Antiarrhythmic Drugs

Singh-Vaughan Williams Classification of Antiarrhythmic Drugs

Antiarrhythmic drugs include a wide range of structural and functional classes. Although not perfect, a highly useful framework for their classification is that proposed by Singh and Vaughan Williams, usually referred to as the Singh-Vaughan Williams (SVW) classification . The SVW classification categorizes antiarrhythmic drugs into four classes ( Table 27.1 ). Class I denotes sodium (Na ⁺ ) channel blocking activity, with resultant delay in phase 0 depolarization and/or altered action potential duration. Class II agents counteract the effects of endogenous catecholamines (in particular epinephrine and norepinephrine) by antagonism of β adrenergic receptors, and hence are known as β blockers or β antagonists. Class III antiarrhythmics prolong the action potential and refractory period, acting primarily by potassium (K ⁺ ) channel blockade. Class IV agents reduce heart rate, primarily by L-type calcium (Ca ²⁺ ) channel blockade, which slows conduction through the sinoatrial (SA) and atrioventricular (AV) nodes. (For a more detailed discussion of cardiac electrophysiology and myocyte depolarization, see Chapter 23 ).

The primary reason that the SVW classification is to some extent imperfect is that many, if not most, antiarrhythmic drugs exhibit physiologically significant actions in more than one of the four classes. Typically the classification is based on the action that was first described for each drug, which is generally but not necessarily the most therapeutically significant mechanism of action for all types of arrhythmia for which the drug is prescribed. This discrepancy can occur because of nonspecificity of the drug, or because major metabolites of the drug exhibit different activity than that of the drug itself. Furthermore, some important antiarrhythmic drugs, including digoxin and adenosine, do not fit primarily into any of the four classes. However, the SVW classification is still the most useful framework for describing antiarrhythmic drugs, especially if one considers the net effect of each drug rather than all its specific molecular targets. The clinical pharmacology of the major antiarrhythmic drugs is summarized in Table 27.2 .

Sodium Channels and Class I Antiarrhythmic Drugs

Voltage-gated Na ⁺ (Na _v ) channels open in response to cell membrane depolarization to permit influx of Na ⁺ , further depolarizing the cell. The predominant Na _v channel in human cardiac electrophysiology is Na _v 1.5, which is encoded by the SCN5A gene. In most cardiac myocytes, Na _v 1.5 activation mediates the upstroke in phase 0 of the action potential, in which the membrane potential is depolarized from around −70 mV to +20 mV as the result of Na ⁺ influx ( Fig. 27.1A ). Na _v 1.5, like other mammalian Na _v channels, inactivates rapidly, which together with the transient outward K ⁺ current, I _to , shapes the notch in phase 1 at the beginning of the human cardiac myocyte action potential.

Class I antiarrhythmics exhibit Na _v channel blocking activity. These drugs are often referred to as “membrane stabilizing agents” because by blocking cardiac Na _v channels, which mediate myocyte depolarization, they reduce cellular excitability. Na _v channels are composed of a 24-transmembrane-segment pore-forming α subunit that consists of four homologous domains (DI-DIV) and also bears four voltage-sensing domains and one or more inactivation gates ( Fig. 27.2A ). The Na _v 1.5 α subunits form cardiac ion channel complexes with single-transmembrane segment β subunits, encoded by the SCNxB genes, which modify Na _v 1.5 function and pharmacology.

Class I antiarrhythmic drugs are thought to bind in the inner pore vestibule of Na _v channels, with drugs from different structural classes, including lidocaine, flecainide, and quinidine, and the anticonvulsant phenytoin, binding to an overlapping but nonidentical site influenced experimentally by mutations in the S6 transmembrane segment of domain IV. Drugs in class I are subcategorized as Ia, Ib, or Ic depending on their effects on Na ⁺ channel conduction and resultant effects on action potentials in cardiac myocytes expressing the predominant cardiac form of Na _v channel, Na _v 1.5 ( Fig. 27.2B ).

Na _v 1.5 is relatively insensitive to the canonical and lethal Na _v channel antagonist tetrodotoxin (TTX), a nerve toxin present, among other animal sources, in the nerves, skin, and gonads of the Japanese puffer fish, Fugu rubripes . F. rubripes is saddled with the dual distinction of having the shortest known genome of any vertebrate organism and being a delicacy in Japan that must be prepared by certified chefs to decrease the chances of ingestion of lethal doses of TTX by adventurous diners. Compared with its effects on Na _v 1.5, TTX acts with up to 10 ³ -fold higher potency on the predominantly neuronal Na _v channel subtypes Na _v 1.1-1.3 (SCN1A-3A) and the skeletal muscle–expressed channel Na _v 1.4 (SCN4A) . TTX, nicknamed “zombie powder” because its paralysis- and coma-inducing effects have led to its use in voodoo ceremonies (it does not cross the blood-brain barrier; those ingesting sublethal doses are potentially conscious while paralyzed), is therefore not a useful antiarrhythmic (owing to its inefficacy and lethality at low doses). However, TTX prolongs the local anesthetic effect of bupivacaine when the two are coadministered. A recent high-resolution x-ray crystallographic structure of an Na _v channel from Arcobacter butzleri has provided a first look at this class of proteins essential to function of nerve and muscle, and will likely enhance future class I antiarrhythmic development (see Chapter 20 ).

