@prefix vivo: . @prefix edm: . @prefix ns0: . @prefix dcterms: . @prefix dc: . @prefix skos: . vivo:departmentOrSchool "Medicine, Faculty of"@en, "Anesthesiology, Pharmacology and Therapeutics, Department of"@en ; edm:dataProvider "DSpace"@en ; ns0:degreeCampus "UBCV"@en ; dcterms:creator "Kertesz, Ron"@en ; dcterms:issued "2009-06-11T18:56:12Z"@en, "1998"@en ; vivo:relatedDegree "Master of Science - MSc"@en ; ns0:degreeGrantor "University of British Columbia"@en ; dcterms:description """The selection of an appropriate therapy for ventricular arrhythmias is complicated by use of therapeutic agents which have several modes of action as well as numerous toxic effects. Attempts to improve existing agents by systematic modification of their chemical structure has had little success. Chronic use of currently available agents is not recommended, and most of these are reserved for acute use in individuals with recent infarcts. The purpose of this study was to examine a new family of ion channel blocking agents, developed by Rhythm Search Developments Ltd. (RSD), with possible application as antiarrhythmic agents. RSD's investigators discovered that a known κ-opioid agonist, U50,488H (RSD 925), had sodium channel blocking actions. RSD then developed a series of compounds which show some limited 3- dimensional similarity to RSD925. Seven of these compounds (RSD 939, RSD 952, RSD 961, RSD 969, RSD 971, RSD 981, and RSD 988) were examined in an attempt to correlate changes at the two structural hetero atoms (labeled as X1 or X2) with changes in sodium and potassium channel blocking activity, toxicity, and electrophysiological effects on isolated whole cells. For this study RSD939 was used as the prototype compound to which the other compounds were compared. The pharmacological profile of the compounds was determined by in vivo experiments on rats and mice. Drug effects were determined by recording blood pressure (BP), heart rate, (HR), and electrocardiogram (ECG) in pentobarbitone-anaesthetized rats given bolus administrations of drug. The same measurements were made in anaesthetized rats receiving infusions of drug. In addition, the infused animals were also subjected to electrical stimulation of the heart in order to measure several of its electrical parameters. The electrical parameters measured were: the threshold current for capture (iT), the threshold pulse duration for capture (tT), the threshold current for induction of ventricular fibrillation (VFT), the effective refractory period (ERP), and the maximum following frequency (MFF). Lethality of the compounds was tested using both rats and mice. Finally, effects of the compounds on single ion channels were assessed in vitro using patch clamp techniques. Changes to the amide nitrogen at the XI position had effects on the pharmacological profile of the compounds. Changing the tertiary amide (RSD 939) to a secondary amide (RSD988) seemed to decrease both sodium channel and potassium channel blocking activity. The secondary amide was also less toxic, but appeared to be a more potent affector of blood pressure (BP) and heart rate (HR) changes. Substitution of an ester linkage (RSD 952) for the amide linkage of RSD 939 reduced potency for ion channel blockade even more than the change to a secondary amide. This ester was less toxic than either RSD 939 or RSD 988. RSD 952 also had less effect on HR and BP than either RSD 939 or RSD 988. Changes to the substitution pattern on the nitrogen at X2 also had distinct effects. Converting the tertiary pyrrollidino amine nitrogen (RSD939) to a quaternary methyl pyrrollidinium nitrogen (RSD971) decreased ion channel blocking activity while leaving toxicity (lethality and morbidity) either unchanged or slightly increased. It appeared that the most important determinants of activity for the compounds examined in this study were the water solubility of the compounds and the pKa of the nitrogen at the X2 position. It is evident that changes to the regions around the nitrogens at positions X1 and X2 affected the activity of these antiarrhythmic agents. It is also evident that a separation of the pharmacophores for Na and K channels, as well as those for morbidity and lethality effects, is possible. Further examination of a larger number of compounds will lead to a better understanding of the chemical structure requirements for effective antiarrhythmic design."""@en ; edm:aggregatedCHO "https://circle.library.ubc.ca/rest/handle/2429/8970?expand=metadata"@en ; dcterms:extent "7534434 bytes"@en ; dc:format "application/pdf"@en ; skos:note "The Effects of Changes at the Nitrogens on the Pharmacological Profile of a Series of Six Analogues of RSD939 by Ron Kertesz B.Sc. University of Toronto, Ontario, 1992 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES DEPARTMENT OF PHARMACOLOGY & THERAPEUTICS FACULTY OF MEDICPNE We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA December 1998 © Ron Kertesz, 1998 ....... -y^m^. ., 01/04/99 18:00 FAX 604 822 9587 SPECIAL COLLECTIONS ill002 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly, purposes may be granted by the head of my department or by his or her representatives. It Is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of p/te'maco la^ v itern^f^'^^ The University of British Columbia Vancouver, Canada Date U^utry V>*//9M DE-6 (2/88) II Abstract The selection of an appropriate therapy for ventricular arrhythmias is complicated by use of therapeutic agents which have several modes of action as well as numerous toxic effects. Attempts to improve existing agents by systematic modification of their chemical structure has had little success. Chronic use of currently available agents is not recommended, and most of these are reserved for acute use in individuals with recent infarcts. The purpose of this study was to examine a new family of ion channel blocking agents, developed by Rhythm Search Developments Ltd. (RSD), with possible application as antiarrhythmic agents. RSD's investigators discovered that a known K-opioid agonist, U50,488H (RSD 925), had sodium channel blocking actions. RSD then developed a series of compounds which show some limited 3-dimensional similarity to RSD925. Seven of these compounds (RSD 939, RSD 952, RSD 961, RSD 969, RSD 971, RSD 981, and RSD 988) were examined in an attempt to correlate changes at the two structural hetero atoms (labeled as XI or X2) with changes in sodium and potassium channel blocking activity, toxicity, and electrophysiological effects on isolated whole cells. For this study RSD939 was used as the prototype compound to which the other compounds were compared. The pharmacological profile of the compounds was determined by in vivo experiments on rats and mice. Drug effects were determined by recording blood pressure (BP), heart rate, (HR), and electrocardiogram (ECG) in pentobarbitone-anaesthetized rats given bolus administrations of drug. The same measurements were made in anaesthetized rats receiving infusions of drug. In addition, the infused animals were also subjected to electrical stimulation of the heart in order to measure several of its electrical parameters. The electrical parameters measured were: the threshold current for capture (iT), the threshold pulse duration for capture (tT), the threshold current for induction of ventricular fibrillation (VFT), the effective refractory period (ERP), and the maximum following frequency (MFF). Lethality of the compounds was tested using both rats and mice. Finally, effects of the compounds on single ion channels were assessed in vitro using patch clamp techniques. Changes to the amide nitrogen at the XI position had effects on the pharmacological profile of the compounds. Changing the tertiary amide (RSD 939) to a secondary amide (RSD988) seemed to decrease both sodium channel and potassium channel blocking activity. The secondary amide was also less toxic, but appeared to be a more potent affector of blood pressure (BP) and heart rate (HR) changes. Substitution of an ester linkage (RSD 952) for the amide linkage of RSD 939 reduced potency for ion channel blockade even more than the change to a secondary amide. This ester was less toxic than either RSD 939 or RSD 988. RSD 952 also had less effect on HR and BP than either RSD 939 or RSD 988. Changes to the substitution pattern on the nitrogen at X2 also had distinct effects. Converting the tertiary pyrrollidino amine nitrogen (RSD939) to a quaternary methyl pyrrollidinium nitrogen (RSD971) decreased ion channel blocking activity while leaving toxicity (lethality and morbidity) either unchanged or slightly increased. It appeared that the most important determinants of activity for the compounds examined in this study were the water solubility of the compounds and the pKa of the nitrogen at the X2 position. It is evident that changes to the regions around the nitrogens at positions XI and X2 affected the activity of these antiarrhythmic agents. It is also evident that a separation of the pharmacophores for Na and K channels, as well as those for morbidity and lethality effects, is possible. Further examination of a larger number of compounds will lead to a better understanding of the chemical structure requirements for effective antiarrhythmic design. V TABLE OF CONTENTS Abstract 11 Table of Contents V List of Tables Vii List of Figures V?u Abbreviations »X Acknowledgments X 1. INTRODUCTION J. 1.1 The Need for a New and Effective Antiarrhythmic £ 1.2 The Normal Cardiac Action Potential 3 1.3 Abnormal Rhythm and Arrhythmias (o 1.3.1 Abnormal Impulse Generation \"7 1.3.2 Abnormal Conduction of Impulses Ai-1.3.3 Electrophysiological Effects of Myocardial Ischaemia 3& 1.4 Mechanisms of Antian-hythmic Drug Action l l 1.4.1 Classification of Antiarrhythmic Drugs Jg> 1.4.2 Sodium Channel Structure V\\ 1.4.3 Class I Agents £1 1.4.4 The S AR of existing Class I Agents 1.4.5 Potassium Channel Structure £y 1.4.6 Class IH Agents %£> 1.4.7 The S AR of existing Class IH Agents do 1.5 Opioid Receptor Agonists and Ion Channel Blockade 1.6 Objectives 26 2. MATERIALS AND METHODS c5> 2.1 Compound Preparation 4.2.1 The X2 Position and Sodium Channel Block SOZ 422 The X2 Position and Potassium Channel Block ±05 4.2.3 The X2 Position and Effects on Blood Pressure and Heart 107 Rate 4.2.4 The X2 Position Lethality J-Gl? 4.3 Conclusions -lO i^-5. REFERENCES Ji3 vfi LIST OF TABLES Pg-1. The chemical names and formulas of the compounds used in this study 3R 2. Descriptive statistics for D25%s of PR Interval effects for RSD939 and RSD961 in the in vivo screen 3. Data from preliminary in vivo screening 34 4. Descriptive statistics for D25%s of VFT effects for RSD939 and fcfe RSD961in the electrical stimulation screen 5. ECG data from electrical stimulation screen 6. Electrical parameter data from electrical stimulation screen (^ (\\ 7. Data from mouse toxicity studies 8. Patch clamp study data 03 9. Summary of data for apparent sodium channel blocking effects &\\ 10. Summary of data for apparent potassium channel blocking effects Q$ 11. Summary of morbidity and mortality data 5L? 12. Relative potency of compounds for measurements which would gg> indicate sodium channel blockade 13. Relative potency of compounds for measurements which would &] indicate potassium channel blockade 14. Relative potency of compounds for measurements which would ^0 indicate toxicity LIST OF FIGURES 1. Cardiac action potential and ion permeabilties 2. The effects of ischaemia on conduction in the heart 3. Proposed transmembrane arrangement of the sodium channel alpha subunit and proposed tertiary structure of the sodium channel 4. General structure of the compounds examined in this study 5. Structure of the compounds examined in this study 6. Typical Rat ECG 7. Dose-response curves for effects on PR interval from bolus administration of RSD939 8. Dose-response curves for effects on PR interval from bolus administration of RSD961 9. D25%s for adverse effects of XI series compounds in the in vivo screen 10. D25%s for ECG effects of XI series compounds in the in vivo screen 11. D25%s for adverse effects of X2 series compounds in the in vivo screen 12. D25%s for ECG effects of X2 series compounds in the in vivo screen 13. Dose-response curves for effects on VFT interval from infusion of RSD939 14. Dose-response curves for effects on VFT interval from infusion of RSD961 15. D25%s for adverse effects of XI series compounds in the electrical stimulation screen 16. D25%s for sodium channel blocking effects of XI series compounds in the electrical stimulation screen 17. D25%s for potassium channel blocking effects of XI series compounds in the electrical stimulation screen 18. D25%s for adverse effects of X2 series compounds in the electrical stimulation screen 19. D25%s for sodium channel blocking effects of X2 series compounds in the electrical stimulation screen 20. D25%s for potassium channel blocking effects of X2 series compounds in the electrical stimulation screen Abbreviations AP action potential APD action potential duration AV atrio-ventricular BP blood pressure ECG electrocardiogram D25% dose of drug producing a 25% change from control ERP effective refractory period HR heart rate Hz Hertz iT current threshold i.v. intravenous i.p. intraperitoneal K potassium MI Myocardial infarction MTD Maximum Tolerated Dose of drug mmHg millimeters of mercury ms millisecond(s) Ltmoles/kg micromoles/kilogram (imoles/kg/min micromoles/kilogram/minute MFF maximum following frequency Na sodium pH hydrogen ion concentration PVC premature ventricular contraction QSAR Quantitative Structure Activity Relationships RSD Rhythm Search Developments SA sino-atrial SEM standard error of the mean tT duration threshold VF ventricular fibrillation VFt ventricular fibrillation threshold VT ventricular tachycardia Acknowledgments I would like to thank Dr. Michael Walker who has put up with a lot of my nonsense over the years and, despite the grey hairs added to his head, gave me opportunities I never expected and taught me a great deal more than pharmacology. To my numerous compatriots who shared the room 413 experience, Mike, Eric, Sandro, Paul, Maria, Weiqun, Terry and numerous others that only stayed with us for a short period of time. We had our share of good times and rough times but we all learned from each other and hopefully we helped create something significant. Thanks also to Allen for the occasional kick in the behind and the chance to try (and hopefully succeed at) something completely different. Last but not least, great thanks go to my family for their support over the years. Thanks especially to my mother for always caring and struggling to provide the best for her children so that they would not have to struggle themselves. 1 1 Introduction The purpose of the studies described in this thesis was to examine certain structure activity relationships of a series of conformationally restricted arylacetamides and esters. Arylacetamides and their analogs are being developed as potential antiarrhythmics and accordingly there is a need to know the relationship between the structure of these compounds and their actions on the heart and the rest of the cardiovascular system. The basic hypothesis tested in this thesis was the following: \"that changes at either of the two nitrogens in a series of arylacetamides and analogs would change potency for various cardiovascular, ECG and indices of electrical stimulation in rats and mice.\" This hypothesis was tested by using a series of related compounds synthesized by RSD and testing their effects on blood pressure (BP), heart rate(HR), ECG and responsiveness to electrical stimulation using anaesthetized rats as the principal screen, and mice for testing lethality. This thesis contains information regarding the need for newer and better antiarrhythmics, the pathology of ischaemic arrhythmias, the mechanisms of action of class I and class III antiarrhythmics, and arylacetamides (and analogs) and their opioid and ion channel blocking actions. An attempt was made to correlate the structural changes in the different arylacetamides to changes in their potency for effects on the cardiovascular system, ECG and indices of electrical stimulation of the heart. 1.1 The Need for a New and Effective Antiarrhythmic The incidence of coronary heart disease in the United States is 110 cases per 10,000 person years for males and 64 cases per 10,000 person years for females. Death from coronary heart disease occurs at a rate of 16.6% for males and 12.5% for females. Of these deaths acute myocardial infarction (MI) accounts for 41.3% in males and 29.7% in females. This translates to approximately 1.5 million cases of acute MI, of which, 550,000 will die (350,000 prior to hospitalization) every year. Heart disease accounts for about 33% of deaths in the United States. Health care costs associated with coronary heart disease exceed 170 billion dollars annually (Biomedical Market Newsletter, 1992). Many agents have been developed for the prevention of ventricular tachycardia and fibrillation. However, these agents are limited to fairly short-term treatment due to severe toxic side effects which usually include neuronal effects and arrhythmogenesis. There is, therefore, a need for new effective and safe antiarrhythmic agents. A therapy which could improve survival by only 20% would prevent over 100,000 deaths in the U.S. each year, and, assuming the world-wide incidence is at least three times that in the U.S., would prevent over 300,000 deaths world-wide each year. The monetary savings to already strained health care systems would likewise be increased significantly if effective chronic treatments for the prevention of arrhythmias due to myocardial ischaemia were found. 6 1.2 The Normal Cardiac Action Potential Normal cardiac rhythm is the product of electrical impulses which travel through the heart via a specific path. This electrical signal, or action potential (AP), is generated by the movement of sodium and calcium ions into the cell, and the movement of potassium ions out of the cell. The differential distribution of ions between the outside and inside aspects of the cell membrane produces a resting potential across the cell membrane. Sodium is present at much higher concentrations outside the cell, 145 mM outside versus 10 mM inside. Likewise the calcium ion concentration is 2 mM outside the cell and normally only 0.0001 mM inside the cell. Potassium shows an opposite distribution with a concentration of 135 mM inside the cell and only 4 mM outside the cell. This resting potential provides the driving force for the flow of ions through cell membranes. The resting potential for cardiac cells is somewhere near -90mV. The action potential in the heart is very similar to that in the nerve but lasts for much longer. An action potential is initiated when the fast sodium channels open suddenly, allowing a rapid influx of sodium into the cell. The resulting depolarization (the potential across the cell membrane rises to approximately +20 mV) causes slow sodium-calcium channels and potassium channels to open. The calcium current is responsible for the plateau phase of the cardiac action potential during which time the potassium permeability actually decreases. The calcium not only maintains a positive membrane potential but is also involved in the excitation-contraction coupling in the cell. Another factor which contributes to the elongated plateau phase is the low chloride permeability of the heart, so that there is little short circuiting of the potential during the plateau phase. The potassium channels then open allowing the cell to return to a negative resting potential and thus ending the action potential. The cardiac action potential and ion permeabilities are illustrated in Figure 1. Time (milliseconds) Figure 1. Top shows a typical cardiac action potential illustrating rapid depolarization (phase 0), a brief period of repolarization (phase 1), followed by a plateau (phase 2) persisting for 0.1 to 0.2 seconds. There then occurs a period of rapid repolarization (phase 3) until the interval preceding the next depolarization (phase 4). Just below is a diagram of the ion permeability changes during this period. (From Fozzard, 1979) The cardiac muscle, like other excitable tissues, is refractory to restimulation during the action potential. The refractory period is therefore the time interval during which a cardiac cell cannot be re-excited by another impulse. There is, in addition, a relative refractory period during which the cell is less responsive to stimulation but can, nonetheless, be excited by a premature signal. 