The first forward trafficking sequence was identified in the inward rectifying potassium channel Kir2.1 (160, 262). Several members of the Kir family have been shown to be trafficked to the cell surface at varying levels in Xenopus oocytes. To better understand the sequence determinants which contribute to the differences in expression levels among the Kir channels, the COOH-terminus of Kir2.1 was replaced with the COOHterminus of other Kir channels. This study demonstrated that chimeric Kir channels containing the COOH-terminus of Kir2.1 were more efficiently trafficked to the cell surface. The presence of a forward trafficking motif as opposed to a sequence altering channel assembly or folding was 38  verified through careful examining of total protein expression, ER-Golgi trafficking, cell surface expression and channel functioning. The COOH-terminus of Kir2.1 was analyzed in further detail through a series of truncation and mutant constructs which led to the identification of the FYCENE sequence motif (160). This motif was shown to be sufficient for the enhancement of cell surface expression in Kir channels. Interestingly, this sequence motif was shown not to be position dependent; inclusion of the motif in either the NH2- or COOH-terminal of non-channel proteins significantly enhanced their cell surface expression. The FYCENE forward signal cannot override the RXR retention signal in Kir6.2 channels, however. If the retention signal in these channels is removed, however, through a specific truncation (KirΔ36), the presence of FYCENE can further enhance the cell surface expression of these channels. FYCENE enhances surface expression of other channels, as well. Typically, cell surface expression of Kv1.2 is low in the absence of its accessory subunit Kvβ2, which is believed to facilitate the folding and maturation of Kv1.2 in the ER. However, the insertion of FYCENE into this channel enhanced the cell surface expression of Kv1.2 to the same extent as Kvβ2, and occluded the ability of Kvβ2 to further enhance that surface expression (160). Although, the FYCENE is perhaps the best characterized forward trafficking motif in ion channels several other motifs have also been identified as likely forward signaling motifs. Various other channels have also been shown to contain different forward trafficking motifs. Kir1.1 another inward rectifying channel was shown to contain a sequence motif, VLSEVDETD which promotes channel surface expression. Both VLSEVDETD and FYCENE contain diacidic motifs – EVD and ENE for Kir1.1 and Kir2.1 (87), respectively. Similar diacidic motifs (DXE) have been identified in non-channel proteins and have also been shown to be important forward trafficking determinants in these proteins. Kv1 channels contain several different forward trafficking motifs. Kv1.4 contains a VXXSL motif in its COOH-terminus and 39  is effectively trafficked to the cell surface (314). In contrast Kv1.5 also contains a LXXSL motif in its COOH-terminus; but it does not reach the cell surface with the same efficiency (143, 146). Interchanging the two motifs between the channels reverses the effects on cell surface expression. Thus, the first amino acid of each motif is responsible for the differences in cell surface expression observed. The final amino acid in the motif also appears important. Both Kv1.4 and Kv1.5 are able to traffic to the cell surface in the absence of the Kvβ accessory subunit. Kv1.2 contains a VXXSN motif in its COOH-terminus and requires Kvβ for effective surface trafficking. Mutation of the N to L, allows surface expression of this channel even in the absence of the Kvβ . Although, the presence of the Kvβ subunit is able to further enhance trafficking, this provides strong evidence that forward trafficking motifs in Kv1 channels are able to promote surface expression independently of the Kvβ subunit. Although, this discussion of trafficking motifs has focused on short amino acid sequences, entire domains of specific channel have also been identified to promote forward trafficking. Most notably, the CNBD of the hERG, ERG and HCN2 channels have been shown to greatly enhance surface trafficking (3, 217). In the thesis we will further discuss the importance of CNBD in the trafficking of HCN2 channels. In particular, we demonstrate the importance of a four amino acid motif (EEYP) in the B-helix of the HCN2 CNBD which serves as a forward trafficking motif and contributes to the surface trafficking of HCN2 (192). Undoubtedly, forward trafficking or export motifs along with accessory subunits play important roles in modulating ER-Golgi trafficking. However, the proteins responsible for the packaging of ion channels need to be addressed. Membrane traffic between the ER and Golgi is controlled by many molecular players. The transport of cargo from the donor (ER) to the acceptor (Golgi) is defined by several stages (137). The first stage involves the formation of coat-protein-mediated cargo and vesicle budding. It has widely been characterized that COPII 40  and COPI regulate the inter-compartmental trafficking between the ER and Golgi. Anterograde trafficking is under the control of COPII, while retrograde transport of cargo is regulated by COPI (11). Anterograde trafficking to the Golgi is a guanosine triphosphatase (GTPase) dependent process which has been shown to require the activation of Sar1-GTPase. Sar1-GDP is converted to the activated form of Sar1-GTP through a guanine exchange factor (GEF) SEC12. This activation of Sar1 at the ER promotes COPII coat recruitment to the ER membrane. Continued activation of Sar1 and the cargo selection properties of COPII result in a vesicle bud. Specifically, it has been suggested that forward trafficking motifs in Kv1.1 and Kv1.4 may interact with COPII. Also, dileucine motifs, which are present in many different channel types, have been shown to bind COPII. The Sar1-GTP:COPII coated vesicle buds from the ER and connects with cytoskeleton for anterograde transport to the cis-Golgi (11, 94). ER to Golgi trafficking of vesicles is dependent on the integrity of microtubules, specifically the microtubuleminus-end-directed motor kinesin. Microtubule-plus-end-directed motor dynein regulates Golgi to ER trafficking (see section 1.3.4 for more details on microtubule motors) (155). Recruitment of Rab proteins to the vesicle and interactions with various tethering proteins, including but not limited to soluble N-ethylmaleimide-sensitive factor-attachment protein receptors (SNAREs) and syntaxin allow fusion of the cargo-containing vesicles to the Golgi membrane (52, 116). 1.3.5 Golgi to plasma membrane trafficking The mechanisms by which ion channels traffic between the Golgi and plasma membrane have been explored only to a limited extent, certainly much less than ER-Golgi trafficking. The detailed sorting of ion channels is likely to occur at the level of the trans-Golgi network. How do ion channels target themselves to specific sub-domains within the plasma membrane? Addressing this question will certainly prove to be difficult and will require careful elucidation of all the cellular components involved. In neurons the targeting of ion channels has been 41  demonstrated to be highly specific (135). For example, Kv4.x channels have been shown to selectively traffic to the distal dendritic regions of the neuron, while Kv1 channels have been shown to localize at juxtaparanodal regions (135). Undoubtedly, the trafficking of vesicles carrying ion channels to the plasma membrane will be found to involve a number of cytoskeletal players. It has been demonstrated that disruption of the actin cytoskeleton results in significant changes in potassium channel expression in cardiac myocytes. Specifically, both Kv1.5 and Kv4.2 have shown significant increases in surface expression when the actin cystoskeleton was disrupted in both heterologous cells and cardiac myocytes (51, 279). Dyneins and Kinesins are two motor protein types which transport cargo along microtubules throughout the cell. Dynein moves cargo towards the minus-end of the microtubule (towards the center of the cell) while kinesins move cargo towards the positive-end of microtubules (towards the periphery of the cell) (127). Several studies have examined the roles of these proteins in the Golgi to membrane trafficking of ion channels. Kif17 a neuron-specific kinesin isoform has been shown to interact with Kv4.2 in both brain lysates and dissociated cortical neurons. The expression of a dominant negative Kif17 mutant was shown to dramatically reduce the localization of both transfected and endogenous Kv4.2 in the dendrites of rat cortical neurons (43). The blockage of other kinesin isoforms in these same neurons did not disrupt Kv4.2 localization, suggesting that Kv4.2 is trafficked specifically by Kif17. Interestingly, the deletion of a dileucine motif in the COOH-terminus of Kv4.2, previously demonstrated to be important in the targeting of the channel in neurons, was shown not to be necessary in its interaction with Kif17. However, the removal of the dileucine motif did result in disrupting the specific dendritic localization of the channel (43).  42  Other potassium channels have also been shown to interact with Kif protein isoforms. Several Kv1 isoforms interact with Kif5b in neurons and regulate their axonal targeting. One particular study has shown that Kv1.5 surface expression requires Kif5b function in both HEK293 cells and H9c2 cardiomyoblasts (307). Co-expression with wildtype Kif5b was shown to significantly enhance anterograde trafficking as demonstrated through an increase in Kv1.5 current density. The dominant negative form of Kif5b completely blocked the surface expression of Kv1.5 when channel expression was induced subsequent to Kif5b dominant negative expression. 1.3.6 Integration, distribution and compartmentalization at the cell surface Upon their arrival at the plasma membrane ion channels must be inserted into the membrane and, as necessary, undergo specific compartmentalization therein. Presumably, once vesicles containing ion channels reach the plasma membrane, the cell employs common SNARE fusion proteins including synaptosomal-associated protein 25 (SNAP-25) and Syntaxin 1A (Syn1A) in order to incorporate channels within the plasma membrane (116). The interaction of these proteins with ion channels has only been limitedly explored. Two potassium channels that have been examined in considerable detail include Kv1.1 and Kv2.1. Briefly, it has been demonstrated that Syn-1A causes inhibition of both Kv1.1 and Kv2.1 channel activity (75, 142). In the case of Kv1.1, Syn-1A is believed to interact with the channels NH2-terminus causing an increase in inactivation (75). Kv2.1 on the other hand interacts with Syn-1A through its COOH-terminus and results in a decrease in overall current amplitude and an increased voltage sensitivity of steady-state inactivation (142). SNAP-25 has also been shown to interact with Kv2.1, specifically in the NH2-terminus, and also causes an inhibition in current amplitude (163). The changes in channel kinetic activity due to the interaction with these specific membrane fusion  43  proteins is interesting and leads one to speculate that these proteins may play an important modulatory role in both membrane integration as well as overall channel functioning. Other channels shown to be modulated by SNARE proteins include Kv4.2 (294), KATP (207), and several voltage-dependent calcium channels (37, 301). With regards to Kv4.2 channels, Syn-1A interacts with the NH2-terminus of the channel and effectively inhibits the binding of KChIP2 to the channel’s NH2-terminus (294). This has been shown to result in an overall decrease in the amount of Ito current, which is in accordance with the role of KChIP2 in enhancing Kv4.2 trafficking to the plasma membrane. In pancreatic β-cells KATP (Kir6.2) plays an important role in coupling cell metabolism with the electrical excitability of these cells. In low glucose conditions adenosine triphospate/adenosine diphosphate (ATP/ADP) ratios are low, which allow KATP channels to remain open, thus preventing β-cell depolarization (98, 147). In high glucose situations this action is reversed, allowing voltage-dependent calcium channels to open, which allows an influx of calcium and exocytosis of insulin. Syn-1A has been shown to modulate Kir6.2 through interaction with the accessory subunit SUR1 (207). Similar to previous examples Syn-1 has an inhibitory role in Kir6.2 expression. As such, in cases of diabetic rats and humans where Syn-1 has been shown to be down-regulated, the corresponding elevated levels of Kir6.2 and enhanced β-cell hyperpolarization can reduce glucose-stimulated insulin secretion (98). Finally, in examples involving VDCC, Syn-1A has been shown to consistently interact through a common conserved cytoplasmic domain linking the transmembrane repeats II and III (37). Both SNAP-25 and Syn-1A play important roles in the coupling of calcium influx through the VDCCs and subsequent vesicle exocytosis (37, 302). Important to any discussion involving the distribution of ion channels within the plasma membrane is the topic of lipid rafts. Rafts exhibit a composition of molecules which make it 44  unique among the surrounding phospholipids of the membrane bi-layer (151). Typically, lipid rafts are comprised of two main components: sphingolipids and cholesterol (151). Lipid rafts exhibit unique properties in comparison to the surrounding phospholipid environment. Firstly, they are resistant to detergent solubilization and secondly, they exhibit different buoyant density when compared with detergent soluble membrane fractions. Based on these characteristics, lipid rafts can be isolated through biochemical methods which separate detergent-soluble from detergent-resistant membrane (DRM) fractions. The debate however continues as to whether rafts are in fact integral structural components of the plasma membrane in vivo. A great deal of research has attempted to decipher the different types of rafts that may exist. Membrane fractions containing the caveolins exhibit many of the same biochemical properties seen in lipid rafts, but are believed to be a separate subpopulation of rafts. Therefore, rafts are now generally characterized as caveolae and non-caveolae lipid rafts and although biochemically these membrane fractions exhibit undistinguishable membrane properties, they differ in their stability, shape and protein content (151, 204). The role of rafts in regulating ion channel distribution and function has been well documented. The first report of an ion channel contained within lipid rafts involved the shaker-like K+ channel (171). Since then several other potassium channels, including, Kv1.3, Kv1.5, Kv2.1, BKca, and Kir3.1 have all been shown to localize at least in part to rafts (29, 54, 95, 172, 201, 239). In most of these studies, association of channels with lipid rafts has been shown to modulate their function. In the case of Kv1.5, caveolin was attributed to trafficking and localization channels to cholesterol-rich membrane microdomains and to the induction of depolarizing shifts in both the steady-state activation and inactivation properties (172). Kv2.1 has been shown to be associated with rafts in both transfected fibroblasts and rat brain. In experiments involving cholesterol depletion and subsequent raft disruption, Kv2.1 channels were shown to exhibit hyperpolarizing shifts in their steady-state activation properties 45  (239). This hyperpolarizing shift is not seen in all channels examined, in the case of Kv4.2, no modulation of channel function was seen in the same cholesterol depleted cell type (201). Other ion channels outside of Kv family have also been associated with rafts, including Nav, Cav and HCN channels, but, only to a limited extent. 1.4 ER ASSOCIATED DEGRADATION The orchestrated balance between protein synthesis and degradation requires exquisite regulation to ensure cell viability. In the case of ion channels, these processes are important in regulating the availability of channels at the cell surface and, therefore, controling cellular excitability. To date, there have been only a limited number of studies focused on addressing the mechanisms which police these processes. Most often the targeting of ion channels for degradation occurs through mechanisms involving ubiquitination of lysine residues within the intracellular COOH-terminus. Ubiquitination of channels has been shown to regulate levels of internationalization and recycling back to the cell surface (107, 119). In addition, channels that are misfolded or incorrectly assembled at the level of the ER also undergo ubiquitination, which results in ER associated degradation (ERAD). As previously discussed the translocation of unfolded proteins into the ER serves as the entry point and marks the beginning of the secretory pathway (58). In order for ion channels to continue along their path to the cell surface, they must undergo proper folding and attain native confirmation. This process is facilitated by ER resident chaperone proteins found in the lumen of the ER (33, 161, 244). In the case of multimeric protein complexes, the proper assembly and stoichiometry of subunits is often crucial for forward trafficking to the Golgi. In situations where proteins do not attain their native structure or are misfolded, quality-control mechanisms have evolved in mammalian cells to tag these proteins and target them for premature degradation. The majority of this quality control occurs at the level of the ER in mammalian cells. Proteins 46  destined for degradation are retranslocated out of the ER in order to undergo ERAD, a process that is regulated through the UPS (174). Ubiquitin is a 76-amino acid protein which is ubiquitously expressed in mammalian cells; the covalent attachment of lysine residues within ubiquitin to proteins destined for degradation occurs post-translationally and involves a series of successive enzymatic reactions (233). The first step involves ubiquitin activating enzyme E1, followed by the ubiquitin-carrier enzyme E2 and finally ubiquitin-protein ligase E3. The extent of protein ubiquitination (i.e. mono, multi or polyubiquitination), and the lysine residues involved has been demonstrated to contribute to the overall fate of the protein. Interestingly, ubiquitination is one of the few post-translational modifications which can occur reversibly through the deubiquitinating enzymes (233). There have been only a limited number of studies examining the regulation of Kv channels via ERAD. One study examined the hERG channel and demonstrated that many of the 200 mutations identified in the gene which encoding for hERG and have been linked to congenital long-QT syndrome, generate misfolded proteins which are targeted for degradation through ERAD (86). Another study examining the KCNA1 gene, which codes for Kv1.1, demonstrated that a nonsense mutation resulting in a COOH-terminally truncated channel exhibited characteristics suggestive of a misfolded channel protein, including intracellular retention, detergent insolubility and rescue at lower temperatures (166). Furthermore, aggregated channels were found to interact with ubiquitin, suggesting that these misfolded subunits also undergo ERAD (166). 1.5 OVERVIEW OF N-LINKED GLYCOSYLATION The addition of sugar moieties through covalent attachment has evolved into the most common post-translational modification of proteins in eukaryotic cells (104). N-linked glycosylation is the most common such modification. In fact, the majority of proteins 47  synthesized in the ER undergo N-linked glycosylation (104). The process begins with the addition of an oligosaccharide sugar moiety to an Asn residue within a specific sequence context (Asn-X-Ser/Thr, where X represents any amino acid except proline) (128). This N-linked glycan is added en bloc as a presynthesized oligosaccharide which is widely conserved among eukaryotes. After several rounds of addition and trimming the final branched core glycan consists of three glucoses, nine mannoses and two N-acetyl-glucosamines (136). The process occurs co-translationally and plays an important role in ensuring the success of the translation and folding processes (40, 109). Upon completion of translation and assembly, the glycoprotein is shuttled to the cis-golgi where the N-glycan will undergo further trimming and branching as it progresses to the trans-golgi. It is during this phase that the glycans will develop in a diverse and complex manner into mature oligosaccharides. From the trans-golgi the mature glycoprotein will proceed to its final destination. In the case of ion channels this is most often the plasma membrane. The role of N-glycosylation in regulating the trafficking and function of ion channels has been widely studied, particularly within the Kv channel superfamily. 1.5.1 Regulation of ion channel trafficking and function through N-glycosylation In addition to its already stated role in promoting protein folding during translation in the ER, N-glycosylation has been shown to alter the function and/or trafficking of many ion channels. One of the first reported studies examined the role of glycosylation in the regulation of sodium channel function in neuroblastoma cells (275). In this study incubation of cells with tunicamycin (an inhibitor of N-glycosylation) resulted in an enhancement of channel turnover and overall decrease in channel surface expression. The role of N-glycosylation has since been extensively examined within the Kv channel family. In most cases it has been shown that N-glycosylation promotes proper folding, enhances overall surface trafficking and alters channel function (129, 211, 235, 283). Interestingly, the 48  location of N-glycosylation differs between the different channel subtypes. In the case of the Kv1.x channels, with the exception of Kv1.6, N-glycosylation occurs within the extracellular linker between the S1 and S2 transmembrane helices (88, 282, 313). Interruption of Nglycosylation either through pharmacological inhibition or mutation of the appropriate asparagine residues results in a shift in the voltage dependence of activation to more positive potentials. This is thought to be related to the influence of the cloud of negatively charged sialic acid residues on the movement of the positively charged S4 transmembrane segment (247, 272). Kv2.x channels also have a consensus N-glycosylation sequence between S1 and S2; however, these channels do not undergo N-glycosylation (254). Two separate studies have revealed that Kv3.x channels do undergo N-glycosylation and the location is again between the S1/S2 extracellular linker (31, 36). A less well-studied family is the EAG-type Kv channels. Unlike the aforementioned subtypes of Kv channels, these channels are N-glycosylated within the pore region of the channel, between the S5 and S6 transmembrane helices. Studies examining human Erg (Kv11.1 or hERG) showed that disruption of N- glycosylation resulted in a decrease in cell surface trafficking (85, 211). Additionally, several mutations within hERG which prevent channel glycosylation have been associated with Long QT syndrome (6). In another channel belonging to the ether-à-go-go (EAG) -type family, Kv10.1, it was revealed that, of the two glycosylation sites found within the pore forming domain, only one was required to undergo N-glycosylation to ensure channel trafficking (190). Similarly, Kv12.2 contains three N-glycosylation consensus sequence sites within the S5-P-loop segment, but glycosylation of only one of the three sites is necessary for proper trafficking to the cell surface (198), brining into question the role of the multiple sites for N-glycosylation in these channels? It has been shown in several Kv, as well as Nav channels, that the extent of channel glycosylation plays a direct role in modulating the 49  voltage dependent properties of these channels. Negatively charged sialic acid residues, which constitute a major component of N-linked glycans, may influence the local environment and electric field which is sensed by the voltage sensing domain, primarily the positively charged S4 transmembrane helix. 1.5.2 Glycosylation of ion channels and disease The aberrant regulation of glycosylation in humans has been found to account for approximately thirty known diseases. Congenital disorders of glycosylation (CDGs) form a growing class of diseases associated with a decrease in glycoprotein sialylation and to date afflict nearly 20 million people worldwide (77). In many CDGs the primary target organ is the heart, resulting in increased susceptibility to conduction defects, arrhythmias, and heart failure (77). As such, detailed knowledge of the mechanisms through which glycosylation regulates ion channel trafficking and function is important for better understanding the association of CDGs with various cardiac disorders. Several studies in fact have shown a direct association between the glycosylation state of ion channels and the manifestation of disease (6, 228). These examine the glycosylation state of the hERG channel and its role in long-QT syndrome. Increased efforts will need to be devoted to furthering our understanding of how deglycosylation of ion channels may contribute to various disease states. 1.6 OVERVIEW OF PALMITOYLATION The modification of proteins through the addition of fatty acids can occur via a number of different mechanisms. One such modification, protein palmitoylation, involves the addition of the fatty acid palmitate to cysteine residues (39, 59). Protein palmitoylation most often occurs via a reversible thioester linkage between the palmitate and cysteine residue and therefore is often referred to as S-palmitoylation. The less frequently observed form of palmitoylation 50  involves the attachment of palmitate to NH2-terminal cysteine residues, which then undergo rearrangement to form an irreversible amide linkage (N-palmitoylation) (149, 274). For the purposes of studies included in this thesis only S-palmitoylation will be discussed further. In general the palmitoylation of proteins serves a wide variety of roles. Firstly, the addition of a hydrophobic fatty acid such as palmitate to a protein increases its overall hypdrophobicity and contributes towards facilitating membrane interaction. In the case of otherwise cystosolic proteins, palmitoylation can result in their tethering to intracellular membranes (149). Secondly, the reversible nature of this modification suggests that palmitoylation cycling in proteins may play a key role in the targeting of proteins to different sub-cellular compartments (115, 150, 169, 218, 296). This can also include segregation to specific microdomains within membranes, such as lipid rafts. Thirdly, through the facilitation of membrane interaction, palmitoylation may result in significant consequences to protein function and behavior (72, 113, 266). Due to its inherent reversible nature, the process of palmitoylation needs to be tightly regulated through the action of on/off enzymes (270). Some proteins that are palmitoylated may remain in this state for the duration of their lifetime, while the palmitoylation status of other proteins may be heavily regulated and the protein may undergo a series of palmitoylation and depalmitoylation cycles during the course of its lifetime (59). Studies involving the Gα subunits have shown that the depolmitoylation of these proteins is regulated through the activation of G-protein-coupled receptors (218). Protein acyltransferases (PATs), the enzymes responsible for the attachment of palmitate to cysteine residues and acylprotein thioesterases (APTs), the enzymes responsible for the depalmitoylation of proteins, together are responsible for the cycling of protein palmitoylation within the cell (82). At least twenty-three mammalian genes likely to encode for PATs have been identified and each has been determined to contain a consensus DHHC motif, required for PAT activity both in vitro 51  and in vivo. In the case of APTs, only two isoforms have been identified. The first, APT1, contributes to the cycling status of palmitoylated proteins. The second form APT is found within lysosomes and is responsible for the removal of fatty acids from proteins prior to degradation. This APT is also often referred to as protein palmitoylthioesterase-1 (PPT-1) (81, 270). Overall protein palmitoylation is a dynamic regulator of protein sorting, localization and function and, as such, further understanding of the mechanisms which govern protein palmitoylation is essential for our greater understanding of the molecular processes regulating proteins within a cell. 1.6.1 Regulation of ion channels through palmitoylation Although, numerous cases of integral and transmembrane proteins have been shown to undergo palmitoylation to varying degrees, ion channels remain a class of proteins which have only been limitedly explored. The sodium channel alpha subunit was the first ion channel demonstrated to serve as a substrate for palmitoylation (241). Since then several potassium channels have also been shown to be palmitoylated in heterologous expression systems including Kv1.1 and Kv1.5 (89, 124, 311). In the case of Kv1.1 palmitoylation was shown to occur within an intracellular domain of the channel, specifically the intracellular loop connecting the second and third transmembrane helices. This study demonstrated that palmitoylation of these channels results in a depolarizing shift in the voltage dependence of activation (89). In the case of Kv1.5 palmitoylation was shown to not modulate any voltage dependent process, but rather played a significant role in the regulation of channel surface expression (124). These studies represent the only examples of Kv channels undergoing palmitoylation in heterologous expression systems. Other ion channels outside of the Kv superfamily have also been shown to undergo palmitoylation including the NMDA and AMPA receptors. In both of these examples palmitoylation has been shown to modulate synaptic targeting in vivo (99, 100). In addition to modulation of ion channel function through direct palmitoylation of alpha pore forming subunits, 52  palmitoylation of accessory subunits have also been shown to modulate ion channel activity. There are several studies demonstrating this form of channel regulation. Dual palmitoylation of two NH2-terminal cysteine residues in the Cav beta subunit β2a results in a dramatic depolarizing shift in the voltage dependence of channel inactivation (113). Specific KChIP isoforms have also been shown to undergo palmitoylation within their NH2-terminus. Through mutational analysis it was demonstrated that palmitoylation of these intracellular cysteine residues plays a critical role in the ability of KChIPs to enhance surface expression of Kv4.3 (263). Finally, palmitoylation of scaffolding proteins such as PSD-95 have been shown to play a role in the interaction with Kv1.4 channels at the plasma membrane (266). Overall, the number of studies examining palmitoylation as a regulator of ion channel behavior is limited and will need to be further explored. 1.6.2 Identification of palmitoylated proteins One major difficulty and roadblock in the identification of palmitoylated proteins has been the lack of a clear palmitoylation consensus sequence. Palmitoylation predictor programs have been designed and are based on computer algorithms which utilize the protein sequences of previously characterized palmitoylated proteins to provide statical scores of potentially palmitoylated cysteine residues of uncharacterized proteins (225, 293). Although these prediction programs have contributed to the identification of candidate cysteine residues in any particular protein, they remain limited in their ability to clearly identify which proteins are capable of undergoing palmitoylation. Another primary issue with the identification and characterization of palmitoylated proteins is the lack of sensitive and quantitative assays. Classically, the most widely used assay for protein palmitoylation has involved the incubation cultured cells with radiolabeled palmitate. After which labeled proteins can be purified and run on an SDS-PAGE for analysis. Most often 3H-palmitate is used as the radiolabeled form of 53  palmitate. There are problems with this mode of detection including, lengthy incubation periods and the low expression of many proteins, which often make their detection through radiolabeled palmitate difficult. Recently newly developed approaches have greatly advanced the means by which palmitoylated protein both in vivo and in vitro can be identified. Fatty acyl exchange chemistry uses an alternative strategy where the fatty acid group at the site of palmitoylation is replaced with a more readily detectable reagent (69, 70). This technology has recently been employed in the large scale identification of palmitoylated proteins in yeast and the rat brain (125, 231). In eliminating the use of radioactivity this method avoids the prolonged exposure time’s necessary with the use of radioactive palmitate without sacrificing assay sensitivity and specificity.  54  1.7 SCOPE OF THE THESIS This thesis examines the structural determinants which contribute to the regulation of cell surface trafficking, expression and function in tetrameric ion channels. Specifically, the experimental work included in this thesis revolves around ion channel isoforms belonging to the HCN and Kv4 families. Both of these channel families belong to the superfamily of voltagegated potassium channels that are widely expressed throughout the cardiovascular and central nervous systems. In the heart, HCN channels produce an inward positively charged current (If) which plays an important role in regulating the diastolic depolarizing phase of the action potential in both SA node and ventricular myocytes. Kv4 channels in the heart are responsible for the transient outward directed currents (Ito) which play an important role during the repolarization phase of the action potential in cardiomyocytes. In the CNS both HCN and Kv4 channels are instrumental in the regulation of neuronal excitability and result in the generation of the currents Ih and IA respectively. Specifically, the work presented in this thesis provides novel insight into the structural determinants and post-translational modifications which contribute to the regulation of the HCN1, HCN2 and Kv4.2 ion channels. In chapter 2 of this thesis we describe a specific four amino acid motif (EEYP) within the CNBD of the HCN2 channel that strongly promotes channel export from the ER and surface expression but does not contribute to the inhibition of channel opening in these channels. Previous studies have demonstrated that removal of the CNBD results in a complete abolishment of channel surface expression. Our findings here provide novel insight into specific amino acids within the CNBD which are responsible for this regulation of channel surface expression conferred by the CNBD. Chapter 3 of this thesis examines the role of N-linked glycosylation within different HCN isoforms. All four mammalian HCN isoforms (HCN1-4) contain the consensus N-linked 55  glycosylation sequon in their S5-S6 extracellular linker near the channel pore region. Previously, it has been demonstrated that all four HCN isoforms undergo N-glycosylation in the brain. It has also been shown that elimination of N-glycosylation in HCN2 eliminates functional channel expression at the plasma membrane. In our study we demonstrate for the first time that Nglycosylation is not required for HCN2 function, but that it is nevertheless tightly coupled to HCN2 cell surface expression. The functional role of N-glycosylation was also examined for the first time in the HCN1 isoform. We demonstrate that HCN1 is much less dependent on Nglycosylation for its functional expression at the cell surface. In Chapter 4 we examine the role of palmitoylation in the regulation of Kv4 channels. We demonstrate that Kv4.2 serves as a substrate for palmitoylation in both the rat brain and when transiently transfected in COS-7 cells. We demonstrate for the first time the palmitoylation of a Kv channel outside of the Kv1 channel family. Additionally, this study provides the first piece of direct evidence for the presence of a palmitoylated Kv channel in vivo.  56  1.8 REFERENCES  1. Abbott GW, Ramesh B and Srai SK. Secondary structure of the MiRP1 (KCNE2) potassium channel ancillary subunit. Protein Pept Lett 15: 63-75, 2008. 2. Accili EA, Robinson RB and DiFrancesco D. Properties and modulation of If in newborn versus adult cardiac SA node. The American journal of physiology 272: H1549-52, 1997. 3. Akhavan A, Atanasiu R, Noguchi T, Han W, Holder N and Shrier A. Identification of the cyclic-nucleotide-binding domain as a conserved determinant of ion-channel cell-surface localization. J Cell Sci 118: 2803-2812, 2005. 4. Alig J, Marger L, Mesirca P, Ehmke H, Mangoni ME and Isbrandt D. Control of heart rate by cAMP sensitivity of HCN channels. Proc Natl Acad Sci U S A 106: 12189-12194, 2009. 5. An WF, Bowlby MR, Betty M, Cao J, Ling HP, Mendoza G, Hinson JW, Mattsson KI, Strassle BW, Trimmer JS and Rhodes KJ. Modulation of A-type potassium channels by a family of calcium sensors. Nature 403: 553-556, 2000. 6. Anderson CL, Delisle BP, Anson BD, Kilby JA, Will ML, Tester DJ, Gong Q, Zhou Z, Ackerman MJ and January CT. Most LQT2 mutations reduce Kv11.1 (hERG) current by a class 2 (trafficking-deficient) mechanism. Circulation 113: 365-373, 2006. 7. Anderson D, Mehaffey WH, Iftinca M, Rehak R, Engbers JD, Hameed S, Zamponi GW and Turner RW. Regulation of neuronal activity by Cav3-Kv4 channel signaling complexes. Nat Neurosci 13: 333-337, 2010. 8. Angelova PR and Muller WS. Arachidonic acid potently inhibits both postsynaptic-type Kv4.2 and presynaptic-type Kv1.4 IA potassium channels. Eur J Neurosci 29: 1943-1950, 2009. 9. Aoe T, Lee AJ, van Donselaar E, Peters PJ and Hsu VW. Modulation of intracellular transport by transported proteins: insight from regulation of COPI-mediated transport. Proc Natl Acad Sci U S A 95: 1624-1629, 1998. 10. Apkon M and Nerbonne JM. Alpha 1-adrenergic agonists selectively suppress voltagedependent K+ current in rat ventricular myocytes. Proc Natl Acad Sci U S A 85: 8756-8760, 1988. 11. Aridor M, Bannykh SI, Rowe T and Balch WE. Sequential coupling between COPII and COPI vesicle coats in endoplasmic reticulum to Golgi transport. J Cell Biol 131: 875-893, 1995. 12. Bahring R, Boland LM, Varghese A, Gebauer M and Pongs O. Kinetic analysis of open- and closed-state inactivation transitions in human Kv4.2 A-type potassium channels. J Physiol 535: 65-81, 2001. 57  13. Bahring R, Dannenberg J, Peters HC, Leicher T, Pongs O and Isbrandt D. Conserved Kv4 N-terminal domain critical for effects of Kv channel-interacting protein 2.2 on channel expression and gating. J Biol Chem 276: 23888-23894, 2001. 14. Bal T and McCormick DA. What stops synchronized thalamocortical oscillations? Neuron 17: 297-308, 1996. 15. Bal T, von Krosigk M and McCormick DA. Synaptic and membrane mechanisms underlying synchronized oscillations in the ferret lateral geniculate nucleus in vitro. J Physiol 483 ( Pt 3): 641-663, 1995. 16. Barnwell LF, Lugo JN, Lee WL, Willis SE, Gertz SJ, Hrachovy RA and Anderson AE. Kv4.2 knockout mice demonstrate increased susceptibility to convulsant stimulation. Epilepsia 2009. 17. Barry DM, Xu H, Schuessler RB and Nerbonne JM. Functional knockout of the transient outward current, long-QT syndrome, and cardiac remodeling in mice expressing a dominant-negative Kv4 alpha subunit. Circ Res 83: 560-567, 1998. 18. Baukrowitz T and Yellen G. Modulation of K+ current by frequency and external [K+]: a tale of two inactivation mechanisms. Neuron 15: 951-960, 1995. 19. Beck H, Clusmann H, Kral T, Schramm J, Heinemann U and Elger CE. Potassium currents in acutely isolated human hippocampal dentate granule cells. J Physiol 498 ( Pt 1): 7385, 1997. 20. Bender RA and Baram TZ. Hyperpolarization activated cyclic-nucleotide gated (HCN) channels in developing neuronal networks. Prog Neurobiol 86: 129-140, 2008. 21. Bertaso F, Sharpe CC, Hendry BM and James AF. Expression of voltage-gated K+ channels in human atrium. Basic Res Cardiol 97: 424-433, 2002. 22. Bichet D, Cornet V, Geib S, Carlier E, Volsen S, Hoshi T, Mori Y and De Waard M. The I-II loop of the Ca2+ channel alpha1 subunit contains an endoplasmic reticulum retention signal antagonized by the beta subunit. Neuron 25: 177-190, 2000. 23. Birnbaum SG, Varga AW, Yuan LL, Anderson AE, Sweatt JD and Schrader LA. Structure and function of Kv4-family transient potassium channels. Physiological reviews 84: 803-33, 2004. 24. Blunck R, McGuire H, Hyde HC and Bezanilla F. Fluorescence detection of the movement of single KcsA subunits reveals cooperativity. Proc Natl Acad Sci U S A 105: 2026320268, 2008. 25. Bois P, Renaudon B, Baruscotti M, Lenfant J and DiFrancesco D. Activation of fchannels by cAMP analogues in macropatches from rabbit sino-atrial node myocytes. J Physiol 501 ( Pt 3): 565-571, 1997. 58  26. Bosch RF, Scherer CR, Rub N, Wohrl S, Steinmeyer K, Haase H, Busch AE, Seipel L and Kuhlkamp V. Molecular mechanisms of early electrical remodeling: transcriptional downregulation of ion channel subunits reduces I(Ca,L) and I(to) in rapid atrial pacing in rabbits. J Am Coll Cardiol 41: 858-869, 2003. 