β Receptors and Class II Antiarrhythmics

Class II antiarrhythmic drugs, also known as β blockers, antagonize the β-adrenergic receptor (β receptor). This β blockade prevents activation of adenylyl cyclase and the consequent increase in intracellular cyclic adenosine monophosphate (cAMP), and thus also activation of its principal downstream target cAMP-dependent protein kinase (PKA), and the promotion of maximal myocardial performance that normally results from the enhancement of sympathetic nervous tone ( Fig. 27.3 ; see Chapters 13 and 14 ).

β Blockers are selective, in that they do not block other receptors, and specific, in that they do not antagonize cardiac stimulation and vasodilatation elicited by agents other than β agonists. All β blockers share the basic structure of a β sympathomimetic side chain, which confers affinity for the receptor, along with an aromatic substituent, which determines potency; most are derivatives of the class-defining agent propranolol (see Fig. 27.3B ). β Blockers are effective as antiarrhythmic agents because, by blocking the action of the sympathetic nervous system on the heart, they depress SA and AV node function, decrease conduction and automaticity, and prolong atrial refractory periods. β Blockers also reduce blood pressure, probably arising from a combination of reduced cardiac output, renal renin release, and perhaps even effects within the central nervous system (see Chapter 26 ). Inasmuch as an exaggerated increase in sympathetic tone typifies the deleterious reaction to congestive heart failure, β blockers, particularly the cardioselective ones, are used successfully also in the treatment of this ailment. Propranolol is marketed as a D,L -racemic mixture (see Chapter 1 for a discussion of the pharmacologic implications of chirality), the reason being that the L -form is the β blocker while the D -form is a “membrane stabilizer,” which adds antiarrhythmic effect to the β-blocking properties of the L -enantiomer.

Potassium Channels and Class III Antiarrhythmic Drugs

Class III antiarrhythmic drugs are defined by their ability to block K ⁺ channels. This activity increases action potential duration in cardiac myocytes and prolongs the refractory period—that is, it extends the period during which the heart is refractory to premature electrical stimuli. Cardiac K ⁺ channels exhibit a much wider variety than other cardiac ion channels. The two most physiologically and therapeutically important families of K ⁺ channels in the heart, based on current understanding, are the voltage-gated K ⁺ (K _v ) channels and inward rectifier potassium (K _ir ) channels. Channels in both these families are predominantly involved in the repolarization phases of the cardiac myocyte action potential, because both K _v and K _ir channels pass inward K ⁺ currents only when the cell membrane potential is negative to the K ⁺ equilibrium potential, which is around −80 mV under physiologic conditions.

K _v channels are each composed of several subunit types. Similar to Na _v channels, the pore-forming α subunits of K _v channels are arranged in a 24-transmembrane-segment array forming an aqueous central pore with external voltage-sensing modules ( Fig. 27.4A ). However, in K _v channels, this is composed of a tetramer of noncovalently linked α subunits (each subunit having six transmembrane segments) rather than one contiguous α subunit with four homologous six-segment domains as in Na _v and Ca _v channels. In K _v channel α subunits, the fourth transmembrane segment (S4) bears the basic residues that confer voltage sensitivity, while S6 lines the pore ( Fig. 27.4A ). High-resolution x-ray crystallographic structures of bacterial and eukaryotic K _v channels have revolutionized the study of these ubiquitous and essential proteins, including current understanding of drug binding sites.

The most important K _v α subunits in human ventricular repolarization are the ether-à-go-go related gene product (hERG; also named KCNH2 ) and KCNQ1 . Tetramers of each of these coassemble with multiple single-transmembrane-domain ancillary or β subunits from the KCNE gene family. KCNQ1-KCNE1 complexes primarily generate the slowly activating I _Ks current; hERG- KCNE2 complexes generate I _Kr ; and each of the five KCNE proteins probably regulates these and other α subunits in the heart too. I _Kr and I _Ks are crucial to phase 3 repolarization (see Fig. 27.1A ); therefore blocking these currents delays ventricular repolarization, which can be proarrhythmic or antiarrhythmic depending on the disease state and other genetic and environmental factors. hERG is particularly sensitive to drug block by a wide range of drugs, owing to its bearing an unusual array of hydrophobic residues, not conserved among other K _v channels, in an internal pore cavity also predicted to be wider than in other K _v channels ( Fig. 27.4B ).

