Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Binding and structural insights of the ryanodine receptor Lau, Kelvin 2014

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

Item Metadata


24-ubc_2014_november_lau_kelvin.pdf [ 25.2MB ]
JSON: 24-1.0166052.json
JSON-LD: 24-1.0166052-ld.json
RDF/XML (Pretty): 24-1.0166052-rdf.xml
RDF/JSON: 24-1.0166052-rdf.json
Turtle: 24-1.0166052-turtle.txt
N-Triples: 24-1.0166052-rdf-ntriples.txt
Original Record: 24-1.0166052-source.json
Full Text

Full Text

BINDING AND STRUCTURAL INSIGHTS OF THE RYANODINE RECEPTOR  by KELVIN LAU Hon. B.Sc. (Co-operative), University of Waterloo, 2008   A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Biochemistry and Molecular Biology)   THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) September 2014   © Kelvin Lau, 2014  ii Abstract Ryanodine Receptors (RyR) are large ion channels that are responsible for the release of Ca2+ from the sarco/endoplasmic reticulum. The channel consists of a large cytosolic cap which functions as a giant allosteric protein, capable of being modulated by an assortment of binding partners and small molecules. To understand its function and mechanisms one needs to dissect the channel to its smallest parts. Using a combination of isothermal titration calorimetry and x-ray crystallography, two areas have been analyzed: binding by calmodulin  (CaM) and the structure of a RyR domain, SPRY2. Calmodulin (CaM) is a Ca2+ binding protein that can regulate RyR under conditions of both high and low Ca2+ by tuning their Ca2+ sensitivity to channel opening and closing in an isoform-specific manner. I analyze the binding of CaM and its individual domains to three different RyR CaM binding regions using isothermal titration calorimetry. I compared binding to skeletal muscle (RyR1) and cardiac (RyR2) isoforms, under both Ca2+-loaded and Ca2+ free conditions.  I find that CaM is able to bind all three regions, but with different binding modes, between the isoforms. Disease mutations target one of the three sites and affect CaM binding and energetics.  The SPRY2 domain is one of three repeats of the same fold that are present within the RyR. It has been suggested as a key protein interaction site with dihydropyridine receptors to mediate excitation-contraction coupling in skeletal muscle tissue. RyR1 and RyR2 SPRY2 domains were crystallized and reveal differences with several other known SPRY domain structures. Docking of the RyR1 SPRY2 structure places it in between the central rim and the clamp region. The structure of a disease mutant causing cardiomyopathy is also determined and shows local misfolding. Finally, RyR1 SPRY2 binding to the DHPR II-III loops is undetectable by isothermal titration calorimetry.  iii Preface This thesis is composed of parts previously published or currently in the submission process. Chapter 1 consists of parts of a book chapter written by my supervisor and myself (Lau, K., Van Petegem, F. Encyclopedia of Signaling Molecules, Chapter: Ryanodine Receptor (RyR). (2013, 1st Edition) Springer Press. Editor: Sangdun Choi. ISBN 978-1-4419-0460-7. Pp. 1704-1709) with new appended sections. Chapter 2 is adapted from a published article Lau, K., Chan, M.M.Y, Van Petegem, F. Lobe-Specific Calmodulin Binding to Different Ryanodine Isoforms. Biochemistry. (2014) 53(5) 932-946. Mandy Chan performed preliminary cloning, expression, purification and ITC experiments on CaMBD2 and CaMBD1 peptides. I cloned, expressed, purified and ITC experiments on CaMBD1-3 peptides and their mutants. I prepared all the figures and jointly wrote the manuscript. My supervisor designed, supervised and jointly wrote the manuscript.  Chapter 3 is adapted from work that has accepted for publication in ‘Nature Communications’ as Lau, K., Van Petegem, F. Crystal structures of wild type and disease mutant forms of the Ryanodine Receptor SPRY2 domain. I cloned, expressed, purified, performed mutagenesis, crystallized both RyR1 and RyR2 constructs and related mutants, collected their diffraction data and solved the structures, analyzed the structure and performed thermal melting experiments. I prepared all the figures, except Figure 3.5 which was done by my supervisor, and jointly wrote the manuscript. My supervisor designed and supervised the project      iv Table of Contents Abstract ......................................................................................................................... ii Preface .......................................................................................................................... iii Table of Contents ......................................................................................................... iv List of Tables ................................................................................................................ vi List of Figures .............................................................................................................. vii List of Abbreviations ..................................................................................................... ix List of Symbols .............................................................................................................. x Acknowledgements ....................................................................................................... xi Dedication ................................................................................................................... xii Chapter 1 - Introduction ................................................................................................ 1 1.1 Calcium as a Biological Signal ..................................................................................... 2 1.2 Calcium and Muscles ................................................................................................... 2 1.3 Ryanodine Receptors: Discovery and Early Insights .................................................. 6 1.4 Disease Mutations ........................................................................................................ 7 1.4.1 Malignant Hyperthermia ............................................................................................................ 7 1.4.2 CCD .............................................................................................................................................. 8 1.4.3 CPVT ............................................................................................................................................ 8 1.5 Structure ........................................................................................................................ 9 1.5.1 Cryo-EM ..................................................................................................................................... 11 1.5.2 Domains of the Ryanodine Receptor .................................................................................... 13 The N-terminal Domains ................................................................................................................ 14 Phosphorylation Domain ................................................................................................................ 18 SPRY Domains ................................................................................................................................. 20 Calcium Binding Domains/CaM-like Domains .......................................................................... 22 1.6 Modulation and Regulation of Ryanodine Receptors ................................................ 22 1.6.1 FKBP .......................................................................................................................................... 23 1.6.2 Phosphorylation ........................................................................................................................ 24 1.6.3 EF-Hand Proteins ..................................................................................................................... 25      Calmodulin ........................................................................................................................................ 27 CaM Target Motifs ........................................................................................................................ 29 CaM Binding Modes ..................................................................................................................... 30     S100A1 ............................................................................................................................................... 33 1.6.4 Calmodulin and Ryanodine Receptors .................................................................................. 34 Binding Sites of Calmodulin on Ryanodine Receptors .............................................................. 35      Structural Biology of CaMBDs ...................................................................................................... 36 1.6.5 Skeletal Muscle DHPR α1S Subunit ........................................................................................ 39 1.6.6 β Subunit .................................................................................................................................... 41 1.7 Research Question ...................................................................................................... 41 Chapter 2 - Lobe-Specific Calmodulin Binding to Different Ryanodine Receptor Isoforms ....................................................................................................................... 43 2.1 Abstract ....................................................................................................................... 44 2.2 Introduction ................................................................................................................ 45 2.3 Methods and Materials ............................................................................................... 48  v 2.3.1 Cloning and Purification of CaMBDs ................................................................................... 48 2.3.2 Cloning and Purification of Calmodulin and its Lobes ...................................................... 51 2.3.3 Isothermal Titration Calorimetry ............................................................................................ 51 2.4 Results ........................................................................................................................ 52 2.4.1  Binding to CaMBD2 and its Mutants ................................................................................... 52 2.4.2  Binding to CaMBD1 ................................................................................................................ 64 2.4.3  Binding to CaMBD3 ............................................................................................................... 65 2.4.4  Disease Mutations in RyR1-CaMBD3 .................................................................................. 67 2.5 Discussion .................................................................................................................. 70 2.5.1 Ca2+/N-lobe is Loosely Associated with CaMBD2 ............................................................ 71 2.5.2 CaMBD1 Only Binds Ca2+/CaM ........................................................................................... 71 2.5.3 CaMBD3 Binds Strongly to Both ApoCaM and Ca2+/CaM ............................................. 72 2.5.4 Mutations in RyR1-CaMBD3 Affect the Energetics and Mode of Binding .................... 72 2.5.5 Multiple CaMBDs are Compatible with Known Structural Data ..................................... 72 2.5.6     CaMBD2 and CaMBD3 may be Accessible ......................................................................... 74 2.5.7     Calmodulation of RyR May Be Involved in More than Just Channel Activity ............... 75 2.5.8 CaMBDs Present Additional Platforms for Binding to Other EF-hand Proteins ......... 76 Chapter 3 - Crystal Structures of Wild Type and Disease Mutant Forms of the Ryanodine Receptor SPRY2 Domain .......................................................................... 77 3.1 Abstract ....................................................................................................................... 78 3.2 Introduction ................................................................................................................ 79 3.3 Methods and Materials ............................................................................................... 80 3.3.1 Cloning, Expression, Purification and Crystallization of SPRY2 ...................................... 80 3.3.2 Data Collection and Structure Determination ..................................................................... 81 3.3.3     Cryo-EM Docking .................................................................................................................... 82 3.3.4 Thermofluor Thermal Shift Assays ........................................................................................ 83 3.3.5 Isothermal Titration Calorimetry ............................................................................................ 83 3.4 Results ........................................................................................................................ 83 3.4.1 The SPRY2 Domains in RyR1 and RyR2 Consist of Modular Subdomains. .................. 83 3.4.2 Loss-of-Function Mutations in the RyR2 SPRY2 Domain ................................................ 92 3.4.3 Mutations in the RyR1 SPRY2 Domain ................................................................................ 95 3.4.4 Does the RyR1 SPRY2 Domain Form an Interaction Site with the II-III Loop? ......... 99 3.4.5 Location of the SPRY2 Domain in Full-length RyR1 ...................................................... 101 3.5 Discussion ................................................................................................................ 105 3.5.1 RyR SPRY2 Contains Modules that are Required For Folding and Not Present in Other SPRY Domains. ..................................................................................................................................... 105 3.5.2 Mutations in SPRY2 Induces Local Misfolding or Alter Interdomain Interactions. ... 105 3.5.3 The RyR1 SPRY2 Domain Binding to the CaV1.1 II-III Loop is Undetectable by Isothermal Titration Calorimetry. ....................................................................................................... 107 3.5.4     SPRY2 is a Domain in the Clamp Region of RyR ............................................................. 108 Chapter 4 – Future Directions and Conclusions ........................................................ 111 4.1 Calmodulin Can Bind More than One Site, but What Site? ...................................... 113 4.2 Calmodulin CPVT Mutants and CaMBDs ............................................................... 115 4.3 Calmodulin and S100A1 ............................................................................................. 115 4.4 Do CaMBDs Serve as Internal EF-Hand Binding Sites? ......................................... 116 4.4 SPRY1 and 3 Domains of RyR .................................................................................. 117 4.5 Tandem Domains ...................................................................................................... 118 4.6 FRET Localization of SPRY2 ................................................................................... 119 4.7 Conclusions .............................................................................................................. 120 References ................................................................................................................... 121  vi  List of Tables Table 1.1 Representative examples of CaM target motifs based upon their anchor spacing. 30	  Table 2.1 Thermodynamic Parameters of Binding between Ca2+/CaM and its Lobes and CaMBDs in 10 mM CaCl2 ................................................................................................................. 53	  Table 2.2 Thermodynamic Parameters of Binding between Ca2+/CaM and its Lobes in Competition Experiments and CaMBDs in 10 mM CaCl2 .......................................................... 55	  Table 2.3 Thermodynamic Parameters of Binding between Ca2+/CaM and its Lobes in Mutant CaMBDs in 10 mM CaCl2 ................................................................................................... 59	  Table 2.4 Thermodynamic Parameters of Binding between apoCaM and its Lobes and CaMBDs in 10 mM EDTA .............................................................................................................. 61	  Table 3.1 Data collection and refinement statistics ...................................................................... 82	                     vii List of Figures Figure 1.1 An overview of processes involved in RyR activation in cardiac myocytes. ............. 5	  Figure 1.2 A 12Å cryo-EM map of RyR1. ..................................................................................... 12	  Figure 1.3 The primary structure of RyR1 and its domains. ....................................................... 14	  Figure 1.4 The N-terminal domains and their location within the Ryanodine Receptor. ...... 15	  Figure 1.5 Domain A Δexon3 mutant is rescued by exon4. ........................................................ 17	  Figure 1.6 The N-terminus of the IP3 receptor superimposed on Domains ABC ................. 18	  Figure 1.7 The phosphorylation domain of RyR2 and its docked location in a cryo-EM map ............................................................................................................................................................... 19	  Figure 1.8 SPRY domains of different classes ............................................................................... 21	  Figure 1.9 The Ryanodine Receptor is modulated by diverse proteins and small molecules. 23	  Figure 1.10 The Ca2+ efflux rate constant from Ryanodine Receptors is dependent on external Ca2+ levels. ............................................................................................................................ 25	  Figure 1.11 A typical EF-hand ......................................................................................................... 26	  Figure 1.12 Apocalmodulin and Ca2+/Calmodulin ....................................................................... 28	  Figure 1.13 Calmodulin binds to its targets in a variety of modes ............................................. 32	  Figure 1.14 Ca2+/CaM bound to amino acids 3614-3643 from RyR1. ...................................... 37	  Figure 1.15 Ca2+/S100A1 bound to amino acids 3616-3627 from RyR1. ................................ 38	  Figure 1.16 Localization of CaM and potential CaMBD by difference Cryo-EM densities. . 39	  Figure 2.1 Sequences of three CaM binding domains (CaMBDs). ............................................. 47	  Figure 2.2 Size exclusion chromatograms and SDS-PAGE gels of purified and synthesized CaMBDs. ............................................................................................................................................. 50	  Figure 2.3 Ca2+/CaM and its lobes bind to all three RyR1 CaMBDs. ....................................... 54	  Figure 2.4 Competition between N-lobe and C-lobes of CaM in all 3 CaMBDs. ................... 56	  Figure 2.5 Ca2+/CaM and its lobes bind to all three RyR2 CaMBDs. ....................................... 58	  Figure 2.6 RyR2 CaMBD2 W3586A and F3602A mutants alter lobe binding. ....................... 60	  Figure 2.7 Binding of apoCaM and its lobes to RyR1 CaMBDs. ............................................... 62	  Figure 2.8 Binding of apoCaM and its lobes to RyR2 CaMBDs. ............................................... 63	  Figure 2.9 Overview of Ca2+/CaM and apoCaM binding to CaMBDs. .................................... 66	  Figure 2.10 R4325D and LRR duplication mutants in RyR1 CaMBD3 .................................... 68	  Figure 2.11 ApoCaM and its lobes bind to RyR1 CaMBD3 LRRdup and R4325D mutants with different energetics than wild-type. ........................................................................................ 69	  Figure 3.1 Superposition of chains A, B and C (blue, green, purple, respectively) of RyR1 SPRY2. ................................................................................................................................................. 84	  Figure 3.2 Overall structure of the SPRY2 domain. ..................................................................... 86	  Figure 3.3 Swapping of the tail region in the crystal lattice. ........................................................ 88	  Figure 3.4 Interactions between the tail and the N-terminal extension strand in RyR2. ........ 88	  Figure 3.5 Comparison with other SPRY structures. ................................................................... 90	  Figure 3.6 Sequence alignment of the three predicted SPRY domains in rabbit RyR1. ......... 91	  Figure 3.7 Disease mutations within SPRY2. ................................................................................ 93	  Figure 3.8 Clash induced by the A1107M mutation in RyR2 SPRY2. ....................................... 94	  Figure 3.9 SPRY2 mutants are monomeric. ................................................................................... 94	  Figure 3.10  Tail Interactions with Arg1076. ................................................................................. 97	  Figure 3.11 Effect of the G1165D disease mutation. ................................................................... 98	  Figure 3.12 Absence of interactions between the RyR1 SPRY2 domain and the CaV1.1 II-III loop. .................................................................................................................................................... 100	   viii Figure 3.13 Location of the SPRY2 domain in intact RyR1 ..................................................... 102	  Figure 3.14 Docking of the RyR1 SPRY2 domain into the RyR1 cryo-EM map with EMDB entry 1275. ......................................................................................................................................... 103	                     ix List of Abbreviations ADP+Pi Adenosine diphosphate + inorganic phosphate ApoCaM Apocalmodulin ARVD2 Arrhythmogenic Right Ventricular Dysplasia Type 2 ATP Adenosine triphosphate BME β-mercaptoethanol Ca2+ Calcium ion Ca2+/CaM Calcium-bound calmodulin CaM Calmodulin CaMBD Calmodulin Binding Domain CaMKII Ca2+/CaM dependent kinase II CaV Voltage-gated calcium channel CCD Central core disease cDNA Complementary deoxyribonucleic acid CICR Calcium-induced calcium release CPVT Catecholaminergic Polymorphic Ventricular Tachycardia Cryo-EM Cryogenic-electron microscopy CV Column volume E-C Excitation contraction  EDTA Ethylenediaminetetraacetic acid EMDB Electron microscopy databank ER Endoplasmic reticulum Da Dalton DHPR Dihydropyridine receptor DR Divergent region FKBP FK-506 binding protein FRET Förster Resonance Energy Transfer GFP Green fluorescent protein  HEPES 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid II-III loop Dihydropyridine receptor loop between domains 2 and 3 IP3  Inositol 1,4,5-triphosphate IP3R  Inositol 1,4,5-triphosphate receptor IPTG Isopropyl-β-D-thiogalactoside ITC  Isothermal titration calorimetry K+ Potassium ions MALDI-TOF Matrix-assisted laser desorption ionization-time-of-flight MBP Maltose binding protein Mg2+ Magnesium ion MH Malignant hyperthermia Na+ Sodium ion NaV Voltage-gated sodium channel NCX Sodium-calcium exchanger OD600 Optical density at 600 nm RyR Ryanodine receptors SERCA sarco/endoplasmic reticulum calcium ATPase SR sarcoplasmic reticulum  x List of Symbols Kd Dissociation constant ∆G Gibbs free energy ∆H Change in enthalphy ∆S Change in entropy                         xi Acknowledgements I would like to thank my supervisor Dr. Filip Van Petegem for the wonderful opportunity to be part of his lab. His mentorship and guidance will be forever valued and I am extremely grateful. I would also like to thank the members of my supervisory committee Dr. Lawrence McIntosh and Dr. Masa Numata for their roles in guiding me towards this point. Second, I would like to thank all the members of my lab as well as colleagues in other labs within the Life Sciences Centre that have made my graduate career an extremely fun and rewarding experience. Special thanks goes especially to Jett Tung, Lynn Kimlicka, Dr. Michael Yuchi, Paolo Lobo, Dr. Sarah Chow and Dr. Maen Sarhan for their invaluable assistance and for endless hours of discussion.   Finally I would like to thank my parents and my family for their unconditional support throughtout my graduate career. I thank them for their boundless love and persistent encouragement to pursue my dreams and interests.            xii Dedication  This dissertation is dedicated to my grandmother who passed away shortly before the completion of this document. You raised me ever since I was a child and I know you are proud of me.       1            Chapter 1 - Introduction          This chapter has been adapted from the original publication: Lau, K., Van Petegem, F. Encyclopedia of Signaling Molecules, Chapter: Ryanodine Receptor (RyR). (2013, 1st Edition) Springer Press. Editor: Sangdun Choi. ISBN 978-1-4419-0460-7. Pp. 1704-1709  2 1.1 Calcium as a Biological Signal The formation of membranes that separated external and internal environments from the cytosol allowed for the formation of large concentration gradients between compartments. The change in concentrations of molecules in time and space allows for distinct messages or signals to be transmitted.1 Calcium ions (Ca2+) have been exploited during evolution as one of several secondary messengers that are critical for signalling within a cell.2 Initially, Ca2+, did not take on this role as early cells only maintained strict control over cytosolic Ca2+ levels as an excess is toxic due to the formation of phosphate precipitates, aggregation of nucleic acids, and disruption of the cell membrane.3 The cell thus expends tremendous resources to maintain a Ca2+ homeostatic system that differs by up to 20,000 times between the intracellular (~10-100nM) and extracellular environments (mM).2 Ca2+ is actively transported out of the cell by pumps and transporters or into intracellular stores.1 Early cells eventually exploited this large concentration difference by evolving proteins that can sense Ca2+  fluctuations over a concentration range of (nM to mM).4 These proteins functioned to buffer Ca2+ levels but also serve as triggers for cellular processes such as cell division, motility, immunity, muscle contraction and apoptosis.4  1.2 Calcium and Muscles Skeletal and cardiac muscles are made up of muscle fibres that compose a bundle known as a fascicle. Each fibre’s entire length contains thread-like sub-structures known as myofilaments composed of actin which alternates with bundles of myosin filaments.5 It was known as early as the 1960s that Ca2+ released from the sarcoplasmic reticulum was the trigger that allowed for the contraction of muscles.6 Seemingly, upon membrane depolarization, an electrical signal could be transmitted into a cell that subsequently released  3 Ca2+.7 The identity of the molecules and signalling pathways was unknown at the time. Another mystery was how a signal could diffuse through the diameter of a muscle fibre (50-100µm) in only 1-2msec to the centre of a fibre to trigger a contraction.6 Invaginations of the plasma membrane or sarcolemma into the muscle fibres provided an answer. By folding inwards, the interior of the fibre is never more than 1µm away from a membrane, similar to the christae found in mitochondria.8 These depressions known as T-tubules are continuous with the surface and allow for signals to penetrate deep within muscle fibres.8-10 In the 1970s, the T-tubules were discovered to be rich in voltage-gated calcium channels (Cav), also known as dihydropyridine receptors (DHPR), that open when depolarized and allow for the rapid influx of extracellular Ca2+ upon the arrival of an action potential.6 The presence of these channels led to the hypothesis that they were the voltage sensors that detect membrane depolarization.11-13 However, a component was still missing because the DHPRs are located on T-tubules and thus could not directly mediate SR Ca2+ release.14 The missing link was a calcium release channel that was first observed in electron micrographs as a large electron dense foot located on the SR.15-17 The molecular identity of these gateways that release SR Ca2+ was not even known until one was cloned later in 1989.18,19  These large channels, now called Ryanodine Receptors (RyR) are the gateways in which the calcium stores of the SR are released into the cytosol when they are activated. Thus, the electrical stimulus via DHPR acts as a signal to open the RyR that is then triggered to release Ca2+ from intracellular stores.  The actual contracture is described by the sliding filament theory where the rapid change in Ca2+ is first sensed by troponin C.5,20 Upon binding, troponin and tropomyosin, located on actin-containing filaments of myofibrils, undergo a conformational change revealing myosin binding sites on actin. Myosin forms bridges with actin and repeated cycles  4 of release of ADP+Pi and hydrolysis of ATP allow for myosin head movement along an actin filament producing force and motion ultimately leading to contracture. As available Ca2+ is reduced due to active pumping by SERCA, troponin returns back to its resting state resulting in the loss of myosin binding sites on actin and relaxation.5 The entire process from the propagation of the action potential down the sarcolemma to the muscle movement is called excitation-contraction coupling (E-C coupling).6 Different types of muscle tissue undergo E-C coupling in distinct ways. In cardiac muscle, Ca2+ influx from DHPR is sensed by RyRs that then act as a signal amplifier to increase amounts of cytosolic Ca2+ levels in a process called Ca2+-induced-Ca2+-release (CICR). (Figure 1.1)            5  Figure 1.1 An overview of processes involved in RyR activation in cardiac myocytes.  