27. Bou-Abboud E and Nerbonne JM. Molecular correlates of the calcium-independent, depolarization-activated K+ currents in rat atrial myocytes. J Physiol 517 ( Pt 2): 407-420, 1999. 28. Brandt MC, Endres-Becker J, Zagidullin N, Motloch LJ, Er F, Rottlaender D, Michels G, Herzig S and Hoppe UC. Effects of KCNE2 on HCN isoforms: distinct modulation of membrane expression and single channel properties. Am J Physiol Heart Circ Physiol 297: H355-63, 2009. 29. Bravo-Zehnder M, Orio P, Norambuena A, Wallner M, Meera P, Toro L, Latorre R and Gonzalez A. Apical sorting of a voltage- and Ca2+-activated K+ channel alpha -subunit in Madin-Darby canine kidney cells is independent of N-glycosylation. Proc Natl Acad Sci U S A 97: 13114-13119, 2000. 30. Brewster AL, Chen Y, Bender RA, Yeh A, Shigemoto R and Baram TZ. Quantitative analysis and subcellular distribution of mRNA and protein expression of the hyperpolarizationactivated cyclic nucleotide-gated channels throughout development in rat hippocampus. Cereb Cortex 17: 702-712, 2007. 31. Brooks NL, Corey MJ and Schwalbe RA. Characterization of N-glycosylation consensus sequences in the Kv3.1 channel. The FEBS journal 273: 3287-300, 2006. 32. Brown HF, DiFrancesco D and Noble SJ. How does adrenaline accelerate the heart? Nature 280: 235-236, 1979. 33. Buchner J. Supervising the fold: functional principles of molecular chaperones. FASEB J 10: 10-19, 1996. 34. Cabrera M, Muniz M, Hidalgo J, Vega L, Martin ME and Velasco A. The retrieval function of the KDEL receptor requires PKA phosphorylation of its C-terminus. Mol Biol Cell 14: 4114-4125, 2003. 35. Callsen B, Isbrandt D, Sauter K, Hartmann LS, Pongs O and Bahring R. Contribution of N- and C-terminal Kv4.2 channel domains to KChIP interaction [corrected]. The Journal of physiology 568: 397-412, 2005. 36. Cartwright TA, Corey MJ and Schwalbe RA. Complex oligosaccharides are N-linked to Kv3 voltage-gated K+ channels in rat brain. Biochimica et biophysica acta 1770: 666-71, 2007. 37. Catterall WA. Interactions of presynaptic Ca2+ channels and snare proteins in neurotransmitter release. Ann N Y Acad Sci 868: 144-159, 1999.  59  38. Cha A, Snyder GE, Selvin PR and Bezanilla F. Atomic scale movement of the voltagesensing region in a potassium channel measured via spectroscopy. Nature 402: 809-813, 1999. 39. Charollais J and Van Der Goot FG. Palmitoylation of membrane proteins (Review). Molecular membrane biology 26: 55-66, 2009. 40. Chavan M and Lennarz W. The molecular basis of coupling of translocation and Nglycosylation. Trends in biochemical sciences 31: 17-20, 2006. 41. Chen L, Tian L, MacDonald SH, McClafferty H, Hammond MS, Huibant JM, Ruth P, Knaus HG and Shipston MJ. Functionally diverse complement of large conductance calcium- and voltage-activated potassium channel (BK) alpha-subunits generated from a single site of splicing. J Biol Chem 280: 33599-33609, 2005. 42. Choi KL, Aldrich RW and Yellen G. Tetraethylammonium blockade distinguishes two inactivation mechanisms in voltage-activated K+ channels. Proc Natl Acad Sci U S A 88: 50925095, 1991. 43. Chu PJ, Rivera JF and Arnold DB. A role for Kif17 in transport of Kv4.2. J Biol Chem 281: 365-373, 2006. 44. Chung WK, Shin M, Jaramillo TC, Leibel RL, LeDuc CA, Fischer SG, Tzilianos E, Gheith AA, Lewis AS and Chetkovich DM. Absence epilepsy in apathetic, a spontaneous mutant mouse lacking the h channel subunit, HCN2. Neurobiol Dis 33: 499-508, 2009. 45. Clark BD, Kwon E, Maffie J, Jeong HY, Nadal M, Strop P and Rudy B. DPP6 Localization in Brain Supports Function as a Kv4 Channel Associated Protein. Front Mol Neurosci 1: 8, 2008. 46. Coetzee WA, Amarillo Y, Chiu J, Chow A, Lau D, McCormack T, Moreno H, Nadal MS, Ozaita A, Pountney D, Saganich M, Vega-Saenz de Miera E and Rudy B. Molecular diversity of K+ channels. Ann N Y Acad Sci 868: 233-285, 1999. 47. Cook PR. A model for all genomes: the role of transcription factories. J Mol Biol 395: 110, 2010. 48. Craven KB, Olivier NB and Zagotta WN. C-terminal movement during gating in cyclic nucleotide-modulated channels. J Biol Chem 283: 14728-14738, 2008. 49. Craven KB and Zagotta WN. CNG and HCN channels: two peas, one pod. Annu Rev Physiol 68: 375-401, 2006. 50. Craven KB and Zagotta WN. Salt bridges and gating in the COOH-terminal region of HCN2 and CNGA1 channels. J Gen Physiol 124: 663-677, 2004. 51. Cukovic D, Lu GW, Wible B, Steele DF and Fedida D. A discrete amino terminal domain of Kv1.5 and Kv1.4 potassium channels interacts with the spectrin repeats of alphaactinin-2. FEBS Lett 498: 87-92, 2001. 60  52. Dascher C, Matteson J and Balch WE. Syntaxin 5 regulates endoplasmic reticulum to Golgi transport. J Biol Chem 269: 29363-29366, 1994. 53. Day M, Carr DB, Ulrich S, Ilijic E, Tkatch T and Surmeier DJ. Dendritic excitability of mouse frontal cortex pyramidal neurons is shaped by the interaction among HCN, Kir2, and Kleak channels. J Neurosci 25: 8776-8787, 2005. 54. Delling M, Wischmeyer E, Dityatev A, Sytnyk V, Veh RW, Karschin A and Schachner M. The neural cell adhesion molecule regulates cell-surface delivery of G-proteinactivated inwardly rectifying potassium channels via lipid rafts. J Neurosci 22: 7154-7164, 2002. 55. Deschenes I, DiSilvestre D, Juang GJ, Wu RC, An WF and Tomaselli GF. Regulation of Kv4.3 current by KChIP2 splice variants: a component of native cardiac I(to)? Circulation 106: 423-429, 2002. 56. Deschenes I, DiSilvestre D, Juang GJ, Wu RC, An WF and Tomaselli GF. Regulation of Kv4.3 current by KChIP2 splice variants: a component of native cardiac I(to)? Circulation 106: 423-429, 2002. 57. Deschenes I and Tomaselli GF. Modulation of Kv4.3 current by accessory subunits. FEBS Lett 528: 183-188, 2002. 58.  Deutsch C. The birth of a channel. Neuron 40: 265-276, 2003.  59. Dietrich LE and Ungermann C. On the mechanism of protein palmitoylation. EMBO reports 5: 1053-7, 2004. 60. DiFrancesco D. The role of the funny current in pacemaker activity. Circ Res 106: 434446, 2010. 61. DiFrancesco D. Characterization of single pacemaker channels in cardiac sino-atrial node cells. Nature 324: 470-473, 1986. 62. DiFrancesco D, Ducouret P and Robinson RB. Muscarinic modulation of cardiac rate at low acetylcholine concentrations. Science 243: 669-671, 1989. 63. DiFrancesco D, Ducouret P and Robinson RB. Muscarinic modulation of cardiac rate at low acetylcholine concentrations. Science 243: 669-671, 1989. 64. DiFrancesco D, Ferroni A, Mazzanti M and Tromba C. Properties of the hyperpolarizing-activated current (if) in cells isolated from the rabbit sino-atrial node. The Journal of physiology 377: 61-88, 1986. 65. DiFrancesco D and Ojeda C. Properties of the current if in the sino-atrial node of the rabbit compared with those of the current iK, in Purkinje fibres. J Physiol 308: 353-367, 1980. 66. DiFrancesco D and Tortora P. Direct activation of cardiac pacemaker channels by intracellular cyclic AMP. Nature 351: 145-7, 1991. 61  67. Dixon JE, Shi W, Wang HS, McDonald C, Yu H, Wymore RS, Cohen IS and McKinnon D. Role of the Kv4.3 K+ channel in ventricular muscle. A molecular correlate for the transient outward current. Circ Res 79: 659-668, 1996. 68. Doyle DA, Morais Cabral J, Pfuetzner RA, Kuo A, Gulbis JM, Cohen SL, Chait BT and MacKinnon R. The structure of the potassium channel: molecular basis of K+ conduction and selectivity. Science 280: 69-77, 1998. 69. Drisdel RC, Alexander JK, Sayeed A and Green WN. Assays of protein palmitoylation. Methods (San Diego, Calif 40: 127-34, 2006. 70. Drisdel RC and Green WN. Labeling and quantifying sites of protein palmitoylation. BioTechniques 36: 276-285, 2004. 71. Dun W, Chandra P, Danilo P,Jr, Rosen MR and Boyden PA. Chronic atrial fibrillation does not further decrease outward currents. It increases them. Am J Physiol Heart Circ Physiol 285: H1378-84, 2003. 72. El-Husseini AE, Craven SE, Chetkovich DM, Firestein BL, Schnell E, Aoki C and Bredt DS. Dual palmitoylation of PSD-95 mediates its vesiculotubular sorting, postsynaptic targeting, and ion channel clustering. J Cell Biol 148: 159-172, 2000. 73. England SK, Uebele VN, Shear H, Kodali J, Bennett PB and Tamkun MM. Characterization of a voltage-gated K+ channel beta subunit expressed in human heart. Proc Natl Acad Sci U S A 92: 6309-6313, 1995. 74. Ficker E, Dennis AT, Wang L and Brown AM. Role of the cytosolic chaperones Hsp70 and Hsp90 in maturation of the cardiac potassium channel HERG. Circ Res 92: e87-100, 2003. 75. Fili O, Michaelevski I, Bledi Y, Chikvashvili D, Singer-Lahat D, Boshwitz H, Linial M and Lotan I. Direct interaction of a brain voltage-gated K+ channel with syntaxin 1A: functional impact on channel gating. J Neurosci 21: 1964-1974, 2001. 76. Fiset C, Clark RB, Shimoni Y and Giles WR. Shal-type channels contribute to the Ca2+-independent transient outward K+ current in rat ventricle. J Physiol 500 ( Pt 1): 51-64, 1997. 77. Footitt EJ, Karimova A, Burch M, Yayeh T, Dupre T, Vuillaumier-Barrot S, Chantret I, Moore SE, Seta N and Grunewald S. Cardiomyopathy in the congenital disorders of glycosylation (CDG): a case of late presentation and literature review. J Inherit Metab Dis 2009. 78. Francis J, Jugloff DG, Mingo NS, Wallace MC, Jones OT, Burnham WM and Eubanks JH. Kainic acid-induced generalized seizures alter the regional hippocampal expression of the rat Kv4.2 potassium channel gene. Neurosci Lett 232: 91-94, 1997. 79. Frere SG, Kuisle M and Luthi A. Regulation of recombinant and native hyperpolarization-activated cation channels. Mol Neurobiol 30: 279-305, 2004. 62  80. Fuenzalida M, Fernandez de Sevilla D, Couve A and Buno W. Role of AMPA and NMDA receptors and back-propagating action potentials in spike timing-dependent plasticity. J Neurophysiol 103: 47-54, 2010. 81. Fukata M, Fukata Y, Adesnik H, Nicoll RA and Bredt DS. Identification of PSD-95 palmitoylating enzymes. Neuron 44: 987-996, 2004. 82. Fukata Y and Fukata M. Protein palmitoylation in neuronal development and synaptic plasticity. Nat Rev Neurosci 11: 161-175, 2010. 83. Gagnon DG and Bezanilla F. A single charged voltage sensor is capable of gating the Shaker K+ channel. J Gen Physiol 133: 467-483, 2009. 84. Gallego M, Alday A, Urrutia J and Casis O. Transient outward potassium channel regulation in healthy and diabetic hearts. Can J Physiol Pharmacol 87: 77-83, 2009. 85. Gong Q, Anderson CL, January CT and Zhou Z. Role of glycosylation in cell surface expression and stability of HERG potassium channels. American journal of physiology 283: H77-84, 2002. 86. Gong Q, Keeney DR, Molinari M and Zhou Z. Degradation of trafficking-defective long QT syndrome type II mutant channels by the ubiquitin-proteasome pathway. J Biol Chem 280: 19419-19425, 2005. 87. Griffith LC. Potassium channels: the importance of transport signals. Curr Biol 11: R2268, 2001. 88. Grupe A, Schroter KH, Ruppersberg JP, Stocker M, Drewes T, Beckh S and Pongs O. Cloning and expression of a human voltage-gated potassium channel. A novel member of the RCK potassium channel family. EMBO J 9: 1749-1756, 1990. 89. Gubitosi-Klug RA, Mancuso DJ and Gross RW. The human Kv1.1 channel is palmitoylated, modulating voltage sensing: Identification of a palmitoylation consensus sequence. Proceedings of the National Academy of Sciences of the United States of America 102: 5964-8, 2005. 90. Gulbis JM, Mann S and MacKinnon R. Structure of a voltage-dependent K+ channel beta subunit. Cell 97: 943-952, 1999. 91. Guo W, Li H, Aimond F, Johns DC, Rhodes KJ, Trimmer JS and Nerbonne JM. Role of heteromultimers in the generation of myocardial transient outward K+ currents. Circ Res 90: 586-593, 2002. 92. Guo W, Malin SA, Johns DC, Jeromin A and Nerbonne JM. Modulation of Kv4encoded K(+) currents in the mammalian myocardium by neuronal calcium sensor-1. J Biol Chem 277: 26436-26443, 2002.  63  93. Guo W, Malin SA, Johns DC, Jeromin A and Nerbonne JM. Modulation of Kv4encoded K(+) currents in the mammalian myocardium by neuronal calcium sensor-1. J Biol Chem 277: 26436-26443, 2002. 94. Gurkan C, Stagg SM, Lapointe P and Balch WE. The COPII cage: unifying principles of vesicle coat assembly. Nat Rev Mol Cell Biol 7: 727-738, 2006. 95. Hajdu P, Varga Z, Pieri C, Panyi G and Gaspar R,Jr. Cholesterol modifies the gating of Kv1.3 in human T lymphocytes. Pflugers Arch 445: 674-682, 2003. 96. Han W, Nattel S, Noguchi T and Shrier A. C-terminal domain of Kv4.2 and associated KChIP2 interactions regulate functional expression and gating of Kv4.2. The Journal of biological chemistry 281: 27134-44, 2006. 97. Harzheim D, Pfeiffer KH, Fabritz L, Kremmer E, Buch T, Waisman A, Kirchhof P, Kaupp UB and Seifert R. Cardiac pacemaker function of HCN4 channels in mice is confined to embryonic development and requires cyclic AMP. EMBO J 27: 692-703, 2008. 98. Hashimoto T, Nakamura T, Maegawa H, Nishio Y, Egawa K and Kashiwagi A. Regulation of ATP-sensitive potassium channel subunit Kir6.2 expression in rat intestinal insulin-producing progenitor cells. J Biol Chem 280: 1893-1900, 2005. 99. Hayashi T, Rumbaugh G and Huganir RL. Differential regulation of AMPA receptor subunit trafficking by palmitoylation of two distinct sites. Neuron 47: 709-23, 2005. 100. Hayashi T, Thomas GM and Huganir RL. Dual palmitoylation of NR2 subunits regulates NMDA receptor trafficking. Neuron 64: 213-26, 2009. 101. Hedley PL, Jorgensen P, Schlamowitz S, Wangari R, Moolman-Smook J, Brink PA, Kanters JK, Corfield VA and Christiansen M. The genetic basis of long QT and short QT syndromes: a mutation update. Hum Mutat 30: 1486-1511, 2009. 102. Hegle AP, Nazzari H, Roth A, Angoli D and Accili EA. Evolutionary emergence of Nglycosylation as a variable promoter of HCN channel surface expression. Am J Physiol Cell Physiol 298: C1066-76, 2010. 103. Heinemann SH, Rettig J, Wunder F and Pongs O. Molecular and functional characterization of a rat brain Kv beta 3 potassium channel subunit. FEBS Lett 377: 383-389, 1995. 104. Helenius A and Aebi M. Roles of N-linked glycans in the endoplasmic reticulum. Annual review of biochemistry 73: 1019-49, 2004. 105. Hemond P, Migliore M, Ascoli GA and Jaffe DB. The membrane response of hippocampal CA3b pyramidal neurons near rest: Heterogeneity of passive properties and the contribution of hyperpolarization-activated currents. Neuroscience 160: 359-370, 2009.  64  106. Herrmann S, Stieber J and Ludwig A. Pathophysiology of HCN channels. Pflugers Arch 454: 517-522, 2007. 107. Hicke L. Gettin' down with ubiquitin: turning off cell-surface receptors, transporters and channels. Trends Cell Biol 9: 107-112, 1999. 108. Hoffman DA and Johnston D. Downregulation of transient K+ channels in dendrites of hippocampal CA1 pyramidal neurons by activation of PKA and PKC. J Neurosci 18: 3521-3528, 1998. 109. Holst B, Bruun AW, Kielland-Brandt MC and Winther JR. Competition between folding and glycosylation in the endoplasmic reticulum. The EMBO journal 15: 3538-46, 1996. 110. Huang B, Qin D and El-Sherif N. Spatial alterations of Kv channels expression and K(+) currents in post-MI remodeled rat heart. Cardiovasc Res 52: 246-254, 2001. 111. Huang B, Qin D and El-Sherif N. Early down-regulation of K+ channel genes and currents in the postinfarction heart. J Cardiovasc Electrophysiol 11: 1252-1261, 2000. 112. Huang J, Huang A, Zhang Q, Lin YC and Yu HG. Novel mechanism for suppression of hyperpolarization-activated cyclic nucleotide-gated pacemaker channels by receptor-like tyrosine phosphatase-alpha. The Journal of biological chemistry 283: 29912-9, 2008. 113. Hurley JH, Cahill AL, Currie KP and Fox AP. The role of dynamic palmitoylation in Ca2+ channel inactivation. Proceedings of the National Academy of Sciences of the United States of America 97: 9293-8, 2000. 114. Isacoff EY, Jan YN and Jan LY. Putative receptor for the cytoplasmic inactivation gate in the Shaker K+ channel. Nature 353: 86-90, 1991. 115. Izawa I, Nishizawa M, Hayashi Y and Inagaki M. Palmitoylation of ERBIN is required for its plasma membrane localization. Genes Cells 13: 691-701, 2008. 116. Jahn R and Scheller RH. SNAREs--engines for membrane fusion. Nat Rev Mol Cell Biol 7: 631-643, 2006. 117. Jahnsen H and Llinas R. Electrophysiological properties of guinea-pig thalamic neurones: an in vitro study. J Physiol 349: 205-226, 1984. 118. Jahnsen H and Llinas R. Ionic basis for the electro-responsiveness and oscillatory properties of guinea-pig thalamic neurones in vitro. J Physiol 349: 227-247, 1984. 119. Jentsch TJ, Hubner CA and Fuhrmann JC. Ion channels: function unravelled by dysfunction. Nat Cell Biol 6: 1039-1047, 2004. 120. Jerng HH and Covarrubias M. K+ channel inactivation mediated by the concerted action of the cytoplasmic N- and C-terminal domains. Biophys J 72: 163-174, 1997. 65  121. Jerng HH, Pfaffinger PJ and Covarrubias M. Molecular physiology and modulation of somatodendritic A-type potassium channels. Molecular and cellular neurosciences 27: 343-69, 2004. 122. Jerng HH, Shahidullah M and Covarrubias M. Inactivation gating of Kv4 potassium channels: molecular interactions involving the inner vestibule of the pore. J Gen Physiol 113: 641-660, 1999. 123. Jiang M, Xu X, Wang Y, Toyoda F, Liu XS, Zhang M, Robinson RB and Tseng GN. Dynamic partnership between KCNQ1 and KCNE1 and influence on cardiac IKs current amplitude by KCNE2. J Biol Chem 284: 16452-16462, 2009. 124. Jindal HK, Folco EJ, Liu GX and Koren G. Posttranslational modification of voltagedependent potassium channel Kv1.5: COOH-terminal palmitoylation modulates its biological properties. American journal of physiology 294: H2012-21, 2008. 125. Kang R, Wan J, Arstikaitis P, Takahashi H, Huang K, Bailey AO, Thompson JX, Roth AF, Drisdel RC, Mastro R, Green WN, Yates JR,3rd, Davis NG and El-Husseini A. Neural palmitoyl-proteomics reveals dynamic synaptic palmitoylation. Nature 456: 904-9, 2008. 126. Kanyshkova T, Pawlowski M, Meuth P, Dube C, Bender RA, Brewster AL, Baumann A, Baram TZ, Pape HC and Budde T. Postnatal expression pattern of HCN channel isoforms in thalamic neurons: relationship to maturation of thalamocortical oscillations. J Neurosci 29: 8847-8857, 2009. 127. Karcher RL, Deacon SW and Gelfand VI. Motor-cargo interactions: the key to transport specificity. Trends Cell Biol 12: 21-27, 2002. 128. Kasturi L, Chen H and Shakin-Eshleman SH. Regulation of N-linked core glycosylation: use of a site-directed mutagenesis approach to identify Asn-Xaa-Ser/Thr sequons that are poor oligosaccharide acceptors. The Biochemical journal 323 ( Pt 2): 415-9, 1997. 129. Khanna R, Myers MP, Laine M and Papazian DM. Glycosylation increases potassium channel stability and surface expression in mammalian cells. The Journal of biological chemistry 276: 34028-34, 2001. 130. Kim J, Jung SC, Clemens AM, Petralia RS and Hoffman DA. Regulation of dendritic excitability by activity-dependent trafficking of the A-type K+ channel subunit Kv4.2 in hippocampal neurons. Neuron 54: 933-47, 2007. 131. Kimura K, Kitano J, Nakajima Y and Nakanishi S. Hyperpolarization-activated, cyclic nucleotide-gated HCN2 cation channel forms a protein assembly with multiple neuronal scaffold proteins in distinct modes of protein-protein interaction. Genes Cells 9: 631-640, 2004. 132. Kole MH, Brauer AU and Stuart GJ. Inherited cortical HCN1 channel loss amplifies dendritic calcium electrogenesis and burst firing in a rat absence epilepsy model. J Physiol 578: 507-525, 2007. 66  133. Kuryshev YA, Gudz TI, Brown AM and Wible BA. KChAP as a chaperone for specific K(+) channels. Am J Physiol Cell Physiol 278: C931-41, 2000. 134. Kuryshev YA, Wible BA, Gudz TI, Ramirez AN and Brown AM. KChAP/Kvbeta1.2 interactions and their effects on cardiac Kv channel expression. Am J Physiol Cell Physiol 281: C290-9, 2001. 135. Lai HC and Jan LY. The distribution and targeting of neuronal voltage-gated ion channels. Nat Rev Neurosci 7: 548-562, 2006. 136. Lau KS, Partridge EA, Grigorian A, Silvescu CI, Reinhold VN, Demetriou M and Dennis JW. Complex N-glycan number and degree of branching cooperate to regulate cell proliferation and differentiation. Cell 129: 123-34, 2007. 137. Lee MC, Miller EA, Goldberg J, Orci L and Schekman R. Bi-directional protein transport between the ER and Golgi. Annu Rev Cell Dev Biol 20: 87-123, 2004. 138. Lee SY, Lee A, Chen J and MacKinnon R. Structure of the KvAP voltage-dependent K+ channel and its dependence on the lipid membrane. Proc Natl Acad Sci U S A 102: 1544115446, 2005. 139. Leicher T, Bahring R, Isbrandt D and Pongs O. Coexpression of the KCNA3B gene product with Kv1.5 leads to a novel A-type potassium channel. J Biol Chem 273: 35095-35101, 1998. 140. Leresche N, Jassik-Gerschenfeld D, Haby M, Soltesz I and Crunelli V. Pacemakerlike and other types of spontaneous membrane potential oscillations of thalamocortical cells. Neurosci Lett 113: 72-77, 1990. 141. Leresche N, Lightowler S, Soltesz I, Jassik-Gerschenfeld D and Crunelli V. Lowfrequency oscillatory activities intrinsic to rat and cat thalamocortical cells. J Physiol 441: 155174, 1991. 142. Leung YM, Kang Y, Gao X, Xia F, Xie H, Sheu L, Tsuk S, Lotan I, Tsushima RG and Gaisano HY. Syntaxin 1A binds to the cytoplasmic C terminus of Kv2.1 to regulate channel gating and trafficking. J Biol Chem 278: 17532-17538, 2003. 143. Levitan ES and Takimoto K. Surface expression of Kv1 voltage-gated K+ channels is governed by a C-terminal motif. Trends Cardiovasc Med 10: 317-320, 2000. 144. Lewis AS, Schwartz E, Chan CS, Noam Y, Shin M, Wadman WJ, Surmeier DJ, Baram TZ, Macdonald RL and Chetkovich DM. Alternatively spliced isoforms of TRIP8b differentially control h channel trafficking and function. J Neurosci 29: 6250-65, 2009. 145. Li CH, Zhang Q, Teng B, Mustafa SJ, Huang JY and Yu HG. Src tyrosine kinase alters gating of hyperpolarization-activated HCN4 pacemaker channel through Tyr531. Am J Physiol Cell Physiol 294: C355-62, 2008. 67  146. Li D, Takimoto K and Levitan ES. Surface expression of Kv1 channels is governed by a C-terminal motif. J Biol Chem 275: 11597-11602, 2000. 147. Li L, Rojas A, Wu J and Jiang C. Disruption of glucose sensing and insulin secretion by ribozyme Kir6.2-gene targeting in insulin-secreting cells. Endocrinology 145: 4408-4414, 2004. 148. Lin L, Sun W, Wikenheiser AM, Kung F and Hoffman DA. KChIP4a regulates Kv4.2 channel trafficking through PKA phosphorylation. Mol Cell Neurosci 43: 315-325, 2010. 149. Linder M and Deschenes R. Protein palmitoylation. Methods (San Diego, Calif 40: 1256, 2006. 150. Linder ME and Deschenes RJ. Palmitoylation: policing protein stability and traffic. Nat Rev Mol Cell Biol 8: 74-84, 2007. 151. Lingwood D and Simons K. Lipid rafts as a membrane-organizing principle. Science 327: 46-50, 2010. 152. Liu W, Yang J, Hu D, Kang C, Li C, Zhang S, Li P, Chen Z, Qin X, Ying K, Li Y, Li Y, Li Z, Cheng X, Li L, Qi Y, Chen S and Wang Q. KCNQ1 and KCNH2 mutations associated with long QT syndrome in a Chinese population. Hum Mutat 20: 475-476, 2002. 153. Liu Y, Jurman ME and Yellen G. Dynamic rearrangement of the outer mouth of a K+ channel during gating. Neuron 16: 859-867, 1996. 154. Livingstone M, Atas E, Meller A and Sonenberg N. Mechanisms governing the control of mRNA translation. Phys Biol 7: 021001, 2010. 155. Loewen ME, Wang Z, Eldstrom J, Dehghani Zadeh A, Khurana A, Steele DF and Fedida D. Shared requirement for dynein function and intact microtubule cytoskeleton for normal surface expression of cardiac potassium channels. Am J Physiol Heart Circ Physiol 296: H71-83, 2009. 156. Long SB, Campbell EB and Mackinnon R. Crystal structure of a mammalian voltagedependent Shaker family K+ channel. Science 309: 897-903, 2005. 157.  Lossin C. A catalog of SCN1A variants. Brain Dev 31: 114-130, 2009.  158. Ludwig A, Zong X, Jeglitsch M, Hofmann F and Biel M. A family of hyperpolarization-activated mammalian cation channels. Nature 393: 587-91, 1998. 159. Luthi A and McCormick DA. Modulation of a pacemaker current through Ca(2+)induced stimulation of cAMP production. Nat Neurosci 2: 634-641, 1999. 160. Ma D, Zerangue N, Lin YF, Collins A, Yu M, Jan YN and Jan LY. Role of ER export signals in controlling surface potassium channel numbers. Science 291: 316-319, 2001. 68  161. Ma Y and Hendershot LM. ER chaperone functions during normal and stress conditions. J Chem Neuroanat 28: 51-65, 2004. 162. Maccaferri G, Mangoni M, Lazzari A and DiFrancesco D. Properties of the hyperpolarization-activated current in rat hippocampal CA1 pyramidal cells. J Neurophysiol 69: 2129-2136, 1993. 163. MacDonald PE, Wang G, Tsuk S, Dodo C, Kang Y, Tang L, Wheeler MB, Cattral MS, Lakey JR, Salapatek AM, Lotan I and Gaisano HY. Synaptosome-associated protein of 25 kilodaltons modulates Kv2.1 voltage-dependent K(+) channels in neuroendocrine islet betacells through an interaction with the channel N terminus. Mol Endocrinol 16: 2452-2461, 2002. 164. Mackinnon R. Structural biology. Voltage sensor meets lipid membrane. Science 306: 1304-1305, 2004. 165. Magee JC and Carruth M. Dendritic voltage-gated ion channels regulate the action potential firing mode of hippocampal CA1 pyramidal neurons. Journal of neurophysiology 82: 1895-901, 1999. 166. Manganas LN, Akhtar S, Antonucci DE, Campomanes CR, Dolly JO and Trimmer JS. Episodic ataxia type-1 mutations in the Kv1.1 potassium channel display distinct folding and intracellular trafficking properties. J Biol Chem 276: 49427-49434, 2001. 167. Mangoni ME and Nargeot J. Genesis and regulation of the heart automaticity. Physiol Rev 88: 919-982, 2008. 168. Mannuzzu LM and Isacoff EY. Independence and cooperativity in rearrangements of a potassium channel voltage sensor revealed by single subunit fluorescence. J Gen Physiol 115: 257-268, 2000. 169. Marino M, Ascenzi P and Acconcia F. S-palmitoylation modulates estrogen receptor alpha localization and functions. Steroids 71: 298-303, 2006. 170. Marionneau C, Couette B, Liu J, Li H, Mangoni ME, Nargeot J, Lei M, Escande D and Demolombe S. Specific pattern of ionic channel gene expression associated with pacemaker activity in the mouse heart. The Journal of physiology 562: 223-34, 2005. 171. Martens JR, Navarro-Polanco R, Coppock EA, Nishiyama A, Parshley L, Grobaski TD and Tamkun MM. Differential targeting of Shaker-like potassium channels to lipid rafts. J Biol Chem 275: 7443-7446, 2000. 172. Martens JR, Sakamoto N, Sullivan SA, Grobaski TD and Tamkun MM. Isoformspecific localization of voltage-gated K+ channels to distinct lipid raft populations. Targeting of Kv1.5 to caveolae. J Biol Chem 276: 8409-8414, 2001. 173. McCrossan ZA, Roepke TK, Lewis A, Panaghie G and Abbott GW. Regulation of the Kv2.1 potassium channel by MinK and MiRP1. J Membr Biol 228: 1-14, 2009. 69  174. Meusser B, Hirsch C, Jarosch E and Sommer T. ERAD: the long road to destruction. Nat Cell Biol 7: 766-772, 2005. 175. Michels G, Er F, Khan IF, Endres-Becker J, Brandt MC, Gassanov N, Johns DC and Hoppe UC. K+ channel regulator KCR1 suppresses heart rhythm by modulating the pacemaker current If. PLoS One 3: e1511, 2008. 176. Milanesi R, Baruscotti M, Gnecchi-Ruscone T and DiFrancesco D. Familial sinus bradycardia associated with a mutation in the cardiac pacemaker channel. N Engl J Med 354: 151-157, 2006. 177. Mistrik P, Mader R, Michalakis S, Weidinger M, Pfeifer A and Biel M. The murine HCN3 gene encodes a hyperpolarization-activated cation channel with slow kinetics and unique response to cyclic nucleotides. J Biol Chem 280: 27056-27061, 2005. 178. Moosmang S, Biel M, Hofmann F and Ludwig A. Differential distribution of four hyperpolarization-activated cation channels in mouse brain. Biol Chem 380: 975-980, 1999. 179. Moosmang S, Stieber J, Zong X, Biel M, Hofmann F and Ludwig A. Cellular expression and functional characterization of four hyperpolarization-activated pacemaker channels in cardiac and neuronal tissues. European journal of biochemistry / FEBS 268: 1646-52, 2001. 180. Moroni A, Barbuti A, Altomare C, Viscomi C, Morgan J, Baruscotti M and DiFrancesco D. Kinetic and ionic properties of the human HCN2 pacemaker channel. Pflugers Arch 439: 618-626, 2000. 181. Moroni A, Gorza L, Beltrame M, Gravante B, Vaccari T, Bianchi ME, Altomare C, Longhi R, Heurteaux C, Vitadello M, Malgaroli A and DiFrancesco D. Hyperpolarizationactivated cyclic nucleotide-gated channel 1 is a molecular determinant of the cardiac pacemaker current I(f). The Journal of biological chemistry 276: 29233-41, 2001. 182. Much B, Wahl-Schott C, Zong X, Schneider A, Baumann L, Moosmang S, Ludwig A and Biel M. 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: 43781-6, 2003. 183. Muller F, Scholten A, Ivanova E, Haverkamp S, Kremmer E and Kaupp UB. HCN channels are expressed differentially in retinal bipolar cells and concentrated at synaptic terminals. Eur J Neurosci 17: 2084-2096, 2003. 184. Nadal MS, Ozaita A, Amarillo Y, Vega-Saenz de Miera E, Ma Y, Mo W, Goldberg EM, Misumi Y, Ikehara Y, Neubert TA and Rudy B. The CD26-related dipeptidyl aminopeptidase-like protein DPPX is a critical component of neuronal A-type K+ channels. Neuron 37: 449-461, 2003. 185. Nakahira K, Shi G, Rhodes KJ and Trimmer JS. Selective interaction of voltagegated K+ channel beta-subunits with alpha-subunits. J Biol Chem 271: 7084-7089, 1996. 70  186. Nakamoto T. Mechanisms of the initiation of protein synthesis: in reading frame binding of ribosomes to mRNA. Mol Biol Rep 2010. 187. Nakamura TY, Coetzee WA, Vega-Saenz De Miera E, Artman M and Rudy B. Modulation of Kv4 channels, key components of rat ventricular transient outward K+ current, by PKC. Am J Physiol 273: H1775-86, 1997. 188. Nakamura TY, Pountney DJ, Ozaita A, Nandi S, Ueda S, Rudy B and Coetzee WA. A role for frequenin, a Ca2+-binding protein, as a regulator of Kv4 K+-currents. Proc Natl Acad Sci U S A 98: 12808-12813, 2001. 189. Nakamura TY, Sturm E, Pountney DJ, Orenzoff B, Artman M and Coetzee WA. Developmental expression of NCS-1 (frequenin), a regulator of Kv4 K+ channels, in mouse heart. Pediatr Res 53: 554-557, 2003. 190. Napp J, Monje F, Stuhmer W and Pardo LA. Glycosylation of Eag1 (Kv10.1) potassium channels: intracellular trafficking and functional consequences. The Journal of biological chemistry 280: 29506-12, 2005. 191. Nazzari H, Angoli D, Chow SS, Whitaker G, Leclair L, McDonald E, Macri V, Zahynacz K, Walker V and Accili EA. 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: C642-52, 2008. 192. Nazzari H, Angoli D, Chow SS, Whitaker G, Leclair L, McDonald E, Macri V, Zahynacz K, Walker V and Accili EA. Regulation of cell surface expression of functional pacemaker channels by a motif in the B-helix of the cyclic nucleotide-binding domain. Am J Physiol Cell Physiol 295: C642-52, 2008. 193. Nerbonne JM. Molecular basis of functional voltage-gated K+ channel diversity in the mammalian myocardium. J Physiol 525 Pt 2: 285-298, 2000. 194. Niwa N and Nerbonne JM. Molecular determinants of cardiac transient outward potassium current (I(to)) expression and regulation. J Mol Cell Cardiol 48: 12-25, 2010. 195. Nof E, Luria D, Brass D, Marek D, Lahat H, Reznik-Wolf H, Pras E, Dascal N, Eldar M and Glikson M. 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, 2007. 196. Nof E, Luria D, Brass D, Marek D, Lahat H, Reznik-Wolf H, Pras E, Dascal N, Eldar M and Glikson M. 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, 2007. 197. Nolan MF, Malleret G, Dudman JT, Buhl DL, Santoro B, Gibbs E, Vronskaya S, Buzsaki G, Siegelbaum SA, Kandel ER and Morozov A. A behavioral role for dendritic 71  integration: HCN1 channels constrain spatial memory and plasticity at inputs to distal dendrites of CA1 pyramidal neurons. Cell 119: 719-732, 2004. 198. Noma K, Kimura K, Minatohara K, Nakashima H, Nagao Y, Mizoguchi A and Fujiyoshi Y. Triple N-glycosylation in the long S5-P loop regulates the activation and trafficking of the Kv12.2 potassium channel. J Biol Chem 284: 33139-33150, 2009. 199. Norris AJ and Nerbonne JM. Molecular dissection of I(A) in cortical pyramidal neurons reveals three distinct components encoded by Kv4.2, Kv4.3, and Kv1.4 alpha-subunits. J Neurosci 30: 5092-5101, 2010. 200.  Nufer O and Hauri HP. ER export: call 14-3-3. Curr Biol 13: R391-3, 2003.  201. O'Connell KM and Tamkun MM. Targeting of voltage-gated potassium channel isoforms to distinct cell surface microdomains. J Cell Sci 118: 2155-2166, 2005. 202. Ogielska EM, Zagotta WN, Hoshi T, Heinemann SH, Haab J and Aldrich RW. Cooperative subunit interactions in C-type inactivation of K channels. Biophys J 69: 2449-2457, 1995. 203. O'Kelly I, Butler MH, Zilberberg N and Goldstein SA. Forward transport. 14-3-3 binding overcomes retention in endoplasmic reticulum by dibasic signals. Cell 111: 577-588, 2002. 204. Ostrom RS and Liu X. Detergent and detergent-free methods to define lipid rafts and caveolae. Methods Mol Biol 400: 459-468, 2007. 205. Pape HC. Queer current and pacemaker: the hyperpolarization-activated cation current in neurons. Annu Rev Physiol 58: 299-327, 1996. 206. Pape HC. Specific bradycardic agents block the hyperpolarization-activated cation current in central neurons. Neuroscience 59: 363-373, 1994. 207. Pasyk EA, Kang Y, Huang X, Cui N, Sheu L and Gaisano HY. Syntaxin-1A binds the nucleotide-binding folds of sulphonylurea receptor 1 to regulate the KATP channel. J Biol Chem 279: 4234-4240, 2004. 208. Patel SP, Campbell DL, Morales MJ and Strauss HC. Heterogeneous expression of KChIP2 isoforms in the ferret heart. J Physiol 539: 649-656, 2002. 209. Perez-Garcia MT, Lopez-Lopez JR and Gonzalez C. Kvbeta1.2 subunit coexpression in HEK293 cells confers O2 sensitivity to kv4.2 but not to Shaker channels. J Gen Physiol 113: 897-907, 1999. 210. Petersen KR and Nerbonne JM. Expression environment determines K+ current properties: Kv1 and Kv4 alpha-subunit-induced K+ currents in mammalian cell lines and cardiac myocytes. Pflugers Arch 437: 381-392, 1999. 72  211. Petrecca K, Atanasiu R, Akhavan A and Shrier A. N-linked glycosylation sites determine HERG channel surface membrane expression. The Journal of physiology 515 ( Pt 1): 41-8, 1999. 212. Petrecca K, Miller DM and Shrier A. Localization and enhanced current density of the Kv4.2 potassium channel by interaction with the actin-binding protein filamin. J Neurosci 20: 8736-8744, 2000. 213. Petrecca K, Miller DM and Shrier A. Localization and enhanced current density of the Kv4.2 potassium channel by interaction with the actin-binding protein filamin. J Neurosci 20: 8736-8744, 2000. 214. Phuket TR and Covarrubias M. Kv4 Channels Underlie the Subthreshold-Operating A-type K-current in Nociceptive Dorsal Root Ganglion Neurons. Front Mol Neurosci 2: 3, 2009. 215. Poolos NP, Bullis JB and Roth MK. Modulation of h-channels in hippocampal pyramidal neurons by p38 mitogen-activated protein kinase. J Neurosci 26: 7995-8003, 2006. 216. Proenza C, Angoli D, Agranovich E, Macri V and Accili EA. Pacemaker channels produce an instantaneous current. J Biol Chem 277: 5101-5109, 2002. 217. Proenza C, Tran N, Angoli D, Zahynacz K, Balcar P and Accili EA. 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: 29634-42, 2002. 218. Qanbar R and Bouvier M. Role of palmitoylation/depalmitoylation reactions in Gprotein-coupled receptor function. Pharmacology & therapeutics 97: 1-33, 2003. 219. Qin D, Zhang ZH, Caref EB, Boutjdir M, Jain P and el-Sherif N. Cellular and ionic basis of arrhythmias in postinfarction remodeled ventricular myocardium. Circ Res 79: 461-473, 1996. 220. Qu J, Kryukova Y, Potapova IA, Doronin SV, Larsen M, Krishnamurthy G, Cohen IS and Robinson RB. MiRP1 modulates HCN2 channel expression and gating in cardiac myocytes. J Biol Chem 279: 43497-43502, 2004. 221. Qu J, Mulo I and Myhr KL. The development of Kv4.2 expression in the retina. Neurosci Lett 464: 209-213, 2009. 222. Radicke S, Cotella D, Sblattero D, Ravens U, Santoro C and Wettwer E. The transmembrane beta-subunits KCNE1, KCNE2, and DPP6 modify pharmacological effects of the antiarrhythmic agent tedisamil on the transient outward current Ito. Naunyn Schmiedebergs Arch Pharmacol 379: 617-626, 2009. 223. Radicke S, Vaquero M, Caballero R, Gomez R, Nunez L, Tamargo J, Ravens U, Wettwer E and Delpon E. Effects of MiRP1 and DPP6 beta-subunits on the blockade induced by flecainide of Kv4.3/KChIP2 channels. Br J Pharmacol 154: 774-786, 2008. 73  224. Rajan S, Preisig-Muller R, Wischmeyer E, Nehring R, Hanley PJ, Renigunta V, Musset B, Schlichthorl G, Derst C, Karschin A and Daut J. Interaction with 14-3-3 proteins promotes functional expression of the potassium channels TASK-1 and TASK-3. J Physiol 545: 13-26, 2002. 225. Ren J, Wen L, Gao X, Jin C, Xue Y and Yao X. CSS-Palm 2.0: an updated software for palmitoylation sites prediction. Protein Eng Des Sel 21: 639-644, 2008. 226. Rettig J, Heinemann SH, Wunder F, Lorra C, Parcej DN, Dolly JO and Pongs O. Inactivation properties of voltage-gated K+ channels altered by presence of beta-subunit. Nature 369: 289-294, 1994. 227. Rhodes KJ, Carroll KI, Sung MA, Doliveira LC, Monaghan MM, Burke SL, Strassle BW, Buchwalder L, Menegola M, Cao J, An WF and Trimmer JS. KChIPs and Kv4 alpha subunits as integral components of A-type potassium channels in mammalian brain. J Neurosci 24: 7903-7915, 2004. 228. Robertson GA and January CT. HERG trafficking and pharmacological rescue of LQTS-2 mutant channels. Handb Exp Pharmacol (171): 349-355, 2006. 229. Roepke TK, Kontogeorgis A, Ovanez C, Xu X, Young JB, Purtell K, Goldstein PA, Christini DJ, Peters NS, Akar FG, Gutstein DE, Lerner DJ and Abbott GW. Targeted deletion of kcne2 impairs ventricular repolarization via disruption of I(K,slow1) and I(to,f). FASEB J 22: 3648-3660, 2008. 230. Rosati B, Pan Z, Lypen S, Wang HS, Cohen I, Dixon JE and McKinnon D. Regulation of KChIP2 potassium channel beta subunit gene expression underlies the gradient of transient outward current in canine and human ventricle. J Physiol 533: 119-125, 2001. 231. Roth AF, Wan J, Bailey AO, Sun B, Kuchar JA, Green WN, Phinney BS, Yates JR,3rd and Davis NG. Global analysis of protein palmitoylation in yeast. Cell 125: 1003-13, 2006. 232. Rudy B. Molecular diversity of ion channels and cell function. Ann N Y Acad Sci 868: 112, 1999. 233. Salomons FA, Acs K and Dantuma NP. Illuminating the ubiquitin/proteasome system. Exp Cell Res 316: 1289-1295, 2010. 234. Sanguinetti MC. Reduced transient outward K+ current and cardiac hypertrophy: causal relationship or epiphenomenon? Circ Res 90: 497-499, 2002. 235. Santacruz-Toloza L, Huang Y, John SA and Papazian DM. Glycosylation of shaker potassium channel protein in insect cell culture and in Xenopus oocytes. Biochemistry 33: 560713, 1994. 236. Santoro B, Grant SG, Bartsch D and Kandel ER. Interactive cloning with the SH3 domain of N-src identifies a new brain specific ion channel protein, with homology to eag and 74  cyclic nucleotide-gated channels. Proceedings of the National Academy of Sciences of the United States of America 94: 14815-20, 1997. 237. Santoro B, Wainger BJ and Siegelbaum SA. Regulation of HCN channel surface expression by a novel C-terminal protein-protein interaction. J Neurosci 24: 10750-10762, 2004. 238. Sato A, Arimura T, Makita N, Ishikawa T, Aizawa Y, Ushinohama H, Aizawa Y and Kimura A. Novel mechanisms of trafficking defect caused by KCNQ1 mutations found in long QT syndrome. J Biol Chem 284: 35122-35133, 2009. 239. Scannevin RH, Murakoshi H, Rhodes KJ and Trimmer JS. Identification of a cytoplasmic domain important in the polarized expression and clustering of the Kv2.1 K+ channel. J Cell Biol 135: 1619-1632, 1996. 240. Schaletzky J and Rapoport TA. Ribosome binding to and dissociation from translocation sites of the endoplasmic reticulum membrane. Mol Biol Cell 17: 3860-3869, 2006. 241. Schmidt JW and Catterall WA. Palmitylation, sulfation, and glycosylation of the alpha subunit of the sodium channel. Role of post-translational modifications in channel assembly. J Biol Chem 262: 13713-13723, 1987. 242. Schrader LA, Anderson AE, Mayne A, Pfaffinger PJ and Sweatt JD. PKA modulation of Kv4.2-encoded A-type potassium channels requires formation of a supramolecular complex. J Neurosci 22: 10123-33, 2002. 243. Schrader LA, Anderson AE, Mayne A, Pfaffinger PJ and Sweatt JD. PKA modulation of Kv4.2-encoded A-type potassium channels requires formation of a supramolecular complex. J Neurosci 22: 10123-10133, 2002. 244. Schrag JD, Bergeron JJ, Li Y, Borisova S, Hahn M, Thomas DY and Cygler M. The Structure of calnexin, an ER chaperone involved in quality control of protein folding. Mol Cell 8: 633-644, 2001. 245. Schroeter A, Walzik S, Blechschmidt S, Haufe V, Benndorf K and Zimmer T. Structure and function of splice variants of the cardiac voltage-gated sodium channel Na(v)1.5. J Mol Cell Cardiol 2010. 246. Schulze-Bahr E, Neu A, Friederich P, Kaupp UB, Breithardt G, Pongs O and Isbrandt D. Pacemaker channel dysfunction in a patient with sinus node disease. J Clin Invest 111: 1537-1545, 2003. 247. Schwetz TA, Norring SA and Bennett ES. N-glycans modulate K(v)1.5 gating but have no effect on K(v)1.4 gating. Biochim Biophys Acta 1798: 367-375, 2010. 248. Scott DB, Blanpied TA and Ehlers MD. Coordinated PKA and PKC phosphorylation suppresses RXR-mediated ER retention and regulates the surface delivery of NMDA receptors. Neuropharmacology 45: 755-767, 2003. 75  249. Seikel E and Trimmer JS. Convergent modulation of Kv4.2 channel alpha subunits by structurally distinct DPPX and KChIP auxiliary subunits. Biochemistry 48: 5721-30, 2009. 250. Selth LA, Sigurdsson S and Svejstrup JQ. Transcript Elongation by RNA Polymerase II. Annu Rev Biochem 2010. 