Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Voltage-gated potassium ion channels : evolution, functional variation and human disease Jackson, Heather Ann 2012

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

Item Metadata

Download

Media
24-ubc_2012_fall_jackson_heather.pdf [ 10.48MB ]
Metadata
JSON: 24-1.0073194.json
JSON-LD: 24-1.0073194-ld.json
RDF/XML (Pretty): 24-1.0073194-rdf.xml
RDF/JSON: 24-1.0073194-rdf.json
Turtle: 24-1.0073194-turtle.txt
N-Triples: 24-1.0073194-rdf-ntriples.txt
Original Record: 24-1.0073194-source.json
Full Text
24-1.0073194-fulltext.txt
Citation
24-1.0073194.ris

Full Text

    VOLTAGEGATED POTASSIUM ION CHANNELS: EVOLUTION, FUNCTIONAL VARIATION AND HUMAN DISEASE   by   Heather Ann Jackson   BSc., Simon Fraser University, 2001   A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF   DOCTOR OF PHILOSOPHY  in  THE FACULTY OF GRADUATE STUDIES  (Physiology)   THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)    August, 2012   © Heather Ann Jackson, 2012  ii ABSTRACT Hyperpolarizationactivated cyclic nucleotide gated (HCN) channels are structurally similar to voltagegated potassium channels and play pivotal roles in cellular pacemaking. Their physiological relevance is illustrated by the fact that genetic mutations in HCN channels are associated with cardiac arrhythmias. In this study, we performed indepth evolutionary analyses of HCN channels and functionally characterized the biophysical properties of two novel HCN clones (ciHCNa and ciHCNb) from the urochordate, Ciona intestinalis; a species emerging at the pivotal evolutionary period of invertebrate and vertebrate divergence that occurred approximately 550MYA. We have expanded the list of known HCN sequences by identifying and annotating 31 novel genes from invertebrates, urochordates, fish, amphibians, birds, and mammals. Our data suggest that the four vertebrate HCN isoforms arose via three duplication and diversification events from a single ancestral gene following the divergence of urochordates. Functional analyses of the two ciHCN channels further support this evolutionary trajectory, suggesting that the common single HCN ancestor of urochordates and vertebrates had a mammalianlike channel phenotype. Lineagespecific duplication and diversification events and 550MY of independent evolution has lead to two Ciona HCN channels with distinct biophysical properties. The voltagegated potassium channels, HERG and KCNQ1, play a key role in cardiac repolarization. Mutations in these delayed rectifier channels are also associated with cardiac arrhythmias, including long QT syndrome and sudden death. Taking advantage of the >200 disease mutations in both of these channels, we performed the first quantitative evolutionary and chemical severity analysis of arrhythmiaassociated mutations (AAMs). Unlike nonsynonymous polymorphisms (nsSNPs), AAMs are preferentially located to the evolutionarily conserved and functionally important sites and regions within HERG and KCNQ1. The mutations are also chemically more severe than changes which occur throughout evolution. In conjunction with previous studies, our findings suggest that novel diseaseassociated mutations can be identified by surveying the naturally occurring variation that exists among species. Overall, this thesis contributes to the current knowledge of the interdependent relationships that exist among ion channel evolution, ion channel function, and human disease.  iii PREFACE CHAPTER 2: A version of this chapter has been published. Jackson, HA, Marshall, CM, Accili, EA. (2007) Evolution and structural diversification of hyperpolarizationactivated cyclic nucleotidegated channel genes. Physiological Genomics, 29(3): 23145.  Contributions: HJ was responsible for experimental design and work, data analysis and writing of the manuscript in partnership with EAA. CM provided guidance and contributed to the original design of the project. All revisions required for publication and all figures and tables were produced by HJ.  CHAPTER 3: A version of this chapter will be submitted for publication. Jackson, HA, Hegle, A, Nazzari, H, Jegla, T, Accili EA (2011). “Duplication and asymmetric diversification of HCN channel isoforms in Ciona intestinalis.”  Contributions: HJ was responsible for equipment setup, experimental design and work, data analysis and writing of the manuscript in partnership with EAA. TJ collected and generated the clones used in this experiment, contributed to the original design of the project and provided feedback on experimental results. HN generated tagged constructs for Nglycosylation experiments. AH assisted with the experimental design for the western blots performed to generate Figure 1D. AR provided the two new sequences included in the evolutionary analysis used in Figure 1A and 1B. All figures and tables were produced by HJ.  CHAPTER 4: A version of this chapter has been published. Jackson, HA and Accili, EA. (2008). Evolutionary analyses of KCNQ1 and HERG voltagegated potassium channel sequences reveal location specific susceptibility and augmented chemical severities of arrhythmogenic mutations. BMC Evolutionary Biology 8: 188.   iv Contributions: HJ was responsible for equipment setup, experimental design and work, data analysis and writing of the manuscript in partnership with EAA. All revisions required for publication and all figures and tables were produced by HJ.  v TABLE OF CONTENTS ABSTRACT ................................................................................................................................... ii PREFACE ..................................................................................................................................... iii TABLE OF CONTENTS ............................................................................................................. v LIST OF TABLES ..................................................................................................................... viii LIST OF FIGURES ..................................................................................................................... ix ABBREVIATIONS ...................................................................................................................... xi ACKNOWLEDGEMENTS ...................................................................................................... xiii 1. INTRODUCTION................................................................................................................. 1 1.1 VoltageGated Ion Channels ................................................................................................. 1 1.2 VoltageGated Potassium Channel Family ........................................................................... 4 1.2.1 Overview ........................................................................................................................ 4 1.2.2 Kv Structure/Function .................................................................................................... 4 1.2.3 Kv Channelopathies ........................................................................................................ 8 1.2.4 Kv Channel Relatives ................................................................................................... 11 1.2.4.1 CNG Channels .......................................................................................................... 11 1.2.4.2 KAT1 and AKT1 Channels ...................................................................................... 11 1.3 HCN Channels .................................................................................................................... 13 1.3.1 Funny Current (If) ........................................................................................................ 13 1.3.2 HCN Mammalian Tissue Expression........................................................................... 15 1.3.3 HCN Structure ............................................................................................................. 16 1.3.4 HCN Channel Heteromerization .................................................................................. 19 1.3.5 HCN VoltageDependent Channel Function ............................................................... 19 1.3.5.1 HCN Ionic Nature ..................................................................................................... 20 1.3.5.2 HCN Instantaneous Current ...................................................................................... 21 1.3.5.3 HCN Voltage Sensing and Activation ...................................................................... 23 1.3.5.4 HCN Blockers ........................................................................................................... 24 1.3.6 HCN Cyclic Nucleotide Modulation ........................................................................... 26 1.3.7 Other HCN Modulation ............................................................................................... 27 1.3.8 HCN NLinked Glycosylation ..................................................................................... 29 1.3.9 HCN Mouse Models .................................................................................................... 30 1.3.10 HCN Channelopathies ............................................................................................... 32 1.4 Molecular Evolution ........................................................................................................... 34 1.4.1 Channel Origins and Phylogeny .................................................................................. 34 1.4.2 HCN Channel Evolution .............................................................................................. 35 1. 5 Evolution and Disease........................................................................................................ 37 1.6 Scope of Thesis ................................................................................................................... 38 2. THE EVOLUTION AND STRUCTURAL DIVERSIFICATION OF HYPERPOLARIZATIONACTIVATED CYCLIC NUCLEOTIDEGATED (HCN) CHANNEL GENES .................................................................................................................... 41 2.1 Introduction ......................................................................................................................... 41 2.2 Materials and Methods ........................................................................................................ 43 2.2.1 Sequence Data .............................................................................................................. 43 2.2.2 Multiple Sequence Alignments .................................................................................... 46 2.2.3 Phylogenetic Analyses ................................................................................................. 48  vi 2.3 Results and Discussion ....................................................................................................... 48 2.3.1 HCN Genes are Present in Multiple Copies Across a Wide Spectrum of Species ...... 49 2.3.2 High Sequence Identity Amongst Four Vertebrate HCN Isoforms Within the Core Region ................................................................................................................................... 49 2.3.3 EST Evidence Supports the Validity of Highly Diverged Sequences Identified in Urochordates ......................................................................................................................... 52 2.3.4 Three Different HCN Duplication Events Occurred Prior to the Divergence of the Fish Lineage .................................................................................................................................. 52 2.3.5 Fish Lineage Show Evidence of Duplicate HCN Genes ............................................. 57 2.3.6 Ciona Genes Most Likely Arose Through LineageSpecific Duplication Events ....... 57 2.3.7 Predicted Phylogenetic Patterns are Supported by Exon Boundary Structure ............ 58 2.3.8 The Evolution of Key Residues in the Voltage Sensing Domain and Pore Region .... 62 2.3.9 The Evolution of the Cyclic Nucleotide Binding and Modulatory Domains .............. 68 2.3.10 Sequence Variability and Functional Divergence of Vertebrate HCN Paralogs ....... 71 2.3.11 Vertebrate IsoformSpecific Alignments of Sequences Spanning 450MY, Reveal Conserved Motifs in the N and CTermini .......................................................................... 73 2.3.12 Summary and Perspectives ........................................................................................ 79 3. DUPLICATION AND ASYMMETRIC DIVERSIFICATION OF HCN CHANNEL ISOFORMS IN CIONA INTESTINALIS ................................................................................ 81 3.1 Introduction ......................................................................................................................... 81 3.2 Materials and Methods ........................................................................................................ 83 3.2.1 HCN Sequence Collection and Analysis ..................................................................... 83 3.2.2 Cloning and Epitope Tagging of Two Ciona HCN Channels ..................................... 84 3.2.3 Identification of NGlycosylation of Ciona HCN Channels in Xenopus Oocytes and Chinese Hamster Ovary Cells ............................................................................................... 84 3.2.4 Electrophysiological Analysis of Ciona HCNs in Xenopus Oocytes .......................... 85 3.3 Results ................................................................................................................................. 88 3.3.1 Molecular Cloning ....................................................................................................... 88 3.3.2 Phylogenetic Pattern and Tracking of the Pore NLinked Glycosylation Sequon Suggests Selective Loss of this Modification Among Ciona HCNs .................................... 88 3.3.3 NGlycosylation Status When Expressed in Cells ....................................................... 89 3.3.4 Two Ciona HCNs Form Channels Which are Variably Opened by Hyperpolarization and Blocked by Cesium ........................................................................................................ 92 3.3.5 Permeation of Cations Through Ciona HCNs is VertebrateLike ............................... 95 3.3.6 Slow Opening of ciHCNs is Facilitated by Raising Intracellular Cyclic AMP ......... 100 3.3.7 The Time Course of Slow Opening for ciHCNa is More Complex than ciHCNb .... 102 3.3.8 Fast and Slow Opening Contribute Equally to Ion Flow Through ciHCNb Channels ............................................................................................................................................. 104 3.4 Discussion ......................................................................................................................... 106 4. EVOLUTIONARY ANALYSES OF KCNQ1 AND HERG VOLTAGEGATED POTASSIUM CHANNEL SEQUENCES REVEAL LOCATIONSPECIFIC SUSCEPTIBILITY AND AUGMENTED CHEMICAL SEVERITIES OF ARRHYTHMOGENIC MUTATIONS .................................................................................. 109 4.1 Introduction ....................................................................................................................... 109 4.2 Methods............................................................................................................................. 111 4.2.1 Sequence Alignment .................................................................................................. 111  vii 4.2.2 Collection of ArrhythmiaAssociated Mutations (AAMs) and NonSynonymous Polymorphisms (nsSNPs) ................................................................................................... 111 4.2.3 Phylogenetic Analysis and Determination of Interspecific Variability ..................... 113 4.2.4 Association Between AAMs or nsSNPs and Evolutionarily Conserved Sites .......... 113 4.2.5 Determination of Codon Evolutionary Rate of Change ............................................. 114 4.2.6 Distribution of AAMs Among Functionally Important Regions of the Channels ..... 114 4.2.7 Chemical Severity of Amino Acid Changes .............................................................. 115 4.2.8 Weighted Average for Amino Acid Expected Chemical Severity ............................ 115 4.3 Results ............................................................................................................................... 116 4.3.1 Channel Structure and Mutation Mapping ................................................................. 116 4.3.2 AAMs Occur Preferentially at Sites Conserved Throughout Vertebrate Evolution and at those with Lower Evolutionary Rates of Change ........................................................... 121 4.3.3 Disease Mutations are Not Equally Distributed Among Functional Regions of the Channels .............................................................................................................................. 125 4.3.4 The Chemical Severities of AAMs are Different than Changes Observed Throughout Evolution ............................................................................................................................. 127 4.3.5 Involvement of Specific Amino Acids in Arrhythmogenic Disease ......................... 132 4.4 Discussion ......................................................................................................................... 134 4.5 Conclusion ........................................................................................................................ 137 5. GENERAL DISCUSSION ............................................................................................... 138 5.1 Overview ........................................................................................................................... 138 5.2 Molecular Evolution ......................................................................................................... 138 5.3 HCN Evolution ................................................................................................................. 139 5.3.1 Urochordate and Vertebrate HCN Channels Arose From a Single Ancestor ............ 140 5.3.2 HCN LineageSpecific Duplications ......................................................................... 141 5.3.3 Future Evolutionary Analyses of HCN ...................................................................... 142 5.4 HCN Channels in Ciona intestinalis ................................................................................. 143 5.4.1 ciHCNa Displays a MammalianLike Phenotype ...................................................... 143 5.4.2 The Function of ciHCNb has Diverged ..................................................................... 144 5.4.3 Future Directions for HCN in Ciona intestinalis ....................................................... 145 5.5 Evolution and Disease Susceptibility ................................................................................ 147 5.6 HCN in Health and Disease .............................................................................................. 148 5.7 Summary ........................................................................................................................... 149 REFERENCES .......................................................................................................................... 151 APPENDIX ................................................................................................................................ 178  viii LIST OF TABLES Table 2 1: List of HCN Sequences Used in Analyses .................................................................. 45 Table 4 1 Breakdown of Clinical Phenotype of Disease Mutations Included in Analyses ........ 112 Table 4 2: Disease Mutation Distribution by Channel Region ................................................... 120 Table 4 3 Average Chemical Differences of Amino Acid Changes........................................... 128  ix LIST OF FIGURES Figure 1.1: Voltagegated Ion Channel Superfamily ...................................................................... 3 Figure 1.2: Ribbon Diagram of KcSA Crystal Structure. ............................................................... 6 Figure 1.3: Ventricular Action Potential and ECG. ...................................................................... 10 Figure 1.4: Heart Sinoatrial Node Pacemaker Current ................................................................. 14 Figure 1.5: Crystal Structure of Mouse HCN2 Clinker and CNBD ............................................ 18 Figure 2.1: Schematic Representation of the Overall Topology of HCN Channels. .................... 47 Figure 2.2: High Sequence Conservation Observed Between Vertebrate Isoforms. .................... 51 Figure 2.3: A Maximum Parsimony (MP) Consensus Tree of the HCN Family. ........................ 54 Figure 2.4: Conservation of Exon Structure Reveals Evolutionary Patterns of HCN Genes. ...... 60 Figure 2.5: Sequence Comparison of the Voltage Sensing Domain and Pore Region. ................ 64 Figure 2.6 Sequence Comparison of the Cyclic Nucleotide Binding Domain ............................. 69 Figure 2.7: Individual Isoform Alignments of the Distal NTermini of HCN Channels. ............. 76 Figure 2.8 Individual isoform alignments of the distal Ctermini of HCN channels. .................. 78 Figure 3.1 HCN Glycosylation Arose Prior to the Divergence of Urochordate Lineage and is Found in the Common Ancestor of Urochordates and Vertebrates. ............................................. 90 Figure 3.2 Variable Opening by Hyperpolarization and Cesium Block of Two Ciona HCNs .... 94 Figure 3.3: Potassium Passes Through Both Ciona HCNs and Enhances Current Flow ............. 96 Figure 3.4: Sodium Passes Through Ciona HCNs but Does Not Enhance Current Flow ............ 99 Figure 3.5: Opening of ciHCNs is Facilitated by Rises in Intracellular 8bromo cAMP. .......... 101 Figure 3.6: The Kinetics of Slow Opening is Different Between ciHCNa and ciHCNb ............ 103 Figure 3.7: The ciHCNb Current is Distributed Equally Between Instantaneous and Slow Components ................................................................................................................................ 105 Figure 4.1 Location of ArrhythmiaAssociated Mutations (AAMs) and NonSynonymous Single Nucleotide Polymorphisms (nsSNPs) in Human HERG and KCNQ1 Subunits ........................ 117 Figure 4.2 Cladograms of Vertebrate HERG and KCNQ1 Protein Sequences. ......................... 122 Figure 4.3 ArrhythmiaAssociated Mutations in HERG and KCNQ1 are Overrepresented at Evolutionarily Conserved and Slowly Evolving Sites. ............................................................... 124  x Figure 4.4 ArrhythmiaAssociated Mutations are Unevenly Distributed Among Functionally Conserved Regions of HERG and KCNQ1 Even After Accounting for Total Length and Evolutionary Conservation of Individual Sites Therein. ............................................................ 126 Figure 4.5 ArrhythmiaAssociated Mutation and Interspecific Chemical Severities Correlate with the Expected Chemical Severity of the Reference Codon in HERG and KCNQ1, but Not With Each Other. ................................................................................................................................. 130 Figure 4.6 Mutations Occur Predominantly at Arginine and Glycine Residues. ........................ 133 Figure A.1 : NeighborJoining (NJ) Phylogram of the HCN Family. ........................................ 178 Figure A.2 : Maximum Likelihood Phylogram of the HCN Family. ......................................... 180  xi ABBREVIATIONS  VGF    Voltage
gated ion channel superfamily DNA    Deoxyribonucleic acid Cav    Voltage
gated calcium Nav    Voltage
gated sodium TRP    Transient receptor potential channel Kv    Voltage
gated potassium CNG    Cyclic nucleotide gated HCN    Hyperpolarization
activated cyclic nucleotide M1 or M2   Membrane subunit 1 or 2 TM    Transmembrane KCNQ    Potassium voltage
gated channel, subfamily KQT
like subfamily EAG    Ether
à
go
go KvAP    Archaebacterial voltage
gated K +  channel Kv1.2    Potassium voltage
gated channel, also known as KCNA2 KcSA    K +  channel from Streptomyces lividans VSD    Voltage sensing domain KCNH    Potassium voltage
gated channel, subfamily H HERG    Human ether
à
go
go related gene KvLQT1 Voltage
gated potassium channel associated with LQT1, also known as KCNQ1 LQTS    Long
QT Syndrome QT    QT interval on an electrocardiogram ECG    Electrocardiogram KAT1    Inward
rectifying potassium channel from Arabidopsis thaliana AKT1    Inward
rectifying potassium channel from Arabidopsis thaliana cGMP    Cyclic guanosine monophosphate cAMP    Cyclic adenosine monophosphate CNBD    Cyclic nucleotide binding domain Ih    Hyperpolarization current If    Funny current SA Node   Sino
atrial node PBC    Phosphate binding cassette Iinst    Instantaneous current VIC    Voltage
independent current V1/2    Half
activation threshold  τ (act)    Time constant of activation MYA    Million years ago AAM    Arrhythmia
associated mutations DAM    Disease
associated mutations nsSNP    Non
synonymous single nucleotide polymorphism mRNA    Messenger ribonucleic acid ZD7288 4
(N
ethyl
N
phenylamino)
1,2
dimethyl
6
(methylamino) pyrimidinium chloride  xii PKA    Protein kinase A PKC    Protein kinase C MIRP1   minK
related peptide ER    Endoplasmic reticulum NJ    Neighbour joining ML    Maximum likelihood MP    Maximum parsimony EST    Expressed sequence tag CAP    Catabolite gene activator protein PCR    Polymerase chain reaction CHO    Chinese hamster ovary cDNA    complementary DNA, synthesized from mRNA PNGase   Peptide: N
Glycosidase F dN/dS    non
synonymous change versus synonymous change TdP    Torsades de Pointes SIDS    Sudden infant death syndrome SUDS    Sudden unexplained death syndrome SNP    Single nucleotide polymorphism   xiii ACKNOWLEDGEMENTS First and foremost, I would like to say thank you to my supervisor, Dr. Eric Accili. This has been an extended journey for both of us and I appreciate your patience. Over the years, your advice, feedback and overall guidance has helped me reach this very important milestone in my research career and has helped pave the way for my future endeavors. I would like to thank my supervisory committee, Dr. Chris Ahern, Dr. Ken Baimbridge, and Dr. Steve Kehl for their support and feedback throughout the years. Finally, I would like to give two special notes of thanks: 1) Dr. Andrew Spencer, who unfortunately passed away before the completion of this thesis, and 2) Dr. Tim Jegla. Together, you played an instrumental role in the development and execution of the third chapter of this thesis. Thank you for your guidance and inspiration.  I would like to say thank you to the many Accili lab members and colleagues that I have been fortunate enough to meet along the way. You have provided me with a lifetime of memories. To all of you that are outside of this academic family, thank you for providing me with an ear and a shoulder when it was needed most and ultimately pushing me to get this done. I am forever grateful.  Thank you to the National Sciences and Engineering Research Council of Canada for my Doctoral funding, as well as the University of British Columbia and the Division of Physiology in the Department of Cellular and Physiological Sciences for financial support.  Lastly, I would like to thank my family. Brian and Ken, you set the bar high but your encouragement and motivation kept me moving forward and helped me reach this goal. Thank you for the inspiration. To my parents, Susan and Peter, thank you for being who you are and for teaching me the true meaning of dedication, love and commitment. I could not have done this without you.  1 1. INTRODUCTION Ion channels are evolutionarily conserved integral macromolecular complexes that play a pivotal role in the normal functioning of cells. These pore forming proteins, strategically positioned across the lipid bilayer of the cell membrane, regulate the movement of specific ions between intra and extracellular environments.  Coordinated movement of ions through these channels leads to changes in the electrical environment that are central to the physiological processes necessary to sustain life, including celltocell communication, secretion and cell proliferation.   In humans, ion channels play a fundamental role in virtually every system and their importance in functional homeostasis is verified by the evergrowing list of syndromes (or “channelopathies”) that are caused by ion channel impairment. Examples include cardiac arrhythmias, epilepsy, migraines, blindness, deafness, pain, diabetes and movement disorders. As new channels are discovered and new roles of existing channels are identified, the direct association of conditions related to their disruption is being unraveled. The ability to combine the knowledge of the structure/function relationship of ion channels with the identification of the evolutionary changes that have been tolerated over time is helping us to better understand human disease. Additionally, discoveries of new channelopathies and diseaseassociated mutations are providing insight into the basic function and vital importance of these macroscopic proteins. 1.1 VoltageGated Ion Channels Voltagegated ion channels are ubiquitously expressed proteins that play critical roles in numerous physiological processes. While our understanding of them has grown immensely over the past two decades, the original experiments leading to their discovery unfolded over a century ago. By the beginning of the 20 th  century ((Hille, 2001) and references therein), it was clear that ion movement across the plasma membrane generated electrical activity; however, the mechanism of transport and the existence of pores was not suggested until the mid1950s. Historical experiments by Hodgkin, Huxley, Katz, Curtis and Cole (Cole and Curtis, 1939; Hodgkin and Huxley, 1952; Hodgkin, et al., 1952; Hodgkin and Katz, 1949) suggested that ion permeation resulted from movement through independent ion channels within the plasma membrane, which was ultimately responsible for generating axonal action potentials. Over the following three decades, patch clamp electrophysiology (Neher and Sakmann, 1976) enabled researchers to measure ionic currents, at the cellular or singlechannel level, in different  2 environments. Functional identification and extensive electrophysiological characterization of numerous different types of currents in several species and tissues slowly emerged. The discovery of ion channels greatly expanded, however, following DNA cloning of the first voltagegated potassium channel primary sequences in the 1980s (Butler, et al., 1989; Papazian, et al., 1987; Pongs, et al., 1988). Two decades of ion channel cloning and characterization experiments, combined with the more recent advancements in ion channel crystallization (Doyle, et al., 1998), have created a clear image of the structural and functional relationships of the voltagegated ion channel superfamily (VGF). With monumental advancements in genomics, ion channel genes are now known to represent approximately 1% of the human genome (Jegla, et al., 2009) and the VGF contains more than 143 members, making it one of the largest groups of signal transduction proteins (Yu and Catterall, 2004) (Figure 1.1). Comprised of sodium (Na + ), calcium (Ca 2+ ), potassium (K + ) and mixed ion channels, the basal underlying function common to these proteins is to provide an ion permeation pathway through a predominantly impermeable lipid bilayer. When activated, or opened, permeable ions flow passively between the cytoplasm and extracellular space, down their electrochemical gradient at a rate of >10 6  ions/second, close to the aqueous diffusion rate (Hille, 2001). Despite their diverse channel characteristics, their abilities to carry different ions, and their independent evolution over time, the VGF possess only four general structural themes (Yu and Catterall, 2004): 1) a single subunit of four homologous domains (Na +  and Ca +  channels); 2) tetrameric proteins that are formed by four individual subunits, each of which contains one homologous domain (K +  voltagedependent outward rectifiers and relatives); 3) two transmembrane domains (K +  inward rectifiers) and 4) twopore channels that contain two pore motifs linked together (Yu and Catterall, 2004). With their origins arising early in evolution in the prokaryotes (Jegla, et al., 2009), the overall conservation of the tertiary structures within the family suggest a strong evolutionary pressure and implicates its fundamental importance to overall voltagegated channel function.  3   Figure 1.1: Voltagegated Ion Channel Superfamily Representation of the amino acid sequence relations of the pore regions of the voltagegated ion channel superfamily. This global view of the 143 members of the structurally related ion channel genes highlights seven groups of ion channel families and their membrane topologies. Four domain channels (CaV and NaV) are shown as blue branches; potassiumselective channels are shown as red branches, cyclic nucleotidegated channels are shown as magenta branches; and transient receptor potential (TRP) and related channels are shown as green branches. Background colors separate the ion channel proteins into related groups: lightblue (CaV and NaV); lightgreen (TRP channels); lightred (potassium channels, except KV10–12, which have a cyclic nucleotide– binding domain and are more closely related to CNG and HCN channels); lightorange (KV10– 12 channels and cyclic nucleotide–modulated CNG and HCN channels). Minimal pore regions bounded by the transmembrane segments M1 or S5 and M2 or S6 of 143 ion channel members were aligned. The pore regions of the fourth homologous domain of NaV and CaV channels, the second domain of TPC, and the first pore regions of the K2P channels were used to assemble the alignment. Bootstrap values from 1000 replicates of neighborjoining analysis are shown by the colors of the branches that represent each ion channel family: colored lines (bootstrap values >50%); black lines (bootstrap values <50%). The scale bar represents the tree distance corresponding to 0.05 substitutions per site in the sequence. Adapted from (Yu and Catterall, 2004).  4 1.2 VoltageGated Potassium Channel Family 1.2.1 Overview One subfamily of the VGF is the voltagegated potassium (Kv) channels. By providing a pathway for the efflux of K + , Kv channels play a major role in the maintenance of resting membrane potentials, the termination of action potentials in excitable cells, and K +  recycling. Their physiological significance is illuminated by their ubiquitous expression in most cell types and by the fact that at least one K +  channel gene has been found in every genome fully sequenced to date, regardless of its eukaryotic, eubacterial, or archael origin (Hille, 2001)((Miller, 2000) and references therein). The Kv channel family is expansive and includes at least 22 different mammalian genes encoding for voltagegated 6TM proteins (Yellen, 2002). These include the fastdelayed rectifiers (i.e. Shaker or Kv14) and slowdelayed rectifiers (i.e. KCNQ1 and EAG). Like all voltagegated ion channels, the Kv channels have three main functions: 1) sensing changes in transmembrane voltage, 2) opening and closing to enable and regulate the passage of ions (gating), and 3) when open, providing a relatively selective pathway for the rapid movement of ions across an otherwise impermeable lipid bilayer (ion permeation) (Yellen, 2002).  1.2.2 Kv Structure/Function Originally, voltagegated potassium channels were cloned from the Shaker locus of the Drosophila melanogaster (Butler, et al., 1989; Papazian, et al., 1987; Pongs, et al., 1988). Shown to be a fastinactivating Kv channel, Shaker is now the beststudied member of the Kv family, and the prototype of all voltagegated ion channels. The primary amino acid sequence was reflective of six putative transmembrane (TM) domains, with the N and Cterminal portions of the channel predicted to be located on the cytoplasmic surface of the membrane. The fourth TM helix (S4) contains lysine and arginine residues at every third position, and a potassium channel ‘signature sequence’ of eight amino acids (TXXTVGYG) located between the fifth and sixth helices (S5 and S6) (Heginbotham, et al., 1994). Mutations in this region resulted in channels being nonselective to monovalent cations, thus identifying it as the potassium channel selectivity filter (Heginbotham, et al., 1994). Kv channels are formed by the assembly of four individual subunits, identical or similar, arranged in fourfold symmetry around an aqueous ionconducting pore. The subunit  5 stoichiometry and tetrameric assembly was originally inferred from sitedirected mutagenesis experiments and interactions with scorpion toxins (MacKinnon, 1991). Later, these findings were confirmed by the crystal structure of a bacterial 2TM K +  channel from Streptomyces lividans (KcSA) (Doyle, et al., 1998) (Figure 1.2). Based on this original crystal structure, which shares high sequence identity with all known Kv channels (Doyle, et al., 1998), the Kv channel pore region was predicted to consist of an inner and outer helix from each subunit (referred to as M2 and M1 in 2TM channels or S6 and S5 in 6TM channels, respectively). The four M2 helices form an inverted teepee in fourfold symmetry. The assumption that the 6TM Kv channels are similar has been more recently confirmed by the crystal structure of KvAP and Kv1.2 (Jiang, et al., 2003a; Long, et al., 2005a). A reentrant pore loop between the inner and outer helices contains an outer turret, a pore helix and the potassium selectivity signature sequence.  6     Figure 1.2: Ribbon Diagram of KcSA Crystal Structure. Ribbon representation of the KcsA tetramer viewed from the A) extracellular side or B) perpendicular to the membrane. The four subunits are distinguished by colour. Adapted from (Doyle, et al., 1998).  7 Together, the crystal structures of various K +  channels have revealed that the ion permeation pathway consists of an inner waterfilled vestibule and a narrow outer region containing the selectivity filter. K +  pass through this filter in a single file and alternate through four binding sites, with two sites being occupied at any given time (Doyle, et al., 1998; Jiang, et al., 2002a; Jiang, et al., 2002b).  Kv channels open in response to changes in transmembrane voltage, a mechanism referred to as activation or gating. The structure that moves to produce the openings, or that closes to block the ion permeation pathway, is referred to as the activation gate. By comparing the structures of open (Jiang, et al., 2002a; Jiang, et al., 2002b; Jiang, et al., 2003a) and closed (Doyle, et al., 1998) state channels, an intracellular gate of Kv channels was proposed to be formed by a bundle crossing of the four S6 transmembrane helices. When closed, ion movement from the cytoplasm into the aqueous vestibule is prevented (Liu, et al., 1997). A bending at a ‘hinge’ located in the lower regions of the S6 helices is believed to lead to gate opening. The gating hinge in prokaryotic channels is produced by a glycine residue in the lower portion of M2. In the mammalian Kv1.2, however, the bend is produced by a prolinevalineproline sequence and the hinge mechanism is suggested to be conserved (Long, et al., 2005a). The voltage sensing domain (VSD) of the 6TM Kv channels is formed by the S1S4 region. Charged lysine or arginine residues, located every three positions in S4, a feature which is conserved throughout the VGF, is termed the voltage sensor (Yellen, 2002). In response to changes in membrane voltage, the sensor moves and transfers charges across the electric field, leading to an activatednotopen state of the channel. In turn, these movements result in a mechanical force and conformational change that causes the channel to switch to an open state. While a consensus has not been reached for the exact mechanism involved, it has been modeled as a transporter (with relatively little movement of 23 angstroms combined with helical rotation, tilt, and a reshaping of the electric field), a helical screw (with movement of 513 angstroms and rotation), or a paddle (with large 1520 angstrom movement across the membrane) (Tombola, et al., 2006) and (Bezanilla, 2008) and references therein). For the voltage sensor to exert an opening effect on the pore region of the channel, it must be linked or coupled to the activation gate. The crystal structure of the mammalian Kv1.2 first shed light on the relationship between the VSD module and that of the pore region in mammalian channels (Long, et al., 2005a; Long, et al., 2005b). Surprisingly, the four VSDs were  8 found to be positioned outside of the pore region of an adjacent subunit, but linked to their intrasubunit pore region via their S4S5 linkers, alphahelices running parallel to and along the cytoplasmic side of the plasma membrane. Thus, the VSD may be able to facilitate both inter and intrasubunit interactions in response to voltage (Tombola, et al., 2006), and specifically, because the linker region crosses over the Cterminal end of its own S6 helix, a mechanical push could be generated and contribute to the gating process (Long, et al., 2005b).  1.2.3 Kv Channelopathies Diseaseassociated genetic mutations identified in ion channels have illuminated the pivotal relationship between the protein structure and function (Ashcroft, 2006). The increasing number of rare diseases that are now linked to an underlying channelopathy reveals the importance of ion channels in the homeostasis of regular physiological function and human health. To date, over 60 ion channel genes are known to cause human disease (Ashcroft, 2006). For example, the importance of Kv channels was delineated by the discovery of inherited mutations causing ventricular cardiac arrhythmias and sudden death (Curran, et al., 1995; Wang, et al., 1996). In general, the spectrum of channel disruptions caused by a genetic mutation is broad, but can involve: 1) misfolded proteins that are retained in the endoplasmic reticulum, and therefore do not reach the cell surface to perform their normal function (trafficking defective), 2) gating disruptions, 3) disruptions to the voltage dependence of activation or inactivation, or 4) changes in ion selectivity. One of the regular physiological functions of Kv channels is to provide a pathway for the efflux of potassium at the termination of cardiac action potentials. The ventricular action potential occurs in multiple stages: Phase 0 (upstroke), Phase 1 (early repolarization), Phase 2 (plateau), Phase 3 (repolarization) and Phase 4 (resting membrane) (Figure 1.3). The repolarization phase results in the termination of the action potential and is a direct result of potassium efflux through the Kv delayed rectifier channels, IKs and IKr, the alpha subunits which are encoded by KCNQ1 (KVLQT1) and KCNH2 (HERG), respectively. Long QT Syndrome (LQTS) is due to a disruption in this repolarization of ventricular myocytes. A prolongation of the repolarization phase predisposes individuals to arrhythmia, syncope and sudden death. LQTS is diagnosed and defined by a prolongation of the QT interval on an electrocardiogram (ECG), which directly corresponds to Phase 3 repolarization of the  9 ventricles. Prolongation and variability of repolarization can lead to a lethal cardiac arrhythmia known as Torsades de Pointes and possible ventricular fibrillation. Recent studies suggest that LQTS affects 1 in 2500 individuals (Schwartz, et al., 2009) and to date, 12 genes have been implicated in the pathogenesis of the disease (Goldenberg, et al., 2008). Lossoffunction mutations found in the Kv delayed rectifier genes, KCNQ1 and KCNH2, together comprise the majority of these diseasecausing mutations, with more than 200 variants being found to date.  10      Figure 1.3: Ventricular Action Potential and ECG. The body surface electrocardiogram (ECG) can detect electrical gradients in the myocardium and is a reflection of the underlying cellular ionic current gradients. A: illustration of a single cardiac cycle ECG detected as electrical gradients on the body surface. B: schematic representation of the ventricular action potential gradients detected on the body surface ECG. C: ionic currents responsible for the different phases of the action potential and the genes that encode for them. Darkened shapes are indicative of relative current amplitude, duration, and direction. The shape of the current is also aligned with its approximate time of action during the ECG (A) and the cardiac ventricular action potential (B).  Adapted from (Clancy and Kass, 2005).  11  1.2.4 Kv Channel Relatives Three families of ion channels that are structurally and evolutionarily related to the Kv channels are the 6TM cyclic nucleotide gated channels (CNG), the 6TM potassium inward rectifiers from plants, KAT1 and AKT1 and the hyperpolarization activated cyclic nucleotide gated channel, HCN (described in the Section 1.3).  1.2.4.1 CNG Channels CNG channels, originally identified in the retinal rod cells and the cilia of olfactory receptors (Fesenko, et al., 1985; Nakamura and Gold, 1987), are vital components of the sensory transduction pathway in the visual and olfactory systems (Zagotta and Siegelbaum, 1996). Additionally, they have been identified in the testis, kidney, colon, pancreas, hippocampus, heart and adrenal gland, though their functional role in these tissues remains to be established (Kaupp and Seifert, 2002). These nonselective cation channels (Goulding, et al., 1992; Kaupp, et al., 1989) open in response to direct binding of cyclic nucleotides, including both cyclic guanosine monophosphate (cGMP) and cyclic adenosine monophosphate (cAMP), but show very little response to membrane voltage, despite the retention of the voltage sensing domain in vertebrate and invertebrate isoforms. Cyclic nucleotides bind to a conserved cyclic nucleotide binding domain (CNBD), which causes an increased probability of opening through the increased stabilization of the open state of the channel (Goulding, et al., 1994). Under physiological conditions, the channels predominantly pass inward Na +  and Ca 2+  currents (Craven and Zagotta, 2006). Similar to the Kv channels, mutations in CNG channels have been linked to human channelopathies, including retinitis pigmentosa and achromatopsia (Ashcroft, 2006).  