REVIEW www.nature.com/clinicalpractice/neuro Mechanisms of Disease: sodium channels and neuroprotection in multiple sclerosis—current status Stephen G Waxman SUMMARY INTRODUCTION The past decade has seen increasing interest in Sodium channels can provide a route for a persistent influx of sodium ions the possibility of neuroprotective therapy with into neurons. Over the past decade, it has emerged that sustained sodium sodium channel blockers in multiple sclerosis influx can, in turn, trigger calcium ion influx, which produces axonal injury (MS), as a result of recognition that axon degen- in neuroinflammatory disorders such as multiple sclerosis (MS). The development of sodium channel blockers as potential neuroprotectants eration is a major contributor to disability in in MS has proceeded rapidly, and two clinical trials are currently ongoing. MS, and the demonstration of a critical role The route from the laboratory to the clinic includes some complex turns, for sodium channels in degeneration of CNS however, and a third trial was recently put on hold because of new data axons. This area is one of rapid flux—5 years that suggested that sodium channel blockers might have multiple, complex ago sodium channel blockers had not been actions. This article reviews the development of the concept of sodium studied in animal models of MS, and 2 years channel blockers as neuroprotectants in MS, the path from laboratory to ago clinical studies had not begun to assess the clinic, and the current status of research in this area. protective effects of sodium channel blockers in humans with MS. Now, however, data are avail- KEYWORDS multiple sclerosis, neuroprotection, sodium channel blockers able from animal models on the effects of four REVIEW CRITERIA sodium channel blockers, all of which are in PubMed was searched for articles published up to October 2007, including routine clinical use for other indications, and electronic early release publications. Search terms included “multiple sclerosis”, two clinical studies of sodium channel blockers “neuroprotection”, “axonal injury”, “sodium channels” and “sodium channel in patients with MS are ongoing. This article blockers”. Relevant articles were retrieved and prioritized for inclusion in the Review and their references were checked for additional material when will review the development of the concept of appropriate. In addition, the author used his own files of references. sodium channel blockers as neuroprotectants in MS, the path of translation from laboratory to clinic, and the current status of this field. EARLY USE OF SODIUM CHANNEL BLOCKERS IN MULTIPLE SCLEROSIS Early experience of neurologists in the use of sodium channel blockers in MS centered largely on the use of carbamazepine—and in some cases other sodium channel blockers—to treat posi- tive phenomena such as tonic flexion spasms, Lhermitte’s sign and, most commonly, tri geminal neuralgia.1,2 Phenytoin, and lidocaine and its orally absorbed derivative mexiletine, have also SG Waxman is the Bridget Flaherty Professor of Neurology, Neurobiology been used for these applications in patients with 3 and Pharmacology at Yale Medical School, New Haven, CT, and Director MS with some degree of success. of the Center for Neuroscience and Regeneration Research, VA Connecticut The pathophysiological basis for posi- Healthcare, West Haven, CT, USA. tive phenomena such as trigeminal neuralgia in MS is not fully understood. It is known, Correspondence Department of Neurology, LCI 707, Yale Medical School, PO Box 20818, New Haven, however, that sodium channels contribute CT 06520–8018, USA a transmembrane current that can produce [email protected] oscillations in membrane potential, resulting in ectopic firing in demyelinated axons.4 The use Received 21 August 2007 Accepted 23 November 2007 Published online 29 January 2008 www.nature.com/clinicalpractice of sodium channel blockers as treatments for doi:10.1038/ncpneuro0735 positive clinical phenomena in MS, therefore, MARCH 2008 VOL 4 NO 3 NATURE CLINICAL PRACTICE NEUROLOGY 159 NNature.indtature.indt 1 228/11/078/11/07 99:46:50:46:50 aamm REVIEW www.nature.com/clinicalpractice/neuro has a rational basis. In addition to dampen ing Na+,K+-ATPase pump to extrude sodium, abnormal ectopic activity, however, the blocking thereby promoting calcium-importing ‘reverse’ of sodium channels would be expected to operation of the Na+/Ca2+ exchanger, which decrease the safety factor (the ratio of current results in elevated, injurious levels of intra- available to stimulate a given node of Ranvier axonal calcium. There is evidence that the versus the current needed to generate an action increased intra-axonal calcium levels can acti- potential at the node) for trans mission of vate intracellular mechanisms that increase the normal action potentials. This drug-induced amplitude of—and prolong—the sodium decrease in safety factor would not be func- current flowing through Nav1.6 channels situ- tionally important in healthy myelinated ated along myelinated and demyelinated axons, fibers in which