251. Serodio P, Kentros C and Rudy B. Identification of molecular components of A-type channels activating at subthreshold potentials. J Neurophysiol 72: 1516-1529, 1994. 252. Serodio P and Rudy B. Differential expression of Kv4 K+ channel subunits mediating subthreshold transient K+ (A-type) currents in rat brain. Journal of neurophysiology 79: 1081-91, 1998. 253. Shi G, Nakahira K, Hammond S, Rhodes KJ, Schechter LE and Trimmer JS. Beta subunits promote K+ channel surface expression through effects early in biosynthesis. Neuron 16: 843-852, 1996. 254. Shi G and Trimmer JS. Differential asparagine-linked glycosylation of voltage-gated K+ channels in mammalian brain and in transfected cells. The Journal of membrane biology 168: 265-73, 1999. 255. Shi W, Wymore R, Yu H, Wu J, Wymore RT, Pan Z, Robinson RB, Dixon JE, McKinnon D and Cohen IS. Distribution and prevalence of hyperpolarization-activated cation channel (HCN) mRNA expression in cardiac tissues. Circulation research 85: e1-6, 1999. 256. Shibata R, Misonou H, Campomanes CR, Anderson AE, Schrader LA, Doliveira LC, Carroll KI, Sweatt JD, Rhodes KJ and Trimmer JS. A fundamental role for KChIPs in determining the molecular properties and trafficking of Kv4.2 potassium channels. J Biol Chem 278: 36445-36454, 2003. 257. Shin KS, Maertens C, Proenza C, Rothberg BS and Yellen G. Inactivation in HCN channels results from reclosure of the activation gate: desensitization to voltage. Neuron 41: 73744, 2004. 258. Shin M and Chetkovich DM. Activity-dependent regulation of h channel distribution in hippocampal CA1 pyramidal neurons. J Biol Chem 282: 33168-33180, 2007. 259. Standley S, Roche KW, McCallum J, Sans N and Wenthold RJ. PDZ domain suppression of an ER retention signal in NMDA receptor NR1 splice variants. Neuron 28: 887898, 2000. 260. Steriade M and Timofeev I. Neuronal plasticity in thalamocortical networks during sleep and waking oscillations. Neuron 37: 563-576, 2003. 261. Stieber J, Herrmann S, Feil S, Loster J, Feil R, Biel M, Hofmann F and Ludwig A. 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: 15235-40, 2003. 76  262. Stockklausner C, Ludwig J, Ruppersberg JP and Klocker N. A sequence motif responsible for ER export and surface expression of Kir2.0 inward rectifier K(+) channels. FEBS Lett 493: 129-133, 2001. 263. Takimoto K, Yang EK and Conforti L. Palmitoylation of KChIP splicing variants is required for efficient cell surface expression of Kv4.3 channels. The Journal of biological chemistry 277: 26904-11, 2002. 264. Tao X, Lee A, Limapichat W, Dougherty DA and MacKinnon R. A gating charge transfer center in voltage sensors. Science 328: 67-73, 2010. 265. Thomsen MB, Sosunov EA, Anyukhovsky EP, Ozgen N, Boyden PA and Rosen MR. Deleting the accessory subunit KChIP2 results in loss of I(to,f) and increased I(K,slow) that maintains normal action potential configuration. Heart Rhythm 6: 370-377, 2009. 266. Topinka JR and Bredt DS. N-terminal palmitoylation of PSD-95 regulates association with cell membranes and interaction with K+ channel Kv1.4. Neuron 20: 125-134, 1998. 267. Tran N, Proenza C, Macri V, Petigara F, Sloan E, Samler S and Accili EA. A conserved domain in the NH2 terminus important for assembly and functional expression of pacemaker channels. The Journal of biological chemistry 277: 43588-92, 2002. 268. Trimmer JS. Peering into the birth canal during ion channel parturition. Neuron 44: 214216, 2004. 269. Tsaur ML, Sheng M, Lowenstein DH, Jan YN and Jan LY. Differential expression of K+ channel mRNAs in the rat brain and down-regulation in the hippocampus following seizures. Neuron 8: 1055-1067, 1992. 270. Tsutsumi R, Fukata Y and Fukata M. Discovery of protein-palmitoylating enzymes. Pflugers Arch 456: 1199-1206, 2008. 271. Ueda K, Nakamura K, Hayashi T, Inagaki N, Takahashi M, Arimura T, Morita H, Higashiuesato Y, Hirano Y, Yasunami M, Takishita S, Yamashina A, Ohe T, Sunamori M, Hiraoka M and Kimura A. Functional characterization of a trafficking-defective HCN4 mutation, D553N, associated with cardiac arrhythmia. J Biol Chem 279: 27194-27198, 2004. 272. Ufret-Vincenty CA, Baro DJ and Santana LF. Differential contribution of sialic acid to the function of repolarizing K(+) currents in ventricular myocytes. American journal of physiology 281: C464-74, 2001. 273. Van Wagoner DR, Pond AL, McCarthy PM, Trimmer JS and Nerbonne JM. Outward K+ current densities and Kv1.5 expression are reduced in chronic human atrial fibrillation. Circ Res 80: 772-781, 1997. 274. Veit M, Ponimaskin E and Schmidt MF. Analysis of S-acylation of proteins. Methods in molecular biology (Clifton, N J 446: 163-82, 2008. 77  275. Waechter CJ, Schmidt JW and Catterall WA. Glycosylation is required for maintenance of functional sodium channels in neuroblastoma cells. J Biol Chem 258: 5117-5123, 1983. 276. Wainger BJ, DeGennaro M, Santoro B, Siegelbaum SA and Tibbs GR. Molecular mechanism of cAMP modulation of HCN pacemaker channels. Nature 411: 805-10, 2001. 277. Wang H, Yan Y, Liu Q, Huang Y, Shen Y, Chen L, Chen Y, Yang Q, Hao Q, Wang K and Chai J. Structural basis for modulation of Kv4 K+ channels by auxiliary KChIP subunits. Nature neuroscience 10: 32-9, 2007. 278. Wang X, Bao J, Zeng XM, Liu Z and Mei YA. Elevation of intracellular Ca2+ modulates A-currents in rat cerebellar granule neurons. J Neurosci Res 81: 530-540, 2005. 279. Wang Z, Eldstrom JR, Jantzi J, Moore ED and Fedida D. Increased focal Kv4.2 channel expression at the plasma membrane is the result of actin depolymerization. Am J Physiol Heart Circ Physiol 286: H749-59, 2004. 280. Wang Z, Feng J, Shi H, Pond A, Nerbonne JM and Nattel S. Potential molecular basis of different physiological properties of the transient outward K+ current in rabbit and human atrial myocytes. Circ Res 84: 551-561, 1999. 281. Watabe AM, Zaki PA and O'Dell TJ. Coactivation of beta-adrenergic and cholinergic receptors enhances the induction of long-term potentiation and synergistically activates mitogenactivated protein kinase in the hippocampal CA1 region. J Neurosci 20: 5924-5931, 2000. 282. Watanabe I, Wang HG, Sutachan JJ, Zhu J, Recio-Pinto E and Thornhill WB. Glycosylation affects rat Kv1.1 potassium channel gating by a combined surface potential and cooperative subunit interaction mechanism. The Journal of physiology 550: 51-66, 2003. 283. Watanabe I, Zhu J, Recio-Pinto E and Thornhill WB. Glycosylation affects the protein stability and cell surface expression of Kv1.4 but Not Kv1.1 potassium channels. A pore region determinant dictates the effect of glycosylation on trafficking. The Journal of biological chemistry 279: 8879-85, 2004. 284. Weake VM and Workman JL. Inducible gene expression: diverse regulatory mechanisms. Nat Rev Genet 2010. 285. Weber IT, Gilliland GL, Harman JG and Peterkofsky A. Crystal structure of a cyclic AMP-independent mutant of catabolite gene activator protein. J Biol Chem 262: 5630-5636, 1987. 286. Wible BA, Yang Q, Kuryshev YA, Accili EA and Brown AM. Cloning and expression of a novel K+ channel regulatory protein, KChAP. J Biol Chem 273: 11745-11751, 1998. 287. Wickenden AD, Jegla TJ, Kaprielian R and Backx PH. Regional contributions of Kv1.4, Kv4.2, and Kv4.3 to transient outward K+ current in rat ventricle. Am J Physiol 276: H1599-607, 1999. 78  288. Wickenden AD, Lee P, Sah R, Huang Q, Fishman GI and Backx PH. Targeted expression of a dominant-negative K(v)4.2 K(+) channel subunit in the mouse heart. Circ Res 85: 1067-1076, 1999. 289. Winder DG, Martin KC, Muzzio IA, Rohrer D, Chruscinski A, Kobilka B and Kandel ER. ERK plays a regulatory role in induction of LTP by theta frequency stimulation and its modulation by beta-adrenergic receptors. Neuron 24: 715-726, 1999. 290. Xia S, Wang Y, Zhang Y, Deng SB, Du JL, Wang XC and She Q. Dynamic changes in HCN2, HCN4, KCNE1, and KCNE2 expression in ventricular cells from acute myocardial infarction rat hearts. Biochem Biophys Res Commun 2010. 291. Xu H, Guo W and Nerbonne JM. Four kinetically distinct depolarization-activated K+ currents in adult mouse ventricular myocytes. J Gen Physiol 113: 661-678, 1999. 292. Xu H, Li H and Nerbonne JM. Elimination of the transient outward current and action potential prolongation in mouse atrial myocytes expressing a dominant negative Kv4 alpha subunit. J Physiol 519 Pt 1: 11-21, 1999. 293. Xue Y, Chen H, Jin C, Sun Z and Yao X. NBA-Palm: prediction of palmitoylation site implemented in Naive Bayes algorithm. BMC Bioinformatics 7: 458, 2006. 294. Yamakawa T, Saith S, Li Y, Gao X, Gaisano HY and Tsushima RG. Interaction of syntaxin 1A with the N-terminus of Kv4.2 modulates channel surface expression and gating. Biochemistry 46: 10942-10949, 2007. 295. Yanagihara K and Irisawa H. Inward current activated during hyperpolarization in the rabbit sinoatrial node cell. Pflugers Arch 385: 11-19, 1980. 296. Yanai A, Huang K, Kang R, Singaraja RR, Arstikaitis P, Gan L, Orban PC, Mullard A, Cowan CM, Raymond LA, Drisdel RC, Green WN, Ravikumar B, Rubinsztein DC, El-Husseini A and Hayden MR. Palmitoylation of huntingtin by HIP14 is essential for its trafficking and function. Nature neuroscience 9: 824-31, 2006. 297. Yang EK, Alvira MR, Levitan ES and Takimoto K. Kvbeta subunits increase expression of Kv4.3 channels by interacting with their C termini. J Biol Chem 276: 4839-4844, 2001. 298. Yao JA, Jiang M, Fan JS, Zhou YY and Tseng GN. Heterogeneous changes in K currents in rat ventricles three days after myocardial infarction. Cardiovasc Res 44: 132-145, 1999. 299. Yashiro K and Philpot BD. Regulation of NMDA receptor subunit expression and its implications for LTD, LTP, and metaplasticity. Neuropharmacology 55: 1081-1094, 2008. 300. Ying SW, Abbas SY, Harrison NL and Goldstein PA. Propofol block of I(h) contributes to the suppression of neuronal excitability and rhythmic burst firing in thalamocortical neurons. Eur J Neurosci 23: 465-480, 2006. 79  301. Yokoyama CT, Myers SJ, Fu J, Mockus SM, Scheuer T and Catterall WA. Mechanism of SNARE protein binding and regulation of Cav2 channels by phosphorylation of the synaptic protein interaction site. Mol Cell Neurosci 28: 1-17, 2005. 302. Yokoyama CT, Myers SJ, Fu J, Mockus SM, Scheuer T and Catterall WA. Mechanism of SNARE protein binding and regulation of Cav2 channels by phosphorylation of the synaptic protein interaction site. Mol Cell Neurosci 28: 1-17, 2005. 303. Yu H, Wu J, Potapova I, Wymore RT, Holmes B, Zuckerman J, Pan Z, Wang H, Shi W, Robinson RB, El-Maghrabi MR, Benjamin W, Dixon J, McKinnon D, Cohen IS and Wymore R. MinK-related peptide 1: A beta subunit for the HCN ion channel subunit family enhances expression and speeds activation. Circ Res 88: E84-7, 2001. 304. Yu X, Chen XW, Zhou P, Yao L, Liu T, Zhang B, Li Y, Zheng H, Zheng LH, Zhang CX, Bruce I, Ge JB, Wang SQ, Hu ZA, Yu HG and Zhou Z. Calcium influx through If channels in rat ventricular myocytes. Am J Physiol Cell Physiol 292: C1147-55, 2007. 305. Yuan LL, Adams JP, Swank M, Sweatt JD and Johnston D. Protein kinase modulation of dendritic K+ channels in hippocampus involves a mitogen-activated protein kinase pathway. J Neurosci 22: 4860-4868, 2002. 306. Yuan LL, Adams JP, Swank M, Sweatt JD and Johnston D. Protein kinase modulation of dendritic K+ channels in hippocampus involves a mitogen-activated protein kinase pathway. J Neurosci 22: 4860-4868, 2002. 307. Zadeh AD, Cheng Y, Xu H, Wong N, Wang Z, Goonasekara C, Steele DF and Fedida D. Kif5b is an essential forward trafficking motor for the Kv1.5 cardiac potassium channel. J Physiol 587: 4565-4574, 2009. 308. Zagotta WN, Olivier NB, Black KD, Young EC, Olson R and Gouaux E. Structural basis for modulation and agonist specificity of HCN pacemaker channels. Nature 425: 200-205, 2003. 309. Zerangue N, Schwappach B, Jan YN and Jan LY. A new ER trafficking signal regulates the subunit stoichiometry of plasma membrane K(ATP) channels. Neuron 22: 537-548, 1999. 310. Zha Q, Brewster AL, Richichi C, Bender RA and Baram TZ. Activity-dependent heteromerization of the hyperpolarization-activated, cyclic-nucleotide gated (HCN) channels: role of N-linked glycosylation. Journal of neurochemistry 105: 68-77, 2008. 311. Zhang L, Foster K, Li Q and Martens JR. S-acylation regulates Kv1.5 channel surface expression. Am J Physiol Cell Physiol 293: C152-61, 2007. 312. Zhao C, Wang L, Netoff T and Yuan LL. Dendritic mechanisms controlling the threshold and timing requirement of synaptic plasticity. Hippocampus 2010.  80  313. Zhu J, Recio-Pinto E, Hartwig T, Sellers W, Yan J and Thornhill WB. The Kv1.2 potassium channel: the position of an N-glycan on the extracellular linkers affects its protein expression and function. Brain Res 1251: 16-29, 2009. 314. Zhu J, Watanabe I, Gomez B and Thornhill WB. Trafficking of Kv1.4 potassium channels: interdependence of a pore region determinant and a cytoplasmic C-terminal VXXSL determinant in regulating cell-surface trafficking. Biochem J 375: 761-768, 2003. 315. Zolles G, Wenzel D, Bildl W, Schulte U, Hofmann A, Muller CS, Thumfart JO, Vlachos A, Deller T, Pfeifer A, Fleischmann BK, Roeper J, Fakler B and Klocker N. Association with the auxiliary subunit PEX5R/Trip8b controls responsiveness of HCN channels to cAMP and adrenergic stimulation. Neuron 62: 814-825, 2009. 316. Zong X, Eckert C, Yuan H, Wahl-Schott C, Abicht H, Fang L, Li R, Mistrik P, Gerstner A, Much B, Baumann L, Michalakis S, Zeng R, Chen Z and Biel M. A novel mechanism of modulation of hyperpolarization-activated cyclic nucleotide-gated channels by Src kinase. J Biol Chem 280: 34224-34232, 2005.  81  CHAPTER 2: REGULATION OF CELL SURFACE EXPRESSION OF FUNCTIONAL PACEMAKER CHANNELS BY A MOTIF IN THE BHELIX OF THE CYCLIC NUCLEOTIDE-BINDING DOMAIN  A version of this chapter has been accepted for publication. Nazzari, H., Angoli, D., Chow, S.S., Whitaker, G., Leclair, L., McDonald, E., Macri, V., Zahynacz, K., Walker, V., 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 – Cell Physiology. 2008 Sep;295(3):C642-52. 82  2.1 INTRODUCTION Hyperpolarization-activated cyclic nucleotide-gated "pacemaker" channels (HCN1-4) contribute to the regulation of spontaneous activity and membrane potential in mammalian cardiac conduction tissue and brain (21). The number of HCN channels on the cell surface is critical to these functions, but the factors that determine their supply to this region of the cell have not been extensively studied. In general, export of plasma membrane-bound ion channels from the endoplasmic reticulum (ER) to the Golgi is limited by multiple quality control mechanisms (4, 6, 7, 22). The export of properly folded and assembled channels from the ER is also regulated and may depend on anterograde signals (6, 14, 15, 17, 24). In HCN2 channels, the cyclic nucleotide-binding domain (CNBD), located in the COOH terminus, appears to be an important determinant of cell surface expression, in addition to its better known role as regulator of channel opening (31), based on two studies (1, 20). First, complex glycosylation is abolished in HCN2 mutants lacking the CNBD, which supports its necessity for export of the channel from the ER (1). Second, we identified a subdomain of the CNBD that strongly promotes cell surface and functional expression; mutants lacking this subdomain do not generate current and are retained intracellularly (20). This same subdomain, which consists of the A and B helices and the interceding β-barrel, exerts tonic inhibition of channel opening in response to hyperpolarization (5). Whether the complete subdomain is required for both functions, perhaps by conferring a shared conformational change, or includes distinct regions that contribute to each function is not known. The mechanism by which the CNBD subdomain promotes cell surface expression is poorly understood. To date, it has been shown that HCN2 channels lacking the entire CNBD do not appear to form functional channels when expressed in mammalian cells (20, 25), but do in Xenopus laevis oocytes (31). Moreover, in mammalian cells these same mutant channels form 83  homotetramers that do not exit the ER (1) but can be rescued by coassembly with wild-type HCN1 subunits to form functional, heteromeric channels (20). Together, these data suggest that the CNBD promotes ER export and cell surface expression in at least two ways. First, the efficiency with which channels correctly fold, assemble, and move through quality control pathways could be enhanced. Second, a step in forward trafficking may be augmented.  Studies  by our group and by Akhavan et al. (1) have narrowed the region of the HCN2 CNBD subdomain that promotes cell surface expression to the B-helix (1, 20). We showed that a COOH-terminal truncation mutant that lacks the B-helix eliminates functional expression and thus has the same phenotype as the complete CNBD truncation mutant, whereas that which retains the B-helix produces wild-type levels of current (20). Akhavan et al. (1) showed that the proline at the end of the B-helix is required for complex glycosylation, which is presumably indicative of cell surface expression. Whether the B-helix contributes to inhibition of channel opening is not known and remains to be investigated. The hypothesis that determinants within the B-helix of the HCN2 channel play a role in both cell surface expression and inhibition of hyperpolarization-activated opening forms the basis of the current study. Herein, we identify a four-amino acid motif (EEYP) that promotes ER export and cell surface expression of functional HCN2 channels by augmenting a step in forward trafficking and/or enhancing the efficiency of folding and assembly. However, the EEYP motif does not contribute to the inhibition of channel opening, indicating that cell surface expression and inhibition of channel opening are carried out, at least in part, by separate regions of the CNBD subdomain.  84  2.2 MATERIALS AND METHODS 2.2.1 Mutagenesis and expression COOH-terminal CNBD truncation mutants were constructed by engineering stop codons into mouse HCN2 cDNA using overlapping mutagenic PCR, and included (amino acid site of stop codon indicated in parentheses) HCN2- CNBD (525) and HCN2- EEYP-C-term (615). Other CNBD mutants constructed were as follows: HCN2-4A, generated by replacing EEYP (616–619) with four alanines using overlapping mutagenic PCR; HCN2-4A- C-term, which has a stop codon inserted immediately after the alanine string (620) in the HCN2-4A construct; and HCN2-EEAA, HCN2-AAYP, and HCN2-EEYM [identified as P619M by Akhavan et al. (1)] by overlapping mutagenic PCR. NH2-terminal deletion mutants, some of which were characterized in previous studies (20, 26), were used comparatively with the COOH-terminal mutants. These included HCN2- 2-130, HCN2- 2-137, HCN2- 2-138, HCN2- 2-143, HCN2- 2-154, and HCN2- 2-182, which were constructed by replacing an EcoRI-AccI restriction fragment of the wild-type HCN2 channel with a PCR product lacking the coding sequence for residues 2-130, 2-137, 2-138, 2-143, 2-154, and 2-182, respectively. All constructs were cloned into the pcDNA3.1 mammalian expression vector (Invitrogen, Burlington, ON, Canada). A glycosylation-deficient HCN2 mutant (HCN2-N380Q) was also created by replacing Asn380 with Gln using overlapping mutagenic PCR. In addition, COOH-terminal-myc (c-myc)tagged versions of wild-type HCN2 and the NH2-terminal deletion mutant channels were constructed using PCR, with BamHI and EcoRI common restriction sites incorporated on the ends of the primers, to allow subcloning into pcDNA 3.1-myc such that the myc protein is expressed on the COOH-terminal end of the resulting fusion protein.  85  Lastly, HCN2 channels were constructed with an extracellular hemagglutinin (HA) tag between the third and fourth transmembrane segments (S3-S4). An HCN2 construct with the HA epitope in pBluescript (a kind gift from Michael Sanguinetti, University of Utah) was excised from pBluescript and ligated into pcDNA3.1 for expression in Chinese hamster ovary (CHO) cells (HCN2-HA). The HA epitope was also inserted into the correlate domain of HCN2CNBD, HCN2- 2-182, HCN2- EEYP-C-term, HCN2-4A- C-term, and HCN2-4A, which were each then cloned into pcDNA3.1 using the same restriction enzyme sites. All constructs were confirmed by automated DNA sequencing (Nucleic Acid Protein Services, University of British Columbia). The HA- and c-myc-tagged wild-type channels were found to be functional using whole cell patch-clamp electrophysiology (see below). CHO-K1 cells (American Tissue Type Culture Collection, Manassas, VA) were maintained in Ham's F-12 medium supplemented with penicillin, streptomycin, and 10% fetal bovine serum and were incubated at 37°C with 5% CO2. Cells were plated onto glass coverslips in 35-mm dishes. One day after plating, mammalian expression vectors encoding wild-type or mutant HCN2 channels (2 µg/dish) were transiently cotransfected into the cells along with the green fluorescent protein reporter plasmid (0.5–0.7 µg/dish) using the FuGENE 6 transfection reagent (Roche Molecular Biochemicals, Indianapolis, IN). Cells expressing the transfected DNA were identified by the appearance of green fluorescence 24–48 h after transfection. 2.2.2 Whole cell patch-clamp electrophysiology One to two days following transfection, a shard of coverslip plated with cells was transferred to a recording chamber ( 200 µl vol) and continually perfused (0.5–1.0 ml/min) with a low K+ extracellular solution (5.4 mM KCl, 135 mM NaCl, 0.5 mM MgCl2, 1.9 mM CaCl2, and 5 mM HEPES, adjusted to pH 7.4 with NaOH). Following rupture of the patch membrane, this was switched to a high K+ extracellular solution (135 mM KCl, 5.4 mM NaCl, 0.5 mM 86  MgCl2, 1.9 mM CaCl2, and 5 mM HEPES, adjusted to pH 7.4 with KOH) to maximize current amplitude. The patch pipettes were filled with a solution containing 130 mM potassium aspartate, 10 mM NaCl, 0.5 mM MgCl2, 5 mM HEPES, and 1 mM EGTA and adjusted to pH 7.4 with KOH. Whole cell currents were measured using borosilicate glass electrodes (Sutter, Novato, CA), which had a resistance of 2.0–4.0 M when filled with the intracellular solution. Currents were recorded using an Axopatch 200B amplifier and Clampex software (Axon Instruments). Data were filtered at 2 kHz and were analyzed using Clampfit (Axon Instruments) and Origin (Microcal) software. All experiments were conducted at room temperature (20–22°C). Series resistance was not compensated, and currents were not leak-subtracted. The voltage dependence of activation was determined from tail currents at –65 mV following 2–3 s test pulses; interpulse intervals were 10–15 s and were followed by a 200–500 ms pulse to +5 mV to ensure complete channel deactivation. The resting current was always at its baseline value before subsequent voltage (V) pulses. Normalized tail current amplitudes were plotted as a function of test potential, and values were fitted with a Boltzmann function  to determine the midpoint of activation (V1/2) and slope factor (k). Statistical comparisons were performed using a student's t-test or a one-way ANOVA followed by Tukey's post hoc analysis; significance was assumed if the P value was <0.05. Data are reported as means ± SE, and n values represent the number of cells tested, which were from a minimum of three separate transfections for each value reported. 2.2.3 Immunocytochemistry and microscopy For these experiments, HCN2 constructs containing an HA-epitope inserted between the third and fourth transmembrane segments were used to identify relative cell surface expression. 87  Two to three days after transfection, cells on coverslips were washed with phosphate-buffered saline (PBS) and fixed in 2% paraformaldehyde in PBS for 5 min. Thereafter, they were washed with PBS, were either permeabilized using 0.2% Triton X-100 or left unpermeabilized, and were then blocked with 10% normal goat serum (NGS). After one wash with PBS containing 1% NGS, cells were incubated with a mouse monoclonal antibody specific to the HA-epitope (Sigma, Oakville, ON) at a dilution of 1:500 in PBS with 1% NGS for 1 h at room temperature. The antibody-containing solution was removed, cells were washed with PBS/NGS 1%, and then incubated with a goat anti-mouse secondary antibody tagged with Alexa 488 (Invitrogen) at a dilution of 1:1,500 in PBS with 1% NGS for 1 h at room temperature in the dark. This solution was removed, and then, after being washed with PBS/NGS 1%, the coverslips were mounted on slides using Gelmount (Sigma) and sealed with clear nail polish. Cells were examined using structured illumination (Zeiss Apotome Imager Z1) with a 63x oil immersion objective lens at wavelengths specific for the Alexa 488 fluorescent protein tag. 2.2.4 Western blot analysis Each sample subjected to Western blot analysis was derived from cells on 35-mm plates that had been lysed in 100 µl of lysis buffer containing 50 mM Tris at pH 8.0, 1% NP40, 150 mM NaCl, 1 mM EDTA, 1 mM PMSF, 2 mM each of Na3VO4 and NaF, and 10 µg/ml each of aprotinin, pepstatin, and leupeptin. Samples were left on ice for 30 min, during which time they were vortexed every 5 min for 5 s. After centrifugation to remove cell debris (25,000 g, 25 min), protein concentration of the supernatant was determined by Bradford assay. Samples of supernatant (20 µg) were fractionated by sodium dodecyl sulphate-polyacrylamide gel electrophoresis (8%) and electroblotted to polyvinylidene difluoride (PVDF) membrane (BioRad, Mississauga, ON, Canada), unless otherwise indicated. Blots were washed three times in TBST (50 mM Tris, pH 7.4, 150 mM NaCl, and 0.1% Tween 20) and then blocked with 5% 88  nonfat dry milk (Bio-Rad) in TBST for 1 h at room temperature. Blots were then incubated with one of the following antibodies: 1) a rabbit polyclonal antibody specific to the COOH terminus of HCN2 (Affinity Bioregions, Golden, CO); 2) a rabbit polyclonal antibody specific to the NH2 terminus of HCN2 (Alomone Laboratories, Jerusalem, Israel); or 3) a mouse monoclonal antibody (Invitrogen) specific to the myc epitope—at dilutions of 1:500, 1:200, and 1:3,000 in TBST with 5% nonfat dry milk, respectively, for 2.5 h at room temperature. Blots were washed in TBST for 10 min, three times, and then incubated with horseradish peroxidase-conjugated to either goat anti-rabbit or goat anti-mouse IgG, accordingly, at 1:3,000 dilution in 5% nonfat dry milk with TBST for 1 h at room temperature; they were subsequently washed three times in TBST. Signals were obtained with ECL Western Blotting Detection Reagents (GE Healthcare, Baie d'Urfe, QC, Canada). Protein loading was controlled by probing all Western blots with either goat anti-GAPDH antibody or rabbit anti-ACTIN (both from Santa Cruz Biotechnology, Santa Cruz, CA). In some experiments, densitometry was carried out to determine the intensities of bands of interest in the Western blots, using ImageJ software (http://rsb.info.nih.gov/ij/). A rectangle of fixed size was centered on the band of interest. Within each designated region the density was determined and corrected for background. 2.2.5 Proteinase K treatment Twenty-four hours after transfection, cells were washed in ice-cold PBS three times and incubated with or without 20 µg/ml proteinase K (PK) (BioShop, McGill University, Montreal, QC, Canada) in PK buffer (10 mM HEPES, 150 mM NaCl, and 2 mM CaCl2) at 37°C for 30 min. To stop the PK reaction, a blocking buffer (25 mM EDTA and 20 mM PMSF) was added to all samples for 10 min at 4°C. Cells were harvested by centrifugation at 5,000 rpm for 5 min at 4°C  89  and were then washed with ice-cold PBS twice. Pellets were resuspended in lysis buffer and processed for Western blotting as described above. 2.2.6 Sucrose gradient analysis Cell lysates of HCN2 and HCN2-4A were subjected to high-speed centrifugation at 140,000 g for 45 min at 4°C. The supernatants (volume chosen to obtain 200 µg protein following Bradford assay) were layered on top of a 5–40% nondenaturing continuous sucrose gradient, made with lysis buffer. Molecular mass protein standards, including bovine serum albumin (66 kDa), alcohol dehydrogenase (151 kDa), and thyroglobulin (669 kDa), 200 µg each, were each layered on separate sucrose gradients. Samples were centrifuged for 16 h at 106,000 g at 4°C, and 13 equal volume fractions were collected serially from the bottom of the gradient. Subsequently, each set of fractions was subjected to Western blotting as described above, with the following changes. Proteins were transferred onto Immobilon-FL PVDF membranes (Millipore, Bedford, MA). Following incubation with the rabbit polyclonal antibody specific for the HCN2 COOH terminus, the blots were washed in TBST and then labeled for 1 h with a secondary anti-rabbit antibody conjugated to Alexa Fluor 680 (1:40,000) (Molecular Probes, Eugene, OR) for analysis on a LI-COR Odyssey imager (Lincoln, NE). For each blot, pixel intensity of each fraction was determined by densitometry and normalized to the highest intensity.  90  2.3 RESULTS 2.3.1 Identification of a motif in the CNBD B-helix potentially important for channel trafficking and function The regulation of cell surface trafficking was an unexpected property of the CNBD, which is better known for its ability to bind cyclic nucleotides in both channels and cytoplasmic proteins. In a previous study, we showed that a COOH-terminal truncation mutant that lacks the B-helix eliminates functional expression and thus has the same phenotype as the complete CNBD truncation mutant; however, that which retains the B-helix produces wild-type levels of current (20). To identify domains within the CNBD, and specifically the B-helix, that might confer channel-specific properties such as efficient cell surface trafficking and inhibition of channel opening, we compared the structures of the CNBDs found in different channels and cytoplasmic proteins. The tertiary structure of the CNBD in HCN2 is conserved among certain channels [e.g., cyclic nucleotide-gated (CNG) channels, ether-a-go-go (ERG)-related channels] and cytoplasmic proteins (e.g., protein kinases, catabolic activating peptide) (32). The CNBD in these structures is composed of both β-sheets and -helices (Fig.2.1). An eight-stranded β-barrel in the center of the CNBD forms a basket in which cAMP binds and is shielded from solvent and phosphodiesterases. The most conserved feature among the CNBDs is contained within this barrel—the phosphate binding cassette (PBC), composed of β-strand 6, a short helix, and βstrand 7. A buried arginine that binds to the exocyclic phosphate of cAMP and a glutamate that binds the ribose 2'-OH are conserved features of the PBC. A sequence alignment of the CNBDs from three ion channels, HCN2, CNGA3, and ERG1, and two cytoplasmic protein kinases, RII and RIβ, all from mouse, shows both similarities and differences among them (Fig. 2.1). The similarities are greatest in two regions. 91  First, as discussed, regions corresponding to the PBC contain key conserved residues. Second, are regions between β-strand 1 and 4, which may be important for shielding cAMP from solvent and from phosphodiesterases. Sequence alignments, as well as more sophisticated structural alignments, of protein kinases and HCN2 show that the similarities in the CNBD fall off in the middle of the B-helix (3). Our alignment, which includes CNGA3 and ERG1 ion channel sequences, revealed a conserved four-amino acid motif in the distal B-helix among the channels that was not found in the protein kinases. This motif corresponds to EEYP in HCN2, TEYP in CNGA3, and DMYP in ERG1. The presence of this conserved hydrophilic motif in these ion channels but not in the protein kinases raises the possibility that it may confer the channelspecific functions of inhibition of channel opening and cell surface expression attributed to the CNBD. Support for the role of EEYP in channel cell surface expression, in particular, comes from a study in which mouse HCN2 channels possessing a methionine in place of the proline (P619M) reduces complex glycosylation (1); however, the effects of this mutation on folding, assembly, and channel function was not assessed. 2.3.2 The EEYP motif promotes ER export and cell surface expression of mature HCN2 To investigate whether the EEYP motif promotes the expression of functional channels, we constructed a series of HCN2 COOH-terminal deletion and substitution mutants that target this region (Fig. 2.2). The deletion mutants were HCN2- EEYP-C-term (the EEYP motif and downstream COOH terminus is deleted, eliminating residues 616-863), HCN2-4A- C-term (the EEYP motif is replaced by four alanines and the remainder of the COOH terminus is deleted), and finally, HCN2-4A (full-length HCN2 with four alanines in place of the EEYP motif). To determine whether the EEYP motif promotes export of HCN2 from the ER, Western blots from cells transfected with wild-type HCN2, the EEYP mutant channels, and HCN2N380Q, the glycosylation-deficient HCN2 mutant, were compared to evaluate relative levels of 92  complex glycosylation (Fig. 2.3A). In cells expressing wild-type HCN2, two major bands were identified at 136 kDa (mature) and 114 kDa (immature), whereas only the immature form was identified in cells expressing HCN2-N380Q, as expected (16). We found that the mature band is consistently heavier by 22 kDa. In HCN2-4A, the same sequence length as wild-type HCN2, a band at 114 kDa alone was identified. In cells expressing HCN2- EEYP-C-term and HCN2-4A- C-term, two bands at 90 and 83 kDa were identified. Importantly, a band corresponding to a mature form of EEYP mutant channels was not seen in any of the Western blots (12/12 transfections), whereas a mature form of the wild-type channel was present in blots from the same transfections. Because the 22-kDa increase in molecular mass seen upon complex glycosylation of HCN2 was not seen in the three EEYP mutant channels, these likely remain in a simple glycosylated and/or unglycosylated state. Thus, the EEYP motif promotes complex glycosylation of HCN2 channels and thus ER export. (Fig. 2.3A). The augmentation of complex glycosylation suggests that the EEYP motif might also promote cell surface localization of HCN2 channels. To test this possibility, we next conducted experiments with the wild-type and the three EEYP mutant channels, tagged with an extracellular HA epitope and imaged by immunofluorescence (Fig. 2.3B). In nonpermeabilized CHO cells expressing HCN2-HA, significant levels of immunofluorescence were seen, as in Fig. 2.1. In contrast, the three EEYP HA-tagged mutants showed relatively reduced levels using the same exposure time. (Fig. 2.3B, top). While these levels were reduced, they were distinguishable and above the level of background (data not shown). In permeabilized cells, significant and comparable levels are seen among wild-type and EEYP HA-tagged mutant channels (Fig. 2.3B, bottom). This suggests that the EEYP motif is not only required for ER export and complex glycosylation, but also for cell surface localization.  93  To link directly EEYP-mediated changes in complex glycosylation, and thus ER export, to cell surface localization, PK experiments were carried out on intact cells transfected with HCN2, HCN2-4A, or HCN2-N380Q, and complex glycosylation was examined (Fig. 2.3C). Exposure of intact cells to PK would be expected to cleave extracellular regions of only those channels localized to the plasma membrane. As shown, exposure to PK resulted in the elimination of the mature complex glycosylated form of the wild-type channel and produced an additional lower-molecular-mass broad band indicative of cleaved by-products. Exposure of cells expressing HCN2-4A or HCN2-N380Q to PK did not produce any additional bands. Notably, there was no effect of PK on the immature (114 kDa) form of these channels. These data not only show that the mature complex glycosylated form of the channel is present on the cell surface, as was appreciated by Ahkavan et al. (1), but, specifically, that the EEYP motif promotes this pattern of cell surface expression. 2.3.3 The EEYP motif is not required to form functional channels and does not contribute to inhibition of channel opening The reduction in mature protein for the mutant EEYP channels (above) suggests that they are retained in the ER, but it does not provide information about the ability of the mutant channels to fold, assemble, and function. Are the mutant channels irreversibly misfolded? The parallel observation that small amounts of EEYP-lacking protein traffic to the cell surface (Fig. 2.3B) suggested that the EEYP motif is not required to form channels that reach the plasma membrane. To test whether these channels reach the plasma membrane in functional form, the densities of hyperpolarization-activated current (If) were determined in transiently transfected CHO cells using whole cell patch-clamp electrophysiology. If densities produced by HCN2EEYP-C-term, HCN2-4A- C-term, and HCN2-4A were significantly reduced compared with those of wild-type HCN2 channels. This is illustrated in Fig. 2.4A, where representative current 94  traces elicited by a 2-s hyperpolarizing pulse to –150 mV (at which the channels were at, or close to, full activation) from a holding potential of –35 mV are shown. If densities at –150 mV are shown in the bar graph in Fig. 2.4B. These data are consistent with the notion that the EEYP motif promotes export of channels from the ER to the plasma membrane and, importantly, is not required to form functional channels. Thus, a small number of functional EEYP mutant channels are able to make it to the plasma membrane, as foreshadowed by Fig. 2.3B. The lack of the mature forms of the EEYP mutants by Western blot (Fig. 2.3A) suggests that the levels of mature protein were below detection or that the immature forms contribute to the observed current. The latter explanation seems less likely given that the immature bands are not modified by PK (Fig. 2.3C). Removal of the EEYP motif could also reduce channel activity, which would contribute to lower levels of If. To test whether the amount of mature mutant protein was below the level of detection, we repeated Western blotting experiments using HCN2 and HCN2-4A and found that this form became detectable when the amount of protein loaded was increased by over four times (Fig. 2.4C). This fits very well with the reduced level of current produced by this mutant, which is <30% of that produced by the wild-type channel. We also found that the mature form of the wild-type channel is no longer apparent in Western blots when the amount of protein loaded was reduced by four times (data not shown). Together, the results suggest that the amount of mature protein produced by HCN2-4A is reduced compared with the wild-type channel, which, in turn, leads to a corresponding reduction in current. Next, we wanted to determine whether the EEYP motif contributes to inhibition of channel opening by hyperpolarization. A limiting factor was that the EEYP mutant channels produced levels of If that were often too small to analyze. This issue could be circumvented by selection of relatively large CHO cells that produced adequate levels of If to permit accurate 95  analysis. We thus repeated our measurements of If in large CHO cells transfected with HCN2EEYP-C-term, HCN2-4A- C-term, and HCN2-4A (Fig. 2.5A). To determine whether the EEYP contributes to basal inhibition of channel opening, If activation curves were generated from the ratio of tail current amplitudes elicited at the –65 mV voltage step (Fig. 2.5A, arrows; 2.5B). If the EEYP contributes to the inhibition of channel opening by the CNBD, then we would expect the activation curves to be shifted in the positive direction in channels lacking this motif. However, we found that the activation curves for all three EEYP mutants were not statistically different from wild type. The similarity in channel activation curves indicates that the EEYP does not contribute to basal inhibition of channel opening by the CNBD. Notably, If activation curves from cells expressing HCN2-4A were difficult to obtain, which could impact determination of If density. First, upon hyperpolarization, the cells often did not return to baseline, and thus activation curves could not be determined. Second, when activation curves could be determined, they were shifted in the negative direction, although this was not statistically significant (Fig. 2.5B). Nonetheless, a negative shift could lead to an underestimation of the If density determined at –150 mV in cells transfected with HCN2-4A. To correct for this possibility, prepulses more negative than –150 mV were sometimes required to ensure complete activation for this mutant. We thus repeated our measurements of If density, determined from tail currents recorded at –65 mV following the appropriate prepulse, in HCN2 and all three EEYP mutants (Fig. 2.5C). These data confirm that the EEYP motif increases If density. 