Early studies in EAE demonstrated a loss of axons (Raine and Cross 1989; White et al. 1992; Kornek et al. 2000; Onuki et al. 2001; Wujek et al. 2002), which was correlated with the accumulation of nonremitting disability (Wujek et al. 2002). To determine whether the improved neurological status observed in EAE with sodium channel blocking agents was accompanied by protection of axons from degeneration, we assessed the magnitude of axonal degeneration in spinal cords of treated versus untreated mice and rats with EAE. Our results with the two distinct EAE models, and with three different sodium channel blocking agents, demonstrated consistent protection of axons in these animals. In C57/Bl6 mice, untreated EAE was accompanied by significant loss of axons within the dorsal corticospinal tract (CST, a descending tract) and the dorsal funiculus (DF, ascending tracts) (Fig. 8.2a; Black et al. 2007), with an approximately 60% loss of axons in the CST and a 40% loss in the DF compared to control mice, as assessed at 28–30 days post-MOG inoculation (Fig. 8.2a; Lo et al. 2003). Interestingly, in a separate study, axonal loss in the CST and DF at 180 days post-MOG inoculation were similar to those at 28–30 days post-MOG injection (Black et al. 2007), indicating that axonal degeneration in this murine EAE model occurs primarily during the initial phase of the disease. In contrast to untreated mice with EAE, treatment with phenytoin at clinically relevant levels significantly attenuated the loss of axons in mice with EAE at 28–30 days post-MOG inoculation, such that there was only a 28% loss of axons in the corticospinal tract and a 17% loss in the dorsal funiculus compared with control values (Fig. 8.2a). Significantly, the axonal protection afforded by phenytoin treatment persisted during longer treatment periods, for at least 180 days (Black et al. 2007).

Similar protective effects of sodium channel blockade were seen in DA rats with chronic-relapsing EAE treated with flecainide or lamotrigine. DA rats inoculated with spinal cord homogenate exhibited prominent axonal loss and ongoing axonal degeneration, particularly within the medial region of the dorsal columns (fasciculus gracilis) (Fig. 8.2b, c; Bechtold et al. 2004), that was closely associated with the peak deficit scores exhibited by these rats. DA rats with EAE that had peak scores ≥8 (on 15-point scale) displayed an approximately 40% loss of axons relative to control rats. In contrast, both flecainide (Bechtold et al. 2002, 2004) and lamotrigine (Bechtold et al. 2006) treatments provided significant protection of spinal cord axons when administered to DA rats with EAE (Fig. 8.2b, c). In comparison with the number of axons in normal rats, flecainide treatment commencing either 3 days prior to or 7 days following homogenate injection provided near complete axonal protection, with axonal counts remaining at 98%  ±  20% and 83%  ±  14% of control values, respectively (Bechtold et al. 2004). Similar results were obtained with lamotrigine treatment of DA rats with EAE, such that, when assaying degeneration as a percentage of total area measured, lamotrigine-treated rats exhibited axonal degeneration in only 9% of the area, while untreated rats displayed degeneration in 31% of the area (Bechtold et al. 2006). The circulating concentration of lamotrigine (mean 11.5 μg/ml) in EAE was within the clinical therapeutic range (1–15 μg/ml) for the treatment of epilepsy. Flecainide therapy was also found to be effective in axonal protection in rats with a related disease, namely, experimental autoimmune neuritis, a model of Guillain–Barré syndrome (Bechtold et al. 2005).

To assess whether the preservation of axons observed with morphological methods in EAE with sodium channel blocking therapy was accompanied by preservation of function in the axons, we performed recordings of the compound action potential (CAP) from the dorsal surface of spinal cords of untreated and sodium channel blocker-treated mice and rats with EAE. Our results in both rodent species with EAE demonstrated that treatment with sodium channel blocking agents provides increased numbers of functional axons (greater CAPs) compared with untreated EAE, consistent with the attenuation of axonal loss and improvement of clinical status in these animals.

C57/Bl6 mice with EAE displayed clear improvements in CAP areas when administered phenytoin compared with untreated mice with EAE. The maximal CAP area in untreated mice with EAE was significantly attenuated compared with control mice (0.038 mV ms in EAE vs. 0.16 mV ms in controls; Fig. 8.3a; Lo et al. 2003). In contrast, phenytoin-treated mice with EAE exhibited maximal CAP areas (0.13 mV ms) that were significantly greater than those in untreated mice with EAE (0.038 mV ms) and that approached control values (0.16 mV ms). In addition, assessment of mean conduction velocities demonstrated significant improvement in conduction in phenytoin-treated versus untreated mice with EAE (control: 11.6  ±  2.1 m/s; EAE: 2.3  ±  1.7 m/s; EAE  +  phenytoin: 13.2  ±  1.1 m/s; Lo et al. 2003).