1.3 Abnormal Rhythm and Arrhythmias Arrhythmias are abnormal rhythms of the heart. The term arrhythmia refers to a deviation from the normal sequence of initiation and conduction of impulses. Some types of arrhythmias, such a premature ventricular contractions (PVC's) are brief and have no overall effect on heart rate. These brief arrhythmias, however can initiate more severe rhythms which can lead to more serious symptoms including unconsciousness and death. Severe arrhythmias can also occur spontaneously with limited or no prior occurrence of PVC's. Ventricular tachycardia (VT) and ventricular fibrillation (VF) are two types of life threatening arrhythmias. In humans tachycardias are arrhythmias with a heart rate greater than 100 beats per minute and may be non-sustained, lasting a few seconds, or sustained , which may last for minutes or hours. Tachycardias may also be subclassified according to their site of origin. A supraventricular tachycardia originates somewhere in the upper heart (either in the atria or the AV node) whereas a ventricular tachycardia originates from somewhere in the ventricles. Fibrillation is a rapid chaotic pattern of electrical activity in the heart. During ventricular fibrillation the ventricles are unable to contract rhythmically and are unable to pump blood to the body. Tachycardias (in the ventricles) and fibrillation in the atria can reduce the heart's ability to pump by interfering with the ventricle's ability to fill with blood. The potential causes of arrhythmias are numerous and include myocardial ischaemia and infarction, imbalances in serum potassium concentration or pH (Curtis, M.J., 1989, Saint, K.M. et al., 1992), excessive sympathetic nervous system discharge or adrenaline release (Szekeres, 1981, Daughery, A. et al., 1986), and drug toxicity such as that produced by digitalis (Hoffman and Dangman, 1987, Hondeghem and Mason, 1989) or alcohol (Koskinen, P. and Kupari, M., 1992). In order to develop a pharmacological route for the prevention of arrhythmias it is necessary to identify the possible underlying mechanism of arrhythmogenesis. Arrhythmias are generally due to either abnormal impulse generation or abnormal impulse conduction (Hoffman and Rosen, 1981; Hondeghem and Mason, 1989) 1.3.1 Abnormal Impulse Generation In normal cardiac rhythm electrical signals originate in the sinoatrial (SA) node. The nodal fiber cells are naturally leaky to ions therefore causing a slow depolarization. Several ionic currents are thought to contribute to the slow depolarization that occurs during phase 4 in the SA node: (1) an inward Na+ current, i f, induced by hyperpolarization; (2) a slow inward current, i s i , comprised of Ca + + and Na+ which becomes activated towards the end of phase 4; and (3) an outward K+ current, i K , that decays steadily throughout phase 4 (thus its opposition to the depolarizing effects of the two inward currents gradually decreases) (Brown, 1982). When the membrane potential reaches the threshold voltage for the slow channels, approximately minus forty millivolts, they become activated and initiate the action potential (Wit et al., 1974). The ends of the sinus node fibers are fused with the surrounding atrial tissue allowing the signal to propagate into the normal atrial muscle. In this manner the signal travels through the entire atrial muscle mass and to the atrioventricular (AV) node. The AV node delays the signal allowing the atria to complete their contraction and finish filling the ventricles with blood (Scheidt, 1983). The AV node is located in the posterior septal wall of the right atrium. The delay in conduction through the AV node is mainly caused by the lower resting potential of the cells and a smaller number of gap junctions between fibers. This results in a lower voltage to drive the flow of ions and greater resistance to the movement of the impulse from cell to cell. This combination of low voltage and high resistance slows the excitation of succeeding fibers in the AV node. The impulse then travels via specialized conduction fibers (Purkinje fibers)which rapidly distribute the signal to the ventricular cardio-myocytes which contract and pump blood from the heart (Scheidt, 1983). The Purkinje fibers, which lead from the AV node, through the AV bundle and into the ventricles, have properties opposite those in the AV node. They are somewhat larger than normal ventricular muscle fibers and show increased ion permeability of the gap junctions, relative to normal myocardial tissue. Purkinje fibers transmit electrical signals six times faster than normal cardiac tissue which provides for rapid transmission of the action potential to the entire ventricular mass. Other areas of the heart such as the AV node and the Purkinje fibers also demonstrate automaticity but their rate of impulse generation is much less than that of the SA node (Hoffman and Dangman, 1987; Wit et al, 1974). The SA node discharges at a rate of 70 to 80 times per minute compared to 40 to 60 discharges per minute in the AV node and 15 to 40 impulses per minute in Purkinje fibers. Impulses from these areas are only important in situations where the cells of the SA node, which functions as the primary pacemaker, are significantly depressed or blocked (Rosen and Spinelli, 1988). There are two major classes of arrhythmias caused by abnormal impulse generation. The first type are those resulting from truly spontaneous impulse generation (or automaticity) and do not rely on a prior impulse. The second type result from triggered activity where the generation of one or more impulses is a consequence of a prior impulse (Hoffman and Rosen, 1981). When the sinus rate is absent other sites may take over as pacemaker or a small area of the heart will sometimes become more excitable and cause abnormal impulses to be generated between normal impulses. Arrhythmias caused by such conditions are said to result from focal mechanisms where a wave of depolarization spreads from this irritable area and initiates a premature beat. The focus of abnormal impulse generation is called an ectopic focus. This type of mechanism is unlikely to be involved in ischaemia-induced arrhythmias because abnormal automaticity is suppressed by elevated extracellular K + (Hoffman and Rosen, 1981; Katzung et al., 1975). It has been demonstrated that extracellular K + rises rapidly in ischaemic myocardium (Hill et al., 1980; Hirche et al., 1980; K16ber, 1983). Triggered activity is a repetitive impulse initiated by a propagated or automatic AP. Triggered activity is thought to be a product of oscillations in membrane potential that follow the AP upstroke (Janse and Wit, 1989). These are called afterdepolarizations and may be either early, occurring during repolarization, or delayed, occurring when the membrane is completely repolarized or nearly so (Cranefield, 1977; Rosen and Reder, 1981). When afterdepolarizations are large enough to reach the threshold level the resultant AP is said to be triggered as they are by definition the product of a previous impulse and thus not automatic (Janse and Wit, 1989, Hoffman and Dangman, 1987). Delayed afterdepolarizations appear to occur more frequently when the preceding stimulation rate is high (Monk and Rosen, 1984) and may be caused by oscillations in the calcium concentration due to calcium loaded sarcoplasmic reticulum (Lazzara and Scherlag, 1988). Triggered activity may play a role in the generation of arrhythmias during ischaemia but direct proof of this is not available (Janse et al., 1982) 1.3.2 Abnormal Conduction of Impulses Abnormal impulse conduction may result from a complete failure of propagation or from unidirectional block and reentry of impulses (Hoffman and Dangman, 1987; Rosen and Spinelli, 1988; Hondeghem and Mason, 1989). Reentry means that during the arrhythmia there is continuous cyclical propagation of the impulse. Simply, reentry occurs when an impulse begins in one part of the heart, spreads in a circuitous pathway around the heart and then returns to the originally excited muscle, which has by that time repolarized, and \"reenters\" this muscle to stimulate it again. The excitatory signal continues again and again around the circle without stopping. Reentry cannot occur if the conduction velocity in the circuit is so fast, or the refractory period so long, as to block the circulating impulse in an area which is refractory. Reentry is only possible if the revolution time in the circuit is longer than the refractory period of the tissues involved (Brugada, 1987) This is illustrated in Figure 2 which depicts cardiac muscle strips as circles. Under normal conditions the muscle at the initiation point of the impulse is still refractory upon the completion of the circle by the impulse. If the length of the path around the circle increases then the site of origin of the impulse may no longer be refractory and may be restimulated. If the length of the pathway remains constant but the rate of conduction is decreased, then the originating tissue may similarly be restimulated. Finally, if the refractory period is decreased then the tissue may be ready to be restimulated when the impulse reaches the end of the circle. Ischaemia usually produces a decreased rate of conduction and increased path length. .13 Site of initiation Tissue still refractory as impulse returns Conduction around pathway slowed Figure 2. The effects of ischaemia on conduction in the heart. Pathways may be either lengthened or conduction may be slowed or both may occur. 14 It is becoming evident that reentry is probably the major mechanism for the generation of arrhythmias in ischaemic tissue in animal models (Lazzara and Scherlag, 1988; Kramer et al., 1985). Numerous studies have shown evidence for reentry by demonstrating continuous electrical activity in extracellular recordings between basic propagated beats and PVCs (Durrer et al., 1971; Boineau and Cox, 1973; Waldo and Kaiser, 1973; El-Sherif et al., 1977). In order to produce the conditions detailed above and thereby initiate and maintain reentry there must be unidirectional block of the impulse in a region of the heart, stable propagation at a sufficiently low velocity, delayed excitation of the tissue distal to the blocked site, and sufficient repolarization in the tissue proximal to the block so that sodium channels are capable of opening when the propagated impulse emerges from around the barrier and enters the region of initial block (Sasyniuk and Mendez, 1971). All the conditions necessary for reentry can be present during myocardial ischaemia: holes (scars or fixed barriers), slow conduction, and abnormal refractoriness (Downar et al., 1977; Lazzara and Scherlag, 1988). In this manner fibrillation may represent chaotic reentrant excitation or multiple continually migrating activation wavefronts (Moe et al., 1964). This has been termed random reentry as opposed to stable reentry based on the fixed anatomical path described above which may be the cause of ventricular tachycardias. In the case of fibrillation the path of excitation changes continuously such that individual groups of fibers may be repeatedly excited (Hoffman and Rosen, 1981). Reentry may also result from the reflection of impulses from an inexcitable tissue segment (Antzeleritch et al., 1985). If an action potential is delayed and then reaches an inexcitable segment it may reflect and re-excite the tissue proximal to the site of block (Hoffman and Dangman, 1987). These reflected impulses could be modulated by changes in the rate and rhythm of the heart (Hoffman and Dangman, 1987). It therefore seems logical to develop pharmacological compounds capable of changing the conditions in myocardial tissue such that they are unfavourable to reentry. Conduction velocity could be increased, refractoriness could be prolonged, or conduction could be blocked in a reentry pathway without prolonging refractoriness (Brugada, 1987) 1.3.3 Electrophysiological Effects of Myocardial Ischaemia Ischaemic myocardium suffers from an accumulation of metabolites and other substances such as protons, prostaglandins, potassium, adenosine, as well as others which can affect the electrophysiological environment (Williams et al., 1974, Bigger et al., 1977, Corr and Sobel, 1979). It is generally agreed that changes in pH and potassium ion concentration are the main contributors to arrhythmia generation (Davies, 1981, Goldstein et al., 1981). Normal metabolism in heart cells generates protons. Cardiac cells extrude protons to maintain intracellular pH at around 7.3 (Ellis and Thomas, 1976). It has generally been accepted that the Na+/H+ exchanger is the primary mechanism by which cardio-myocytes regulate intracellular pH. The Na+/H+ exchanger extrudes protons against an inwardly directed Na+ gradient. The fall in extracellular pH during ischaemia likely results from a combination of two factors. Firstly, there is an accumulation of protons which would be washed away in normally perfused tissue (Case et al., 1979; Ichihara et al., 1984; Gevers, 1977; Seeley, 1980). The second contribution appears to be anaerobic glycolysis in the anoxic cells producing protons, lactate and ATP (Williamson, 1966; Gevers, 1977; Seeley, 1980). Ischaemia also causes myocardial cells to release intracellular potassium (Harris et al., 1954; Hill and Gettes, 1980; Hirche et al., 1980). It is believed that hypoxic conditions impair the function of the Na7K+ pump (K16ber, 1983; Vleugels et al., 1980) thus resulting in a net K + efflux. Increased extracellular potassium is accompanied by a decreased resting membrane potential leading to a decrease in AP amplitude, decreased rise rate of phase 0 depolarization, and decreased action potential duration (APD) (Janse and Wit, 1989). Increased extracellular potassium, acidosis, and lack of oxygen and energy substrates combine to produce these effects (Janse and Wit, 1989). Slow and fast currents are both depressed to an equal degree by ischaemia (Cardinal et al., 1981). Contrary to the expected observation, effective refractory period (ERP) is not abbreviated (due to a decreased APD) but rather prolonged since it continues beyond full repolarization (Lazzara et al., 1978). This is called post-repolarization-refractoriness. Thus conduction delay and block necessary for reentry and resultant reentry arrhythmias are present during myocardial ischaemia (Botting et al., 1986). 1.4 Mechanisms of Antiarrhythmic Drug Action The Vaughan Williams classification (1970; 1984) of antiarrhythmic drugs is based on the electrophysiological actions of antiarrhythmic agents on cardiac tissue. However, the mechanisms of antiarrhythmic drug activity are less clearly defined (Davy et al., 1988). These compounds can affect the potentially arrhythmogenic tissue or reduce the incidence of initiating events such as PVCs or changes in the sinus cycle length (Davy et al., 1988). Nonetheless, use of current antianriythmic agents, with a few exceptions, has had little success at reducing the incidence of life threatening arrhythmias. The current thought on pharmacological intervention for the treatment of arrhythmias focuses on the modification of impaired conduction by increasing refractoriness or on reducing ectopic pacemaker activity. The modification of electrophysiological activity in the myocardium can be accomplished by selective blockade of ion channels in the cardio-myocytes. The majority of antiarrhythmic drugs are agents which block sodium, potassium, or calcium channels. Unfortunately, all of the current drugs have severe side-effects which preclude J5 their long-term prophylactic use in the prevention of arrhythmias (Echt et al., 1991). 1.4.1 Classification of Antiarrhythmic Drugs The most widely accepted classification of antiarrhythmic drug action was initially proposed by Vaughan Williams (1970) and later modified (1984). Vaughan Williams separated agents according to their predominant electrophysiological effects on the action potential rather than separating them according to chemical structure. This is a descriptive classification based on experiments with normal rather than diseased tissue and relevance to the clinical setting has yet to be established (Zipes, 1987; Brugada, 1990). Many drugs have actions that fall into more than one class or subclass. Moreover, many agents have active metabolites with a different class of action than that of the parent molecule. The drugs are generally separated into four classes. Class I agents are predominantly sodium channel blocking agents. p%adrenoceptor antagonists which depress phase 4 depolarization are known as class II drugs. Class III agents are drugs whose main site of action is on potassium channels. Finally calcium channel blocking drugs are grouped into Class IV and act by decreasing the rate of rise of phase 4 depolarization and slowing conduction in tissues dependent on calcium currents (such as the AV node). 1.4.2 Sodium Channel Structure The voltage gated sodium channel is a large (about 1800 amino acid) membrane bound protein. The primary structure of several types of sodium channels has been elucidated by cDNA cloning and sequencing and the secondary and tertiary structure is now becoming more clearly understood (Sato and Matsumoto, 1992). It has generally been agreed that the model proposed by Noda et al (1987) is representative of the secondary structure of the sodium channel. The sodium channel consists of four homologous transmembrane repeats each made up of six oc-helices numbered SI to S6. All of these helices save the S4 segment are hydrophobic or amphipathic. The S4 segment contains a positively charged residue at almost every third position in its sequence and is thought to form the voltage sensor for the channel (Mulvey et al., 1989, Stuhmer et al., 1989). There are four large extracellular loops, one connecting helices five and six in each of the repeats. These loops are proposed to fold down into the pore formed by the helices and form the channel lining and selectivity filter (Sato and Matsumoto, 1992). The selectivity filter is thought to consist of two clusters of amino acids at equivalent positions on the S5-S6 connecting segments which would form two rings of charged amino acids in the channel pore. Cluster one consists of aspartate 384, glutamate 942, methionine 1422, and aspartate 1714. Cluster two is made up of residues glutamate 387, glutamate 945, methionine 1425, and aspartate 1717 (Pusch et al., 1991, Heinemann et al., 1992). -/SO Na* channel Figure 3. Proposed transmembrane arrangement of the sodium channel alpha subunit and proposed tertiary structure of the sodium channel (figure from Noda et al, 1987). 