2.3.4 The EEYP motif regulates cell surface expression by a mechanism that does not lead to substantive degradation or disruption of subunit assembly To this point, the data suggest that there are at least two ways in which the EEYP may regulate ER export and cell surface expression of functional HCN2 channels. First, the efficiency 96  with which the channel folds, assembles, and moves through the trafficking pathway may be facilitated by the EEYP motif and, conversely, compromised by its absence. In this scenario, the majority of EEYP mutant channels would be identified as misfolded. Under some circumstances, and depending on the cellular checkpoint at which the channels are identified, the channels would be subject to rapid degradation (29). Second, the EEYP motif may promote specifically the export of correctly folded channels from the ER to the cell surface. In this case, EEYP mutant channels might remain in a relatively stable state in the ER, Golgi, or other compartment along the trafficking pathway and may not undergo rapid degradation. To limit the mechanisms that might apply, degradation products were evaluated, in HCN2-4A (the full-length representative of the EEYP mutant phenotype) as compared with HCN2 and a set of HCN2 NH2-terminal deletion mutants characterized by altered trafficking, enhanced shunting of channel protein to the degradation pathway and diminished production of current (unpublished observations). HCN2- 2-137 (NH2-terminal amino acids 2-137 removed) was first used in this analysis because it produces levels of If comparable to HCN2-4A. Representative current traces for the NH2-terminal mutant HCN2- 2-137 and HCN2-4A mutant are shown in comparison to wild-type HCN2, in Fig. 2.6A. In both mutants a significant and comparable reduction in If density relative to the wild-type channel was confirmed (Fig. 2.6B); If activation is unaltered (Fig. 2.6C). To examine the degradation profile of HCN2-4A, we compared smaller-molecular-mass fragments among wild-type, HCN2-4A, HCN2- 2-137, and HCN2-N380Q channels by Western blotting, using a COOH-terminally directed antibody for HCN2 (Fig. 2.6D). Mature and immature forms were again identified at 136 kDa and 114 kDa for HCN2, whereas only the immature form was identified for HCN2-4A and HCN2-N380Q. In HCN2- 2-137, a band corresponding to an immature form ( 83 kDa), and an obvious smaller band at 53 kDa were 97  present (Fig. 2.6D). A band at this lower molecular mass was also found in lanes containing HCN2, HCN2-N380Q, and HCN2-4A, but at very low intensity. This 53-kDa band is likely a degraded form of HCN2, and its greater intensity in the lane containing HCN2- 2-137 suggests that this mutant undergoes degradation to a greater extent. Successively larger truncations of the NH2 terminus progressively engage the degradation pathway and decrease If density (Fig. 2.7). This demonstrates that the level of cell surface expression of the NH2-terminal truncation mutants parallels the extent of observed degradation. Thus, the efficiency of channel folding and assembly of these NH2-terminal deletion mutants is likely decreased. On the contrary, for HCN24A, there is a low level of degradation observed that is comparable to the wild-type channel. Thus, the roughly equivalent reduction in functional expression of the HCN2-4A and NH2terminal truncations must therefore come about by different mechanisms. The EEYP mutant channels, which are ostensibly immature, likely reside in an intracellular compartment in relatively stable state. The similar lack of degradation seen in the HCN2-N380Q mutant, which too is immature, suggests that a trafficking step that involves complex glycosylation may be regulated by the EEYP motif. As for the HCN2-4A, bands corresponding to the mature form of the two NH2-terminal mutant channels were apparent when four times more protein was loaded (data not shown); this corresponds to the lower levels of current measured and suggests that the mutant channels were retained to a greater extent compared with wild-type channels. To determine whether the EEYP motif reduces the efficiency of intracellular subunit assembly, we analyzed soluble lysates from cells expressing HCN2 and HCN2-4A using nondenaturing continuous sucrose gradients. As shown in Fig. 2.8, the sedimentation profiles for the immature (intracellular) forms of HCN2 and HCN2-4A are similar. These data suggest that the EEYP motif does not substantially impact subunit assembly.  98  2.3.5 The EE and YP amino-acid couplets have an additive effect on cell surface expression Up to this point, the EEYP motif was disrupted in its entirety to examine its role in HCN2 trafficking. To determine whether the complete EEYP was required, this motif was bisected and each side mutated to a pair of alanines (HCN2-AAYP and HCN2-EEAA). If density for these two mutant channels was compared with wild-type HCN2 and HCN2-4A. Representative current traces for the double alanine mutants are shown in Fig. 2.9A. Not surprisingly, the position of the activation curve was not affected by either couplet substitution (Fig. 2.9B). As seen in Fig. 2.9C, double alanine mutation of the first two amino acids (EE) reduces current by 40%, double alanine mutation of the second two amino acids (YP) reduces current by >60%, and replacement of all four amino acids by alanine reduces current by almost 90%. These data suggest that "YP" contributes to current levels to a greater extent than "EE" and that the combined effects of both couplets are almost completely additive. As a corollary to the above, we characterized HCN2-EEYM [identified as P619M by Akhavan et al. (1)]. Previously, this mutation was shown to reduce complex glycosylation, but its functional consequences were not assessed. These findings suggest that cell surface expression is reduced but do not discriminate between a reduction in functional channel formation versus a complete inability to fold and assemble. Similar to the couplet mutations, the activation curve was not significantly affected by the methionine substitution (Fig. 2.9B). The If density of HCN2-EEYM was significantly lower than wild type but comparable to the couplet mutant channels (Fig. 2.9D). These data show that the P619M mutation does not eliminate the ability of the channels to fold, assemble, and traffic to the cell surface and suggest specific contributions to HCN2 cell surface expression by the individual amino acids in the EEYP motif.  99  2.4 DISCUSSION The CNBD of HCN2 channels has two distinct functions: an inhibitory effect on channel opening and a facilitatory effect on cell surface expression. In this study, we have identified a four amino-acid motif (EEYP) in the B-helix of the CNBD that promotes the cell surface expression of functional channels but does not contribute to the CNBD-mediated inhibition of channel opening. Accordingly, the CNBD may be separated structurally into an a channel opening-inhibitory region and an expression-enhancing region. A separate region of the CNBD was delimited that is responsible for inhibition of channel opening, namely, the A-helix, β-barrel, and proximal B-helix. It will be important to determine the precise element required for this function. Inhibition involves interactions of the COOH terminus with the transmembrane domains (30, 31), but the exact molecular determinants and nature of the interactions are not known. Although the EEYP does not contribute to channel inhibition, small negative shifts in If activation were observed in cells expressing HCN2-4A, HCN2-AAYP and HCN2-EEAA, but not in cells expressing the truncated HCN2- EEYP-Cterm or HCN2-4A- C-term. These data suggest that the EEYP may influence opening of the channel by interacting with downstream elements of the COOH terminus. The C-helix, distal and adjacent to the B-helix, is a probable candidate for this interaction since it is required to transduce the actions of cAMP on channel opening (31). Important clues about the mechanism underlying enhancement of cell surface expression by the EEYP motif are presented. In this study, the observation that functional HCN2 channels can form in the absence of EEYP, albeit at low levels, shows that they are not irreversibly misfolded. Thus, this motif is not an absolute requirement for correct folding, assembly, and delivery of functional channels to the cell surface. Instead, the EEYP motif likely regulates the efficiency of cell surface expression. Unlike HCN2 NH2-terminal truncation mutants, the levels 100  of degradation of HCN2-4A, as determined using antibodies directed to the COOH terminus, were comparable to wild-type, which points toward one of two possibilities. First, the channels may be identified as misfolded and shunted toward a pathway in which degradation does not occur rapidly. To identify misfolded proteins, cells possess multiple folding "checkpoints" and pathways (2, 8, 9, 13), but only some of these are associated with their rapid clearance and degradation (10, 29). Thus, EEYP mutant channels may be misfolded but in a relatively stable state intracellularly. Second, correctly folded and assembled EEYP mutant channels may reside in an intracellular compartment in a relatively stable state (15, 17, 24) and thus may likewise not undergo rapid degradation. The EEYP may actively promote the movement of these correctly folded and assembled channels out of the ER or actively reduce channel retention in the ER, possibly by masking a retention signal. The lack of degradation of the EEYP mutants may be related to the observation that the EEYP motif does not impact the spectrum of oligomeric assembly of intracellular channels, but a determination of whether the intracellularly localized channels are folded and assembled correctly is required to discriminate between the two proposed possibilities. Because the EEYP mutants reduced but did not abolish functional expression, this motif is not the sole determinant of HCN2 cell surface expression. A proximal sequence in the CNBD and/or extant regions is likely of added importance. Interestingly, the primary structure of the entire B-helix region (VDNFNEVLEEYP) contains a number of acidic amino acids—an important feature of anterograde signals that facilitate forward trafficking but do not affect folding or assembly (6, 14, 15, 17, 24). It is also interesting that HCN4 but not HCN2 channels devoid of the CNBD are able to form functional channels in another mammalian cell line (HEK) (25). This is further evidence that regions other than the EEYP, which is conserved among all of  101  the mammalian isoforms, and the CNBD are involved in the regulation of HCN cell surface expression. Unlike the EEYP mutants and wild type channels, the NH2-terminal truncation mutants were subject to significant degradation when expressed in CHO cells, despite the fact that the levels of If for some of the truncation mutants were similar to those of the EEYP mutants. This suggests that the similar reductions in If produced by NH2-terminal truncations and EEYP mutants were the result of different processes. Previously, we proposed that the NH2 terminus plays a role in channel assembly similar to that of the NH2-terminal T1 domains in Shakerrelated voltage-gated potassium channels (20, 26). In Shaker channels, T1 domains facilitate tetramerization by bringing subunits into close proximity at the start of their assembly, which begins while still attached to ribosomes (12), and thereby increase the effective local concentration of compatible subunits (33). This is consistent with findings that show that T1deleted Shaker-related channels require high mRNA concentrations and prolonged times to form channels (11, 27). Taken together, the data suggest that HCN2 subunits lacking portions of the NH2 terminus are identified by a distinct mechanism and degraded during an inefficient folding and assembly process. A revealing finding in our study is that different defects within the same channel, which produce similar reductions in cell surface expression, lead to different end points in a given cell. This is consistent with previous studies that have demonstrated that the location of the lesion within a protein determines the predominant pathway used for protein disposal and that the pathway chosen depends on the ability of the protein to pass through several sequential checkpoints (29). This has important implications for disease-associated mutations in HCN channels and other channels of the voltage-gated potassium channel superfamily. A mutation in the COOH terminus has been proposed to disrupt cell surface expression of HCN4 and cause 102  sinoatrial arrhythmia (28). In addition, many disease-associated mutations in related channels such as the human ether-a-go-go, KCNQ1, and cyclic nucleotide-gated channels lead to disruption of cell surface expression (18, 19, 23). Our findings suggest that cell surface disruption may occur by mutation and defect-specific mechanisms, which then lead to consequences unique to those defects. Multiple mechanisms of protein disposal, and their diverse consequences, likely contribute to the variable and pleiotropic nature of disease-associated mutations with this superfamily of channels. An important future goal will be to elucidate the molecular and cellular details of these diverse mechanisms, and to relate these to specific mutations.  103  2.5 GRANTS These studies were generously supported by a Grant-in-Aid from the Heart and Stroke Foundation of British Columbia and the Yukon. H. Nazzari is the recipient of doctoral scholarships from the Michael Smith Foundation for Health Research (MSFHR) and the National Science and Engineering Research Council. G. Whitaker is the recipient of a Canada Graduate doctoral scholarship from the Canadian Institutes for Health Research (CIHR). V. Macri is the recipient of doctoral scholarships from the MSFHR and CIHR. E. A. Accili is the recipient of a Tier II Canada Research Chair.  104  2.6 ACKNOWLEDGEMENTS We thank Dr. A. Ludwig and Prof. M. Biel (University of Munich) for the HCN2 construct, Prof. H. Ohmori (Kyoto University) for the HCN4 construct, and Prof. M. Sanguinetti (University of Utah) for the HCN2-HA construct. Special thanks go to members of the Molday Lab, especially Laurie Molday, for assistance with sucrose gradient experiments. We also thank Heather Jackson and Chris Peters (Accili Lab) for comments on the manuscript, and Dr. Elizabeth Ross for both editing and comments. Finally, comments from anonymous reviewers were greatly appreciated.  105  Figure 2.1 The distal B-helix is conserved among ion channels but not protein kinases with similar cyclic nucleotide-binding domains (CNBDs). Alignment (by ClustalW 1.8) of the CNBDs of the cyclic nucleotide-gated channel A3 (CNGA3), hyperpolarization-activated cyclic nucleotide-gated channel 2 (HCN2), ether-a-go-go-related channel isoform 1 (ERG1), protein kinase RII, and protein kinase R1β from the mouse is shown. The gray arrows represent β-sheets and the curved lines represent -helices. Amino acids highlighted in black and gray represent complete and conserved identities, respectively. Note that conserved among the channel sequences only is the distal B-helix (-B, in red) motif of EEYP in HCN2, TEYP in CNGA3, and DMYP in ERG1 (indicated by the red bar).  106  Figure 2.2 Schematic representation of CNBD B-helix mutants. Top: schematic representation of wild-type HCN2 with the CNBD shown in bold-edged box. Subdomains of the CNBD are delineated and include the –A helix (A), β-sheet domain (β), -B helix (B), and -C helix (C). The B-helix is composed of 12 residues, with the EEYP shown in bold. Bottom 3 schematics: representations of the B-helix mutants, which include a mutant in which the EEYP has been substituted with 4 alanines, leaving the distal COOH terminus intact (HCN2-4A); an EEYP and distal COOH terminus deletion mutant (HCN2-EEYP-C-term); and this same mutant with 4 alanines in place of the EEYP motif (HCN2-4A-C-term).  107  Figure 2.3 The EEYP motif promotes cell surface expression of the mature form of HCN2. (Legend on the following page).  108  Figure 2.3 The EEYP motif promotes cell surface expression of the mature form of HCN2. A: Western blot probed with a rabbit polyclonal antibody specific for the mammalian HCN2 NH2 terminus. Lane 1, untransfected cells (UT); lane 2, HCN2; lane 3, HCN2-N380Q; lane 4, HCN2EEYP-C-term; lane 5, HCN2-4A-C-term; lane 6, HCN2-4A. The arrows identify the mature complex glycosylated (M, 136 kDa) and immature (I, 114 kDa) forms of the untruncated protein, as well as the immature (90 and 83 kDa) forms of the truncated proteins. The predicted molecular mass values for the unmodified proteins are as follows: 95 kDa for HCN2, HCN2N380Q, and HCN2-4A; 70 kDa for HCN2-EEYP-C-term and HCN2-4A-C-term. Only in the wild-type HCN2 channel is a band corresponding to the mature complex glycosylated form present. In all of the blots probed, a band corresponding to a mature form of the EEYP mutant channels was never observed, whereas that for the wild-type channel, from the same transfections, was always observed. B: images of nonpermeabilized (top) and permeabilized (bottom) Chinese hamster ovary (CHO) cells transfected with HCN2-hemagglutinin (HCN2-HA), HCN2-EEYP-C-term-HA, HCN2-4A-C-term-HA, and HCN2-4A-HA visualized with a mouse anti-HA primary antibody and a donkey anti-mouse Alexa 488 secondary antibody. Scale bars, 10 µm. All images were compared at the same exposure times and are representative of three independent experiments. C: Western blot of HCN2, HCN2-4A, and HCN2-N380Q from cells treated with proteinase K (PK) (+) or left untreated (–), probed with rabbit polyclonal antibody specific for the mammalian HCN2 COOH terminus. Relative to GAPDH, the total amounts of protein expressed (A and C) were comparable. Data shown in A and C are representative of at least three independent experiments.  109  Figure 2.4 The EEYP motif is not required to form functional channels. (Legend on the following page).  110  Figure 2.4 The EEYP motif is not required to form functional channels. A: current traces elicited by single voltage steps to –150 mV from a holding potential of –35 mV in cells expressing HCN2, HCN2-EEYP-C-term, HCN2-4A-C-term, or HCN2-4A. B: average hyperpolarization-activated current (If) densities in response to voltage pulses to –150 mV in cells expressing HCN2, HCN2-EEYP-C-term, HCN2-4A-C-term, or HCN2-4A. *Statistically significant difference from wild-type HCN2 (one-way ANOVA, followed by Tukey's test; P < 0.05). The numbers in parentheses above each bar represent the number of cells assayed in that group. C: Western blot of HCN2 and HCN2-4A, which was probed with a rabbit polyclonal specific for the mammalian HCN2 COOH terminus and in which the lanes were loaded with larger amounts of protein than in Fig.2.3, A and C (80 and 100 µg rather than 20 µg).  111  Figure 2.5 The EEYP motif does not contribute to inhibition of channel opening. (Legend on the following page). 112  Figure 2.5 The EEYP motif does not contribute to inhibition of channel opening. A: current traces recorded from relatively large CHO cells expressing HCN2, HCN2-EEYP-Cterm, HCN2-4A-C-term, or HCN2-4A elicited in response to 2-s voltage steps ranging from –50 mV to –150 mV from a holding potential of –35 mV (protocol is shown below current traces). B: If activation curves determined from the ratio of tail current amplitudes (I/Imax) elicited at the –65 mV voltage step (identified by arrows in Fig. 2.7A) in cells transfected with HCN2, HCN2EEYP-C-term, HCN2-4A-C-term, or HCN2-4A. Boltzmann fitting yielded V1/2 and k values of – 112.9 ± 2.5 mV and 11.8 ± 1.7 (n = 6 cells), –115.1 ± 5.5 mV and 11.5 ± 2.1 (n = 5 cells), –110.8 ± 1.9 mV and 7.7 ± 0.9 (n = 6 cells), and –124.2 ± 6.9 mV and 10.5 ± 1.4 (n = 5 cells), respectively. For V1/2 and k, values were not significantly different among the channels (one-way ANOVA, P > 0.05). C: average If densities of tail currents elicited by voltage step to –65 mV from a hyperpolarizing prepulse step to a fully activating voltage (–150 mV to –170 mV), from a holding potential of –35 mV. The numbers in parentheses above each bar represent the number of cells tested in that group. *Statistically significant difference from wild-type HCN2 (one-way ANOVA, followed by Tukey's test; P < 0.05).  113  Figure 2.6 The EEYP motif does not prevent channel degradation. (Legend on the following page).  114  Figure 2.6 The EEYP motif does not prevent channel degradation. A: representative current traces recorded from relatively large CHO cells expressing HCN2, HCN2-4A, or HCN2-2-137 elicited in response to 2-s voltage steps to –150 mV from a holding potential of –35 mV (protocol is shown below the current traces). B: average If densities in response to voltage pulses to –150 mV in cells expressing HCN2, HCN2-4A, or HCN2-2-137. The numbers in parentheses above each bar represent the number of cells tested in that group. *Statistically significant difference from cells expressing HCN2 (one-way ANOVA, followed by Tukey's test; P < 0.05). C: If activation curves determined from the ratio of tail current amplitudes elicited at the –65 mV voltage step (identified by arrows in A) in cells transfected with HCN2, HCN2-4A, or HCN2-2-137. Boltzmann fitting yielded V1/2 and k values of –110.7 ± 2.0 mV and 13.9 ± 1.2 (n = 6 cells), –124.2 ± 6.9 mV and 10.5 ± 1.4 (n = 5 cells), and –109.6 ± 1.9 mV and 10.8 ± 3.2 (n = 5 cells), respectively. For V and k, values were not significantly different among the channels (one-way ANOVA, P > 0.05). D: Western blot probed with a rabbit antibody specific for the mammalian HCN2 COOH terminus. lane 1, untransfected cells; lane 2, HCN2; lane 3, HCN2-N380Q; lane 4, HCN2-4A; lane 5, HCN2-2-137. The arrows indicate the presence of mature (136 kDa), immature (114 kDa for the three full-length proteins; 83 kDa for NH2-terminal deletion mutant), and degraded (D, 53 kDa) forms of the channel proteins. Note the presence of the degraded form in lanes 2–4 at much lower intensity and its absence in lane 1. The predicted molecular mass values for the unmodified proteins are as follows: 95 kDa for HCN2, HCN2-N380Q, and HCN2-4A; 81 kDa for HCN2-2-137. Note the absence of a complex glycosylated form in the mutant channels. Although a band is present at 83 kDa in all lanes, including the UT lane, its intensity was much greater in the HCN2-2-137 lane, consistent with the presence of the predicted full-length form of this truncated channel. GAPDH was used as a loading control. Relative to GAPDH, the total amounts of protein were not greatly affected. Data shown are representative of three independent experiments.  115  Figure 2.7 Enhanced degradation accompanies reduced If density upon NH2-terminal truncation and thus defines an intracellular fate distinct from that seen in the EEYP mutants. (Legend on the following page). 116  Figure 2.7 Enhanced degradation accompanies reduced If density upon NH2-terminal truncation and thus defines an intracellular fate distinct from that seen in the EEYP mutants. A: Western blot probed with a mouse antibody directed against the myc epitope at the COOH terminus of the wild-type and NH2-terminal deletion mutant proteins. Lane 1, untransfected cells; lane 2, HCN2-myc; lane 3, HCN2-2-130-myc; lane 4, HCN2-2-143-myc; lane 5, HCN2-2-154myc; lane 6, HCN2-2-182-myc. The arrows indicate the presence of mature (136 kDa for HCN2), immature (114 kDa for HCN2, and at 75 kDa or 80 kDa for the NH2-terminal deletion mutants), and degraded (59 kDa) protein forms. The predicted molecular mass values for the unmodified proteins are as follows: 96 kDa for HCN2-myc; 83 kDa for HCN2-2-130-myc; 82 kDa for HCN2-2-143-myc; 81 kDa for HCN2-2-154-myc; and 78 kDa for HCN2-2-182-myc. These data are representative of 6–8 independent experiments. B: average values for the ratio of band intensities of degraded protein to full-length immature protein (mature forms were not seen in the NH2-terminal mutants) produced by HCN2-myc, HCN2-2-130-myc, HCN2-2-143-myc, HCN2-2-154-myc, and HCN2-2-182-myc. Band intensities were determined directly from Western blots using densitometry (see MATERIALS AND METHODS). The numbers in parentheses above each bar refer to the number of separate transfections and Western blot experiments used in the analysis for that group. *Statistically significant difference from the ratio determined for HCN2 (one-way ANOVA followed by Tukey's test; P < 0.05). A monoclonal antibody against myc was used rather than the COOH-terminally directed HCN2 polyclonal antibody to avoid cross-reactivity with proteins endogenous to CHO cells seen in Fig. 2.8D. C: average If densities in response to voltage pulses to –150 mV in cells expressing HCN2-2-130, HCN2-2-137, HCN2-2-138, HCN2-2-143, HCN2-2-154, and HCN2-2-182. The latter three constructs produced no detectable If. The numbers in parentheses above each bar represent the number of cells tested in that group. Like HCN2-2-137, HCN2-2-130 and HCN2-2-138 produced low levels of If, and their activation curves were not different from wild-type HCN2 (data not shown).  117  Figure 2.8 HCN2-4A and wild-type channels assemble to the same extent. (Legend on the following page).  118  Figure 2.8 HCN2-4A and wild-type channels assemble to the same extent. A: Western blot of HCN2 and HCN2-4A probed with a rabbit polyclonal antibody directed to the COOH terminus, following sucrose gradient analysis as described in MATERIALS AND METHODS. Serially collected fractions from the sucrose gradient are numbered below the blot, beginning with the heaviest fractions. For this picture, the blot was processed as for those in the previous figures. M and I indicate the positions of the mature and immature forms of the protein, respectively. B: graph of normalized pixel intensity of the immature fractions versus fraction number from the data in A. Black line, HCN2; gray line, HCN2-4A. Gradient controls are as follows: A, thyroglobulin; B, alcohol dehydrogenase; C, bovine serum albumin. These data are representative of three independent experiments.  119  Figure 2.9 Individual elements within the EEYP contribute to its function. (Legend on the following page).  120  Figure 2.9 Individual elements within the EEYP contribute to its function. A: current traces recorded from relatively large CHO cells expressing HCN2-AAYP or HCN2EEAA elicited in response to 3-s voltage steps to –150 mV from a holding potential of –35 mV (protocol is shown to the right of the current traces). B: representative If activation curves generated from the ratio of tail current amplitudes elicited at the –65 mV voltage step (identified by arrows in A) in cells transfected with HCN2, HCN2-EEAA, HCN2-AAYP, and HCN2EEYM. Boltzmann fitting yielded V1/2 and k values of –110.7 ± 2.0 mV and 13.9 ± 1.2 (n = 6 cells), –119.8 ± 3.4 mV and 9.5 ± 0.5 (n = 8 cells), –120.6 ± 1.0 mV and 12.8 ± 1.0 (n = 8 cells), and –122.5 ± 4.2 mV and 10.0 ± 1.4 (n = 11 cells), respectively. For V1/2 and k, values were not significantly different from one another (one-way ANOVA, P > 0.05). C: average If current density in response to voltage pulses to –150 mV in cells expressing HCN2, HCN2-AAYP, HCN2-EEAA, or HCN2-4A. The numbers in parentheses above each bar represent the number of cells tested in that group. *Statistically significant difference between HCN2-AAYP and HCN2-4A. **Statistically significant difference between HCN2-EEAA and HCN2 and HCN24A. ***Statistically significant difference between HCN-4A and HCN2, HCN2-EEAA, and HCN2-AAYP (one-way ANOVA, followed by Tukey's test; P < 0.05). D: average If current density in response to voltage pulses to –150 mV in cells expressing HCN2 or HCN2-EEYM. The numbers in parentheses above each bar represent the number of cells tested in that group. *Statistically significant difference from HCN2 (one-way ANOVA, followed by Tukey's test, P < 0.05).  121  2.7 REFERENCES 1. Akhavan A, Atanasiu R, Noguchi T, Han W, Holder N, Shrier A. Identification of the cyclic-nucleotide-binding domain as a conserved determinant of ion-channel cell-surface localization. J Cell Sci 118: 2803–2812, 2005. 2. Arvan P, Zhao X, Ramos-Castaneda J, Chang A. Secretory pathway quality control operating in Golgi, plasmalemmal, and endosomal systems. Traffic 3: 771–780, 2002. 3. Berman HM, Ten Eyck LF, Goodsell DS, Haste NM, Kornev A, Taylor SS. The cAMP binding domain: an ancient signaling module. Proc Natl Acad Sci USA 102: 45–50, 2005. 4. Brodsky JL, McCracken AA. ER protein quality control and proteasome-mediated protein degradation. Semin Cell Dev Biol 10: 507–513, 1999. 5. Chen S, Wang J, Siegelbaum SA. Properties of hyperpolarization-activated pacemaker current defined by coassembly of HCN1 and HCN2 subunits and basal modulation by cyclic nucleotide. J Gen Physiol 117: 491–504, 2001. 6. Deutsch C. The birth of a channel. Neuron 40: 265–276, 2003. 7. Ellgaard L, Helenius A. Quality control in the endoplasmic reticulum. Nat Rev Mol Cell Biol 4: 181–191, 2003. 8. Fayadat L, Kopito RR. Recognition of a single transmembrane degron by sequential quality control checkpoints. Mol Biol Cell 14: 1268–1278, 2003. 9. Ferreira T, Mason AB, Pypaert M, Allen KE, Slayman CW. Quality control in the yeast secretory pathway: a misfolded PMA1 H+-ATPase reveals two checkpoints. J Biol Chem 277: 21027–21040, 2002. 10. Huyer G, Piluek WF, Fansler Z, Kreft SG, Hochstrasser M, Brodsky JL, Michaelis S. Distinct machinery is required in Saccharomyces cerevisiae for the endoplasmic reticulumassociated degradation of a multispanning membrane protein and a soluble luminal protein. J Biol Chem 279: 38369–38378, 2004. 11. Kobertz WR, Miller C. K+ channels lacking the ‘tetramerization’ domain: implications for pore structure. Nat Struct Biol 6: 1122–1125, 1999. 12. Lu J, Robinson JM, Edwards D, Deutsch C. T1-T1 interactions occur in ER membranes while nascent Kv peptides are still attached to ribosomes. Biochemistry 40: 10934–10946, 2001. 13. Luo WJ, Gong XH, Chang A. An ER membrane protein, Sop4, facilitates ER export of the yeast plasma membrane [H+]ATPase, Pma1. Traffic 3: 730–739, 2002. 14. Ma D, Jan LY. ER transport signals and trafficking of potassium channels and receptors. Curr Opin Neurobiol 12: 287–292, 2002. 122  15. Ma D, Zerangue N, Lin YF, Collins A, Yu M, Jan YN, Jan LY. Role of ER export signals in controlling surface potassium channel numbers. Science 291: 316–319, 2001. 16. Much B, Wahl-Schott C, Zong X, Schneider A, Baumann L, Moosmang S, Ludwig A, Biel M. Role of subunit heteromerization and N-linked glycosylation in the formation of functional hyperpolarization-activated cyclic nucleotide-gated channels. J Biol Chem 278: 43781–43786, 2003. 17. Nishimura N, Balch WE. A di-acidic signal required for selective export from the endoplasmic reticulum. Science 277: 556–558, 1997. 18. Patel KA, Bartoli KM, Fandino RA, Ngatchou AN, Woch G, Carey J, Tanaka JC. Transmembrane S1 mutations in CNGA3 from achromatopsia 2 patients cause loss of function and impaired cellular trafficking of the cone CNG channel. Invest Ophthalmol Vis Sci 46: 2282– 2290, 2005. 19. Peroz D, Rodriguez N, Choveau F, Baró I, Mérot J, Loussouarn G. Kv7.1 (KCNQ1) properties and channelopathies. J Physiol 586: 1785–1789, 2008. 20. Proenza C, Tran N, Angoli D, Zahynacz K, Balcar P, Accili EA. Different roles for the cyclic nucleotide binding domain and amino terminus in assembly and expression of hyperpolarization-activated, cyclic nucleotide-gated channels. J Biol Chem 277: 29634–29642, 2002. 21. Robinson RB, Siegelbaum SA. Hyperpolarization-activated cation currents: from molecules to physiological function. Annu Rev Physiol 65: 453–480, 2003. 22. Romisch K. Endoplasmic reticulum-associated degradation. Annu Rev Cell Dev Biol 21: 435–456, 2005. 23. Sanguinetti MC, Tristani-Firouzi M. hERG potassium channels and cardiac arrhythmia. Nature 440: 463–469, 2006. 24. Sevier CS, Weisz OA, Davis M, Machamer CE. Efficient export of the vesicular stomatitis virus G protein from the endoplasmic reticulum requires a signal in the cytoplasmic tail that includes both tyrosine-based and di-acidic motifs. Mol Biol Cell 11: 13–22, 2000. 25. Stieber J, Thomer A, Much B, Schneider A, Biel M, Hofmann F. Molecular basis for the different activation kinetics of the pacemaker channels HCN2 and HCN4. J Biol Chem 278: 33672–33680, 2003. 26. Tran N, Proenza C, Macri V, Petigara F, Sloan E, Samler S, Accili EA. A conserved domain in the NH2 terminus important for assembly and functional expression of pacemaker channels. J Biol Chem 277: 43588–43592, 2002. 27. Tu L, Santarelli V, Sheng Z, Skach W, Pain D, Deutsch C. Voltage-gated K+ channels contain multiple intersubunit association sites. J Biol Chem 271: 18904–18911, 1996. 123  28. Ueda K, Nakamura K, Hayashi T, Inagaki N, Takahashi M, Arimura T, Morita H, Higashiuesato Y, Hirano Y, Yasunami M, Takishita S, Yamashina A, Ohe T, Sunamori M, Hiraoka M, Kimura A. Functional characterization of a trafficking-defective HCN4 mutation, D553N, associated with cardiac arrhythmia. J Biol Chem 279: 27194–27198, 2004. 29. Vashist S, Ng DT. Misfolded proteins are sorted by a sequential checkpoint mechanism of ER quality control. J Cell Biol 165: 41–52, 2004. 30. Viscomi C, Altomare C, Bucchi A, Camatini E, Baruscotti M, Moroni A, DiFrancesco D. C terminus-mediated control of voltage and cAMP gating of hyperpolarization-activated cyclic nucleotide-gated channels. J Biol Chem 276: 29930–29934, 2001. 31. Wainger BJ, DeGennaro M, Santoro B, Siegelbaum SA, Tibbs GR. Molecular mechanism of cAMP modulation of HCN pacemaker channels. Nature 411: 805–810, 2001. 32. Zagotta WN, Olivier NB, Black KD, Young EC, Olson R, Gouaux E. Structural basis for modulation and agonist specificity of HCN pacemaker channels. Nature 425: 200–205, 2003. 33. Zerangue N, Jan YN, Jan LY. An artificial tetramerization domain restores efficient assembly of functional Shaker channels lacking T1. Proc Natl Acad Sci USA 97: 3591–3595, 2000.  124  CHAPTER 3: EVOLUTIONARY EMERGENCE OF N-GLYCOSYLATION AS A VARIABLE PROMOTER OF HCN CHANNEL SURFACE EXPRESSION  A version of this chapter has been accepted for publication. Hegle, A.P., Nazzari, H., Roth, A., Angoli, D., Accili, E.A. (2010) Evolutionary Emergence of N-Glycosylation as a Variable Promoter of HCN Channel Surface Expression 125  3.1 INTRODUCTION An outstanding question in HCN channel biology, and indeed in the biology of most intrinsic proteins of the plasma membrane, is how expression at the cell surface is regulated to affect functional heterogeneity in health and pathologic states. The number of HCN channels on the cell surface is a critical determinant of beating frequency in cardiac conduction tissue. HCN isoforms and HCN-mediated currents (Ih) are upregulated in the embryonic and neonatal ventricle (24, 45, 53, 56), in the neonatal sinoatrial node (2) and in beating embryonic stem cells during development in culture (27, 36, 40, 41). However, the factors that determine HCN channel supply at the cell surface during cardiac development have been studied in only a limited fashion. An important determinant of plasma membrane expression of intrinsic membrane proteins is N-linked glycosylation, which promotes proper folding, stability and oligomeric assembly in the endoplasmic reticulum and facilitates transport and targeting to the plasma membrane (14, 15). HCN1 and HCN2 are extensively N-glycosylated in mouse brain (31, 39, 58) and contain only one Asn-X-Ser/Thr consensus sequon for N-glycosylation in the S5 linker, close to the channel pore (Fig 3.1A, B). Mutation of the Asn to Gln in the mouse (m)HCN2 isoform (mHCN2-N380Q) appears to abolish functional expression in HEK 293 cells, unique in the voltage-gated potassium channel (Kv) superfamily (31). This abrogation of mHCN2 function in the absence of N-glycosylation may be due to impaired cell surface expression, as suggested by confocal imaging experiments that showed apparent low levels of GFP-tagged channels at the cell surface following treatment with the N-glycosylation inhibitor tunicamycin (31). However, there has been no direct or quantitative measurement of mHCN2 cell surface protein to ascertain if and to what extent it is influenced by N-glycosylation. Importantly, whether mHCN1 depends on N-glycosylation to the same extent as mHCN2 is also not known. 126  In this study, we were surprised to find that N-glycosylation is not required for mHCN2 function, but that it is nevertheless tightly coupled to mHCN2 cell surface expression. mHCN1 channels are much less dependent on N-glycosylation for cell surface expression, suggesting mHCN1 and mHCN2 genes followed diverse molecular trajectories after duplication from a common ancestor. We also found both channels in embryonic cardiac tissue, predominantly in a fully-N-glycosylated state as they are in the brain.  127  3.2 MATERIALS AND METHODS 3.2.1 Mutagenesis Asn to Gln mutations were engineered at position 327 in mHCN1 via Quickchange (Strategene) and at position 380 in mHCN2 via overlapping PCR. All mutations to Ala were made using Quickchange. For oocyte recordings, HCN2-WT and HCN2-N380Q were subcloned into the pBluescript SK+ vector using HindIII and XbaI. HA-tagged constructs were generated as previously described (53). All mutations were verified by sequencing. 3.2.2Western Blotting CHO cells (American Tissue Type Culture Collection) were maintained at a subconfluent density in F-12 media supplemented with 10% FBS. Transient transfection with FuGENE was performed according to manufacturer protocols (Roche). 24 hrs post-transfection, cells were washed with PBS and lysed for 30 min in RIPA buffer containing, in mM: 50 Tris (pH 8.0), 150 NaCl, 1 EDTA, 1 PMSF, 2 Na3VO4, 2 NaF, 1% NP40 and 10 µg/mL each of aprotinin, pepstatin, and leupeptin. Protein concentrations were determined by Bradford assay (BioRad). All PNGaseF assays were performed by incubating extracts with PNGaseF (New England Biolabs) in the presence of denaturing buffer and NP-40 for 2 hr at 37°C. Proteins were resolved using SDS-PAGE and transferred to polyvinylidene difluoride (PVDF) membranes. Blots were probed with antisera to HCN1 (1:200, NeuroMab), HCN2 (1:400, Alomone), c-myc (1:500, Invitrogen) or actin (1:3000, Invitrogen), followed by HRP-conjugated secondary antibody (1:2000, Santa Cruz), and visualized with ECL (Amersham). Scanned blots were cropped and adjusted for optimum brightness and contrast using Adobe Photoshop CS3. For experiments using rat heart lysates, freshly dissected hearts from embryonic day 18 rats were dissected in Tyrode solution containing, in mM: 140 NaCl, 5.4 KCl, 1.8 CaCl2, 1.0 128  MgCl2, 5 HEPES and 5.5 glucose (pH 7.4), and homogenized in RIPA buffer. For experiments using oocyte lysates, mature decollagenated oocytes were injected with 25 ng of the indicated RNAs and maintained in OR3 media at 18C. 48 hrs post-injection, 40-50 oocytes for each condition were homogenized on ice in HEDP buffer (100 mM HEPES, 1mM EDTA, pH 7.6) with protease inhibitors (10 µg/mL each of aprotinin, pepstatin, and leupeptin and 1 Roche protease inhibitor tablet per 10 mL) and centrifuged 2x at 6,000 RPM for 2 min at 4°C. Supernatants were then overlaid on a 15% sucrose / HEDP cushion and ultracentrifuged at 50,000 RPM for 90 min at 4°C to isolate membrane fractions. Pellets were resuspended in HEDP buffer and protein concentrations were determined by Bradford assay. 3.2.3Immunocytochemistry and microscopy HCN2 constructs containing an HA-epitope inserted between the third and fourth transmembrane segments were used to identify relative cell surface expression. 2-3 days after transfection, cells on coverslips were washed with phosphate-buffered saline (PBS) and fixed in 2% paraformaldehyde in PBS for 5 min. Thereafter, they were washed with PBS, either permeabilized using 0.2% Triton X-100 or left unpermeabilized, and blocked with 10% normal goat serum (NGS). After one wash with PBS containing 1% NGS, cells were incubated with a mouse monoclonal antibody specific to the HA-epitope (1:500, Sigma) for 1 hr at room temperature (RT). Cells were washed with PBS/NGS 1%, and then incubated with a goat antimouse secondary antibody tagged with Alexa 488 (1:1500, Invitrogen) for 1 hr at RT in the dark. After a final wash, the coverslips were mounted using Gelmount (Sigma) and sealed with clear nail polish. Imaging was performed using structured illumination (Zeiss Apotome Imager Z1) with a 63x oil immersion objective lens at wavelengths specific for the Alexa 488 fluorescent protein tag.  129  3.2.4 ELISA mHCN2 constructs were tagged with an HA epitope in the extracellular S3-S4 loop and transfected into CHO cells as described above. Cells were incubated with mouse anti-HA antibody (1:1000, Sigma) for 1 hr at RT followed by a goat anti-mouse HRP conjugated secondary antibody (1:1000, Santa Cruz) for 1 hr at RT. Cells were then incubated with a chemiluminescent HRP-substrate and luminescence values at 425 nm were quantified on a PerkinElmer Victor3V plate reader. For measurements of total luminescence, cells were permeabilized with 0.2% Triton-X-100. Background and untransfected luminescence values were subtracted for statistical comparisons using student's t-tests between constructs. Data are reported as mean ± SEM. 3.2.5 Electrophysiology and data analysis For two-electrode voltage clamp recordings, cRNA was synthesized using a mMessage mMachine T7 kit (Ambion). RNA concentrations were determined using spectrophotometry and gel electrophoresis. Mature Xenopus laevis oocytes were injected with 5-25 ng RNA and maintained in Barth's solution for 24 hrs prior to recording. Recording solution contained, in mM: 93 NaCl, 5 KCl, 2 MgCl2, 10 HEPES (pH 7.4, NaOH). For whole cell patch clamp recordings, CHO cells maintained on coverslips were co-transfected with 0.7 µg of pCDNA3-EGFP and 2 µg of the indicated constructs. 24-48 hrs after transfection, cells was transferred to a recording chamber and perfused with a low K+ extracellular solution containing, in mM: 5.4	
  KCl,	
  135	
  NaCl,	
   0.5	
  MgCl2,	
  1.9	
  CaCl2,	
  5	
  HEPES,	
  (pH	
  7.4,	
  NaOH).	