1.2.4.2 KAT1 and AKT1 Channels Two other interesting relatives of Shaker Kv channels have been identified and cloned from the plant species Arabidopsis thaliana, called AKT1 (Sentenac, et al., 1992) and KAT1 (Anderson, et al., 1992). Functional expression produced hyperpolarizationactivated K +  selective channels, which were characterized and shown to encode inwardrectifying channels (Gaymard, et al., 1996; Schachtman, et al., 1992). Like the other Kv channels, membrane topology was predicted, and later shown (Uozumi, et al., 1998), to contain six transmembrane  12 domains. Nevertheless, similar to the CNG channels, a putative CNBD was located in the cytoplasmic Cterminus of the protein. Interestingly, Cterminal deletions including the CNBD, but not more distal deletions, abolished KAT1 channel function (Marten and Hoshi, 1997), implicating the domain’s involvement in overall channel function. Functionally, KAT1 and AKT1 channels open slowly upon hyperpolarization with a multiexponential time course, are highly selective for potassium and have been shown to behave as inwardrectifying channels (Schachtman, et al., 1992) that can be blocked by cesium (Cs + ) in the pore (Becker, et al., 1996). Furthermore, they have been shown to be sensitive to cGMP (Gaymard, et al., 1996). Interestingly, while these structural relatives of the 6TM Shaker channel behave as inward rectifiers, the rectification is not due to intracellular polyamine block (Hoshi, 1995), as is the case for the 2TM inward rectifiers. Single channel recordings of KAT1 channels in Xenopus oocytes further demonstrated that the rectification property was due to an intrinsic gating mechanism (Zei and Aldrich, 1998). Similar to whole cell experiments, single channel experiments demonstrated that KAT1 channels activate slowly after a noticeable delay, suggesting multiple closed states within the activation pathway (Zei and Aldrich, 1998). Furthermore, the delay was reduced at more negative potentials. Results from gating current measurements, together with the observed delay prior to the onset of ionic current, suggest that the channels transition through multiple closed states prior to channel opening (Latorre, et al., 2003).  Lastly, one of the most interesting aspects of the KAT1 channel is related to the S4 movement. External accessibility experiments of KAT1 channels, similar to those for HCN channels described in Section 1.3.5.3, demonstrate that the S4 voltage sensor moves inward upon hyperpolarization, the same direction as in depolarizationactivated Shaker channels. Nevertheless, while the inward motion in the latter results in channel closure, in KAT1, the same voltage sensor movement causes the channels to open. The difference between Shaker and KAT1 channels, therefore, must be due to differences found within the region that couples voltage sensing to the movement of the activation gate (Latorre, et al., 2003).  13 1.3 HCN Channels 1.3.1 Funny Current (If) The hyperpolarizationactivated current, referred to as Ih (for hyperpolarization) or If (for funny), was first characterized 30 years ago in cardiac conduction tissue (Brown, et al., 1979), and has since been shown to play a vital role in both excitable and nonexcitable cells. Biophysical features were uncharacteristic of other known currents, as they were found to be: 1) a mixed cation current, carried by both sodium and potassium ions; and 2) activated slowly upon hyperpolarization at negative potentials (Brown and Difrancesco, 1980). Found in the Purkinje fibers (DiFrancesco, 1981b; DiFrancesco and Ojeda, 1980) and the sinoatrial node (DiFrancesco, et al., 1986; Yanagihara and Irisawa, 1980) of the mammalian heart, as well as the central and peripheral nervous systems (McCormick and Pape, 1990; Pape, 1996; Takigawa, et al., 1998) and the retinal photoreceptors (Attwell and Wilson, 1980; Bader, et al., 1982; Bader, et al., 1979), the unique features of these currents suggested their role in contributing to the depolarization phase of the pacemaker potential. In the cardiovascular system, spontaneously active pacemaker cells contribute to the generation of the sinus rhythm and rate (DiFrancesco, 1991; DiFrancesco, 1993). In the nervous system, they have been shown to contribute to the oscillation rate or the synchronization of neuronal firing (Pape, 1996). In both cases, the rhythmic firing or spontaneous activity of these cells is possible due to the activation of this mixed cation current upon plasma membrane hyperpolarization. The resultant net inward flow of Na +  ions ultimately leads to membrane depolarization towards the activation threshold of other voltagegated ion channels (Ludwig, et al., 1999a) (Figure 1.4).  14       Figure 1.4: Heart Sinoatrial Node Pacemaker Current Sinoatrial node (black oval) of the heart (right) and the ion channels responsible for pacemaker activity (left). Each channel contributes to a phase of the sinoatrial action potential, indicated by the dotted arrows.  Adapted from (Craven and Zagotta, 2006).  15 In the thalamocortical neurons (McCormick and Pape, 1990), and at physiological resting conditions of nonpacing cells, If currents have been shown to facilitate the regulation of resting membrane properties and limit the extent to which membrane potentials can fluctuate (Pape, 1996). Unique to If at the time, was the apparent dual activation by both voltage and autonomic inputs. Before their molecular identity was elucidated, it was clear that these proteins were under the direct control of autonomic stimulation. Currents were modulated in opposite ways by beta adrenergic and cholinergic inputs via stimulation or inhibition of adenylate cyclase and the resultant effect on the production of cAMP (Brown, et al., 1979; DiFrancesco, 1993; DiFrancesco and Tromba, 1988). Adrenaline, applied to single sinoatrial node cells, demonstrated an increase of If near the halfactivation voltage (V1/2), with an acceleration of kinetics as well as an indirect enhancement of current amplitude due to a positive shift in the voltagedependence of activation (DiFrancesco, 1986; DiFrancesco and Mangoni, 1994). Experiments in excised patches confirmed these findings and showed direct binding of cAMP to the channels that carried If and caused an increase in their open probability (DiFrancesco and Mangoni, 1994; DiFrancesco and Tortora, 1991). In general, the positive shift in voltage dependence of activation leads to more channels available to contribute to the pacemaker potential at any given voltage and the increased current, in turn, results in an increased in firing frequency of the SA node. 1.3.2 HCN Mammalian Tissue Expression The molecular identity of the channels underlying the pacemaker current was originally determined from mammalian species and sea urchin in the late1990s (Gauss, et al., 1998; Ludwig, et al., 1998; Santoro, et al., 1997; Santoro, et al., 1998). The four mammalian HCN isoforms (HCN1HCN4) display partial overlapping tissue distribution in both the heart (Ishii, et al., 1999; Mistrik, et al., 2005; Moosmang, et al., 2001; Shi, et al., 1999) and the central nervous system (Ludwig, et al., 1999a). In rabbit and mouse hearts, HCN4 is the predominant isoform, expressed in the sinoatrial (SA) node, and accounts for approximately 80% of the total HCN mRNA, with the remaining comprised of HCN1 and HCN2 in a speciesspecific manner (Brioschi, et al., 2009; Ishii, et al., 1999; Moosmang, et al., 2001; Peters, et al., 2009; Shi, et al., 1999). HCN4 has been proposed to be a morphological marker of pacemaker tissue (Brioschi, et  16 al., 2009). HCN2 has been identified in mouse atria and ventricles (Moosmang, et al., 2001), along with HCN3 (Mistrik, et al., 2005), albeit the latter is rarely reported. To further complicate these profiles, expression levels fluctuate with species, age and development (Schweizer, et al., 2009; Shi, et al., 1999) (Stieber, et al., 2003a). In the nervous system, all four isoforms are clearly expressed (Ludwig, et al., 1999b): HCN1 expression has been shown in the hippocampus, thalamus and brain stem (Ludwig, et al., 1998), and in the cortex, and cerebellum (Craven and Zagotta, 2006); HCN2 is thought to be ubiquitously expressed throughout the central nervous system (Santoro, et al., 2000), including the hippocampus, superior colliculus, cerebral cortex, cerebellum (Ludwig, et al., 1998), and basal ganglia (Craven and Zagotta, 2006); and both HCN3 and HCN4 expression have been shown in the olfactory bulb (Mistrik, et al., 2005; Santoro, et al., 2000), and the hypothalamus (Mistrik, et al., 2005)  and thalamus (Bender, et al., 2001; Santoro, et al., 2000), respectively. In addition to the heart and central nervous system, HCN channels have been identified in dorsal root ganglia (Moosmang, et al., 2001), pancreas (Zhang, et al., 2009b), enteric nervous system (Xiao, et al., 2004), retinal photoreceptors (Demontis, et al., 2002; Moosmang, et al., 2001), taste receptors (Stevens, et al., 2001), and pain receptors and mechanosensitive neurons (Biel, et al., 2009). 1.3.3 HCN Structure Based on sequence similarity with their CNG and Kv channel (EAG) potassium channel relatives, HCN subunits are predicted to have six transmembrane segments (S1S6) with cytoplasmic amino (N) and carboxyl (C) termini. They possess a voltage sensor domain in S1 S4 and an ion conducting pore between S5 and S6. Together, the crystal structures of the KcSA, KvAP, and Kv1.2 potassium channels (Doyle, et al., 1998; Jiang, et al., 2003a; Long, et al., 2005a) provide insight into the putative tertiary and quaternary structure of HCN channels. Analogous to these proteins, four HCN subunits are proposed to come together to form tetramers around a central pore. Furthermore, the HCN channels appear to exhibit the socalled potassium channel selectivity filter signature sequence (GY/FG) within the putative pore, which is associated with channels that are highly selective for potassium. Lastly, similar to the CNG, EAG, and KAT1 channels, the Cterminus of HCN channels contains a 120 amino acid CNBD.  17 From the original crystal structure of the HCN2 Cterminal region (Zagotta, et al., 2003), solved in the presence of cAMP, an 80amino acid Clinker region was shown to be composed of six helices (A’F’), joining the transmembrane region to a CNBD (Figure 1.5). The CNBD structure is similar to that of other CNBDs, such as the bacterial cataboliteactivating protein and protein kinase A (Berman, et al., 2005), suggesting that the tertiary structure of this domain has been under selective pressure and conserved throughout evolution. The HCN2 CNBD contains four alpha helices and an eightstranded betaroll, arranged with two helices flanking the beta roll (A and B helix) and a third distal helix (C helix). A fourth helix is found between β6 and β7, in a region known as the phosphatebinding cassette (PBC). Cyclic AMP binds to the PBC through electrostatic and hydrophobic interactions and hydrogen bonding and is further stabilized through interactions with the Chelix. Cyclic nucleotides bind in this pocket and are thought to have an allosteric effect on channel opening and closing.  18      Figure 1.5: Crystal Structure of Mouse HCN2 Clinker and CNBD a) Ribbon representation of a single protomer of HCN2J bound with cAMP. b) HCN2 tetramer viewed perpendicular (left) and parallel (right) to the fourfold axis. Each subunit is shown in a different colour. Adapted from (Zagotta, et al., 2003)  19  HCN2 intrasubunit interactions were shown to occur between the Clinker and the Broll of the CNBD, whereas, HCN2 intersubunit interactions were shown to only involve the neighbouring Clinkers (Craven and Zagotta, 2004). More recent crystal structures of the spHCN (Flynn, et al., 2007) and human HCN4 (Xu, et al., 2010) CNBDs, further support the predicted structural motif of fourfold symmetry, conservation of the overall structure of the Cterminal domains, and these intersubunit interactions between neighbouring subunits (Flynn, et al., 2007). 1.3.4 HCN Channel Heteromerization Similar to Kv channels, functional HCN channels are produced by four individual alpha subunits assembling in fourfold symmetry. Nevertheless, because native If current properties have been shown to be intermediate to any of the individual HCN isoforms expressed in heterologous systems, this finding was hypothesized to be due to heteromerization of different isoforms. Concatenated channels, and those formed by coinjection of individual isoforms, produced biophysical properties that were intermediate to those of the homomeric channels alone. Furthermore, the intermediate properties could not be explained by the algebraic sum of independent populations of homomeric channels, however, this could not be conclusively ruled out as a possibility (Altomare, et al., 2003; Chen, et al., 2001b; Ulens and Tytgat, 2001a). In fact, HCN2 and HCN4 channels were recently shown to display equal preference for assembly as heteromultimers and homomultimers (Whitaker, et al., 2007). Furthermore, while almost all dimeric isoform combinations have been shown to produce functional channels in vitro (Much, et al., 2003), the physiological importance of this assembly process has been demonstrated by the finding of heteromeric channels in vivo: in the embryonic mouse heart and rat thalamus (Whitaker, et al., 2007), the brain (Much, et al., 2003), and more specifically in hippocampal tissue following the induction of seizure activity (Brewster, et al., 2005). 1.3.5 HCN VoltageDependent Channel Function In general, HCN channels expressed in heterologous systems display all of the archetypal If biophysical properties (Reviewed in (Biel, et al., 2002; Biel, et al., 2009; Robinson and Siegelbaum, 2003; WahlSchott and Biel, 2009)). The channels are activated upon membrane hyperpolarization and close upon depolarization to more positive potentials. They produce slow timedependent inward currents that are activated by voltages more negative than 40 mV.  20 Although they possess the K +  signature sequence, they carry a mixed cation current permeable to a significant amount of both Na +  and K + . The channels are blocked by millimolar concentrations of extracellular Cs +  and can be modulated by cyclic nucleotides. Presumably due to slight sequence differences between them, the kinetics of opening and closing and the modulatory properties of the four isoforms differ quantitatively (Baruscotti, et al., 2005): HCN1 activates the most quickly and has the most positive voltage dependence of activation, but displays minimal response to cAMP (Santoro, et al., 2000; Santoro, et al., 1998); conversely, HCN2 and HCN4 activate more slowly and are strongly modulated by cAMP (Ishii, et al., 1999; Ludwig, et al., 1998; Ludwig, et al., 1999b; Santoro, et al., 2000; Seifert, et al., 1999). 1.3.5.1 HCN Ionic Nature As shown for the If currents (Bader and Bertrand, 1984; Bader, et al., 1982; DiFrancesco, 1981b; DiFrancesco, 1982; DiFrancesco, et al., 1986; Edman, et al., 1987), fullyactivated HCN channels, expressed in heterologous expression systems, produce inward currents with reversal potentials of ~20mV in physiological solutions (Moroni, et al., 2000). Dependent on both external Na +  and K +  concentrations, fluctuations in reversal potentials suggest that both are contributing to the observed current. Using the GoldmanHodgkinKatz equation to calculate relative permeabilities, the PNa/PK ratio for HCN2 was shown to be 0.41, indicating a selective preference for K +  (Moroni, et al., 2000). Under physiological conditions, however, Na +  ions experience a much larger driving force (Edman, et al., 1987) and therefore, predominate as the ion contributing to the inward currents. Along with the low single channel conductance (<1pS), the relatively high sodium permeability is unusual in the normally highly K + selective ion channels that contain the GY/FG motif in the pore.  In addition to K +  and Na +  ions, other mono and divalent cations display limited permeation through HCN channels. They are slightly permeable to Li +  with a permeability ratio for PLi/PK of ~0.02 (DiFrancesco, 1982). Early experiments with If suggested that the channels were modulated by external calcium (Hagiwara and Irisawa, 1989) but that Ca 2+  did not directly contribute to the current. More recent studies in heterologous expression systems and myocytes, however, have shown that Ca 2+  can, in fact, permeate HCN2 and HCN4 channels, contributing to less than 1% of the total current (Michels, et al., 2008; Yu, et al., 2007; Yu, et al., 2004b).  21 A second feature of HCN and If currents is the K + induced channel activation (DiFrancesco, 1981b). An increase in the external potassium concentration not only causes the expected shift in the reversal potentials in a positive direction, but also augments the channel conductance. This was shown to be due to a depolarizing shift in the V1/2 and a deceleration of channel kinetics (Azene, et al., 2003). Residues in and near the HCN selectivity filter have been found to contribute to this gating effect (Azene, et al., 2005; Azene, et al., 2003). Lastly, an archetypal If channel property, and one similar to KAT1 channels, is the concentration and voltagedependent block by low concentrations of extracellular Cs + . The ability of Cs +  to block If increases at more negative potentials (DiFrancesco, 1982; DiFrancesco, et al., 1986). Preliminary work suggested a blocking site that was accessible to external Cs +  ions and located approximately ¾ of the way across the electric field (DiFrancesco, 1982; Wollmuth and Hille, 1992), indicating that external Cs +  ions entered the channel pore. HCN channels have been shown to display a similar voltagedependent block by extracellular Cs +  (Moroni, et al., 2000), and based on the Woodhull’s block model, the site is located on the internal side of the selectivity filter in the central cavity and approximately 66% of the way across the electric field (δ = 0.659). 1.3.5.2 HCN Instantaneous Current Upon hyperpolarization, HCN channels display two distinct current profiles: a time independent instantaneous current (Iinst) also referred to as voltageindependent current (VIC), and a slowactivating, sigmoidal timedependent current (If), both of which increase with greater hyperpolarization due to increases in the driving force. The slow If currents activate and deactivate following an exponential time course, with the tau of activation peaking at voltages corresponding to the V1/2 (Moroni, et al., 2000). To date, the kinetics have been described by both single (Ludwig, et al., 1998; Santoro, et al., 1998) (Ishii, et al., 2001) (Altomare, et al., 2001) and double (Santoro, et al., 2000) (Moroni, et al., 2000) exponentials. The fitting of activation kinetics; however, consistently begins immediately following an initial period of delay (Altomare, et al., 2001; Santoro, et al., 2000), a characteristic that was originally defined for If currents (DiFrancesco, 1984; Edman, et al., 1987), and which was previously described for KAT1 channels. In both scenarios, the delay is reduced at more hyperpolarized potentials. For  22 the mammalian isoforms, HCN1 and HCN4 channels display shorter delays, whereas HCN2 has the most pronounced delay (Altomare, et al., 2001). The current delay has been suggested to indicate that the channel moves through multiple closed states before channel opening. More recently, the delay observed in macroscopic currents has also been seen in multichannel patch recordings of single HCN2 channels, and reported to be the latency to first opening (Dekker and Yellen, 2006). A lag time of 200ms was reported to occur at 120mV prior to any channel opening. Interestingly, in addition to that observed at more negative potentials, the delay was shown to be reduced by preconditioning steps (Altomare, et al., 2001; DiFrancesco, 1984), thus supporting the notion that prehyperpolarization primes the channel for opening prior to activation (i.e. moves a closed channel towards the open state) (Altomare, et al., 2001; Edman, et al., 1987). Instantaneous currents (Iinst) or voltageindependent current (VIC) have been described previously for both mammalian and echinoderm HCN channels (Chen, et al., 2000; Chen, et al., 2001a; Gauss, et al., 1998; Ishii, et al., 1999; Proenza, et al., 2002a; Proenza and Yellen, 2006). In addition to the slowactivating If currents, mammalian HCN channels produce a small Iinst that is only a fraction of the total ion flow. It has been shown to be associated with the channel surface expression, display similar reversal potentials and correlate to the size of the If current. It was not, however, found to be sensitive to extracellular Cs +  block (Macri and Accili, 2004; Macri, et al., 2002; Proenza, et al., 2002a), suggesting that Iinst in this scenario is flowing through a leaky Cs + insensitive open state of the channel. In a more recent report; however, the instantaneous current (VIC) could be blocked by Cs +  and an HCNspecific pore blocker, suggesting that ion permeation was, in fact, through the same pore as the voltagedependent current (Proenza and Yellen, 2006). These authors concluded that VIC represents a nonzero limiting open probability for HCN channels at positive voltages (Proenza and Yellen, 2006) due to inadequate channel closure. This finding was supported by: 1) a recent alanine scan of the pore region showing a weak interaction between the voltage sensor and the pore resulting in the open probability not reaching zero (Macri, et al., 2009); and 2) the observation that Iinst increased with consecutive hyperpolarizations, an effect that was enhanced in isoforms with slow kinetics (HCN3 and HCN4) and negligible in those with faster kinetics (HCN1 and HCN2) (Mistrik, et al., 2006).  23 1.3.5.3 HCN Voltage Sensing and Activation Similar to Kv channels, HCN channels contain a conserved motif in S4 that is part of the voltage sensor, the movement of which begins the process of channel opening. In HCN channels, S4 consists of 8 or 9 regularlyspaced basic residues, separated into inner and outer halves by a conserved serine residue. Neutralization of the Nterminal basic residues produced channels with a negative shift in their voltage dependence of activation, making the channel more difficult to open, and the effects of each mutation were additive (Chen, et al., 2000). Mutation of the serine residue created a channel that produced predominantly instantaneous currents and very little timedependent current, suggesting that the channel was stuck in a partially opened state. Mutations in the basic residues, located in the inner half of the S4 domain, did not express and were intolerant to change, supporting that they play a vital role in channel function. The mechanism of voltage sensing involves the basic charges moving in the focused electric field of the membrane in response to applied changes in voltage. Somewhat surprisingly, albeit similar to that found for KAT1 channels, solvent accessibility studies demonstrated that the direction of voltage sensor movement is conserved among ion channels. Depolarization drives the sensor outwards whereas hyperpolarization retracts the sensor either due to translational movement (Mannikko, et al., 2002) (Vemana, et al., 2004) or via changes in the shape of the electric field (Bell, et al., 2004). Again, analogous to the KAT1 channels, the way in which the voltage sensor domain is coupled to the activation gate is presumed to enable the channels to open in response to hyperpolarization opposed to depolarization (Bell, et al., 2004; Mannikko, et al., 2002). The S4S5 linker plays a critical role in coupling the voltage sensing to the activation of the intracellular gate in S6 (Chen, et al., 2001a; Macri and Accili, 2004). Mutational analysis of the linker region demonstrated that changes produced large instantaneous currents and led to a disruption in channel closure. Analyses demonstrated that these currents displayed Iflike properties, with a reversal potential that varied with extracellular ion concentrations and was blocked by extracellular Cs + . Furthermore, these mutations caused the Iflike instantaneous current to represent over 70% of the total current. The channel activated at more depolarized potentials and the activation curve was less steep. The S4S5 mutation also caused the deactivation to be significantly slower and was thought to be due to the mutation inhibiting channel closure (Macri and Accili, 2004).  24 Based on findings for Kv HERG channels, where the S4S5 linker is involved in coupling the voltage sensor movement and the gating, through electrostatic interactions with the Clinker (Sanguinetti and Xu, 1999; TristaniFirouzi, et al., 2002), similar interactions were found to stabilize the closed state in mammalian HCN channels (Decher, et al., 2004). Furthermore, cross linking studies between the S4S5 linker and the Clinker regions of spHCN channels confirm the conserved nature of these interactions throughout the HCN family, and suggest that the interactions between these two domains ultimately dictate the polarity of voltage dependence in ion channels (Prole and Yellen, 2006). Similar to depolarizationactivated potassium channels, the activation gate is formed by the intracellular regions of the S6 transmembrane domain (Rothberg, et al., 2002; Rothberg, et al., 2003; Shin, et al., 2001). When the channels are closed, intracellular ions and HCNspecific blockers applied from the intracellular side of the membrane have limited access to the channel pore (Rothberg, et al., 2002; Shin, et al., 2001). The inactivation process observed in spHCN channels, and not in any other HCN channel cloned to date, is thought to be due to a reclosure of this main intracellular activation gate (Shin, et al., 2004). 1.3.5.4 HCN Blockers ZD7288 (4(NethylNphenylamino)1,2dimethyl6(methylamino) pyrimidinium chloride), is an Ifspecific blocker originally shown to slow heart rate in vivo without affecting the force of contraction (Marshall, et al., 1993). Thus, it was classified as a ‘selective bradycardic agent’. ZD7288 was shown to block If in isolated sinoatrial cells in a concentrationdependent manner, by causing a negative shift in the activation curve and by reducing the amplitude of the activation curve without affecting ion selectivity (BoSmith, et al., 1993). Block, due to external drug application, was observed over a range of negative membrane potentials, though the time course of inhibition was slow and the blocking effect took more than 30 minutes. Regardless, the action potential firing rate was shown to decrease in the presence of the drug due to decreased diastolic depolarization (BoSmith, et al., 1993), with minimal effects on IKcurrent, ICa, and inward rectifier IK1. Similar results were found in cardiac Purkinje fibers (Berger, et al., 1994), with the halfmaximal suppression of rate found to be 0.92 µM but with abolishment of diastolic depolarization with 10 µM. Concentrationresponse curves showed an EC50value of 0.85 µM (Berger, et al., 1994).  25  On the other hand, experiments with the intracellular application of ZD7288 led to the notion of an intracellular blocking site and could help to explain slow onset block when the agent is applied to the extracellular surface. The externally applied drug would first have to somehow permeate across the cell membrane (Harris and Constanti, 1995). While the molecule has a positive charge, suggesting it would be impermeable to the cell membrane, the distribution of charge among the different nitrogens of the structure would make it more permeable and therefore, more likely to reach an intracellular binding site in a lipophilic manner. This was supported by the requirement for a long drug application from the external side (Berger, et al., 1994; BoSmith, et al., 1993; Harris and Constanti, 1995), the long washout period (Berger, et al., 1994) and the lack of complete recovery following washout (Harris and Constanti, 1995; Satoh and Yamada, 2000). When applied to the intracellular side, the halfmaximum blocking concentration was 2 WM, with full block occurring after 1015 min of exposure to 50 WM (Harris and Constanti, 1995).  Regardless of whether the blocker was applied externally or internally, the authors found a lack of usedependence, suggesting that the blocker preferentially binds to a closed or resting state of the channel (Berger, et al., 1994; BoSmith, et al., 1993; Harris and Constanti, 1995). Nevertheless, the effect of ZD7288 on cloned HCN channels suggests a slightly different interpretation. First, using excised, insideout patches, Shin and colleagues (Shin, et al., 2001) demonstrated a usedependent block in both mammalian and invertebrate isoforms and that channel opening by hyperpolarization was required for the blockade. Second, the authors found that the blocking agent could be trapped by closure of the channel, suggesting that the blocking site is located internally to the activation gate. Sitedirected mutagenesis identified specific residues in the S6 transmembrane domain that interact with the ZD7288 molecule, indicating that different sensitivities to block are due to slight sequence differences (Chan, et al., 2009; Cheng, et al., 2007). Lastly, the block was found to be rapid and reversible, with a Kd of ~41 WM (Shin, et al., 2001). To explain the lack of usedependence that had been observed, the authors suggested that in the whole cell conditions, if a small portion of the channels were open at the experimental holding potential, the blocker could equilibrate with the binding site over the duration of the blocker application (Shin, et al., 2001).  Similar in the action to ZD7288, Ivabradine has been demonstrated to be an open channel If blocker that accesses the channel from the internal side of the membrane and the agent is now  26 approved for clinical use (Bucchi, et al., 2002; Bucchi, et al., 2006). More specifically, Ivabradine blocks HCN4 channels in a usedependent manner (Bucchi, et al., 2006), in that it was found to preferentially block open channels, with a greater block found with more frequent hyperpolarizations. In turn, the physiological consequence of blocking If is to slow sinoatrial diastolic depolarization, thereby reducing heart rate without any effects on contractility. Overall, this provides a longer time for coronary blood flow and consequently, Ivabradine is now clinically indicated for the symptomatic treatment of angina pectoris in patients with normal sinus rhythm (Prasad, et al., 2009). 1.3.6 HCN Cyclic Nucleotide Modulation Cyclic nucleotides, including cAMP and cGMP, have been shown to be full and partial agonists of HCN channels, respectively (Gauss, et al., 1998). They bind directly to these channels, independently of PKA or PKC, shift the voltagedependence of channel opening and closing to more positive voltages (DiFrancesco and Tortora, 1991; Moroni, et al., 2000; Ulens and Tytgat, 2001b) and increase the maximal open probability at negative potentials (Craven and Zagotta, 2004; Dekker and Yellen, 2006; Shin, et al., 2004). This increases the amount of current available at subthreshold potentials and contributes to the positive chronotropic effect of β adrenergic agonists on the heart. Through sitedirected mutagenesis of the cAMPbinding site, in combination with tandem subunits, a single cAMP molecule was shown to be capable of enhancing gating, though maximal channel enhancement required all four sites to be bound (Ulens and Siegelbaum, 2003). More recent findings; however, from a combined approach using patchclamp flourometry and confocal microscopy, suggests that only two of the four binding sites must be occupied to achieve maximal activation (Kusch, et al., 2010). Furthermore, in addition to the direct binding effect of cyclic nucleotides, cAMP can indirectly modulate HCN channels via PKAdependent channel phosphorylation (Liao, et al., 2010). The extent of cAMP modulation of HCN channels depends on the isoform in question (Wainger, et al., 2001). Whereas HCN1 channels are only shifted by 2 mV (Santoro, et al., 1998), HCN4 displays a shift of up to +23 mV (Ishii, et al., 1999). The human HCN3 channel does not appear to be modulated by cyclic nucleotide binding (Stieber, et al., 2005). Interestingly however, when the HCN3 CNBD was placed in an HCN4 background channel, the channel  27 could bind and respond to the cyclic nucleotides, suggesting that the HCN3 CNBD is still functional but the translation to activation kinetics is affected. Based on deletion and chimera experiments, the CNBD has been proposed to fulfill an autoinhibitory role on channel gating (Wainger, et al., 2001; Wang, et al., 2001). In the absence of cAMP, the Cterminus of the channel interacts with the gating region, shifting the voltage dependence of activation to more negative potentials and making it harder for the channel to open. In the presence of cAMP, or in the absence of the Cterminus, this inhibition is removed, kinetics are enhanced, and the voltagedependence of activation is shifted to more positive potentials. The binding of cAMP to the CNBD is thought to induce a significant conformational change of the entire Cterminus region of the channel, which is coupled to the opening of the channel pore. Furthermore, the cAMP sensitivity and efficacy of modulation have been shown to depend on the CNBD and Clinker regions (Viscomi, et al., 2001; Zhou and Siegelbaum, 2007). The crystal structures of the Cterminal regions of a few channels (Zagotta, et al., 2003) (Flynn, et al., 2007; Xu, et al., 2010) have shed substantial light on the mechanism of cyclic nucleotide modulation. In the absence of cAMP, the CNBD is thought to interact with the upstream Clinker region. This, in turn, applies a force on the S6 domain to inhibit pore opening (Craven and Zagotta, 2004; Zhou and Siegelbaum, 2007). The binding of cAMP is proposed to cause local movements in the CNBD. In turn, this causes direct movement in the distal regions of the intrasubunit Clinker helices, movement in the Clinker of the neighbouring subunit (Zhou and Siegelbaum, 2007), and ultimately, an enhanced channel gating. This is also supported by crosslinking experiments in spHCN (Prole and Yellen, 2006), which demonstrate that movement of the S4S5 region of the transmembrane domain relative to the Clinker region, or vice versa, is required for activation by cAMP. 1.3.7 Other HCN Modulation In addition to cyclic nucleotide modulation and current augmentation by extracellular cations, HCN channels interact with, and can be modulated by, several different proteins and molecules. Although displaying isoform variability, the minkrelated peptide (MIRP1) a known βsubunit of the Kv channel family, has been suggested to be an HCN auxillary subunit. Macroscopic currents reveal that MIRP1 enhances surface protein and current expression, accelerates activation kinetics and increases maximal conductance in HCN1 and HCN2 (Yu, et  28 al., 2001) (Qu, et al., 2004). Nevertheless, MIRP1 has been shown to have either no effect (Altomare, et al., 2003) or a negative effect on HCN4 channels (Decher, et al., 2003). These results may be explained by the findings from single channel currents (Brandt, et al., 2009), which demonstrate that the beta subunit significantly increases the single channel amplitudes and conductance of HCN1, 2 and 4. Mean open time, however, was decreased for HCN4, which could explain the negative effect observed on the macrscopic currents. HCN channels are also modulated by the phosphorylation status of the channel. Before the molecular correlates of If were identified, the current was shown to be sensitive to phosphatase inhibition by calyculin A, suggesting that phosphorylation increased current conductance in a manner distinct from the cyclic nucleotide modulation (Accili, et al., 1997). Soon thereafter, the original cloning of HCN channels was completed and was possible due to an interaction with a Src kinase peptide (Santoro, et al., 1997). Only recently, however, has this interaction been shown to be functionally relevant (Zong, et al., 2005). Through various analyses, Src kinase has been shown to directly bind to the Cterminal portion of HCN2 channels through the SH3 domain, resulting in the phosphorylation of a residue in the Clinker domain (Arinsburg, et al., 2006; Li, et al., 2008; Zong, et al., 2005). While the effects of the phosphorylation status on current density and voltage dependence remain controversial, the activation kinetics are accelerated. Furthermore, phosphorylation has been found to reduce the amount of delay observed upon channel opening (Arinsburg, et al., 2006). In addition, tyrosine phosphorylation indirectly modulates HCN channel function in a similar manner. Using the tyrosine kinase blocker, genistein, phosphorylation was suggested to cause an increase in current amplitude, shift the voltagedependence of activation to more depolarized potentials, and accelerate activation kinetics (Wu and Cohen, 1997; Yu, et al., 2004a). Lastly, protein kinase A was recently shown to regulate If in the SA node and caused a positive shift in the voltage dependence of activation in heterologously expressed HCN4 channels. PKA phosphorylation sites were identified in the N and Ctermini of the HCN4 channel  (Liao, et al., 2010). Excised patch experiments consistently demonstrate a substantial negative shift of the If voltage dependence of activation, compared to whole cell configurations. Most likely, this is due to the loss of intracellular regulators of channel properties. While a fraction of this shift is due to cAMP, and can be reversed by application of cAMP to the bathing solution (DiFrancesco and Tortora, 1991), other modulatory factors must play a role. A membrane phospholipid,  29 phosphatidylinositol 4,5bisphosphate (PIP2), both exogenous and endogenous, has been shown to regulate HCN2 and native sinoatrial HCN currents, and causes a positive shift in the voltage dependence of activation independently of that observed by cAMP (Pian, et al., 2007; Pian, et al., 2006; Zolles, et al., 2006). Interestingly, the effect is opposite to that observed for the evolutionarilyrelated CNG channels that are inhibited by PIP2 (Pian, et al., 2007). Lastly, HCN channels are sensitive to the intracellular acidity level. Changing the intracellular pH from 6.4 to 8.4 shifted the voltage dependence of spHCN activation to more positive potentials. A histidine residue in the Clinker of spHCN channels was found to be critical for this response (Mistrik and Torre, 2004), however, in mammalian channels, this pH sensor is located in the intracellular S4S5 linker (Zong, et al., 2001). 1.3.8 HCN NLinked Glycosylation  Nlinked glycosylation is a posttranslational modification of the channel that adds an N glycan to a consensus sequence of AsnXSer/Thr that is exposed to the extracellular space. In general, glycosylation promotes proper folding, stability, and oligomeric assembly in the endoplasmic reticulum (ER) and facilitates transport to the plasma membrane (Helenius and Aebi, 2001; Helenius and Aebi, 2004). All mammalian HCN channels have one potential N linked glycosylation consensus sequence, in the extracellular linker region between S5 and the pore helix (Jackson, et al., 2007; Proenza, et al., 2002b) and are glycosylated in various tissues and heterologous systems (Hegle, et al., 2010; Much, et al., 2003; Santoro, et al., 1997; Zha, et al., 2008). Furthermore, glycosylation appears to play a crucial role in membrane insertion (Much, et al., 2003; Nazzari, et al., 2008) and heteromerization of different isoforms (Zha, et al., 2008), albeit not essential for mammalian channel function (Hegle, et al., 2010). Finally, the glycosylation of one mammalian isoform has been shown to rescue another nonglycosylated isoform by increasing its stability and membrane insertion (Much, et al., 2003) (Whitaker, et al., 2007). Overall, glycosylation assists the successful coassembly and functioning of heteromeric complexes, an important mechanism in generating ion channel diversity. Interestingly, however, invertebrate HCN channels do not contain the glycosylation sequon and do not undergo this post translational modification (Hegle, et al., 2010) but still manage to assemble correctly, traffick to the cell surface and behave as functional channels. The point during evolutionary history at which HCNs evolved and adopted Nlinked glycosylation remains to be elucidated.  30 1.3.9 HCN Mouse Models As previously mentioned, the physiological role of mammalian HCN channels is to contribute to rhythmic firing in both cardiac and neuronal tissue. The importance of their contribution has been assessed using several knockout models (reviewed in (Ludwig, et al., 2008)). HCN4 was found to be the predominant isoform in the sinoatrial node and in the original HCN4 knockout models (global and cardiac specific HCN4 constitutive knockouts), the channel ablation was lethal in utero at embryonic days 10.511.5 (Stieber, et al., 2003a). This suggested that HCN4 was necessary for sinoatrial node formation and development. Moreover, inducible and/or cardiacspecific HCN4 deletions (Baruscotti, et al., 2010b; Herrmann, et al., 2007; Hoesl, et al., 2008) demonstrate up to 5075% reduction in sinoatrial If with resulting cardiac arrhythmias, characterized by sinus pauses (Herrmann, et al., 2007), bradycardia, and atrioventricular block (Baruscotti, et al., 2010b). Overall, these results suggest that HCN4 is crucial in early development and that it contributes to, but is not the sole determinant of, the generation of normal heart rate in adult mice (Nof, et al., 2010). The knockout models also illustrate that dysfunctional HCN4 channels can be a direct cause of cardiac rhythm disorders. Interestingly, the phenotype of a zebrafish model lacking Ih, created over a decade earlier, shared a few similarities with the HCN4 mouse models (Baker, et al., 1997; Warren, et al., 2001). While the molecular identity of the mutation or channel was not identified, the reduction in Ih caused a reduced heart rate in the embryo, in the absence of any other observed defect or morphological change (Baker, et al., 1997). This translated to an observed chronic bradycardia in the adult fish (Warren, et al., 2001). The knockout mouse model for HCN2 displayed a much different phenotype (Ludwig, et al., 2003). In addition to a cardiac arrhythmia, affected mice demonstrated ataxic gait, tremor, reduced locomotor activity and absence epilepsy. Interestingly, the heart rate response in mutant mice was similar to the wild type controls and suggested that the HCN2 isoform was not required for the chronotropic effect of sympathetic stimulation. Furthermore, a recent mouse model of absence epilepsy, displaying ataxia and brief behavioural arrests and tonicclonic convulsions, was found to be the result of a novel spontaneous mutation in HCN2. A 4base pair insertion, causing a truncated channel and a severe reduction of HCN2 mRNA in brain tissue (Chung, et al., 2009), produced a phenotype that was analogous to that created by the HCN2 knockout.  31 The phenotype of the HCN1 knockout mouse model is probably the most complex. General and forebrainspecific knockouts were generated and the role of HCN1 in the Purkinje cells of the cerebellum (Nolan, et al., 2003) and the hippocampal CA1 pyramidal cells (Nolan, et al., 2004) were assessed. In general, the health and longevity of the mice were unaltered by the absence of HCN1 and no changes were seen in the anatomical features of the brain. Interestingly, the knockout in the different regions had different impacts on learning and memory. The general HCN1 knockout, that included cerebellar and hippocampal regions, had motor learning and memory deficits, especially in relation to speed and repetition of movement (Nolan, et al., 2003) The authors concluded that while spontaneous activity in the Purkinje cells did not require HCN1, the primary role of the channel was to integrate the hyperpolarizing inputs at voltages negative to spiking thresholds. HCN1 knockout mice did not display spontaneous seizure activity; however, when assessed for their susceptibility to seizures (Santoro, et al., 2010), they displayed an increased seizure severity and mortality, which supports a putative role of HCN1 in temporal lobe epilepsy in humans. Lastly, to address the physiological contribution of cAMP binding to the role of HCN in pacemaker activity, Harzheim and colleagues recently created a constitutively expressing HCN4 knockin model of R669Q, a mutation that specifically eliminates cAMP binding without affecting other channel function (Harzheim, et al., 2008). Like the original HCN4 knockout model (Stieber, et al., 2003a), the homozygous mutant was embryonically lethal by E12. Both homozygous and heterozygous mice displayed reduced heart rates during the embryonic stage and had either no or an attenuated response to betaadrenergic stimulation (Harzheim, et al., 2008). Unexpectedly; however, heterozygous adult mice displayed normal heart rates at rest and during exercise. Similar to the findings of Herrmann et al (Herrmann, et al., 2007), the adult mice were susceptible to sinus pauses and conduction block following exercise. Ultimately, the authors summarized that HCN4 is a major target for cAMP binding during the embryonic stage and that this is essential for survival. If the mice survive; however, the severity of the physiological impact is reduced and HCN4 isoforms become a safety mechanism to help stabilize heart rates.  32 1.3.10 HCN Channelopathies As was described for Kv channels, the discovery of associations between genetic mutations in HCN channels and inheritable disease is ongoing. Albeit in only a few cases, the physiological impact of HCN channels’ contribution to the cardiac pacemaking activity of the heart has been confirmed by the association of different mutations in the HCN4 isoform to sinus arrhythmias in humans (Milanesi, et al., 2006; Nof, et al., 2007; SchulzeBahr, et al., 2003; Ueda, et al., 2004). The first known diseasecausing mutation was found in a patient with idiopathic sinus node dysfunction who had experienced severe syncope and atrial fibrillation over two decades and presented with bradycardia and chronotropic incompetence (SchulzeBahr, et al., 2003). An identified frameshift mutation HCN4573X, in the Clinker region of HCN4, resulted in a truncated channel that lacked the Cterminus, including the CNBD. Biophysical analysis revealed that mutant channel function was normal, except for the inability to bind cAMP. A recent transgenic mouse of HCN4573X (Alig, et al., 2009; SchulzeBahr, et al., 2003) displayed a phenotype that was similar to the recent temporallyinduced HCN4 knockout (Baruscotti, et al., 2010b), supporting the clinical symptoms of the mutation carrier. The mutant caused a reduction in basal firing frequency and a reduced maximal HR during exercise, though, the SAN cells were still responsive to beta adrenergic stimulation, suggesting that other pacemaker mechanisms were contributing significantly to the chronotropic effect. The first reported traffickingdefective disease mutation in HCN channels was identified in a 43year old female presenting with sinus node dysfunction, severe bradycardia, recurrent syncope, prolongation of the QT interval (670 ms), and polymorphic ventricular tachycardia following a period of cardiac arrest (Ueda, et al., 2004). A point mutation, D533N, at an evolutionarily conserved residue in the Clinker region of HCN4, cosegregated with the phenotype in the family and three affected relatives were found to be heterozygous, thus, supporting the pathogenicity of this specific HCN4 mutation.  