Particular voltage-gated calcium channels known as dihydropyridine receptors (DHPR) allow an influx of Ca2+ upon the arrival of an action potential. Through CICR, the RyR releases Ca2+ from the sarcoplasmic reticulum, triggering muscle contraction. Also shown is a PKA-dependent phosphorylation pathway that activates RyR upon β-adrenergic stimulation. The Sodium-Calcium Exchanger pumps remove one Ca2+ ion for 3 Na+ in return. The Sarco/endoplasmic Reticulum Ca2+-ATPase (SERCA) allows for reuptake of Ca2+ from the cytosol. It is regulated by phospholamban (PLB) that is also a substrate target for PKA.  6 Skeletal muscle contraction does not require CICR, but can proceed through another mechanism. It is thought that skeletal muscle DHPR and RyR1 are directly coupled at the T-tubule and that the electrical signal that opens DHPR also induces a conformational change within its II-III loop.6 The II-III loop provides a critical link with the RyR allowing it to activate RyRs even in the absence of Ca2+ influx. Many see this coupling as a way for the DHPR to act as a voltage sensor for the RyR that becomes the pore of a multi-transmembrane complex.  1.3 Ryanodine Receptors: Discovery and Early Insights  RyRs play a critical role in muscle contraction and their discovery and naming have only been achieved since the 1980s. The channel was observed in junctional SR vesicles isolated from cardiac and skeletal muscle tissue as electron dense “feet” that resembled objects first seen in intact tissues. 17,21,22 The name is derived from the initial observation that Ca2+efflux from SR vesicles could be enhanced or inhibited depending on the amount applied of an alkaloid insecticide, Ryanodine.15 Further work with SR vesicles showed that Ca2+ release could be modulated by various small molecules like ATP, Mg2+, and Ca2+.23 Early biochemical evidence first showed that the channel exists as a homotetramer, with each monomer having a mass of ~550 kDa. This makes RyRs the largest known ion channels at ~2.2 MDa.24Recordings made in planar lipid bilayers showed that RyRs are permeable only to small inorganic and organic cations and are completely impermeable to anions.25-27 Mammalian organisms contain three known RyR isoforms. In 1989, Takeshima et al. reported the first primary structure and cloning of cDNA of RyR derived from rabbit skeletal muscle (RyR1).19 The following year, the group of MacLennan published the cloning  7 of the human skeletal muscle RyR and rabbit cardiac muscle isoform (RyR2).28 A third isoform (RyR3), distinct from both the cardiac and skeletal muscle isoforms, was originally cloned from brain tissue but is expressed in a variety of other tissues as well.29 The genes of all the isoforms are located on different chromosomes in humans. RyR1 is found on chromosome 19q13.2 spanning 104 exons. The gene encoding RyR2 is on chromosome 1q43 (102 exons) and RyR3 is encoded on 15q13.3-14 (103 exons).30-32 In addition to mammalian organisms, RyRs have also been identified in non-mammalian vertebrates with two isoforms, RyRα (homologous to RyR1) and RyRβ (similar to RyR3).33,34 Moreover, RyR genes have also been identified in invertebrates including Caenorhabditis elegans and Drosophila melanogaster.35,36 The importance of RyRs is evident from several knockout (KO) studies. RyR1 double KO mice die immediately after birth, whereas an RyR2 KO is embryonically lethal. RyR3 KO mice survive, but have impaired learning abilities.37-39  1.4 Disease Mutations Since Ca2+ is a potent second messenger, mutations in proteins involved in Ca2+ release can lead to genetic disorders. As such, numerous disease-causing mutations have been identified in RyR1 and RyR2. RyR1 mutations commonly lead to either malignant hyperthermia or congenital myopathies and RyR2 mutations have been linked to catecholaminergic polymorphic ventricular tachycardia and cardiac hypertrophy.   1.4.1 Malignant Hyperthermia Malignant hyperthermia (MH) is a disorder that requires both the presence of a trigger and a genetic mutation to cause disease.40  The triggers are volatile anaesthetics such  8 as halothane, sevoflurane and desflurane, and the muscle relaxant, succinylcholine.41 An MH episode is characterized by a sudden rise in core body temperature, muscle rigidity, and acidosis.42 This is caused by myocytes being in a hypermetabolic state to continuously remove Ca2+ leaked through the RyR from the cytosol.43 Incidence in the population could be as high as 1:2000; however, a true figure is hard to determine as less than 2% of the population will ever be in a situation that requires any of the pharmacological triggers.44 Although fatal historically, most surgery rooms keep a stock supply of dantrolene, a RyR1 blocker that can rapidly reverse the symptoms. Fatalities are greater in young men with the cause unknown.45 It is thought that this drug affects RyR by inhibiting Ca2+ release; however, evidence for direct binding is controversial.46,47  1.4.2 CCD Mutations in RyR1 also lead to myopathies with central core disease (CCD) comprising the largest group. CCD is typified by extensive muscle weakness, a decrease in muscle tone, skeletal deformities and scoliosis. Diagnosis is usually confirmed by the presence of lesions or “cores” within the centre of myocytes that are devoid of metabolic activity and lack mitochondria.48-50 Multi-mini core disease (MmD) is similar to CCD but patients exhibit multiple cores instead of just one. In addition, CCD is autosomal dominant while MmD is mostly inherited recessively.49  1.4.3 CPVT Catecholaminergic polymorphic ventricular tachycardia (CPVT) is the main disease resulting from mutations in RyR2. It is typically diagnosed in young patients that have family members with known CPVT, and by the presentation of symptoms such as syncope and  9 sudden death.51 Like RyR1, it is thought that CPVT is triggered by exercise or even emotional stress. These activities release catecholamines and trigger a β-adrenergic response leading to direct phosphorylation of RyR2 increasing its activity or through phosphorylation of proteins that modulate RyR2 or other components involved in Ca2+ handling like phospholamban.52,53 In addition to RyR2, mutations in proteins that interact with RyR2 like calsequestrin54, triadin55 and CaM56,57 have also been shown produce CPVT phenotypes.  In both isoforms, it is thought that the presence of mutations cause gain-of-function “leaky” channels allowing for excess Ca2+ release leading to prolonged or premature calcium signals within the cell.52 Mutations that lead to CCD can also be due to loss-of-function.58 In cardiac myocytes, the excess Ca2+ resulting from CPVT mutations, may activate the Na+/Ca2+ exchanger (NCX) on the cytoplasmic membrane that exchanges one Ca2+ out for every three Na+ ions. The new influx of positive charges generates inwardly depolarizing currents causing delayed-after-depolarizations (DAD) leading to cardiac arrhythmias.59 In skeletal muscle myocytes, the massive release of Ca2+ stimulates activity of Ca2+-ATPases, ultimately depleting the ATP reserves and leading the cells into a hypermetabolic state.43 Disease mutations in RyR1 and RyR2 seem to cluster in three different regions of the receptor genes (hot spots). (Figure 1.3) These three locations in both receptors match, suggesting important functional roles for those portions. They are located near the N-terminus, in a central domain region, and a C-terminal region covering the transmembrane segments.60  1.5 Structure RyRs are the largest known ion channels, with molecular weights ~2.2 MDa. They form homotetrameric arrangments. The three isoforms found in mammals (RyR1, RyR2 and  10 RyR3) are highly similar with amino acid sequence identities ~70%. Of note are several divergent regions (DR) with the greatest dissimilarity between isoforms, DR1, DR2 and DR3. DR1 encompasses a large part of the C-terminus and has the greatest dissimilarity.61 Interestingly, RyRs have significant sequence and structural homology to the inositol-1,4,5-triphosphate (IP3) receptor, another intracellular Ca2+ release channel.62,63 Their N-terminal domains are structurally homologous and it has been suggested that both channels evolved from a single ancestor.  Early work utilized negative-stain electron microscopy (EM) to visualize sub-cellular localization of RyRs along with other proteins that compose the EC machinery. 64,65 Images of the junction region between the SR and T-tubules of muscle tissues revealed large “foot” structures. Improvements to resolution revealed that these feet were four-fold symmetric and provided evidence for large-scale organization of RyRs due to their ability to form two-dimensional arrays in micrographs.21 In addition, tetrads of skeletal muscle DHPR co-localized with RyRs, further suggesting a physical connection between these two membrane proteins.66 Cryo-EM has been utilized to provide a higher-resolution structure of the channel, showing that RyRs are mushroom-shaped and composed of multiple globular domains.67,68 Studies have imaged open and closed states in the presence of certain ligands.69-71 However, since RyRs can adopt multiple conformational states that likely become averaged during the imaging, the resolution of the cryo-EM has thus far been limited to ~12Å. Due to their large size and membrane nature RyRs have posed a challenge for high-resolution structural studies.72 Most methods in structural biology require large amounts of protein at high purity. Due to RyRs inherent size and the presence of hydrophobic transmembrane regions, proper folding requires chaperones and post-translational modifications that may not be present in heterologous expression systems.73 Only since 2006  11 have atomic resolution structures of portions of RyR been determined by NMR and X-ray crystallography.  1.5.1 Cryo-EM The cryo-EM structures revealed what typifies RyRs, a large mushroom-cap cytosolic region that consists of 90% of the protein that is connected to a stem that leads to the transmembrane pore. The cap resembles a small prism (290x290x130Å) sitting upon the pore-forming transmembrane region (120x120x70Å).67 Performed on a closed state RyR1, the highest resolution 3D reconstruction to 12Å showed that the cap is a compact assembly of at least 10 globular domains that arranged themselves in a web-like scaffold.74(Figure 1.2) The domains are organized into regions that were described in lower resolution maps: the clamp, handle and central rim.75 Large solvent channels in the assembly account for over 50% of the enclosed volume and this maximizes the total surface area and allows for a variety of ligands and proteins to bind.74,76 Reconstructions of RyR2 and RyR3 show remarkably similar architecture to RyR1 that is expected due to their high amino acid sequence similarity to RyR1.77,78 In addition to the closed state architecture, cryo-EM studies have revealed the structure of RyR1 in the open state at 10.2Å.69 Upon opening, the channel undergoes large conformational changes in the cytosolic region, along with motions in the pore-forming part of the transmembrane region. The RyR is therefore a bona fide allosteric protein, whereby motions in the pore are coupled to regulatory domains in the cytosolic portion.    12  Figure 1.2 A 12Å cryo-EM map of RyR1.  The map shows the channel viewed from the top (A) cytosolic side and from the ‘side’, parallel to the membrane (B). The colouring and numbering represents individual regions of density that form globular regions that may be composed of one or more protein domains. Dotted outlines indicate common names of regions use in describing RyR. Reproduced with permission from (Serysheva et al. 2008)74     13 Binding sites of accessory proteins such as calmodulin (CaM), FKBP12/12.6 and the toxin imperatoxin A have been visualized by examining the difference densities and in the presence and absence of these proteins.70,79-81 Owing to the large size of RyR, attempts have been made to correlate the sequence of the primary structure within an intact tetramer. Recombinant RyRs with green fluorescent protein (GFP) or glutathione S-transferase (GST) insertions at various points have allowed for the visualization of additional density in cryo-EM reconstructions.78,82-86 Antibodies have also been targeted to specific sequences and the extra density owing to the antibody is interpreted as the region where the targeted sequence might be present.87,88   1.5.2 Domains of the Ryanodine Receptor Through primary structure analysis, structural homology to other channels and proteins and secondary structure prediction, RyRs appear to be built up by multiple individual domains.89 (Figure 1.3) Prior to the work described in this thesis, X-ray crystallography and NMR structures have been described for two different regions (N-terminal region and phosphorylation domain) within the RyR sequence.90-95 These structures represent only ~12% of the total channel.   14  Figure 1.3 The primary structure of RyR1 and its domains.  This primary structure map shows the approximate locations of domains. Each coloured region represents a predicted or currently known domain. The names of the domains are listed at the top. Dark blue lines represent the location of known disease causing mutations in RyR1. Dotted red boxes represent three regions of sequence divergence between RyR1 and RyR2 or divergent regions (DR1-3). The N-terminal Domains Domains within the N-terminal region were the first to be described. Initially, single domains of the very N-terminus of both RyR191 and RyR290 were described. Later, structures encompassing the first three domains in RyR192 and RyR296 were determined.  For simplicity, we refer to these three domains as domains A, B and C. The ABC domains comprises the largest fragment that has been solved. It is approximately 550 amino acids in length in both RyR1 and RyR2 or about 10% of the full-length RyR. The region is composed of 3 domains, two β-trefoil domains comprising 12 β-strands each (domains A 1-204, and B 205-394; RyR1 rabbit) and an α-helical bundle resembling armadillo repeat (domain C 395-532).92,96(Figure 1.4) The three domains together form a compact triangular arrangement with significant domain-domain interactions that are mainly hydrophilic in nature.92 In RyR2, an ion pair network, responsible for domain-domain interactions, has been replaced by a central  15 chloride binding site.96 The N-terminal region represents a part of the channel that is a target for 80 mutations, and it has therefore been suggested as a disease 'hot spot'. Several structures of the disease mutants have been determined and show that disruption of salt bridges through mutations causes relative domain reorientations.97   Figure 1.4 The N-terminal domains and their location within the Ryanodine Receptor.  A. Domains A, B and C (PDB 2XOA) are arranged as a compact triangle mediated by hydrophilic interactions at the interface between domains. B. ABC docks within the 10.3Å map (EMDB 5014) in the central rim region of the RyR. It forms a 4-fold symmetric vestibule on the cytosolic face, far away from the pore-forming region.  Docking of ABC into cryo-EM maps reveal that the domain forms a four-fold symmetric ring at the cytosolic surface of the channel forming a central vestibule.92 (Figure 1.4) In addition to the interfaces formed between the three domains, there are 5 additional interfaces with the rest of the receptor. The pseudo-atomic model reveals that, although it is far from the transmembrane region, the N-terminal domains are connected to the pore through electron-dense columns. Docking of the N-terminal domains into different cryo- 16 EM maps and FRET measurements show that the domains are mobile during opening and closing of the channel and that they are allosterically coupled to the pore region.97,98 Some disease-causing mutations are buried within individual domains, but the bulk are found at domain-domain interfaces. This suggests that proper domain-domain interactions are required for a normal functioning RyR. Domains ABC can be further subdivided and purified as either domain A or domains BC. Domain A was the first RyR domain structure to be solved by X-ray crystallography and by NMR.90,91 Even in such a short segment, domain A contains 32 mutations between RyR1 and RyR2. An unusual deletion mutant of domain A also exists where the entirety of exon3 (Δexon3) is missing. Individuals carrying this mutation were found to suffer from a severe form of CPVT.99,100 Surprisingly the structure could be determined and revealed wide-ranging changes. Instead of missing the β-strand structural element that exon 3 provided, a flexible alpha helix encoded by exon 4 replaces the missing strand, rescuing the fold of the protein.101,102 (Figure 1.5) This novel structural rescue mechanism suggests that alternative splicing may be used as a mechanism to fine-tune channel activity.101   17  Figure 1.5 Domain A Δexon3 mutant is rescued by exon4.  A. Domain A (PDB 2MC3) is coloured in blue, exon3 is in orange and exon 4 is in red. B. The Δexon3 is a mutation that removes all of exon3 that encodes a crucial β-strand. Surprisingly in the mutant, the β-strand is replaced by exon4, which undergoes a helix-to-strand conformational change.  The structure of ABC also revealed its similarities with RyR’s sister channel, the IP3 receptor (IP3R). The structures of the first three domains of IP3R1 (isoform 1) have been determined and show remarkable structural homology to domains ABC.63 (Figure 1.6) The first domain of IP3R1 is the “suppressor” domain which is analogous to domain A while the IP3-binding core represents domains BC.63 In IP3Rs, ligand binding occurs within the IP3-binding core which then activates the channel.103 The suppressor domain inhibits this process by forming interactions with the IP3-binding core preventing it from undergoing conformational changes that allow it to bind IP3. Structural and functional parallels to RyRs suggest that mutations or ligand binding that weaken domain interfaces may promote channel opening as well. Supporting this is cryo-EM docking of the N-terminal domains of  18 IP3R1 that also locate them to the corresponding positions in the receptor. Interestingly, chimeras whereby IP3R domain A is replaced by the corresponding domain from RyR2 still produce functional channels.63 Since IP3 binding can stimulate channel opening, a similar ligand may be modulating RyR activity. However, no ligands have been described so far that can bind to the RyR N-terminal region.  Figure 1.6 The N-terminus of the IP3 receptor superimposed on Domains ABC  A. Superposition of the N-terminus of the IP3 receptor (PDB 3UJ4) yields a striking similarity to domains ABC with an RMSD of 1.2Å. The largest difference is the presence of longer helices within the IP3R domain A. Domains ABC are coloured as before and the IP3 receptor is in beige. Phosphorylation Domain The RyR amino acid sequence contains a short region that is repeated 4 times. 104,105 These repeats are found in tandem pairs in two areas (Figure 1.3). The second such tandem pair, encoded by residues 2743-2940 (rabbit RyR1), 2699-2904 (mouse RyR2) and 2597-2800 (human RyR3) contains multiple phosphorylation sites (Ser/Thr/Tyr), and is often referred  19 to as the phosphorylation domain. This domain has two-fold symmetry with each half adopting a helix-strand-helix structure.93,106 It has been a region of intense interest and controversy due to the presence of multiple phosphorylation sites that have been implicated in heart failure.107-114  The phosphorylation domain was crystallized in all three isoforms and revealed phosphorylation sites that are thought to be crucial to channel regulation. In RyR2 several sites are located on a flexible loop that were not well resolved suggesting a highly dynamic nature. (Figure 1.7) Disease mutations in RyR1 cluster around the phosphorylation sites, suggesting that the mutations and phosphorylation may promote similar structural changes.93     Figure 1.7 The phosphorylation domain of RyR2 and its docked location in a cryo-EM map  A. The phosphorylation domain consists of two pseudo-symmetric halves connected by a flexible loop. The latter contains several phosphorylation sites that have been linked to CPVT. Sites capable of being phosphorylated in vitro are in red (PKA) and blue (CaMKII). All of these sites, together with additional ones (no dots), have also been identified by mass spectrometry of native tissue. B. The docking of the phosphorylation domain (black circle) places it right at the tip of the clamp region of the channel. Adapted from Yuchi, Lau and Van Petegem, 2012.     20 Docking of the phosphorylation domain into cryo-EM maps localized it to the tip of the clamp domain. (Figure 1.7) As that is a region that undergoes significant movement as observed between closed/open state maps, phosphorylation may play a role in the transition between the two states. However, no neighbouring domain has been identified.  Interestingly, no other eukaryotic protein is known to contain these tandem repeats, but they have been identified in prokaryotes and viruses, and that is where they are thought to have originated.105 The other tandem repeats, located between the first two SPRY domains, current has an unknown function. (Figure 1.3) SPRY Domains SPRY domains were first identified as a repeat domain found multiple times within a splA kinase from Dictyostelium discoideum and within RyRs.104,115 SPRY domains have now been identified in over 177 mammalian proteins and are usually modules mediating specific protein-protein interactions that play a role in immunity.116 They are composed of a conserved twisted β-barrel core of 12 strands connected by highly variable loops.116 SPRY domains segregate into two families based upon their sequence phylogeny: ‘SPRY-only” and “PRYSPRY/B30.2” domains.117 Historically, the SPRY core was predicted in both families, with an additional Pre-SPRY (PRY) motif in the B30.2 family. Structures have revealed that the SPRY-only family also contains elements that are structurally homologous to the PRY motif, but differ significantly in sequence. These structural features are the addition of two to three helices and three β-strands that are N-terminal to the SPRY core. (Figure 1.8)  21  Figure 1.8 SPRY domains of different classes  Three representative SPRY domains are shown here in cartoon representation with their PDB accession codes below. A. The SPRY domain of Ash2L does not contain the N-terminal helices present in other SPRYs. B & C. Although bioinformatically the ‘SPRY-only’ and B30.2 class are recognized as distinct and separate families, other than the β-barrel core, structural features such as their N-terminal helices are conserved.  Although named partly after RyRs, no SPRY structure from the channel had been solved prior to the work described in this thesis. Within the RyR, there are three repeats (SPRY1 659-797, SPRY2 1084-1207, SPRY3 1430-1570; RyR1 human) (Figure 1.3) that show low sequence identity to known SPRY domains (~20%). Of the three only SPRY2 has been implicated in a RyR function.118-120 SPRY2 is contained within the region that has been implicated as the interaction site for the II-III loop of CaV1.1 mediating excitation-contraction coupling.119 Cryo-EM studies utilizing antibodies against peptides and segments of SPRY2 have visualized extra density near the clamp region.88     22 Calcium Binding Domains/CaM-like Domains The ryanodine receptor not only conducts Ca2+ but is also activated and regulated by Ca2+ through CICR. CaM is able to fine-tune the channel’s Ca2+ sensitivity by binding to a CaM binding domain (CaMBD; 3614-3643; RyR1 rabbit). Depending on the free cytosolic Ca2+ concentration CaM is able to activate or inhibit the channel.121 CaM contains EF-hand motifs which allow it to act as Ca2+-sensors.122 Upon binding Ca2+, the EF-hands undergo conformational changes allowing CaM to bind differently to RyR. 94  Interestingly, the RyR itself contains several EF-hand motifs. Through sequence alignments with known EF-hand proteins, Xiong et al., have identified several regions within RyRs that have a surprising homology to CaM.123,124 In RyR1, the 4064-4210 region could be purified and approximates the size of CaM. Initial biochemical data using equilibrium dialysis suggests that not all the EF-hands are functional, as only two Ca2+ were able to bind.123 These EF-hands may function as the RyR’s intrinsic Ca2+ sensors and be involved in CICR.   1.6 Modulation and Regulation of Ryanodine Receptors Since cytosolic Ca2+ needs to be strictly regulated, RyRs have evolved to become sensitive signal integrators, receiving input from many proteins and small molecules. These mostly bind to the cytosolic cap and provide positive or negative input. In addition proteins in the ER lumen are also able to bind and regulate channel function. The sites of action for most of these interactions are still to be discovered. Figure 1.9 shows the multitude of positive and negative regulators of RyR. A few of the regulators are discussed below.   23  Figure 1.9 The Ryanodine Receptor is modulated by diverse proteins and small molecules.  Regulators in the cytosol and within the SR modulate RyRs. Proteins and compounds within the blue circle are activators and those within the red circle inhibit the channel. Some are able to both activate and inhibit the channel.  1.6.1 FKBP FK506-binding proteins were initially discovered as they are tightly bound to RyR and are copurified.125 They are immunophilins that exhibit cis-trans-peptidyl-prolyl isomerase activity but can also stabilize the closed state of RyR.126,127 A skeletal muscle isoform of 12 kDa and a cardiac muscle isoform of 12.6 kDa isoform exists (FKBP12 and FKBP12.6) and thus the former typically binds RyR1 and the latter binds RyR2. The exact mechanism of action of FKBP12.6 on RyR2 is still unknown but is of intense interest as loss of the protein has been implicated in heart failure and arrhythmias.107,128 In experiments where FKBP12 was absent on RyR1, the channels were observed to enter multiple sub-conductance states, to have higher open probabilities, resulting in increased Ca2+ leak and reduced EC coupling.127,129 One group has reported that FKBP12 promotes coupling between neighbouring RyR allowing for activation of one channel to affect its neighbours.130 Cryo- 24 EM studies have shown difference density in the 16Å RyR1-FKBP12 and 33Å RyR2-FKBP12.6 maps between the handle and clamp regions and show that 4 FKBPs bind to one RyR. 79,131 Through yeast-two hybrid experiments, it has been proposed that RyR1 residues 2361-2496 are involved in binding but further experiments are still required to confirm.107   1.6.2 Phosphorylation RyRs are regulated by various kinases (PKA, PKG and CaMKII) and phosphatases that are able to modulate channel activity and connect RyRs to stress signalling.132-134 Protein kinase A (PKA) has been found to phosphorylate multiple residues on RyR (S2843 in human RyR1; S2030 and S2808 in mouse RyR2).114,133,135 Calmodulin kinase II (CaMKII) is also able to phosphorylate S2808 in RyR2 but also S2814.136,137 Mass spectrometry has shown that an even greater number of sites could possibly be phosphorylated in vivo.138 Several reports have suggested that hyperphosphorylation of S2808 leads to FKBP12.6 loss that results in heart failure.107,135,139 However, conflicting studies have shown either both increased or unchanged activity in single channel recordings of RyR in the presence of PKA or phosphomimetic mutant S2808D.137,140 Similarly, in studies with S2808A and S2808D mice, PKA phosphorylation affected Ca2+ spark frequencies and amplitudes, but this could not be repeated by others.112,135 No study has been able to attribute changes to a particular phosphorylation site because a modification may simultaneously affect other sites. One cannot eliminate the possibility of cross-talk between phosphorylation sites. In addition, several groups have been unable to reproduce the FKBP12.6 dissociation under conditions of heart failure.111,112,141 The effect of phosphorylation on RyR is one of the most discussed topics in the field of E-C coupling.53,110,142   25 1.6.3 EF-Hand Proteins A primary ligand for triggering RyR opening is cytosolic Ca2+. However, as Ca2+ levels in the cytosol increase, RyRs close, suggesting that there are multiple Ca2+ binding sites with different affinities and kinetics. 23,143 A plot of channel activity (as observed through its Ca2+ efflux rate) versus cytosolic Ca2+ concentration shows a bell-shaped curve.144,145 (Figure 1.10)  Figure 1.10 The Ca2+ efflux rate constant from Ryanodine Receptors is dependent on external Ca2+ levels.  This bell-shaped curve shows the dependence of the Ca2+ efflux rate constant on Ca2+ concentrations. Low levels of Ca2+ promote channel activity with maximal rate constant of 1.5s-1 at 5µM. At high concentrations of external Ca2+, efflux is inhibited. This suggests the presence of high-affinity activating and low-infinity inhibitory Ca2+ binding sites. Adapted with permission from Meissner, Darling, Eveleth, 1986.145   Multiple calcium sensors can also bind RyR and modulate its Ca2+ sensitivity. These proteins contain EF-hand motifs that consist of a helix-loop-helix structure. The loop includes key acidic residues and backbone oxygens that mediate Ca2+ binding with pentagonal bipyramidal coordination.122,146,147(Figure 1.11)    26  Figure 1.11 A typical EF-hand  An EF-hand of CaM (PDB 1EXR) consists of a helix-loop-helix that coordinates Ca2+ (green). The ion is bound by a combination of acidic residues, backbone oxygens and waters in a pentagonal bipyramidal geometry.  In addition, pseudo and non-canonical EF-hand proteins also exist whereby their sequences vary significantly from the typical ‘EF-handome’.122,148 Usually present as a pair, EF-hands form structurally stable four-helix arrangements.149 EF-hands are one of the most common structural motifs found in proteins with up to 1000 known and their most common function is to bind protein substrates in a Ca2+-dependent manner.150 The first EF-hand crystal structure was described for parvalbulmin in the 1970s.151 Over the years further structures determined by X-ray crystallography and NMR have provided details on the dynamic nature of Ca2+ binding in these proteins and nuances that allow them to specifically target a protein sequence. 