A similar protective effect on axonal conduction was seen in DA rats with chronic-relapsing EAE. Normal DA rats exhibited supramaximal-stimulated monophasic CAP areas of around 50–55 μV ms (Fig. 8.3b; Bechtold et al. 2004, 2006). The CAP areas of DA rats with EAE were significantly reduced by 40–60% (20–30 μV ms). Consistent with the observed preservation of axons achieved with treatment with sodium channel blocking agents, both flecainide (Bechtold et al. 2002, 2004) and lamotrigine (Bechtold et al. 2006) administration to DA rats with EAE significantly increased the CAP areas compared with vehicle-treated DA rats with EAE (Fig. 8.3b). Flecainide treatment commencing either-3 days prior to or 7 days following spinal cord homogenate inoculation significantly improved the CAP areas (-3-day flecainide: 28 μV ms vs. untreated 21 μV ms; 7-day flecainide: 37 μV ms vs. untreated 20 μV ms; Bechtold et al. 2004). Lamotrigine also preserved axonal conduction in DA rats with EAE, increasing the CAP area to 43 μV ms compared with 31 μV ms for untreated rats (Bechtold et al. 2006).

The onset and progression of EAE is associated with infiltration of immune cells into the CNS (Zamvil and Steinman 1990; Gold et al. 2006). Interestingly, sodium channels are expressed by immune cells, including T lymphocytes (Gallin 1991; Lai et al. 2000; Fraser et al. 2004), microglia (Korotzer and Cotman 1992; Norenberg et al. 1994; Eder 1998; Craner et al. 2005; Black et al. 2009) and macrophages (Carrithers et al. 2007). Sodium currents are suggested to contribute to T lymphocyte activation and co-stimulation (Khan and Poisson 1999; Lai et al. 2000), while phagocytosis by microglia and macrophages and migration by microglia in vitro are attenuated by sodium channel blockade (Craner et al. 2005; Carrithers et al. 2007; Black et al. 2009). Figure 8.4 shows a significant reduction in phagocytosis by ­LPS-activated microglia in vitro when treated with either TTX or phenytoin (Black et al. 2009). Carrithers et al. (2007, 2009) showed that Nav1.5 sodium channels in endosomes provide a route for Na⁺ ion efflux that offsets proton influx during endosomal acidification, a late stage of phagocytosis in macrophages, and demonstrated a role of Nav1.6 sodium channels in controlling macrophage motility via regulation of podosome formation. Microglial migration in vitro is also significantly attenuated by exposure to TTX or phenytoin (Fig. 8.5; Black et al. 2009). Similarly, there is evidence for a role of Nav1.6 in microglial migration (Black et al. 2009).

Craner et al. (2005) demonstrated that treatment with phenytoin reduces the inflammatory infiltrate with the CNS by 75%. Black et al. (2007) more recently carried out additional studies on the effect of treatment with sodium channel blocking agents on immune cell infiltration and/or functions in EAE. Spinal cord sections of control C57/Bl6 mice labeled for CD45 (leukocyte common antigen), which provides a marker for most immune cells (Sedgwick et al. 1991), exhibited relatively few CD45⁺ cells (Fig. 8.6). In contrast, untreated mice with EAE mice at 28 days post-MOG injection displayed substantial infiltration of CD45⁺ immune cells in dorsal, ventral, and lateral funiculi (Fig. 8.6). Mice with EAE that were administered phenytoin (Fig. 8.6) or carbamazepine, however, showed substantially reduced numbers of CD45⁺ cells in their spinal cords (Black et al. 2007).

Sodium channel blockade with flecainide also reduced spinal inflammation significantly in DA rats with EAE (Bechtold 2004). Quantification of ED-1 immunoreactivity (marking activated macrophages/microglia) in the spinal cords of the rats revealed that flecainide therapy effected a significant reduction in this measure of inflammation during the relapse stage of EAE (flecainide: 6.8%  ±  3.0 of spinal cord area; vehicle: 11.5%  ±  4.8; p  <  0.05). The expression of the inducible form of nitric oxide synthase during the relapse period of EAE was also significantly reduced in the rats treated with flecainide (Fig 8.7; 0.4%  ±  1.0), when compared with vehicle-treated animals (2.0%  ±  2.2, p  <  0.05). In this context, it is interesting that lamotrigine also reduced lymph node cell proliferation in response to phytohemagglutinin (Makowska et al. 2004). Together with the results obtained in mice, these observations suggest that immunomodulation of immune cell functions by sodium channel blocking agents may contribute to the improvement of the clinical course of EAE by these agents.