Very weak cell membrane expression of D553N was seen, and when coexpressed with wild type subunits, the mutant behaved in a dominant negative manner. The third HCN4 mutation was also found through the screening of individuals with sinus bradycardia. Of 52 individuals screened, one individual was found to carry a point mutation in the evolutionary conserved CNBD (Milanesi, et al., 2006). Twentyseven members of a four  33 generation family were screened and 15 were found to have the S672R mutation, and presented with asymptomatic sinus bradycardia. Heart rates of the mutation carriers were significantly lower than those of the unaffected relatives; however, no other cardiac investigations were carried out. In vitro analysis showed that the mutation caused a negative shift in the voltage dependence of activation and faster deactivation kinetics. Strikingly, modulation by cyclic nucleotides was not affected when assessed in insideout patches. A more recent identification of a diseasecausing mutation in HCN4 is the most interesting from a biophysical perspective (Nof, et al., 2007). A 16member family with sinus bradycardia was clinically screened and 8 individuals were identified as being affected with the condition. All individuals had experienced bradycardia from a young age, though they were all asymptomatic and had a normal heart response to exercise. Direct sequencing revealed a single point mutation, G480R, located within the selectivity filter of the HCN4 transmembrane region, which showed complete segregation with the phenotype in an autosomal dominant fashion. Interestingly, however, the biophysical analysis demonstrated no change in the reversal potential of the mutant channels, suggesting that the permeability of these channels was not affected. Nevertheless, a negative shift was seen in the voltage dependence of activation, an effect mimicking vagal stimulation, and a reduction in channel synthesis and cell surface expression was seen, as demonstrated by biotinylation and western blot experiments (Nof, et al., 2007). Overall, the authors suggested that, while the location of the mutation is in one of the most critical areas relative to channel function, the impact on clinical outcome was fairly benign. They speculated that compensation by either the second allele of the wild type HCN4 or the HCN2 isoform could be involved. Lastly, a novel HCN4 mutation (A485V) located in the channel pore has been identified in 3 unrelated families of Moroccan Jewish descent that had familial sinus bradycardia (Laish Farkash, et al., 2010). Individuals presented with cardiac arrest or presyncopal episodes and upon further review of family history, sinus bradycardia was evident in other firstdegree relatives. The phenotype seemed to segregate in an autosomal dominant fashion and the identified mutation was not observed in 150 controls. Functional characterization revealed a reduced synthesis and a reduced channel function and was concluded to be the cause of the observed bradycardia. Interestingly, however, these patients also experienced a normal response to exercise.  34 1.4 Molecular Evolution Genetic variation is the foundation of molecular evolution. Variation is caused by the insertion, deletion or substitution of one or more nucleotides, whole gene duplication or deletions, chromosomal rearrangements and complete genome duplications. While two rounds of complete genome duplication have been predicted to have occurred throughout vertebrate evolution (Hughes and Friedman, 2003), gene duplication is thought to be the key player in generating the molecular diversity seen today (Ohno, 1970). Because genetic redundancy carries extreme energetic burdens, three possible consequences follow gene duplication: 1) loss of function in which the accumulation of deleterious mutations ultimately leads to removal of the duplication; 2) neofunctionalization, in which the new duplicates gain a new function through either advantageous mutations and positive Darwinian selection and/or neutral variation; or 3) subfunctionalization – a process in which two genes undergo slight modifications with an ensuing partition of their function or expression profile (Lynch and Conery, 2000; Lynch and Katju, 2004). The eventual consequence is the preservation of only those genes that are non redundant. In general, the most frequent byproduct of gene duplication is loss of function. 1.4.1 Channel Origins and Phylogeny In contrast, the generation of the diversity found within the voltagegated channel superfamily is predicted to be a direct result of several gene duplications and single mutations followed by neofunctionalization (Anderson and Greenberg, 2001) of a common ancestral channel (Hille, 2001). For example, the available crystal structures have striking similarities between the pore regions of the 2TM and 6TM channels, implying that the latter likely arose through the gene duplication and linkage of the 2TM channel with an independent VSD protein (Jegla, et al., 2009). Some of these VSD proteins have been recently identified in urochordates (Ramsey, et al., 2006) (Murata, et al., 2005; Sasaki, et al., 2006). Furthermore, a phylogenetic analysis of the pore regions in the voltagegated channels suggests that the EAG family of ion channels (Kv1012 or ERG, EAG, ELK) are more closely related to CNG and HCN channels (Yu and Catterall, 2004) (Figure 1.1). Unexpectedly, the pore regions of these K +  selective channels are more similar to the pore regions of channels that are either non or weaklyselective for K +  and activated by cyclic nucleotide binding than they are to those of other Kv channels. While the EAG channels do have a CNBD, they show minimal response to nucleotide binding  35 (Cui, et al., 2000). The phylogenetic relationship, therefore, implies that critical residues involved in the link between nucleotide binding and gating located within this pore region most likely arose in a common ancestor of these channels and was evolutionarily conserved, despite the overt loss of the functional connection. As described previously, the KAT1 and AKT1 plant channels share common functional characteristics with both the EAG and the cyclic nucleotide modulated channels. Together, the evolutionary relationships and functional similarities suggest that the common ancestor of the Kv channels likely arose prior to the divergence of the plant lineage. Furthermore, the EAG, KAT1 and AKT1 channel families represent a pivotal evolutionary connection within the Kv superfamily. The sequences of all extant channels found in a diverse range of species are the byproduct of the tolerated genetic variation that has occurred throughout evolutionary history (Jegla, et al., 2009). Therefore, phylogenetic analyses of the Kv superfamily can help reveal the timing of successful duplication events, most of which have occurred more recently in vertebrate evolution. As well, the identity of the particular amino acids retained can be elucidated. Residue conservation over time implies an intolerance to change and sites that are conserved throughout Kv channel evolution are presumably under high selective pressure and likely critical for the general structurefunction of these tetrameric proteins. On the other hand, sites conserved among paralogous gene sequences within a channel subfamily (i.e. HCN channels), but variable throughout the rest of the superfamily, can help identify amino acids that are vital for specific familial function and most likely contributors to the neofunctionalization of the protein. Finally, conservation among orthologs (same gene but found in different species), can help reveal functionally important isoformspecific residues. 1.4.2 HCN Channel Evolution As previously mentioned, HCN channels are members of the voltagegated cation channel superfamily and are most closely related to the cyclic nucleotide gated (CNG) channel and etheragogo (EAG) potassium channel families. The four mammalian paralogs (HCN1 HCN4) exhibit 8090% sequence identity in the core region of the protein (from the start of S1 to the end of the CNBD). This high sequence identity suggests that the four mammalian HCN genes arose from duplications of a single ancestral gene prior to the lineage divergence and that there has been a high degree of selective pressure to conserve the channel amino acid composition.  36 Although the mammalian sequences and biophysical functions are similar, the different phenotypes of the knockout mouse models previously mentioned imply that the sequence divergence that has occurred throughout the evolution of this gene family has been sufficient to differentiate the physiological roles of these four paralogs and minimize functional redundancy. In addition to the four mammalian isoforms, and prior to the work described in this thesis, only a few HCN genes had been cloned or identified from lower vertebrate and invertebrate species. A single HCN1 ortholog has been cloned from the rainbow trout (Cho, et al., 2003) whereas one and two HCN homologs have been cloned from arthropods (Gisselmann, et al., 2005; Gisselmann, et al., 2003; Krieger, et al., 1999; Marx, et al., 1999) and the sea urchin (Galindo, et al., 2005; Gauss, et al., 1998), respectively. Due to the lack of sequence representation from a diverse sampling of species distributed throughout evolutionary history, phylogenetic analyses of the HCN family and its relatives have been limited and have yielded inconsistent patterns of evolution (Craven and Zagotta, 2006; Galindo, et al., 2005). Recently, HCN genes have been identified in more species distributed across the metazoan lineage and they demonstrate considerable identity with the mammalian homologs. While two gene sequences have been found in Nematostella (sea anemone), (Hegle, et al., 2010; Jegla, et al., 2009) a radially symmetric Cnidaria species that diverged very early in Metazoan evolution, HCN is strikingly absent from the genome of C. elegans. Nevertheless, several HCN genes have been identified in the model organisms of the tunicate species Ciona intestinalis and Botryllus schlosseri (Hellbach, et al., 2011; Jackson, et al., 2007; Okamura, et al., 2005) (see Chapters 2 and 3), both of which are urochordate species that evolved at the evolutionary branch point at the divergence of invertebrate and vertebrate lineages. These recent findings suggest that HCN channels play a role in the cardiac pacemaking activity of these organisms and could represent a precursor of the vertebrate conduction system (Hellbach, et al., 2011). This physiological connection supports an even greater need to explore the function of these channels in model organisms, as they could shed light on the ancestral origins of HCN channels and provide a unique and prolific system to explore drug development and human disease mechanisms related to pacemaker dysfunction.   37 1. 5 Evolution and Disease Evolutionary change and Mendelian genetic disorders share remarkable similarities in the mechanisms by which they arise. The primary difference, however, is in the timescale of their occurrence. Compared to the duration of the evolutionary history of life, the timecourse of human disease is infinitesimal. In general, point mutations (substitutions) can be divided into two main types: missense mutations which result in the change of the amino acids at the peptide level, and silent mutations which occur when the DNA change at the codon level does not alter the residue which it encodes. The latter is presumed to escape evolutionary pressures because the resultant phenotype is unaltered. Missense mutations; however, can be further subdivided into two categories, each of which is under different evolutionary constraints. Amino acid substitutions that correspond to no detectable change in the protein’s phenotype are said to be neutral because they evade selective pressures and are the foundation of genetic drift. Missense mutations that cause phenotypic change, in contrast, can experience either positive or negative/purifying selection, depending on the consequential impact on the fitness of the species. A change that causes a negative impact on survival would be removed throughout the course of evolution. A positive impact, on the other hand, would lead to an increase in genotypic frequency in future generations. In general, proteins are under strong purifying selection (Cargill, et al., 1999) and those that are under higher selective pressures are significantly more likely to be associated with human disease (Arbiza, et al., 2006). The majority of the Mendelian diseases are due to single amino acid substitutions, and a strong tendency exists for disease mutations to occur at highly conserved sites (Thomas and Kejariwal, 2004). Unless a disease mutation confers a selective advantage, such as the case of sickle cell anemia with its protective effects from malaria, or is inconsequential for the reproductive fitness of an individual, given enough time it would likely be removed over the course of future generations. In other words, what we experience as human disease is shortterm evolution in progress. Overall, an interdependent relationship exists between protein evolution, protein structurefunction and human disease (FerrerCosta, et al., 2002). Patterns of conservation can help identify residues that are predicted to be important to protein structure and function and intolerant to change. In turn, changes at these residues carry an inherent susceptibility for causing human disease. Evolutionary analyses are therefore important to highlight sites and regions of  38 channels prone to diseasecausing mutations and are being used in medical genetics to help establish pathogenicity of newly identified variants (Chen, et al., 2011; Krishnan, et al., 2011), a common occurrence with the everincreasing use of genetic testing. In turn, mutations that are associated with channelopathies provide us with a direct clinical demonstration of the intolerance to change and help to expand our knowledge of the structurefunction relationship of voltage gated ion channels. 1.6 Scope of Thesis This thesis explores the evolutionary relationships of voltagegated potassium channels and how evolutionary change relates to human disease. Specifically, the first two chapters focus on the evolution and function of hyperpolarizationactivated cyclic nucleotidegated (HCN) channels, while the third utilizes the evolutionary relationships of two related potassium ion channels involved in cardiac tissue excitability to explore the molecular identities underlying cardiac arrhythmias. Phylogenetic analyses are imperative to understand how channel sequences have evolved and to illuminate key sequences, and portions thereof, important for channel structure and function. Prior to the work that encompasses the second chapter of this thesis, the available HCN sequence representation from a diverse range of species was lacking and previous analyses were inconclusive. The specific objective of Chapter 2 was to use available protein and accessible genomic databases to identify new putative HCN sequences, perform a detailed sequence comparison of the channel family and construct an evolutionary history for these genes. Our analyses provide insight into the molecular evolution of the HCN family and provide a valuable tool to aid in the planning of future experiments designed to probe the relationship between structure and function of HCN channels. One key result of the second chapter of this thesis is the identification and analysis of three putative HCN sequences in two tunicate species: Ciona intestinalis and Ciona savignyi. These ascidian species are classified as marine invertebrate chordates (Satoh, et al., 2003) and represent the closest extant ancestor of the vertebrate lineage. They are believed to have diverged from the vertebrate lineage over 550MYA, at the pivotal evolutionary branch point of invertebrates and vertebrates and provide an ancestral genetic reference point prior to the duplication and diversification events that have occurred throughout vertebrate evolution (Corbo,  39 et al., 2001). The sequence conservation between them was high and channel functions were predicted to be similar but not identical. The specific objective of Chapter 3 was therefore to clone and characterize two of these channels from C. intestinalis that were predicted to be functional based on sequence data. Our results suggest that at least one lineagespecific duplication event generated two HCN genes, ciHCNa and ciHCNb, both of which have undergone independent evolution. While preliminary phylogenetic analyses and sequence comparisons suggest that channel function would be similar, in vitro analyses suggest that both genes produce functional HCN channels but with intriguingly different biophysical properties. Our data supports previous findings of the physiological role of HCN channels in tunicate species (Hellbach, et al., 2011) and will provide a template to build future experiments to explore the in vivo role of these channels. Furthermore, they will help shed light on the ancestral reference point of vertebrate HCN channels. With a better understanding of the ancestral state of the vertebrate HCN channel family, we will begin to gain a more indepth understanding of how the four mammalian HCN channels have evolved and ultimately how this evolution corresponds to their physiological function in the human body. Previous studies have demonstrated that amino acid mutations associated with human disease are preferentially located at evolutionarily conserved residues (Briscoe, et al., 2004; Miller and Kumar, 2001). To perform an accurate analysis of the correlation between evolution and diseases related to voltagegated potassium channels, however, we needed to identify an ion channel family that had a significant number of known diseaseassociated mutations. Two channels, HERG and KCNQ1, play a critical role in cardiac repolarization. Disease mutations are associated with Long QT syndrome, atrial fibrillation, sudden infant death syndrome and sudden unexplained death. The objective of Chapter 4 was to take advantage of the >200 mutations reported for each of the channels to quantitatively analyze, for the first time, the distribution of disease mutations associated with these arrhythmias and to determine the predicted chemical severity of these changes in relation to those observed throughout evolution. Our findings demonstrate that arrhythmiaassociated mutations, and not benign polymorphisms, are preferentially located at evolutionarily conserved and functionally important sites and that they result in more drastic changes in the chemical properties of the residue, compared to those changes observed throughout evolution. Together with previous studies, our data suggest that novel disease mutations can be potentially recognized by surveying naturally occurring variation  40 among species (Briscoe, et al., 2004). The location of the mutation may correlate with clinical severity, however, significant phenotypic variation of these conditions confounds this conclusion. Continued discovery and mapping of mutations in relation to evolutionary change, in parallel with the ongoing analysis of genotypephenotype relationships of channelopathies, will be imperative to better predict the outcome of novel disease mutations and aid in development of mutationspecific therapies.  41 2. THE EVOLUTION AND STRUCTURAL DIVERSIFICATION OF HYPERPOLARIZATIONACTIVATED CYCLIC NUCLEOTIDEGATED (HCN) CHANNEL GENES 1  2.1 Introduction Hyperpolarizationactivated currents, commonly referred to as Ih or If, have been identified in species distributed across the metazoan lineage and contribute to membrane potential in excitable and nonexcitable cells. In the mammalian heart and nervous system they contribute to spontaneous beating in pacemaker cells (DiFrancesco, 1993; Pape, 1996). In non pacing cells, Ih facilitates the regulation of resting membrane properties by limiting the fluctuation of membrane potentials (Pape, 1996). The proteins responsible for Ih have been cloned from both invertebrate and vertebrate species and are referred to as hyperpolarization activated cyclic nucleotidegated (HCN) channels (Gauss, et al., 1998; Ishii, et al., 1999; Ludwig, et al., 1998; Santoro, et al., 1998). When expressed in heterologous systems, HCN subunits form cationselective channels that are directly modulated by cyclic nucleotides and exhibit biophysical properties that are similar to those of Ih observed in native tissue. HCN channels are members of the voltagegated cation channel superfamily and based on sequence homology, are most closely related to the cyclic nucleotide gated (CNG) channel and etheragogo (EAG) potassium channel families. Individual subunits (Figure 2.1) are predicted to have six transmembrane segments (S1S6), with a voltage sensor domain in the S4 segment and an ion conducting pore between S5 and S6. Based on the crystal structures of related potassium channels (Doyle, et al., 1998; Jiang, et al., 2003a; Long, et al., 2005b) it is proposed that four HCN subunits come together to form a tetramer around a central pore. The distal termini of each subunit are cytoplasmic and from the crystal structure of the Cterminus (Zagotta, et al., 2003), it is now known that an αhelical linker region joins the transmembrane region to an evolutionarily conserved cyclic nucleotidebinding domain (CNBD). To date, four mammalian isoforms have been cloned (HCN1HCN4) (Ishii, et al., 1999; Ludwig, et al., 1998; Santoro, et al., 1998).  These paralogs exhibit 8090% sequence identity between the start of S1 to the end of the CNBD. Differences in both sequence and length  1  A version of this chapter has been published. Jackson, HA, Marshall, CM Accili, EA. (2007) Evolution and structural diversification of hyperpolarizationactivated cyclic nucleotidegated channel genes. Physiological Genomics, 29(3): 23145.  42 between paralogs occur predominantly in the N and Ctermini (Ludwig, et al., 1999b). This high sequence identity suggests that the four mammalian HCN genes arose from duplications of a single ancestral gene prior to the lineage divergence. However, when these duplication and diversification events occurred remain unknown. The functional properties of the individual isoforms, such as their responses to changes in voltage and modulation of channel opening by cAMP, are similar but not identical. Likewise, the four HCN paralogs display partial overlapping tissue distribution in both the heart (Ishii, et al., 1999; Mistrik, et al., 2005; Moosmang, et al., 2001; Shi, et al., 1999) and the central nervous system (Ludwig, et al., 1999b). Despite what seems like redundant functional and expression profiles, the overall physiological importance of the individual isoforms is clearly evident from the different phenotypes of knockout mouse models. General and forebrain specific HCN1 knockout mice demonstrate learning and movement dysfunction (Nolan, et al., 2003), HCN2deficient mice exhibited absence epilepsy as well as cardiac sinus dysrhythmia (Ludwig, et al., 2003) and HCN4deficient mice died during embryonic development, possibly due to reduced HCNmediated currents and heart rates of the embryos (Stieber, et al., 2003a). Furthermore, the physiological impact of the mammalian HCN4 isoform in the heart has been confirmed by the association of mutations in this isoform to sinus arrhythmias in humans (Milanesi, et al., 2006; SchulzeBahr, et al., 2003; Ueda, et al., 2004). This pattern of functional disruption partially reflects the predominate tissue distribution of the isoforms but also indicates that the sequence divergence that has occurred in these four paralogs has been sufficient to differentiate their physiological roles, thereby reducing functional redundancy and allowing each to have been retained throughout evolution since the duplication process. In addition to the four mammalian isoforms, only a few HCN genes have been cloned from lower vertebrate and invertebrate species. A single HCN1 ortholog has been cloned from the rainbow trout (Cho, et al., 2003) whereas one and two HCN homologs have been cloned from arthropods (Gisselmann, et al., 2005; Gisselmann, et al., 2003; Krieger, et al., 1999; Marx, et al., 1999) and the sea urchin (Galindo, et al., 2005; Gauss, et al., 1998), respectively. These sequences demonstrate considerable identity with the mammalian homologs, as well as some intriguing differences, and thus have offered some preliminary clues about the evolutionary relationships among HCN genes. But due to the lack of sequence representation from a diverse sampling of species distributed throughout evolutionary history, previous phylogenetic analyses  43 of the HCN family and its relatives have been limited and have yielded inconsistent patterns of evolution (Craven and Zagotta, 2006; Galindo, et al., 2005). Furthermore, the current sampling of the four vertebrate isoforms is limited primarily to closelyrelated mammalian sequences. The high residue conservation among these sequences, combined with the lack of sequences from more distantly related vertebrates, renders comparative analyses of the orthologs from the individual vertebrate isoforms ineffective. In this study, we report the first thorough phylogenetic analysis and sequence comparison of the HCN gene family. By performing an extensive search of the available protein and currently accessible wholegenome databases we derived a comprehensive list of known and putative fulllength sequences for HCN homologs from a wide variety of species including urochordates and lower vertebrates. The increased number of sequences and improved species diversification provide information about HCN gene structure at critical periods during its evolutionary history. We identified sequences that are conserved and likely important for general HCN functions, as well as regions that may underlie more subtle differences in function among the different isoforms. Furthermore, analyses of exon structure and genomic organization were carried out. These analyses provide insight into the molecular evolution of this protein within different taxa and support the hypothesis that both lineagespecific and ancestral duplication and divergence events of the HCN genes have occurred throughout its history. 2.2 Materials and Methods 2.2.1 Sequence Data Currently available HCN protein sequences were identified through BLASTP (Altschul, et al., 1990) searches of the National Center for Biotechnology Information (NCBI) and UniProt/SwissProt nonredundant (nr) protein databases. Human HCN2 (Genbank GI no. 21359848), trout HCN1 (Genbank GI no. 33312350) or drosophila HCN (Genbank GI no. 5326833) amino acid sequences were used as queries and default parameters were applied.  To prevent the inclusion of incorrect sequences generated by computer prediction programs, all computerderived annotations were temporarily removed until confirmed by sequence alignment or database annotation described below. Splice variants, short fragments, and other duplicates were also removed.  44 The Ensembl Genome Browser (http://www.ensembl.org/) (Birney, et al., 2004; Hubbard, et al., 2002) was used to determine the genomic position and distribution of the established HCN genes, and either to examine new HCN genes identified by the computer gene prediction programs and annotation process or to identify novel genes. Sequences classified as HCN genes by Ensembl were examined and those that spanned the entire length of known sequences were downloaded. Protein annotations that resembled HCN but either lacked regions of the predicted sequence or showed signs of additional exons or exon fragments were not included.  However, the genomic DNA underlying these protein predictions were used as further reference to help in the manual reannotation of the protein sequence.  TBLASTN (Altschul, et al., 1990) searches of the available genome databases using either the low sensitivity default parameters (optimized for nearexact matches) or medium sensitivity default parameters (optimized to allow for local mismatch) were conducted to identify any further genes or genomic regions that showed significant sequence identity to the peptide query sequence used. In total, thirteen genomes were analyzed, including: Zebrafish (Danio rerio) (v. 35.5b), Japanese pufferfish (Fugu rubripes) (v. 29.2e), Green Pufferfish (Tetraodon) (v. 31.1c), opossum (Monodelphis domestica) (v35.2), dog (Canis familiaris) (v. 35.1d), cow (Bos taurus) (v.36.2), chimpanzee (Pan troglodytes) (v.31.2a), clawed frog (Xenopus tropicalis) (v. 31.1a), chicken (Gallus gallus) (v.35.1k), sea squirt (Ciona intestinalis) (v. 35.195b), pacific sea squirt (Ciona savignyi) (CSAV2.0), mosquito (Anopheles gambiae) (v. 23.2b.1), and worm (Caenorhabditis elegans) (v 29.130). Human HCN2, trout HCN1 and drosophila HCN sequences were used as query sequences for vertebrate, fish and insect genomes, respectively. For the urochordate genomes, sea urchin HCN (Genbank GI no. 74136757), trout HCN1 and human HCN2 sequences were used. In general, similar TBLASTN results were found regardless of the query sequence used, reflecting the high degree of sequence identity found within the core region throughout the HCN family. Full length putative protein sequences were constructed from the conceptual translation of genomic DNA as previously described (Marshall, et al., 2005) with the proposed starting methionine in vertebrates supported by the presence of a consensus start sequence (Kozak, 1996). Genome position and intronexon structure were examined and recorded. A list of the sequences used in the analysis is shown in Table 1.  45 Table 21: List of HCN Sequences Used in Analyses Naming of sequences = HCN isoform followed by the species. () indicate previous names for sequences. For duplicates, ‘a’ and ‘b’ labels were assigned arbitrarily. * indicate those that were reannotated or identified in this study and †  indicates a sequence missing at least one internal exon and was therefore removed from phylogenetic analyses. Abbreviations: Invt., invertebrate; Mam., mammal; Vt., lower vertebrate; Uro., urochordate.  NA = protein identifier was not available for the sequence included in final analysis. (Jackson, et al., 2007)              Name    Organism  Common Name       Taxonomy Identifier          Database HCN_silkmoth (HvIh) HCN_bee (AmIh) HCN_lobster (PaIh) HCN_fly (DmIh) HCN_urchin (SpIh/spHCN) HCN2_urchin (spHCN2) HCN1_mouse (mHCN1) HCN1_trout (tHCN1) HCN1_human (hHCN1) HCN1_rabbit (rbHCN1) HCN1_dog HCN1_rat (rHCN1) HCN1_fugu HCN2_human (hHCN2) HCN2_mouse (mHCN2) HCN2_rat (rHCN2) HCN2_zebrafish HCN3_mouse (mHCN3) HCN3_human (hHCN3) HCN3_rat (rHCN3) HCN3_cow HCN3_dog HCN4_human (hHCN4) HCN4_mouse (mHCN4) HCN4_rabbit (rbHCN4) HCN4_rat (rHCN4) HCN4_dog HCN1_chimpanzee* HCN4_chimpanzee* † HCN1_opposum* HCN3_opposum* HCN4_opposum* HCN2_chicken* † HCN4_chicken* † HCN1_frog* † HCN2_frog* HCN4_frog* † HCN1_green_puffer* HCN4a_green puffer* HCN4b_green_puffer* HCN2a_green_puffer* HCN2b_green puffer* HCN3_green_puffer* HCN3_zebrafish* HCN4_zebrafish* HCN2b_fugu*† HCN2a_fugu* HCN3a_fugu* HCN3b_fugu* HCN4a_fugu* HCN4b_fugu* HCN_mosquito* HCNa_c.intestinalis* HCNb_c.intestinalis* HCNc_c.intestinalis* HCNa_c.savignyi* HCNb_c.savignyi* HCNc_c.savignyi*  Heliothis virescens Apis mellifera Panulirus argus Drosophila Strongylocentrotus purpuratus Strongylocentrotus purpuratus Mus musculus Oncorhynchus mykiss Homo sapiens Oryctolagus cuniculus Canis familiaris Rattus norvegicus Takifugu rubripes Homo sapiens Mus musculus Rattus norvegicus Danio Rerio Mus musculus Homo sapiens Rattus norvegicus Bos taurus Canis familiaris Homo sapiens Mus musculus Oryctolagus cuniculus Rattus norvegicus Canis familiaris Pan troglodytes Pan troglodytes Monodelphis domestica Monodelphis domestica Monodelphis domestica Gallus gallus Gallus gallus Xenopus tropicalis Xenopus tropicalis Xenopus tropicalis Tetraodon nigroviridis Tetraodon nigroviridis Tetraodon nigroviridis Tetraodon nigroviridis Tetraodon nigroviridis Tetraodon nigroviridis Danio Rerio Danio Rerio Takifugu rubripes Takifugu rubripes Takifugu rubripes Takifugu rubripes Takifugu rubripes Takifugu rubripes Anopheles gambiae Ciona intestinalis Ciona intestinalis Ciona intestinalis Ciona savignyi Ciona savignyi Ciona savignyi  silkmoth honey bee spiny lobster fruit fly sea urchin sea urchin mouse rainbow trout human rabbit dog norway rat japanese pufferfish human mouse norway rat zebrafish mouse human norway rat cow dog human mouse rabbit norway rat dog chimpanzee chimpanzee opossum opossum opossum red jungle fowl red jungle fowl frog frog frog spotted green pufferfish spotted green pufferfish spotted green pufferfish spotted green pufferfish spotted green pufferfish spotted green pufferfish zebrafish zebrafish japanese pufferfish japanese pufferfish japanese pufferfish japanese pufferfish japanese pufferfish japanese pufferfish mosquito sea squirt sea squirt sea squirt pacific sea squirt pacific sea squirt pacific sea squirt  Invt. Invt. Invt. Invt. Invt. Invt. Mam. Vt. Mam. Mam. Mam. Mam. Vt. Mam. Mam. Mam. Vt. Mam. Mam. Mam. Mam. Mam. Mam. Mam. Mam. Mam. Mam. Mam. Mam. Mam. Mam. Mam. Vt. Vt. Vt. Vt. Vt. Vt. Vt. Vt. Vt. Vt. Vt. Vt. Vt. Vt. Vt. Vt. Vt. Vt. Vt. Invt. Uro. Uro. Uro. Uro. Uro. Uro.  GI:3970750 GI:33355927 GI:33355925 GI:5326833 GI:47551101 GI:74136757 GI:6754168 GI:33312350 GI:32698746 GI:38605639 GI:73954256 GI:29840774 SINFRUP00000160459 GI:21359848 GI:6680189 GI:29840773 GI:68399930 GI:6680191 GI:38327037 GI:29840772 GI:76612489 GI:73961597 GI:4885407 GI:29840776 GI:38605640 GI:29840771 GI:74000965 NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA  GenBank GenBank GenBank GenBank GenBank GenBank GenBank GenBank GenBank GenBank GenBank GenBank Ensembl GenBank GenBank GenBank GenBank GenBank GenBank GenBank GenBank GenBank GenBank GenBank GenBank GenBank GenBank Ensembl Ensembl Ensembl Ensembl Ensembl Ensembl Ensembl Ensembl Ensembl Ensembl Ensembl Ensembl Ensembl Ensembl Ensembl Ensembl Ensembl Ensembl Ensembl Ensembl Ensembl Ensembl Ensembl Ensembl Ensembl Ensembl Ensembl Ensembl Ensembl Ensembl Ensembl  46  2.2.2 Multiple Sequence Alignments Protein sequence alignments were performed using the default parameters of ClustalX (version 1.83) (Thompson, et al., 1997). Alignments were subsequently examined and edited in GeneDoc (Nicholas, et al., 1997). To ensure that the HCN genes identified were fulllength and nonredundant, those lacking a putative start codon were discarded. Isoform pairs identified in some of the fish genomes were included if they corresponded to a different location in the genome, as they are likely the result of a polyploidization event that is thought to have occurred early in their lineage ((Hoegg, et al., 2004) and references therein).  Partial sequences that lacked a portion of the distal Cterminus corresponding to the last mammalian exon were included as this region was not used in the phylogenetic analyses due to the high degree of length and sequence variability among the different HCN genes. For the N and Cterminal alignments, vertebrate sequences corresponding to the region upstream of S1 or downstream of the CNBD (see Figure 2.1) were individually realigned for HCN14 and were shaded based on sequence conservation of 100% (black), 80% (dark grey), and 60% (light grey).  47                        Figure 2.1: Schematic Representation of the Overall Topology of HCN Channels. Included are the six putative transmembrane segments (S1S6), including an S4 voltage sensor (light grey) and the reentrant Ploop region between S5 and S6, all structural characteristics common to the entire voltagegated potassium channel superfamily, including ERG, CNG and HCN channels. HCN and ERG channels also share the potassium selectivity filter GY/FG. Based on the crystal structure of mHCN2 (Zagotta, et al., 2003), the Clinker region in HCN channels contains six alpha helices (A’E’) and connects the cytoplasmic end of S6 to the start of the cyclic nucleotide binding domain (CNBD). The CNBD consists of two alpha helices (A and B) that flank an eightstranded betaroll followed by a third alpha helix at its Cterminal end (C helix). The cyclic nucleotides bind to pocket that is formed by the betaroll and the Chelix. Thick black lines delineate the region that is included in the phylogenetic analyses. In HCN channels, these lines encompass vertebrate exons 2 through 7. Alignments of HCN vertebrate isoforms specific Nterminal (exon 1) and Cterminal (exon 8) regions include those residues that fall outside of this region. (Jackson, et al., 2007) NH2 COOH S2 S3S1 S4 S5 ++ ++ ++ ++ P S6 C l in k e r C N B D cAMP G Y G  48 2.2.3 Phylogenetic Analyses Sequences were trimmed to produce a core alignment spanning from the start of the transmembrane segment S1 to the end of the CNBD (see Figure 2.1). This region exhibits high sequence identity within the HCN family and among the other related sequences that were used. As there is no known bacterial HCN channel, and based on a previous phylogenetic analysis of HCN and CNG channels (Craven and Zagotta, 2006), KAT1 (Genbank GI no. 44888080), human ERG1 (Genbank GI no. 7531135), human CNGA1 and CNGA3 (Genbank GI no. 2506302 and 13959682) sequences were included in the analyses to serve as an outgroup for the rooted phylogenetic trees of the HCN family. Six sequences that were missing exons due to gaps in the genome assembly were removed from the alignment prior to running the programs. NeighborJoining (NJ) trees were generated using ClustalX, followed by tree evaluation with bootstrap resampling of 1000 times. Additional NJ, Maximum Parsimony (MP) and Maximum Likelihood (ML) trees were created using the Seqboot, Protdist, Neighbor, Protpars, Proml, and Consense programs from the PHYLIP package (version 3.65), bootstrapping with 100 replicates with randomized input order and 10 jumbles (Felsenstein, 1996). The TreeView program (version 1.6.6) (Page, 1996) was used to examine and display all trees. 2.3 Results and Discussion To date, 21 unique fulllength HCN coding sequences have been cloned and deposited in GenBank: 4 from arthropods, 2 from sea urchin, 1 from fish and 14 from mammals, specifically human, rabbit, mouse and rat. Previous phylogenetic analyses of the HCN family and its relatives have been limited and inconsistent, in part due to the lack of available sequences from a diverse sampling of species. For example, the product of the first duplication event has been shown to be either HCN1 (Craven and Zagotta, 2006) or HCN3 (Galindo, et al., 2005). Furthermore, the recently cloned spHCN2 (Galindo, et al., 2005) sequence was shown to group with the HCN family, but was situated at the base of the phylogenetic tree. This is not consistent with species evolution, which would place the sea urchin genes with the other deuterostome species, after the arthropod clade division. Therefore, spHCN2 represents either a new family of channels not yet identified in any other species or an HCN homolog that was the product of a duplication event followed by rapid sequence divergence. These inconsistencies underscore the current limitations and need for more detailed analyses. This study expands the breadth of  49 available HCN sequences through the manual annotation of available genomes and in turn, provides the first comprehensive sequence comparison and phylogenetic analysis of this protein family including sequences from a representative sampling of different species across evolutionary history. 2.3.1 HCN Genes are Present in Multiple Copies Across a Wide Spectrum of Species Using a BLASTP search at NCBI or by searching the genome databases available at the Ensembl website, we identified 58 nonredundant HCN sequences. Twentyseven of these sequences were previously identified by cloning or by computer annotation which were confirmed by genomic data. The remaining thirtyone are novel and were completed by the data mining of genomes available at Ensembl or the manual reannotation of the predicted protein translations. Several of the Ensembl predictions were either inconsistent with known sequences or the predicted gene did not span the estimated length of the transcript. The inherent problems in gene prediction and computer protein annotation methods have been previously described ((Marshall, et al., 2005) and references therein). Therefore, we have corrected for this by multi species comparisons and manual reannotation on a genebygene basis. A complete list of the sequences included in this analysis is shown in Table 1. Included in the final list are: 23 mammalian sequences;  including 2 and 3 new sequences from the chimpanzee and opossum genomes, respectively; 22 lower vertebrate sequences; 6 urochordate sequences, including 3 new sequences from both C. intestinalis and C. savignyi genomes; and 7 invertebrate sequences, with a new annotation of the single HCN gene found in the mosquito genome. Of the lower vertebrates, 2 new sequences were from chicken, 3 from frog, 6 new sequences from both the green pufferfish and the Japanese pufferfish genomes, and 2 new sequences in the zebrafish. No HCN sequences were identified in C. elegans. With this substantial increase in the number of fulllength sequences representing a wide range of species, we were able to reconstruct the evolutionary history of this protein family and probe the relationship of channel structure and function in much greater detail than was previously possible. 2.3.2 High Sequence Identity Amongst Four Vertebrate HCN Isoforms Within the Core Region Due to the length and sequence variability that occurs in the N and Ctermini, these regions were trimmed and a core alignment was produced corresponding to a region between S1  50 and the end of the CNBD and approximately to exon 2 through 7 of the mammalian isoforms. Sequence conservation of the four vertebrate isoforms within this region is high, at least 8090% identity amongst the mammalian sequences and over 90% residue conservation between the newly identified fish sequences and their respective mammalian orthologs (Figure 2.2). This indicates that this region has been slow to evolve during the 450MY that separate the fish and human lineages. Two interesting findings based on this sequence comparison are that HCN4 shares the highest sequence identity with all other vertebrate isoforms and that the mammalian HCN3 and fish HCN3 sequences are as similar to the other vertebrate isoforms as they are to each other. This suggests that HCN4 sequences have diverged the least from a common ancestral sequence, as is further evident from the branch lengths in the NJ and ML phylogenetic trees, and that HCN3 sequences have evolved independently within the fish and mammalian lineages by equal amounts but at different sites. Overall, the sequences of the invertebrates and urochordates display a lower conservation with the mammalian sequences. Arthropods share a general sequence identity of approximately 6065% with the mammalian homologs, whereas one of the sea urchin sequences, spHCN1 (also known as spIH or spHCN), and two of the Ciona homologs, here named HCNb and HCNa, are only 55% identical. The second sea urchin sequence, spHCN2, and the other sequence from the Ciona species (HCNc) are even more diverged.  51                   Figure 2.2: High Sequence Conservation Observed Between Vertebrate Isoforms. Sequence conservation between mammalian HCN14 (M1M4) and fish HCN14 (F1F4) isoforms indicate that mammalian HCN3 and fish HCN3 sequences are as similar to each other as they are to the other isoforms. This suggests that sequence divergence has continued independently in each lineage. Numbers indicate % sequence identity/sequence conservation generated by GeneDoc for the region between S1 and end of CNBD, where the total number of sites used was 457. Fore example, 85/94 correlates to 393 of 457 sites identical and 430 of 457 conserved. Human sequences were used as a representative of mammalian sequences and HCNa_fugu used as an example of fish sequences. ‘B’ sequences tend to show lower conservation. (Jackson, et al., 2007)   F4 M491/ 96 F384/ 94 86/ 95 M382/ 93 85/ 94 83/ 93 F285/ 93 88/ 95 83/ 93 80/ 91 M287/ 93 90/ 96 84/ 94 82/ 92 91/ 96 F182/ 93 86/ 94 82/ 94 79/ 92 83/ 93 85/ 93 M184/ 94 87/ 94 82/ 93 80/ 92 85/ 94 85/ 94 93/ 97 F4M4F3M3F2M2F1M1  52 2.3.3 EST Evidence Supports the Validity of Highly Diverged Sequences Identified in Urochordates Sequences identified from C. intestinalis and C. savignyi are substantially different from those of either invertebrates or vertebrates. Two sequence pairs, here named HCNa and HCNb to limit their association with any of the vertebrate isoforms, only share ~5055% identity or 70 75% conservation with either group. The third sequence, HCNc, is even less similar. It shares only 40% identity and 6164% conservation with all other sequences with the exception of Ciona HCNa, with which it shares a slightly higher identity. The three sequences were found in the genome databases of both Ciona species and the orthologs share a conserved identity of over 90%. This HCN sequence similarity between the two Ciona genome projects makes it very unlikely that any differences between genes or with sequences from either invertebrates or vertebrates are due to errors in sequencing. To further support their validity, HCN expressed sequence tags (ESTs) were found in the database of the Kyoto University Ciona cDNA project (Satou, et al., 2003) (http://ghost.zool.kyotou.ac.jp/indexr1.html). These ESTs span several exons and cover a large portion of the C. intestinalis sequences provided here, suggesting that the three C. intestinalis genes have been correctly identified and that their transcripts are expressed. A recent comprehensive analysis of the ascidian genome revealed the existence of these three putative HCN ion channels (HCN1 = HCNc, HCN2 = HCNa and HCN3 = HCNb) (Okamura, et al., 2005), however, the sequences themselves were not reported.  The sequences identified here were in turn used to correctly annotate those found in the C. savignyi genome. 2.3.4 Three Different HCN Duplication Events Occurred Prior to the Divergence of the Fish Lineage In order to examine the phylogeny of this family, a multiple sequence alignment of the core region of HCN was created using 52 of the 58 identified sequences (see methods). Three different phylogenetic programs, two NJ methods and one MP method, produced similar results. Due to the high sequence conservation among mammalian sequences, ML methods were incapable of analyzing the complete list of sequences. Using a subgroup of 41 sequences, ML methods produced similar results to the NJ and MP trees. Figure 2.3 is a rooted MP consensus tree produced by the PHYLIP program. The NJ rooted phylogram created by ClustalX and a  53 single representative ML phylogram produced by PHYLIP are provided in Appendix Figure 7.1 and Figure 7.2. Four sequences were included as an outgroup: human CNGA1 and CNGA3, human ERG1 and KAT1. Tree topology did not change when any of these outgroup sequences were removed.  54  Figure 2.3: A Maximum Parsimony (MP) Consensus Tree of the HCN Family. 