94,152-158 The following section will provide an overview of two EF-hand proteins before a discussion on EF-hand modulation of RyR.    27 Calmodulin Calmodulin (CaM) is the best known EF-hand containing protein. It is a eukaryotic protein that mediates many Ca2+-dependent processes of target enzymes such as myosin light chain kinase (MLCK), CaM-dependent kinases and phosphatase, nitric oxide synthase and Ca2+-ATPases.159-161  CaM is a small protein of 17 kDa and is composed of 4 EF-hands arranged as two pairs or lobes that can bind up to 4 Ca2+.162 CaM resembles a dumb-bell with its two lobes, commonly termed the N-lobe or C-lobe (N-terminus and C-terminus), connected by a highly flexible linker. The two lobes share 46% sequence identity and 75% sequence homology and their Cα backbone superimposes very well with an RMSD of 0.7Å.163,164 The differences are reflected in their calcium affinities where the C-lobe has a higher calcium affinity (Kd ~ 10-6M) than N-lobe (Kd ~ 10-5M).163,164 There is positive co-operativity between the lobes upon binding the first Ca2+ allowing for CaM to be responsive over a large range of cellular Ca2+ levels.122,165,166 When bound to many target proteins, the affinity of Ca2+ for CaM increases due to a significant decrease in the dissociation rate constant koff.167-169 CaM is essential as knockouts are lethal in model organisms such as yeast and Drosophila.170,171 Mutations in CaM are also deleterious causing recurrent cardiac arrest in infants and CPVT in patients. 57,172 CaM is abundant near membranes (50%) and the nucleus with a low free concentration in the cytosol (50-75nM).173-175 CaM can be tethered to some of its target proteins allowing it to function as a dedicated Ca2+-sensor.176,177 This allows for spatially-segregated, fast CaM-mediated processes as soon as cytosolic Ca2+ levels rise. Its highly conserved amino acid sequence (100%) between all vertebrates underlies its importance for normal cellular processes.178   28 The most important feature of CaM is its conformational plasticity and ability to bind to an abundant assortment of protein targets.163,179 In the absence of Ca2+, CaM exists in a compact conformation with the lobes closed and packed in towards the central linker.180 This limits its ability to bind and apoCaM indeed only associates with a select number of targets.181 Upon binding Ca2+, the lobes undergo a conformational change whereby they open and expose hydrophobic residues that form a concave pocket that is able to accommodate a wide array of substrates.147,166,182,183 (Figure 1.12) The binding of Ca2+ into the EF-hands causes significant alterations in the relative orientations of the helices in each lobe.184 The lobes rotate away from the central linker allowing CaM to adopt an open, extended state which permits a wide diversity of binding modes with high affinities (Kd = 10-7 – 10-11M).185    Figure 1.12 Apocalmodulin and Ca2+/Calmodulin The two forms of CaM exist in very distinct conformations. A. Apocalmodulin (PDB 1CFD) with no Ca2+ bound is compact and closed. Upon binding Ca2+ (green spheres) calmodulin undergoes a conformational change to become Ca2+/calmodulin (B; PDB 1EXR). Hydrophobic residues including numerous methionine residues (orange) are exposed upon this change allowing for target recognition and binding. It should be noted that NMR structures do not show the central linker as a continuous linker.186  29 The high number of methionine residues that form the hydrophobic pocket is uncommon in proteins and accounts for CaM’s promiscuity in target binding.179,187 (Figure 1.12) Sulfur is highly polarizable, more so than carbon, as it contains a larger electron cloud that allows it to interact with a variety of hydrophobic residues through van der Waals interactions but still being compatible with aqueous environments.187 In addition, the sidechain’s flexibility allows the hydrophobic pocket to be conformationally pliable. Charged residues lining the binding pocket stabilize and orient the target via electrostatic interactions. CaM Target Motifs   Classically, CaM interacts with short peptide segments between 16-30 amino acids in length that form amphipathic helices.163,185,188 Bulky hydrophobic residues (Trp, Tyr, Phe, Val, Ille) that provide anchor points for CaM’s hydrophobic clefts are interspersed with basic residues along a helix. It is thought that many of these helices are initially present as intrinsically disordered regions in the host proteins and fold upon binding to CaM.189 As more CaM binding domains (CaMBD) have been identified there have been great efforts to develop consensus sequences that could be predicted to bind CaM. These motifs can be assigned based upon their spacing of the hydrophobic residues ie. 1-10, 1-14, 1-16, which indicates the position of the hydrophobic anchors.163,188 Further subgroups such as 1-5-10 and 1-5-8-14 have been established as intervening hydrophobic residues have been identified to play a role in certain sequences. Table 1.1 provides examples of these motifs.      30 Table 1.1 Representative examples of CaM target motifs based upon their anchor spacing. Target Peptide Motif Spacinga PDB smMLCK 1-14 (1-8-14) 1CDL skMLCK 1-14 (1-5-8-14) 2BBM CaMKII 1-10 (1-5-10) 1CDM CaMKK 1-16 1CKK RyR1 1-17 2BCX Ca2+-ATPase 1-18 2KNE aOverall motif class with parentheses indicating the subgroup if applicable.   The so-called “IQ” motif has been fount to bind apoCaM. It corresponds to a (ILV)Qxxx(RK)Gxxx(RK) consensus sequence where x represents any amino acid.181 IQ motifs share similar features with Ca2+/CaM binding domains as they still contain all the common structural elements: a basic, amphipathic helix that contains several hydrophobic anchor points. Many IQ motifs are able to bind both apoCaM and Ca2+/CaM and some preferentially bind apoCaM. For example, the apoCaM binding to Nav1.2’s IQ domain reduces the C-lobe’s affinity for Ca2+ due to constraints imposed to maintain the apoCaM complex.190 CaM Binding Modes Characterization through fluorescence and small-angle X-ray scattering experiments of Ca2+/CaM and peptides indicated that these ternary complexes have a very compact structure.191-193 This suggested early on that the central linker was required to bend and kink to allow for the lobes to come into closer proximity to form a closed state. The first structures of CaM bound to peptides were of skeletal muscle and smooth muscle myosin light chain kinase (skMLCK and smMLCK) 194,195 and of CaM-dependent kinase IIα (CaMKIIα).196 In the Ca2+/CaM/CaMKIIα structure, the linker is unwound allowing the  31 lobes to rotate by 120° and bend by 100° relative to Ca2+/CaM. The peptide is an α helix with its bulky hydrophobic residues largely buried inside both Ca2+-bound lobes. This example typified most Ca2+/CaM structures but is no longer the case. (Figure 1.13) Structures of Ca2+/CaM complexes have now shown a multitude of binding modes.  CaM has been observed to be able to bridge two peptide segments197,198 as well as bind peptides in an extended open conformation155,156. A single target region is also capable binding multiple CaM molecules.161 CaM is also able of binding deep within a target protein and not an exposed segment.152 Finally, there are instances where binding with both Ca2+ and apo lobes occur simultaneously.158 The diversity of binding modes of CaM with its targets allows for a variety of “custom-tailored” regulatory mechanisms that are fast, highly specific, tightly regulated and dynamic.    32  Figure 1.13 Calmodulin binds to its targets in a variety of modes A. CaM bound to skeletal muscle myosin light chain kinase was the first complex determined (PDB 2BBM). B. CaM bound to CaMKII has a collapsed form onto its target similar to (A) (PDB 1CDM). C. CaM with no target bound. D. An unusual complex where 4 CaMs bridge two peptides that contain two CaM-binding regions within a voltage gated calcium channel (PDB 3OXQ). E. CaM binding a sodium channel peptide with only one lobe in an open conformation (PDB 4DJC). F. Two CaM molecules bound to a single peptide derived from a Ca2+-ATPase (PDB 4AQR). G. CaM activating adenylyl cyclase by binding deep within the protein (PDB 1K9O). H. CaM bridging two peptides from a Ca2+-activated potassium channel but with one lobe Ca2+ bound and one Ca2+ free (PDB 1G4Y). Ca2+ are green spheres.  33 S100A1 The S100 family of proteins are also EF-hand containing proteins that appeared late in evolution and are only expressed in vertebrates.199 They were originally isolated from bovine brain and discovered to be soluble in 100% ammonium sulfate, hence their name ‘S100’.200 Like CaM, S100 proteins contain EF-hands that are able to bind Ca2+ followed by binding to target proteins.201,202 S100A1 is the first member of this family and forms small homodimers of 21 kDa. Unlike CaM lobes, the first EF-hand of each subunit is considered a “pseudo” EF-hand due to its extremely weak Ca2+ binding affinity (Kd ~ 100-500µM) while the second EF-hand binds with a more typical affinity with Kd = 1-50µM.202 The overall apparent Ca2+ affinity is increased by glutathionylation and the presence of target proteins.203,204  S100A1 is highly abundant in skeletal muscle tissue at a concentration of 0.5 – 15µM, which is significantly higher than the concentration of CaM.205 In RyR1, S100A1 and CaM have been shown to compete for a common binding site, encoded by residues 3614-3643.121,204 S100A1 has a subtle effect on excitation-contraction coupling acting as an enhancer of the entire process.206 Knockouts of S100A1 have increased the incidence of heart failure and there is decreased S100A1 in cardiomyopathic hearts, suggesting S100A1 plays a critical role in muscle maintenance.207,208 In S100A1 knockout muscle fibres there are decreased Ca2+ transients upon electrical stimulation but this can be rescued by adenoviral S100A1 transfection.209 Due to the ability of S100A1 to enhance cardiac contractility, it has been suggested as a possible target for treatment of heart failure through gene therapy. 210     34 1.6.4 Calmodulin and Ryanodine Receptors The first isolated preparations of RyR identified CaM as an important protein that was co-purified.132 Further biochemical characterizations showed that CaM binds directly to RyR and modulates the channel both in vivo and in vitro.211-214 As observed by cryo-EM, RyR is able to bind up to 4 CaMs (1/subunit), but it is not known how many are required to modulate the channel.80 Its modulatory effect depends on the free Ca2+ concentration and the RyR isoform. At high Ca2+ levels, CaM can inhibit both RyR1 and RyR2.215 At low Ca2+ levels it activates RyR1 but inhibits RyR2212,216-218, although the inhibition at low Ca2+ concentrations has been controversial.219 In single channel recordings, the presence of 50 nM CaM at low Ca2+ levels (0.4µM) is sufficient to see an effect.215 In RyR1, the open probability (Po) is doubled while in RyR2 it is almost halved. At higher Ca2+ levels (2µM), the Po decreases for both isoforms. Recent experiments using RyR2 have shown that mutations to the N-lobe of CaM increase the threshold for termination of Ca2+ release, whereas mutations on the C-lobe decrease the threshold220, suggesting individual roles and possibly distinct binding sites for the lobes on the full-length channel.  Interestingly, mutations in CaM have also been associated with CPVT57, indicating that a proper regulation of RyR2 by CaM is absolutely required. The mechanism of CaM modulation of RyR is currently unknown. Many sequences on the RyR have been found to bind CaM but not all have been thoroughly investigated. 19,221,222 A crystal structure of CaM in complex with a RyR peptide (3614-3643; mouse RyR1) shows both lobes of CaM anchored. However, NMR data suggests that the Ca2+/N-lobe is only weakly associated and could bind to a different site.94      35 Binding Sites of Calmodulin on Ryanodine Receptors When the full-length sequence of RyR was first determined, analysis revealed the presence of up to 9 different CaMBDs that consisted of a typical sequence (as discussed above) or a section that was predicted to be an amphipatic alpha helix.19,30,221,222 Several labs independently screened fusion protein fragments of RyR to see which can interact with CaM through pull-down experiments; however, results have been conflicting due to identification of different regions. One region that has been the topic of intense research is encoded by RyR1 residues 3614-3643 (3583-3603 in mouse RyR2).223-230 It has been identified as both an apoCaM and a Ca2+/CaM binding site.223 Mutations in the site in RyR2 reduced the total amount of CaM bound on RyR and relieved CaM-dependent effects on the channel such as inhibition at high Ca2+.230 This suggested that a single site was all that was necessary for CaM modulation of RyR. However, questions that are unresolved are: 1) how can a single site be responsible for the separate effects of ApoCaM and Ca2+/CaM? 2) How can both CaM lobes have different functional effects? 3) How can apoCaM exert different effects on RyR1 and RyR2 when the proposed single binding site has a near-identical sequence in both isoforms?  Two views have been proposed by several groups. In one view, the 3614-3643 sequence is the site of action of CaM and all subsequent effects are due to downstream changes.215,229,231 Another view is that a second CaMBD site that is dissimilar between the two isoforms allows for site-specific modulation of the channel.94,225,226 The mechanism of how CaM exerts differential effects on the two isoforms is still unknown. In addition, the RyR is regulated by S100A1 and other EF-hand proteins. There is high redundancy between CaMBDs with other EF-hand proteins suggesting that multiple players may compete for the same sites and play a role in RyR regulation. Unlike CaM,  36 S100A1 acts as an agonist of RyR activity at higher Ca2+ concentrations.95,199,204 The relative amounts of CaM and S100A1 thus may dictate channel activity when amounts of either protein are limiting. Finally, it has been suggested the EF-hand domain of RyRs is able to mimic Ca2+/CaM and bind to the region 3614-3643 serving as a clamp to inhibit channel opening.123,228,232 Structural Biology of CaMBDs The most studied CaMBD of RyR is the 3614-3643 region. An X-ray crystal structure of the region in complex with Ca2+/CaM reveals an arrangement similar to other ternary complexes with a few novelties.94 (Figure 1.14) Two aromatic anchors associate with the CaM lobes: W3620 for the C-lobe, and F3636 for the N-lobe.  They have an unusual 1-17 spacing that had not been previously observed in other CaM:target complexes. 1H-15N RDCs (residual dipolar coupling) revealed that although the conformation of each lobe bound on the peptide is comparable between the crystal structure and in solution, the ordering and motions differ for the lobes in solution. This contrasts with other CaM complexes where solution NMR RDCs provided excellent fits with x-ray data.233,234 As such, it was interpreted that both lobes can bind their anchors independent from one another. With a shortened RyR1 peptide, 3614-3634, CaM is still able to bind with its C-lobe even with one anchor missing and without collapsing onto the peptide. The N-lobe is free to interact elsewhere.94  This opens up the possibility of state-dependent CaM regulation when the N-lobe anchor is conformationally inaccessible.   37   Figure 1.14 Ca2+/CaM bound to amino acids 3614-3643 from RyR1. The structure of Ca2+/CaM bound to the peptide derived from amino acids 3614-3643 (PDB 2BCX) shows a closed compact conformation similar to other CaM complexes but with an atypical 1-17 spacing between the anchor residues Trp3620 and Phe3636. The peptide is bound antiparallel fashion. Ca2+ are green spheres.  The 3616-3627 segment in complex with S100A1 has also been determined by NMR.204 (Figure 1.15) In contrast with other S100A1 structures, the Ca2+/S100A1 is arranged antiparallel to the peptide with respect to its 3rd helix. It engages the same aromatic anchor, W3620, as the Ca2+/C-lobe, suggesting a direct competition with CaM. Binding experiments suggest that both proteins are competitive at similar concentrations allowing for fine-tuning of RyR activity.204   38  Figure 1.15 Ca2+/S100A1 bound to amino acids 3616-3627 from RyR1. S100A1 also binds the same region as CaM on the RyR. Each half of the homodimer binds a short section of RyR (PDB 2K2F). The interaction is mediated by one of the anchor points that also binds CaM, Trp3620. Ca2+ are green spheres.  In addition to atomic models the locations of CaM have been localized on RyR1 and RyR2 by cryo-EM. ApoCaM and Ca2+/CaM occupy overlapping regions on the surface of RyR1. There is a 30Å shift between the centre of mass of apoCaM to the Ca2+/CaM position, suggesting that both states of CaM bind to different sites.70 (Figure 1.16) However, one issue with this study is that there may have been intrinsic effects of Ca2+ on RyR conformation. Follow-up experiments using CaM1234, which is Ca2+-binding deficient, at both low and high Ca2+ levels show no change in the position of apoCaM nor the overall structure of RyR.235 This suggests that the 30Å shift is likely due to CaM movement and not conformational changes of RyRs. Interestingly in RyR2, the apoCaM site coincides with the Ca2+/CaM site in RyR1. The locations have yet to be correlated with the primary structure of RyR or specific functional effects on the channel.  39 GFP/YFP-insertions in RyR2 have attempted to map primary structure on to cryo-EM maps. Two insertions have been made at potential CaMBDs: 1) near R3595, which corresponds to the 3614-3643 region in RyR1 and 2) K4269.236 (Figure 1.15) Both insertions localized near the CaM densities that were determined before, and the study suggest that CaM might be able to bridge the two segments. FRET measurements show that these two sites can undergo relative conformational changes upon channel opening, further suggesting that a bridging CaM may play a role in the transition between open and closed channels.   Figure 1.16 Localization of CaM and potential CaMBD by difference Cryo-EM densities. A. Difference density between RyR1 and RyR1 with apoCaM or Ca2+/CaM reveal two distinct but overlapping locations for CaM. The centre of mass moves 30Å suggesting a large shift in the location of CaM upon binding Ca2+. B. GFP insertions at residues 3595 and 4269, near two putative CaMBD, show difference density near the location of CaM in A. The close proximity suggest that these two sites may be bridged by CaM. Adapted with permission from Samsó, M. and Wagenknect, T. (2002) and from Huang et al. (2013).70,236  1.6.5 Skeletal Muscle DHPR α1S Subunit Excitation-contraction coupling is the process that allows a depolarizing electrical signal to be converted into a chemical signal, allowing for muscle contraction. In cardiac  40 muscle, entry of Ca2+ by activation of DHPR on T-tubules is sufficient to evoke Ca2+ release from RyR2. In skeletal muscle, however, the process is Ca2+ independent as its entry is not required for RyR opening and Ca2+ release from the SR.6 It is thought that instead, conformational changes are evoked in the skeletal muscle DHPR α1S subunit upon depolarization, which directly activates RyR1 through mechanical coupling.11,237,238 This forward signal is termed 'orthograde', but it has also been found that 'retrograde' signals exist, whereby the state of RyR1 can affect the function of the DHPR.239 The different mechanism in E-C coupling between the two muscle types also translates to a physiological difference. In skeletal muscle, DHPRs form a group of four particles or “tetrads” which come into the contact with the corners of RyR. In cardiac muscle the DHPRs are arranged variably.65,66,240,241   The regions of the skeletal muscle DHPR and of RyR1 that mediate this process have been investigated thoroughly but there is still little evidence to suggest that the two proteins are in direct physical contact. In the DHPR, a chimeric and biochemical approach has been used to identify cytosolic loops that may come into contact with RyR. These experiments have consistently shown that a single loop between domains II and II (II-III loop, 720-765) of the DHPR is required for E-C coupling to occur.242-247 Conversely, chimeric RyR1/RyR2 constructs expressed in dyspedic (lacking RyR1) myotubes indicate a wide variety of regions that may mediate signalling with the DHPR.248-250 Structurally, multiple regions of RyR1 also have been implicated for tetrad formation in skeletal muscle but not cardiac muscle.251 However, further analysis of these regions through both in vivo and in vitro methods have provided conflicting results, suggesting that the machinery mediating E-C coupling is complex and may involve multiple sites of contact or low affinity interactions.248  41  1.6.6 β Subunit The skeletal muscle DHPR is composed of multiple subunits, with the α1S forming the transmembrane subunit. The β subunit is an intracellular component and has been found to play a vital role in E-C coupling.252 β1A null mice die perinatally due to asphyxia253 β subunits interact with the I-II loop of the α1S and are known to promote membrane insertion, trafficking and proper DHPR activity.254 In addition, they are essential for the formation of tetrads that are necessary for skeletal muscle E-C coupling.252 However, direct interactions between this subunit and RyR1 have never been reported and currently it is thought that β subunits indirectly influence α1S -RyR1 interactions by stabilizing and tethering the DHPR in the membrane.  1.7 Research Question The Ryanodine Receptor is a large and complex channel that is modulated by a wide assortment of proteins and ligands. Despite intense interest in certain aspects of RyR regulation, there are still many unanswered questions. Much of the underlying uncertainties are due to the lack of both structural and biochemical characterizations of the channel itself. The overall aim of this thesis is to shed light on two aspects of RyR regulation and structure.  The second chapter of the thesis focuses on the binding of CaM to RyR peptides that have been identified as potential effector sites on the channel. Although many sites have been reported in literature, there are still outstanding questions. (1) Which of these sites bind apoCaM and Ca2+/CaM? (2) Are there differences in binding between homologous regions between the isoforms that may mediate isoform-specific functions? (3) Is CaM able to bind to CaMBDs that harbour disease-causing mutations?   42 In the third chapter, the focus is on a domain that has been implicated to mediate excitation-contraction coupling in RyR. Using X-ray crystallography and cryo-EM docking, the structure and its location within the full-length channel are revealed. Disease mutants were also characterized and crystallized. Finally, calorimetric binding experiments were used to test the possibility of interactions with other proteins. My work has required the extensive use of calmodulin and isothermal titration calorimetry. I have been involved in several projects that are not covered within the scope of this thesis. A collaboration with the lab of Dr. Mate Erdélyi was initiated to probe the role of calmodulin binding to volatile anaesthetics. This was initiated to test the hypothesis that CaM could be the target of volatile anaesthetics, which prevents CaM from regulating target proteins such as RyR. This work is also published (Brath U., Lau, K., Van Petegem, F., Erdélyi. M. Mapping the sevoflurane-binding sites of calmodulin. (2014) Pharmacology Research and Perspectives. 2, e00025. Doi:10.10002/prp2.25).  I tested whether FKBP12 (Chapter 1.6.1) could bind to the RyR1 N-terminal region by calorimetry. This work was recently published as well (Kimlicka, L., Lau, K., Ching-Chieh, T., Van Petegem, F. Disease mutations in the ryanodine receptor N-terminal disease hot spot couple to a mobile intersubunit interface. (2013) Nature Communications. 4, Article 1506). Together, the work presented represents an analysis of regions of RyR that have been implicated in modulating the channel. It provides much needed basic biochemical and structural information in regards to two processes that have been widely discussed and are controversial in the field of RyR regulation.     43          Chapter 2 - Lobe-Specific Calmodulin Binding to Different Ryanodine Receptor Isoforms          This chapter has been adapted from the original publication: Lau, K., Chan, M.M.Y, Van Petegem, F. Lobe-Specific Calmodulin Binding to Different Ryanodine Isoforms. Biochemistry. (2014) 53(5) 932-946  44 2.1 Abstract Ryanodine Receptors (RyRs) are large ion channels that are responsible for the release of Ca2+ from the sarco/endoplasmic reticulum. Calmodulin (CaM) is a Ca2+ binding protein that can affect the channel open probability under conditions of both high and low Ca2+, shifting the Ca2+ dependencies of channel opening in an isoform-specific manner. Here we analyze the binding of CaM and its individual domains to three different RyR regions using isothermal titration calorimetry. We compared binding to skeletal muscle (RyR1) and cardiac (RyR2) isoforms, under both Ca2+-loaded and Ca2+-free conditions.  CaM can bind to all three regions in both isoforms, but the binding modes differ appreciably in two segments.  The results highlight a Ca2+/CaM and apoCaM binding site in the C-terminal 1/5th of the channel. This binding site is the target for malignant hyperthermia and central core disease mutations in RyR1, which affect the energetics and mode of CaM binding.               45 2.2 Introduction The primary ligand to trigger RyR opening is cytosolic Ca2+.  Under these circumstances, the RyR acts as a signal amplifier, in a process known as Ca2+ induced Ca2+ release (CICR).14,255  However, as Ca2+ levels in the cytosol rise, RyRs close, indicating that there are multiple Ca2+ binding sites with different affinities and binding kinetics. A plot of the open probability of the channel as a function of Ca2+ concentration therefore displays a bell-shaped curve.145 In addition, luminal Ca2+ levels can regulate RyR opening through the process of store overload induced calcium release (SOICR).256,257 The exact timing of channel opening and closing is crucial. Several genetic diseases result from altered RyR activity, often increasing their sensitivity to cytosolic or luminal Ca2+ and leading to channels with enhanced open probabilities. RyR1 mutations are mostly associated with malignant hyperthermia (MH) 258 and central core disease (CCD)259, whereas RyR2 mutations can cause catecholaminergic polymorphic ventricular tachycardia (CPVT).260 These mutations are found throughout the RyR genes, but many have been found to cluster at domain-domain interfaces.90-93,96,97,102,106 Multiple auxiliary proteins can further tune the sensitivity of RyRs to Ca2+.   A major player is Calmodulin (CaM), a 17-kDa protein with 4 EF-hands able to bind Ca2+.  CaM directly binds to RyRs, but the effect is dependent on Ca2+ concentrations and the exact RyR isoform. At high Ca2+ levels, CaM can inhibit both RyR1 and RyR2. At low Ca2+ levels it activates RyR1 but inhibits RyR2212,216-218, although the inhibition at low Ca2+ concentrations has been controversial.219 Recent experiments using RyR2 have shown that mutations to the N-lobe of CaM increase the threshold for termination of Ca2+ release, whereas mutations on the C-lobe decrease the threshold220, suggesting individual roles and possibly distinct binding sites for the lobes on the full-length channel.  Interestingly,  46 mutations in CaM have also been associated with CPVT57,172, indicating that a proper regulation of RyR2 by CaM is absolutely required.   How exactly CaM exerts its stimulatory or inhibitory effects is currently unknown. Previously, a number of RyR peptides have been suggested to bind CaM based mainly on pull-downs, overlay assays, gel shifts, and fluorescence measurements.221,222,224-226,229,261-263 However, most of these have not been validated using a quantitative method, and multiple ones have been discarded because they have been shown to be inaccessible93 or do not agree with cryo-EM data, as recently discussed by Huang et al.236  Thus far, only three RyR segments remain as candidates to be ‘true’ CaM binding sites.  CaM has often been shown to use its individual lobes to bridge different peptide stretches in its target proteins, so it is likely that more than one of these segments is involved in binding either Ca2+/CaM or apoCaM.  A crystal structure has shown that Ca2+/CaM is able to bind to a peptide in the central region of RyR1 (residues 3614 to 3643 in rabbit RyR1), with both CaM lobes binding the segment simultaneously.94 However, NMR data have shown that the Ca2+/N-lobe is only loosely associated with the peptide, suggesting it is likely to bind elsewhere in full-length channels.94 Cryo-EM studies have shown that only one CaM binds per RyR monomer70, although this does not exclude the possibility that additional CaMs may bind to separate, low affinity sites. The cryo-EM studies also show that the binding site for CaM on RyR1 shifts by ~30Å.70,235 This shift is too large to be attributed to a different binding mode onto the same peptide, and suggests that one or both CaM lobes shift to another sequence stretch upon binding Ca2+. Curiously, these cryo-EM studies have also shown that the apoCaM binding site in RyR2 coincides with the Ca2+/CaM binding site in RyR1. Together with the  47 different functional effects of CaM in the two isoforms, this indicates that the mode of binding in the two isoforms is different. Because of the complexities by which CaM can bind its targets, and the inherent ability of the individual lobes to bridge different sequence stretches in other proteins, we used isothermal titration calorimetry (ITC) to analyze the binding of CaM and its individual lobes under both apo and Ca2+-loaded conditions. We focused on the three remaining CaM binding candidates (CaM binding domains – CaMBDs 1-3; Figure 2.1) in both RyR1 and RyR2. The data show isoform-specific differences and suggest that the highest affinity for apoCaM resides in the C-terminal region of the RyR. The latter is also the target for disease mutations that affect CaM binding.   Figure 2.1 Sequences of three CaM binding domains (CaMBDs).  Residues that are strictly conserved among RyR1−RyR3 are highlighted in gray. Purple residues are the sites of mutation in RyR1CBD3, duplication of L4319−4321 and R4325D. The numbering is for rabbit RyR1, mouse RyR2, and human RyR3.    