Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Understanding the interactions that occur between KCNE1 and KCNQ1 : stoichiometry, gating and dynamic… Westhoff, Maartje Flore Emilie 2019

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

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
24-ubc_2019_may_westhoff_maartje.pdf [ 22.45MB ]
Metadata
JSON: 24-1.0376582.json
JSON-LD: 24-1.0376582-ld.json
RDF/XML (Pretty): 24-1.0376582-rdf.xml
RDF/JSON: 24-1.0376582-rdf.json
Turtle: 24-1.0376582-turtle.txt
N-Triples: 24-1.0376582-rdf-ntriples.txt
Original Record: 24-1.0376582-source.json
Full Text
24-1.0376582-fulltext.txt
Citation
24-1.0376582.ris

Full Text

 i UNDERSTANDING THE INTERACTIONS THAT OCCUR BETWEEN KCNE1 AND KCNQ1: STOICHIOMETRY, GATING AND DYNAMIC MOVEMENTS  by  MAARTJE FLORE EMILIE WESTHOFF  B.Sc., University of Sussex, 2013     A DISSERTATION SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY  in  THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES  (Pharmacology and Therapeutics)   THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)     February 2019  © Maartje Flore Emilie Westhoff, 2019  ii The following individuals certify that they have read, and recommend to the Faculty of Graduate and Postdoctoral Studies for acceptance, the dissertation entitled: Understanding the interactions that occur between KCNE1 and KCNQ1: Stoichiometry, gating and dynamic movements.  submitted by Maartje Flore Emilie Westhoff  in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Pharmacology and Therapeutics  Examining Committee: Dr. David Fedida Supervisor  Dr. Eric Accili Supervisory Committee Member  Dr. Filip Van Petegem Supervisory Committee Member Dr. Ismail Laher University Examiner Dr. Edwin D.W. Moore University Examiner    iii Abstract The IKs current has an important role in repolarizing the cardiac action potential. KCNQ1 subunits form a tetrameric voltage-gated potassium channel, with which an accessory b-subunit, KCNE1, can interact. KCNE1 binding to KCNQ1 remarkably alters channel kinetics by delaying activation, increasing current density and removing inactivation.  There has always been confusion about how many KCNE1 subunits can associate with the KCNQ1 tetramer. Several groups have reported a strict fixed stoichiometry of two KCNE1 subunits to four KCNQ1 subunits. However, others have shown that the ratio varies depending on the concentration of KCNE1 subunits available. Using whole cell and single channel patch clamp recordings of tethered fusion constructs with different ratios of KCNE1:KCNQ1, as well as photo-crosslinking experiments, we show that up to four KCNE1 subunits can associate with the complex. Therefore, in vitro, IKs can have a variable stoichiometry.   In further photo-crosslinking experiments we show that two adjacent residues in KCNE1 interact with KCNQ1 in different channel states, open and closed. Additionally, we show that these interactions are likely taking place with the pore domain of the channel. We also confirm what other groups had proposed, that KCNE1 moves within the cleft of KCNQ1 during channel activation, since the rates of crosslinking change when the channels are held at more depolarized holding potentials.  Finally, we investigate whether or not all four voltage sensors have to activate and the complex undergo a concerted step in order for the pore to conduct current. We employ a mutation, E160R, which applies electrostatic repulsion to the positive charges of the  iv voltage sensor and keeps it in a resting conformation. By locking one, two, three and four voltage sensors down we show, via whole cell and single channel recordings, that the channel can conduct when only one voltage sensor is free to move. From these experiments, we provide additional evidence that the IKs channel gates in an allosteric fashion, where each additional voltage sensor movement results in a progressively higher open probability. Additionally, we propose that there is cooperativity between KCNQ1 subunits, where movement of one voltage sensor facilitates movement of a neighboring voltage sensor.      v Lay Summary Heartbeats are elicited in heart cells through electrical signals. These electrical signals are caused by ion channels: proteins which allow ions to pass into or out of the cell. One such ion channel, IKs, is made up of two protein components, and mutations in both have resulted in irregular heartbeats. The research presented in this thesis aims to elucidate how these two components interact. Firstly, we show that the ratio of the two proteins is variable rather than fixed, allowing for more flexibility. Additionally, our results suggest that the proteins interact differently when channels are open or closed. Finally, we show that the channel does not require a concerted movement to open. These results suggest that the IKs channel is highly flexible and can be adapted in a variety of different ways. This information may aid the development of better pharmaceuticals to restore normal electrical activity in the heart.    vi Preface All of the work presented in this thesis was conducted in the laboratory of Dr. David Fedida at the University of British Columbia in collaboration with Dr. Christopher Murray, Dr. Jodene Eldstrom, Emely Thompson and Robert Emes. Chapter 2: Unnatural amino acid photo-crosslinking of the IKs channel complex demonstrates a KCNE1: KCNQ1 stoichiometry of up to 4:4 In this work I was involved in acquisition of whole cell data, analysis and interpretation of data, producing figures and drafting and revising the manuscript. Christopher Murray was involved in the conception and design of experiments, acquisition of whole cell data, analysis and interpretation of data and drafting and revising the manuscript. Jodene Eldstrom was involved in conception and design of experiments, acquisition of single channel data, analysis and interpretation of data and drafting and revising the manuscript. Emely Thompson was involved in acquisition of single channel data and analysis and interpretation of data. Robert Emes was involved in acquisition of whole cell data and analysis of data. David Fedida was the supervisory author and was involved in the conception and design of experiments, analysis and interpretation of data and drafting and revising the manuscript.  A version of Chapter 2 has been published: Murray, C.I., Westhoff, M., Eldstrom, J., Thompson, E., Emes, R., Fedida, D. Unnatural amino acid photo-crosslinking of the IKs channel complex demonstrates a KCNE1:KCNQ1 stoichiometry of up to 4:4. Elife. 2016 Jan 23; 5: e11815. doi: 10.7554/eLife.11815.  vii Chapter 3: Photo-crosslinking of IKs demonstrates state-dependent interactions between KCNE1 and KCNQ1.  In this work I was involved in conception and design of experiments, acquisition of whole cell data, analysis and interpretation of data, producing figures and drafting and revising the manuscript. Christopher Murray was involved in conception and design of experiments, acquisition of whole cell data, analysis and interpretation of data and drafting and revising the manuscript. Jodene Eldstrom was involved in conception and design of experiments, acquisition of single channel data, analysis and interpretation of data and drafting and revising the manuscript. David Fedida was the supervisory author and was involved in the conception and design of experiments, analysis and interpretation of the data and drafting and revising the manuscript.  A version of Chapter 3 has been published: Westhoff, M., Murray, C.I., Eldstrom J., Fedida, D. Photo-Cross-Linking of IKs Demonstrates State-Dependent Interactions between KCNE1 and KCNQ1. Biophysical Journal. 2017 Jul 25; 113(2): 415-425. doi: 10.1016/j.bpj.2017.06.005. Chapter 4: Individual voltage sensor movements result in IKs channel conductance In this work I was involved in the conception and design of experiments, acquisition of whole cell data, acquisition of two electrode voltage clamp data, analysis and interpretation of data, producing figures and drafting and revising the manuscript. Jodene Eldstrom was involved in conception and design of experiments, acquisition of single channel data, analysis and interpretation of data and drafting and revising the manuscript. Christopher Murray was involved in conception and design of experiments, acquisition of  viii whole cell data, analysis and interpretation of data and revising the manuscript. Emely Thompson was involved in analysis and interpretation of single channel data. David Fedida was the supervisory author and was involved in conception and design of experiments, analysis and interpretation of data and drafting and revising the manuscript. For experiments in Chapter 4, oocytes were prepared in accordance with the University of British Columbia Animal Care Protocol (Certificate Number A18-0074). A version of Chapter 4 has been submitted for peer review. Westhoff, M., Eldstrom, J., Murray, C.I., Thompson, E., Fedida, D. IKs ion channel pore conductance can result from individual voltage sensor movements.     ix Table of Contents  Abstract ........................................................................................................................ iii Lay Summary ................................................................................................................ v Preface .......................................................................................................................... vi Table of Contents ......................................................................................................... ix List of Tables .............................................................................................................. xiv List of Figures ............................................................................................................. xv List of Abbreviations ................................................................................................. xix Acknowledgements ................................................................................................. xxiv Chapter 1: Introduction ................................................................................................ 1 1.1 Voltage-Gated Potassium Channels ................................................................. 1 1.1.1 Physiological role of voltage-gated ion channels ....................................... 1 1.1.2 Structure and basic properties of Kv channels .......................................... 1 1.1.3 KCNQ1 channel ......................................................................................... 4 1.2 KCNE b-subunit interactions with KCNQ1 ........................................................ 5 1.2.1 KCNE1 ...................................................................................................... 5 1.2.2 KCNE2 ...................................................................................................... 9 1.2.3 KCNE3 .................................................................................................... 10 1.2.4 KCNE4 .................................................................................................... 11 1.2.5 KCNE5 .................................................................................................... 12 1.2.6 Structural differences between b-subunits ............................................... 13 1.3 Stoichiometry of IKs ......................................................................................... 14 1.3.1 Fixed Stoichiometry (2 KCNE1:4 KCNQ1)............................................... 14 1.3.2 Variable Stoichiometry ............................................................................. 16 1.4 Localizing interactions between KCNE1 and KCNQ1 ..................................... 17 1.4.1 Pore Domain............................................................................................ 17 1.4.2 Voltage Sensor Domain ........................................................................... 18  x 1.4.3 State-dependent interactions between KCNE1 and KCNQ1 ................... 19 1.4.4 Proposed location of KCNE1 in IKs channel complex ............................... 20 1.5 Gating of KCNQ1 in the presence and absence of KCNE1 ............................ 21 1.5.1 Gating of KCNQ1 alone and with KCNE1 ................................................ 21 1.5.2 ATP is required for pore opening ............................................................. 26 1.5.3 PIP2 is required for VS-PD coupling ........................................................ 27 1.5.4 Effect of KCNE1 on VS-PD coupling ....................................................... 28 1.6 Scope of the thesis ......................................................................................... 30 Chapter 2: Unnatural amino acid photo-crosslinking of the IKs channel complex demonstrates a KCNE1:KCNQ1 stoichiometry of up to 4:4 .................................... 31 2.1  Introduction ..................................................................................................... 31 2.2  Materials and Methods ................................................................................... 33 2.2.1  Reagents ................................................................................................. 33 2.2.2 Molecular Biology .................................................................................... 33 2.2.3  Cell culture and transfections .................................................................. 34 2.2.4 Patch-clamp electrophysiology ................................................................ 35 2.2.5 Solutions .................................................................................................. 36 2.2.6 Data analysis ........................................................................................... 37 2.2.7 Western blot ............................................................................................ 37 2.2.8 Co-immunoprecipitation ........................................................................... 38 2.2.9  Statistics .................................................................................................. 39 2.3 Results ............................................................................................................ 39 2.3.1 Characterization of IKs channel constructs ............................................... 39 2.3.2 Reduced single-channel conductance and latencies of EQQ and EQQQQ IKs channel complexes ........................................................................................... 41 2.3.3 Expression and characterization of Bpa-incorporated IKs ........................ 46 2.3.4 UV-dependent channel crosslinking ........................................................ 49 2.4 Discussion ...................................................................................................... 53  xi 2.4.1 KCNQ1 regulation by KCNE1 .................................................................. 55 2.4.2 Implications of the single channel studies ............................................... 56 2.4.3 Stoichiometry of the IKs channel complex in cardiomyocytes ................... 58 2.4.4 Conclusions ............................................................................................. 59 Chapter 3: Photo-crosslinking of IKs demonstrates state-dependent interactions between KCNE1 and KCNQ1 ...................................................................................... 60 3.1 Introduction ..................................................................................................... 60 3.2 Materials and Methods ................................................................................... 62 3.2.1 Reagents ................................................................................................. 62 3.2.2 Molecular Biology .................................................................................... 63 3.2.3 Cell culture and transfections .................................................................. 63 3.2.4 Patch-clamp electrophysiology ................................................................ 64 3.2.5 Solutions .................................................................................................. 64 3.2.6 Data analysis ........................................................................................... 65 3.2.7 Statistics .................................................................................................. 66 3.3 Results ............................................................................................................ 66 3.3.1 F57Bpa KCNE1 crosslinks to KCNQ1 in the resting state ....................... 66 3.3.2 F57Bpa has a reduced rate of crosslinking in a higher open probability mutant…. ............................................................................................................... 70 3.3.3 The rate of crosslinking with F57Bpa is reduced after longer depolarizing pulses…. ................................................................................................................ 71 3.3.4 F57 KCNE1 moves across the cleft before channel opening................... 72 3.3.5 F56 interacts with KCNQ1 in the open state ............................................ 76 3.3.6 F56 crosslinks to E1R/R4E KCNQ1 ........................................................ 78 3.3.7 F56 does not interact with KCNQ1 in a preopen closed state ................. 79 3.4 Discussion ...................................................................................................... 80 3.4.1  Comparing F56Bpa and F57Bpa ............................................................. 81 3.4.2 Investigating crosslinking with F57Bpa .................................................... 81  xii 3.4.3 Investigating crosslinking with F56Bpa .................................................... 83 3.4.4 Regions of interaction between KCNE1 and KCNQ1 .............................. 84 3.4.5 Regulation of KCNQ1 by KCNE1 ............................................................ 86 3.4.6 Conclusions ............................................................................................. 87 Chapter 4: Individual voltage sensor movements result in IKs channel conductance ................................................................................................................ 89 4.1 Introduction ..................................................................................................... 89 4.2 Materials and Methods ................................................................................... 93 4.2.1 Chemicals ................................................................................................ 93 4.2.2 Molecular Biology .................................................................................... 93 4.2.3 Cell Culture and Transfection .................................................................. 94 4.2.4 Oocyte Preparation .................................................................................. 94 4.2.5 Electrophysiology Solutions ..................................................................... 95 4.2.6 Patch Clamp Electrophysiology ............................................................... 96 4.2.7 TEA+ Experiments.................................................................................... 97 4.2.8 MTSET Experiments ............................................................................... 97 4.2.9 VCF Experiments .................................................................................... 97 4.2.10    Data Analysis ........................................................................................ 98 4.2.11    Modeling ............................................................................................... 99 4.3 Results .......................................................................................................... 100 4.3.1 Channels with the E160R mutation in all four Q1 subunits are non-functional.. ........................................................................................................... 100 4.3.2 IKs with one, two or three VSs held down produce functional channels . 102 4.3.3 Subunits with the E160R mutation are not excluded from channel assembly.. ............................................................................................................ 106 4.3.4 Activation kinetics of IKs when one, two or three VSs are held down. .... 106 4.3.5 Rate of deactivation increases progressively as multiple VSs are held down…… ............................................................................................................. 108  xiii 4.3.6 Single channel recordings show reduced conductance of channels with E160R subunits ................................................................................................... 108 4.3.7 F57W, a depolarizing E1 mutant, corroborates the behavior of E160R . 113 4.3.8 VSs with the E160R mutation are locked down ..................................... 114 4.4 Discussion .................................................................................................... 119 4.4.1 The E160R mutation in one, two or three Q1 subunits holds them in a resting state, but does not prevent IKs current.................................................................. 119 4.4.2 Previous studies and contrasting conclusions. ...................................... 122 4.4.3 Considering the two steps of fluorescence movement. ......................... 123 4.4.4 Loose coupling between VS activation and pore opening and an allosteric model…. .............................................................................................................. 125 4.4.5 Functional implications of an allosteric model of activation gating. ........ 130 4.4.6 Conclusions ........................................................................................... 131 Chapter 5: Discussion .............................................................................................. 132 5.1  Summary of findings ..................................................................................... 132 5.2 Pharmacological relevance of variable stoichiometry ................................... 133 5.3 Stoichiometry of IKs in vivo ............................................................................ 135 5.4 Interaction with several KCNE β-subunits within the same complex............. 137 5.5 Allosteric gating of IKs .................................................................................... 138 5.6  Cooperativity between KCNQ1 subunits ....................................................... 139 5.7 Limitations .................................................................................................... 141 5.8 Future directions ........................................................................................... 142 5.9  Conclusions .................................................................................................. 143 Bibliography .............................................................................................................. 145 Appendix A: Supplementary Figures for Chapter 2 ............................................... 160 Appendix B: Supplementary Figures for Chapter 3 ............................................... 164 Appendix C: Supplementary Figures for Chapter 4 ............................................... 167   xiv List of Tables Table 2.1. Summary of single channel parameters. ...................................................... 46 1 Table 3.1. Crosslinking rate constants for F57Bpa at different membrane potentials. .. 69  Table 4.1. V1/2 of activation for Q1 constructs in the presence and absence of E1. .... 105    xv List of Figures Figure 1.1. Characterization of the IKs channel................................................................ 2 Figure 1.2. Selectivity filter of the KcsA channel. ............................................................ 3 Figure 1.3. Sequence alignment of human KCNE1, KCNE2, KCNE3, KCNE4 and KCNE5. ...................................................................................................................................... 13 Figure 1.4. Charge reversal mutations can lock voltage sensors in intermediate and fully activated states. ............................................................................................................ 25  Figure 2.1. Additional β-subunits can alter IKs channel complex gating properties. ....... 40 Figure 2.2. EQ, EQQ and EQQQQ show clear differences in single channel behavior. 42 Figure 2.3. Co-expression of additional KCNE1 subunits restores wild type IKs single channel behavior to EQQ and EQQQQ. ....................................................................... 44 Figure 2.4. Expression and characterization of F57Bpa KCNE1 in IKs. ......................... 48 Figure 2.5. The IKs channel complex does not have a restricted β-subunit stoichiometry. ...................................................................................................................................... 51 Figure 2.6. Increasing KCNE1 has a linear effect on depolarizing shift in the V1/2 of activation. ...................................................................................................................... 55 a s Figure 3.1. State-dependent crosslinking of F57Bpa IKs shows preferential interaction in the resting state. ........................................................................................................... 68 Figure 3.2. The crosslinking rate of F57Bpa KCNE1 at +60 mV is reduced in a higher open probability mutant. ................................................................................................ 71 Figure 3.3. The crosslinking rate of F57Bpa is reduced during longer depolarizing pulses. ...................................................................................................................................... 73  xvi Figure 3.4. Wild-type and F57Bpa IKs channels activate more rapidly when held at depolarized potentials before activation. ....................................................................... 74 Figure 3.5. F57Bpa KCNE1 has reduced interaction with KCNQ1 when channels are held at depolarizing potentials before opening...................................................................... 75 Figure 3.6. F56Bpa KCNE1 interacts with KCNQ1 in the open state. ........................... 77 Figure 3.7. F56Bpa crosslinks to E1R/R4E KCNQ1 in the open state. ......................... 78 Figure 3.8. F56Bpa does not interact with KCNQ1 in a pre-open closed state. ............ 80 Figure 3.9. Crosslinking rates of F57Bpa plotted against the G-V and F-V of IKs. ......... 87  Figure 4.1. Models of Ion Channel Gating..................................................................... 91 Figure 4.2. E160R EQ* is a non-functional channel. ................................................... 101 Figure 4.3. IKs channel complexes with one, two or three E160R mutations produce currents. ...................................................................................................................... 104 Figure 4.4. Activation and deactivation kinetics of complexes with one, two or three VSs locked down. ............................................................................................................... 107 Figure 4.5. Single channel recordings of channels with one, two or three VSs locked down. .......................................................................................................................... 110 Figure 4.6. Subconductance analysis of a channel with two or three VSs locked down. .................................................................................................................................... 112 Figure 4.7. Extracellular MTSET is unable to modify subunits containing the E160R mutation. ..................................................................................................................... 116 Figure 4.8. Fluorescence changes are not detected in subunits containing the E160R mutation. ..................................................................................................................... 118  xvii Figure 4.9. A simple allosteric model with voltage sensor cooperativity most closely reproduces experimental results. ................................................................................ 128  Appendix A, Figure 1. All-points amplitude histograms. .............................................. 160 Appendix A, Figure 2. Subconductance analysis of EQQ and EQ demonstrates that as the number of KCNE1 subunits decreases so does higher conductance state occupancy. .................................................................................................................................... 160 Appendix A, Figure 3. M62W KCNE1 does not alter IKs channel activation compared to wild type. ..................................................................................................................... 161 Appendix A, Figure 4. F57Bpa KCNE1 does not alter channel gating of KCNQ1, EQQ and EQ compared to wild type KCNE1. ...................................................................... 162 Appendix A, Figure 5. UV-rundown is consistent across all the IKs channel configurations. .................................................................................................................................... 163 Appendix A, Figure 6. F57Bpa KCNE1 covalently crosslinks with KCNQ1. ................ 163 s Appendix B, Figure 1. Incorporation of Bpa at positions F56 and F57 in KCNE1 is well tolerated. ..................................................................................................................... 164 Appendix B, Figure 2. Wild type UV rundown is consistent between -90 mV and +60 mV. .................................................................................................................................... 165 Appendix B, Figure 3. Characterization of E1R/R4E IKs. ............................................. 166 s Appendix C, Figure 1. When expressed without E1, channel complexes with one or two E160R mutations produce functional currents. ........................................................... 167 Appendix C, Figure 2. Wt QQ alone and co-expressed with E1.................................. 168  xviii Appendix C, Figure 3. Activation kinetics of E160R QQQ*Q* + E1 are similar to E160R Q*Q + E1. ................................................................................................................... 169 Appendix C, Figure 4. TEA+ sensitivity reveals mutated E160R subunits are not excluded from channel assembly. .............................................................................................. 170 Appendix C, Figure 5. Representative single exponential fits to activating current waveforms................................................................................................................... 171 Appendix C, Figure 6. Representative single exponential fits to tail currents. ............. 172 Appendix C, Figure 7. Channel complexes with tethered E1s produce similar results to the untethered constructs............................................................................................ 173 Appendix C, Figure 8. Additional single channel information. ..................................... 174 Appendix C, Figure 9. Effect of one, two or four KCNE1 F57W mutations on the IKs channel. ...................................................................................................................... 176 Appendix C, Figure 10. Additional MTSET information. .............................................. 178 Appendix C, Figure 11. VCF recordings of G219C Q-Q + E1 channels. ..................... 179    xix List of Abbreviations a – Alpha b – Beta t – Time constant µA – Microamps °C – Degrees in Celsius DF – Change in fluorescence µM – Micromolar aa – Amino acid  Act – Activation ANOVA – Analysis of variance ATP – Adenosine triphosphate Bpa –  p-Benzoyl-L-phenylalanine C-terminus – Carboxy-terminus of a protein CaCl2 – Calcium dichloride CaM – Calmodulin cAMP – Cyclic adenosine monophosphate CiVSP – Voltage sensitive lipid phosphatase from Ciona intestinalis CTX – Charybdotoxin  D1 – Modeling rate constant DDM – Dodecylmaltoside Deact – Deactivation DMSO – Dimethyl sulfoxide Do – Modeling rate constant DTT – Dithiothreitol E’ – KCNE1 subunits mutated with F57W E1 – KCNE1  xx EGTA – Ethylene glycol-bis(2-aminoethylether)-N,N,N′,N′-tetraacetic acid EQ – KCNE1 tethered to one KCNQ1 (KCNE1-KCNQ1) EQQ – KCNE1 tethered to two KCNQ1s (KCNE1-KCNQ1-KCNQ1) EQQQQ – KCNE1 tethered to four KCNQ1s (KCNE1-KCNQ1-KCNQ1-KCNQ1-KCNQ1) F – Fluorescence F-V – Fluorescence-voltage relationship FAF – Familial Atrial Fibrillation G – Conductance G-V – Conductance-voltage relationship GFP – Green fluorescent protein Go – Modeling rate constant GSH – Glutathione GTP – Guanosine triphosphate HCl – Hydrogen chloride HEK – Human embryonic kidney cells HEPES – 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid hESC-CMs – Cardiomyocytes derived from human embryonic stem cells HRP – Horseradish peroxidase Hrs – Hours Hz – Hertz I-V – Current-voltage relationship IKr – Rapid delayed rectifier potassium current IKs – Slow delayed rectifier potassium current iPSC – Induced pluripotent stem cell JLNS – Jervell and Lange-Nielson Syndrome k – Slope factor  K+ – Potassium   xxi KCl – Potassium chloride KCNE1L – KCNE1-like  kHz – Kilohertz Kir – Inwardly rectifying potassium current KOH – Potassium hydroxide KRD – Rate of rundown Kv – Voltage-gated potassium channel KXL – Rate of crosslinking LM – Mouse ltk- fibroblast cells LQT – Long QT  LQTS – Long QT Syndrome M – Molar MW – Megaohm MES – 2-(N-Morpholino)ethanesulfonic acid MgCl2 – Magnesium chloride ml – Milliliter mM – Millimolar ms – Millisecond MTSET – [2-(Trimethylammonium)ethyl]methanethiosulfonate Bromide mV – Millivolts N-terminus – Amino-terminus of protein nA – Nanoampere Na2-ATP – Adenosine 5’-triphosphate disodium salt NaCl – Sodium chloride NaOH – Sodium hydroxide NMDG – N-Methyl-D-glucamine NMR – Nuclear magnetic resonance  xxii pA – Picoampere PBS – Phosphate-buffered saline PD – Pore domain PIP2 – Phosphatidylinositol 4,5-bisphosphate Po – Open probability pS – picosemens Q – Charge Q-V – Charge-voltage relationship Q* – KCNQ1 subunits mutated with E160R Q1 – KCNQ1 QQ – Two KCNQ1s tethered together (KCNQ1-KCNQ1) QQQQ – Four KCNQ1s tethered together (KCNQ1-KCNQ1-KCNQ1-KCNQ1) RFP – Red fluorescent protein RS – Aminoacyl-tRNA-synthetase s – Seconds SQT – Short QT SQTS – Short QT Syndrome t1/2 – Half maximum time TAG – Amber stop codon TCEP – Tris-(2-carboxyethyl)phosphine TEA+ – Tetraethylammonium chloride TFP – Teal fluorescent protein tRNA – Transfer ribonucleic acid TSA – tsA201 transformed human embryonic kidney 293 cells UAA – Unnatural amino acid UV – Ultraviolet  V – Voltage  xxiii V1/2 – Voltage at half maximal activation VCF – Voltage clamp fluorometry VGKC – Voltage-gated potassium channel VS – Voltage sensor domain W – Watt wt – Wild type X – Stop codon  Amino acid name Abbreviation Alanine Ala (A) Arginine Arg (R) Asparagine Asn (N) Aspartate Asp (D) Cysteine Cys (C) Glutamine Gln (Q) Glutamate Glu (E) Glycine Gly (G) Histidine His (H) Isoleucine Ile (I) Leucine Leu (L) Lysine Lys (K) Methionine Met (M) Phenylalanine Phe (F) Proline Pro (P) Serine Ser (S) Threonine Thr (T) Tyrosine Tyr (Y) Tryptophan Trp (W) Valine Val (V)      xxiv Acknowledgements I would like to thank my supervisor, Dr. David Fedida, for his guidance and support throughout all aspects of my degree. Additionally, I would like to send my gratitude to both Dr. Chris Murray and Dr. Jodene Eldstrom for their mentorship and friendship over the past years. I have learned so much from all of you, and will always appreciate your encouragement.  Thank you to my supervisory committee members, Dr. Eric Accili and Dr. Filip Van Petegem, for their friendly suggestions and discussions at our yearly meetings, as well as for reviewing my thesis.  I would also like to thank UBC for my funding via the Four Year Fellowship program. Thank you to MITACS for funding my three-month stay to perform research at the UPMC in Paris, France via the Globalink Research Award. Thank you to Dr. Stéphane Hatem and Dr. Elise Balse, as well as other members of the Hatem lab, for their support during my stay in Paris. I would like to thank my partners in crime in the lab, Emely and Logan, as well as other members of the Fedida and Kurata labs, for making these past five and a half years an amazing experience. Especially for the fun times we had outside of the lab and at conferences. I have been lucky to call all of you my friends.  Finally, I would like to express my gratitude to my family, especially my parents, for always encouraging me in my goals. I couldn’t have done this without you, and I will always be thankful for your support.    1 Chapter 1: Introduction 1.1 Voltage-Gated Potassium Channels 1.1.1 Physiological role of voltage-gated ion channels Ion channels initiate or support a wide range of cellular processes within the body, including many aspects of cell signaling and excitability (Hille, 2001). Voltage-gated ion channels, in particular, respond to changes in the membrane potential, opening and closing in a time- and voltage-dependent manner (Grant, 2009). During the cardiac action potential, for example, voltage-gated sodium channels open rapidly but transiently, allowing positively charged sodium ions to flow briefly into the cell, depolarizing the membrane (Grant, 2009). This is represented by the initial spike in the action potential (Grant, 2009). In order to bring the membrane back to a resting potential, the depolarization of the membrane causes voltage-gated potassium (Kv) channels, among other ion channels, to open and allow positive potassium (K+) ions to flow out of the cell, which is represented during the repolarization phase of the cardiac action potential (Tristani-Firouzi et al., 2001, Grant, 2009).   1.1.2 Structure and basic properties of Kv channels All Kv channels have a similar structure. The channel subunits are made up of six transmembrane domain helices: S1-S4 form the voltage sensor domain (VS) and S5-S6 form the pore domain (PD) (Fig. 1.1A). Four of these subunits come together to form a tetrameric Kv channel (Kuang et al., 2015).   2 The VS of Kv channels detect changes in the membrane potential, due to the positive charges located within the S4 (Kuang et al., 2015). These changes in membrane potential lead to a conformational change of the VS, which is translated to the PD, allowing the pore to open or close (Kuang et al., 2015, Jensen et al., 2012).    Figure 1.1. Characterization of the IKs channel. (A) Cartoon diagram of KCNQ1 and KCNE1 subunits. (B) Representative current traces of KCNQ1 alone (upper) and co-expressed with KCNE1 (lower) to form the IKs current. The currents were obtained via a 4 s isochronal activation protocol from -80 to +100 mV, with a tail current at -40 mV. (C) Conductance-Voltage (G-V) relationship of KCNQ1 alone (open green squares) and co-expressed with KCNE1 (closed green squares) (n = 5-11).   A region located near the extracellular surface of the pore contributes to the high selectivity for potassium over other ions in Kv channels (Heginbotham et al., 1994, Doyle et al., 1998). This region, known as the selectivity filter, is made up of five amino acids which are conserved among potassium channels, TVGYG (As shown for the bacterial KcsA channel in Fig. 1.2)(Heginbotham et al., 1994, Doyle et al., 1998, Kuang et al., 2015). Mutations in this region, have resulted in a decrease in the selectivity for K+ ions  3 (Heginbotham et al., 1994). As K+ ions enter the selectivity filter, water molecules dissociate, allowing interactions to take place between the K+ ions and the carbonyl oxygen atoms within the filter (Doyle et al., 1998, Kuang et al., 2015). The hydrophobic nature of the pore between the cytoplasm and the filter is thought to provide a “low resistance pathway” for efficient movement of K+ ions (Doyle et al., 1998).   Figure 1.2. Selectivity filter of the KcsA channel. (A) Two domains of the tetrameric bacterial KcsA channel (PDB: 1BL8)(Doyle et al., 1998) are displayed using Pymol. (B) Zoomed in image of the selectivity filter. The KcsA domains are in green, the residues of the selectivity filter are in orange and the potassium ions in the filter are represented by the purple spheres.  