52 HCN sequences were included in this analysis. The MP tree was constructed by resampling 100 datasets (bootstrap) with randomized input order of 10 jumbles using Seqboot, Protpars and Consense programs in the PHYLIP software. To control for length differences between the different isoforms, only the region between the start of S1 and the end of the CNBD was used. The tree is rooted with an outgroup of KAT1, hERG1, CNGA1 and CNGA3, 4 sequences that are known to be related to the HCN family. Similar to the previous studies, HCN2_urchin is shown to branch before the invertebrate clade and may be consequence of the MP method’s sensitivity to unequal rates of evolution. The six urochordate sequences from C. intestinalis and C. savignyi fall according to their evolutionary position, between the invertebrates and vertebrate species, and are supported by high bootstrap values. Because they group within their own species, they are not considered orthologous to any of the mammalian isoforms and are therefore named HCNa, HCNb and HCNc. HCN3 is shown to be the product of the first duplication event, however, a split between the fish and mammalian sequences is observed but the partition is not well defined. HCN4 is shown to be the product of the second duplication event, followed by the emergence of HCN2 and HCN1. Low bootstrap values within isoform clades are most likely a consequence of the high sequence identity within the region used and in turn, the lack of informative sites required by the MP method. Duplicate copies of mammalian paralogs in the fish species are denoted with ‘a’ or ‘b’ in their sequence names. * indicate sequences annotated in this study. Numbers indicate bootstrap values and represent % support for the respective partitions. (Jackson, et al., 2007)  55  100 hCNGA3 hCNGA1 KAT1 hERG1 HCN2_urchin HCN_urchin HCN_lobster HCN_mosquito HCN_silkmoth HCN_bee HCN_fly HCNb_c.intestinalis* HCNb_c.savignyi* HCNa_c.intestinalis* HCNa_c.savignyi* HCNc_c.intestinalis* HCNc_c.savignyi* HCN3_green puffer* HCN3a_fugu* HCN3b_fugu* HCN3_zebrafish* HCN3_opossum* HCN3_mouse HCN3_rat HCN3_cow HCN3_human HCN3_dog HCN4b_fugu* HCN4b_green puffer* HCN4_zebrafish HCN4a_green puffer* HCN4a_fugu* HCN4_opossum* HCN4_mouse HCN4_rat HCN4_rabbit HCN4_dog HCN4_human HCN2a_green puffer* HCN2a_fugu* HCN2b_green puffer* HCN2_zebrafish HCN2_frog* HCN2_rat HCN2_human HCN2_mouse HCN1_trout HCN1_fugu HCN1_green puffer* HCN1_mouse HCN1_rat HCN1_opossum HCN1_chimp HCN1_dog HCN1_human HCN1_rabbit Vertebrates Invertebrates Urochordates Outgroup HCN3 HCN4 HCN2 HCN1 100 100 100 96 83.7 100 72.1 56.6 97 71.5 94.9 69.7 56.5 100 94.5 100 98.5 25.1 26.4 62 29.5 31.1 95 48.1 93.4 41.2 44.5 100 92.9 66.1 41.5 38 41.3 30.6 100 100 75.8 92.5 37.1 28 33.3 100 100 82.7 98 98 100 51.3 46.9 100 100  56  HCN sequences did not group according to species but rather within isoform clades, indicating that the HCN14 isoforms are paralogs. Genes from fish, birds and amphibians partitioned within the four mammalian clades which strongly suggests that the four isoforms originated prior to the origin of the vertebrate clade and were present in the common ancestor of fish and tetrapods. From the tree topology, we also predict that an HCN ancestor underwent three separate gene duplication events prior to the divergence of the fish lineage, over 450 MYA. Our data are inconsistent with two rounds of wholegenome duplication and suggests that the HCN family evolved through independent gene duplications or chromosomal block duplication (Hughes and Friedman, 2003). However, due to a relatively low sequence resolution, the 2R hypothesis can not be ruled out. Of the four HCN isoforms, the clade belonging to HCN3 was the first to diverge and represents the product of the first duplication event. This position of HCN3 is supported by all four phylogeny methods and by a high bootstrap value, and supports the findings of Galindo and colleagues (Galindo, et al., 2005). Functional data will be required to further explain this branching position. The low bootstrap value at the division of the HCN3 clade in the MP and NJ trees indicates that this subdivision remains unresolved. We hypothesize that this clade division may be due to the independent sequence evolution which has occurred in the teleost and tetrapod HCN3 genes following the lineage divergence. Interestingly, the HCN3 clade is not divided in the ML tree for which the JonesTaylorThornton model of evolution was imposed. We can not rule out the possibility that this difference is due to the decreased sequences used in the ML analysis. The resolution of the HCN3 clade will only be improved with the inclusion of more complete sequences from species which evolved following the emergence of the four HCN paralogs but prior to the divergence of mammals. Functional data of the HCN3 channels from lower vertebrates will also be required to conclusively determine whether a functional divergence occurred within this clade. Our data also suggests that HCN4 is the product of the second duplication event followed by the emergence of HCN1 and HCN2. This proposed evolutionary pattern is evident in all three phylogenetic trees and is further supported by the sequence conservation pattern shown in Figure 2.2. Based on phylogenetic results, the division between HCN1 and HCN2 remains unresolved. Nevertheless, functional data for the mammalian isoforms suggests a clear difference between these two isoforms. The discrepancy with the predicted order of species evolution within each  57 mammalian clade is likely due to the high sequence conservation seen within this core region. This results in a limited number of informative sites and produces a low phylogenetic signal (Gallin, 1998). The lack of sequence divergence in the mammalian clades may be due to insufficient evolutionary time to permit the accumulation of mutations and/or a strong selective pressure to retain the conserved sequence of this channel. 2.3.5 Fish Lineage Show Evidence of Duplicate HCN Genes Genome data mining revealed multiple HCN genes in both the Fugu rubripes and Tetraodon species, in both cases exceeding the number found in the mammalian lineage. From the branching order in the phylogenetic tree shown in Figure 2.3, it is clear that fish gene pairs group together within the clades of individual mammalian orthologs. This suggests that the common ancestor of teleost fish and tetrapods had 4 HCN genes, and that these underwent further duplications independently in the fish lineage. These sequences have therefore been named according to their mammalian ortholog and subsequently designated ‘a’ or ‘b’. This pattern is clearly evident in the HCN2 and HCN4 clades, where the green puffer and fugu ‘a’ and ‘b’ sequences are grouped together and are separated from each other by high bootstrap values. Because the teleost fish are predicted to have undergone a complete genome duplication early in their own lineage ((Hoegg, et al., 2004) and references therein), this distribution pattern is not unexpected. However, because some of the genes identified were not full length and some showed evidence of intron insertion, further analyses is required to determine whether these duplicates are expressed and functional or have become pseudogenes (Vanin, 1985). 2.3.6 Ciona Genes Most Likely Arose Through LineageSpecific Duplication Events In contrast to the observed pattern of the multiple fish genes, the multiple copies of HCN sequences found within the urochordate species do not partition with any of the mammalian isoform clades. However, each gene did show a highly significant pairing between C. intestinalis and C. savignyi, indicating that these duplication events occurred before these two species diverged. The timing of this duplication is unknown, but with the presence of multiple HCN genes previously identified in the sea urchin, two different scenarios are possible. First, one duplication of an ancestral gene may have occurred prior to the divergence of the deuterostomes and given rise to the multiple HCN genes seen in the sea urchin and urochordate species. A lineage specific duplication then occurred in the urochordates to produce the third Ciona HCN  58 gene. Subsequently, the genes evolved rapidly and independently in the different lineages and thus no longer resemble each other or any specific HCN isoform. The loss of one ancestral gene and the duplication and diversification of the other, would then have given rise to the four isoforms now common to all chordates. A second, and more parsimonious, pattern of evolution is that lineagespecific duplication events of a single HCN ancestor occurred and gave rise to the three HCN homologs within the urochordate lineage. Further ancestral duplication events occurred within the vertebrate lineage, after the divergence of the urochordates and prior to the fish lineage, giving rise to the four known mammalian isoforms. Lineagespecific gene duplication in Ciona has also been shown for the evolution of sodium gene family. Two sodium channel genes have been identified but one possesses a sequence that has diverged considerably from both its paralogous pair and from the other known sodium channel gene sequences. The authors concluded that the duplication events occurred just prior to fish lineage (Lopreato, et al., 2001).  Similarly, independent lineagespecific duplication was suggested for the ankyrin gene family based on their phylogenetic results and the differences in gene sequences between Ciona and the vertebrate homologs (Cai and Zhang, 2006). The general branching order between the Ciona HCN homologs, in which HCNa and HCNc group together and HCNb is independent, is consistent among the different trees produced. Their position within the tree, however, is variable and is most likely a result of the different tree building methods used. It may also reflect the amount of lineagespecific sequence divergence that has occurred in these HCN genes which has caused them to evolve independently of both the invertebrate and vertebrate clades and has blurred their phylogenetic position. 2.3.7 Predicted Phylogenetic Patterns are Supported by Exon Boundary Structure Our phylogenetic analyses indicate that the four vertebrate isoforms arose via three duplications of an ancestral HCN gene. This is supported by the exon structure of their coding sequences. The four human HCN genes are located on different chromosomes: 1q21.2 (HCN3), 5p12 (HCN1), 15q24q25 (HCN4) (Seifert, et al., 1999) and 19p13 (HCN2) (Vaccari, et al., 1999). Using the genome databases, we found this pattern of differential localization for all vertebrate HCN genes, fish through mammals. Furthermore, the overall exon structure of the four isoforms has remained consistent since the duplication and divergence events (Figure 2.4). They are comprised of 8 exons, with highly conserved length and sequence in exons 2 through 7.  59 An exception to this is observed in both of the fish HCN3b genes which are predicted to have an intron of <75bp inserted in the middle of exon 2. Exon 1 and 8, which directly correspond with the distal N and Ctermini, vary in both length and sequence for all vertebrate genes analyzed. In addition, exon boundary positions are highly conserved throughout the vertebrate lineage. This suggests that the extant vertebrate HCN sequences are derived from a single ancestral gene that had an exon structure similar to current mammalian HCN genes and that the duplication events occurred after the intron positions were fixed in the linear sequence.  60    Figure 2.4: Conservation of Exon Structure Reveals Evolutionary Patterns of HCN Genes. The general regions of the HCN protein are outlined in the grey box, not drawn to scale. These include: the Nterminal region, six transmembrane segments (16), a pore region (P) that forms the pore and selectivity filter in the quaternary structure of the channel, Clinker, CNBD, and the distal Cterminus. Below, the vertical lines represent exon boundary position of the coding sequence: black lines indicate an exon boundary position that is conserved with the human HCN isoforms, dark grey lines indicated conservation with invertebrate boundary positions and light grey indicates boundary position specific to the individual protein. The exon number is shown above the horizontal black line and amino acid length of the protein corresponding to the exon is shown below. The exon structure of most vertebrate isoforms (HCN14) is conserved with that of human (exception is fish HCN3b genes, see text for details), consisting of 8 exons with 2 through 7 corresponding to the region between the start of S1 and the end of the CNBD. The sequence diverges in length between isoforms in the N and Ctermini. A general HCN_human structure is shown as a representative with length differences of the four isoforms in exon 1 and 8 indicated by a range. The total lengths of the predicted exons and the exon boundary structure is variable among the three different C. intestinalis HCN coding sequences identified in this analysis. HCNb_c.intestinalis is composed of 15 exons, sharing all vertebrate HCN channels exon boundary positions, with an additional six unique boundaries and one boundary shared with the invertebrate HCN proteins of the fly and mosquito. HCNa_c.intestinalis is composed of 10 exons but only 4 boundaries are conserved with vertebrates. HCNc_c.intestinalis is predicted to have 14 exons, and shows no similarity in exon structure or boundary position to either of the other Ciona proteins or vertebrate HCN. In general, the mosquito and fly proteins have a higher number of exons compared to vertebrate HCN, sharing only a few boundary positions. (Jackson, et al., 2007)  6 1  HCN 184_human (77481203aa) HCNb_c.intestinalis (804aa) HCNa_c.intestinalis (688aa) HCNc_c.intestinalis (608aa) 1 2 103 4 8 9 11 4541 53 44 49 40 56 19754 5 44 29 6 7 42 1512 13 14 38 45 27 1 2 103 4 5 6 7 61219 54 75 50 44 48 5243 42 8 9 1 2 83 4 5 6 7 1418143938262 54 73 49 80 56 2258488 HCN_fly (945aa) 1 2 83 4 5 6 7 268 50 26 104 32 29 4455 80 55 45 51 1312 106 9 10 11 HCN_mosquito (945aa) Nterminus 1 2 3 4 5 6 CterminusCNBDClinkerP 1 72 3 4 5 6 50 26 57 32 29 4179 80 55 45 46 12 1413 43 9 10 118 47 64 1 18 2 143 4 5 6 7 8 9 10 131211 29 492543 60 74 3750 54 4643 53 27  62  Our phylogenetic results suggest that the invertebrate and urochordate HCN genes are homologs of the four vertebrate isoforms but, because they group outside the vertebrate clades, they are not orthologous to any one in particular. The presence of multiple genes in the urochordate species, therefore, raised the question as to where and when duplications occurred and what course of evolution lead from the invertebrate to vertebrate gene structure. To address these questions, we compared the exon structures of invertebrates and urochordates to the highly conserved vertebrate structure. The exon structure of the single HCN gene in the arthropod lineage is quite different from the mammalian structure. Using the fly and mosquito genes as representatives, the invertebrate gene is comprised of 13 or 14 shorter exons, separated by short introns. Of the seven boundary positions conserved throughout the vertebrate lineage, only 24 are conserved in these species. The exon structures of the urochordate genes, on the other hand, show similarities to both the invertebrate and vertebrate genes. The HCNb gene contains 15 exons separated by short introns, which is similar to the invertebrate structure. Nevertheless, seven of the exon boundaries parallel all of those found in the vertebrate structure. The other two urochordate genes, HCNa and HCNc, are different from both the vertebrate and invertebrate genes. HCNa contains 10 exons and only four boundaries are conserved with the vertebrate structure and none with the invertebrate genes. The 13 exon boundaries of HCNc are not conserved with any other gene, including paralogs from within the same species.  In conjunction with sequence identity, residue conservation and phylogenetic analyses, these findings further support the hypothesis that HCNb represents the homolog most similar to the ancestral HCN gene of the urochordates, and that HCNa and HCNc arose through lineagespecific duplication processes. Furthermore, we suggest that the ancestral gene of the HCN family contained numerous introns which were then lost during the course of evolution. Exons became fused to form the structure now seen within the vertebrate lineage. 2.3.8 The Evolution of Key Residues in the Voltage Sensing Domain and Pore Region HCN channels open in response to changes in membrane voltage and allow for the passage of Na + and K +  ions across the plasma membrane. The transmembrane domains of the individual subunits, which form tetramers around a central pore, are primarily responsible for these functions. As might be expected from the natural constraints of the hydrophobic bilayer, sequence conservation is abundant in areas predicted to correspond to the six transmembrane  63 segments. Similar to depolarizationactivated K +  channels, the fourth transmembrane segment (S4) contains positively charged residues and is likely to sense changes in voltage across the cell membrane (Figure 2.5A, ‘a’). In contrast to depolarizationactivated K +  channels, however, HCN channels open instead of close in response to hyperpolarization. Interestingly, HCN channels possess an additional 4 or 5 charged residues at the Nterminal end of S4 (Figure 2.5A ‘b’). Together, in response to changes in voltage, these charges have been shown to move in the same direction as that in depolarizationactivated K +  channels (Bell, et al., 2004). Therefore, it has been suggested that the movement of the voltagesensing domain is coupled to channel opening in the opposite way (Mannikko, et al., 2002). Throughout the HCN family, there is high sequence identity in this region amongst the vertebrate isoforms and high conservation across all sequences. The more diverged sequences from the sea urchin and urochordate species, spHCN2, Ciona HCNa and Ciona HCNc, which are predicted to be the products of lineagespecific duplication, possess only three of the upstream positive charges. The effect of this loss of charge awaits functional characterization, but could be due to relaxed constraints in these duplicate genes.  64  Figure 2.5: Sequence Comparison of the Voltage Sensing Domain and Pore Region. A) Sequence of voltage sensing domain conserved throughout evolution. The fly, human and fugu sequences were chosen to represent arthropod, mammal and fish species, respectively. Sequences are organized into invertebrates, urochordates and vertebrate groups. A high sequence identity is notable amongst all vertebrate sequences. Shading indicates four levels of sequence conservation (see methods), excluding the assembly gap regions. Black stars indicate regions of assembly gaps in genome database and the arrow indicates the conserved exon boundary between mammalian exon 2 and 3. a = region of charge common to all potassium channel voltage sensing domains, b = region of charge unique to HCN channels and c = region associated with normal channel closure. B) A sequence comparison of the HCN pore region. The same representative sequences as in 5A were used, except HCN4_chicken. This was omitted due to a gap in the genome assembly across this region. The selectivity filter sequence CIGYG (a) is conserved throughout the HCN family, with the exception of two duplicate genes from sea urchin and urochordate species. The NXT/S glycosylation site (b) is located immediately prior to the pore helix and is conserved in urochordate and vertebrate sequences. Two conserved glycine residues (c and d) are located in S6, one of which was hypothesized to be the hinge that obstructs that pore. More recent evidence suggests that the hinge is located further downstream between the threonine and glutamine residues (e) at the Cterminal end of S6 in which a dipeptide is conserved throughout the entire HCN family. (f) The residue identified as responsible for the difference in chloride sensitivity in HCN2 vs. HCN1. (Jackson, et al., 2007)  65    Vertebrate HCN Urochordate HCN Invertebrate HCN * * b a c S48S5 linker S4 – voltage sensor Vertebrate HCN Urochordate HCN Invertebrate HCN Pore S6 ab c d ef  66  Based on their similarity to K +  channels (Yifrach and MacKinnon, 2002), we predict that the HCN channels are in their lowest energy state when closed and that the voltagesensors work to open the gate upon hyperpolarization of the membrane potential. The S4S5 linker couples the sensing of membrane voltage by S4 to the opening of the channel gate, which is located at the inner portion of the S6 segment. Channel opening is highly sensitive to mutations in the S4S5 segment. Within this region (Figure 2.5A ‘c’), a dipeptide of a tyrosine followed by an acidic residue is conserved across the vertebrate and urochordate channels, but is not present in the invertebrates. Mutation of the tyrosine residue to polar or acidic amino acids in the mammalian HCN2 isoform disrupts channel closure (Chen, et al., 2001a; Macri and Accili, 2004). Similar results were shown for the nearby arginine residue located at the end of this segment which is conserved throughout the HCN family. These experiments confirm that this region is critical for the regulation of channel opening by changes in voltage. The lack of complete conservation within this region therefore suggests that the strength of coupling may be variable among the different HCN genes. For example, spHCN1 channels exhibit an inactivation process that is due to a desensitization of the opening of the intracellular gate to voltage, which may reflect a weaker coupling of the voltage sensor and gate (Rothberg, et al., 2002). Based on sequence homology to the crystal structure of a bacterial K +  channel (Doyle, et al., 1998), the region between S5 and the end of S6 is believed to form the ion conduction pore in HCN channels. It contains a pore helix and selectivity filter and is involved in both ion selectivity and transport. In K +  channels, the selectivity filter exhibits the K +  signature sequence (GY/FG), which is thought to confer their K +  selective nature (Doyle, et al., 1998). In the HCN family, this tripeptide has been conserved but the channels allow the passage of significant amounts of both Na +  and K + . In all but two HCN genes, the sequence motif that corresponds to the selectivity filter is CIGYG (Biel, et al., 2002) (Figure 2.5b ‘a’). The conservation of this sequence only differs from K +  selective channels at the cysteine residue, implicating this site’s involvement in the reduced K +  permeability. The two exceptions are again found in gene duplicates from sea urchin and urochordate species. spHCN2 has a filter sequence of SIGFG (Galindo, et al., 2005), which makes it more similar to the filter sequences of channels in the EAG family. Functional data does not exist for the spHCN2 channel, so whether this difference affects ion selectivity is not known. In Ciona HCNc genes, the selectivity filter motif is CIGYS.  67 In mammalian HCN channels, substitution of the second glycine residue by serine (G404S in HCN2), reduced the slow activating current (Macri, et al., 2002). Evidence of channel function is required to determine if this residue difference in the urochordate channel is involved in gene silencing. If Ciona HCNc is functional, a different mutational tolerance at this particular site seems likely and may reflect an adaptive process that has enabled these channels to fill a different functional niche specific to these species. Recently, an Nlinked glycosylation site (NXT/S) located in the outer turret between the end of S5 and the pore helix (Figure 2.5b ‘b’) has been shown to play a role in membrane expression of mammalian isoforms (Much, et al., 2003). Similar to residues in the S4S5 linker, this motif emerges with the urochordate sequences and is conserved in two of the three urochordate genes and throughout all vertebrate sequences. To understand the necessity of this motif and the role it plays in channel function throughout evolution, further studies are needed to determine if glycosylation does occur in these Ciona sequences and if it, or some other compensatory posttranslational modification that allows the channel to mature through the ER/golgi process, occurs elsewhere in the invertebrate sequences. Based on the K +  channel crystal structure (Doyle, et al., 1998), the S6 segments of HCN likely form the inner vestibule of the pore and the gate. They show high sequence conservation with other protein relatives, such as ERG and CNG, and are almost completely conserved throughout the HCN family. Not surprisingly, the S6 segments of Ciona HCNa and HCNc and spHCN2, which are predicted to be the result of lineagespecific duplication, are the exceptions. Two glycine residues (Figure 2.5b ‘c’ and ‘d’) are completely conserved among the HCN genes. Based on sequence homology with voltagegated K+ channels (Shealy, et al., 2003), it has been suggested that one of these may form a glycine hinge involved in the opening of the channel gate in response to the movement of the S4S5 linker. However, more recent experimental and homology modeling evidence (Giorgetti, et al., 2005; Rothberg, et al., 2002; Rothberg, et al., 2003) has shown that a threonine residue, positioned after the glycines in the linear sequence and on the intracellular side of the channel, is only accessible in the open state. Furthermore, a glutamine residue at the end of S6 is accessible in both the open and closed positions, suggesting that the putative hinge position is located between these two residues (Figure 2.5b ‘e’).  68 2.3.9 The Evolution of the Cyclic Nucleotide Binding and Modulatory Domains Cyclic nucleotides are known to bind directly to the channel (DiFrancesco and Tortora, 1991) in the CNBD on the intracellular Cterminus and modify channel opening. The CNBD is an evolutionarily conserved domain that is found in several cyclic nucleotide binding proteins, including the bacterial catabolic activating peptide (CAP) and the protein kinase A (PKA) family (Berman, et al., 2005). The crystal structure of the Cterminus of mouse HCN2 has been solved (Zagotta, et al., 2003). Despite a low overall sequence conservation, the tertiary structure of its CNBD is highly conserved with the crystal structures of other CNBDs (Berman, et al., 2005). Furthermore, residues identified as being critical to structure and nucleotide binding are conserved throughout the HCN family (Figure 2.6). This includes the phosphate binding cassette (PBC), the most conserved feature of the CNBD which makes direct contact with cAMP (Diller, et al., 2001), and the hinge region, important for the capping of cAMP by the Chelix of the CNBD. The middle of the PBC has diverged in one of the sea urchin genes and all of the urochordate genes, but these residues are also variable in the other related crystal structures, undermining the importance of their identity to cyclic nucleotide binding. Because the HCN family possesses these conserved key residues, it seems probable that all of their CNBDs can stabilize the tertiary structure and bind cyclic nucleotides. One exception to this is seen in urochordate HCNc genes. These two sequences are missing a key hydrophobic residue in the PBC (Figure 2.6 ‘↑’) that forms a conserved interaction with cAMP (Passner, et al., 2000). In these genes, this residue is threonine, a hydrophilic amino acid that could disrupt this cAMP interaction. This difference is consistent with our hypothesis that the Ciona HCNc genes have undergone diversification following a lineagespecific duplication. Whether this difference also corresponds to a functional change in cAMP binding is an interesting question that will require functional studies to confirm.   69  Figure 2.6 Sequence Comparison of the Cyclic Nucleotide Binding Domain A sequence alignment of the CNBDs from representative sequences of the HCN family indicates that functionally important residues in the CNBD are conserved throughout evolution. Shading indicates four levels of sequence conservation (see methods). Evolutionarily conserved glycine residues known to be important for maintaining the fold of the βroll found in all CNBDs (located between A and B) and those residues known to interact with the cyclic nucleotide are indicated by stars below the alignment. The phosphate binding cassette (PBC) and hinge region common to all CNBDs are identified by the black lines below the sequence. The arrow indicates the hydrophobic site where divergence has occurred in Ciona HCNc genes (as explained in text). Cylinders above the sequence represent helices and arrows = bstrands that together form the B roll. Cyclic nucleotides are known to bind to the second conserved arginine in the PBC. (Jackson, et al., 2007)  7 0  PBC Hinge * * * * ***** * * *** * B CPA ↑  71  The effects of cAMPbinding on HCN channel opening are variable among isoforms and depend on the CNBD as well as the Clinker region which connects it to the transmembrane domain. For the mammalian HCN1, HCN2 and HCN4 isoforms, the binding of cAMP relieves a tonic inhibition of the transmembrane domain by the CNBD (Wainger, et al., 2001). The extent to which this relief occurs (HCN4 = HCN2 > HCN1) depends in part on the Clinker  region which transmits the effect of cAMP onto the gating machinery (Wang, et al., 2001). In contrast, cAMP binding has no effect on mouse HCN3 (Mistrik, et al., 2005) and enhances the inhibition of channel opening of human HCN3 (Stieber, et al., 2005). Whether the sequence differences within the Clinker are responsible for these differences is not known. Further complicating the issue is that the effects of cAMP also depend upon regions in the transmembrane domains (Wang, et al., 2001) which may lead to more complex interactions with the cytoplasmic portions of the gating elements. The complexity of these interactions is especially apparent in the spHCN1 channel in which channel inactivation/desensitization after first opening is eliminated by cAMP binding (Shin, et al., 2004). Clinker sequences show considerable divergence throughout the HCN family, including differences among mammalian paralogs, and even between mammalian and nonmammalian orthologs. This divergence is consistent with the variability in cAMPmediated effects on channel opening. Overall, the role of the Clinker in HCN channels is unusual compared to other regions in the channel. Its sequence and function show variation throughout the family, but it connects two domains that are themselves highly conserved in sequence throughout evolution and carry out highly conserved, but distinct, functions in a cooperative manner (transmembrane = voltage sensing, channel opening and ion permeation, CNBD = cAMP binding and modulation of channel open by the transmembrane domain). Because of its position between these two domains, the sequence variability in the Clinker may be, in part, the result of an adaptive process that has enabled the diversification of cyclic nucleotide binding and the modulation of channel opening by the CNBD within the HCN family. 2.3.10 Sequence Variability and Functional Divergence of Vertebrate HCN Paralogs From our data, we predict that the four vertebrate isoforms are paralogs of each other resulting from three duplication events which occurred before the divergence of the teleost and tetrapod lineages. At the time of their origin, the gene pairs produced by these events would have  72 been functionally redundant. Because the vast majority of duplicate genes are silenced throughout the course of evolution (Lynch and Conery, 2000), the retention of all four is probably due to the acquisition of unique functional characteristics and/or expression patterns which result from tolerated mutations specific to each paralog (Sidow, 1996). Throughout the core region, there is high sequence identity among the four vertebrate HCN isoforms. Some of these invariant residues probably contribute to functional properties common among all vertebrate channels. In contrast, some of the residues that vary among paralogs within this conserved region probably contribute to the more subtle isoformspecific differences in function (Li and Gallin, 2005). The search has begun to identify which residues underlie differences in function among the four mammalian isoforms but approaches have been limited to direct sequence comparisons followed by single site mutagenesis and/or construction of chimeras, and functional assays. Using these approaches, three studies have identified residues or regions responsible for phenotypic variation among the four mammalian HCN channels. First, differences in the rates of channel opening and cyclic nucleotide modulation between the HCN2 and HCN4 isoforms were localized to a variant residue in the S1 segment (Stieber, et al., 2003b). Second, a single residue difference in the inner selectivity filter was shown to confer chloride sensitivity to the HCN2 channel (WahlSchott, et al., 2005) (see Figure 2.5b). Lastly, differences in the Clinker were shown to contribute to differences in cAMP efficacy between HCN1 and HCN2, although the specific residues involved were not identified (Wang, et al., 2001). From these few examples, it is clear that the functional consequences of residue variation among the four vertebrate isoforms, which may encompass not only overt differences in channel opening and closing but also differences in permeation, cyclic nucleotide affinity, homo and hetero tetrameric assembly, glycosylation status, proteinchannel interactions and abilities to traffick to the cell surface, cannot be easily predicted simply by its location within the channel. Furthermore, differences in function may involve multiple residues and domains located throughout the channel. Therefore, sitedirected mutagenesis and/or the construction of chimeras between vertebrate isoforms may not be sufficient to determine all differences, especially in the absence of high resolution three dimensional structures for each of the four paralogs. In this study, we expanded the list of sequences for each of the four HCN paralogs by the addition of sequences from lower vertebrates. By broadening the evolutionary representation of the four vertebrate isoforms, the sequence signaltonoise ratio is improved and the identification  73 of residues that are conserved among orthologs, but differ among paralogs, is enhanced. However, genes from lower vertebrates (eg. fish) and mammals have continued to evolve under different selective pressures, since their divergence from a common ancestor, several million years ago. Therefore, information derived from this expanded list of sequences is complex. We identified three general groups of residues based on similarities among vertebrate orthologs. First, sites may be conserved among orthologs and could contribute to a function specific to each vertebrate isoform. Second, sites may be conserved in the mammalian and fish orthologs but differ between these two groups. These sites may contribute to functional differences between the mammalian and fish channels of a particular isoform. Finally, sites may be conserved in orthologs of mammals or fish, but not both. These sites could be involved in speciesspecific functions or paralogspecific functions within each species. Alternatively, they may be the result of relaxed evolutionary constraints. By analyzing these different sets of conserved sites together with functional characterization of the various channels belonging to each of the vertebrate isoforms, sites that may contribute to paralogspecific differences in channel function can be more easily identified. An informative subset of sites identified when comparing HCN genes from lower vertebrates and mammals are those that are conserved with a paralog rather than its own ortholog. If we assume that the probability of identical sites mutating to the same residues independently following duplication and species divergence is low, then these probably represent sites that have been retained from ancestral genes. Differential retention of sites from ancestral genes among orthologs suggests that channel phenotypes may not be completely conserved among them. In conjunction with the identification of conserved and nonconserved sites, the functional analysis of HCN channels from lower vertebrates, and comparisons of their properties with those of mammalian channels, will help to identify residues that contribute to different phenotypes, and will also have the potential to shed light on the sequences and functions of ancestral HCN channels. 2.3.11 Vertebrate IsoformSpecific Alignments of Sequences Spanning 450MY, Reveal Conserved Motifs in the N and CTermini The large number of vertebrate HCN sequences assembled in this study has greatly increased the power to identify isoformspecific motifs in the N and Ctermini that may  74 contribute to unique functions and specific patterns of expression within cells and tissues. The high sequence conservation among the four vertebrate isoforms extends beyond the core region analyzed above and includes regions just upstream of the S1 segment and downstream of the CNBD. The more distal N and Ctermini, however, vary in both length and sequence among paralogs and do not align well. The sequences of the mammalian orthologs identified prior to this study were too close in evolutionary time to reveal sequence divergence in the distal N and C termini. On the other hand, the four paralogs from a single species were too diverged to align reliably. By expanding the taxa sampling to include species from different time periods of vertebrate evolution, the signaltonoise ratio that is inherent in the sequence information was considerably improved (Anderson and Greenberg, 2001). Alignments of the distal termini, using this expanded list of vertebrate sequences, revealed several isoformspecific blocks of conserved sequence interspersed with diverged regions of variable length (Figures 2.7 and 2.8). In the distal N and C termini, multiple regions specific to each of the four vertebrate paralogs were identified and were seen to be interspersed with regions of variable length and sequence. The conservation of these motifs throughout 450MY of evolution highlights a selective constraint placed upon them and suggests that they confer isoformspecific properties. These properties remain to be identified but there is some evidence to support a role for these motifs in the regulation of cell surface expression. For example, a single block of five residues is conserved in the Nterminus of HCN2 in fish, frog, chicken and mammals (Figure 2.7 ‘a’). This motif is not analogous to any known consensus sequence. However, the deletion of the first 130 amino acids of the mouse HCN2 Nterminus, which includes this conserved block of five amino acids, reduces cell surface expression and has only minor effects on channel opening and closing (Tran, et al., 2002). A second example is a block of residues found in the Cterminus of HCN1 channels (Figure 2.8 ‘a’). This region is responsible for binding of filamin A to HCN1, and not the other isoforms (Gravante, et al., 2004), and its removal abolishes the effects of filamin on channel cell surface expression. Overall, the motifs conserved within the termini of an individual isoform are more likely involved in roles specific to the function or expression of that paralog or may be involved in protein interactions specific to the tissues where these proteins are expressed. The regions of variable sequence and length observed between these conserved motifs result from DNA insertions and/or deletions. This variability may represent either relaxed constraints and/or the beginnings of speciesspecific adaptation processes. These adaptations may involve  75 properties such as temperature sensitivity (Marshall, et al., 2002; Marshall, et al., 2005) or binding to regulatory proteins with speciesspecific sequences.  76 Figure 2.7: Individual Isoform Alignments of the Distal NTermini of HCN Channels. Sequence alignments of available vertebrate sequences for the Nterminal region of three of the four individual isoforms were produced using ClustalX and edited and shaded using GenDoc. Shading indicates four levels of sequence conservation (see methods). a = example of conserved motif in vertebrate HCN2 sequences that may play a role in channel surface expression. b = region conserved amongst all vertebrate HCN sequences that may be involved in tetrameric interactions in HCN2 and may play a similar role in HCN1, 3 and 4. Arrows between conserved blocks represent variable regions with the range of amino acid length variation listed above. Numbers at the start and end of the sequences correspond to their position in the full length sequences (Jackson, et al., 2007).  7 7    22829 AAs 20841 AAs 718110 AAs 18835 AAs 7818 AAs 21849 AAs ba  78    Figure 2.8 Individual isoform alignments of the distal Ctermini of HCN channels. Sequence alignments of all available vertebrate sequences for the Cterminal region of three of the four individual isoforms were produced using ClustalX and edited/shaded using GenDoc. Shading indicates four levels of sequence conservation (see methods). a = example of conserved motif in vertebrate HCN1 sequences has been shown to interact with Filamin A. b = region conserved amongst all vertebrate HCN sequences downstream of the CNBD. c = tripeptide sequence shown to be involved in proteinprotein interactions with scaffolding proteins in vivo and to regulate cell surface expression. Dotted line indicates conserved region amongst four vertebrate isoforms previously identified but of unknown function. Arrows between conserved blocks represent variable regions with the range of amino acid length variation listed above. Numbers at the start and end of the sequences correspond to their position in the full length sequences (Jackson, et al., 2007).   8814 AAs 1918560 AAs 668112 AAs b c B 21835 AAs 548103 AAs b c 578113 AAs 13814 AAs 10811 AAs 29841 AAs b c a  79  Finally, both the N and Ctermini possess regions of high sequence conservation among the four vertebrate isoforms, in addition to those of the core region used for the phylogenetic analyses. These regions probably confer properties that are not unique to any individual isoform, but are important for all vertebrate HCN channels. In the Nterminus, a region of approximately 50 residues immediately upstream of the start of S1 is conserved in all four vertebrate isoforms (Figure 2.7 ‘b’, HCN3 not shown). In mouse HCN2, this region interacts with itself and thus may be involved in intersubunit interactions of tetrameric assembly (Tran, et al., 2002). Furthermore, the removal of this region, along with the rest of the Nterminus, prevents the formation of functional channels. Based on the high level of sequence conservation among the four isoforms, it seems probable that this region provides similar interactions for HCN1, 3 and 4. If this is true, the few differences among paralogs within this region may modify intersubunit interactions and thus regulate homomeric and/or heteromeric assembly. In the Cterminus, a block of residues conserved among the vertebrate HCN1, 2 and 4 isoforms was identified immediately downstream of the CNBD, which corresponds to the start of the last exon (Figure 2.8 ‘b’). Deletion of this region, together with the entire distal Cterminus and the Chelix of the CNBD, does not affect HCN1, HCN2 or HCN4 channel cell surface expression or function in heterologous systems (Proenza, et al., 2002b; Wainger, et al., 2001). A function for this conserved block of residues is not known but, whatever this function may be, based on the lack of sequence conservation, it is not likely possessed by HCN3 channels. Finally, a motif found at the distal end of the Cterminus, SNL, is conserved in HCN1, 2 and 4 (Figure 2.8 ‘c’). This motif, which qualifies as a consensus PDZbinding domain, interacts with PDZ containing proteins in vitro (Kimura, et al., 2004), and also with the TRP8 protein in vitro and in vivo, where it regulates channel cell surface expression (Santoro, et al., 2004). The absence of this motif in the HCN3 genes suggests that either this isoform does not interact with PDZ containing proteins or that these interactions take place at different locations within the channel. 2.3.12 Summary and Perspectives The availability of an increasing number of genome sequences has enabled us to generate a list of putative HCN coding sequences that has doubled the number of those previously known and covers a significantly greater portion of evolutionary time. With this improved species representation, we were able to more accurately complete sequence comparisons, phylogenetic  80 analyses and exon structure comparisons of the HCN gene family and put forward a model of its molecular evolution. We propose that the vertebrate isoforms evolved from a single ancestral sequence that had an exon structure similar to current mammalian HCN genes and that the duplication events occurred after the intron positions were fixed in the linear sequence. We also suggest that HCN duplications have occurred independently in the sea urchin, urochordate and fish lineages. Increasing the evolutionary distance between the sequences for each HCN isoform provided a good contrast and enabled us to unmask and identify regions putatively important for isoformspecific, as well as speciesspecific, functions. Together, our study provides a strong basis from which to refine the proposed model of evolution as more genomes become available and as the functional analysis of HCN genes progresses. Finally, our study provides a valuable tool to aid in the planning of experiments designed to probe the relationship between structure and function of HCN channels, and to determine the functional significance of sequence similarities and differences among them.   81 3. DUPLICATION AND ASYMMETRIC DIVERSIFICATION OF HCN CHANNEL ISOFORMS IN CIONA INTESTINALIS2 3.1 Introduction HCN channels are voltagegated cation channels with an unusual set of functional properties that make them critical for regulating membrane potential in electrically active cells such as neurons and cardiomyocytes (Robinson and Siegelbaum, 2003). Principally, HCNs are opened by hyperpolarization of the membrane potential; allow both sodium and potassium ions to pass and most isoforms open more easily when cyclic AMP is bound to their cytoplasmic surface. To date, HCNs have been cloned from mammals (HCN1HCN4) (Ishii, et al., 1999; Ludwig, et al., 1998; Santoro, et al., 1998), arthropods (one gene) (Gisselmann, et al., 2005; Gisselmann, et al., 2003; Marx, et al., 1999; Ouyang, et al., 2007) and sea urchins (two genes) (Galindo, et al., 2005; Gauss, et al., 1998).   All known HCN genes are predicted to form protein subunits that will be comprised of a voltage sensing domain, an ion conducting pore region and a cyclic nucleotide binding domain in the Cterminus. Functionally, HCNs cloned from mammals and invertebrates differ quantitatively (Baruscotti, et al., 2005). Among the four mammalian HCNs, opening occurs at different rates and, notably, is facilitated strongly by cAMP in only two isoforms (HCN2 and HCN4) (Altomare, et al., 2001; Chen, et al., 2001b; Mistrik, et al., 2005; Moosmang, et al., 2001; Santoro, et al., 2000; Stieber, et al., 2005). The one sea urchin isoform whose function is known opens and closes comparably quickly (Gauss, et al., 1998) but its internal gate recloses during hyperpolarizationinduced activation (Shin, et al., 2004); this slippage of the activation gate does not occur when cAMP is bound. Other invertebrate HCNs display a more vertebratelike phenotype; they do not reclose and open more easily when cAMP is elevated (Gisselmann, et al., 2005; Gisselmann, et al., 2003; Ouyang, et al., 2007). Finally, the vertebrate, but not invertebrate, HCNs also contain a putative Nlinked glycosylation site near the pore (Jackson, et al., 2007), which has been shown to control cell surface expression variably among the four mammalian isoforms (Hegle, et al., 2010; Much, et al., 2003).  2  A version of this chapter will be submitted for publication. Jackson, HA, Hegle, A, Nazzari, H, Jegla, T, Accili EA (2011). Duplication and asymmetric diversification of HCN channel isoforms in Ciona intestinalis.  