1.4.3 Class I Agents The classification of antiarrhythmic agents based on classical ion channel blocking properties is complex. Nevertheless, these agents have consistently demonstrated antifibrillatory and antiarrhythmic actions against ischaemia-induced arrhythmias. Class I agents have a long history of use in the treatment of arrhythmias and of development by modification of chemical structure (Courtney, 1988). Unfortunately, endeavours to improve on prototypical class I drugs have met with little success (Schlepper, 1989). Upon the realization that many of the class I antiarrhythmics differed in their mode of action on Na+ channels, a subclassification of the class I agents was suggested (Vaughan Williams, 1984; Harrison, 1985). The class I agents were subclassified according to their different effects at clinical concentrations. Class la compounds slow the rate of rise of phase 0, thereby slowing conduction, and causing prolongation of refractoriness resulting in a widening of QRS and prolongation of Q-T interval (Harrison, 1985). The prototypical class la drugs include quinidine, procainamide, and disopyramide. Class lb drugs which slow conduction slightly with no effect on refractoriness exhibit limited (if any) effect on QRS duration at normal heart rate while shortening the QT interval and AP duration and elevating the fibrillation threshold (Harrison, 1985). Drugs which fall into this class include lidocaine and mexilitine. Class Ic agents such as flecainide and encainide exhibit marked slowing of conduction with slight prolongation of refractoriness (Harrison, 1985). These electrophysiological differences are also seen clinically (Milne et al., 1984). Later studies examined the onset kinetics of rate dependent block by class I agents, showing that the three classes could also be separated on the basis of their rate of onset. Class lb agents showed fast onset kinetics, class la drugs exhibited intermediate kinetics, while class Ic antiarrhythmics had slow onset kinetics (Campbell, 1983a, 1983b). Class I agents have been limited in their long term use by numerous toxic side-effects. These drugs show poor selectivity for cardiac sodium channels and often block neuronal channels both in the peripheral and central nervous systems. In addition, sodium channel blockers may be proarrhythmic and increase the likelihood of a fatal arrhythmia (Selzer, 1964; Bergey, 1982; McKibbin, 1984; Carson, 1986). This is due to the normal function of the heart which relies upon defined routes, rates of conduction, and refractoriness to produce a single orderly sinus beat and then extinction of the action potential after its propagation (Scheidt, 1983). Any drugs which modify action potential and refractory period could induce as well as prevent arrhythmias (Singer at al., 1969; Cranefield et al., 1971; Hope etal., 1974). 1.4.4 The SAR of existing Class I Agents Class I antiarrhythmics in clinical use have diverse structures. However there are certain structural similarities amongst them. Class I drugs generally have a relatively large lipophilic constituent, often an aromatic or alicyclic/heterocyclic ring. In most cases the alkyl linking chain is attached to the lipophilic parts by an ester or amide linkage. The structures usually incorporate a hydrophilic group at the end of the alkyl linking chain, often an amino nitrogen which is protonated at normal physiological pH. Current models of antiarrhythmic drug activity assume that class I drugs bind to a specific site on cardiac sodium channels causing blockade of the channel and that these agents have different binding affinities for different states of the Na+ channel (Hille, 1977; Hondeghem and Katzung, 1977; 1984; Starmerm 1984). Binding studies have shown evidence for a specific binding site for Class I drugs on the cardiac sodium channel that may overlap or is associated with the binding site for batrachotoxin (Sheldon et al., 1991). Several studies have between them correlated activities of many antiarrhythmic drugs with their physiochemical characteristics (Courtney et al., 1978a,b; Courtney, 1979; 1980a,b,c; Sada and Ban, 1980; 1981a,b,c). These studies demonstrated that the lipid solubility at physiological pH (estimated by logP and pKa) of class I drugs plays an important role in their blocking of sodium channels in the resting state. More lipophilic drugs seem to be more potent at blocking activated channels. More recently 4D-Quantitative Structure Activity relationships (QSAR) studies have also shown that lipid solubility, supplemented by spatial descriptors, is predictive of in vivo antiarrhythmic activity in a series of substituted N,N-Dimethyl-3-phenylpropyl-amines (Klein and Hopfinger, 1998). There was also a correlation between molecular weight and onset of frequency dependence such that lower molecular weight drugs had a faster onset than higher molecular weight compounds. In addition these studies showed that the rate of recovery from frequency dependent block correlated well with molecular weight. This correlation was improved when logP and pKa were taken into account (permanently charged molecules showed very slow recovery regardless of molecular weight). More recent studies with derivatives of lidocaine (Sheldon et al., 1991) and tocainide (Sheldon and Thakore, 1995) have shown that there are several interactions between these compounds and their binding sites. The drug binding site recognizes an aryl group linked by an alkyl chain to an amine. The orientation of the aryl group and amine group to the rest of the molecule is also of importance. There is a hydrophobic pocket at the amino terminal end that accommodates four or more aminoalkyl carbons. There is a voltage sensitive site near this pocket that binds the amine. Next to this site is a sterically limited pocket that accommodates an alkyl link two carbons in length. There is a link domain which can be an ether, ester, or amide link between the two carbon alkyl link and the aryl ring. Near this link domain is a hydrophobic pocket for at least six carbons with several subdomains around this pocket. It appears that there are hydrophilic subdomains at the ortho sites and hydrophobic pockets at the meta and para positions (Sheldon and Thakore, 1995). Another study (Courtney, 1987) elucidated the role of the linkage between the alkyl link and the aryl ring. It appears that ether-linked drugs such as mexilitine are more potent than equivalently lipid soluble amide-linked drugs such as tocainide. This pattern is also evident in the |3-blockers alprenolol and propranalol which are ether-linked and more potent than the equivalent amides. Steric factors also appear to be of some importance in channel blocking ability. Quinidine, which has a complex heterocyclic structure, is less potent than an equivalently lipid soluble amide linked structure (Courtney, 1987). This, along with results from studies on myelinated nerve and skeletal muscle (Courtney, 1983), suggests that bulk around the linkage between the lipophilic ring and the tertiary nitrogen might interfere with binding. Further studies with a series of lidocaine and procainamide derivatives which explored all of the structural differences in various combinations also showed that no single structural change accounted for the difference in the activity of these agents. There was a strong correlation between lipid solubility and channel blocking activity (Ehring et al., 1988). Similarly, this and other studies have also shown a strong correlation between molecular weight, and also molecular size, and kinetics of offset from sodium channels. Smaller compounds seem to exhibit more rapid dissociation 2la from the channel than larger molecules. This is of importance as kinetically slower drugs appear to be more cardiotoxic. 1.4.5 Potassium Channel Structure There are many different types of potassium channels found in numerous tissue types (Palotta and Wagoner, 1992). At least seven different types of potassium channels have been identified in the heart. The structure of these channels, namely, the amino acid sequence, has allowed them to be grouped in one of three super families; the Shaker type, the Mini, and the inward rectifier/ATP sensitive families (Hoshi and Zagotta, 1993; Jan and Jan, 1992). The Shaker channels have a structure very similar to sodium channels, but instead of being made up of a single protein, Shaker channels are homotetramers with each subunit consisting of six transmembrane alpha helices, SI to S6 (Mackinnon, 1991). The pore is formed, as in sodium channels, by a folding in of the S5-S6 connecting region and the S4 segments are also implicated as voltage sensors (Mackinnon, 1991; Liman et al., 1991). An eight amino acid sequence found on the S5-S6 connecting segment is highly conserved amongst the potassium channels and is thought to make up the selectivity filter (Mackinnon et al., 1988; Heginbotham et al., 1994). The potassium channels from Streptomyces lividans have been shown to possess sequence similarity to known K + channels, particularly in the pore region, (Mackinnon et al, 1998). Recent x-ray crystallography studies on these channels has produced more information on the structure of the pore region(Doyle et al, 1998). X-ray analysis with data to 3.2 angstroms shows that there are four identical subunits which create an inverted cone, cradling the selectivity filter of the pore in its outer end. The narrow selectivity filter is 12 angstroms long, whereas the remainder of the pore is wider and lined with hydrophobic amino acids. A water-filled cavity and helix dipoles are positioned so as to overcome electrostatic destabilization of an ion in the pore at the center of the bilayer. Main chain carbonyl oxygen atoms line the selectivity filter to coordinate K + ions but not smaller Na+ ions. The selectivity filter itself contains two K + ions about 7.5 angstroms apart which promote ion conduction by exploiting electrostatic repulsive forces to overcome attractive forces between K + ions and the selectivity filter. The architecture of the pore establishes the physical principles underlying selective K + conduction. Mini channels consist of several small (130 amino acid) subunits assembled into a functional channel (Takumi et al., 1988). It is not known how many subunits are found in the complete channel but it is thought that at least four are required (Goldstein and Miller, 1991). Subunits consist of a single transmembrane segment with the amino terminus found on the extracellular side and the carboxyl terminus exposed to the intracellular environment (Wilson et al., 1994). Inward rectifier and ATP-sensitive channels are also tetramers composed of subunits containing two transmembrane segments separated by a pore-forming connecting region (Tygat et al., 1994). There is sequence homology between these subunits and the S5-S6 region of Shaker type channels (Kerr and Sansom, 1995). These may be seen as a Shaker type channel without the S1-S4 sequences. The lack of the S4 segment probably accounts for the lack of voltage sensitivity of these channels. 1.4.6 Class III Agents If the reentrant circuit has a zone of depressed conduction it appears logical that further depression of conduction with class I drugs may prevent or terminate the arrhythmia. If the reentrant circuit has a short excitable gap, the most obvious goal for treatment would be to prolong the refractory period (Task Force on Arrhythmias, 1991). Prolongation of the duration of the action potential with an increase in the absolute and effective refractory periods is the mode of action of Class III agents (Singh and Vaughan Williams, 1970; Hondeghem and Snyders, 1990). Refractoriness is prolonged by delaying repolarization of myocardial cells thus delaying the voltage dependent recovery of Na+ channels. If all other properties of a reentry remain equal, prolongation of the refractory period would make initiation of reentry more difficult since the wavefront would encounter a wall of refractory tissue where previously was re-excitable tissue. Prolongation of refractoriness would prolong the wavelength and when the wavelength becomes larger than the available tissue, the reentry is terminated. The principle has been demonstrated experimentally in several models (Spinelli and Hoffman, 1989; Hoffman and Spinelli, 1989; Zuanetti and Corr, 1991). There are at least seven types of potassium channels in cardiac cells which may play a role in the action potential. The most important target for class III antiarrhythmic drugs appears to be the delayed rectifier Ik (Colatsky et al., 1990). There are two reasons why blocking Ik may be successful. Firstly, the contribution of Ik to repolarization should be greater at shorter cycle lengths because the slow deactivation- of Ik during diastole will result in a progressive increase in the number of open channels during the action potential at high heart rates (as in tachycardia). This is known as negative use-dependency. Secondly, the blocking effects of drugs should be increased in partially depolarized tissue such as ischaemic tissue. Experimental evidence shows that blocking Ik may be effective in preventing ischaemia-induced VF (Lynch et al., 1985). In addition, amiodarone and sotalol, which block Ik channels have been shown to have efficacy in the prevention of life threatening arrhythmias, presumably due to APD prolongation (Nademanee et al., 1985; Taggert et al., 1985; Johnston et al., 1985; Nademanee, 1992). Difficulties are encountered when considering the first case. In practice, Ik blockers appear to function more effectively at slow heart rates and to be less effective at high heart rates (Roden et al., 1988). This phenomenon has been named \"reverse use dependence\" (Hondeghem and Snyders, 1990). Further confusion is added when one considers that many class I drugs also appear to block Ik currents. The mechanism is not clear but it has been suggested that at rapid rates the contribution of other channels, such as calcium channels, to repolarization may increase and overrule the role of Ik (Colatsky et al., 1990). The above presents problems because prolongation of the action potential at long cycle lengths increases the probability of afterdepolarizations and triggered arrhythmias (Jackman et al., 1988; Carlsson et al., 1993, O'Rourke et al., 1994; Opthof, 1994). The most frequently observed form of arrhythmia associated with class III toxicity is a polymorphic VT called torsade des pointes (Schwartz, 1985; Surawicz and Knoebel, 1985; Jackman et al, 1988). 1.4.7 The SAR of existing Class III Agents Studies with local anaesthetics and class III antiarrhythmic drugs have suggested that electron withdrawing substituents on aryl groups may be important in diminishing sodium channel activity (Phillips et al., 1990). One example is procainamide which is a class la agent with well defined effects on conduction and refractoriness. Procainamide's primary acetylated metabolite, N-acetyl-procainamide (NAPA) is able to prolong the time course of repolarization without affecting conduction (Bagwell, E.E. at al., 1976). Sematilide, an investigational drug with a pure class III electrophysiological profile, similarly replaces the para-NH 2 group of procainamide with a methylsulfonamido group (Lumma et al., 1987). A similar pattern is seen in a series of related compounds in which the parent molecule (Wy-47,804) has a class I profile. Substitution of an electron-withdrawing 4-N02 group for an electron donating 4-NH2 group produces a class III agent (Buzby, 1987; Colatsky, 1987; Follmer, 1988, 1989). Even though there are large electronic changes to the molecule which affect the channel selectivity, these substitution did not have a great effect on lipid solubility and molecular weight. Examination of these investigational agents suggests that the site of action at the sodium channel is sterically restricted and electron deficient. Studies have not revealed a clear SAR relationship between class III activity and molecular size or lipid solubility as they have for some of the class I agents. However evidence exists for hydrophobic interactions between class III compounds and their binding site in the potassium channel. Studies were done on the binding of dofetilide, a potent blocker of the 1^ potassium channel, to cloned human potassium channels subjected to site-directed mutagenesis (Kiehn et. al., 1996). A single mutation changing the isoleucine residue at position 177 to a cysteine reduced the hydrophobicity at this location and added hydrogen bonding capability to this residue. Dofetilide has strongly hydrophobic aromatic functions and this mutation resulted in a much higher rate of unblocking of the channels after block by dofetilide. Since there was no effect on the voltage dependence of block it would appear that the hydrophobic rings in dofetilide are responsible for the strength of the binding between drug and binding site but do not affect its initial interaction and block of the channel. Studies examining the effects of point mutations to human potassium channels on the binding of quinidine, a class I compound with class III activity, have also demonstrated the importance of hydrophobic interactions (Yeola et.al., 1996). Mutations which increased the hydrophobicity at residue 505 of the h.Kvl.5 potassium channel enhanced quinidine binding. This is consistent with the hypothesis that quinidine binding is also stabilized by hydrophobic interactions. 1.5 Arylacetamide Opioid Receptor Agonists and Ion Channel Blockade A number of arylacetamide opioid receptor agonists, especially the K -receptor agonists, appear to posses cardiovascular actions which may be mediated by direct actions on the heart and blood vessels. A number of studies have demonstrated that some opioid agonists, such as the selective K-receptor agonist U50,488H, have antiarrhythmic actions in rats subjected to coronary artery ligation (Fagbemi et al., 1983; Pugsley et al., 1992a, 1992c). The non-opioid properties of many of these agents may be due to blockade of sodium channels. It was previously shown that U50,488H has sodium channel blocking activity in neuronal cells (Alzheimer and Bruggencate, 1990). In addition studies have shown that the anticonvulsant K-receptor agonist U54,494A is also capable of blocking sodium channels in neuroblastoma cells (Zhu and Irn, 1992). More recently, effects on the cardiac sodium channel and antiarrhythmic properties have been demonstrated with U50,488H (Sarne et al., 1991, Pugsley et al., 1992c, Zhu et al., 1992) as well as with the K-agonists PD 129290 and PD 129289 (Pugsley et al., 1993). These effects were not modulated by opioid receptor antagonists further suggesting a direct effect on the ion channels rather than any involvement of opiate receptors. 