In the simplest model, Kv channels classically visit three gating states: resting, where VSs are down and the channel pore is closed, activated, where VSs have moved and the channel pore is open, and inactivated, where channels are activated but non-conducting. Three types of inactivation exist for potassium channels: N-type, C-type and U-type. N-type inactivation, also known as ball and chain inactivation, is where the N-terminus of the channel subunit occludes the intracellular side of the pore, and thus makes the  4 channel unable to conduct current (Hoshi et al., 1991, Gebauer et al., 2004). The slower C-type inactivation is thought to be caused by conformational changes in the selectivity filter in the outer part of the pore (Hoshi et al., 1990, López-Barneo et al., 1993, Cuello et al., 2010). U-type inactivation is another slow form of inactivation. It is represented by the presence of a U-shaped relationship between inactivation and voltage in Kv2.1, Kv3.1 and Shaker channels. This suggests that inactivation is faster at a more hyperpolarized membrane potentials, when channels are in a closed state (Klemic et al., 2001, Kurata et al., 2002, Cheng et al., 2011). The mechanism of U-type inactivation is not yet well understood, however the N-terminus (Kurata et al., 2002) as well as residues in the S3-S4 and S5-S6 linkers (Cheng et al., 2011) have been shown to play a role.   1.1.3 KCNQ1 channel The KCNQ family is a group of Kv channels, made up of five members, KCNQ1-5, showing strong sequence homology. KCNQ1, also known as KvLQT1 and Kv7.1, was first discovered in 1996 by Wang et al., via positional cloning methods. The gene was found on chromosome 11, as mutations were associated with cardiac abnormalities (Wang et al., 1996b). KCNQ1 is expressed throughout the body, including the heart, stomach, thyroid, ear among other tissues (Wrobel et al., 2012, Lundquist et al., 2005, Fröhlich et al., 2011). Due to its role in repolarizing the cardiac action potential, several loss-of-function mutations in KCNQ1 alone lead to long QT (LQT) syndrome type 1, whereby the QT interval is prolonged, leading to an increased susceptibility to arrhythmias (Wu et al.,  5 2016). Additionally, homozygous knockout KCNQ1-/- mice have hypothyroidism (Fröhlich et al., 2011), an enlarged stomach, low stomach acid, and hearing loss (Lee et al., 2000). When expressed alone, KCNQ1 produces a fast activating and deactivating current, which undergoes inactivation (Fig. 1.1B). The current produced is also of a very low conductance (Fig. 1.1B). However, no endogenous current in the body seems to be represented by KCNQ1 channels alone (Abbott, 2014). Instead, KCNQ1’s functional and physiological relevance becomes apparent through its association with several b-subunits, KCNE1-5, which alter channel kinetics in a variety of different ways (See Section 1.2).    1.2 KCNE b-subunit interactions with KCNQ1 1.2.1 KCNE1 Initially, when KCNE1, also known as MinK and Isk, was first discovered in 1988 it was thought to be a potassium channel in its own right, albeit with a very unusual single transmembrane domain structure when compared with all other known Kv channels (Takumi et al., 1988). Due to the presence of endogenous KCNQ1 channels in Xenopus oocytes, the injection of KCNE1 led to the appearance of slow delayed IKs currents, leading researchers to believe KCNE1 alone was sufficient to conduct potassium ions (Takumi et al., 1988). Later on, by expressing KCNE1 in cells lacking endogenous KCNQ1, as well as co-localization of KCNE1 and KCNQ1 in many tissues throughout the body including the heart, inner ear and kidney, it became clear that KCNQ1 forms the channel alpha subunit, and the association of a b-subunit, KCNE1, produces IKs currents  6 (Barhanin et al., 1996, Sanguinetti et al., 1996). Unlike the KCNQ1 current alone, the whole cell recordings of IKs current show slow activation, a depolarized shift in voltage dependence, increased conductance, and removal of channel inactivation (Fig. 1.1B-C)(Barhanin et al., 1996, Sanguinetti et al., 1996, Yang and Sigworth, 1998, Pusch, 1998, Tristani-Firouzi and Sanguinetti, 1998). At the single channel level, IKs currents were shown to have a flickering phenotype as well as a long latency to opening (>1.5 s at +60 mV and room temperature) and an extremely low open probability (approximately 15% after 4 s at +60 mV) (Yang and Sigworth, 1998, Werry et al., 2013). Additionally, during an activating pulse, channels are able to visit multiple sub-conducting levels, prior to full IKs conductance, indicating a complex gating pathway  (Yang and Sigworth, 1998, Werry et al., 2013).  Unlike other Kv channels, KCNE1 assembles to KCNQ1 post-translationally, either in the endoplasmic reticulum or the golgi apparatus (Krumerman et al., 2004, Chandrasekhar et al., 2006, Vanoye et al., 2010). Additionally, post-translational modifications of the channel complex are required to maintain normal channel function. For example, in vivo, KCNE1 is glycosylated at residue T7. When this site is mutated to prevent glycosylation (T7A), trafficking of the IKs complex is impaired (Chandrasekhar et al., 2011).  Heart Examining mRNA levels of KCNE1 shows that it expressed alongside KCNQ1 in the ventricles and atria of the human heart (Lundquist et al., 2005). Interestingly, in cardiomyopathic heart tissue, KCNE1 expression significantly increases, suggesting it is upregulated in the failing heart (Lundquist et al., 2005). The IKs current is important in  7 regulation of normal heart rhythm. This is due to its role, alongside other voltage gated potassium currents, to move positive potassium ions out of the cell during the repolarization phase of the cardiac action potential (Grant, 2009).  Due to the role of the IKs current in cardiac repolarization, mutations in both KCNE1 and KCNQ1 cause disruption of the normal heart rhythm. Loss-of-function mutations in KCNE1 result in long QT (LQT) syndrome type 5. Electrophysiology recordings of these mutations sometimes show a reduced IKs current density (Moss and Kass, 2005, Harmer et al., 2010, Wu et al., 2006b), or a shift in the voltage dependence to more positive potentials (Wu et al., 2006b). Predominantly, though, LQT mutations are thought to cause their actions via reduced trafficking of IKs channel (Bianchi et al., 1999, Krumerman et al., 2004), or impaired assembly of KCNE1 with KCNQ1 (Harmer et al., 2010).  IKs has been found to be most important at higher heart rates. Here, the fight or flight response releases adrenaline to activate b-adrenergic receptors, leading to a cascade whereby the IKs current increases (Madamanchi, 2007, Marx et al., 2002). This increase in current sets up a repolarization reserve, which in turn, shortens the duration of the action potential (Stengl et al., 2003, Jost et al., 2005, Silva and Rudy, 2005). Due to this, symptoms of a cardiac abnormality may only be displayed when an individual is exposed to exercise or high stress (Wit et al., 1975, Hekkala et al., 2010). Inner Ear In the inner ear, the fluid in the cochlear duct, or endolymph, has a very high concentration of potassium ions, this sets up a gradient to allow potassium to flow into hair cells, depolarizing them; a process required for normal hearing transduction (Zdebik et al.,  8 2009). It is thought that IKs is responsible for maintaining the potassium level in the endolymph (Marcus and Shen, 1994, Vetter et al., 1996, Zdebik et al., 2009). Thus, disruption of KCNE1 can be associated with loss of hearing. For example, homozygous mutations in KCNE1 can lead to Jervell and Lange-Nielson syndrome (JLNS), a type of LQT syndrome associated with deafness (Jervell and Lange-Nielsen, 1957, Schulze-Bahr et al., 1997, Tyson et al., 1997, Neyroud et al., 1997). Homozygous KCNE1 knockout mice also exhibit deafness (Vetter et al., 1996), with a similar pathology to that seen in the inner ear of JLNS patients (Vetter et al., 1996, Friedmann et al., 1966). Kidney KCNE1 is also located within the proximal tubule of the kidney, where it is thought to assemble with KCNQ1 (Vallon et al., 2001) or other potassium channels (Neal et al., 2011). Sodium coupled transport of glucose in the proximal tubule depolarizes the membrane potential of the epithelial cells (Lang and Rehwald, 1992, Vallon et al., 2001). To counteract this and maintain normal absorption and secretion, potassium channels, such as IKs, allow potassium ions to flow from the proximal tubule to the lumen, repolarizing the cells (Lang and Rehwald, 1992, Vallon et al., 2001). To further underlie the important role of potassium channels in the kidney, KCNE1-/- knockout mice are hypokalemic (Arrighi et al., 2001, Warth and Barhanin, 2002). This leads to an imbalance of ions, which causes a disparity between absorption and secretion in the kidney, resulting in dehydration due to water loss (Arrighi et al., 2001, Warth and Barhanin, 2002).    9 1.2.2 KCNE2 KCNE2, also known as MiRP1, when co-expressed with KCNQ1, produces very small constitutively active currents in mammalian cells (Bendahhou et al., 2005, Tinel et al., 2000). In oocytes, it was shown that increasing the amount of KCNE2 injected leads to subsequently smaller current amplitudes (Wu et al., 2006a).  As with KCNE1 and KCNQ1, KCNE2 has also been localized to the human ventricle (Jiang et al., 2004). Co-precipitation experiments of cells co-expressing KCNQ1 with KCNE1 and KCNE2 show that both b-subunits can interact with KCNQ1 (Tinel et al., 2000). Additionally, in adult rat ventricular myocytes, KCNE2 can be colocalized with the IKs channel (Wu et al., 2006a). Whole cell recordings of KCNQ1 + KCNE1 + KCNE2, shows a shift of activation towards positive potentials as well as a faster rate of deactivation (Toyoda et al., 2006). Therefore, KCNE2 may play an important role in suppressing the IKs current in the heart.  The physiological role of KCNE2 in the heart is represented through several disease-causing mutations. A gain-of-function mutation in KCNE2, R27C, has been reported in patients with familial atrial fibrillation (Yang et al., 2004). Additionally, I57T KCNE2, a long QT mutation, was shown to reduce the constitutive nature of the KCNE2 + KCNQ1 current and introduce voltage dependence of current activation (Tinel et al., 2000, Abbott et al., 1999).  KCNE2 is also found alongside KCNQ1 in non-excitable cells within the stomach and thyroid of mice (Roepke et al., 2006, Roepke et al., 2009, Purtell et al., 2012). Here, the constitutively active “leak” behavior of the KCNE2 + KCNQ1 current, allows potassium  10 ions to continuously efflux from the cell, maintaining a hyperpolarized membrane potential (Purtell et al., 2012, Liin et al., 2015). In thyroid epithelial cells, this negative membrane potential is required in order to allow uptake of the iodide ion, a process required for normal thyroid function (Purtell et al., 2012).  1.2.3 KCNE3 Co-expression of KCNE3, also known as MiRP2, with KCNQ1 results in a large constitutively active current (Bendahhou et al., 2005), with channels only closing at extremely negative membrane potentials (even more hyperpolarized than -120 mV), far from the physiological range (-80 mV to +40 mV)(Barro-Soria et al., 2015). High levels of KCNE3 mRNA were found in the human kidney, prostate and liver, with low expression in the heart (Lundquist et al., 2006). Despite a low expression of KCNE3 in the heart, mutations in this subunit have been implicated in atrial fibrillation (V17M) and Brugada syndrome (R99H), however only due to its interaction with Kv4.3, not KCNQ1 (Lundby et al., 2008, Delpón et al., 2008).  The constitutive nature of the KCNE3 + KCNQ1 current allows KCNQ1 to play a role in non-excitable cells (as discussed for KCNE2 in 1.2.2)(Liin et al., 2015, Kroncke et al., 2016). The voltage-insensitive KCNE3 + KCNQ1 complex is required for potassium efflux from epithelial cells (Preston et al., 2010, Kroncke et al., 2016). This allows the Na+/K+ pump to pump potassium back in and sodium ions out of the cell, by a process known as “potassium recycling” (Preston et al., 2010, Kroncke et al., 2016). Following this, the Na+/K+/Cl- co-transporter allows chloride ions to enter the cell, and subsequently be  11 secreted into the lumen, which is required for many processes within the body (Preston et al., 2010, Kroncke et al., 2016). Due to the expression of KCNE3 and KCNQ1 in the epithelia of the trachea in mice (Grahammer et al., 2001, Preston et al., 2010), disruption of KCNE3, and thereby the “potassium recycling” to allow secretion of chloride ions, may be a factor in the development of cystic fibrosis (Preston et al., 2010, Kroncke et al., 2016).  1.2.4 KCNE4 KCNE4, also known as MiRP3, and KCNE5 are both less studied than KCNE1-3. KCNE4 acts as an inhibitor of the KCNQ1 channel, with the channel complex conducting current only at potentials greater than +50 mV (Bendahhou et al., 2005, Grunnet et al., 2002). In mammalian cells, co-assembly of KCNE4 with KCNQ1 results in a similar depolarizing effect on activation as seen with IKs (Bendahhou et al., 2005). In oocytes, KCNE4 still inhibits channel conductance, yet does not affect the activation kinetics (Grunnet et al., 2005).  Details of the physiological role of KCNE4 are still relatively limited. By investigating mRNA levels, Radicke et al. (2006) showed that there was an extremely high expression of KCNE4 in human non-failing hearts (approximately 60 times higher than KCNE1, KCNE3 and KCNE5)(Radicke et al., 2006). Interestingly, in failing hearts this was significantly reduced, indicating that there is perhaps a regulatory mechanism involved in the expression of b-subunits in failing hearts (Radicke et al., 2006).   12 Additionally, a single nucleotide polymorphism in KCNE4, E145D, was discovered to be a risk factor for atrial fibrillation in the Chinese Han and Uyghur populations (Zeng et al., 2005, Zeng et al., 2007, Mao et al., 2013). E145D, is proposed to overturn the inhibitory role of KCNE4, increasing the conductance of KCNQ1 (Ma et al., 2007).    1.2.5 KCNE5 When KCNE5 was first discovered, it was originally called the KCNE1-like (KCNE1L) gene, due to its similarities with KCNE1 (Piccini et al., 1999). However, more like KCNE4, in mammalian cells, co-expression of KCNE5 with KCNQ1 results in a suppressed current, which is slow to activate, has a large depolarizing shift in the activation curve and fast deactivation kinetics (Bendahhou et al., 2005, Angelo et al., 2002). KCNE5 can be found throughout the human heart (Piccini et al., 1999) and mutations have been implicated in cardiac arrhythmia syndromes. For example, mutations in KCNE5 were found to be risk factors for sudden cardiac death (Lin et al., 2018), as well as Brugada syndrome (Ohno et al., 2011) Interestingly, when co-expressing KCNE5 into an IKs stable cell line, KCNE5 was found to suppress the IKs current (Ravn et al., 2008, Lundquist et al., 2005). Since these subunits are found within the same tissues, this suggests that KCNE5 plays an important role in regulating the cardiac action potential, perhaps by competing with KCNE1 (Lundquist et al., 2005, Ravn et al., 2008, Abbott, 2016). An atrial fibrillation mutation found in KCNE5, L65F, reduces the regulatory effect KCNE5 has on IKs, further supporting the importance of the interplay between b-subunits to maintain normal cardiac activity (Ravn et al., 2008).  13 1.2.6 Structural differences between b-subunits Although all five KCNE b-subunits have a similar single transmembrane domain structure, they all have unique effects on KCNQ1 (See 1.2.1-1.2.5). This is likely due to the differences in their amino acid (aa) sequences (ranging in size from 103 to 221 aa), as shown in Figure 1.3. Interestingly, regions of higher conservation between b-subunits are localized within the transmembrane domain regions (Fig. 1.3).    Figure 1.3. Sequence alignment of human KCNE1, KCNE2, KCNE3, KCNE4 and KCNE5. Alignments were performed using www.uniprot.org/align. The transmembrane domains of KCNEs are highlighted in purple. Positions with a fully conserved residue are denoted by an asterisk (*). Positions with conservation of residues with strongly similar properties are denoted by two dots (:). Positions with conservation of residues with weakly similar properties are denoted by a single dot (.).  The activation triplet, located within the transmembrane domain (residues 57-59), is responsible for the slowing effect that KCNE1 has on the activation kinetics of the IKs channel (Melman et al., 2001). When these three residues were replaced by their KCNE3  14 counterparts (residues 71-73) in a KCNE1/KCNE3 chimera, the behavior of the currents produced resembled that of KCNE3 + KCNQ1 much more closely, and vice versa.  A similar chimera approach was used to examine the differences between KCNE1 and KCNE2 (Li et al., 2014). When the KCNE1 N-terminus was replaced with that of KCNE2, co-expression with KCNQ1 resulted in a dramatic decrease in current density, indicating that the N-terminus of KCNE2 plays an important role in suppressing KCNQ1 current (Li et al., 2014).   1.3 Stoichiometry of IKs 1.3.1 Fixed Stoichiometry (2 KCNE1:4 KCNQ1) The ratio of KCNE1 to KCNQ1 subunits has been debated for many years. In 1995, Wang and Goldstein were the first to investigate the stoichiometry of the channel complex. Through co-expression with different ratios of wild type KCNE1 and D77N KCNE1, a mutation which inhibits IKs current, it was proposed that only two KCNE1 subunits were required for normal channel activity (Wang and Goldstein, 1995).  Chen et al. approached the question by incorporating mutations into KCNQ1 to increase affinity to charybdotoxin (CTX) binding, resulting in reduced current density (Chen et al., 2003a). The inhibition seen in separately expressed KCNE1 and mutated KCNQ1 subunits was very similar to a construct where one KCNE1 subunit was tethered to two mutated KCNQ1 subunits (2:4 stoichiometry)(Chen et al., 2003a). Additionally, an antibody was used to count the number of KCNE1 subunits, suggesting that, when expressed separately, only 2 KCNE1s associate with the channel complex (Chen et al.,  15 2003a). This did not change when more or less KCNE1 was expressed (Chen et al., 2003a). Morin and Kobertz (2008), synthesized a specific cleavable CTX which modifies cysteines to count the number of KCNE1 subunits in the channel complex. A cysteine was incorporated into the N-terminus of KCNE1 (T14C). Addition of the cleavable CTX to the bath blocked the pore of KCNQ1, causing an inhibition of current. A disulphide bond formed between the CTX and T14C on one of the KCNE1 subunits, and washing off the solution did not reverse channel inhibition. This disulphide bond was then cleaved with Tris-(2-carboxyethyl)phosphine (TCEP) and the current returned to normal. The same phenomenon occurred in the second stage of CTX addition. In the third stage, the CTX was removed upon wash-off and there was no need for addition of TCEP to cleave CTX from a third KCNE1 subunit. Since only two exposure/wash-off cycles were required to induce a reversible CTX effect, the results support a fixed 2 KCNE1:4 KCNQ1 stoichiometry (Morin and Kobertz, 2008).   In 2014, Plant et al. used total internal reflection fluorescence (TIRF) microscopy to quantify the subunit stoichiometry. KCNE1 and KCNQ1 were tethered to fluorescent proteins (TFP and RFP respectively)(Plant et al., 2014). Through TIRF microscopy, which shows channel complexes at or near the surface of the cell, the number of photo-bleaching steps for each subunit was counted (Plant et al., 2014). Four photo-bleaching steps were observed for KCNQ1, and only two for KCNE1 (Plant et al., 2014).  Additionally, in support of a 2:4 stoichiometry, a structural model of IKs proposed that when two KCNE1 subunits bind, the helices of the b-subunits may be positioned so that they  16 prohibit binding of more KCNE1 subunits into the unoccupied clefts of KCNQ1 (Kang et al., 2008).  1.3.2 Variable Stoichiometry Several groups have provided data in favor of a variable KCNE1:KCNQ1 stoichiometry, where between one and four KCNE1s can bind to KCNQ1. In 1994, Cui et al. showed that the kinetics and conductance of IKs changes when different amounts of KCNE1 mRNA were injected into Xenopus oocytes (Cui et al., 1994). This finding was supported by Morokuma et al. (2008), who also showed that with more KCNE1, there was an increasing hyperpolarizing shift in the V1/2 of activation, suggesting that additional KCNE1 subunits can modify channel characteristics. In addition, tandem constructs where one KCNE1 is tethered to one KCNQ1 (4:4 stoichiometry, E1-Q1) have similar activation kinetics to separately expressed KCNE1 and KCNQ1 (Wang et al., 1998). Furthermore, in a construct where one KCNE1 is tethered to two KCNQ1s (2:4 stoichiometry, E1-Q1-Q1), channels activate at more hyperpolarized potentials (Nakajo et al., 2010). By co-expressing KCNE1 along with the E1-Q1-Q1 tandem, activation kinetics were delayed, showing that KCNE1 can bind to the unoccupied clefts (Nakajo et al., 2010). Using the same TIRF microscopy method, Nakajo et al. (2010) showed the opposite phenomenon to that presented in Plant et al. (2014). In their experiments, depending on the concentration of KCNE1 available, one, two, three and four photo-bleaching steps were observed for KCNE1 (Nakajo et al., 2010). This suggests that, when more KCNE1 is available, channels tend to become more saturated (3:4 or 4:4 stoichiometries).  17 Cardiomyocytes derived from human embryonic stem cells (hESC-CMs) were used to investigate IKs channel properties (Wang et al., 2011a). It was found that when the hESC-CMs are transfected with KCNE1, the endogenous IKs activation kinetics were shifted towards more depolarized potentials (Wang et al., 2011a). This suggests that exogenous KCNE1 may occupy spare clefts in order to fully saturate the complex. Finally, structural modeling by Strutz-Seebohm et al. (2011) was based on a 2:4 stoichiometry; however, they stated that there was still an ability for additional KCNE1 subunits to bind to the channel’s two unoccupied clefts.  1.4 Localizing interactions between KCNE1 and KCNQ1 1.4.1 Pore Domain In 2004, Melman et al. used chimeras of KCNQ1 and KCNQ4 to show that the slowing effect that KCNE1 has on the activation kinetics of the channel is reliant on the pore of KCNQ1, particularly residues S338, F339 and F340 in the S6 domain (Melman et al., 2004). Cysteine crosslinking experiments showed that two members of an activation triplet (residues 57-59) of KCNE1, T58 and L59 in KCNE1, interact with F340 and F339 in KCNQ1 respectively (Panaghie et al., 2006). Furthermore, F57 KCNE1 was thought to be in close proximity to F270, a residue in the S5 domain of KCNQ1 (Strutz-Seebohm et al., 2011).  Several other cysteine crosslinking experiments have shown links between KCNE1 and the pore domain of KCNQ1. In 2012, Wang et al. showed that C331 in the S6 domain of KCNQ1 forms a disulphide bond with F54C in KCNE1 (Wang et al., 2012). Additionally,  18 in 2010, Lvov et al. showed the ability of bonds to form between H363C in the S6 domain of KCNQ1 and three KCNE1 residues (H73C, S74C and D76C) (Lvov et al., 2010). Furthermore, similar disulphide links were visualized when cysteines were incorporated into residues W323 and V324 in the S6 domain, and L42 and E43 in KCNE1 (Chung et al., 2009).  1.4.2 Voltage Sensor Domain Using the same technique as above, cysteine crosslinking experiments showed that disulphide bonds can also take place between residues in KCNE1 and those in the voltage sensor domain (VS) of KCNQ1. Cysteines introduced into the extracellular end of KCNE1 (L40 and L41) interact and form crosslinks with I145C at the extracellular end of the S1 of KCNQ1 (Xu et al., 2008, Chung et al., 2009).  Via a mutagenesis scan, Shamgar et al. (2008) showed that by introducing a tryptophan into different locations in the S4 of KCNQ1, they were able to mimic the gating effect seen in the presence of KCNE1 as well as other KCNE subunits. Some of these KCNQ1 S4 mutations, for example G229W at the N-terminus of the S4, prevented KCNE1 from slowing the rate of activation (Shamgar et al., 2008).  Furthermore, there are a number of mutations in the VS of KCNQ1 which are associated with a variety of diseases, and only display a mutated phenotype in the presence of KCNE1 (Chouabe et al., 2000, Franqueza et al., 1999, Chan et al., 2012). For example, R243C KCNQ1 produces currents that are very similar to wt KCNQ1, however in the presence of KCNE1, conductance is greatly reduced (Franqueza et al., 1999).    19 1.4.3 State-dependent interactions between KCNE1 and KCNQ1 Several groups have shown that there are changing interactions between residues in KCNE1 and KCNQ1, depending on the gating state of the channel. For example, in experiments that investigated disulfide bond formation between cysteines, F54C KCNE1 was found to be in close proximity to C331 KCNQ1 when channels occupied an open state, while L59C KCNE1 was found to be close to C331 KCNQ1 when channels were in a closed state (Wang et al., 2012). Additionally, KCNE1 residues 36-43 were able to interact with several of KCNQ1’s S1 residues (T144, I145 and Q147) at varying times during the activation process (Wang et al., 2011b, Xu et al., 2008). Crosslinking pairs proposed to interact in two opposite KCNE1 orientations, depending on the state of the channel, were used to examine the state-dependent interactions between KCNE1 and KCNQ1 (Chung et al., 2009). Disulphide bonds between K41C KCNE1 and I145C KCNQ1 (K41C-I145C) as well as L42C KCNE1 and V324C KCNQ1 (L42C-V324C) were proposed to occur in one KCNE1 orientation. While K41C KCNE1 and V324C KCNQ1 (K41C-V324C) as well as L42C KCNE1 and I145C KCNQ1 (L42C-I145C) were proposed to interact in an opposite KCNE1 orientation. K41C-I145C slowed deactivation and L42C-V324C eliminated deactivation, resulting in a constitutively active current, indicating a preference for the open state in this orientation. On the other hand, K41C-V324C slowed activation, while L42C-I145C increased the rate of deactivation, indicating a preference for the closed state. These state-dependent interactions that occur between KCNE1 and KCNQ1 led these groups to believe that KCNE1 is not a static player in the IKs channel complex (Wang et al., 2011b, Wang et al., 2012, Xu et al., 2008, Chung et al., 2009). Instead, it was  20 proposed that KCNE1 subunits are able undergo an approximately 120° counterclockwise rotation in the cleft of KCNQ1 when the channel transitions from a resting to an activated state, as well as a reverse clockwise rotation when channels return to a closed conformation (Wang et al., 2012).  1.4.4 Proposed location of KCNE1 in IKs channel complex Although a cryo-EM structure (Sun and MacKinnon, 2017) has recently been published for KCNQ1 with calmodulin (CaM), structural characterization of the IKs channel complex remains extremely difficult, and has yet to be performed. However, experimental investigation of direct interactions between KCNE1 and KCNQ1 (see 1.4.1-1.4.3), led several groups to hypothesize the location of KCNE1 within the IKs channel complex. For example, Xu et al. (2008) suggested that KCNE1 interacts with the S1, S4 and S6 of three different KCNQ1 subunits. Similarly, Shamgar et al. (2008) proposed that KCNE1 may be located between the S1 and S4 of two different KCNQ1 subunits, and Chung et al. (2009) also suggested the possibility of multiple KCNQ1 subunits interacting with one KCNE1 subunit.   Additionally, through model docking studies, the location of KCNE1 within the cleft of KCNQ1 can be studied. In 2009, Kang et al. used nuclear magnetic resonance (NMR) spectroscopy to produce a structure of KCNE1. Using this structure, they docked the transmembrane domain of KCNE1 to a KCNQ1 model based on crystal structures of other Kv channels (Kang et al., 2008). Here, they found that the intracellular end of the KCNE1 TMD is localized to the S4-S5 linker in a closed state model, proposing that this allows  21 KCNE1 to restrain VS movement, slowing activation of the channel (Kang et al., 2008). Additionally, in an open-state configuration of KCNQ1, KCNE1 rotates and now occupies a “gain-of-function cleft”, proposed to result in increased channel conductance (Kang et al., 2008, Van Horn et al., 2011).   1.5 Gating of KCNQ1 in the presence and absence of KCNE1 1.5.1 Gating of KCNQ1 alone and with KCNE1 As discussed in 1.1.2, KCNQ1 has a classic six-transmembrane domain Kv channel structure. The voltage sensor domain (VS) is made up of S1-S4 and the pore domain (PD) is formed by S5-S6 (Kuang et al., 2015). In order for the channel pore to conduct current, the channel undergoes a gating process whereby the VS moves up into the extracellular space, by a process known as activation. Much like other Kv channels, the S4 of KCNQ1 contains several positively charged arginine residues, with a net total charge of +3. These positive charges interact with the negative charge at the top of the S2 domain, E160, to allow the VSs to translocate to different positions during the activation process (Fig. 1.4A) (Zaydman et al., 2014, Wu et al., 2010). VS activation links to the pore via voltage sensor domain-pore domain (VS-PD) coupling to allow the pore to open or close (Larsson, 2002, Cui, 2016). KCNQ1 channels activate to reach a steady state within a second or two, and activation is overlain by channel inactivation (Fig. 1.1B, upper). Recovery from inactivation is seen as a hook in the tail current (Fig. 1.1B, upper), and this allows the inference that inactivation has actually occurred. In the presence of KCNE1, on the other hand,  22 activation is dramatically delayed and does not saturate to reach a steady state even after several seconds of depolarization (Fig. 1.1B, lower). Therefore, a steady-state value for half-maximal activation (V1/2) cannot readily be obtained. Thus, experiments done on IKs essentially employ isochronal activation protocols whereby the tail current amplitude is measured in order to obtain a conductance-voltage (G-V) relationship from which an accurate V1/2 can be attained (Fig. 1.1C). Due to the lack of hooked tail currents in IKs, it is usually assumed that KCNE1 prevents the inactivation process from occurring within the complex. In 2005, Silva and Rudy developed a Markov model with the aim of describing the gating processes of KCNQ1 and IKs. In this model, all four VSs go through two initial transitions prior to channel opening, leading to a model with 15 closed states. Once all four VSs have reached a fully activated state, the model proposes that a concerted movement is required to lead to an open pore configuration (Silva and Rudy, 2005). When IKs channels are held at more depolarized holding potentials they likely transition through the various pre-open closed states, where several VSs have moved to an activated conformation, but the pore is not yet conducting current (Tzounopoulos et al., 1998, Osteen et al., 2010, Silva and Rudy, 2005). From these pre-open closed states, channel activation is much more rapid once a depolarized activating pulse is applied (Cole and Moore, 1960, Tzounopoulos et al., 1998, Osteen et al., 2010).  The Silva and Rudy Markov model was adapted by Werry et al. (2013) in order to account for the single channel subconductance behaviors seen in IKs (Werry et al., 2013, Yang and Sigworth, 1998). In this model, four VSs undergo a first transition, however, a  23 subconducting level is reached after at least one VS undergoes a second transition to a fully activated conformation (Werry et al., 2013). In order to better understand the relationship between VS displacement and pore opening, methods to follow VS movement were developed. Gating currents obtained using the cut-open voltage clamp technique on Xenopus oocytes allow direct measurement of the translocation of charges, and therefore movement of VSs (Stefani and Bezanilla, 1998). Gating charge movements for KCNQ1 alone are almost thirty-fold slower than Shaker channels, and are difficult to resolve for IKs due to the enhanced delay of channel activation in the presence of KCNE1 (Ruscic et al., 2013).  Voltage clamp fluorometry (VCF) is another method used to track movement of VSs in channels expressed in Xenopus oocytes. Here, a fluorescent probe is incorporated into the VS of the channel and the fluorescence-voltage (F-V) and G-V relationships are measured at the same time. When KCNQ1 is expressed alone, the F-V and G-V curves overlap. This suggests that VS movement and pore opening occur simultaneously (Osteen et al., 2010). Additionally, it is thought that the channel can reach an open conformation regardless of the positions of the voltage sensors, which is supported by small constitutively active currents visualized with KCNQ1 alone (Ma et al., 2011). Furthermore, when a mutation is introduced to enhance the open probability of the KCNQ1 pore, L353K, the VSs are still able to visit a fully resting conformation; indicating that when the pore is open VSs can reach a resting state (Zaydman et al., 2013). Altogether, these results suggest that there is weak coupling between the VS and the PD in KCNQ1 channels (Kasimova et al., 2015, Osteen et al., 2012). Thus, in KCNQ1, the pore can open even when all four VSs are in a resting conformation, but the open  24 probability greatly increases as more VSs activate, which supports an allosteric action of activated VS subunits on pore coupling/opening (Osteen et al., 2012).  For IKs currents, G-V curves are displaced to more positive potentials, whereas the F-V is now divided into two phases (Barro-Soria et al., 2014). The first fluorescence phase occurred prior to pore opening, while the second phase of the F-V overlapped the G-V curve. From this, it was suggested that the VSs are required to undergo two steps before the channel pore conducts. The first fluorescence phase represents individual movement of the VSs to an intermediate level. The second phase represents a concerted movement of the four VSs to a fully activated conformation, after which the pore can conduct current (Barro-Soria et al., 2014), similar to that proposed by Silva and Rudy (2005). A concerted step of VSs required to conduct current has been described before in the Shaker Kv channel (Schoppa and Sigworth, 1998, Zagotta et al., 1994). Alternatively, several groups propose that independent movement, rather than a concerted step, of VSs also results in pore opening. To investigate the implications of VS movement during the two transitions during activation, Zaydman et al. (2014) used charge reversal mutations between the negative charge at the top of the S2 domain, E160, and the positive arginines in the S4 domain to fix VSs in an intermediate (E160R/R231E; E1R/R2E; Fig. 1.4B) or a fully activated (E160R/R237E; E1R/R4E; Fig. 1.4C) conformation (Zaydman et al., 2014). In KCNQ1 alone, opening is potentiated in the intermediate E1R/R2E state, and in the fully active E1R/R4E conformation, there is an attenuation of conductance. When co-expressed with KCNE1, there is no longer any visible current in the intermediate state, suggesting that KCNE1 prevents channel conductance when VSs are in an intermediate state. For VSs in an active conformation,  25 E1R/R4E, IKs channel conductance is enhanced. From these experiments, this group developed a model of IKs activation where one VS must exist in a fully activated conformation for the pore to open, regardless of the position of the other VSs (Zaydman et al., 2014).  Figure 1.4. Charge reversal mutations can lock voltage sensors in intermediate and fully activated states. (A) In wild type KCNQ1, the negative charge, E160, at the top of the S2, interacts with the positive arginines in the S4 to allow translocation of the VS from a resting conformation to a fully active conformation. (B) By reversing the charge at the top of the S2 into a positive charge, E160R, and the second arginine in the S4 into a negative charge, R231E, the voltage sensor is locked in an intermediate conformation (E1R/R2E). (C) Charge reversal at the top of the S2, E160R, and of the arginine at the bottom of the S4, R237E, locks the voltage sensor in a fully active conformation (E1R/R4E). These diagrams have been adapted from Zaydman et al., (2014).  Meisel et al. (2012) used thermodynamic mutant cycle analysis to investigate the gating of IKs. Gain-of-function (R231W) and loss-of-function (R234W) mutations were implemented into one, two, three or four KCNQ1s of constructs where four subunits were tethered together. R231W stabilized the channels in an open configuration, whereas R234W stabilized the closed state of the channel. For both, they found either a linear +-+++ ++-+++-A B CWild Type E1R/R2E E1R/R4ES2 S4 26 hyperpolarizing (R231W) or depolarizing shift (R234W) in the V1/2 of activation with each additional mutation. From these linear relationships, this group proposed that VS movement in the IKs complex can occur independently, and thus suggested that the channel complex does not require a concerted activation step (Meisel et al., 2012). In support of this non-concerted gating model, simulated computer modeling by Ramasubramanian and Rudy showed that by restraining movement of one, two and three VSs, channels were able to conduct current, albeit to lower subconducting levels (Ramasubramanian and Rudy, 2018).   1.5.2 ATP is required for pore opening Adenosine triphosphate (ATP) and phosphatidylinositol 4,5-bisphosphate (PIP2; See 1.5.3) have been implicated as important factors in IKs channel gating. Intracellular ATP acts to regulate ion channels and transporters via several mechanisms including membrane trafficking, protein kinases and direct modulation of channel/transporter function (Hilgemann, 1997). In 1986, Bezanilla et al. showed that the conductance of the delayed rectifier potassium current in perfused squid giant axons decreased when ATP was dialyzed from the cell, and increased upon reintroduction of 2 mM ATP (Bezanilla et al., 1986). Additionally, in inside-out patch clamp recordings, when the intracellular components of the cell were dialyzed out, the rundown of IKs current could be prevented by maintaining the ATP levels in the solution (Loussouarn et al., 2003, Li et al., 2013). Via photo-crosslinking experiments, an ATP analog was found to crosslink directly to KCNQ1 channels, particularly to three residues in the C-terminus, suggesting that KCNQ1 channels are directly modified by ATP (Li et al., 2013). This ATP-modulated effect on IKs,  27 is proposed to enhance pore opening, rather than voltage sensor movement or VS-PD coupling (Li et al., 2013). The gain-of-function effect that ATP has on IKs channels is important for maintaining a normal cardiac rhythm. For example, a LQT mutation in KCNQ1, Q357R, acts to reduce the channel’s sensitivity to ATP (Li et al., 2013). Additionally, in failing human hearts, ATP levels are significantly decreased (Beer et al., 2002) as well as cardiac action potentials being prolonged (Beuckelmann et al., 1993), suggesting that there is an important interplay between the two components (Li et al., 2013).   1.5.3 PIP2 is required for VS-PD coupling  Phosphatidylinositol 4,5-bisphosphate (PIP2) is a lipid located in the cytoplasmic part of the plasma membrane, which plays an important role in regulation of a variety of ion channels and transporters (Bian et al., 2001, Suh and Hille, 2008, Kim et al., 2017). PIP2 has an enhancing effect on KCNQ1 alone, and the addition of KCNE1 makes the complex more sensitive to PIP2 (Li et al., 2011, Loussouarn et al., 2003). In 2003, Loussouarn et al. showed that rundown of IKs currents in excised patch experiments could be slowed upon addition of PIP2 to the bath, consistent with a stabilization of the open state (Loussouarn et al., 2003).  Cells already exposed to ATP showed a further increase in current upon addition of PIP2, suggesting that the two substrates work in different ways (Loussouarn et al., 2003). VCF experiments in the presence of PIP2 showed normal VS movement and pore opening for KCNQ1 (Zaydman et al., 2013). On the other hand, addition of a voltage-sensitive lipid  28 phosphatase, CiVSP, to cleave PIP2, maintained normal VS movement, while pore opening was disrupted (Zaydman et al., 2013). This suggests that PIP2 is required for coupling between the VS and the pore (Zaydman et al., 2013). In support of the role of PIP2 in coupling, a recent cryo-EM structure of KCNQ1 channels, collected in the absence of PIP2, showed that the VSs were in a fully activated conformation, whereas the pore was closed (Sun and MacKinnon, 2017).  As expected, several LQT mutations have been found to act in a way whereby the channel’s affinity to PIP2 is impaired (Park et al., 2005, Dvir et al., 2014). This further supports the important role the lipid plays in maintaining normal channel function.   1.5.4 Effect of KCNE1 on VS-PD coupling As discussed in section 1.4, KCNE1 is proposed to interact with several regions of KCNQ1, with sites of interaction found between KCNE1 and both the VS and PD. KCNE1 is thought to interfere with VS movement (Shamgar et al., 2008, Ruscic et al., 2013), pore opening (Barro-Soria et al., 2017) or perhaps both, via altering VS-PD coupling to result in the delayed activation of IKs (Barro-Soria et al., 2017).  In 2017, Barro-Soria et al. used a KCNQ1 mutant, F351A, to separate S4 movement from ionic current conductance when expressed alone. It was found that co-expression of KCNE1 with F351A KCNQ1, still shifted the G-V and the F-V curves, and thus, KCNE1 acts to slow both VS movement and pore opening (Barro-Soria et al., 2017). Furthermore, as discussed above in section 1.5.3, co-expression with KCNE1 makes the channel more sensitive to PIP2, a lipid required for VS-PD coupling (Li et al., 2011,  29 Loussouarn et al., 2003), with the highest affinity proposed to occur when the channel is in the fully activated state (E1R/R4E) (Zaydman et al., 2014). This suggests that KCNE1 intensifies the coupling between the VS and the PD in the fully activated state (Zaydman et al., 2014, Cui, 2016).     