Inhibitors of K _v channels tend to bind in one of three distinct sites: the outer vestibule, the inner vestibule (drugs in these classes would be considered pore blockers), or on the voltage sensor, which is rarer among small molecules but is seen with some toxins from venomous animals, such as hanatoxin and SGTx from tarantula spiders. The canonical inhibitors of hERG and KCNQ1 are E-4031 and chromanol 293B, respectively, both of which bind in the inner vestibule and show relatively high specificity for their targets. Additionally, both drugs are sensitive to the presence of KCNE subunits in complex with the α subunits. E-4031 will not be used in the future for antiarrhythmic therapy because pure hERG antagonists such as this cause dangerous repolarization delays. I _Ks inhibitors are still being considered because it might be therapeutically useful to counteract the upregulation of I _Ks during periods of high sympathetic activity associated with arrhythmia triggering. Other important K _v channel targets of class III antiarrhythmic drugs include the K _v 4 α subunits that generate human ventricular I _to (particularly active during phase 1 of the ventricular action potential), and K _v 1.5, which generates the ultrarapidly activating K ⁺ current, I _Kur , important in atrial myocyte repolarization. K _v 4 and K _v 1.5 α subunits are also regulated by the KCNE subunits. In addition, all the K _v channels discussed herein are regulated by a host of cytoplasmic β subunits and other regulatory proteins ( Fig. 27.4C ), each of which can affect channel pharmacology either directly or indirectly, because changes in gating alter drug binding kinetics.

K _ir channels do not possess a voltage sensor but exhibit inward rectification because they are inhibited at more positive membrane potentials by intracellular constituents, including magnesium (Mg ²⁺ ) and polyamines such as spermine. These channels are composed of a tetramer of α subunits, each with only two transmembrane segments, around a central, aqueous, K ⁺ -selective pore ( Fig. 27.4D ). The ventricular inward rectifier current I _K1 is generated by K _ir 2 family α subunits and contributes to stabilizing the cardiac potential, passing current at either end but less so when the myocyte is strongly depolarized (see Fig. 27.1A ). The K _ATP channel, which also probably contributes to cardiac excitability, is an octamer of four K _ir 6 α subunits and four membrane-spanning sulfonylurea receptor subunits that render it sensitive to adenosine triphosphate (ATP). A host of class III drugs, including amiodarone and dofetilide, inhibit K _ir channels by direct pore block within the inner vestibule.

Calcium Channels and Class IV Antiarrhythmics

Class IV agents slow AV nodal conduction, primarily by L-type Ca ²⁺ channel (LTCC) antagonism. LTCCs mediate the upstroke of nodal cell action potentials, unlike true atrial and ventricular myocyte action potentials in which the upstroke is mediated primarily by the faster activating and inactivating Na _v channel, Na _v 1.5. Hence, nodal action potentials exhibit a much slower upstroke than that of typical atrial and ventricular myocytes (see Chapter 23 ). The transmembrane topology of voltage-gated Ca ²⁺ channel (Ca _v ) α subunits mirrors that of the Na _v channel α subunits: 24 transmembrane segments around an aqueous pore. LTCCs also incorporate a requisite array of ancillary subunits: the cytoplasmic β subunit, transmembrane δ and γ subunits, and extracellular α ₂ subunit ( Fig. 27.5A ). In cardiac muscle, the LTCC Ca _v 1.2 α subunit is located in the T-tubules and is activated by cellular depolarization, via its voltage-sensing apparatus, which moves upon membrane depolarization and opens the discrete but connected pore. Ca ²⁺ influx through Ca _v 1.2, down the Ca ²⁺ concentration gradient, helps depolarize the cell and increase cytosolic [Ca ²⁺ ] directly, but it also works indirectly by activating the sarcoplasmic reticulum–located ryanodine receptor (RyR2 in cardiac muscle) through Ca ²⁺ -activated Ca ²⁺ release. In skeletal muscle, the mechanism is somewhat different: Ca _v 1.1 is mechanically linked to RyR1 and acts as the latter's voltage sensor. Thus in skeletal muscle, RyR1 is activated primarily by membrane depolarization, with the Ca _v 1.1 S4 domains acting as the RyR1 voltage sensors (see Fig. 27.5B ).

Most clinically relevant Ca _v channel blockers are in one of three chemical classes: the dihydropyridines, which are not generally indicated for arrhythmias; the phenylalkylamines, exemplified by verapamil; and the benzothiazepines, exemplified by diltiazem. Diltiazem and verapamil (see Fig. 27.5C ) are thought to bind overlapping but distinct sites within the S6 segments of repeats III and IV, and the Ca ²⁺ selectivity filter of the α ₁ subunit of the cardiac LTCC Ca _v 1.2.

Categories of Arrhythmogenic Mechanisms

The mechanistic bases for most, if not all, arrhythmias can be placed in one of three categories. These are discussed with the most common modes of treatment.