  Following	
  rupture	
  of	
  the	
  patch	
  membrane,	
   this	
  was	
  switched	
  to	
  a	
  high	
  K+	
  extracellular	
  solution	
  containing,	
  in	
  mM:	
  135	
  KCl,	
  5.4	
  NaCl,	
   0.5	
  MgCl2,	
  1.9	
  CaCl2,	
  5	
  HEPES	
  (pH	
  7.4,	
  KOH).	
  Fully	
  activated	
  current	
  protocols	
  were	
  carried	
   out	
  in	
  the	
  low	
  K+	
  extracellular	
  solution.	
  Briefly,	
  prepulses	
  to	
  -­‐150	
  mV	
  were	
  applied	
  for	
  1s	
   followed	
  by	
  pulses	
  to	
  test	
  voltages	
  ranging	
  from	
  -­‐120	
  mV	
  to	
  +30	
  mV.	
  The pipette solution 130  contained, in mM: 130 K-aspartate, 10 NaCl, 0.5 MgCl2, 5 HEPES, and 1 EGTA (pH 7.4, KOH). Pipette resistances ranged from 2-4 MΩ. spIH	
  currents	
  were	
  recorded	
  in	
  low	
  K+	
  extracellular	
   solution	
  with	
  saturating	
  (2mM)	
  levels	
  of	
  cAMP	
  added	
  to	
  the	
  pipette	
  solution.	
  Voltage dependence of activation was determined from tail currents at -35 mV following test pulses of varying lengths of time in order to reach steady state; interpulse intervals were followed by a 100 ms pulse to +5 mV to ensure complete channel deactivation. Resting current was always at baseline before subsequent voltage pulses. Normalized tail current amplitudes were plotted as a function of test potential, and fitted with a Boltzmann function, f(V) = Imax/(1 + e(V½-V)/k) to determine the midpoint of activation (V½) and slope factor (k). Activation at -150 mV was fitted with a double exponential; deactivation at -35 mV was fitted with a single exponential. Statistical comparisons were performed using student's t-test or one-way ANOVA followed by Tukey's post-hoc analysis. Data are reported as mean ± SEM. 3.2.6 HCN sequence collection and analysis With the exceptions of Ciona intestinalis sequences, all DNA sequences were retrieved from the ENSEMBL (v55), UCSC Genome Browser, JGI Genome Portal and NCBI Entrez nonredundant nucleotide databases (8, 18, 20, 21). Ciona sequences have been previously annotated (19). (See Supplementary Table for detailed sequence information). Translated protein sequences were aligned with MCoffee (50); these were then used to align the nucleotide sequences by codon. C- and N-terminal regions were subsequently removed from the aligned nucleotide sequences, leaving the S1 through cyclic-nucleotide binding domain (cNBD) region. Predicted extracellular domains, based on annotations in the UniProt database (1), were individually scanned for the Asn-X-Ser/Thr motif, in which X can be any amino acid except Pro. All figures were prepared using Adobe Illustrator CS3. 131  3.3 RESULTS 3.3.1 HCN2 channels undergo N-glycosylation in native cardiac tissue Both HCN1 and HCN2 are predominantly N-glycosylated in brain tissue (31, 39, 58), but whether these channels are N-glycosylated in cardiac tissue is unknown. HCN1 and HCN2 are expressed in conduction tissue of the adult heart (28-30, 43), and are found in various regions of embryonic and neonatal hearts (45, 56) and in cardiac embryonic stem cells (36, 40, 41, 55). Studies from our lab and others have shown that HCN2 contributes to the generation and modulation of heart rate in the embryonic stage (16, 24, 26, 36, 45, 53, 56), and knockout of HCN2 leads to significant reductions in If amplitudes recorded from individual embryonic cardiac myocytes (26). Therefore, we probed whole heart lysates from embryonic day 18 (E18) rats with an antibody directed to an N-terminal epitope and examined N-glycosylation by treating lysates with peptide N-glycosidase F (PNGase F), an enzyme that cleaves all N-linked glycans. Figure 3.1C shows a single band corresponding to HCN2 near 100 kD. In the presence of PNGaseF, the HCN2 band is completely shifted to ~90 kD, suggesting that cardiac HCN2 channels are fully N-glycosylated in vivo. These molecular weights are consistent with a nonproteolyzed form of HCN2 as has been reported for rat and human ventricles (17, 36, 46, 57, 59). To compare N-glycosylation of HCN2 in heart with that of heterologously expressed HCN2, we transfected mouse (m)HCN2 into CHO cells and performed Western blots as was done with heart lysates. We also expressed a mHCN2 channel in which the predicted Asn was substituted by Gln, for direct comparison with wild type mHCN2 in the presence or absence of PNGaseF (Fig 3.1D). Western blots of CHO cell extracts expressing wild type mHCN2 reveal two distinct molecular forms of the channel: a PNGaseF-insensitive band near 105 kD, likely corresponding to immature, non-glycosylated channels, and a PNGaseF-sensitive band near 125 kD, indicating N-glycosylated channels (Fig 3.1D). Notably, the shift in molecular mass of the 132  upper band following PNGaseF treatment (~20 kD) was larger than that observed in heart lysates (~10 kD), possibly reflecting a higher level of complexity and/or difference in structure of the Nglycans attached to the channel when heterologously expressed in CHO cells. This difference was not unexpected, as the N-glycosylation of recombinant proteins in mammalian cell systems is highly sensitive to culture conditions and often varies between cell lines (4). As shown previously in HEK cells, mHCN2-N380Q is not sensitive to PNGaseF and is comparable in mass to the PNGaseF-treated wild type mHCN2. Although we sometimes noticed an apparent reduction in mHCN2-N380Q band intensity, this was not a consistent finding among all blots. 3.3.2 N-glycosylation is required for mHCN2 function in multiple expression systems No Kv isoforms examined to date require N-glycosylation to fold correctly and form functional channels (5-7, 9, 13, 23, 31, 32, 34, 37, 38, 48, 51, 52, 54). In contrast, mHCN2N380Q reportedly fails to produce Ih when transfected in HEK 293 cells (31). Nevertheless, Nglycosylation mutants of other channels are sometimes functional only in certain cell lines or expression systems. Therefore, we expressed mHCN2-N380Q alongside the wild type channel in Xenopus oocytes. This system was chosen for two reasons: first, non-glycosylated Shaker, the prototypical channel in the Kv superfamily, trafficks to the plasma membrane much more efficiently in oocytes than in HEK cells (6, 38). Second, certain HCN2 trafficking mutants that do not express functionally in mammalian cells (33, 35), do so in oocytes (49). For comparison, we also expressed mHCN2-N380Q or wild type mHCN2 in CHO cells. Although wild type mHCN2 produced robust Ih (~3-6 µA) after injections of 5 ng RNA, we consistently found no detectable levels of this current for mHCN2-N380Q over several individual rounds of RNA synthesis, using injections of 5-25 ng (Fig 3.2A). Severe reductions in current were also found in recordings from CHO cells (Fig 3.2B). Western blots of oocyte membrane extracts (Fig 3.2C) show that mHCN2-N380Q protein is synthesized, but, as in CHO 133  cells, the level was sometimes reduced compared to wild type mHCN2. These blots also show that membrane-associated wild type mHCN2 is sensitive to PNGaseF, yielding a band at the same position as mHCN2-N380Q, as was expected. The shift in molecular mass conferred by the absence of N-glycosylation in oocytes was more subtle than in hearts or CHO cells, again suggesting differences in the composition and/or size of the N-glycans among expression systems. Taken together, these data strongly suggest that N-glycosylation promotes the functional expression of mHCN2, either by enhancing function or by increasing the number of channels at the cell surface. 3.3.2 N-glycosylation promotes mHCN2 cell surface expression An outstanding question arising from the above results, and from previous work in HEK cells (31), is whether the increase in mHCN2 functional expression by N-glycosylation is due to an enhancement of function or to an increase in the cell surface expression of channel protein. To answer this question, we performed immunolabeling and imaging experiments to compare surface and total protein distribution of wild type mHCN2 and mHCN2-N380Q. The channels were tagged with an HA epitope in the extracellular S3-S4 loop and expressed in CHO cells in the presence or absence of Triton-X-100 to permeabilize cells. In the absence of Triton X-100, only channels localized to the cell surface could be visualized. In these non-permeabilized cells (Fig. 3.3A, upper panels), an antibody to the HA tag reveals that wild type mHCN2 is strongly expressed at the cell surface, whereas mHCN2-N380Q is only minimally present. In permeabilized cells (Fig. 3.3A, lower panels), wild type mHCN2 and mHCN2-N380Q were visible throughout the cytoplasm as evenly spread puncta. This pattern of intracellular protein distribution was not notably different between CHO cells expressing wild type mHCN2 and mHCN2-N380Q.  134  To confirm that the reduction in cell surface expression of mHCN2-N380Q was due to disruption of N-glycosylation and not a result of a sensitivity of the region to the Gln substitution, we created additional constructs individually substituting each residue in the Asn-X-Ser sequon to Ala, generating mHCN2-N380A, mHCN2-H381A and mHCN2-S382A. Previous studies with the rabies virus glycoprotein have shown that N-glycan attachment is sensitive to single Ala substitutions in the Asn and Ser/Thr residues of the sequon, but that the efficiency of Nglycosylation is close to maximal when X is either His or Ala (22, 42). Therefore, we expected that mHCN2-N380A and mHCN2-S382A, like mHCN2-N380Q, would have limited Nglycosylation and cell surface expression, whereas mHCN2-H381A would present a pattern not unlike the wild type channel. These expectations were confirmed using Western blotting and immunolabeling combined with imaging to examine N-glycosylation and cell surface expression (Fig 3.3A, E). 3.3.3 The effect of N-glycosylation on cell surface expression correlates only partly with an increase in total protein To confirm and quantify the effect of N-glycosylation on cell surface mHCN2 and compare this with changes in total protein, we used a modified enzyme-linked immunosorbant assay (ELISA) to detect HA-tagged channels on the plasma membrane of transfected CHO cells. Figure 3.3B (left) shows that the surface expression of mHCN2-N380Q is significantly decreased compared to wild type mHCN2. Consistent with the importance of Asn and Ser residues for N-glycosylation efficiency, cell surface expression is low, but not abolished, for both mHCN2-N380A and mHCN2-S382A (Fig 3.3B, left), and comparable to that of an Nterminal-truncated mHCN2 that poorly localizes to the cell surface (33, 47). As predicted, mHCN2-H381A surface expression was not reduced, and was in fact significantly greater than that of wild type mHCN2 (Fig 3.3B, left). 135  Differences in protein expression among mHCN2 constructs were also observed in ELISA measurements of total protein in permeabilized cells (Fig 3.3B, center). Compared to wild type mHCN2, total protein was significantly decreased for all non-glycosylated mutants, and increased for mHCN2-H381A. To determine whether differences in total protein levels can account for the changes observed at the cell surface, we calculated the ratio of surface to total luminescence values for each construct (Fig 3.3B, right). Compared to wild type mHCN2, the surface fraction of non-glycosylated mutants was reduced by approximately 50%, indicating that overall decreases in protein levels cannot fully explain the lower number of channels at the cell surface. In contrast, the surface fraction of mHCN2-H381A was similar to wild type mHCN2, suggesting that the increased surface expression of this mutant corresponds to an overall increase in protein. 3.3.4 HCN2 channels lacking N-glycosylation form functional channels An important and broad conclusion made in a previous study of mHCN2 in HEK cells was that N-glycosylation is required to obtain functional HCN channels. However, in our ELISA experiments, we noted that cell surface expression of mHCN2 was not eliminated by disruption of N-glycosylation in CHO cells. To determine whether the surface-localized mutant channels were functional, we performed whole-cell patch clamp experiments in CHO cells (Fig 3.3C). The cells transfected with mHCN2-N380A, mHCN2-S382A or mHCN2-N380Q were very fragile, and most often could not be reliably hyperpolarized beyond -100 mV. In cells expressing these channels, Ih was never detectable at voltages less negative than -100 mV (data not shown). However, in a small number of cells, hyperpolarizing pulses to voltages more negative than -100 mV were tolerated and gave rise to a current with clear features of Ih. Because such a slowlyactivating inward current in response to hyperpolarization of the membrane potential was never observed in untransfected or mock-transfected CHO cells, we are confident that it comes from 136  the mutant mHCN2 subunits. The low level of current expression and the fragility of transfected cells precluded a more comprehensive functional analysis; therefore, it is not clear whether the lack of N-glycosylation affects channel function. Nevertheless, the identification of a distinct Ih, even if only in a small number of cells, is still convincing evidence that N-glycosylation is not required for correctly folded and functioning mHCN2 channels. A strong influence of Nglycosylation on cell surface expression, as shown in Figure 3.3A, is the best explanation for the corresponding decrease in current density found here and previously (31). In contrast, the current density of mHCN2-H381A was significantly larger than that of wild-type mHCN2 (Fig 3.3C, D), consistent with the greater overall cell surface expression observed for this mutant in ELISA experiments. The voltage dependence of activation did not noticeably differ between this mutant and the wild type: half activation values were -113.0 mV +/- 4.848 (n=3) and -113.3 mV +/- 10.32 (n=3) for mHCN2-WT and mHCN2-H381A, respectively, implying no change in function. These data again suggest that the reduction in cell surface expression is due to disruption of N-glycosylation and not a consequence of mutation of the sequon. 3.3.5 N-glycosylation of HCN channels emerged in an ancestor common to chordates to promote cell surface expression To date, a functional analysis of putative N-glycosylation sequons has been carried out only with mHCN2, but all four mammalian isoforms contain the pore-associated sequon. Are they all as dependent on N-glycosylation of this sequon for cell surface expression? Evidence of HCN evolution suggests that they are not. We have previously shown that, unlike most chordate and urochordate HCN sequences (19), known invertebrate sequences do not have the poreassociated sequon. Yet the invertebrate channels that have been cloned, which include those from Drosophila melanogaster, Apis mellifera, Panulirus argus and Strongylocentrotus 137  purpuratus (spIH), all produce Ih in HEK cells (10-12, 44). Together these data suggest that an ancestor common to all chordates first acquired the pore-associated sequon but that glycosylation at this site was not required for cell surface expression. Therefore, we might expect that one or more of vertebrate HCN isoforms that arose from this common ancestor also does not require Nglycosylation of the pore-associated sequon for efficient cell surface expression. A weakness of our previous analysis is that it was based on a limited representation of major groups of organisms. Fortunately, the sequencing of genomes is continuing at an accelerated rate, with many new HCN sequences becoming available from a much broader range of major invertebrate and vertebrate lineages. In order to more stringently test our hypothesis that the pore-associated sequon arose just prior to the chordate/urochordate lineage, we took advantage of the newly available sequences and performed a more extensive analysis in which the representation of major metazoan groups has been expanded. Importantly, cnidarian, mollusc and annelid sequences have been added to our list of invertebrate sequences, while additional amphibian, fish and mammalian sequences absent from the original analysis have now also been analyzed. (19). The Supplementary Figure shows a representative alignment of the S5 to pore region of 61 metazoan HCN sequences. Only chordate and urochordate HCN sequences contain the pore-associated sequon. These findings, summarized in Figure 3.4, strongly support our hypothesis that N-glycosylation at this site is not needed for surface localization of invertebrate HCN channels (10-12, 44). Although invertebrate HCN sequences do not possess the pore-associated Asn-X-Ser/Thr sequon, none have been explicitly tested for N-glycan addition. To confirm that an invertebrate HCN isoform is both non-glycosylated and functional in our system, we expressed the sea urchin HCN channel (spIH) in CHO cells. Western blot analysis of spIH-transfected CHO cells reveals a PNGaseF-insensitive band at ~110 kD, confirming that spIH is not N-glycosylated (Fig 3.4B). 138  Nevertheless, this channel produces robust currents (Fig 3.4C), as it does in HEK cells (10, 44), indicating that N-glycosylation is not required for surface localization of functional channels. 3.3.7 N-glycosylation is not required for efficient mHCN1 cell surface expression and function Because spIH and other invertebrate channels form functional channels at the plasma membrane, it seems plausible that an ancestor common to both invertebrate and mammalian channels but lacking the pore-associated sequon was also able to do so. This line of reasoning led us to begin examining other mammalian HCN isoforms, with the idea that one of them may have retained the ability to traffick and function without N-glycan addition. To this end, we tested the mHCN1 isoform, which undergoes N-glycosylation in the brain (31, 39, 58). Re-probing the E18 rat heart extracts (Fig 3.1C) for HCN1 using a C-terminal antibody shows that it is also present predominantly as a ~100 kD N-glycosylated protein in embryonic cardiac tissue (Fig 3.5A). To verify that mHCN1 is N-glycosylated at the predicted pore-associated sequon, we mutated the conserved Asn to Gln. Western blots of CHO cell extracts expressing wild type mHCN1 reveal two bands, at approximately 110 kD and 120 kD, whereas mHCN1-N327Q produced only a single band near 110 kD (Fig. 3.5B). In addition to being insensitive to PNGaseF and comparable in mass to the lower band of the PNGaseF-treated wild type channel, the mHCN1-N327Q bands are less dense overall (Fig 3.5B), suggesting that, as was observed for non-glycosylated mHCN2 mutant channels, the total protein level is reduced. N-glycosylation at N327 in wild type mHCN1 is strongly supported by these data. In stark contrast to non-glycosylated mHCN2 mutants, expression of mHCN1-N327Q in CHO cells results in robust Ih; current density is reduced by only 50% relative to wild type mHCN1 (Fig 3.5C, D). Moreover, cells expressing mHCN1-N327Q did not exhibit the fragility caused by non-glycosylated mHCN2 mutants. Together, these data strongly suggest isoformspecific sensitivity to N-glycosylation in the trafficking of mHCN1 and mHCN2 to the cell 139  surface, in which the expression of mHCN1 is less tightly coupled to N-glycosylation than mHCN2. Because the conserved N-glycosylation site is located in the outer portion of the pore and very near the selectivity filter in mammalian HCN channels, N-glycosylation could directly affect channel function. However, a comparison of reversal potentials between mHCN1-N327Q and wild type mHCN1 using physiological ionic concentrations shows no significant difference (Fig 3.6A), suggesting that selectivity and permeation are not altered by N-glycosylation despite the proximity of the sequon to the outer pore. We also found that the V1/2 of Ih activation is not significantly different between wild type mHCN1 and mHCN1-N327Q (Fig 3.6B), although a small, but significant, difference in slope was observed (Fig 3.6C). Thus, the decrease in Ih density produced by the N327Q mutation, determined at -150 mV, is not due to a negative shift in the activation curve and a corresponding reduction of channel availability at that voltage. Moreover, analysis of channel kinetics reveals that the rates of slow and fast channel activation, as well as that of deactivation, are all unaffected in mHCN1-N327Q (Fig. 3.6D). Because N327Q abolishes N-glycosylation completely (Fig. 3.5), these findings suggest that N-glycan addition does not greatly affect mHCN1 function.  140  3.4 DISCUSSION Previous studies have suggested that N-glycosylation is required for HCN channel function based only on experiments using the mHCN2 isoform. However, the absence of Nglycosylation in invertebrate isoforms, together with their ability to form functional channels in mammalian cells, suggested to us that the influence of N-glycosylation might have evolved in a variable manner among vertebrates such that, following gene duplication, not all of the resulting HCN isoforms became as heavily dependent on N-glycosylation for function. This was confirmed here, where we show that both mHCN1 and mHCN2 are N-glycosylated at a site between S5 and the pore and that, surprisingly, neither requires this posttranslational modification for function. Notably, we do find that mHCN1, unlike mHCN2, is robustly expressed at the plasma membrane without N-glycan addition. These findings are consistent with an evolutionary trajectory whereby a HCN common ancestor acquired the S5-pore Nglycosylation site, and vertebrate isoforms later accumulated amino acid changes that variably altered dependence on N-glycosylation for cell surface expression. Hence, mHCN2 may have become more reliant on N-glycans for surface expression over the course of evolution, whereas mHCN1 remained less so. Although we show that cell surface expression of mHCN2 is more dependent on Nglycosylation than mHCN1, the molecular basis underlying this dependence and the difference between the isoforms remains to be established. Our data suggest that factors independent of total protein levels contribute to the maintenance of cell surface expression. In general, Nglycosylation yields glycoproteins that are recognized by ER chaperones during folding. Once the native conformation is attained, the glycoprotein is released and proceeds to the Golgi; misfolded proteins are retained in the ER and may be targeted for degradation (3, 14). Future experiments will examine the role of N-glycosylation on the interaction of HCN channels with 141  chaperones and their relative stability in the cell, and identify the regions that are responsible for the differences among HCN isoforms. A novel finding in this paper was the apparent lack of strong effects of N-glycosylation on mHCN1 channel function. Despite the proximity of the sequon to the selectivity filter and outer pore, the N327Q mutation had no discernible effects upon mHCN1 Ih activation and deactivation kinetics, or on selectivity and permeation (Fig 3.6). Likewise, the position of the Ih activation curve does not change, although the slope is somewhat steepened. Thus, our data suggest that N-glycosylation only minimally impact mHCN1 function. We found that both HCN1 and HCN2 are present in embryonic cardiac tissue in significant amounts, consistent with previous reports. Both isoforms are present in predominantly N-glycosylated and non-proteolyzed forms, although the nature of N-glycan attachment may be different from that found in other cell types such as CHO cells. Together, our data suggest that cell surface expression of HCN1 in native tissue may also be less dependent on the extent of N-glycosylation as compared to HCN2. It will be important to determine how tightly expression of the various HCN isoforms is coupled to N-glycosylation in vivo, and whether this coupling is regulated to affect excitability in cardiac myocytes and neurons.  142  3.5 ACKNOWLEDGEMENTS We would like to thank Prof. Martin Biel (University of Munich) and Dr. Andreas Ludwig (University of Erlangen) for the mHCN1 and mHCN2 constructs, Dr. Gary Yellen (Harvard University) for the spIH construct, Dr. Michael Sanguinetti (University of Utah) for the HA-tagged mHCN2 construct, and Dr. David Steele (University of British Columbia) for helpful comments. These studies were generously supported by a Grant-in-Aid from the Heart and Stroke Foundation of British Columbia & the Yukon (EAA). EAA is the recipient of a Tier II Canada Research Chair. APH is supported by a Heart and Stroke Foundation of Canada Research Fellowship. HN is supported by the National Sciences and Engineering Research Council of Canada and the Michael Smith Foundation for Health Research. AR is supported by a National Sciences and Engineering Research Council Undergraduate Award.  143  Figure 3.1 Mouse (m) hyperpolarization-activated cyclic nucleotide-modulated (HCN)2 channels undergo N-glycosylation in native cardiac tissue. A: predicted membrane topology of an HCN channel subunit. N-glycosylation at the S5-pore linker is represented as a branched chain of core N-glycans. S1-S6, transmembrane domains; P, pore domain; cNBD, cyclic nucleotide binding domain; Cyto, cytoplasm. B: amino acid sequence alignment of the S5-pore regions of the 4 mHCN isoforms and the sea urchin HCN homolog (spIH). The Asn-X-Ser N-glycosylation sequon is shaded; note the absence of this motif in spIH. The selectivity filter sequence, CIGYG, is also shaded for reference. C: Western blot of embryonic day (E)18 rat heart lysates in the presence or absence of peptide N-glycosidase (PNGaseF) to cleave N-glycans. HCN2 is shifted to a lower molecular mass in the presence of PNGaseF. D: Western blot of Chinese hamster ovary (CHO) cell extracts expressing wild-type (WT) mHCN2 or mHCN2-N380Q in the presence or absence of PNGaseF. Note that mHCN2N380Q does not produce a mature band (arrow). Similar results were obtained from at least 4 separate transfections.  144  Figure 3.2 N-glycosylation is required for mHCN2 function in multiple expression systems. (Legend on the following page).  145  Figure 3.2 N-glycosylation is required for mHCN2 function in multiple expression systems. A, left: representative mHCN2-WT current traces obtained from 2-electrode recordings in Xenopus oocytes; n = 20 from 4 separate RNA batches. Right: mHCN2 current was absent from oocytes expressing mHCN2-N380Q; n = 24 from 4 separate RNA batches. B, left: representative mHCN2-WT current traces obtained from whole cell patch-clamp recordings of transfected CHO cells. Right: mHCN2 current is dramatically reduced in CHO cells expressing mHCN2N380Q. Pulse protocols are indicated below each trace. Dashed line represents zero current levels. Similar results were obtained from at least 8 separate transfections of each construct. C: Western blot of oocyte membrane fractions expressing mHCN2-WT and mHCN2-N380Q in the presence or absence of PNGaseF. Note that mHCN2 is present primarily in a single band that shifts to a lower molecular mass in the presence of PNGaseF. mHCN2-N380Q is present at the lower mass and is insensitive to PNGaseF.  146  Figure 3.3 N-glycosylation promotes mHCN2 cell surface expression. (Legend on the following page). 147  Figure 3.3 N-glycosylation promotes mHCN2 cell surface expression. A: images showing immunolocalization of mHCN2 constructs in CHO cells. Channels were hemagglutinin (HA) tagged at the extracellular S3-S4 loop and fluorescently labeled in the presence or absence of Triton X-100 (TX-100) to permeabilize cells. Nonglycosylated channels do not reach the plasma membrane but are distributed similarly to mHCN2-WT in the cytoplasm of permeabilized cells. Exposure time was kept constant among all constructs for each condition. B, left and center: surface expression and total expression, respectively, of HA-tagged mHCN2 constructs in CHO cells. Cells expressing mHCN2 channels with disrupted N-glycosylation have greatly reduced luminescence compared with those expressing mHCN2-WT; n = 4. Right: ratio of surface to total luminescence for cells expressing each construct. *P < 0.05. C: representative HCNmediated current (Ih) traces (leak subtracted) recorded from CHO cells expressing WT and mutant mHCN2 channels. Note the difference in scale for mHCN2-S382A. D: comparison of Ih density of mHCN2-WT with mHCN2-H381A in CHO cells. *P = 0.017. Ih density for mHCN2N380Q, mHCN2-N380A, and mHCN2-S382A was most often negligible. Numbers of cells are indicated above each bar. E: Western blot of mHCN2 constructs expressed in CHO cells. Note the increased mature band for mHCN2-H381A compared with that of mHCN2-WT. Similar results were observed for at least 3 separate transfections.  148  Figure 3.4 N-glycosylation of HCN channels emerged in an ancestor common to chordates. A: cladogram of metazoans showing the emergence of N-glycosylation prior to the divergence of chordate (vertebrate) and urochordate (tunicate) HCN lineages. Branches corresponding to sequences that contain the N-glycosylation consensus sequence are designated with solid lines; those lacking the consensus sequence are designated with dashed lines. Adapted from the Tree of Life web project (25). B: Western blot of CHO cells expressing WT spIH in the presence or absence of PNGaseF, which does not affect molecular mass. C: representative spIH current traces obtained in whole cell patch-clamp recordings of transfected CHO cells according to the protocol shown. Dashed line represents zero current levels. Similar results were obtained in 6 cells. 149  Figure 3.5 N-glycosylation is not required for efficient mHCN1 cell surface expression. A: Western blot of E18 rat heart lysates in the presence or absence of PNGaseF to cleave Nglycans. HCN1 is shifted to a lower molecular mass in the presence of PNGaseF. B: Western blot of CHO cell extracts expressing mHCN1-WT or mHCN1-N327Q in the presence or absence of PNGaseF. Note that mHCN1-N327Q does not produce a mature band (arrow). Similar results were obtained from at least 4 separate transfections. C: representative Ih traces obtained from CHO cells expressing mHCN1-WT and mHCN1-N327Q. A staggered pulse protocol was used to allow the currents to approach steady state at each potential. Dashed line represents zero current levels. Similar results were obtained from at least 8 separate transfections of each construct. D: Ih density of mHCN1-N327Q compared with mHCN1-WT. *P = 0.0071. Numbers of cells are indicated above each bar. 150  Figure 3.6 Disruption of N-glycosylation minimally impacts mHCN1 function. A: plots of fully activated Ih vs. test voltage obtained from CHO cells expressing mHCN1 as described in the text and mHCN1-N327Q. Reversal potentials were –29.2 ± 3.7 mV and –22.8 ± 4.3 mV, respectively, and did not significantly differ; P = 0.34. B: activation curves obtained by fitting plots of normalized tail current (I/Imax) amplitudes vs. voltage with a Boltzmann equation. Midpoint of activation (V1/2) values were –98.7 ± 3.1 mV and –93.3 ± 3.0 mV for mHCN1-WT and mHCN1-N327Q, respectively, and did not significantly differ; P = 0.58. C: comparison of slope factor of mHCN1-N327Q to mHCN1-WT. *P = 0.049. D, left: comparison of fast and slow components of activation at –150 mV between mHCN1 and mHCN1-N327Q. *P = 0.15 and 0.092, respectively. Right: comparison of deactivation at –35 mV between mHCN1 and mHCN1-N327Q. *P = 0.81. Numbers of cells are indicated in parentheses. 151  3.6 REFERENCES 1.  The universal protein resource (UniProt). Nucleic acids research 36: D190-195, 2008.  2. Accili EA, Robinson RB, and DiFrancesco D. Properties and modulation of If in newborn versus adult cardiac SA node. The American journal of physiology 272: H1549-1552, 1997. 3. Burda P, and Aebi M. The dolichol pathway of N-linked glycosylation. Biochimica et biophysica acta 1426: 239-257, 1999. 4. Butler M. Optimisation of the cellular metabolism of glycosylation for recombinant proteins produced by Mammalian cell systems. Cytotechnology 50: 57-76, 2006. 5. Cartwright TA, Corey MJ, and Schwalbe RA. Complex oligosaccharides are N-linked to Kv3 voltage-gated K+ channels in rat brain. Biochimica et biophysica acta 1770: 666-671, 2007. 6. de Souza NF, and Simon SM. Glycosylation affects the rate of traffic of the Shaker potassium channel through the secretory pathway. Biochemistry 41: 11351-11361, 2002. 7. Erler I, Al-Ansary DM, Wissenbach U, Wagner TF, Flockerzi V, and Niemeyer BA. Trafficking and assembly of the cold-sensitive TRPM8 channel. The Journal of biological chemistry 281: 38396-38404, 2006. 8. Flicek P, Aken BL, Beal K, Ballester B, Caccamo M, Chen Y, Clarke L, Coates G, Cunningham F, Cutts T, Down T, Dyer SC, Eyre T, Fitzgerald S, Fernandez-Banet J, Graf S, Haider S, Hammond M, Holland R, Howe KL, Howe K, Johnson N, Jenkinson A, Kahari A, Keefe D, Kokocinski F, Kulesha E, Lawson D, Longden I, Megy K, Meidl P, Overduin B, Parker A, Pritchard B, Prlic A, Rice S, Rios D, Schuster M, Sealy I, Slater G, Smedley D, Spudich G, Trevanion S, Vilella AJ, Vogel J, White S, Wood M, Birney E, Cox T, Curwen V, Durbin R, Fernandez-Suarez XM, Herrero J, Hubbard TJ, Kasprzyk A, Proctor G, Smith J, Ureta-Vidal A, and Searle S. Ensembl 2008. Nucleic acids research 36: D707-714, 2008. 9. Freeman LC, Lippold JJ, and Mitchell KE. Glycosylation influences gating and pH sensitivity of I(sK). The Journal of membrane biology 177: 65-79, 2000. 10. Gauss R, Seifert R, and Kaupp UB. Molecular identification of a hyperpolarizationactivated channel in sea urchin sperm. Nature 393: 583-587, 1998. 11. Gisselmann G, Gamerschlag B, Sonnenfeld R, Marx T, Neuhaus EM, Wetzel CH, and Hatt H. Variants of the Drosophila melanogaster Ih-channel are generated by different splicing. Insect biochemistry and molecular biology 35: 505-514, 2005.  152  12. Gisselmann G, Marx T, Bobkov Y, Wetzel CH, Neuhaus EM, Ache BW, and Hatt H. Molecular and functional characterization of an I(h)-channel from lobster olfactory receptor neurons. The European journal of neuroscience 21: 1635-1647, 2005. 13. Gong Q, Anderson CL, January CT, and Zhou Z. Role of glycosylation in cell surface expression and stability of HERG potassium channels. American journal of physiology 283: H77-84, 2002. 14. Helenius A, and Aebi M. Intracellular functions of N-linked glycans. Science (New York, NY 291: 2364-2369, 2001. 15. Helenius A, and Aebi M. Roles of N-linked glycans in the endoplasmic reticulum. Annual review of biochemistry 73: 1019-1049, 2004. 16. Herrmann S, Stieber J, Stockl G, Hofmann F, and Ludwig A. HCN4 provides a 'depolarization reserve' and is not required for heart rate acceleration in mice. The EMBO journal 26: 4423-4432, 2007. 17. Huang J, Huang A, Zhang Q, Lin YC, and Yu HG. Novel mechanism for suppression of hyperpolarization-activated cyclic nucleotide-gated pacemaker channels by receptor-like tyrosine phosphatase-alpha. The Journal of biological chemistry 283: 29912-29919, 2008. 18. Hubbard TJ, Aken BL, Ayling S, Ballester B, Beal K, Bragin E, Brent S, Chen Y, Clapham P, Clarke L, Coates G, Fairley S, Fitzgerald S, Fernandez-Banet J, Gordon L, Graf S, Haider S, Hammond M, Holland R, Howe K, Jenkinson A, Johnson N, Kahari A, Keefe D, Keenan S, Kinsella R, Kokocinski F, Kulesha E, Lawson D, Longden I, Megy K, Meidl P, Overduin B, Parker A, Pritchard B, Rios D, Schuster M, Slater G, Smedley D, Spooner W, Spudich G, Trevanion S, Vilella A, Vogel J, White S, Wilder S, Zadissa A, Birney E, Cunningham F, Curwen V, Durbin R, Fernandez-Suarez XM, Herrero J, Kasprzyk A, Proctor G, Smith J, Searle S, and Flicek P. Ensembl 2009. Nucleic acids research 37: D690-697, 2009. 19. Jackson HA, Marshall CR, and Accili EA. Evolution and structural diversification of hyperpolarization-activated cyclic nucleotide-gated channel genes. Physiological genomics 29: 231-245, 2007. 20. JGI. JGI Genome Portal The Regents of the University of California. http://genome.jgipsf.org/. 21. Karolchik D, Kuhn RM, Baertsch R, Barber GP, Clawson H, Diekhans M, Giardine B, Harte RA, Hinrichs AS, Hsu F, Kober KM, Miller W, Pedersen JS, Pohl A, Raney BJ, Rhead B, Rosenbloom KR, Smith KE, Stanke M, Thakkapallayil A, Trumbower H, Wang T, Zweig AS, Haussler D, and Kent WJ. The UCSC Genome Browser Database: 2008 update. Nucleic acids research 36: D773-779, 2008. 22. Kasturi L, Chen H, and Shakin-Eshleman SH. Regulation of N-linked core glycosylation: use of a site-directed mutagenesis approach to identify Asn-Xaa-Ser/Thr sequons that are poor oligosaccharide acceptors. The Biochemical journal 323 ( Pt 2): 415-419, 1997. 153  23. Khanna R, Myers MP, Laine M, and Papazian DM. Glycosylation increases potassium channel stability and surface expression in mammalian cells. The Journal of biological chemistry 276: 34028-34034, 2001. 24. Kuwahara K, Saito Y, Takano M, Arai Y, Yasuno S, Nakagawa Y, Takahashi N, Adachi Y, Takemura G, Horie M, Miyamoto Y, Morisaki T, Kuratomi S, Noma A, Fujiwara H, Yoshimasa Y, Kinoshita H, Kawakami R, Kishimoto I, Nakanishi M, Usami S, Saito Y, Harada M, and Nakao K. NRSF regulates the fetal cardiac gene program and maintains normal cardiac structure and function. The EMBO journal 22: 6310-6321, 2003. 25. Letunic I, and Bork P. Interactive Tree Of Life (iTOL): an online tool for phylogenetic tree display and annotation. Bioinformatics (Oxford, England) 23: 127-128, 2007. 26. Ludwig A, Budde T, Stieber J, Moosmang S, Wahl C, Holthoff K, Langebartels A, Wotjak C, Munsch T, Zong X, Feil S, Feil R, Lancel M, Chien KR, Konnerth A, Pape HC, Biel M, and Hofmann F. Absence epilepsy and sinus dysrhythmia in mice lacking the pacemaker channel HCN2. The EMBO journal 22: 216-224, 2003. 27. Maltsev VA, Wobus AM, Rohwedel J, Bader M, and Hescheler J. Cardiomyocytes differentiated in vitro from embryonic stem cells developmentally express cardiac-specific genes and ionic currents. Circulation research 75: 233-244, 1994. 28. Marionneau C, Couette B, Liu J, Li H, Mangoni ME, Nargeot J, Lei M, Escande D, and Demolombe S. Specific pattern of ionic channel gene expression associated with pacemaker activity in the mouse heart. The Journal of physiology 562: 223-234, 2005. 29. Moosmang S, Stieber J, Zong X, Biel M, Hofmann F, and Ludwig A. Cellular expression and functional characterization of four hyperpolarization-activated pacemaker channels in cardiac and neuronal tissues. European journal of biochemistry / FEBS 268: 16461652, 2001. 30. Moroni A, Gorza L, Beltrame M, Gravante B, Vaccari T, Bianchi ME, Altomare C, Longhi R, Heurteaux C, Vitadello M, Malgaroli A, and DiFrancesco D. Hyperpolarizationactivated cyclic nucleotide-gated channel 1 is a molecular determinant of the cardiac pacemaker current I(f). The Journal of biological chemistry 276: 29233-29241, 2001. 31. Much B, Wahl-Schott C, Zong X, Schneider A, Baumann L, Moosmang S, Ludwig A, and Biel M. 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, 2003. 32. Napp J, Monje F, Stuhmer W, and Pardo LA. Glycosylation of Eag1 (Kv10.1) potassium channels: intracellular trafficking and functional consequences. The Journal of biological chemistry 280: 29506-29512, 2005. 33. Nazzari H, Angoli D, Chow SS, Whitaker G, Leclair L, McDonald E, Macri V, Zahynacz K, Walker V, and Accili EA. Regulation of cell surface expression of functional pacemaker channels by a motif in the B-helix of the cyclic nucleotide-binding domain. Am J Physiol Cell Physiol 295: C642-652, 2008. 154  34. Petrecca K, Atanasiu R, Akhavan A, and Shrier A. N-linked glycosylation sites determine HERG channel surface membrane expression. The Journal of physiology 515 ( Pt 1): 41-48, 1999. 35. Proenza C, Tran N, Angoli D, Zahynacz K, Balcar P, and Accili EA. 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, 2002. 36. Qu Y, Whitaker GM, Hove-Madsen L, Tibbits GF, and Accili EA. Hyperpolarization-activated cyclic nucleotide-modulated 'HCN' channels confer regular and faster rhythmicity to beating mouse embryonic stem cells. The Journal of physiology 586: 701716, 2008. 37. Rho S, Lee HM, Lee K, and Park C. Effects of mutation at a conserved Nglycosylation site in the bovine retinal cyclic nucleotide-gated ion channel. FEBS letters 478: 246-252, 2000. 38. Santacruz-Toloza L, Huang Y, John SA, and Papazian DM. Glycosylation of shaker potassium channel protein in insect cell culture and in Xenopus oocytes. Biochemistry 33: 56075613, 1994. 39. Santoro B, Grant SG, Bartsch D, and Kandel ER. 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, 1997. 40. Sartiani L, Bettiol E, Stillitano F, Mugelli A, Cerbai E, and Jaconi ME. Developmental changes in cardiomyocytes differentiated from human embryonic stem cells: a molecular and electrophysiological approach. Stem cells (Dayton, Ohio) 25: 1136-1144, 2007. 41. Sartiani L, Bochet P, Cerbai E, Mugelli A, and Fischmeister R. Functional expression of the hyperpolarization-activated, non-selective cation current I(f) in immortalized HL-1 cardiomyocytes. The Journal of physiology 545: 81-92, 2002. 42. Shakin-Eshleman SH, Spitalnik SL, and Kasturi L. The amino acid at the X position of an Asn-X-Ser sequon is an important determinant of N-linked core-glycosylation efficiency. The Journal of biological chemistry 271: 6363-6366, 1996. 43. Shi W, Wymore R, Yu H, Wu J, Wymore RT, Pan Z, Robinson RB, Dixon JE, McKinnon D, and Cohen IS. Distribution and prevalence of hyperpolarization-activated cation channel (HCN) mRNA expression in cardiac tissues. Circulation research 85: e1-6, 1999. 44. Shin KS, Maertens C, Proenza C, Rothberg BS, and Yellen G. Inactivation in HCN channels results from reclosure of the activation gate: desensitization to voltage. Neuron 41: 737744, 2004. 