30 1.6 Scope of the thesis The research in this thesis provides new insights into the stoichiometry, dynamic interactions and gating of the IKs channel. Gaining a better understanding of these components of channel function, helps shed light on how the channel might act to maintain a normal heart rhythm, and how it can be pharmacologically manipulated.  As discussed in Section 1.3 of the Introduction, the stoichiometry of the IKs channel complex remains controversial. In Chapter 2, the aim of the experiments is to investigate the stoichiometry of the IKs channel via whole cell and single channel patch clamp recordings of fusion constructs with one, two or four KCNE1 subunits assembling with the channel construct. In addition, unnatural amino acid (UAA) photo-crosslinking is used as a technique to investigate crosslinking between the open clefts of the fusion constructs and free KCNE1. The location of interactions between KCNE1 and KCNQ1 remains unclear (see Section 1.4). The aim of the experiments in Chapter 3 is to implement UAA photo-crosslinking to investigate the state-dependent interactions between two adjacent residues in KCNE1 and KCNQ1. In addition, we will investigate the changes in crosslinking during the channel’s activation process.  Finally, it remains unknown if the IKs channel requires a concerted step to opening or whether individual VS movements lead to pore opening (see Section 1.5). Due to this, the aim of the experiments performed in Chapter 4 is to investigate this issue. This will be achieved by using a mutation to lock one, two, three and four VSs in a resting conformation, followed by whole cell and single channel patch clamp recording.    31 Chapter 2: Unnatural amino acid photo-crosslinking of the IKs channel complex demonstrates a KCNE1:KCNQ1 stoichiometry of up to 4:4 2.1  Introduction Kv7.1 (KCNQ1) is a voltage gated potassium channel expressed throughout the body. When paired with the accessory β-subunit, KCNE1, it has unique properties in several tissues, particularly the myocardium and inner ear. In the heart, KCNQ1 and KCNE1 together conduct the slow delayed rectifier current (IKs). This current is primarily responsible for stabilizing repolarization and shortening action potential duration at high heart rates (Marx et al., 2002, Ackerman, 1998). Several inherited mutations in KCNQ1 or KCNE1 result in the life-threating arrhythmogenic syndromes long QT (LQT) types 1 and 5, short QT type 2 or familial atrial fibrillation (Splawski et al., 2000, Bellocq et al., 2004, Chen et al., 2003b). In addition to their role in the heart, KCNQ1 and KCNE1 are also involved in hearing, where mutations have been linked with Jervell and Lange-Nielsen syndrome, an inherited form of LQT syndrome accompanied by deafness (Jervell and Lange-Nielsen, 1957, Neyroud et al., 1997, Schulze-Bahr et al., 1997). KCNQ1 and KCNE1 complexes have also been found in the proximal tubule of the kidney, where they participate in secretory transduction (Vallon et al., 2001). KCNQ1 has a classic six transmembrane domain structure consisting of a voltage sensor domain (VS; S1-S4) and pore domain (S5-S6), which assemble into a tetrameric channel. Unlike other Kv channels, KCNQ1 has flexible gating characteristics and the association with accessory β-subunits can drastically modulate its gating behavior (Eldstrom and Fedida, 2011, Liin et al., 2015). The best characterized of these accessory subunits is KCNE1, a single transmembrane domain protein proposed to reside in the exterior cleft  32 formed between the voltage sensor domains of the α-subunits (Sanguinetti et al., 1996, Barhanin et al., 1996, Osteen et al., 2010, Kang et al., 2008, Van Horn et al., 2011). The association of KCNE1 has been found to increase channel conductance, remove inactivation, and depolarize the voltage dependence of activation. These actions have a profound effect on the rate of channel activation and its response to drugs and hormones (Sanguinetti et al., 1996, Barhanin et al., 1996, Sesti and Goldstein, 1998, Yang and Sigworth, 1998, Pusch, 1998, Tristani-Firouzi and Sanguinetti, 1998, Nerbonne and Kass, 2005, Lerche et al., 2000, Yu et al., 2013). Despite the significant role of this ion channel complex in health and disease, the stoichiometry of the binding between KCNQ1 and KCNE1 remains controversial. Since the identification of the components that comprise this channel almost 20 years ago, several studies have reported a ratio of two β-subunits to every four α-subunits (Chen et al., 2003a, Kang et al., 2008, Wang and Goldstein, 1995, Morin and Kobertz, 2007, Morin and Kobertz, 2008, Plant et al., 2014). However, other experiments have suggested a more flexible stoichiometry with anywhere from one to four β-subunits associating per KCNQ1 tetramer depending on their concentration (Yu et al., 2013, Cui et al., 1994, Wang et al., 1998, Morokuma et al., 2008, Zheng et al., 2010, Wang et al., 2011a, Strutz-Seebohm et al., 2011, Nakajo et al., 2010). Most notably, two recent reports both utilizing total internal reflection fluorescence microscopy to count β-subunits by single molecule photo-bleaching reached opposing conclusions. Working in oocytes, Nakajo and colleagues characterized a flexible stoichiometry up to 4:4 that was dependent on the relative density of KCNE1 in the membrane (Nakajo et al., 2010). Using a similar approach in mammalian cells, Plant et al. (2014) reported that KCNQ1 tetramers were  33 only present on the cell surface in a 2:4 or 0:4 stoichiometry with KCNE1 (Plant et al., 2014). They also found that the 2:4 ratio was not exceeded even in the presence of increasing KCNE1 concentrations. Their observation of a maximum of 2 KCNE1 subunits per tetramer suggests an intrinsic cooperative mechanism preventing the association of a third or fourth β-subunit during channel assembly. Here, we sought to clarify this issue using fusion channels where KCNE1 was linked to one (EQ - 4:4), two (EQQ – 2:4) or four (EQQQQ - 1:4) KCNQ1s. The channel complexes were evaluated in mammalian cells on their own or in combination with additional independent KCNE1 subunits using whole cell patch clamp, single-channel recordings and UV-crosslinking unnatural amino acid approaches. In each instance, we were able to demonstrate that all four of KCNQ1’s exterior clefts are accessible to KCNE1, confirming the variable stoichiometry model for channel complex assembly. In addition, the single channel analysis suggests that a 4:4 stoichiometry of KCNQ1 to KCNE1 maximizes channel conductance.  2.2  Materials and Methods 2.2.1  Reagents Bpa was obtained from Bachem. All other reagents and chemicals were from Sigma-Aldrich (St. Louis, MO, USA). 2.2.2 Molecular Biology A human KCNE1-Flag-GFP pcDNA3 construct was mutated using the QuikChange II Site-Directed Mutagenesis Kit (Agilent Technologies, Santa Clara, CA, USA) using the  34 following primers F57TAG 5'-ggtactgggattcttcggcttctagaccctgggcatc-3', M62W 5’-cttcaccttgggcatctggctgagctac-3’ and Y40TAG-GFP 5’-agggcgatgccacctagggcaagctg-3’ according to the manufacturer’s protocol. The EQ, EQQ and EQQQQ constructs were generated by linking the C-terminus of KCNE1 with the N-terminus of one, two or four human KCNQ1 sequences. The linker between KCNE1 and KCNQ1 consisted of 31 amino acids including a flag and V5 epitopes (UV experiments) or 52 amino acids, which included an additional 21 aa (SRGGSGGSGGSGGSGGSGGRS) inserted after the flag sequence (whole cell and single channel experiments). For EQQQQ and EQQ, the linker between KCNQ1’s was 22 amino acids including a V5 tag. All mutations were confirmed by sequencing. 2.2.3  Cell culture and transfections tsA201 transformed human embryonic kidney (HEK) 293 or ltk- mouse fibroblast (LM) cells were grown in Minimum Eagle Medium (Thermo Fisher Scientific, Waltham, MA, USA) supplemented with 10% fetal bovine serum (Thermo Fisher Scientific), 100 U/ml penicillin and 100 ug/ml streptomycin (Thermo Fisher Scientific). Cells were maintained at 37°C in a humidified atmosphere containing 5% CO2. One day prior to transfection, cells were lifted with 1 min exposure to trypsin/EDTA and re-plated on 25 mm2 glass coverslips. IKs, wild type or Bpa-incorporated, channels were over-expressed by transient transfection using Turbofect (Thermo Fisher Scientific) in tsA201 cells according to the manufacturer’s protocol in the presence of media supplemented with 1 mM Bpa (a membrane permeable photo-crosslinking UAA). Turbofect yielded a higher efficiency transfection than lipofectamine for HEK cells. In some cases, a Y40TAG-GFP construct was used to indicate a successful transfection. The F57TAG KCNE1-Y40TAG GFP,  35 KCNQ1 (or EQQQQ/EQQ/EQ), were co-transfected with a mutated tRNA, and RS pair; (Ye et al., 2008) in a 5:1:1:1 ratio, while wild type KCNE1-GFP and KCNQ1 (or EQQ/EQ) plasmids were transfected in a 3:1 ratio. For transfections lacking Bpa, wild type GFP was used at a ratio of 0.05 to indicate transfected cells. For single channel recordings constructs were transfected in LM cells using Lipofectamine2000 (Thermo Fisher Scientific). All recordings were performed 24–48 hr post-transfection. Adherent cells on coverslips were removed from their media, washed with external solution and transferred to the recording chamber containing external solution. All recordings were performed at room temperature. In preliminary rescue experiments, there were occasional instances of GFP positive cells in the absence of Bpa. To ensure accurate reporting of full length KCNE1 expression, a TAG site was also introduced at position 40 in the GFP sequence. The F57TAG/M62W KCNE1 – Y40TAG GFP construct was used in all experiments (abbreviated to F57Bpa KCNE1) unless otherwise indicated. 2.2.4 Patch-clamp electrophysiology Whole-cell current recordings were obtained from isolated and unconnected GFP-positive cells using an Axopatch 200B amplifier and Digidata 1440A controlled by pClamp10 software (Molecular Devices, Sunnyvale, CA, USA). Patch electrodes with resistance 1–3 MΩ were made from borosilicate glass (World Precision Instruments, Sarasota, FL, USA) using a linear multi-stage puller (Sutter Instruments, Novato, CA, USA) and fire polished before use. Currents were filtered at 2–5 kHz and sampled at 10 kHz. Series resistance compensation of >80% was applied to all recordings. UV-irradiation was performed simultaneously with voltage commands using a TTL controlled UVICO  36 continuous UV light source equipped with an automated shutter (Rapp OptoElectronic, Hamburg, Germany). For single channel recordings, methods of cell-attached recordings are as previously published (Werry et al., 2013, Eldstrom et al., 2015) except that the patch electrodes were fabricated with thick-walled borosilicate glass (Sutter Instruments) and a change was made to the pipette solution as detailed below. In initial experiments, a UV-specific rundown was observed in wild type channels. We hypothesized this was due to the formation of oxidative radicals upon UV irradiation. However, supplementing the internal and external solutions with reactive oxygen species scavengers (DTT, DMSO, ascorbate, GSH) had no effect on the level of rundown (data not shown). 2.2.5 Solutions For whole-cell ionic current recordings the intracellular pipette solution contained the following (mM): 130 KCl, 5 EGTA, 1 MgCl2, 4 Na2-ATP, 0.1 GTP, 10 HEPES, pH adjusted to 7.2 with KOH. The extracellular bath solution contained (mM): 135 NaCl, 5 KCl, 1 MgCl2, 2.8 NaAcetate, 10 HEPES, pH adjusted to 7.4 with NaOH. For single channel recordings the patch pipette solution contained (mM): 6 NaCl, 129 MES sodium salt, 1 MgCl2, 10 HEPES, 5 KCl, 1 CaCl2 and was adjusted to pH 7.4 with NaOH. The bath solution contained (mM): 135 KCl, 1 MgCl2, 1 CaCl2, 10 HEPES and was adjusted to pH 7.4 with KOH. Due to the high amount of chloride channel contamination in our single channel recordings, we adjusted the original solution from 135 mM NaCl to contain 6 mM NaCl and 129 mM MES sodium salt. We found that this combination allowed us to limit chloride  37 currents seen in single channel patches. In our single channel high K+ bath solution, the cell resting potential is assumed to be depolarized to 0 mV. 2.2.6 Data analysis G-V relations were obtained from normalized tail current amplitudes. V1/2 and k-factors were obtained by fitting the data from each cell with a Boltzmann sigmoidal function. The effect of UV-crosslinking was expressed in diary plots normalized to the peak current prior to UV exposure for each cell. To account for the UV-rundown, normalized wild type currents vs. cumulative UV exposure for each cell were fit with a one-phase decay equation to determine a mean rate constant (KRundown (RD): KCNQ1+KCNE1, 0.1864 s-1; EQQQQ+KCNE1, 0.1588 s-1; EQQ+KCNE1, 0.1049 s-1; EQ+KCNE1, 0.1259 s-1). The rate of crosslinking (KXL) was determined by plotting normalized F57Bpa containing currents vs. cumulative UV and fit with a two-phase decay equation where the slow rate constant was set to the KRD obtained for each construct. Recordings where peak currents were greater than 20 nA or smaller than 2 nA or with a ratio of peak current/initial current < 5 prior to UV exposure were excluded from analysis. 2.2.7 Western blot To confirm rescue expression of F57Bpa KCNE1, tsA201 cells were transfected as described above in the presence and absence of 1 mM Bpa. After 48 hr of expression, cells were washed in PBS, lifted and pelleted by centrifugation for 1 min at 1000 x g. Pellets were stored at -80°C until used. Cells were lysed in 50 mM tris-HCl pH 7.4, 300 mM NaCl, 1% (w/v) DDM and protease inhibitors (Roche, Basel, Switzerland) by sonication. Lysates were clarified by centrifugation for 5 min at 2000 x g. Resulting supernatants were quantified and diluted into SDS-PAGE sample buffer. 3 μg of KCNE1- 38 GFP and 50 μg of KCNE1-Y40TAG GFP or F57TAG KCNE1-Y40TAG GFP lysates were separated on a 12% bis-tris gel using MES running buffer. Proteins were then transferred to nitrocellulose for 1 hr at 100 V in NuPage transfer buffer. Blots were blocked in TBS-t containing 5% (w/v) milk powder, probed with an anti-GFP monoclonal (GF28R; Thermo Fisher Scientific) primary antibody and a goat anti-mouse HRP conjugated secondary antibody (Jackson Immunoresearch, West Grove, PA, USA) for 1 hr each at room temp. Exposures were made on film (Eastman Kodak, Rochester, NY, USA) using the Western Lightning Plus-ECL substrate (Perkin Elmer, Waltham, MA, USA). 2.2.8 Co-immunoprecipitation F57Bpa KCNE1-GFP or wild type KCNE1-GFP were co-expressed with KCNQ1 containing an N-terminal T7 affinity epitope in a tsA201 cell line stably expressing the inward rectifier Kir4.1 (Gift from Dr. C. Ahern). Cells were washed and lifted in a low potassium hyperpolarizing solution (-90 mV; as determined by the Nernst equation) (mM): 4 KCl, 140 NMDG, 10 HEPES, 1.5 CaCl2, 1 MgCl2 pH 7.4. Cells were exposed to UV irradiation using a 500 W arc lamp with a Mercury-Xenon bulb (Newport Corporation, Irvine, CA, USA) for 0, 2 or 5 min. Cells were collected by centrifugation and stored at -80°C until used. Cell pellets were lysed as described above. 500 μg of each supernatant was diluted to 0.5% (w/v) DDM and 2 μg of T7-tag monoclonal (69522; Millipore, Etobicoke, ON, Canada) was added. Samples were rotated for 1 hr at room temp. 25 μl of washed, packed protein G-agarose beads (Roche) were added to each sample and rotated for 1 hr at room temp. Beads were washed with PBS containing 0.5% (w/v) DDM and 600 mM NaCl. Protein complexes were eluted with PBS containing 2% (w/v) SDS and 50 mM DTT and separated on a 7% tris-acetate gel. Western blots were performed  39 as described above using the anti-GFP monoclonal or T7-tag monoclonal (69522; Millipore, Etobicoke, ON, Canada) antibodies. 2.2.9  Statistics Results are expressed as mean ± S.E. and represent data from at least 3 independent experiments. Difference of means testing was performed using one-way ANOVA with Tukey’s post hoc test or a two-tailed Student’s t-test. A p-value of less than 0.05 was considered to be statistically significant. Non-linear regressions were performed on G-V relations using the Boltzmann sigmoidal function, and on UV-crosslinking data using the one- or two-phase decay functions in Prism 6 (GraphPad Software, La Jolla, CA, USA).  2.3 Results 2.3.1 Characterization of IKs channel constructs To help differentiate possible channel complex stoichiometries, we created three IKs fusion constructs (Fig. 2.1A). The C-terminus of KCNE1 was connected with the N-terminus of KCNQ1 via a flexible linker (EQ), forcing a 4:4 stoichiometry of β and α subunits. An additional one or three KCNQ1 subunits were linked to the construct to make EQQ and EQQQQ, respectively (Yu et al., 2013, Wang et al., 1998, Nakajo et al., 2010). A priori, EQQ, establishes a 2:4 KCNE1 to KCNQ1 subunit stoichiometry, and leaves two unoccupied clefts on each channel’s exterior, while EQQQQ has a 1:4 ratio with three unoccupied clefts. KCNQ1, EQQQQ, EQQ and EQ were expressed independently in mammalian tsA201 cells and the resulting currents were characterized by whole cell patch clamp (Fig. 2.1B). A 4 s isochronal activation protocol ranging from -80 to +100 mV  40 was used to determine the V1/2 of activation (Fig. 2.1D). As is known, KCNQ1 expressed alone had a V1/2 of -15.5 mV, while the presence of additional tethered β-subunits resulted in a progressive depolarization of the V1/2 of activation to a maximum of ~+30 mV, (Fig. 2.1E and H). This indicates that the linked subunits contribute to channel complex regulation.  Figure 2.1. Additional β-subunits can alter IKs channel complex gating properties. (A) Channel topology diagrams indicate the configuration and proposed stoichiometry of the channels. The KCNE1 sequences are shaded. Representative whole cell patch clamp recordings are shown for KCNQ1 (black circle), EQQQQ (green diamond), EQQ (blue square) and EQ (orange triangle), expressed alone (B; open symbols) or in combination with wild type KCNE1-GFP (C; filled symbols). (D) Currents were elicited using a 4 s isochronal activation protocol. Only odd numbered sweeps are shown for clarity. Tail current G-V plots are shown comparing the response to increasing number of β-subunits either by fusion (E) or by co-expression with KCNE1 (F). (G) G-V plots comparing each channel complex with and without co-expression of KCNE1. (H) Summary of each channel’s V1/2 of activation (n = 3–11; *p<0.05).   41 Next, KCNQ1, EQQQQ, EQQ and EQ were co-expressed with a KCNE1-GFP construct. GFP was included in the construct to verify that KCNE1 was expressed and localized to the plasma membrane in each cell patched. The resulting currents were characterized using the same protocol (Fig. 2.1C). In the presence of excess additional β-subunits, all the α-subunit complex configurations had V1/2’s equivalent to that of EQ (Fig. 2.1F and H). This convergence of the V1/2’s of activation to ~+30 mV, particularly the depolarizing shift of EQQQQ and EQQ, suggests that the independent GFP-tagged β-subunits can participate in channel gating. 2.3.2 Reduced single-channel conductance and latencies of EQQ and EQQQQ IKs channel complexes Current records in Figure 2.2A illustrate selected bursts of channel openings from the EQ, EQQ and EQQQQ (4:4, 2:4 and 1:4) channel complexes during 4 s depolarizations to +60 mV and repolarization to -40 mV to observe tail currents. The EQ 4:4 complexes open after a variable delay period and show fairly consistent bursts of opening that are close to 0.5 pA in amplitude. Similar single channel behavior is seen with KCNQ1+KCNE1, (Werry et al., 2013) as also shown in Figure 2.3A. In comparison, it can be seen that EQQ and EQQQQ channels open to lower levels than the EQ channels. In fact, EQQ channels show extremely long residence times in lower sublevels and rarely, or in the case of EQQQQ never, reach the fully open amplitudes seen in the EQ tracings. These observations are confirmed in the all-events amplitude histograms for the constructs, and the current-voltage relationships (Fig. 2.2B and C). While for EQ it can be seen that the peak of the events amplitude histogram is at ~0.45 pA, it is only 0.18 pA for   42  Figure 2.2. EQ, EQQ and EQQQQ show clear differences in single channel behavior. Membrane patches containing a single IKs channel made up of EQ, EQQ, or EQQQQ were stepped from a holding potential of -60 mV (-80 mV for EQQQQ) to 60 mV for 4 s and then to -40 mV for 0.75 s as indicated in the protocol at top. (A) Representative traces of single channel recordings from cells expressing EQ (top), EQQ (middle) or EQQQQ (bottom). (B) All-points amplitude histograms of only the active single channel sweeps from a file of 100 sweeps, blank sweeps were removed to limit the 0 amplitude peak. All-points amplitude histograms containing all the blank sweeps are presented in Appendix A, Figure 1. For EQQQQ the histogram represents data from 3 cells. Arrows indicate the peak conductance as determined by Gaussian fits using Clampfit. (C) The peak open amplitudes for several voltages were plotted with an extrapolated K+ channel reversal potential to derive a slope conductance for each construct. The slope conductance for IKs made up of unlinked KCNE1 and KCNQ1, as previously published, (Werry et al., 2013) is indicated by a dashed line. Amplitudes for EQQQQ were too small to make analysis of conductance meaningful.  EQQ, and barely separable from the closed events distribution in the case of EQQQQ, despite very large numbers of events. The dotted line in the plotted current-voltage relationships (Fig. 2.2C) refers to a slope conductance of 3.2 pS observed by Werry et al. Amplitude (pA)0.0 0.2 0.4 0.6 0.8 1.0Events020000400006000060 mV -40 mV 4321Time (s) Sw eep:36 Visible:1 of 103IN 0(pA)00.51 3 424321Time (s) Sw eep:34 Visible:1 of 103IN 0(pA)00.51 3 424321Time (s) Sw eep:27 Visible:1 of 103IN 0(pA)00.51 3 42420Time (s) Sw eep:55 Visible:1 of 55Ipatch(pA)00.42 3414321Time (s) Sw eep:54 Visible:1 of 101IN 0(pA)00.511.51 3 4 24321Time (s) Sw eep:87 Visible:1 of 101IN 0(pA)-0.500.511 23 44321Time (s) Sw eep:67 Visible:1 of 101IN 0(pA)-0.500.511 23 44321Time (s) Sw eep:62 Visible:1 of 101IN 0(pA)-0.500.511 23 44321Time (s) Sw eep:60 Visible:1 of 101IN 0(pA)-0.500.511 23 4-60 mV 60 mV -40 mV 0.5 0.5 0.5 0.5 0 0 0 0 EQ 0.5 0.5 0.5 0.5 0 0 0 0 EQQ A B C Amplitude (pA)0.0 0.2 0.4 0.6 0.8 1.0Events0500010000150002000025000-100 -60 -20 20 60 1000.00.10.20.30.40.50.6Voltage (mV)Amplitude (pA)-100 -60 -20 20 60 1000.00.10.20.30.40.50.6Voltage (mV)Amplitude (pA)Amplitude (pA)0.0 0.2 0.4 0.6 0.8 1.0Events02000004000006000008000000.4 0.4 0.4 0.4 0 0 0 0 420Time (s) Sw eep:44 Visible:1 of 55Ipatch(pA)00.42 341420Time (s) Sw eep:46 Visible:1 of 55Ipatch(pA)00.42 341420Time (s) Sw eep:52 Visible:1 of 55Ipatch(pA)00.42 341-80 mV 0.45 pA 0.18 pA EQQQQ  43 (2013) from separately co-expressed KCNQ1 and KCNE1 (Werry et al., 2013). Clearly, the data from the EQ construct overlay the prior published data, and the fit line gives a mean slope conductance of 3.0 pS. However, for EQQ, the mean conductance data gives a value of only 1.3 pS, well below that observed for KCNE1 and KCNQ1 co-expressed separately (Fig. 2.3C) or as EQ together (p<0.01). Due to the small amplitudes and brevity of EQQQQ openings it was not possible to derive a conductance value. In our prior analysis of wild type IKs, we described the presence of multiple sub-opening levels of the channel complex that were sometimes long-lived. Here we have analyzed single channel data from EQ and EQQ to compare their striking subconductance occupancy behavior (Appendix A, Figure 2). The all-points and idealized histograms are shown overlaying each other in Appendix A, Figure 2A, and the tabulated data below. In EQ we observed the same five sublevels described before for E1 and Q1 expressed separately, plus an additional sixth smaller opening level at 0.08 pA that we did not see as clearly before. Visualizing this level was only possible due to improved signal to noise ratio in the present experiments. Again, EQ had its longest open residency in the second highest sublevel at 0.44 pA (36% of the time spent open), and split its remaining open time fairly evenly across the other sublevels. EQQ visited the same subconductance levels, but spent most of its time at the lower levels, only reaching the 0.44 pA level for 0.6% of the time it spent open. The most commonly occupied subconductance state for EQQ was the 0.29 pA sublevel, which it occupied 34% of the total open time. The data suggest that the open pore architecture of the IKs, channel complex is not altered by the presence of only two E1 subunits in EQQ, just that the stability of higher subconductance levels is reduced to such an extent that their occupancy becomes probabilistically  44 unlikely. This is the explanation for the lower observed single channel conductance of EQQ (Fig. 2.2), and likely EQQQQ as well.  Figure 2.3. Co-expression of additional KCNE1 subunits restores wild type IKs single channel behavior to EQQ and EQQQQ. (A) Representative traces of single channel recordings from cells expressing EQQ + KCNE1-GFP (top), EQQQQ + KCNE1-GFP (middle) or KCNQ1 + KCNE1-GFP (bottom). (B) All-points amplitude histograms of only the active single channel sweeps as described in Figure 2.2. Arrows indicate the peak conductance as determined by Gaussian fits using Clampfit. (C) The peak open amplitudes for several voltages were plotted as in Figure 2.2.  We also found changes in the activation kinetics of EQQ and EQQQQ vs. EQ, as shown in Table 2.1. Consistent with more rapid activation of the EQQ and EQQQQ channel complexes, the first latency to opening of single channels was reduced compared with Amplitude (pA)-0.2 0.0 0.2 0.4 0.6 0.8Events0500010000150002000025000Amplitude (pA)-0.2 0.0 0.2 0.4 0.6 0.8Events020000400006000080000-60 mV 60 mV -40 mV 0.5 0.5 0.5 0.5 0 0 0 0 A B C EQQ + E1 4321Time (s) Sw eep:64 Visible:1 of 108IN 0(pA)-0.500.511 23 44321Time (s) Sw eep:62 Visible:1 of 108IN 0(pA)-0.500.511 23 44321Time (s) Sw eep:96 Visible:1 of 108IN 0(pA)-0.500.511 23 44321Time (s) Sw eep:52 Visible:1 of 108IN 0(pA)-0.500.511 23 4Q1 + E1 0.5 0.5 0.5 0.5 0 0 0 0 4321Time (s) Sw eep:61 Visible:1 of 108IN 0(pA)00.512 3414321Time (s) Sw eep:62 Visible:1 of 108IN 0(pA)00.512 3414321Time (s) Sw eep:63 Visible:1 of 108IN 0(pA)00.512 3414321Time (s) Sw eep:69 Visible:1 of 108IN 0(pA)00.512 341Amplitude (pA)-0.2 0.0 0.2 0.4 0.6 0.8Events05000100001500020000250003000035000-100 -60 -20 20 60 1000.00.20.40.6Voltage (mV)Amplitude (pA)EQQQQ + E1 -100 -60 -20 20 60 1000.00.20.40.6Voltage (mV)Amplitude (pA)420Time (s) Sweep:111 Visible:1 of 201IN 0(pA)-1012420Time (s) Sweep:169 Visible:1 of 201IN 0(pA)-1012420Time (s) Sweep:133 Visible:1 of 201IN 0(pA)-1012420Time (s) Sweep:97 Visible:1 of 201IN 0(pA)-10120.5 0.5 0.5 0.5 0 0 0 0 0.45 pA 0.42 pA 0.41 pA -100 -60 -20 20 60 1000.00.20.40.6Voltage (mV)Amplitude (pA) 45 EQ, from 1.48 ± 0.18 s to 0.94 ± 0.07 s (p<0.05), and to 0.81 ± 0.07 s (p<0.01), respectively. These numbers may be compared with independently-expressed KCNE1 and KCNQ1, which shows a mean first latency of 1.50 ± 0.12 s (Werry et al., 2013) (Table 2.1). Interestingly, it was possible to increase the amplitude of EQQ and EQQQQ construct channel openings and delay their first latency to opening by co-expressing them with additional KCNE1 (Fig. 2.3 and Table 2.1). These data are shown in a similar format and now it can be clearly seen that the current openings of the resulting EQQ+KCNE1 and EQQQQ+KCNE1 channel complexes routinely reach 0.4–0.5 pA, also confirmed by the open events histograms (Fig. 2.3B) and the current-voltage relationships (Fig. 2.3C). The slope of the current-voltage relationships for EQQ+KCNE1 and EQQQQ+KCNE1 were 3.0 ± 0.18 pS, and 2.9 ± 0.15 pS, respectively, which were not significantly different from KCNQ1 and KCNE1 expressed separately (2.9 ± 0.12 pS; P>0.05). The mean first latencies to opening of EQQ+KCNE1 and EQQQQ+KCNE1 were now 1.43 ± 0.08 s, and 1.44 ± 0.14 s, respectively, very similar to that reported before for KCNE1+KCNQ1 and measured again here, 1.50 ± 0.12 s.      46 Construct First Latency (s) n (cells) # Active Sweeps Conductance (pS) n (cells) KCNQ1 + KCNE1 1.50 ± 0.12 3 71 2.9 ± 0.12 2 EQQQQ 0.81 ± 0.07 3 124 n.d.  EQQ 0.94 ± 0.07 3 128 1.3 ± 0.01 3-6 EQ 1.48 ± 0.18 3 36 3.0 ± 0.11 2-4 EQQ + KCNE1 1.43 ± 0.08 4 178 3.0 ± 0.18 2-5 EQQQQ + KCNE1 1.44 ± 0.14 3 52 2.9 ± 0.15 1-4  Table 2.1. Summary of single channel parameters. P values for first latency: EQQ vs. E1+Q1, EQ, EQQQQ+E1 and EQQ+E1 p<0.05; EQQQQ vs. E1+Q1, EQ, EQQQQ+E1 and EQQ+E1 p<0.05. P values for conductance: EQQ vs. Q1+E1, EQQQQ+E1, EQ and EQQ+E1 p<0.001  2.3.3 Expression and characterization of Bpa-incorporated IKs To determine if the independently expressed β-subunits are directly interacting with the channel complex we employed a photo-crosslinking unnatural amino acid approach (Farrell et al., 2005, Hino et al., 2005, Klippenstein et al., 2014). p-Benzoylphenylalanine (Bpa) can form a covalent crosslink following excitation with UV light (350–380 nm) (Fig. 2.4A). The UV light excites the side chain’s central ketone, generating a triplet oxygen radical. This short-lived radical can abstract a hydrogen atom from a carbon atom within a 3.1-angstrom radius. The resulting alkyl radical then rapidly reacts with the remaining ketyl radical in Bpa to form a covalent bond (Dormán and Prestwich, 1994). Bpa was incorporated into the KCNE1 sequence using the amber stop codon (TAG) suppression system (Fig. 2.4A) (Ye et al., 2008, Xie and Schultz, 2005, Young and Schultz, 2010, Chin  47 and Schultz, 2002). F57 was selected for mutation as it is located within the KCNE1 transmembrane domain and has been previously suggested to interact with KCNQ1 in the channel’s closed state although no physical association has been established (Melman et al., 2001, Chen and Goldstein, 2007). Methionine 62 was also mutated to a tryptophan as a precaution against an inappropriate start site downstream of the TAG site. The M62W KCNE1 mutation has been previously shown not to impact channel gating, (Chen and Goldstein, 2007) and the same lack of effect was observed here (Appendix A, Figure 3). To evaluate the effect of Bpa incorporation on IKs channel gating, the F57TAG/M62W KCNE1-Y40TAG GFP (F57Bpa KCNE1) construct was co-expressed with KCNQ1 and the tRNA and aminoacyl-tRNA-synthetase (RS) pair in media supplemented with 1 mM Bpa. This resulted in characteristic IKs currents indicating the successful rescue expression of full length KCNE1 (Fig. 2.4B, upper). Plots of tail current G-V relations demonstrated that F57Bpa KCNE1 showed only a 2 mV depolarizing shift in the V1/2 of activation compared to wild type IKs (wild type, 26.1 ± 2.2 mV vs. F57Bpa, 28.2 ± 6.8 mV; n = 3–11) (Fig. 2.4C). The same was found for F57Bpa KCNE1 co-expressed with EQQQQ, EQQ and EQ channels compared to wild type KCNE1 (Appendix A, Figure 4). This demonstrates that incorporation of Bpa at position 57 does not significantly alter expression or channel gating properties.  48  Figure 2.4. Expression and characterization of F57Bpa KCNE1 in IKs. (A) Schematic showing the structure of the UV- activated crosslinking amino acid, Bpa, and the position of F57 within the transmembrane domain of KCNE1. Bpa was also incorporated at position 40 in the GFP sequence. (B) Currents were obtained using the isochronal activation protocol from F57TAG/M62W KCNE1-Y40TAG GFP constructs cultured in the presence (upper) or absence (lower) of 1 mM Bpa. Only odd numbered sweeps are presented for clarity. (C) G-V relations are shown from wild type (black circles) and F57Bpa KCNE1 IKs (red triangles) (n = 3–11). (D) Western blot depicting the expression of wild type KCNE1-GFP, KCNE1-Y40TAG GFP and F57TAG/M62W KCNE1-Y40TAG GFP constructs in the presence and absence of 1 mM Bpa.  In the absence of Bpa, translation is terminated at the TAG codon, resulting in the expression of a truncated 1–56 β-subunit (Fig. 2.4B, lower). Currents in the absence of Bpa were similar to KCNQ1 alone. This indicated that there was good fidelity at the TAG site and that the truncated KCNE1 had no significant impact on the regulation of the channel complex.  49 To further validate Bpa incorporation, Western blot analysis was also performed (Fig. 2.4D). In the absence of Bpa, no full-length KCNE1 was observed (lanes 2 and 4). Full-length expression of KCNE1-GFP was observed in the presence of Bpa for both the F57TAG KCNE1-Y40TAG GFP and wild type KCNE1-Y40TAG GFP constructs indicating good suppression of the amber stop codons by the tRNA and RS. The lower bands present in lanes 2 and 3 represent the truncated expression of KCNE1-Y40TAG GFP. In the absence of Bpa (lane 2), all translation ends at the Y40 stop codon resulting in a 20 kDa product that does not fluoresce. In the presence of Bpa, individual cells transfected with all 3 plasmids (KCNE1, tRNA, RS) have the complement to suppress the amber stop codon and incorporate Bpa, resulting in full-length expression (lane 3, upper band). Some cells will be transfected with the KCNE1-Y40TAG GFP plasmid but fail to receive the tRNA and/or RS. These cells cannot suppress the stop codon and express the truncated form (lane 3, lower band). 2.3.4 UV-dependent channel crosslinking To determine if F57Bpa directly interacts with KCNQ1, a 300 ms flash of UV light (365 nm) was applied while holding at -90 mV followed by a 4 s activation step (+60 mV) and repeated with each successive sweep (Fig. 2.5A). The application of UV light to F57Bpa channels in the closed state resulted in a rapid and complete loss of peak current (Fig. 2.5B, left). A diary plot of the normalized peak currents shows the establishment of a stable baseline (five sweeps) followed by the UV-dependent inhibition of current (Fig. 2.5B, right). Co-expression of KCNQ1 with wild type KCNE1 also exhibited a UV-dependent progressive decline, or rundown, in peak current (Fig. 2.5B, right, open symbols). This relatively slow loss of current was observed equally across all IKs channel  50 configurations (Appendix A, Figure 5); The Bpa-specific UV-crosslinking effect was found to be much faster and more complete than the wild type rundown (Fig. 2.5B). To further confirm covalent crosslinking between F57Bpa KCNE1 and KCNQ1, co-immunoprecipitation was performed (Appendix A, Figure 6). F57Bpa or wild type KCNE1-GFP was co-expressed with KCNQ1 and exposed to UV light. Western blot analysis indicated a clear increase in the co-association of F57Bpa KCNE1-GFP with KCNQ1 in the presence of longer UV irradiation compared to wild type KCNE1-GFP. These findings, in combination with the UV-current recordings, indicate that F57Bpa forms a UV-dependent covalent crosslink with KCNQ1 that traps the channels in the closed state. The EQQ and EQQQQ channel complexes have two and three voltage sensor clefts that are not occupied by β-subunits, respectively. The suggestion that IKs can only adopt a 2:4 stoichiometry implies that access to the unoccupied clefts is restricted, while in a variable stoichiometry model they would be accessible (Plant et al., 2014, Nakajo et al., 2010). Since F57Bpa interacts with KCNQ1 in the channel’s closed state, we used this observation to test whether an independently expressed F57Bpa KCNE1 β-subunit could enter an open cleft in EQQQQ or EQQ and crosslink with the channel. UV irradiation of EQQQQ or EQQ co-expressed with F57Bpa KCNE1 resulted in an accelerated rate of peak current decay compared to expression of these constructs with wild type KCNE1-GFP (Fig. 2.5C and D). The results clearly suggest that F57Bpa can access unoccupied clefts in the EQQQQ and EQ constructs to allow UV-crosslinking to occur.   51  Figure 2.5. The IKs channel complex does not have a restricted β-subunit stoichiometry.  (A) Schematic of the UV-voltage protocol. A flash of UV light (purple line) is applied once per sweep for 300 ms at -90 mV before a 4s activation step to +60 mV. Representative currents are shown for KCNQ1 (B), EQQQQ (C), EQQ (D) and EQ (E) co-expressed with F57Bpa KCNE1 (left). For all recordings, UV was applied at sweep 6 after a stable baseline had been established. Sweeps 1 and 5–25 are presented. Diary plots (right) of the UV treated normalized peak currents for each channel construct co-expressed with F57Bpa KCNE1 (filled symbols) or wild type KCNE1 (open symbols) (n = 3–7; *p<0.01; #p>0.05). (F) Plots of the normalized peak current vs. cumulative UV exposure for KCNQ1 + F57Bpa KCNE1 (red circle), EQQQQ + F57Bpa KCNE1 (green diamond), EQQ + F57Bpa KCNE1 (blue square) and EQ + F57Bpa KCNE1 (purple triangle). *p<0.05 comparing normalized peak currents after 2.1 s of UV exposure (n = 5–7). (G) Summary of the crosslinking rates obtained from each cell (n = 5–7; *p<0.05).  52 We also considered that an independently expressed β-subunit could displace one of the tethered β-subunits in the channel complex and crosslink with KCNQ1. To test this possibility, we co-expressed F57Bpa KCNE1 with the EQ fusion construct. The EQ channel complex is proposed to have all the voltage sensor clefts occupied. UV-treatment did not cause any crosslinking-dependent current decrease compared to EQ co-expressed with wild type KCNE1 (Fig. 2.5E). This finding indicates that there is minimal displacement of the tethered β-subunits in a voltage sensor cleft by an independent one. To evaluate the differences in crosslinking, the mean rate constant (KXL) was determined by double exponential decay fit of the normalized peak current vs. cumulative UV exposure for each of the replicates, where the slow rate constant was set to the value obtained from the wild type rundown (KRD) (Fig. 2.5F and G and Materials and Methods 2.2.6). Analysis of the fast rate constants revealed more rapid F57Bpa KCNE1 crosslinking as the number of potentially available clefts increased (EQQ – 2 clefts, 0.55 ± 0.06 s-1; EQQQQ – 3 clefts, 0.63 ± 0.06 s-1; KCNQ1 – 4 clefts, 1.12 ± 0.09 s-1 (n = 5–7)) (Fig. 2.5G). These findings demonstrate that more F57Bpa KCNE1-GFP subunits are interacting with the KCNQ1 channel complexes, as more clefts are available. EQ, which has no available clefts, had a UV decay rate no different than the rundown observed for all wild type channels, and significantly slower than other channel configurations (KXL: EQ + F57Bpa, 0.09 ± 0.09 s-1: p<0.01 (n = 5), Fig. 2.5G). The intermediate rates of crosslinking for EQQQQ and EQQ between KCNQ1 and EQ indicate that F57Bpa KCNE1 can access the unoccupied clefts in the channel complexes. These findings demonstrate that there is no intrinsic restriction of β-subunit association and confirms a variable stoichiometry model for IKs channel complex composition.  53 2.4 Discussion Our study is the first to apply single-channel analysis and unnatural amino acid substitution to address the question of β-subunit stoichiometry within the IKs channel complex. Using these approaches we were able to determine that up to four KCNE1’s can interact per KCNQ1 tetramer. Whole cell patch clamp studies indicated that EQQQQ and EQQ have a hyperpolarized voltage dependence of activation compared to EQ or wild type IKs, which could be normalized by co-expression of additional KCNE1. We also observed the same phenomenon in single channels, where EQQQQ and EQQ had a shorter first latency to open and reduced conductance that equaled EQ and wild type IKs channels when KCNE1 was co-transfected. Finally, our key experiment using a photo-crosslinking unnatural amino acid, unequivocally demonstrated that KCNE1 can incorporate into the EQQQQ and EQQ channel complexes. Taken together, our findings verify that up to four KCNE1 subunits can regulate KCNQ1. As summarized in the Introduction, several previous studies support these findings, in suggesting that the ratio is variable ranging from 1-4 KCNE1’s depending on their availability. However, this debate has recently coalesced around an opposing idea that IKs has a strict 2:4 ratio of KCNE1’s:KCNQ1’s (Kang et al., 2008, Morin and Kobertz, 2008, Plant et al., 2014). This interaction has been investigated using a variety of approaches, including macroscopic currents, pharmacology and most recently by total internal reflection fluorescence microscopy. The report by Plant et al. (2014) utilizing single molecule photo-bleaching counted a maximum ratio of 2 β-subunits per KCNQ1 tetramer in opposition to a previous study using a similar approach (Nakajo et al., 2010). They suggest a possible explanation for the different conclusions was their use of  54 mammalian cells, representing a better model system than Xenopus oocytes for IKs assembly. Our findings using three different techniques in mammalian cells do not support this difference. We clearly show that an independently expressed KCNE1 can associate with EQQ, so that a stoichiometry greater than 2:4 is possible, and indeed that there is no intrinsic mechanism limiting the association of KCNE1 with the channel complex. In the current study we utilized fusion proteins, different versions of which have been used previously. It is notable that the use of these proteins has not always yielded the same result. For example, in the first report of IKs fusion channels, no significant difference between EQQ and EQ channels was found, although a slowing of activation kinetics was observed when EQ and KCNE1 were co-expressed (Wang et al., 1998). A similar result was also found using EQ and EQQ variants that had 30 mV hyperpolarizing shifts in their V1/2 (Chen et al., 2003b). More recently, two other studies have reported differences in the current-voltage relationship between EQ and EQQ that are consistent with our findings (Yu et al., 2013, Nakajo et al., 2010). Nakajo et al. (2010) also reported a small depolarizing shift in the V1/2 when EQQ was co-expressed with KCNE1 in oocytes. One possible explanation for the differences in behavior could be the composition of the KCNE1-KCNQ1 linker in the fusion proteins. The earlier reports appear to use constructs where the C- and N-termini of KCNE1 and KCNQ1 are directly fused while later studies have inserted flexible linkers. Here, we constructed channels containing 52 and 31 amino acid linkers both of which exhibited a difference in V1/2 between EQQ and EQ, and when EQQ was co-expressed with KCNE1 (Fig. 2.1 and Appendix A, Figure 4). This is consistent with the other longer E1-Q1 linker studies and suggests that restricting the C-  55 (KCNE1) or N-termini (KCNQ1) may have subtle effects on channel gating that were not observed in basic macroscopic characterizations. 