1.6 Objectives It is evident that the selection of an appropriate therapy for ventricular arrhythmias is complicated by compounds with several modes of action and numerous toxic effects. Attempts to improve on existing agents through chemical modification so far have met with little success. Use of current ion channel blocking antiarrhythmics for long periods of time in people who have not demonstrated a recent infarct but may be at risk is not recommended. Most of the current agents are reserved for fairly acute use in individuals with recent infarcts. Rhythm Search Developments Ltd. is a company developing new antiarrhythmic agents. In order to develop safer and more effective drugs four properties required to produce an \"ideal\" antiarrhythmic drug were identified: 1. Selectivity for cardiac ion channels 2. Mixed channel blockade (i.e. blockade of sodium and potassium channels) 3. Selective blockade of channels under ischaemic conditions 4. Positive frequency dependence, i.e. more ion channel blockade at higher heart rates. The purpose of this study was to examine a new family of ion channel blocking agents, which show some limited 3-dimensional similarity to U50,488H (RSD 925), in order to assess their potential application to antiarrhythmic therapy. The study focused on the second ideal property from the list above, that is, the profile of sodium and potassium channel blocking activity. The first property from the list above was partially examined by assessment of the toxicity of the compounds. Compounds with greater selectivity should show decreased toxicity relative to their ability to block ion channels in the heart. A series of six compounds with the same alkoxyaryl group (3,4-dichlorophenoxymethylene) were examined in an attempt to correlate structural changes at the two linking hetero atoms with changes in sodium and potassium channel blocking activity, toxicity, and electrophysiological effects on isolated whole cells. The general structure of the series and the principal sites of modification, called XI and X2, are shown in figure 4. These compounds were tested in a number of whole animal and single cell preparations in order to examine their effects from an organ/system level down to the cellular level. An attempt was then made to correlate the structural chemical differences with observed differences in pharmacological activity. Figure 4. Specific configuration of the R,R-trans form of the compounds used in this study showing the XI and X2 sites which were modified. All compounds tested were equimolar mixtures of R,R and S,S enantiomers. 3b 2. Materials and Methods 2.1 Compound Preparation All compounds were synthesized by Rhythm Search Developments Ltd. for Structure Activity Relationship studies as part of the company's antiarrhythmic project. Figure 5 illustrates the structures of the compounds examined. The logP and pKa values were calculated by quantum mechanical methods using Cache software and Tallas software version 1.2 respectively. RSD 939 has an ether linkage between the dichlorobenzene ring and the rest of this molecule. RSD 939 served as a prototype compound for this study. RSD 988 and RSD 952 are two analogues of RSD 939 which possess a secondary nitrogen or an oxygen respectively at the XI position rather than the tertiary nitrogen of RSD 939. These compounds were synthesized in order to examine changes in steric bulk and electron density around the XI hetero atom. The compound RSD 969 examines the effects of lowering the pKa of the amino nitrogen at X2 (the pyrollidinyl nitrogen) by substitution of a morpholino for the pyrollidino group. RSD 971 is an analogue of RSD 939 with a quaternary nitrogen atom at the X2 position thus providing insight into the effect of permanent positive charge at this site as opposed to RSD 969 which will only be fully protonated (positively charged) at pH significantly below physiological pH (such as that seen in ischaemic tissue). RSD 961 has an azepine ring rather than a pyrollidine ring at the X2 position thus examining the effect of increased bulk in this region of the molecule. Bulk is further increased and pKa reduced (compared to RSD929) in RSD 981 which possesses a di methoxyethyl amino group at the X2 position. The chemical names and formulas of the RSD compounds tested are illustrated in table 1. as Figure 5. Structures of the Compounds examined in this study Table 1. The chemical names and formulas of the compounds used in this study Chemical Formula C 1 9 H 2 6 N 2 0 2 C 1 2 Compound Chemical Name RSD939 (±)-rra«5-N-methyl-N-[2-(l-pyrrolidinyl)cyclohexyl] (3,4-dichlorophenoxy)acetamide RSD952 (±-)rra«i-[2-(l-pyrrolidinyl)cyclohexyl] (3,4-dichlorophenoxy)acetate RSD961 (±)-fram-N-methyl-N-[2-(l-hexahydroazepinyl) cyclohexyl] (3,4-dichlorophenoxy)acetamide RSD969 (±)-rra«i-[2-(morpholinyl)cyclohexyl] (3,4-dichlorophenoxy)acetate RSD971 (±)-rra/w-N-methyl-N-[2-(l-methylpyrrolidinium) cyclohexyl] (3,4-dichlorophenoxy)acetamide RSD981 (±)-fran*-N-mediyl-N-[2-(bis(2-methoxyemyl)arrunyl) C 2 1 H 3 2 N 2 0 4 C 1 2 cyclohexyl] (3,4-dichlorophenoxy)acetamide RSD988 (±)-rra«5-[2-(l-pyrrolidinyl)cyclohexyl] (3,4-dichlorophenoxy)acetamide C 1 8 H 2 3 N 0 3 C 1 2 C 2 i H 3 0 N 2 O 2 C l 2 C 1 9 H 2 6 N 2 0 3 C 1 2 C 2 o H 2 9 N 2 0 2 C l 2 Ci gH 2 4 N 2 0 2 Cl 2 - t o 2.2 Preliminary In Vivo Screening This screen gave an initial profile of the acute cardiovascular actions of the compounds. The primary testing was used to determine effects on blood pressure (BP), heart rate (HR), and ECG. In addition, the cause of death was noted. 2.2.1 Preparation and Experimentation Male Sprague-Dawley rats (250-350g)were anaesthetized with pentobarbitone (65 mg/kg i.p.). The carotid artery was cannulated for measurement of blood pressure. The cannula was connected to a pressure transducer monitored by a Grass Polygraph (model 79D) and kept open with a leak pump attached in series. The jugular vein was cannulated for drug administration. ECG was measured with subcutaneous needle electrodes in a Lead II configuration. ECG leads were also connected to the Grass Polygraph. A tracheotomy was performed with animals left to breathe spontaneously. After a control period of at least 15 minutes to obtain stability, animals (n=5) were administered drug. Drug was injected in bolus doses up to a maximum volume of 1.0 ml/lOOg body weight. Doses were successively doubled until the death of the animal. After each dose a trace at a speed of lOOmm/sec for 2 seconds, was obtained 30 seconds, 1, 2,4, and 8 minutes after administration. The dose at which the animal died and the cause of death, either by cardiac arrhythmia, an irreversible decline in blood pressure, or respiratory failure was noted. 2.2.2 Data Analysis The experimental trace was analyzed for the mean arterial BP, HR, and ECG parameters at each recorded time for every dose. The ECG was analyzed for drug effects on various parameters which are indirect measures of ion channel blocking activity. Changes in PR and QRS intervals were taken as indicative of sodium channel blockade. Changes to another ECG measure called RSh have also been found to correlate with sodium channel blockade (Penz et al., 1992). Due to the unique nature (lack of a clear T wave and the asymmetric nature of the T wave) of the rat ECG measuring QT intervals can be problematic. The rat action potential is characterized by a rapid repolarization phase which has been postulated to be due to the I t 0 current (Josephson, et al., 1984). Most animals larger than a rat show a prolonged plateau in the action potential prior to repolarization such as that seen in guinea pig myocardium (Beatch, et al., 1990). The QT interval in the rat is not as well defined as in other species (Budden, et al., 1981), and thus two measures were made. QT1 was measured from the Q wave to the peak of the ST segment and QT2 was measured from the Q wave to the inflection point of the down stroke of the ST complex. Drug induced widening of the QT intervals were taken as being indicative of potassium channel blockade. Figure 6 illustrates a typical rat ECG and the points used to measure various intervals. Figure 6. A representation of the rat ECG with intervals measured. Dose response curves for the various measures were plotted for each animal individually. A line of best fit to a simple exponential model was determined by computer and doses for a 25% change from control (D25%) were determined from the equations used to model the dose response curves. The mean D25% and S.E.M. for each measure was calculated from each group of five animals. The D25% values of the various compounds were compared with each other using analysis of variance (ANOVA). 2.3 Electrical Stimulation Studies Electrical stimulation studies were performed to further examine the electrophysiological actions of the drugs in intact rats. Drugs were administered as infusions while monitoring BP, HR, and ECG, as well as examining for signs of toxicity. In addition a program of electrical stimulation provided information on possible effects on myocardial sodium and potassium channels. It has been shown that drugs which block sodium channels increase the threshold current (iT) and pulse width (tT) needed for the capture of single beats. These drugs also increase the threshold current for the induction of ventricular fibrillation (VFt) (Vaughan-Williams and Szekeres, 1961). Drugs which block potassium channels also suppress the induction of VF by making the heart refractory to higher frequency stimuli (Winslow, 1984). Potassium channel blockers would thus prolong the effective refractory period 4+ (ERP) and decrease the rate at which the heart could follow an external stimulus (maximum following frequency - MFF) (Vaughan-Williams, 1970). 2.3.1 Preparation and Experimentation Pentobarbitone anaesthetized (65 mg/kg i.p.) male Sprague-Dawley rats (250-350g) underwent carotid artery cannulation for measurement of blood pressure. The cannula was connected to a pressure transducer monitored by a Grass Polygraph (model 79D) and kept open with a leak pump attached in series. Drug was administered via a jugular vein cannula. ECG was measured with subcutaneous needle electrodes in a Lead II configuration. ECG leads were also connected to the Grass polygraph. A tracheotomy was performed and animals ventilated if respiration became impaired. The method for electrical stimulation of the heart in intact rats has been described elsewhere (Howard and Walker, 1990). Briefly, ventricular stimulating electrodes were constructed from Teflon coated silver wire and inserted into the left ventricular mass by trans-thoracic puncture with a 27 gauge needle. Electrodes were typically placed close to the apex of the ventricle and approximately 1-2 mm apart. Placement was confirmed by post-experimental dissection. Stimulation, according to the protocols given above, was applied every five minutes until three sets of stable control values were obtained. Drug was administered to animals (n=5) by infusion for five minutes with electrical stimulation testing after three minutes of infusion. Infusion rates were then doubled and measurements repeated until the death of the animal. After each dose a trace at a speed of lOOmm/sec for 2 seconds was obtained 1, 2, and 3 minutes after drug administration. The dose at which the animal died and the cause of death, either by cardiac arrhythmia or an irreversible decline in blood pressure, was noted. 2.3.2 Stimulation Protocol Endpoints for electrical measures were determined using a Honeywell E for M oscilloscope and square wave pulses were used to determine electrical stimulation variables. Each variable was measured three times and a mean value recorded. Threshold current (iT) for capture was determined at a 1 ms pulse duration and 7.5 Hz by uniformly increasing the applied current until capture. Threshold pulse width (tT) was determined by uniformly increasing pulse duration at 7.5 Hz and twice the threshold current. Maximum following frequency (MFF) was determined at twice the pulse width and current thresholds. Frequency of stimulation was increased from a frequency of 7 to 20 Hz. MFF was taken as the point at which the heart failed to follow, on a 1:1 basis, the stimulating pulse. This was evident as an upward spike in the blood pressure after a sustained decrease upon initiation of stimulation. The ventricular fibrillation threshold (VFT) was determined at twice the threshold duration and at 50 Hz. The minimum current required to produce sustained fibrillation, as seen by a sustained and precipitous fall in blood pressure, was taken as the threshold. Effective refractory period (ERP) was determined by the extra stimulus method. An extra single electrical stimulus was applied to the heart at various intervals behind a pacing stimulus (7.5 Hz). The shortest interval between the pacing interval and the extra stimulus in which an extra-systole was obtained was determined to be the ERP. 2.3.3 Data Analysis As in the above experiment, the trace was analyzed for mean arterial BP, HR, and ECG at all recorded times for each dose. Again the ECG parameters measured were PR, QRS, and QT1 and QT2 intervals and RSh. Dose response curves for the various measures were constructed for each animal using the values at three minutes after drug infusion at a particular dose. A line of best fit to a simple exponential model was determined by computer and doses for a 25% change from control (D25%) were determined from the equations used to model the dose response curves. The mean D25% and S.E.M. for each measure was calculated based on the five samples. The D25% values of the different compounds were compared with each other using analysis of variance 41 (ANOVA) in order to determine whether differences existed in the activity of the different compounds. 2.4 Lethality Studies in Mice Acute maximum tolerated dose (MTD) was determined in conscious male albino mice (weights were approximately 30g). CD-I mice between 20-25 g were warmed with a heating lamp to produce venodilation in the tail veins. Injections of drugs (volumes < 0.2 ml.) were made in the distal part of the tail with lcc syringes and 27 gauge needles. Experiments were begun with high doses in two mice and if both mice died, the dose was decreased 10 fold. This was repeated until one or both mice survived. Once one or both mice survived another two mice were injected with the highest minimum lethal dose. Groups of seven mice were then injected with diluted concentrations (1:1 dilution) of drug until no mortality was seen. The doses and mortality were then entered into a computer program for the calculation of MTD. 2.5 Patch Clamp Studies Compounds were examined using various patch-clamp techniques in order to further examine their electrophysiological effects on ion channels. The main experiments were on isolated rat myocytes to ascertain the intracellular and extra cellular sodium channel blocking effects of these drugs. The drugs were also investigated to determine effects on some of the currents mediated by potassium channels. These experiments were conducted by members of Dr. J.G. MacLarnon's research group in his laboratory at the University of British Columbia. The data was used to ascertain whether the ion channel blocking activity seen in the in vivo experiments of this study correlated with that produced in vitro. 3. Results 3.1 In Vivo Screening Examples of the data generated in the in vivo screens are shown in Figures 7 and 8, the dose response curves for the PR interval of RSD939 and RSD961. Following the curves is a comparison of the D25% values for these two compounds (table 2). 5 0 Oi 40' y = 64.209 » IO0 0 0 6 * r 2 = 0.758 y = 67.035 * l o 0 0 1 0 x r 2 = 0.981 y = 64.684 * 10°0l0x r 2 = 0.839 y = 63.919 * 10\"' = 0.982 y = 66.689 * I00 0 0 9 x r 2 = 0.870 O » o A C o n t r o l 1 Dose (u.mole/kg) 10 Figure 7. Dose response curves for the effect of bolus administration of RSD 939 on the PR interval in five rats. A dose dependent prolongation of the PR interval was evident. Shown in the legend are the equations for the line of best fit to a simple exponential model. RSD939 was administered in doses ranging from 1 to 16 Limole/kg. The D25%s for the five rats were 7.4, 7.9, 5.4, 22, and 7 umole/kg. The mean D25% ± S.E.M. was 9.9 ±3.0 Limole/kg. t 5L 90-t 50H 4 0 H 30J y = 57.859 * IO 0 0 1 2 * r 2 = 0.946 y = 58.328 * 10°°<«* r 2 = 0.764 y = 65.864 * 100-005* r 2 = 0.932 y = 65.022 » 10 0 0 0 7* r 2 = 0.978 y = 61.539 * IO 0' 0 0 9* r 2 = 0.982 » 20 J Control 1 0 Dose (|imole/kg) Figure 8. Dose response curves for the effect of RSD 961 on PR interval in five rats. Bolus administration produced a dose dependent prolongation of the PR interval. Shown in the legend are the equations for the line of best fit to a simple exponential model. RSD961 was administered in doses ranging from 0.5 to 16 fxmole/kg. The D25%s for the five rats were 8.1, 11.2, 20.4, 15.2, and 11 fxmole/kg. The mean D25% + S.E.M. was 13.2 ±2.1 (xmole/kg. 51 Table 2. Statistics for D25%s for prolongation of the PR interval generated from the dose response curves in figures 7 and 8. Sample variance was relatively large. The data for RSD939 was more skewed and had a wider range than that for RSD961. Compound RSD939 RSD961 D25%s for PR Interval (Limole/kg) 7.4 8.1 7.9 11 5.4 20 22 15 7.0 11 Mean 9.9 13 S.E.M. 3.0 2.1 Median 7.4 11 Standard Deviation 6.8 4.8 The results of the in vivo screening are summarized in table 3 which shows the D25%s for the various measures. Many of the tested compounds had little effect on QRS over the dose range tested. Only RSD 939 showed a 25% change in QRS before death. Other drugs did not produce a 25% change in QRS interval before the death of the animal thus an D25% was not estimated. RSD 981 was approximately equipotent for effects on the ECG indicative of blockade of sodium and potassium channels or but shows a tendency to greater potency for effects on the QT intervals which correlates with potassium channel blockade. The other compounds were more potent for effects on the PR interval and RSh which would suggest greater potency for sodium channel than potassium channels blockade. All compounds, except RSD 961 which was lethal via cardiac toxicity, killed the animals by respiratory depression. Morbidity effects on blood pressure (BP) and heart rate (HR) varied. RSD 939, 988, and 961 were about equipotent at reducing BP and HR. RSD 939 and RSD 961 appeared to block sodium channels at slightly lower doses than those required to affect blood pressure or heart rate but required higher doses to elicit changes suggestive of block of potassium channels. RSD 988 affected BP and HR at lower doses than those required to produce ECG changes suggestive of ion channel blockade. RSD 952, 971, and 969 were more potent in reducing BP than heart rate. RSD 981 reduced heart rate more than blood pressure. 