CaMBD1CaMBD2CaMBD3ID% = 64 (RyR1 v. RyR2)ID% = 89ID% = 26 •1975                         •1985                         •1985         199919651867 •3614                         •3624                         •3634      364036063495RyR1RyR2RyR3 •4295                         •4305                         •4315   4325427641785RyR1RyR2RyR3RyR1RyR2RyR3SRYALLMRAFTMSAAETARRTREFRFRYNEVMQALNMSAALTARKTREFRFRYNELMQALNMSAALTARKTREFRKSKKAVWHKLLSKQRRRAVVACFRMTPRSKKAVWHKLLSKQRKRAVVACFRMAPRSKKAVWHKLLSKQRKRAVVACFRMAPRLAAAAARALRGLSYRSLRRRVRRLRRLTARALRYNVLTLVRMLSLKSLKKQMKRMKKMTVKSVKRNVTDFLKRATLKNLRKQYRNVKKMTAK 48 2.3 Methods and Materials 2.3.1 Cloning and Purification of CaMBDs Sequences encompassing rabbit RyR1 (CaMBD1: residues 1975-1999; CaMBD2: 3614-3640) and rat RyR2 (CaMBD1: 1941-1965; CaMBD2: 3580-3606; CaMBD3: 4246-4276) were cloned in a pET24a vector containing a C-terminal hexahistidine tag using NdeI and XhoI cut sites. At the N-terminus, maltose binding protein (MBP) was attached, along with a cleavage site for tobacco etch virus (TEV) protease. The presence of affinity tags on both sides allows us to extract only peptides without degradation within the CaMBD sequence. RyR1 CaMBD3 (4295-4325) could not be cloned using routine PCR and instead synthetic peptides (Lifetein) were used for wild type and disease mutant versions. In order to protect the fusion proteins from proteolytic degradation, they were co-expressed with human CaM, which binds and prevents access to proteases. The CaM was co-expressed in a peGST vector264, using Escherichia coli Rosetta (DE3) pLacI at 37°C, induced at an OD600 of ~0.6 by addition of 0.2 mM IPTG for 4 hours.  Cells were lysed by sonication in buffer A (250 mM KCl, 10 mM CaCl2 and 10 mM HEPES, pH 7.4) supplemented with 25 mg ml-1 DNase I, 25 mg ml-1 lysozyme, 1 mM iodoacetamide and 1 mM phenylmethylsulphonyl fluoride.   The lysate was applied to a 25 ml Poros MC column (Tosoh Bioscience), washed with 5CV of buffer A and 5CV of 2% (vol/vol) buffer B (250 mM KCl, 10 mM CaCl2 and 500 mM imidazole, pH 7.4) and eluted with 30% (vol/vol) buffer B. The protein was diluted two-fold with buffer A and then applied to a 25 ml amylose column (New England Biolabs). The protein was then washed with 20CV of buffer A plus 10 mM EDTA, 2CV of buffer A and eluted with buffer C (buffer A plus 10 mM maltose). The wash step with EDTA completely removed the co-expressed CaM. The protein was diluted two-fold with buffer A and applied once again to a 25 ml Poros MC column, washed and eluted as before.  For the  49 CaMBD2 constructs, which contain cysteines, 10 mM β-mercaptoethanol (BME) was present in all buffers except for ones used for the Poros MC column.   The fusion proteins were confirmed to be monomeric using size exclusion chromatography on a Superdex200 column (GE Healthcare) (Figure 2.2). The molecular weights were confirmed by MALDI-TOF on a Voyager-DE STR (Applied Biosystems) as fusion proteins and as peptides after cleavage with TEV protease.   50  Figure 2.2 Size exclusion chromatograms and SDS-PAGE gels of purified and synthesized CaMBDs.  A,B. Size exclusion chromatograms of purified CaMBDs on a Superdex200 at 150 mM KCl, 10 mM HEPES 7.4, 0.1% NaN3 and (A) 10 mM CaCl2 or (B) 10 mM EDTA. C. 15% SDS-PAGE gel of purified CaMBDs. D,E. Size exclusion chromatograms of synthetic CaMBDs on a Superdex75 at 150 mM KCl, 10 mM HEPES 7.4, 0.1% NaN3 and (D) 10 mM CaCl2 or (E) 10 mM EDTA.   51 2.3.2 Cloning and Purification of Calmodulin and its Lobes Full-length Human CaM, its N-terminal lobe (N-lobe, residues 1-78) and C-terminal lobe (C-lobe, residues 79-149) were cloned as described265 into a modified pET28 vector containing an N-terminal hexahistidine tag and a tobacco etch virus (TEV) protease cleavage site. Purification was like for the CaMBD constructs, with the following exception. After the first Poros MC column, the protein was cleaved with his-tagged TEV protease overnight at room temperature while dialyzed against buffer A. The tagged protease and cleavage product were removed with an additional PorosMC column.  The flow through was applied to a Phenyl-Sepharose HP column (GE Healthcare) equilibrated with 150 mM KCl, 20 mM HEPES, pH 7.4, 10 mM CaCl2.  The protein was eluted with the same buffer containing 10mM EDTA instead of CaCl2, and was applied to a HiLoad Q-Sepharose HP column equilibrated with 20 mM HEPES pH 7.4, 10 mM EDTA, and eluted with a gradient of 20% to 40% of buffer containing an additional 1M of KCl over 14 CV.  Molecular weights for all proteins were confirmed by MALDI-TOF on a Voyager-DE STR (Applied Biosystems).  2.3.3 Isothermal Titration Calorimetry The purified MBP-CaMBD fusion proteins were dialyzed against 150 mM KCl, 10 mM HEPES, pH 7.4, 10 mM 2-mercaptoethanol, 10 mM CaCl2 or 10 mM EDTA at 4°C.  Peptides that were synthesized and delivered as lyophilized powder (Lifetein) were dissolved directly in dialysis buffer. Peptide oligomeric state was confirmed to be monomeric on a Superdex75 column (GE Healthcare) (SI Figure 1). Concentrations were determined using the calculated extinction coefficient at 280nm in the presence of 6M Guanidine.266 Titrations consisted of 20 injections of 2µL titrant (CaM or lobes) into the cell containing a CaMBD at a 10-fold lower concentration.  Typical concentrations for the titrant were between 100- 52 500µM for experiments in 10 mM CaCl2 and between 700-2000µM for experiments in 10 mM EDTA, depending on the affinity.   Experiments were performed at 25°C and a stirring speed of 1000 rpm on an ITC200 instrument (GE Healthcare).  Control experiments titrating CaM or its lobes (2000µM) into the hexahistidine-MBP without CaMBD2 (200µM)  in 10 mM CaCl2 or 10 mM EDTA showed heats that were indistinguishable from buffer injections (data not shown). In addition, control experiments titrating N-lobe (2000µM) into C-lobe (200µM)  in the presence of 10 mM CaCl2 or 10 mM EDTA show heats that are indistinguishable from buffer injections (data not shown). The data were processed using Origin 7.0 and fit to a single- or two-site fitting model after background buffer subtraction.    2.4 Results 2.4.1  Binding to CaMBD2 and its Mutants A previous crystal structure has reported a complex between Ca2+/CaM and a peptide in the central region of RyR1 (aa 3614 to 3643).94 We termed this ‘CaMBD2’ simply to reflect its position in the sequence relative to other CaM binding domains investigated in this study (Figure 2.1). We utilized ITC to explore the thermodynamic parameters that underlie the binding.  Because physiological concentrations of Ca2+ would yield mixtures of Ca2+-free and Ca2+-occupied lobes, the experiments were performed in the presence of excess Ca2+ (10 mM) to saturate all Ca2+ binding sites and to isolate binding of a fully Ca2+-occupied CaM.  The interaction of Ca2+/CaM is strong, with a Kd of 46 nM (Figure 2.3, Table 2.1).  The binding is driven by enthalpy and has an unfavorable entropic contribution.  In order to determine the contribution of each CaM lobe, we also investigated their binding individually.  Ca2+/C-lobe binds much stronger (Kd ~800 nM) than the Ca2+/N-lobe (Kd  53 ~33µM).  The Ca2+/C-lobe is driven by both favorable entropy and enthalpy while Ca2+/N-lobe binding is driven by enthalpy alone.   Table 2.1 Thermodynamic Parameters of Binding between Ca2+/CaM and its Lobes and CaMBDs in 10 mM CaCl2a      y g 2construct lobe no. of sites (N) Kd (μM) ΔH (kcal/mol) ΔS (cal mol−1 deg−1) nRYR1CaMBD1 CaM 0.81 ± 0.22 2.05 ± 0.96 −9.36 ± 1.25 −5.21 ± 3.6 4N 0.79 ± 0.33 5.95 ± 2.07 −8.81 ± 1.82 −5.58 ± 5.4 2C 0.88 ± 0.35 7.88 ± 0.74 −1.49 ± 0.04 18.4 ± 0.4 2CaMBD2 CaM 0.75 ± 0.14 0.046 ± 0.032 −16.8 ± 2.9 −18.3 ± 6.3 4N 0.87 ± 0.11 33.3 ± 10.7 −10.7 ± 2.1 −15.4 ± 7.7 3C 0.80 ± 0.03 0.80 ± 0.29 −5.94 ± 0.57 8.09 ± 1.83 4CaMBD3 CaM 1.02 ± 0.19 0.052 ± 0.022 −5.00 ± 1.19 16.7 ± 3.5 6N 1.07 ± 0.08 4.19 ± 1.44 −4.18 ± 2.27 10.7 ± 8.0 6C 1.16 ± 0.27 2.62 ± 1.22 −2.77 ± 1.95 16.5 ± 7.1 6RYR2CaMBD1 CaM 0.71 ± 0.04 7.59 ± 1.05 −6.09 ± 0.14 3.01 ± 0.76 2N 0.75 ± 0.11 26.7 ± 8.7 −5.52 ± 0.74 2.47 ± 2.66 2C binding detected but not quantifiable 4CaMBD2 CaM 0.77 ± 0.13 0.047 ± 0.017 −17.2 ± 1.1 −24 ± 2.8 3N 1.13 ± 0.28 18.9 ± 7.9 −7.25 ± 0.47 −2.44 ± 4.13 4C 1.16 ± 0.41 1.03 ± 0.75 −5.09 ± 0.18 10.6 ± 6.1 3CaMBD3 CaM complex binding 4N, Kd1 1.06 ± 0.11 0.052 ± 0.015 −1.28 ± 0.16 29.1 ± 0.9 4N, Kd2 1.14 ± 0.08 4.6 ± 0.4 3.79 ± 0.18 37.2 ± 0.5 4C 1.04 ± 0.01 3.09 ± 1.12 1.99 ± 0.18 31.7 ± 0.5 4aValues are averages of n measurements. Errors are standard deviations. 54  Figure 2.3 Ca2+/CaM and its lobes bind to all three RyR1 CaMBDs.  A-I. ITC binding isotherms show the interaction between CaM or its lobes titrated into each CaMBD in the presence of 10 mM CaCl2. The columns indicate the titrant, the rows show the CaMBD in the cell.  Solid lines represent the fit. Concentrations used are (titrant into cell) A: 500µM in 50µM, B: 1330µM in 133µM, C: 1330µM in 133µM, D: 150µM in 15µM, E: 1000µM in 80µM, F: 800µM in 80µM, G: 223µM in 22.3µM, H:  490.9µM in 31.5µM, G: 315µM in 31.5µM. Affinity and thermodynamic parameters are shown in Table 1.  RyR1 N CCaMBD2CaMBD3N CCaMBD1A B CD E FG H I 55 As it is possible that both lobes compete for an overlapping binding site, we performed competition experiments with the Ca2+/C-lobe pre-mixed with RyR1 CaMBD2 (Figure 2.4, Table 2.2). These show that the affinity of CaMBD2 for the Ca2+/N-lobe, once the Ca2+/C-lobe has been bound, is even lower (Kd ~60µM). The ability of the weaker Ca2+/N-lobe to still bind in the presence of excess Ca2+/C-lobe indicates that the binding sites are not mutually exclusive, but the weaker apparent affinity shows there are some unfavourable interactions when both lobes bind.  Table 2.2 Thermodynamic Parameters of Binding between Ca2+/CaM and its Lobes in Competition Experiments and CaMBDs in 10 mM CaCl2a           construct lobe no. of sites (N) Kd (μM) ΔH (kcal/mol) ΔS (cal mol−1 deg−1) nRYR1CaMBD1 N > C 0.71 ± 0.18 22.4 ± 6.8 −6.44 ± 0.17 −0.27 ± 0.05 2C > N heats indistinguishable from background 1CaMBD2 N > C 0.67 ± 0.05 60.7 ± 27.2 −6.40 ± 1.94 −2.04 ± 7.31 3C > N 0.44 ± 0.02 8.00 ± 1.35 −1.08 ± 0.07 19.7 1RYR2CaMBD3 N > C 0.51 ± 0.20 5.86 ± 4.54 −2.50 ± 0.80 16.0 ± 4.3 3C > N heats indistinguishable from background 1aValues are averages of n measurements. Errors are standard deviations (for n = 1, errors are deviations of the fit from data). 56  Figure 2.4 Competition between N-lobe and C-lobes of CaM in all 3 CaMBDs.  A-F. ITC binding isotherms show the interaction between either N-lobe or C-lobe titrated into CaMBD with pre-incubated C-lobe or N-lobe, respectively, in the presence of 10 mM CaCl2. The columns indicate the experiment, the rows show the CaMBD in the cell. Solid lines represent the fit. Concentrations used are (titrant in cell with pre-incubated lobe) A: 1330µM in 133µM with 222µM, B: 1330µM in 133µM with 222µM, C: 1000µM in 80µM with 133µM, D: 800µM in 80µM with 167µM, E: 1000µM in 100µM with 167µM, F: 661µM in 63µM with 185µM. Affinity and thermodynamic parameters are shown in Table 2.2. RyR1CaMBD1CaMBD2CaMBD3A BC DE FRyR1RyR2N CCaMBDCCaMBDNC 57 As Ca2+/CaM has different effects on RyR1 and RyR2, we wondered whether there are any differences in the binding affinities or energetic signatures of either lobe between both isoforms.  However, the corresponding peptide in RyR2 binds with a very similar affinity, enthalpy and entropy, and as for RyR1, the Ca2+/C-lobe forms the major interaction (Fig. 2.5, Table 2.1). Thus, in agreement with the high amount of sequence conservation among the different RyR isoforms, the overall binding mode on CaMBD2 is likely to be very similar.   58  Figure 2.5 Ca2+/CaM and its lobes bind to all three RyR2 CaMBDs.  A-I. ITC binding isotherms show the interaction between CaM or its lobes titrated into each CaMBD in the presence of 10 mM CaCl2. The columns indicate the titrant, the rows show the CaMBD in the cell. Solid lines represent the fit. Concentrations used are (titrant into cell) A: 1100µM in 110µM, B: 1132µM in 110µM, C: 1132µM in 110µM, D: 150µM in 15µM, E: 788µM in 79µM, F: 674µM in 67µM, G: 630µM in 63µM, H:  1109µM in 63µM, G: 661µM in 63µM. Affinity and thermodynamic parameters are shown in Table 2.1. Binding isotherm in Panel C could not be accurately fit due to the low affinity. RyR2 N CCaMBD1CaMBD2CaMBD3N CA B CD E FG H I 59 In the crystal structure of the RyR1 CaMBD2 complex94,  a Trp residue forms the main anchor point for the Ca2+/C-lobe, whereas a Phe residue provides the major contacts with the Ca2+/N-lobe.  We used individual lobes with mutant CaMBD2s to further dissect this interaction. In agreement with this, the W3586A mutation in RyR2 completely abolishes Ca2+/C-lobe binding to CaMBD2, while the F3602A mutation has no effect (Figure 2.6, Table 2.3).  However, both W3586A and F3602A still allow for Ca2+/N-lobe binding, but with different affinities and enthalpic components, suggesting that the Ca2+/N-lobe can bind to two different sites.  When one Ca2+/N-lobe site is abolished, it can still bind another, a type of behavior that has been observed in other CaM targets.265,267,268 These data are in agreement with previous studies showing that neither the equivalent W3586A or F3602A mutations individually can knock out Ca2+/CaM binding in full-length RyR2.229 Overall, these data suggest that CaMBD2 primarily forms a high affinity Ca2+/C-lobe binding site, whereas the binding site for Ca2+/N-lobe, as observed in the crystal structure, is weak and mobile.  Table 2.3 Thermodynamic Parameters of Binding between Ca2+/CaM and its Lobes in Mutant CaMBDs in 10 mM CaCl2a   construct lobe no. of sites (N) Kd (μM) ΔH (kcal/mol) ΔS (cal mol−1 deg−1) nRYR2CaMBD2W3586A CaM 0.91 ± 0.16 1.01 ± 0.43 −11.9 ± 2.6 −12.3 ± 9.3 4N 0.69 ± 0.39 18.9 ± 7.6 −12.8 ± 0.7 −21.3 ± 2.0 3C heats indistinguishable from background 4CaMBD2F3602A CaM 1.10 ± 0.02 0.32 ± 0.09 −12.7 ± 0.4 −13.0 1N 0.96 ± 0.02 6.86 ± 1.43 −1.86 ± 0.02 17.4 ± 1.8 2C 0.78 ± 0.50 0.44 ± 0.29 −7.66 ± 0.09 3.8 ± 3.6 3RYR1CaMBD3LRRdup CaM 0.79 ± 0.14 0.014 ± 0.008 −10.3 ± 1.4 1.6 ± 3.9 3N 0.82 ± 0.05 2.35 ± 0.67 −7.37 ± 0.31 1.1 ± 1.6 2C 0.88 ± 0.25 1.21 ± 0.16 −8.21 ± 2.18 −0.5 ± 7 3CaMBD3R4325D CaM 0.64 ± 0.02 0.12 ± 0.005 −9.03 ± 0.06 1.4 ± 0.2 3N 0.83 ± 0.03 5.11 ± 0.46 −6.35 ± 0.17 2.9 ± 0.7 3C 1.01 ± 0.09 1.54 ± 0.11 −6.04 ± 0.03 6.4 ± 0.12 3aValues are averages of n measurements. Errors are standard deviations (for n = 1, errors are deviations of the fit from data). 60  Figure 2.6 RyR2 CaMBD2 W3586A and F3602A mutants alter lobe binding.  A-F. ITC binding isotherms show the interaction between CaM or its lobes titrated into each mutant CaMBD in the presence of 10 mM CaCl2.  The columns indicate the titrant, the rows show the CaMBD in the cell. Solid lines represent the fit. Concentrations used are (titrant into cell) A: 500µM in 50µM, B: 1000µM in 100µM, C: 521µM in 52µM, D: 500µM in 50µM, E: 788µM in 79µM, F: 674µM in 67µM.  Affinity and thermodynamic parameters are shown in Table 2.33.   We also analyzed apoCaM binding to CaMBD2, by repeating the ITC experiments in the absence of Ca2+ and the presence of excess EDTA (Figure 2.7 and 2.8, Table 2.4). ApoCaM is able to bind CaMBD2 from either isoform, but significantly weaker than Ca2+/CaM (Kd 50-70µM).  The individual lobes are also capable of binding independently, RyR2mutantsN CCaMBD2WFCaMBD2WFXXN CA B CD E FW3586AF3602A 61 but the binding was too weak to be quantified reliably.  These data show that both apo-lobes contribute to the binding of apoCaM to CaMBD2.  Table 2.4 Thermodynamic Parameters of Binding between apoCaM and its Lobes and CaMBDs in 10 mM EDTAa      y g pconstruct lobe no. of sites (N) Kd (μM) ΔH (kcal/mol) ΔS (cal mol−1 deg−1) nRYR1CaMBD1 apoCaM heats indistinguishable from background 3apo-N-lobe heats indistinguishable from background 1apo-C-lobe not performed -CaMBD2 apoCaM b 1 46.5 ± 22.5 1.59 ± 0.09 26 ± 2 3apo-N-lobe 1.09 ± 0.10 55.6 ± 36.4 0.34 ± 0.09 20.8 1apo-C-lobe binding detected but not quantifiable 1CaMBD3 apoCaM c 0.38 ± 0.09 27.1 ± 14.6 1.5 ± 0.5 26 ± 1 4apo-N-lobe 0.90 ± 0.40 22.7 ± 7.7 0.90 ± 0.46 24.3 ± 2.4 3apo-C-lobe heats indistinguishable from background 2RYR2CaMBD1 apoCaM heats indistinguishable from background 5apo-N-lobe heats indistinguishable from background 3apo-C-lobe heats indistinguishable from background 2CaMBD2 apoCaM 0.89 ± 0.82 72.3 ± 28.6 2.2 ± 2.2 26.3 ± 4.7 2apo-N-lobe heats indistinguishable from background 3apo-C-lobe binding detected but not quantifiable 2CaMBD3 apoCaM 0.87 ± 0.26 5.11 ± 2.69 13.2 ± 7.8 68.8 ± 26.7 4apo-N-lobe 1.25 ± 0.07 49.3 ± 29.9 2.2 ± 1.7 27.3 ± 2.1 2apo-C-lobe 1.01 ± 0.13 62.1 ± 33 1.7 ± 0.2 25.2 ± 1.7 2aValues are averages of n measurements. Errors are standard deviations (for n = 1, errors are deviations of the fit from data). b This value was forcedto be 1 to allow for fitting. c This lower value suggests that one calmodulin binds two peptides (i.e., N = 0.5). This is possible as apo-C-lobe binding isenthalpically silent and undetectable and overlaps with the apo-N-lobe site. 62  Figure 2.7 Binding of apoCaM and its lobes to RyR1 CaMBDs.  A-G. ITC binding isotherms show the interaction between CaM or its lobes titrated into each CaMBD in the presence of 10 mM EDTA. The columns indicate the titrant, the rows show the CaMBD in the cell.  Solid lines represent the fit. Concentrations used are (titrant into cell) A: 1000µM in 82µM, B: 1500µM in 150µM, C: 2264µM in 200µM, D: 1435µM in 200µM, E: 1500µM in 150µM, F: 1500µM in 150µM, G: 1500µM in 150µM. Affinity and thermodynamic parameters are shown in Table 2.4. N CN CRyR1CaMBD1CaMBD2CaMBD3AB C DE F G 63  Figure 2.8 Binding of apoCaM and its lobes to RyR2 CaMBDs. A-G. ITC binding isotherms show the interaction between CaM or its lobes titrated into each CaMBD in the presence of 10 mM EDTA. The columns indicate the titrant, the rows show the CaMBD in the cell.  Solid lines represent the fit. Concentrations used are (titrant into cell) A: 760µM in 60µM, B: 1000µM in 100µM, C: 1000µM in 100µM, D: 1000µM in 100µM, E: 250µM in 25µM, F: 750µM in 75µM, G: 750µM in 75µM. Affinity and thermodynamic parameters are shown in Table 2.4. RyR2 N CCaMBD1CaMBD2CaMBD3N CAB C DE F G 64 2.4.2  Binding to CaMBD1 As the Ca2+/N-lobe affinity for the CaMBD2 is low, especially in the presence of a prebound Ca2+/C-lobe, we tested the binding to another segment, CaMBD1 (aa 1975 to 1999 in RyR1), which was previously suggested to form a binding site based on gel shift assays.226 Cross-linking studies also imply that it is close to CaMBD2.226 This site is less conserved than CaMBD2 between the isoforms (Figure 2.1), with 64% sequence identity between RyR1 and RyR2.  In RyR1, CaMBD1 binds Ca2+/CaM with a Kd of ~2µM. Both the Ca2+/N-lobe and Ca2+/C-lobe can bind, but their affinities are weaker than for full-length Ca2+/CaM, indicating that both contribute to the avidity (Figure 2.3, Table 2.1).  In the presence of pre-bound Ca2+/C-lobe, the affinity of Ca2+/N-lobe for CaMBD1 is weakened ~3.5 fold (Figure 2.4, Table 2.2), suggesting that the binding of both lobes simultaneously induces some strain. The CaMBD1 sequences of RyR1 and RyR2 differ in 9 positions (Figure 2.1), and this affects the binding of Ca2+/CaM. In RyR2, Ca2+/CaM binds ~4-fold weaker, with an affinity of ~7.6µM. Ca2+/N-lobe binds with a ~27µM Kd, but Ca2+/C-lobe titrations yielded a flat binding isotherm that could not be fit reliably, suggesting the interaction is extremely weak (Figure 2.5, Table 2.1). As in RyR1, the affinity of Ca2+/CaM is higher than for either lobe, indicating that both lobes can still contribute to the binding.  The main difference between RyR1 and RyR2 thus seems to be a higher affinity for CaMBD1 of RyR1, and only a very small contribution of the Ca2+/C-lobe to RyR2 CaMBD1 binding. Unlike CaMBD2, apoCaM was not observed to bind to CaMBD1 from either isoform (Figure 2.7 and 2.8, Table 2.4).  65 In summary, CaMBD1 forms a Ca2+/CaM binding site that is mainly driven by the Ca2+/N-lobe.  This lobe binds CaMBD1 stronger than it binds CaMBD2 (when a Ca2+/C-lobe is prebound to it), suggesting that Ca2+/CaM could bridge these 2 segments.    2.4.3 Binding to CaMBD3 Yet another segment, located in the C-terminal 1/5th of the channel, has been suggested to form a putative CaM binding site.221,262 The analyzed segment, termed CaMBD3, spans residues 4295 to 4325 in RyR1.  There is only ~26% sequence identity between RyR1 and RyR2 in this region (Figure 2.1).  In RyR1, CaMBD3 forms a high affinity Ca2+/CaM binding site (Kd ~52 nM) with favorable enthalpy and entropy (Figure 2.3, Table 2.1). The individual lobes bind as well, but with weaker affinities (2-4µM Kd), indicating that both contribute to Ca2+/CaM binding.   The corresponding region in RyR2 interacts with Ca2+/CaM in a completely different manner as it exhibits a complex binding isotherm that cannot be fit using a simple 1:1 binding model (Figure 2.5, Table 2.1).  Such complex isotherms have been observed before for CaM-peptide interactions in other ion channels and is likely due to a single CaM binding two peptides at the initial stage of the ITC experiment (when sufficient free peptide is available), followed by unbinding of one peptide and a 1:1 stoichiometry at the later stages (when free peptide is scarce).265  Despite our inability to fit this isotherm, the sharp transition suggests that the affinity is very high.  In order to deconstruct this complex curve, we also analyzed the binding of the individual CaM lobes. The Ca2+/C-lobe binds with a Kd ~3µM, similar to RyR1-CaMBD3. However, the Ca2+/N-lobe titrations show two transitions, revealing a high-affinity site with a Kd of 52nM driven by both favorable entropy and enthalpy and a second low-affinity site (Kd ~ 4.6µM) with unfavorable enthalpy. We used a  66 competition experiment to determine whether both sites are available in the presence of Ca2+/C-lobe (Figure 2.4, Table 2.2).  In the presence of Ca2+/C-lobe, only one exothermic Ca2+/ N-lobe binding event was observed with a Kd of 5.9µM. Overall, the binding mode and energetics of Ca2+/CaM binding to CaMBD3 in both isoforms thus appears to be very different .(Figure 2.9)   Figure 2.9 Overview of Ca2+/CaM and apoCaM binding to CaMBDs.  Bar graphs comparing the Kd and ∆G of binding for CaM (A) and the individual Ca2+/lobes (B) to the three CaMBDs. Error bars indicate S.D. AB 67 Importantly, the CaMBD3 also forms a binding site for apoCaM in both RyR1 (Kd ~27µM) and RyR2 (Kd ~5µM) (Figures 2.7 and 2.8, Table 2.4).  These values show that apoCaM can bind CaMBD3 significantly stronger than CaMBD2, suggesting that CaMBD3 may form the primary binding site for apoCaM. The two isoforms differ appreciably in their apoCaM binding. In RyR1, the binding seems to involve only the apo-N-lobe (Kd ~23µM), whereas in RyR2, both lobes contribute to apoCaM binding (Kd ~50µM and ~60µM for apo-N-lobe and apo-C-lobe, respectively).  In conclusion, CaMBD3 forms the strongest apoCaM binding site of the peptides tested, and forms an additional high-affinity site for Ca2+/CaM. Both the sequence and binding mode differ appreciably between both isoforms.  2.4.4 Disease Mutations in RyR1-CaMBD3 Interestingly, two disease mutations map into the CaMBD3 sequence of RyR1 (Figure 2.1). R4325D has been linked to CCD and multiminicore disease.269 The second mutation involves a duplication event from L4319 - R4321, which repeats the sequence LRR and has been linked to increased serum creatine kinase and potentially causes MH.269  Compared to wild type CaMBD3, the R4325D mutant weakly affects Ca2+/CaM binding (Kd 120 nM versus 52 nM) (Figure 2.10, Table 2.3). However, apoCaM binding appears stronger (Kd ~7µM versus ~27µM) (Figure 2.11).       68  Figure 2.10 R4325D and LRR duplication mutants in RyR1 CaMBD3  A-F. ITC binding isotherms show the interaction between CaM or its lobes titrated CaMBD3 mutants in the presence of 10 mM CaCl2. The columns indicate the titrant, the rows show the CaMBD in the cell. Solid lines represent the fit. Concentrations used are (titrant into cell) A: 200µM in 20µM, B: 400µM in 40µM, C: 400µM in 40µM, D: 400µM in 35µM, E: 400µM in 35µM, F: 400µM in 35µM. Affinity and thermodynamic parameters are shown in Table 2.3.      RyR1mutantsN CN CA B CD E FCaMBD3^LRRCaMBD3DR4325DLRRdup 69  Figure 2.11 ApoCaM and its lobes bind to RyR1 CaMBD3 LRRdup and R4325D mutants with different energetics than wild-type.  A-F. ITC binding isotherms show the interaction between CaM or its lobes titrated into RyR1 CaMBD3 mutants in the presence of 10 mM EDTA. The columns indicate the titrant, the rows show the CaMBD in the cell.  Solid lines represent the fit. Concentrations used are (titrant in cell) A: 1500µM in 150µM, B: 2000µM in 200µM, C: 1500µM in 150µM, D: 1500µM in 150µM, E: 2000µM in 200µM, F: 2000µM in 200µM. Affinities are B: 57±9µM, C: 65.4±9.8µM, D: 7±2µM, E: 101±19µM. Binding isotherm in Panel A could not be fit due to complex binding.  In contrast, the LRR duplication mutant binds Ca2+/CaM ~3.5 fold stronger than wild type (14 nM versus 52nM), and the affinities of each lobe individually have changed (Figure 2.10, Table 3).  Under Ca2+-free conditions, the LRR duplication mutant produces a N CN CA B CCaMBD3^LRRLRRdupRyR1mutantD E FCaMBD3DR4325D 70 complex binding isotherm with both endothermic and exothermic components (Figure 2.11).  Experiments with the individual lobes show that apo-N-lobe binding is endothermic, whereas apo-C-lobe binding is exothermic.  This is different from the WT CaMBD3, which shows a single endothermic binding event, and where direct apo-C-lobe binding could not be detected. In conclusion, both disease mutations have direct effects on the binding of both apoCaM and Ca2+/CaM to RyR1 CaMBD3.   2.5 Discussion CaM has been identified as a resident Ca2+ sensor for multiple ion channel families, including voltage-gated calcium and sodium channels, small conductance Ca2+ activated potassium channels, KCNQ potassium channels and many more.270-273 In many cases, the primary role of CaM seems to be affecting the gating properties (opening, closing, inactivation) of these channels, but it has also been involved in channel trafficking.  RyRs are sensitive signal amplifiers: increases in cytosolic Ca2+ concentrations trigger their opening, through a phenomenon known as Ca2+-dependent Ca2+ release. Although the skeletal muscle isoform (RyR1) can be activated through direct mechanical coupling with L-type Ca2+ channels11,66,237,274, they are also gated by Ca2+ ions. CaM can fine-tune the precise sensitivity to Ca2+ concentrations, in a manner that appears to be isoform-specific.  How exactly this happens remains unknown, but cryo-EM images have shown that the cytoplasmic cap of RyRs undergoes large conformational changes during opening and closing. Both apoCaM and Ca2+/CaM have been shown to bind to the lateral side of the cap70, and are likely to interfere with these allosteric motions. For example, by stabilizing the  71 closed state or destabilizing the open state, Ca2+/CaM would exert an inhibition. In both RyR1 and RyR2, Ca2+/CaM seems to inhibit the channel.212,216-218  2.5.1 Ca2+/N-lobe is Loosely Associated with CaMBD2 In order to understand the mechanisms that underlie CaM regulation of RyRs, it is necessary to know which regions of the RyR sequence can bind CaM and under which conditions.  Several RyR regions have been proposed as CaM binding sites221,222,224-226,229,261-263, but in many cases these sites were not validated using a quantitative method. In fact, most of the initially identified sites have been shown to be inaccessible within a folded domain93, or were found to be located too far from the CaM binding site identified in cryo-EM studies.236 A crystal structure of Ca2+/ CaM bound to the CaMBD2 of RyR1 has shown that this area forms a strong Ca2+/C-lobe binding site. However, the Ca2+/N-lobe is only loosely associated, and is likely to bind elsewhere.94 We confirm these observations using ITC as the Ca2+/C-lobe binds significantly stronger than Ca2+/N-lobe, with the latter affinity decreasing even further once the Ca2+/C-lobe is already bound.  Therefore it is possible that the Ca2+/N-lobe is associated with a different segment. 94,226 Here we also analyzed the binding of apoCaM and Ca2+/CaM to the two remaining candidates, which we named CaMBD1 (corresponding to RyR1 residues 1975-1999) and CaMBD3 (RyR1 4295-4325).   2.5.2 CaMBD1 Only Binds Ca2+/CaM CaMBD1 was first identified as a potential apoCaM binding site in RyR1.226 However, using our ITC experiments we could not detect any apoCaM binding to CaMBD1 for either RyR1 or RyR2. Instead, CaMBD1 only forms an alternative Ca2+/CaM binding site, with a higher affinity for the Ca2+/N-lobe than for the Ca2+/C-lobe. Our results may suggest that  72 previous pull-downs may have been false positives due to improperly folded protein. This is especially true in RyR2, where Ca2+/C-lobe binding to CaMBD1 individually could not be detected. A simple interpretation would then be that Ca2+/CaM could bridge CaMBD1 and CaMBD2 via its N-lobe and C-lobe, respectively.   2.5.3 CaMBD3 Binds Strongly to Both ApoCaM and Ca2+/CaM However, CaMBD3, a poorly conserved segment in the C-terminal 1/5th of the channel, forms an additional binding site for both Ca2+/CaM and apoCaM. In fact, it forms the highest affinity binding site we detected for CaM in the absence of Ca2+, suggesting that it may form the primary apoCaM binding site. In both RyR1 and RyR2, CaMBD3 strongly bound apoCaM and is mainly mediated by the apo-N-lobe. The Ca2+/CaM binding is almost as strong as for CaMBD2, but the energetic profile differs substantially between RyR1 and RyR2, with even 2 binding sites existing for the Ca2+/N-lobe in RyR1.  2.5.4 Mutations in RyR1-CaMBD3 Affect the Energetics and Mode of Binding We also investigated physiological mutations in RyR1-CaMBD3 which have been linked to CCD, mmCD (R4235D) and to increased creatine kinase levels/MH (LRR duplication). Neither mutation abolishes CaM binding, but both affect the energetics and likely also the binding mode, suggesting that interfering with CaM regulation at this site may be the primary cause of the disease for these specific mutations.   2.5.5 Multiple CaMBDs are Compatible with Known Structural Data A crucial question is whether the CaMBDs described here also bind CaM when present in their native context of a full-length channel. In the absence of a high-resolution  73 structure of the full-length protein, this is a difficult question to answer.   CaM binds to RyR subunits with a 1:1 stoichiometry, which would seem at odds with the presence of 3 sites that can all bind Ca2+/CaM.  