2.4.1 KCNQ1 regulation by KCNE1 While determining the subunit stoichiometry was our primary goal, there were some additional observations of interest for KCNE1’s role in modulating channel gating. The V1/2’s of the current-voltage relationships for KCNQ1 alone, EQQQQ, EQQ, and EQ, increased linearly with increasing KCNE1 presence (Fig. 2.6), as can also be observed using the data presented by Yu et al. (2013). Each β-subunit that associates appears to have a similar contribution in regulating the midpoint of the voltage dependence of activation, and the linear change in the V1/2 does not support cooperative gating effects of additional KCNE1 association.   Figure 2.6. Increasing KCNE1 has a linear effect on depolarizing shift in the V1/2 of activation. Plot of the mean tail current G-V relations comparing KCNQ1 alone (0:4), EQQQQ (1:4), EQQ (2:4) and EQ (4:4) for fusion proteins with 31 amino acid and 52 amino acid linker as well as mean data obtained from Yu et al. (2013). R2 values: Yu et al, 0.94; 31aa, 0.97; 52aa, 0.99.   56 The results of our crosslinking study also implicate KCNE1 in regulating the pore domain by showing that residue F57 in KCNE1 interacts with KCNQ1 in the resting state in a conformation that prevents channel opening (Fig. 2.5). F57 has been identified as part of the ‘activation triplet’ consisting of F57, T58 and L59, which has been found to interact with the pore region of the channel (Wang et al., 1996a, Tapper and George, 2001, Melman et al., 2004). A crosslinking event between KCNE1 and the pore region could restrict its movement and prevent channel opening. There are several theories for how KCNE1 delays channel opening, including that KCNE1 inhibits the movement of the S4 domain in the voltage sensor (Nakajo and Kubo, 2007, Ruscic et al., 2013), or that it slows the opening of the pore domain (Rocheleau and Kobertz, 2008, Osteen et al., 2010, Barro-Soria et al., 2014). Our findings suggesting that KCNE1 potentially impacts both are supported by a recent paper by Zaydman et al. (2014). They propose that KCNE1 does not directly affect VS activation or pore opening, and instead suggest that KCNE1 alters the state-dependent interactions linking the VS and pore in KCNQ1 channels (Zaydman et al., 2014) . If KCNE1 affects the interactions between the two domains, it is plausible that KCNE1 not only independently modulates the voltage dependence of its own α-subunit, but also participates in interactions that, when covalently crosslinked, can prevent channel opening. Clearly additional studies are required to better understand the detailed nature of these interactions. 2.4.2 Implications of the single channel studies One consequence of the prolonged debate over IKs stoichiometry has been to suppress investigation of sub-saturating KCNE1 on the regulation of KCNQ1. Since the nature of the channel complex has been in doubt, it has been difficult to consider the effects of a  57 partially chaperoned channel. By confirming that IKs has a variable stoichiometry up to 4:4, our analysis of EQQ provides some insights into the channel’s regulation. In addition to a hyperpolarizing shift in V1/2 (Fig. 2.1), our single channel analysis indicated that EQQ and EQQQQ channels had a reduced first latency to open at +60 mV compared with wild type (Table 2.1). Also, unsaturated channels rarely reach a fully conducting state, and instead they primarily reside in subconducting conformations  (Figs. 2.2 and Appendix A, Figure 2). The modulation of both these parameters is consistent with the kind of regulatory model proposed by Zaydman et al. (2014), in which KCNQ1 has intermediate and activated voltage sensor states, both of which lead to channel opening, but where KCNE1 suppresses opening in the intermediate state. In this scenario, EQQ and EQQQQ would represent hybrid channels; QQ subunits lacking E1 could open from an earlier gating state while those associated with E1 open from a more fully activated state. This would change the interacting forces on the subunits, move channel opening to an earlier conformation in the activation pathway, and likely left-shift the V1/2 of activation and reduce the first latency for opening as we observed.  The same hybrid characteristics can be applied to the behavior of the pore. We have previously observed that wild type channels show a mean conductance of 3.2 pS and also exhibit several sub-conducting levels (Werry et al., 2013). On its own, KCNQ1 is thought to have a reduced channel conductance, although this has not been confirmed by single channel analysis (Yang and Sigworth, 1998, Pusch, 1998). Our findings indicate that in the presence of only two KCNE1 subunits, the pore is still capable of achieving a wild type conducting state, but much less frequently such that a mean conductance of ~1.3 pS was observed. In the presence of only one E1 subunit (EQQQQ) the effect was  58 even greater and points to a destabilized open pore configuration in the absence of KCNE1. Others have reported that KCNE1 can interact and modulate the characteristics of the pore domain but this is the first direct demonstration that sub-saturating levels of KCNE1 lowers mean single channel conductance, and it indicates that 4 KCNE1’s preferentially stabilize the ion throughput rate of the open pore (Wang et al., 1996a, Tapper and George, 2001, Melman et al., 2004). These findings indicate that a full complement of KCNE1 enhances KCNQ1 channel activity. 2.4.3 Stoichiometry of the IKs channel complex in cardiomyocytes While the majority of studies have focused on the composition of IKs in model systems, the stoichiometry on the surface of native cardiomyocytes has not yet been specifically addressed. The implication of variable stoichiometry suggests a distribution of stoichiometries depending on the expression level (Nakajo et al., 2010). A characterization of IKs currents in dog cardiac myocytes determined the V1/2 at 37°C for IKs across different regions of the ventricular wall to be ~25 mV (Liu and Antzelevitch, 1995). Additionally, co-localization studies of KCNE1 and KCNQ1 in rat cardiomyocytes indicated that KCNE1 is present in excess (Wang et al., 2013). Our results and these previous findings suggest that under these conditions, it is likely that many of the IKs channel complexes have a β-subunits configuration exceeding 2:4. A variable stoichiometry model for IKs has some interesting implications. A fully saturated channel has a depolarizing shift in the current-voltage relationship, extremely slowed activation and increased conductance (Figs. 2.1 and 2.2). However, during a normal cardiac action potential this would result in limited activation. If the stoichiometry can be regulated based on availability (Nakajo et al., 2010), then altering the expression could  59 change the distribution of channel composition, shifting the overall activation kinetics of the current. Additionally, there is evidence for the association of other β-subunits. All five of the KCNE genes (KCNE1-5) can be expressed in cardiomyocytes (Lundquist et al., 2006, Bendahhou et al., 2005, Radicke et al., 2006). Each of these accessory proteins has been found to alter KCNQ1 channel gating in distinct ways (Eldstrom and Fedida, 2011, Liin et al., 2015). There is also potential for different KCNE’s associating with the same channel to modulate its function (Manderfield and George, 2008, Jiang et al., 2009). Flexibility in KCNQ1’s β-subunits association, both in number and in type, allows for a powerful mechanism to modulate repolarization in the heart through changes in expression. 2.4.4 Conclusions The number of β-subunits present in the IKs channel complex is critical for its regulation. Here, we have shown that up to four KCNE1 subunits can associate with the KCNQ1 tetramer. Variable association allows for greater flexibility in the modulation of IKs current but also potentially increases the heterogeneity of channel species in the membrane. The development of models of cardiac function, IKs channel structure and activity, pharmaceutical screens and transgenic animals should include consideration of this potential diversity in IKs channel configuration to best reflect the underlying physiology.    60 Chapter 3: Photo-crosslinking of IKs demonstrates state-dependent interactions between KCNE1 and KCNQ1 3.1 Introduction The slow delayed rectifier channel (IKs) generates one of the key repolarizing potassium currents in the cardiac action potential. It is composed of four KCNQ1 α-subunits, each consisting of a classic six-transmembrane domain (TMD) structure with a voltage sensor domain (VS; S1–S4) and a pore domain (S5–S6), which associate to form a tetrameric voltage-gated potassium channel. A single transmembrane domain spanning accessory β-subunit, KCNE1, is proposed to reside and interact within exterior clefts created by coassembly of KCNQ1 subunits (Barhanin et al., 1996, Kang et al., 2008, Osteen et al., 2010, Sanguinetti et al., 1996). Heteromultimerization with KCNE1 results in an increase in channel conductance, a slowing of activation, depolarizing shift in the voltage dependence of activation, and removal of channel inactivation (Barhanin et al., 1996, Sanguinetti et al., 1996, Yang and Sigworth, 1998, Pusch, 1998, Tristani-Firouzi and Sanguinetti, 1998). Mutations in either subunit have been shown to result in life-threatening cardiac rhythm disorders such as long QT syndrome types 1 and 5, short QT syndrome type 2 and familial atrial fibrillation (Bellocq et al., 2004, Chen et al., 2003b, Splawski et al., 2000). The channel complex is not restricted to activity in the heart; it has also been shown to be prevalent in the kidney, where the complex is involved in secretory transduction (Vallon et al., 2001), and the ear, where mutations have been linked to deafness (Jervell and Lange-Nielsen, 1957, Neyroud et al., 1997). Although IKs plays an important role in health and disease, the interactions between KCNQ1 and KCNE1 are not yet fully understood.  61 Various studies have suggested regulatory interactions occur between KCNE1 and the KCNQ1 tetramer at both the VS and the pore domain. Using a chimeric approach and scanning mutagenesis, interactions have been identified between KCNE1 and the pore domain of KCNQ1 (Melman et al., 2004, Panaghie et al., 2006, Wang et al., 2012). Whereas the extracellular end of the β-subunit transmembrane domain was shown to interact with the extracellular end of the S1 domain of KCNQ1 (Xu et al., 2008), cysteine scanning mutagenesis experiments and computational models have shown that KCNE1 comes into close contact with the voltage sensor, S4, in KCNQ1 (Nakajo and Kubo, 2007, Shamgar et al., 2008, Strutz-Seebohm et al., 2011). Additionally, there are several disease-causing KCNQ1 mutations in the VS that only become evident when KCNE1 is present (Chan et al., 2012, Chouabe et al., 2000, Franqueza et al., 1999). Finally, a recent study by Zaydman et al. (2014) has shown that KCNE1 is involved in coupling the VS to the pore (Zaydman et al., 2014). Taken together, these reports indicate that KCNE1 sits in a region where it is able to interact with various elements of KCNQ1 channel complex. In addition to the regions of interaction between the subunits, the dynamics of KCNE1 within the channel’s cleft have been key to understanding its regulation of KCNQ1. It has been proposed that the transmembrane helix of KCNE1 rotates during channel gating from a closed to an open state, and that the midsection of the helix leans toward KCNQ1 in a state-dependent manner (Wang et al., 2012). Supporting a rotational movement of KCNE1 within the cleft of KCNQ1, Xu et al. (2008) showed by investigating disulfide bond that the interactions between S1 of KCNQ1 and KCNE1 varied depending on the state that the channel was in (Xu et al., 2008).  62 It has been clearly demonstrated that mutations within the transmembrane domain of KCNE1 have a significant effect on KCNQ1, particularly at residues F57, T58, and L59. These three residues, known as the activation triplet, are proposed to be responsible for the characteristic delaying effect KCNE1 has on IKs channel kinetics (Melman et al., 2001). The F57 amino acid residue was also identified by Chen and Goldstein (2007) as a likely site of strong interaction with KCNQ1 using a serial tryptophan screen of the transmembrane domain (Chen and Goldstein, 2007). Furthermore, molecular dynamics simulations suggested that F57, and perhaps F56, are possible sites of interaction with KCNQ1 (Xu et al., 2008). Recently, our lab showed that when a photo-crosslinking unnatural amino acid was incorporated at position F57 in KCNE1, it was possible to use UV light to crosslink channels held in a resting state conformation (Murray et al., 2016)(See Chapter 2). Here, we employ the same unnatural amino acid crosslinking technique to understand more about the structure-function relationship of KCNQ1 and KCNE1 at sites F57 and the lesser-studied F56. This technique allowed us to investigate the interactions between KCNE1 and KCNQ1, and the state-dependent movement of KCNE1 across the cleft of the channel. The results increase our understanding of the dynamic interactions that occur between KCNE1 and KCNQ1 during channel gating.  3.2 Materials and Methods 3.2.1 Reagents All reagents, except Bpa (Bachem, Bubendorf, Switzerland), were obtained from Sigma-Aldrich (St. Louis, MO).  63 3.2.2 Molecular Biology To choose sites for unnatural amino acid incorporation, we targeted residues with a similar bulky, hydrophobic side chain as Bpa (i.e., phenylalanine, F), to ensure the most conserved mutation. As well as F56 and F57, we investigated several other sites around the activation triplet of KCNE1 to incorporate our unnatural amino acid (F54, T58, and L59). Unfortunately, at these locations, Bpa was not well tolerated and did not produce functional IKs currents. The KCNE1-Flag-GFP pcDNA3 construct was mutated with the same primers for F57TAG, M62W, and Y40TAG-GFP as published previously (Murray et al., 2016). KCNE1-Flag pcDNA3 was mutated with a primer for F56TAG 5′-CTGGGATTCTTCGGCTAGTTCACCCTGGGCATCA-3′. E160R/R237E (E1R/R4E) KCNQ1 was a gift from Dr. Jianmin Cui. All mutations were confirmed by sequencing. 3.2.3 Cell culture and transfections tsA201 transformed human embryonic kidney 293 or ltk-mouse fibroblast cells were cultured at 37°C in an air/5% CO2 incubator. Cells were grown in Minimum Eagle Medium (Thermo Fisher Scientific, Waltham, MA) containing 10% fetal bovine serum (Thermo Fisher Scientific), 100 UI/mL penicillin and 100 μg/mL streptomycin (Thermo Fisher Scientific). Twenty-four hours ahead of transfection, cells were lifted using trypsin/EDTA (Thermo Fisher Scientific) and replated on glass coverslips in 35-mm dishes. For whole cell experiments, channel constructs were transiently transfected in tsA201 cells using Turbofect (Thermo Fisher Scientific) according to the manufacturer’s protocol. For unnatural amino acid rescue experiments, the media was supplemented with 1 mM Bpa. The F57TAG/M62W KCNE1-Y40TAG GFP or wild-type (wt) KCNE1-Y40TAG GFP constructs were transfected along with wt KCNQ1 or E160R/R237E KCNQ1 and a  64 coevolved orthogonal tRNA and amino-acyl-tRNA synthetase pair (Ye et al., 2008) in a 5:1:1:1 ratio. M62W was incorporated into E1-GFP to prevent an inappropriate start site downstream of the TAG site in KCNE1. We have previously shown that this does not alter channel kinetics (Murray et al., 2016). F56TAG KCNE1-Flag and wt KCNE1-Flag were transfected with wt KCNQ1 or E160R/R237E KCNQ1, tRNA, amino-acyl-tRNA-synthetase, and Y40TAG-GFP in a 4:1:1:1:1 ratio. A Y40TAG site was incorporated into GFP to indicate a successful transfection in the presence of Bpa. To indicate transfected cells in the absence of Bpa, wt GFP was used. For single channel experiments, wt KCNE1 and E160R/R237E KCNQ1 were transiently transfected in ltk-mouse fibroblast cells at a 3:1 ratio using Lipofectamine 2000 (Thermo Fisher Scientific). All experiments were performed at room temperature, 24–48 h post-transfection. 3.2.4 Patch-clamp electrophysiology Whole-cell current recordings were made using an Axopatch 200B amplifier and acquired using pClamp 10 software with a Digidata 1440A (Molecular Devices, Sunnyvale, CA). Pipettes were pulled from borosilicate glass (World Precision Instruments, Sarasota, FL) using a linear multistage electrode puller (Sutter Instrument, Novato, CA), to a resistance of 1–3 MΩ, and fire-polished before use. Currents were sampled at 10 kHz, filtered at 2–5 kHz, and a series resistance compensation of >80% was applied. Using a UVICO continuous UV light source with an automated shutter (Rapp OptoElectronic, Hamburg, Germany), application of UV-irradiation accompanied voltage commands. Single channel recordings were completed as previously published (Murray et al., 2016, Werry et al., 2013, Eldstrom et al., 2015). 3.2.5 Solutions  65 For whole-cell current recordings, the intracellular pipette solution contained 130 mM KCl, 5 mM EGTA, 1 mM MgCl2, 4 mM Na2-ATP, 0.1 mM GTP, and 10 mM HEPES, with pH adjusted to 7.2 with KOH. The extracellular bath solution contained 135 mM NaCl, 5 mM KCl, 1 mM MgCl2, 2.8 mM NaAcetate, and 10 mM HEPES, with pH adjusted to 7.4 with NaOH. For single channel current recordings, the pipette solution contained 6 mM NaCl, 129 mM MES, 1 mM MgCl2, 10 mM HEPES, 5 mM KCl, and 1 mM CaCl2, with pH adjusted to 7.4 with NaOH. The bath solution contained 135 mM KCl, 1 mM MgCl2, 1 mM CaCl2, and 10 mM HEPES, with pH adjusted to 7.4 with KOH. 3.2.6 Data analysis Tail current amplitudes obtained from current-voltage protocols were normalized to produce G-V relations. Data was fit with a Boltzmann sigmoidal function (Prism 6; GraphPad Software, La Jolla, CA), from which the V1/2 could be extracted. Diary plots normalized to peak current were used to measure the effect of UV-crosslinking in each cell. One-phase decay functions were used to fit the resulting plots (Prism 6; GraphPad Software). In these experiments, exposure to UV caused all IKs currents to run down over time (Fig. 3.1A, lower). To correct for rundown in short pulse F57Bpa data, we fitted normalized non-Bpa currents versus UV exposure with a single exponential decay equation (Prism 6; GraphPad Software): 𝑖 = 𝐼$𝑒&'	(*+,) . The rates of rundown (KRD) for the constructs were found to be: KCNQ1 + E1 (−90 mV), 0.1851 s−1; KCNQ1 + E1 (+60 mV), 0.1898 s−1; and E1R/R4E KCNQ1 + E1 (+60 mV), 0.0461 s−1. Crosslinking rates were subsequently calculated using the KRD for each construct as detailed below.  66 F57Bpa currents were normalized, plotted versus UV exposure time, and fit with a single exponential decay equation: 	𝑖 = 𝐼$𝑒&'	(*./012) . Where KTotal = KRD + KXL. KTotal was extracted from each single exponential decay and KXL was calculated by KXL = KTotal − KRD. For long pulse F57Bpa recordings, the portion of the sweep where UV was applied was extracted and normalized to the peak current. The curves were fit with a single exponential equation, from which a rate of crosslinking (KXL) could be extracted (Prism 6; GraphPad Software). For F56Bpa rundown correction, the wt rundown was fit with a single exponential where each data point was normalized to the initial peak current to produce a rundown correction factor. This correction factor was added to each KCNQ1 + F56Bpa +60 mV data point to produce a single exponential curve. We did not extract a rate of crosslinking from F56Bpa data. 3.2.7 Statistics All results are given as mean ± SE. Statistical analysis was performed using either a two-tailed Student’s t-test or one-way ANOVA followed by Tukey’s post hoc test. P values of <0.05 were considered statistically significant.  3.3 Results 3.3.1 F57Bpa KCNE1 crosslinks to KCNQ1 in the resting state The photo-crosslinking unnatural amino acid, Bpa, was incorporated at position F57 in the transmembrane domain of KCNE1. As described previously (Murray et al., 2016) and in Chapter 2, the incorporation of Bpa at position F57 in KCNE1 is well tolerated  67 (Appendix B, Fig. 1C and D). In the absence of Bpa, the current is similar to KCNQ1 alone, indicating that either the truncated F57TAG E1 does not interact with the channel or cannot modulate KCNQ1. Culture in media supplemented with 1 mM Bpa rescued full-length expression of the β-subunit, resulting in a current that is very similar to wt IKs. G-V relationships show that there is no significant shift in the V1/2 of activation compared to wt IKs (wt IKs, 26.1 ± 2.2 mV; F57Bpa IKs, 28.2 ± 6.8 mV; Appendix B, Fig. 1D). Previously, residue F57 had been shown to interact with KCNQ1 in the resting state of the channel (Melman et al., 2001, Chen and Goldstein, 2007). As KCNE1 is proposed to shift and rotate within the clefts of the KCNQ1 tetramer during activation (Wang et al., 2012, Xu et al., 2008), the unnatural amino acid crosslinking technique was used to investigate the state-dependent interactions that might occur between these subunits. Upon excitation with UV light (350–380 nm), Bpa forms a covalent crosslink with a C-H bond within a 3.1 Å radius (Dormán and Prestwich, 1994). We have previously used this technique to crosslink F57Bpa to KCNQ1 (Murray et al., 2016)(See Chapter 2).Here, we again show that when a 300 ms flash of UV was applied at −90 mV to cells expressing F57TAG/M62W KCNE1-Y40TAG GFP (F57Bpa) and KCNQ1, there is a decline in peak current (Fig. 3.1A and C). This suggests when channels are held in a resting state there is an interaction between F57 and KCNQ1. G-V relationships recorded before and part-way through full UV modification of KCNQ1 + F57Bpa at -90 mV were unchanged, with a mean shift in V1/2 of 1.4 ± 1.9 mV (n = 4). This suggests that crosslinking F57Bpa at −90 mV traps channels in a closed conformation, removing them from the pool of channels available to activate, rather than subtly biasing the rate constants for channel opening to more positive potentials.  68  Figure 3.1. State-dependent crosslinking of F57Bpa IKs shows preferential interaction in the resting state. (A, upper) Shown here is a schematic of the -90 mV UV-voltage protocol. A 300 ms flash of UV light (bold black line) is applied each sweep at -90 mV followed by a 4 s voltage step to +60 mV. The interval between each sweep for all whole cell experiments is 15 s. Representative currents are shown for KCNQ1 + F57Bpa (A, middle) and KCNQ1 + wt E1 (A, lower). (B, upper) Shown here is a schematic of the +60 mV UV-voltage protocol. A 300 ms flash of UV light (bold black line) is applied at each sweep at 3.7 s depolarization. A representative current is shown for KCNQ1 + F57Bpa (B, lower), with the first UV sweep in grey. Note the current decrease when UV is applied. For all recordings, UV was applied at either sweep 6 (-90 mV) or sweep 5 (+60 mV), after a stable baseline had been established. Sweeps 1 and 5-25 are shown. (C) Given here is a diary plot of the UV-treated normalized peak current from KCNQ1 + F57Bpa -90 mV (black circles), +60 mV (grey triangles) and wt E1 (open squares) (n = 6-7). Peak current was either measured at the end of the 4 s pulse (-90 mV) or immediately before UV application (+60 mV). (D) Given here is a plot of normalized peak current versus UV exposure fit with a single exponential (n = 6-7). (E) Shown here is a summary of the crosslinking rate (KXL) at each potential (n = 6-8; * = p <0.05) (see 3.2.6 Materials and Methods for how KXL was obtained).   69  Table 3.1. Crosslinking rate constants for F57Bpa at different membrane potentials. Crosslinking rates (KXL) for KCNQ1 + F57Bpa at -110, -90, -70, -50, and +60 mV; long pulse KXL for KCNQ1 + F57Bpa at 0, +30, and +60 mV; and the KXL for E1R/R4E KCNQ1 + F57Bpa at +60 mV (see 3.2.6 Materials and Methods for how KXL was obtained).  To investigate whether F57Bpa could crosslink with KCNQ1 in the open state, UV flashes were applied toward the end of a 4.3 s activating pulse to +60 mV with the expectation that crosslinks would produce constitutively open channels. In this case, UV irradiation at the end of 4 s depolarizations resulted in an immediate downward deflection in current, followed by a progressive decline in peak current but no increase in the instantaneous current in subsequent pulses to +60 mV (Fig. 3.1B and C), none of which are consistent with an open state crosslink. Calculated as described in 3.2.6 Materials and Methods, the rundown rate (KRD) for wt IKs was shown to be consistent between -90 and +60 mV (Appendix B, Figure 2A-C). For F57Bpa IKs the UV-crosslinking rate (KXL) during depolarization was found to be significantly slower than crosslinking in the resting state (-90 mV, 0.81 ± 0.12 s-1; +60 mV, 0.41 ± 0.05 s-1; p = 0.01; Fig. 3.1D and E; Table 3.1). It is possible that crosslinking open channels renders them nonfunctional, but IKs channels Construct Membrane Potential KXL (s-1) n (cells)Q1 + F57Bpa-110 mV 0.78 ± 0.08 5-90 mV 0.81 ± 0.12 7-70 mV 0.55 ± 0.07 7-50 mV 0.44 ± 0.08 7+60 mV 0.41 ± 0.05 6Long pulse 0 mV 0.33 ± 0.03 4Long pulse +30 mV 0.35 ± 0.06 5Long pulse +60 mV 0.20 ± 0.05 5E1R/R4E Q1 + F57Bpa +60 mV 0.17 ± 0.05 5 70 have been found to have a relatively low open probability (Po < 0.2) after 4 s depolarizations to +60 mV at room temperature (Werry et al., 2013). Thus, ∼80% of the channels remain in a closed state, making it difficult to clearly establish whether F57Bpa interacts with KCNQ1 in the open state. 3.3.2 F57Bpa has a reduced rate of crosslinking in a higher open probability mutant To investigate this further, E160R/R237E (E1R/R4E) KCNQ1 was used—a charge reversal mutant channel known to increase the open probability (Po) of the channel, reportedly by keeping the voltage sensor in an upward position (Zaydman et al., 2014, Wu et al., 2010). Recordings of E1R/R4E KCNQ1 + wt E1 (Appendix B, Fig. 3B) and F57Bpa E1 (Appendix B, Fig. 3D) show constitutively active currents between -80 and +80 mV (Appendix B, Fig. 3E). Single channel recordings of E1R/R4E KCNQ1 + wt E1 (Appendix B, Fig. 3F) have a high Po (∼0.8 after 4 s from the ensemble average) at +60 mV compared with ∼0.2 from the wt IKs data we have previously published (Werry et al., 2013, Eldstrom et al., 2015). A downward deflection and a progressive decline in peak current was still observed when a 300 ms UV flash was applied at the end of a 4.3 s +60 mV pulse to E1R/R4E KCNQ1 + F57Bpa constructs, much like in wt KCNQ1 + F57Bpa at + 60 mV (Fig. 3.2A and B). However, the rate of crosslinking obtained at +60 mV was significantly slower (KCNQ1 + F57Bpa, KXL = 0.41 ± 0.05 s-1; E1R/R4E KCNQ1 + F57Bpa, KXL = 0.17 ± 0.05 s-1; p < 0.01; Fig. 3.2C and D; Table 3.1). This reduced rate of crosslinking in E1R/R4E suggests that F57Bpa predominately interacts with KCNQ1 in the closed state.  71  Figure 3.2. The crosslinking rate of F57Bpa KCNE1 at +60 mV is reduced in a higher open probability mutant. (A) Shown here is a schematic of the +60 mV UV-voltage protocol (upper). A 300 ms flash of UV light (bold black line) is applied each sweep at 3.7 s of the voltage step to +60 mV. A representative set of currents is shown for E1R/R4E KCNQ1 + F57Bpa, with the first UV sweep in grey. Note the current decrease when UV is applied. For all recordings, UV was applied at sweep 5 after a stable baseline had been established. Sweeps 1 and 5-25 are shown. (B) Given here is a diary plot of the UV-treated normalized peak current from E1R/R4E KCNQ1 + F57Bpa (black diamonds) and E1R/R4E KCNQ1 + wt E1 (open diamonds) (n = 4-5). Peak current was measured immediately before UV application. (C) Given here is a plot of normalized peak current versus UV exposure at +60 mV of E1R/R4E KCNQ1 + F57Bpa (black diamonds) and wt KCNQ1 + F57Bpa (grey triangles) fit with a single exponential (n = 5-6). (D) Shown here is a summary of crosslinking rates (KXL) at +60 mV for KCNQ1 + F57Bpa and E1R/R4E KCNQ1 + F57Bpa (n = 5-6, * = p < 0.01) (see 3.2.6 Materials and Methods for how KXL was obtained).  3.3.3 The rate of crosslinking with F57Bpa is reduced after longer depolarizing pulses UV was applied after long depolarizations of 7.5-9 s to 0 mV, +30, and +60 mV (Fig. 3.3) to enhance activation of KCNQ1 + F57Bpa channels. At each potential, a downward deflection was seen in peak current in the first sweep with UV, and a greatly reduced  72 effect in the second sweep. This indicated that most of the crosslinking occurred within the first 10 s of UV exposure. Long pulse recordings of KCNQ1 + wt E1 (Appendix B, Fig. 2D) show that long UV exposure does not result in a deflection in peak current in the first sweep; however, in sweep 2 there is an initial decrease in the size of the current, representing UV-induced wild-type rundown. The UV-induced deflection of the first long pulse sweep of F57Bpa was extracted and the normalized mean peak current versus time of UV exposure plotted (Fig. 3.3D-G) and fitted to extract KXL. The rate of crosslinking with KCNQ1 + F57Bpa at +60 mV was not significantly slower than at 0 and +30 mV (0 mV, 0.33 ± 0.03 s-1; +30 mV, 0.35 ± 0.06 s-1; +60 mV, 0.20 ± 0.05 s-1; Fig. 3.3H; Table 3.1). The KXL after 8 s at +60 mV is half that after 4 s (0.41 from Fig. 3.1;  p = 0.01, Fig. 3.3I). This reduction in KXL further suggests that crosslinking with F57Bpa is most efficient when IKs is in the resting state. 3.3.4 F57 KCNE1 moves across the cleft before channel opening Although the crosslinking rates for F57Bpa at -90 mV and +60 mV are different, a simple two-state model of KCNE1 interaction with KCNQ1 cannot explain them. Crosslinking during depolarizations is not equal to the product of the resting state rate and the channel closed state probability at +60 mV (0.81 s-1∗0.8 ≠ 0.41 s-1). IKs is known to have multiple closed states (Osteen et al., 2010, Tzounopoulos et al., 1998) with KCNQ1/KCNE1 reorientation likely taking place throughout that might lead to changes in crosslinking efficiency. We confirmed the presence of multiple closed states by holding the channels at progressively more depolarized, but non-activating membrane potentials (-110 to -50 mV) for 2 s followed by a 4 s pulse to +60 mV (Fig. 3.4A). Consistent with previous reports (Osteen et al., 2010, Tzounopoulos et al., 1998), a Cole-Moore shift of activation  73   Figure 3.3. The crosslinking rate of F57Bpa is reduced during longer depolarizing pulses. (A, upper) Given here is a UV protocol where 12.5 s of UV is applied after a 7.5 s pulse to 0 mV. (A, lower) A representative set of long pulse currents of KCNQ1 + F57Bpa at 0 mV. (B, upper) Given here is a UV protocol where 10 s of UV is applied after a 9 s pulse to +30 mV. (B, lower) Shown here is a representative set of long pulse currents of UV KCNQ1 + F57Bpa at +30 mV. (C, upper) Shown here is a UV protocol where 10 s of UV is applied after an 8 s pulse to +60 mV. (C, lower) Shown here is a representative set of long pulse currents of KCNQ1 + F57Bpa at +60 mV. Plots of mean normalized peak current versus UV exposure (solid black line) at long pulse 0 mV (D), +30 mV (E), and +60 mV (F) data fit with a single exponential (dashed grey line) (n = 4-5). Mean ± SE is represented by the grey region. (G) Given here is a plot of single exponential fits of normalized peak current versus UV exposure at 0 mV (dashed grey line), +30 mV (solid black line), and +60 mV (solid grey line). (H) Shown here is a summary of the crosslinking rates (KXL) for each long pulse potential (n = 4-5) (see 3.2.6 Materials and Methods for how KXL for long pulse data was obtained). (I) Given here is a graph comparing the crosslinking rates (KXL) of KCNQ1 + F57Bpa when UV is applied after a +60 mV pulse for either 4 or 8 s (n = 5-6; * = p < 0.05).   74 is seen in both wt IKs (Fig. 3.4B) and F57Bpa IKs (Fig. 3.4C) channels. There is a significant decrease of the t1/2 of activation for -70 and -50 mV compared to -110 mV (Fig. 3.4D). No significant difference in the t1/2 between wt and F57Bpa IKs channels was observed for each prepulse potential.    Figure 3.4. Wild-type and F57Bpa IKs channels activate more rapidly when held at depolarized potentials before activation.  (A) Shown here is a voltage-clamp protocol where channels are held at increasingly depolarized holding potentials (−110 to −50 mV) for 2 s before pulsing to +60 mV for 4 s. Representative currents are shown for wt IKs (B, left and inset) and KCNQ1 + F57Bpa (C). (D) Shown here is a change in time to half-maximum current normalized to −110 mV for wt (black circles) and F57Bpa IKs (grey triangles) (n = 7-8; * = p<0.01 compared to -110 mV).     75  Figure 3.5. F57Bpa KCNE1 has reduced interaction with KCNQ1 when channels are held at depolarizing potentials before opening. Shown here is a schematic of the UV-voltage protocol where a 300 ms flash of UV light (bold black line) is applied each sweep during a 2 s hold at -110 mV (A, upper), -70 mV (B, upper), or -50 mV (C, upper), followed by a 4 s step to +60 mV. Representative currents are shown for KCNQ1 + F57Bpa with UV at -110 mV (A, lower), -70 mV (B, lower), and -50 mV (C, lower). For all recordings UV was applied at sweep 6, after a stable baseline had been established. Sweeps 1 and 5-25 are shown. (D) Given here is a diary plot of UV-treated normalized peak current from KCNQ1 + F57Bpa -110 mV (open grey hexagons), -70 mV (grey diamonds), -50 mV (black squares), and wt IKs (open black squares) (n = 5-7). Peak current was measured at the end of the 4 s pulse. (E) Given here is a plot of normalized peak current versus UV exposure fit with a single exponential for KCNQ1 + F57Bpa -110 mV (open grey hexagons), -90 mV (black circles), -70 mV (grey diamonds), -50 mV (black squares), and +60 mV (grey triangles) (n = 5-7). (F) Shown here is a summary of the crosslinking rates (KXL) at each potential (n = 5-7;	* = p < 0.05) (see 3.2.6 Materials and Methods for how KXL was obtained).  To evaluate the extent of F57Bpa interaction as the channels progress through their activation pathway, UV-irradiation was applied in the final 300 ms of the 2 s prepulse at different holding potentials (-110, -70, and -50 mV; Fig. 3.5A-C).  As anticipated, a steady decrease in crosslinking rate was seen as holding potential decreased from -90 to -50 mV and channels populate preopen closed states (-90 mV, 0.81 ± 0.12 s-1; -70 mV, 0.55 ±  76 0.07 s-1; -50 mV, 0.44 ± 0.08 s-1; +60 mV, 0.41 ± 0.05 s-1; Fig. 3.5D-F, Table 3.1). There was no difference in the crosslinking rate between -90 and -110 mV, suggesting that there was little additional reorientation taking place at more hyperpolarized potentials (-110 mV, 0.78 ± 0.08 s-1; Fig. 3.5F; Table 3.1).   3.3.5 F56 interacts with KCNQ1 in the open state Because F57Bpa KCNE1 was found to interact primarily in the resting state of the channel, the equivalent analysis was carried out on the adjacent residue F56 (F56TAG KCNE1-Flag). When F56Bpa KCNE1 was co-expressed with KCNQ1 it produced functional currents (Appendix B, Fig. 1B, upper) with a similar V1/2 of activation to wt IKs (wt IKs, 26.1 ± 2.2 mV; F56Bpa IKs, 25.1 ± 6.9 mV; Appendix B, Fig. 1D). This indicates that Bpa is well tolerated at position F56 in KCNE1. Furthermore, when expressed in the absence of the unnatural amino acid, the currents resembled KCNQ1 alone (Appendix B, Fig. 1B, lower).  Application of a 300 ms flash of UV at −90 mV produced a current decrease similar to that seen in wt IKs (Fig. 3.6A and C), indicating that F56Bpa does not interact with KCNQ1 when the channels are closed. Interestingly, application of UV light during depolarizations to +60 mV produced an immediate upward deflection in current and preserved peak current levels over time (Fig. 3.6B and C). This increase in peak current, the absence of a speeding of current activation, and the failure of a constitutive current to appear—a feature of E1R/R4E + KCNE1 (Appendix B, Fig. 3B)—suggests that crosslinking is impacting pore behavior rather than the voltage sensor. To further demonstrate open- 77 state crosslinking with F56Bpa, the wt rundown was subtracted from the crosslinking observed (see 3.2.6 Materials and Methods). The resulting plot showed an upward single exponential curve (Fig. 3.6D), indicating that F56Bpa interacts with KCNQ1 in the open state.   Figure 3.6. F56Bpa KCNE1 interacts with KCNQ1 in the open state. (A) Shown here is a schematic of the -90 mV UV-voltage protocol (upper). A 300 ms flash of UV light (bold black line) is applied each sweep at -90 mV followed by a 4 s step to +60 mV. A representative set of currents is shown for KCNQ1 + F56Bpa (lower). (B) Shown here is a schematic of the +60 mV UV-voltage protocol (upper). A 300 ms flash of UV light (bold black line) is applied each sweep at the end of a 4 s voltage step. A representative current is shown for KCNQ1 + F56Bpa with the first UV sweep in grey (lower). Note the upturn of current upon UV exposure (inset). For all recordings, UV was applied at either sweep 6 (-90 mV) or sweep 5 (+60 mV), after a stable baseline had been established. Sweeps 1 and 5-25 are shown. (C) Given here is a diary plot of the UV-treated normalized peak current from KCNQ1 + F56Bpa -90 mV (black triangles), +60 mV (black squares), and KCNQ1 + wt E1 (open squares) (n = 3-10). Peak current was either measured at the end of the 4 s step (-90 mV) or immediately before UV application (+60 mV). (D) Given here is a plot of the rundown corrected and normalized peak current versus UV exposure for KCNQ1 + F56Bpa + 60 mV (black squares) and wt E1 (open squares) (n = 5-10) (see 3.2.6 Materials and Methods for rundown correction).  78 3.3.6 F56 crosslinks to E1R/R4E KCNQ1 E1R/R4E KCNQ1 + F56Bpa (Appendix B, Fig. 3C and E) channels are constitutively active like the other E1R/R4E constructs used in the study (Appendix B, Fig. 3). Three-hundred millisecond UV flashes at the end of 4.3 s pulses to +60 mV increased E1R/R4E KCNQ1 + F56Bpa peak currents over time (Fig. 3.7A), compared with the rundown seen in E1R/R4E KCNQ1 + wt E1 (Fig. 3.7B). Net crosslinking was obtained by subtraction of wt KCNE1 rundown (see 3.2.6 Materials and Methods) and appeared very similar to that seen in wt KCNQ1 + F56Bpa (Fig. 3.7C). The similar amplitudes and rates of current increase seen whether the voltage sensors are held in a constitutively active conformation (E1R/R4E KCNQ1), or are in various states of activation (wt KCNQ1; Fig. 3.6D), suggest that the F56Bpa crosslinking effects are mediated via the pore of the channel rather than the VS.  Figure 3.7. F56Bpa crosslinks to E1R/R4E KCNQ1 in the open state. (A) Shown here is a schematic of the +60 mV UV-voltage protocol (upper). A 300 ms flash of UV light (bold black line) was applied each sweep after a 3.7 s step to +60 mV. A representative  79 current is shown for E1R/R4E KCNQ1 + F56Bpa, with the first UV sweep in grey (lower). For all recordings, UV was applied at sweep 5 after a stable baseline had been established. Sweeps 1 and 5-25 are shown. (B) Shown here is a diary plot of the +60 mV UV-treated normalized peak current from E1R/R4E KCNQ1 + F56Bpa (black squares) (n = 4) and E1R/R4E KCNQ1 + wt E1 (open diamonds) (n = 4). Peak current was measured immediately before UV application. (C) Given here is a plot of the rundown corrected and normalized peak current versus UV exposure for E1R/R4E KCNQ1 + F56Bpa (black squares) (n = 4) and wt E1 (open diamonds) (n = 4).   3.3.7 F56 does not interact with KCNQ1 in a preopen closed state The incorporation of Bpa at position F56 did not disrupt the presence of multiple closed states during the activation process in KCNQ1. Holding KCNQ1 + F56Bpa at progressively more depolarized, but nonactivating membrane potentials (-110 to -50 mV) for 2 s, followed by a pulse to +60 mV for 4 s, decreased the t1/2 of activation for -70 and -50 mV when compared to -110 mV. This indicated the existence of preopen closed states in F56Bpa IKs, with very similar activation kinetics to wt and F57Bpa (Fig. 3.8B).  