82 Ciona intestinalis is a marine organism whose tadpolelike larval stage undergoes a phase of metamorphosis and gives way to a sessile adult stage adapted for filter feeding (Suzuki, et al., 2005). They are a deuterostome and are classified as a marine invertebrate chordate, or urochordate, (Satoh, et al., 2003), due to the presence of a notochord at the tadpole stage. C. intestinalis displays a simple adult body plan, which includes a heart with an open circulatory system, a neural complex, and digestive and reproductive organs. This organism diverged from the vertebrate lineage over 550MYA and is part of the chordate phylum along with other urochordates (ascidians, larvaceans and thaliaceans), cephalochordates (amphioxus) and vertebrates (Delsuc, et al., 2006). Urochordates are now thought to be the closest chordate relatives to vertebrates, supplanting the fishlike amphioxus from this privileged position (Delsuc, et al., 2006; Delsuc, et al., 2008). Thus, urochordates likely provide the closest ancestral genetic reference point prior to the duplication and diversification events that have occurred in vertebrate genomes (Corbo, et al., 2001). A comprehensive analysis of the C. intestinalis genome revealed 160 ion channel genes with homologs in mammals (Okamura, et al., 2005), which included a minimal set of voltage gated ion channel genes. Among these voltagegated channels were three HCN channels (Okamura, et al., 2005) (named ciHCN1, ciHCN2, and ciHCN3), each of which have supporting EST evidence and suggest an in vivo physiological role in the ascidian species. More recently, a study identified HCN channels in a related tunicate species, Botryllus schlosseri, and found high expression patterns in the heart tube and a physiological role in its cardiac pacemaking activity was revealed (Hellbach, et al., 2011). Previously, we identified and analyzed the HCN homolog sequences in two tunicate genomes, C. intestinalis and C. savignyi (Jackson, et al., 2007), which display high sequence identity between species and phylogenetically group together as an independent clade, rather than as orthologs of the four vertebrate isoforms. To avoid reference to the vertebrate homologs, we termed these channels ciHCNa (ciHCN2), ciHCNb (ciHCN3) and ciHCNc (ciHCN1). The branching pattern suggests that the three Ciona HCN genes arose via lineagespecific duplications prior to divergence of C. intestinalis and C. savignyi. Based on exon structure, sequence identity and phylogenetic position, we suggested that ciHCNb is evolutionarily closest to the common ancestor, while ciHCNa and ciHCNc arose through subsequent lineagespecific duplications (Jackson, et al., 2007). ciHCNa and ciHCNb share approximately 55% sequence identity with invertebrate and vertebrate HCNs, as well as with  83 each other. ciHCNc shares approximately 40% and 60%, with invertebrate and vertebrate HCNs, respectively. Although the specific function of the three Ciona HCN channels was not known, we suspected that the function of ciHCNa and ciHCNb would be similar to the vertebrate HCNs because the narrowest part of the pore regions, responsible for ion selectivity and flow, contain the identical amino acids. But, which of these two channels will be the most similar to the vertebrate isoforms is hard to predict based on overall sequence conservation. Although ciHCNb possesses an exon structure more similar to the vertebrate HCNs, ciHCNa possesses the pore associated Nglycosylation sequon. Here, we show that ciHCNa function is most similar to the mammalian HCNs and also undergoes Nglycosylation at a sequon near the pore. ciHCNb does not undergo Nglycosylation, lacks the poreassociated sequon, and possesses a unique phenotype, in which the channels remain partially open across all voltages tested. We also found that the Nglycosylation sequon is retained in the one HCN gene of Oikopleura dioica, a urochordate closely related to Ciona, but is lacking in the single HCN gene of Branchiostoma floridae, a representative of cephalochordata, also known as amphioxus or lancelet. Together with a refined sequence analysis and comparison, the data support close phylogenetic ties between urochordates and vertebrates and suggest duplication of a vertebratelike ancestral gene in Ciona to yield two genes that share roughly the same sequence identity with vertebrate homologs: one with minimally evolved functions and another in which the gate is half open and the poreassociated Nglycosylation sequon has been lost. 3.2 Materials and Methods 3.2.1 HCN Sequence Collection and Analysis With the exception of the putative lancelet and Oikopleura HCN sequences, the collection of HCN sequences used in this analysis has been previously described (Jackson, et al., 2007)}. Using hHCN2 as a query, the Oikopleura and Lancelet sequences were obtained by a BLAST/BLAT search of their respective genomic databases (Oikopleura dioica v1 at www.genoscope.cns.fr) (Karolchik, et al., 2008). A segment of the genomes near the search hits was downloaded and run through a genewise prediction program. Translated protein sequences were aligned with ClustalX (Thompson, et al., 1997), and the NH2 and COOH terminal regions  84 were subsequently removed. The region between the S1 and the end of the CNBD were used to generate the cladogram displayed with TreeView (Page, 1996). 3.2.2 Cloning and Epitope Tagging of Two Ciona HCN Channels We initially cloned ciHCNa and ciHCNb because, unlike ciHCNc, both retain pore sequences that are highly conserved among all known HCNs. However, only ciHCNa possesses a putative Nglycosylation sequon near the pore, as do most other vertebrate sequences. Sequences were identified by BLAST search of the Ciona intestinalis genome sequence (Dehal, et al., 2002) using mouse HCN1 as a query. The ciHCNa coding sequence was determined by comparison of gene predictions and sequenced cDNA clones, and cloned by standard RTPCR from Ciona intestinalis total RNA isolated from adult siphon. Briefly, 5 µg of total RNA was reverse transcribed using Superscript III (Invitrogen, Carlsbad, CA) and oligodT priming. 1/50 th  of this reaction was used for each RTPCR. The gene was amplified in two pieces, assembled by standard overlap PCR and cloned into pOX for expression in oocytes (Jegla and Salkoff, 1997; Zhang, et al., 2009a), or pCDNA3.1 derivative, which expresses eGFP off the same transcript under control of the CMV promoter for expression in mammalian cells. The 5’ end of ciHCNb could not be determined from gene predictions or cDNA clones and was identified by 5’ RACE PCR using the Marathon cDNA synthesis kit (Clontech, Palo Alto, CA) with total siphon RNA as a template. Full length ciHCNb clones for expression were then assembled from three pieces by overlap PCR and cloned as described above for ciHCNa. 4 separate clones were sequence verified for ciHCNa and ciHCNb; only clones that fully matched the consensus amino acid sequence determined for each gene were used for functional experiments. A Cterminal V5tagged ciHCNa and ciHCNb was constructed by overlapping PCR mutagenesis and digested back into the original pOX vector for oocyte expression or into a pcDNA3.1 myc (Invitrogen) vector using BamHI and Not1 restriction enzymes. The N380Q ciHCNaV5 mutant was then constructed using QuickChange mutatgenesis. Resulting sequences were confirmed by automated sequencing facility (UBC). 3.2.3 Identification of NGlycosylation of Ciona HCN Channels in Xenopus Oocytes and Chinese Hamster Ovary Cells To identify Nglycosylated protein, channels were expressed in Xenopus oocytes or Chinese Hamster ovary cells. Expression plasmids were linearized with Not1 and mRNA was  85 produced by in vitro transcription from the T3 promoter (mMessagae mMachine kit, Ambion). Ovarian lobes were surgically removed from adult Xenopus laevis and separated into small clumps of 510 oocytes. Oocytes were washed with OR2 solution (82.5mM NaCl, 2.5mM KCl, 1mM MgCl2 and 5mM HEPES, pH 7.6 with NaOH) and dissociated using 2mg/ml collagenase 1A (Sigma Aldrich) in OR2 for 1hr. Oocytes were then washed 3x with OR2 solution, followed by 3x with OR3 solution and incubated in OR3 solution for at least 1hr. Mature decollagenated oocytes were injected with 50 nl of mRNA (550 ng total) of the indicated RNAs and maintained in OR3 media at 18C. 48 hrs postinjection, 4050 oocytes for each condition were homogenized on ice in HEDP buffer (100 mM HEPES, 1mM EDTA, pH 7.6) with protease inhibitors (10 Wg/mL each of aprotinin, pepstatin, and leupeptin and 1 Roche protease inhibitor tablet per 10 mL) and centrifuged 2x at 6,000 RPM for 2 min at 4C. Supernatants were then overlaid on a 15% sucrose / HEDP cushion and ultracentrifuged at 50,000 RPM for 90 min at 4C to isolate membrane fractions. Pellets were resuspended in HEDP buffer and protein concentrations were determined by Bradford assay. Chinese hamster ovaryK1 (CHO) cells (American Tissue Type Culture Collection) were maintained at a subconfluent density in F12 media supplemented with 10% FBS and were transiently transfected with indicated cDNAs with FuGENE, according to manufacturer protocols (Roche). After 24 hrs, cells were washed with PBS and lysed for 30 min in RIPA buffer containing, in mM: 50 Tris (pH 8.0), 150 NaCl, 1 EDTA, 1 PMSF, 2 Na_3 VO_4 , 2 NaF, 1% NP40 and 10 Wg/mL each of aprotinin, pepstatin, and leupeptin. Protein concentration was determined by Bradford assay (BioRad). Proteins were resolved using SDSPAGE and transferred to polyvinylidene difluoride (PVDF) membranes. Blots were probed with antibodies for V5 (1:2000) and actin (1:5000), followed by HRPconjugated secondary antibody (1:2000), and visualized with ECL (Amersham). For PNGaseF assays (New England Biolabs), extracts were first incubated with PNGaseF for 1 hr at 37°C before SDSPAGE. 3.2.4 Electrophysiological Analysis of Ciona HCNs in Xenopus Oocytes In oocytes expressing either Ciona HCN isoform, ionic current  was recorded with the twomicroelectrode voltage clamp technique  using an OC725C voltage clamp (Warner Instruments, Hamden,  CT), 13 days after injection. Borosilicate glass microelectrodes (Sutter Instruments Co.) were filled with 3M KCl and had a resistance of 0.5 to 1.5 Mohms. Liquid  86 junction potential was compensated prior to oocyte impalement. Data were acquired using an amplifier and 1440 analogue/digital converter, along with pClamp 10.0 software. All experiments were carried out at room temperature. Analysis was done using Clampfit 10 (Molecular Devices), Origin (Microcal) and GraphPad software. Oocytes were perfused with a high K +  solution (96mM KCl, 2mM NaCl, 2mM MgCl 2  and 10mM HEPES) to maximize current amplitude. For both channels, I/V relationships were determined by a multistep protocol. Oocytes were held at 10mV and given a preliminary 500ms pulse to a range of test voltages from +30mV to 70mV in 20mV increments, followed by a 3 second holding step at 10mV, a 3 second prepulse to 70mV and a similar 500ms test pulse from +30mV to 70mV in 20mV increments. To obtain the amplitude of slowlyactivating If, ‘leak’ values from the preliminary pulse (Iinst) were subtracted from the instantaneous current amplitudes for each potential and plotted against voltage. Reversal potential and slope conductance for Iinst and If in individual oocytes were determined by linear regression of the linear portion of the I/V curve that crosses the xaxis. Average values are reported at mean ± SEM and compared with an unpaired ttest (p<0.05). PNa/PK ratios were determined using the GoldmannHodgkinKatz equation (Hille, 2001)  Erev = RT/F ln ((PK[Ko]+PNa[Nao])/(PK[Ki]+PNa[Nai]))   (Equation 1)  and the calculated average reversal potentials in low (5mM) K +  solutions. The intracellular concentrations of  K +  and Na +  were assumed to be 102 and 10mM, respectively (Weber, 1999) and the additional assumption was made that chloride does not contribute to the current. Deactivation kinetics were recorded in the high K +  solution and were fit with a single exponential. Low K +  solutions contained 5mM KCl, 93mM NaCl, 2mM MgCl 2  and 10mM HEPES). The voltage dependence of activation was determined in a high K +  solution using a staggered protocol. Tail currents were measured at 30mV (ciHCNa) or +30mV (ciHCNb) after hyperpolarizing test pulses of varying lengths to ensure more complete current activation from  100mV (ciHCNa) and 80mV (ciHCNb) in 10mV increments. Each test pulse was followed by a sufficient time to allow the current to return to baseline levels. Tail currents were measured as the difference between the peak of the tail and the steady state value at the end of the tail decay  87 to account for a nonzero holding current. Normalized tail currents were plotted as a function of test voltage as provided from voltage clamp data and individual activation curves were fitted by the following Boltzmann function (GraphPad),  f(V) = Imax/(1 + e (V1/2V)/k) (Equation 2)  Where V1/2 is the midpoint activation voltage and k is the slope factor (mV). V1/2 and k values are reported as mean ± SEM. Time constants to assess the rates of activation were measured over the entire duration of the pulse for all voltages between 80 and 40mV and fit with either a single or double exponential following an initial delay (Altomare, et al., 2001; Santoro, et al., 2000). The effects of cAMP on the voltage dependence of activation were established using 10mM 8bromo cyclic AMP (Sigma Aldrich) (Ouyang, et al., 2007). Eggs were incubated in the low K +  solution previously described, with or without 10mM 8bromo cAMP, for 45min. Following incubation, the staggered protocol was run in a high K +  solution containing the 10mM 8bromo cAMP. For pharmacological blocking experiments, oocytes were held at 10mV and given a single 6 second pulse to 70mV and back to 30mV for 10 seconds. Oocytes were then perfused with the blocking solution (dissolved in high K +  ND96 solution) and incubated for 3 minutes (5mM Cs + ) or 15min (100WM ZD7288). Following incubation, a second test pulse was given; current amplitudes before and after drug application were statistically compared using a paired t test. Ion selectivity experiments were performed using a similar twostep protocol as described for the reversal potential calculations in high and low K + . Oocytes were perfused with one of five solutions that were made up of: 20KCl/80XCl, 2MgCl 2  and 10mM HEPES, where X was: K, Na, NMDG, Li or Cs. All solutions were pH 7.4 with their respective XOH, except for NMDG which was titrated with HCl. Linear regression was used to generate individual reversal potentials and slope conductances; the average values (± SEM) for each condition were compared using a oneway ANOVA and Tukey’s post test (p<0.05).  88 3.3 Results 3.3.1 Molecular Cloning Two urochordate cDNA homologs of the mammalian hyperpolarizationactivated cyclic nucleotide modulated (HCN) channel family were cloned from the adult siphon of Ciona intestinalis using genomic comparisons and PCR strategies. We have termed these clones ciHCNa and ciHCNb to adequately differentiate them from the HCN14 mammalian isoforms and they have been deposited in GenBank (Accession numbers: ciHCNa JN815316 and ciHCNb JN815317). These channels share approximately 50% sequence identity with the mammalian isoforms between the beginning of the first transmembrane segment and the end of the cyclic nucleotide binding. ciHCNa is 671aa long with a predicted molecular weight of 78.8 kDa. When compared to the predicted gene XP_002131312 (JGIv1 ID: ci0100152577), named from manual annotation as ciHCN2, the cloned ciHCNa protein is 17 amino acids shorter due to a deletion in the Cterminal region of the channel. This deletion, therefore, represents a splice variant of this channel. Otherwise, the cloned sequence is 97% identical to that predicted in the genome and 90% identical with the putative sequence from C. savignyi between the S1 and the end of the CNBD. The cloned sequence of ciHCNb is 802aa long with a predicted molecular weight of 93.3 kDa and is also 97% identical to the genome predicted sequence (JGIv1 ID: ci0100130432) and 93% identical with that from C. savignyi. The two genes are approximately 55% identical to each other within this conserved region and show less conservation with the putative HCNc gene. Interestingly, when the orthologs from C. intestinalis and C. savignyi are analyzed for selection using MEGA 3.1 software, results indicate that all three gene pairs are undergoing independent purifying selection. That is, using the Fisher’s exact test for positive selection on the DNA alignments, a probability value of 1.0 is returned. This value is assigned by the program to indicate purifying selection based on the ratio of nonsynonomous and synonomous changes (dN/dS) values. 3.3.2 Phylogenetic Pattern and Tracking of the Pore NLinked Glycosylation Sequon Suggests Selective Loss of this Modification Among Ciona HCNs A striking finding of our previous studies was that an Nlinked glycosylation site (NXS/T), in the extracellular region near the pore, is found in HCNs of vertebrates and two of  89 three Ciona HCNs, but not in cnidarians, arthropods, annelids or mollusks (Hegle, et al., 2010; Jackson, et al., 2007). Here, we refine the phylogenetic pattern of HCN evolution among chordates, by generating an unrooted cladogram of the HCN family that includes sequences from the tunicate Oikoplerua dioica and from amphioxus, Brachio cephalochordata. The cladogram shows that the sole HCN channel from B. cephalochordata falls between that of urochordates and echinoderms, while an alignment of key sequences shows it does not have the poreassociated N glycosylation sequon found in other vertebrate and Ciona HCNs (Figure 3.1). An N glycosylation sequon near the pore is also found in the sole putative HCN channel mined from the genome of Oikoplerua dioica, although the sequon is located Nterminally and directly adjacent to the conserved location. These findings strengthen our previous suggestion (Jackson, et al., 2007) that the N glycosylation sequon arose in a common ancestor of urochordates and vertebrates, but was lost in ciHCNb. The more distant position of amphioxus within the tree, and its lack of the N glycosylation sequon, lend strong support to a recent study suggesting that tunicates, and not amphioxus, are the closest vertebrate relatives (Delsuc, et al., 2006). 3.3.3 NGlycosylation Status When Expressed in Cells To confirm whether either ciHCNa or ciHCNb can be Nglycosylated, the clones were tagged with a Cterminal V5 epitope. Following expression in both oocytes and mammalian cells (Figure 3.1C and D), membrane fractions were either left untreated or treated with PNGaseF and run on an SDSPAGE gel (Figure 3.1C and D). Indeed, in both CHO cells and Xenopus oocytes, ciHCNb is present as one band which is insensitive to PNGaseF treatment indicating no N glycosylation, as would be expected given the absence of the poreassociated sequon. In contrast,  90 Figure 3.1 HCN Glycosylation Arose Prior to the Divergence of Urochordate Lineage and is Found in the Common Ancestor of Urochordates and Vertebrates. A) Alignment of the pore region produced by ClustalX and generated by GeneDoc. Red box indicates known HCN Nlinked glycosylation site and the yellow line indicates the division between the presence and absence of this functional site. B) An HCN cladogram generated by aligned sequences between the start of S1 and the end of the CNBD in ClustalX and produced by TreeView. * indicates predicted HCN genes from the lancelet and Oikopleura genomes (Hegle, et al., 2010). X indicates the predicted emergence of Nlinked glycosylation in the HCN family. C) Western blot of membrane fractions from oocytes injected with cRNA of ciHCNa, ciHCNa N380Q or ciHCNb, untreated () or treated (+) with PNGaseF and probed with antiV5 antibody. Antiactin was used as a loading control. ciHCNa, but not ciHCNaN380Q or ciHCNb, is shifted to a lower molecular mass in the presence of PNGaseF. D) Western blot of whole cell lysates from CHO cells transfected with 1.5ug cDNA of ciHCNa, ciHCNaN380Q or ciHCNb, either treated (+) or untreated () with PNGaseF and probed for using an antiV5 antibody. Indicated weights are 95 kDa (top), 72 kDa (middle) and 34 kDa (bottom). Predicted weights of Ciona HCN clones are 78.8 kDa and 93.3 kDa for ciHCNa and ciHCNb, respectively.  9 1  92 ciHCNa is predominantly one band on the gel, which is shifted completely to a lower molecular weight when PNGase is added, suggesting almost complete Nglycosylation of this isoform in both cell types. Mutation of the putative Nglycosylated asparagine to glutamine also shifted the resulting ciHCNa band to the same weight as the wild type band after PNGase treatment, and its position was unaltered by exposure to that agent. This is good evidence that ciHCNa is N glycosylated at the same poreassociated sequon as for the vertebrate isoforms. 3.3.4 Two Ciona HCNs Form Channels Which are Variably Opened by Hyperpolarization and Blocked by Cesium Although both ciHCNa and ciHCNb are phylogenetically equidistant from, and share approximately the same sequence identity with vertebrate HCNs (Jackson, et al., 2007), N glycosylation of ciHCNa suggests it is phenotypically closer to vertebrate HCNs. To compare their function with vertebrate HCNs, properties of the two Ciona HCN clones were established using twoelectrode voltage clamp on Xenopus oocytes expressing either channel. In response to hyperpolarization of the membrane potential, both ciHCNs produced currents that share features with the other known HCN channels. From a holding potential of 10 mV, hyperpolarizing steps elicited complex inward currents with two main parts; a very fast or “instantaneous” current (Iinst) followed by a slowlyactivating current (If) (Figure 3.2).  Two important differences in the appearance of the current waveforms are that ciHCNa produces a very small, and voltageindependent Iinst and a proportionally larger If (Figure 3.2 A). Furthermore, the ciHCNa If current displays the characteristic brief delay upon activation. ciHCNb, on the other hand, produces a relatively large Iinst, which is about equal in size to the slower If that follows (Figure 3.2 B). Furthermore, there is no apparent delay in the onset of If. On average, the total current produced by ciHCNb is also small compared to the ciHCNa currents, as shown by a difference in scale in Figure 3.2.  Mammalian HCN channels are blocked by the application of extracellular cesium (Cs + ) (DiFrancesco, 1982). To determine whether the two Ciona HCN channels also shared this characteristic, a single 6second pulse to 70mV before and after the application of 5mM +  Cs +  in 96mM K +  was performed. Similar to other mammalian HCN channels, the slowlyactivating current of both channels is almost completely eliminated by 5mM Cs + . ciHCNa If displayed a significant reduction in the timedependent If from 4.188 ± 0.750 uA to 0.412 ± 0.112 WA  93 (Figure 3.2 C) and ciHCNb If showed a decrease from 1.142 ± 0.157 WA to 0.0173 ± 0.0098 WA (Figure 3.2 D). Additionally, the small fast current of ciHCNa was not reduced by Cs + , consistent with our findings in mammalian HCNs (Macri, et al., 2002; Proenza, et al., 2002a). This Cs + insensitive fast current may reflect contributions from current actually passing through ciHCNa channels, together with current flowing through the endogenous channels of the oocyte. In contrast, the large Iinst of ciHCNb is blocked by extracellular Cs + . The Iinst of ciHCNb was significantly reduced from 2.097 ± 0.279 WA (n=18) to 0.922 ± 0.192 WA.  94 Figure 3.2 Variable Opening by Hyperpolarization and Cesium Block of Two Ciona HCNs A) ciHCNa and B) ciHCNb  current traces elicited by a single 6second hyperpolarizing pulse to –70mV from a holding potential of 10mV (inset) followed by a pulse to 30mV  in the presence of 96mM K +  (black), followed by a 3minute incubation period in 5mM Cs + /96mM K +  while clamped at 10mV, and a second pulse protocol in the presence of the incubation solution (gray). C) and D) Plots of amplitudes of instantaneous and slow currents before and during perfusion with Cs + . Instantaneous current was measured between the zero line (dashed) and the start of timedependent current C) ciHCNa, n=32 and D) ciHCNb, n=18 * indicates p<0.05 in a two tailed paired ttest.   A. B. C. D. uA uA  95 3.3.5 Permeation of Cations Through Ciona HCNs is VertebrateLike Known vertebrate and invertebrate HCN channels allow the passage of both potassium and sodium ions; but, they are more permeable to potassium, and sodium permeation  has been shown to require the presence of external potassium ions (DiFrancesco, 1981a; DiFrancesco, 1981b). In addition to the predicted positive shift in reversal potential, raising extracellular K + , unlike Na + , also notably enhances current flow in HCNs (DiFrancesco, 1982). To examine these characteristics in ciHCNa and ciHCNb, we used a twostep pulse protocol (Figure 3.3, inset) and determined the currentvoltage relationships for If, from which contaminating ion flow could be easily separated. This relation was determined from the amplitude of current flowing through the pore upon opening by hyperpolarization, subtracted from the instantaneous component, and plotted against test voltage in the presence of low potassium and high potassiumcontaining extracellular solutions (Figure 3.3 A and B). Similar to mammalian HCN channels, increasing extracellular K +  significantly shifted the reversal potential positively for both ciHCNa and ciHCNb and suggests that the HCN current in these channels is carried in part by K +  (Figure 3.3 CF). The relations, which were approximately linear for both solutions, were steeper and crossed the xaxis at less negative voltages in the high potassium containing solutions, as has been described for other HCNmediated currents (DiFrancesco, 1982). Using the GoldmannHodgkinKatz equation (Hille, 2001) and the calculated reversal potentials in low (5mM) K +  solutions, the PNa/PK ratios were determined for both cloned channels. The intracellular concentrations were assumed to be 102mM K +  and 10mM Na +  (Weber, 1999) and the additional assumption was made that chloride does not contribute to the current. PNa/PK ratios were calculated to be 0.25 for ciHCNa and 0.38 for ciHCNb in 5mM K + . Both values are similar to those reported for cloned mammalian HCN channels in physiological solutions (Ludwig, et al., 1998).  96 Figure 3.3: Potassium Passes Through Both Ciona HCNs and Enhances Current Flow Using a twostep pulse protocol, instantaneous current levels were recorded at a series of test voltages following a prepulse to 70mV in A) ciHCNa and B) ciHCNb, in low 5mM K +  (light gray/squares), 10mM K +  (not shown) and high 96mM K +  (black/triangles) bath solutions. Zero current levels are indicated by the dashed line. C) and D) Plots of instantaneous current amplitudes versus voltage for If recorded in low and high concentrations of extracellular potassium. Currents from individual oocytes were measured by fitting a straight line through the linear portion of the curve across the xaxis. E) and F) The average of these reversal potentials and slopes are reported as mean ± SEM and compared using a oneway ANOVA and postTukey multiple comparison test (* p<0.05  compared to 5mM and # p<0.05 compared to 10mM). The reversal potentials of ciHCNa If are 25.63 ± 1.16 mV (n=20) and 3.40 ± 0.89 mV (n=12) in 5mM and 96mM K + , respectively. The reversal potential of ciHCNb shifted from 32.59 ± 3.50 mV (n=6) to 3.76 ± 0.98 mV (n=21) for If in low and high K + , respectively.  The slope conductance of If shifted for ciHCNa from 0.022 ± 0.002 in 5mM K +  to 0.069 ± 0.012 in 96mM K +  and from 0.0036 ± 0.0012 in low K +  to 0.0153 ± 0.0019 in high K +  for ciHCNb. The I/V graphs (C and D) depict the average data for each voltage step.     97  A. B. C. D. E. F.  98  To assess the permeability of ciHCNa and ciHCNb to different ions and to estimate permeability ratios, ion substitution experiments were performed using the twostep voltage protocol described in Figure 3.3. Because sodium requires potassium to pass, the experimental extracellular solutions used contained a constant level of KCl (20 mM) supplemented with 80 mM of one of KCl, NaCl, NMDGCl or LiCl to produce a 100mM solution. Reversal potentials and slope conductances were calculated for individual oocytes perfused with one of the five solutions. Average reversal potentials for the five solutions were calculated by fitting a straight line through the linear portion of the current/voltage curve across the xaxis of individual oocytes (Figure 3.4). The relations were approximately linear for all solutions, and were steeper and crossed the xaxis at the least negative voltages in the high potassiumcontaining solutions (100 mM KCl). The reversal potentials of LiCl and NMDG were the most negative and were not significantly different from each other. Assuming that NMDG does not cross the membrane, our data suggest that LiCl also does not contribute to the inward or outward current in these channels. The reversal potential in the presence of 80NaCl is significantly shifted in the positive direction from LiCl and NMDG conditions, consistent with a contribution of Na +  to If. Interestingly, the slope conductances in the 80 NMDG, 80NaCl or 80LiCl conditions are not significantly different from each other, suggesting that none of these extracellular ions augment channel conductance (Figure 3.4). Increasing extracellular K +  concentrations to 100KCl caused a further positive shift in the reversal potential and an increase slope, suggesting that the current is carried by both Na +  and K +  and that extracellular K +  enhances channel conductance.   99  Figure 3.4: Sodium Passes Through Ciona HCNs but Does Not Enhance Current Flow Using a similar twostep protocol as shown in Figure 3.3, oocytes were perfused with one of five solutions: 100 KCl (n=13,ciHCNa and n=18, ciHCNb), 20mM KCl/80 NMDG (n=9, ciHCNa and n=12, ciHCNb), 20 KCl/80 NaCl (n=7, ciHCNa and n=16, ciHCNb), 20 KCl/80 LiCl (n=8, ciHCNa and n=11, ciHCNb) or 20KCl/80 CsCl (data not shown).  ciHCNa (A) and ciHCNb (B) Iinst measured during a prepulse were subtracted and the remaining current was plotted as a function of test pulse voltage. Reversal potentials and slope conductances for the If of ciHCNa (C and E) and ciHCNb (D and F) in individual oocytes were calculated by fitting a straight line through the linear portion of the curve across the xaxis. The average of these reversal potentials and slopes are reported as mean ± SEM and compared using a oneway ANOVA and Tukey’s post hoc test (*  p<0.05 when compared to all other conditions, ** not significantly different from other **, but p<0.05 when compared to * or ***, *** p<0.05 when compared to all other conditions). The I/V graphs (A) depict the average data for each voltage step. B.A. D.C. F.E.  100 3.3.6 Slow Opening of ciHCNs is Facilitated by Raising Intracellular Cyclic AMP HCN channels possess a domain in the Cterminus to which the intracellular messenger cyclic AMP binds (Zagotta, et al., 2003). In some vertebrate and invertebrate HCNs, binding facilitates slow opening by relieving a tonic inhibition by the Cterminus; notably, however, cAMP itself does not open the channels. To determine whether cAMP facilitates slow opening of ciHCNs, the voltage dependence of channel opening (V1/2) for both clones was established using a staggered voltage protocol (Figure 3.5). To maximize current size, recordings were made in oocytes perfused with a 96mM K +  solution, either with or without 10mM 8bromo cAMP, a plasma membranepermeable analog of cAMP. For both conditions, oocytes were incubated for 45 minutes in a 5mM K +  solution, with or without cAMP. Boltzmann fits of normalized tail current (If) amplitudes versus test voltage yielded a curve showing that ciHCNa If began to activate around 40mV, whereas ciHCNb If began to activate at a more positive potential.  The average V1/2 of ciHCNa is 61.4 ± 1.1 mV (black, n=14) and 56.7 ± 1.1 mV in the absence or presence of 10mM 8bromo cAMP (red, n=11) and represents a significant (* p<0.05) difference of the means of 4.8 ± 1.6 as determined by a twotailed unpaired ttest. A similar significant shift of +6.7 mV (n=10) was found for ciHCNa when individual eggs were recorded in 96mM K +  and then incubated in the 10mM cAMP/96mM K +  for 45 minutes while clamped at 10mV (data not shown). Similarly, the average V1/2 of ciHCNb is 43.3 ± 1.8 mV (black, n=8) and 36.2 ± 1.7 mV in the presence of 10mM 8bromo cAMP (grey, n=8). This represents a significant (* p<0.05) difference of the means of 7.1 ± 2.5 as determined by a twotailed unpaired ttest. The positive shift of the V1/2 by raising intracellular cAMP is consistent with a direct facilitatory action on both channels. Nevertheless, for future studies, direct application of cyclic nucleotides will have to be carried out to rule out an effect of protein kinase A, or other downstream effectors of cAMP, on the Ciona HCN channels.   101    Figure 3.5: Opening of ciHCNs is Facilitated by Rises in Intracellular 8bromo cAMP. A staggered voltage protocol was used to determine voltage dependence of activation. Tail currents were measured at 30mV for ciHCNa and +30mV for ciHCNb. A) An average fit of the voltage dependence of activation of ciHCNa in the absence (black) and presence (grey) of 10mM 8bromo cAMP (cAMP) in 96mM K + . Oocytes were incubated for 45 minutes in a 5mM K +  solution, either with or without cAMP. The average V1/2 of ciHCNa is 61.4 ± 0.9 mV (black, n=14) and 56.7 ± 1.1 mV in the presence of cAMP (grey, n=11). Error bars represent the standard error of the mean. This represents a significant (* p<0.05) difference of the means of 4.7 ± 1.4 as determined by a twotailed unpaired ttest. The average slope factor of ciHCNa is 4.5 ± 0.3 (black, n=14) and 4.0 ± 0.3 in the presence of cAMP (grey, n=11). Error bars represent the standard error of the mean. The difference is not significant at p<0.05.  B) An average fit of V½ of ciHCNb in the absence (black) and presence (grey) of cAMP in 96mM K + . For both conditions, oocytes were incubated for 45 minutes in a 5mM K +  solution, with or without cAMP. The average V1/2 of ciHCNb is 43.3 ± 1.8 mV (black, n=8) and 36.2 ± 1.7 mV in the presence of cAMP (grey, n=8). Error bars represent the standard error of the mean. This represents a significant (* p<0.05) difference of the means of 7.1 ± 2.5 as determined by a twotailed unpaired ttest. The average slope factor of ciHCNa is 7.4 ± 1.0 (black, n=8) and 6.3 ± 0.9 in the presence of cAMP (grey, n=8). Error bars represent the standard error of the mean. The difference is not significant at p<0.05. B.A.  102 3.3.7 The Time Course of Slow Opening for ciHCNa is More Complex than ciHCNb The differences in the kinetics of channel opening between ciHCNa and ciHCNb are depicted in Figure 3.2 and Figure 3.3. To quantify these apparent differences, we determined the rates of slow opening from the same staggered protocol used to ascertain the range of voltages over which opening occurred. For ciHCNa, If could be fit best following an initial delay, using a double exponential function, with the time constant (τ) for the slower component reaching 45 seconds at its slowest. For ciHCNb, a single exponential function was used to fit the time dependence of If, with τ value reaching 56 seconds at its slowest. Both channels show faster activation kinetics at more hyperpolarized potentials (Figure 3.6), with the slowest rate of activation corresponding to voltages similar to their respective V1/2 values. Average τ(act) values for ciHCNa and ciHCNb at 80mV are 586.82 ± 83.49 ms (n=11) and 461.53 ± 45.64 (n=14).  103    Figure 3.6: The Kinetics of Slow Opening is Different Between ciHCNa and ciHCNb A) A sample trace from the protocol used to generate activation kinetics of ciHCNa in 96mM K +  B) Average taus of activation for the fast and slow (or single) components are plotted with respect to voltage. Activation kinetics were fit over the complete duration of the pulse following the initial lag phase of HCN activation, using a double exponential for voltages between 100 and 70mV and a single exponential for the 60mV pulse. A sample fit of the two components (Afast and Aslow) at 90mV are shown in dark and light grey, respectively, and the residual is shown as the dotted grey line above (A, inset). Error bars represent SEM (n=30). C) Ratio of fast and slow components of activation kinetics are plotted with respect to voltage. For ciHCNa, the ratio remains stable over all tested voltages (n=30). D) A sample trace of the staggered voltage protocol used to determine activation kinetics of ciHCNb.  E) Activation kinetics were fit with a single exponential over the complete duration of the pulse following the initial lag phase of HCN activation for all voltages between 80 and 40mV. A sample fit and the residual of fit at 70mV are shown in dark and light grey, respectively (inset D). Error bars represent standard errors of the mean (n=20).    104 3.3.8 Fast and Slow Opening Contribute Equally to Ion Flow Through ciHCNb Channels ciHCNb produced a proportionately larger fast current component (Iinst), which is blocked by Cs +  and about equal in size to the slowlyactivating current (If) at 70 mV. Because other fast activating currents, such as those through Kir2 channels, are susceptible to block by Cs + , we utilized ZD7288, a drug that more specifically blocks HCN channels (BoSmith, et al., 1993). A similar 6second protocol as outlined in Figure 3.2 was utilized to examine the effect of ZD7288, following a 15 minute incubation of 100uM ZD7288. ZD7288 significantly blocked both Iinst and If in oocytes expressing ciHCNb (Figure 3.7). A plot of the amplitude of ZDsensitive instantaneous current measured at 70mV against that for the ZDsensitive slow current at 70mV from the same oocytes showed a high correlation (r=0.911) between them. Together, the block of both instantaneous and slow currents by both Cs +  and ZD, and the high degree of correlation between ZDsensitive instantaneous and slow currents are strong evidence that both result from ions flowing specifically through the ciHCNb channel pore. Finally, the data show that the ZDsensitive fast and slow current amplitudes at  70mV are approximately equal in size, suggesting the gate is always at least half open.  105    Figure 3.7: The ciHCNb Current is Distributed Equally Between Instantaneous and Slow Components A similar protocol to the cesium experiments shown in Figure 3.2 was used. A single 6 second pulse to 70mV from a holding potential of 10mV was used, followed by a 15min incubation in 10uMZD7288/96mM K +  and then repeated. A) A sample trace from a single oocyte before (black) and after (gray) treatment. B) Average instantaneous and timedependent current recorded at 70mV before and after drug application. Error bars represent standard error of the mean and * is p<0.05 in a twotailed paired ttest (n=9). Values reduced from 1.63 ± 0.16mV to 0.445 ± 0.084mV and from 0.898 ± 0.116 to 0.0466 ± 0.0184 for Iinst and If, respectively. C) Bar graph of the portion of total Cssensitive or ZDsensitive current that is due to Iinst. Values are 0.484 ± 0.032 and 0.578 ± 0.025 for Cs+ (n=18) and ZD (n=9), respectively. D) Correlation plot between ZDsensitive Iinst and ZDsensitive If in ciHCNb. A significant positive correlation (r=0.911) is shown, similar to that found with 5mM Cs +  (data not shown). B.A. D.C.  106  3.4 Discussion To help establish the trajectory of HCN evolution from invertebrates to vertebrates, we have refined the phylogeny of HCNs and functionally characterized two novel HCN clones from Ciona intestinalis. ciHCNa undergoes Nglycosylation at a sequon near the pore and functions similarly to the mammalian HCNs, strongly suggesting that the common ancestor of tunicates and vertebrates also possessed similar properties. In contrast, ciHCNb does not undergo N glycosylation, consistent with the absence of the sequon, and it possesses an unusual gating phenotype in which the channel is constitutively open. Other key features of HCN function, including hyperpolarizationsensitive opening, cAMP facilitation of opening, permeability to sodium and potassium and block of slow opening by cesium, are retained in both Ciona isoforms. A voltageindependent conductance through HCNs has been appreciated only relatively recently, when they were cloned and expressed heterologously in Xenopus Oocytes and mammalian cell lines. This allowed the instantaneous current to be clearly delineated from endogenous currents and reliably studied (Chen, et al., 2000; Chen, et al., 2001a; Gauss, et al., 1998; Ishii, et al., 1999; Proenza, et al., 2002a; Proenza and Yellen, 2006). In most known HCNs, a voltageindependent conductance represents only a very small fraction of the total ion flow. A somewhat larger voltageindependent current has been observed in SPIH, one of two HCNs found in sea urchin, which represents about four percent of the total current flowing through it (Proenza and Yellen, 2006). In contrast, the voltageindependent current makes up about half of the total current through ciHCNb channels at 70mV, and was also noted when using more positive holding potentials (data not shown). To better understand the origin of VIC in ciHCN, we utilized a known HCN ion channel blocker, ZD7288. A single pulse to 70mV following an extended drug incubation period, resulted in close to complete block of both Iinst and If components. This finding suggests that both currents flow through the same ion conducting pore. Previous studies utilizing excisedpatch experiments, have demonstrated that ZD7288 blocks the HCN ion conducting pore in a usedependent manner at a blocking site located internal of the activation gate (Shin, et al., 2001). Therefore, channel activation is required to  107 open the gate to allow drug access to the binding site. The original work of ZD7288, however, demonstrated a nonusedependent block of If when applied to the external surface in a whole cell configuration (Berger, et al., 1994; BoSmith, et al., 1993; Harris and Constanti, 1995). Shin and colleagues suggested that this difference in mode of ZD7288 block could be explained if a portion of the channels were open at the experimental holding potential, as the blocker could equilibrate with the binding site over the duration of drug incubation (Shin, et al., 2001). An alanine scan of the pore region, showing that the voltage sensor and the pore interact weakly, support the notion that open probability does not reach zero (Macri, et al., 2009). More recent work on the voltageindependent current of the sea urchin isoform, spHCN (Proenza and Yellen, 2006) investigated the notion that VIC represents a nonzero limiting open probability for HCN channels at positive voltages. Proenza and colleagues (Proenza and Yellen, 2006), demonstrated that both VIC and If were blocked by ZD7288 and, in a cysteinesubstituted pore, by Cd 2+ . This suggests that both components result from ion movement through the channel pore. Curiously, however, they found that the kinetics of block, and the degree of block over short pulses, for these agents were different between VIC and If, as though they resulted from channels that were not in rapid equilibrium with each other. Thus, they concluded that VIC and If, result from two separate populations of spHCN channels. Whether this is true for VIC in our studies of ciHCNb remains to be determined. Structural insight into the origins of VIC has come from studies using a combined electrophysiological and mutagenesis approach, which have shown that the region connecting the putative fourth and fifth transmembrane segments of HCNs controls the degree of voltage insensitive opening (Chen, et al., 2001a; Macri and Accili, 2004); this region is highly divergent between ciHCNa and ciHCNb and will be a focus of future investigations of the structural basis for this gating phenotype. Interestingly, the HCNrelated hyperpolarizationactivated potassium channels from plants (e.g. KAT1, AKT1 and AKT2) also demonstrate variable degrees of constitutive opening (Gaymard, et al., 1996; Lacombe, et al., 2000; Schachtman, et al., 1992), suggesting that this dual mode of gating was integral in a common ancestral ion channel. All mammalian HCN channels have one potential Nlinked glycosylation consensus sequence in the extracellular linker region between S5 and the pore helix ((Jackson, et al., 2007; Proenza, et al., 2002b), and have been shown to be glycosylated in various tissues and heterologous systems (Hegle, et al., 2010; Much, et al., 2003; Santoro, et al., 1997; Zha, et al.,  108 2008). Furthermore, glycosylation has been shown to play a crucial role in membrane insertion (Much, et al., 2003; Nazzari, et al., 2008) and heteromerization of different isoforms (Zha, et al., 2008). It has been shown that glycosylation of one mammalian isoform can rescue another non glycosylated isoform by increasing stability and membrane insertion of (Much, et al., 2003) (Whitaker, et al., 2007) and suggests that coassembly of homologous subunits to heteromeric complexes serves as an important mechanism in generating ion channel diversity. It is plausible that heteromerization evolved with the presence of duplication and glycosylation status. Further experiments will need to address whether these two Ciona clones can coassemble, the characteristics of the heteromer and if glycosylation plays any role in this process. The one HCN gene mined from the amphioxus genome falls between tunicates and echinoderms in the phylogenetic tree and the sequence lacks the poreassociated Nglycosylation sequon. Together, these findings support other molecular evidence suggesting that urochordates, and not amphioxus, are the closest chordate relatives to vertebrates (Delsuc, et al., 2006; Delsuc, et al., 2008). Tunicates diverged from a common ancestor with vertebrates over 550MYA and have undergone rapid evolution that includes dramatic rearrangement and loss of genomic material (Delsuc, et al., 2006; Gee, 2006; Putnam, et al., 2008; Schubert, et al., 2006; Swalla, et al., 2000).  Selective and rapid evolution could explain the diverse evolutionary trajectories of ciHCNa and ciHCNb following a lineagespecific duplication from a common ancestor as well as the finding that the sequences of the two Ciona clones are as similar to each as they are to those of the vertebrate HCNs. Future studies will compare the physiological roles for the Ciona HCN isoforms, which are likely different because of their contrasting gating phenotypes. Interestingly, a putative ortholog of ciHCNa was recently identified in a related tunicate species, Botryllus schlosseri (Hellbach, et al., 2011) and was shown to be strongly expressed and functional in the cardiac pacemaking tissue. This provides strong support for a putative role of ciHCNa in the cardiac function of Ciona intestinalis. Cumulatively, over one billion years of independent evolution have occurred between the extant ascidians and modern vertebrates. Features shared between the tunicate and mammalian channels can shed light on the ancestral condition (Gee, 2006) of the HCN channel function prior to the family expansion in the vertebrate lineage and help us begin to understand how channel diversification has occurred throughout evolution.  