Two sites could bind Ca2+/CaM simultaneously through the ability of individual CaM lobes to bind targets separately, a feature that has been observed in several CaM:ion channel complexes.270,273 In addition, the sites could also be mutually exclusive, whereby binding to one site occludes the other one through direct steric hindrance or allosterically. This allows redundancy in CaM binding, without affecting the 1:1 stoichiometry. Cryo-EM studies have shown that the center of mass of CaM undergoes a ~30Å shift on RyR1. Because CaMBD2 is able to bind both apoCaM and Ca2+/CaM, it has been suggested that CaM shifts along this helix upon binding Ca2+.223,224,275  However, mere shifts within the CaMBD2 are unlikely to create the large movements observed in the cryo-EM studies. Instead, it is more likely that either one or both lobes of CaM shift to another CaMBD upon associating with Ca2+. FRET measurements between CaM and FKBP on RyRs however do not reveal a large shift in apoCaM and Ca2+/CaM.276 To reconcile these differences it has been proposed that CaM moves in an arc with constant radius relative to the donor FKBP in the FRET measurements. The concentration of free CaM in the cytosol of cardiac myocytes has been measured at ~ 50-75nM.175 In addition, the affinity of CaM for intact RyR2 near cardiac myocyte Z-lines has been found to have a Kd ~10-20nM277, comparable to the Kd values obtained by measuring [35S]CaM binding to RyRs in SR vesicles (~20-30nM), even at Ca2+ concentrations below 0.01µM.278 One could therefore question the relevance of individual CaMBDs with Kd values higher than 1µM.  However, functional studies with voltage-gated calcium channels have shown that CaM concentrations in the immediate vicinity of some targets can be enriched several orders of magnitude.279 More importantly, due to the inherent  74 ability of the CaM lobes to bind segments individually, it is likely that higher affinity sites are formed by allowing the lobes to bridge non-contiguous CaMBDs, capitalizing on additive affinities.  This would be most important for apoCaM, since the highest affinity we observed is ~5µM for RyR2 CaMBD3. Since CaMBD1 does not bind apoCaM in our experiments, an attractive possibility is thus that apoCaM bridges CaMBD2 and CaMBD3.   2.5.6 CaMBD2 and CaMBD3 may be Accessible Importantly, for the CaMBDs to be ‘true’ CaM binding sites, they also have to be solvent accessible within full-length RyR. Cryo-EM reconstructions using GFP insertions in full-length RyR2 localized CaMBD3 to the cytosolic surface, near the side of domain 3 and close to the cleft formed by domains 7 and 8a.236 This is in proximity to the Ca2+/CaM binding site of RyR1 and the apoCaM site in RyR2.235  Importantly, the GFP insertion did not create any visible conformational changes, suggesting that the site could be solvent exposed.  Similarly, GFP insertion studies show that CaMBD2 is located near domains 3 and 8a, again close to both apoCaM and Ca2+/CaM sites of RyR1, and to the apoCaM site of RyR2.3 On the other hand, two GFP insertions near CaMBD1, at positions T1874 and T2023 of RyR2, localized to domains 9 and 4, respectively.83,85 These were interpreted to be too far from the identified CaM binding sites in cryo-EM studies. Although this would argue against the involvement of CaMBD1 in CaM binding, no insertion was done directly within the CaMBD1 sequence (residues 1941-1965 in RyR2), and the longer linker length used for inserting GFP (nine residues on either side of the insertion) may have created difference densities further away from the insertion site. Importantly, the insertions in CaMBD2 and CaMBD3 had used shorter linkers, so the locations of these sites are more reliable.  75 In addition to cryo-EM studies, the functional effect of CaM on full-length RyRs has been analyzed on a series of chimeras, deletion mutants and point mutations. Introduction of mutations in RyR2 CaMBD2 has been shown to cause early cardiac hypertrophy in mice and in neonatal cardiomyocytes, as well as an impaired inhibition by CaM.230 This study highlights a clear role for CaMBD2 in mediating CaM regulation.  2.5.7 Calmodulation of RyR May Be Involved in More than Just Channel Activity  For CaMBD1, no deletion or CaM binding knockout mutations have been generated, so a direct test of this segment in modulation by CaM has not been performed yet. The CaMBD1 sequences of RyR1 and RyR2 differ in 10 positions, but swapping these did not have an effect on the channel open probability at either high or low Ca2+.215 The difference in modulation of RyR1 and RyR2 by apoCaM is therefore unlikely to be due to any sequence differences in CaMBD1, but this does not preclude a role for CaMBD1 in binding and modulation by CaM.  No deletion has been made for CaMBD3 individually, but deletion experiments have been performed on a much larger area (RyR1 4274-4535), which encompasses CaMBD3.275 Overlay binding assays showed that apoCaM and Ca2+/CaM can still bind to this deletion mutant. Ca2+/CaM inhibited, and apoCaM activated this mutant with potencies similar to wild type RyR1.  These observations argue against an involvement of CaMBD3 in mediated CaM regulation of RyR1.  However, CaMBD3 may still have a function. First, no deletion experiments have been performed on CaMBD3 in RyR2, and given the differences in both binding and functional effects of CaM in RyR1 and RyR2, it may still be directly involved in CaM regulation of RyR2. Second, similar to other ion channels, there may be additional functions for CaM beyond regulation of the open probability, such as mediating proper  76 folding and trafficking, and these may involve CaMBD3. CaM may also be able to further affect Po in the presence of additional RyR post-translational modifications or auxiliary proteins, and this may be dependent on particular CaMBDs.    2.5.8 CaMBDs Present Additional Platforms for Binding to Other EF-hand Proteins  Finally, one should also consider the possibility of an EF-hand redundancy. It is frequently observed that CaM target peptides are also able to bind other EF-hand containing proteins. For example, the IQ domain of voltage-gated calcium channels forms a known binding site for CaM,265,267,268 but is also able to bind other members of the CaBP family,280 and even an EF-hand containing region of RyRs.123  Within RyRs, CaMBD2 has also been found to associate with S100A1, a dimeric EF-hand containing protein that is thought to compete with CaM for binding RyRs.95,204 In addition, CaMBD2 has been found to bind the same RyR EF hand-containing region.123,228 Many other EF-hand containing proteins simply have not been tested yet. It is therefore possible that one or several of the RyR CaMBDs are general EF-hand binding domains, with the exact binding depending on Ca2+ levels and relative availabilities of the various EF-hand containing proteins.        77         Chapter 3 - Crystal Structures of Wild Type and Disease Mutant Forms of the Ryanodine Receptor SPRY2 Domain          A version of this chapter has been accepted for publication: Lau, K., Van Petegem, F. Crystal structures of wild type and disease mutant forms of the Ryanodine Receptor SPRY2 domain. Nature Communications (September 2014)  78 3.1 Abstract Ryanodine receptors (RyRs) form channels responsible for the release of Ca2+ from the endoplasmic and sarcoplasmic reticulum.  The SPRY2 domain in the skeletal muscle isoform (RyR1) has been proposed as a direct link with L-type calcium channels (CaV1.1), allowing for direct mechanical coupling between plasma membrane depolarization and Ca2+ release. Here we present the crystal structures of the SPRY2 domain from RyR1 and RyR2 at 1.34-1.84Å resolution. They form two antiparallel β sheets establishing a core, and four additional modules of which several are required for proper folding.  A buried disease mutation, linked to hypertrophic cardiomyopathy and loss-of-function, induces local misfolding and strong destabilization. Isothermal titration calorimetry experiments negate the RyR1 SPRY2 domain as the major link with CaV1.1. Instead, docking into full-length RyR1 cryo-electron microscopy maps shows that the SPRY2 domain forms a link between the N-terminal gating ring and the clamp region.             79 3.2 Introduction Ryanodine receptors represent the pinnacle of ion channel complexity: with molecular weights exceeding 2.2 MDa, they form docking sites for tens of auxiliary proteins and small molecules that can regulate their activities.  Cryo-electron microscopy (cryo-EM) studies have shown that the RyRs form tetrameric assemblies that resemble the shape of a mushroom69,75,76,281: the stalk region is thought to traverse the membrane of the ER or SR, and the large cap is located entirely in the cytosol. The cap regulates the ability of the channel to open or close, turning RyRs into enormous allosteric membrane proteins. A number of crystal structures have been reported, focusing mostly on the N-terminal three domains90-92,96,97,101,102, and a single domain containing a phosphorylation hot spot loop.93,106 The primary signal for opening of RyRs is Ca2+, the very same ion it permeates.14 In cardiac myocytes, the triggering Ca2+ primarily originates from opening of DHPR located in the plasma membrane. In skeletal muscle, however, it has been shown that depolarization of the plasma membrane can trigger Ca2+ release even in the absence of extracellular Ca2+. Many studies suggest that RyR1 and CaV1.1 form direct physical contacts, with CaV1.1 acting as the voltage sensor for RyR1 through direct mechanical coupling.11,66,238,274,282 In addition, ‘retrograde’ signalling has been reported, whereby changes in RyR1 can affect the function of CaV1.1.242,283 Understanding how CaV1.1 and RyR1 can influence one another’s behaviour is one of the major puzzles in the field of excitation-contraction coupling. On the CaV1.1 side, several studies have shown the importance of the cytosolic loop connecting transmembrane repeats II and III (‘II-III loop’).237,243,284 Within RyR1, several regions of RyR1 that may bind this loop have been proposed through the use of pull-downs and chimeras between RyR1  80 and RyR2. Several interaction sites have been suggested249-251 and a major focus has been on the SPRY2 domain.88,118-120,285,286 SPRY domains lend their name to splA kinase and RyRs, where they were first identified.115 RyRs contain three such SPRY domains, conserved among vertebrates and invertebrates. Although they are founders for this family, no high-resolution structure has been reported for any of them.  Here we describe crystal structures of the SPRY2 domain of both RyR1 and RyR2, showing they consist of additional modules not previously predicted. Their location in full-length RyR1 suggests a role for linking motions in the central gating ring, formed by the N-terminal domains, with the mobile corner region.  We also analyse the effect of disease-causing mutations on the structure and stability.  3.3 Methods and Materials 3.3.1 Cloning, Expression, Purification and Crystallization of SPRY2 Rabbit RyR1 (1070-1246, 1085-1208) and mouse RyR2 (1080-1253) were cloned into the pET28 HMT vector as previously described.91 Mutations were introduced by the Quikchange method (Stratagene). Human CaV1.1 II-III loop (666-787) was cloned in a pET24a vector containing a C-terminal hexahistidine tag using NdeI and XhoI restriction sites. At the N-terminus, maltose-binding protein (MBP) was attached along with a Tobacco Etch Virus (TEV) protease cleavage site. Cultures were grown in 2xYT broth to an OD600 ~ 0.8-1 at 37°C. The temperature was then lowered to 18°C, cells were induced with 0.2 mM IPTG and allowed to grow for a total of 22-24 hours.  Two liter cell pellets were frozen at -20°C. Cell pellets were lysed by sonication after addition of 5 mL glycerol, 40 mL Buffer A (250 mM KCl, 10 mM HEPES 7.4) supplemented with 25 µg/mL DNase I, 25 µg/mL  81 lysozyme, 1 mM PMSF and 10 mM β-mercaptoethanol (bME). The lysate was centrifuged for 30 minutes at 35,000xg and a clarified supernatant was loaded onto a PorosMC (Tosoh Biosep) or TALON (GE Healthcare) column. The column was washed with 5-10 column volumes (CV) of Buffer A followed by 5 CV of 2% Buffer B (250 mM KCl, 500 mM imidazole pH 7.4). The 2% Buffer B wash was omitted for all RyR2 constructs. Protein was eluted with 30% Buffer B, TEV protease added and immediately placed in a dialysis containing Buffer A at room temperature for 4 hours or overnight. TEV protease was removed by loading the cleavage mixture onto a second PorosMC or Talon column with the flow-through collected. Excess MBP was removed by passing the flow-through over an amylose column.  The flowthrough was diluted three-fold and further purified by anion-exchange chromatography with a Hiload-Q (GE Healthcare) equilibrated with Buffer C (20 mM HEPES pH 7.4) then on a gradient from 15 to 30 % Buffer D (1M KCl, 20 mM HEPES pH 7.4) over 12 column volumes. The protein was concentrated and polished by gel-filtration chromatography on a Superdex 200 (GE Healthcare). Protein crystals were grown by the hanging-drop method at 4°C for RyR1 and at room temperature for RyR2. RyR1 SPRY2 domain (5-10 mg/mL) was crystallized in 0.9-1.1 M sodium citrate pH 6.5. RyR2 SPRY2 domain (4 mg/mL) and its mutant were crystallized in 0.2 M sodium formate, 18%-25% PEG3350.   3.3.2 Data Collection and Structure Determination Crystals were harvested and flash-frozen in solutions containing the original growth conditions supplemented with 30% glycerol (RyR1) or 25% isopropyl alcohol (RyR2). Diffraction data were collected at the Canadian Light Source (CLS) beamline 08ID-1 and the Advanced Photon Source (APS) beamline 23-ID-D-GM/CA. Data were processed with  82 XDS.287 Initial phases were calculated by Sulfur-SAD with Autosol288 using native sulfur atoms on the RyR2 SPRY2 WT structure collected at 6.5 keV. The initial structure was further refined against a 1.34 Å dataset using PHENIX289 and manual model building using COOT290. The RyR2 SPRY2 structure was used as a search model for molecular replacement to solve the RyR1 structure and RyR2 SPRY2 A1107M mutant. Table 3.1 contains the statistics for data collection and refinement.  Table 3.1 Data collection and refinement statistics ss 3.3.3 Cryo-EM Docking The ‘colores’ program in the SITUS suite291 was used to dock the RyR1 SPRY2 domain, containing the C-terminal tail in the antiparallel configuration, into four different RyR1 cryo-EM maps with EMDB entries 1606, 1607, 1275 and 5014. Six dimensional  RyR1 1070-1246 RyR2 1080-1253 (Native) RyR2 1080-1253 (SAD) RyR2 1080-1253 (A1107M) Data collection     Space group P1 P22121 P22121 P22121 Cell dimensions       a, b, c (Å) 37.6, 37.6, 95.8 36.3, 66.2, 67.0 36.6, 66.7, 67.2 36.7, 67.0, 67.2  , ,   (°)  89.97, 89.97, 120.05 90, 90, 90 90, 90, 90 90, 90, 90 Wavelength (Å) 1.03 0.98 1.91 1.03 Resolution (Å) 32.55  - 1.84 (1.91  - 1.84) 47.09  - 1.34 (1.39 - 1.34) 36.55  - 2.26 (2.34  - 2.26) 47.45  - 1.44 (1.49  - 1.44) Rmerge (%) 10.0 (61.6) 7.2 (45.0) 4.9 (6.4) 4.0 (6.1) Rpim(%) 10.0 (61.5) 3.0 (29.0) 1.0 (1.6) 1.7 (38.2) I / I 8.1 (1.4) 17.6 (3.9) 65.1 (34.7) 24.3 (1.9) Completeness (%) 94.0 (92.4) 91.6 (56.0) 91.7 (54.0) 92.6 (59.2) Redundancy 2.0 (2.0) 6.0 (3.1) 25.2 (15.9) 6.4 (3.3) CC* 0.997 (0.837) 0.999 (0.951) 1 (0.999) 1 (0.937)      Refinement     Resolution (Å) 32.55  - 1.84 47.09  - 1.34  47.45  - 1.44 No. reflections 73737  203975  184743 Rwork / Rfree 16.3/22.0 13.8/16.3  15.1/18.4 No. atoms     Protein 4188 1291  1285 Ligand/ion 0 0  0 Water 391 227  199 B-factors     Protein 18.8 14.3  20.5 Water 25.3 28.0  34.2 R.m.s. deviations     Bond lengths (Å) 0.013 0.007  0.008 Bond angles (°) 1.36 1.17  1.14  83 searches were performed with a rotational sampling of 3 degrees and with off-lattice Powell optimization. No density cutoff was applied, and a Laplacian filter was used throughout.   3.3.4 Thermofluor Thermal Shift Assays Protein stability was measured by fluorescence as described before.91,292 10 µL of 1.5 mg/mL protein was mixed with 5 µL of a 1/500 dilution of SYPRO orange (Invitrogen) and diluted to 50 µL with buffer (150mM KCl, 10mM sodium phosphate pH 7.4, 14.3 mM bME). Thermal melts were performed eight fold using a DNA Engine Opticon 2 real-time PCR machine (Biorad) with the SYBR green filter. The temperature was increased from 25°C to 95°C in 0.5°C increments with each step held constant for 15 seconds. Curves were normalized and the midpoints were taken as the melting points.  3.3.5 Isothermal Titration Calorimetry Proteins were dialyzed against 150 mM KCl, 10 mM HEPES (pH 7.4), 10 mM bME and 1 mM NaN3 at 4°C. Concentrations were determined by the method of Edelhoch64. Titrations consisted of 20 injections of 2 µL RyR1 SPRY2 domain into the cell containing either buffer or II-III loop. Concentrations used are noted in the figure legends. Experiments were performed at 25°C and a stirring speed of 1000 rpm on an ITC200 instrument (GE Healthcare). All data were processed using Origin 7.0.  3.4 Results 3.4.1 The SPRY2 Domains in RyR1 and RyR2 Consist of Modular Subdomains.  We solved the crystal structures of the SPRY2 domains of two RyR isoforms, encoded by residues 1070-1246 in rabbit RyR1 and 1080-1253 in mouse RyR2. The RyR2  84 SPRY2 structure was solved via sulphur-SAD phasing, which was then used as a molecular replacement model for the RyR1 SPRY2 domain structure. The final resolutions were 1.84 Å (RyR1) and 1.34 Å (RyR2). Three molecules are present in the asymmetric unit for the RyR1 structure. As these are nearly identical (Figure 3.1), we performed all analyses with chain A.   Figure 3.1 Superposition of chains A, B and C (blue, green, purple, respectively) of RyR1 SPRY2.  The chains look nearly identical, with RMSD values for Cα atoms corresponding to: A vs B: 0.09Å; A vs C: 0.13Å; B vs C: 0.12Å.  Figure 3.2 shows the overall structures of the SPRY2 domains. It is immediately clear that these consist of several individual structural elements that make them substantially different from previously predicted SPRY2 domain structures. The core of the protein  85 consists of a stretch of 10 β strands (β 1- β 10), arranged in two antiparallel sheets, which was previously assumed to constitute the entire SPRY2 domain.88,118,285 However, at the N-terminus an extension is present, adding a supplementary beta strand (β 0) and a short 310 helix.  A 10-residue loop (‘insertion loop’) is inserted in what is normally strand β 8, splitting it up into two smaller strands β 8a and β 8b. C-terminal to the last β strand of the core is a region we named the ‘lid’, containing two 310 helices and a short β strand that forms β sheet interactions with the first strand of the core. The lid makes substantial hydrophobic interactions with both the N-terminal extension and the core, burying ~820 Å2 of surface area.  The core, lid and insertion loop are in the exact same position in RyR1 and RyR2 (Figure 3.2C).  Because an identical arrangement is found in two structures for which the crystallization conditions and crystal contacts are entirely different, these arrangements appear stable and likely also occur in full-length RyRs.            86  Figure 3.2 Overall structure of the SPRY2 domain.  A. Two views of the RyR1 SPRY2 domain. Secondary structure elements are labelled. The core of the structure is shown in purple. Extra modules are shown in different colours (blue: N-terminal extension; yellow: lid; cyan: insertion loop; green: tail). B. Two views of the RyR2  87 SPRY2 domain with colouring and labels as in panel A.   C. Superposition of the SPRY2 domains from RyR1 (colours) and RyR2 (grey).   D. Surface representation of the RyR1 SPRY2 domain with each module in a different color, showing the extensive interactions between the modules.   E. Sequence alignment of the SPRY2 domains in RyR1, RyR2, and RyR3, with the secondary structures indicated. Colours represent the modules. Loop regions with structures are indicated by a solid line. Dotted lines represent parts of the construct that lack interpretable electron density. The red letters indicate positions of disease-causing mutations.    Immediately following this lid is an extended region (‘tail’).  This unusual conformation in both structures is the direct result of crystal contacts (Figure 3.3A,B). In both RyR1 and RyR2, the tail interacts with the same surface of a neighbouring molecule, although the orientation is different: parallel in RyR1, but forming an additional antiparallel β strand in RyR2.  Despite these extensive crystal contacts, the constructs behave as monomeric species using size exclusion chromatography, suggesting that in solution, the tail is more probable to form an intramolecular, rather than intermolecular interaction. Only the antiparallel arrangement can be formed, extending the antiparallel β sheet (Figure 3.3C). A parallel intramolecular arrangement is sterically not allowed due to the limited length of the tail. The tail contains a Trp that forms extensive interactions and that is highly conserved among RyRs (Figure 3.4).           88  Figure 3.3 Swapping of the tail region in the crystal lattice. Two neighbouring chains in the crystal lattice are shown for the SPRY2 domain of A. RyR1 and B. RyR2, indicating that the tail mediates crystal packing contacts with strand β0.  In RyR2, the tail extends an antiparallel sheet. C. Likely intramolecular arrangement for the tail in solution, shown here for the RyR2 SPRY2 domain.   Figure 3.4 Interactions between the tail and the N-terminal extension strand in RyR2.   A. Sequence alignment between selected RyR sequences shows the conservation of a Trp residue in the tail. B. Detail of the structure showing that the tail Trp resides in a small pocket formed by the remainder of the SPRY2 domain.  The tail peptide (green) extends one of the β sheets through interactions with β0 of the N-terminal extension. Colours are as in Figure 3.2.   89 In comparison with SPRY domains from other proteins, several elements appear unique to the SPRY2 domain (Figure 3.5). The closest structural homolog is the SPRY domain from Ash2L, which also contains a ‘lid’, but still misses the insertion loop and contains two extra β sheets that replace the N-terminal and C-terminal extension strands. Two other SPRY domains are present within the RyR sequence. Both share low sequence identity with SPRY2 (~23% and 17% for SPRY1 and SPRY3, respectively) and several insertions and deletions are present (Figure 3.6). Neither of them seems to contain the insertion loop and the sequences within the additional modules are particularly dissimilar, suggesting that the three SPRY domains within RyRs have diverged substantially.  90   Figure 3.5 Comparison with other SPRY structures.  Shown are superpositions of SPRY2 (colours) with (white) SPRY domains of (A) Trim5α (PDB 4B3N), (B) SPRYD3 (PDB 2YYO) and (C) Ash2L (PDB 3TOJ).  The labeled portions indicate parts that are different in the superposed structures. Compared to all existing SPRY structures, the insertion loop, which cuts SPRY2 strand β8 in two halves, is not observed in any other SPRY structure. No other SPRY domain has a C-terminal extension (shown here in the extended conformation, but likely to form an additional beta strand interaction with the SPRY2 core). A lid is observed in Ash2L, but obtains a different conformation. Many other SPRY structures contain an extra β sheet at the N-terminus, also known as a PRY domain that is absent in SPRY2.   91  Figure 3.6 Sequence alignment of the three predicted SPRY domains in rabbit RyR1.  The shaded regions show strictly conserved residues. The secondary structure of rabbit RyR1 SPRY2 is shown on top for reference. Overall, the sequence identities between the three domains are very low, with multiple insertions and deletions.    92 3.4.2 Loss-of-Function Mutations in the RyR2 SPRY2 Domain The SPRY2 domain of RYR2 harbours the position of a very unusual disease mutation (T1107M) that has been linked to hypertrophic cardiomyopathy.293 Importantly, functional experiments have indicated a loss-of-function phenotype for the T1107M mutant, with a reduction in fractional Ca2+ release and an increase in the threshold for its termination.294 Since most mutations characterized in the cytosolic portions of RyRs are instead linked to gain-of-function phenotypes, we decided to take a closer look at this loss-of-function mutation. The corresponding residue in the mouse RyR2 (A1107) is buried, with the Cα atom pointing towards a hydrophobic core (Figure 3.7A,C). There is space for a Thr at this position, but a Met side chain would completely clash with Trp1156 (Figure 3.8). Surprisingly, the A1107M mutation is tolerated within the individual domain, as indicated by the absence of any aggregation in size exclusion chromatography (Figure 3.9). However, the substitution causes a significant destabilization of the domain, with the melting temperature decreasing ~9°C compared to wild type (Fig. 3.7B). At a physiological temperature of 37°C, ~22% of the domain is unfolded. Although the actual melting temperature will be different within full-length RyR2, where domain interactions may confer additional stability, there is a clear trend towards destabilization of the fold.        93  Figure 3.7 Disease mutations within SPRY2.  A. Relative positions of disease-causing mutations (black sticks) in the RyR1 and RyR2 SPRY2 domains. Labels indicate the mutation in human RyRs, with the corresponding rabbit RyR1 and mouse RyR2 residue in brackets. B. Thermal melting curves, as determined by thermofluor assays, for WT and mutant SPRY2 domains. The curves are the average of 8 independent measurements. The obtained melting temperatures are (with standard deviations): RyR1 SPRY2: 43.2±0.3°C, RyR2 SPRY2: 47.9±0.3°C, RyR1 R1076W: 40.2±0.5°C, RyR2 A1107M: 39.1±0.2°C. C. Close-up of the superposition of WT (colours) and A1107M SPRY2 (white) in RyR2. Ala1107 is shown in black. The arrows show the relative positional changes. D. Close-up of Arg1076 (black) in RyR1 SPRY2. A Trp side chain (white) at this site would cause a steric clash with neighbouring hydrophobic residues.    94  Figure 3.8 Clash induced by the A1107M mutation in RyR2 SPRY2.    Shown is the wild type RyR2 SPRY2 crystal structure with the Ala1107 residue replaced by a Met residue. This would cause a steric clash with Trp1156. The Van der Waals surfaces for Met1107 and Trp1156 are shown transparent to highlight the clash.    Figure 3.9 SPRY2 mutants are monomeric. Superposed size exclusion chromatograms (Superdex200 analytical, GE) for wild-type and disease mutant forms of the RyR1 and RyR2 SPRY2 domains.     95  We crystallized the corresponding A1107M mutant and solved its structure at 1.44 Å. In order to allow the Met side chain to fit in the hydrophobic core, the backbone of residue 1107 is shifted over 1.6 Å. This shift now results in several changes at the surface. It affects the main chain positions of the neighbouring residues Glu1106 and Val1108, whose side chains adopt different conformations, resulting in positional shifts up to 3.1Å (Fig. 3.7C). In the wild type structure, Glu1106 forms a salt bridge with Arg1214, but in the mutant this salt bridge is lost and the electron density for the Arg1214 is absent, indicating that this residue has become flexible. In conclusion, the mutation induces a local misfolding resulting in surface changes at low temperatures, and has a substantial degree of unfolding at 37°C.  These combined effects likely underlie the loss-of-function phenotype associated with this mutation.   3.4.3 Mutations in the RyR1 SPRY2 Domain The RyR1 SPRY2 domain contains locations for 5 disease mutations (Fig. 3.7A). This includes R1075W (human numbering), identified in two patients from one family who died shortly after birth. Muscle biopsies indicated central core disease (CCD).295 The residue is strictly conserved, and is located at the end of the N-terminal extension strand (β0), making multiple interactions with neighbouring residues. The stalk portion of the side chain is in hydrophobic contact with several hydrophobic residues (Fig. 3.7D), whereas the positively charged guanidinium group is exposed to the surface.  In a model of the RyR1 SPRY2 domain where the tail is added in the antiparallel, intramolecular arrangement, the Arg1075 is also in contact with the tail (Figure 3.10). Without substantial rearrangements, there would be no room for a bulkier Trp side chain,  96 even in the absence of the tail strand.  We tested the impact of the corresponding R1076W mutation in the rabbit RyR1 SPRY2 domain. The protein can still fold, as it does not cause aggregation as measured by size exclusion chromatography (Figure 3.9). However, thermofluor experiments show that the mutation destabilizes the domain, reducing the melting temperature by ~3°C (Fig. 3.7B). Although the effect on thermal stability is thus smaller than for the A1107M mutation in RyR2, the RyR1 SPRY2 domain is already less stable than in RyR2, and the melting temperature of the R1076W mutant SPRY2 is very similar to the RyR2 A1107M SPRY2. Our attempts to crystallize the mutant failed, suggesting that the structural rearrangements are substantial. Based on these observations, we predict that the human R1075W mutant may lead to a loss-of-function phenotype, consistent with CCD.   97  Figure 3.10 Tail Interactions with Arg1076.   The human RyR1 disease mutation R1075W corresponds to position Arg1076 in rabbit RyR1. This residue (black) is part of strand β0 in the N-terminal extension (blue) and forms interactions with the core (purple) and the tail (green, modelled for the RyR1 SPRY2 domain based on the antiparallel interaction seen for the RyR2 SPRY2 domain).  A. and B. The structure without and with the tail, respectively. Both in the absence and presence of the tail, several residues would have to rearrange to accommodate the bulky Trp side chain.  Two other mutations in the RyR1 SPRY2 domain have also been identified in patients with CCD.296 However, in both cases, an additional mutation was found, so the phenotype could be due to either one or their combination. The G1165D is located in a turn  98 preceding strand β 8a in a region that interacts with the lid module (Figure 3.7A and 11). An Asp residue at this position is expected to interfere with the interactions, as it would cause clashes with 2 hydrophobic residues of the lid (Figure 11). The mutation is therefore highly likely to affect function of RyR1. In contrast with this, the R1179W mutation is exposed at the surface of the protein, and is thus not expected to interfere with folding (Figure 3.7A). It can only be responsible for a disease phenotype if it is located at an interface with an auxiliary protein or another domain.   Figure 3.11 Effect of the G1165D disease mutation.  