To examine the interaction of F56Bpa with KCNQ1 as channels progress through the activation pathway, UV was applied in the final 300 ms of a 2 s prepulse to -50 mV (Fig. 3.8C). Current decreased when UV was applied to KCNQ1 + F56Bpa channels at -50 mV at a similar rate as when holding at -90 mV and during rundown of KCNQ1 + wt E1 channels (Fig. 3.8D). This indicates that although F56 has likely moved positions within the channel complex at this preopen closed state, as shown with F57, Bpa has not come into close enough proximity to KCNQ1 for a functional crosslink to form.   80  Figure 3.8. F56Bpa does not interact with KCNQ1 in a pre-open closed state.  (A, upper) Shown here is a voltage-clamp protocol where channels are held at increasingly depolarized holding potentials (-110 to -50 mV) for 2 s before pulsing to +60 mV for 4 s. (A, lower and inset) A representative set of currents is shown for KCNQ1 + F56Bpa. (B) Given here is a change in time to half-maximum current normalized to -110 mV for wt (black circles) and F56Bpa IKs (grey squares) (n = 5-7; * = p < 0.01 compared to -110 mV). (C, upper) Given here is a schematic of the UV-voltage protocol where a 300 ms flash of UV light (bold black line) is applied each sweep during a 2 s hold at -50 mV. (C, lower) A representative current is shown for KCNQ1 + F56Bpa. UV was applied at sweep 6, after a stable baseline had been established. Sweeps 1 and 5-25 are shown. (D) Given here is a diary plot of the UV-treated normalized peak current from wild-type IKs (open squares), KCNQ1 + F56Bpa +60 mV (black squares), -90 mV (black triangles), and -50 mV (grey diamonds) (n = 3-10).  3.4 Discussion Neighboring residues in the transmembrane region  (F57 and F56) of KCNE1 interact with KCNQ1 and can be specifically crosslinked at rest and when the channel is activated during depolarizations,  respectively. By focusing on the closed state interaction of F57, reorientation of the KCNE1/KCNQ1 complex before channel opening was demonstrated.  81 Additionally, results indicate that both F56 and F57 modulate pore domain activity of KCNQ1, rather than regulating the activation state of the VS. 3.4.1  Comparing F56Bpa and F57Bpa An interesting outcome of the unnatural amino acid approach was that two residues one-helix position away from each other in the transmembrane domain of KCNE1, F56, and F57, exhibited opposite crosslinking effects. This difference in state-dependent crosslinking is consistent with the model presented by Wang et al. (2012), where KCNE1 rotates while traversing the KCNQ1 exterior cleft during channel activation and opening. At rest, F57 interacts closely with KCNQ1, whereas F56 does not. Upon depolarization of the membrane, rotation of KCNE1 would put F56 into closer contact with KCNQ1 and move F57 so that it can no longer be efficiently crosslinked to KCNQ1. Other sites of interest within the transmembrane domain of KCNE1 for unnatural amino acid studies included the final two members of the activation triplet, T58 and L59. Unfortunately, the incorporation of Bpa at these two sites was not well tolerated, and did not produce functional IKs currents. 3.4.2 Investigating crosslinking with F57Bpa It has previously been shown that F57 interacts with KCNQ1 in the closed state of the channel (Chen and Goldstein, 2007, Murray et al., 2016), but the state-dependence of interactions with this residue have not been reported. UV applied when KCNQ1 + F57Bpa channels are at rest (−90 mV; Fig. 3.1) prevented channel opening and channels held at more depolarized holding potentials (−70 to −50 mV) showed a progressive decrease in the rate of F57Bpa crosslinking (Fig. 3.5). The differences in crosslinking rates at these various preopen closed states are consistent with the proposed rotation of KCNE1 as it  82 is displaced across the cleft by KCNQ1 (Wang et al., 2012, Xu et al., 2008). Thus, the proximity of F57Bpa KCNE1 to KCNQ1 decreases from a plateau at −110 mV as the channel transitions out of deeper closed states between −90 and −50 mV. When a flash of UV was applied to F57Bpa during depolarizations, a downward deflection and a progressive decline of F57Bpa IKs current was seen. This is still consistent with closed-state crosslinking, and one explanation for the downward deflection is that channels are being crosslinked closed and are thus removed from the pool of channels available to open. These might be channels visiting a closed conformation after a burst or channels that are activated but not yet open. As discussed earlier, at the end of a 4 s pulse to +60 mV a majority of the channels remain in a closed state conformation (Werry et al., 2013). Additional studies, perhaps at the single channel level, are required to better understand the downward current deflection at +60 mV. Due to the low open probability of wt channels at the end of the 4 s steps to +60 mV (Werry et al., 2013), it was difficult to be sure that F57Bpa did not interact with KCNQ1 in the open state. Long pulse UV recordings of KCNQ1 + F57Bpa showed that when UV was applied during 8 s depolarizing pulses to +60 mV, when the Po is higher, the rate of crosslinking was significantly reduced from that after 4 s (Fig. 3.3). Additionally, a KCNQ1 mutant with a high Po of 0.8 at +60 mV after 4 s, E160R/R237E (E1R/R4E) was employed. E1R/R4E consists of two charge reversal mutations, where a negatively charged glutamic acid in the S2 domain of KCNQ1 is mutated to a positive arginine, and the opposite occurs to the fourth arginine in S4 (Wu et al., 2010). These mutations result in a constitutively active IKs channel, in which the voltage sensors are thought to be held in an activated conformation (Zaydman et al., 2014, Wu et al., 2010) (Appendix B, Figure  83 3). In this mutant, current decrease as a result of UV crosslinking was still observed, albeit at a significantly reduced rate of 0.17 ± 0.05 s-1 (Fig. 3.2). This finding suggests that crosslinking can directly modify pore opening and closing, and calculation of a crosslinking rate based on Po predicts a KXL of 0.162 s-1 in this construct (from the KXL at −90 mV, 0.81 * 0.2, the closed probability for E1R/R4E IKs at +60 mV). This is very close to the rate of crosslinking observed (KXL = 0.17 ± 0.05 s-1) and suggests that F57Bpa predominantly interacts with KCNQ1 in the closed state of the channel. 3.4.3 Investigating crosslinking with F56Bpa When UV was applied at +60 mV to F56Bpa IKs, there was an immediate upward deflection in current, followed by a preservation of peak current over wt rundown (Fig. 3.6). The UV-induced rundown seen in wt channels and the dynamic nature of the KCNQ1 pore (Werry et al., 2013) is likely why the crosslinking effect seen with F56Bpa KCNE1 at +60 mV was mild. When the wt UV-specific rundown was accounted for in the KCNQ1 + F56Bpa +60 mV data, crosslinking clearly resulted in more current (Fig. 3.6). As a result of open-state crosslinking, along with the upward deflection in peak current, one would expect to see an increase in the fast activating current, similar to a previous study where cysteines introduced at positions I145 in the VS of KCNQ1 and G40 in KCNE1 resulted in an instantaneous current (Xu et al., 2008). The fact that there was no accumulation of a constitutive current suggests that crosslinking did not constrain the voltage sensor. Additionally, E1R/R4E KCNQ1 + F56Bpa had a similar rate of crosslinking. These lines of evidence all indicate that the crosslinking preferentially modifies pore activity.  84 The UV-induced upward deflection of peak current in F56Bpa IKs suggests that the conductance of the pore is increased. One of the main functions of KCNE1 is to enhance conductance of KCNQ1, perhaps by reducing the flexibility of the selectivity filter such that K+ ions are better coordinated (Wang et al., 2012, Xu et al., 2015), and reducing the flicker behavior of the channel (Werry et al., 2013). Although we do not know where F56 is interacting with the pore, it is of interest that the selectivity filter is sensitive to introduction of cysteines throughout the transmembrane regions of KCNE subunits to alter the GRB/GK conductance ratio (Wang et al., 2012). Additionally, our lab has previously demonstrated that the IKs channel has multiple subconducting states (Werry et al., 2013) and stabilizing a higher subconducting state and/or preventing the channel from entering fast closed states might also underlie the small upward deflection in the peak current upon UV exposure. 3.4.4 Regions of interaction between KCNE1 and KCNQ1 The transmembrane domain of KCNE1 has previously been demonstrated to be an important factor in the regulation of KCNQ1. Using a chimeric approach, Tapper and George (Tapper and George, 2000) showed that by replacing the transmembrane domain of KCNE1 with that of KCNE2, there was a loss of IKs-like regulation. This was later shown to be conferred by the activation triplet, residues 57–59 (Melman et al., 2001). A set of three residues in the S6 domain of KCNQ1, 338–340, have also been identified as a site of specific interaction with the transmembrane domain of KCNE1, especially with T58 (Panaghie et al., 2006). Additionally, cysteine scanning mutagenesis found that the midsection of the KCNE1 transmembrane helix interacts with the S6 hinge, and the inner part of the helix interacts with the S4–S5 linker (Wang et al., 2012). Although there are  85 clear examples of the KCNE1 transmembrane domain interacting with the pore domain, it does not seem to be limited to this region. Xu et al. (2008) showed that residues in the extracellular end of the S1 domain of KCNQ1 and the extracellular end of the transmembrane domain of KCNE1 can form disulphide bonds with one another. From this study, they propose that KCNE1 is located between the S1, S4, and S6 of three individual KCNQ1 subunits (Xu et al., 2008). As well, cysteine crosslinking has been used to suggest that residues K41 and L42 in KCNE1 interact with the S1 and S6 domains, respectively, depending on the state the channel is in (Chung et al., 2009). Multiple studies reveal that KCNE1 can interact with both voltage sensor and pore domains of KCNQ1, so that a reduction in current could occur from F57Bpa crosslinking with the voltage sensor, trapping it in a downward position, or it might result from crosslinking in the pore domain of KCNQ1 to hold it closed. The E1R/R4E KCNQ1 + F57Bpa data offered some insight into these possibilities. Although the voltage sensor is held up, the channel can still close occasionally during depolarizations (Appendix B, Figure 3) and the slower rate of crosslinking is consistent with the decreased closed state probability as discussed above. Crosslinking was also observed with E1R/R4E KCNQ1 + F56Bpa (Fig. 3.7), and the occurrence of crosslinking in channels where the voltage sensor is prevented from returning to its closed conformation suggests that F56Bpa and F57Bpa may not crosslink to the voltage sensor, but interact with the pore domain of KCNQ1 instead. This idea is supported by Strutz-Seebohm et al. (2011), who placed F57 into close proximity with the pore. They noted that there was a strong energetic perturbation with F270A, a residue located in S5 of KCNQ1. From this, they proposed that F270 could perhaps π-stack with F57 (Strutz-Seebohm et al., 2011). Additionally,  86 Melman et al. (2004) used chimeras of KCNQ1 and other potassium channels that do not interact with KCNE1, along with cysteine-scanning methods, to propose that residues S338, F339, and F340 in the S6 of KCNQ1 interact with the activation triplet (residues 57–59) of KCNE1. 3.4.5 Regulation of KCNQ1 by KCNE1 Several models for how KCNE1 inhibits KCNQ1 activation have been proposed, including a detailed model put forward by the Sanders group. Using an experimentally restrained homology model of KCNE1 docking, Kang et al. (2008) proposed that significant adherent contacts between KCNE1 and KCNQ1 in the closed state form the basis for the delayed activation kinetics of IKs. When KCNE1 is docked to an open-state KCNQ1, KCNE1 displaces across the channel to reside in a “gain-of-function cleft” that helps stabilize the open state and increase the unitary conductance of the channel (Kang et al., 2008, Van Horn et al., 2011). Additionally, Barro-Soria et al. (2014) report that KCNE1 divides KCNQ1 voltage sensor movement into two components: an initial step, followed by a KCNE1-mediated concerted final step that results in channel opening (Barro-Soria et al., 2014). In Figure 3.9, the rates of crosslinking for F57Bpa IKs at -110, -90, -70, -50, and +60 mV are plotted against the conductance-voltage relationship of F57Bpa IKs and the fluorescence-voltage relationship of wt IKs obtained from Barro-Soria et al. (2014). The fastest rates of crosslinking (-110 and -90 mV) occur during the first component of the F-V, when the channel pore is closed and the voltage sensors are thought to be transitioning from a fully rested to an intermediate position. At -70 and -50 mV, the rate of crosslinking decreases as KCNE1 moves away from its optimum crosslinking alignment with KCNQ1. The rate of crosslinking at -50 mV (0.44 ± 0.08 s-1), when the channels are still closed,  87 is very similar to the rate at +60 mV (0.41 ± 0.05 s−1, p = 0.9996), when the channels have undergone the second voltage sensor movement and are opening. This indicates that most of the displacement of KCNE1 occurs before the channel opens, within the first component of the F-V.  Figure 3.9. Crosslinking rates of F57Bpa plotted against the G-V and F-V of IKs.  Given here is a plot comparing the rates of crosslinking of KCNQ1 + F57Bpa (-110, -90, -70, -50, and +60 mV) (grey triangles) to the G-V relationship of KCNQ1 + F57Bpa (solid black line; extracted from Appendix B, Fig. 1D) and the F-V of KCNQ1 + wt E1 (dashed black line). The F-V was extracted from Barro-Soria et al., 2014.  3.4.6 Conclusions Using an unnatural amino acid approach, the data show that residues F56 and F57 in the transmembrane domain of KCNE1 interact with KCNQ1 in a state-dependent manner. Additionally, the results confirm that the interactions between KCNQ1 and KCNE1 F57 change in a voltage-dependent manner before channel opening due to movement of KCNE1 as a result of conformational changes in KCNQ1. Finally, these results indicate that F56 and F57 modulate the pore domain activity of KCNQ1. Understanding more  88 about the interactions that occur between KCNE1 and KCNQ1 is important for the development of targeted therapeutics that can regulate repolarization in the heart.    89 Chapter 4: Individual voltage sensor movements result in IKs channel conductance 4.1 Introduction KCNQ channels are a family of five voltage-gated potassium channels (VGKCs) found throughout the body, whose members all play important roles in health and disease (Liin et al., 2015). KCNQ1 (Q1) in the heart is required for regulation of cardiac action potential duration, and thus cardiac inotropy and dromotropy. Loss- and gain-of-function mutations have been implicated in long and short QT-interval syndromes, respectively, (Bellocq et al., 2004, Franqueza et al., 1999, Splawski et al., 2000) which can lead to serious ventricular arrhythmias and sudden death. Q1 channels are known to associate with KCNE1 (E1) subunits in the heart to form the IKs current, and with other KCNE subtypes in many other tissues, although the full nature and diversity of their heteromultimeric interactions have still not been clearly elucidated (Liin et al., 2015).  The ratio of the number of Q1 to E1 subunits has been studied extensively, and while some report a fixed stoichiometry of four Q1 subunits for every two E1s (Plant et al., 2014, Wang and Goldstein, 1995, Morin and Kobertz, 2008, Chen et al., 2003a), we and other groups have shown that a variable stoichiometry is possible, with anywhere between one and four E1s associating within the channel complex, depending on the concentration of E1 available (Nakajo et al., 2010, Murray et al., 2016, Wang et al., 2011a, Wang et al., 1998). If channel complexes do exist with multiple E1:Q1 configurations, this would allow for greater flexibility upon modulation. This is because the presence of E1 subunits in the channel complex has a dramatic effect on all the physiologically-relevant properties of the channel including “voltage-dependence, current kinetics, inactivation, single channel  90 conductance, selectivity and pharmacology” (Zaydman et al., 2014). The profound changes that E1 can impose on Q1 channel properties have prompted a series of probing experimental and modeling studies over many years from a number of different laboratories to understand the gating and pore functions of the Q1 channel, with and without E1.  Electrophysiological data have been used to define state models of Q1 (Silva and Rudy, 2005) in the presence or absence of E1. These studies have been interpreted using the principles established by Hodgkin and Huxley (Hodgkin and Huxley, 1952), for potassium channels, where gating particles move independently, and once activated lead to pore opening via a concerted step, but do not affect ion permeation once the pore is open. This can be seen in the development of the models for Shaker channel gating, a member of the VGKC family (simplified in Fig. 4.1A)(Zagotta et al., 1994, Schoppa and Sigworth, 1998). The voltage sensor domains (VS) undergo two independent transitions to reach an activated conformation, before all four subunits undergo one, or multiple, concerted transitions (Schoppa and Sigworth, 1998) at which point the pore is able to conduct current. The presence of a concerted transition between VS activation and pore opening is reflected experimentally in the degree of voltage displacement between direct gating current measurements of VS charge movement (Q-V) (Bezanilla et al., 1994), or fluorescence measurements of VS movement (Mannuzzu and Isacoff, 2000, Pathak et al., 2005, Cha and Bezanilla, 1997) and the voltage dependence of the conductance-voltage (G-V) relationship (Mannuzzu and Isacoff, 2000). Similar observations of fluorescence-voltage (F-V) and G-V non-concordance have been made in the IKs channel complex (Osteen et al., 2010, Barro-Soria et al., 2014) and led these authors to  91 incorporate a concerted opening step in channel gating into their IKs models, with the implication that the VSs in all four subunits must be activated before pore opening can occur.  Figure 4.1. Models of Ion Channel Gating Simplified Markov schemes of IKs channel gating. (A) Independent gating domains each undergo two transitions, followed by a concerted step involving all four domains that leads to channel opening. Resting and intermediate states are represented by ‘R1’ and ‘R2’, fully activated states are represented by ‘A’ and the open state is represented by ‘O’. Forward rate constants are represented by ‘α’ and ‘γ’, while reverse rate constants are represented by ‘β’ and ‘δ’. This diagram was adapted from (Zagotta et al., 1994). (B) Allosteric model of channel gating in which activation of individual subunits can lead to opening. Closed states are represented by ‘C’, and the open state is represented by ‘O’. Forward and reverse rate constants between closed states are represented by ‘α’ and ‘β’ respectively. ‘Dn’ represents the allosteric coupling factor between closed and open states, with n referring to the number of activated subunits. As used in Fig. 9A, G0 = 37 s-1, D0 = 0.074 s-1, D1 = 7 s-1. The diagram was adapted from (Horrigan et al., 1999).   In contrast to this scheme of gating, others have suggested that the IKs channel pore may open before all VSs are activated. Using thermodynamic mutant cycle analysis with mutations that disrupt IKs channel opening, in either one, two, three or all four S4 domains, Meisel et al. noted that individual subunit movement contributed to channel conductance,  92 rather than a concerted step being required (Meisel et al., 2012). They likened IKs gating to the allosteric model of gating (simplified in Fig. 4.1B), proposed for Q1 alone (Osteen et al., 2010) where the pore can open or close regardless of the position of the VSs, even when they are at rest (Ma et al., 2011). As more VSs are activated, the pore has a higher open probability, exemplified by voltage activation of the BK channel, a voltage- and calcium-activated potassium channel (Horrigan et al., 1999). A logical extension of this reasoning is that the physical association of E1 (or PIP2) significantly influences but does not materially change the underlying allosteric gating mechanism of Q1 (Zaydman et al., 2014, Cui, 2016), and a concerted gating transition is not obligatory for pore opening. Recent support for an allosteric gating pathway comes from the modeling studies of Ramasubramanian and Rudy (2018) in which pore openings to small or intermediate subconductance levels are shown to be possible over a range of early and intermediate S4 translations (Ramasubramanian and Rudy, 2018). In this study we have experimentally addressed the question discussed above; whether the IKs channel complex can open and conduct current before all four VSs have moved. We utilized the E160R and F57W mutations (Zaydman et al., 2014, Restier et al., 2008, Chen and Goldstein, 2007) in Q1 and E1 respectively, in tandem or quadruple constructs, to restrain between one to four VSs in their resting conformations, and studied them using whole cell and single channel patch clamp techniques. Our results suggest that movement of all four VSs is not required in order to allow the IKs channel to conduct current, and thus that IKs channels do not need an obligatory final concerted transition to open like some other VGKCs, but rather are better represented by a model where the pore can open from multiple closed channel conformations. In addition, our data suggest,  93 in contrast to other allosteric models proposed for the IKs channel, that as well as VS movement and pore opening being coupled allosterically, closed-state transitions themselves may not be independent, and that inter-subunit cooperativity may contribute to transient and steady-state current kinetics.  4.2 Materials and Methods 4.2.1 Chemicals [2-(Trimethylammonium)ethyl]methanethiosulfonate Bromide (MTSET) was obtained from Toronto Research Chemicals Inc (North York, ON, Canada). Alexa Fluor 488 C5-maleimide was obtained from Thermo Fisher Scientific (Waltham, MA, USA). All other chemicals were obtained from Sigma-Aldrich (St. Louis, MO, USA). 4.2.2 Molecular Biology The EQ, EQQ, EQQQQ constructs were generated as previously described (Murray et al., 2016)(See Chapter 2.2.2). E1 was removed from EQQ and EQQQQ constructs, to produce QQ and QQQQ respectively. The E160R mutation was incorporated into the first Q1 of the fusion constructs via a gBlocks Gene Fragment (Integrated DNA Technologies, Coralville, IA, USA) with compatible restriction sites. The V319Y mutation, to increase TEA+ sensitivity of KCNQ1, was incorporated into the first Q1 of wt EQQ and E160R EQQ via a FragmentGENE (GENEWIZ, South Plainfield, NJ, USA) with compatible restriction sites. The tandem MTSET and fluorescence constructs were constructed by Applied Biological Materials Inc (Richmond, BC, Canada). All mutations were confirmed by sequencing. The F57W mutation was incorporated into E1 using the QuikChange II Site- 94 Directed Mutagenesis Kit (Agilent Technologies, Santa Clara, CA, USA) via the manufacturer’s protocol. 4.2.3 Cell Culture and Transfection tsA201 transformed human embryonic kidney 293 (TSA; whole cell experiments) or ltk-mouse fibroblast cells (LM; single channel experiments) were cultured and plated for experiments as previously described (Murray et al., 2016, Westhoff et al., 2017). Cells were transfected using Lipofectamine2000 (Thermo Fisher Scientific) as per the manufacturer’s protocol. The wild type and mutated (E)QQ/(E)QQQQ and EQ constructs were transfected with GFP in a 3:1 ratio. When co-expressed with E1-GFP (hereafter referred to as E1), the ratio was 3 E1:1 Q1 construct. Throughout the paper, the E160R mutation in Q1 is denoted by an asterisk (Q*) and the F57W mutation in E1 is denoted by an apostrophe (E’).  All experiments were performed 24-48 hrs post transfection at room temperature. Successfully transfected cells were identified by GFP-fluorescence. For recordings in the presence of E1, currents less than 1 nA or not expressing  IKs were discarded (except for E160R EQ*).  4.2.4 Oocyte Preparation Mature female Xenopus laevis frogs (Xenopus 1, Dexter, MI, USA) were anaesthetized in a solution containing: 2 g/L tricaine methanesulfonate, 2 g/L HEPES (pH 7.4 with NaOH). Under anesthesia, the animal was euthanized in accordance with the University of British Columbia animal care protocols. The ovarian lobes were extracted, divided into smaller sections and digested for 2-4 hours in a solution containing: 3 mg/mL collagenase  95 type 4 (Worthington Biochemical Corporation, Lakewood, NJ, USA), 82.5 mM NaCl, 2.5 mM KCl, 1 mM MgCl2, 5 mM HEPES (pH 7.6 with NaOH). The oocytes were washed and stored in a media containing: 500 ml Leibovitz’s L-15 medium (Thermo Fisher Scientific), 15 mM HEPES, 1 mM glutamine, 500 μM gentamycin, brought up to 1 L with distilled water (pH 7.6 with NaOH). Stage IV and V oocytes were selected and stored at 18°C. Oocytes were injected with cRNA synthesized using the Ambion mMessage mMachine T7 transcription kit (Applied Biosystems, Foster City, California, USA). 10 ng of C214A/G219C/C331A Q1 pcDNA3.1+ (G219C Q; a gift from Dr. Jianmin Cui) cRNA was injected, while 50 ng of the tandem constructs in pGEMHE was used [C214A/G219C/C331A Q-C214A/C331A Q (G219C Q-Q), C214A/G219C/C331A Q-E160R/C214A/C331A Q* (G219C Q-E160R Q*; G219C Q-Q*) and E160R/C214A/G219C/C331A Q-C214A/C331A Q (E160R/G219C Q*-Q; G219C Q*-Q)]. The pGEMHE vector was a gift from Dr. Yoshihiro Kubo. All constructs were co-injected with 5 ng E1 pBSTA. Experiments were performed 3-4 days post injection at room temperature. 4.2.5 Electrophysiology Solutions For whole cell recordings, the bath solution contained (in mM): 135 NaCl, 5 KCl, 1 MgCl2, 2.8 NaAcetate, 10 HEPES (pH 7.4 with NaOH). The pipette solution contained (in mM): 130 KCl, 5 EGTA, 1 MgCl2, 4 Na2-ATP, 0.1 GTP, 10 HEPES (pH 7.2 with KOH).  For single channel recordings, the bath solution contained (in mM): 135 KCl, 1 MgCl2, 1 CaCl2, 10 HEPES (pH 7.4 with KOH). The pipette solution contained (in mM): 6 NaCl, 129 MES, 1 MgCl2, 5 KCl, 1 CaCl2, 10 HEPES (pH 7.4 with NaOH).  96 For two electrode voltage clamp fluorometry (VCF) experiments, the bath solution contained (in mM): 96 NaCl, 3 KCl, 1 MgCl2, 2 CaCl2, 0.1 LaCl3, 5 HEPES (pH 7.4 with NaOH). The pipette solution contained 3 M KCl.   4.2.6 Patch Clamp Electrophysiology Whole cell and single channel currents were acquired using an Axopatch 200B amplifier, Digidata 1440A and pClamp 10 software (Molecular Devices, San Jose, CA, USA). For whole cell recordings, a linear multistage electrode puller (Sutter Instrument, Novato, CA, USA) was used to pull electrode pipettes from thin-walled borosilicate glass (World Precision Instruments, Sarasota, FL, USA) (Murray et al., 2016, Westhoff et al., 2017). Pipettes were fire-polished prior to use. Electrode resistances for whole cell recordings were between 1-2 MW, with series resistances <4 MW. Series resistance compensation of ~80% was applied to all whole cell recordings, with a calculated voltage error of ~1 mV/nA current. Whole cell currents were sampled at 10 kHz and filtered at 2-5 kHz (Murray et al., 2016, Westhoff et al., 2017).  Single channel electrodes were pulled from thick-walled borosilicate glass (Sutter Instrument) using the linear multistage electrode puller (Sutter Instrument). Electrodes were coated with Sylgard (Dow Corning, Midland, MI, USA). After fire polishing, single channel electrode resistances were between 40 and 60 MW (Thompson et al., 2017, Murray et al., 2016, Werry et al., 2013, Eldstrom et al., 2015). Two electrode voltage clamp experiments were performed using an Oocyte Clamp OC-725C (Warner Instruments, Hamden, CT, USA) and digitized using a Digidata 1440A (Molecular Devices) using the pClamp 10 software (Molecular Devices).   97 4.2.7 TEA+ Experiments Wt EQQ + E1, V319Y EQQ + E1 and E160R/V319Y EQ*Q (V319Y EQ*Q) + E1 were expressed in TSA cells. Tetraethylammonium chloride (TEA+) was made up to the desired stock concentration in whole cell patch clamp bath solution. After currents reached a stable baseline of 5 sweeps, 2 mM or 50 mM TEA+ was added to the bath. After the effect of TEA+ had stabilized, the TEA+ was washed off with the external bath solution. 4.2.8 MTSET Experiments C214A/G229C/C331A Q-E160R/C214A/C331A Q* (G229C Q-E160R Q*; G229C Q-Q*) + E1 or E160R/C214A/G229C/C331A Q*-C214A/C331A Q (E160R/G229C Q*-Q; G229C Q*-Q) + E1 were expressed in TSA cells. MTSET was made up to the desired stock concentration in whole cell external bath solution and aliquots were flash frozen in liquid nitrogen and stored at -80°C before use. While waiting for the current to reach a stable baseline, MTSET aliquots were removed from the freezer and kept on ice. 2 mM MTSET was added to the bath after the current maintained a stable baseline for five sweeps. 4.2.9 VCF Experiments Oocytes were labelled with 10 μM Alexa Fluor 488 C5-maleimide in a depolarizing high potassium solution containing (in mM): 98 KCl, 1.8 CaCl2, 5 HEPES (pH 7.6 with KOH) for 30 minutes on ice. The oocytes were washed with the bath solution and left on ice prior to recording. Fluorescence and ionic recordings were obtained simultaneously as previously described (Es-Salah-Lamoureux et al., 2010) with an Omega XF100-2 filter set (Omega Optical Inc, Brattleboro, VT, USA). Fluorescence recordings from the same oocyte were averaged to reduce signal noise errors. To correct for photobleaching, a  98 fluorescence signal was recorded in the absence of a voltage step, and was subtracted from the signal.  4.2.10    Data Analysis Conductance-Voltage (G-V) plots were obtained from tail current amplitudes. A Boltzmann sigmoidal equation was used to fit G-Vs (Prism 7, GraphPad Software, La Jolla, CA) to obtain the V1/2 and slope factor. All the V1/2s of activations and k-factors for the Q1 constructs in the presence and absence of E1 are shown in Table 4.1. Fluorescence-Voltage (F-V) plots were fit with a double Boltzmann equation (Prism 7, GraphPad Software). Current activation waveforms were fit with a single exponential equation in Prism 7, with the fit starting at 0.5 s. Deactivation curves were fit with a single exponential curve in Prism 7 for all 4 s of the deactivating pulse to extract time constants.  Single-channel current records were low-pass filtered at 2 kHz at acquisition using a −3dB, four-pole Bessel filter, sampled at 10 kHz, and digitally filtered at 200 Hz before analysis. Subconductance analysis was completed as previously described (Thompson et al., 2017) using the single channel search function in Clampfit (pClamp 10 Software, Molecular Devices) at 200 Hz. The levels were set to 0, 0.13, 0.19, 0.29, 0.44 and 0.66 pA following the “three-halves (3/2) rule” (Pollard et al., 1994). The ‘update level automatically’ function was not used. The search function detects each event, its amplitude level and event length. A representation of this idealization process was previously shown in Figure S8 of Thompson et al. (Thompson et al., 2017). The information obtained from the single channel search function was then transferred into spreadsheets for further analysis. Events briefer than 3.5 ms (2 x 1.75 ms (rise time))  99 were excluded from the subconductance analysis. When comparing closed dwell times greater than 30 ms, the first latency of the recording was not included. All results are reported as mean ± SE, unless otherwise stated. Statistical comparison was performed using either a one-way ANOVA with Tukey’s post hoc test or a two-tailed Student’s t-test. p-values less than 0.05 were considered to be statistically significant.  4.2.11    Modeling Markov modeling in Figure 4.9 was performed using the IonChannelLab software (Santiago-Castillo et al., 2010) incorporating Q-matrix solutions to the differential equations defining the kinetic behavior of rate transitions (Colquhoun and Hawkes, 1995). The model was constructed with as few variables as possible and microscopic reversibility was maintained. Simulations were generally run as a 4 s depolarization and 2 s repolarization to match experimental protocols, and so data tended to be isochronal and not steady state. Data from simulations were exported to Excel (Microsoft Office, Redmond, WA, USA) for fitting and analysis. The voltage-dependence of rates between channel closed conformations was initially set using the experimental data for isochronal activation of the wt channel at 4 s; V1/2 (31.5 mV), and activation and deactivation time constants. The pore opening and closing rate constants, ‘G0’ , ‘D0’, and ‘D1’, were set given the very low open probability of resting IKs channels, and a latency to first opening of ~1.5 s (Werry et al., 2013). The VS coupling to closed channel conformations was set via the rate constants ‘l’ and ‘b’. When the values for ‘l’ and ‘b’ become larger, the effective rigidity of coupling between the VSs and the different channel closed conformations increases, and in the limit, VS conformations completely determine the accessibility to closed channel conformations, such that stabilized E160R subunits completely prevent  100 access to corresponding closed, and open channel conformations. A geometric progression of the form ‘arn-1’ was used for the exponents of ‘l’ and ‘b’, where ‘a’= 1 and ‘r’ = 3, to model cooperativity between closed states and ‘n’ indicates the number of up to four E160R subunits in the complex. This geometric progression allows the model to follow the non-linear shift in V1/2 seen experimentally, whereby there is no shift when one VS is locked down, versus a  shift of ~30 mV when three of four VSs are locked down. Additionally, this was necessary to ensure that no significant current was observed from EQ* at potentials >+100 mV. Exploration of a range of values for ‘l’ and ‘b’ allowed us to simulate scenarios that provided a good fit to the experimental data. Values for rate constants may be found in Figure 4.9 legend.  4.3 Results 4.3.1 Channels with the E160R mutation in all four Q1 subunits are non-functional In wild type (wt) channels, the VS of Q1 respond to changes in the membrane potential in a manner that eventually increases channel pore open probability. Interactions between the negative charge, E160, in the S2 domain, with the positive charges in the S4 domain facilitate the movement of the VS (Fig. 4.2A, left)(Zaydman et al., 2014). Movement of the VS, via its coupling to the pore allows the pore to open and conduct current (Zaydman et al., 2013). By converting the negatively charged glutamic acid in the S2 of Q1 into a positively charged arginine, E160R, it is proposed that the VS is held in a resting conformation via electrostatic repulsion (Zaydman et al., 2014) (Fig. 4.2A, right).    101  Figure 4.2. E160R EQ* is a non-functional channel. (A) Cartoons representing the putative interactions between charges within the S2 and S4 domains of Q1 during normal activation in wt channels (left) and when S4 movement is impeded by the presence of E160R (right). Adapted from (Zaydman et al., 2014). (B) Currents were obtained using an isochronal 4 s activation protocol with pulses from -80 to +100 mV, or higher. Sweep to sweep intervals were 15 s. Representative currents are shown for wt EQ (C), E160R EQ* (D) and untransfected cells (E). Cartoons describe the mutations in the constructs: E1 (small grey circles), wt Q1 (unfilled circles) and E160R Q1 (red circles). Black lines indicate tethers between subunits. (F) Plot representing the peak current density of wt EQ (black circles, n = 3), E160R EQ* (red triangles, n = 42) and untransfected cells (black open squares, n = 4) at various membrane potentials. All error bars denote mean ± SE, unless otherwise stated.  In order to investigate channel function when all VSs are held in the resting state, a 1:1 construct where the C-terminus of E1 was tethered to the N-terminus of Q1 (KCNE1-KCNQ1; EQ) was used. To determine the V1/2 of activation, a 4 s activation protocol ranging from -80 mV to +100 mV, or higher, was used (Fig. 4.2B). Wt EQ (31.5 mV) produces currents with a similar V1/2 of activation to wt Q1 + E1 (26.1 mV) (Fig.  102 4.2C)(Murray et al., 2016). E160R EQ* (EQ*) channels have the E160R mutation in all four Q1 subunits and do not produce a recognizable signal over the endogenous current of untransfected TSA cells (Figs. 4.2D-E), as well as having a similar current density (Fig. 4.2F). This phenomenon was previously described for E160R Q1* +/- E1 expressed in oocytes, by others (Restier et al., 2008, Zaydman et al., 2014). Additionally, it has previously been shown that this lack of current is not due to a trafficking problem, as a biotinylation assay indicated that the mutant channels are able to reach the membrane (Wu et al., 2010). 4.3.2 IKs with one, two or three VSs held down produce functional channels To investigate what happens to channels when one, two or three VSs are held down, constructs with tethered Q1 and/or E1 subunits were used ((E)QQ or (E)QQQQ). The E160R mutation was incorporated into Q1 subunits of each construct to produce a channel complex where either one (E160R Q*QQQ; hereafter referred to as Q*QQQ), two (E160R Q*Q; hereafter referred to as Q*Q) or three (E160R EQ*QQ*Q*; hereafter referred to as EQ*QQ*Q*) VSs are held down. In all of these situations, robust IKs-like currents were observed (Fig. 4.3), with at first glance only minor differences between the mutant and wt currents.  When expressed in the absence of E1, wt QQQQ and wt QQ (Appendix C, Figs. 1A and 2A) produced currents with a similar V1/2 to that previously reported for Q1 alone (-21.9 and -19.6 mV; Appendix C, Fig. 1E and Table 4.1)(Murray et al., 2016). When one VS is locked down by E160R (Q*QQQ) there was no significant shift in the V1/2 of activation compared with wt, with a V1/2 of -18.0 mV, but when two VSs are held down (Q*Q), the mean V1/2 of activation was significantly depolarized by 17 mV to -3.0 mV (Appendix C,  103 Figs. 1B-E and Table 4.1). Expression of wt constructs with E1-GFP (hereafter referred to as E1), wt QQQQ + E1 (Fig. 4.3A) and wt QQ + E1 (Appendix C, Fig. 2B), produced normal-looking IKs currents with similar V1/2s to that previously reported for fully saturated IKs (32.1 and 34.1 mV; Table 4.1)(Murray et al., 2016). Interestingly, Q*QQQ + E1 (Fig. 4.3B) currents have a similar V1/2 to wt (33.6 mV; Figs. 4.3E-F and Table 4.1). With Q*Q + E1 (Fig. 4.3C), the channel pore was still able to conduct current; however, as for constructs without E1, the mean V1/2 of activation was significantly depolarized by 16 mV to 49.9 mV (Figs. 4.3E-F and Table 4.1). Next, the location of the two E160R mutations within the tandem construct was altered from opposing (Q*Q) to adjacent Q1 subunits (E160R QQQ*Q*; QQQ*Q*). QQQ*Q* + E1 had a similar depolarized V1/2 of activation (55.6 mV) as Q*Q + E1 (49.9 mV) (p-value = 0.31), indicating that there was no significant difference in the V1/2 of activation when changing the location of the two E160R subunits (Appendix C, Fig. 3A-C and Table 4.1).  When three VSs are locked down (EQ*QQ*Q* + E1) (Fig. 4.3D), the currents produced had a very low macroscopic conductance and a further depolarized V1/2 of activation (61.6 mV; Fig. 4.3E-F and Table 4.1). There is a clear non-linear relationship between the V1/2s of activation for wt IKs currents, and those containing one, two, or three mutant subunits.   104  Figure 4.3. IKs channel complexes with one, two or three E160R mutations produce currents. Currents were obtained using the same isochronal activation protocol shown in Figure 4.2B (E160R EQ*QQ*Q* + E1 was recorded in 20 mV steps). Representative currents are shown to +90 mV for wt QQQQ + E1 (A), E160R Q*QQQ + E1 (B), E160R Q*Q + E1 (C) and  +100 mV for E160R EQ*QQ*Q* + E1 (D). Cartoons describe the mutations in the constructs: E1 (small grey circles), wt Q1 (unfilled circles) and E160R Q1 (red circles). Black lines indicate tethers between subunits. (E) G-V plots of wt QQQQ + E1 (purple triangles), E160R Q*QQQ + E1 (purple circles), E160R Q*Q + E1 (green circles) and E160R EQ*QQ*Q* + E1 (blue circles) (F) Summary of the V1/2 of activation for each construct (n = 3-8; * = p<0.05; Table 4.1).    105 Construct V1/2 (mV) mean ± SEM k-factor mean ± SEM n (cells) Wt QQQQ -21.9 ± 3.8 13.6 ± 3.9 3 Wt QQQQ + E1 32.1 ± 2.9 19.9 ± 0.7 7 Wt EQQQQ + E1† 30.9 ± 1.4 18.0 ± 1.6 5     Wt QQ -19.6 ± 3.9 14.5 ± 1.4 4 Wt QQ + E1 34.1 ± 1.9 19.7 ± 1.7 3 Wt EQQ + E1† 31.9 ± 3.0 20.2 ± 1.4 4     Wt EQ† 31.5 ± 0.5 20.0 ± 1.2 3     E160R Q*QQQ -18.0 ± 3.5 13.1 ± 0.2 4 E160R Q*QQQ + E1 33.6 ± 1.9 17.6 ± 0.9 5 E160R EQ*QQQ + E1 32.0 ± 3.1 17.9 ± 0.8 4     E160R Q*Q -3.0 ± 3.9 18.6 ± 2.2 3 E160R Q*Q + E1 49.9 ± 2.8 20.0 ± 0.9 8 E160R EQ*Q + E1 46.2 ± 1.9 21.4 ± 1.4 3 E160R QQQ*Q* + E1 55.6 ± 4.6 21.6 ± 2.7 3     E160R EQ*QQ*Q* + E1 61.6 ± 0.4 22.4 ± 1.1 3     F57W E'QQQQ + E1 34.8 ± 2.1 17.3 ± 0.8 3 F57W E'QQ + E1 42.3 ± 2.9 20.0 ± 0.5 3 F57W E'Q 110.7 and 101.3  31.6 and  36.1 2     G229C Q-E160R Q* + E1 39.8 ± 2.3 23.0 ± 1.3 3 E160R/G229C Q*-Q + E1 41.6 ± 4.7 20.5 ± 0.8 4     G219C Q + E1 20.9 ± 6.4 21.3 ± 0.8 5 G219C Q-Q + E1 10.7 and  1.5 16.5 and  13.8 2 G219C Q-E160R Q* + E1 23.5 ± 1.8 16.5 ± 1.1 6 E160R/G219C Q*-Q + E1 19.4 ± 6.5 17.3 ± 2.5 3 † denotes data from Murray et al, 2016.  