5+ Table 3. Data from preliminary in vivo screening. Except for MTD (the Maximum Tolerated Dose in ^mole/kg) all values are D25% +S.E.M. (Dose required to produce a 25% change from control) in |xmole/kg , NE designating a non-estimable value. R or C following the MTD represents the mode of death, either respiratory or cardiac. Compound BP HR PR ORS RSh QTl QT2 MTD 939 7.4±2.1 5.8±0.8 9.9±3.0 21±4 2.3±1.1 1312 1714 1610 R 988 4.110.8 5.8±0.5 8.9±0.3 NE 10±2 2915 2715 2913 R 952 15±6 23±3 19±5 NE 16±5 4319 3418 42110 R 971 5.0±0.3 24±5 44±6 NE 34±14 NE NE 3816 R 961 14±4 17±6 13±2 NE 1.7±0.6 1715 2312 2913 C 981 33±4 18±5 45±15 NE 9.714.3 1314 8.112.4 102116R 969 25+1 44±7 NE NE 1616 NE NE 115+13 R Figure 9 shows a plot of the D25% values for the toxic effects of RSD939, RSD988, and RSD952. These compounds have changes at the XI position of the molecule. RSD988 was not significantly different from RSD939. Blood pressure effects for RSD952 were not significantly different from RSD939. However, RSD952 was significantly less potent than RSD939 in reducing heart rate. RSD952 was also significantly less potent than RSD939 in reducing both blood pressure and heart rate. • B P M H R H MTD ioon 939 988 952 Compound Figure 9. Data for toxic effects of RSD939, RSD988, and RSD952. An * denotes a significant difference from RSD939 (p<0.05). An # denotes a significant difference from RSD988 (p<0.05). The D25%s for reduction of blood pressure were 7.4±2.1, 4.1±0.8, and 15±6 |imole/kg respectively. RSD988 and RSD 952 were not significantly different from RSD939, whereas RSD952 was significantly different than RSD988 for reduction of blood pressure. For heart rate the D25%s were 5.8±0.8, 5.8±0.5, and 23±0.3 i^mole/kg respectively. The MTDs were 16±0 iimole/kg for RSD939, 29+3 ixmole/kg for RSD988, and 42±10 ^mole/kg for RSD952. RSD 952 and RSD 988 were not statistically different, however both were significantly different than RSD939. Figure 10 shows the effects of RSD939, RSD988, and RSD952 on the ECG. Both RSD988 and RSD952 were significantly less potent than RSD939 for increasing the RSh of the ECG. RSD988 and RSD952 were also significantly less potent than RSD939 for prolongation of the QT1 interval. There was no significant difference between RSD988 and RSD952 for effects on the ECG. •r' E] PR Compound Figure 10. Data for ECG effects of RSD939, RSD988, and RSD952. An * denotes a significant difference from RSD939 (p<0.05). The D25%s for prolongation of the PR interval were 9.9+3.0, 8.9±0.03, and 19+5 |Limole/kg respectively. A 25% increase of the RSh occurred at doses of 2.3+1.1, 10+.2, and 16±5 fimole/kg respectively. The D25%s for prolongation of the QTl interval were 13±2 |imole/kg , 29+5 ixmole/kg, and 43±9 famole/kg for RSD939, RSD988, and RSD952. Similarly, the D25%s for prolongation of the QT2 interval were 17±4 |amole/kg for RSD939, 27±5 (xmole/kg for RSD988, and 34±8 for RSD952. Figure 11 shows the D25%s for toxic effects of the compounds with changes around the nitrogen at the X2 position. RSD 971 was significantly less potent than RSD 939 in reducing of heart rate but was not significantly different for reduction of blood pressure. RSD 961 was not significantly different than RSD 939 for effects on blood pressure or heart rate. Both RSD 981 and RSD969 were significantly less potent than both RSD 939 and RSD961 for reduction of blood pressure. RSD 969 was also significandy less potent than RSD939, RSD961 and RSD981 for reduction of heart rate. In terms of Maximum Tolerated Dose RSD971, RSD961, RSD981, and RSD969 exhibited a statistically significant decrease in toxicity relative to RSD939. In addition RSD981 and RSD969, while not significandy different from each other, were significandy less toxic than RSD971 and RSD961. LO 1000-, Figure 11. Data for toxic effects of RSD939, RSD971, RSD961, RSD981, and RSD969. An * denotes a significant difference from RSD939 (p<0.05). An % denotes a significant difference from RSD971. An @ denotes a significant difference from RSD961 (p<0.05), and an $ denotes a significant difference from RSD981. The D25%s for reduction of blood pressure were 7.4±2.1, 5.0±0.3, 14±4, 33±4, and 25±1 Limole/kg respectively. For heart rate the D25%s were 5.8±0.8, 24±5, 17±6, 18±5, and 44±7 Limole/kg respectively. The MTDs were 16±0 Limole/kg for RSD939, 38±6 Limole/kg for RSD971, 29±3 Limole/kg of RSD961,102±16 Limole/kg for RSD981, and 115+13 Limole/kg for RSD969. The D25%s for ECG effects of compounds with changes around the nitrogen at the X2 position are shown in Figure 12. RSD 971 was significantly less potent than RSD939 for prolongation of the PR interval and increasing RSh. RSD 971 had no effect on the QT intervals. RSD 961 was not significantly different from RSD 939 for any ECG effects. RSD 981 was significantly less potent than RSD939 for PR interval prolongation and for increasing RSh, but not for prolonging QT intervals. RSD 981 was also significantly less potent than RSD961 in prolonging PR interval and increasing RSh. RSD 981 was also significantly more potent than RSD 961 for prolongation of the QT2 interval but not QTl interval. RSD969 was significantly less potent than both RSD939 and RSD961 for increasing RSh but had no effect on any other ECG parameters. 100 Q PR El RSh M QTl • QT2 I E 969 Figure 12. Data for ECG effects of RSD939, RSD971, RSD961, RSD981, and RSD969. An * denotes a significant difference from RSD939 (p<0.05). An @ denotes a significant difference from RSD961 (p<0.05). The D25%s for prolongation of the PR interval were 9.9+3.0, 44+6, 13±2, 45±15, and NE (xmole/kg respectively. A 25% increase of the RSh occurred at doses of 2.3+1.1, 34+14, 1.7±0.6, 9.7±4.3, and 16±6 ^mole/kg respectively. The D25%s for prolongation of the QTl interval were 13+2 |xmole/kg , NE, 17+5 (xmole/kg, 13±4 (xmole/kg and NE for RSD939, RSD971, RSD961, RSD981, and RSD969 respectively. Similarly, the D25%s for prolongation of the QT2 interval were 17±4 iimole/kg for RSD939, 23±2 iimole/kg for RSD961, and 8.1+2.4 for RSD981. RSD 971 and RSD 969 had no effect on the QT2 interval. k3 3.2 Electrical Stimulation Studies Examples of the data generated by the electrical stimulation screens are shown in Figures 13 and 14, the dose response curves for the VFT of RSD939 and RSD969. Following the curves a comparison of the D25% values for these two compounds (Table 4). It should be noted that these studies were carried out using an infusion regimen for drug administration. Although the infused doses (in Limoles/kg/min) may appear to be lower than the bolus doses in the in vivo study (in Limoles/kg), the total amount of drug administered at a particular infusion rate was higher than an apparently similar bolus dose (i.e. 2 Limoles/kg vs. 2(j,moles/kg/min infused for 5 minutes giving a cumulative dose of 10 Limoles/kg). y = 92.952 * 1 0 0 1 5 8 x r 2 = 0.893 * y = 142.080 * I 0 0 1 0 8 x r 2 = 0.998 o y = 82.809 * 10° 1 2 6 1 r 2 = 0.970 A y = 212.445 * 1 0 0 0 6 7 * r 2 = 0.975 * Control 1 10 Dose (p:mole/kg/min) Figure 13. Dose response curves for the effect of RSD 939 administered by infusion on the ventricular fibrillation threshold (VFT) in five rats. A dose dependent increase of the VFT was evident. Shown in the legend are the equations used to model the line of best fit for each curve. The D25%s for the five rats were 1.8, 1.2, 1.2, 1.2, and 2.2 Limole/kg/min. The mean D25% ± S.E.M. was 1.5 ±0.2 Limole/kg/min. Control 1 Dose (umole/kg/rnin) 1 0 Figure 14. Dose response curves for the effect of infused RSD 961 on the Ventricular Fibrillation Threshold (VFT) in five rats. Administration by infusion produced a dose dependent increase in VFT. Shown in the legend are the equations used to model the line of best fit for each curve. The D25%s for the five rats were 0.54, 0.35, 0.35, 0.54, and 0.33 Limole/kg/min. The mean D25% ± S.E.M. was 0.42 ±0.05 Limole/kg/min. Table 4. A summary of the descriptive statistics for the D25%s for increase in ventricular fibrillation threshold (VFT) generated from the dose response curves in figures 13 and 14. Sample variance was relatively small. The data for RSD939 was slightly more skewed and had a wider range than that for RSD961. Compound RSD939 RSD961 D25%s for VFT (umole/kg/min) 1.8 0.54 1.2 0.35 1.2 0.35 1.2 0.54 2.2 0.33 Mean 1.5 0.42 S.E.M. 0.20 0.048 Median 1.2 0.35 Standard Deviation 0.46 0.11 Effects on QRS were easier to measure in this infusion study because the animals were ventilated to prevent death due to respiratory failure. This allowed higher doses to be administered and thus measurement of the QRS effects was easier. These results along with other data from the electrical stimulation screen are presented in tables 5 and 6. Once again, when looking at ECG effects, RSD 981 appeared to be equally or slightly more potent for affecting QT intervals whereas the other compounds seemed to preferentially influence PR interval and RSh. Effective doses were lower in this screen when compared to the preliminary screen because drug was continuously infused rather than delivered as a bolus. In this electrical stimulation study RSD988 was significantly more potent in reducing heart rate than either RSD939 or RSD952. In addition RSD 988 was significantly more potent than RSD952 in reducing blood pressure. Table 5. ECG data from electrical stimulation studies. Values are D25% ± S.E.M. (dose required to produce a 25% change from control) in Limole/kg/min, NE designated a non-estimable value. MTD is the Maximum Tolerated Dose in Limole/kg/min. All deaths were due to cardiac failure because animals were ventilated. Drug BP HR PR ORS RSh OT1 OT2 MTD 939 6.9±2.4 3 .6±0.5 3 . 4 1 0 . 6 NE 1 . 2 1 0 . 2 6 .7+2.6 5 . 1 1 0 . 6 1 0 1 2 988 3.0+0.7 1.9+0.3 7 .1+1.0 15+7 2 .0+0.6 7 .0+1.0 7 .9+1.4 19+3 952 7.1 ±1.2 4 . 2 1 0 . 9 1 111 NE 1 0 1 4 2 6 1 7 1 1 1 3 2 6 1 4 971 2.2+1.0 6 .0±2.3 1 0 1 3 1211 2 . 5 1 0 . 6 1 6 1 4 1 7 1 3 1 1 1 2 961 19±2 2 . 3 1 0 . 3 2 . 5 1 0 . 9 1 9 1 1 3 1 . 4 1 0 . 4 4 . 4 1 0 . 9 5 . 2 1 0 . 9 7 . 2 1 0 . 8 981 12±4 8 . 1 1 1 . 4 1 4 1 3 4 6 1 1 7 1 9 1 4 2 1 1 6 5 . 7 1 1 . 7 6 4 1 0 969 25±6 1 7 1 0 . 6 4 9 1 9 4 . 1 1 0 . 4 6 . 4 1 1 . 2 6 . 0 1 1 . 0 5 1 1 8 Table 6. Electrical parameters data from electrical stimulation studies. Values are D25% ± S.E.M. in Limole/kg/min, NE designating a non-estimable value. All deaths were due to cardiac failure because animals were ventilated. Drug iT tT V F T E R P M F F 939 1.7+0.2 3.8+1.6 1.5±0.2 2.2±0.3 3.710.7 988 2.6+0.6 1.7±0.4 1.0+0.2 1.9+0.4 3.411.2 952 7.3±1.9 7.8+2.4 2.6±0.2 4.511.3 5.311.2 971 6.4+1.2 5.2+1.7 2.3±0.8 5.910.8 9.7+2.6 961 1.1 ±0.1 1.6±0.4 0.42+0.05 0.96+0.22 1.7+0.1 981 2.3±0.4 2.5±0.3 4.7+1.4 3.110.9 6.311.1 969 12+3 49+7 4.5±1.7 6.711.1 1211 Figure 15 shows the D25%s and MTDs for the effects of RSD939, RSD988, and RSD 952 on blood pressure, heart rate, and lethality. RSD 988 was more potent than RSD939 for reduction of blood pressure (although not significantly so) and significantly more potent for the reduction of heart rate. The MTD for RSD988, however, appeared to be higher than that of RSD939. RSD 952 was not significantly different than RSD939 but was less significantly potent than RSD988 for decreasing both blood pressure and heart rate. Figure 16 shows the D25%s for the effects of RSD939, RSD988, and RSD952 on measures indicative of sodium channel blockade (PR interval prolongation, RSh increase, increase in iT, prolongation of tT, and increased VFT). RSD 988 was significantly less potent than RSD 939 for PR interval prolongation but not significantly different for effects on other ECG measures which might be produced by sodium channel blockade. RSD 952 was significantly less potent than RSD939 for all measures suggestive of sodium channel blockade except tT. RSD 952 was also significantly less potent than RSD988 for all measures suggestive of sodium channel blockade except PR interval prolongation. Figure 17 shows the effects of these three compounds on measures which correlate with potassium channel blocking activity (prolongation of the QT intervals and ERP and a decreased MFF), RSD 988 showed no difference in potency for these measures versus RSD939. RSD 952 was significantly less 7 1 potent than RSD939 for prolongation of the QT intervals and significantly less potent than RSD988 for prolongation of the QT1 interval. 939 988 952 Compound Figure 15. Data for toxic effects of infused RSD939, RSD988, and RSD952. An * denotes a significant difference from RSD939 (p<0.05). An # denotes a significant difference from RSD988 (p<0.05). The D25%s for reduction of blood pressure were 6.9±2.4, 3.0+0.7, and 7.1+1.2 Limole/kg/min respectively. For heart rate the D25%s were 3.6±0.5, 1.9±0.3, and 4.2+0.9 Limole/kg/min respectively. The MTDs were 10±2 Limole/kg/min for RSD939, 19±3 Limole/kg/min for RSD988, and 26+4 Limole/kg/min for RSD952. RSD952 and RSD988 were both significantly different from RSD939 but did not differ from each other. Compound Figure 16. Data for ECG and electrical stimulation measurements which indicate possible sodium channel blocking effects of RSD939, RSD988, and RSD952. An * denotes a significant difference from RSD939 (p<0.05). An # denotes a significant difference from RSD988 (p<0.05). The D25%s for prolongation of the PR interval were 3.4+0.6, 7.1+1.0, and 11+1 Limole/kg/min respectively. A 25% increase of the RSh occurred at doses of 1.2+0.2, 2.0+0.6, and 10+4 u.mole/kg/min respectively. The D25%s for affecting iT were 1.7+0.2, 2.6+0.6, and 7.3+1.9 for RSD939, RSD988, and RSD952 respectively. Similarly tT was affected at doses of 3.8+1.6, 1.7+0.4, and 7.8+2.4 Limole/kg/min and VFT was increased at 1.5+0.2, 1.0+0.2,and 2.6+0.2 Limole/kg/min. 1 0 0 -I c S 3L S1 988 Compound Figure 17. Data for ECG and electrical stimulation measurements which indicate possible potassium channel blocking effects of RSD939, RSD988, and RSD952. An * denotes a significant difference from RSD939 (p<0.05). An # denotes a significant difference from RSD988 (p<0.05). The D25%s for prolongation of the QTl interval were 6.7±2.6 (xmole/kg , 7.0+.1.0 (imole/kg, and 26±7 |xmole/kg/min for RSD939, RSD988, and RSD952. Similarly, the D25%s for prolongation of the QT2 interval were 5.1±0.6 (imole/kg/min for RSD939, 7.9±1.4 (imole/kg/min for RSD988, and 11±3 for RSD952. ERP was increased at doses of 2.2±0.3, 1.9±0.4, and 4.5±1.3 |xmole/kg/min. A 25% decrease in MFF was seen at infusion rates of 3.7±0.7, 3.4±1.2, and 5.3±1.2 (xmole/kg/min. Figure 18 illustrates how changes around the Nitrogen at the X2 site affects the D25%s for adverse effects on blood pressure, heart rate, and lethality. RSD971 was significantly more potent for reduction of blood pressure than was RSD939. There was no difference between RSD971 and RSD939 for effects on heart rate and MTD. RSD 961 was significantly less potent than RSD939 for reducing blood pressure and significantly more potent for reduction of heart rate. MTD was similar for these two compounds. RSD 981 was significantly less potent than both RSD939 and RSD961 for reduction of heart rate. The MTD of RSD981 was also higher than that for RSD961 and RSD939. RSD 969 was also significantly less potent than RSD939 for reduction of blood pressure and heart rate and had a lower MTD as well. RSD 969 was also significantly less potent than both RSD961 and RSD 981 for reduction of the heart rate. Figure 19 shows the effects of infused RSD939, RSD971, RSD961, RSD981, and RSD969 on ECG and electrical stimulation measurements which indicate possible sodium channel blockade. RSD971 was significantly less potent than RSD939 for PR interval prolongation and increase in iT. RSD 961 was significantly more potent than RSD939 for increasing both iT and VFT. RSD 981 was significantly less potent than both RSD939 and RSD 961 for prolonging the PR interval and increasing the RSh and iT. RSD981 was also significantly less potent than RSD961 for increases to VFT. RSD969 was significantly less potent than RSD939 for all measures which might correlate with sodium channel blockade except VFT. RSD969 was also significantly less potent for all measures which would indicate sodium channel blockade versus RSD961. RSD969 was also less potent than RSD981 for PR interval prolongation, and increasing iT and tT. However, RSD969 was less potent than RSD981 for increasing RSh. 77 939 971 961 981 969 Compound Figure 18. Data for toxic effects of infused RSD939, RSD971, RSD961, RSD981, and RSD969. An * denotes a significant difference from RSD939 (p<0.05). An % denotes a significant difference from RSD971 (p<0.05). An @ denotes a significant difference from RSD961 (p<0.05). An $ shows a significant difference fromRSD981. 18 • PR 939 971 961 981 969 Compound Figure 19. Data for ECG and electrical stimulation measurements which indicate possible sodium channel blocking effects of infused RSD939, RSD971, RSD961, RSD981, and RSD969. An * denotes a significant difference from RSD939 (p<0.05). An @ denotes a significant difference from RSD961 (p<0.05). An $ shows a significant difference from RSD981. The D25%s for prolongation of the PR interval were 3.4+0.6, 10+3, 2.5±0.9, 13±3, and 49±9 Limole/kg/min respectively. A 25% increase of the RSh occurred at doses of 1.2+0.2, 2.5+0.6, 1.4+0.4, 19±4, and 4.1±0.4 Limole/kg/min respectively. The D25%s for affecting iT were 1.7±0.2, 6.4+1.2, 1.1+0.1, 2.3+0.4, and 12+3 for RSD939, RSD971, RSD961, RSD988, and RSD969 respectively. Similarly tT was affected at doses of 3.8+1.6, 5.2+1.7, 1.6±0.4, 2.5+0.3, and 49±7 (imole/kg/min and VFT was increased at 1.5±0.2, 2.3+0.8, 0.42±0.04,4.7±1.4,and 4.5±1.7 Limole/kg/min. 1<\\ 100 -, Figure 20. Data for ECG and electrical stimulation measurements which indicate possible potassium channel blocking effects of infused RSD939, RSD971, RSD961, RSD981, and RSD969. An * denotes a significant difference from RSD939 (p<0.05). An @ denotes a significant difference from RSD961 (p<0.05). An $ shows a significant difference from RSD981. The D25%s for prolongation of the QTl interval were 6.7±2.6 jimole/kg/min , 16±4(imole/kg/min, 4.4+0.9 |imole/kg/min, 21+6 (imole/kg/min, and 6.41.2 |xmole/kg for RSD939, RSD971, RSD961, RSD981, and RSD969. Similarly, the D25%s for prolongation of the QT2 interval were 5.1+0.6 |imole/kg/min for RSD939, 17±3 (imole/kg/min for RSD971, 5.2+0.9 ^ mole/kg/min for RSD961, 5.7±1.7 ixmole/kg/min for RSD981, and 6.1±1.0 for RSD969. ERP was increased at doses of 2.2±0.3, 5.9±0.8, 0.96±0.22, 3.1±0.9, and 6.7±1.1 i^mole/kg/min. A 25% decrease in MFF was seen at infusion rates of 3.7+0.7, 9.7±2.6, 1.710.1, 6.311.1, and 1211 ixmole/kg/min. 3.3 Mouse Toxicity Experiment Results The MTDs for I.V. administration in mice are shown in table 7. RSD 971 was the most lethal compound; RSD 969 and RSD 952 were the least lethal. All the compounds examined appeared to kill by respiratory depression save RSD 961 which was cardiotoxic. 81 Table 7. Data from mouse toxicity studies. Values are Maximum Tolerated Dose in (xmole/kg. Compound MTD Confidence Interval (95%) 939 58 R 52-66 988 64 R 62 - 65 952 104 R 86 - 128 971 12 R 11.8 - 13.0 961 57 C 49 - 66 981 93 R 6 0 - 1 5 4 969 120 R 57 - 252 3.4 Data from Patch Clamp Studies The concentrations for a 50% reduction in sodium or potassium current are shown in table 8. RSD 939, 952, and 981 are more potent sodium channel blockers than potassium channel blockers. RSD 971 was not found to have any effects over the dose range tested. Potassium channel data were not available for RSD 988 and 961. RSD 969 has a greater effect on potassium channels than on sodium channels. In some cases compounds were applied to the inside of the membrane as well as the outside in order to test the ability of the compound to block from the inside aspect of the sodium channel. RSD939 was more than 10 times more potent for block of sodium channels when applied to the outside of the membrane than when applied to the inside surface of the membrane. RSD981 was at least twice as potent when applied to the outside of the channel as when applied to the inside. In some cases the data indicates a value greater than a particular concentration. This is because there was no effect up to that concentration and higher concentrations were unable to be tested because there was an insufficient quantity of compound available or the compounds were insoluble at higher concentrations. 83 Table 8. Patch clamp study data. Values are EC50 (the concentration required to produce a 50% change from control) in (xMolar. Compound Na+ Inside Na+ Outside K + Outside 939 >50 5 >30 988 NA 30 NA 952 NA 150 >500 971 >500 >500 >500 961 NA 7 NA 981 >100 50 100 969 >150 >150 150 3.5 Summary of Data Table 9. Summary of data for measurements which would indicate sodium channel blockade. Values are D25%±S.E.M. in Limole/kg for the bolus dosing study and Limole/kg/min for the electrical stimulation study. Compound In Vivo Screen Electrical Stimulation Screen PR RSh PR QRS RSh IT tT V F T 939 9.9±3.0 2.3±1.1 3.410.6 NE 1.210.2 1.710.2 3.811.6 1.510.2 988 8.9±0.3 10±2 7.111.0 1517 2.010.6 2.610.6 1.710.4 1.010.2 952 19±5 16±5 11+1 NE 1014 7.311.9 7.812.4 2.610.2 971 44±6 34±14 1013 1211 2.510.6 6.411.2 5.211.7 2.310.8 961 13+2 1.7±0.6 2.510.9 19113 1.410.4 1.110.1 1.610.4 .421.05 981 45±15 9.7±4.3 1413 46117 1914 2.310.4 2.510.3 4.711.4 969 NE 16±6 4919 NE 4.110.4 1213 4917 4.511.7 Table 10. Summary of data for measurements which indicate potassium channel blockade. Values are D25%±S.E.M. in Limole/kg for the bolus dosing study and Limole/kg/min for the electrical stimulation study. Compound In Vivo Screen Electrical Stimulation Screen QTl QT2 QTl QT2 ERP MFF 939 13±2 17±4 6.7+2.6 5.1+0.6 2.2+0.3 3.7+0.7 988 29±5 27±5 7.0+1.0 7.9+1.4 1.910.4 3.4+1.2 952 43±9 34+8 26+7 11+3 4.5+1.3 5.3+1.2 971 NE NE 16+4 17+3 5.9+0.8 9.7+2.6 961 17+5 23±2 4.4+0.9 5.2+0.9 0.96+0.22 1.7+0.1 981 13+4 8.1+2.4 21+6 5.7+1.7 3.1+0.9 6.3+1.1 969 NE NE 6.4+1.2 6.0+1.0 6.7+1.1 12+1 Table 11. Summary of data for measurements of Toxicity. For blood pressure and heart rate values are D25%±S.E.M. in |i,mole/kg for the bolus dosing study and fimole/kg/min for the electrical stimulation screen. MTD is the maximum tolerated dose in (amole/kg for the bolus dosing study and mouse data or (imole/kg/min for the electrical stimulation study. Compound In Vivo Screen Electrical Stimulation Screen Mouse BP HR MTD BP HR MTD MTD 939 7.4±2.1 5.810.8 16+0 R 6.9+2.4 3.6+0.5 1012 58 R 988 4; 1±0.8 5.8+0.5 29+3 R 3.0+0.7 1.9+0.3 1913 64 R 952 15±6 23±3 42+10R 7.1+1.2 4.2+0.9 2614 104 R 971 5.010.3 24+5 38+6 R 2.2+1.0 6.0+2.3 11+2 12R 961 1414 17+6 2913C 19+2 2.3+0.3 7.2+0.8 57 C 981 33+4 18+5 102+16R 12+4 8.111.4 64+0 93 R 969 25+1 44+7 115+13R 25+6 17+0.6 51+8 120 R Tables 12, 13 and 14 show the relative potency of compounds with respect to RSD939. With RSD939 assigned a potency of 1, the relative potency of the other compounds for each measure was calculated using the following formula: 1 (D25% RSD9XX/D25% RSD939) Thus with RSD939 having a potency of 1, a value greater than one indicates more potency than RSD939 and a value of less than 1 indicates a less potent compound than RSD939. 