155  45. Stieber J, Herrmann S, Feil S, Loster J, Feil R, Biel M, Hofmann F, and Ludwig A. 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, 2003. 46. Stillitano F, Lonardo G, Zicha S, Varro A, Cerbai E, Mugelli A, and Nattel S. Molecular basis of funny current (If) in normal and failing human heart. Journal of molecular and cellular cardiology 45: 289-299, 2008. 47. Tran N, Proenza C, Macri V, Petigara F, Sloan E, Samler S, and Accili EA. A conserved domain in the NH2 terminus important for assembly and functional expression of pacemaker channels. The Journal of biological chemistry 277: 43588-43592, 2002. 48. Urbach S, Cherel I, Sentenac H, and Gaymard F. Biochemical characterization of the Arabidopsis K+ channels KAT1 and AKT1 expressed or co-expressed in insect cells. Plant J 23: 527-538, 2000. 49. Wainger BJ, DeGennaro M, Santoro B, Siegelbaum SA, and Tibbs GR. Molecular mechanism of cAMP modulation of HCN pacemaker channels. Nature 411: 805-810, 2001. 50. Wallace IM, O'Sullivan O, Higgins DG, and Notredame C. M-Coffee: combining multiple sequence alignment methods with T-Coffee. Nucleic acids research 34: 1692-1699, 2006. 51. Watanabe I, Wang HG, Sutachan JJ, Zhu J, Recio-Pinto E, and Thornhill WB. Glycosylation affects rat Kv1.1 potassium channel gating by a combined surface potential and cooperative subunit interaction mechanism. The Journal of physiology 550: 51-66, 2003. 52. Watanabe I, Zhu J, Recio-Pinto E, and Thornhill WB. Glycosylation affects the protein stability and cell surface expression of Kv1.4 but Not Kv1.1 potassium channels. A pore region determinant dictates the effect of glycosylation on trafficking. The Journal of biological chemistry 279: 8879-8885, 2004. 53. Whitaker GM, Angoli D, Nazzari H, Shigemoto R, and Accili EA. 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, 2007. 54. Xu H, Fu Y, Tian W, and Cohen DM. Glycosylation of the osmoresponsive transient receptor potential channel TRPV4 on Asn-651 influences membrane trafficking. Am J Physiol Renal Physiol 290: F1103-1109, 2006. 55. Yano S, Miake J, Mizuta E, Manabe K, Bahrudin U, Morikawa K, Arakawa K, Sasaki N, Igawa O, Shigemasa C, Yamamoto Y, Morisaki T, Hidaka K, Kurata Y, Yoshida A, Shiota G, Higaki K, Ninomiya H, Lee JK, Shirayoshi Y, and Hisatome I. Changes of HCN gene expression and I(f) currents in Nkx2.5-positive cardiomyocytes derived from murine embryonic stem cells during differentiation. Biomedical research (Tokyo, Japan) 29: 195-203, 2008. 156  56. Yasui K, Liu W, Opthof T, Kada K, Lee JK, Kamiya K, and Kodama I. I(f) current and spontaneous activity in mouse embryonic ventricular myocytes. Circulation research 88: 536-542, 2001. 57. Ye B, and Nerbonne JM. Proteolytic processing of HCN2 and co-assembly with HCN4 in the generation of cardiac pacemaker channels. The Journal of biological chemistry 284: 25553-25559, 2009. 58. Zha Q, Brewster AL, Richichi C, Bender RA, and Baram TZ. Activity-dependent heteromerization of the hyperpolarization-activated, cyclic-nucleotide gated (HCN) channels: role of N-linked glycosylation. Journal of neurochemistry 105: 68-77, 2008. 59. Zhang Q, Huang A, Lin YC, and Yu HG. Associated changes in HCN2 and HCN4 transcripts and I(f) pacemaker current in myocytes. Biochimica et biophysica acta 1788: 11381147, 2009.  157  CHAPTER 4: COOH-TERMINAL PALMITOYLATION OF THE KV4.2 VOLTAGE-GATED POTASSIUM CHANNEL REGULATES CELL SURFACE EXPRESSION  A version of this chapter will be submitted for publication. Nazzari, H., Kang, R., Cheng, Y., Fedida, D., Accili, E.A. (2010) 158  4.1 INTRODUCTION The molecular mechanisms which govern the precise trafficking, localization and functioning of ion channels at the cell surface is critical for the regulation of cellular excitability. The accurate orchestration of ion channels is essential for ensuring the proper governing of physiological processes including action potential shape and duration, and vesicle exocytosis from pre-synaptic membranes. The superfamily of voltage-gated potassium (Kv) channels serve as critical players in ensuring the proper regulation of cellular excitability and a great deal of effort has been devoted towards fully understanding the mechanisms which govern Kv channel trafficking and function. However, one area that has only been limitedly explored is the posttranslational modifications of Kv channels. Two well documented modifications include channel phosphorylation and glycosylation (13, 19, 22, 25, 26, 32, 37, 41, 43). These protein modifications have been shown to regulate the sub-cellular localization and functioning of these Kv channel sub-types. Interestingly, one particular post-translational modification which has been shown to modulate the activity of a large number of integral membrane proteins, but has only been limitedly explored in ion channels is palmitoylation (4, 14). Protein palmitoylation is a dynamic and reversible post-translational modification (20). It involves the addition of palmitate, a 16-Carbon fatty acid moiety to cysteine residues through a thioester linkage and is therefore often referred to as S-palmitoylation (11, 20). The addition of palmitate results in enhanced protein hydrophobicity and facilities lipid bilayer interaction, which can in turn, drastically alter protein-membrane interactions. The palmitoylation and depalmitoylation of proteins is regulated by palmitoyl acyltranferase (PAT) and acylprotein thioesterase (APT) enzymes respectively (6), and it is this reversibility that has been shown to modulate the dynamic intracellular compartmental trafficking of proteins and localization to specific membrane microdomains (14, 15). 159  Specifically focusing on Kv channels, only two members of the Kv1 sub-family have been identified as substrates for palmitoylation. Kv1.1 was the first potassium channel of any type to be shown to undergo palmitoylation, specifically at residue C243 of the intracellular S2S3 linker, resulting in a depolarizing shift in the voltage dependent properties of channel activation (12). This suggests that protein-membrane interactions through palmitoylation of C243 may play an important role in the voltage sensing mechanisms within these channels. In the case of Kv1.5, palmitoylation does not contribute to the modulation of channel voltage dependence but results in the reduced stabilization of channels at the cell surface (16). It is important to note that in neither of these studies was the palmitoylation status of these channels in vivo ascertained. The identification and classification of all mammalian palmitoylated proteins (palmitoylproteome) continues to prove to be a difficult and complicated endeavor due to several factors. Firstly, there is as of yet no well defined consensus sequence for protein palmitoylation. Palmitoylation prediction software has been developed to aid in the identification of potential candidates for palmitoylation (28, 42). Additionally, the presence of hydrophobic residues surrounding candidate cysteine has also often been used as an indicator for likely palmitoylation. Neither of these methods has been proven to be entirely accurate. Secondly, the technologies which have previously existed have been insufficient for the identification of palmitoylated proteins on a large scale. Recently developed methodologies have greatly improved the process by which palmitoylated proteins are identified from native tissue (17). These technological advances utilize acylbiotin exchange chemistry (ABE) in combination with mass spectroscopic protein analysis and have led to the systematic and comprehensive classification of the palmitoyl-proteome of the yeast S. cerevisiae (29) and the neuronal palmitoyl-proteome of the rat brain (17). 160  Kv4.x channels are responsible for the generation of the transient outward current (Ito) which contributes to the repolarization of cardiomyocytes and the A-type sub-threshold current (IA) which regulates excitability in specific neurons of the central nervous system (3). In this study we demonstrate that the Kv4.2 channel is palmitoylated in embryonic rat cortical neurons and in COS-7 cells using the ABE assay. The palmitoylation of Kv4.2 results in an overall significant increase in the number of channels expressed at the cell surface, but does not contribute to the modulation of either activation or inactivation gating parameters. To the best of our knowledge this represents the first demonstration of a palmitoylated Kv channel outside of the Kv1 family and the first time direct evidence has been provided for the presence of a palmitoylated Kv channel in vivo.  161  4.2 MATERIALS AND METHODS 4.2.1 Mutagenesis and expression Rat Kv4.2 cDNA was PCR amplified and subcloned into pEGFP-C2 using BamHI and SmaI. Thirteen putative cytoplasmic cysteine residues were identified using the Kyte-Doolittle hydrophobicity secondary protein structure predictor programs. Mutagenesis of cysteine residues was completed using the Quick Change Site-Directed Mutagenesis Kit was purchased from Stratagene (La Jolla, CA). Three region specific cysteine mutant channels were constructed to determine the region of palmitoylation. These included mutation of three N-terminal cysteine residues: C111, C132, C133 (Kv4.2-3NT), two cysteines residues within the linker connecting transmembrane helices 2 and 3: C232, CC236 (Kv4.2-2-3linker) and finally all seven cysteine residues within the cytoplasmic COOH-terminus: C484, C503, C529, C530, C563, C583, C588 (Kv4.2-7CT). Wild type and mutant channels were transiently transfected into COS-7 cells and assayed for 48h post-transfection. 4.2.2 Electrophysiological and data analysis The standard bath solution contained (in mM): NaCl, 135; KCl, 5; MgCl2, 1; sodium acetate, 2.8; HEPES, 10; CaCl2, 1; adjusted to pH 7.4 using NaOH. The standard pipette filling solution contained (in mM): KCl, 130; EGTA, 5; MgCl2, 1; HEPES, 10; Na2ATP, 4; GTP, 0.1; adjusted to pH 7.2 with HCl. All chemicals were from Sigma (Mississauga, ON, Canada). For 2bromopalmitate experiments, 200 mM 2-bromopalmitate stock solution was prepared in 100% EtOH and stored at 4°C. Cells were incubated with 1mM 2-bromopalmitate or 100% EtOH as vehicle control for 36 hours prior to electrophysiological recordings. Whole cell current recording and data analysis were done using an Axopatch 200B amplifier and pClamp 9 software (Axon Instruments, Foster City, CA). Patch electrodes were fabricated using thin-walled 162  borosilicate glass (World Precision Instruments, Sarasota, FL) and polished by heating. Electrodes had resistances of 1-3 MΩ when filled with intracellular solution. Analogue capacity compensation and 80% series resistance compensation were used during whole cell measurements. Activation and inactivation protocols were run. To activate Kv4.2 channels, the cells were depolarized from −80 to +80 mV in 10 mV, 200 ms steps and repolarized to -80 mV between pulses. In the inactivation protocol, the cells were depolarized from -40 to +50 mV in 10 mV, 800 ms steps, followed by a second step to +60 mV for 50 ms, and finally repolarized to -80 mV between pulses. The peak current generated by the +60 mV second step of each pulses were normalized to the maximum current and plotted against membrane voltage to create an inactivation curve. Data were sampled at 10 kHz and filtered at 2 kHz. All recordings were performed at room temperature (20-23°C). Current density measurements were obtained through normalizing of peak currents at each test potential with cell capacitance. Current density values were compared using a one-way anova with a Bonferroni post test, significant differences were concluded for p values <0.05. 4.2.3 Cortical neuron preparation Cortical neuronal cultures were prepared from embryonic day 18 (E18) rat. High density (750-1000 cells/mm2) neurons were plated and grown in neurobasal medium (Invitrogen). Neurons were plated on 10 cm plates (BD bioscience) and grew for 14-16 days. Pharmacological treatment with 100µM 2-bromopalmiate (Sigma) was done at day in vitro 14 ( DIV14). COS-7 cells were transfected with various Kv4.2 constructs by lipid-mediated gene transfer kit Lipofectamine 2000 (Invitrogen). Cells were harvested after 24-48h and proteins were immunoprecipatated and subject to the ABE assay.  163  4.2.4 Cell lysate preparation and immunoprecipitation COS-7 cells were transfected with different constructs for 36-48 hours. Cells were washed with ice-cold PBS and re-suspended in 0.1 ml of lysis buffer containing TEE (50 mM Tris-HCl [pH 7.4], 1 mM EDTA, 1 mM EGTA), 150 mM NaCl, and 1% SDS. After extracting the proteins for 5 min at 40C, triton X-100 was added to 1% to neutralize the SDS in a final volume of 0.5 ml. Insoluble material was removed by centrifugation at 10,000 g for 10 min at 40C. For immunoprecipitation, samples were incubated with a GFP antibody for 1 hr at 40C. After addition of 20 µl protein G sepharose beads (Pharmacia), samples were incubated for 1 hr at 40C. Immunoprecipitates were analysed using the ABE assay. 4.2.5 Acyl-Biotin-Exchange (ABE) chemistry assay Cultured cortical neurons were subjected to the three chemical treatment steps of the acyl-biotinyl exchange (ABE) protocol as described before (Roth et al, 2006). In briefly, synaptosome extracts were treated with 1M N- ethylmaleimide (NEM) to block free Cys residues for 1 hour. Half of the sample was then treated with 1M hydroxylamine for 1 hour to remove palmitate from Cys residues (HAM+), whereas the other half was the control sample (HAM-) without hydroxylamine treatment. Both samples were then labeled with the Cys-crosslinking reagent 100mM Biotin-BMCC (1-biotinamido-4-[4'-(maleididomethyl) cyclohexanecarboxamido] butane) (Pierce). Biotinylated proteins were then pulled down by a streptavidin column and eluted by incubation with DTT.  164  4.3 RESULTS 4.3.1 Kv4.2 is palmitoylated in E18 rat embryonic cortical neurons To determine if Kv4.2 serves as a substrate for palmitoylation in vivo we employed the previously established acyl-biotinyl exchange (ABE) assay to screen for palmitoylated Kv4.2 from E18 embryonic rat cortical neurons. This method has been well characterized and proven to be highly specific and effective in the identification of palmitoylated proteins from complex protein extracts (17). The essential steps and general methodology of the ABE assay are highlighted in Figure 4.1. Briefly, the first step of this assay involves blockage of all free sulfhydryls with saturating concentrations of N-ethylmaleimide (NEM). The second step employs the use of hydroxylamine (HAM) which specifically cleaves the thioester bond that covalently attaches the fatty acyl group to the cysteine residue at the site of palmitoylation. Finally, the newly exposed cysteinyl thiols can be marked with a non-radioactive thiol-specific biotinylation reagent, in this case biotin-BMCC. Pull down of biotinylated proteins with avidin beads allows for detection of target proteins using protein specific antibodies. To confirm and validate the effectiveness of the ABE assay in this particular study we screened for the presence of palmitoylated post-synaptic density protein-95 (PSD-95). PSD-95 is a major neuronal scaffolding protein which has been previously shown to be palmitoylated on residues C3 and C5 (8, 9). The palmitoylation of PSD-95 is critical to ensuring that it undergoes proper clustering and localization at the post-synaptic density (5, 8). Using the ABE assay we were able to detect a signal in the HAM (+) treated lane (Fig. 4.2A), indicating that PSD-95 is palmitoylated in embryonic rat cortical neurons, thus validating our method for detecting palmitoylated proteins from this protein extract. In contrast, synapse-associated proteins, specifically SAP102 has been well documented as a non-palmitoylated protein (17, 45) and in our study we did not detect any significant positive signal from HAM (+) treated samples (Fig, 4.2A). In the case of Kv4.2 the 165  presence of a band in the HAM (+) lane strongly suggests that Kv4.2 is present in a palmitoylated form in embryonic rat cortical neurons (Fig, 4.2B). To the best of our knowledge this is the first time direct evidence has been provided for the presence of a palmitoylated Kv channel in vivo. 4.3.2 2-Bromopalmitate treatment of rat cortical neurons reduces the total amount of palmitoylated Kv4.2 To examine the turnover rate of palmitoylation for both of PSD-95 and Kv4.2 in vivo, we treated E18 embryonic rat cortical neurons with 100µM of 2-bromopalmitate (2-BP) for 5h. 2BP is a pharmacological agent which has been shown to be a selective inhibitor of protein palmitoylation. In the case of both PSD-95 and Kv4.2, treatment with 2-BP resulted in a significant reduction in the levels of palmitoylated protein, as indicated by the reduced signals in HAM (+) treated lanes (Fig. 4.3). Treatment with 2-BP did not alter overall protein levels for either PSD-95 or Kv4.2. Taken together these results suggest that the majority of Kv4.2palmitate bound protein undergoes significant turnover in rat cortical neurons within 5h. 4.3.3 Rat Kv4.2 and Kv4.3 are palmitoylated in COS-7 cells To readily characterize the functional role of palmitoylation in Kv4 channels we transiently expressed the rat Kv4.2 and Kv4.3 isoforms in COS-7 cells. The palmitoylation of proteins is accomplished through the action of palmitoyl acyltranferases (PATs). To date a significant number of mammalian proteins have been identified as likely candidates to serve as PATs. However, the characterization of each these proteins and their potential cell-type specifity has not been fully elucidated. Therefore, to determine if Kv4.2 and Kv4.3 can undergo palmitoylation in COS-7 cells we again utilized the ABE assay. Kv4.2 and Kv4.3 are closely related proteins which share a significant amount of homology at the amino acid level. As 166  demonstrated in Figure 4.4, a strong signal is present for Kv4.2 and Kv4.3 HAM (+) treated lanes. This demonstrates that both of these channels are present in a palmitoylated form in the COS-7 cell line, thus validating our use of these cells as a suitable model expression system to characterize the functional consequence of palmitoylation in Kv4 channels. 4.3.4 Kv4.2 is palmitoylated within the intracellular COOH-terminus To confirm that the signal detection in the HAM (+) treated samples is the result of a thioester modification involving a single or multiple cysteine residues, and to localize the site of palmitoylation within Kv4.2, we created a series of mutant Kv4.2 proteins. In the case of the rat Kv4.2 channel there are twelve intracellular cysteine residues (Fig. 4.5A) which may serve as potential candidate sites for channel palmitoylation. These cysteine residues are conserved throughout all mammalian isoforms of Kv4.2, and the majority of these residues are also conserved within the closely related Kv4.3 channel. We proceeded to create mutant Kv4.2 channels as follows: Kv4.2-3NT involves mutation of the three intracellular NH2-terminal cysteine residues (C111, C129, C130), Kv4.2-2-3linker eliminates the two cysteine residues within the intracellular linker connecting the S2 and S3 transmembrane helices (C232, C236), and finally Kv4.2-7CT which has the seven cysteine residues found in the intracellular COOHterminus mutated (C484, C503, C529, C530, C563, C583, C588). A comparison of the palmitoylation status of Kv4.2-WT with each of these three mutant constructs expressed in COS7 cells is shown in Figure 4.5B. In the case of both Kv4.2-3NT and Kv4.2-2-3linker we see the presence of a band in HAM (+) treated samples similar to that observed in Kv4.2-WT. Notably, Kv4.2-7CT results in a complete elimination of any detectable band in the HAM (+) lane, strongly suggesting that palmitoylation of Kv4.2 involves one or more cysteine residues within the intracellular COOH-terminus.  167  4.3.5 Palmitoylation does not modulate gating parameters of Kv4.2 To examine the functional role that palmitoylation plays in modulating the gating parameters of Kv4.2 channels, we again expressed Kv4.2-WT and Kv4.2-7CT (palmitoylation deficient mutant) in COS-7 cells in presence of 10µM 2-BP or vehicle control and performed whole-cell patch clamp electrophysiology experiments 48h post transfection. Analysis of the steady state V1/2 of activation and inactivation between the four test groups examined did not yield any significant differences (Figure 4.6A, B). The V1/2 of activation was 11.63 ± 1.367 mV and 13.25 ± 1.779 mV for Kv4.2-WT in the presence or absence of 2-BP respectively. While, cells expressing Kv4.2-7CT had V1/2 of activation values of 7.038 ± 1.345 mV and 11.62 ± 2.080 mV for 2-BP and vehicle control respectively. V1/2 of inactivation values were -66.19 ± 0.6114 and -68.93 ± 1.069 mV for Kv4.2-WT in the presence or absence of 2-BP respectively, while Kv4.2-7CT inactivation values were -68.88 ± 1.577 mV and -73.48 ± 1.211 mV in the presence or absence of 2-BP respectively. Overall, these results demonstrate that palmitoylation of Kv4.2 channels does not play any significant role in either activation or inactivation gating parameters. 4.3.6 Palmitoylation of Kv4.2 increase current density measurements To examine what role palmitoylation plays in the cell surface expression of Kv4.2 channels we again utilized whole-cell patch clamp electrophysiology to compare current density measurements between cells expressing Kv4.2-WT and Kv4.2-7CT in the presence or absence of 2-BP. Sample current traces for each of these test groups are provided in Figure 4.7A. Current density profiles from each test potential examined revealed that cells expressing Kv4.2-WT in the presence of 2-BP had similar current density profiles to Kv4.2-7CT expressing cells in the presence or absence of 2-BP. Each of these test groups had current density values which were significantly reduced when compared to cells expressing Kv4.2-WT in presence of a vehicle control (Fig. 4.7B). These data suggest that inhibition of palmitoylation through either mutation 168  of COOH-terminal cysteine residues or pharmacological inhibition with 2-BP results in an overall decrease in the number of channels expressed at the cell surface. The agreement in results between the two modes of palmitoylation inhibition supports that notion that the observed decrease in overall current density observed in the Kv4.2-7CT expressing cells is not due to any drastic alteration in protein folding due to the mutation of any of the seven cysteine residues.  169  4.4 DISCUSSION Lipid modifications of both soluble and membrane-bound proteins contribute to an enhancement of their hydrophobicity and promote their interaction with intracellular and plasma membranes (2, 21). Myristoylation, prenylation and palmitoylation are all commonly observed lipid modifications. Of these, palmitoylation represents one of the most frequently observed types (4). In addition to its obvious role in promoting membrane tethering of proteins, palmitoylation also serves a variety of other regulatory roles. These include targeting of proteins to specific membrane microdomains, promoting protein stability and regulating protein-protein interactions. The extent of palmitoylation in ion channels has not been explored extensively. In comparison to the large number of studies available on other common post-translational modifications, such as phosphorylation (1, 13, 19, 23, 24) and glycosylation (10, 18, 27, 30, 33, 35, 39, 40), palmitoylation of ion channels has only been reported in a small subset of channels (12, 16, 44). The first documented case involved the alpha-subunit of the sodium channel (31); but the functional consequence of this modification has yet to be investigated. The subsequent identification of Kv1.1 and Kv1.5 as substrates of palmitoylation, and the modulatory roles this modification has been shown to play in both channel trafficking and voltage-sensing, has provided further confirmation for the need to characterize the role of palmitoylation in the regulation of other ion channel sub-types. Recent advances in technologies relevant to the identification of palmitoylated proteins have paved the way for more detailed studies of palmitoylated ion channels (7, 17). In this study we set out to investigate whether Kv4 channels serve as a substrate for palmitoylation and, if so, what role palmitoylation plays in the regulation of its biological properties.  170  Utilizing ABE chemistry, we were able to demonstrate, for the first time, the presence of a palmitoylated Kv channel in vivo. Using cultured embryonic rat cortical neurons we show that Kv4.2 serves as a substrate for palmitoylation. Furthermore, we localized the site of palmitoylation to the intracellular COOH-terminus of the channel using mutational analysis of candidate cysteine residues. This study also demonstrates for the first time the presence of a palmitoylated Kv channel outside of the Kv1 family. Similar to Kv1.5, we show that palmitoylation of Kv4.2 plays an important role in the regulation of channel surface trafficking. In the case of Kv1.5, palmitoylation decreased overall surface trafficking, while in our study palmitoylation of Kv4.2 resulted in dramatic increases in overall surface expression, as revealed by enhanced current density measurements. Moreover, palmitoylation of Kv4.2 did not play a significant role in either the gating or voltage sensing properties of these channels. This is in contrast to the finding involving Kv1.1, where palmitoylation resulted in significant depolarizing shifts in channel voltage dependence. The similarities between the role of palmitoylation in both Kv4.2 and Kv1.5 and their differences to Kv1.1 may be attributed to location of channel palmitoylation. In the case of both Kv1.5 and Kv4.2 the cysteine residues which were identified to serve as the site of palmitoylation were found within the intracellular COOH-terminus, whereas in Kv1.1 the palmitoylated cysteine residues were located within the intracellular linker connecting the transmembrane helices S2 and S3. Palmitoylation of this domain may result in an overall increase in the anchoring and stabilization of the voltage sensing domain which may restrict movement of the S4 transmembrane segment. Additionally, palmitoylation of the more freely flexible intracellular COOH-terminus may facilitate protein-lipid interaction without significant voltage-sensing modulations.  171  Previous reports have demonstrated the ability of palmitoylation to facilitate proteinprotein interactions (21, 34, 36, 38). It would therefore be interesting to examine a possible role for palmitoylation of Kv4 channels in promoting the interaction with several well characterized binding partners, including potassium channel interaction proteins (KChIPs).In this respect, it is also of interest to note that palmitoylation of KChIPs can influence its ability to interact with Kv4.3 channels (34). In summary, our study has, for the first time, provided direct evidence for the presence of a palmitoylated Kv channel in vivo. Further, we have provided evidence that, as a result of palmitoylation, Kv4.2 channels display enhanced levels of surface trafficking in COS-7 cells. Although, the physiological relevance of this modification in vivo was not examined in this study, future experiments will be aimed at addressing this issue. Kv4 channels represent the molecular correlate responsible for production of the IA current in neurons. IA has been shown to play an important role in the regulation of neuronal excitability and synaptic plasticity. It will be interesting to examine the role in which palmitoylation of Kv4 channels plays in contributing to the regulation of these physiological processes.  172  4.5 ACKNOWLEDGEMENTS These studies were generously supported by a Grant-in-Aid from the Heart and Stroke Foundation of British Columbia & the Yukon (EAA). EAA is the recipient of a Tier II Canada Research Chair. HN is supported by the National Sciences and Engineering Research Council of Canada and the Michael Smith Foundation for Health Research.  173  Figure 4.1 Schematic representation of acyl-biotinyl exchange (ABE) assay. Illustrated is the sequence of experimental steps required for the identification of palmitoylated proteins using the ABE assay. Initially, free sulfhydryl groups are blocked using NEM. Following NEM blockage, HAM treatment results in the breakage of the thioester bond which links the palmitate to the cysteine residue. The newly freed sulfhydryl residues can then undergo a reaction with non-radioactive thiol-specific biotinylation reagents. 174  Figure 4.2 Kv4.2 is palmitoylated in E18 embryonic rat cortical neurons. ABE purification of palmitoylated proteins and western blot analysis was used to assay the palmitoylation status of A, PSD-95, SAP102 and B, Kv4.2 using protein specific antibodies from cultured rat embryonic cortical neurons. Experiment was performed the absence (-) or presence (+) of HAM. Total protein levels were confirmed in each protein sample to ensure equal protein loading between the different treatments.  175  Figure 4.3 2-bromopalmitate reduces Kv4.2 palmitoylation in rat cortical neurons. Example western blots are shown for 2-BP treatment of cortical neurons. 5-h treatment of cortical neurons with 100µM 2-BP to was performed to assess the amount of palmitoyl-turnover for both PSD-95 and Kv4.2. Total palmitoylated proteins, ABE-purified from the 2-BP treated and parallel, untreated neuronal cultures were blotted with the indicated specific antibodies. To control for 2BP effects on test protein expression levels, the initial, unpurified protein extracts also were blotted (total protein).  176  Figure 4.4 Kv4.2 and Kv4.3 are palmitoylated in COS-7 cells. Wild type Kv4.2 and Kv4.3 channels were transiently expressed in COS-7 cells for 48h and assayed for HAM sensitive thioester bonds. Cells were lysed in the presence of saturating concentration of NEM (50mM). Lysates were then incubated in the absence (-) or presence (+) of 1M HAM and then labelled with biotin-BMCC. Protein samples were subjected to SDSPAGE, transferred to a nitrocellulose paper and probed with streptavidin which is conjugated to a fluorescent probe. The same blot was then striped and reprobed with an anti-GFP antibody to examine total protein levels.  177  A. Extracellular  + + + + + Intracellular 484  133 132  231  236  503 529 530  GFP  111  563  583  588  er  B.  HAM  -  4.2 v K + -  nk i l 3  NT CT 3 2 7 2 2 2 4. 4. 4. v v v K K K + - + - +  anti-GFP  Figure 4.5 Kv4.2 is palmitoylated within the intracellular COOH-terminus. A, Schematic representation of the 12 intracellular cysteine residues which represent potential sites of palmitoylation in the rat Kv4.2 channel. B, Mutant channels in which cysteine residues are replaces with serines in specific domains within the channel were constructed. Wild type Kv4.2, Kv4.2-3NT (C111S, C132S, C133S), Kv4.2-2-3linker (C231S, C236S) and Kv4.2-7CT (C484S, C503S, C529S, C530S, C563S, C583S, C588S) were transiently expressed in COS-7 cells and analyzed using the previously described ABE chemistry to assay for the presence of palmitoylated protein in the absence (-) or presence (+) of HAM. 178  Figure 4.6 Palmitoylation does not modulate steady-state activation or inactivation properties. Expression of Kv4.2 and Kv4.2-7CT in COS-7 cells in the presence of 2-BP or vehicle control. A, Steady-state activation curves for Kv4.2 and Kv4.2-7CT. B, Steady-state inactivation curves for Kv4.2 and Kv4.2-7CT. Both activation and inactivation curves were fit using the Boltzmann equation.  179  Figure 4.7 Palmitoylation of Kv4.2 results in enhanced expression of channels at the cell surface. Mutation of intracellular COOH-terminal cysteine residues and treatment of cells expressing wild type Kv4.2 with 2-BP results in an overall decrease in current density measurements. A, Representative whole-cell current recordings (-80mV holding potential followed by +10mV steps to +60mV) from COS-7 cells transiently expressing Kv4.2-WT or Kv4.2-7CT channels in the absence (-) or presence (+) of 10µM 2-BP for 48h. B, Total current values at each test potential were normalized for variance in cell size and plotted as means ± SE for wild type Kv4.2 (n=8), Kv4.2-7CT (n=6), wild type Kv4.2 in the presence of 10µM 2-BP (n=6) and Kv4.2-7CT in the presence of 10µM 2-BP (n=6). Mean current density values of wild type Kv4.2 were significantly larger than the other three test groups at each test potential are represented with an asterisk (*P < 0.05). 180  4.6 REFERENCES 1. Anderson AE, Adams JP, Qian Y, Cook RG, Pfaffinger PJ and Sweatt JD. Kv4.2 phosphorylation by cyclic AMP-dependent protein kinase. J Biol Chem 275: 5337-5346, 2000. 2. Bhatnagar RS and Gordon JI. Understanding covalent modifications of proteins by lipids: where cell biology and biophysics mingle. Trends Cell Biol 7: 14-20, 1997. 3. Birnbaum SG, Varga AW, Yuan LL, Anderson AE, Sweatt JD and Schrader LA. Structure and function of Kv4-family transient potassium channels. Physiological reviews 84: 803-33, 2004. 4. Charollais J and Van Der Goot FG. Palmitoylation of membrane proteins (Review). Molecular membrane biology 26: 55-66, 2009. 5. Craven SE, El-Husseini AE and Bredt DS. Synaptic targeting of the postsynaptic density protein PSD-95 mediated by lipid and protein motifs. Neuron 22: 497-509, 1999. 6. Dietrich LE and Ungermann C. On the mechanism of protein palmitoylation. EMBO reports 5: 1053-7, 2004. 7. Drisdel RC, Alexander JK, Sayeed A and Green WN. Assays of protein palmitoylation. Methods (San Diego, Calif 40: 127-34, 2006. 8. El-Husseini AE, Craven SE, Chetkovich DM, Firestein BL, Schnell E, Aoki C and Bredt DS. Dual palmitoylation of PSD-95 mediates its vesiculotubular sorting, postsynaptic targeting, and ion channel clustering. J Cell Biol 148: 159-172, 2000. 9. El-Husseini A, Schnell E, Dakoji S, Sweeney N, Zhou Q, Prange O, GauthierCampbell C, Aguilera-Moreno A, Nicoll RA and Bredt DS. Synaptic strength regulated by palmitate cycling on PSD-95. Cell 108: 849-863, 2002. 10. Freeman LC, Lippold JJ and Mitchell KE. Glycosylation influences gating and pH sensitivity of I(sK). The Journal of membrane biology 177: 65-79, 2000. 11. Fukata Y and Fukata M. Protein palmitoylation in neuronal development and synaptic plasticity. Nat Rev Neurosci 11: 161-175, 2010. 12. Gubitosi-Klug RA, Mancuso DJ and Gross RW. The human Kv1.1 channel is palmitoylated, modulating voltage sensing: Identification of a palmitoylation consensus sequence. Proceedings of the National Academy of Sciences of the United States of America 102: 5964-8, 2005. 13. Hattan D, Nesti E, Cachero TG and Morielli AD. Tyrosine phosphorylation of Kv1.2 modulates its interaction with the actin-binding protein cortactin. J Biol Chem 277: 38596-38606, 2002. 181  14. Hayashi T, Rumbaugh G and Huganir RL. Differential regulation of AMPA receptor subunit trafficking by palmitoylation of two distinct sites. Neuron 47: 709-23, 2005. 15. Hayashi T, Thomas GM and Huganir RL. Dual palmitoylation of NR2 subunits regulates NMDA receptor trafficking. Neuron 64: 213-26, 2009. 16. Jindal HK, Folco EJ, Liu GX and Koren G. Posttranslational modification of voltagedependent potassium channel Kv1.5: COOH-terminal palmitoylation modulates its biological properties. American journal of physiology 294: H2012-21, 2008. 17. Kang R, Wan J, Arstikaitis P, Takahashi H, Huang K, Bailey AO, Thompson JX, Roth AF, Drisdel RC, Mastro R, Green WN, Yates JR,3rd, Davis NG and El-Husseini A. Neural palmitoyl-proteomics reveals dynamic synaptic palmitoylation. Nature 456: 904-9, 2008. 18. Khanna R, Myers MP, Laine M and Papazian DM. Glycosylation increases potassium channel stability and surface expression in mammalian cells. The Journal of biological chemistry 276: 34028-34, 2001. 19. Levin G, Chikvashvili D, Singer-Lahat D, Peretz T, Thornhill WB and Lotan I. Phosphorylation of a K+ channel alpha subunit modulates the inactivation conferred by a beta subunit. Involvement of cytoskeleton. J Biol Chem 271: 29321-29328, 1996. 20. Linder M and Deschenes R. Protein palmitoylation. Methods (San Diego, Calif 40: 125-6, 2006. 21. Linder ME and Deschenes RJ. Palmitoylation: policing protein stability and traffic. Nat Rev Mol Cell Biol 8: 74-84, 2007. 22. Mason HS, Latten MJ, Godoy LD, Horowitz B and Kenyon JL. Modulation of Kv1.5 currents by protein kinase A, tyrosine kinase, and protein tyrosine phosphatase requires an intact cytoskeleton. Mol Pharmacol 61: 285-293, 2002. 23. Misonou H, Mohapatra DP, Park EW, Leung V, Zhen D, Misonou K, Anderson AE and Trimmer JS. Regulation of ion channel localization and phosphorylation by neuronal activity. Nat Neurosci 7: 711-718, 2004. 24. Mohapatra DP and Trimmer JS. The Kv2.1 C terminus can autonomously transfer Kv2.1-like phosphorylation-dependent localization, voltage-dependent gating, and muscarinic modulation to diverse Kv channels. J Neurosci 26: 685-695, 2006. 25. Napp J, Monje F, Stuhmer W and Pardo LA. Glycosylation of Eag1 (Kv10.1) potassium channels: intracellular trafficking and functional consequences. The Journal of biological chemistry 280: 29506-12, 2005. 26. Noma K, Kimura K, Minatohara K, Nakashima H, Nagao Y, Mizoguchi A and Fujiyoshi Y. Triple N-glycosylation in the long S5-P loop regulates the activation and trafficking of the Kv12.2 potassium channel. J Biol Chem 284: 33139-33150, 2009. 182  27. Petrecca K, Atanasiu R, Akhavan A and Shrier A. N-linked glycosylation sites determine HERG channel surface membrane expression. The Journal of physiology 515 ( Pt 1): 41-8, 1999. 28. Ren J, Wen L, Gao X, Jin C, Xue Y and Yao X. CSS-Palm 2.0: an updated software for palmitoylation sites prediction. Protein Eng Des Sel 21: 639-644, 2008. 29. Roth AF, Wan J, Bailey AO, Sun B, Kuchar JA, Green WN, Phinney BS, Yates JR,3rd and Davis NG. Global analysis of protein palmitoylation in yeast. Cell 125: 1003-13, 2006. 30. Santacruz-Toloza L, Huang Y, John SA and Papazian DM. Glycosylation of shaker potassium channel protein in insect cell culture and in Xenopus oocytes. Biochemistry 33: 560713, 1994. 31. Schmidt JW and Catterall WA. Palmitylation, sulfation, and glycosylation of the alpha subunit of the sodium channel. Role of post-translational modifications in channel assembly. J Biol Chem 262: 13713-13723, 1987. 32. Schwetz TA, Norring SA and Bennett ES. N-glycans modulate K(v)1.5 gating but have no effect on K(v)1.4 gating. Biochimica et biophysica acta 2009. 33. Sutachan JJ, Watanabe I, Zhu J, Gottschalk A, Recio-Pinto E and Thornhill WB. Effects of Kv1.1 channel glycosylation on C-type inactivation and simulated action potentials. Brain Res 1058: 30-43, 2005. 34. Takimoto K, Yang EK and Conforti L. Palmitoylation of KChIP splicing variants is required for efficient cell surface expression of Kv4.3 channels. The Journal of biological chemistry 277: 26904-11, 2002. 35. Thornhill WB, Wu MB, Jiang X, Wu X, Morgan PT and Margiotta JF. Expression of Kv1.1 delayed rectifier potassium channels in Lec mutant Chinese hamster ovary cell lines reveals a role for sialidation in channel function. J Biol Chem 271: 19093-19098, 1996. 36. Topinka JR and Bredt DS. N-terminal palmitoylation of PSD-95 regulates association with cell membranes and interaction with K+ channel Kv1.4. Neuron 20: 125-134, 1998. 37. Utsunomiya I, Tanabe S, Terashi T, Ikeno S, Miyatake T, Hoshi K and Taguchi K. Identification of amino acids in the pore region of Kv1.2 potassium channel that regulate its glycosylation and cell surface expression. Journal of neurochemistry 2009. 38. Washbourne P. Greasing transmission: palmitoylation at the synapse. Neuron 44: 901902, 2004. 39. Watanabe I, Wang HG, Sutachan JJ, Zhu J, Recio-Pinto E and Thornhill WB. Glycosylation affects rat Kv1.1 potassium channel gating by a combined surface potential and cooperative subunit interaction mechanism. The Journal of physiology 550: 51-66, 2003. 183  40. Watanabe I, Zhu J, Recio-Pinto E and Thornhill WB. Glycosylation affects the protein stability and cell surface expression of Kv1.4 but Not Kv1.1 potassium channels. A pore region determinant dictates the effect of glycosylation on trafficking. The Journal of biological chemistry 279: 8879-85, 2004. 41. Watanabe I, Zhu J, Sutachan JJ, Gottschalk A, Recio-Pinto E and Thornhill WB. The glycosylation state of Kv1.2 potassium channels affects trafficking, gating, and simulated action potentials. Brain Res 1144: 1-18, 2007. 42. Xue Y, Chen H, Jin C, Sun Z and Yao X. NBA-Palm: prediction of palmitoylation site implemented in Naive Bayes algorithm. BMC Bioinformatics 7: 458, 2006. 43. Yang JW, Vacher H, Park KS, Clark E and Trimmer JS. Trafficking-dependent phosphorylation of Kv1.2 regulates voltage-gated potassium channel cell surface expression. Proc Natl Acad Sci U S A 104: 20055-20060, 2007. 44. Zhang L, Foster K, Li Q and Martens JR. S-acylation regulates Kv1.5 channel surface expression. Am J Physiol Cell Physiol 293: C152-61, 2007. 45. Zheng CY, Petralia RS, Wang YX, Kachar B and Wenthold RJ. SAP102 is a highly mobile MAGUK in spines. J Neurosci 30: 4757-4766, 2010.  