Table 4.1. V1/2 of activation for Q1 constructs in the presence and absence of E1.  106 4.3.3 Subunits with the E160R mutation are not excluded from channel assembly. In order to confirm that the relatively mild changes in the channel kinetics with additional E160R mutations were not due to the exclusion of E160R subunits from the channel complex, the tetraethylammonium chloride (TEA+) sensitivity of channels was examined (Appendix C, Fig. 4). Previously, others have shown that wt Q1 + E1 channels are insensitive to addition of up to 50 mM TEA+ extracellularly (Kurokawa et al., 2001), and introduction of the V319Y mutation into the outer pore of Q1 significantly increases the channel’s sensitivity to TEA+ (Kurokawa et al., 2001). Here, we too show that addition of 2 mM and 50 mM TEA+ to wt EQQ + E1 channels (Appendix C, Fig. 4A and 4D; n = 4; p-value between control and 50 mM = 0.9209) does not affect the peak current. Inclusion of V319Y into the first Q of EQQ (V319Y EQQ), increased the sensitivity to TEA+ with 2mM TEA+ causing an approximately 25% reduction in peak current (n = 5; p-value = <0.0001), an effect which was reversed upon wash off. E160R/V319Y EQ*Q has the E160R and V319Y mutations in the same subunit, and produces currents with a very low conductance. Addition of 2 mM TEA+ to E160R/V319Y EQ*Q + E1 had a similar effect to that on the V319Y mutant alone, with a reduction in peak current of approximately 27% (n = 4; p-value = <0.0001), which was reversed upon wash off with control solution (Appendix C, Fig. 4C and 4D). This result indicates that subunits containing the E160R mutation are not excluded from IKs channel complexes during assembly.  4.3.4 Activation kinetics of IKs when one, two or three VSs are held down. To investigate whether holding one, two or three VSs down alters the activation kinetics of the IKs channel, current waveforms between +50 mV and +160 mV were fit with a single  107 exponential starting at 0.5 s (Appendix C, Fig. 5A-F) and activation time constants (τact) extracted and plotted against the membrane potential. Noting that this method of fitting omits most of the delay during channel activation, there was no difference between the τact-V curve for wt QQQQ + E1, wt QQ + E1 and wt EQQQQ + E1 (Appendix C, Fig. 5G). For channels where one VS was held down (Q*QQQ + E1), the τact-V curve overlies that of wt QQQQ + E1 (Fig. 4.4A). Whereas, when two VSs were held down (Q*Q + E1), the τact-V curve was depolarized compared with wt, consistent with the approximately 16 mV depolarizing shift in the V1/2 of activation. For three VSs locked down (EQ*QQ*Q* + E1), the τact-V curve overlaps with that of two VSs locked down. If this curve were to be corrected for the approximately 30 mV depolarization in V1/2 from wt, it would be hyperpolarized to the wt τact-V curve, which suggests an acceleration of current activation kinetics in EQ*QQ*Q* + E1 (Fig. 4.4A).   Figure 4.4. Activation and deactivation kinetics of complexes with one, two or three VSs locked down. The time constants of IKs current activation (τact) (A) and deactivation (τdeact) (B) at various membrane potentials were plotted for wt QQQQ + E1 (purple triangles), E160R Q*QQQ + E1 (purple circles), E160R Q*Q + E1 (green circles) and E160R EQ*QQ*Q* + E1 (blue circles) (n = 3-10). τact vs. membrane potential were fit with a single exponential equation to produce a τact-V curve for wt QQQQ + E1 (purple line), E160R Q*QQQ + E1 (dashed purple line), E160R Q*Q + E1 (dashed green line) and E160R EQ*QQ*Q* + E1 (dashed blue line).    108 4.3.5 Rate of deactivation increases progressively as multiple VSs are held down. In order to investigate changes in deactivation rate, the protocol shown in Appendix C, Figure 6A was used. The resulting tail currents were fit with a single exponential curve from which a deactivation time constant (τdeact) was extracted (Appendix C, Fig. 6B-G), and plotted against the membrane potential. Wt QQQQ + E1, wt QQ + E1 and wt EQQQQ + E1 had very similar τdeact at different membrane potentials (Appendix C, Fig. 6H), indicating that there was no difference between the deactivation rates when two or four Q1s are linked together. Q*QQQ + E1 channels deactivated faster than wt (Fig. 4.4B and Appendix C, Fig. 6E), significantly so at -50 and -60 mV. With Q*Q + E1, the rate of deactivation was further increased (Fig. 4.4B and Appendix C, Fig. 6F), and was significantly different from wt at -50 to -90 mV (not -70 mV). Deactivation rates for EQ*QQ*Q* + E1 were even faster, significantly so at -50 to -100 mV (not -80 mV) (Fig. 4.4B and Appendix C, Fig. 6G). These data indicate that as more VSs are held down, there is a progressive increase in the rate of deactivation, suggesting perhaps that either the restrained VSs can exert physical effects on the open conformations of the channel to accelerate closing, or that there are fewer mobile VSs to stabilize the open pore.   4.3.6 Single channel recordings show reduced conductance of channels with E160R subunits To further understand the effects of locking one, two or three VSs down on the IKs channel at the microscopic level, single channel recording was used. Constructs with and without tethered E1’s ((E)Q*QQQ + E1, (E)Q*Q + E1 and EQ*QQ*Q* + E1) were utilized. At the whole cell level, the tethered E1 does not interfere with the activation kinetics of IKs (  109 Appendix C, Fig. 7 and Table 4.1). Previously, we have reported the single channel characteristics of the IKs channel (Murray et al., 2016, Werry et al., 2013), and in Figure 4.5B, we again show that when cells are pulsed to +60 mV for 4 s, wt EQ channels open through multiple sublevels, with the histogram of opening event amplitudes of the representative sweep and average of 49 sweeps (Fig. 4.5F, G, black) showing a dominant peak at 0.42 pA. Channel complexes where one VS is locked down, EQ*QQQ + E1, produced currents with a flickering phenotype (Fig. 4.5C). While most events were to smaller conductance levels, the channel can infrequently open to the main level seen in the wt EQ channel (Fig. 4.5F, purple). The all-points histogram of 49 sweeps of Q*QQQ + E1 shows a similar trend, with a reduced number of events at larger amplitudes and an increase in those at lower conductance levels (Fig. 4.5G, purple).  Ensemble averages of 17 active wt EQ sweeps and 17 active Q*QQQ + E1 sweeps are shown in Appendix C, Figure 8A and although single channel recordings from Q*QQQ + E1 differ significantly from wt EQ (Fig. 4.5C, F, G and Appendix C, Fig. 8B), when averaged, the latency and waveform of averaged channel activity of both channel complexes are not dissimilar (Appendix C, Fig. 8A), consistent with whole cell currents (Fig. 4.3F).   110   Figure 4.5. Single channel recordings of channels with one, two or three VSs locked down. (A) Single channel currents were obtained using a protocol where cells were stepped from -80 to +60 mV for 4 s, followed by a tail current at -40 mV. Sweep to sweep intervals for single channel recordings were 10 s. Representative single channel traces are shown for wt EQ (B), E160R EQ*QQQ + E1 (C), E160R EQ*Q + E1 (D) and E160R EQ*QQ*Q* + E1 (E). Cartoons describe the mutations in the constructs: E1 (small grey circles), wt Q1 (unfilled circles) and E160R Q1 (red circles). Black lines indicate tethers between subunits. (F) All-points histograms for the displayed single sweep of wt EQ (black), E160R EQ*QQQ + E1 (purple), E160R EQ*Q + E1 (green) and E160R EQ*QQ*Q* + E1 (blue) are shown. Arrows highlight the amplitude peak in the open event distribution of each histogram based on Gaussian fits in Clampfit (0.42 pA for wt EQ, 0.25 pA for E160R EQ*QQQ + E1, 0.2 pA for E160R EQ*Q + E1 and 0.08 pA for E160R EQ*QQ*Q* + E1). (G) All-points histograms for 49 active sweeps of wt EQ (black line, n = 5),  E160R Q*QQQ + E1 (purple line, n = 3), E160R EQ*Q + E1 (green line, n = 3) and E160R EQ*QQ*Q* + E1 (blue line, n = 2).  111 When two VSs were locked down, conductance was further reduced (Fig. 4.5D), with the majority of the events occurring at approximately 0.2 pA and smaller (Fig. 4.5F, green), better highlighted by the all-points histogram of multiple sweeps shown in Figures 4.5G and 4.6A. A similar result is seen when two E160R mutations are located in adjacent subunits (QQQ*Q* + E1) (Appendix C, Fig. 3D-E), correlating with the whole cell data (Appendix C, Fig. 3A-C). Interestingly, although the channels with two E160R mutations reach lower amplitudes most of the time, they are occasionally still able to reach higher conductance levels, with events occurring above 0.4 pA (Fig. 4.5F and G, green).  In channel complexes with three VSs locked down, there was a continuing trend toward brief, flickering bursts of opening to lower conductance levels (Fig. 4.5E). The histogram of the representative sweep (Fig. 4.5F, blue) shows openings to a main peak of 0.08 pA, as does the all-points histogram of 49 sweeps of E160R EQ*QQ*Q* + E1 (Fig. 4.5G, blue).  Shown in Figure 4.6, are all-points and idealized amplitude event histograms comparing 49 active sweeps of wt EQ from five cells, EQ*Q + E1 from three cells and EQ*QQ*Q* + E1 from two cells. Idealization involved separately binning event data from each of the three constructs into one closed and five open subconducting levels (0.13 pA, 0.19 pA, 0.29 pA, 0.44 pA and 0.66 pA). Total dwell time data for the closed and open levels are shown in Figure 4.6C, and for closed durations longer than 30 ms, in Appendix C, Figure 8E. When two VSs are locked down, there was increased occupancy of the closed state, with 87.1% of the total dwell time occurring in the closed state, as opposed to 75.3% in wt EQ (Fig. 4.6C). For three VSs locked down, this was further increased to 92.8% (Fig. 4.6C), which is significantly different from wt EQ (p-value = 0.0307; Appendix C, Fig. 8D).  112 Additionally, EQ*Q + E1 more frequently occupied lower subconductance levels, while the time spent occupying the 0.44 pA level decreased from 9.3% in wt to 1.0% (p-value = 0.0177; Fig. 4.6C and Appendix C, Fig. 8D), with a further reduction to 0.1% when three VSs are locked down (p-value compared to wt EQ = 0.0207). When two VSs are locked down, channels were again still able to occupy the highest conductance level, with 0.2% of the total dwell time occurring at 0.66 pA, albeit reduced from the time spent by wt EQ (1.9%, Fig. 4.6C). For three VSs locked down, meaningful occupancy of this level was not achieved.   Figure 4.6. Subconductance analysis of a channel with two or three VSs locked down. Raw all-points amplitude histograms (A) and idealized histograms (B) from 49 active sweeps of wt EQ (black line, n = 5), E160R EQ*Q + E1 (green line, n = 3) and E160R EQ*QQ*Q* + E1 (blue line, n = 2) at +60 mV. The thresholds used for the idealization were: 0.13, 0.19, 0.29, 0.44 and 0.66 pA. (C) Total dwell times for each threshold, the percentage of time spent at each level and  113 the number of events at each threshold are shown for wt EQ (upper), E160R EQ*Q + E1 (middle) and E160R EQ*QQ*Q* + E1 (lower). The summed total dwell time for all states was 140292 ms for wt EQ, 148368 ms for E160R EQ*Q + E1 and 176797 ms for E160R EQ*QQ*Q* + E1. These data were filtered at 200 Hz. The bin width used was 0.01 pA. Events <3.5 ms were excluded.  Individual idealized amplitude event histograms of the three EQ*Q + E1 cells and two EQ*QQ*Q* + E1 cells included in the subconductance analysis show that each individual recording had a similar trend in the occupancy of the subconductance states (Appendix C, Fig. 8C). The mean duration of long closings (closed dwell times >30 ms) not including the first latencies is compared for different constructs in Appendix C, Figure 8E. For 49 sweeps of wt EQ, mean duration was 139 ms. This was unchanged when two VSs were held down (110 ms; p-value = 0.63) or when three VSs were held down (157 ms; p-value = 0.86). However, the total duration of closed times >30 ms increased by approximately 14 s in EQ*Q + E1 and approximately 38 s in EQ*QQ*Q* + E1, compared with wt (42 s). This was accounted for by an increase in the number of closed events >30 ms from 334 in 49 wt sweeps to 497 events with two VSs locked down, and 570 with three VSs locked down. These data suggest that locking two or three VSs down leads to a less stable open state and increases the overall time the channels spent closed.  4.3.7 F57W, a depolarizing E1 mutant, corroborates the behavior of E160R Whole cell recordings from IKs complexes containing different numbers of the E1 mutant, F57W, which is known to drastically depolarize the V1/2 of IKs  (Chen and Goldstein, 2007), provide supportive data (Appendix C, Fig. 9) to our results with E160R. Replacement of  114 phenylalanine with the bulkier tryptophan in F57W is likely causing steric hindrance that must be overcome to drive the channel complex from the closed to open state.  One E1 mutated with F57W (F57W E’QQQQ + E1) had a similar V1/2 to wt (34.8 mV; Appendix C, Fig. 9A-D and Table 4.1), whereas the V1/2 of two mutated E1 (F57W E’QQ + E1) was depolarized by approximately 10 mV (to 42.3 mV). Unlike EQ*, four F57W E1s in the complex (F57W E’Q) still allowed the channel to conduct, but with a V1/2 of 106 mV.  At the single channel level, there were subtle differences between complexes containing the E160R mutation and those containing the F57W mutation.  Like the E160R containing mutants, the F57W mutants showed a reduction in the overall conductance levels with increasing number of mutated subunits. Appendix C, Figure 9E shows examples of active single channel sweeps of Q*Q + E1 and F57W E’QQ +E1. While both channel types largely occupy lower conductance levels, it is clear that F57W-containing channels showed a greater number of brief visits to higher conductance levels (Appendix C, Fig. 9F).  4.3.8 VSs with the E160R mutation are locked down To ensure that the E160R mutation locks the VS in a closed conformation, extracellular MTSET modification was used to track the external displacement of the VSs during channel activation in mammalian cells. To investigate this, two constructs were created. For both, a QQ fusion construct with a cysteine-less background (C214A/C331A) in both Q1 subunits was used. In the first construct, the MTSET G229C probe was in the first Q1, and the second Q1 contained the E160R mutation (G229C Q-Q*). The second construct  115 had both the E160R and G229C mutations in the first Q1 (G229C Q*-Q). When both MTSET fusion constructs were expressed in the presence of E1, they produced currents with a similar V1/2 of activation, 40-42 mV (Appendix C, Figs. 10A-D and Table 4.1). However, the G229C Q-Q* + E1 currents clearly had a faster rate of activation and produced currents that were less sigmoidal. This is likely due to the effect of the G229C mutation to delay deactivation, as increasing the sweep to sweep interval to 60 s resulted in a more sigmoidal activation (Appendix C, Fig. 10E-G). It is known that MTSET modification of G229C Q1 + E1 results in currents that do not fully deactivate at -90 mV, and thus show loss of sigmoidicity during activation and a constitutive current (Rocheleau and Kobertz, 2008). In Figure 4.7A, cells were pulsed to +60 mV for 4 s. After reaching a stable baseline, 2 mM MTSET was added to the bath at sweep 6. For G229C Q-Q* + E1, the instantaneous current dramatically increased, while the shape also became less sigmoidal. A diary plot of MTSET modification shows that, after addition of 2 mM MTSET, the time to half-maximum activation (t1/2) decreased (Fig. 4.7C; red squares). In Figure 4.7D, the t1/2 was shown to be significantly reduced when comparing before (sweep 5) and after (sweep 20) addition of MTSET. This indicates that MTSET is able to modify the two freely moving VSs in G229C Q-Q* + E1. Addition of 2 mM MTSET to cells expressing G229C Q*-Q + E1 did not alter the activation time course or amount of instantaneous current (Fig. 4.7B). The decrease in peak current seen towards the end of the recording is similar to rundown observed in wt channels in the absence of MTSET (Appendix C, Fig. 10H). Furthermore, there was no decrease in the t1/2 after addition of MTSET (Fig. 4.7C; blue triangles), with no significant change in the t1/2 before (sweep 5) or after (sweep 20) addition of 2 mM MTSET (Fig. 4.7D). This result  116 shows that the E160R mutation in the same subunits as G229C prevents the G229C from being modified by MTSET when channels are activated. This suggests that the presence of the E160R mutation in a VS is able to prevent its outward translation during channel activation, and supports the idea that VSs containing E160R are ‘locked down’.   Figure 4.7. Extracellular MTSET is unable to modify subunits containing the E160R mutation. Diary plot currents were obtained via a protocol where cells were held at -90 mV, pulsed to +60 mV for 4 s, and -40 mV for 0.9 s. The sweep to sweep interval was 20 s. Representative currents are shown for G229C Q-E160R Q* + E1 (A) and E160R/G229C Q*-Q + E1 (B). Sweep 5 is highlighted in grey. After a stable baseline of 5 sweeps, 2 mM MTSET was applied at sweep 6. Cartoons describe the mutations in the constructs: E1 (small grey circles), wt Q1 (unfilled circles), E160R Q1 (red circles), G229C Q1 (blue circles), E160R/G229C Q1 (blue circles with red border). Black lines indicate tethers between subunits. (C) A diary plot showing time to half maximum peak current (t1/2) plotted versus sweep number for G229C Q-E160R Q* + E1 (red squares) and E160R/G229C Q*-Q + E1 (blue triangles) (n = 6-8). (D) Plot comparing t1/2 for both G229C Q-E160R Q* + E1 and E160R/G229C Q*-Q + E1 before (sweep 5) and after (sweep 20) addition of MTSET (* = p<0.05).  117 Voltage clamp fluorometry (VCF) was used as an alternative method to follow the displacement of VSs during activation. Previously, other groups had shown that the labelling of a cysteine at the top of the S4 of Q1, G219C, with Alexa 488 C5-maleimide, allows tracking of VS movement alongside the conductance of the IKs channel in oocytes (Zaydman et al., 2014, Barro-Soria et al., 2014). When a channel where G219C is located in all four VSs with a cysteine-less background (G219C Q) was expressed with E1 in oocytes, a current with a V1/2 of 20.9 mV resulted (Fig. 4.8B upper, Fig. 4.8E, and Table 4.1). As previously shown (Barro-Soria et al., 2014), fluorescence movement (Fig. 4.8B lower) occurs in two steps, with one step occurring at quite negative potentials, and the second step occurring at higher voltages where pore opening occurs (Fig. 4.8E).  In order to confirm that E160R locks VSs down, QQ fusion constructs with a cysteine-less background were used. In the first construct, G219C was in one Q1 and the other Q1 contained the E160R mutation (G219C Q-Q*). A second construct was made where the G219C and E160R mutations were in the same Q1 (G219C Q*-Q). Expression of both of these constructs in oocytes, with E1 (Fig. 4.8C, 4.8D, upper and Fig. 4.8E), produced currents with a similar V1/2 of activation to G219C Q + E1 (19-24 mV; Table 4.1), and were both depolarized from a similar QQ fusion construct without the E160R mutation, with a V1/2 of 6.1 mV (G219C Q-Q; Appendix C, Fig. 11 and Table 4.1). Fluorescence recordings of G219C Q-Q* + E1 (Fig. 4.8C, lower), were smaller than that seen in G219C Q + E1, which is to be expected as two fewer VSs are being tracked. A similar two-step VS movement was detected, though, albeit with a shallower slope (Fig. 4.8E). For G219C Q*-Q + E1, fluorescence changes were not observed, indicating that the presence of the E160R mutation was impeding VS movement (Fig. 4.8D, lower).   118  Figure 4.8. Fluorescence changes are not detected in subunits containing the E160R mutation. Recordings were obtained using the protocol in (A). Sweep to sweep intervals were 15 s. Representative currents (upper) and fluorescence recordings (lower) are shown for G219C Q + E1 (B), G219C Q-E160R Q* + E1 (C) and E160R/G219C Q*-Q + E1 (D). Cartoons describe the mutations in the constructs: E1 (small grey circles), wt Q1 (unfilled circles), E160R Q1 (red circles), G219C Q1 (teal circles), E160R/G219C Q1 (teal circles with red border). Black lines indicate tethers between subunits. (E) G-V and F-V plots of G219C Q + E1 (G-V: closed black circles and F-V: open black circles), G219C Q-E160R Q* + E1 (G-V: closed red squares and F-V: open red squares) and E160R/G219C Q*-Q + E1 (G-V: closed blue triangles) (n = 3-8).      119 4.4 Discussion 4.4.1 The E160R mutation in one, two or three Q1 subunits holds them in a resting state, but does not prevent IKs current. In order to investigate whether currents were inhibited if one, two, three, or four VSs are locked down, the E160R mutation was incorporated into the S2 domain of Q1. Previously, E160R homotetramers expressed in oocytes had been suggested to produce a channel complex in which the VSs were restrained, creating a very low open probability at the pore (Zaydman et al., 2014). This is due to a disruption of charge interactions that normally take place in wt channels, wherein the negative glutamic acid at residue 160 interacts with sequential positive charges in the S4, allowing the VS to move in between resting and activated conformations (Fig. 4.2A)(Zaydman et al., 2014). By incorporating this mutation into various Q1s of fusion constructs, channel conformations were created where either one, two, three, or four VSs were locked down (Q*QQQ, Q*Q, QQQ*Q*, EQ*QQ*Q*, and EQ* respectively). For channels expressed both in the absence (Appendix C, Fig. 1) and presence (Figs. 4.3 and 4.5) of E1, functional currents were visualized, except in the case of EQ* (Fig. 4.2). These results suggest that at least one VS is required to move within the IKs channel complex to permit detectable ion conduction at the whole cell level, but that neither Q1 alone, nor the IKs complex, require a concerted transition of all VSs to occur in order for the pore to open.  Several lines of evidence from our data support the notion that the E160R mutation does indeed restrain VSs from moving. By incorporating E160R into all four subunits, a non-functional channel was produced (Fig. 4.2). This lack of current was not due to a trafficking problem (Wu et al., 2010), instead, it is likely due to the very low open probability of the  120 channel pore when all VSs are in a resting conformation. Additionally, single channel recordings show that with each additional E160R mutation in the IKs channel complex, there is a decreased conductance due to a destabilization of the open pore when not all of the VSs can activate (Figs. 4.5 and 4.6). Similar results were seen from F57W (Appendix C, Fig. 9) although there are subtle differences at the single channel level. In recordings from the E160R mutant, longer inter-burst closings are observed, as well as small subconductance events that never lead to larger openings (Appendix C, Fig. 9E, e.g. second sweep on left).  As the F57W mutation does not immobilize the VS via electrostatic repulsion like E160R, but favours the closed conformation as a result of steric considerations, the result is a channel that is more likely to continue to burst open once the pore begins to conduct, and one that can more frequently visit higher conducting states (Appendix C, Fig. 9E, right). Of course, when there are four copies of the mutation in the complex the differences become even more obvious. At this point E160R currents are undetectable (Figs. 4.2D and 4.2F) while those of F57W can be observed with sufficient driving force (Appendix C, Fig. 9). MTSET modification was used to track VS movement directly in mammalian cells (Fig. 4.7). VSs with a G229C probe in the opposing Q1 subunit from E160R (G229C Q-Q* + E1) were successfully modified by MTSET, demonstrating their freedom to move into the external space. When the E160R mutation and G229C probe were in the same subunit (G229C Q*-Q + E1), modification was no longer observed (Fig. 4.7B), as would be expected from a restrained VS. Additionally, the shape of the activating currents of G229C Q-Q* + E1 was dramatically different from G229C Q*-Q + E1, with the former activating at an increased rate (Appendix C, Figs. 10A-B). This is likely due to the G229C mutation  121 altering the kinetics of the channel as it activates. As the VSs with the G229C mutation in G229C Q*-Q + E1 are restrained from moving, there is no effect of G229C on the activation of the channel.  VCF experiments confirmed the MTSET results (Fig. 4.8). Channels with G219C in either the opposite subunit (G219C Q-Q*) or the same subunit (G219C Q*-Q) as E160R were expressed with E1 in oocytes. Voltage-dependent VS fluorescence was detected in G219C Q-Q* + E1 (Fig. 4.8C) but not in G219C Q*-Q + E1 (Fig. 4.8D), indicating that the E160R mutation does restrain the VS. Additionally, the G219C mutation seems to increase the rate of deactivation when it is free to move (G219C Q + E1, G219C Q-Q* + E1 and G219C Q-Q + E1; Figs. 4.8B, 4.8C and Appendix C, Fig. 11A). However, when E160R is in the same subunit, the deactivation kinetics at -40 mV appear to be slower (Fig. 4.8D), indicating that G219C is unable to alter channel kinetics due to those VSs being locked in a resting conformation.  It was important to establish that E160R subunits incorporated normally when expressed as linked constructs with wt subunits. Using TEA+, we have shown that E160R mutated subunits do indeed assemble within the channel complex as expected (Appendix C, Fig. 4). As well, the order of the presence of E160R subunits within the tetramers makes no difference to the kinetic properties of expressed constructs (Appendix C, Fig. 3), which again supports the belief that these linked subunits assemble as expected. We also note potential limitations associated with MTSET and VCF recordings. It is possible that the presence of the E160R mutation could prevent G229C modification or G219C labeling by modifying solvent accessibility to these cysteines. Additionally, the E160R mutation might  122 alter the fluorophore environment, so that fluorescence changes upon S4 translation are no longer visible. 4.4.2 Previous studies and contrasting conclusions. When Q1 is expressed without E1, the voltage-dependence of VS movement, measured using fluorescence, overlaps that of channel conductance, indicating that movement of individual VSs can lead to channel pore opening (Osteen et al., 2010) and current flow before all VSs have moved, obviating the need for a concerted transition of all activated subunits for this to occur. Furthermore, Q1 has a constitutively active current even at very low voltages (Ma et al., 2011), which was suggested to represent transition to an open state that is independent of any VS movement. In 2013, Zaydman et al. used a mutation to lock the Q1 pore open. It was noted that although the pore remained locked open, the VSs were able to move to a resting conformation (Zaydman et al., 2013). A similar phenomenon was also observed in HCN channels (Ryu and Yellen, 2012). Altogether, these results suggest that Q1 pore opening and VS activation are not rigidly coupled via a concerted gating step(s), but instead are rather loosely coupled during channel activation and deactivation, as represented by the model in Figure 4.1B. In the presence of E1, however, there are contrasting reports. Barro-Soria et al. showed in 2014, again using fluorescence, that E1 divides VS movement into two steps (Barro-Soria et al., 2014). They suggested that there are independent VS movements of four Q1 subunits, followed by a second concerted conformational step involving all four VSs that allows the pore of the channel to open. This process where four VSs are required to move in order to allow the pore to conduct is exemplified in an IKs model set forth by Silva and Rudy in 2005 (Silva and Rudy, 2005), and its equivalent is presented in simplified form in  123 Figure 4.1A. Alternatively, Meisel et al. used thermodynamic mutant cycle analysis to investigate the effect of channel gating in the presence of a loss-of-function mutation located at the bottom of the S4 domain, R243W (Meisel et al., 2012). When this mutation was incorporated into one, two, three, or all four Q1 subunits, there was a progressive displacement of channel opening to more depolarized potentials in the presence of E1. They concluded that the linear correlation between the depolarizing shift of the V1/2  and the number of mutated subunits indicated independent VS movement, and that channel opening did not require a concerted activation step (Meisel et al., 2012).  The model of allosteric coupling of pore open probability to VS activation (Fig. 4.1B), rather than coupling through a concerted step, may also be supported by the presence of subconductance states visualized in IKs single channel recordings (Murray et al., 2016, Yang and Sigworth, 1998, Werry et al., 2013), as well as in the present experiments where different numbers of VS are restrained (Figs. 4.5 and 4.6), and in recent modeling studies (Ramasubramanian and Rudy, 2018) where individual VS movements are suggested to result in channel opening to lower conducting states.  4.4.3 Considering the two steps of fluorescence movement. As mentioned above, the first step of voltage-dependent fluorescence movement in IKs occurs at more negative potentials than channel conductance, while a second takes place at channel opening potentials and has been suggested to indicate the presence of a concerted final step to channel pore opening (Barro-Soria et al., 2014). However, our study has shown that pore opening and the second fluorescence step remain even when two VSs are restrained (Figs. 4.3 and 4.8), so it is relevant to consider what else the two fluorescence movements might represent. In the model suggested by Zaydman et al.  124 (2014), charge reversal mutations between the negative E160 in the S2 domain and the positive charges of the S4 domain were used to produce channel conformations where VSs were in either intermediate (E160R/R231E), or fully activated states (E160R/R237E)(Zaydman et al., 2014). For Q1 alone, the intermediate state promoted the opening of the pore, represented by the overlap between the G-V and the first component of the F-V curves. In the presence of E1, the pore was prevented from opening at this intermediate state. However, when the VS reached the fully activated state, E1 potentiated pore opening, represented as an overlap between the G-V and the second component of the F-V. The suggestion is then that the first component of the F-V may represent VS movement from a resting to an intermediate state, while the second component perhaps represents movement to an activated conformation. There is consensus in all these studies that the first step of fluorescence represents the major translocation of the VS, and this is supported by our experiments which demonstrate a shallower slope of this region of the F-V relation when two VSs contain the E160R mutation and are restrained (Fig. 4.8E). Although the second fluorescence step suggests that IKs channels require at least one VS to be in a fully activated conformation to allow the pore to conduct, it does not indicate that all four VSs are required to be activated simultaneously for pore openings to occur. Other contributions to a secondary fluorescence movement might include a secondary rearrangement of the channel complex at higher membrane potentials, or perhaps even displacement of E1 altering the fluorescence signal. It seems that additional experiments are required to fully understand this two-step fluorescence signal.   125 4.4.4 Loose coupling between VS activation and pore opening and an allosteric model. There seems little doubt from the data presented that up to three VS can be locked in resting conformations, while the channel may still open and produce currents that appear relatively normal in their activation sigmoidicities, and voltage-dependent kinetics. The depolarizing shifts in the isochronal activation V1/2s seem relatively mild, with little change when one VS is immobilized, and a depolarization of only ~30 mV when three of four are immobilized (Table 4.1). As noted by Meisel et al. (Meisel et al., 2012) who measured very low inter-subunit coupling free energies for multiple R243W mutations, these results suggest very loose VS coupling to overall closed and open channel conformations. In their data, inclusion of a loss-of-function mutation, R243W, located at the bottom of the S4 domain, into one, two or three subunits, caused a mild linear depolarizing shift in the V1/2 of activation with each additional mutation (Meisel et al., 2012). This linear shift was ascribed to independent VS movement within the channel complex.  Unlike the data from R243W, with either E160R or F57W there was no compelling relationship between the number of VSs locked down and the change in V1/2 (Fig. 4.3 and Appendix C, Fig. 9). With one VS locked down (Q*QQQ) in the presence or absence of E1, there was no statistically significant shift in the V1/2 of activation (Fig. 4.3 and Appendix C, Fig. 1). With two VSs locked down (Q*Q) in the presence or absence of E1, there was a depolarizing shift of approximately 16 mV, and a further 12 mV change between two and three VSs locked down (Q*Q + E1 vs. EQ*QQ*Q* + E1) (Fig. 4.3). Similarly, for F57W in E1, which is known to shift the IKs V1/2 of activation to greater than 100 mV when present in all four copies in the octameric complex (Chen and Goldstein, 2007), compared with  126 control, when one, two or four F57W mutations are present in the IKs channel complex, the G-V curve V1/2s were unchanged (34.8 mV), depolarized by ~8 mV (42.3 mV), or to 106.0 mV, respectively (Appendix C, Fig. 9A-D and Table 4.1).  We were prompted to construct a simple model of allosteric gating for IKs as shown in Figure 9 to understand the implications of the experimental observations of this study, in terms of non-linear changes in the V1/2 of activation (Fig. 4.3 and Appendix C, Figs. 1 and 9D), the relatively unchanged activation delays and overall activation time course (Fig. 4.4A), the speeding of tail currents seen with increasing numbers of immobilized VSs (Fig. 4.4B), and the decreased single channel conductance seen with increasing numbers of locked VSs (Fig. 4.5). The diagram in Figure 4.9A represents the expansion of the scheme in Fig. 4.1B when two VSs are wt (R or A), and two contain the E160R mutation (L). The scheme was modified accordingly to simulate one or three stabilized VSs. The upper layer of states represents closed channel conformations coupled to the VSs as they undergo voltage-dependent activation (rightwards across the layer). Channel conformations in the upper layer do not reflect VS conformations directly, but rather represent conflation of the multiple VS transitions loosely coupled into closed conformations from which the channel can open, independent of the VS (even locked) configuration. The coupling factors, ‘l’ and ‘b’ determine the strength of E160R VS coupling to these states. High values for ‘l’ and ‘b’ invoke tight coupling to locked VSs with little occupancy of states not coupled to activatable VSs, and result in simulations that show hyperpolarized G-Vs and very steep current activation profiles with fewer activatable subunits. Such simulations did not reflect experimental observations and thus confirm looser coupling between VS and open channel conformations (and lower values  127 of ‘l’ and ‘b’), which is also suggested by single channel data showing that full pore openings can occur even when two or three E160R subunits are present in the channel complex, albeit much less commonly than in wt channels (Fig. 4.5). Transitions to the lower layer of open states are constrained by rate constants ‘G0’, ‘D0’, and ‘D1’as defined in Figure 4.1B and legend. When the channel complex occupies progressively more activated conformations (to the right in Fig. 4.9A), this is assumed to allosterically increase the probability of pore opening by a factor ‘Dn’, vertically between closed and open states (Fig. 4.1B), in a similar manner to that suggested by others, in BK, and KCNQ channels (Horrigan et al., 1999, Horrigan and Aldrich, 1999, Osteen et al., 2012, Zaydman et al., 2014), reflecting the promotion of pore opening from more activated channel conformations.  This simple model was able to simulate the minimal change in V1/2 of activation with one E160R subunit, but did not simulate so well the change in V1/2 seen between two and three E160R subunits, nor the changes in tail current decay rates (Fig. 4.9B).  The model was improved by including VS cooperativity assuming that the forward and backwards rate constants between closed conformations were modified exponentially by increases in ‘l’ and ‘b’ for each VS that is held down. Results from these simulations are shown in Figure 4.9 as G-V relations (Fig. 4.9C), families of currents from -80 to +100 mV (Fig. 4.9D), and current overlays at their respective V1/2s of activation comparing modeling and experimental data (Fig. 4.9E-F). The major experimental findings at the whole cell level could be well reproduced by this relatively simple state model, and suggest that when one or more VSs are locked down, there are steric effects between the subunits that restrict activation. However, these kinetic effects are relatively mild for one locked  128 subunit when translated to channel opening, but increase dramatically for two to four E160R subunits.    Figure 4.9. A simple allosteric model with voltage sensor cooperativity most closely reproduces experimental results. (A) Markov model of IKs gating. Channels progress through various closed states, represented by Cn, in a voltage-dependent manner, and are able to reach an open state, represented by On, from each closed conformation. Resting VSs are represented by ‘R’, activated VSs are represented by ‘A’, and locked VSs are represented by ‘L’. Forward and reverse rate constants are ‘α’ and ‘β’ respectively, while ‘l’ and ‘b’ represent VS-closed state coupling factors. For clarity these are only shown on representative transitions. The rate constants ‘α’ and ‘β’ were formulated as exponential functions of voltage; α = 0.1*EXP(zVF/RT) s-1, and β = 0.022*EXP(-zVF/RT) s-1,  where V is potential and z, F, R, and T have their usual meanings, and z = 0.52. The vertical coupling constants between open and closed states are as shown in Fig. 4.1B. (B) G-V curves are shown for the simulated data from an allosteric model for IKs channels with no (black line), one (purple line), two (green line) and three (blue line) VSs locked down, and without  129 intersubunit cooperativity: ‘l’ = 1.3 s-1, ‘b’ = 4 s-1. (C) G-V curves are shown for the simulated data from an allosteric model with VS cooperativity for channels with no (black line, V1/2  = 32.1 mV, k = 20.6), one (purple line, V1/2  = 34.9 mV, k = 20.9), two (green line, V1/2  = 46.9 mV, k = 20.4), three (blue line, V1/2  = 61.3 mV, k = 17.9) and four (red line) VSs locked down. G-V curves were normalized to themselves, except for four VSs locked down, which was normalized to no VSs locked down. (D) Simulated IKs currents between -80 and +100 mV are shown for an allosteric model with VS cooperativity (see       4.2.11 Materials and Methods) for channels with no (black traces), one (purple traces), two (green traces) and three (blue traces) VSs locked down: ‘l’ = 1.1 s-1, ‘b’ = 1.5 s-1. The protocol was set from -80 to +100 mV for 4 s, with a 2 s tail current at -40 mV. Simulated (E) and experimental whole cell data (F) normalized traces are shown for no (black), one (purple), two (green) and three (blue) VSs locked down at their respective V1/2s.  Subconductance levels associated with transitions that occur between closed to fully open states have been studied in detail in Kv2.1 (Chapman et al., 1997). Partially activated channels are able to visit subconducting states, so that movement of a single VS to an activated conformation is sufficient for the channel to conduct current, with each additional VS movement increasing conductance. Furthermore, it was noted that when VSs are all in the same ‘homomeric’ position (four resting or four activated), they are in their most stable conformation (Chapman et al., 1997). Whereas, when one VS activates, a ‘heteromeric’ conformation is achieved that results in much less stable current sublevels (Chapman et al., 1997, Chapman and VanDongen, 2005), implying that there are stabilizing interactions that occur between the four subunits (Chapman et al., 1997). These findings provide a conceptual framework for the present experiments. As more VSs are locked down by the E160R mutation, currents are still visible, but conductance is greatly reduced and open state stability is reduced (Figs. 4.3 and 4.5). The depolarizing shift in V1/2 (Fig. 4.3), the increased rate of deactivation (Fig. 4.4B) and the increased flickering phenotype in single channel recordings (Fig. 4.5) with each additional E160R  130 mutation all suggest that although VSs move independently to allow channels to conduct, they do still maintain stabilizing cooperative inter-subunit interactions. The existence of VS cooperativity is further supported by the presence of a shallower slope in the fluorescence movement when two VSs are restrained by the E160R mutation (Fig. 4.8E and Appendix C, Fig. 11B), seen especially at negative potentials. Such a mechanism where there are cooperative interactions between subunits, yet no concerted step to opening has also been previously shown in KcsA channels (Blunck et al., 2008). While the simulations shown in Figure 4.9 did not incorporate changes in pore conductance associated with stabilized VSs, those experimental results (Fig. 4.5) may be simulated by assigning different pore conductances to open states associated with different numbers of activated subunits. 