80 Table 12. Relative potency of compounds for measurements which indicate sodium channel blockade. The order of potency is split into two groups, one for changes at the XI position and one for changes at the X2 position. The Mean ± S.E.M. relative potency is also indicated. Compound In Vivo Screen Electrical Stimulation Screen PR RSh PR RSh iT tT VFT Mean 939 1 1 1 1 1 1 1 1 988 1.11 0.25 0.48 .60 0.65 2.24 1.50 .98±.24 952 0.52 0.14 0.31 0.12 0.23 0.49 0.58 .34±.07 939 1 1 1 1 1 1 1 1 971 0.22 0.07 0.34 0.48 0.26 0.73 0.65 .39+.08 961 0.76 1.35 1.36 0.86 1.54 2.38 3.57 1.69±.34 981 .22 .24 .24 .06 .74 1.52 .32 .48±.18 969 - .14 .07 .29 .14 .08 .33 .18+04 Table 13. Relative potency of compounds for measurements which indicate potassium channel blockade. The order of potency is split into two groups, one for changes at the XI position and one for changes at the X2 position. The Mean ± S.E.M. relative potency is also indicated. Compound In Vivo Screen Electrical Stimulation Screen Q T l QT2 Q T l QT2 ERP M F F Mean 939 1 1 1 1 1 1 1 988 .45 .63 .96 .64 1.16 1.09 .821.11 952 .30 .50 .26 .46 .49 .70 .451.06 939 1 1 1 1 1 1 1 971 - - .42 .3 .37 .38 .37±.02 961 .76 .74 1.52 .98 2.29 2.18 1.41+26 981 1.00 2.10 .32 .89 .71 .59 .94±.23 969 - - 1.05 .85 .33 .31 .641.16 Table 14. Relative potency of compounds for measurements which would indicate toxicity. The order of potency is split into two groups, one for changes at the XI position and one for changes at the X2 position. Compound In Vivo Screen Electrical Stimulation Screen Mouse BP HR MTD BP HR MTD MTD 939 1 1 1 1 1 1 1 988 1.8 1 .55 2.3 1.89 .52 .9 952 .49 .25 .38 .97 .86 .38 .56 939 1 1 1 1 1 1 1 971 1.5 .24 .42 3.1 .6 .91 4.8 961 .53 .34 .55 .36 1.6 1.4 1.0 981 .22 .32 .16 .58 .44 .16 .62 969 .3 .13 .14 .28 .21 .20 .48 4. Discussion Chemical changes to the structure of RSD939, the parent compound of this study, had definite effects on the pharmacology of the series. However, conclusions can only be made with a low probability of accuracy due to the small number of compounds used to explore the pharmacophore and the small difference in activity between the compounds in this series. A more extensive study of a larger number of structurally similar molecules with greater differences in activity would be necessary for more confidence in the conclusions. Nevertheless it is possible to make some conclusions which give an indication of the SAR of these molecules. 4.1 Changes to the Nitrogen at the XI Position Changes to the hetero atom at the XI position have noticeable effects on the pharmacological as well as the physiochemical profiles of these compounds. Substitution of the methyl amide nitrogen of RSD939 with a protonated amide nitrogen, as in RSD 988, has two major effects. The hydrogen is less electronegative than carbon and would donate more charge towards the nitrogen than carbon. This is somewhat offset by the derealization of charge within the amide linkage. Nonetheless, the secondary amide nitrogen will be slightly more negatively charged than a tertiary amide nitrogen. In addition, the secondary amide is less sterically hindered than the methyl substituted tertiary amide. The ester linkage seen in RSD952 further enhances these changes. Oxygen is more electronegative than nitrogen, therefore this region of the molecule will be more negatively charged than the corresponding amide and will reduce the basicity of the nitrogen at X2. There is also some derealization of charge within the ester group. This region of the molecule is also less sterically hindered than that in RSD 939 with the tertiary amide. 4.1.1 The XI Site and activity indicative of Na+ Channel Blockade These RSD compounds share some structural similarities with known class I antiarrhythmics and local anaesthetics. The XI nitrogen may be seen as equivalent to the amide nitrogen group of the local anaesthetic drugs (procainamide, lidocaine). Classical procaine-like local anaesthetics have a two carbon linkage terminating in an amido or oxycarbonyl connection to an aromatic ring as in the compounds of this study. Incorporated in the two carbon linkage is an alicyclic ring which may be functionally similar to the four carbon \"backside\" of the piperidine ring of flecainide and encainide that is thought to fit into a hydrophobic pocket. However, rather than a bond between the amide nitrogen or carbonyl, and the aromatic ring, the tested compounds have a methyloxy extension of the aryl-X2 nitrogen linkage. Previously published assumptions about the binding of class I drugs suggest that this linking region fits into a hydrophobic pocket in the channel pore. However, the amide or ester carbonyl group would be fairly hydrophilic and would be expected to reduce binding in this hydrophobic component of the class I binding site. It is possible that the carbonyl group is able to orient itself in such a fashion as to project away from the hydrophobic pocket and thus not affect binding. Further studies with compounds that do not incorporate the carbonyl substituent would be necessary to test this hypothesis. Changing the tertiary amide nitrogen of RSD 939 to a secondary amide nitrogen (RSD 988) appeared to have mixed effects on measures of sodium channel blocking activity. The potency for effects on ECG parameters indicative of Na channel block (RSh and PR) was reduced approximately two fold in the electrical stimulation screen (constant infusion of drug). In in vivo screening there was no difference in potency for effects on PR interval but a four fold decrease in potency for effects on RSh. The effect on RSh was not mirrored in the electrical stimulation parameters (iT, tT, VFT) in which RSD988 was more potent than RSD939 for effects on tT and VFT and less potent for effects on iT. The mean of all of the relative potencies for RSD988 was approximately 1, suggesting that overall there was no difference from RSD939 and that the differences in individual parameters might be due to biological variability. If these compounds are binding to the putative class I antiarrhythmic binding site then it would be expected that both the tertiary and secondary amide nitrogens might have similar interactions with a voltage sensitive segment of the binding site. Replacing the amide nitrogen by an ester (RSD 952) increased the D25% (decreased potency) for changes to PR interval and RSh in both screens and increased the D25% for electrical stimulation parameters (approximately a two to three fold decrease). Thus RSD952 appeared to be less potent than RSD939 for effects on these measures of sodium channel blockade. However, the ester linkage is sufficiently different from the amide linkage in terms of its electronic properties that one would expect a greater reduction in potency than the observed two to three fold difference, Thus it is possible that these RSD compounds are acting at a binding site that is different than the class I antiarrhythmic binding site proposed in previous studies (Sheldon et al., 1991). Previous studies, where amide and ether linked drugs were compared, might predict that the secondary amide RSD988 and the ester RSD952, which is similar to the ethers, would be more potent than the tertiary amide RSD 939. However these previous studies did not compare amide and ether linked drugs of equivalent structure but compared compounds with similar lipophilicity but different overall molecular structure. In those studies the ether-linked compounds were more potent than the amide linked compounds of similar lipophilicity. In addition, the more lipid soluble compounds in these earlier studies were more potent (Courtney et al., 1978a,b; Courtney, 1979; 1980a,b,c; Sada and Ban, 1980; 1981a,b,c). It is difficult to make comparisons of these three compounds (RSD939, RSD52 and RSD988) based upon existing models because they are based on comparing ethers and amides of similar lipophilicity rather than similar molecular structures(Courtney et al., 1978a,b; Courtney, 1979; 1980a,b,c; Sada and Ban, 1980; 1981a,b,c). There are also some marked structural differences between the compounds in this series and the current antiarrhythmic drugs that are known to bind to sodium channels. It would be necessary to conduct binding and comparative studies between the RSD compounds and existing drugs to determine their sites of action. It is evident from the data in this study that simple single changes in the molecular structure of these compounds affects both their lipophilicity and their apparent sodium channel blocking activity in a way not predicted by the existing model for the S AR of class I antiarrhythmic action. The existing model for sodium channel blockade also postulates that the site of action of the local anaesthetics is on the intracellular aspect of the ion channel. Those channel blockers must cross the cell membrane to reach their site of action. If the model applies, the more lipid soluble compounds would cross the membrane more easily and thus be more potent. RSD961 is the most lipid soluble and also appears to be the most potent compound of this series. However, the differences in potency for sodium and potassium channel blockade between RSD961 and the other compounds from this series are small and the differences in lipophilicity between the compounds are, with the exception of RSD971, small. The data from this study do not show a correlation between lipophilicity and activity. It is possible that the compounds examined in this study are acting at a novel site on the sodium channel that is less sterically restricted and also more accessible to the aqueous environment since lipid solubility does not appear to correlate to channel block and more steric bulk around the nitrogen does not decrease channel block. The electrophysiological data, although incomplete, indicates that these compounds are more effective at blocking sodium channels from the outside. Therefore the binding site for the RSD compounds may be on the extracellular aspect of the ion channel where lipid solubility would not influence the RSD compounds access to the binding site. It would also appear that hydrogen bonding or some sort of dipole interaction with this part of the molecule is not important since RSD952 and RSD988, the most likely compounds of this series to form hydrogen bonds, are less potent than the relatively non-polar tertiary amide, RSD939. 4.1.2 The XI Site and activity indicative of K+ Channel Blockade The study of class III drug action is special in rats. This is because in most mammals the repolarization of the myocytes is dominated by the Ik (delayed rectifier) current. Rat cardio-myocytes do not contain a significant amount of Ik type potassium channels. Rather, the repolarization of rat cardio-myocytes is dominated by the Ito (transient outward) current. Thus the effects on QT intervals, MFF and ERP seen in this rat study probably indicate block of Ito type channels and may differ from data generated for these compounds in other species where Ik currents dominate. The change from a tertiary to a secondary amide and then to an ester shows a pattern of potassium channel blockade similar to that of sodium channel blockade. Examining the relative potencies of these compounds for effects on the QT intervals of the ECG shows that in both the bolus dosing and constant infusion regimens RSD988 was less potent than RSD939. However, RSD988 was roughly equipotent to RSD939 for effects on the electrical stimulation parameters ERP and MFF. RSD952 was less potent than the other two compounds for effects on all the measures indicative of potassium channel blockade. Previous examinations of the SAR of class III agents have centered around substituents on aromatic rings and also primarily in tissues where the Ik type channel dominates (Phillips et al., 1990). Compounds which contain electron withdrawing substituents on the rings appear to be the most potent potassium channel blockers. However, in the case of the compounds in this study, RSD952 contains the most electron withdrawing type of linkage but was the least potent compound. It is possible that these drugs are acting on a different site of action on the potassium channel than those in the previous SAR studies (Phillips et. al., 1990, Bagwell et al., 1976, Lumma et al., 1987, Buzby, 1987, Colatsky, 1987, Follmer, 1988, 1989) and thus the requirements for binding are quite different. It is also possible that the differences between the Ik channel and the Ito channel are such that the requirements for binding are different. Further electrophysiological studies on the different potassium channel subtypes as well as in other species would help determine whether these compounds act in a more classical fashion on the Ik channel. However, given the similar relative potencies for both sodium and potassium channel blockade for these compounds it would appear that the site of action on the I t 0 channel may be structurally similar to that on the sodium channel. 4.1.3 The XI Site and Effects on Heart Rate and Blood Pressure RSD988 was approximately equipotent to RSD939 for effects on heart rate when given as a bolus dose. However when administered by constant infusion RSD988 was nearly twice as potent as RSD939. It is possible that the lower lipid solubility of RSD988 leads to a greater amount of drug accumulating in the aqueous environments such as plasma and extracellular fluid during infusion than with RSD939. RSD939 will be rapidly taken up into fat and may cross the cell membrane more easily and thus have a lower effective plasma concentration than RSD988. However, if this were true then the pattern of potencies for effects on the ECG should also be different between bolus and infusion. The pattern is not similar for effects on ECG and electrical stimulation parameters thus the mechanism of bradycardia may be more complex. Further studies examining the pharmacokinetics of these compounds would need to be performed to examine this hypothesis. It is also possible that RSD988 has a greater effect on ion channels in the SA node than in other cardiac tissue thus the higher relative potency for decreasing heart rate than for blocking conduction in the ventricle. A similar pattern was seen for RSD952 which was 25% as potent as RSD939 in reducing heart rate when administered by bolus dosing but was only slightly less potent than RSD939 when infused. The lipid solubility of RSD952 is similar to that of RSD939 and the other compounds in this series (except for RSD971) thus the hypothesis that an accumulation in plasma versus other depots when infused does not hold. This further indicates a direct bradycardic action, probably in the SA node. The hypotensive nature of these compounds adds further support for this hypothesis. It would be expected that the falling blood pressure upon administration of these compounds would lead to a reflex tachycardia. The RSD compounds appear to overwhelm this reflex and cause a decrease in heart rate. Because RSD952 is less potent than RSD939 for effects on heart rate and RSD988 is more potent than RSD939, and, if it is assumed that these compounds are having a direct effect on the SA node, then it is possible to make some predictions about the binding site. The site of action on these channels appears to be on the extracellular surface because lipid solubility does not affect potency. Furthermore, this site appears to be somewhat sterically hindered because RSD939 is less potent than RSD988. The pattern was somewhat different when examining effects on blood pressure. RSD988 was approximately twice as potent as RSD939 for hypotensive actions while RSD952 was less potent than RSD939 when administered by bolus, and equipotent when infused. Since RSD988 is a more potent hypotensive agent than RSD939 but is less potent for effects on sodium and potassium channels in the ventricle it is likely that the hypotension is due to effects on the vasculature rather than on the heart, i.e. reduced cardiac output. It is possible that sodium channels in the peripheral vasculature are more similar to channels in the SA node than to channels in the ventricle and thus RSD988 is a more potent blocker of channels in both the SA node and the peripheral vasculature. It is also possible that RSD988 has affects on calcium channels in the vascular tissue and is thus affecting blood pressure via calcium channel blockade. Further experiments would be required to determine effects on calcium channels as that was not part of this study. RSD952 is less potent in reducing blood pressure than RSD939 but to a lesser degree than that for its actions on ECG and electrical stimulation parameters. If, as postulated for RSD988, the site of action is on the peripheral vasculature then it is possible that the ester linkage affects binding to these ion channels more so than effects on cardiac or neuronal channels. The lower potency in bolus dose studies also lends further evidence for effects on the vasculature itself. If respiratory depression were the cause of the hypotension then the compounds would be more potent in the bolus dosing screen where the animals are breathing spontaneously than in the electrical stimulation screen where the animals are ventilated. These compounds were more potent in the ventilated 401 animals than in the unventilated ones and thus the hypotensive mechanism is likely not respiratory depression. 4.1.4 The XI Site and Lethality In rats the tertiary amide RSD 939 was the most lethal drug with the secondary amide RSD 988 being half as toxic as RSD939 and the ester RSD 952 approximately one third as toxic as RSD 939. In mice RSD988 and RSD939 were equipotent and RSD952 was half as toxic as the other two. All three drugs appeared to kill by respiratory depression. Nevertheless their pattern of toxicity was the same in the electrical stimulation study (drugs administered by infusion), where death was due to cardiac mechanisms and the pattern of relative potencies also mirrored effects on ventricular ion channels. This suggests that sodium channel blockade in peripheral nerves may be responsible for the respiratory depression rather than a CNS mechanism of action. This also argues for a novel site of action on the ion channel that is somewhat similar in both neuronal and cardiac channels. However, it is possible that the similarity of these compounds to known K-agonists (which are also potent Li-agonists) produces a CNS mediated respiratory depression via actions on Li-receptors. This could be tested in further experiments to determine if naloxone, given to prevent any opioid actions, has an affect on the MTD for these compounds. 