184  CHAPTER 5: CONCLUSIONS AND GENERAL DISCUSSION  185  The regulation of cellular excitability is critically dependent on the cells ability to carefully orchestrate the biogenesis, trafficking and function of a large number of protein players. Ion channels represent the most critical player. From the beating of our hearts to our conscious thoughts, the proper regulation of these channels is essential in ensuring the regulated flow of ions in and out of our cells, which ultimately is responsible for the generation of carefully tuned action potentials. Those are responsible for the transmission of electrical impulses throughout our bodies, and are essential for sustaining life. This thesis describes studies that have examined the structural and molecular determinants regulating surface trafficking and function of ion channels belonging to the superfamily of voltage-gated potassium channels. Specifically, we have focused our efforts on the HCN1, HCN2 and Kv4.2 channels. Each of these channels plays an important role in the generation of action potentials within the cardiac and central nervous systems. Therefore knowledge of the molecular determinants which regulate their trafficking and function in these tissues is crucial for our understanding of how these proteins contribute to the physiological and pathophysiological states which originate in these complex systems. We considered three primary hypotheses. Firstly, determinants within the B-helix of the HCN2 channel play a role in both cell surface expression and inhibition of hyperpolarizationactivated opening. Secondly, glycosylation modifies HCN channel function in an isoform specific manner. Thirdly, Kv4 channels serve a substrate for palmitoylation and this modification contributes to the regulation of channel surface trafficking. Each of these hypotheses originates from unanswered questions related to the regulation of these channel subtypes. To date, little is still known about the molecular or structural determinants which regulate HCN and Kv4 channel function and trafficking. By addressing these three specific hypotheses in this thesis we provide 186  novel insight into several important factors which contribute to the regulation of these channels specifically and also to Kv-related channels in general. The EEYP motif within the B-helix of the HCN2 channel is required for cell surface expression but does not contribute to the inhibition of channel gating The experiments presented in chapter two describe a specific four amino acid motif (EEYP) within the B-helix of CNDB in the HCN2 channel. Previously, it was demonstrated that the CNBD domain of HCN2 is essential in ensuring forward trafficking of this channel to the cell surface (1, 13). It was also demonstrated that the 12 amino acids which comprise the B-helix of the CNBD plays a particularly important role in this cell surface regulation of HCN2 channels (13). Using a variety of biochemical and electrophysiological approaches we demonstrate that the EEYP motif within the B-helix is critical in promoting forward trafficking of HCN2 channels to the cell surface, but interestingly does not contribute to the inhibition of channel opening which is conferred by the CNBD in the absence of cyclic nucleotides (cAMP, cGMP) (18, 19). Through sequence alignment we compared the CNBD of the HCN2 channel with other proteins containing CNBDs. These included cytoplasmic protein kinases and several ion channels. We were able to conclude that among all the proteins examined the CNBD is well conserved up until the distal portion of the B-helix. Focusing our attention to this region revealed a four amino acid motif at the distal end of the B-helix which was well conserved among the three ion channels under study but not in the cytoplasmic protein kinases. This suggested the possible importance of this motif (EEYP in the case of HCN2) in promoting the trafficking of these channels to the plasma membrane. Through mutational analysis of this motif, includeding replacement of EEYP with four alanines in the full length channel (HCN2-4A), we were able to demonstrate that these channels failed to traffic to the cell surface to the same extant as wild type HCN2. In the case of HCN2187  4A, the absence of mature complex glycosylated channels in western blot analysis, suggests that removal of the EEYP results in an increased accumulation of channels at the level of the ER. There are a number of potential causes for accumulation of channels within the ER, the most likely of which is the gross misfoldeding of the channels which are then unable to traffic past the level of ER. Eventually these misfolded channels will be targeted for ERAD (10). Often times ERAD may result in activation of pro-apoptotic signalling events and eventual cell death (14, 15). Although we did not directly examine either of these possibilities for the ER retained HCN2-4A channels, we did compare this mutant with another mutant form of HCN2 known to undergo significant amounts of protein degradation. This mutant, which lacks a large portion of the NH2-terminus, fails to traffic to the cell surface and is shown through western blot analysis to reside in the cell in a degraded form. Through western blot analysis we were able to demonstrate that HCN2-4A does not undergo significant amounts of degradation and therefore is most likely residing within the ER in a relatively stable state. An important question addressed in this study was the functional state of the of HCN24A channels that were able to traffic to the cell surface. To do this we performed whole-cell patch clamp electrophysiology on CHO cells expressing HCN2-4A channels. Upon analysis of these cells we were able to detect the presence of small but appreciable amounts of If current production. Although these current levels were decreased nearly nine fold when compared to wild type HCN2, it provided evidence that HCN2-4A channels are able to undergo proper folding and trafficking to the plasma membrane, albeit at a significantly reduced amount. To better address the folding status of intracellular HCN2-4A channels we employed the use of sucrose density gradients. Examining the sedimentation profiles between wild type HCN2 and HCN2-4A revealed no significant difference. This observation coupled with the presence of mature glycosylated HCN2-4A channels in protein overloaded western blot experiments 188  provided strong evidence that channels which are retained within the ER are found in a properly folded and assembled state, while the limited number of channels that are able to traffic to the cell surface undergo mature complex glycosylation prior to their arrival. The exact mechanism by which EEYP promotes channel surface expression remains to be elucidated and our study leaves us with some very intriguing questions. Firstly, does the EEYP truly represent a forward trafficking motif? Although we have provided evidence in support of this idea, we have not conclusively demonstrated this to be true. To definitively define the EEYP motif in HCN2 as a novel forward trafficking motif it would be useful to insert this motif in a different ER retained channel to determine if its addition is able to restore surface expression in this channel. Such elegant studies were completed using a forward trafficking motif from an inward rectifying potassium channel (9). Secondly, does the removal of the EEYP motif result in the unmasking of an ER retention motif? ER retention motifs have been well documented in a number of different channel sub-types (3, 8, 17, 20). In the case of Kir6.2 channels it has been reported that dissociation between alpha pore forming subunit and beta subunit (SUR1) leads to the unmasking of an ER retention motif (17, 20) which ensures that only properly assembled channel complexes with the correct alpha:beta stoichiomety are trafficked out of the ER. To date, specific ER retention motifs have not been identified for HCN channels, but nonetheless remains a possible explanation for the ER retention of HCN2-4A channels. Finally, removal of the EEYP in full length channels may result in minor structural rearrangements, which do not prevent tetrameric assembly of channels, but may diminish the ability of these channels to interact with auxiliary chaperones and subunits which are critical for promoting cell surface trafficking. Overall, the experiments and results in chapter 2 demonstrate that the EEYP is an essential component in the forward trafficking of HCN2 channels to the plasma membrane, but 189  does not play any role in the inhibition of channel gating within these channels. The mechanistic details of this regulation forms the basis for future studies devoted to further examining structural determinants within HCN channels which regulate their trafficking and function. N-linked glycosylation modulates the trafficking and function of HCN channels in an isoform specific manner In chapter 3 of this thesis, we examine the role of N-linked glycosylation in the regulation of two specific pacemaker channel isoforms: HCN1 and HCN2. Previously it had been demonstrated that N-glycosylation is a critical determinant in both the trafficking and function of the HCN2 isoform (11). The authors of this study concluded that N-glycosylation serves as an essential requirement for the trafficking and function of all four HCN isoforms. We set out to determine the extent through which N-glycosylation modulates these parameters in both the HCN1 and HCN2 channel isoforms. Our results showed that both HCN1 and HCN2 are able to form functional channels in the absence of N-glycosylation, although expression levels of both channels at the cell surface was significantly reduced. In the case of the HCN2, elimination of channel glycosylation through mutation of the putative glycosylated asparagine residue within the extracellular S5-S6 linker (N380) results in a significantly marked reduction in overall If current density when expressed in either CHO cells or Xenopus oocytes. In the case of HCN1, mutation of the equivalent asparagine residue (N327) also resulted in significant reductions in If current densities, albeit to a much lesser degree than HCN2. These results demonstrated, for the first time, the ability of non-glycosylated HCN channels to traffic to the plasma membrane in a functional manner. A generally unanswered question in the regulation of membrane proteins and ion channels specifically, is how alterations in the levels of N-glycosylation can influence protein trafficking to the cell surface? We attempted to address this issue by generating a series of 190  glycosylation sequon mutant HCN2 channels. In doing so we discovered one particular mutation which enhanced HCN2 ability to undergo N-glycosylation. This particular mutant HCN2-H381A demonstrated significant increases in overall surface expression and current density levels. These results provide novel insight towards our understanding of N-glycosylation as a regulator/promoter of ion channel trafficking. For the first time, we have shown in HCN channels that increasing the efficiency of channel glycosylation can have significant impacts on overall cell surface expression. Overall, the reduction and enhancement in HCN channel surface expression through modulation of N-glycosylation levels may be attributed to the folding process of these channels at the level of the ER. The addition of oligosaccharides to proteins plays an important role in monitoring the overall folding process (5, 6, 16). Glycosylation of channels allows for their interaction with the ER chaperone protein calnexin, which in turn contributes to the prevention of grossly misfolded proteins escaping from the ER. In the absence of N-glycosylation misfolded proteins will not be able to undergo this regulatory checkpoint and therefore will not be able to undergo proper folding which will eventually lead to channel degradation. In our study, HCN2H381A demonstrated an enhanced efficiency for undergoing N-glycosylation and as a result enhanced surface expression. Therefore, it is very likely that a greater number of these channels undergo proper folding and forward trafficking as a result of this increased glycosylation efficiency. An outstanding question which remains from the studies described in chapter 3, is to what extent N-glycosylation contributes to channel assembly and tetramerization? Addressing this question is not trivial and will require co-expression studies involving wild type and glycosylation mutant channels. The rescue of glycosylation mutant channels by co-expression with wild type subunits would suggest that N-glycosylation of all four subunits in not required 191  for proper assembly and trafficking. There already exists evidence for this in the case of HCN1 and HCN2 (12, 21). However, further investigation is necessary in order to fully address this question. The entire data set examining the functional consequence of N-glycosylation in HCN channels was completed using heterologous expression systems and therefore represents a limitation in the interpretation of our findings. Although we were able to demonstrate that both HCN1 and HCN2 are present as glycosylated proteins in the embryonic mouse heart the functional consequence of this modification in this particular tissue can only be speculated upon. Future studies will be focused on understanding the role of N-glycosylation in native tissue. Does the requirement of N-glycosylation for HCN surface trafficking and function differ between the four HCN isoforms in native tissue? Again, addressing this question will prove difficult and will certainly involve the use of pharmacological agents which disrupt Nglycosylation in native cells. Palmitoylation of Kv4.2 and its role in surface expression and function In the final results chapter (chapter 4) we present experiments which demonstrate, for the first time, the presence of a palmitoylated Kv channel in native tissue. Only recently has the palmitoylation status of ion channels been explored (4, 7). Identifying which channels serve as substrates for palmitoylation has the potential to provide insight into novel mechanisms that regulate their trafficking and function. The identification of palmitoylated channels has become more feasible due to the many recent advances in the technological methods used for the identification of palmitoylated proteins. The leading assay which has emerged is based on ABE chemistry and has been shown to be highly effective in large scale identification of palmitoylated proteins from native tissue (2). Using this technology we were able to identify Kv4.2 as a palmitoylated protein from embryonic rat cortical neurons. Through mutational analysis of 192  candidate cysteine residues and transient expression of these mutant channels in heterologous expression systems, we were able to localize the domain of palmitoylation to the intracellular COOH-terminus. Furthermore, we demonstrated that inhibition of palmitoylation through either mutation of COOH-terminal cysteine residues or the pharmacological agent 2-bromopalmitate resulted in dramatic reductions in overall current density measurements. Although our findings provide evidence for a novel mechanism through which Kv4 channels may be regulated, our study leaves with some intriguing unanswered questions. Firstly, what is the physiological relevance of Kv4.2 palmitoylation in cortical neurons? IA current measurements from neurons incubated in the presence of 2-bromopalmitate will be useful in addressing this question. However, these types of experiments need to be approached with the caveat that the application of a global inhibitor of palmitoylation may lead to non-specific effects which may indirectly alter the trafficking of Kv4.2 in these cells. Transient expression of a palmitoylation deficient Kv4.2 channel is another approach; however, caveats related to this experimental approach include low protein expression levels and the teasing out of currents originating from transfected channels from those of endogenously expressed channels. Secondly, what is the cause for the overall reduction in current density measurements? Although, our study did not directly compare the surface expression of palmitoylation deficient and wild type Kv4.2 channels, the absence of any functional modulation (activation or inactivation) coupled with the overall decrease in current density suggests a decrease in the number of channels trafficked to the cell surface. However, we cannot rule out the possibility of enhanced channel endocytosis due to the removal of palmitoylation. Thirdly, how does palmitoylation of Kv4.2 influence its ability to interact with potential accessory proteins? Kv4 channels have been demonstrated to interact with a number of different auxiliary subunits. Many of these result in modulations in channel trafficking and/or function, it would be interesting to 193  determine if palmitoylation of the intracellular COOH-terminus in Kv4.2 results in a disruption of these interactions. Overall the findings of this thesis provide novel insight into the structural and molecular determinants which regulate the trafficking and function of several voltage-gated channels belonging to the Kv channel superfamily. We have shown through several different approaches that structural domains with the CNBD of the HCN2 channel, specifically a four amino acid motif in the B-helix plays an important role in promoting forward trafficking of channels to the cell surface. Through the examination of N-glycosylation in two different HCN isoforms, we have gained novel insight into the role of this modification in regulating ion channel surface expression and function. Finally, we show that palmitoylation of Kv4 channels occur in vivo and that this fatty acid modification contributes in the regulation of channel surface expression. This is of particular interest since palmitoylation of ion channels, in general, has been examined to a very limited extent.  194  5.1 REFERENCES 1. Akhavan A, Atanasiu R, Noguchi T, Han W, Holder N and Shrier A. Identification of the cyclic-nucleotide-binding domain as a conserved determinant of ion-channel cell-surface localization. J Cell Sci 118: 2803-2812, 2005. 2. Drisdel RC, Alexander JK, Sayeed A and Green WN. Assays of protein palmitoylation. Methods (San Diego, Calif 40: 127-34, 2006. 3. Griffith LC. Potassium channels: the importance of transport signals. Curr Biol 11: R226-8, 2001. 4. Gubitosi-Klug RA, Mancuso DJ and Gross RW. The human Kv1.1 channel is palmitoylated, modulating voltage sensing: Identification of a palmitoylation consensus sequence. Proceedings of the National Academy of Sciences of the United States of America 102: 5964-8, 2005. 5. Helenius A and Aebi M. Roles of N-linked glycans in the endoplasmic reticulum. Annual review of biochemistry 73: 1019-49, 2004. 6. Helenius A and Aebi M. Intracellular functions of N-linked glycans. Science (New York, N Y 291: 2364-9, 2001. 7. Jindal HK, Folco EJ, Liu GX and Koren G. Posttranslational modification of voltagedependent potassium channel Kv1.5: COOH-terminal palmitoylation modulates its biological properties. American journal of physiology 294: H2012-21, 2008. 8. Ma D and Jan LY. ER transport signals and trafficking of potassium channels and receptors. Curr Opin Neurobiol 12: 287-292, 2002. 9. Ma D, Zerangue N, Lin YF, Collins A, Yu M, Jan YN and Jan LY. Role of ER export signals in controlling surface potassium channel numbers. Science 291: 316-319, 2001. 10. Meusser B, Hirsch C, Jarosch E and Sommer T. ERAD: the long road to destruction. Nat Cell Biol 7: 766-772, 2005. 11. Much B, Wahl-Schott C, Zong X, Schneider A, Baumann L, Moosmang S, Ludwig A and Biel M. 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: 43781-6, 2003. 12. Much B, Wahl-Schott C, Zong X, Schneider A, Baumann L, Moosmang S, Ludwig A and Biel M. Role of subunit heteromerization and N-linked glycosylation in the formation of functional hyperpolarization-activated cyclic nucleotide-gated channels. J Biol Chem 278: 43781-43786, 2003. 195  13. Proenza C, Tran N, Angoli D, Zahynacz K, Balcar P and Accili EA. 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: 29634-42, 2002. 14. Rao RV, Ellerby HM and Bredesen DE. Coupling endoplasmic reticulum stress to the cell death program. Cell Death Differ 11: 372-380, 2004. 15. Rao RV, Hermel E, Castro-Obregon S, del Rio G, Ellerby LM, Ellerby HM and Bredesen DE. Coupling endoplasmic reticulum stress to the cell death program. Mechanism of caspase activation. J Biol Chem 276: 33869-33874, 2001. 16. Schrag JD, Bergeron JJ, Li Y, Borisova S, Hahn M, Thomas DY and Cygler M. The Structure of calnexin, an ER chaperone involved in quality control of protein folding. Mol Cell 8: 633-644, 2001. 17. Sharma N, Crane A, Clement JP,4th, Gonzalez G, Babenko AP, Bryan J and Aguilar-Bryan L. The C terminus of SUR1 is required for trafficking of KATP channels. J Biol Chem 274: 20628-20632, 1999. 18. Wainger BJ, DeGennaro M, Santoro B, Siegelbaum SA and Tibbs GR. Molecular mechanism of cAMP modulation of HCN pacemaker channels. Nature 411: 805-10, 2001. 19. Zagotta WN, Olivier NB, Black KD, Young EC, Olson R and Gouaux E. Structural basis for modulation and agonist specificity of HCN pacemaker channels. Nature 425: 200-205, 2003. 20. Zerangue N, Schwappach B, Jan YN and Jan LY. A new ER trafficking signal regulates the subunit stoichiometry of plasma membrane K(ATP) channels. Neuron 22: 537-548, 1999. 21. Zha Q, Brewster AL, Richichi C, Bender RA and Baram TZ. Activity-dependent heteromerization of the hyperpolarization-activated, cyclic-nucleotide gated (HCN) channels: role of N-linked glycosylation. Journal of neurochemistry 105: 68-77, 2008.  196  APPENDICES  197  APPENDIX A: CONTRIBUTIONS TO OTHER PUBLISHED MATERIAL  1. Macri V, Nazzari H, McDonald E, Accili EA. 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. 2009 Jun 5;284(23):15659-67. 2. Whitaker GM, Angoli D, Nazzari H, Shigemoto R, Accili EA. HCN2 and HCN4 isoforms self-assemble and co-assemble with equal preference to form functional pacemaker channels. Journal of Biological Chemistry. 2007 Aug;282(31):22900-9. 3. Peters CJ, Chow SS, Angoli D, Nazzari H, Cayabyab FS, Morshedian A, Accili EA. In situ co-distribution and functional interactions of SAP97 with sinoatrial isoforms of HCN channels. J Mol Cell Cardiol. 2009 May;46(5):636-43.  198  APPENDIX B: ALANINE SCANNING OF THE S6 SEGMENT REVEALS A UNIQUE AND CAMP –SENSITIVE ASSOCIATION BETWEEN THE PORE AND VOLTAGEDEPENDENT OPENING IN HCN CHANNELS  199  THE JOURNAL OF BIOLOGICAL CHEMISTRY VOL. 284, NO. 23, pp. 15659 –15667, June 5, 2009 © 2009 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in the U.S.A.  Alanine Scanning of the S6 Segment Reveals a Unique and cAMP-sensitive Association between the Pore and Voltage-dependent Opening in HCN Channels*□ S  Received for publication, December 5, 2008, and in revised form, January 30, 2009 Published, JBC Papers in Press, March 6, 2009, DOI 10.1074/jbc. M809164200  Vincenzo Macri1, Hamed Nazzari2, Evan McDonald, and Eric A. Accili3 From the Department of Cellular and Physiological Sciences, University of British Columbia, Vancouver, British Columbia V6T 1Z3, Canada Hyperpolarization-activated cyclic nucleotide-modulated (HCN) channels resemble Shaker K؉ channels in structure and function. 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. However, HCNs open when hyperpolarized, whereas Shaker opens 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 Shaker, an alanine/valine scan of residues across the pore, by single point mutation, showed that most mutations made the channel easier to open and steepened the response of the channel to changes in voltage. Because most mutations likely destabilize protein packing, the Shaker pore is most stable when closed, and the voltage sensor works to open it. In HCN channels, the pore energetics and vector of work by the voltage sensor are unknown. Accordingly, we performed a 22-residue alanine/valine scan of the distal pore of the HCN2 isoform and show that the effects of mutations on channel opening and on the steepness of the response of the channel to voltage are mixed and smaller than those in Shaker. These data imply that the stabilities of the open and closed pore are similar, the voltage sensor must apply force to close the pore, and the interactions between the pore and voltage sensor are weak. Moreover, cAMP binding to the channel heightens the effects of the mutations, indicating stronger interactions between the pore and voltage sensor, and tips the energetic balance toward a more stable open state.  Hyperpolarization-activated cyclic nucleotide-modulated (HCN)4 channels are similar in structure and function to Shaker Kϩ channels (1–3). As in Shaker, HCN channels are  * This work was supported in part by grants from the Heart and Stroke Foundation of British Columbia & the Yukon (to E. A. A.). The on-line version of this article (available at http://www.jbc.org) contains supplemental Fig. S1. 1 Recipient of doctoral scholarships from the Michael Smith Foundation for Health Research and the Canadian Institutes for Health Research. 2 Recipient of doctoral scholarships from the Michael Smith Foundation for Health Research and the National Sciences and Engineering Research Council of Canada. 3 Recipient of a Tier II Canada Research Chair. To whom correspondence should be addressed: Dept. of Cellular and Physiological Sciences, University of British Columbia, 2350 Health Sciences Mall, Vancouver, British Columbia V6T 1Z3, Canada. Tel.: 604-822-6900; Fax: 604-822-6048; E-mail: eaaccili@interchange.ubc.ca. 4 The abbreviations used are: HCN, hyperpolarization-activated cyclic nucleotide-modulated; CNBD, cyclic nucleotide-binding domain; pF, picofarad. □ S  JUNE 5, 2009 • VOLUME 284 • NUMBER 23  comprised of four 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 (4 – 6), 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 (7–10). 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 intrinsically most stable when closed and that the voltage sensor works to open the pore during depolarization (11, 12). 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 response of the channel 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 (13–15). 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 (11). 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 stabilities 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 JOURNAL OF BIOLOGICAL CHEMISTRY  15659  S6 Alanine Scanning in HCN Channels to close it. Thus, Shaker is closed and HCN2 is open in the absence of input from the voltage sensor. Moreover, cAMP 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 toward a more stable open state.  EXPERIMENTAL PROCEDURES 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, bp 1172–2216 of the mouse HCN2 template were amplified by PCR primers containing distal EcoRI and BamHI sites and subcloned into pBluescript. QuikChange (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 (Nucleic Acid Protein Service Unit facility, University of British Columbia). Tissue Culture and Expression of HCN2 Constructs—Chinese hamster ovary (CHO-K1) cells (ATCC, Manassas, VA) were maintained in Ham’s F-12 media supplemented with antibiotics and 10% fetal bovine serum (Invitrogen) and maintained at 37 °C 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 reporter plasmid (0.3 ␮g per dish), were transiently cotransfected into the cells using the FuGene6 transfection reagent (Roche Applied Science). Whole Cell Patch Clamp Electrophysiology—Cells expressing green fluorescent protein were chosen for whole cell patch clamp recordings 24 – 48 h post transfection. The pipette solution contained (in mM): 130 potassium 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 (sodium 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 ϭ 2.5– 4.5 M⍀). Data Analysis—Data were filtered at 2-kHz and were analyzed using Clampfit (Axon Instruments), Origin (MicroCal, Northampton, MA), and Excel (Microsoft, Seattle, WA) software. If activation curves were determined from tail currents at a 2-s pulse to Ϫ35 mV following 3- to 15-s test pulses ranging from Ϫ150 to Ϫ10 mV, in 20-mV steps. Single tail current test pulses were followed by a 500-ms pulse to ϩ5 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,  15660 JOURNAL OF BIOLOGICAL CHEMISTRY  f ͑ V ͒ ϭ I max/͑1 ϩ e͑V͓1/ 2͔ Ϫ V͒/k͒  (Eq. 1)  to determine the midpoint of activation (V1⁄2) and slope factor (k). The effective charge (Z) was calculated using the equation Z ϭ RT/kF, where T ϭ 295 K and R and F have their usual thermodynamic meanings. Changes in free energy between open and closed states were given by ϪZFV1⁄2. The perturbation in free energy produced by introduction of the point mutations (⌬(ZFV1⁄2)) was given by ϪF(ZmutV1⁄2mut Ϫ ZwtV1⁄2wt). The standard errors for ⌬(ZFV1⁄2) were calculated using ␣⌬(ZFV1/2) ϭ (␣2ZFV1/2,wt ϩ ␣2ZFV1/2,mut)1/2. Differences in values for V1⁄2, Z, and ZFV1⁄2 between the wild-type channel and mutant channels were determined independently using an unpaired t test (p Ͻ 0.05 was considered significant). Western Blot Analysis—Each sample was derived from cells on 35-mm plates that had been lysed in 100 ␮l of lysis buffer containing 50 mM Tris at pH 8.0, 1% Nonidet P-40, 150 mM NaCl, 1 mM EDTA, 1 mM phenylmethylsulfonyl fluoride, 2 mM each of Na3VO4 and NaF, and 10 ␮g/ml each of aprotinin, pepstatin, and leupeptin. Samples were left on ice for 30 min, during which time they were vortexed every 5 min for ϳ5 s. After centrifugation to remove cell debris (25,000 ϫ g, 25 min), protein concentration of the supernatant was determined by Bradford assay. 20-␮g samples of supernatant were fractionated by SDS-PAGE (8%) and electroblotted to polyvinylidene fluoride membrane (Bio-Rad, Mississauga, Ontario, Canada). 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% nonfat dry milk (Bio-Rad) in TBST for 1 h at room temperature. Blots were then incubated with a rabbit polyclonal antibody specific to the C terminus of HCN2 (Affinity Bioreagents, Golden, CO), at a dilution of 1:500 in TBST with 5% nonfat dry milk for 2.5 h at room temperature. Blots were washed in TBST for 10 min, three times, and then incubated with horseradish peroxidase conjugated to goat anti-rabbit 1:3000 dilution in 5% nonfat dry milk with TBST for 1 h at room temperature; they were subsequently washed three times in TBST. Signals were obtained with ECL Western blotting Detection Reagents (Amersham Biosciences). Protein loading was controlled by probing all Western blots with goat anti-glyceraldehyde-3-phosphate dehydrogenase antibody (Santa Cruz Biotechnology, Santa Cruz, CA).  RESULTS 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 singlepoint alanine/valine scan of the C-terminal 22 amino acids of the S6 segment in HCN2 (Ile422–Asp443) and examined channel opening, as described previously in Shaker (11). We hypothesized that, as for Shaker channels, the values for V1⁄2 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 conformaVOLUME 284 • NUMBER 23 • JUNE 5, 2009  S6 Alanine Scanning in HCN Channels tion, 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 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 1 (Experimental Procedures). From this fit, values for V1⁄2 and Z were determined to thereby allow calculation of perturbation energies (Table 1, A and B). Gating parameters and perturbation energies of wildtype channels were compared with those of the mutant channels using an unpaired t test. 18 of 22 single-point mutations expressed measurable levels of If from which activation curves could be derived (Fig. 1, A and B). Levels of If for G424A, A425V, T426A, and Y428A were not detectable. Unexpectedly, more mutants had a V1⁄2 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. 1C, upper). With one exception, all Z values of mutants were unchanged from that of wild type (Fig. 1C, lower). Finally, with the exception of three values, the free energies of mutants were unchanged from that of wild type (Fig. 1D). The mix of positive and negative shifts in V1⁄2 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 S6 in HCN2 expressed in Xenopus oocytes, which showed that most mutations shifted the opening of the channel to more negative potentials or had no effect; however, the energetic repercussions of these changes on gating were not explored (16). cAMP Shifts the Balance of Perturbation Energies of the S6 Mutations toward Negative Values— cAMP 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 (17–22). 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 (20, 23). cAMP 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 V1⁄2 and Z and a shift in perturbation energies toward 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 activaJUNE 5, 2009 • VOLUME 284 • NUMBER 23  TABLE 1 The effects of S6 pore mutations on voltage-dependent gating at basal (A) and saturating (2 mM; B) levels of cAMP The V[1/2] and Z values are from fits of activation curves with Equation 1 for wild-type and mutant channels (A, basal cAMP; B, 2 mM cAMP). The free energy of the open or closed state is shown as ϪZFV[1/2]. The difference in free energy between each mutant channel relative to wild type is indicated by ⌬(ZFV[1/2]). Data are presented as the mean Ϯ S.E. Asterisks represent significant differences from wild type.  tion curves could be determined (Fig. 2, A and B). For the wildtype HCN2 channel, V1⁄2 was shifted ϩ10.1 mV, and Z was decreased 0.4 compared with the values determined at basal cAMP (Table 1, A and B). The majority of V1⁄2 values in the mutant channels were more negative (6/21) or unchanged (9/21) compared with wild-type, whereas fewer values were more positive (6/21) (Fig. 2C, upper). The majority of Z values were larger (6/21) or unchanged (14/21) compared with wild type, whereas only one value was smaller (Fig. 2C, lower). A majority of free energies were more negative (9/21) or unchanged (11/21) compared with wild type, but only one value was more positive (Fig. 2D). Comparing free energies in saturating cAMP with those in basal cAMP (Figs. 1D and 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 JOURNAL OF BIOLOGICAL CHEMISTRY  15661  S6 Alanine Scanning in HCN Channels  15662 JOURNAL OF BIOLOGICAL CHEMISTRY  VOLUME 284 • NUMBER 23 • JUNE 5, 2009  S6 Alanine Scanning in HCN Channels 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 toward 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 Ϫ150 mV). In elevated cAMP, the activation range would have 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 (20), 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 with the wild-type channel (wt HCN2, Ϫ421 Ϯ 98 pA/picofarad (pF), n ϭ 8; A425V, Ϫ71 Ϯ 8 pA/pF, n ϭ 3; T426A, Ϫ116 Ϯ 22 pA/pF, n ϭ 4; Y428A, Ϫ100 Ϯ 16 pA/pF, n ϭ 5; all of the mutants are significantly different from wild-type HCN2, p Ͻ 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 (16). 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. 3). These data suggest that G424 is important for plasma membrane localization of functional channels. 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 V1⁄2 values became more negative (11). This relationship is consistent with effects on the final closed to open step in a linear gating scheme in which each of the four voltage sensors moves independently, and, once all sensors reach the permissive state, the pore opens by a voltage-independent concerted transition (24, 25). For HCN2, we were struck by the mutation-induced changes in Z, because they were very small compared with 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 (26) (Scheme 1). 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 Po ϭ  1 ͑1 ϩ 1/K͑V͒͒ 1 ϩ L͑V͒ ͑1 ϩ 1/aK͑V͒͒  ͩ  ͪ  4  (Eq. 2)  where K(V) and L(V) are the equilibrium constants for voltagesensor 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. (26) 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 V1⁄2 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 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. Select Po curves that spanned a similar range of voltages as those determined experimentally were then fitted with Equation 1 to yield theoretical values for Z and V1⁄2, which were then plotted in Fig. 4 (A and B). Both the Z values obtained by varying L(V) and those observed experimentally do not vary greatly with V1⁄2; this held true at basal and at saturating levels of cAMP (in Fig. 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 black lines, respectively). In contrast, the Z values obtained by varying K(V) in the model increase at more negative voltages and plateau in the 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. 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 gray lines, respectively). Furthermore, when K(V) was  FIGURE 1. Alanine scanning of the HCN2 S6 segment minimally perturbs the energetics of channel opening. 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 Ϫ150 mV to Ϫ10 mV, in 20-mV steps from a holding potential of Ϫ35 mV. The tail currents were elicited at Ϫ35 mV. B, representative If activation curves determined by plotting tail current amplitudes, which were normalized to their maximum value (I/Imax), versus test voltages (squares, HCN2; circles, Q440A; upright triangles, C427A; and inverted triangles, L438A). The curved lines represent fitting by Equation 1 (see “Experimental Procedures”). C, bar graphs depicting the changes in V1⁄2 (upper) and Z (lower) values for each mutant channel relative to wild-type. D, bar graph depicting change in perturbation of free energy, ⌬(ZFV1⁄2), 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).  JUNE 5, 2009 • VOLUME 284 • NUMBER 23  JOURNAL OF BIOLOGICAL CHEMISTRY  15663  S6 Alanine Scanning in HCN Channels  2 mM cAMP A.  C. HCN2  Q440A  30 * 20  ∆V1/2  *  10  *  *  *  *  0 -10  1 nA  1 nA -20  3s  3s  -30  T426A  *  *  *  *  I422A V423A G424A A425V T426A C427A Y428A A429V M430A F431A I432A G433A H434A A435V T436A A437V L438A I439A Q440A S441A L442A D443A  A437V  **  4 3 1 nA  0.5 nA 3s  2 3s  ∆Z  B.  ** *  1  *  * *  0 *  -1 -2 1.0  HCN2  8 6  Q440A  0.6  A437V T426A  0.4 0.2  ∆ZFV1/2  4 2  *  0 -2 -4  0.0  -6 -150 -130 -110 -90 -70 -50 -30 -10 10  voltage (mV)  15664 JOURNAL OF BIOLOGICAL CHEMISTRY  * *  **  * *  *  **  -8 I422A V423A G424A A425V T426A C427A Y428A A429V M430A F431A I432A G433A H434A A435V T436A A437V L438A I439A Q440A S441A L442A D443A  normalized current  D. 0.8  VOLUME 284 • NUMBER 23 • JUNE 5, 2009  S6 Alanine Scanning in HCN Channels TABLE 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. (26) as initial values, which were determined for the wild type human HCN2 channel.  FIGURE 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 HCN2G424A (lower trace) in response to hyperpolarizing voltage pulses to Ϫ150 mV from a holding potential of Ϫ35 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, ϳ136 kDa), immature (I, ϳ114 kDa) protein forms. These data are representative of three independent experiments. Note the absence of a mature form of HCN2 in lanes containing HCN2 N380Q (as demonstrated previously (37, 38)) and HCN2 G424A.  SCHEME 1  decreased in the model, the activation curves reached a point at which Z and V1⁄2 values changed very little, even with very small values for K(V). Consequently, there are no model Z values at voltages less negative than ϳϪ95 mV in Fig. 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. 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.  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 the voltage sensor works to open the pore (11). 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 (3, 27–30) and the frequent observation that artificial perturbations to HCN lead to even larger constitutively active currents. A resting conductance of ϳ2% has been estimated for HCN2 channels, whereas a value between 4 and 8% has been estimated for sea urchin HCN channels, without and with cAMP, respectively (28). 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 Clinkers (30 –33), and upon cadmium binding to cysteine substitutions near the intracellular side of the pore (8), 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  FIGURE 2. Saturating levels of cAMP (2 mM) shift the balance of perturbation energies to more negative values. 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 Ϫ150 mV to Ϫ10 mV, in 20-mV steps from a holding potential of Ϫ35 mV. The tail currents were elicited at Ϫ35 mV. B, Representative If activation curves determined by plotting tail current amplitudes which were normalized to their maximum value (I/Imax), versus test voltages (squares, HCN2; circles, Q440A; upright triangles, A437V; and inverted triangles, T426A). The curved lines represent fitting by Equation 1 (see “Experimental Procedures”). C, bar graphs depicting the changes in V1⁄2 (upper) and Z (lower) values for each mutant channel relative to wild type. D, bar graph depicting change in perturbation of free energy, ⌬(ZFV1⁄2), in mutant channels relative to wild type. One mutant channel did not yield measurable levels of If (solid line through numbered residue, x axis).  