4.4.5 Functional implications of an allosteric model of activation gating. Allosteric gating of IKs potentially allows for a channel that can be highly regulated at different stages in the activation process by a variety of different factors. This is exemplified by the actions of a cAMP analogue on IKs, which can be localized to the VS, where the V1/2 is hyperpolarized, current density is increased, and Po is increased (Thompson et al., 2017, Terrenoire et al., 2005). As well, given the flexible stoichiometry of IKs, where up to four E1 subunits can complex with the Q1 channel tetramer (Murray et al., 2016, Nakajo et al., 2010), loose coupling between VSs and pore opening, and the lack of a requirement for a concerted step during opening, suggests that the effect that E1 has on a single VS activation (Barro-Soria et al., 2014) will not constrain the behavior of the other three VSs, and the subsequent nature of pore conductance. This is of further interest as Q1 can be regulated by a variety of different accessory subunits (KCNE1-5)  131 (Liin et al., 2015) which all have different actions on the VSs. For example, in contrast to E1, KCNE3 promotes VS activation and produces channels with a very high open probability at more hyperpolarized potentials (Barro-Soria et al., 2015). As several of the accessory subunits have been co-localized within the same tissues (Bendahhou et al., 2005), it is possible that there could be a mixed population of KCNE subunits interacting with the same Q1 tetramer, with each KCNE exerting its own effects on individual VSs. 4.4.6 Conclusions By studying the electrophysiology of IKs complexes where one, two, three or four VSs were restrained, we have shown that not all four VSs are required to move and act in a concerted manner in order to allow the Q1 pore to conduct current in the presence and absence of E1. Instead, Q1 and IKs channels gate allosterically, while individual and loosely coupled VS movements allow for highly flexible and regulated opening of the pore. Additionally, we have shown that although the channels gate in an allosteric manner, there may be cooperative interactions between the subunits themselves that occur during activation gating.    132 Chapter 5: Discussion 5.1  Summary of findings The stoichiometry of the IKs channel has been a controversial topic, with groups proposing either a fixed (2 KCNE1:4 KCNQ1) (Chen et al., 2003a, Wang and Goldstein, 1995, Plant et al., 2014) or a variable ratio (1-4 KCNE1:4 KCNQ1)(Cui et al., 1994, Wang et al., 1998, Nakajo et al., 2010) of the alpha and beta subunits.  In Chapter 2, whole cell and single channel patch clamp recordings of fusion constructs with different ratios of KCNE1:KCNQ1 (1:4, 2:4 and 4:4) show that more than two KCNE1 subunits can assemble with and modify the channel complex. Additionally, introduction of an unnatural amino acid (UAA) into KCNE1 shows that when two clefts are occupied with a tethered wild type KCNE1, a free KCNE1 is able to associate with the unoccupied clefts and crosslink to the channel. From this, it can be concluded that the IKs channel has a variable stoichiometry, where the number of KCNE1 subunits occupying the channel cleft is dependent upon its concentration.  In Chapter 3, UAA photo-crosslinking was further implemented in order to investigate the different interactions between two adjacent residues in KCNE1, F56 and F57 and KCNQ1. Our experiments illustrate that F56Bpa interacts with KCNQ1 in the open state, whereas F57Bpa interacts in the closed state. In addition, the rate of crosslinking of F57Bpa changes when channels are in pre-open closed states, indicating that KCNE1 moves within the cleft before the channel opens. Finally, crosslinking is still visualized with both residues when VSs are locked in a fully activated conformation (E1R/R4E KCNQ1), suggesting that these two residues are crosslinking to the pore of KCNQ1.    133 Finally, in Chapter 4, experiments were done to examine whether the four VSs move in a concerted step (Barro-Soria et al., 2014) or individually (Meisel et al., 2012, Ramasubramanian and Rudy, 2018) in order to allow the pore to open. We introduced a positively charged arginine into the top of the S2 of KCNQ1, E160R, in order to repel the positive charges in the S4, and keep the VS locked in a resting conformation. When four VSs are locked with E160R, no current is visualized at the whole cell level. With one, two and three locked down, the channel is able to produce current, however with a reduced conductance and a more unstable pore configuration. MTSET and fluorescence experiments were used to confirm that E160R does keep VSs in a resting conformation. These results suggest that the IKs channel does not undergo a concerted step of all four VSs in order to allow the channel to conduct current. Instead, the channel pore can open from any state, however the probability of pore opening increases as more VSs move to an activated conformation.   5.2 Pharmacological relevance of variable stoichiometry As loss-of-function mutations in KCNQ1 and KCNE1 lead to LQT syndrome types 1 and 5 respectively, it becomes of interest to investigate IKs channel activators. However, to date many of the known activators have different actions depending on the number of KCNE1 subunits in the channel complex. I have already described, in Chapter 2 and above, that in vitro IKs may show a variable stoichiometry with up to four KCNE1 associated with KCNQ1.   134 A natural activator of the channel, cyclic adenosine monophosphate (cAMP) is known to increase current density and hyperpolarize the V1/2 of activation of IKs due to activation of protein kinase A, which in turn phosphorylates KCNQ1 (Kurokawa et al., 2003, Terrenoire et al., 2005, Lopes et al., 2007, Thompson et al., 2017). When a cAMP analogue is applied to cells expressing KCNQ1 alone, there is no effect (Kurokawa et al., 2003, Haitin et al., 2009). Previously, our lab has shown that the effect of cAMP increases as more KCNE1 subunits are present (Thompson et al., 2018). For EQ, the V1/2 of activation shifts from approximately 27 mV before to 4 mV after exposure to the cAMP analogue (Thompson et al., 2018). When two KCNE1s are present, EQQ, the V1/2 shifts from approximately 10 mV to -3 mV, and for one KCNE1 present, EQQQQ, it goes from 6 mV to -3 mV (Thompson et al., 2018).  A similar effect is observed with mefenamic acid, a chloride channel blocker which is used as a nonsteroidal anti-inflammatory drug (Busch et al., 1994, Busch et al., 1997). Addition of mefenamic acid to cells expressing KCNQ1 alone does not affect the channel kinetics (Busch et al., 1997, Toyoda et al., 2006). Contrarily, IKs current becomes stabilized in the open state conformation in the presence of the drug; with an increase in current density and delayed tail current deactivation (Busch et al., 1994, Busch et al., 1997, Toyoda et al., 2006). Native IKs current in the guinea pig ventricular cardiomyocyte also shows a larger current density and slower deactivation in the presence of mefenamic acid (Toyoda et al., 2006). ML277 is a small compound which acts to preferentially activate KCNQ1, rather than the IKs current (Yu et al., 2013). Addition of ML277 to cells expressing KCNQ1 alone, causes a dramatic increase in current density, hyperpolarizing shift in the V1/2 of activation and  135 slowed deactivation kinetics (Yu et al., 2013). When KCNE1 is co-expressed at saturating levels, ML277 does not alter activation kinetics, but may slightly increase the current density (Yu et al., 2013, Xu et al., 2015). However, when a lower amount of KCNE1 is expressed alongside KCNQ1, the ML277 effect progressively increases, suggesting that the number of β-subunits occupying the cleft of KCNQ1 changes the potency of the drug (Yu et al., 2013).  As discussed above, the role of these activators all depend on either the presence or absence of KCNE1. Therefore, it is important to understand the physiological stoichiometry of IKs in order to design a channel activator which can effectively enhance IKs activity in a loss-of-function situation such as LQT syndrome.    5.3 Stoichiometry of IKs in vivo In Chapter 2, it is shown that a variable stoichiometry, where up to four KCNE1 subunits bind to the channel complex, is possible. However, it remains unknown as to what the physiological stoichiometry of KCNE1 and KCNQ1 is in vivo.  At 35°C, the V1/2 of IKs in ventricular cardiomyocytes of the Japanese White rabbit is approximately -1 mV, and for the guinea pig it is approximately 18 mV (Lu et al., 2001). At this temperature, the V1/2 of transfected IKs in a mammalian cell line is hyperpolarized by 10 mV to ~18 mV (Eldstrom et al., 2010). This suggests that guinea pig IKs may be predominantly fully saturated, while rabbit IKs may mainly exist in a 2:4 stoichiometry of KCNE1:KCNQ1.   136 In cardiomyocytes derived from human embryonic stem cells (hESC-CM), the V1/2 of activation of IKs is ~9 mV at room temperature (Wang et al., 2011a), comparable with the V1/2 of EQQ (11.6 mV; see Chapter 2), this would suggest that the these cells may have a native 2:4 ratio of KCNE1:KCNQ1. Interestingly, when the hESC-CMs were transfected with additional KCNE1, the V1/2 shifted to ~34 mV, similar to EQ (31.5 mV; see Chapter 2). As levels of KCNE1 are thought to increase in cardiomyopathic tissue, this would likely mean that IKs channels in the failing heart are in a fully saturated 4:4 stoichiometry, and might contribute to the observed prolonged action potential duration due to the depolarizing shift in activation V1/2 (Lundquist et al., 2005, Watanabe et al., 2007, Wang et al., 2011a). Since the ML277 activator becomes less potent with each additional KCNE1 subunit (see Section 5.2), it can be used as a tool to predict the stoichiometry of the channel complex. Action potentials produced by induced pluripotent stem cell (iPSC)-derived human cardiomyocytes show a shortening upon addition of ML277, suggesting that the majority of the channels do not exist in a fully saturated 4:4 stoichiometry (Yu et al., 2013). A similar phenomenon was observed in guinea pig and canine ventricular myocytes (Xu et al., 2015).   In the same way, the effect of cAMP can be used to probe the stoichiometry. As discussed in Section 5.2, the stoichiometry of IKs alters the range of shift in the V1/2 of activation induced by phosphorylation of KCNQ1 after addition of a cAMP analogue (Thompson et al., 2018). The largest shift in V1/2 was seen in a fully saturated channel, EQ, which went from ~27 mV before to ~4 mV after addition of cAMP, however if left longer it could go to ~-10 mV (Thompson et al., 2018). Interestingly, the V1/2 of activation of IKs in healthy  137 human ventricular cardiomyocytes, recorded in the presence of forskolin, which increases cAMP levels in the cell, was found to be approximately 5 mV (Virág et al., 2001). This correlates with the results seen with EQ, and thus it can be speculated that IKs may exist in its fully saturated form in human ventricular cardiomyocytes (Thompson et al., 2018).   5.4 Interaction with several KCNE β-subunits within the same complex KCNQ1 is able to associate with several KCNE β-subunits (KCNE1-5), which all have unique effects on channel kinetics (Bendahhou et al., 2005)(See Section 1.2). Additionally, mRNA from all five KCNE β-subunits has been localized to the heart with varying levels of expression (Radicke et al., 2006, Lundquist et al., 2006). Due to the variable stoichiometry of IKs channels, it is very possible that clefts unoccupied by KCNE1 can be filled by other subunits. In addition, as discussed in Chapter 4, we show that IKs and KCNQ1 alone both gate through an allosteric gating pathway, where individual VS movements lead to channel opening (Meisel et al., 2012, Ramasubramanian and Rudy, 2018). A variable stoichiometry and allosteric gating pathway, both allow for a greater level of flexibility in modulation by different KCNE subunits.  For example, in 2010, Nakajo et al. used a 2:4 stoichiometry fusion construct, EQQ, and co-expressed it with KCNE3. KCNE3 is still successfully able to alter channel kinetics of EQQ, and thus is able to bind within the open clefts of IKs (Nakajo et al., 2010). KCNE1 and KCNE3 both have different actions on the VS of KCNQ1. KCNE1 is thought to constrain VS movement (Shamgar et al., 2008) whereas KCNE3 promotes VS activation  138 (Barro-Soria et al., 2015). Thus, each individual VS can be modulated by its own neighboring KCNE subunit.   5.5 Allosteric gating of IKs KCNQ1 and IKs channel gating have been proposed to occur via two different gating pathways. KCNQ1 alone is thought to follow an allosteric gating pathway (Osteen et al., 2012), whereas IKs has been proposed to require a concerted movement of the channel complex prior to channel opening (Barro-Soria et al., 2014). In order to accommodate subconductance levels, our lab has previously developed a model of gating where all four VSs have to undergo an initial transition, followed by at least one VS reaching a fully activated conformation before the channel can conduct (Werry et al., 2013). As more VSs reach a fully activated conformation, the channel transitions through various subconducting levels (Werry et al., 2013). However, more recently, it has been proposed that IKs channels do not require movement of all four VSs, and rather opening occurs through a similar allosteric gating pathway to KCNQ1 (Meisel et al., 2012, Ramasubramanian and Rudy, 2018). In Chapter 4, we show that although one, two or three VSs are immobilized by E160R, currents are still produced, suggesting that a concerted step to opening is indeed not obligatory.  Simulations of IKs channel activation, generated by machine learning, are able to reproduce characteristic IKs traits observed experimentally, such as the two step movement in fluorescence and the presence of subconductance levels (Ramasubramanian and Rudy, 2018). Using these simulations, one, two, three, and four  139 VSs were immobilized and the results showed a progressive linear decline in current density, not consistent with a concerted model of gating (Ramasubramanian and Rudy, 2018). In our results, we too observe a reduction in current density at the whole cell and single channel levels with each additional E160R mutation (Figs. 4.3 and 4.5). Furthermore, Ramasubramanian and Rudy (2018) discuss that higher subconductance levels are only attainable when all four VSs move to an activated conformation. In our single channel subconductance analysis of the E160R mutation, we show that each additional E160R mutation indeed creates an environment where it becomes increasingly difficult to reach the higher subconductance levels. For example, wt channels spend approximately 9% of the time at 0.44 pA; however this is reduced to 1%, and 0.1% (when two or three subunits are locked down respectively, Fig. 4.6). It would be interesting to see what the simulated single channels look like when one, two or three VSs are restrained and whether we can correlate this behavior to that seen experimentally with E160R. However, these simulations have yet to be published.  Due to the allosteric nature of the gating, it is proposed that IKs channels have a loose, more flexible coupling between the VSs and the pore than previously thought (Ramasubramanian and Rudy, 2018, Meisel et al., 2012).  Such a flexible model of gating means that IKs activation can be modulated by a variety of different factors.   5.6  Cooperativity between KCNQ1 subunits In Chapter 4, cooperativity is discussed to describe potential interactions taking place between KCNQ1 subunits to stabilize the channel. Support for cooperativity comes from  140 the increase in the rate of deactivation with each additional locked VS, suggesting that the steric hindrance of the neighboring locked VS is acting to pull the freely moving VS back into a resting conformation (Fig. 4.4). An unstable open state is also represented by the flickering behavior and longer inter-burst closings of the single channel recordings (Fig. 4.5). Additionally, there is a shallower slope in the F-V when two VSs are locked down (Fig. 4.8).  Cooperativity has previously been investigated in other ion channels. As discussed earlier (See Section 4.4.4), Chapman et al. suggested that subconductance states of Kv2.1 represented movements of individual VSs to unstable ‘heteromeric’ conformations (Chapman et al., 1997, Chapman and VanDongen, 2005). A stable conformation was only maintained when all four VSs were in the same position, suggesting that there are interactions taking place between subunits (Chapman et al., 1997, Chapman and VanDongen, 2005).  Blunck et al. investigated the prokaryotic KcsA channel, which is simply made up of pore forming subunits (Blunck et al., 2008). Similar to what we propose with IKs, these KcsA channels have cooperative interactions that take place between subunits, however a conformational change of the channel complex is not required for the pore to open (Blunck et al., 2008). As with our modelling in Chapter 4, introduction of a coupling constant to the Kv2.1 and KcsA channel gating models to represent cooperativity, greatly improved their ability to reproduce experimental data (Chapman and VanDongen, 2005, Blunck et al., 2008).   141 With the goal of enhancing the IKs current in situations where for example loss-of-function mutations result in LQT syndrome, these stabilizing cooperative behaviors between subunits could potentially be interesting targets for drug development.   5.7 Limitations  One limitation of these experiments is that we use non-naturally occurring tethered fusion constructs to test our hypotheses. Initially, we used a shorter 31 aa linker between KCNE1 and KCNQ1, which at the whole cell level produced normal currents. However, at the single channel level, we found inconsistencies between our recordings. After extending the linker between KCNE1 and KCNQ1 to 52 aa, the inconsistencies between the single channel recordings disappeared, and the currents observed were very similar to those when KCNE1 and KCNQ1 were co-expressed separately. This finding suggests that our tandem constructs with longer linkers are as close to a natural conformation as possible.  Additionally, a limitation with our photo-crosslinking experiments is that there is a UV-induced wild type rundown of IKs currents. As discussed in Section 2.2.4, we had thought that this could be due UV irradiation forming oxidative radicals, however addition of free radical scavengers did not change the rundown observed. Another hypothesis is that UV irradiation could either destroy PIP2 or disrupt binding of PIP2 to the channel complex, making our rundown perhaps comparable to that observed when PIP2 is diffused out of the cell (Loussouarn et al., 2003). Additional experiments are clearly required to fully understand this. Although wild type rundown does occur within these channel complexes,  142 we were able to easily distinguish crosslinking with F56Bpa and F57Bpa from UV-induced wild type rundown (See Methods 2.2.6 and 3.2.6).   5.8 Future directions As our experiments suggest that the second component of the F-V does not represent a concerted movement of all four VSs, it would be interesting to investigate what it does describe. In section 4.4.3, we propose that the change in fluorescence could be due to movement of KCNE1. As shown by our experiments in Chapter 3 and by other groups (Wang et al., 2012, Xu et al., 2008), KCNE1 is thought to move and perhaps rotate within the exterior cleft during the activation process. This movement could potentially be mapped alongside pore opening by incorporating a fluorophore into the N-terminus region of KCNE1, and obtaining F-V and G-V curves.  Another region of interest for future experiments would be to investigate what happens when VSs are locked in different stages of activation. As described by Zaydman et al. (2014), E1R/R2E locks VSs in an intermediate state. In KCNQ1 alone, E1R/R2E conducts current, but in the presence of KCNE1 pore opening is prevented (Zaydman et al., 2014). It would be interesting to compare the current at the whole cell and single channel when one, two, three or four VSs are in an intermediate conformation in the presence and absence of KCNE1. This would allow us to understand why IKs channels with four VSs in an intermediate conformation do not conduct current. Additionally, several combinations between VSs in resting, intermediate and active can be investigated. Such studies may  143 also help shed a light on the potentially cooperative interactions that take place between the KCNQ1 subunits.   5.9  Conclusions The main aim of the research presented in this thesis was to gain a better understanding of the interactions that occur between KCNQ1 and an accessory β-subunit, KCNE1. KCNE1 is known to drastically alter the kinetics of KCNQ1, producing the IKs current, which plays an important role in repolarizing phase of the cardiac action potential. Several mutations in either subunit have led to cardiac arrhythmia syndromes.  Our experiments confirm that the IKs channel complex exists in a variable stoichiometry, where the amount of KCNE1 associated with KCNQ1 depends on the concentration of KCNE1 available. Additionally, our results suggest that KCNE1 moves within the channel cleft prior to opening, as well as two adjacent residues in KCNE1 interacting with KCNQ1 in either an open or closed state. Furthermore, we provide evidence for KCNE1 interacting with the pore of KCNQ1. Finally, we also show that IKs channels do not require a concerted movement of all four VSs to an activated conformation before the channel passes current. Instead, we confirm that gating of the channel occurs via an allosteric mechanism, where the pore is loosely coupled to the voltage sensors and can open before any VSs have moved. The probability of the pore opening, however, progressively increases with the translocation of an additional VS. Our results also suggest, that although there is no concerted step, there is cooperativity between VSs.  144 The results of this thesis provide new information about the interplay between KCNE1 and KCNQ1. This knowledge will be of importance in understanding the electrical activity of the heart, as well as designing novel pharmaceuticals which aim to maintain a normal heart rhythm.     145 Bibliography ABBOTT, G. W. 2014. Biology of the KCNQ1 Potassium Channel. New Journal of Science, 2014, 1-26. ABBOTT, G. W. 2016. KCNE4 and KCNE5: K(+) channel regulation and cardiac arrhythmogenesis. Gene, 593, 249-60. ABBOTT, G. W., SESTI, F., SPLAWSKI, I., BUCK, M. E., LEHMANN, M. H., TIMOTHY, K. W., KEATING, M. T. & GOLDSTEIN, S. A. 1999. MiRP1 forms IKr potassium channels with HERG and is associated with cardiac arrhythmia. Cell, 97, 175-87. ACKERMAN, M. J. 1998. The long QT syndrome: ion channel diseases of the heart. Mayo Clin Proc, 73, 250-69. ANGELO, K., JESPERSEN, T., GRUNNET, M., NIELSEN, M. S., KLAERKE, D. A. & OLESEN, S. P. 2002. KCNE5 induces time- and voltage-dependent modulation of the KCNQ1 current. Biophys J, 83, 1997-2006. ARRIGHI, I., BLOCH-FAURE, M., GRAHAMMER, F., BLEICH, M., WARTH, R., MENGUAL, R., DRICI, M. D., BARHANIN, J. & MENETON, P. 2001. Altered potassium balance and aldosterone secretion in a mouse model of human congenital long QT syndrome. Proc Natl Acad Sci U S A, 98, 8792-7. BARHANIN, J., LESAGE, F., GUILLEMARE, E., FINK, M., LAZDUNSKI, M. & ROMEY, G. 1996. K(V)LQT1 and lsK (minK) proteins associate to form the I(Ks) cardiac potassium current. Nature, 384, 78-80. BARRO-SORIA, R., PEREZ, M. E. & LARSSON, H. P. 2015. KCNE3 acts by promoting voltage sensor activation in KCNQ1. Proc Natl Acad Sci USA, 112, E7286-92. BARRO-SORIA, R., RAMENTOL, R., LIIN, S. I., PEREZ, M. E., KASS, R. S. & LARSSON, H. P. 2017. KCNE1 and KCNE3 modulate KCNQ1 channels by affecting different gating transitions. Proc Natl Acad Sci U S A, 114, E7367-E7376. BARRO-SORIA, R., REBOLLEDO, S., LIIN, S. I., PEREZ, M. E., SAMPSON, K. J., KASS, R. S. & LARSSON, H. P. 2014. KCNE1 divides the voltage sensor movement in KCNQ1/KCNE1 channels into two steps. Nat Commun, 5, 3750-3750. BEER, M., SEYFARTH, T., SANDSTEDE, J., LANDSCHÜTZ, W., LIPKE, C., KÖSTLER, H., VON KIENLIN, M., HARRE, K., HAHN, D. & NEUBAUER, S. 2002. Absolute concentrations of high-energy phosphate metabolites in normal, hypertrophied, and failing human myocardium measured noninvasively with (31)P-SLOOP magnetic resonance spectroscopy. J Am Coll Cardiol, 40, 1267-74. BELLOCQ, C., VAN GINNEKEN, A. C. G., BEZZINA, C. R., ALDERS, M., ESCANDE, D., MANNENS, M. M. A. M., BARÓ, I. & WILDE, A. A. M. 2004. Mutation in the KCNQ1 gene leading to the short QT-interval syndrome. Circulation, 109, 2394-2397. BENDAHHOU, S., MARIONNEAU, C., HAUROGNE, K., LARROQUE, M. M., DERAND, R., SZUTS, V., ESCANDE, D., DEMOLOMBE, S. & BARHANIN, J. 2005. In vitro molecular interactions and distribution of KCNE family with KCNQ1 in the human heart. Cardiovasc Res, 67, 529-38. BEUCKELMANN, D. J., NÄBAUER, M. & ERDMANN, E. 1993. Alterations of K+ currents in isolated human ventricular myocytes from patients with terminal heart failure. Circ Res, 73, 379-85.  146 BEZANILLA, F., CAPUTO, C., DIPOLO, R. & ROJAS, H. 1986. Potassium conductance of the squid giant axon is modulated by ATP. Proc Natl Acad Sci U S A, 83, 2743-5. BEZANILLA, F., PEROZO, E. & STEFANI, E. 1994. Gating of Shaker K+ channels: II. The components of gating currents and a model of channel activation. Biophys J, 66, 1011-21. BIAN, J., CUI, J. & MCDONALD, T. V. 2001. HERG K(+) channel activity is regulated by changes in phosphatidyl inositol 4,5-bisphosphate. Circ Res, 89, 1168-76. BIANCHI, L., SHEN, Z., DENNIS, A. T., PRIORI, S. G., NAPOLITANO, C., RONCHETTI, E., BRYSKIN, R., SCHWARTZ, P. J. & BROWN, A. M. 1999. Cellular dysfunction of LQT5-minK mutants: Abnormalities of I(Ks), I(Kr) and trafficking in long QT syndrome. Human Molecular Genetics, 8, 1499-1507. BLUNCK, R., MCGUIRE, H., HYDE, H. C. & BEZANILLA, F. 2008. Fluorescence detection of the movement of single KcsA subunits reveals cooperativity. Proc Natl Acad Sci USA, 105, 20263-8. BUSCH, A. E., BUSCH, G. L., FORD, E., SUESSBRICH, H., LANG, H. J., GREGER, R., KUNZELMANN, K., ATTALI, B. & STÜHMER, W. 1997. The role of the IsK protein in the specific pharmacological properties of the IKs channel complex. Br J Pharmacol, 122, 187-9. BUSCH, A. E., HERZER, T., WAGNER, C. A., SCHMIDT, F., RABER, G., WALDEGGER, S. & LANG, F. 1994. Positive regulation by chloride channel blockers of IsK channels expressed in Xenopus oocytes. Mol Pharmacol, 46, 750-3. CHA, A. & BEZANILLA, F. 1997. Characterizing voltage-dependent conformational changes in the Shaker K+ channel with fluorescence. Neuron, 19, 1127-40. CHAN, P. J., OSTEEN, J. D., XIONG, D., BOHNEN, M. S., DOSHI, D., SAMPSON, K. J., MARX, S. O., KARLIN, A. & KASS, R. S. 2012. Characterization of KCNQ1 atrial fibrillation mutations reveals distinct dependence on KCNE1. J Gen Physiol, 139, 135-144. CHANDRASEKHAR, K. D., BAS, T. & KOBERTZ, W. R. 2006. KCNE1 subunits require co-assembly with K+ channels for efficient trafficking and cell surface expression. J Biol Chem, 281, 40015-23. CHANDRASEKHAR, K. D., LVOV, A., TERRENOIRE, C., GAO, G. Y., KASS, R. S. & KOBERTZ, W. R. 2011. O-glycosylation of the cardiac I(Ks) complex. J Physiol, 589, 3721-30. CHAPMAN, M. L. & VANDONGEN, A. M. 2005. K channel subconductance levels result from heteromeric pore conformations. J Gen Physiol, 126, 87-103. CHAPMAN, M. L., VANDONGEN, H. M. & VANDONGEN, A. M. 1997. Activation-dependent subconductance levels in the drk1 K channel suggest a subunit basis for ion permeation and gating. Biophys J, 72, 708-19. CHEN, H. & GOLDSTEIN, S. A. N. 2007. Serial perturbation of MinK in IKs implies an alpha-helical transmembrane span traversing the channel corpus. Biophys J, 93, 2332-2340. CHEN, H., KIM, L. A., RAJAN, S., XU, S. & GOLDSTEIN, S. A. 2003a. Charybdotoxin binding in the I(Ks) pore demonstrates two MinK subunits in each channel complex. Neuron, 40, 15-23.  147 CHEN, Y., WANG, X., WANG, Y., XU, W., JIN, H., SUN, H., SU, X., ZHUANG, Q., YANG, Y., LI, Y., LIU, Y., XU, H. & LI, X. 2003b. KCNQ1 Gain-of-Function Mutation in Familial Atrial Fibrillation. Science, 299, 251-254. CHENG, Y. M., AZER, J., NIVEN, C. M., MAFI, P., ALLARD, C. R., QI, J., THOUTA, S. & CLAYDON, T. W. 2011. Molecular determinants of U-type inactivation in Kv2.1 channels. Biophys J, 101, 651-61. CHIN, J. W. & SCHULTZ, P. G. 2002. In vivo photocrosslinking with unnatural amino acid mutagenesis. ChemBioChem, 3, 1135-1137. CHOUABE, C., NEYROUD, N., RICHARD, P., DENJOY, I., HAINQUE, B., ROMEY, G., DRICI, M. D., GUICHENEY, P. & BARHANIN, J. 2000. Novel mutations in KvLQT1 that affect I(ks) activation through interactions with Isk. Cardiovascular Research, 45, 971-980. CHUNG, D. Y., CHAN, P. J., BANKSTON, J. R., YANG, L., LIU, G., MARX, S. O., KARLIN, A. & KASS, R. S. 2009. Location of KCNE1 relative to KCNQ1 in the I(KS) potassium channel by disulfide cross-linking of substituted cysteines. Proc Natl Acad Sci USA, 106, 743-8. COLE, K. S. & MOORE, J. W. 1960. Potassium ion current in the squid giant axon: dynamic characteristic. Biophysical journal, 1, 1-14. COLQUHOUN, D. & HAWKES, A. G. 1995. A Q-matrix cookbook: how to write only one program to calculate the single-channel and macroscopic predictions for any kinetic mechanism. In: SAKMANN, B. & NEHER, E. (eds.) Single-Channel Recording. 2nd ed.: Springer. CUELLO, L. G., JOGINI, V., CORTES, D. M. & PEROZO, E. 2010. Structural mechanism of C-type inactivation in K(+) channels. Nature, 466, 203-8. CUI, J. 2016. Voltage-Dependent Gating: Novel Insights from KCNQ1 Channels. Biophys J, 110, 14-25. CUI, J., KLINE, R. P., PENNEFATHER, P. & COHEN, I. S. 1994. Gating of IsK expressed in Xenopus oocytes depends on the amount of mRNA injected. J Gen Physiol, 104, 87-105. DELPÓN, E., CORDEIRO, J. M., NÚÑEZ, L., THOMSEN, P. E., GUERCHICOFF, A., POLLEVICK, G. D., WU, Y., KANTERS, J. K., LARSEN, C. T., HOFMAN-BANG, J., BURASHNIKOV, E., CHRISTIANSEN, M. & ANTZELEVITCH, C. 2008. Functional effects of KCNE3 mutation and its role in the development of Brugada syndrome. Circ Arrhythm Electrophysiol, 1, 209-18. DORMÁN, G. & PRESTWICH, G. D. 1994. Benzophenone photophores in biochemistry. Biochemistry, 33, 5661-5673. DOYLE, D. A., MORAIS CABRAL, J., PFUETZNER, R. A., KUO, A., GULBIS, J. M., COHEN, S. L., CHAIT, B. T. & MACKINNON, R. 1998. The structure of the potassium channel: molecular basis of K+ conduction and selectivity. Science, 280, 69-77. DVIR, M., STRULOVICH, R., SACHYANI, D., BEN-TAL COHEN, I., HAITIN, Y., DESSAUER, C., PONGS, O., KASS, R., HIRSCH, J. A. & ATTALI, B. 2014. Long QT mutations disrupt IKS regulation by PKA and PIP2 at the same KCNQ1 helix C-KCNE1 interface. Journal of cell science, 127, 3943-3955.  148 ELDSTROM, J. & FEDIDA, D. 2011. The voltage-gated channel accessory protein KCNE2: multiple ion channel partners, multiple ways to long QT syndrome. Expert Rev Mol Med, 13, e38-e38. ELDSTROM, J., WANG, Z., WERRY, D., WONG, N. & FEDIDA, D. 2015. Microscopic mechanisms for long QT syndrome type 1 revealed by single-channel analysis of IKs with S3 domain mutations in KCNQ1. Heart Rhythm, 12, 386-394. ELDSTROM, J., XU, H., WERRY, D., KANG, C., LOEWEN, M. E., DEGENHARDT, A., SANATANI, S., TIBBITS, G. F., SANDERS, C. & FEDIDA, D. 2010. Mechanistic basis for LQT1 caused by S3 mutations in the KCNQ1 subunit of IKs. J Gen Physiol, 135, 433-448. ES-SALAH-LAMOUREUX, Z., FOUGERE, R., XIONG, P. Y., ROBERTSON, G. A. & FEDIDA, D. 2010. Fluorescence-tracking of activation gating in human ERG channels reveals rapid S4 movement and slow pore opening. PLoS One, 5, e10876. FARRELL, I. S., TORONEY, R., HAZEN, J. L., MEHL, R. A. & CHIN, J. W. 2005. Photo-cross-linking interacting proteins with a genetically encoded benzophenone. Nat Methods, 2, 377-84. FRANQUEZA, L., LIN, M., SPLAWSKI, I., KEATING, M. T. & SANGUINETTI, M. C. 1999. Long QT syndrome-associated mutations in the S4-S5 linker of KvLQT1 potassium channels modify gating and interaction with minK subunits. J Biol Chem, 274, 21063-21070. FRIEDMANN, I., FRASER, G. R. & FROGGATT, P. 1966. Pathology of the ear in the cardioauditory syndrome of Jervell and Lange-Nielsen (recessive deafness with electrocardiographic abnormalities). J Laryngol Otol, 80, 451-70. FRÖHLICH, H., BOINI, K. M., SEEBOHM, G., STRUTZ-SEEBOHM, N., URECHE, O. N., FÖLLER, M., EICHENMÜLLER, M., SHUMILINA, E., PATHARE, G., SINGH, A. K., SEIDLER, U., PFEIFER, K. E. & LANG, F. 2011. Hypothyroidism of gene-targeted mice lacking Kcnq1. Pflugers Arch, 461, 45-52. GEBAUER, M., ISBRANDT, D., SAUTER, K., CALLSEN, B., NOLTING, A., PONGS, O. & BÄHRING, R. 2004. N-type inactivation features of Kv4.2 channel gating. Biophys J, 86, 210-23. GRAHAMMER, F., WARTH, R., BARHANIN, J., BLEICH, M. & HUG, M. J. 2001. The small conductance K+ channel, KCNQ1: expression, function, and subunit composition in murine trachea. J Biol Chem, 276, 42268-75. GRANT, A. O. 2009. Cardiac ion channels. Circ Arrhythm Electrophysiol, 2, 185-94. GRUNNET, M., JESPERSEN, T., RASMUSSEN, H. B., LJUNGSTRØM, T., JORGENSEN, N. K., OLESEN, S. P. & KLAERKE, D. A. 2002. KCNE4 is an inhibitory subunit to the KCNQ1 channel. J Physiol, 542, 119-30. GRUNNET, M., OLESEN, S. P., KLAERKE, D. A. & JESPERSEN, T. 2005. hKCNE4 inhibits the hKCNQ1 potassium current without affecting the activation kinetics. Biochem Biophys Res Commun, 328, 1146-53. HAITIN, Y., WIENER, R., SHAHAM, D., PERETZ, A., COHEN, E. B.-T., SHAMGAR, L., PONGS, O., HIRSCH, J. A. & ATTALI, B. 2009. Intracellular domains interactions and gated motions of I(KS) potassium channel subunits. The EMBO journal, 28, 1994-2005.  149 HARMER, S. C., WILSON, A. J., ALDRIDGE, R. & TINKER, A. 2010. Mechanisms of disease pathogenesis in long QT syndrome type 5. Am J Physiol Cell Physiol, 298, C263-73. HEGINBOTHAM, L., LU, Z., ABRAMSON, T. & MACKINNON, R. 1994. Mutations in the K+ channel signature sequence. Biophys J, 66, 1061-7. HEKKALA, A. M., VIITASALO, M., VÄÄNÄNEN, H., SWAN, H. & TOIVONEN, L. 2010. Abnormal repolarization dynamics revealed in exercise test in long QT syndrome mutation carriers with normal resting QT interval. Europace, 12, 1296-301. HILGEMANN, D. W. 1997. Cytoplasmic ATP-dependent regulation of ion transporters and channels: mechanisms and messengers. Annu Rev Physiol, 59, 193-220. HILLE, B. 2001. Ion channels of excitable membranes, Sunderland, Mass., Sinauer. HINO, N., OKAZAKI, Y., KOBAYASHI, T., HAYASHI, A., SAKAMOTO, K. & YOKOYAMA, S. 2005. Protein photo-cross-linking in mammalian cells by site-specific incorporation of a photoreactive amino acid. Nat Methods, 2, 201-6. HODGKIN, A. L. & HUXLEY, A. F. 1952. A quantitative description of membrane current and its application to conduction and excitation in nerve. J Physiol, 117, 500-44. HORRIGAN, F. T. & ALDRICH, R. W. 1999. Allosteric voltage gating of potassium channels II. Mslo channel gating charge movement in the absence of Ca(2+). J Gen Physiol, 114, 305-36. HORRIGAN, F. T., CUI, J. & ALDRICH, R. W. 1999. Allosteric voltage gating of potassium channels I. Mslo ionic currents in the absence of Ca(2+). J Gen Physiol, 114, 277-304. HOSHI, T., ZAGOTTA, W. N. & ALDRICH, R. W. 1990. Biophysical and molecular mechanisms of Shaker potassium channel inactivation. Science, 250, 533-8. HOSHI, T., ZAGOTTA, W. N. & ALDRICH, R. W. 1991. Two types of inactivation in Shaker K+ channels: effects of alterations in the carboxy-terminal region. Neuron, 7, 547-56. JENSEN, M., JOGINI, V., BORHANI, D. W., LEFFLER, A. E., DROR, R. O. & SHAW, D. E. 2012. Mechanism of voltage gating in potassium channels. Science, 336, 229-33. JERVELL, A. & LANGE-NIELSEN, F. 1957. Congenital deaf-mutism, functional heart disease with prolongation of the Q-T interval, and sudden death. Am Heart J, 54, 59-68. JIANG, M., XU, X., WANG, Y., TOYODA, F., LIU, X. S., ZHANG, M., ROBINSON, R. B. & TSENG, G. N. 2009. Dynamic partnership between KCNQ1 and KCNE1 and influence on cardiac IKs current amplitude by KCNE2. J Biol Chem, 284, 16452-62. JIANG, M., ZHANG, M., TANG, D. G., CLEMO, H. F., LIU, J., HOLWITT, D., KASIRAJAN, V., POND, A. L., WETTWER, E. & TSENG, G. N. 2004. KCNE2 protein is expressed in ventricles of different species, and changes in its expression contribute to electrical remodeling in diseased hearts. Circulation, 109, 1783-8. JOST, N., VIRÁG, L., BITAY, M., TAKÁCS, J., LENGYEL, C., BILICZKI, P., NAGY, Z., BOGÁTS, G., LATHROP, D. A., PAPP, J. G. & VARRÓ, A. 2005. Restricting excessive cardiac action potential and QT prolongation: A vital role for IKs in human ventricular muscle. Circulation, 112, 1392-1399.  150 KANG, C., TIAN, C., SÖNNICHSEN, F. D., SMITH, J. A., MEILER, J., GEORGE, A. L., VANOYE, C. G., HAK, J. K. & SANDERS, C. R. 2008. Structure of KCNE1 and implications for how it modulates the KCNQ1 potassium channel. Biochemistry, 47, 7999-8006. KASIMOVA, M. A., ZAYDMAN, M. A., CUI, J. & TAREK, M. 2015. PIP₂-dependent coupling is prominent in Kv7.1 due to weakened interactions between S4-S5 and S6. Scientific Reports, 5, 1-9. KIM, R. Y., PLESS, S. A. & KURATA, H. T. 2017. PIP2 mediates functional coupling and pharmacology of neuronal KCNQ channels. Proc Natl Acad Sci U S A, 114, E9702-E9711. KLEMIC, K. G., KIRSCH, G. E. & JONES, S. W. 2001. U-type inactivation of Kv3.1 and Shaker potassium channels. Biophys J, 81, 814-26. KLIPPENSTEIN, V., GHISI, V., WIETSTRUK, M. & PLESTED, A. J. 2014. Photoinactivation of glutamate receptors by genetically encoded unnatural amino acids. J Neurosci, 34, 980-91. KRONCKE, B. M., VAN HORN, W. D., SMITH, J., KANG, C., WELCH, R. C., SONG, Y., NANNEMANN, D. P., TAYLOR, K. C., SISCO, N. J., GEORGE, A. L., MEILER, J., VANOYE, C. G. & SANDERS, C. R. 2016. Structural basis for KCNE3 modulation of potassium recycling in epithelia. Sci Adv, 2, e1501228. KRUMERMAN, A., GAO, X., BIAN, J. S., MELMAN, Y. F., KAGAN, A. & MCDONALD, T. V. 2004. An LQT mutant minK alters KvLQT1 trafficking. Am J Physiol Cell Physiol, 286, C1453-63. KUANG, Q., PURHONEN, P. & HEBERT, H. 2015. Structure of potassium channels. Cell Mol Life Sci, 72, 3677-93. KURATA, H. T., SOON, G. S., ELDSTROM, J. R., LU, G. W., STEELE, D. F. & FEDIDA, D. 2002. Amino-terminal determinants of U-type inactivation of voltage-gated K+ channels. J Biol Chem, 277, 29045-53. KUROKAWA, J., CHEN, L. & KASS, R. S. 2003. Requirement of subunit expression for cAMP-mediated regulation of a heart potassium channel. Proc Natl Acad Sci USA, 100, 2122-2127. KUROKAWA, J., MOTOIKE, H. K. & KASS, R. S. 2001. TEA(+)-sensitive KCNQ1 constructs reveal pore-independent access to KCNE1 in assembled I(Ks) channels. J Gen Physiol, 117, 43-52. LANG, F. & REHWALD, W. 1992. Potassium channels in renal epithelial transport regulation. Physiol Rev, 72, 1-32. LARSSON, H. P. 2002. The search is on for the voltage sensor-to-gate coupling. J Gen Physiol, 120, 475-81. LEE, M. P., RAVENEL, J. D., HU, R. J., LUSTIG, L. R., TOMASELLI, G., BERGER, R. D., BRANDENBURG, S. A., LITZI, T. J., BUNTON, T. E., LIMB, C., FRANCIS, H., GORELIKOW, M., GU, H., WASHINGTON, K., ARGANI, P., GOLDENRING, J. R., COFFEY, R. J. & FEINBERG, A. P. 2000. Targeted disruption of the Kvlqt1 gene causes deafness and gastric hyperplasia in mice. J Clin Invest, 106, 1447-55. LERCHE, C., SEEBOHM, G., WAGNER, C. I., SCHERER, C. R., DEHMELT, L., ABITBOL, I., GERLACH, U., BRENDEL, J., ATTALI, B. & BUSCH, A. E. 2000. Molecular impact of MinK on the enantiospecific block of I(Ks) by chromanols. Br J Pharmacol, 131, 1503-6.  