The human RyR1 mutation G1165D corresponds to residue Gly1166 in rabbit RyR1, and is located at the interface between the SPRY2 core (purple) and the lid module (yellow). A simple introduction of an Asp residue would cause clashes with hydrophobic residues (Phe1213 and Ile1216) in the lid. The modelled aspartic acid is shown in black sticks, with its Van der Waal’s surface transparent.    In addition, two mutations in the SPRY2 domain have been linked to malignant hyperthermia, a disorder linked to RyR1 gain-of-function.49,297 The R1127H and R1140C  99 mutations (human numbering) are both exposed at the surface of the protein (Figure 3.7A). The substitutions are not predicted to create any steric hindrance, and therefore would only affect the surface properties of the domain. Similar to the R1179W mutation, they can only create a disease phenotype if they are located at an interface. Finally, the P1144L mutation was found in exome sequencing but it is unclear whether it is causative of MH.298 Its corresponding residue is a Ser in the rabbit RyR1 sequence, and is exposed at the surface.  In order to consider the further impact of these disease mutations, we explored the potential role of the SPRY2 domain by analysing its proposed binding and its location in full-length RyRs.  3.4.4 Does the RyR1 SPRY2 Domain Form an Interaction Site with the II-III Loop? Several studies have supported a role for the RyR1 SPRY2 domain in direct interactions with the cytosolic II-III loop of CaV1.1.  With a very soluble and well-behaved RyR1 SPRY2 domain in hand, we tested its ability to bind the II-III loop using isothermal titration calorimetry (ITC).  However, even at high concentrations, the heat signals were indistinguishable from background titrations (Figure 3.12A), indicating that any binding between the two, if at all, would be extremely weak (Kd > 1 mM) , or proceed with extremely small enthalpic changes. However, pull-down experiments we performed also did not indicate any interaction (results not shown).        100  Figure 3.12 Absence of interactions between the RyR1 SPRY2 domain and the CaV1.1 II-III loop.  A. Raw ITC data comparing the titrations of RyR1 SPRY2 (887 µM) into buffer or the II-III loop (89 µM), showing no significant difference in heat signals. B. Size exclusion chromatograms (analytical Superdex200) of MBP fusions of RyR1 1070-1246 and 1085-1208. Removal of the MBP tag from the 1085-1208 construct led to complete loss of the protein.   C. Structure of the RyR1 SPRY2 domain, highlighting residues 1076-1112 (human numbering, green) and the ‘F-loop’, encoded by residues 1107-1121 (human numbering, orange). Alternating colours show regions of overlap. Both are involved in extensive interactions with the remainder of the domain and are unlikely to adopt a native conformation in isolation. D. Detail of the RyR1 F-loop (orange), indicating its hydrophobic side chains in green sticks. These interact extensively with hydrophobic residues on the rest of the SPRY2 domain (purple sticks).   Previous biochemical experiments were performed using only the ‘core’ of the SPRY2 domain, consisting of residues 1085-1208 in RyR1.120,285,286 This represents the major module in the structure, but upon expression and purification of such a construct, it proved to be poorly soluble and caused major aggregation as indicated by size exclusion  101 chromatography (Fig. 3.12B). This aggregation is observed even despite the presence of a stabilizing maltose binding protein (MBP) fused at the N-terminus. Proteolytic removal of the MBP (see Methods) caused complete loss of the protein. Even constructs in which only the N-terminal extension or lid was deleted separately appeared to be misfolded.  This is in agreement with the extensive amount of hydrophobic surface that is buried by these extra modules.   As such, any previous studies using only the SPRY2 core or portions thereof are likely to yield unfortunate false positives because these missed major elements required for stability of the SPRY2 domain. We conclude that the RyR1 SPRY2 domain is unlikely to be a major determinant of the RyR1-CaV1.1 interaction. As a result, the disease-causing mutations also are unlikely to affect binding to CaV1.1 directly. We next explored a possible role for the SPRY2 domain by analyzing its location.  3.4.5 Location of the SPRY2 Domain in Full-length RyR1 We attempted to locate the position of the RyR1 SPRY2 domain in available cryo-EM maps of full-length RyR1. We used an unbiased approach, employing systematic 6-dimensional searches (3 rotational and 3 translational parameters) as implemented in Situs.291 We analysed and ranked the cross-correlation coefficients for each of the tested locations. As noted before, the addition of a Laplacian filter appeared crucial to obtain sensible docking results.92 In three of the tested maps (EMDB accession numbers 5014,1606, and 1607)69,131, the top hit consistently docked to the same location and, reassuringly, in the same orientation (Figure 3.13A).  Plotting the correlation coefficients of the top hits shows that there is some contrast, especially in the EMD5014 map where the top hit is separated by the mean of the next 9 hits by up to ~8 standard deviations.   102  Figure 3.13 Location of the SPRY2 domain in intact RyR1 A. Top docking solutions for the RyR1 SPRY2 domain in three different cryo-EM maps of full-length RyR1. The colours for the SPRY2 domain are as in Figure 3.2.  The positions of the N-terminal three domains (blue, green, red) are also shown for reference. The bar graphs below show the normalized correlation coefficients for the top 10 unique hits in each map. Due to the four-fold symmetry, each solution comes up 4 times, but is only shown once in the bar graphs. The filled circles in the top left panel represent the locations for difference density observed for RyR2-GFP insertions near RyR2 residues 846 (yellow circle)299 and 1366 (orange circle).84 The purple dotted circle represents the area where difference densities were observed for cryo-EM studies of anti-SPRY2 antibody bound to RyR1.88 B) ‘side view’ of the docked SPRY2 domain, showing its solvent accessibility at the periphery of the RyR (arrow). C) Close-up showing that 2 solvent channels (arrows) line the docked SPRY2.  The loop encoded by residues 1107-1121 is shown in orange. It is accessible from both solvent channels.  Searches in a fourth map (EMDB 1275) only yielded nonsense positions, such as the column positions, transmembrane area, and the position where the N-terminal area was  103 found to be located.92,97,98 As expected with nonsense solutions, there was no contrast between the top hit and the following solutions (Figure 3.14). Figure 3.14 Docking of the RyR1 SPRY2 domain into the RyR1 cryo-EM map with EMDB entry 1275.  A. and B. 2 views of the RyR1 cryo-EM map (mesh). The positions of the N-terminal three domains, which have been validated, are shown in black. The different coloured domains correspond to the top 10 ranked docking results. Because of the 4-fold symmetry, each hit would appear 4 times, but is only shown once for clarity. All top 10 solutions dock to nonsense regions located either within the central columns, or clashing with the N-terminal domains. None of these are compatible with previous GFP insertion studies in areas close to the SPRY2 domain.84,299 C) Normalized correlation coefficients for these top 10 unique hits. There is no large ‘contrast’ between these hits, which is typical of spurious docking results.  Because there are two other SPRY domains within the RyR polypeptide, there is the inherent possibility that the top hit could correspond to either SPRY1 or SPRY3.  In the  104 absence of crystal structures for these, this cannot be formally tested, but docking of homology models in the EMDB 5014 map yielded entirely different positions. In addition, docking of other known SPRY structures also yielded different positions, suggesting that there is sufficient resolution in the maps to discriminate between proteins of the same fold. We therefore propose the location shown in Figure 3.13 as the most likely position for the SPRY2 domain. This setting is strategically located between the corner, an area shown to undergo large conformational changes during opening and closing69, and the central rim, formed by the N-terminal three domains establishing a ‘gating ring’ that is thought to modulate channel open probability.97 The SPRY2 domain is likely required to couple the conformational changes in both areas, and disease-causing mutations on its surface may interfere with this process. Importantly, this location is in agreement with several previous studies. First, several antibodies raised against RyR1 SPRY2 have been shown to access their epitopes in native, folded RyR1.88 In the proposed dock, the SPRY2 domain is accessible via 2 solvent channels and from the periphery (Figure 3.13B,C). Cryo-EM reconstructions of antibody-bound complexes showed difference density in the clamp region, in an area compatible with our proposed SPRY2 position,  especially given the ~160Å dimensions of antibodies (Figure 3.13A, left panel, dotted circles). One of the antibodies was raised against residues 1107-1121, and this loop is accessible via 2 solvent channels. Second, several cryo-EM studies have been performed on GFP fusions of RyR2. In one of these, GFP was inserted after RyR2 residue Thr1366, ~110 residues downstream of the SPRY2 sequence, and difference density was shown in subregion 6.84 Another study inserted GFP near residue Tyr846, ~200 residues upstream of the SPRY2 domain, yielding difference density in the top portion of subregion 9.299 So although no GFP insertion studies have been reported for the SPRY2 domain,  105 insertion studies in the neighbouring domains within the sequence appear compatible with the proposed SPRY2 position, which is located directly in between these two difference densities (Fig. 3.13A, left panel).  3.5 Discussion 3.5.1 RyR SPRY2 Contains Modules that are Required For Folding and Not Present in Other SPRY Domains. The Ryanodine Receptor is a large allosteric membrane protein.  Thanks to the presence of multiple domains located in the cytosolic cap, this membrane protein giant can receive input from tens of proteins and small molecules.  One type of domain, the so-called ‘SPRY’ domain is repeated three times in the RyR sequence300, and these are conserved from mammalian species all the way down to Drosophila and C. elegans. However, up to now, no structure from any SPRY domain of any RyR had been reported.  Here we show high resolution crystal structures of the SPRY2 domain of both RyR1 and RyR2. Contrary to previous predictions, this domain consists of several distinct modules (N-terminal extension, lid, insertion loop, and tail) that add to what was previously thought to contain the SPRY2 domain. At least two of these modules, the N-terminal extension and lid, are absolutely required for proper folding of the domain.   3.5.2 Mutations in SPRY2 Induces Local Misfolding or Alter Interdomain Interactions.  RyRs are the target for disease-causing mutations that mostly result in Malignant Hyperthermia301 or Central Core Disease302 for RyR1 or in Catecholaminergic Polymorphic Ventricular Tachycardia260 for mutations in RyR2.   A common theme is that most mutations  106 on the cytosolic cap have been linked to a gain-of-function, resulting in premature or prolonged release of Ca2+ into the cytosol. The SPRY2 structures allowed us to map six mutations. Interestingly, they segregate into two groups, with three mutations either fully or partially buried within the domain, and three fully exposed at the surface.  The buried T1107M mutation targets RyR2 and has been linked to hypertrophic cardiomyopathy293, as well as CPVT.303 It is unclear how it could result in two different phenotypes, but functional experiments have clearly shown it to confer a rare loss-of-function phenotype with early termination of Ca2+ release.294 The mutation causes a significant thermal destabilization, resulting in 22% unfolding at physiological temperatures. In addition, it induces local misfolding, which results in conformational changes at the surface. Although the folding in intact RyR2 can be stabilized, the 9°C shift in melting temperature stability suggests a trend towards significant destabilization, suggesting that misfolding of the SPRY2 domain may generally lead to loss-of-function phenotypes. Two RyR1 mutations (G1165D, R1075W) also affect partially buried residues. Interestingly, both of them have been linked to CCD, a disorder that can be associated with loss-of-function phenotypes in RyR1. Although both mutations were also found in conjunction with mutations elsewhere in RyR1, we predict they interfere with proper folding and either contribute or are fully responsible for the disease phenotype.  Three other mutations are found at the surface. Two of these have been linked to MH, a disorder characterized by gain-of-function of RyR1, and another has been linked to CCD.  Since CCD can be linked to both gain and loss of function, a possible interpretation here is that mutations interfering with folding of SPRY2 create loss-of-function, whereas mutations on the surface create gain-of-function phenotypes. For the three mutations at the surface to be disease-causing, they have to lie at an interface with an auxiliary protein or  107 other domain. In order to identify what this might be, we analysed its interaction with a proposed binding partner and its location in full-length RyR.  3.5.3 The RyR1 SPRY2 Domain Binding to the CaV1.1 II-III Loop is Undetectable by Isothermal Titration Calorimetry. In the absence of a high-resolution structure, it was previously assumed that the SPRY2 domain only consisted of a smaller region, encoded by human RyR1 residues 1085-1208.  However, it is now clear that the SPRY2 domain consists of additional portions that add ~43% of extra sequence, and our data show that the shorter construct causes aggregation, in stark contrast with the high solubility and monomeric nature of the crystallized 1070-1246 construct. In the presence of these additional modules, our ITC experiments fail to recapitulate the interaction with the CaV1.1 II-III loop, indicating that the SPRY2 domain cannot be the major determinant of the RyR1-CaV1.1 interaction.  Previous experiments also suggested that an even smaller portion of the SPRY2 domain, encoded by residues 1076-1112, could bind to the II-III loop.119 However, it is clear from the structure that such a fragment would not be able to fold in its native conformation, as it isolates single β strands from two different β-sheets (Figure 3.12C). Other experiments have further narrowed down the interaction to a loop encoded by residues 1107-1121, and showed that this peptide, in isolation, can affect the activity of RyR1 in planar lipid bilayers.118 However, it is clear that this loop, which previously was thought to be extended,285 makes substantial interactions with the remainder of the structure, burying hydrophobic side chains (Fig. 3.12D). The loop is highly unlikely to be adopting this stabilized conformation in isolation. Most likely, the interactions between RyR1 and Cav1.1, whether direct or via an intermediate  108 protein, involve one or multiple RyR1 segments that have previously been suggested to be important for E-C coupling.249-251,304 A popular method for studying RyR structure-function is the use of isolated RyR peptides and to measure their effect on RyR function. It was shown that the ‘F-loop’, encoded by RyR1 residues 1107-1121, could increase the activity of RyR1 in planar lipid bilayers.118 However, this loop forms extensive hydrophobic interactions with the remainder of the SPRY2 domain (Figure 3.12D), and is thus very unlikely to adopt a native structure in isolation.  It is unclear then why such a peptide could cause a functional effect. The same is true for a peptide in the N-terminal region of RyR2 (residues 163-195), which was shown to increase leak of Ca2+ through RyR2,305 but is largely an integral part of the β-trefoil core of domain A91,96, showing that it is impossible to adopt the same conformation in isolation. It may simply be that RyRs are sensitive to several unfolded peptides that do not necessarily mimic a native structure found in intact RyRs.  3.5.4 SPRY2 is a Domain in the Clamp Region of RyR Obtaining a high-resolution structure of the entire RyR has thus far been unsuccessful. The next best thing is to obtain high-resolution structures of individual domains or domain clusters, and then to locate these by finding the best position in full-length RyR cryo-EM maps.  RyR1 reconstructions are available in the 10-12 Å range69,74-76, and these have allowed locating the N-terminal three domains of RyR1 with high accuracy to the central rim.92,97 Thanks to the larger size, locating a three-domain structure appeared straightforward, with high contrast between the correlation coefficient of the top hit compared to the next hits in the ranking. The position of the N-terminal region was also confirmed via difference cryo-EM and FRET measurements.98 However, with single-domain  109 structures, the contrast is typically lower.93 Locating the SPRY2 domain yielded consistent results among three out of four RyR1 cryo-EM maps (Figure 3.13A). Given that the refined positions appear to adopt the exact same relative orientation of the domain in all three maps, we propose this location as the most likely site for the SPRY2 domain.   There currently is no report of any cryo-EM study using the SPRY2 domain as a position for insertions. However, the proposed location of the SPRY2 domain is in agreement with previous cryo-EM studies, which used insertions of GFP  ~200 residues upstream299 or ~110 residues downstream84 of the SPRY2 domain. The proposed location of SPRY2 is directly in between the two difference densities (Figure 3.13A, left panel). In addition, cryo-EM reconstructions of RyR1 bound to antibodies raised against the SPRY2 domain are also consistent with this location.88  One antibody was raised against the ‘F-loop’, which is accessible in the docked SPRY2 via two solvent channels. However, some structural changes may be necessary to allow access to the antibodies, which may explain why most of the RyR1 particles did not have antibodies bound to each subunit simultaneously.88 Interestingly, antibodies raised against the SPRY2 domain were found to increase RyR1 activity in planar lipid bilayers, which would be expected if conformational changes are required for antibody binding.  Importantly, this location is directly in between the corner region and the central rim. Both areas have been shown to undergo large conformational changes during channel opening.69,97 The N-terminal three domains have been shown to form a ring near the four-fold symmetry axis92, and opening of the channel is coupled to disruption of an intersubunit interface.97  The corner positions have been shown to move ‘downward’ towards the SR membrane during channel opening.69 The SPRY2 domain, together with other domains, is likely involved in linking these transitions and the disease-causing mutations may then  110 interfere with this process.  One of the contacts seems to involve the insertion loop, a loop that cuts strand β8, normally found in canonical SPRY folds, into two smaller strands. Interestingly, two of the mutations located at the surface (R1179W,R1140C) flank this loop, and may thus interfere directly with this domain-domain interaction. The identities of the domains directly contacting SPRY2 remain to be found. Importantly, the RyR sequence contains two additional predicted SPRY domains. Homology models for these domains did not dock to the proposed SPRY2 location, which is not surprising given their low degree of sequence identity with SPRY2 (23% and 17% for SPRY1 and SPRY3, respectively), and the multiple insertions and deletions (Supplementary Fig. S4). Similarly, we have previously shown that the cryo-EM maps can discriminate between domains A and B, which both form β-trefoil folds with twelve β strands each.92 Despite the same fold, and a near perfect superposition of the strands, the loop regions of these domains largely determine the overall shape, which determines their docked location. In order to further confirm the SPRY2 location, FRET experiments and further difference cryo-EM studies will be invaluable for experimental verification and to suggest possible interacting domains.     111            Chapter 4 – Future Directions and Conclusions              112 Ever since the Ryanodine Receptor was first cloned it has proven to be a major challenge for structural. Due to its large size, teasing apart the intricate web of interactions between its domains and modulators is a tough puzzle to solve. In addition, its membrane protein nature and size make it a difficult protein to overexpress and purify.  Studying the modulation and control of RyR has been of interest because its dysfunction leads to multiple diseases. Currently, the mechanism of action of triggers that activate RyRs, such as volatile anaesthetics, or compounds that can inhibit the channel, such as dantrolene, are not known. In a small protein, one can assay for a functional endpoint or try to detect binding via biochemical methods such as fluorescence, NMR, calorimetry and surface plasmon resonance amongst others. For RyRs, most of these methods prove impossible due the limited availability of recombinant protein. Even with known ligands that modulate the channel it has been a challenge to locate their exact binding sites. Initial attempts have been made to overcome this limitation by expressing portions of the channel with fusion proteins.221 Although this methodology is quick and easy, it is undoubtedly prone to false positives. As no structural data was available, and often no quality control of the proteins is performed, these fusion proteins may have contained only parts of domains. The resulting misfolded protein constructs then often expose hydrophobic regions that would otherwise be buried, causing non-specific interactions with several RyR modulators. A more elegant method that has been used in other membrane proteins that can be used on RyR is through its limited proteolysis.306 Some domains that are part of the receptor are inherently stable and treatment with proteases can provide soluble domains. Sequencing of these fragments can then lead to the identity of these islands of protein stability. Even with the current domain structures of the N-terminal region, phosphorylation domain and SPRY2, the binding sites of RyR ligands, including: Ca2+, Mg2+, ruthenium red,  113 FKBP12/12.6, ATP, dantrolene, ryanodine, caffeine and DHPR II-III loop are still not known. Sites that have previously been proposed have also never been independently confirmed.143,221,249,307,308 Calmodulin remains one exception where there is both structural and biochemical evidence. However, the field is still debating about its exact binding site almost 20 years after the discovery of its role in RyR modulation.   4.1 Calmodulin Can Bind More than One Site, but What Site?  The work presented in this thesis covered the binding analysis of three separate CaMBDs that are located within the RyR. The rationale was to provide clear biochemical evidence of either binding or absence of binding by CaM to these regions. There has been great controversy within the field as to the number of sites that CaM might act on within RyR. Although cryo-EM suggests that 4 CaM molecules bind to equivalent sites within an intact channel, previous reports have shown that there are up to 9 different possible binding sites.70  I determined that indeed more than one site could bind both apoCaM and Ca2+/CaM. Importantly, there are differences between CaM binding to homologous regions of different isoforms, suggesting that these could underlie the different regulation of RyR1 and RyR2. Although it is well established now that the 3614-3643 site, or CaMBD2, is an important site, it does not exclude the potential contributions of other CaMBDs. A key question is now whether the identified CaMBDs can also bind within the context of intact channels. In isolation, peptides and RyR domains that contain CaMBDs are made accessible and can interact with CaM. However, within the folded channels these sites may not be available. This is already apparent in the phosphorylation domain, which contains three putative CaMBDs, none of which are accessible when mapped on the structure. During the  114 preparation of Chapter 2, I tested CaM binding to this domain which could not be detected by ITC, in agreement with the fact that they are buried within the domain.93  One way to answer this question is through the use of purified channels. Recombinant RyR channels with mutations or deletions in CaMBDs can be generated. By making specific mutants within CaMBDs that prevent CaM binding, we can eliminate known sites (CaMBD2) and examine the contributions to binding by other CaMBDs. Furthermore, mutations can also be made in CaMBD3 to analyze its contribution to CaM binding, since it is the strongest apoCaM binder and because it is adjacent to a region that has recently been shown to be surface exposed.236 These channels would then have to be purified in high quantity for binding experiments through ITC or a complementary technique like SPR. However, the yields from heterologous expression are typically very low, so an alternative option is to monitor the functional effect of CaM on the CaMBD mutants. Using planar lipid bilayer electrophysiology, measurements of RyR channel opening and closing can be observed.   Due to RyR’s high affinity for both apoCaM and Ca2+/CaM one could try photo-crosslinking experiments to further localize its binding sites. Affinity tagged-CaM could be labelled with widely available compounds such as biotin-hydrazone and bound to RyR at different Ca2+ levels. Regions close to CaM would be cross-linked to it upon UV light exposure. The channel can then be subjected to limited proteolysis and the fragments that are cross-linked to CaM enriched using the biotin affinity tag. CaM is extremely stable even in the presence of proteases. The fragment can then be sequenced by mass spectrometry to provide a map of nearby residues. This would further aid the localization of CaM.    115 4.2 Calmodulin CPVT Mutants and CaMBDs  Recently, CaM missense mutants, two of which are N98S and N54I, have been isolated that cause CPVT.57 There are conflicting reports showing that the CPVT mutants have either impaired Ca2+ binding or improved affinity compared to WT-CaM.56,57 Functionally, the mutants become stimulatory instead of inhibitory towards RyR at high Ca2+ levels. This presents a conundrum as it poses the question as to whether CaM’s binding affinity to RyR2 translates to either an activating or inhibiting effect. To test whether binding affinity has changed due to the CaM mutants, a recent undergraduate, Stephanie Ng, and I have cloned and expressed the CaM mutants and their associated lobes (N-lobe for N54I and C-lobe for N98S).  In preliminary experiments, we could not detect a difference in binding affinity to CaMBD2 between the mutants versus the wild-type. However, further follow-up experiments are needed. If binding affinity is affected, it would be critical to determine the binding mode to CaMBD2. As our mutational analysis with CaMBD2 revealed, Ca2+/N-lobe is able to bind both of the anchor residues. It has also been proposed that the lobes themselves may be responsible for the different effects of CaM: the C-lobe is inhibitory while the N-lobe is activating.225 CPVT mutations in CaM may be affecting the way the lobes are interacting with CaMBD2’s anchor points. As one group also reported no difference in mutant CaM binding affinity to CaMBD2, the mutations may instead be tuning CaM’s affinity for the other CaMBDs.   4.3 Calmodulin and S100A1  It is accepted that at high Ca2+ levels CaM inhibits and S100A1 enhances RyR activity. Both of these proteins are also thought to bind the same regulatory site, CaMBD2. Little  116 work has been done to further understand how these proteins, both of which have high affinities for CaMBD2, can compete for the same site.204 A criticism of current S100A1 experiments with CaMBD2 is that they do not use the entire region, but just a fragment that contains a canonical S100 binding sequence. What is crucial is the fact that S100A1, as shown from NMR structures, only binds via the Trp in CaMBD2 as the Phe anchor is not present. Using ITC, CaMBD2 mutants that lack the Trp or Phe anchor could be used to test if S100A1 prefers binding to the Trp as observed by NMR. How the two proteins are able to compete for the same site can also be tested through competition experiments where S100A1 or CaM is pre-incubated with CaMBD2 and is titrated with the competing protein to determine the apparent affinity. Finally, an alternative possibility is that both proteins can bind simultaneously, with the S100A1 binding via the Trp and CaM binding through the Phe.  4.4 Do CaMBDs Serve as Internal EF-Hand Binding Sites?   An overlooked component in CaM modulation of RyR is the functional role of the predicted EF-hands in the RyR. Classically, these domains have been thought of as simple Ca2+ sensors that would undergo conformational changes upon binding Ca2+ increasing RyR channel activity. Only recently has it been proposed that EF-hands may transduce this effect through more CaM-like means.123,228,309 The EF-hands are predicted to be folded and functional and bind CaMBD2. Two reports suggest an interplay between the EF-hands, CaM and CaMBDs.123,228 They suggest that the interactions between EF-hands and CaMBD2, depending on channel state, promote either its inhibition or activation. CaM competition with the EF-hands for CaMBD2 would disrupt this link. In addition, the EF-hands are capable of binding to additional CaMBDs in other proteins.  117 A previous graduate student in the lab, Paolo Lobo, has tested an EF-hand construct in RyR1, 4071-4138, and tested binding to the CaMBD3LRR mutant. Surprisingly, in high Ca2+ the EF-hand showed binding to the peptide with a Kd  ~ 7 µM. I have now cloned and purified additional EF-hand constructs in RyR2 and RyR3 for further biochemical characterization. The proteins are highly soluble but do not crystallize. Presently, others are attempting to determine the structure through mutagenesis to reduce the flexibility of surface residues and to eliminate charged residues that may reduce protein solubility. Further binding experiments of the EF hands with CaMBDs are also forthcoming.  