109 4. EVOLUTIONARY ANALYSES OF KCNQ1 AND HERG VOLTAGEGATED POTASSIUM CHANNEL SEQUENCES REVEAL LOCATIONSPECIFIC SUSCEPTIBILITY AND AUGMENTED CHEMICAL SEVERITIES OF ARRHYTHMOGENIC MUTATIONS 3  4.1 Introduction Two voltagegated potassium ion channel genes, KCNQ1 and KCNH2 (HERG), encode for channels that underlie the slowly and rapidlyactivated delayed rectifier potassium currents (IKs and IKr), respectively (Barhanin, et al., 1996; Sanguinetti, et al., 1996; Sanguinetti, et al., 1995; Trudeau, et al., 1995). Efflux of potassium ions through these channels is critical for repolarization of the cardiac action potential. Mutations that disrupt normal biosynthesis and function of KCNQ1 and HERG have been associated with three cardiac arrhythmias: Short QT syndrome (SQTS) (Bellocq, et al., 2004; Brugada, et al., 2004; Gaita, et al., 2003), atrial fibrillation (Chen, et al., 2003; Hong, et al., 2005) and Long QT syndrome (LQTS) (Curran, et al., 1995; Wang, et al., 1996). To date, there are close to 200 reported arrhythmiaassociated mutations (AAMs) in each channel, with more than 95% of which are linked to LQTS. This arrhythmia, which affects an estimated 1 in 500010000 people worldwide, is characterized by a prolongation of the QT interval on an electrocardiogram and can lead to Torsade de Pointes (TdP), ventricular fibrillation and sudden cardiac death (Roden, 2008; Schwartz, 2005). More recently, the progression from LQTS into TdP has been proposed as a cause of sudden infant death syndrome (SIDS) (Arnestad, et al., 2007; Schwartz, et al., 2001; Wang, et al., 2007) and sudden unexplained death syndrome (SUDS) (Tester and Ackerman, 2007). Ninety percent of all known LQTSassociated mutations occur in HERG and KCNQ1. In studies of larger groups of proteins, or individual proteins other than HERG and KCNQ1, diseaseassociated amino acid mutations (DAMs) have been analyzed according to the chemical severity of the change, as determined from Grantham’s Scale (Grantham, 1974), and the location  and/or context of the altered amino acid (Botstein and Risch, 2003). An up to two fold increase in clinically observable disease occurs in parallel with increases in the amount of chemical change (Krawczak, et al., 1998). In many proteins, DAMs are chemically more severe  3  A version of this chapter has been published. Jackson, HA and Accili, EA. (2008). Evolutionary analyses of KCNQ1 and HERG voltagegated potassium channel sequences reveal locationspecific susceptibility and augmented chemical severities of arrhythmogenic mutations. BMC Evolutionary Biology 8: 188.  110 than changes over the course of evolution (called interspecific changes) and polymorphic changes (Briscoe, et al., 2004; Miller and Kumar, 2001).  In rhodopsin, the chemical severity of DAMs also correlates with the expected chemical severity of a given codon (Briscoe, et al., 2004), as determined by comparison with the normally occurring human codon. The importance of an amino acid to protein function can be inferred from its conservation over the course of metazoan evolution. In many proteins, DAMs are overabundant at evolutionarily conserved and slowly evolving sites (Briscoe, et al., 2004; Miller and Kumar, 2001; Mooney and Klein, 2002; Subramanian and Kumar, 2006), presumably because sites that have experienced little interspecific variation are critical for function. Mutations at these sites are likely deleterious and would be removed from the population by natural selection, if given enough time. In some proteins, DAMs are unevenly distributed among functionally conserved domains (Miller, et al., 2003), even after accounting for the length and number of evolutionarily conserved amino acids. These data imply that functionally conserved domains, like conserved sites, are less tolerant to mutations because of their greater importance to the overall protein function. However, according to the Neutral Theory of Molecular Evolution, most nucleotide substitutions are phenotypically neutral and avoid natural selection (Nei, 2005). Most nucleotide changes, except for diseaseassociated changes, might then be expected to distribute randomly throughout a protein. In a study of a large number of proteins, synonymous single nucleotide polymorphisms (sSNPs) were shown to distribute randomly consistent with a neutral phenotype (Subramanian and Kumar, 2006). Initial studies of nonsynonymous (ns) SNPs in the cystic fibrosis transmembrane regulator and tuberous sclerosis complex 2 gene, were also shown to distribute randomly (Miller and Kumar, 2001). However, a more recent, study, using many proteins, showed that nsSNPs preferentially locate at variable sites and sites with high evolutionary rates, and are underrepresented at sites that are evolutionarily conserved and have low evolutionary rates (Subramanian and Kumar, 2006). This discrepancy may be due to differences in the nature of the nsSNPs available e.g. ns SNPs are less disruptive when located at conserved sites in some proteins versus others, or to a difference in the nature of the conserved sites e.g. conserved sites in some proteins are less tolerant to amino acid changes. In this study, we take advantage of the >200 mutations reported per channel to quantitatively analyze, for the first time, the distribution of AAMs and nonsynonymous SNPs  111 within HERG and KCNQ1 and to determine the chemical severity of these changes, as compared to interspecific changes. 4.2 Methods 4.2.1 Sequence Alignment Human protein and mRNA sequences for HERG and KCNQ1 were obtained from NCBI. Other vertebrate sequences were collected from a BLASTP (Altschul, et al., 1990) search at ENSEMBL (default parameters) and the NCBI nonredundant database. To eliminate the inclusion of sequences from closely related paralogs, a manual reciprocal besthit analysis was used. Sequences were aligned by ClustalX (Thompson, et al., 1997) using default settings. Less conserved regions in the N and/or Ctermini were removed, leaving sequences corresponding to residues 1906 and 83599 of human HERG and human KCNQ1, respectively. 4.2.2 Collection of ArrhythmiaAssociated Mutations (AAMs) and NonSynonymous Polymorphisms (nsSNPs) HERG and KCNQ1 AAMs were obtained from the NCBI OMIM database, the Human Gene Mutation Database (HGMD), the European Society of Cardiology Working Group on Arrhythmias (WGA) LQTS gene database  and from primary literature (Arnestad, et al., 2007; Aydin, et al., 2005; MankSeymour, et al., 2006; Millat, et al., 2006; Tester and Ackerman, 2007).  Only missense mutations that fell within the aligned region were used, and each was included only once, to eliminate frequency bias, but multiple disease mutations could be associated with a single site. nsSNPs were collected from NCBI dbSNP, HAPMAP international and primary literature and were assumed to be phenotypically neutral based on their inclusion in the different databases. nsSNPs that overlapped with identified disease mutations were removed. The final data set included 172 AAMs and 16 nsSNPs for HERG, and 174 AAMs (Table 1) and 12 nsSNPs for KCNQ1.   112  Table 41 Breakdown of Clinical Phenotype of Disease Mutations Included in Analyses   HERG KCNQ1 Long QT 166 169 Short QT 1 1 Atrial Fibrillation 0 1 SIDS 3 2 Long QT/SIDS 2 1 Total 172 174   113  4.2.3 Phylogenetic Analysis and Determination of Interspecific Variability Aligned protein sequences, with gaps removed, were used to construct a phylogenetic tree using the maximum likelihood analysis in PHYLIP (Felsenstein, 1996). SEQBOOT, PROML, and CONSENSE programs were used with a JTT model of evolution. Ancestral sequences were determined using the maximum likelihood method from PAMLv3.15 (Yang, 1997) under a Poisson model of amino acid evolution. Discrete values for interspecific variability (0 through n) were determined for each residue in the protein from differences between the ancestral and descendent sequences throughout the provided tree. 4.2.4 Association Between AAMs or nsSNPs and Evolutionarily Conserved Sites To determine whether AAMs and nsSNPs are preferentially associated with evolutionarily conserved sites, we compared the number of observed mutations in HERG and KCNQ1 to the number of mutations expected at each site in both proteins based on neutral substitution (Miller and Kumar, 2001). Sites were binned according to counts of interspecific variability (0 through i) which were determined using PAML. The expected number of mutations was determined using the following equation  Di expected  = D total  x Ni/N (Equation 1)  where Di expected  is the expected number of mutations at sites that have undergone i substitutions, D total  is the total number of disease mutations for each channel given by ΣDi observed , Ni is the observed number of sites in the alignment that have undergone i substitutions and N is the total number of residues in the gene being examined. Therefore, Ni/N is the fraction of sites in the gene that belong to a particular variability class (i), and if disease mutations distribute randomly throughout the protein, then Di expected  will be proportional to the fraction of the total sites and the total number of disease mutations observed in each gene. To determine whether differences in the distribution pattern between observed values and those expected from neutral theory were significant, the Χ 2 statistic was calculated (Miller and Kumar, 2001) and compared to a critical value for the given degrees of freedom (i1) using the following equation,  114  Χ 2 = Σ0 i  (D observed D expected ) 2 /D expected   (Equation 2)  4.2.5 Determination of Codon Evolutionary Rate of Change  The evolutionary rates of change for codons were estimated using the maximum likelihood method implemented in the CODEML program of PAML (Subramanian and Kumar, 2006; Yang, 1997), using a discrete gamma model (eight categories). The shape parameter was either fixed or free to vary and a likelihood ratio test was performed to evaluate model fitting. Evolutionary rates based on a Poisson model of evolution were established for every site and normalized to the maximum rate observed for each protein. Values between 0 and 1 were binned into eight different categories and used to represent eight different levels of evolutionary change. The analysis was performed on nucleotide sequence alignments of core regions, with gaps removed, of human and three closely related vertebrate orthologs, guided by protein sequence alignments. The expected number of mutations at the codons belonging to each rate category was calculated using a modification of Eq. 1 where Ni is the number of codons belonging to i category, N is the total number of amino acid positions, and D total  is the total number of disease mutations for each protein used in the analysis.  4.2.6 Distribution of AAMs Among Functionally Important Regions of the Channels To quantify over or underrepresentation of AAMs in functionally conserved regions of the channels, we compared their distribution to that expected from a uniform or evolutionary hypothesis (Miller, et al., 2003). First, we tested whether AAMs were distributed uniformly across the protein. The expected number of mutations in a given region was determined using the following equation  Dj expected  = (Rj/R) x D total  (Equation 3)  where Dj expected  is the expected number of disease mutations in a particular region, j, Rj is the number of residues found in region j, R is the total number of residues used in the analysis (ΣRj) and D total = ΣDj observed  or the total number of disease mutations used in the analysis.  115  The Χ 2 statistic was calculated and compared to a critical value for the given degrees of freedom equal to j1, where j is the number of different regions in the channel being analyzed, using the following equation,  Χ 2 = Σj (D observed D expected ) 2 /D expected  (Equation 4)  Second, we tested whether the distribution of AAMs in different regions was related to the distribution of evolutionarily conserved sites. If AAMs are overrepresented at conserved sites, the number of AAMs for a given region will be proportional to the number of conserved sites found within that region. The expected number of mutations per region was determined using the following equation,  Dj expected  = Σ0 i  ((aij/ai) x Di observed ) (Equation 5)  where ai is the total number of sites in a protein belonging to variability class i, aij is the number of sites of variability class i found within the region j, and Di observed  is the total number of disease mutations found at variability sites, i, across the entire protein. 4.2.7 Chemical Severity of Amino Acid Changes The interspecific chemical severity of a given site was determined by the average severity (according to Grantham’s Scale (Grantham, 1974)) of all ancestordescendent amino acid differences at that site throughout the tree, as reported by PAML. Only those interspecific changes that result from a single point mutation were included and each type of amino acid change at a given site was counted once to account for common ancestry (Miller and Kumar, 2001). The expected chemical severity at each site in HERG and KCNQ1 was determined by computing the average severity of all nonsynonymous changes produced by a single point mutation from the human reference codon (Briscoe, et al., 2004). 4.2.8 Weighted Average for Amino Acid Expected Chemical Severity To examine the involvement of specific amino acid residues in disease, the proportion of AAMs at a particular amino acid was calculated as percentage of the total AAMs in both  116 channels. To determine whether these findings were due to an overrepresentation of certain amino acids in the proteins, data was normalized for the total number of codons for a particular residue. Finally, the weighted average for amino acid expected chemical severity was calculated by the sum of the average of each of the individual expected codon chemical severities of the residue multiplied by its contribution to the total number of codons for the residue in the two channels combined. 4.3 Results 4.3.1 Channel Structure and Mutation Mapping HERG and KCNQ1 channels are likely formed by the tetrameric assembly of individual alpha subunits, each of which is composed of six transmembrane segments and cytosolic N and Ctermini (Figure 4.1a, b). The voltage sensing domain (VSD) is composed of the first four transmembrane segments and a pore region is composed of S5, a reentrant Ploop containing the selectivity filter, and S6. Mapping AAMs onto the sequences and predicted topologies of HERG and KCNQ1 subunits yielded some common distribution patterns as well as some unique to the individual proteins. The final data set used for HERG consists of 172 AAMs at 134 sites and 16 nsSNPs at 16 individual sites. Of 30 sites harboring multiple AAMs, 24 sites had two, 4 sites had three and 2 sites had four. For KCNQ1, the final data set includes 174 AAMs mapping to 130 sites and 12 nsSNPs to 12 individual sites (see Additional file 1). Thirtysix sites had multiple AAMs associated with them: 30 sites with two, 5 with three and 1 with five (Figure 4.1).  117 Figure 4.1 Location of ArrhythmiaAssociated Mutations (AAMs) and NonSynonymous Single Nucleotide Polymorphisms (nsSNPs) in Human HERG and KCNQ1 Subunits a) Schematic of the human HERG subunit. Included are 172 AAMs at 134 sites and 16 nsSNPs at 16 sites. b) Schematic of KCNQ1 subunit. Included are 174 disease mutations at 130 sites and 12 nsSNPs at 12 sites. Blue region = voltage sensing domain (S1S4), pink region = pore forming domain (S5S6), yellow = PAS domain or CNBD domain in cytosolic regions of the channel. Red X’s delimit the region used in analysis. Circles: red = 1 AAMs, yellow = 2 mutations/site, green = 3 mutations/site, blue = 4 mutations/site, purple = 5 mutations/site, blue outline = 1 nsSNP/site, green outline = overlap of disease mutation and evolutionary change, red outline = overlap of AAM, evolutionary change and nsSNP. (Jackson and Accili, 2008).  118    119    Between the two channels, some similarities exist in the distribution patterns of AAMs in HERG and KCNQ1, supporting results gathered when much fewer mutations were known (Splawski, et al., 2000). For example, both channels contain a large number of mutations within the pore region (23% and 29% for HERG and KCNQ1, respectively) (Table 2). On the other hand, differences in percentages of AAMs do exist in the intra and extracellular linker regions as well as the two cytosolic termini. In HERG, 23% of all known disease mutations are found in the extracellular linkers (between S1 and S2, S3 and S4, S5 and the Ploop, and between the P loop and S6). In KCNQ1, only 6% are found extracellularly whereas 20% are found in the intracellular linkers (between S2S3 and S4S5). These differences suggest that different linker regions contribution to overall function may be channel specific. Another difference occurs in the distal termini. In HERG, 27% of disease mutations are located in the Nterminus, compared to only 2% in KCNQ1.  However, three quarters of these are located in the PAS (PerArntSim) domain, a basic helixloophelix domain that is unique to the HERG Nterminus and regulates channel closing (Sanguinetti and TristaniFirouzi, 2006).  120  Table 42: Disease Mutation Distribution by Channel Region Region HERG KCNQ1 Nterm 6% 2% PAS 21% NA VSD 9% 11% ICL 3% 20% ECL 23% 6% Pore 23% 29% Cterm 15% 32% PAS = PerArntSim domain in HERG, Nterm = Nterminal region used in analysis (excluding PAS domain in HERG), ECL = extracellular linker regions between S1S2, S3S4, S5P loop and PloopS6, ICL = intracellular linker regions between S2S3 and S4S5, VSD = voltage sensing domain (only transmembrane portions), Pore = pore region including S5, Ploop and S6, Cterm = Cterminal portion of channels used in this analysis. (Jackson and Accili, 2008).  121  Nonsynonymous SNPs distribute differently from AAMs in HERG and KCNQ1 (Figure 4.1). In HERG, nsSNPs are more commonly found in the cytosolic regions (15/16) compared to the transmembrane regions (1/16). In KCNQ1, 8/12 nsSNPs are found in the cytosolic regions and 4/12 nsSNPs are found in the transmembrane region. Most of these occur at sites that do not have associated disease mutations, suggesting that location is an important determinant of disease. In both channels, however, two sites situated in cytosolic regions harbor both disease and polymorphic mutations, suggesting that amino acid identity plays a role in channel dysfunction at certain sites. 4.3.2 AAMs Occur Preferentially at Sites Conserved Throughout Vertebrate Evolution and at those with Lower Evolutionary Rates of Change The majority of AAMs mapped to sites completely conserved throughout the evolution of the respective channel: 146/172 mutations (84.9%) in HERG, mapping to 114/906 (12.6%) of the sites used in the analysis, and 153/174 (87.9%) in KCNQ1, mapping to 110/517 (21.3%) of the sites used in the analysis. Conversely, 22/172 AAMs in HERG and 20/174 AAMs in KCNQ1, map to sites with interspecific variation (Figure 4.1). Only two sites in HERG and one in KCNQ1 involve an amino acid change that is observed in both the AAM and in interspecific change.  Because the amino acid sequences of HERG and KCNQ1 are highly conserved (over 65% complete identity between fish and human sequences in both channels for the regions used), a neutral mutational process would still produce a large number of mutations at conserved sites. To determine whether AAMs and nsSNPs occur preferentially at evolutionarily diverse sites in HERG and KCNQ1, we utilized a quantitative approach developed previously (Subramanian and Kumar, 2006). The evolutionary relationships and the number of interspecific changes at each site were determined (Figure 4.2) and in both channels, the largest number of interspecific changes observed at any site was five. Therefore variability data was binned into six categories, ranging from completely conserved sites (0) to highly variable (5).  122                               Figure 4.2 Cladograms of Vertebrate HERG and KCNQ1 Protein Sequences. a) HERG: GenBank accession numbers/Ensembl identifiers for protein sequences used: human (NP_000229.1), dog (NP_001003145.1), zebrafish (NP_998002), monkey (ENSMMUP00000026900), rabbit (Q8WNY2), mouse (NP_038597.1), rat (NP_446401.1), opossum (ENSMODP00000004651), fugu (NEWSINFRUP00000161615), tetraodon (GSTENP00027811001). b) KCNQ1: GenBank accession numbers/Ensembl identifiers for KCNQ1 protein sequences used: human (NP_000209.2), mouse (NP_032460.2), rat (NP_114462.1), chicken (ENSGALP00000010441), fugu (NEWSINFRUP00000143259), zebrafish (ENSDARP00000077951). See methods for details. (Jackson and Accili, 2008).   123 A greater proportion of AAMs are found at evolutionarily conserved sites, and a smaller proportion are found at variable sites, than would be expected by an underlying neutral process (Figure 4.3a). Using X 2  analysis, the difference between the distributions of observed and expected disease mutations were statistically significant in both channels, even when only the numbers of disease harboring sites were analyzed (ruling out an effect of multiple mutations at highly mutable sites) or when data were pooled to account for low numbers of expected disease mutations in higher variability classes (data not shown). In KCNQ1, nsSNPs distribute randomly but, in HERG, they were significantly underrepresented at completely conserved sites and overabundant at variable positions (Figure 4.3b). To ascertain directly whether AAMs associate preferentially with sites that experience low rates of evolutionary change, we utilized an approach similar to the site analysis used above together with codon evolutionary rate obtained from CODEML in PAML (Subramanian and Kumar, 2006). For both channels, AAMs are found at sites with lower evolutionary rates (Figure 4.3c). In both channels, certain AAMs were located at codons belonging to the 1 st  and 2 nd  evolutionary rate categories as well as to variability classes 0, 1 and 2. This implies that, when AAMs are found at variable sites, they may preferentially occur at those with low evolutionary rates.  124  Figure 4.3 ArrhythmiaAssociated Mutations in HERG and KCNQ1 are Overrepresented at Evolutionarily Conserved and Slowly Evolving Sites. a) Counts of observed (white bars) and expected (black bars) numbers of AAMs at amino acid sites in HERG and KCNQ1 which have undergone different numbers of substitutions among species (see methods). . Because of a low number of expected mutations in the more variable positions, the X 2 statistic was also calculated for pooled data with two bins, 0 and 1+, with 1 degree of freedom. The number of disease mutations observed at completely conserved sites (0 class) in both HERG and KCNQ1 is significantly higher than by chance alone: HERG, Χ 2  (5 df) = 37.41, p<0.001 or Χ 2  (1 df) = 34.65, p<0.001; KCNQ1, Χ 2  (5 df) = 50.45, p<0.001 or Χ 2  (1 df) = 49.37, p<0.001. b) Counts of observed and expected numbers of nonsynonymous single nucleotide polymorphisms (nsSNPs). In HERG, fewer nsSNPs occur at completely conserved sites than expected by chance alone (Χ 2  (5 df) = 22.94, p<0.001 or Χ 2  (1 df) = 10.07, p<0.05) whereas in KCNQ1, the distribution is not significantly different from the expected count of neutral variation (Χ 2  (5 df) = 1.04, p>0.05). c) Data were pooled to account for low numbers of expected AAMs at variable sites and significance was confirmed. The distribution of AAMs was significantly different than what would be expected by random chance for both HERG (Χ 2  (7df) = 26.10, p<0.001 or Χ 2  (1 df) = 14.17, p<0.001) and KCNQ1 (Χ 2  (7 df) = 34.74, p<0.001 or Χ 2  (1 df) = 18.15, p<0.001). (Jackson and Accili, 2008).  125  4.3.3 Disease Mutations are Not Equally Distributed Among Functional Regions of the Channels Because both channels possess functional regions that are well conserved among voltage gated channels, we tested for uneven domain distribution of AAMs. KCNQ1 and HERG were divided into six or seven regions, respectively: Nterminus, PAS domain (HERG only), VSD (S1 through S4 transmembrane regions only), pore region (excluding outer turret), extracellular linkers, intracellular linkers and the Cterminus (see Figure 4.1). Based on the X 2  analysis, AAMs in both channels are unevenly distributed among the defined functional domains and, in general, do not support either a uniform pattern (in which the number of randomly occurring mutations are proportional to the total number of residues) or evolutionary pattern (in which the number of randomly occurring mutations are proportional to the total number of conserved residues) (Figure 4.4a). For both channels, AAMs are overrepresented in the pore region. AAMs are found preferentially in the intracellular (IC) linker of KCNQ1 and the extracellular (EC) linker of HERG, as well as the PAS domain of HERG, but are underrepresented in the N and C termini of both channels. This finding is especially striking for KCNQ1 considering 32% of its disease mutations are found in the Cterminus. The number of AAMs in the VSD of HERG and KCNQ1 were not different from the expected number based on a uniform or evolutionary distribution. To examine whether the overall conservation of a region (average variability/site in domain) exerts a nonadditive influence on the disease susceptibility (Miller, et al., 2003), the average number of AAMs per site in a given region of HERG and KCNQ1 was plotted against its average variability per site. These values were correlated for KCNQ1, but not for HERG (Figure 4.4b). The slope of this relationship in KCNQ1 was greater than those based on the uniform or evolutionary hypothesis. This indicates that AAMs are overabundant in conserved regions and less than expected in variable regions, which suggests that the regional sequence conservation may play a role in KCNQ1 disease susceptibility.  126    Figure 4.4 ArrhythmiaAssociated Mutations are Unevenly Distributed Among Functionally Conserved Regions of HERG and KCNQ1 Even After Accounting for Total Length and Evolutionary Conservation of Individual Sites Therein. a) Counts of observed number of AAMs per region (white), counts of expected number of AAMS based on a uniform distribution across each gene (black) and expected number of disease mutations based on an evolutionary distribution within each region (gray). Disease mutations are unevenly distributed among different regions of the channel: HERGuniform (Χ 2  (6 df) = 145.10, p<0.001), HERGevolutionary (Χ 2  (6 df) = 116.55, p<0.001), KCNQ1uniform (Χ 2  (5 df) = 81.59, p<0.001) and KCNQ1evolutionary (Χ 2  (5 df) = 37.39, p<0.001). b) Scatter plots showing the relationship between channel region conservation (average variability/site within domain) and the average number of observed disease mutations per site (diamonds) or expected number of disease mutations per site based on a uniform (circles) or evolutionary (triangles) distribution. Dotted and dashed lines indicate fits for expected uniform and evolutionary distribution, respectively. Solid lines represent bestfit regression of observed data. The correlation is significant for KCNQ1 but not for HERG, and the best fit of the KCNQ1 data is significantly different from the other two hypotheses. (Jackson and Accili, 2008).  127  4.3.4 The Chemical Severities of AAMs are Different than Changes Observed Throughout Evolution Two sites in each channel are associated with both an AAM and nsSNP. In each case, the AAM and nsSNP involve different amino acid substitutions. Additionally, of the 42 disease sites that overlap with interspecific change, only two sites in HERG and one site in KCNQ1 display an AAM that is the same as an interspecific change, suggesting that the identity of the amino acid contributes to susceptibility to disease. In previous studies, the chemical severity of disease causing mutations, determined by the Grantham’s Scale, were on average larger than interspecific changes, and not correlated with evolutionary changes found at the same sites (Briscoe, et al., 2004; Miller and Kumar, 2001). The average chemical severity of AAMs in HERG and KCNQ1 were also larger than those observed throughout evolution (Table 3). Furthermore, we found no correlation between the chemical severities of both types of change in HERG and KCNQ1 at variable sites that harbor a disease mutation (Figure 4.5a).  128 Table 43 Average Chemical Differences of Amino Acid Changes  Gene Disease nsSNP Interspecific HERG 93.63 (3.99) n = 172 75.50 (10.55) n = 16 67.37 (1.97) * N = 296 KCNQ1 88.29 (3.75) n = 174 52.25 (8.66) † n = 12 58.32 (2.74) * N = 168 *, † indicate a significant difference from disease severity († p<0.0001, *p<0.008) as determined by a MannWhitney U test. There was no significant difference between the average chemical differences observed in disease vs. polymorphic changes (p = 0.218) for HERG. There was no significant difference between the average chemical differences observed in polymorphic and interspecific amino acid changes in either HERG (p = 0.1441) or KCNQ1 (p = 0.6671). (Jackson and Accili, 2008).  129  The chemical severity of amino acid changes may depend on the expected chemical severity of mutations that can arise from a single nucleotide substitution in the codon involved (Briscoe, et al., 2004). For KCNQ1 and HERG, the chemical severity of AAMs and expected chemical severity were correlated (Figure 4.5b). Interspecific and expected chemical severities, however, were not correlated for KCNQ1, and had a very small correlation coefficient for HERG (Figure 4.5c) (which becomes nonsignificant when three outliers are removed; not shown). These findings are consistent with previous studies, which suggest that interspecific chemical severity is not influenced by the expected chemical severity of the codon involved but rather by the process of natural selection (Briscoe, et al., 2004). Nonetheless, the chemical severities of changes tolerated at variable sites may be influenced by the codon’s expected chemical severity. When completely conserved sites were removed (Figure 4.5d), the expected chemical severity of the involved codon was correlated with interspecific chemical severity, for both HERG and KCNQ1, but to a lesser extent than with disease chemical severity based on slope and correlation coefficient.  130  Figure 4.5 ArrhythmiaAssociated Mutation and Interspecific Chemical Severities Correlate with the Expected Chemical Severity of the Reference Codon in HERG and KCNQ1, but Not With Each Other. a) Plot of AAM vs. interspecific chemical severities. No significant correlation is observed for either HERG (p = 0.20) or KCNQ1 (p=0.10). b) Plots of AAM vs. expected chemical severities. A significant correlation is observed for both HERG (p<0.0001) and KCNQ1 (p<0.0001). c) Plots of interspecific vs. expected chemical severities (all sites). A significant correlation is observed for HERG (p=0.0148) but not for KCNQ1 (p=0.9918). For both proteins, the correlation (r) statistic, and the slopes of the regression, for interspecific vs. expected chemical severities are significantly smaller (p<0.001) than those for AAM vs. expected chemical severities (‘b’, above). d) Plots of interspecific vs. expected chemical severities using only variable sites. A significant correlation is observed for both HERG (p<0.0001) and KCNQ1 (p<0.0001). The correlations and slopes of the linear regression are significantly larger (p<0.001) than those using all sites (‘c’, above). For both proteins, the slopes of the linear regression are significantly smaller than those for AMM vs. expected chemical severities (‘b’, above) The correlation (r) is significantly smaller than that for AMM vs. expected chemical severities (‘b’, above) for HERG (p<0.05) and not for KCNQ1 (p=0.058). Significant differences in the average chemical severity were tested using the MannWhitney U test whereas the correlation differences between expected chemical severity and disease or interspecific severity were tested for significance using the zscore calculations. (Jackson and Accili, 2008).   131    132  We next compared the average chemical severities of AAMs and nsSNPs (Table 3). We found a significant difference between AAMs and nsSNPs in KCNQ1, but not in HERG, suggesting that other factors, such as location, may play a larger role in causing channel dysfunction for AAMs in HERG compared to KCNQ1. There were no significant differences between nsSNP and interspecific severities in either channel. 4.3.5 Involvement of Specific Amino Acids in Arrhythmogenic Disease We next determined which amino acids in HERG and KCNQ1 are targets in arrhythmia causing mutations. Figure 4.6 displays the amino acid spectrum of residues involved in disease for both genes. We found that 28% of all AAMs in these two channels have occurred at either a glycine or arginine residue (Figure 4.6a). This is similar to a broader protein analysis reported previously (Vitkup, et al., 2003). The contribution of a particular residue to the overall disease spectrum may be influenced by the proportion of the given amino acid in these proteins. Therefore, we determined the number of disease mutations at a given residue as a proportion of the total number of the residue in both proteins (Figure 4.6b). Arginine and glycine residues remain highly represented suggesting they are more likely to be involved in a disease phenotype compared to other residues. The proportion of tryptophan residues involved in disease is also high, although the total number of tryptophan residues is low.  133                                     Figure 4.6 Mutations Occur Predominantly at Arginine and Glycine Residues. a) Percentage of AAMs that occur at a given amino acid residue. b) Relative contribution of a given amino acid residue in HERG and KCNQ1 to AAMs. The total number of AAMs that occur at a given amino acid was divided by the total number of residue sites in the combination of the two protein regions used in the analysis. c) The number of AAMs in KCNQ1 and HERG at each of the twenty residues, proportional to their occurrence in the two channels, is significantly correlated (p=0.001) with the amino acid’s weighted average expected chemical severity. (Jackson and Accili, 2008).  134  The high involvement of some residues in disease may be due to a high mutation rate that occurs at CpG dinucleotides (Briscoe, et al., 2004) which is a result of a cytosine to thymine transition. This transition is possible in triplets coding for only five of the twenty amino acids, which are: arginine (4/6 codons), serine (1/6 codons), proline (1/4 codons), threonine (1/4 codons) and alanine (1/4 codons). Of the total number of AAMs in HERG and KCNQ1, only 13.3% are due to a C/T transition at a CpG dinucleotide. Of all arginine mutations resulting in disease, 47% are due to this specific nucleotide transition. These numbers suggest that CpG dinucleotides may contribute to the disease process but that this is not the only factor responsible for the high numbers of mutations in this subset of amino acids. A role for factors other that CpG dinucleotide hypermutability is also supported by the high involvement of glycines and tryptophans in disease (Figure 4.6b), which do not possess CpG dinucleotides. Our findings that the overall chemical severity of AAMs is greater than those of interspecific variation and SNPs, and that the chemical severity of AAMs correlates with expected chemical severity at those individual codons, suggest that chemical severity is on a continuum and that a threshold severity exists which, once crossed, results in disease. We might then expect that whether mutations at specific amino acids cause disease at all also depends on the expected chemical severity of the involved codon. To examine this, the proportion of total AAMs at a particular residue was plotted against the calculated weighted average of expected chemical severity (Figure 4.6c). A significant, positive correlation was found between these suggesting that expected chemical severity of a site contributes to the probability of obtaining a disease mutation, as well as to the severity of that mutation. 4.4 Discussion In this study, we show that that AAMs are overabundant at evolutionarily conserved and slowly evolving sites, which are likely critical for channel function and thus intolerant of changes in amino acid sequence. Because KCNQ1 and HERG are highly conserved, an underlying neutral mutational process could produce large numbers of mutations at conserved sites. Our data provide the quantitative backing to support an over representation of AAMs at evolutionarily conserved positions in these channels. A smaller than expected, but still substantial, numbers of AAMs were found at sites that show interspecific variation. However, only two sites in HERG and one in KCNQ1 are converted to residues that are both an AAM and  135 an interspecific change. Thus, the identity of the residue, rather than its location, is most responsible for producing the disease at sites that have undergone evolutionary change. HERG and KCNQ1 possess structurally defined domains with specific functions that have been strongly preserved throughout the course of evolution (Jespersen, et al., 2005; Sanguinetti and TristaniFirouzi, 2006).  We found that AAMs are found preferentially in some domains, even after accounting for their size and evolutionary conservation. In both channels, the pore region possesses an overabundance of mutations, while both N and Ctermini possess an underabundance of mutations. The extracellular linker region had an overabundance of mutations in HERG, whereas the intracellular linker had an overabundance of mutations in KCNQ1. This implies that the extracellular linker may be more important to overall function in HERG, whereas the intracellular linker is more critical for function in KCNQ1. The overabundance of mutations in the PAS domain, not present in KCNQ1, highlights its functional importance and suggests that that the addition of a functionally important domain in a protein can increase susceptibility to disease. Overall regional conservation, which takes into account the average variability per site in a domain, contributes to the uneven regional distribution of diseasecausing mutations in some proteins (Miller, et al., 2003), but we found this only for KCNQ1 (Figure 4.4b). Therefore, factors other than domain size, site conservation and regional conservation must influence the domain specific distribution of AAMs in HERG, and possibly in KCNQ1 as well. Nonsynonymous SNPs are underabundant at conserved sites in HERG, but distribute randomly in KCNQ1. The latter finding may be due to a small sample size, although the numbers of nsSNPs analyzed in HERG were similarly small. This difference between the channels may be because nsSNPs are less disruptive when located at conserved sites in KCNQ1, or because conserved sites in HERG are less tolerant to amino acid changes. Nevertheless, the distribution of analyzed nsSNPs in both channels suggests that they are phenotypically neutral. The different patterns of nsSNP distribution between the KCNQ1 and HERG underscore the need to identify and quantitatively analyze the distribution of more nsSNPs on each protein, and to ascertain their impact on channel function and arrhythmia susceptibility. In HERG, it is known that some, but not all, polymorphisms may alter channel function (Anson, et al., 2004), and also contribute to an increased QT interval duration (NewtonCheh, et al., 2007).  136 In both HERG and KCNQ1, AAMs are chemically more severe than interspecific changes tolerated throughout evolution or polymorphisms that are not associated with disease. Codons with a higher expected chemical severity are associated with disease mutations with a high chemical severity. These data are in keeping with those found previously in rhodopsin (Briscoe, et al., 2004) and suggest that the intrinsic potential of the involved codon contributes to disease chemical severity. We also provide novel evidence that the expected chemical severity of a codon contributes to the overrepresentation of certain amino acids in HERG and KCNQ1 AAMs, especially arginine, tryptophan and glycine. In a recent study of 437 proteins, these three amino acids were also highly overrepresented in DAMs (Vitkup, et al., 2003). Therefore, we predict that expected chemical severity plays an important role in determining the propensity of a given codon to cause disease in many proteins, in addition to other factors such as the residue’s roles in biosynthesis, function and stability of the channels. Finally, the chemical severities of interspecific changes in KCNQ1 and HERG also correlate with those expected for the codons, when only evolutionarily variable sites are considered. These novel data argue that, in addition to a predicted role of natural selection, the expected chemical severity of the codon contributes to variation observed over the course of evolution in these channels. Despite the presumed predominance of natural selection, the genetic code has been shown to influence the mutational process when evolutionary divergence is low (Benner, et al., 1994).  These data are significant given the uncertainty as to the role of natural selection versus nonadaptive forces in shaping genotypic and phenotypic variation (Lynch, 2007; Yi, 2006). Our analyses may be influenced by the fact that some mutations may, in a systematic way, never be detected. For example, AAMs that lead to death before natural birth, or to SUDs, may never be identified unless the fetus or victim is subsequently screened for arrhythmogenic mutations. This could, in turn, reduce the number of observed mutations unique to certain sites or functional domains. The evaluation of AAMs has broadened, and identification of mutations in KCNQ1 and HERG, and in other candidate genes, associated with SIDs and SUDs has been carried out (Arnestad, et al., 2007; Schwartz, et al., 2001; Tester and Ackerman, 2007; Wang, et al., 2007). Identification of AAMs in a more broad population may reveal a different subset of mutations that localize to unique regions within the channels, or more strongly support the  137 susceptibility of sites and functional domains identified in this study to arrhythmogenic mutations. 4.5 Conclusion Our study represents the first quantitative evolutionary and chemical severity analysis of AAMs in the HERG and KCNQ1 potassium channel genes. Unlike nsSNPs, AAMs preferentially locate to evolutionarily conserved, and functionally important, sites and regions within HERG and KCNQ1, and are chemically more severe than changes which occur in evolution. Expected chemical severity may contribute to the overrepresentation of certain residues in AAMs, as well as to changes observed throughout evolution. Our findings, together with those from other studies, suggest that novel DAMs and AAMs may be recognized quickly by surveying naturally occurring variation among species (Briscoe, et al., 2004). If a SNP identified in an individual does not appear in other species at that position, then it is likely to be diseasecausing. The location of AAMs (to conserved or variable regions and/or residues) may correlate with clinical severity or other characteristics of the diseases. In the case of Long QT syndrome, genotype and specific mutations have been shown to contribute to phenotype (Brink, et al., 2005; Crotti, et al., 2007; Moss, et al., 2002; Tan, et al., 2006) and the underlying genetic defects contribute to risk stratification, prevention and therapy (Priori, et al., 2003; Schwartz, 2005).  Unfortunately, there is still considerable variation in Long QT phenotype and age of disease onset (Roden, 2008). Therefore, continued discovery and mapping of mutations, as done in this study, along with parallel studies on disease phenotype will ultimately lead to a better understanding of the genotypephenotype relationship, help to better predict the outcome of novel disease mutations and aid in development of mutation specific therapies.     138  5. GENERAL DISCUSSION 5.1 Overview Molecular evolution results from the trialanderror process of genetic variation. The amino acid sequences of extant proteins represent the accumulation of changes that have been tolerated throughout evolution, and the conserved sites reflect residues that are important for protein structure and/or function. The overall goal of this thesis was to expand our current understanding of the evolution of voltagegated ion channels and the relation to channel function and human disease. Specifically, in Chapter 2, the sequence list of HCN channels was expanded, using the available genomic data of a broad range of species distributed across the metazoan lineage. We used the newly annotated sequences to construct an extensive phylogenetic history of the HCN family. In Chapter 3, we explored two of the newly identified sequences and cloned and functionally characterized two novel HCN genes from the urochordate species Ciona intestinalis. In Chapter 4, we examined the correlation between the evolutionary changes of voltagegated ion channels with disease mutations associated with cardiac arrhythmias. To do this, we identified ion channels with a significant number of known diseaseassociated mutations. Furthermore, by taking advantage of the >200 mutations reported in HERG and KCNQ1 that are associated with Long QT syndrome, atrial fibrillation, sudden infant death syndrome, and sudden unexplained death, we quantitatively analyzed their distribution and chemical severity in relation to those observed throughout evolution. 5.2 Molecular Evolution Genetic mutation drives evolutionary innovation and gene duplication is thought to be the major player in generating the molecular diversity seen today (Ohno, 1970). The direct consequence of duplication; however, is genetic redundancy, which carries with it high energetic costs. Therefore, in following a gene duplication event, the gene products can have three possible outcomes: nonfunctionalization, neofunctionalization, and subfunctionalization (Lynch and Conery, 2000; Lynch and Katju, 2004). Loss of function will occur through the accumulation of deleterious mutations, which will remove the gene from the pressures of natural selection. The latter two consequences, either the emergence of new function or the diversification of an  139 existing function, likely arise through advantageous mutations and positive Darwinian selection and/or neutral variation/genetic drift. The ultimate consequence is the preservation of only those genes that are nonredundant. The voltagegated ion channel superfamily is predicted to have formed from several gene duplications and single mutations, resulting in neofunctionalization (Anderson and Greenberg, 2001) of a common ancestral channel. As two rounds of complete genome duplication likely occurred throughout vertebrate evolution (Hughes and Friedman, 2003), several families of the VGF contain multiple vertebrate paralogs. Phylogenetic analyses are needed to determine whether the paralogous gene families arose through genomewide duplication or through individual gene duplication events. Regardless of their evolutionary trajectory, the functional retention of the paralogs throughout evolutionary history implies their physiological importance. By comparing the sequences of the extant channels within each subfamily found in a diverse range of species, the amino acids that have been conserved throughout evolutionary history can be identified. Sites that are intolerant to change are likely under strong evolutionary constraints and crucial to protein function. 