JUNE 5, 2009 • VOLUME 284 • NUMBER 23  JOURNAL OF BIOLOGICAL CHEMISTRY  15665  S6 Alanine Scanning in HCN Channels whether the gating model developed in that study predicts the small changes in Z seen in our study. cAMP 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 (20, 23). Our data showing a significant shift of perturbation energies to more negative values by FIGURE 4. Experimental and model Z values are comparable and change minimally over the range of mutations in the S6 are consistent observed mid-activation voltages. Plots of Z values versus V1⁄2 values for wild-type HCN2 channels and each with this proposed action of cAMP mutant channel examined, at basal (A) and 2 mM cAMP (B). Each line is derived from paired Z and V1⁄2 values determined from model Po curves at varying L(V) (black) and K(V) (gray) (see “Results”). Also shown are individ- and identify a cluster of residues ual values for Z and V1⁄2 obtained experimentally for wild-type (filled black diamonds), mutants that are signif- around the proposed activation gate icantly different from wild-type (filled red or blue diamonds, which are smaller or larger than wild-type, respec(35) that are modified by the inhibitively) and mutants that are not significantly different from wild-type (open squares). tory action of the CNBD (suppleopen and closed states. Nevertheless, a constitutively open mental Fig. S1). Our data are also consistent with previous work channel may not necessarily be an inevitable consequence of a in sea urchin HCN wherein mutation of a single residue in S6 pore that is more stable when open. At more positive voltages, (F459L) produced an equivalent effect to cAMP on gating (36). the voltage sensor could actively keep the channel shut. This is The corresponding site in mouse HCN2 (F431) is one of the ten the opposite of what happens in a channel with a pore that is cAMP-sensitive sites identified in our study. Our data suggest that the primary effect of the S6 mutations more stable when closed, like Shaker, in which the voltage senis on the closed to open step, the final step of the activation sors work to keep the channel open. Perturbation energies induced by the S6 mutations in HCN2 process, which seems reasonable for several reasons. First, the were smaller than those in Shaker (11), which suggest weaker mutations that are energetically sensitive cluster in a region of interactions between the voltage-sensing elements and the the S6 that likely forms the activation gate (7, 8, 35). Second, the pore. Loose coupling between the voltage sensor and pore, as small effects of the mutations on effective charge can be mostly, might be expected from a weak structural interaction, has been although not completely, explained by effects on the poreproposed recently for HCN channels (34). These authors opening step. Third, cAMP, which releases the inhibitory influshowed that the energetics of voltage-sensor movement is little ences on pore opening, significantly shifts perturbation eneraffected in sea urchin HCN channels that have been “locked gies toward the negative, suggesting that both the mutations open,” as opposed to the energetics of voltage-sensor move- and the CNBD target the same region. Nevertheless, an alloment in locked open Shaker channels, which are significantly steric effect of the mutations on voltage-sensor movement affected. The lack of apparent coupling in a locked open HCN could have contributed to the observed alterations in gating. channel is completely consistent with the notion that the pore We found that the significant effects on the effective charge (Z) is naturally open without input from the voltage-sensing produced by some of the mutations could not be explained by an allosteric model in which only the pore-opening step, or only elements. A difference in gating dynamics of HCN2 from Shaker is also the voltage-sensor movement, was altered. Other strategies are suggested by our finding that the effective charge Z, determined required to determine whether the voltage-sensing elements of from the slope of the activation curve, was changed only mini- HCN channels contribute to the observed effects of the S6 mally by the single-point S6 mutations. In contrast, single-point mutations on gating. It is important to note that the perturbamutations in the S6 of Shaker altered Z and perturbation energy tion energies of the S6 mutations in HCN2 are small relative to to a much greater extent, and the Z values increased as V1⁄2 those in the prototypical Shaker channel, especially at basal values became more negative (11). This difference in observed levels of cAMP; therefore, neither the pore or voltage sensor are Z between these two channels may arise from the fact that, in apparently affected despite mutations in and around the actiHCN2, the closed to open transition as well as the movement of vation gate. These small perturbation energies, along with their the voltage sensor may be voltage-dependent (11, 26). Thus, the shift toward the negative by cAMP, are strong support for both slope of the HCN2 activation curve would reflect contributions a weak interaction between the pore and voltage sensor, comfrom both processes, whereas that of Shaker would reflect a pared with Shaker, and a pore that is not at its energetic minicontribution primarily from voltage-sensor movement. It mum when closed. The evidence demonstrating that the effects should be noted that in 2007 a study on HCN2 channels sug- of the mutations on perturbation energy in saturating cAMP gested that the closed to open transition may instead be volt- levels are larger, and shifted toward negative, greatly strengthage-independent (21). It will be interesting to determine ens this conclusion.  15666 JOURNAL OF BIOLOGICAL CHEMISTRY  VOLUME 284 • NUMBER 23 • JUNE 5, 2009  S6 Alanine Scanning in HCN Channels 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 (4 – 6). 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. Acknowledgments—We thank Patrick Fletcher for help with Matlab and Martin Biel (Munich) for mouse HCN2 cDNA. REFERENCES 1. Santoro, B., Liu, D. T., Yao, H., Bartsch, D., Kandel, E. R., Siegelbaum, S. A., and Tibbs, G. R. (1998) Cell 93, 717–729 2. Ludwig, A., Zong, X., Jeglitsch, M., Hofmann, F., and Biel, M. (1998) Nature 393, 587–591 3. Gauss, R., Seifert, R., and Kaupp, U. B. (1998) Nature 393, 583–587 4. Mannikko, R., Elinder, F., and Larsson, H. P. (2002) Nature 419, 837– 841 5. Bell, D. C., Yao, H., Saenger, R. C., Riley, J. H., and Siegelbaum, S. A. (2004) J. Gen. Physiol. 123, 5–19 6. Vemana, S., Pandey, S., and Larsson, H. P. (2004) J. Gen. Physiol. 123, 21–32 7. Shin, K. S., Rothberg, B. S., and Yellen, G. (2001) J. Gen. Physiol. 117, 91–101 8. Rothberg, B. S., Shin, K. S., and Yellen, G. (2003) J. Gen. Physiol. 122, 501–510 9. Macri, V., Proenza, C., Agranovich, E., Angoli, D., and Accili, E. A. (2002) J. Biol. Chem. 277, 35939 –35946 10. Giorgetti, A., Carloni, P., Mistrik, P., and Torre, V. (2005) Biophys. J. 89, 932–944 11. Yifrach, O., and MacKinnon, R. (2002) Cell 111, 231–239 12. Hackos, D. H., Chang, T. H., and Swartz, K. J. (2002) J. Gen. Physiol. 119, 521–532  JUNE 5, 2009 • VOLUME 284 • NUMBER 23  13. Doyle, D. A., Morais Cabral, J., Pfuetzner, R. A., Kuo, A., Gulbis, J. M., Cohen, S. L., Chait, B. T., and MacKinnon, R. (1998) Science 280, 69 –77 14. Jiang, Y., Lee, A., Chen, J., Cadene, M., Chait, B. T., and MacKinnon, R. (2002) Nature 417, 515–522 15. Jiang, Y., Lee, A., Chen, J., Cadene, M., Chait, B. T., and MacKinnon, R. (2002) Nature 417, 523–526 16. Cheng, L., Kinard, K., Rajamani, R., and Sanguinetti, M. C. (2007) J. Pharmacol. Exp. Ther. 322, 931–939 17. DiFrancesco, D., and Tortora, P. (1991) Nature 351, 145–147 18. DiFrancesco, D. (1999) J. Physiol. 515, 367–376 19. Wainger, B. J., DeGennaro, M., Santoro, B., Siegelbaum, S. A., and Tibbs, G. R. (2001) Nature 411, 805– 810 20. Craven, K. B., and Zagotta, W. N. (2004) J. Gen. Physiol. 124, 663– 677 21. Chen, S., Wang, J., Zhou, L., George, M. S., and Siegelbaum, S. A. (2007) J. Gen. Physiol. 129, 175–188 22. Barbuti, A., Baruscotti, M., Altomare, C., Moroni, A., and DiFrancesco, D. (1999) J. Physiol. 520, 737–744 23. Zhou, L., and Siegelbaum, S. A. (2007) Structure 15, 655– 670 24. Zagotta, W. N., Hoshi, T., and Aldrich, R. W. (1994) J. Gen. Physiol. 103, 321–362 25. Schoppa, N. E., and Sigworth, F. J. (1998) J. Gen. Physiol. 111, 313–342 26. Altomare, C., Bucchi, A., Camatini, E., Baruscotti, M., Viscomi, C., Moroni, A., and DiFrancesco, D. (2001) J. Gen. Physiol. 117, 519 –532 27. Proenza, C., Angoli, D., Agranovich, E., Macri, V., and Accili, E. A. (2002) J. Biol. Chem. 277, 5101–5109 28. Proenza, C., and Yellen, G. (2006) J. Gen. Physiol. 127, 183–190 29. Ishii, T. M., Takano, M., Xie, L. H., Noma, A., and Ohmori, H. (1999) J. Biol. Chem. 274, 12835–12839 30. Chen, J., Mitcheson, J. S., Tristani-Firouzi, M., Lin, M., and Sanguinetti, M. C. (2001) Proc. Natl. Acad. Sci. U. S. A. 98, 11277–11282 31. Macri, V., and Accili, E. A. (2004) J. Biol. Chem. 279, 16832–16846 32. Decher, N., Chen, J., and Sanguinetti, M. C. (2004) J. Biol. Chem. 279, 13859 –13865 33. Chen, J., Mitcheson, J. S., Lin, M., and Sanguinetti, M. C. (2000) J. Biol. Chem. 275, 36465–36471 34. Bruening-Wright, A., Pandey, S., and Larsson, P. (2008) Biophys. J. 94, 119 35. Rothberg, B. S., Shin, K. S., Phale, P. S., and Yellen, G. (2002) J. Gen. Physiol. 119, 83–91 36. Shin, K. S., Maertens, C., Proenza, C., Rothberg, B. S., and Yellen, G. (2004) Neuron 41, 737–744 37. Nazzari, H., Angoli, D., Chow, S. S., Whitaker, G., Leclair, L., McDonald, E., Macri, V., Zahynacz, K., Walker, V., and Accili, E. A. (2008) Am. J. Physiol. 295, C642–C652 38. Much, B., Wahl-Schott, C., Zong, X., Schneider, A., Baumann, L., Moosmang, S., Ludwig, A., and Biel, M. (2003) J. Biol. Chem. 278, 43781– 43786  JOURNAL OF BIOLOGICAL CHEMISTRY  15667  APPENDIX C: HCN2 AND HCN4 ISOFORMS SELF-ASSEMBLE AND CO-ASSEMBLE WITH EQUAL PREFERENCE TO FORM FUNCTIONAL PACEMAKER CHANNELS  200  THE JOURNAL OF BIOLOGICAL CHEMISTRY VOL. 282, NO. 31, pp. 22900 –22909, August 3, 2007 © 2007 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in the U.S.A.  HCN2 and HCN4 Isoforms Self-assemble and Co-assemble with Equal Preference to Form Functional Pacemaker Channels* Received for publication, November 29, 2006, and in revised form, June 4, 2007 Published, JBC Papers in Press, June 6, 2007, DOI 10.1074/jbc.M610978200  Gina M. Whitaker‡, Damiano Angoli‡, Hamed Nazzari‡, Ryuichi Shigemoto§, and Eric A. Accili‡1 From the ‡Department of Cellular and Physiological Sciences, University of British Columbia, Vancouver, British Columbia V6T 1Z3, Canada and the §Division of Cerebral Structure, National Institute for Physiological Sciences, Myodaiji, Okazaki 444-8787, Japan  Hyperpolarization-activated cyclic nucleotide-modulated (HCN)2 channels, which underlie hyperpolarization-activated or funny currents (Ih or If) in excitable cells, are thought to be made up of subunits that assemble as tetramers to form functional channels (1). Four mammalian HCN isoforms (HCN1 to -4) (2– 6) possess various overlapping patterns of expression in the heart and throughout the central nervous system, suggesting  * This work was supported by grants from the Heart and Stroke Foundation of British Columbia and the Yukon and the Canadian Institutes for Health Research (to E. A. A.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1 To whom correspondence should be addressed: Dept. of Cellular and Physiological Sciences, University of British Columbia, 2350 Health Sciences Mall, Vancouver, British Columbia V6T 1Z3, Canada. Tel.: 604-822-6900; Fax: 604-822-6048; E-mail: eaaccili@interchange.ubc.ca. 2 The abbreviations used are: HCN, hyperpolarization-activated cyclic nucleotide-modulated; CHO, Chinese hamster ovary; BRET2, bioluminescence resonance energy transfer; Rluc, Renilla luciferase; GFP, green fluorescent protein; PBS, phosphate-buffered saline; NGS, normal goat serum; CNBD, cyclic nucleotide binding domain; HA, hemagglutinin.  22900 JOURNAL OF BIOLOGICAL CHEMISTRY  that they form heteromeric channels in these tissues (1, 7, 8). Previous studies suggest that the following combinations of HCN isoforms co-assemble and form functional channels in heterologous expression systems: HCN1 with HCN2 (9 –12) and HCN1 with HCN4 (11). On the other hand, whether HCN2 and HCN4 isoforms co-assemble and form functional channels has not been shown and is an important objective of the present experiments. The best evidence for co-assembly of HCN2 and HCN4 in native tissue comes from studies in the embryonic heart and adult thalamus. In the embryonic mouse heart, mRNA for HCN2 and both mRNA and protein for HCN4 have been found (13–16). Knock-out of HCN4 reduces, but does not abolish, If and speeds up rates of If activation in cardiomyocytes, consistent with the presence of other HCN isoforms in these cells (16). Immunohistochemical approaches in rats and mice have demonstrated HCN2 and HCN4 protein in thalamocortical relay nuclei (17, 18) and colocalization in cells of the ventrobasal complex and reticular nucleus of the thalamus (19). Knock-out of HCN2 reduces, If in thalamocortical neurons, consistent with the presence of other HCN isoforms in these cells (20). Co-assembly of HCN2 and HCN4 is supported by evidence of interaction of the two isoforms in heterologous expression systems. When co-expressed in human embryonic kidney cells, HCN2 and HCN4 were found to colocalize and co-immunoprecipitate, and cell surface fluorescence of an HCN2 trafficking mutant channel was rescued when co-expressed with HCN4 (21). Reductions in HCN4 current density by co-expression with a nonfunctional HCN2 pore mutant in Chinese hamster ovary (CHO) cells (22) also suggests that HCN2 and HCN4 interact in a physical and/or functional way. Although these data could be explained by co-assembly of HCN2 and HCN4, they can also be readily explained by other interactions. For example, functional and physical associations between ion channel subunits have been reported between voltage-gated potassium channels that are distantly related in primary structure and do not co-assemble. One example is between the human-ether-a-go-go-related gene and KCNQ1, which exist as separate groups of homomeric channels within macromolecular complexes both in native tissue and when overexpressed in mammalian cells (23). Together, the data support interactions between HCN2 and HCN4 but do not demonstrate the formation of functional heteromeric channels. Evidence in favor of functional co-assembly of HCN2 and HCN4 in live cells comes from single If channel recordings in VOLUME 282 • NUMBER 31 • AUGUST 3, 2007  Downloaded from www.jbc.org at University of British Columbia on October 18, 2008  Hyperpolarization-activated cyclic nucleotide-modulated (HCN) “pacemaker” channel subunits are integral membrane proteins that assemble as tetramers to form channels in cardiac conduction tissue and nerve cells. Previous studies have suggested that the HCN2 and HCN4 channel isoforms physically interact when overexpressed in mammalian cells, but whether they are able to co-assemble and form functional channels remains unclear. The extent to which co-assembly occurs over self-assembly and whether HCN2-HCN4 heteromeric channels are formed in native tissue are not known. In this study, we show co-assembly of HCN2 and HCN4 in live Chinese hamster ovary cells using bioluminescence resonance energy transfer (BRET2), a novel approach for studying tetramerization of ion channel subunits. Together with results from electrophysiological and imaging approaches, the BRET2 data show that HCN2 and HCN4 subunits self-assemble and co-assemble with equal preference. We also demonstrate colocalization of HCN2 and HCN4 and a positive correlation of their intensities in the embryonic mouse heart using immunohistochemistry, as well as physical interactions between these isoforms in the rat thalamus by coimmunoprecipitation. Together, these data support the formation of HCN2-HCN4 heteromeric channels in native tissue.  HCN2 and HCN4 Channel Co-assembly  EXPERIMENTAL PROCEDURES Molecular Biology—The construction of an HCN2 C-terminal deletion (lacking the cyclic nucleotide binding domain and the remainder of the C terminus distal to it, HCN2⌬CNBD) and another mutant lacking the entire N terminus was described previously (12, 30). An extracellular HA epitope was inserted between the S3 and S4 transmembrane domain of HCN2⌬CNBD (Fig. 4A). This construct was made by digesting HCN2-HA (kind gift from M. Sanguinetti) at common HCN2 restriction sites, such that the HA tag was removed and placed in HCN2⌬CNBD. For BRET2 constructs, Renilla luciferase (Rluc) or green fluorescent protein (GFP) tags were added to the N-terminal end of HCN2, HCN2⌬N, HCN4, or Kv1.5 cDNA. BRET2 vectors (pRlucC and pGFPC; PerkinElmer Life Sciences) were digested at restriction sites complementary to those present on 5Ј and 3Ј ends of HCN2, HCN2⌬N, HCN4, or Kv1.5, such that the tags were expressed in frame with the channel cDNA when ligated. Resulting sequences were confirmed by automated DNA sequencing (DNA sequencing core facility, Vancouver, Canada). All tagged constructs, except those made with HCN2⌬N, which does not form functional channels, were tested by patch clamp electrophysiology and produced currents characteristic of their wild type counterparts (data not shown). Cell Culture and Transfection—CHO-K1 cells (American Type Culture Collection, Manassas, VA) were maintained in Ham’s F-12 medium (Invitrogen) supplemented with 50 ␮g/ml penicillin/streptomycin (Invitrogen) and 10% fetal bovine serum (Invitrogen) and incubated at 37 °C with 5% CO2. For electrophysiology and immunocytochemistry, cells were plated onto glass coverslips in 35-mm dishes. After 48 h, CHO cells AUGUST 3, 2007 • VOLUME 282 • NUMBER 31  were transiently transfected with mammalian expression vectors encoding wild type and/or mutant channels (2 ␮g/dish) using FuGene6 transfection reagent (Roche Applied Science). For electrophysiology, cells were also co-transfected with the GFP reporter plasmid for identification using fluorescent microscopy (0.5– 0.7 ␮g/dish). Whole Cell Patch Clamp Electrophysiology and Analysis— One to 2 days following transfection, a shard of coverslip plated with cells was transferred to a recording chamber (ϳ200-␮l volume) and continually perfused (0.5–1.0 ml/min) with a low Kϩ extracellular solution (5.4 mM KCl, 135 mM NaCl, 0.5 mM MgCl2, 1.8 mM CaCl2, 5 mM HEPES, pH to 7.4 with NaOH). Following rupture of the patch membrane, the solution was changed to a high Kϩ recording solution (135 mM KCl, 5.4 mM NaCl, 0.5 mM MgCl2, 1.8 mM CaCl2, 5 mM HEPES, pH to 7.4 with KOH) to maximize current amplitude. The patch pipettes were filled with a solution containing 130 mM potassium aspartate, 10 mM NaCl, 0.5 mM MgCl2, 5 mM HEPES, and 1 mM EGTA and adjusted to a pH of 7.4 with KOH. In some experiments, this solution was supplemented with 2 mM cAMP (as noted). Whole cell hyperpolarization-activated currents (If) were measured using borosilicate glass electrodes (Sutter Instrument Co.), which had a resistance of 2.0 – 4.0 megaohms when filled with the intracellular solution. Currents were recorded using an Axopatch 200B amplifier and Clampex software (Axon Instruments Inc.). Data were filtered at 2 kHz and were analyzed using Clampfit (Molecular Devices) and Origin (Microcal) software. All experiments were conducted at room temperature (20 –22 °C). Currents were not leak-subtracted. The voltage dependence of If activation was determined from tail currents at Ϫ65 mV following 2-s test pulses ranging from Ϫ50 to Ϫ150 mV in 20-mV steps. Normalized tail current amplitudes were plotted as a function of test potential, and values were fit with a Boltzmann function as follows, f ͑ V ͒ ϭ I max/͑1 ϩ e͑V1/ 2 Ϫ V͒/k͒  (Eq. 1)  to determine the midpoint of activation (V1⁄2) and slope factor (k). Each test pulse was followed by a 200 –500-ms pulse to ϩ5 mV to ensure complete channel deactivation, and the resting current was allowed to return to its base-line value before subsequent voltage pulses. Time constants to assess rates of If activation were generated using a single exponential fitting procedure. An initial delay during If activation was not well described by a single exponential function and therefore was not used in our fits (17, 31). Immunocytochemistry, Fluorescence Microscopy with Structured Illumination, and Determination of Pearson Correlation Coefficients—Forty-eight hours after transfection, cells on coverslips were washed briefly with PBS and fixed with 2% paraformaldehyde in PBS for 5 min. Fixed cells were washed twice with PBS, some were permeabilized for 10 min using 0.2% Triton X-100, and all were blocked with 10% normal goat serum (NGS) and 10% bovine serum albumin for 10 min. After one wash with PBS containing 1% NGS, cells were incubated with primary antibodies for 1 h at room temperature. Anti-HA (Sigma) or anti-Rluc (Chemicon International Inc.) mouse monoclonal antibodies were used as needed at a dilution of JOURNAL OF BIOLOGICAL CHEMISTRY  22901  Downloaded from www.jbc.org at University of British Columbia on October 18, 2008  co-transfected CHO cells. These single channels possessed intermediate activation kinetics, in theory reflecting contributions from both subunits (24). However, whether the measured channels in that study truly represent HCN channels remains controversial (25–28). Among the HCN isoforms that do co-assemble, whether they have a preference for self-assembly (homomerization) over co-assembly (heteromerization) is not known. Because the primary amino acid sequences of the four mammalian HCN isoforms are very similar, it is possible that their abilities to selfassemble or to co-assemble may also be similar. After exclusion of the nonconserved regions of the cytoplasmic N and C termini, the amino acid homology among the four mammalian isoforms is very high (Ն94% conserved), and HCN2 and HCN4 isoforms bear the strongest conservation of primary sequence among the four isoforms (Ն96% conserved) (29). Nevertheless, the differences between HCN2 and HCN4, both those within the conserved region and those within the cytoplasmic N and C termini, could modify the extent to which these isoforms coassemble under different conditions. In this study, we demonstrate that co-assembly of HCN2 and HCN4 in live CHO cells occurs with equal preference compared with self-assembly, using multiple approaches, including bioluminescence resonance energy transfer (BRET2), a novel approach to study the association of ion channel subunits. We also provide evidence in support of co-assembly of HCN2 and HCN4 in the rat thalamus and embryonic mouse heart.  HCN2 and HCN4 Channel Co-assembly  ͚ ͑͑ GVC1 Ϫ MVCI͒ ϫ ͑GVC2 Ϫ MVC2͒͒  ͱ͚͑͑GVC1 Ϫ MVC1͒2 ϫ ͚͑GVC2 Ϫ MVC2͒2͒  (Eq. 2)  where GV represents Gray value, MV is mean value, and C is channel. The values range from Ϫ1 to ϩ1, representing an increasing correlation of the intensities measured in two channels. In other words, the Pearson correlation coefficient describes the interdependence of varying intensities of fluorescence between two proteins of interest throughout a cell. Bioluminescence Resonance Energy Transfer (BRET2) Analysis—BRET2 experimental methods were based on a recent study examining the dimerization of G-protein-coupled receptors (32). CHO cells were transiently transfected with a constant amount (0.5 ␮g) of Rluc-tagged and varying amounts of GFP-tagged constructs (0.5–2.0 ␮g) in order to measure optimal expression ratios for BRET2 experiments. Forty-eight hours after transfection, cells were washed twice with Dulbecco’s PBS (Invitrogen), detached with 0.05% trypsin-EDTA (Invitrogen) and resuspended in Dulbecco’s PBS supplemented with 2 ␮g/ml aprotinin. Approximately 100,000 cells were then distributed into individual wells of 96-well Optiplates (PerkinElmer Life Sciences). Using a Victor 3V plate reader (PerkinElmer Life Sciences), expression of GFP-tagged constructs was assessed by directly exciting GFP with a 400 – 410-nm excitation filter. Expression of Rluc-tagged constructs was assessed using luminescence values obtained in the BRET2 assay. -Fold level over background of emission was determined for both Rluc- and GFP-tagged constructs by comparing luminescence and fluorescence values with background values in untransfected cells. Although we varied the amount of cDNA of the GFP-tagged constructs, the resulting expression levels did not vary greatly, and therefore all samples were used in subsequent BRET2 experiments. For BRET2 measurements, DeepBlueC substrate (PerkinElmer Life Sciences) was added to cells (final concentration of 5 ␮M), and Rluc emission was measured through a 370 – 450-nm filter. Resulting GFP emission was in turn measured through a 500 –530-nm filter. BRET2 ratios  22902 JOURNAL OF BIOLOGICAL CHEMISTRY  were calculated by the ratio of GFP emission to Rluc emission. Ratios were corrected for background by subtracting emission ratios from untransfected cells after the addition of DeepBlueC substrate. BRET2 values are expressed as the mean Ϯ S.D. for three or six independent experiments (as noted), performed in duplicate, using different Rluc-tagged/GFP-tagged transfection ratios in each experiment. Graphing and statistical analysis were done using Prism 4 software (GraphPad Software). Immunoprecipitation and Western Blotting of Rat Thalamus— Rat thalami were obtained as previously described (33). Tissue was processed in radioimmune precipitation lysis buffer (50 mM Tris at pH 8.0, 1% Nonidet P-40, 150 mM NaCl, 1 mM EDTA, 1 mM PMSF, 2 mM each Na3VO4 and NaF, and 10 ␮g/ml each aprotinin, pepstatin, and leupeptin) by multiple passes through 16-gauge and then 23-gauge syringes and incubated on ice for 30 min, followed by centrifugation to remove cell debris. Lysates were precleared with normal rabbit IgG and protein-A beads (Sigma) for 2 h at 4 °C and then incubated with primary antibodies (rabbit anti-HCN2 or rabbit anti-HCN4; Alomone Laboratories) and protein A beads overnight at 4 °C. For negative controls, primary antibodies were preincubated with supplied antigens for 1 h prior to incubation with precleared lysates. Beads were collected by centrifugation and washed three times in ice-cold PBS and then boiled in sample buffer with 2-mercaptoethanol for 10 min and loaded into 8% SDSpolyacrylamide gels. Gels were transferred to polyvinylidene difluoride membranes, and blots were washed two times in TBS-T and then blocked with 5% nonfat dry milk in TBS-T for 1 h. Blots were incubated with primary antibody overnight at 4 °C (guinea pig anti-HCN4 (1:500) or rabbit anti-HCN2 (1:400)) (18) in 5% nonfat milk with 3% bovine serum albumin. After three washes with TBS-T, blots were incubated with horseradish peroxidase-conjugated secondary antibodies at a dilution of 1:3000 in 5% nonfat milk for 1 h at room temperature. After three washes with TBS-T, signals were obtained with ECL detection reagents (GE Healthcare). All IP experiments were carried out three times (n ϭ 3 rats). Immunohistochemistry of Mouse Tissue Sections—Mouse day 18 embryos were obtained from CD1 mice. Isolated tissue was cryosectioned into 10-␮m sections, mounted on poly-Llysine-treated slides (Wax-it histology services; UBC), and stored at Ϫ80 °C. Sections were fixed in 4% paraformaldehyde for 10 min and then washed twice with PBS. Sections were then permeabilized with 0.2% Triton X-100 for 15 min, washed three times in PBS, and then blocked with 10% normal donkey serum in PBS for 1 h at room temperature. Sections were then incubated with primary antibody overnight at 4 °C. Primary antibody dilutions were as follows: guinea pig anti-HCN2 and antiHCN4 (18), 1:500; goat anti-HCN4 (Santa Cruz Biotechnology, Inc., Santa Cruz, CA), 1:50; and rabbit anti-HCN2 (Alomone Laboratories), 1:200. For negative controls, tissue sections were incubated with either goat or guinea pig IgG or with rabbit anti-HCN2 preincubated with its supplied antigen (3:1 antigen/ antibody ratio). Tissue sections were washed three times in PBS, followed by incubation with Alexa 488 or 555-tagged secondary antibodies raised in donkeys (1:1000; Molecular Probes). After three washes in PBS, 4Ј,6-diamidino-2-phenylindole (1:50,000; Molecular Probes) was added to tissue for 5 VOLUME 282 • NUMBER 31 • AUGUST 3, 2007  Downloaded from www.jbc.org at University of British Columbia on October 18, 2008  1:500 in 1% NGS, PBS. Cells were subsequently washed with PBS three times and incubated with Alexa-555-tagged goat anti-mouse secondary antibodies (Molecular Probes, Inc., Ontario, Canada) at a dilution of 1:1500 in PBS with 1% NGS plus bovine serum albumin for 1 h at room temperature in the dark. After washing three times in PBS, coverslips were rinsed in double-distilled H2O and mounted on slides using Gel Mount (Sigma). Cells were visualized using a Zeiss Axiovert 200 fluorescence microscopy with an Apotome structured illumination module and with a ϫ63 oil immersion objective lens. Results reported represent a minimum of four transfections for each set of the imaging experiments described. To correlate intensities of fluorescence for each pair of proteins, Pearson correlation coefficients were calculated from individual cells co-transfected with different combinations of Kv1.5 and/or HCN isoforms, tagged with GFP or Rluc. Calculation of the Pearson correlation coefficient from captured images of individual cells was determined by the following equation (Axiovision User Guide),  HCN2 and HCN4 Channel Co-assembly  min, followed by two washes in PBS. Coverslips were mounted on tissue using GelMount (Sigma), and heart tissue was visualized as described for transfected cells. These experiments were carried out on sections of heart tissue from two embryonic mice. In each field of view, four areas were analyzed. Pearson correlation coefficients were determined in each area and averaged to yield one value per field of view. Values for Pearson correlation coefficients are presented as means Ϯ S.E. in n number of fields of view.  RESULTS Tetrameric Assembly of HCN2 and HCN4 Isoforms—Because of their similarity to voltage-gated potassium channels, HCN channels are probably tetrameric, and their co-assembly can be described by a binomial distribution. This distribution is determined from the proportion of subunits present to the nth power (pn), where n is equal to the number of different subunit isoforms in a given channel (34). When co-expressed, two isoforms with equal preference for homomeric and heteromeric assembly will co-assemble according to a binomial distribution, whereas any preference toward homomeric or heteromeric assembly would alter this distribution pattern. Here, we sought to determine whether HCN2 and HCN4 subunits co-assemble to form heteromeric channels and whether they exhibit a preference for homomeric versus heteromeric assembly. This was accomplished using two approaches in which HCN2 and HCN4 were tagged to Rluc or GFP epitopes. Co-transfection allowed us to obtain measurements from channels made up of a mixture of subunits possessing each tag (Fig. 1, bracketed section). Thus, both self-assembly and co-assembly could be determined independently and compared. These assays cannot measure channels made up of subunits containing the same tag or differentiate among channels AUGUST 3, 2007 • VOLUME 282 • NUMBER 31  JOURNAL OF BIOLOGICAL CHEMISTRY  22903  Downloaded from www.jbc.org at University of British Columbia on October 18, 2008  FIGURE 1. Tetrameric assembly of Rluc- and GFP-tagged subunits. Subunits tagged with Rluc and GFP form tetrameric channels according to a binomial distribution. Depicted are 24, or 16, total channels from four possible combinations of subunit assembly (shown above each bar), in which Rlucand GFP-tagged subunits assemble without preference. The number of channels per combination, of 16 total channels, is plotted on the y axis. In this diagram, Rluc-tagged subunits are white, whereas GFP-tagged subunits are black. The bracket denotes the group of possible stoichiometric Rluc-GFP combinations that can be detected by the imaging and BRET2 approaches used in this study.  that exhibit different stoichiometries, but they can report on the sum of all channels possessing both subunits. We utilized the voltage-gated channel subunit Kv1.5 as a negative control. Kv1.5 is found in the same superfamily as the HCN channel family and has a similar overall structure but is not expected to co-assemble with HCN subunits. However, when co-expressed in CHO cells, Kv1.5 is highly colocalized with both HCN2 and HCN4, as determined by measuring colocalization coefficients in the present imaging studies (see below). Thus, Kv1.5 provided a critical negative control for potential interactions among subunits that did not involve coassembly. These types of interactions include both specific interactions (e.g. between adjacent but separate tetrameric channels) and nonspecific interactions possibly due to overexpression of protein in intracellular compartments. Co-transfection of differentially tagged HCN2 and HCN4 with themselves provided positive controls for co-assembly. Homomeric and Heteromeric Combinations of HCN2 and HCN4 Produce Equally High Pearson Correlation Coefficients in CHO Cells—We first acquired immunofluorescent images of CHO cells co-expressing constructs tagged with GFP or Rluc (Fig. 2A) and determined the extents to which the intensities of their fluorescence correlated. To quantify this, we utilized a Pearson correlation coefficient (see “Experimental Procedures”). For two subunits that co-assemble to form a tetrameric ion channel, their varying levels of expression throughout a cell would be expected to be interdependent. Therefore, the variation of their intensities of fluorescence would correlate to a greater extent than for two subunits that associate in another way (e.g. two subunits that form separate channels which are localized to a similar region) or that do not associate at all. We co-transfected CHO cells with HCN2-GFP or HCN4GFP with HCN2-Rluc, HCN4-Rluc, or Kv1.5-Rluc and correlated the intensities of fluorescence for each combination of two isoforms. A strength of this approach is that the same molecules (GFP and Rluc) were used for all correlations; thus, there is no variability in our measurements due to the use of different antibodies. We hypothesized that HCN2 or HCN4, when cotransfected with themselves or with each other, would yield Pearson correlation coefficients that were significantly higher than those produced by co-transfection of either HCN2 or HCN4 with Kv1.5, if they formed homomeric or heteromeric channels. We found that Pearson correlation coefficients were significantly larger in cells co-transfected with HCN2-GFP and HCN2-Rluc, HCN2-GFP and HCN4-Rluc, HCN4-GFP and HCN4-Rluc, or HCN4-GFP and HCN2-Rluc than those determined from cells that were co-transfected with either HCN2GFP or HCN4-GFP and Kv1.5-Rluc (Fig. 2, B and C). Since HCN isoforms do not co-assemble with Kv1.5, the larger Pearson correlation coefficients determined for the different combinations of HCN subunits suggest co-assembly among them. This conclusion is strengthened by our finding that coefficients determined from cells co-transfected with HCN2-GFP and HCN2-Rluc or with HCN4-GFP and HCN4-Rluc (which are expected to self-assemble) were not significantly different from those determined from cells co-transfected with HCN2-GFP and HCN4-Rluc or with HCN4-GFP and HCN2-Rluc, respec-  HCN2 and HCN4 Channel Co-assembly  22904 JOURNAL OF BIOLOGICAL CHEMISTRY  VOLUME 282 • NUMBER 31 • AUGUST 3, 2007  Downloaded from www.jbc.org at University of British Columbia on October 18, 2008  ficking, and thus they may be localized to similar areas of cells, especially when overexpressed. Homomeric and Heteromeric Combinations of HCN2 and HCN4 Produce Equally High BRET2 Ratios in CHO Cells—In order to study coassembly of HCN2 and HCN4 channels in live cells, we utilized BRET2 technology. This approach, which has been widely used to examine and demonstrate receptor-protein interactions and receptor dimerization (35), offers several advantages for the analysis of protein interactions. BRET2 measures two proteins that are located within 10 nm of each other. Like its methodological cousin, fluorescence energy transfer, BRET2 allows for measurements in live cells and avoids a number of issues associated with more invasive approaches. Unlike fluorescence energy transfer, BRET2 does not require an initial light source and thus avoids photobleaching. An added and powerful advantage is that BRET2 measurements are FIGURE 2. Correlation of fluorescent intensities in co-transfected CHO cells supports similar levels of self-assembly and co-assembly of HCN2 and HCN4. A, images of CHO cells co-transfected with combina- taken from a large population of live tions of HCN2, HCN4, and Kv1.5 constructs. Rluc-tagged constructs are shown in red, and GFP-tagged con- cells rather than from individual structs are shown in green, with areas of colocalization depicted in yellow. Scale bars, 10 ␮m. B, colocalization cells (36). Thus, BRET2 seems well scatterplots of representative images in A of cells co-transfected with HCN2-GFP and Kv1.5-Rluc, HCN2-Rluc, or HCN4-Rluc. Intensity of GFP fluorescence is represented on the x axis, and intensity of Alexa555 (or Rluc) suited for the determination of tetfluorescence is shown on the y axis. The frequency of pixel overlap at a given relative fluorescence intensity is rameric ion channel assembly. plotted with greater frequency of overlap in red (toward the origin of axes) and lower in blue. Pearson correlaOnce both Rluc- and GFP-tagged tion coefficients determined from the values of intensities in each graph are shown. C, bar graph of Pearson correlation coefficients determined from cells co-transfected with different construct combinations. Values constructs are expressed in CHO represent means Ϯ S.D. for each set of constructs, and n values depict number of cells used to determine mean cells, the addition of DeepBlueC Pearson correlation coefficients over a minimum of four independent transfections. Statistical comparisons were carried out using a one-way analysis of variance followed by Bonferroni’s multiple comparison post-test substrate to live, intact cells is oxicomparing all pairs. The asterisks indicate significant differences from the correlation coefficients produced by dized by Rluc, causing emission at co-transfection of Kv1.5-Rluc and HCN2-GFP or HCN4-GFP. All combinations of HCN construct co-expression 380 nm. Emission at this wavelength show Pearson correlation coefficients that are not significantly different from each other (p Ͼ 0.05) but are significantly greater than those determined from combinations of Kv1.5RLuc with the HCN constructs excites GFP if the two proteins are (p Ͻ 0.001). within 10 nm of each other (Fig. 3, A and B). To test the ability of our system to detect BRET2, we carried out tively. These similarities also suggest that HCN2 and HCN4 experiments using CHO cells co-transfected with ␤1-adrenerisoforms form homomeric channels and heteromeric channels gic receptors tagged with GFP or Rluc. We were able to reprowith equal preference. duce previously published results showing high BRET2 ratios It should be noted that Pearson correlation coefficients give with ␤1-adrenergic receptors (data not shown), which supports more information about the nature of interaction between two the formation of homodimers (32). We hypothesized that proteins than simply quantifying the extent of colocalization. HCN2 or HCN4, when co-transfected with themselves or with For example, the coefficients were significantly lower for com- each other, would yield BRET2 values that were significantly binations of Kv1.5 and HCN isoforms, but the colocalization higher than those produced by co-transfection of either HCN2 coefficient (which measures pixel overlap between two or HCN4 with Kv1.5. channels) was high (r Ͼ 0.85) for all combinations, including We compared BRET2 ratios from cells co-transfected with those with Kv1.5 and HCN isoforms. This suggests that the same combinations of tagged constructs used in the imagalthough Kv1.5 does not co-assemble with HCN2 or HCN4 ing experiments (above). In addition, we used CHO cells transisoforms, they do colocalize to similar regions within the fected with Kv1.5-Rluc to determine background levels of emiscell. This is not unexpected, because separate homomeric sion at 510 nm. We found that BRET2 ratios determined from channels may share similar pathways of biogenesis and traf- cells co-transfected with HCN2-GFP and HCN2-Rluc, with  HCN2 and HCN4 Channel Co-assembly  HCN2-GFP and HCN4-Rluc, with HCN4-GFP and HCN4-Rluc, or with HCN4-GFP and HCN2-Rluc were significantly larger that those obtained from cells co-transfected with either HCN2-GFP or HCN4-GFP and with Kv1.5-Rluc (Fig. 3, C and D). Furthermore, we found that the BRET2 ratios determined from cells cotransfected with HCN2-GFP or HCN4-GFP and Kv1.5-Rluc were not significantly different from those determined from cells transfected with only Kv1.5-Rluc, indicating that our negative control produced nearly background levels of BRET2. These data strongly suggest that HCN2 and HCN4 co-assemble to form functional channels in live CHO cells. In addition, the levels of BRET2 determined from cells co-transfected with HCN2-GFP and HCN4-Rluc or with HCN4-GFP and HCN2-Rluc were not significantly different from homomeric combinations of Rluc- and GFP-tagged constructs. These data further support an equal preference for homomeric and heteromeric channel assembly in live CHO cells. Importantly, BRET2 values were similar between cells cotransfected with HCN4-GFP and HCN4-Rluc and with HCN2AUGUST 3, 2007 • VOLUME 282 • NUMBER 31  GFP and HCN2-Rluc (between cells containing homomeric HCN2 channels or HCN4 channels), suggesting that the distances between GFP and Rluc tags in both homomeric channels were similar. Levels of emission among GFP-tagged constructs and among Rluc-tagged constructs were similar, also suggesting that the overall levels of expression among constructs were similar (for both BRET2 and imaging experiments). As an additional negative control, we co-transfected CHO cells with wild type HCN2 or HCN4 and an N-ter