4.4 SPRY1 and 3 Domains of RyR   With the successful determination of the structure of SPRY2, another significant piece of the RyR puzzle has been obtained. SPRY2 being one of three repeats can provide clues in solving the other two domains. Structural alignments of SPRY2 with the predicted stretches of SPRY1 and SPRY3 reveal large sequence differences within the N-terminus region and the loops connecting β-strands. (Figure 3.6) These differences in structural elements suggest that multiple approaches may need to be used to crystallize these remaining domains. As there are insertions in loops, which extend long stretches in SPRY2, constructs that delete portions of the loops predicted to be highly disordered may aid in crystallization.  The N- and C- terminus structural elements, which are novel in the SPRY2 structure, may also be present in the SPRY1 and SPRY3 domains. Evolutionarily it is much more likely that the three repeats were all derived through duplication from an initial SPRY domain. It is possible that through time significant changes have occurred in these modules due to the different pressures faced by SPRY1 and SPRY3. However, the structural feature with the addition of a short β strand and an α helix at the N-terminus and a lid covering hydrophobic  118 residues at the C-terminus may still be present in these SPRYs. New constructs that have variable termini may aid in finding a well-behaved domain amenable to crystallization.  4.5 Tandem Domains  Having defined boundaries of a well-behaved SPRY2 construct, one can systematically add increasing amounts of sequence to either terminus to attempt to generate larger constructs. These tandem domains are beneficial in many ways. First, RyRs are built up by independent domains that can have intricate interdomain interactions and understanding their arrangement in the superstructure is crucial. An example is the N-terminal region, consisting of three domains that adopt a compact triangular arrangement. Second, having a larger construct increases the reliability of docking into cryo-EM maps. Although SPRY2 was placed with a high docking contrast for a protein of its size, there are concerns that proteins with similar folds (ie. SPRY1 or SPRY33) may also dock to the same location. With Laplacian filtering, the edges of an image or body are enhanced. The more discrete a protein structure is, the more reliable the dock. For example, a small globular protein with limited surface features may resemble many domains observed on a cryo-EM map resulting in a low docking contrast. A small protein with a unique extension or a large multi-domain construct would provide enough details to fit the contours of a map more reliably. Finally, larger constructs may provide the necessary interdomain surfaces that mediate the plethora of RyR interactions that have not been confirmed. In the case of SPRY2, a larger construct may allow for detection of binding to the DHPR II-III loop by ITC.   Attempts have been made to extend SPRY2 further from both ends. Holding the C-terminus fixed, I have extended it N-terminally to include the first predicted tandem repeat  119 and SPRY1 domain that are present before SPRY2 to create Repeat1-SPRY2 or SPRY1-Repeat1-SPRY2 constructs (Figure 1.3). The region C-terminal to SPRY2 is predicted to have the propensity to form a small domain of approximately 50-70 amino acids that is composed of β-strands. I have systematically added residues in groups of 10. Beyond this subdomain lies a region of divergence, divergent region 2 (DR2). DR2 may play a role in EC coupling, DHPR II-III loop binding and tetrad formation.84,251,310 I have attempted to express this region alone but have been unsuccessful at obtaining well-behaved pure protein. I am now extending SPRY2 into the DR2 region with the hope of obtaining a stable longer construct.   The next domain after DR2 is SPRY3 and attempts are also being made into successfully expressing a SPRY2-DR2-SPRY3 construct. If this is unsuccessful an interesting possibility to generate a deletion mutant is available. In RyR3, which is expressed widely in all tissues, there is no DR2 suggesting that it is not crucial at least in this isoform. Constructs in RyR3 or deletions of the DR2 region in RyR1 and RyR2 may be possible in producing a soluble SPRY2-SPRY3 tandem domain construct.    4.6 FRET Localization of SPRY2  A recent development in the field is using FRET to confirm the location of docked atomic structures of RyR.311 Based upon a docking, His insertions into the flexible loops of a protein domain are made in full-length RyR. The His insertions can bind to an NTA dye which can participate in FRET with a donor fluorophore attached to a protein at a known location on RyR such as FKBP.311 Through the FRET efficiencies and the known locations of the donor fluorophore, distance restraints can be calculated to triangulate the location of the acceptor on the domain. The docking solution provides several locations on SPRY2 that  120 may be amenable to the insertion of a His sequence. We are currently planning for insertions to be placed in loops facing the two solvent channels (Gly1185 and His1127) according to our proposed SPRY2 docking site, as well as an insertion in a loop exposed to the lateral surface (Arg1212).   4.7 Conclusions  The work here has only shone a small laser light on to the vast Ryanodine Receptor. Using ITC, thermodynamics of binding of CaM have now been described to three different potential CaMBD on the channel. Importantly, it reveals a possible new apoCaM binding site that is the target for disease-causing mutations. It remains to be seen what the roles are for each of these sites in regulating RyR.  The SPRY2 structure from RyR1 and RyR2 provides the first evidence that it may play a role in channel gating as it is located between the central vestibule formed by the N-terminal region (domains ABC) and the clamp region. The lack of binding to the DHPR II-III loop also suggests that the interaction is located elsewhere or requires a surface that is formed by multiple domains.               121 References 1. Berridge, M. J., Bootman, M. D. & Roderick, H. L. Calcium signalling: dynamics, homeostasis and remodelling. Nature Reviews Molecular Cell Biology 4, 517–529 (2003). 2. Case, R. M. et al. Evolution of calcium homeostasis: from birth of the first cell to an omnipresent signalling system. Cell Calcium 42, 345–350 (2007). 3. Jaiswal, J. K. Calcium - how and why? J. Biosci. 26, 357–363 (2001). 4. Clapham, D. E. Calcium Signaling. Cell 131, 1047–1058 (2007). 5. Hopkins, P. M. Skeletal muscle physiology. Contin Educ Anaesth Crit Care Pain 6, 1–6 (2006). 6. Dulhunty, A. F. Excitation-Contract Coupling From the 1950s Into The New Millenium. Clin. Exp. Pharmacol. Physiol. 33, 763–772 (2006). 7. Sandow, A. Excitation-contraction coupling in muscular response. The Yale Journal of Biology and Medicine 25, 176–201 (1952). 8. Adrian, R. H., Costantin, L. L. & Peachey, L. D. Radial spread of contraction in frog muscle fibres. J Physiol 204, 231–257 (1969). 9. Eisenberg, R. S. & Gage, P. W. Ionic Conductances of the Surface and Transverse Tubular Membranes of Frog Sartorius Fibres. The Journal of General Physiology 53, 279–297 (1969). 10. Dulhunty, A. F. & Gage, P. W. Electrical properties of toad sartorius muscle fibres in summer and winter. J Physiol 230, 619–641 (1973). 11. Rios, E. & Brum, G. Involvement of dihydropyridine receptors in excitation–contraction coupling in skeletal muscle. Nature 325, 717–720 (1987). 12. Lamb, G. D. Asymmetric charge movement in polarized and depolarized muscle fibres of the rabbit. J Physiol 383, 349–367 (1987). 13. Ríos, E. & Pizarro, G. Voltage sensor of excitation-contraction coupling in skeletal muscle. Physiol Rev 71, 849–908 (1991). 14. Fabiato, A. & Fabiato, F. Calcium release from the sarcoplasmic reticulum. Circ Res 40, 119–129 (1977). 15. Meissner, G. Ryanodine activation and inhibition of the Ca2+ release channel of sarcoplasmic reticulum. J Biol Chem 261, 6300–6306 (1986). 16. Smith, J. S., Coronado, R. & Meissner, G. Sarcoplasmic reticulum contains adenine nucleotide-activated calcium channels. Nature 316, 446–449 (1985). 17. Fleischer, S. & Ogunbunmi, E. M. Localization of Ca2+ release channels with ryanodine in junctional terminal cisternae of sarcoplasmic reticulum of fast skeletal muscle. Proc Natl Acad Sci USA 82, 7526–7259 (1985). 18. Caswell, A. H. & Brandt, N. R. Does muscle activation occur by direct mechanical coupling of transverse tubules to sarcoplasmic reticulum? Trends in Biochemical Sciences 14, 161–165 (1989). 19. Takeshima, H. et al. Primary structure and expression from complementary DNA of skeletal muscle ryanodine receptor. Nature 339, 439–445 (1989). 20. Huxley, A. F. & Niedergerke, R. Structural changes in muscle during contraction. Nature 173, 971–973 (1954). 21. Franzini-Armstrong, C. & Protasi, F. Ryanodine receptors of striated muscles: a complex channel capable of multiple interactions. Physiol Rev 77, 699–729 (1997). 22. Ferguson, D. G., Schwartz, H. W. & Franzini-Armstrong, C. Subunit structure of junctional feet in triads of skeletal muscle: a freeze-drying, rotary-shadowing study. J  122 Cell Biol 99, 1735–1742 (1984). 23. Dias, J. M., Szegedi, C., Jona, I. & Vogel, P. D. Insights into the regulation of the ryanodine receptor: differential effects of Mg2+ and Ca2+ on ATP binding. 45, 9408–9415 (2006). 24. Lai, F. A., Erickson, H. P., Rousseau, E., Liu, Q. Y. & Meissner, G. Purification and reconstitution of the calcium release channel from skeletal muscle. Nature 331, 315–319 (1988). 25. Lindsay, A. R., Manning, S. D. & Williams, A. J. Monovalent cation conductance in the ryanodine receptor-channel of sheep cardiac muscle sarcoplasmic reticulum. J Physiol 439, 463–480 (1991). 26. Smith, J. S. et al. Purified ryanodine receptor from rabbit skeletal muscle is the calcium-release channel of sarcoplasmic reticulum. The Journal of General Physiology 92, 1–26 (1988). 27. Tinker, A. & Williams, A. J. Divalent cation conduction in the ryanodine receptor channel of sheep cardiac muscle sarcoplasmic reticulum. The Journal of General Physiology 100, 479–493 (1992). 28. Otsu, K. et al. Molecular cloning of cDNA encoding the Ca2+ release channel (ryanodine receptor) of rabbit cardiac muscle sarcoplasmic reticulum. J Biol Chem 265, 13472–13483 (1990). 29. Hakamata, Y., Nakai, J., Takeshima, H. & Imoto, K. Primary structure and distribution of a novel ryanodine receptor/calcium release channel from rabbit brain. FEBS letters 312, 229–235 (1992). 30. Zorzato, F. et al. Molecular cloning of cDNA encoding human and rabbit forms of the Ca2+ release channel (ryanodine receptor) of skeletal muscle sarcoplasmic reticulum. J Biol Chem 265, 2244–2256 (1990). 31. Tunwell, R. E. et al. The human cardiac muscle ryanodine receptor-calcium release channel: identification, primary structure and topological analysis. Biochem J 318 ( Pt 2), 477–487 (1996). 32. Leeb, T. & Brenig, B. cDNA cloning and sequencing of the human ryanodine receptor type 3 (RYR3) reveals a novel alternative splice site in the RYR3 gene. FEBS letters 423, 367–370 (1998). 33. Ottini, L., Marziali, G., Conti, A., Charlesworth, A. & Sorrentino, V. Alpha and beta isoforms of ryanodine receptor from chicken skeletal muscle are the homologues of mammalian RyR1 and RyR3. Biochem J 315, 207–216 (1996). 34. Oyamada, H. et al. Primary structure and distribution of ryanodine-binding protein isoforms of the bullfrog skeletal muscle. J Biol Chem 269, 17206–17214 (1994). 35. Maryon, E. B., Coronado, R. & Anderson, P. unc-68 encodes a ryanodine receptor involved in regulating C. elegans body-wall muscle contraction. J Cell Biol 134, 885–893 (1996). 36. Takeshima, H. et al. Isolation and characterization of a gene for a ryanodine receptor/calcium release channel in Drosophila melanogaster. FEBS letters 337, 81–87 (1994). 37. Takeshima, H. et al. Excitation-contraction uncoupling and muscular degeneration in mice lacking functional skeletal muscle ryanodine-receptor gene. Nature 369, 556–559 (1994). 38. Takeshima, H., Komazaki, S. & Hirose, K. Embryonic lethality and abnormal cardiac myocytes in mice lacking ryanodine receptor type 2. EMBO J 17, 3309–3316 (1998).  123 39. Balschun, D. et al. Deletion of the ryanodine receptor type 3 (RyR3) impairs forms of synaptic plasticity and spatial learning. EMBO J 18, 5264–5273 (1999). 40. Mickelson, J. R. & Louis, C. F. Malignant hyperthermia: excitation-contraction coupling, Ca2+ release channel, and cell Ca2+ regulation defects. Physiol Rev 76, 537–592 (1996). 41. Klingler, W. et al. Functional and genetic characterization of clinical malignant hyperthermia crises: a multi-centre study. Orphanet J Rare Dis 9, 8 (2014). 42. Monnier, N., Procaccio, V., Stieglitz, P. & Lunardi, J. Malignant-hyperthermia susceptibility is associated with a mutation of the alpha 1-subunit of the human dihydropyridine-sensitive L-type voltage-dependent calcium-channel receptor in skeletal muscle. Am J Hum Genet 60, 1316–1325 (1997). 43. Ørding, H. Pathophysiology of malignant hyperthermia. Ann Fr Anesth Reanim 8, 411–416 (1989). 44. Monnier, N. et al. Presence of two different genetic traits in malignant hyperthermia families: implication for genetic analysis, diagnosis, and incidence of malignant hyperthermia susceptibility. Anesthesiology 97, 1067–1074 (2002). 45. Strazis, K. P. & Fox, A. W. Malignant hyperthermia: a review of published cases. Anesthesia & Analgesia 77, 297–304 (1993). 46. Parness, J. & Palnitkar, S. S. Identification of dantrolene binding sites in porcine skeletal muscle sarcoplasmic reticulum. J Biol Chem 270, 18465–18472 (1995). 47. Zhao, F., Li, P., Chen, S. R., Louis, C. F. & Fruen, B. R. Dantrolene inhibition of ryanodine receptor Ca2+ release channels. Molecular mechanism and isoform selectivity. J Biol Chem 276, 13810–13816 (2001). 48. Zhang, Y. et al. A mutation in the human ryanodine receptor gene associated with central core disease. Nature genetics 5, 46–50 (1993). 49. Robinson, R., Carpenter, D., Shaw, M. A., Halsall, J. & Hopkins, P. Mutations in RYR1 in malignant hyperthermia and central core disease. Hum. Mutat. 27, 977–989 (2006). 50. Kossugue, P. M. et al. Central core disease due to recessive mutations in RYR1 gene: Is it more common than described? Muscle & Nerve 35, 670–674 (2007). 51. Priori, S. G. et al. Clinical and molecular characterization of patients with catecholaminergic polymorphic ventricular tachycardia. Circulation 106, 69–74 (2002). 52. Kontula, K. et al. Catecholaminergic polymorphic ventricular tachycardia: recent mechanistic insights. Cardiovascular Research 67, 379–387 (2005). 53. Houser, S. R. Role of RyR2 Phosphorylation in Heart Failure and Arrhythmias: Protein Kinase A-Mediated Hyperphosphorylation of the Ryanodine Receptor at Serine 2808 Does Not Alter Cardiac Contractility or Cause Heart Failure and Arrhythmias. Circ Res 114, 1320–1327 (2014). 54. Lahat, H. et al. A missense mutation in a highly conserved region of CASQ2 is associated with autosomal recessive catecholamine-induced polymorphic ventricular tachycardia in Bedouin families from Israel. Am J Hum Genet 69, 1378–1384 (2001). 55. Roux-Buisson, N. et al. Absence of triadin, a protein of the calcium release complex, is responsible for cardiac arrhythmia with sudden death in human. Human Molecular Genetics 21, 2759–2767 (2012). 56. Hwang, H.-S. et al. Divergent regulation of ryanodine receptor 2 calcium release channels by arrhythmogenic human calmodulin missense mutants. Circ Res 114, 1114–1124 (2014). 57. Nyegaard, M. et al. Mutations in Calmodulin Cause Ventricular Tachycardia and  124 Sudden Cardiac Death. The American Journal of Human Genetics 91, 703–712 (2012). 58. Avila, G., O'Brien, J. J. & Dirksen, R. T. Excitation–contraction uncoupling by a human central core disease mutation in the ryanodine receptor. Proc Natl Acad Sci USA 98, 4215–4220 (2001). 59. Lehnart, S. E. et al. Sudden Death in Familial Polymorphic Ventricular Tachycardia Associated With Calcium Release Channel (Ryanodine Receptor) Leak. Circulation 109, 3208–3214 (2004). 60. Yano, M. Ryanodine receptor as a new therapeutic target of heart failure and lethal arrhythmia. Circ J 72, 509–514 (2008). 61. Hamilton, S. Ryanodine receptors. Cell Calcium 38, 253–260 (2005). 62. Yuchi, Z. & Van Petegem, F. Common allosteric mechanisms between ryanodine and inositol-1,4,5-trisphosphate receptors. Channels 5, 120–123 (2011). 63. Seo, M.-D. et al. Structural and functional conservation of key domains in InsP3 and ryanodine receptors. Nature 483, 108–112 (2012). 64. Franzini-Armstrong, C. Studies of the Triad IV. Structure of the Junction in Frog Twitch Fibres. J Cell Biol 47, 488–499 (1970). 65. Protasi, F., Franzini-Armstrong, C. & Allen, P. Role of ryanodine receptors in the assembly of calcium release units in skeletal muscle. J Cell Biol 140, 831–842 (1998). 66. Block, B. A., Imagawa, T., Campbell, K. P. & Franzini-Armstrong, C. Structural evidence for direct interaction between the molecular components of the transverse tubule/sarcoplasmic reticulum junction in skeletal muscle. J Cell Biol 107, 2587–2600 (1988). 67. Radermacher, M. et al. Cryo-electron microscopy and three-dimensional reconstruction of the calcium release channel/ryanodine receptor from skeletal muscle. J Cell Biol 127, 411–423 (1994). 68. Serysheva, I. I., Orlova, E. V., Sherman, M. B., Hamilton, S. L. & van Heel, M. Electron cryomicroscopy and angular reconstitution used to visualize the skeletal muscle calcium release channel. Nat Struct Biol 2, 18–24 (1995). 69. Samsó, M., Feng, W., Pessah, I. N. & Allen, P. D. Coordinated movement of cytoplasmic and transmembrane domains of RyR1 upon gating. PLoS Biol. 7, e85 (2009). 70. Samso, M. & Wagenknecht, T. Apocalmodulin and Ca 2+-calmodulin bind to neighboring locations on the ryanodine receptor. Journal of Biological Chemistry 277, 1349–1353 (2002). 71. Wagenknecht, T., Berkowitz, J., Grassucci, R., Timerman, A. P. & Fleischer, S. Localization of calmodulin binding sites on the ryanodine receptor from skeletal muscle by electron microscopy. Biophys J 67, 2286–2295 (1994). 72. Sonoda, Y. et al. Tricks of the trade used to accelerate high-resolution structure determination of membrane proteins. FEBS letters 584, 2539–2547 (2010). 73. Mus-Veteau, I. Heterologous expression of membrane proteins for structural analysis. Methods Mol. Biol. 601, 1–16 (2010). 74. Serysheva, I. I. et al. Subnanometer-resolution electron cryomicroscopy-based domain models for the cytoplasmic region of skeletal muscle RyR channel. Proc Natl Acad Sci USA 105, 9610–9615 (2008). 75. Samsó, M., Wagenknecht, T. & Allen, P. D. Internal structure and visualization of transmembrane domains of the RyR1 calcium release channel by cryo-EM. 12, 539–544 (2005). 76. Ludtke, S. J., Serysheva, I. I., Hamilton, S. L. & Chiu, W. The pore structure of the  125 closed RyR1 channel. Structure 13, 1203–1211 (2005). 77. Sharma, M. R., Jeyakumar, L. H., Fleischer, S. & Wagenknecht, T. Three-dimensional structure of ryanodine receptor isoform three in two conformational states as visualized by cryo-electron microscopy. J Biol Chem 275, 9485–9491 (2000). 78. Liu, Z. et al. Three-dimensional reconstruction of the recombinant type 3 ryanodine receptor and localization of its amino terminus. Proc Natl Acad Sci USA 98, 6104–6109 (2001). 79. Sharma, M., Jeyakumar, L., Fleischer, S. & Wagenknecht, T. Three-dimensional visualization of FKBP12. 6 binding to an open conformation of cardiac ryanodine receptor. Biophys J 90, 164–172 (2006). 80. Wagenknecht, T. et al. Locations of calmodulin and FK506-binding protein on the three-dimensional architecture of the skeletal muscle ryanodine receptor. J Biol Chem 272, 32463–32471 (1997). 81. Samso, M., Trujillo, R. & Gurrola, G. Three-dimensional location of the imperatoxin A binding site on the ryanodine receptor. J Cell Biol (1999). 82. Liu, Z., Zhang, J., Li, P., Chen, S. R. W. & Wagenknecht, T. Three-dimensional reconstruction of the recombinant type 2 ryanodine receptor and localization of its divergent region 1. J Biol Chem 277, 46712–46719 (2002). 83. Zhang, J. et al. Three-dimensional localization of divergent region 3 of the ryanodine receptor to the clamp-shaped structures adjacent to the FKBP binding sites. J Biol Chem 278, 14211–14218 (2003). 84. Liu, Z., Zhang, J., Wang, R., Wayne Chen, S. R. & Wagenknecht, T. Location of divergent region 2 on the three-dimensional structure of cardiac muscle ryanodine receptor/calcium release channel. J Mol Biol 338, 533–545 (2004). 85. Jones, P. P. et al. Localization of PKA phosphorylation site, Ser2030, in the three-dimensional structure of cardiac ryanodine receptor. Biochem J 410, 261 (2008). 86. Wang, R. et al. Localization of an NH(2)-terminal disease-causing mutation hot spot to the ‘clamp’ region in the three-dimensional structure of the cardiac ryanodine receptor. J Biol Chem 282, 17785–17793 (2007). 87. Benacquista, B. L. et al. Amino acid residues 4425-4621 localized on the three-dimensional structure of the skeletal muscle ryanodine receptor. Biophys J 78, 1349–1358 (2000). 88. Perálvarez-Marín, A. et al. 3D Mapping of the SPRY2 Domain of Ryanodine Receptor 1 by Single-Particle Cryo-EM. PLoS ONE 6, e25813 (2011). 89. Hamilton, S. L. & Serysheva, I. I. Ryanodine receptor structure: progress and challenges. J Biol Chem 284, 4047–4051 (2009). 90. Amador, F. J. et al. Crystal structure of type I ryanodine receptor amino-terminal beta-trefoil domain reveals a disease-associated mutation ‘hot spot’ loop. Proc Natl Acad Sci USA 106, 11040–11044 (2009). 91. Lobo, P. A. & Van Petegem, F. Crystal Structures of the N-Terminal Domains of Cardiac and Skeletal Muscle Ryanodine Receptors: Insights into Disease Mutations. Structure 17, 1505–1514 (2009). 92. Tung, C.-C., Lobo, P. A., Kimlicka, L. & Van Petegem, F. The amino-terminal disease hotspot of ryanodine receptors forms a cytoplasmic vestibule. Nature 468, 585–588 (2010). 93. Yuchi, Z., Lau, K. & Van Petegem, F. Disease mutations in the ryanodine receptor central region: crystal structures of a phosphorylation hot spot domain. Structure 20, 1201–1211 (2012).  126 94. Maximciuc, A. A., Putkey, J. A., Shamoo, Y. & Mackenzie, K. R. Complex of calmodulin with a ryanodine receptor target reveals a novel, flexible binding mode. Structure 14, 1547–1556 (2006). 95. Prosser, B. L. et al. S100A1 Binds to the Calmodulin-binding Site of Ryanodine Receptor and Modulates Skeletal Muscle Excitation-Contraction Coupling. J Biol Chem 283, 5046–5057 (2007). 96. Kimlicka, L. et al. The Cardiac Ryanodine Receptor N-Terminal Region Contains an Anion Binding Site that Is Targeted by Disease Mutations. Structure 21, 1440–1449 (2013). 97. Kimlicka, L., Lau, K., Tung, C.-C. & Van Petegem, F. Disease mutations in the ryanodine receptor N-terminal region couple to a mobile intersubunit interface. Nat Commun 4, 1506 (2013). 98. Zhong, X. et al. Conformational Dynamics inside Amino-Terminal Disease Hotspot of Ryanodine Receptor. Structure 21, 1–10 (2013). 99. Bhuiyan, Z. A. et al. Expanding Spectrum of Human RYR2-Related Disease New Electrocardiographic, Structural, and Genetic Features. Circulation 116, 1569–1576 (2007). 100. Marjamaa, A. et al. Search for cardiac calcium cycling gene mutations in familial ventricular arrhythmias resembling catecholaminergic polymorphic ventricular tachycardia. BMC Medical Genetics 10, 12 (2009). 101. Lobo, P. A., Kimlicka, L., Tung, C.-C. & Van Petegem, F. The deletion of exon 3 in the cardiac ryanodine receptor is rescued by β strand switching. Structure 19, 790–798 (2011). 102. Amador, F. J. et al. Type 2 ryanodine receptor domain A contains a unique and dynamic α-helix that transitions to a β-strand in a mutant linked with a heritable cardiomyopathy. J Mol Biol 425, 4034–4046 (2013). 103. Taylor, C., Da Fonseca, P. & Morris, E. IP3 receptors: the search for structure. Trends in Biochemical Sciences 29, 210–219 (2004). 104. Ponting, C. P. Novel repeats in ryanodine and IP3 receptors and protein O-mannosyltransferases. Trends in Biochemical Sciences 25, 48–50 (2000). 105. Ryanodine receptor calcium release channels: an evolutionary perspective. 740, 159–182 (2012). 106. Sharma, P. et al. Structural determination of the phosphorylation domain of the ryanodine receptor. FEBS J. 279, 3952–3964 (2012). 107. Marx, S. O. et al. PKA phosphorylation dissociates FKBP12.6 from the calcium release channel (ryanodine receptor): defective regulation in failing hearts. Cell 101, 365–376 (2000). 108. Marx, S. O. et al. Phosphorylation-dependent regulation of ryanodine receptors: a novel role for leucine/isoleucine zippers. J Cell Biol 153, 699–708 (2001). 109. Respress, J. L. et al. Role of RyR2 Phosphorylation at S2814 During Heart Failure Progression. Circ Res 110, 1474–1483 (2012). 110. Valdivia, H. H. Ryanodine Receptor Phosphorylation and Heart Failure Phasing Out S2808 and ‘Criminalizing’ S2814. Circ Res 110, 1398–1402 (2012). 111. George, C. H., Higgs, G. V. & Lai, F. A. Ryanodine receptor mutations associated with stress-induced ventricular tachycardia mediate increased calcium release in stimulated cardiomyocytes. Circ Res 93, 531–540 (2003). 112. Zhang, H. et al. Hyperphosphorylation of the cardiac ryanodine receptor at serine  127 2808 is not involved in cardiac dysfunction after myocardial infarction. Circ Res 110, 831–840 (2012). 113. Xiao, B. et al. Characterization of a novel PKA phosphorylation site, serine-2030, reveals no PKA hyperphosphorylation of the cardiac ryanodine receptor in canine heart failure. Circ Res 96, 847–855 (2005). 114. Xiao, B. et al. Ser-2030, but not Ser-2808, is the major phosphorylation site in cardiac ryanodine receptors responding to protein kinase A activation upon beta-adrenergic stimulation in normal and failing hearts. Biochem J 396, 7–16 (2006). 115. Ponting, C., Schultz, J. & Bork, P. SPRY domains in ryanodine receptors (Ca(2+)-release channels). Trends in Biochemical Sciences 22, 193–194 (1997). 116. D'Cruz, A. A., Babon, J. J., Norton, R. S., Nicola, N. A. & Nicholson, S. E. Structure and function of the SPRY/B30.2 domain proteins involved in innate immunity. Protein Science 22, 1–10 (2012). 117. Perfetto, L. et al. Exploring the diversity of SPRY/B30.2-mediated interactions. Trends in Biochemical Sciences 38, 38–46 (2013). 118. Tae, H. et al. The elusive role of the SPRY2 domain in RyR1. Channels 5, 148–160 (2011). 119. Leong, P. & MacLennan, D. H. A 37-Amino Acid Sequence in the Skeletal Muscle Ryanodine Receptor Interacts with the Cytoplasmic Loop between Domains II and III in the Skeletal Muscle Dihydropyridine Receptor. J Biol Chem 273, 7791–7794 (1998). 120. Cui, Y. et al. A dihydropyridine receptor alpha1s loop region critical for skeletal muscle contraction is intrinsically unstructured and binds to a SPRY domain of the type 1 ryanodine receptor. The International Journal of Biochemistry & Cell Biology 41, 677–686 (2009). 121. Bers, D. M. Calmodulin and S100A1 fine tune skeletal muscle E-C coupling. Focus on "Modulation of sarcoplasmic reticulum Ca2+ release in skeletal muscle expressing ryanodine receptor impaired in regulation by calmodulin and S100A1". Am J Physiol, Cell Physiol 300, C974–5 (2011). 122. Gifford, J. L., Walsh, M. P. & Vogel, H. J. Structures and metal-ion-binding properties of the Ca2+-binding helix-loop-helix EF-hand motifs. Biochem J 405, 199–221 (2007). 123. Xiong, L., Zhang, J.-Z., He, R. & Hamilton, S. L. A Ca2+-binding domain in RyR1 that interacts with the calmodulin binding site and modulates channel activity. Biophys J 90, 173–182 (2006). 124. Gangopadhyay, J. P. & Ikemoto, N. Aberrant interaction of calmodulin with the ryanodine receptor develops hypertrophy in the neonatal cardiomyocyte. Biochem J 438, 379–387 (2011). 125. Xin, H. B., Timerman, A. P., Onoue, H., Wiederrecht, G. J. & Fleischer, S. Affinity purification of the ryanodine receptor/calcium release channel from fast twitch skeletal muscle based on its tight association with FKBP12. Biochem Biophys Res Commun 214, 263–270 (1995). 126. Futer, O., DeCenzo, M. T., Aldape, R. A. & Livingston, D. J. FK506 binding protein mutational analysis. Defining the surface residue contributions to stability of the calcineurin co-complex. J Biol Chem 270, 18935–18940 (1995). 127. Ahern, G. P., Junankar, P. R. & Dulhunty, A. F. Subconductance states in single-channel activity of skeletal muscle ryanodine receptors after removal of FKBP12. Biophys J 72, 146–162 (1997).  128 128. Marx, S. O. & Marks, A. R. Dysfunctional ryanodine receptors in the heart: new insights into complex cardiovascular diseases. J Mol Cell Cardiol 58, 225–231 (2013). 129. Shou, W. et al. Cardiac defects and altered ryanodine receptor function in mice lacking FKBP12. Nature 391, 489–492 (1998). 130. Marx, S. O., Ondrias, K. & Marks, A. R. Coupled gating between individual skeletal muscle Ca2+ release channels (ryanodine receptors). Science 281, 818–821 (1998). 131. Samsó, M., Shen, X. & Allen, P. Structural characterization of the RyR1–FKBP12 interaction. J Mol Biol (2006). 132. Seiler, S., Wegener, A. D., Whang, D. D., Hathaway, D. R. & Jones, L. R. High molecular weight proteins in cardiac and skeletal muscle junctional sarcoplasmic reticulum vesicles bind calmodulin, are phosphorylated, and are degraded by Ca2+-activated protease. J Biol Chem 259, 8550–8557 (1984). 133. Suko, J. et al. Phosphorylation of serine 2843 in ryanodine receptor-calcium release channel of skeletal muscle by cAMP-, cGMP- and CaM-dependent protein kinase. Biochim Biophys Acta 1175, 193–206 (1993). 134. Takasago, T., Imagawa, T., Furukawa, K., Ogurusu, T. & Shigekawa, M. Regulation of the cardiac ryanodine receptor by protein kinase-dependent phosphorylation. J Biochem 109, 163–170 (1991). 135. Wehrens, X. H. T. et al. Ryanodine receptor/calcium release channel PKA phosphorylation: A critical mediator of heart failure progression. Proc Natl Acad Sci USA 103, 511–518 (2006). 136. Witcher, D. R., Kovacs, R. J., Schulman, H., Cefali, D. C. & Jones, L. R. Unique phosphorylation site on the cardiac ryanodine receptor regulates calcium channel activity. J Biol Chem 266, 11144–11152 (1991). 137. Wehrens, X. H. T., Lehnart, S. E., Reiken, S. R. & Marks, A. R. Ca2+/Calmodulin-Dependent Protein Kinase II Phosphorylation Regulates the Cardiac Ryanodine Receptor. Circ Res 94, e61–e70 (2004). 138. Huttlin, E. L. et al. A tissue-specific atlas of mouse protein phosphorylation and expression. Cell 143, 1174–1189 (2010). 139. Blayney, L., Jones, J., Griffiths, J. & Lai, F. A mechanism of ryanodine receptor modulation by FKBP12/12.6, Protein Kinase A and K201. Cardiovascular Research (2009). 