4.2 Changes at the X2 Position Changes to the group at X2 also had distinct effects on the pharmacology of this series of compounds. Changing from a pyrollidine to a quaternary methylpyrollidinium (RSD 939 vs. RSD 971) places a permanent positive charge on this site. In consequence, RSD971 is significantly less lipid soluble than the other compounds from this study. Changing the pyrollidine ring of RSD 939 to an azepine ring in RSD 961 does not change the electron density of this region of the molecule but does add steric bulk. Steric bulk is also added when the pyrollidine is changed to a morpholine as in RSD 969. In addition the morpholine group adds an electron withdrawing oxygen atom thus changing the pKa of the Nitrogen such that this site will be less likely to be charged than the nitrogen of the pyrollidine or azepine rings. The di-methoxyethylamino substitution in RSD 981 also adds more bulk than the azepine or the morpholine and also reduces pKa of the nitrogen relative to RSD 939 but not as low as the pKa of the morpholino nitrogen. 4.2.1 The X2 Position and activity indicative of Sodium Channel Blockade Permanently charging this site, as in RSD 971, decreased potency for cardiac sodium channel blockade. Potency for effects on all measures of the ECG and electrical stimulation parameters were lower for RSD 971 than for RSD 939. The potencies of RSD971 for effects on parameters indicative of sodium channel blockade were from 30% to 90% lower than for RSD939, and the mean relative potency for sodium channel blockade was 60% lower than RSD939. It is possible that the charge on RSD971 makes it act more like guanidium toxins such as tetrodotoxin and saxitoxin that are also permanently charged. Cardiac sodium channels are less sensitive to these toxins than are neuronal channels. If the tested compounds, as postulated above, act on a site on the external aspect of the channels (the permanently charged guanidium toxins are known to act on an external binding site) it may be that RSD971 binds to the same (or a similar) site as that acted on by the toxins and thus has a reduced affinity for cardiac ion channels. Binding studies with the guanidium toxins and the RSD compounds might resolve some of these questions. It is also possible that the permanent positive charge on RSD971 inhibits its ability to cross the membrane. It is believed that the class I antiarrhythmic binding site as well as the predominant site of action for local anaesthetics is on the intracellular side of the ion channel. However, the evidence from the electrophysiological studies indicates that the RSD compounds have a site of action on the extracellular side of the ion channel. Increasing bulk at the X2 site showed indications of greater sodium channel blockade. RSD 961 is more potent than RSD 939 for effects on parameters indicative of sodium channel blockade in both the bolus and electrical stimulation screens, and on both ECG and electrical stimulation parameters. However, the increased bulk in RSD 969 and RSD 981 does not seem to increase potency for sodium channel blockade. These latter two compounds are both less potent sodium channel blockers than RSD 939 with RSD 981 being slighdy more potent than RSD 969. However, it is difficult to separate the effects of increased bulk from those of electron density since RSD981 and RSD969 differ from RSD939 in bulk, charge density and pKa. Further studies with X2 substituents that have different sized rings or other bulky constituents, yet similar pKa and electron density would be necessary to understand the role of bulk around the X2 amino nitrogen. The pKa of the RSD compounds appears to be a major determinant of their ion channel blocking actions. The order of potency for the compounds from most potent to least potent is 961>939>981>971>969. While the order of these compounds from highest pKa to lowest pKa is 961=939>981>969>971. These two sets show essentially the same rank order RSD971 is permanently charged and cannot generate a neutral species, hence the activity of this molecule may be more like that of guanidium toxins than of the other compounds in this study and RSD971 may even act at a different binding site. However, the order of potency for the other compounds is virtually identical to the order of pKa indicating that a higher pKa appears to increase potency for sodium channel blockade. Assuming that the initial long range interactions between the drug and its binding site is based on electrostatic attractions, then any molecule being attracted to this binding site would likely be in an ionized (positively charged) form. It is believed that an ionizable component is necessary for the binding of many local anaesthetics and antiarrhythmic compounds to ion channels. This assumption fits with the observed data from this study that the compounds with a higher pKa, and thus a higher proportion of the molecules in a charged form at physiological pH, were more potent. It is thus evident that this region of the molecule has a complex relationship with the binding site whereby electron withdrawing substituents may decrease potency, and potency is greater in compounds with a higher pKa. The presence of any steric restriction at the active site for these compounds is not clear from this study. 4.2.2 The X2 Position and activity indicative of Potassium Channel Blockade Permanently charging this structural component in RSD971 had a similar effect to that seen for sodium channel block on potassium channels. Potency for apparent potassium channel blockade was also reduced to the same extent as that seen for sodium channel blockade. This is may be due to some functional similarities between potassium and sodium channel binding sites for these compounds. Previous studies have shown that electron withdrawing substituents on drugs tend to favour potassium channel block. Whether this is due to changes in lipophilicity or charge density is not known but it is obvious that with this series of compounds regions of strong positive charge do not favour channel blockade. According to existing SAR RSD 969 and RSD981, which have an electron withdrawing oxygen atoms close enough to the X2 amino nitrogen to affect its basicity, might be expected to be more potent potassium channel blockers than RSD 939. RSD 969 and RSD981 are actually less potent potassium channel blockers than RSD 939. RSD969 is also less potent than RSD981 and this may be due to the nature of the electron withdrawing group being part of a ring resulting in derealization of charge within the ring structure via hybridization of pi orbitals. However, the previous studies indicated that potassium channel blockade was increased by electron withdrawing substituents on an aryl ring. The electron withdrawing components of the RSD compounds in the X2 region of the molecule are not on the molecules aryl ring, thus it is difficult to make conclusions about the RSD compounds based on the SAR of other class III antiarrhythmics. The data for this series of RSD compounds does not fit with the current model of potassium channel blockade by class III drugs. Perhaps the low pKa of this series of compounds plays a more important role in ion channel blockade than does the presence of electron withdrawing groups. These data also support the hypothesis that the RSD compounds act at a site on the ion channel that is different than that for the existing class III agents. However, since the potency of RSD969 and RSD981 on measurements indicative of potassium channel blockade was not decreased as much as potency for measurements suggesting sodium channel blockade, it is possible that the electron withdrawing nature of these substituents does contribute to increased potassium channel activity. Just as for sodium channel blockade the two most potent compounds were RSD939 and RSD961. These compounds do not have any electron withdrawing groups but they do have the highest pKa of all the tested compounds. Thus it seems that pKa is also a major determinant of potassium channel blockade for this family of compounds. It is evident that the interaction of this region of the molecule with the potassium channel is a fairly complex process and this small series of drugs does not allow us to sufficiendy separate the different effects of steric bulk, charge distribution, pKa and lipophilicity. 4.2.3 The X2 Position and Effects on Blood Pressure and Heart Rate RSD 971, the permanendy charged molecule in this test series was slighdy more potent in reducing blood pressure than RSD 939 but less potent for reducing heart rate. This may be due to increased potency on sodium channels in vascular smooth muscle, or for neuronal sodium channels relative to potency for cardiac channels. However, if the test compounds are acting like guanidium toxins it would be expected that they would reduce blood pressure more in unventilated animals (bolus study) and less so in the ventilated animals used for the electrical stimulation study. In fact, RSD971 was more potent for effects on blood pressure in the ventilated animals than the unventilated ones. This indicates that the hypotension is likely due to direct actions on the peripheral nerves or vasculature. Hypotension due to decreased cardiac output is also not likely since RSD971 exhibited low potency for blockade of cardiac ion channels. RSD 961 was less potent than RSD 939 at reducing blood pressure and heart rate in the in vivo screen. In the electrical stimulation screen RSD961 was less potent than RSD939 for effects on blood pressure but more potent for effects on heart rate. The average relative potency for heart rate effects in the bolus and electrical stimulation studies is close to one, indicating that RSD961 may not differ from RSD939 in its effects on heart rate. It therefore appears that steric bulk around the X2 amino function does not play as important a role in effects on heart rate as on blood pressure. Both RSD 969 and 981 effect blood pressure and heart rate to a lesser degree than RSD 939 and RSD961. This may be due to their lower pKa. Lipophilicity does not seem to play a role as RSD 981 is more lipophilic than RSD 939 and RSD 969 is less lipid soluble than RSD 939, yet both RSD981 and RSD969 are less toxic than RSD939. 4.2.4 The X2 Position and Lethality Permanent charge at this location produces a more toxic agent which appears to kill by virtue of respiratory depression. RSD971 was significantly more lethal than the other compounds in mice. The pattern in both ventilated and unventilated rats was different, since RSD971 was less lethal than several other compounds in rats. It is possible that RSD971 has a different mechanism of action than the other compounds examined in this study. Once again, steric factors seem to be less important than electron density. RSD 961 has only increased steric bulk at this site and is roughly equally toxic to RSD 939. It is interesting to note that RSD 961 appears to kill via a cardiac rather than a respiratory mechanism. It may be that the site on the cardiac sodium channel is slighdy less sterically restricted than neuronal sodium channels. pKa also seems to be important as RSD 969 and 981 are less toxic and have a lower pKa than RSD939 and RSD961. This indicates that death may occur via an ion channel mechanism since the order of potency for lethality is similar to that for ion channel blockade. Possibly these compounds are acting in a fashion similar to the guanidium toxins and thus the compounds with a lower pKa and therefore a lower proportion of charged species are less toxic. 4.3 Conclusions In conclusion, it is evident that even small changes to the regions around the hetero atoms at positions XI and X2 of the test compounds influence the activity of these novel putative antiarrhythmic agents. It is also evident that a separation of the pharmacophores for effects which suggest blockade of Na+ and K + channels as well as those for morbidity and lethality effects should be possible. Examination of a larger number of compounds with a greater range of activities would lead to a better understanding of the structural requirements of effective class I or class III antiarrhythmic agents. In summary, changes to the hetero atom at the XI position have noticeable effects on the pharmacological profile of the test compounds. Changing the tertiary amide nitrogen (RSD 939) to a secondary amide nitrogen (RSD988) seemed to decrease both sodium channel and potassium channel blocking activity. The effects on toxicity were somewhat more complex, thus MTD was increased twofold in rats but less so in mice. That is, RSD988 was two time less toxic, however it appeared to be more potent on blood pressure and heart rate. Decreasing HR may be a desirable side effect of arrhythmia treatment but decreased blood pressure is an undesirable effect. Further changing the character of this region of the test compounds by substituting an ester (RSD952) for the tertiary amide of RSD939 reduced potency for ion channel blockade more than did the change to a secondary amide. The test ester was less toxic than RSD 939, but was more toxic than RSD 988. RSD 952 had, however, less effect on HR and BP and a higher MTD than either RSD 939 or RSD 988. Thus it would seem that the pharmacophore for lethality and that for morbidity (effects on HR and BP) are somewhat similar to that for ECG and electrical changes. In summary it appears that the part of the active site interacting with the XI region of the test molecules is not sterically hindered nor does it favour ligands capable of hydrogen bonding. Lipophilicity did correlate with the potency of these molecules indicating that close-range bonding interactions are important to interactions with the site of action of these test compounds. Changes to the substitution pattern on the amino nitrogen at X2 also had distinct effects. Changing from a pyrollidine to a quaternary methylpyrollidinium group (RSD 939 vs. RSD 971) decreases ion channel blocking activity while increasing toxicity (lethality and morbidity). Thus a permanent positive charge at the X2 location seems to favour unwanted side effects producing a compound which behaved more like guanidium toxins. Increasing steric bulk at the X2 location as in RSD 961 appeared to make the molecule a more potent sodium channel blocker with lesser effects on apparent potassium channel blockade, or toxicity. RSD 969 and RSD 981 also possess increased steric bulk at this site, however this increase is complicated by changes in the pKa and electron distribution. Both of these compounds have pKa's which are 2-3 pH units lower than the 9.3 value for RSD939 as a consequence of oxygen atoms P to their amino nitrogens. RSD969 and RSD981 show effects which suggest that they are less potent sodium channel and potassium channel blocking agents than RSD939. Thus, the region of the active site interacting with this part of the test compounds appears to favour some degree of positive charge at the point of connection of the X2 component to the test molecules. This is evident from the greater potency of compounds with higher pKa. However, a permanent charge at this location is less favoured and produces a compound that appears less potent for ion channel blockade, and more potent for adverse effects. This part of the binding site also appears to be sterically unhindered such that the bulkiest ligands tested do not seem to show different activity than less bulky ones. However, increased polarity adjacent to the amino group seems to generally decrease potency while favouring effects which would indicate potassium channel blockade over sodium channel blockade. The effects of polarity were less clear because they were inevitably confused with changes in the basicity of the X2 nitrogen. B5 5. References Aldrich, R.W.: Ion channels: mutating and gating. Nature. 1989; 339: 578-579 Bagwell, E.E., Walle, T., Drayer, D.E., reidenberg, M.M., Pruett, J.K.: Correlation of the electrophysiological and antiarrhythmic properties of N-acetyl metabolite of procainamide with plasma and tissue drug concentrations in the dog.. / . Pharmacol. Exp. Ther., 1976; 197: 38-48 Bean, B.: Calcium channels: gating for the physiologist.. Nature, 1990; 348: 192-193 Beatch, G.N., MacLeod, B.A., Abraham, S., Walker, M.J.A.: The in vivo electrophysiological actions of the new potassium channel blockers, tedisamil and UK 68.798. Proceedings of the Western Pharmacology Society, 1990; 33:5-8 Beatch, G.N., Abraham, S., Macleod, B.A., Walker, M.J.A.: Antiarrhythmic properties of tedisamil (KC8857). a putative transient outward K+ channel blocker.. Br. J. Pharmacol., 1991; 182:13-18 Bergey, J.L., Nocella, K., McCalleum, J.D.: Acute coronary artery occlusion-reperfusion-induced arrhythmias in rats, dogs, and pigs: antiarrhythmic evaluation of quinidine. procainamide and lidocaine.. Eur. J. Pharmacol, 1982; 81: 205-216 Bigger, J.T., Dresdale, R.J., Heissenbuttel, R.H., Weld, F.M., Wit, A.L.: Ventricular arrhythmias in ischaemic heart disease: Mechanism, prevalence, significance, and management.. Progr. Cardiovasc. Dis., 1977; 19: 255-300 Boineau, J,P., Cox, J.L.: Slow ventricular activation in acute myocardial infarction. A source of re-entrant premature ventricular contraction.. Circulation, 1973; 48: 702-713 Botting, J.H., Curtis, M.J., Walker, M.J.A.: Arrhythmias associated with myocardial ischaemia and infarction.. Molec. Aspects Med., 1986; 8: 311-422 Brown, H.F.: Electrophysiology of the sinoatrial node. Physiol. Rev., 1982; 62:505 Brugada, P.: New antiarrhythmic drugs: Still looking for the ideal one.. European Heart J., 1987; 8: 91-94 Brugada, P.: The Vaughan-Williams classification of antiarrhytmic drugs. Why don't we find its clinical counterpart?. PACE, 1990; 13: 339-343 i i 4 Budden, R., Buschmann, G., Kiihl, U.G.: The rat ECG in acute pharmacology and toxicology.. In R. Budden, D.K. Detweiler, and G. Zbinden (eds), The Rat Electrocardiogram in Pharmacology and Toxicology., Permagon Press, New York, 1981: pp.41-81 Buzby, G.C. Jr., Colatsky, T.J.: 4-Aminobenzenesulfonamides. U.S. patent No. 4,698,445; 1987 Campbell, T.J.: Resting and rate dependent depression of maximum rate of depolarisation (Vmax) in guinea-pig ventricular action potentials by mexilitine. disopyramide. and encainide.. /. cardiovasc. Pharmacol, 1983a; 5: 291-296 Campbell, T.J.: Kinetics of onset of rate-dependent effectsof Class I antiarrhythmic drugs are important in determining their effects on refractoriness in guinea-pig ventricle, and provide a theoretical basis for their subclassification.. Cardiovasc. Res., 1983b; 17: 344-352 Campbell, T.J.: Importance of physio-chemical properties in determining the kinetics of the effects of class I antiarrhythmic drugs on maximum rate of depolarization in guinea-pig ventricle.. Br. J. Pharmacol., 1983c; 80: 33-40 Cardinal, R., Janse, M.J., Van Eeden, I., Werner, G., Nuamann d'Alnoncourt, C , Durrer, D.: The effects of lidocaine on intracellular and extracellular potentials, activation, and ventricular arrhythmias during acute regional iscaemia in the isolated porcine heart.. Circ Res., 1981; 49: 792-806 Carlsson, L., Drews, L., Goran, D., Shiller-Linhardt, G.: Attenuation of proarrhythmias related to delayed repolarization by low-dose lidocaine in anesthetized rabbit.. J. Pharmacol. Exp. Ther., 1993; 267: 1076-1080 Carson, D.L., Cardinal, R., Savard, P., Vasseur,C, Nattel, S., Lambert, C , Nadeau, R.: Relationship between an arrhythmogenic action of lidocaine and its effect on excitation patterns in acutely ischemic porcine myocardium.. / . Cardiovasc. Pharmacol., 1986; 8: 126-136 Case, R.B., Felix, A., Castellana, F.S.: Rate of rise of myocardial Pco2 during early myocardial