151 LI, P., LIU, H., LAI, C., SUN, P., ZENG, W., WU, F., ZHANG, L., WANG, S., TIAN, C. & DING, J. 2014. Differential modulations of KCNQ1 by auxiliary proteins KCNE1 and KCNE2. Sci Rep, 4, 4973. LI, Y., GAO, J., LU, Z., MCFARLAND, K., SHI, J., BOCK, K., COHEN, I. S. & CUI, J. 2013. Intracellular ATP binding is required to activate the slowly activating K+ channel I(Ks). Proc Natl Acad Sci U S A, 110, 18922-7. LI, Y., ZAYDMAN, M. A., WU, D., SHI, J., GUAN, M., VIRGIN-DOWNEY, B. & CUI, J. 2011. KCNE1 enhances phosphatidylinositol 4,5-bisphosphate (PIP2) sensitivity of IKs to modulate channel activity. Proc Natl Acad Sci USA, 108, 9095-9100. LIIN, S. I., BARRO-SORIA, R. & LARSSON, H. P. 2015. The KCNQ1 channel - remarkable flexibility in gating allows for functional versatility. J Physiol, 593, 2605-2615. LIN, Y., HUANG, J., HE, S., FENG, R., ZHONG, Z., LIU, Y., YE, W., LI, X., LIAO, H., FEI, H., RAO, F., SHAN, Z., DENG, C., ZHAN, X., XUE, Y., LIU, H., ZHANG, B., WANG, K., ZHANG, Q., WU, S. & LIN, X. 2018. Case report of familial sudden cardiac death caused by a DSG2 p.F531C mutation as genetic background when carrying with heterozygous KCNE5 p.D92E/E93X mutation. BMC Med Genet, 19, 148. LIU, D. W. & ANTZELEVITCH, C. 1995. Characteristics of the delayed rectifier current (IKr and IKs) in canine ventricular epicardial, midmyocardial, and endocardial myocytes. A weaker IKs contributes to the longer action potential of the M cell. Circ Res, 76, 351-65. LOPES, C. M. B., REMON, J. I., MATAVEL, A., SUI, J. L., KESELMAN, I., MEDEI, E., SHEN, Y., ROSENHOUSE-DANTSKER, A. & LOGOTHETIS, D. E. 2007. Protein Kinase A Modulates PLC-Dependent Regulation and PIP2-Sensitivity of K+ Channels. Channels, 1, 124-134. LOUSSOUARN, G., PARK, K. H., BELLOCQ, C., BARÓ, I., CHARPENTIER, F. & ESCANDE, D. 2003. Phosphatidylinositol-4,5-bisphosphate, PIP2, controls KCNQ1/KCNE1 voltage-gated potassium channels: a functional homology between voltage-gated and inward rectifier K+ channels. EMBO J, 22, 5412-21. LU, Z., KAMIYA, K., OPTHOF, T., YASUI, K. & KODAMA, I. 2001. Density and kinetics of I(Kr) and I(Ks) in guinea pig and rabbit ventricular myocytes explain different efficacy of I(Ks) blockade at high heart rate in guinea pig and rabbit: implications for arrhythmogenesis in humans. Circulation, 104, 951-6. LUNDBY, A., RAVN, L. S., SVENDSEN, J. H., HAUNS, S., OLESEN, S. P. & SCHMITT, N. 2008. KCNE3 mutation V17M identified in a patient with lone atrial fibrillation. Cell Physiol Biochem, 21, 47-54. LUNDQUIST, A. L., MANDERFIELD, L. J., VANOYE, C. G., ROGERS, C. S., DONAHUE, B. S., CHANG, P. A., DRINKWATER, D. C., MURRAY, K. T. & GEORGE, A. L. 2005. Expression of multiple KCNE genes in human heart may enable variable modulation of I(Ks). J Mol Cell Cardiol, 38, 277-87. LUNDQUIST, A. L., TURNER, C. L., BALLESTER, L. Y. & GEORGE, A. L. 2006. Expression and transcriptional control of human KCNE genes. Genomics, 87, 119-28. LVOV, A., GAGE, S. D., BERRIOS, V. M. & KOBERTZ, W. R. 2010. Identification of a protein-protein interaction between KCNE1 and the activation gate machinery of KCNQ1. J Gen Physiol, 135, 607-618.  152 LÓPEZ-BARNEO, J., HOSHI, T., HEINEMANN, S. H. & ALDRICH, R. W. 1993. Effects of external cations and mutations in the pore region on C-type inactivation of Shaker potassium channels. Receptors Channels, 1, 61-71. MA, K. J., LI, N., TENG, S. Y., ZHANG, Y. H., SUN, Q., GU, D. F. & PU, J. L. 2007. Modulation of KCNQ1 current by atrial fibrillation-associated KCNE4 (145E/D) gene polymorphism. Chin Med J (Engl), 120, 150-4. MA, L. J., OHMERT, I. & VARDANYAN, V. 2011. Allosteric features of KCNQ1 gating revealed by alanine scanning mutagenesis. Biophys J, 100, 885-94. MADAMANCHI, A. 2007. Beta-adrenergic receptor signaling in cardiac function and heart failure. MJM, 10, 99-104. MANDERFIELD, L. J. & GEORGE, A. L. 2008. KCNE4 can co-associate with the I(Ks) (KCNQ1-KCNE1) channel complex. FEBS J, 275, 1336-49. MANNUZZU, L. M. & ISACOFF, E. Y. 2000. Independence and cooperativity in rearrangements of a potassium channel voltage sensor revealed by single subunit fluorescence. J Gen Physiol, 115, 257-68. MAO, T., MIAO, H. J., XU, G. J., GAN, T. Y., ZHOU, X. H., ZHANG, J., LI, F. P. & TANG, B. P. 2013. [Association of single nucleotide polymorphism of KCNE1 and KCNE4 gene with atrial fibrillation in Xinjiang Uygur and Han population]. Zhonghua Xin Xue Guan Bing Za Zhi, 41, 916-21. MARCUS, D. C. & SHEN, Z. 1994. Slowly activating voltage-dependent K+ conductance is apical pathway for K+ secretion in vestibular dark cells. Am J Physiol, 267, C857-64. MARX, S. O., KUROKAWA, J., REIKEN, S., MOTOIKE, H., ARMIENTO, J. D., MARKS, A. R. & KASS, R. S. 2002. Requirement of a Macromolecular Signaling Complex for B Adrenergic Receptor Modulation of the KCNQ1- KCNE1 Potassium Channel. Science, 295, 496-499. MEISEL, E., DVIR, M., HAITIN, Y., GILADI, M., PERETZ, A. & ATTALI, B. 2012. KCNQ1 channels do not undergo concerted but sequential gating transitions in both the absence and the presence of KCNE1 protein. J Biol Chem, 287, 34212-24. MELMAN, Y. F., DOMÈNECH, A., DE LA LUNA, S. & MCDONALD, T. V. 2001. Structural Determinants of KvLQT1 Control by the KCNE Family of Proteins. J Biol Chem, 276, 6439-6444. MELMAN, Y. F., UM, S. Y., KRUMERMAN, A., KAGAN, A. & MCDONALD, T. V. 2004. KCNE1 binds to the KCNQ1 pore to regulate potassium channel activity. Neuron, 42, 927-937. MORIN, T. J. & KOBERTZ, W. R. 2007. A derivatized scorpion toxin reveals the functional output of heteromeric KCNQ1-KCNE K+ channel complexes. ACS Chem Biol, 2, 469-73. MORIN, T. J. & KOBERTZ, W. R. 2008. Counting membrane-embedded KCNE beta-subunits in functioning K+ channel complexes. Proc Natl Acad Sci U S A, 105, 1478-82. MOROKUMA, J., BLACKISTON, D., ADAMS, D. S., SEEBOHM, G., TRIMMER, B. & LEVIN, M. 2008. Modulation of potassium channel function confers a hyperproliferative invasive phenotype on embryonic stem cells. Proc Natl Acad Sci U S A, 105, 16608-13.  153 MOSS, A. J. & KASS, R. S. 2005. Long QT syndrome: from channels to cardiac arrhythmias. J Clin Invest, 115, 2018-24. MURRAY, C. I., WESTHOFF, M., ELDSTROM, J., THOMPSON, E., EMES, R. & FEDIDA, D. 2016. Unnatural amino acid photo-crosslinking of the IKs channel complex demonstrates a KCNE1:KCNQ1 stoichiometry of up to 4:4. eLife, 5, 1-18. NAKAJO, K. & KUBO, Y. 2007. KCNE1 and KCNE3 stabilize and/or slow voltage sensing S4 segment of KCNQ1 channel. J Gen Physiol, 130, 269-281. NAKAJO, K., ULBRICH, M. H., KUBO, Y. & ISACOFF, E. Y. 2010. Stoichiometry of the KCNQ1 - KCNE1 ion channel complex. Proc Natl Acad Sci USA, 107, 18862-18867. NEAL, A. M., TAYLOR, H. C., MILLAR, I. D., KIBBLE, J. D., WHITE, S. J. & ROBSON, L. 2011. Renal defects in KCNE1 knockout mice are mimicked by chromanol 293B in vivo: identification of a KCNE1-regulated K+ conductance in the proximal tubule. J Physiol, 589, 3595-609. NERBONNE, J. M. & KASS, R. S. 2005. Molecular physiology of cardiac repolarization. Physiol Rev, 85, 1205-53. NEYROUD, N., TESSON, F., DENJOY, I., LEIBOVICI, M., DONGER, C., BARHANIN, J., FAURÉ, S., GARY, F., COUMEL, P., PETIT, C., SCHWARTZ, K. & GUICHENEY, P. 1997. A novel mutation in the potassium channel gene KVLQT1 causes the Jervell and Lange-Nielsen cardioauditory syndrome. Nat Genet, 15, 186-189. OHNO, S., ZANKOV, D. P., DING, W. G., ITOH, H., MAKIYAMA, T., DOI, T., SHIZUTA, S., HATTORI, T., MIYAMOTO, A., NAIKI, N., HANCOX, J. C., MATSUURA, H. & HORIE, M. 2011. KCNE5 (KCNE1L) variants are novel modulators of Brugada syndrome and idiopathic ventricular fibrillation. Circ Arrhythm Electrophysiol, 4, 352-61. OSTEEN, J. D., BARRO-SORIA, R., ROBEY, S., SAMPSON, K. J., KASS, R. S. & LARSSON, H. P. 2012. Allosteric gating mechanism underlies the flexible gating of KCNQ1 potassium channels. Proc Natl Acad Sci U S A, 109, 7103-8. OSTEEN, J. D., GONZALEZ, C., SAMPSON, K. J., IYER, V., REBOLLEDO, S., LARSSON, H. P. & KASS, R. S. 2010. KCNE1 alters the voltage sensor movements necessary to open the KCNQ1 channel gate. Proc Natl Acad Sci USA, 107, 22710-5. PANAGHIE, G., TAI, K.-K. & ABBOTT, G. W. 2006. Interaction of KCNE subunits with the KCNQ1 K+ channel pore. J Physiol, 570, 455-467. PARK, K. H., PIRON, J., DAHIMENE, S., MÉROT, J., BARÓ, I., ESCANDE, D. & LOUSSOUARN, G. 2005. Impaired KCNQ1-KCNE1 and phosphatidylinositol-4,5-bisphosphate interaction underlies the long QT syndrome. Circ Res, 96, 730-9. PATHAK, M., KURTZ, L., TOMBOLA, F. & ISACOFF, E. 2005. The cooperative voltage sensor motion that gates a potassium channel. J Gen Physiol, 125, 57-69. PICCINI, M., VITELLI, F., SERI, M., GALIETTA, L. J., MORAN, O., BULFONE, A., BANFI, S., POBER, B. & RENIERI, A. 1999. KCNE1-like gene is deleted in AMME contiguous gene syndrome: identification and characterization of the human and mouse homologs. Genomics, 60, 251-7. PLANT, L. D., XIONG, D., DAI, H. & GOLDSTEIN, S. A. N. 2014. Individual IKs channels at the surface of mammalian cells contain two KCNE1 accessory subunits. Proc Natl Acad Sci USA, 111, E1438-46.  154 POLLARD, J. R., ARISPE, N., ROJAS, E. & POLLARD, H. B. 1994. A geometric sequence that accurately describes allowed multiple conductance levels of ion channels: the "three-halves (3/2) rule". Biophys J, 67, 647-55. PRESTON, P., WARTOSCH, L., GÜNZEL, D., FROMM, M., KONGSUPHOL, P., OUSINGSAWAT, J., KUNZELMANN, K., BARHANIN, J., WARTH, R. & JENTSCH, T. J. 2010. Disruption of the K+ channel beta-subunit KCNE3 reveals an important role in intestinal and tracheal Cl- transport. J Biol Chem, 285, 7165-75. PURTELL, K., PARODER-BELENITSKY, M., REYNA-NEYRA, A., NICOLA, J. P., KOBA, W., FINE, E., CARRASCO, N. & ABBOTT, G. W. 2012. The KCNQ1-KCNE2 K⁺ channel is required for adequate thyroid I⁻ uptake. FASEB J, 26, 3252-9. PUSCH, M. 1998. Increase of the single-channel conductance of KvLQT1 potassium channels induced by the association with minK. Pflugers Archiv European Journal of Physiology, 437, 172-174. RADICKE, S., COTELLA, D., GRAF, E. M., BANSE, U., JOST, N., VARRÓ, A., TSENG, G. N., RAVENS, U. & WETTWER, E. 2006. Functional modulation of the transient outward current Ito by KCNE beta-subunits and regional distribution in human non-failing and failing hearts. Cardiovasc Res, 71, 695-703. RAMASUBRAMANIAN, S. & RUDY, Y. 2018. The Structural Basis of IKs Ion-Channel Activation: Mechanistic Insights from Molecular Simulations. Biophys J, 114, 2584-2594. RAVN, L. S., AIZAWA, Y., POLLEVICK, G. D., HOFMAN-BANG, J., CORDEIRO, J. M., DIXEN, U., JENSEN, G., WU, Y., BURASHNIKOV, E., HAUNSO, S., GUERCHICOFF, A., HU, D., SVENDSEN, J. H., CHRISTIANSEN, M. & ANTZELEVITCH, C. 2008. Gain of function in IKs secondary to a mutation in KCNE5 associated with atrial fibrillation. Heart Rhythm, 5, 427-35. RESTIER, L., CHENG, L. & SANGUINETTI, M. C. 2008. Mechanisms by which atrial fibrillation-associated mutations in the S1 domain of KCNQ1 slow deactivation of IKs channels. J Physiol, 586, 4179-91. ROCHELEAU, J. M. & KOBERTZ, W. R. 2008. KCNE Peptides Differently Affect Voltage Sensor Equilibrium and Equilibration Rates in KCNQ1 K+ Channels. J Gen Physiol, 131, 59-68. ROEPKE, T. K., ANANTHARAM, A., KIRCHHOFF, P., BUSQUE, S. M., YOUNG, J. B., GEIBEL, J. P., LERNER, D. J. & ABBOTT, G. W. 2006. The KCNE2 potassium channel ancillary subunit is essential for gastric acid secretion. J Biol Chem, 281, 23740-7. ROEPKE, T. K., KING, E. C., REYNA-NEYRA, A., PARODER, M., PURTELL, K., KOBA, W., FINE, E., LERNER, D. J., CARRASCO, N. & ABBOTT, G. W. 2009. Kcne2 deletion uncovers its crucial role in thyroid hormone biosynthesis. Nat Med, 15, 1186-94. RUSCIC, K. J., MICELI, F., VILLALBA-GALEA, C. A., DAI, H., MISHINA, Y. & BEZANILLA, F. 2013. IKs channels open slowly because KCNE1 accessory subunits slow the movement of S4 voltage sensors in KCNQ1 pore-forming subunits. Proc Natl Acad Sci USA, 110, E559-E566. RYU, S. & YELLEN, G. 2012. Charge movement in gating-locked HCN channels reveals weak coupling of voltage sensors and gate. J Gen Physiol, 140, 469-79.  155 SANGUINETTI, M. C., CURRAN, M. E., ZOU, A., SHEN, J., SPECTOR, P. S., ATKINSON, D. L. & KEATING, M. T. 1996. Coassembly of KvLQT1 and minK (IsK) proteins to form cardiac IKs potassium channel. Nature, 384, 80-83. SANTIAGO-CASTILLO, J. A., COVARRUBIAS, M., SÁNCHEZ-RODRÍGUEZ, J. E., PEREZ-CORNEJO, P. & ARREOLA, J. 2010. Simulating complex ion channel kinetics with IonChannelLab. Channels (Austin), 4, 422-8. SCHOPPA, N. E. & SIGWORTH, F. J. 1998. Activation of Shaker potassium channels. III. An activation gating model for wild-type and V2 mutant channels. J Gen Physiol, 111, 313-42. SCHULZE-BAHR, E., WANG, Q., WEDEKIND, H., HAVERKAMP, W., CHEN, Q., SUN, Y., RUBIE, C., HÖRDT, M., TOWBIN, J. A., BORGGREFE, M., ASSMANN, G., QU, X., SOMBERG, J. C., BREITHARDT, G., OBERTI, C. & FUNKE, H. 1997. KCNE1 mutations cause jervell and Lange-Nielsen syndrome. Nat Genet, 17, 267-8. SESTI, F. & GOLDSTEIN, S. A. 1998. Single-channel characteristics of wild-type IKs channels and channels formed with two minK mutants that cause long QT syndrome. J Gen Physiol, 112, 651-663. SHAMGAR, L., HAITIN, Y., YISHAREL, I., MALKA, E., SCHOTTEINDREIER, H., PERETZ, A., PAAS, Y. & ATTALI, B. 2008. KCNE1 constrains the voltage sensor of Kv7.1 K+ channels. PLoS ONE, 3. SILVA, J. & RUDY, Y. 2005. Subunit interaction determines IKs participation in cardiac repolarization and repolarization reserve. Circulation, 112, 1384-1391. SPLAWSKI, I., SHEN, J., TIMOTHY, K. W., LEHMANN, M. H., PRIORI, S., ROBINSON, J. L., MOSS, A. J., SCHWARTZ, P. J., TOWBIN, J. A., VINCENT, G. M. & KEATING, M. T. 2000. Spectrum of mutations in long-QT syndrome genes. KVLQT1, HERG, SCN5A, KCNE1, and KCNE2. Circulation, 102, 1178-1185. STEFANI, E. & BEZANILLA, F. 1998. Cut-open oocyte voltage-clamp technique. Methods Enzymol, 293, 300-18. STENGL, M., VOLDERS, P. G. A., THOMSEN, M. B., SPÄTJENS, R. L. H. M. G., SIPIDO, K. R. & VOS, M. A. 2003. Accumulation of slowly activating delayed rectifier potassium current (IKs) in canine ventricular myocytes. J Physiol, 551, 777-786. STRUTZ-SEEBOHM, N., PUSCH, M., WOLF, S., TAPKEN, D., GERWERT, K. & ATTALI, B. 2011. Structural Basis of Slow Activation Gating in the Cardiac I Ks Channel Complex. Cell Physiol Biochem, 27, 443-452. SUH, B. C. & HILLE, B. 2008. PIP2 is a necessary cofactor for ion channel function: how and why? Annu Rev Biophys, 37, 175-95. SUN, J. & MACKINNON, R. 2017. Cryo-EM Structure of a KCNQ1/CaM Complex Reveals Insights into Congenital Long QT Syndrome. Cell, 169, 1042-1050.e9. TAKUMI, T., OHKUBO, H. & NAKANISHI, S. 1988. Cloning of a membrane protein that induces a slow voltage-gated potassium current. Science, 242, 1042-5. TAPPER, A. R. & GEORGE, A. L. 2000. MinK subdomains that mediate modulation of and association with KvLQT1. J Gen Physiol, 116, 379-90. TAPPER, A. R. & GEORGE, A. L. 2001. Location and orientation of minK within the I(Ks) potassium channel complex. J Biol Chem, 276, 38249-54.  156 TERRENOIRE, C., CLANCY, C. E., CORMIER, J. W., SAMPSON, K. J. & KASS, R. S. 2005. Autonomic control of cardiac action potentials: role of potassium channel kinetics in response to sympathetic stimulation. Circ Res, 96, e25-34. THOMPSON, E., ELDSTROM, J., WESTHOFF, M., MCAFEE, D., BALSE, E. & FEDIDA, D. 2017. cAMP-dependent regulation of IKs single-channel kinetics. J Gen Physiol, 149, 781-798. THOMPSON, E., ELDSTROM, J., WESTHOFF, M., MCAFEE, D. & FEDIDA, D. 2018. The   IKs   Channel Response to cAMP Is Modulated by the KCNE1:KCNQ1 Stoichiometry. Biophys J, 115, 1731-1740. TINEL, N., DIOCHOT, S., BORSOTTO, M., LAZDUNSKI, M. & BARHANIN, J. 2000. KCNE2 confers background current characteristics to the cardiac KCNQ1 potassium channel. EMBO J, 19, 6326-30. TOYODA, F., UEYAMA, H., DING, W. G. & MATSUURA, H. 2006. Modulation of functional properties of KCNQ1 channel by association of KCNE1 and KCNE2. Biochem Biophys Res Commun, 344, 814-20. TRISTANI-FIROUZI, M., CHEN, J., MITCHESON, J. S. & SANGUINETTI, M. C. 2001. Molecular biology of K(+) channels and their role in cardiac arrhythmias. Am J Med, 110, 50-9. TRISTANI-FIROUZI, M. & SANGUINETTI, M. C. 1998. Voltage-dependent inactivation of the human K+ channel KvLQT1 is eliminated by association with minimal K+ channel (minK) subunits. Journal of Physiology, 510, 37-45. TYSON, J., TRANEBJAERG, L., BELLMAN, S., WREN, C., TAYLOR, J. F., BATHEN, J., ASLAKSEN, B., SØRLAND, S. J., LUND, O., MALCOLM, S., PEMBREY, M., BHATTACHARYA, S. & BITNER-GLINDZICZ, M. 1997. IsK and KvLQT1: mutation in either of the two subunits of the slow component of the delayed rectifier potassium channel can cause Jervell and Lange-Nielsen syndrome. Hum Mol Genet, 6, 2179-85. TZOUNOPOULOS, T., MAYLIE, J. & ADELMAN, J. P. 1998. Gating of I(sK) channels expressed in Xenopus oocytes. Biophysical journal, 74, 2299-305. VALLON, V., GRAHAMMER, F., RICHTER, K., BLEICH, M., LANG, F., BARHANIN, J., VOLKL, H. & WARTH, R. 2001. Role of KCNE1-dependent K(+) fluxes in mouse proximal tubule. Journal of the American Society of Nephrology, 12, 2003-2011. VAN HORN, W. D., VANOYE, C. G. & SANDERS, C. R. 2011. Working model for the structural basis for KCNE1 modulation of the KCNQ1 potassium channel. Current Opinion in Structural Biology, 21, 283-291. VANOYE, C. G., WELCH, R. C., TIAN, C., SANDERS, C. R. & GEORGE, A. L., JR. 2010. KCNQ1/KCNE1 assembly, co-translation not required. Channels (Austin, Tex.), 4, 108-114. VETTER, D. E., MANN, J. R., WANGEMANN, P., LIU, J., MCLAUGHLIN, K. J., LESAGE, F., MARCUS, D. C., LAZDUNSKI, M., HEINEMANN, S. F. & BARHANIN, J. 1996. Inner ear defects induced by null mutation of the isk gene. Neuron, 17, 1251-64. VIRÁG, L., IOST, N., OPINCARIU, M., SZOLNOKY, J., SZÉCSI, J., BOGÁTS, G., SZENOHRADSZKY, P., VARRÓ, A. & PAPP, J. G. 2001. The slow component of the delayed rectifier potassium current in undiseased human ventricular myocytes. Cardiovasc Res, 49, 790-7.  157 WANG, K., TERRENOIRE, C., SAMPSON, K. J., IYER, V., OSTEEN, J. D., LU, J., KELLER, G., KOTTON, D. N. & KASS, R. S. 2011a. Biophysical properties of slow potassium channels in human embryonic stem cell derived cardiomyocytes implicate subunit stoichiometry. J Physiol, 589, 6093-104. WANG, K. W. & GOLDSTEIN, S. A. 1995. Subunit composition of minK potassium channels. Neuron, 14, 1303-9. WANG, K. W., TAI, K. K. & GOLDSTEIN, S. A. 1996a. MinK residues line a potassium channel pore. Neuron, 16, 571-7. WANG, Q., CURRAN, M. E., SPLAWSKI, I., BURN, T. C., MILLHOLLAND, J. M., VANRAAY, T. J., SHEN, J., TIMOTHY, K. W., VINCENT, G. M., DE JAGER, T., SCHWARTZ, P. J., TOUBIN, J. A., MOSS, A. J., ATKINSON, D. L., LANDES, G. M., CONNORS, T. D. & KEATING, M. T. 1996b. Positional cloning of a novel potassium channel gene: KVLQT1 mutations cause cardiac arrhythmias. Nat Genet, 12, 17-23. WANG, W., XIA, J. & KASS, R. S. 1998. MinK-KvLQT1 fusion proteins, evidence for multiple stoichiometries of the assembled IsK channel. J Biol Chem, 273, 34069-74. WANG, Y., ZANKOV, D. P., JIANG, M., ZHANG, M., HENDERSON, S. C. & TSENG, G. N. 2013. [Ca2+]i elevation and oxidative stress induce KCNQ1 protein translocation from the cytosol to the cell surface and increase slow delayed rectifier (IKs) in cardiac myocytes. J Biol Chem, 288, 35358-71. WANG, Y., ZHANG, M., XU, Y., JIANG, M., ZANKOV, D. P., CUI, M. & TSENG, G. N. 2012. Probing the structural basis for differential KCNQ1 modulation by KCNE1 and KCNE2. J Gen Physiol, 140, 653-669. WANG, Y. H., JIANG, M., XU, X. L., HSU, K. L., ZHANG, M. & TSENG, G. N. 2011b. Gating-related molecular motions in the extracellular domain of the IKs channel: implications for IKs channelopathy. J Membr Biol, 239, 137-56. WARTH, R. & BARHANIN, J. 2002. The multifaceted phenotype of the knockout mouse for the KCNE1 potassium channel gene. Am J Physiol Regul Integr Comp Physiol, 282, R639-48. WATANABE, E., YASUI, K., KAMIYA, K., YAMAGUCHI, T., SAKUMA, I., HONJO, H., OZAKI, Y., MORIMOTO, S., HISHIDA, H. & KODAMA, I. 2007. Upregulation of KCNE1 induces QT interval prolongation in patients with chronic heart failure. Circ J, 71, 471-8. WERRY, D., ELDSTROM, J., WANG, Z. & FEDIDA, D. 2013. Single-channel basis for the slow activation of the repolarizing cardiac potassium current, I(Ks). Proc Natl Acad Sci USA, 110, E996-1005. WESTHOFF, M., MURRAY, C. I., ELDSTROM, J. & FEDIDA, D. 2017. Photo-Cross-Linking of I. Biophys J, 113, 415-425. WIT, A. L., HOFFMAN, B. F. & ROSEN, M. R. 1975. Electrophysiology and pharmacology of cardiac arrhthmias IX. Cardiac electrophysiologic effects of beta adrenergic receptor stimulation and blockade. Part C. Am Heart J, 90, 795-803. WROBEL, E., TAPKEN, D. & SEEBOHM, G. 2012. The KCNE tango - how KCNE1 interacts with Kv7.1. Frontiers in Pharmacology, 3 AUG, 1-14.  158 WU, D., DELALOYE, K., ZAYDMAN, M. A., NEKOUZADEH, A., RUDY, Y. & CUI, J. 2010. State-dependent electrostatic interactions of S4 arginines with E1 in S2 during Kv7.1 activation. J Gen Physiol, 135, 595-606. WU, D. M., JIANG, M., ZHANG, M., LIU, X. S., KOROLKOVA, Y. V. & TSENG, G. N. 2006a. KCNE2 is colocalized with KCNQ1 and KCNE1 in cardiac myocytes and may function as a negative modulator of I(Ks) current amplitude in the heart. Heart Rhythm, 3, 1469-80. WU, D. M., LAI, L. P., ZHANG, M., WANG, H. L., JIANG, M., LIU, X. S. & TSENG, G. N. 2006b. Characterization of an LQT5-related mutation in KCNE1, Y81C: implications for a role of KCNE1 cytoplasmic domain in IKs channel function. Heart Rhythm, 3, 1031-40. WU, J., DING, W. G. & HORIE, M. 2016. Molecular pathogenesis of long QT syndrome type 1. J Arrhythm, 32, 381-388. XIE, J. & SCHULTZ, P. G. 2005. An expanding genetic code. Methods, 36, 227-38. XU, X., JIANG, M., HSU, K. L., ZHANG, M. & TSENG, G. N. 2008. KCNQ1 and KCNE1 in the IKs channel complex make state-dependent contacts in their extracellular domains. The Journal of general physiology, 131, 589-603. XU, Y., WANG, Y., ZHANG, M., JIANG, M., ROSENHOUSE-DANTSKER, A., WASSENAAR, T. & TSENG, G. N. 2015. Probing binding sites and mechanisms of action of an IKs activator by computations and experiments. Biophysical Journal, 108, 62-75. YANG, Y. & SIGWORTH, F. J. 1998. Single-Channel Properties of I Ks Potassium Channels. Journal of general physiology, 112. YANG, Y., XIA, M., JIN, Q., BENDAHHOU, S., SHI, J., CHEN, Y., LIANG, B., LIN, J., LIU, Y., LIU, B., ZHOU, Q., ZHANG, D., WANG, R., MA, N., SU, X., NIU, K., PEI, Y., XU, W., CHEN, Z., WAN, H., CUI, J. & BARHANIN, J. 2004. Identification of a KCNE2 gain-of-function mutation in patients with familial atrial fibrillation. Am J Hum Genet, 75, 899-905. YE, S., KÖHRER, C., HUBER, T., KAZMI, M., SACHDEV, P., YAN, E. C. Y., BHAGAT, A., RAJBHANDARY, U. L. & SAKMAR, T. P. 2008. Site-specific incorporation of keto amino acids into functional G protein-coupled receptors using unnatural amino acid mutagenesis. Journal of Biological Chemistry, 283, 1525-1533. YOUNG, T. S. & SCHULTZ, P. G. 2010. Beyond the canonical 20 amino acids: expanding the genetic lexicon. J Biol Chem, 285, 11039-44. YU, H., LIN, Z., MATTMANN, M. E., ZOU, B., TERRENOIRE, C., ZHANG, H., WU, M., MCMANUS, O. B., KASS, R. S., LINDSLEY, C. W., HOPKINS, C. R. & LI, M. 2013. Dynamic subunit stoichiometry confers a progressive continuum of pharmacological sensitivity by KCNQ potassium channels. Proc Natl Acad Sci U S A, 110, 8732-7. ZAGOTTA, W. N., HOSHI, T. & ALDRICH, R. W. 1994. Shaker potassium channel gating. III: Evaluation of kinetic models for activation. J Gen Physiol, 103, 321-62. ZAYDMAN, M. A., KASIMOVA, M. A., MCFARLAND, K., BELLER, Z., HOU, P., KINSER, H. E., LIANG, H., ZHANG, G., SHI, J., TAREK, M. & CUI, J. 2014. Domain-domain interactions determine the gating, permeation, pharmacology, and subunit modulation of the IKs ion channel. eLife, 3, 1-19.  159 ZAYDMAN, M. A., SILVA, J. R., DELALOYE, K., LI, Y., LIANG, H., LARSSON, H. P., SHI, J. & CUI, J. 2013. Kv7.1 ion channels require a lipid to couple voltage sensing to pore opening. Proc Natl Acad Sci USA, 110, 13180-5. ZDEBIK, A. A., WANGEMANN, P. & JENTSCH, T. J. 2009. Potassium ion movement in the inner ear: insights from genetic disease and mouse models. Physiology (Bethesda), 24, 307-16. ZENG, Z., TAN, C., TENG, S., CHEN, J., SU, S., ZHOU, X., WANG, F., ZHANG, S., GU, D., MAKIELSKI, J. C. & PU, J. 2007. The single nucleotide polymorphisms of I(Ks) potassium channel genes and their association with atrial fibrillation in a Chinese population. Cardiology, 108, 97-103. ZENG, Z. Y., PU, J. L., TAN, C., TENG, S. Y., CHEN, J. H., SU, S. Y., ZHOU, X. Y., ZHANG, S., LI, Y. S., WANG, F. Z. & GU, D. F. 2005. [The association of single nucleotide polymorphism of slow delayed rectifier K+ channel genes with atrial fibrillation in Han nationality Chinese]. Zhonghua Xin Xue Guan Bing Za Zhi, 33, 987-91. ZHENG, R., THOMPSON, K., OBENG-GYIMAH, E., ALESSI, D., CHEN, J., CHENG, H. & MCDONALD, T. V. 2010. Analysis of the interactions between the C-terminal cytoplasmic domains of KCNQ1 and KCNE1 channel subunits. Biochem J, 428, 75-84.     160 Appendix A: Supplementary Figures for Chapter 2  Appendix A, Figure 1. All-points amplitude histograms. All-points amplitude histograms of single channel records during a 60 mV step, filtered at 200 Hz for EQ (left) and EQQ (right) including sweeps with no channel activity. These are the same single channel current records as shown in Figure 2.2B.   Appendix A, Figure 2. Subconductance analysis of EQQ and EQ demonstrates that as the number of KCNE1 subunits decreases so does higher conductance state occupancy. (A) All-points amplitude histograms, filtered at 200 Hz, from 11 active sweeps of EQQ (left) and EQ (right) are plotted in blue with idealized histograms of the same data superimposed in red. The left axis represents number of events for the all-points amplitude histogram and right axis for the idealized histogram. Events were idealized using six separate thresholds (0.08, 0.13, 0.19, 0.29, 0.44 and 0.66 pA) as indicated in the tables below and are set based on our earlier work (Werry et al., 2013) and the 3/2 rule for subconductance behavior (Pollard et al., 1994). (B) Total and average dwell times (ms) in each sublevel were calculated as well as the percent dwell time and event numbers. The mean dwell time from the closed state does not include the first latency. Amplitude (pA)-0.2 0.0 0.2 0.4 0.6 0.8Events050000100000150000200000250000300000350000Amplitude (pA)-0.2 0.0 0.2 0.4 0.6 0.8Events0100000200000300000400000500000600000Amplitude (pA)-0.2 0.0 0.2 0.4 0.6 0.8Events050000100000150000200000250000300000350000Amplitude (pA)0.0 0.2 0.4 0.6 0.8Events050001000015000200002500030000Amplitude (pA)0.0 0.2 0.4 0.6 0.8Events0100200300400500600Amplitude (pA)0.0 0.2 0.4 0.6 0.8Events010000200003000040000Events02004006008001000120014001600EQQ EQ   Closed 0.08 pA 0.13 pA 0.19 pA 0.29 pA 0.44 pA 0.66 pA All Total Dwell Time 22684.5 5300.84 5690.32 7274.07 2795.81 135.224 43880.7 Mean 12.245 2.080 1.732 2.792 2.931 2.940 % of total 51.7% 12.1% 13.0% 16.6% 6.4% 0.3% # events 968 2549 3285 2605 954 46 0 10407   Closed 0.08 pA 0.13 pA 0.19 pA 0.29 pA 0.44 pA 0.66 pA All Total Dwell Time 22042.33 2701.09 1979.97 2503.21 4170.01 7986.11 2366.96 43749.69 Mean 13.852 2.311 1.391 1.613 2.202 4.930 4.304 % of total 50.4% 6.2% 4.5% 5.7% 9.5% 18.3% 5.4% # events 477 1169 1423 1552 1894 1620 550 8685 A B  161  Appendix A, Figure 3. M62W KCNE1 does not alter IKs channel activation compared to wild type. (A) Representative whole cell patch clamp recording is shown of M62W KCNE1 co-expressed with KCNQ1. Current-voltage relations were elicited using a 4 s isochronal activation protocol. Only odd numbered sweeps are presented for clarity. (B) Plot of the tail current G-V relations for KCNQ1 + M62W (blue triangle) or wild type (black circle) KCNE1 (n = 9–10).        162  Appendix A, Figure 4. F57Bpa KCNE1 does not alter channel gating of KCNQ1, EQQ and EQ compared to wild type KCNE1. (A) Channel diagrams indicate the configuration and proposed stoichiometry of the channels. Representative currents are shown of KCNQ1 (black circle), EQQQQL (green diamond), EQQS (blue square) and EQS (orange triangle) (sub-S or –L denotes the 31 aa or 52 aa linker between KCNE1 and KCNQ1 in the fusion proteins) alone (B; open), co-expressed with KCNE1- GFP (C; filled) or F57Bpa KCNE1-GFP (D; light filled). (E) Currents were elicited using a 4s isochronal activation protocol. Only odd numbered sweeps are shown for clarity. (F) Tail current G-V plots are shown comparing the response to increasing number of β-subunits either by fusion (top) or by co-expression of KCNE1/F57Bpa KCNE1 (middle and bottom). (G) G-V plots comparing each channel complex with and without co-expression of KCNE1/F57Bpa KCNE1. (H) Summary of each channel’s V1/2 of activation with and without KCNE1 (n = 3–16).     163  Appendix A, Figure 5. UV-rundown is consistent across all the IKs channel configurations. (A) Schematic of the UV-voltage protocol. Representative currents depicting UV-rundown are shown for EQQ (B) and EQ (C). (D) Combined diary plot of UV-treated normalized peak currents for EQQ (green square), EQ (orange triangle), as well as the data presented in Figure 2.5; KCNQ1 (red circle), EQQQQ (green diamond) EQQ (blue square) and EQ (purple triangle) all co-expressed with wild type KCNE1. (E) Summary of the rundown rates obtained from each cell (n = 3–5).    Appendix A, Figure 6. F57Bpa KCNE1 covalently crosslinks with KCNQ1. Western blot of the immunoprecipitation of KCNQ1 co-expressed with F57Bpa KCNE1-GFP (left) or wild type KCNE1-GFP (right) after 0, 2 or 5 min of UV-irradiation. Western blots of the immunoprecipitation elution fractions were probed for GFP (upper) or T7-KCNQ1 (lower) (n = 2). Arrow indicates the crosslinked KCNE1-KCNQ1 complex.    164 Appendix B: Supplementary Figures for Chapter 3   Appendix B, Figure 1. Incorporation of Bpa at positions F56 and F57 in KCNE1 is well tolerated. (A) Currents were obtained using 4 s pulses to potentials between -80 and +100 mV. Representative currents of KCNQ1+F56 (B) and KCNQ1+F57 (C) are shown from cells cultured in the presence (upper) and absence (lower) of 1 mM Bpa. Only odd numbered sweeps are presented for clarity. (D) G-V relationship of KCNQ1+wt E1 (open squares), KCNQ1+F56Bpa (black triangles) and KCNQ1+F57Bpa (grey triangles) (n = 3-11).     165  Appendix B, Figure 2. Wild type UV rundown is consistent between -90 mV and +60 mV. (A) Schematic of the +60 mV UV-voltage protocol. A 300 ms flash of UV light (bold black line) is applied each sweep after a 3.7 s voltage step to +60 mV. A representative current is shown for KCNQ1+wt E1. UV was applied at sweep 5, after a stable baseline had been established. Sweeps 1 and 5-25 are shown. (B) Diary plot of the UV-treated normalized peak current from KCNQ1+wt E1 +60 mV (open grey triangles) and -90 mV (open black squares) (n = 5-6). Peak current was either measured at the end of the 4 s pulse (-90 mV) or immediately before UV application (+60 mV). (C) Summary of the rundown rates (KRD) at each potential (n = 5-6, p-value = 0.80). (D, upper) UV protocol where 6 s of UV is applied after 8 s at +60 mV. (D, lower) A representative set of long pulse currents of KCNQ1+wt E1 at +60 mV.   166  Appendix B, Figure 3. Characterization of E1R/R4E IKs.  (A) Currents were obtained using 4 s pulses to potentials between -80 and +80 mV. Representative currents of E1R/R4E KCNQ1+wt E1 (B), E1R/R4E KCNQ1+F56Bpa (C) and E1R/R4E KCNQ1+F57Bpa (D) are shown. Only odd numbered sweeps are presented for clarity. (E) G-V relationship of wt KCNQ1+wt E1 (open squares), E1R/R4E KCNQ1+wt E1 (open diamonds), E1R/R4E KCNQ1+F56Bpa (grey squares) and E1R/R4E KCNQ1+F57Bpa (black diamonds) (n = 3-11). For single channel recordings, cells were held at -80 mV, followed by a 4 s pulse to +60 mV, with the tail current recorded at -40 mV (F, upper). Representative single channel recordings of E1R/R4E KCNQ1+wt KCNE1 are shown (F, lower).   167 Appendix C: Supplementary Figures for Chapter 4   Appendix C, Figure 1. When expressed without E1, channel complexes with one or two E160R mutations produce functional currents. Currents were obtained using the same isochronal activation protocol as shown in Figure 4.2B. Representative currents are shown for wt QQQQ (A), E160R Q*QQQ (B), and E160R Q*Q (C). Cartoons describe the mutations in the constructs: Wt Q1 (unfilled circles) and E160R Q1 (red circles). Black lines indicate tethers between subunits. (D) G-V plots of wt QQQQ (open purple triangles), E160R Q*QQQ (open purple circles), wt QQ (open green diamonds) and E160R Q*Q (open green circles). (E) Summary of the V1/2s of activation for each construct (n = 3-4; * = p<0.05, Table 4.1).    168  Appendix C, Figure 2. Wt QQ alone and co-expressed with E1. Currents were obtained using the same isochronal activation protocol described in Figure 4.2B. A representative set of currents is shown for wt QQ alone (A) and wt QQ + E1 (B). Cartoons describe the constructs: E1 (small grey circles) and wt Q1 (unfilled circles). Black lines indicate tethers between subunits. (C) G-V plots of wt QQ alone (open green diamonds) and wt QQ + E1 (closed green diamonds). (D) Summary of the V1/2s of activation of each construct (n = 3-4; Table 4.1).   169   Appendix C, Figure 3. Activation kinetics of E160R QQQ*Q* + E1 are similar to E160R Q*Q + E1. Currents were obtained using the same isochronal activation protocol described in Figure 4.2B. (A) A representative set of currents is shown for E160R QQQ*Q* + E1. Cartoons describe the mutations in the constructs: E1 (small grey circles), wt Q1 (unfilled circles) and E160R Q1 (red circles). Black lines indicate tethers between subunits. (B) G-V plots of E160R Q*Q + E1 (green circles) and E160R QQQ*Q* + E1 (open black triangles). (C) Summary of the V1/2 of activation for each construct (n = 3-8; p-value = 0.31; Table 4.1). (D) Representative single channel sweeps of E160R QQQ*Q* + E1. Cells were pulsed to +60 mV for 4 s, followed by a tail current at -40 mV. (E) All points-histogram of 39 active sweeps of E160R EQ*Q + E1 (green line; n = 3) and 39 active sweeps of E160R QQQ*Q* + E1 (dashed black line; n = 1).     170  Appendix C, Figure 4. TEA+ sensitivity reveals mutated E160R subunits are not excluded from channel assembly. Currents were obtained by pulsing to +60 mV or +80 mV for 4 s, followed by -40 mV for 0.9 s. The sweep to sweep interval was 15 s. Representative current diary plots for wt EQQ + E1 (A), V319Y EQQ + E1 (B) and E160R/V319Y EQ*Q + E1 (C) showing peak current at +60 or +80 mV vs sweep number. The solid black line indicates addition of 2 mM TEA+ and the dashed line indicates addition of 50 mM TEA+. Cartoons describe the mutations in the constructs: E1 (small grey circles), wt Q1 (unfilled circles), V319Y Q1 (blue circles) and E160R/V319Y Q1 (blue circles with red border). Black lines indicate tethers between subunits. (D) Plot comparing percentage peak current for wt EQQ + E1 (control (n = 4), 2 mM (n =1) and 50 mM (n = 4)(p-value between control and 50 mM = 0.9209), V319Y EQQ + E1 (control and 2 mM, n = 5; p-value = <0.0001) and E160R/V319Y EQ*Q + E1 (control and 2 mM, n = 4; p-value = <0.0001).    171  Appendix C, Figure 5. Representative single exponential fits to activating current waveforms. Representative sets of currents during activation to a range of potentials, fit with a single exponential function (red line) starting at 0.5 s for wt QQQQ + E1 (A), wt QQ + E1 (B), wt EQQQQ + E1 (C), E160R Q*QQQ + E1 (D), E160R Q*Q + E1 (E) and E160R EQ*QQ*Q* + E1 (F). Membrane current at each membrane potential was fit with a single exponential equation to produce a time constant of activation (𝜏act) for each construct at each voltage. Cartoons describe the mutations in the constructs: E1 (small grey circles), wt Q1 (unfilled circles) and E160R Q1 (red circles). Black lines indicate tethers between subunits. (G) The time constants of activation (𝜏act) at various membrane potentials are plotted for wt QQQQ + E1 (purple triangles), wt QQ + E1 (green diamonds) and wt EQQQQ + E1 (blue squares) (n = 3-7). No significant differences were found between values for different wt constructs.    172  Appendix C, Figure 6. Representative single exponential fits to tail currents.  (A) Deactivation protocol used to obtain tail currents. Cells were held at -80 mV, and pulsed to +60 mV for 4 s followed by a 4 s pulse to a range of potentials from -30 to -120 mV in 10 mV steps, to record tail currents. Representative set of tail currents with a single exponential fit (blue line) for wt QQQQ + E1 (B), wt QQ + E1 (C), wt EQQQQ + E1 (D), E160R Q*QQQ + E1 (E), E160R Q*Q + E1 (F) and E160R EQ*QQ*Q* + E1 (G). Cartoons describe the mutations in the constructs: E1 (small grey circles), wt Q1 (unfilled circles) and E160R Q1 (red circles). Black lines indicate tethers between subunits. (H) Time constants of deactivation (𝜏deact) at various membrane potentials were plotted for wt QQQQ + E1 (purple triangles), wt QQ + E1 (green diamonds) and wt EQQQQ + E1 (blue squares) (n = 4-7). No significant differences were found between values for different wt constructs.  173   Appendix C, Figure 7. Channel complexes with tethered E1s produce similar results to the untethered constructs. (A) Summary of the V1/2 of activation for channels containing one E160R mutant compared to wt: wt EQQQQ + E1 (orange triangles), E160R EQ*QQQ + E1 (orange circles), wt QQQQ + E1 (purple triangles) and E160R Q*QQQ + E1 (purple circles) (n = 4-7; Table 4.1). (B) Summary of the V1/2 of activation for channels containing two E160R mutants compared to wt: wt EQQ + E1 (pink triangles), E160R EQ*Q + E1 (pink circles), wt QQ + E1 (green diamonds) and E160R Q*Q + E1 (green circles) (n = 3-8; * = p<0.05; Table 4.1). Cartoons describe the mutations in the constructs: E1 (small grey circles), wt Q1 (unfilled circles) and E160R Q1 (red circles). Black lines indicate tethers between subunits.       174   Appendix C, Figure 8. Additional single channel information. (A) Ensemble averages of 17 active wt EQ sweeps (black trace) and 17 active E160R Q*QQQ + E1 sweeps (purple trace), each from a single patch. (B) All-points histogram for 17 active wt EQ sweeps (black line) and 17 active E160R Q*QQQ + E1 sweeps (purple line). (C, left) Individual idealized histograms for E160R EQ*Q + E1 of 22 active sweeps from Cell 1 (blue line), 18 active sweeps from Cell 2 (green line) and 9 active sweeps from Cell 3 (pink line). (C, right) Individual idealized histograms for E160R EQ*QQ*Q* + E1 of 29 active sweeps from Cell 1 (blue line) and 20 active sweeps from Cell 2 (green line). The bin width used was 0.01 pA and events <3.5 ms duration were excluded from analysis. (D) P-values obtained by one-way ANOVA comparing percentage of total dwell time spent at each subconductance level are shown for the different constructs (EQ vs.  175 E160R EQ*Q + E1 vs. E160R EQ*QQ*Q* + E1). Significant p-values are displayed in green. (E) Average closed dwell times greater than 30 ms are displayed for wt EQ (left), E160R EQ*Q + E1 (middle) and E160R EQ*QQ*Q* + E1 (right).    176  Appendix C, Figure 9. Effect of one, two or four KCNE1 F57W mutations on the IKs channel.  Currents were obtained using the same isochronal activation protocol described in Figure 4.2B. Representative currents are shown for F57W E’QQ + E1 (A) and F57W E’Q (B). Cartoons describe the mutations in the constructs: Wt E1 (small grey circles), F57W E1 (small orange circles), E160R Q1 (red circles) and wt Q1 (unfilled circles). Black lines indicate tethers between subunits. (C) G-V plots of wt EQ (black circles), F57W E’QQQQ + E1 (open red squares), F57W E’QQ + E1 (red circles) and F57W E’Q (open red triangles). (D) Summary of the V1/2s of activation of wt, F57W and E160R constructs (n = 2-5; Table 4.1). (E)  Representative single channel recordings of E160R Q*Q + E1 (left)  177 and F57W E’QQ + E1 (right). Membrane patches containing a single mutant IKs channel were stepped from -80 to +60 mV for 4 s and then to -40 mV for 0.8 s as shown in Figure 5A. (F) All-points amplitude histogram of 10 active E160R Q*Q + E1 sweeps (green line) and 10 active F57W E’QQ + E1 sweeps (dashed red line).     178  Appendix C, Figure 10. Additional MTSET information. Currents were obtained using the same isochronal activation protocol as shown in Figure 4.2B. The sweep to sweep interval was 20 s. Representative currents are shown for G229C Q-E160R Q* + E1 (A) and E160R/G229C Q*-Q + E1 (B). Cartoons describe the mutations in the constructs: E1 (small grey circles), wt Q1 (unfilled circles), E160R Q1 (red circles), G229C Q1 (blue circles), E160R/G229C Q1 (blue circles with red border).  179 Black lines indicate tethers between subunits. (C) G-V plots of G229C Q-E160R Q* + E1 (red squares) and E160R/G229C Q*-Q + E1 (blue triangles). (D) Summary of the V1/2 of activation for each construct (n = 3-4; Table 4.1). Holding G229C Q-E160R Q* + E1 channels longer between sweeps results in more sigmoidal currents. Currents were obtained using a similar isochronal activation protocol as shown in Figure 4.2B, however voltage steps were 20 mV and the sweep to sweep interval was 60 s. (E) A representative set of currents is shown for G229C Q-E160R Q* + E1. (F) G-V plots of G229C Q-E160R Q* + E1 with sweep to sweep intervals of 20 s (red squares) and 60 s (open black squares). (G) Summary of the V1/2 of activation for each construct (n = 2-3). (H) Normalized peak current over 20 sweeps for wt Q1 + E1 in the absence of MTSET (black circles, n = 4) and E160R/G229C Q*-Q + E1 in the presence of 2 mM MTSET after sweep 5 (blue triangles, n = 6) is shown.      Appendix C, Figure 11. VCF recordings of G219C Q-Q + E1 channels. VCF recordings were obtained using the same protocol as described in Figure 4.8A. Representative current (upper) and fluorescence (lower) recordings are shown for G219C Q-Q + E1 (A). Cartoons describe the mutations in the constructs: E1 (small grey circles), wt Q1 (unfilled circles), G219C Q1 (teal circles) and E160R Q1 (red circles). Black lines indicate tethers between subunits. (B) G-V and F-V plots are shown for G219C Q-Q + E1 (G-V: closed purple diamonds and F-V: open purple diamonds) and G219C Q-E160R Q* + E1 (G-V: closed red squares and F-V: open red squares) (n = 2-7).   

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0376582/manifest

Comment

Related Items