140. Xiao, B. et al. Functional consequence of protein kinase A-dependent phosphorylation of the cardiac ryanodine receptor: sensitization of store overload-induced Ca2+ release. J Biol Chem 282, 30256–30264 (2007). 141. Guo, T. et al. Kinetics of FKBP12.6 Binding to Ryanodine Receptors in Permeabilized Cardiac Myocytes and Effects on Ca Sparks. Circ Res 106, 1743–1752 (2010). 142. Dobrev, D. & Wehrens, X. H. T. Role of RyR2 Phosphorylation in Heart Failure and Arrhythmias: Controversies Around Ryanodine Receptor Phosphorylation in Cardiac Disease. Circ Res 114, 1311–1319 (2014). 143. Meissner, G. Ryanodine receptor/Ca2+ release channels and their regulation by endogenous effectors. Annual review of physiology 56, 485–508 (1994). 144. Meissner, G., Rios, E., Tripathy, A. & Pasek, D. A. Regulation of skeletal muscle Ca2+ release channel (ryanodine receptor) by Ca2+ and monovalent cations and anions. J Biol Chem 272, 1628–1638 (1997). 145. Meissner, G., Darling, E. & Eveleth, J. Kinetics of rapid Ca2+ release by sarcoplasmic reticulum. Effects of Ca2+, Mg2+, and adenine nucleotides.  129 Biochemistry 25, 236–244 (1986). 146. Strynadka, N. C. & James, M. N. Crystal structures of the helix-loop-helix calcium-binding proteins. Annu Rev Biochem 58, 951–998 (1989). 147. Wilson, M. A. & Brunger, A. T. The 1.0 Å crystal structure of Ca2+-bound calmodulin: an analysis of disorder and implications for functionally relevant plasticity. J Mol Biol 301, 1237–1256 (2000). 148. Haiech, J., Moulhaye, S. B. M. & Kilhoffer, M.-C. The EF-Handome: combining comparative genomic study using FamDBtool, a new bioinformatics tool, and the network of expertise of the European Calcium Society. Biochim Biophys Acta 1742, 179–183 (2004). 149. Chazin, W. J. Relating Form and Function of EF-Hand Calcium Binding Proteins. Acc. Chem. Res. 44, 171–179 (2011). 150. Henikoff, S. ENCODE and our very busy genome. Nature genetics 39, 817–818 (2007). 151. Kretsinger, R. H. & Nockolds, C. E. Carp muscle calcium-binding protein. II. Structure determination and general description. J Biol Chem 248, 3313–3326 (1973). 152. Drum, C. et al. Structural basis for the activation of anthrax adenylyl cyclase exotoxin by calmodulin. Nature 415, 396–402 (2002). 153. Yuan, T. & Vogel, H. J. Calcium-calmodulin-induced dimerization of the carboxyl-terminal domain from petunia glutamate decarboxylase. A novel calmodulin-peptide interaction motif. J Biol Chem 273, 30328–30335 (1998). 154. Fisher, P. J. et al. Calmodulin interacts with amphiphilic peptides composed of all D-amino acids. Nature 368, 651–653 (1994). 155. Sarhan, M. F., Tung, C.-C., Van Petegem, F. & Ahern, C. A. Crystallographic basis for calcium regulation of sodium channels. Proc Natl Acad Sci USA (2012). 156. Liu, Y. et al. Crystal structure of calmodulin binding domain of orai1 in complex with Ca2+ calmodulin displays a unique binding mode. Journal of Biological Chemistry 287, 43030–43041 (2012). 157. Juranic, N. et al. Calmodulin wraps around its binding domain in the plasma membrane Ca2+ pump anchored by a novel 18-1 motif. Journal of Biological Chemistry 285, 4015–4024 (2010). 158. Schumacher, M. A., Rivard, A. F., Bächinger, H. P. & Adelman, J. P. Structure of the gating domain of a Ca2+-activated K+ channel complexed with Ca2+/calmodulin. Nature 410, 1120–1124 (2001). 159. Blumenthal, D. K. & Stull, J. T. Activation of skeletal muscle myosin light chain kinase by calcium(2+) and calmodulin. Biochemistry 19, 5608–5614 (1980). 160. Rellos, P. et al. Structure of the CaMKIIδ/Calmodulin Complex Reveals the Molecular Mechanism of CaMKII Kinase Activation. PLoS Biol. 8, e1000426 (2010). 161. Tidow, H. et al. A bimodular mechanism of calcium control in eukaryotes. Nature 491, 468–472 (2012). 162. Vogel, H. J. The Merck Frosst Award Lecture 1994. Calmodulin: a versatile calcium mediator protein. Biochem. Cell Biol. 72, 357–376 (1994). 163. Yamniuk, A. P. & Vogel, H. J. Calmodulin's flexibility allows for promiscuity in its interactions with target proteins and peptides. Mol Biotechnol 27, 33–57 (2004). 164. Linse, S., Helmersson, A. & Forsén, S. Calcium binding to calmodulin and its globular domains. J Biol Chem 266, 8050–8054 (1991). 165. Ogawa, Y. & Tanokura, M. Calcium binding to calmodulin: effects of ionic strength,  130 Mg2+, pH and temperature. J Biochem 95, 19–28 (1984). 166. Zhang, M. et al. Structural basis for calmodulin as a dynamic calcium sensor. Structure 20, 911–923 (2012). 167. Peersen, O. B., Madsen, T. S. & Falke, J. J. Intermolecular tuning of calmodulin by target peptides and proteins: differential effects on Ca2+ binding and implications for kinase activation. Protein Sci 6, 794–807 (1997). 168. Brown, S. E., Martin, S. R. & Bayley, P. M. Kinetic control of the dissociation pathway of calmodulin-peptide complexes. J Biol Chem 272, 3389–3397 (1997). 169. Mirzoeva, S., Weigand, S., Lukas, T. J. & Shuvalova, L. Analysis of the functional coupling between calmodulin's calcium binding and peptide recognition properties. 38, 3936–3947 (1999). 170. Heiman, R. G. et al. Spontaneous avoidance behavior in Drosophila null for calmodulin expression. Proc Natl Acad Sci USA 93, 2420–2425 (1996). 171. Davis, T. N., Urdea, M. S., Masiarz, F. R. & Thorner, J. Isolation of the yeast calmodulin gene: Calmodulin is an essential protein. Cell 47, 423–431 (1986). 172. Crotti, L. et al. Calmodulin mutations associated with recurrent cardiac arrest in infants. Circulation 127, 1009–1017 (2013). 173. Maier, L. S. & Bers, D. M. Calcium, calmodulin, and calcium-calmodulin kinase II: heartbeat to heartbeat and beyond. J Mol Cell Cardiol 34, 919–939 (2002). 174. Harper, J. F. et al. Localization of calmodulin in rat tissues. Proc Natl Acad Sci USA 77, 366–370 (1980). 175. Xu Wu, D. M. B. Free and Bound Intracellular Calmodulin Measurements in Cardiac Myocytes. Cell Calcium 41, 353–364 (2007). 176. Hudmon, A. et al. CaMKII tethers to L-type Ca2+ channels, establishing a local and dedicated integrator of Ca2+ signals for facilitation. J Cell Biol 171, 537–547 (2005). 177. Pitt, G. S. et al. Molecular basis of calmodulin tethering and Ca2+-dependent inactivation of L-type Ca2+ channels. J Biol Chem 276, 30794–30802 (2001). 178. Friedberg, F. & Rhoads, A. R. Evolutionary Aspects of Calmodulin. IUBMB Life 51, 215–221 (2001). 179. Vetter, S. W. & Leclerc, E. Novel aspects of calmodulin target recognition and activation. Eur J Biochem 270, 404–414 (2003). 180. Zhang, M., Tanaka, T. & Ikura, M. Calcium-induced conformational transition revealed by the solution structure of apo calmodulin. Nat Struct Biol 2, 758–767 (1995). 181. Rhoads, A. Sequence motifs for calmodulin recognition. The FASEB journal (1997). 182. YS, B. et al. Three-dimensional structure of calmodulin. Nature 315, 37–40 (1984). 183. Fallon, J. L. & Quiocho, F. A. A Closed Compact Structure of Native Ca2+-Calmodulin. Structure 11, 1303–1307 (2003). 184. Nelson, M. R. & Chazin, W. J. Structures of EF-hand Ca 2+-binding proteins: Diversity in the organization, packing and response to Ca 2+ Binding. Biometals 11, 297–318 (1998). 185. Crivici, A. & Ikura, M. Molecular and structural basis of target recognition by calmodulin. Annu Rev Biophys Biomol Struct 24, 85–116 (1995). 186. Chou, J. J., Li, S., Klee, C. B. & Bax, A. Solution structure of Ca(2+)-calmodulin reveals flexible hand-like properties of its domains. Nat Struct Biol 8, 990–997 (2001). 187. Gellman, S. H. On the role of methionine residues in the sequence-independent recognition of nonpolar protein surfaces. Biochemistry 30, 6633–6636 (1991). 188. Tidow, H. & Nissen, P. Structural diversity of calmodulin binding to its target sites.  131 FEBS J. 280, 5551–5565 (2013). 189. Radivojac, P. et al. Calmodulin signaling: analysis and prediction of a disorder-dependent molecular recognition. Proteins 63, 398–410 (2006). 190. Feldkamp, M. D., Yu, L. & Shea, M. A. Structural and energetic determinants of apo calmodulin binding to the IQ motif of the Na(V)1.2 voltage-dependent sodium channel. Structure 19, 733–747 (2011). 191. Chapman, E. R., Alexander, K., Vorherr, T., Carafoli, E. & Storm, D. R. Fluorescence energy transfer analysis of calmodulin-peptide complexes. 31, 12819–12825 (1992). 192. Small, E. W. & Anderson, S. R. Fluorescence anisotropy decay demonstrates calcium-dependent shape changes in photo-cross-linked calmodulin. 27, 419–428 (1988). 193. Seaton, B. A., Head, J. F., Engelman, D. M. & Richards, F. M. Calcium-induced increase in the radius of gyration and maximum dimension of calmodulin measured by small-angle x-ray scattering. Biochemistry 24, 6740–6743 (2002). 194. Ikura, M. et al. Solution structure of a calmodulin-target peptide complex by multidimensional NMR. Science 256, 632–638 (1992). 195. Meador, W., Means, A. & Quiocho, F. Target enzyme recognition by calmodulin: 2.4 A structure of a calmodulin-peptide complex. Science 257, 1251 (1992). 196. Meador, W. E., Means, A. R. & Quiocho, F. A. Modulation of Calmodulin Plasticity in Molecular Recognition on the Basis of X-Ray Structures. Science 262, 1718–1721 (1993). 197. Yap, K. L., Yuan, T., Mal, T. K., Vogel, H. J. & Ikura, M. Structural basis for simultaneous binding of two carboxy-terminal peptides of plant glutamate decarboxylase to calmodulin. J Mol Biol 328, 193–204 (2003). 198. Kim, E. Y. et al. Multiple C-terminal tail Ca(2+)/CaMs regulate Ca(V)1.2 function but do not mediate channel dimerization. EMBO J 29, 3924–3938 (2010). 199. Prosser, B. L., Hernández-Ochoa, E. O. & Schneider, M. F. S100A1 and calmodulin regulation of ryanodine receptor in striated muscle. Cell Calcium 50, 323–331 (2011). 200. Moore, B. W. A soluble protein characteristic of the nervous system. Biochem Biophys Res Commun 19, 749–744 (1965). 201. Rustandi, R. R. et al. Three-Dimensional Solution Structure of the Calcium-Signaling Protein Apo-S100A1 As Determined by NMR. 41, 788–796 (2002). 202. Wright, N. T. et al. The three-dimensional solution structure of Ca(2+)-bound S100A1 as determined by NMR spectroscopy. J Mol Biol 353, 410–426 (2005). 203. Goch, G., Vdovenko, S., Kozłowska, H. & Bierzyñski, A. Affinity of S100A1 protein for calcium increases dramatically upon glutathionylation. FEBS J. 272, 2557–2565 (2005). 204. Wright, N. T. et al. S100A1 and calmodulin compete for the same binding site on ryanodine receptor. J Biol Chem 283, 26676–26683 (2008). 205. Haimoto, H. & Kato, K. S100a0 (alpha alpha) protein, a calcium-binding protein, is localized in the slow-twitch muscle fibre. Journal of Neurochemistry 48, 917–923 (1987). 206. Treves, S. et al. Interaction of S100A1 with the Ca2+ release channel (ryanodine receptor) of skeletal muscle. 36, 11496–11503 (1997). 207. Most, P. et al. Cardiac S100A1 protein levels determine contractile performance and propensity toward heart failure after myocardial infarction. Circulation 114, 1258–1268 (2006).  132 208. Remppis, A. et al. Altered expression of the Ca(2+)-binding protein S100A1 in human cardiomyopathy. Biochim Biophys Acta 1313, 253–257 (1996). 209. Pleger, S. T. et al. S100A1 Gene Therapy Preserves in Vivo Cardiac Function after Myocardial Infarction. Molecular Therapy 12, 1120–1129 (2005). 210. Ritterhoff, J. & Most, P. Targeting S100A1 in heart failure. Gene Ther. 19, 613–621 (2012). 211. Yang, H. C., Reedy, M. M., Burke, C. L. & Strasburg, G. M. Calmodulin interaction with the skeletal muscle sarcoplasmic reticulum calcium channel protein. 33, 518–525 (1994). 212. Fuentes, O., Valdivia, C., Vaughan, D., Coronado, R. & Valdivia, H. H. Calcium-dependent block of ryanodine receptor channel of swine skeletal muscle by direct binding of calmodulin. Cell Calcium 15, 305–316 (1994). 213. Ikemoto, T., Takeshima, H., Iino, M. & Endo, M. Effect of calmodulin on Ca 2+ -induced Ca 2+ release of skeletal muscle from mutant mice expressing either ryanodine receptor type 1 or type 3. Pflugers Arch 437, 43–48 (1998). 214. Meissner, G. Evidence of a role for calmodulin in the regulation of calcium release from skeletal muscle sarcoplasmic reticulum. Biochemistry 25, 244–251 (1986). 215. Yamaguchi, N., Xu, L., Evans, K. E., Pasek, D. A. & Meissner, G. Different regions in skeletal and cardiac muscle ryanodine receptors are involved in transducing the functional effects of calmodulin. J Biol Chem 279, 36433–36439 (2004). 216. Ikemoto, T., Iino, M. & Endo, M. Enhancing effect of calmodulin on Ca(2+)-induced Ca2+ release in the sarcoplasmic reticulum of rabbit skeletal muscle fibres. J Physiol 487 ( Pt 3), 573–582 (1995). 217. Tripathy, A., Xu, L., Mann, G. & Meissner, G. Calmodulin activation and inhibition of skeletal muscle Ca2+ release channel (ryanodine receptor). Biophys J 69, 106–119 (1995). 218. Buratti, R., Prestipino, G., Menegazzi, P., Treves, S. & Zorzato, F. Calcium dependent activation of skeletal muscle Ca2+ release channel (ryanodine receptor) by calmodulin. Biochem Biophys Res Commun 213, 1082–1090 (1995). 219. Fruen, B. R., Bardy, J. M., Byrem, T. M., Strasburg, G. M. & Louis, C. F. Differential Ca(2+) sensitivity of skeletal and cardiac muscle ryanodine receptors in the presence of calmodulin. AJP - Cell Physiology 279, C724–33 (2000). 220. Tian, X., Tang, Y., Liu, Y., Wang, R. & Chen, S. R. W. Calmodulin modulates the termination threshold for cardiac ryanodine receptor-mediated Ca2+ release. Biochem J 455, 367–375 (2013). 221. Chen, S. & MacLennan, D. Identification of calmodulin-, Ca (2+)-, and ruthenium red-binding domains in the Ca2+ release channel (ryanodine receptor) of rabbit skeletal muscle sarcoplasmic reticulum. J Biol Chem 269, 22698 (1994). 222. Menegazzi, P. et al. Identification and characterization of three calmodulin binding sites of the skeletal muscle ryanodine receptor. Biochemistry 33, 9078–9084 (1994). 223. Moore, C. P. et al. Apocalmodulin and Ca2+ calmodulin bind to the same region on the skeletal muscle Ca2+ release channel. Biochemistry 38, 8532–8537 (1999). 224. Rodney, G. et al. Calcium binding to calmodulin leads to an N-terminal shift in its binding site on the ryanodine receptor. J Biol Chem 276, 2069–2074 (2001). 225. Xiong, L.-W. et al. Lobe-dependent regulation of ryanodine receptor type 1 by calmodulin. J Biol Chem 277, 40862–40870 (2002). 226. Zhang, H., Zhang, J.-Z., Danila, C. I. & Hamilton, S. L. A noncontiguous, intersubunit binding site for calmodulin on the skeletal muscle Ca2+ release channel.  133 J Biol Chem 278, 8348–8355 (2003). 227. Fruen, B. R. et al. Direct detection of calmodulin tuning by ryanodine receptor channel targets using a Ca2+-sensitive acrylodan-labeled calmodulin. Biochemistry 44, 278–284 (2005). 228. Gangopadhyay, J. & Ikemoto, N. Interaction of the Lys3614-Asn3643 calmodulin-binding domain with the Cys4114-Asn4142 region of the type 1 ryanodine receptor is involved in the mechanism of Ca2+/agonist-induced channel activation. Biochem J (2008). 229. Yamaguchi, N., Xu, L., Pasek, D. A., Evans, K. E. & Meissner, G. Molecular basis of calmodulin binding to cardiac muscle Ca(2+) release channel (ryanodine receptor). J Biol Chem 278, 23480–23486 (2003). 230. Yamaguchi, N., Takahashi, N., Xu, L., Smithies, O. & Meissner, G. Early cardiac hypertrophy in mice with impaired calmodulin regulation of cardiac muscle Ca2+ release channel. J. Clin. Invest. 117, 1344–1353 (2007). 231. Yamaguchi, N. et al. Calmodulin regulation and identification of calmodulin binding region of type-3 ryanodine receptor calcium release channel. Biochemistry 44, 15074–15081 (2005). 232. Gomez, A. C. & Yamaguchi, N. Two regions of the ryanodine receptor calcium channel are involved in Ca(2+)-dependent inactivation. Biochemistry 53, 1373–1379 (2014). 233. Mal, T. K., Skrynnikov, N. R., Yap, K. L., Kay, L. E. & Ikura, M. Detecting Protein Kinase Recognition Modes of Calmodulin by Residual Dipolar Couplings in Solution NMR. Biochemistry 41, 12899–12906 (2002). 234. Contessa, G. M. et al. Structure of calmodulin complexed with an olfactory CNG channel fragment and role of the central linker: residual dipolar couplings to evaluate calmodulin binding modes outside the kinase family. J Biomol NMR 31, 185–199 (2005). 235. Huang, X., Fruen, B., Farrington, D. T., Wagenknecht, T. & Liu, Z. Calmodulin-binding locations on the skeletal and cardiac ryanodine receptors. J Biol Chem 287, 30328–30335 (2012). 236. Huang, X. et al. Two potential calmodulin-binding sequences in the ryanodine receptor contribute to a mobile, intra-subunit calmodulin-binding domain. Journal of Cell Science 126, 4527–4535 (2013). 237. Tanabe, T., Beam, K. G., Adams, B. A., Niidome, T. & Numa, S. Regions of the skeletal muscle dihydropyridine receptor critical for excitation-contraction coupling. Nature 346, 567–569 (1990). 238. Adams, B. A., Tanabe, T., Mikami, A., Numa, S. & Beam, K. G. Intramembrane charge movement restored in dysgenic skeletal muscle by injection of dihydropyridine receptor cDNAs. Nature 346, 569–572 (1990). 239. Grabner, M., Dirksen, R. T., Suda, N. & Beam, K. G. The II-III Loop of the Skeletal Muscle Dihydropyridine Receptor Is Responsible for the Bi-directional Coupling with the Ryanodine Receptor. J Biol Chem 274, 21913–21919 (1999). 240. Franzini-Armstrong, C. & Nunzi, G. Junctional feet and particles in the triads of a fast-twitch muscle fibre. J Muscle Res Cell Motil 4, 233–252 (1983). 241. Takekura, H., Sun, X. & Franzini-Armstrong, C. Development of the Excitation-Contraction Coupling Apparatus in Skeletal-Muscle - Peripheral and Internal Calcium-Release Units Are Formed Sequentially. J Muscle Res Cell Motil 15, 102–118 (1994).  134 242. Nakai, J. et al. Enhanced dihydropyridine receptor channel activity in the presence of ryanodine receptor. Nature 380, 72–75 (1996). 243. Nakai, J., Tanabe, T., Konno, T., Adams, B. & Beam, K. G. Localization in the II-III loop of the dihydropyridine receptor of a sequence critical for excitation-contraction coupling. J Biol Chem 273, 24983–24986 (1998). 244. Leong, P. & MacLennan, D. H. The cytoplasmic loops between domains II and III and domains III and IV in the skeletal muscle dihydropyridine receptor bind to a contiguous site in the skeletal muscle ryanodine receptor. J Biol Chem 273, 29958–29964 (1998). 245. Lu, X., Xu, L. & Meissner, G. Activation of the skeletal muscle calcium release channel by a cytoplasmic loop of the dihydropyridine receptor. J Biol Chem 269, 6511–6516 (1994). 246. Saiki, Y., El-Hayek, R. & Ikemoto, N. Involvement of the Glu724-Pro760 region of the dihydropyridine receptor II-III loop in skeletal muscle-type excitation-contraction coupling. J Biol Chem 274, 7825–7832 (1999). 247. Casarotto, M. G., Green, D., Pace, S., Young, J. & Dulhunty, A. F. Activating the ryanodine receptor with dihydropyridine receptor II-III loop segments: size and charge do matter. Front Biosci 9, 2860–2872 (2004). 248. Proenza, C. et al. Identification of a region of RyR1 that participates in allosteric coupling with the alpha(1S) (Ca(V)1.1) II-III loop. J Biol Chem 277, 6530–6535 (2002). 249. Nakai, J., Sekiguchi, N., Rando, T. A., Allen, P. D. & Beam, K. G. Two Regions of the Ryanodine Receptor Involved in Coupling withl-Type Ca2+ Channels. J Biol Chem 273, 13403–13406 (1998). 250. Sheridan, D. C. et al. Bidirectional signaling between calcium channels of skeletal muscle requires multiple direct and indirect interactions. Proc Natl Acad Sci USA 103, 19760–19765 (2006). 251. Protasi, F. et al. Multiple Regions of RyR1 Mediate Functional and Structural Interactions with α1S-Dihydropyridine Receptors in Skeletal Muscle. Biophys J 83, 3230–3244 (2002). 252. Schredelseker, J., Dayal, A., Schwerte, T., Franzini-Armstrong, C. & Grabner, M. Proper restoration of excitation-contraction coupling in the dihydropyridine receptor beta1-null zebrafish relaxed is an exclusive function of the beta1a subunit. J Biol Chem 284, 1242–1251 (2009). 253. Gregg, R. G. et al. Absence of the beta subunit (cchb1) of the skeletal muscle dihydropyridine receptor alters expression of the alpha 1 subunit and eliminates excitation-contraction coupling. Proc Natl Acad Sci USA 93, 13961–13966 (1996). 254. Van Petegem, F., Clark, K. A., Chatelain, F. C. & Minor, D. L. Structure of a complex between a voltage-gated calcium channel beta-subunit and an alpha-subunit domain. Nature 429, 671–675 (2004). 255. Endo, M., Tanaka, M. & Ogawa, Y. Calcium Induced Release of Calcium from the Sarcoplasmic Reticulum of Skinned Skeletal Muscle Fibres. Nature 228, 34–36 (1970). 256. Palade, P., Mitchell, R. D. & Fleischer, S. Spontaneous calcium release from sarcoplasmic reticulum. General description and effects of calcium. J Biol Chem 258, 8098–8107 (1983). 257. Jiang, D. et al. Enhanced store overload-induced Ca2+ release and channel  135 sensitivity to luminal Ca2+ activation are common defects of RyR2 mutations linked to ventricular tachycardia and sudden death. Circ Res 97, 1173–1181 (2005). 258. Fujii, J. et al. Identification of a mutation in porcine ryanodine receptor associated with malignant hyperthermia. Science 253, 448–451 (1991). 259. Kausch, K. et al. Evidence for linkage of the central core disease locus to the proximal long arm of human chromosome 19. Genomics 10, 765–769 (1991). 260. Priori, S. G. et al. Mutations in the cardiac ryanodine receptor gene (hRyR2) underlie catecholaminergic polymorphic ventricular tachycardia. Circulation 103, 196–200 (2001). 261. Guerrini, R. et al. Calmodulin binding sites of the skeletal, cardiac, and brain ryanodine receptor Ca2+ channels: modulation by the catalytic subunit of cAMP-dependent protein kinase? Biochemistry 34, 5120–5129 (1995). 262. Balshaw, D., Xu, L., Yamaguchi, N., Pasek, D. & Meissner, G. Calmodulin binding and inhibition of cardiac muscle calcium release channel (ryanodine receptor). Journal of Biological Chemistry 276, 20144 (2001). 263. Zhu, X., Ghanta, J., Walker, J., Allen, P. & Valdivia, H. The calmodulin binding region of the skeletal ryanodine receptor acts as a self-modulatory domain. Cell Calcium 35, 165–177 (2004). 264. Kholod, N. & Mustelin, T. Novel vectors for co-expression of two proteins in E. coli. BioTechniques 31, 322–3– 326–8 (2001). 265. Van Petegem, F., Chatelain, F. C. & Minor, D. L. Insights into voltage-gated calcium channel regulation from the structure of the CaV1.2 IQ domain-Ca2+/calmodulin complex. 12, 1108–1115 (2005). 266. Edelhoch, H. Spectroscopic determination of tryptophan and tyrosine in proteins. Biochemistry 6, 1948–1954 (1967). 267. Mori, M. X., vander Kooi, C. W., Leahy, D. J. & Yue, D. T. Crystal structure of the CaV2 IQ domain in complex with Ca2+/calmodulin: high-resolution mechanistic implications for channel regulation by Ca2. Structure 16, 607–620 (2008). 268. Kim, E. Y. et al. Structures of CaV2 Ca2+/CaM-IQ domain complexes reveal binding modes that underlie calcium-dependent inactivation and facilitation. Structure 16, 1455–1467 (2008). 269. Ibarra M, C. A. et al. Malignant hyperthermia in Japan: mutation screening of the entire ryanodine receptor type 1 gene coding region by direct sequencing. Anesthesiology 104, 1146–1154 (2006). 270. Kovalevskaya, N. V. et al. Structural analysis of calmodulin binding to ion channels demonstrates the role of its plasticity in regulation. Pflugers Arch 465, 1507–1519 (2013). 271. Saimi, Y. & Kung, C. Calmodulin as an ion channel subunit. Annual review of physiology 64, 289–311 (2002). 272. Pitt, G. S. Calmodulin and CaMKII as molecular switches for cardiac ion channels. Cardiovascular Research 73, 641–647 (2007). 273. Van Petegem, F., Lobo, P. A. & Ahern, C. A. Seeing the forest through the trees: towards a unified view on physiological calcium regulation of voltage-gated sodium channels. Biophys J 103, 2243–2251 (2012). 274. Tanabe, T., Beam, K. G., Powell, J. A. & Numa, S. Restoration of excitation-contraction coupling and slow calcium current in dysgenic muscle by dihydropyridine receptor complementary DNA. Nature 336, 134–139 (1988). 275. Yamaguchi, N., Xin, C. & Meissner, G. Identification of apocalmodulin and Ca2+- 136 calmodulin regulatory domain in skeletal muscle Ca2+ release channel, ryanodine receptor. J Biol Chem 276, 22579–22585 (2001). 276. Cornea, R. L. et al. FRET-based mapping of calmodulin bound to the RyR1 Ca2+ release channel. Proc Natl Acad Sci USA 106, 6128–6133 (2009). 277. Yang, Y. et al. Cardiac myocyte Z-line calmodulin is mainly RyR2-bound, and reduction is arrhythmogenic and occurs in heart failure. Circ Res 114, 295–306 (2014). 278. Meissner, G. et al. Thermodynamics of calmodulin binding to cardiac and skeletal muscle ryanodine receptor ion channels. Proteins 74, 207–211 (2009). 279. Mori, M. X. Functional stoichiometry and local enrichment of calmodulin interacting with Ca2+ channels. Science 304, 432–435 (2004). 280. Findeisen, F. & Minor, D. L., Jr. Structural Basis for the Differential Effects of CaBP1 and Calmodulin on CaV1.2 Calcium-Dependent Inactivation. Structure 18, 1617–1631 (2010). 281. Kimlicka, L. & Van Petegem, F. The structural biology of ryanodine receptors. Sci China Life Sci 54, 712–724 (2011). 282. Takekura, H., Bennett, L., Tanabe, T., Beam, K. G. & Franzini-Armstrong, C. Restoration of junctional tetrads in dysgenic myotubes by dihydropyridine receptor cDNA. Biophys J 67, 793–803 (1994). 283. Avila, G. & Dirksen, R. T. Functional impact of the ryanodine receptor on the skeletal muscle L-type Ca2+ channel. The Journal of General Physiology 115, 467–479 (2000). 284. Kugler, G., Weiss, R. G., Flucher, B. E. & Grabner, M. Structural requirements of the dihydropyridine receptor alpha1S II-III loop for skeletal-type excitation-contraction coupling. J Biol Chem 279, 4721–4728 (2004). 285. Tae, H.-S. et al. Molecular Recognition of the Disordered Dihydropyridine Receptor II-III loop by a Conserved SPRY Domain of the Type 1 Ryanodine Receptor. Clin. Exp. Pharmacol. Physiol. 36, 346–349 (2009). 286. Tae, H.-S. et al. Cyclization of the intrinsically disordered α1S dihydropyridine receptor II-III loop enhances secondary structure and in vitro function. J Biol Chem 286, 22589–22599 (2011). 287. Kabsch, W. XDS. Acta Crystallogr. D Biol. Crystallogr. 66, 125–132 (2010). 288. Terwilliger, T. C. et al. Decision-making in structure solution using Bayesian estimates of map quality: the PHENIX AutoSol wizard. Acta Crystallogr. D Biol. Crystallogr. 65, 1–20 (2009). 289. Adams, P. D. et al. PHENIX: a comprehensive Python-based system for macromolecular structure solution. Acta Crystallogr. D Biol. Crystallogr. 1–9 (2010). doi:10.1107/S0907444909052925 290. Emsley, P. & Cowtan, K. Coot: model-building tools for molecular graphics. Acta Crystallogr. D Biol. Crystallogr. 1–7 (2004). doi:10.1107/S0907444904019158 291. Wriggers, W., Milligan, R. A. & McCammon, J. A. Situs: A package for docking crystal structures into low-resolution maps from electron microscopy. Journal of Structural Biology 125, 185–195 (1999). 292. Nettleship, J. E., Brown, J., Groves, M. R. & Geerlof, A. Methods for protein characterization by mass spectrometry, thermal shift (ThermoFluor) assay, and multiangle or static light scattering. Methods Mol. Biol. 426, 299–318 (2008). 293. Fujino, N., Ino, H., Hayashi, K., Uchiyama, K. & Yamagishi, M. A Novel Missense Mutation in Cardiac Ryanodine Receptor Gene as a Possible Cause of Hypertrophic  137 Cardiomyopathy: Evidence From Familial Analysis. Circulation 114, (2006). 294. Tang, Y., Tian, X., Wang, R., Fill, M. & Chen, S. R. W. Abnormal termination of Ca2+ release is a common defect of RyR2 mutations associated with cardiomyopathies. Circ Res 110, 968–977 (2012). 295. Böhm, J. et al. An Integrated Diagnosis Strategy for Congenital Myopathies. PLoS ONE 8, e67527 (2013). 296. Klein, A. et al. Clinical and genetic findings in a large cohort of patients with ryanodine receptor 1 gene-associated myopathies. Hum. Mutat. 33, 981–988 (2012). 297. Tammaro, A. et al. Novel missense mutations and unexpected multiple changes of RYR1 gene in 75 malignant hyperthermia families. Clin. Genet. 79, 438–447 (2011). 298. Gonsalves, S. G. et al. Using exome data to identify malignant hyperthermia susceptibility mutations. Anesthesiology 119, 1043–1053 (2013). 299. Wang, R. et al. Localization of the dantrolene-binding sequence near the FK506-binding protein-binding site in the three-dimensional structure of the ryanodine receptor. Journal of Biological Chemistry 286, 12202–12212 (2011). 300. Van Petegem, F. Ryanodine receptors: structure and function. J Biol Chem 287, 31624–31632 (2012). 301. Gillard, E. F. et al. A substitution of cysteine for arginine 614 in the ryanodine receptor is potentially causative of human malignant hyperthermia. Genomics 11, 751–755 (1991). 302. Jungbluth, H., Sewry, C. A. & Muntoni, F. What's new in neuromuscular disorders? The congenital myopathies. Eur. J. Paediatr. Neurol. 7, 23–30 (2003). 303. Medeiros-Domingo, A. et al. The RYR2-encoded ryanodine receptor/calcium release channel in patients diagnosed previously with either catecholaminergic polymorphic ventricular tachycardia or genotype negative, exercise-induced long QT syndrome: a comprehensive open reading frame mutational analysis. J Am Coll Cardiol 54, 2065–2074 (2009). 304. Perez, C. F., Mukherjee, S. & Allen, P. D. Amino Acids 1–1,680 of Ryanodine Receptor Type 1 Hold Critical Determinants of Skeletal Type for Excitation-Contraction Coupling ROLE OF DIVERGENCE DOMAIN D2. J Biol Chem 278, 39644–39652 (2003). 305. Tateishi, H. et al. Defective domain-domain interactions within the ryanodine receptor as a critical cause of diastolic Ca2+ leak in failing hearts. Cardiovascular Research (2008). 306. Abu-Abed, M., Mal, T. K., Kainosho, M., MacLennan, D. H. & Ikura, M. Characterization of the ATP-binding domain of the sarco(endo)plasmic reticulum Ca(2+)-ATPase: probing nucleotide binding by multidimensional NMR. 41, 1156–1164 (2002). 307. Bultynck, G. et al. Characterization and mapping of the 12 kDa FK506-binding protein (FKBP12)-binding site on different isoforms of the ryanodine receptor and of the inositol 1,4,5-trisphosphate receptor. Biochem J 354, 413–422 (2001). 308. Popova, O. B., Baker, M. R., Tran, T. P., Le, T. & Serysheva, I. I. Identification of ATP-Binding Regions in the RyR1 Ca2+ Release Channel. PLoS ONE 7, e48725 (2012). 309. Xiong, H. et al. Identification of a two EF-hand Ca2+ binding domain in lobster skeletal muscle ryanodine receptor/Ca2+ release channel. Biochemistry 37, 4804–4814 (1998). 310. Yamazawa, T., Takeshima, H., Shimuta, M. & Iino, M. A region of the ryanodine  138 receptor critical for excitation-contraction coupling in skeletal muscle. J Biol Chem 272, 8161–8164 (1997). 311. Girgenrath, T. et al. N-terminal and central segments of the type 1 ryanodine receptor mediate its interaction with FK506-binding proteins. Journal of Biological Chemistry 288, 16073–16084 (2013).  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items