5.3 HCN Evolution The phylogenetic analysis of protein families has become an essential tool for understanding protein structure and function as well as evolutionary trajectories of protein families. Examples of earlier studies include the sodium channel family (Lopreato, et al., 2001), the cadherins (Gallin, 1998), the sodiumcalcium exchanger (Marshall, et al., 2005; On, et al., 2008) and the ankyrin family (Cai and Zhang, 2006). These studies used phylogenetic analyses to assess rate of evolutionary change and the different selective pressures on different paralogs, to suggest the origins of vertebrate paralogs and the patterns of the duplication process, as well as to identify novel gene paralogs and orthologs in different species. In Chapter 2, the increasing availability of various assembled genome sequences was used to search for and annotate putative HCN genes in a variety of species. We generated an expanded list of putative HCN coding sequences and doubled the number of previously known HCN genes, with a greater representation of evolutionary time. In doing so, we more accurately completed sequence comparisons, phylogenetic analyses, and exon structure comparisons of the HCN gene family, and put forward a model of its molecular evolution.  140 5.3.1 Urochordate and Vertebrate HCN Channels Arose From a Single Ancestor For many ion channels, invertebrate species contain fewer duplicate genes (e.g. Na channel family) than vertebrates. Our findings on HCN channels are consistent with this pattern and suggest that the four vertebrate and three urochordate HCN isoforms evolved from a single ancestral sequence via independent duplication and diversification events. Our first data to support this came from the exon structure analysis. Exonintron structure comparisons had been previously used to assist in the evolutionary reconstruction of protein families (Rogozin, et al., 2005). Based on our exon comparisons of the HCN family, described in Chapter 2, the ancestral gene most likely had an exon structure similar to the current mammalian HCN genes. The duplication events likely occurred after the intron positions were fixed in the linear sequence. The second piece of evidence comes from the consensus of the various phylogenetic analyses described in Chapter 2. Regardless of the phylogenetic method or evolutionary model used to establish the events, urochordate HCNs arose through lineagespecific duplication events of a single common gene, and the vertebrate HCNs are predicted to have evolved in the order of HCN3, HCN4, and HCN1/HCN2 via three duplication events of the ancestral sequence. Alternatively, the presence of multiple vertebrate paralogs could be explained by one or more duplication events occurring prior to the divergence of urochordates and vertebrates, with subsequent gene loss occurring within the urochordate lineage. This hypothesis is plausible as significant gene loss has occurred in the genomic evolution of Ciona intestinalis (Hughes and Friedman, 2005); however, our data from Chapters 2 and 3 suggest otherwise. The sequence comparisons suggest that the extant HCN4 is most similar to that of the ancestral sequence, though future experiments are needed to explore this in greater detail. Lastly, the proposed evolutionary trajectory is well supported by our functional data generated in Chapter 3. The shared functional properties of the newly cloned ciHCNa and the previously characterized mammalian HCN channels suggest that the ancestor to the urochordate and vertebrate HCN channels likely displayed a mammalianlike phenotype, which supports our hypothesis of a single common origin.   141 5.3.2 HCN LineageSpecific Duplications As suggested for the sodium channel (Lopreato, et al., 2001) and ankyrin families (Cai and Zhang, 2006), we propose that HCN duplications occurred  within the urochordate lineages and that the three gene products have undergone independent and rapid sequence diversification. Three putative urochordate HCN sequences had been previously identified from Ciona intestinalis (Okamura, et al., 2005; Satoh, et al., 2003). Our detailed annotation and analysis of these sequences demonstrated that each of these sequences were as similar to each other as they were to the vertebrate homologs, suggesting that they have undergone a similar amount of evolutionary change in the 550 MY since the lineage divergence as the vertebrate sequences have experienced in over one billion years of cumulative evolutionary history. One plausible explanation for this finding is the rapid evolutionary rate known to be present in the Ciona species (Holland and GibsonBrown, 2003). Furthermore, our analysis of the HCN sequences from Ciona savignyi provided us with orthologous HCN gene pairs to better assess their evolutionary trajectory throughout the urochordate lineage. Phylogenetic analyses suggest that the three Ciona HCN genes arose via two duplication events prior to the divergence of C. intestinalis and C. savignyi. Based on exon structure, sequence identity, and phylogenetic position, we proposed that ciHCNb was evolutionarily closest to the common ancestor, while ciHCNa and ciHCNc arose subsequently through lineagespecific duplications (Jackson, et al., 2007). Our functional data, shown in Chapter 3, suggests that channel phenotype is not in parallel with these findings and provide a great example of the importance of functional data to support phylogenetic findings. Overall, ciHCNb may share more common gene and sequence features with the ancestral state; however, 550MY of independent evolution has produced channels with distinct biophysical properties. In general, the extant ciHCNa is likely more phenotypically similar to the putative ancestral state. Our phylogenetic analyses suggest that duplications also occurred independently in the fish lineages, resulting in several HCN coorthologs. This resembles the pattern for ion channels in general, where over half of the ion channel genes in Fugu had undergone a lineagespecific duplication event to produce coorthologs (Jegla, et al., 2009). Because teleost fish are predicted to have undergone a complete genome duplication early in their own lineage, approximately 225 300 MYA ((Hoegg, et al., 2004) and references therein), this distribution pattern is not unexpected. Nevertheless, because some of the putative HCN sequences were incompletely  142 assembled or showed evidence of intron insertion, further analyses are required to determine whether the duplicates are actually expressed and functional or whether they have become pseudogenes (Vanin, 1985). Genetic duplication is costly and the most common outcome is nonfunctionalization with the eventual loss of one of the duplicate copies (Lynch and Conery, 2000; Lynch and Katju, 2004). Still, gene duplication is also the source of genetic innovation and evolutionary novelty (Taylor and Raes, 2004). While one gene copy can maintain the original function, the other copy is free to accumulate amino acid changes (Li and Gojobori, 1983). If the changes lead to a functional divergence, duplicate genes can still be retained. In fact, this has been observed for the voltagegated sodium channel family (Arnegard, et al., 2010; Novak, et al., 2006; Widmark, et al., 2011) where four duplicate pairs have been conserved and remain functional in modern teleosts. This would support the relevance for exploring the role of HCN duplicate pairs in fish species. 5.3.3 Future Evolutionary Analyses of HCN Our studies provide a strong basis from which to refine the proposed model of HCN evolution. A key limitation in our study was the availability of assembled genomes and the sampling of HCN sequences from a diverse range of species. As more genomes become available, especially of species evolving intermediate to the invertebrates and mammals, and as the functional analysis of these novel HCN genes progresses, more elaborate analyses will be possible. For example, some of the branches in our phylogenetic analysis remain uncertain (e.g. HCN3 clade). It is likely that some of these discrepancies will disappear as new sequences are added however; recent studies indicate that resolution will not always be possible (Philippe, et al., 2011). In addition to the phylogenetic analyses presented here, an increased number and sampling of HCN sequences will provide the foundation for a number of other evolutionary experiments, including: ancestral protein reconstruction and crystallization (Dean and Thornton, 2007; Ortlund, et al., 2007; Thornton, 2004; Yang, et al., 1995), analysis of positive and/or purifying selection and estimates of evolutionary rates (Aguileta, et al., 2006; Bielawski and Yang, 2004; Huelsenbeck and Dyer, 2004; Hughes and Criscuolo, 2008; Nei, 2005; Suzuki and Gojobori, 1999; Yang and Nielsen, 2002), and synteny/chromosomal location comparisons of HCN genes and adjacent gene families to examine timing of duplication events, evolutionary origins and gene trajectories (Chopra, et al., 2007; Novak, et al., 2006; Widmark, et al., 2011;  143 Zakon, et al., 2011). Lastly, further exploration of available bacterial genomes will provide opportunities to identify bacterial HCN genes and elucidate the true evolutionary origins of this gene family. Currently, a single potassium selective, cyclic nucleotide gated channel has been identified and studied from the bacterial species, Mesorhizobium loti (Nimigean, et al., 2004; Schunke, et al., 2011). Nevertheless, the evolutionary connection between this gene and that of HCN gene family remains to be solved. 5.4 HCN Channels in Ciona intestinalis Ciona intestinalis is a urochordate species that diverged from the vertebrate lineage over 550MYA, serving as an example of the closest extant ancestor of the vertebrate lineage. Evolving at the pivotal evolutionary branch point of invertebrates and vertebrates, urochordates provide an ancestral genetic reference point for the vertebrate lineage (Corbo, et al., 2001). Three urochordate HCN sequences had been reported (Okamura, et al., 2005) and ESTs were identified for each sequence (Satoh, et al., 2003). In Chapter 3, cloning and functional characterization of two of the channels, ciHCNa and ciHCNb, from the siphon of an adult sample is described and discussed. Our initial phylogenetic analyses and sequence comparisons predicted that the channel functions would be similar, as many regions of the channels known to be responsible for certain functions are highly conserved between Ciona and vertebrate sequences. Indeed, the in vitro analyses demonstrated that both genes produced channels with some general biophysical properties associated with those known for HCN channels.  However, one of the isoforms displayed an unusual gating phenotype and did not possess an Nlinked glycosylation sequon near the pore. The other was similar to the vertebrate isoforms, suggesting that it may be phenotypically closer to the common HCN ancestor. 5.4.1 ciHCNa Displays a MammalianLike Phenotype Inconsistent with our predictions from our phylogenetic analysis, ciHCNa shared more phenotypic features with mammalian HCN channels, compared to ciHCNb. In particular, ciHCNa possessed the HCN Nglycosylation sequon near the channel pore and underwent post translational modification in vitro. Also, similar to mammalian HCN channels, the ciHCNa channels produced a very small voltageindependent Iinst and a relatively large voltagedependent If, the latter being characteristically blocked by extracellular Cs +  (DiFrancesco, 1982). Both  144 channel types were augmented by extracellular K +  (DiFrancesco, 1982) and modulated by a plasmamembrane permeant cAMP analogue, which had been previously used for HCN channel characterization (Ouyang, et al., 2007). For the cyclic nucleotide experiments, our preliminary goal was to determine whether or not the channels were sensitive to cyclic nucleotides.  Our sequence analysis in Chapter 2 demonstrated that, while the CNBD in Ciona was only 5060% identical with that of the vertebrate isoforms, key residues involved in nucleotide binding were conserved. Therefore, we hypothesized that these channels would bind the cyclic nucleotides. However, whether the binding would lead to channel modulation was unknown. In Chapter 3, we show cyclic nucleotide modulation of the channels, however; the interpretation of our findings was limited by the use of a high concentration of cAMP analogue and the duration of drug application required to observe a response. While our results are consistent with a direct effect of cAMP, it is possible that the cAMP effect could have been indirect through a protein other than the HCN channel (e.g. PKA or Epac). Therefore, excisedpatch experiments in which cAMP can be applied to the intracellular surface of the membrane are necessary to determine if these HCN isoforms are sensitive to this cyclic nucleotide. 5.4.2 The Function of ciHCNb has Diverged Unlike ciHCNa, ciHCNb does not undergo Nlinked glycosylation and produces large voltageindependent currents across a wide range of membrane potentials. The lack of N glycosylation is unusual, as many ion channels and other intrinsic proteins of the plasma membrane possess this posttranslational modification, which, in many cases, contributes to the efficiency of cell surface expression. Nevertheless, the invertebrate forms of HCN do not possess this modification, yet express robust currents (e.g. spIH) (Hegle, et al., 2010). The voltageindependent current comprised approximately 50% of the total current and was significantly reduced in the presence of external Cs+ and ZD7288, two known HCN channel blockers (BoSmith, et al., 1993; DiFrancesco, 1982), indicating that the current was passing through the same HCN channel pore. Interestingly, ZD7288 is known to block HCN channels at an intracellular binding site (Chan, et al., 2009; Cheng, et al., 2007; Shin, et al., 2001). Thus, when the drug is applied to the extracellular surface, it must first permeate across the cell  145 membrane, which may explain the extended incubation required for the onset of block (Berger, et al., 1994; BoSmith, et al., 1993; Harris and Constanti, 1995). Some authors have suggested that VIC represents a nonzero limiting open probability for HCN channels at positive voltages. Inadequate channel closure has been supported by an alanine scan of the pore region, revealing a weak interaction between the voltage sensor and the pore, resulting in an open probability that does not reach zero (Macri, et al., 2009). Furthermore, HCN mutations that result in channels locked in an open state (Rothberg, et al., 2003) also support the idea that the coupling between the pore and voltagesensor may not be that strong. Overall, the characteristics of ciHCNb are parallel to those observed for HCN2 channels with mutations in the S4S5 linker, which exhibit large voltageindependent currents due to incomplete channel closure (Chen, et al., 2001a; Macri and Accili, 2004). A more complex interpretation suggests that VIC and If result from channels that are not in rapid equilibrium with each other and belong to separate populations (Proenza and Yellen, 2006). Future experiments, including an examination of the kinetics of block, will be required to assess this interpretation in ciHCNb. Further analyses and sitedirected mutagenesis experiments focusing on the S4S5 linker region of the Ciona HCNs may also provide insight into the regions that contribute to VIC and/or responsible for channel closure. Finally, it will be of interest to examine cells of Ciona, which express ciHCNb in vivo, in which a constant depolarizing current in the absence of any hyperpolarizing stimulus would be predicted. Physiologically, this would likely lead to cells with an increased susceptibility to arrhythmogenic activity. 5.4.3 Future Directions for HCN in Ciona intestinalis In Chapter 2, three different HCN genes from urochordate species were analyzed. In Chapter 3, the cloning and characterization of two of the channels was presented, with predictions about their functionality based on sequence analysis. Nevertheless, EST evidence suggests that ciHCNc is also expressed in vivo (Satoh, et al., 2003), and therefore, may retain some physiological function. Key sequence differences exist in the channel selectivity filter, including a G/S substitution in the potassium channel signature sequence (CIGYS). In mammalian HCN channels, substitution of the second glycine residue by serine (G404S in HCN2), reduced the slow activating current (Macri, et al., 2002). If Ciona HCNc is functional, a different mutational tolerance in the selectivity filter seems likely, possibly reflecting an adaptive  146 process that has enables these channels to fill a different functional niche in these species. It is likely that the channel will display significantly different permeation or gating functions. If, however, the channel is found to be nonfunctional, from a biophysical perspective, further analyses would be warranted to determine if this third channel might act as an accessory subunit. In Chapter 3, the biophysical characteristics of ciHCNa and ciHCNb were presented for in vitro experiments. The role that these channels play in vivo, and their tissue distribution, remains to be elucidated. As mammalian HCN channels have been shown to coassemble and heteromerize, both in vitro and in vivo (Much, et al., 2003; Whitaker, et al., 2007), co immunoprecipitation experiments from Ciona tissue will shed light on the occurrence of this for these constructs. Additionally, coinjection and/or cotransfection in vitro experiments or the generation of concatenated channels., similar to those done for mammalian HCN channels (Altomare, et al., 2003; Chen, et al., 2001b; Ulens and Tytgat, 2001a), should reveal the ability of the channels to coassemble. Recently, a putative ortholog of ciHCNa was identified in a related tunicate species, Botryllus schlosseri (Hellbach, et al., 2011) and was shown to be strongly expressed and functional in the cardiac pacemaking tissue. This provides strong support for a putative role of ciHCNa in the cardiac function of Ciona intestinalis and could represent a precursor of the vertebrate conduction system. This physiological link supports an even greater need to explore the function of these channels in the tunicate species, as they could shed light on the ancestral origins of HCN channels and provide a unique and prolific system to explore drug development and human disease mechanisms related to pacemaker dysfunction.  Additionally, an independent voltage sensing domain protein has been identified in Ciona intestinalis and was shown to behave as a proton permeation pathway (Ramsey, et al., 2010; Ramsey, et al., 2006; Sasaki, et al., 2006). According to a hypothesis of ion channel evolution, the six transmembrane voltagegated ion channels arose from the fusion of a twotransmembrane channel, similar to the K +  inward rectifier channels, and a fourtransmembrane voltage sensing domain protein. The discovery of proton transport provides the first evidence of a functional role for the latter of these two protein types. In future studies, the pH sensitivity for the novel ciHCN clones could be examined to determine whether or not proton permeation occurs in these channels, in addition to the mixed cation current passing through the HCN ion conducting pore.  147  In conjunction with experiments previously described for the HCN channel family as a whole, experiments that explore the rate of evolutionary change that has occurred in these three genes will provide some insight into the selective pressures they have faced throughout evolution. When orthologs from C. intestinalis and C. savignyi were analyzed for selection, preliminary results suggested that all three gene pairs were undergoing purifying selection. Supported by the EST evidence, sufficient functional divergence has likely occurred following HCN duplication in the urochordates to enable the retention of each of the genes and suggests they play an in vivo role in this ascidian species. 5.5 Evolution and Disease Susceptibility The importance of an amino acid residue for overall protein function can be inferred by its conservation over the course of metazoan evolution. Previous studies have demonstrated that amino acid mutations associated with human disease are preferentially located at evolutionarily conserved residues (Briscoe, et al., 2004; Miller and Kumar, 2001; Mooney and Klein, 2002; Subramanian and Kumar, 2006). The objective of Chapter 4 was to perform an accurate analysis of the correlation between evolutionary change and diseases associated with voltagegated potassium channels. We took advantage of the >200 mutations reported for HERG and KCNQ1, two proteins that play critical roles in cardiac repolarization; for which disease mutations are associated with Long QT syndrome, atrial fibrillation, sudden infant death syndrome, and sudden unexplained death. For the first time, we quantitatively analyzed the distribution of disease mutations associated with these arrhythmias. Our findings suggest that arrhythmiaassociated mutations, and not benign polymorphisms, are preferentially located at evolutionarily conserved and functionally important sites. In addition, they are chemically more severe than the changes observed throughout evolution. We were also interested in determining whether or not specific amino acids were targets for AAMs. We found that 28% of all AAMs occurred at either glycine or arginine residues, a finding that is supported by earlier reports (Vitkup, et al., 2003). We hypothesized that this could be due to a higher prevalence of these two residues in these channels; however, once we corrected for this, these two residues remained highly represented. In a more recent analysis, transmembrane regions were found to bury more residues and thereby evolve at slower rates. In turn, transmembrane residues have an increased susceptibility to diseasecausing polymorphisms  148 (Oberai, et al., 2009). Followup experiments will be to break down the amino acid distribution by protein location to see if the increased prevalence of arginine and glycine residues is due to the transmembrane location. Another explanation for arginine involvement is the high mutation rate at CpG dinucleotides (Briscoe, et al., 2004). In fact, our data shows that 47% of the mutations at arginine residues are due to CpG mutations (Suzuki, et al., 2009). Overall, our analyses are influenced by the fact that some mutations will never be detected; either through lack of testing, variable phenotypic expression, and therefore, lack of identification, or because the mutation was lethal and lead to death prior to natural birth. Also, in this study, we have only addressed nonsynonomous changes as they are the primary basis of for genetic evolution. Nevertheless, synonomous changes at splice sites and RNA editing could also play roles, and should be assessed in future analyses. 5.6 HCN in Health and Disease Mutations in the HCN channels have been associated with sinus arrhythmias and are frequently associated with sinus node dysfunction (SND) (LaishFarkash, et al., 2010; Milanesi, et al., 2006; Nof, et al., 2007; SchulzeBahr, et al., 2003; Ueda, et al., 2004). Collectively, these studies suggest that a reduction in cardiac If due to lossoffunction mutations in HCN4 is associated with cardiac rhythm dysfunction (Baruscotti, et al., 2010a). In addition, alterations in expression profiles have been associated with other cardiac conditions such as Brugada Syndrome (Ueda, et al., 2009), cardiac hypertrophy (WeiQing, et al., 2011) and in animal models of congestive heart failure (Zicha, et al., 2005) and SND (Yeh, et al., 2009). Several variants have also been identified through mutational screening of HCN1 and HCN2 in individuals with epilepsy (Tang, et al., 2008), though their pathogenicity remains to be determined. Overall, HCN channels clearly play a role in health and disease, though the exact connections have only recently been identified. A natural extension of this thesis would be to look at HCN channel evolution and the relation to disease susceptibility. Unfortunately, analyses such as those described in Chapter 4 are limited by the low number of HCN diseasecausing mutations that have been identified. Several reasons are possible for the low prevalence of known mutations, one of which is that they are benign and do not cause disease. Another explanation is that HCN mutations are embryonically lethal, thereby escaping detection, as indicated by various HCN4 knockout mouse  149 models (Baruscotti, et al., 2010b; Harzheim, et al., 2008; Herrmann, et al., 2007; Hoesl, et al., 2008; Nof, et al., 2010; Stieber, et al., 2003a). Another explanation is that the mutations are not identified because they are not routinely investigated. Sinus node dysfunction is associated with symptoms ranging from weakness and palpitations to syncope. While SND patients may be asymptomatic and unlikely to undergo clinical investigation, syncope accounts for 1.5% of all emergency room visits (Moya, et al., 2009), and should not be seen as a benign phenotype. Unlike the situation for Long QT syndrome and several other cardiac arrhythmia conditions, there is no common clinical genetic test to identify potential molecular causes of SND. Until genetic screening is readily available for not only those individuals with putative inherited forms of bradycardia, heart block, and sinus node dysfunction (Baruscotti, et al., 2010a), but also with familial forms of epilepsy (Huang, et al., 2009; Jung, et al., 2010; Noam, et al., 2011; Reid, et al., 2011) or other neurological conditions associated with HCN channels such as chronic pain (Lewis and Chetkovich, 2010; Luo, et al., 2007; Momin, et al., 2008; Papp, et al., 2010; Takasu, et al., 2010) and Parkinson’s disease (Chan, et al., 2011), our knowledge of the prevalence of disease mutations in HCN will remain limited. For example, a recent postmortem genetic analysis of sudden unexpected death in epilepsy revealed six novel and 3 previously reported nonsynonomous mutations in each of the HCN isoforms (Tu, et al., 2011). When these genetic screens become more accessible, either through research or clinical avenues, and our list of known diseasecausing mutations in HCN expands, analyses such as that presented in Chapter 4, will be essential for interpreting results and assess the clinical implications of any genetic findings. 5.7 Summary In this thesis, we performed indepth evolutionary analyses of voltagegated potassium channels, including HCN and the delayed rectifiers. Our data from the HCN phylogenetic and sequence analyses will provides a strong foundation for refining the proposed model of HCN evolution. As more genomes become available, and with advances in the functional analysis of HCN genes, our study will be a valuable tool to aid in the planning of experiments for exploring the structure/function relationship of HCN channels. Data from the functional analyses of HCN channels in Ciona intestinalis shed light on the ancestral origin of vertebrate HCN function. To understand the functional divergence that occurred within the vertebrate paralogs, and to assess  150 when and how the duplications took place, it is essential to have this genetic reference point. In turn, by gaining an understanding of how the four vertebrate paralogs evolved independently, and by examining the relationship to physiological function, we can further understand their expression in the human body. We have clearly demonstrated a strong interdependent relationship among molecular evolution, channel function, and human disease, such as cardiac channelopathies. The ongoing mapping of newly discovered ion channel mutations in relation to evolutionary change is key to assessing disease susceptibility and the accurate determination of genotypephenotype relationships. Personalized medicine is the future of health care and these analyses will ultimately aid in the production and implementation of mutationspecific therapies for individuals affected by cardiac channelopathies.     151 REFERENCES Inherited Arrhythmias Database. http://www.fsm.it/cardmoc/  Accili EA, Redaelli G, DiFrancesco D. 1997. Differential control of the hyperpolarization activated current (i(f)) by cAMP gating and phosphatase inhibition in rabbit sinoatrial node myocytes. J Physiol 500 (Pt 3):64351.  Aguileta G, Bielawski JP, Yang Z. 2006. Evolutionary rate variation among vertebrate beta globin genes: implications for dating gene family duplication events. Gene 380(1):219.  Alig J, Marger L, Mesirca P, Ehmke H, Mangoni ME, Isbrandt D. 2009. Control of heart rate by cAMP sensitivity of HCN channels. Proc Natl Acad Sci U S A 106(29):1218994.  Altomare C, Bucchi A, Camatini E, Baruscotti M, Viscomi C, Moroni A, DiFrancesco D. 2001. Integrated allosteric model of voltage gating of HCN channels. J Gen Physiol 117(6):51932.  Altomare C, Terragni B, Brioschi C, Milanesi R, Pagliuca C, Viscomi C, Moroni A, Baruscotti M, DiFrancesco D. 2003. Heteromeric HCN1HCN4 channels: a comparison with native pacemaker channels from the rabbit sinoatrial node. J Physiol 549(Pt 2):34759.  Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ. 1990. Basic local alignment search tool. J Mol Biol 215(3):40310.  Anderson JA, Huprikar SS, Kochian LV, Lucas WJ, Gaber RF. 1992. Functional expression of a probable Arabidopsis thaliana potassium channel in Saccharomyces cerevisiae. Proc Natl Acad Sci U S A 89(9):373640.  Anderson PA, Greenberg RM. 2001. Phylogeny of ion channels: clues to structure and function. Comp Biochem Physiol B Biochem Mol Biol 129(1):1728.  Anson BD, Ackerman MJ, Tester DJ, Will ML, Delisle BP, Anderson CL, January CT. 2004. Molecular and functional characterization of common polymorphisms in HERG (KCNH2) potassium channels. Am J Physiol Heart Circ Physiol 286(6):H243441.  Arbiza L, Duchi S, Montaner D, Burguet J, PantojaUceda D, PinedaLucena A, Dopazo J, Dopazo H. 2006. Selective pressures at a codonlevel predict deleterious mutations in human disease genes. J Mol Biol 358(5):1390404.  Arinsburg SS, Cohen IS, Yu HG. 2006. Constitutively active Src tyrosine kinase changes gating of HCN4 channels through direct binding to the channel proteins. J Cardiovasc Pharmacol 47(4):57886.  Arnegard ME, Zwickl DJ, Lu Y, Zakon HH. 2010. Old gene duplication facilitates origin and diversification of an innovative communication systemtwice. Proc Natl Acad Sci U S A 107(51):221727.  152  Arnestad M, Crotti L, Rognum TO, Insolia R, Pedrazzini M, Ferrandi C, Vege A, Wang DW, Rhodes TE, George AL, Jr. and others. 2007. Prevalence of longQT syndrome gene variants in sudden infant death syndrome. Circulation 115(3):3617.  Ashcroft FM. 2006. From molecule to malady. Nature 440(7083):4407.  Attwell D, Wilson M. 1980. Behaviour of the rod network in the tiger salamander retina mediated by membrane properties of individual rods. J Physiol 309:287315.  Aydin A, Bahring S, Dahm S, Guenther UP, Uhlmann R, Busjahn A, Luft FC. 2005. Single nucleotide polymorphism map of five longQT genes. J Mol Med 83(2):15965.  Azene EM, Sang D, Tsang SY, Li RA. 2005. Poretogate coupling of HCN channels revealed by a pore variant that contributes to gating but not permeation. Biochem Biophys Res Commun 327(4):113142.  Azene EM, Xue T, Li RA. 2003. Molecular basis of the effect of potassium on heterologously expressed pacemaker (HCN) channels. J Physiol 547(Pt 2):34956.  Bader CR, Bertrand D. 1984. Effect of changes in intra and extracellular sodium on the inward (anomalous) rectification in salamander photoreceptors. J Physiol 347:61131.  Bader CR, Bertrand D, Schwartz EA. 1982. Voltageactivated and calciumactivated currents studied in solitary rod inner segments from the salamander retina. J Physiol 331:25384.  Bader CR, Macleish PR, Schwartz EA. 1979. A voltageclamp study of the light response in solitary rods of the tiger salamander. J Physiol 296:126.  Baker K, Warren KS, Yellen G, Fishman MC. 1997. Defective "pacemaker" current (Ih) in a zebrafish mutant with a slow heart rate. Proc Natl Acad Sci U S A 94(9):45549.  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(6604):7880.  Baruscotti M, Bottelli G, Milanesi R, Difrancesco JC, Difrancesco D. 2010a. HCNrelated channelopathies. Pflugers Arch.  Baruscotti M, Bucchi A, Difrancesco D. 2005. Physiology and pharmacology of the cardiac pacemaker ("funny") current. Pharmacol Ther 107(1):5979.  Baruscotti M, Bucchi A, Viscomi C, Mandelli G, Consalez G, GnecchiRusconi T, Montano N, Casali KR, Micheloni S, Barbuti A and others. 2010b. Deep bradycardia and heart block caused by inducible cardiacspecific knockout of the pacemaker channel gene Hcn4. Proc Natl Acad Sci U S A.   153 Becker D, Dreyer I, Hoth S, Reid JD, Busch H, Lehnen M, Palme K, Hedrich R. 1996. Changes in voltage activation, Cs+ sensitivity, and ion permeability in H5 mutants of the plant K+ channel KAT1. Proc Natl Acad Sci U S A 93(15):81238.  Bell DC, Yao H, Saenger RC, Riley JH, Siegelbaum SA. 2004. Changes in local S4 environment provide a voltagesensing mechanism for mammalian hyperpolarizationactivated HCN channels. J Gen Physiol 123(1):519.  Bellocq C, van Ginneken AC, Bezzina CR, Alders M, Escande D, Mannens MM, Baro I, Wilde AA. 2004. Mutation in the KCNQ1 gene leading to the short QTinterval syndrome. Circulation 109(20):23947.  Bender RA, Brewster A, Santoro B, Ludwig A, Hofmann F, Biel M, Baram TZ. 2001. Differential and agedependent expression of hyperpolarizationactivated, cyclic nucleotide gated cation channel isoforms 14 suggests evolving roles in the developing rat hippocampus. Neuroscience 106(4):68998.  Benner SA, Cohen MA, Gonnet GH. 1994. Amino acid substitution during functionally constrained divergent evolution of protein sequences. Protein Eng 7(11):132332.  Berger F, Borchard U, Gelhaar R, Hafner D, Weis T. 1994. Effects of the bradycardic agent ZD 7288 on membrane voltage and pacemaker current in sheep cardiac Purkinje fibres. Naunyn Schmiedebergs Arch Pharmacol 350(6):67784.  Berman HM, Ten Eyck LF, Goodsell DS, Haste NM, Kornev A, Taylor SS. 2005. The cAMP binding domain: an ancient signaling module. Proc Natl Acad Sci U S A 102(1):4550.  Bezanilla F. 2008. How membrane proteins sense voltage. Nat Rev Mol Cell Biol 9(4):32332. Biel M, Schneider A, Wahl C. 2002. Cardiac HCN channels: structure, function, and modulation. Trends Cardiovasc Med 12(5):20612.  Biel M, WahlSchott C, Michalakis S, Zong X. 2009. Hyperpolarizationactivated cation channels: from genes to function. Physiol Rev 89(3):84785.  Bielawski JP, Yang Z. 2004. A maximum likelihood method for detecting functional divergence at individual codon sites, with application to gene family evolution. J Mol Evol 59(1):12132.  Birney E, Andrews TD, Bevan P, Caccamo M, Chen Y, Clarke L, Coates G, Cuff J, Curwen V, Cutts T and others. 2004. An overview of Ensembl. Genome Res 14(5):9258.  BoSmith RE, Briggs I, Sturgess NC. 1993. Inhibitory actions of ZENECA ZD7288 on whole cell hyperpolarization activated inward current (If) in guineapig dissociated sinoatrial node cells. Br J Pharmacol 110(1):3439.  Botstein D, Risch N. 2003. Discovering genotypes underlying human phenotypes: past successes for mendelian disease, future approaches for complex disease. Nat Genet 33 Suppl:22837.  154  Brandt MC, EndresBecker J, Zagidullin N, Motloch LJ, Er F, Rottlaender D, Michels G, Herzig S, Hoppe UC. 2009. Effects of KCNE2 on HCN isoforms: distinct modulation of membrane expression and single channel properties. Am J Physiol Heart Circ Physiol 297(1):H35563.  Brewster AL, Bernard JA, Gall CM, Baram TZ. 2005. Formation of heteromeric hyperpolarizationactivated cyclic nucleotidegated (HCN) channels in the hippocampus is regulated by developmental seizures. Neurobiol Dis 19(12):2007.  Brink PA, Crotti L, Corfield V, Goosen A, Durrheim G, Hedley P, Heradien M, Geldenhuys G, Vanoli E, Bacchini S and others. 2005. Phenotypic variability and unusual clinical severity of congenital longQT syndrome in a founder population. Circulation 112(17):260210.  Brioschi C, Micheloni S, Tellez JO, Pisoni G, Longhi R, Moroni P, Billeter R, Barbuti A, Dobrzynski H, Boyett MR and others. 2009. Distribution of the pacemaker HCN4 channel mRNA and protein in the rabbit sinoatrial node. J Mol Cell Cardiol 47(2):2217.  Briscoe AD, Gaur C, Kumar S. 2004. The spectrum of human rhodopsin disease mutations through the lens of interspecific variation. Gene 332:10718.  Brown H, Difrancesco D. 1980. Voltageclamp investigations of membrane currents underlying pacemaker activity in rabbit sinoatrial node. J Physiol 308:33151.  Brown HF, DiFrancesco D, Noble SJ. 1979. How does adrenaline accelerate the heart? Nature 280(5719):2356.  Brugada R, Hong K, Dumaine R, Cordeiro J, Gaita F, Borggrefe M, Menendez TM, Brugada J, Pollevick GD, Wolpert C and others. 2004. Sudden death associated with shortQT syndrome linked to mutations in HERG. Circulation 109(1):305.  Bucchi A, Baruscotti M, DiFrancesco D. 2002. Currentdependent block of rabbit sinoatrial node I(f) channels by ivabradine. J Gen Physiol 120(1):113.  Bucchi A, Tognati A, Milanesi R, Baruscotti M, DiFrancesco D. 2006. Properties of ivabradine induced block of HCN1 and HCN4 pacemaker channels. J Physiol 572(Pt 2):33546.  Butler A, Wei AG, Baker K, Salkoff L. 1989. A family of putative potassium channel genes in Drosophila. Science 243(4893):9437.  Cai X, Zhang Y. 2006. Molecular evolution of the ankyrin gene family. Mol Biol Evol 23(3):5508.  Cargill M, Altshuler D, Ireland J, Sklar P, Ardlie K, Patil N, Shaw N, Lane CR, Lim EP, Kalyanaraman N and others. 1999. Characterization of singlenucleotide polymorphisms in coding regions of human genes. Nat Genet 22(3):2318.   155 Chan CS, Glajch KE, Gertler TS, Guzman JN, Mercer JN, Lewis AS, Goldberg AB, Tkatch T, Shigemoto R, Fleming SM and others. 2011. HCN channelopathy in external globus pallidus neurons in models of Parkinson's disease. Nat Neurosci 14(1):8592.  Chan YC, Wang K, Au KW, Lau CP, Tse HF, Li RA. 2009. Probing the bradycardic drug binding receptor of HCNencoded pacemaker channels. Pflugers Arch 459(1):2538.  Chen J, Mitcheson JS, Lin M, Sanguinetti MC. 2000. Functional roles of charged residues in the putative voltage sensor of the HCN2 pacemaker channel. J Biol Chem 275(46):3646571.  Chen J, Mitcheson JS, TristaniFirouzi M, Lin M, Sanguinetti MC. 2001a. The S4S5 linker couples voltage sensing and activation of pacemaker channels. Proc Natl Acad Sci U S A 98(20):1127782.  Chen J, Weber M, Yon Um S, Walsh CA, Tang Y, McDonald TV. 2011. A Dual Mechanism for I(Ks) Current Reduction by the Pathogenic Mutation KCNQ1S277L. Pacing Clin Electrophysiol 34(12):16521664.  Chen S, Wang J, Siegelbaum SA. 2001b. Properties of hyperpolarizationactivated pacemaker current defined by coassembly of HCN1 and HCN2 subunits and basal modulation by cyclic nucleotide. J Gen Physiol 117(5):491504.  Chen YH, Xu SJ, Bendahhou S, Wang XL, Wang Y, Xu WY, Jin HW, Sun H, Su XY, Zhuang QN and others. 2003. KCNQ1 gainoffunction mutation in familial atrial fibrillation. Science 299(5604):2514.  Cheng L, Kinard K, Rajamani R, Sanguinetti MC. 2007. Molecular mapping of the binding site for a blocker of hyperpolarizationactivated, cyclic nucleotidemodulated pacemaker channels. J Pharmacol Exp Ther 322(3):9319.  Cho WJ, Drescher MJ, Hatfield JS, Bessert DA, Skoff RP, Drescher DG. 2003. Hyperpolarizationactivated, cyclic AMPgated, HCN1like cation channel: the primary, full length HCN isoform expressed in a saccular haircell layer. Neuroscience 118(2):52534.  Chopra SS, Watanabe H, Zhong TP, Roden DM. 2007. Molecular cloning and analysis of zebrafish voltagegated sodium channel beta subunit genes: implications for the evolution of electrical signaling in vertebrates. BMC Evol Biol 7:113.  Chung WK, Shin M, Jaramillo TC, Leibel RL, LeDuc CA, Fischer SG, Tzilianos E, Gheith AA, Lewis AS, Chetkovich DM. 2009. Absence epilepsy in apathetic, a spontaneous mutant mouse lacking the h channel subunit, HCN2. Neurobiol Dis 33(3):499508.  Clancy CE, Kass RS. 2005. Inherited and acquired vulnerability to ventricular arrhythmias: cardiac Na+ and K+ channels. Physiol Rev 85(1):3347.   156 Cole KS, Curtis HJ. 1939. Electric Impedance Of The Squid Giant Axon During Activity. J Gen Physiol 22(5):64970.  Corbo JC, Di Gregorio A, Levine M. 2001. The ascidian as a model organism in developmental and evolutionary biology. Cell 106(5):5358.  Craven KB, Zagotta WN. 2004. Salt bridges and gating in the COOHterminal region of HCN2 and CNGA1 channels. J Gen Physiol 124(6):66377.  Craven KB, Zagotta WN. 2006. CNG AND HCN CHANNELS: Two Peas, One Pod. Annu Rev Physiol 68:375401.  Crotti L, Spazzolini C, Schwartz PJ, Shimizu W, Denjoy I, SchulzeBahr E, Zaklyazminskaya EV, Swan H, Ackerman MJ, Moss AJ and others. 2007. The common longQT syndrome mutation KCNQ1/A341V causes unusually severe clinical manifestations in patients with different ethnic backgrounds: toward a mutationspecific risk stratification. Circulation 116(21):236675.  Cui J, Melman Y, Palma E, Fishman GI, McDonald TV. 2000. Cyclic AMP regulates the HERG K(+) channel by dual pathways. Curr Biol 10(11):6714.  Curran ME, Splawski I, Timothy KW, Vincent GM, Green ED, Keating MT. 1995. A molecular basis for cardiac arrhythmia: HERG mutations cause long QT syndrome. Cell 80(5):795803.  Dean AM, Thornton JW. 2007. Mechanistic approaches to the study of evolution: the functional synthesis. Nat Rev Genet 8(9):67588.  Decher N, Bundis F, Vajna R, Steinmeyer K. 2003. KCNE2 modulates current amplitudes and activation kinetics of HCN4: influence of KCNE family members on HCN4 currents. Pflugers Arch 446(6):63340.  Decher N, Chen J, Sanguinetti MC. 2004. Voltagedependent gating of hyperpolarization activated, cyclic nucleotidegated pacemaker channels: molecular coupling between the S4S5 and Clinkers. J Biol Chem 279(14):1385965.  Dehal P, Satou Y, Campbell RK, Chapman J, Degnan B, De Tomaso A, Davidson B, Di Gregorio A, Gelpke M, Goodstein DM and others. 2002. The draft genome of Ciona intestinalis: insights into chordate and vertebrate origins. Science 298(5601):215767.  Dekker JP, Yellen G. 2006. Cooperative gating between single HCN pacemaker channels. J Gen Physiol 128(5):5617.  Delsuc F, Brinkmann H, Chourrout D, Philippe H. 2006. Tunicates and not cephalochordates are the closest living relatives of vertebrates. Nature 439(7079):9658.   157 Delsuc F, Tsagkogeorga G, Lartillot N, Philippe H. 2008. Additional molecular support for the new chordate phylogeny. Genesis 46(11):592604.  Demontis GC, Moroni A, Gravante B, Altomare C, Longoni B, Cervetto L, DiFrancesco D. 2002. Functional characterisation and subcellular localisation of HCN1 channels in rabbit retinal rod photoreceptors. J Physiol 542(Pt 1):8997.  DiFrancesco D. 1981a. A new interpretation of the pacemaker current in calf Purkinje fibres. J Physiol 314:35976.  DiFrancesco D. 1981b. A study of the ionic nature of the pacemaker current in calf Purkinje fibres. J Physiol 314:37793.  DiFrancesco D. 1982. Block and activation of the pacemaker channel in calf purkinje fibres: effects of potassium, caesium and rubidium. J Physiol 329:485507.  DiFrancesco D. 1984. Characterization of the pacemaker current kinetics in calf Purkinje fibres. J Physiol 348:34167.  DiFrancesco D. 1986. Characterization of single pacemaker channels in cardiac sinoatrial node cells. Nature 324(6096):4703.  DiFrancesco D. 1991. The contribution of the 'pacemaker' current (if) to generation of spontaneous activity in rabbit sinoatrial node myocytes. J Physiol 434:2340.  DiFrancesco D. 1993. Pacemaker mechanisms in cardiac tissue. Annu Rev Physiol 55:45572.  DiFrancesco D, Ferroni A, Mazzanti M, Tromba C. 1986. Properties of the hyperpolarizing activated current (if) in cells isolated from the rabbit sinoatrial node. J Physiol 377:6188.  DiFrancesco D, Mangoni M. 1994. Modulation of single hyperpolarizationactivated channels (i(f)) by cAMP in the rabbit sinoatrial node. J Physiol 474(3):47382.  DiFrancesco D, Ojeda C. 1980. Properties of the current if in the sinoatrial node of the rabbit compared with those of the current iK, in Purkinje fibres. J Physiol 308:35367.  DiFrancesco D, Tortora P. 1991. Direct activation of cardiac pacemaker channels by intracellular cyclic AMP. Nature 351(6322):1457.  DiFrancesco D, Tromba C. 1988. Muscarinic control of the hyperpolarizationactivated current (if) in rabbit sinoatrial node myocytes. J Physiol 405:493510.  Diller TC, Madhusudan, Xuong NH, Taylor SS. 2001. Molecular basis for regulatory subunit diversity in cAMPdependent protein kinase: crystal structure of the type II beta regulatory subunit. Structure 9(1):7382.   158 Doyle DA, Morais Cabral J, Pfuetzner RA, Kuo A, Gulbis JM, Cohen SL, Chait BT, MacKinnon R. 1998. The structure of the potassium channel: molecular basis of K+ conduction and selectivity. Science 280(5360):6977.  Edman A, Gestrelius S, Grampp W. 1987. Current activation by membrane hyperpolarization in the slowly adapting lobster stretch receptor neurone. J Physiol 384:67190.  Felsenstein J. 1996. Inferring phylogenies from protein sequences by parsimony, distance, and likelihood methods. Methods Enzymol 266:41827.  FerrerCosta C, Orozco M, de la Cruz X. 2002. Characterization of diseaseassociated single amino acid polymorphisms in terms of sequence and structure properties. J Mol Biol 315(4):771 86.  Fesenko EE, Kolesnikov SS, Lyubarsky AL. 1985. Induction by cyclic GMP of cationic conductance in plasma membrane of retinal rod outer segment. Nature 313(6000):3103.  Flynn GE, Black KD, Islas LD, Sankaran B, Zagotta WN. 2007. Structure and rearrangements in the carboxyterminal region of SpIH channels. Structure 15(6):67182.  Gaita F, Giustetto C, Bianchi F, Wolpert C, Schimpf R, Riccardi R, Grossi S, Richiardi E, Borggrefe M. 2003. Short QT Syndrome: a familial cause of sudden death. Circulation 108(8):96570.  Galindo BE, Neill AT, Vacquier VD. 2005. A new hyperpolarizationactivated, cyclic nucleotidegated channel from sea urchin sperm flagella. Biochem Biophys Res Commun 334(1):96101.  Gallin WJ. 1998. Evolution of the "classical" cadherin family of cell adhesion molecules in vertebrates. Mol Biol Evol 15(9):1099107.  Gauss R, Seifert R, Kaupp UB. 1998. Molecular identification of a hyperpolarizationactivated channel in sea urchin sperm. Nature 393(6685):5837.  Gaymard F, Cerutti M, Horeau C, Lemaillet G, Urbach S, Ravallec M, Devauchelle G, Sentenac H, Thibaud JB. 1996. The baculovirus/insect cell system as an alternative to Xenopus oocytes. First characterization of the AKT1 K+ channel from Arabidopsis thaliana. J Biol Chem 271(37):2286370.  Gee H. 2006. Evolution: careful with that amphioxus. Nature 439(7079):9234.  Giorgetti A, Carloni P, Mistrik P, Torre V. 2005. A homology model of the pore region of HCN channels. Biophys J 89(2):93244.   159 Gisselmann G, Marx T, Bobkov Y, Wetzel CH, Neuhaus EM, Ache BW, Hatt H. 2005. Molecular and functional characterization of an I(h)channel from lobster olfactory receptor neurons. Eur J Neurosci 21(6):163547.  Gisselmann G, Warnstedt M, Gamerschlag B, Bormann A, Marx T, Neuhaus EM, Stoertkuhl K, Wetzel CH, Hatt H. 2003. Characterization of recombinant and native Ihchannels from Apis mellifera. Insect Biochem Mol Biol 33(11):112334.  Goldenberg I, Zareba W, Moss AJ. 2008. Long QT Syndrome. Curr Probl Cardiol 33(11):629 94.  Goulding EH, Ngai J, Kramer RH, Colicos S, Axel R, Siegelbaum SA, Chess A. 1992. Molecular cloning and singlechannel properties of the cyclic nucleotidegated channel from catfish olfactory neurons. Neuron 8(1):4558.  Goulding EH, Tibbs GR, Siegelbaum SA. 1994. Molecular mechanism of cyclicnucleotide gated channel activation. Nature 372(6504):36974.  Grantham R. 1974. Amino acid difference formula to help explain protein evolution. Science 185(4154):8624.  Gravante B, Barbuti A, Milanesi R, Zappi I, Viscomi C, DiFrancesco D. 2004. Interaction of the pacemaker channel HCN1 with filamin A. J Biol Chem 279(42):4384753.  