the safety factor is 5–6, but providing a positive feedback loop that imports it could produce conduction block in some still more calcium, thereby further amplifying de myelinated fibers in which the safety factor the damage.10 Injury-induced sodium influx was already reduced by demyelination to can also trigger release of calcium from intra- around 1.0, with no room for further compro- cellular stores within white matter axons.11 The mise.5 Impaired impulse conduction resulting increased intra-axonal calcium concentration from the reduction in safety factor produced by that results from these movements of calcium sodium channel blockers would be expected to into the axoplasm activates multiple injurious be transient, and to be reversed on unbinding pathways that involve calpain and other degra- and metabolism of the drug. Consistent with dative enzymes.12 This injurious cascade can these observations in the laboratory, transient occur in normal myelinated axons (in which worsening of negative symptoms (for example, sodium channels are clustered at nodes of weakness) has been noted in a small number of Ranvier13) when they are subjected to energy patients with MS during treatment for positive deprivation (for example, through anoxia). In symptoms with carbamazepine but, importantly, individuals with MS, it is expected that it was reported to reverse within a few days demyelinated axons, some of which display long after cessation of treatment.6 In an attempt to expanses of membrane that express sodium capitalize on this phenomenon, Sakurai et al.7 channels,14 would be especially susceptible to demonstrated that lidocaine can unmask silent this mode of injury. Remyelinated axons, in demyelinating lesions in MS and proposed that which nodes of Ranvier can be very closely this might provide a useful diagnostic test. spaced,15,16 might be expected to be at A 3-year follow-up study of patients with increased risk.17 MS treated with sodium channel blockers for Building upon these observations, subse- neuropathic pain or other paroxysmal symp- quent experiments demonstrated that the toms revealed adverse effects that mimicked sodium blockers tetrodotoxin (TTX)9, lidocaine, a relapse in 12 out of 36 patients treated with procaine,18 mexiletine,19 phenytoin, and carba- carbamazepine. In these patients, clinical status mazepine20 can protect white matter axons from returned to the pretreatment level once the drug anoxic and ischemic injury in vitro. Importantly, was discontinued.8 No studies have so far been protection could be achieved in these in vitro carried out to determine whether there are long- experiments with sodium channel blockers at term effects of treatment with sodium channel concentrations that did not compromise the blockers in MS, and no case reports of long-term conduction of action potentials.18,21 changes have been published. An important link to axonal injury in MS was established with the discovery that nitric SODIUM CHANNELS AS DRIVERS oxide (NO) is present at increased levels OF AXONAL INJURY IN THE CNS within MS lesions, where, among other effects, Early evidence that voltage-gated sodium chan- it can trigger axon degeneration similar to nels have a role in degeneration of CNS axons the axonal injury produced by anoxia.22,23 was gained from an in vitro model of CNS This action can be attributed in part to dele- anoxia.9 It was shown that axon degeneration terious effects of NO on mitochondrial func- within white matter could be triggered by a tion,24,25 which result in a reduction in ATP cascade (Figure 1) involving sodium influx levels and a rundown of Na+,K+-ATPase, through noninactivating sodium channels. This thereby compromising the axon’s ability to influx overwhelms the ability of the ATP-fueled maintain normal trans membrane sodium 160 NATURE CLINICAL PRACTICE NEUROLOGY WAXMAN MARCH 2008 VOL 4 NO 3 NNature.indtature.indt 1 228/11/078/11/07 99:46:50:46:50 aamm REVIEW www.nature.com/clinicalpractice/neuro Microglia Hypoxia/ Inflammation Altered gene ischemia expression Nav + 2+ + + Na /Ca exchanger Nitric oxide Na ,K -ATPase Nav1.6 Ca2+ + 2+ Rundown Persistent Na+ Na Ca influx of ATPase influx Calcium-induced calcium release Energy failure Mitochondrial Ca2+ Depolarization Activation of nitric oxide synthase Activation of proteases Axon Activation of lipases Figure 1 The central role of sodium channels in the axon degeneration cascade. A number of factors (nitric oxide; ischemia that results from inflammation of small blood vessels; and decreased expression of genes that encode mitochondrial redox carriers) contribute to energy failure and subsequent rundown of the Na+,K+-ATPase pump, with subsequent depolarization and loss of capacity to maintain transmembrane ion gradients. The depolarization activates sodium channels (e.g. Nav1.6), which provide a route for persistent sodium influx. This process, in turn, drives the Na+/Ca2+ exchanger to operate in a calcium-importing mode. The rise in intracellular calcium induces a further increase in calcium levels via calcium-induced calcium release.