ischemia in the dog.. Circ. Res., 1979; 45: 324-330 Colatsky, T.J., Buzby, G.C. Jr.: N-substituted-4(3Vnitrobenzene sulfonamides. U.S. Patent No. 4,665, 227; 1987 Colatsky, T.J., Follmer, C.H., Starmer, C.F.: Channel specificity in antiarrhytmic drug action. Mechanism of potassium channel block and its role in suppressing and aggravating cardiac arrhythmias.. Circulation, 1990; 82: 2235-2242 Corr, P.B., Sobel, B.E.: The importanceof metabolites in the genesis of ventricular dysrhythmia induced by ischemia. I. Electrophysiological considerations.. Mod. Concepts Cardiovasc. Dis., 1979; 48: 43-47 Courtney, K.R.: Fast frequency-dependent block of action potential upstroke in rabbit atrium by small local anaesthetics. Life Sci., 1979; 24: 1581-1588 Courtney, K.R.: Antiarrhythmic drug design: frequency-dependent block in myocardium. In Fink, B.R. (ed): Molecular Mechanisms in Anesthesia, 1980a; pp. 111-118 Courtney, K.R., Etter, E.F.: Modulated anticonvulsant block of sodium channels in nerve and muscle. Eur. J. Pharmacol., 1983; 88: 1-9 Courtney, K.R.: Why do some drugs preferentially block open sodium channels?. /. Mol. Cell. Cardiol., 1988; 20: 461-464 Courtney, K.R.: Review: Quantitative structure/activity relations based on use-dependent block and repriming kinetics in myocardium.. / . Mol. Cell. Cardiol., 1987; 19: 319-330 Courtney, K.R.: Why do some drugs preferentially block open sodium channels?. /. Mol. Cell. Cardiol, 1988; 20: 461-464 Cranefield, P.F., Klein, H.O., Hoffman, B.F.: Conduction of the cardiac impulse. I. Delay block and one way block in depressed cardiac Purkinje fibers.. Circ. Res., 1971, 38: 302 Cranefield, P.F.: Action potentials, after potentials and arrhythmias.. Circ. Res., 1977; 38: 302 Curtis, M.J.: Regional elevation of extracellular K+ concentration in the absence of ischemia elicits ventricular arrhythmias: relevance to arrhythmogenesis during ischemia.. J. Mol. Cell Cardiol, 1989; 21 (suppl II): S138 Davies, M.J.: Pathological view of sudden cardiac death.. Br. Heart J., 1981; 45: 88-96 Daugherty, A., Frayn, K.N., Redfern, W.S., Woodward, W.: The role of catecholamines in the production of ischaemia-induced ventricular arrhythmias in the rat in vivo and in vitro.. Br. J. Pharm., 1986; 87: 265-277 Downar, E., Janse, M.J., Durrer, D.: The effect of coronary artery occlusion on subepicardial transmembrane potentials in the intact porcine heart.. Circulation, 1977, 56: 217-224 Doyle D.A., Cabral J.M., Pfuetzner R.A., Kuo A., Gulbis J.M., Cohen S.L., Chait B.T., MacKinnon R .: The structure of the potassium channel: molecular basis of K+ conduction and selectivity.. Science, 1998, 280(5360):69-77 Durrer, D., Van Dam, R.T., Freud, G.E., Janse, M.J., Jeijler, F.L., Arzbaecher, R.C: Total excitation of the isolated human heart.. Circulation, 1971; 41: 899-912 Echt, D.S., Liebson, P.R., Mitchell, L.B., et al.: Mortality and morbidity in patients receiving encainide. flecainide. or placebo: the Cardiac Arrhythmia Suppresion Trial.. N. England j. med., 1991; 324: 781-788 Ehring, G.R., Moyer, J.W., Hondeghem, L.M.: Quantitative structure activity studies of antiarrhythmic properties of lidocaine and procainamide derivatives.. J. Pharmacol. Exp. Ther., 1988; 244: 479-492 Ellis, D., Thomas, R.C: Direct measurement of the intracellular pH of mammalian cardiac muscle.. J. Physiol., 1976; 262: 755-771 El-Sherif, N., Scherlag, B.J., Lazzara, R., Hope, R.R.: Re-entrant ventricular arrhythmias in the late myocardial infarction period. 1. Conduction characteristics in the infarction zone.. Circulation,\\911; 55: 686-702 Follmer, C.H., Buzby, G.C. Jr., Jurkiewicz, N.K., Stein, R.P., Colatsky, T.J.: Selective block of Na+ and K+ channels in feline ventricular myocytes.. Biophys.J., 1988; 53:424 Follmer, C.H.,Cullman, C.A., Pirozzi, C , Colatsky, T.J.: Physico-chemical correlates of Na+ channel block in cardiac myocytes studies in a series of benzene sulfonamide analogues.. /. Moi. cell. Cardiol, 1989; 21: S112 Fozzard, H.A.: Conduction of the action potential. In Handbook of physiology; Section 2: The cardiovascular system - the heart, vol. I, American Physiological Society, 1979. Jil Gevers, W.: Generation of protons by metabolic processes in heart cells.. / . Moi. Cell. Cardiol, 1977; 9: 867-874 Goldstein, S., Landis, J.R., Leighton, R., Ritten, G., Vacu, C.M., Lantis, A., Serokiman, R.: Characterization of the resucitated out-of-hospital cardiac arrest victim with coronary heart disease.. Circ, 1981; 64: 977-984 Goldstein, S.A., Miller, C : Site specific mutations in a minimal voltage dependent K+ channel alter ion selectivity and open channel block.. Neuron, 1991; 7: 403-408 Harris, A.S., Bisteni, A., Russel, A., Brigham, J.C, Firestone, J.E.: Excitatory factors in ventricular tachycardia resulting from myocardial ischemia: potassium as a major excitant.. Science, 1954; 200-203 Harrison, D.C: Antiarrhythmic drug classification: New science and practical applications.. Amer. J. cardiol, 1985; 56:185-187 Heginbotham, L„ Lu, Z., Abramson, T., Mackinnon, R.: Mutations in the K+ channel signature sequence.. Biophysical J., 1994; 66: 1061-1067 Heinemann, S.H., Terlau, H , Stiihmer, W., Imoto, K., Numa, S: Calcium channel characteristics conferred on the sodium channel by single mutations.. Nature, 1992; 356: 441-443 Hill, H.J., Gettes, L.S.: Effect of acute coronary artery occlusion on local myocardial extracellular K+ activity in Swine.. Circulation, 1980; 61: 768-778 Hille, B.: Local anaesthetics: Hydrophilic and hydrophobic pathways for the drug-receptor reaction., gen. Physiol, 1977; 69: 497-515 Hirche, H , Franz, C , Bos, L., Bissig, R., Lang, Schramm, M.: Myocardial extracellular K+ and H+ increase and noradrenaline release as possible causes of early arrhythmias following acute coronary artery ligation in pigs.. / . Moi. Cell. Cardiol, 1980; 12: 579-593 Hodgkin, A.L., Huxley, A.F.: A quantitative description of membrane current and its application to conduction and excitation in nerve.. /. Physiol, 1952; 117: 500-544 Hoffman, B.F., Cranefield, PiF.: Physiological basis of cardiac arrhythmias.. Amer. J. Med., 37: 670 Hoffman, B.F., Dangman, K.H.: Mechanisms for cardiac arrhythmias.. Experientia; 1987; 43: 1039-1130 Hoffman, B.F., Rosen, M.R.: Cellular mechanisms for cardiac arrhythmias.. Circ. Res.; 1981; 49: 1-15 Hoffman, B.F., Spinelli, W.: Pharmacological mechanisms in the treatment of arrhythmias due to ichemia and infarction.: In Rxisen, M.R., Palti, Y. (eds.) Lethal arrhythmias resulting from myocardial ischemia and infarction.., 1989; pp 281-292 Hondeghem, L.M., Katzung, B.G.: Time- and voltage-dependent interactions of antiarrhythmic drugs with cardiac sodium channels.. Biochem. Biophys. Acta., 1977,477:373-398 Hondeghem, L.M. , Katzung, B.G.: Anti-arrhythmic agents: The modulated receptor mechanism of action of sodium and calcium-channel blocking drugs.. Ann. Rev. Pharmacol. Toxicol, 1984, 24: 387-423 Hondeghem, L.M., Mason, J.W.: Agents used in cardiac arrhythmias.. In B.G. Katzung (ed): Basic and Clinical Pharmacology, Appleton and Lange Medical Publications, Norwalk, 1989: pp. 165-197 Hondeghem, L.M., Snyders, D.J!: Class III antiarrhythmic agents have a lot of potential but a long way to go: Reduced effectiveness and dangers of reverse use dependence. Circulation, 1990; 81: 686-690 Hope, R.R., Williams, D.O., El-Sherif, N., Lazara, R., Scherlag B.J.: The efficacy of antiarrhythmic agents during acute myocardial ischemia and the role of heart rate.. Circulation, 1974; 50: 507 Hoshi, T., Zagotta, W.N.: Recent advances in understanding of potassium channel function.. Neurobiolgy, 1993; 3: 283-290 Howard, P.G., Walker, M.J.A.: Electrical stimulation studies with quinicainol. a putitave Ic agent, in the anaesthetised rat.. Proc. West. Pharmacol Soc, 1990; 33: 123-127 Ichihara, K., Haga, N., Abilo, Y.: Is ischemia-induced pH decrease of dogmyocardium respiratory or metabolic acidosis?. Am. J. Physiol, 1984; 246: H652-H657 Jackman, W.H., Friday, K.J., Anderson, J.L., Aliot, E.M., Clark,M., Lazzara, R.: The long QT syndromes: A critical review, new clinical observations and a unifying hypothsis.. Prog, cardiovasc. Dis., 1988; 31:115-172 Jan, L.Y., Jan, Y.N.: Structural elements involved in specific K+ channel functions. Ann. Rev. Physiol, 1992; 54: 537-555 Janse, M.J., Van Capelle, F.J.L.: Electronic interactions accross an inexcitable region as a cause of ectopic activity in acute regional myocardial ischemia. A study of intact porcine and canine hearts and computer models.. Circ. Res., 1982; 50: 527-537 Janse, M.J., Wit, A.L.: Electrophysiological mechanisms of ventricular arrhythmias resulting from myocardial ischemia and infarction.. Physiol. Rev., 1989;69:1049-1169 Josephson, I.R., Sanchez-Chapula, J., Brown, A.M.: Early outward current in rat singular ventricular cells.. Circ. Res., 1984; 54: 157-162 Katzung, B.G., Hondeghem, L.M., Grant, A.O.: Cardiac ventricular automaticity induced by current of injury.. Pflugens Arch., 1975; 360: 193-197 Kerr, LD., Sansom, M.S.P.: Cation selectivity in ion channels.. Nature, 1995; 373: 112 Kleber, A.G.: Resting membrane potential, extracellular potassium activity and intracellular sodium activity during acute global ischemia in isolated perfused guinea-pig hearts.. Circ. Res., 1983; 52: 442-450 Klein, C.D.P., Hopfinger, A.J.: Pharmacological Activity and Membrane Interactions of Antiarrhythmics: 4D-OSAR/OSPR Analysis.. Pharm. Res.. 1998; 15: 303-311 Koskinen, P., Kupari, M.: Alcohol and cardiac arrhythmias.. BMJ, 1992; 304: 1394-1395 Kramer, J.B., Saffitz, J.E., Witkowski, F.X., Corr, P.B.: Intramural reentry as a mechanism of ventricular tachycardia during evolving canine myocardial infarction.. Circ. Res., 1985; 27 (suppl):83-89 Lazzara, R., El-Sherif, N., Hope, R.R., Scherlag, B.J.: Ventricular arrhythmias and electrophysiological consequences of myocardial ischemia and infarction.. Circ. Res., 1978; 42: 740-749 Lazzara, R., Scherlag, B.J.: Generation of arrhythmias in myocardial iscaemia and infarction.. Am. J. Cardiol, 1988; 61: 20A-26A Liman, E.R., Hess, P., Weaver, F., Koren, G.: Voltage-sensing residues in the S4 region of a mammalian K+ channel.. Nature, 1991; 353: 752-756 Lummam W.C. Jr., Wohl, R.A., Darvey, D.D.: Rational design of 4-fCmethylsulfonyDaminolbenzamides as class III antiarhythmic agents.. J. Med. Chem., 1987; 30: 758-763 Lynch, J.J., Coskey, L.A., Montgomery, D.G., Lucchesi, B.R.: Prevention of ventricular fibrillation by dextrorotary sotalol in a conscious canine model of sudden coronary death.. Am. Heart J., 1985; 109: 949-958 Mackinnon, R, Reinhart, P.H., White, M.M.: Charybdotoxin block of shaker K+ channels suggests that different types of K+ channels share common structural features.. Neuron, 1988; 1: 997-1001 Mackinnon R.: Determination of the subunit stoichiometry of a voltage-activated potassium channel.. Nature, 1991; 350: 232-235 MacKinnon R., Cohen S.L., Kuo A., Lee A., Chait B.T.: Structural conservation in prokaryotic and eukaryotic potassium channels.. Science, 1998; 280(5360): 106-9 McKibbin, J.K., Pocock, W.A., Barlow, J.B., Millar, R.N.S., Obel, I.W.P.: Sotalol hypokalemia, svncope and torsade de pointes.. Br. Heart J., 1984; 51: 157-162 Moe, G.K., Rheinboldt, W.C, Abildskov, J.A.: A computer model of atrial fibrillaion.. Am. Heart J., 1964; 67: 200-220 Milne, J.R., Hellestrand, K.J., Bexton, R.S., Burnett, P.J., Debbas, N.M.G., Camm, A.J.: Class I antiarrhythmic drugs - Characteristic electrocardiographic differences when assessed by atrial and ventricular pacing.. Eur. Heart J., 1984; 5: 99-107 Mulvey, D., King, G.F., Cooke, R.M., Doak, D.G., Harvey, T.S., Campbell, I.D.: High resolution IH NMRstudy of the solution structure of the S$ segment of the sodium channel protein.. FEBS Letters, 1989; 257: 113-117 Nademanee, K., Feld, G., Hendrickson, J., Singh, P.N., Singh, B.N.: Electrophysiological and antiarrhythmic effects of sotalol in patients with life-threatening ventricular tachyarrhythmias..Circulation. 1985; 72: 555-564 4Z1 Nademanee, K.: The amiodarone odyssey.. /. Am. Coll. Cardiol, 1992; 20: 1063-1065 Noda, M., Ikeda, T., Kayano, T.: Existence of distinct sodium channel messenger RNAs in rat brain.. Nature. 1986; 320: 188-192 Noda M., Numa, S.: Structure and function of sodium channel.. J. Receptor Res., 1987; 7: 467-497 O'Rourke, B., Ramza, B.M., Marban, E.: Oscillations of membrane current and excitability driven by metaboloic oscillations in heart cells.. Science, 1994; 265: 962-966 Opthof, T.: IK1 blockade is unlikely to be a useful antiarrhythmic mechanism.. Cardiovasc. Res.,1994; 28: 420 Palotta, B.S., Wagonr, P.K.: Voltage dependentpotassium channels since Hodgkin and Huxlev.. Physiol. Rev., 1992; 72 (suppl.): S49-S67 Penz, W., Pugsley, M., Hsieh, M.Z., Walker, M.J.A.: A new measure (RSh) for detecting possible sodium channel blockade in vivo in rats.. / . Pharmacol. Methods, 1992; 27: 51-58 Pugsley, M.K., Penz, W.P., Walker, M.J.A., Wong, T.M.: Cardiovascular actions of the k-agonist. U-50.488H. in the absence and presence of opioid receptor blockade.. Br. J. Pharmacol, 1992; 105: 521-526 Pusch, M., Noda, M., Stuhmer, W., Numa, S., Conti, F.: Single point mutations of the sodium channel drastically reduce the pore permeability without preventing its gating.. Eur. Biophys. J., 1991; 20: 127-133 Roden, D.M., Bennett, P.B., Snyder, D.J., Balser, J.R., Hondeghem, L.M.: Quinidine delays Ik activation in guinea pig ventricular myocytes.. Circ. Res., 1988; 62: 1055-1058 Rosen, M.R., Reder, R.F.: Does triggered activity have a role in the genesis of cardiac arhythmias?. Ann. Intern. Med., 1981; 94: 794-801 Rosen, M.R., Spinelli, W.: Some recent concepts concerning the mechanisms of action of antiarrhythmic drugs.. PACE. 1988; 11: 1485-1498 Saint, K.M., Abraham, S., Macleod, B.A., McGough, J., Yoshida, N., Walker, M.J.A.: Ischemic but not reperfusion arrhythmias depend upon serum potassium concentration.. J. Moi Cell Cardiol, 1992; 24: 701-710 Sasyniuk, B.I., Mendez, C : A mechanism for reentry in canine ventricular tissue.. Circ. Res., 1981; 28: 3-15 Sato, C , Matsumoto, G.: Proposed tertiary structure of the sodium channel.. Biocem. Biophys. Res. Comm., 1992; 186: 1158-1167 Scheidt, S.: Basic electrocardiograpgy: Lead, axes, and arrhythmias.. In W. Heidel (ed.), Clinical Symposia, CIBA Pharmaceutical Company, ON., 1983; 35(3): pp 1-16 Schlepper, M.: cardioprotective effects of antiarrhythmic drugs.. Eur. Heart J., 1989; 10:73-80 Schwartz, P.H.: Idiopathic long QT syndrome: progress and questions.. Am. Heart J., 1985; 109: 399-412 Seeley, R.N.: Proton generation and control during anaerobic glycolysis in heart cells.. /. Moll. Cell cardiol, 1980; 12: 14-83-1486 Selzer, A., Wray, H.W.: Quinidine syncope. Paroxysmal ventricular fibrillation occuring during treatment of chronic atrial arrhythmias.. Circulation, 1964; 30: 17-26 Sheldon, R.S., Duff, H.J., Hill, R.J.: Class I anti-arrhythmic drugs: structure and function at the cardiac sodium channel.. Clin . Invest. Med., 1991; 14: 458-465 Sheldon, R.S., Thakore, E.: Ring and Link Requirements for Tocainide Binding to the Class I Antiarrhythmic Drug Receptor on Rat Cardiac Myocytes.. / . Pharm. Exp. Ther., 1995; 272: 1005-1010 Singer, D.H., Ten Eick, R.E.: Pharmacology of cardiac arrhythmias.. Progr. Cardiovasc. Disc., 1969; 11: 488 Singh, B.N, Vaughan Williams, E.M.: A third class of antiarrhythmic action. Effects on atrial and ventricular intracellular potentials, and other pharmacologic! actions on cardiac muscle, of MJ 1999 and AH 3474.. Br. J. Pharmacol, 1970; 39: 675-687 1*3 Spinelli, W., Hoffman, B.F.,: Mechanisms of termination of reentrant atrial arrhythmias by class I and class III antiarrhythmic agents.. Circ. Res., 1989; 65: 1565-1579 Starmer, C.F., Grant, A.O., Strauss, H.C.: Mechanisms of use-dependent block of sodium channels in excitable membranes by local anaesthetics.. Biophys. J., 1984; 46: 15-27 Stiihmer, W., Conti, F., Suzuki, H.: Structural parts involved in activation and inactivation of the sodium channel.. Nature, 1986; 339: 597-603 Surawicz, B., Knoebel, S.: Long Qt: good, bad or indifferent?. / . Am. Coll. Cardiol, 1985; 4: 494-516 Szekeres, L.: Role of the autonomic nervous system in the mechanism of cardiac rhythm disturbances.. In A.G.B Kovach, P. Sandor, M. Kollai (eds) cardiovascular Physiology. Neural Control Mechanisms., Adv. Physiol. Sci., 1981; 9: 57-65 Takumi, T., Ohkubo, H., Nakanishi, S.: Cloning of a membrane protein that induces slow voltage-gated potassium current.. Science, 1988; 242: 1042-1045 Task Force of the Working Group on Arrhythmias of the European Society of Cardiology: The Sicilian Gambit. A new approach to the classification of antiarrhytmic drugs based on their actions on arrhythmogenie mechanisms.. Circulation, 1991; 84: 1831-1851 Tygat, J., Vereecke, J., Carmeleit, E.: Reversal of rectification and alteration of selectivity and pharmacology in a mammalianKvl.l potassium channel by deletion of domains SI to S4.. /. Physiol, 481: 7-13 Vleugels, A., Vereecke, J., Carmeleit, E.E.: Ionic currents during hypoxia in voltage clamped cat ventricular muscle.. Circ. Res., 1980; 47: 501-508 Vaughan-Williams, E.M., Szekeres, L.: A comparison of tests for antifibrillatory action.. BritJ. Pharmacol, 1961; 17: 424-432 Vaughan-Williams, E.M.: Classification of anti-arrhythmic drugs.. In Sandoe, E., Flensted-Jensen, E., Olsen, K.H., (eds.) Symposium on cardiac arrhythmias. Astra, Elsinore, Denmark, 1970: pp. 449-501 Vaughan-Williams, E.M.: A classification of antiarrhythmic actions reassessed after a decade of new drugs.. /. Clin. Pharmacol, 1984a; 24: 129-147 Waldo, A.L., Kaiser, G.A.: A study of ventricular arrhythmias associated with acute myocardial infarction in the canine heart.. Circulation, 1973; 47: 1222-1228 Williams, D.O., Scherlag, B.J., Hope, R.R., El-Sherif, N., Lazzara, R.: The pathophysiology of malignant ventricular arrhythmias during acute myocardial ischemia.. Circulation, 1974; 50: 1163-1172 Williamson, J.R.: glycolytic control mechanisms: II. Kinetics of intermediate changes during the aerobic anoxic transition in the perfused rat heart.. / . Biol. Chem., 1966; 12: 1483-1486 Wilson, G.G., Sivaprasadarao, A., Findlay, J.B.C., Wray, D.: Changes in activation gating of IsK potassium currents brought about by mutations in the transmembrane sequence.. FEBS Letters, 1994; 353: 251-254 Wit, A.L., Rosen, M.R., Hoffman, B.F.: Electrophysiology and pharmacology of cardiac arrhythmias. II. Relationship of normal and abnormal electrical activity of cardiac fibers to the genesis of arrhythmias.. Am. Heart J., 1974; 88: 515-524 Yeola ,S.W., Rich, T.C., Uebele, V.N., Tamkun, M.M., Snyders, D.J.: Molecular analysis of a binding site for quinidine in a human cardiac delayed rectifier K+ channel. Role of S6 in antiarrhythmic drug binding.. Circ. Res., 1996; 78(6): 1105-14 Zipes, D.P.: Proarhythmic effects of antiarrhythmic drugs.. Am. J. Cardiol, 1987; 59: 26E-31E Zuanetti, G., Corr, P.B.: Antiarrhythmic efficacy of a new class III agent. UK-68.798. during chronic myocardial infarction: evaluation using three-dimensional mapping.. J. Pharm. Exp. Ther., 1991; 256: 325-334 "@en ; edm:hasType "Thesis/Dissertation"@en ; vivo:dateIssued "1999-05"@en ; edm:isShownAt "10.14288/1.0088914"@en ; dcterms:language "eng"@en ; ns0:degreeDiscipline "Pharmacology"@en ; edm:provider "Vancouver : University of British Columbia Library"@en ; dcterms:publisher "University of British Columbia"@en ; dcterms:rights "For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use."@en ; ns0:scholarLevel "Graduate"@en ; dcterms:title "The effects of changes at the nitrogens on the pharmacological profile of a series of six analogues of rsd939"@en ; dcterms:type "Text"@en ; ns0:identifierURI "http://hdl.handle.net/2429/8970"@en .