Hagiwara N, Irisawa H. 1989. Modulation by intracellular Ca2+ of the hyperpolarization activated inward current in rabbit single sinoatrial node cells. J Physiol 409:12141.  Harris NC, Constanti A. 1995. Mechanism of block by ZD 7288 of the hyperpolarization activated inward rectifying current in guinea pig substantia nigra neurons in vitro. J Neurophysiol 74(6):236678.  Harzheim D, Pfeiffer KH, Fabritz L, Kremmer E, Buch T, Waisman A, Kirchhof P, Kaupp UB, Seifert R. 2008. Cardiac pacemaker function of HCN4 channels in mice is confined to embryonic development and requires cyclic AMP. Embo J 27(4):692703.  Heginbotham L, Lu Z, Abramson T, MacKinnon R. 1994. Mutations in the K+ channel signature sequence. Biophys J 66(4):10617.  Hegle AP, Nazzari H, Roth A, Angoli D, Accili EA. 2010. Evolutionary emergence of N glycosylation as a variable promoter of HCN channel surface expression. Am J Physiol Cell Physiol 298(5):C106676.  Helenius A, Aebi M. 2001. Intracellular functions of Nlinked glycans. Science 291(5512):2364 9.   160 Helenius A, Aebi M. 2004. Roles of Nlinked glycans in the endoplasmic reticulum. Annu Rev Biochem 73:101949.  Hellbach A, Tiozzo S, Ohn J, Liebling M, De Tomaso AW. 2011. Characterization of HCN and cardiac function in a colonial ascidian. J Exp Zool A Ecol Genet Physiol 315(8):47686.  Herrmann S, Stieber J, Stockl G, Hofmann F, Ludwig A. 2007. HCN4 provides a 'depolarization reserve' and is not required for heart rate acceleration in mice. Embo J 26(21):442332.  Hille B. 2001. Ionic channels of excitable membranes.  Hodgkin AL, Huxley AF. 1952. A quantitative description of membrane current and its application to conduction and excitation in nerve. J Physiol 117(4):50044.  Hodgkin AL, Huxley AF, Katz B. 1952. Measurement of currentvoltage relations in the membrane of the giant axon of Loligo. J Physiol 116(4):42448.  Hodgkin AL, Katz B. 1949. The effect of sodium ions on the electrical activity of giant axon of the squid. J Physiol 108(1):3777.  Hoegg S, Brinkmann H, Taylor JS, Meyer A. 2004. Phylogenetic timing of the fishspecific genome duplication correlates with the diversification of teleost fish. J Mol Evol 59(2):190203.  Hoesl E, Stieber J, Herrmann S, Feil S, Tybl E, Hofmann F, Feil R, Ludwig A. 2008. Tamoxifen inducible gene deletion in the cardiac conduction system. J Mol Cell Cardiol 45(1):629.  Holland LZ, GibsonBrown JJ. 2003. The Ciona intestinalis genome: when the constraints are off. Bioessays 25(6):52932.  Hong K, Bjerregaard P, Gussak I, Brugada R. 2005. Short QT syndrome and atrial fibrillation caused by mutation in KCNH2. J Cardiovasc Electrophysiol 16(4):3946.  Hoshi T. 1995. Regulation of voltage dependence of the KAT1 channel by intracellular factors. J Gen Physiol 105(3):30928.  Huang Z, Walker MC, Shah MM. 2009. Loss of dendritic HCN1 subunits enhances cortical excitability and epileptogenesis. J Neurosci 29(35):1097988.  Hubbard T, Barker D, Birney E, Cameron G, Chen Y, Clark L, Cox T, Cuff J, Curwen V, Down T and others. 2002. The Ensembl genome database project. Nucleic Acids Res 30(1):3841.  Huelsenbeck JP, Dyer KA. 2004. Bayesian estimation of positively selected sites. J Mol Evol 58(6):66172.  Hughes AL, Friedman R. 2003. 2R or not 2R: testing hypotheses of genome duplication in early vertebrates. J Struct Funct Genomics 3(14):8593.  161  Hughes AL, Friedman R. 2005. Loss of ancestral genes in the genomic evolution of Ciona intestinalis. Evol Dev 7(3):196200.  Hughes J, Criscuolo F. 2008. Evolutionary history of the UCP gene family: gene duplication and selection. BMC Evol Biol 8:306.  Ishii TM, Takano M, Ohmori H. 2001. Determinants of activation kinetics in mammalian hyperpolarizationactivated cation channels. J Physiol 537(Pt 1):93100.  Ishii TM, Takano M, Xie LH, Noma A, Ohmori H. 1999. Molecular characterization of the hyperpolarizationactivated cation channel in rabbit heart sinoatrial node. J Biol Chem 274(18):128359.  Jackson HA, Marshall CR, Accili EA. 2007. Evolution and structural diversification of hyperpolarizationactivated cyclic nucleotidegated channel genes. Physiol Genomics 29(3):231 45.  Jegla T, Salkoff L. 1997. A novel subunit for shal K+ channels radically alters activation and inactivation. J Neurosci 17(1):3244.  Jegla TJ, Zmasek CM, Batalov S, Nayak SK. 2009. Evolution of the human ion channel set. Comb Chem High Throughput Screen 12(1):223.  Jespersen T, Grunnet M, Olesen SP. 2005. The KCNQ1 potassium channel: from gene to physiological function. Physiology (Bethesda) 20:40816.  Jiang Y, Lee A, Chen J, Cadene M, Chait BT, MacKinnon R. 2002a. Crystal structure and mechanism of a calciumgated potassium channel. Nature 417(6888):51522.  Jiang Y, Lee A, Chen J, Cadene M, Chait BT, MacKinnon R. 2002b. The open pore conformation of potassium channels. Nature 417(6888):5236.  Jiang Y, Lee A, Chen J, Ruta V, Cadene M, Chait BT, MacKinnon R. 2003a. Xray structure of a voltagedependent K+ channel. Nature 423(6935):3341.  Jiang Y, Ruta V, Chen J, Lee A, MacKinnon R. 2003b. The principle of gating charge movement in a voltagedependent K+ channel. Nature 423(6935):428.  Jung S, Bullis JB, Lau IH, Jones TD, Warner LN, Poolos NP. 2010. Downregulation of dendritic HCN channel gating in epilepsy is mediated by altered phosphorylation signaling. J Neurosci 30(19):667888.  Karolchik D, Kuhn RM, Baertsch R, Barber GP, Clawson H, Diekhans M, Giardine B, Harte RA, Hinrichs AS, Hsu F and others. 2008. The UCSC Genome Browser Database: 2008 update. Nucleic Acids Res 36(Database issue):D7739.  162  Kaupp UB, Niidome T, Tanabe T, Terada S, Bonigk W, Stuhmer W, Cook NJ, Kangawa K, Matsuo H, Hirose T and others. 1989. Primary structure and functional expression from complementary DNA of the rod photoreceptor cyclic GMPgated channel. Nature 342(6251):7626.  Kaupp UB, Seifert R. 2002. Cyclic nucleotidegated ion channels. Physiol Rev 82(3):769824.  Kimura K, Kitano J, Nakajima Y, Nakanishi S. 2004. Hyperpolarizationactivated, cyclic nucleotidegated HCN2 cation channel forms a protein assembly with multiple neuronal scaffold proteins in distinct modes of proteinprotein interaction. Genes Cells 9(7):63140.  Kozak M. 1996. Interpreting cDNA sequences: some insights from studies on translation. Mamm Genome 7(8):56374.  Krawczak M, Ball EV, Cooper DN. 1998. Neighboringnucleotide effects on the rates of germ line singlebasepair substitution in human genes. Am J Hum Genet 63(2):47488.  Krieger J, Strobel J, Vogl A, Hanke W, Breer H. 1999. Identification of a cyclic nucleotide and voltageactivated ion channel from insect antennae. Insect Biochem Mol Biol 29(3):25567.  Krishnan Y, Zheng R, Walsh C, Tang Y, McDonald TV. 2011. Partially Dominant Mutant Channel Defect Corresponding with Intermediate LQT2 Phenotype. Pacing Clin Electrophysiol.  Kusch J, Biskup C, Thon S, Schulz E, Nache V, Zimmer T, Schwede F, Benndorf K. 2010. Interdependence of receptor activation and ligand binding in HCN2 pacemaker channels. Neuron 67(1):7585.  Lacombe B, Pilot G, Michard E, Gaymard F, Sentenac H, Thibaud JB. 2000. A shakerlike K(+) channel with weak rectification is expressed in both source and sink phloem tissues of Arabidopsis. Plant Cell 12(6):83751.  LaishFarkash A, Glikson M, Brass D, MarekYagel D, Pras E, Dascal N, Antzelevitch C, Nof E, Reznik H, Eldar M and others. 2010. A novel mutation in the HCN4 gene causes symptomatic sinus bradycardia in moroccan jews. J Cardiovasc Electrophysiol 21(12):136572.  Latorre R, Olcese R, Basso C, Gonzalez C, Munoz F, Cosmelli D, Alvarez O. 2003. Molecular coupling between voltage sensor and pore opening in the Arabidopsis inward rectifier K+ channel KAT1. J Gen Physiol 122(4):45969.  Lewis AS, Chetkovich DM. 2010. HCN channels in behavior and neurological disease: too hyper or not active enough? Mol Cell Neurosci 46(2):35767.  Li B, Gallin WJ. 2005. Computational identification of residues that modulate voltage sensitivity of voltagegated potassium channels. BMC Struct Biol 5:16.   163 Li CH, Zhang Q, Teng B, Mustafa SJ, Huang JY, Yu HG. 2008. Src tyrosine kinase alters gating of hyperpolarizationactivated HCN4 pacemaker channel through Tyr531. Am J Physiol Cell Physiol 294(1):C35562.  Li WH, Gojobori T. 1983. Rapid evolution of goat and sheep globin genes following gene duplication. Mol Biol Evol 1(1):94108.  Liao Z, Lockhead D, Larson ED, Proenza C. 2010. Phosphorylation and modulation of hyperpolarizationactivated HCN4 channels by protein kinase A in the mouse sinoatrial node. J Gen Physiol 136(3):24758.  Liu Y, Holmgren M, Jurman ME, Yellen G. 1997. Gated access to the pore of a voltage dependent K+ channel. Neuron 19(1):17584.  Long SB, Campbell EB, Mackinnon R. 2005a. Crystal structure of a mammalian voltage dependent Shaker family K+ channel. Science 309(5736):897903.  Long SB, Campbell EB, Mackinnon R. 2005b. Voltage sensor of Kv1.2: structural basis of electromechanical coupling. Science 309(5736):9038.  Lopreato GF, Lu Y, Southwell A, Atkinson NS, Hillis DM, Wilcox TP, Zakon HH. 2001. Evolution and divergence of sodium channel genes in vertebrates. Proc Natl Acad Sci U S A 98(13):758892.  Ludwig A, Budde T, Stieber J, Moosmang S, Wahl C, Holthoff K, Langebartels A, Wotjak C, Munsch T, Zong X and others. 2003. Absence epilepsy and sinus dysrhythmia in mice lacking the pacemaker channel HCN2. Embo J 22(2):21624.  Ludwig A, Herrmann S, Hoesl E, Stieber J. 2008. Mouse models for studying pacemaker channel function and sinus node arrhythmia. Prog Biophys Mol Biol 98(23):17985.  Ludwig A, Zong X, Hofmann F, Biel M. 1999a. Structure and function of cardiac pacemaker channels. Cell Physiol Biochem 9(45):17986.  Ludwig A, Zong X, Jeglitsch M, Hofmann F, Biel M. 1998. A family of hyperpolarization activated mammalian cation channels. Nature 393(6685):58791.  Ludwig A, Zong X, Stieber J, Hullin R, Hofmann F, Biel M. 1999b. Two pacemaker channels from human heart with profoundly different activation kinetics. Embo J 18(9):23239.  Luo L, Chang L, Brown SM, Ao H, Lee DH, Higuera ES, Dubin AE, Chaplan SR. 2007. Role of peripheral hyperpolarizationactivated cyclic nucleotidemodulated channel pacemaker channels in acute and chronic pain models in the rat. Neuroscience 144(4):147785.  Lynch M. 2007. The evolution of genetic networks by nonadaptive processes. Nat Rev Genet 8(10):80313.  164  Lynch M, Conery JS. 2000. The evolutionary fate and consequences of duplicate genes. Science 290(5494):11515.  Lynch M, Katju V. 2004. The altered evolutionary trajectories of gene duplicates. Trends Genet 20(11):5449.  MacKinnon R. 1991. Determination of the subunit stoichiometry of a voltageactivated potassium channel. Nature 350(6315):2325.  Macri V, Accili EA. 2004. Structural elements of instantaneous and slow gating in hyperpolarizationactivated cyclic nucleotidegated channels. J Biol Chem 279(16):1683246.  Macri V, Nazzari H, McDonald E, Accili EA. 2009. Alanine scanning of the S6 segment reveals a unique and cAMPsensitive association between the pore and voltagedependent opening in HCN channels. J Biol Chem 284(23):1565967.  Macri V, Proenza C, Agranovich E, Angoli D, Accili EA. 2002. Separable gating mechanisms in a Mammalian pacemaker channel. J Biol Chem 277(39):3593946.  MankSeymour AR, Richmond JL, Wood LS, Reynolds JM, Fan YT, Warnes GR, Milos PM, Thompson JF. 2006. Association of torsades de pointes with novel and known single nucleotide polymorphisms in long QT syndrome genes. Am Heart J 152(6):111622.  Mannikko R, Elinder F, Larsson HP. 2002. Voltagesensing mechanism is conserved among ion channels gated by opposite voltages. Nature 419(6909):83741.  Marshall C, Elias C, Xue XH, Le HD, Omelchenko A, Hryshko LV, Tibbits GF. 2002. Temperature dependence of cardiac Na+/Ca2+ exchanger. Ann N Y Acad Sci 976:10912.  Marshall CR, Fox JA, Butland SL, Ouellette BF, Brinkman FS, Tibbits GF. 2005. Phylogeny of Na+/Ca2+ exchanger (NCX) genes from genomic data identifies new gene duplications and a new family member in fish species. Physiol Genomics 21(2):16173.  Marshall PW, Rouse W, Briggs I, Hargreaves RB, Mills SD, McLoughlin BJ. 1993. ICI D7288, a novel sinoatrial node modulator. J Cardiovasc Pharmacol 21(6):9026.  Marten I, Hoshi T. 1997. Voltagedependent gating characteristics of the K+ channel KAT1 depend on the N and C termini. Proc Natl Acad Sci U S A 94(7):344853.  Marx T, Gisselmann G, Stortkuhl KF, Hovemann BT, Hatt H. 1999. Molecular cloning of a putative voltage and cyclic nucleotidegated ion channel present in the antennae and eyes of Drosophila melanogaster. Invert Neurosci 4(1):5563.  McCormick DA, Pape HC. 1990. Properties of a hyperpolarizationactivated cation current and its role in rhythmic oscillation in thalamic relay neurones. J Physiol 431:291318.  165  Michels G, Brandt MC, Zagidullin N, Khan IF, Larbig R, van Aaken S, Wippermann J, Hoppe UC. 2008. Direct evidence for calcium conductance of hyperpolarizationactivated cyclic nucleotidegated channels and human native If at physiological calcium concentrations. Cardiovasc Res 78(3):46675.  Milanesi R, Baruscotti M, GnecchiRuscone T, DiFrancesco D. 2006. Familial sinus bradycardia associated with a mutation in the cardiac pacemaker channel. N Engl J Med 354(2):1517.  Millat G, Chevalier P, RestierMiron L, Da Costa A, Bouvagnet P, Kugener B, Fayol L, Gonzalez Armengod C, Oddou B, Chanavat V and others. 2006. Spectrum of pathogenic mutations and associated polymorphisms in a cohort of 44 unrelated patients with long QT syndrome. Clin Genet 70(3):21427.  Miller C. 2000. An overview of the potassium channel family. Genome Biol 1(4):REVIEWS0004.  Miller MP, Kumar S. 2001. Understanding human disease mutations through the use of interspecific genetic variation. Hum Mol Genet 10(21):231928.  Miller MP, Parker JD, Rissing SW, Kumar S. 2003. Quantifying the intragenic distribution of human disease mutations. Ann Hum Genet 67(Pt 6):56779.  Mistrik P, Mader R, Michalakis S, Weidinger M, Pfeifer A, Biel M. 2005. The murine HCN3 gene encodes a hyperpolarizationactivated cation channel with slow kinetics and unique response to cyclic nucleotides. J Biol Chem 280(29):2705661.  Mistrik P, Pfeifer A, Biel M. 2006. The enhancement of HCN channel instantaneous current facilitated by slow deactivation is regulated by intracellular chloride concentration. Pflugers Arch 452(6):71827.  Mistrik P, Torre V. 2004. Histidine 518 in the S6CNBD linker controls pH dependence and gating of HCN channel from seaurchin sperm. Pflugers Arch 448(1):7684.  Momin A, Cadiou H, Mason A, McNaughton PA. 2008. Role of the hyperpolarizationactivated current Ih in somatosensory neurons. J Physiol 586(Pt 24):591129.  Mooney SD, Klein TE. 2002. The functional importance of diseaseassociated mutation. BMC Bioinformatics 3:24.  Moosmang S, Stieber J, Zong X, Biel M, Hofmann F, Ludwig A. 2001. Cellular expression and functional characterization of four hyperpolarizationactivated pacemaker channels in cardiac and neuronal tissues. Eur J Biochem 268(6):164652.   166 Moroni A, Barbuti A, Altomare C, Viscomi C, Morgan J, Baruscotti M, DiFrancesco D. 2000. Kinetic and ionic properties of the human HCN2 pacemaker channel. Pflugers Arch 439(5):618 26.  Moss AJ, Zareba W, Kaufman ES, Gartman E, Peterson DR, Benhorin J, Towbin JA, Keating MT, Priori SG, Schwartz PJ and others. 2002. Increased risk of arrhythmic events in longQT syndrome with mutations in the pore region of the human etheragogorelated gene potassium channel. Circulation 105(7):7949.  Moya A, Sutton R, Ammirati F, Blanc JJ, Brignole M, Dahm JB, Deharo JC, Gajek J, Gjesdal K, Krahn A and others. 2009. Guidelines for the diagnosis and management of syncope (version 2009). Eur Heart J 30(21):263171.  Much B, WahlSchott C, Zong X, Schneider A, Baumann L, Moosmang S, Ludwig A, Biel M. 2003. Role of subunit heteromerization and Nlinked glycosylation in the formation of functional hyperpolarizationactivated cyclic nucleotidegated channels. J Biol Chem 278(44):437816.  Murata Y, Iwasaki H, Sasaki M, Inaba K, Okamura Y. 2005. Phosphoinositide phosphatase activity coupled to an intrinsic voltage sensor. Nature 435(7046):123943.  Nakamura T, Gold GH. 1987. A cyclic nucleotidegated conductance in olfactory receptor cilia. Nature 325(6103):4424.  Nazzari H, Angoli D, Chow SS, Whitaker G, Leclair L, McDonald E, Macri V, Zahynacz K, Walker V, Accili EA. 2008. Regulation of cell surface expression of functional pacemaker channels by a motif in the Bhelix of the cyclic nucleotidebinding domain. Am J Physiol Cell Physiol 295(3):C64252.  Neher E, Sakmann B. 1976. Singlechannel currents recorded from membrane of denervated frog muscle fibres. Nature 260(5554):799802.  Nei M. 2005. Selectionism and neutralism in molecular evolution. Mol Biol Evol 22(12):2318 42.  NewtonCheh C, Guo CY, Larson MG, Musone SL, Surti A, Camargo AL, Drake JA, Benjamin EJ, Levy D, D'Agostino RB, Sr. and others. 2007. Common genetic variation in KCNH2 is associated with QT interval duration: the Framingham Heart Study. Circulation 116(10):1128 36.  Nicholas KB, Jr. NHB, Deerfield DWI. 1997. GeneDoc: Analysis and Visualization of Genetic Variation. EMBNEW NEWS 4:14.  Nimigean CM, Shane T, Miller C. 2004. A cyclic nucleotide modulated prokaryotic K+ channel. J Gen Physiol 124(3):20310.   167 Noam Y, Bernard C, Baram TZ. 2011. Towards an integrated view of HCN channel role in epilepsy. Curr Opin Neurobiol.  Nof E, Antzelevitch C, Glikson M. 2010. The Contribution of HCN4 to normal sinus node function in humans and animal models. Pacing Clin Electrophysiol 33(1):1006.  Nof E, Luria D, Brass D, Marek D, Lahat H, ReznikWolf H, Pras E, Dascal N, Eldar M, Glikson M. 2007. Point mutation in the HCN4 cardiac ion channel pore affecting synthesis, trafficking, and functional expression is associated with familial asymptomatic sinus bradycardia. Circulation 116(5):46370.  Nolan MF, Malleret G, Dudman JT, Buhl DL, Santoro B, Gibbs E, Vronskaya S, Buzsaki G, Siegelbaum SA, Kandel ER and others. 2004. A behavioral role for dendritic integration: HCN1 channels constrain spatial memory and plasticity at inputs to distal dendrites of CA1 pyramidal neurons. Cell 119(5):71932.  Nolan MF, Malleret G, Lee KH, Gibbs E, Dudman JT, Santoro B, Yin D, Thompson RF, Siegelbaum SA, Kandel ER and others. 2003. The hyperpolarizationactivated HCN1 channel is important for motor learning and neuronal integration by cerebellar Purkinje cells. Cell 115(5):55164.  Novak AE, Jost MC, Lu Y, Taylor AD, Zakon HH, Ribera AB. 2006. Gene duplications and evolution of vertebrate voltagegated sodium channels. J Mol Evol 63(2):20821.  Oberai A, Joh NH, Pettit FK, Bowie JU. 2009. Structural imperatives impose diverse evolutionary constraints on helical membrane proteins. Proc Natl Acad Sci U S A 106(42):1774750.  Ohno S. 1970. Evolution by Gene Duplication.  Okamura Y, Nishino A, Murata Y, Nakajo K, Iwasaki H, Ohtsuka Y, TanakaKunishima M, Takahashi N, Hara Y, Yoshida T and others. 2005. Comprehensive analysis of the ascidian genome reveals novel insights into the molecular evolution of ion channel genes. Physiol Genomics 22(3):26982.  On C, Marshall CR, Chen N, Moyes CD, Tibbits GF. 2008. Gene structure evolution of the Na+ Ca2+ exchanger (NCX) family. BMC Evol Biol 8:127.  Ortlund EA, Bridgham JT, Redinbo MR, Thornton JW. 2007. Crystal structure of an ancient protein: evolution by conformational epistasis. Science 317(5844):15448.  Ouyang Q, Goeritz M, HarrisWarrick RM. 2007. Panulirus interruptus Ihchannel gene PIIH: modification of channel properties by alternative splicing and role in rhythmic activity. J Neurophysiol 97(6):388092.   168 Page RD. 1996. TreeView: an application to display phylogenetic trees on personal computers. Comput Appl Biosci 12(4):3578.  Papazian DM, Schwarz TL, Tempel BL, Jan YN, Jan LY. 1987. Cloning of genomic and complementary DNA from Shaker, a putative potassium channel gene from Drosophila. Science 237(4816):74953.  Pape HC. 1996. Queer current and pacemaker: the hyperpolarizationactivated cation current in neurons. Annu Rev Physiol 58:299327.  Papp I, Hollo K, Antal M. 2010. Plasticity of hyperpolarizationactivated and cyclic nucleotid gated cation channel subunit 2 expression in the spinal dorsal horn in inflammatory pain. Eur J Neurosci 32(7):1193201.  Passner JM, Schultz SC, Steitz TA. 2000. Modeling the cAMPinduced allosteric transition using the crystal structure of CAPcAMP at 2.1 A resolution. J Mol Biol 304(5):84759.  Peters CJ, Chow SS, Angoli D, Nazzari H, Cayabyab FS, Morshedian A, Accili EA. 2009. In situ codistribution and functional interactions of SAP97 with sinoatrial isoforms of HCN channels. J Mol Cell Cardiol 46(5):63643.  Philippe H, Brinkmann H, Lavrov DV, Littlewood DT, Manuel M, Worheide G, Baurain D. 2011. Resolving difficult phylogenetic questions: why more sequences are not enough. PLoS Biol 9(3):e1000602.  Pian P, Bucchi A, Decostanzo A, Robinson RB, Siegelbaum SA. 2007. Modulation of cyclic nucleotideregulated HCN channels by PIP(2) and receptors coupled to phospholipase C. Pflugers Arch 455(1):12545.  Pian P, Bucchi A, Robinson RB, Siegelbaum SA. 2006. Regulation of gating and rundown of HCN hyperpolarizationactivated channels by exogenous and endogenous PIP2. J Gen Physiol 128(5):593604.  Pongs O, Kecskemethy N, Muller R, KrahJentgens I, Baumann A, Kiltz HH, Canal I, Llamazares S, Ferrus A. 1988. Shaker encodes a family of putative potassium channel proteins in the nervous system of Drosophila. Embo J 7(4):108796.  Prasad UK, Gray D, Purcell H. 2009. Review of the If selective channel inhibitor ivabradine in the treatment of chronic stable angina. Adv Ther 26(2):12737.  Priori SG, Schwartz PJ, Napolitano C, Bloise R, Ronchetti E, Grillo M, Vicentini A, Spazzolini C, Nastoli J, Bottelli G and others. 2003. Risk stratification in the longQT syndrome. N Engl J Med 348(19):186674.  Proenza C, Angoli D, Agranovich E, Macri V, Accili EA. 2002a. Pacemaker channels produce an instantaneous current. J Biol Chem 277(7):51019.  169  Proenza C, Tran N, Angoli D, Zahynacz K, Balcar P, Accili EA. 2002b. Different roles for the cyclic nucleotide binding domain and amino terminus in assembly and expression of hyperpolarizationactivated, cyclic nucleotidegated channels. J Biol Chem 277(33):2963442.  Proenza C, Yellen G. 2006. Distinct populations of HCN pacemaker channels produce voltage dependent and voltageindependent currents. J Gen Physiol 127(2):18390.  Prole DL, Yellen G. 2006. Reversal of HCN channel voltage dependence via bridging of the S4 S5 linker and PostS6. J Gen Physiol 128(3):27382.  Putnam NH, Butts T, Ferrier DE, Furlong RF, Hellsten U, Kawashima T, RobinsonRechavi M, Shoguchi E, Terry A, Yu JK and others. 2008. The amphioxus genome and the evolution of the chordate karyotype. Nature 453(7198):106471.  Qu J, Kryukova Y, Potapova IA, Doronin SV, Larsen M, Krishnamurthy G, Cohen IS, Robinson RB. 2004. MiRP1 modulates HCN2 channel expression and gating in cardiac myocytes. J Biol Chem 279(42):43497502.  Ramsey IS, Mokrab Y, Carvacho I, Sands ZA, Sansom MS, Clapham DE. 2010. An aqueous H+ permeation pathway in the voltagegated proton channel Hv1. Nat Struct Mol Biol 17(7):86975.  Ramsey IS, Moran MM, Chong JA, Clapham DE. 2006. A voltagegated protonselective channel lacking the pore domain. Nature 440(7088):12136.  Reid CA, Phillips AM, Petrou S. 2011. HCN channelopathies: Pathophysiology in genetic epilepsy and therapeutic implications. Br J Pharmacol.  Robinson RB, Siegelbaum SA. 2003. Hyperpolarizationactivated cation currents: from molecules to physiological function. Annu Rev Physiol 65:45380.  Roden DM. 2008. Clinical practice. LongQT syndrome. N Engl J Med 358(2):16976.  Rogozin IB, Sverdlov AV, Babenko VN, Koonin EV. 2005. Analysis of evolution of exonintron structure of eukaryotic genes. Brief Bioinform 6(2):11834.  Rothberg BS, Shin KS, Phale PS, Yellen G. 2002. Voltagecontrolled gating at the intracellular entrance to a hyperpolarizationactivated cation channel. J Gen Physiol 119(1):8391.  Rothberg BS, Shin KS, Yellen G. 2003. Movements near the gate of a hyperpolarization activated cation channel. J Gen Physiol 122(5):50110.  Sanguinetti MC, Curran ME, Zou A, Shen J, Spector PS, Atkinson DL, Keating MT. 1996. Coassembly of K(V)LQT1 and minK (IsK) proteins to form cardiac I(Ks) potassium channel. Nature 384(6604):803.   170 Sanguinetti MC, Jiang C, Curran ME, Keating MT. 1995. A mechanistic link between an inherited and an acquired cardiac arrhythmia: HERG encodes the IKr potassium channel. Cell 81(2):299307.  Sanguinetti MC, TristaniFirouzi M. 2006. hERG potassium channels and cardiac arrhythmia. Nature 440(7083):4639.  Sanguinetti MC, Xu QP. 1999. Mutations of the S4S5 linker alter activation properties of HERG potassium channels expressed in Xenopus oocytes. J Physiol 514 (Pt 3):66775.  Santoro B, Chen S, Luthi A, Pavlidis P, Shumyatsky GP, Tibbs GR, Siegelbaum SA. 2000. Molecular and functional heterogeneity of hyperpolarizationactivated pacemaker channels in the mouse CNS. J Neurosci 20(14):526475.  Santoro B, Grant SG, Bartsch D, Kandel ER. 1997. Interactive cloning with the SH3 domain of Nsrc identifies a new brain specific ion channel protein, with homology to eag and cyclic nucleotidegated channels. Proc Natl Acad Sci U S A 94(26):1481520.  Santoro B, Lee JY, Englot DJ, Gildersleeve S, Piskorowski RA, Siegelbaum SA, Winawer MR, Blumenfeld H. 2010. Increased seizure severity and seizurerelated death in mice lacking HCN1 channels. Epilepsia 51(8):16247.  Santoro B, Liu DT, Yao H, Bartsch D, Kandel ER, Siegelbaum SA, Tibbs GR. 1998. Identification of a gene encoding a hyperpolarizationactivated pacemaker channel of brain. Cell 93(5):71729.  Santoro B, Wainger BJ, Siegelbaum SA. 2004. Regulation of HCN channel surface expression by a novel Cterminal proteinprotein interaction. J Neurosci 24(47):1075062.  Sasaki M, Takagi M, Okamura Y. 2006. A voltage sensordomain protein is a voltagegated proton channel. Science 312(5773):58992.  Satoh N, Satou Y, Davidson B, Levine M. 2003. Ciona intestinalis: an emerging model for wholegenome analyses. Trends Genet 19(7):37681.  Satoh TO, Yamada M. 2000. A bradycardiac agent ZD7288 blocks the hyperpolarization activated current (I(h)) in retinal rod photoreceptors. Neuropharmacology 39(7):128491.  Satou Y, Kawashima T, Kohara Y, Satoh N. 2003. Large scale EST analyses in Ciona intestinalis: its application as Northern blot analyses. Dev Genes Evol 213(56):3148.  Schachtman DP, Schroeder JI, Lucas WJ, Anderson JA, Gaber RF. 1992. Expression of an inwardrectifying potassium channel by the Arabidopsis KAT1 cDNA. Science 258(5088):1654 8.   171 Schubert M, Escriva H, XavierNeto J, Laudet V. 2006. Amphioxus and tunicates as evolutionary model systems. Trends Ecol Evol 21(5):26977.  SchulzeBahr E, Neu A, Friederich P, Kaupp UB, Breithardt G, Pongs O, Isbrandt D. 2003. Pacemaker channel dysfunction in a patient with sinus node disease. J Clin Invest 111(10):1537 45.  Schunke S, Stoldt M, Lecher J, Kaupp UB, Willbold D. 2011. Structural insights into conformational changes of a cyclic nucleotidebinding domain in solution from Mesorhizobium loti K1 channel. Proc Natl Acad Sci U S A 108(15):61216.  Schwartz PJ. 2005. Management of long QT syndrome. Nat Clin Pract Cardiovasc Med 2(7):34651.  Schwartz PJ, Priori SG, Bloise R, Napolitano C, Ronchetti E, Piccinini A, Goj C, Breithardt G, SchulzeBahr E, Wedekind H and others. 2001. Molecular diagnosis in a child with sudden infant death syndrome. Lancet 358(9290):13423.  Schwartz PJ, StrambaBadiale M, Crotti L, Pedrazzini M, Besana A, Bosi G, Gabbarini F, Goulene K, Insolia R, Mannarino S and others. 2009. Prevalence of the congenital longQT syndrome. Circulation 120(18):17617.  Schweizer PA, Yampolsky P, Malik R, Thomas D, Zehelein J, Katus HA, Koenen M. 2009. Transcription profiling of HCNchannel isotypes throughout mouse cardiac development. Basic Res Cardiol.  Seifert R, Scholten A, Gauss R, Mincheva A, Lichter P, Kaupp UB. 1999. Molecular characterization of a slowly gating human hyperpolarizationactivated channel predominantly expressed in thalamus, heart, and testis. Proc Natl Acad Sci U S A 96(16):93916.  Sentenac H, Bonneaud N, Minet M, Lacroute F, Salmon JM, Gaymard F, Grignon C. 1992. Cloning and expression in yeast of a plant potassium ion transport system. Science 256(5057):6635.  Shealy RT, Murphy AD, Ramarathnam R, Jakobsson E, Subramaniam S. 2003. Sequence function analysis of the K+selective family of ion channels using a comprehensive alignment and the KcsA channel structure. Biophys J 84(5):292942.  Shi W, Wymore R, Yu H, Wu J, Wymore RT, Pan Z, Robinson RB, Dixon JE, McKinnon D, Cohen IS. 1999. Distribution and prevalence of hyperpolarizationactivated cation channel (HCN) mRNA expression in cardiac tissues. Circ Res 85(1):e16.  Shin KS, Maertens C, Proenza C, Rothberg BS, Yellen G. 2004. Inactivation in HCN channels results from reclosure of the activation gate: desensitization to voltage. Neuron 41(5):73744.   172 Shin KS, Rothberg BS, Yellen G. 2001. Blocker state dependence and trapping in hyperpolarizationactivated cation channels: evidence for an intracellular activation gate. J Gen Physiol 117(2):91101.  Sidow A. 1996. Gen(om)e duplications in the evolution of early vertebrates. Curr Opin Genet Dev 6(6):71522.  Splawski I, Shen J, Timothy KW, Lehmann MH, Priori S, Robinson JL, Moss AJ, Schwartz PJ, Towbin JA, Vincent GM and others. 2000. Spectrum of mutations in longQT syndrome genes. KVLQT1, HERG, SCN5A, KCNE1, and KCNE2. Circulation 102(10):117885.  Stevens DR, Seifert R, Bufe B, Muller F, Kremmer E, Gauss R, Meyerhof W, Kaupp UB, Lindemann B. 2001. Hyperpolarizationactivated channels HCN1 and HCN4 mediate responses to sour stimuli. Nature 413(6856):6315.  Stieber J, Herrmann S, Feil S, Loster J, Feil R, Biel M, Hofmann F, Ludwig A. 2003a. The hyperpolarizationactivated channel HCN4 is required for the generation of pacemaker action potentials in the embryonic heart. Proc Natl Acad Sci U S A 100(25):1523540.  Stieber J, Stockl G, Herrmann S, Hassfurth B, Hofmann F. 2005. Functional expression of the human HCN3 channel. J Biol Chem 280(41):3463543.  Stieber J, Thomer A, Much B, Schneider A, Biel M, Hofmann F. 2003b. Molecular basis for the different activation kinetics of the pacemaker channels HCN2 and HCN4. J Biol Chem 278(36):3367280.  Subramanian S, Kumar S. 2006. Evolutionary anatomies of positions and types of disease associated and neutral amino acid mutations in the human genome. BMC Genomics 7:306.  Suzuki MM, Nishikawa T, Bird A. 2005. Genomic approaches reveal unexpected genetic divergence within Ciona intestinalis. J Mol Evol 61(5):62735.  Suzuki Y, Gojobori T. 1999. A method for detecting positive selection at single amino acid sites. Mol Biol Evol 16(10):131528.  Suzuki Y, Gojobori T, Kumar S. 2009. Methods for incorporating the hypermutability of CpG dinucleotides in detecting natural selection operating at the amino acid sequence level. Mol Biol Evol.  Swalla BJ, Cameron CB, Corley LS, Garey JR. 2000. Urochordates are monophyletic within the deuterostomes. Syst Biol 49(1):5264.  Takasu K, Ono H, Tanabe M. 2010. Spinal hyperpolarizationactivated cyclic nucleotidegated cation channels at primary afferent terminals contribute to chronic pain. Pain 151(1):8796.   173 Takigawa T, Alzheimer C, Quasthoff S, Grafe P. 1998. A special blocker reveals the presence and function of the hyperpolarizationactivated cation current IH in peripheral mammalian nerve fibres. Neuroscience 82(3):6314.  Tan HL, Bardai A, Shimizu W, Moss AJ, SchulzeBahr E, Noda T, Wilde AA. 2006. Genotype specific onset of arrhythmias in congenital longQT syndrome: possible therapy implications. Circulation 114(20):2096103.  Tang B, Sander T, Craven KB, Hempelmann A, Escayg A. 2008. Mutation analysis of the hyperpolarizationactivated cyclic nucleotidegated channels HCN1 and HCN2 in idiopathic generalized epilepsy. Neurobiol Dis 29(1):5970.  Taylor JS, Raes J. 2004. Duplication and divergence: the evolution of new genes and old ideas. Annu Rev Genet 38:61543.  Tester DJ, Ackerman MJ. 2007. Postmortem long QT syndrome genetic testing for sudden unexplained death in the young. J Am Coll Cardiol 49(2):2406.  Thomas PD, Kejariwal A. 2004. Coding singlenucleotide polymorphisms associated with complex vs. Mendelian disease: evolutionary evidence for differences in molecular effects. Proc Natl Acad Sci U S A 101(43):15398403.  Thompson JD, Gibson TJ, Plewniak F, Jeanmougin F, Higgins DG. 1997. The CLUSTAL_X windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res 25(24):487682.  Thornton JW. 2004. Resurrecting ancient genes: experimental analysis of extinct molecules. Nat Rev Genet 5(5):36675.  Tombola F, Pathak MM, Isacoff EY. 2006. How does voltage open an ion channel? Annu Rev Cell Dev Biol 22:2352.  Tran N, Proenza C, Macri V, Petigara F, Sloan E, Samler S, Accili EA. 2002. A conserved domain in the NH2 terminus important for assembly and functional expression of pacemaker channels. J Biol Chem 277(46):4358892.  TristaniFirouzi M, Chen J, Sanguinetti MC. 2002. Interactions between S4S5 linker and S6 transmembrane domain modulate gating of HERG K+ channels. J Biol Chem 277(21):18994 9000.  Trudeau MC, Warmke JW, Ganetzky B, Robertson GA. 1995. HERG, a human inward rectifier in the voltagegated potassium channel family. Science 269(5220):925.  Tu E, Waterhouse L, Duflou J, Bagnall RD, Semsarian C. 2011. Genetic analysis of hyperpolarizationactivated cyclic nucleotidegated cation channels in sudden unexpected death in epilepsy cases. Brain Pathol 21(6):6928.  174  Ueda K, Hirano Y, Higashiuesato Y, Aizawa Y, Hayashi T, Inagaki N, Tana T, Ohya Y, Takishita S, Muratani H and others. 2009. Role of HCN4 channel in preventing ventricular arrhythmia. J Hum Genet 54(2):11521.  Ueda K, Nakamura K, Hayashi T, Inagaki N, Takahashi M, Arimura T, Morita H, Higashiuesato Y, Hirano Y, Yasunami M and others. 2004. Functional characterization of a trafficking defective HCN4 mutation, D553N, associated with cardiac arrhythmia. J Biol Chem 279(26):271948.  Ulens C, Siegelbaum SA. 2003. Regulation of hyperpolarizationactivated HCN channels by cAMP through a gating switch in binding domain symmetry. Neuron 40(5):95970.  Ulens C, Tytgat J. 2001a. Functional heteromerization of HCN1 and HCN2 pacemaker channels. J Biol Chem 276(9):606972.  Ulens C, Tytgat J. 2001b. Gi and Gscoupled receptors upregulate the cAMP cascade to modulate HCN2, but not HCN1 pacemaker channels. Pflugers Arch 442(6):92842.  Uozumi N, Nakamura T, Schroeder JI, Muto S. 1998. Determination of transmembrane topology of an inwardrectifying potassium channel from Arabidopsis thaliana based on functional expression in Escherichia coli. Proc Natl Acad Sci U S A 95(17):97738.  Vaccari T, Moroni A, Rocchi M, Gorza L, Bianchi ME, Beltrame M, DiFrancesco D. 1999. The human gene coding for HCN2, a pacemaker channel of the heart. Biochim Biophys Acta 1446(3):41925.  Vanin EF. 1985. Processed pseudogenes: characteristics and evolution. Annu Rev Genet 19:253 72.  Vemana S, Pandey S, Larsson HP. 2004. S4 movement in a mammalian HCN channel. J Gen Physiol 123(1):2132.  Viscomi C, Altomare C, Bucchi A, Camatini E, Baruscotti M, Moroni A, DiFrancesco D. 2001. C terminusmediated control of voltage and cAMP gating of hyperpolarizationactivated cyclic nucleotidegated channels. J Biol Chem 276(32):299304.  Vitkup D, Sander C, Church GM. 2003. The aminoacid mutational spectrum of human genetic disease. Genome Biol 4(11):R72.  WahlSchott C, Baumann L, Zong X, Biel M. 2005. An arginine residue in the pore region is a key determinant of chloride dependence in cardiac pacemaker channels. J Biol Chem 280(14):13694700.  WahlSchott C, Biel M. 2009. HCN channels: structure, cellular regulation and physiological function. Cell Mol Life Sci 66(3):47094.  175  Wainger BJ, DeGennaro M, Santoro B, Siegelbaum SA, Tibbs GR. 2001. Molecular mechanism of cAMP modulation of HCN pacemaker channels. Nature 411(6839):80510.  Wang DW, Desai RR, Crotti L, Arnestad M, Insolia R, Pedrazzini M, Ferrandi C, Vege A, Rognum T, Schwartz PJ and others. 2007. Cardiac sodium channel dysfunction in sudden infant death syndrome. Circulation 115(3):36876.  Wang J, Chen S, Siegelbaum SA. 2001. Regulation of hyperpolarizationactivated HCN channel gating and cAMP modulation due to interactions of COOH terminus and core transmembrane regions. J Gen Physiol 118(3):23750.  Wang Q, Curran ME, Splawski I, Burn TC, Millholland JM, VanRaay TJ, Shen J, Timothy KW, Vincent GM, de Jager T and others. 1996. Positional cloning of a novel potassium channel gene: KVLQT1 mutations cause cardiac arrhythmias. Nat Genet 12(1):1723.  Warren KS, Baker K, Fishman MC. 2001. The slow mo mutation reduces pacemaker current and heart rate in adult zebrafish. Am J Physiol Heart Circ Physiol 281(4):H17119.  Weber W. 1999. Ion currents of Xenopus laevis oocytes: state of the art. Biochim Biophys Acta 1421(2):21333.  WeiQing H, QingNuan K, Lin X, ChengHao G, QiYi Z. 2011. Expression of hyperpolarizationactivated cyclic nucleotidegated cation channel (HCN4) is increased in hypertrophic cardiomyopathy. Cardiovasc Pathol.  Whitaker GM, Angoli D, Nazzari H, Shigemoto R, Accili EA. 2007. HCN2 and HCN4 isoforms selfassemble and coassemble with equal preference to form functional pacemaker channels. J Biol Chem 282(31):229009.  Widmark J, Sundstrom G, Ocampo Daza D, Larhammar D. 2011. Differential evolution of voltagegated sodium channels in tetrapods and teleost fishes. Mol Biol Evol 28(1):85971.  Wollmuth LP, Hille B. 1992. Ionic selectivity of Ih channels of rod photoreceptors in tiger salamanders. J Gen Physiol 100(5):74965.  Wu JY, Cohen IS. 1997. Tyrosine kinase inhibition reduces i(f) in rabbit sinoatrial node myocytes. Pflugers Arch 434(5):50914.  Xiao J, Nguyen TV, Ngui K, Strijbos PJ, Selmer IS, Neylon CB, Furness JB. 2004. Molecular and functional analysis of hyperpolarisationactivated nucleotidegated (HCN) channels in the enteric nervous system. Neuroscience 129(3):60314.  Xu X, Vysotskaya ZV, Liu Q, Zhou L. 2010. Structural basis for the cAMPdependent gating in the human HCN4 channel. J Biol Chem 285(47):3708291.   176 Yanagihara K, Irisawa H. 1980. Inward current activated during hyperpolarization in the rabbit sinoatrial node cell. Pflugers Arch 385(1):119.  Yang Z. 1997. PAML: a program package for phylogenetic analysis by maximum likelihood. Comput Appl Biosci 13(5):5556.  Yang Z, Kumar S, Nei M. 1995. A new method of inference of ancestral nucleotide and amino acid sequences. Genetics 141(4):164150.  Yang Z, Nielsen R. 2002. Codonsubstitution models for detecting molecular adaptation at individual sites along specific lineages. Mol Biol Evol 19(6):90817.  Yeh YH, Burstein B, Qi XY, Sakabe M, Chartier D, Comtois P, Wang Z, Kuo CT, Nattel S. 2009. Funny current downregulation and sinus node dysfunction associated with atrial tachyarrhythmia: a molecular basis for tachycardiabradycardia syndrome. Circulation 119(12):157685.  Yellen G. 2002. The voltagegated potassium channels and their relatives. Nature 419(6902):35 42.  Yi SV. 2006. Nonadaptive evolution of genome complexity. Bioessays 28(10):97982.  Yifrach O, MacKinnon R. 2002. Energetics of pore opening in a voltagegated K(+) channel. Cell 111(2):2319.  Yu FH, Catterall WA. 2004. The VGLchanome: a protein superfamily specialized for electrical signaling and ionic homeostasis. Sci STKE 2004(253):re15.  Yu H, Wu J, Potapova I, Wymore RT, Holmes B, Zuckerman J, Pan Z, Wang H, Shi W, Robinson RB and others. 2001. MinKrelated peptide 1: A beta subunit for the HCN ion channel subunit family enhances expression and speeds activation. Circ Res 88(12):E847.  Yu HG, Lu Z, Pan Z, Cohen IS. 2004a. Tyrosine kinase inhibition differentially regulates heterologously expressed HCN channels. Pflugers Arch 447(4):392400.  Yu X, Chen XW, Zhou P, Yao L, Liu T, Zhang B, Li Y, Zheng H, Zheng LH, Zhang CX and others. 2007. Calcium influx through If channels in rat ventricular myocytes. Am J Physiol Cell Physiol 292(3):C114755.  Yu X, Duan KL, Shang CF, Yu HG, Zhou Z. 2004b. Calcium influx through hyperpolarization activated cation channels (I(h) channels) contributes to activityevoked neuronal secretion. Proc Natl Acad Sci U S A 101(4):10516.  Zagotta WN, Olivier NB, Black KD, Young EC, Olson R, Gouaux E. 2003. Structural basis for modulation and agonist specificity of HCN pacemaker channels. Nature 425(6954):2005.   177 Zagotta WN, Siegelbaum SA. 1996. Structure and function of cyclic nucleotidegated channels. Annu Rev Neurosci 19:23563.  Zakon HH, Jost MC, Lu Y. 2011. Expansion of voltagedependent Na+ channel gene family in early tetrapods coincided with the emergence of terrestriality and increased brain complexity. Mol Biol Evol 28(4):141524.  Zei PC, Aldrich RW. 1998. Voltagedependent gating of single wildtype and S4 mutant KAT1 inward rectifier potassium channels. J Gen Physiol 112(6):679713.  Zha Q, Brewster AL, Richichi C, Bender RA, Baram TZ. 2008. Activitydependent heteromerization of the hyperpolarizationactivated, cyclicnucleotide gated (HCN) channels: role of Nlinked glycosylation. J Neurochem 105(1):6877.  Zhang X, Bursulaya B, Lee CC, Chen B, Pivaroff K, Jegla T. 2009a. Divalent cations slow activation of EAG family K+ channels through direct binding to S4. Biophys J 97(1):11020.  Zhang Y, Liu Y, Qu J, Hardy A, Zhang N, Diao J, Strijbos PJ, Tsushima R, Robinson RB, Gaisano HY and others. 2009b. Functional characterization of hyperpolarizationactivated cyclic nucleotidegated channels in rat pancreatic beta cells. J Endocrinol 203(1):4553.  Zhou L, Siegelbaum SA. 2007. Gating of HCN channels by cyclic nucleotides: residue contacts that underlie ligand binding, selectivity, and efficacy. Structure 15(6):65570.  Zicha S, FernandezVelasco M, Lonardo G, L'Heureux N, Nattel S. 2005. Sinus node dysfunction and hyperpolarizationactivated (HCN) channel subunit remodeling in a canine heart failure model. Cardiovasc Res 66(3):47281.  Zolles G, Klocker N, Wenzel D, WeisserThomas J, Fleischmann BK, Roeper J, Fakler B. 2006. Pacemaking by HCN channels requires interaction with phosphoinositides. Neuron 52(6):1027 36.  Zong X, Eckert C, Yuan H, WahlSchott C, Abicht H, Fang L, Li R, Mistrik P, Gerstner A, Much B and others. 2005. A novel mechanism of modulation of hyperpolarizationactivated cyclic nucleotidegated channels by Src kinase. J Biol Chem 280(40):3422432.  Zong X, Stieber J, Ludwig A, Hofmann F, Biel M. 2001. A single histidine residue determines the pH sensitivity of the pacemaker channel HCN2. J Biol Chem 276(9):63139.    178  APPENDIX Figure A.1 : NeighborJoining (NJ) Phylogram of the HCN Family. The NJ tree was constructed with resampling (bootstrap, 1000 datasets) using ClustalX (excluding positions with gaps and corrected for multiple substitutions). Only the region between the start of S1 and the end of the CNBD was used, to control for length differences between the different isoforms. The tree is rooted with an outgroup of KAT1, hERG1, CNGA1 and CNGA3, four sequences that are known to be distantly related to the HCN family. Tree topology indicates that HCN3 was the first product of duplication events in the vertebrate lineage. There is a split between mammalian and fish HCN3 clades, which may be expected based on the sequence conservation shown in Figure 2, but the low bootstrap value indicates that this intraisoform division remains unresolved. HCN4 is represented as the product of the second duplication event, followed by the emergence of HCN2 and HCN1.  Similar to the HCN2_urchin, the urochordates, C. intestinalis and C. savignyi, are shown to branch before the invertebrate clade. This position, however, is not conserved in the MP tree in which the 6 sequences fall according their evolutionary position, between the invertebrates and vertebrates. Furthermore, the low bootstrap values at the nodal partitions throughout the invertebrate/urochordate group indicate that the topology of these divisions remains unclear with this tree building method. This may be a consequence of excluding alignment positions with gaps and that insertions/deletions need to be included to resolve the evolutionary order of these primitive sequences. * indicate sequences identified in this study. Numbers indicate bootstrap values and represent number of trees containing the particular division.  179 0.1 hCNGA1 hCNGA3 1000 hERG1 KAT1 870 998 HCN2_urchin HCNa_c.intestinalis* HCNa_c.savingyi* 1000 1000 691 1000 HCN_urchin HCN_lobster HCN_fly HCN_mosquito* HCN_bee HCN_silkmoth 733 604 978 1000 963 HCN3_opossum* HCN3_mouse HCN3_rat 966 HCN3_human HCN3_cow HCN3_dog 279 505 681 1000 HCN3_green puffer* HCN3a_fugu* 1000 HCN3_zebrafish* HCN3b_fugu* 1000 896 HCN4a_green puffer* HCN4a_fugu* 1000 HCN4_zebrafish HCN4b_green puffer* HCN4b_fugu* 1000 709 985 HCN4_opossum* HCN4_rabbit HCN4_human HCN4_dog796 HCN4_mouse HCN4_rat 923 735 1000 820 HCN2_frog* HCN2_human HCN2_mouse HCN2_rat 960 991 872 HCN2a_green puffer* HCN2a_fugu* 1000 HCN2b_green puffer* HCN2_zebrafish 970 849 982 HCN1_trout HCN1_green puffer* HCN1_fugu 1000 1000 HCN1_opossum* HCN1_mouse HCN1_rat 649 HCN1_rabbit HCN1_dog HCN1_human HCN1_chimpanzee* 932 594 642 912 1000 1000 506 869 501 1000 710 747 651 HCNc c.intestinalis* HCNc c.savignyi* HCNb c.intestinalis* HCNb c.savignyi* HCN3 HCN1 Vertebrates HCN2 HCN4 Invertebrates & Urochordates  180  Figure A.2 : Maximum Likelihood Phylogram of the HCN Family. Due to the high sequence conservation and program limitations, the maximum likelihood phylogram was constructed using a subgroup of 41 sequences, resampling of 100 datasets and a randomized input order of 10 jumbles. Seqboot, Proml and Consense programs from PHYLIP were used and the JTT model of evolution was employed. A single representative tree similar in topology to consense tree was used to provide indication of branch length divergence. Numbers indicate bootstrap values for consense tree node divisions and represent % support for the respective partition. Consistent with the NJ and MP trees generated, HCN2_urchin is at the base of the phylogram, and is separated by a high bootstrap value. Similar to the NJ tree, Ciona HCNb genes are independent from the HCNa and HCNc and the invertebrates/urochordates divisions remain undefined. HCN3 is separated from the other three vertebrate isoforms by a high bootstrap value, whereas the order of the other two duplication events remains unresolved by this method. As ML tree results are dependent on the evolutionary model imposed, this undefined partition suggests that the JTT model can not conclusively generate the sequence evolution observed within the HCN family.   Based on branch length information, mammalian HCN4 and HCN2 sequences seem to have evolved the least from a common ancestral sequence, whereas all fish sequences and those in the HCN1 and HCN3 clades seem to have undergone more evolutionary change. Similar results were suggested by the sequence identity profiles shown in Figure 2.   181  0.1 HCN2_urchin HCNb_c.intestinalis HCNb_c.savignyi HCNc_c.intestinalis HCNc_c.savignyi HCNa_c.savignyi HCNa_c.intestinalis HCN4a_fugu HCN4b_fugu HCN4_zebrafish HCN4_mouse HCN4_opposum HCN4_human HCN2_frog HCN2_mouse HCN2_human HCN2a_green puffer HCN2_zebrafish HCN2b_green puffer HCN1_trout HCN1_fugu HCN1_opposum HCN1_mouse HCN1_human HCN3_green puffer HCN3a_fugu HCN3_zebrafish HCN3b_fugu HCN3_opposum HCN3_human HCN3_mouse HCN_urchin HCN_lobster HCN_fly HCN_mosquito HCN_bee HCN_silkmoth KAT1 hERG1 hCNGA1 hCNGA3 100 94.0 79.0 62.0 85.0 74.0 57.0 100 100 86.0 63.0 55.0 42.0 96.0 64.0 74.0 71.0 99.0 68.0 60.0 100 89.0 88.0 100 99.0 96.0 100 100 100 98.0 67.0 100 56.0 100 67.0 43.0 98.0

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0073194/manifest

Comment

Related Items