Clinical abnormalities in multiple sclerosis (MS) have classically been considered to be caused by demyelination and͞or axonal degeneration; the possibility of molecular changes in neurons, such as the deployment of abnormal repertoires of ion channels that would alter neuronal electrogenic properties, has not been considered. Sensory Neuron-Specific sodium channel SNS displays a depolarized voltage dependence, slower activation and inactivation kinetics, and more rapid recovery from inactivation than classical ''fast'' sodium channels. SNS is selectively expressed in spinal sensory and trigeminal ganglion neurons within the peripheral nervous system and is not expressed within the normal brain. Here we show that sodium channel SNS mRNA and protein, which are not present within the cerebellum of control mice, are expressed within cerebellar Purkinje cells in a mouse model of MS, chronic relapsing experimental allergic encephalomyelitis. We also demonstrate SNS mRNA and protein expression within Purkinje cells from tissue obtained postmortem from patients with MS, but not in control subjects with no neurological disease. These results demonstrate a change in sodium channel expression in neurons within the brain in an animal model of MS and in humans with MS and suggest that abnormal patterns of neuronal ion channel expression may contribute to clinical abnormalities such as ataxia in these disorders.demyelinating diseases ͉ gene transcription ͉ ion channels M ultiple sclerosis (MS) has classically been considered to be a demyelinating disease in which clinical abnormalities are the result of damage to the insulating myelin sheath, which causes axonal conduction block (1, 2). Recently, increased attention has been focused on axonal degeneration in MS, and it has been suggested that axonal loss can produce persistent clinical deficits in MS (3, 4). Demyelination and axonal degeneration, however, do not necessarily underlie all of the clinical abnormalities that are observed in MS. The question of whether there are any changes in neurons that might affect their function in this disorder remains unanswered. In this regard, the possibility of molecular changes within neurons, such as deployment of abnormal repertoires of ion channels that would alter neuronal electrical signaling capability, has not received attention.The generation and transmission of electrical impulse activity in neurons depend on ion f luxes through voltage-gated sodium channels, which consist of large (230 -260 kDa) poreforming ␣-subunits and smaller modulatory ␤-subunits. At least 10 different genes encode distinct sodium channel ␣-subunits with a common overall motif (5) but with different amino acid sequences that underlie different voltage dependences and kinetic properties. The expression of different sodium channel isotypes in different types of neurons endows them with distinct patterns of electroresponsiveness and impulse generation (6 -11).Sensory neuron-specific sodium channel SNS (alternatively termed PN3, SCN1OA, and Nav 1.8) is a tet...

The expression of sodium channels in morphologically and antigenically distinct astrocytes derived from neonatal rat spinal cords was examined at various times in culture. During the course of this study [2-40 days in vitro (DIV)], nine morphologies of glial fibrillary acidic protein (GFAP)+ cells were distinguished: 1a) flat, fibroblast-like; 1b) elongated, with generally few, short processes; 1c) triangular soma with three short, stubby processes; 1d) bipolar with long, slender processes; 1e) bipolar with broad, flared processes; 1f) stellate with radially oriented slender processes extending from a small to moderate-sized soma; 1g) multiple short, stubby processes extending from a moderate-sized soma; 1h) flat, roundish shape with either a smooth edge ("pancake"-like) or numerous very short processes; and 1i) broad, elongated cell body with orthogonally oriented short, spike-like processes. Not all cell types were present at all times in culture. Each type of astrocyte displayed sodium channel immunoreactivity at some time in culture; however, different types of astrocytes exhibited different patterns, over time, of sodium channel staining. Sodium channel immunoreactivity in all astrocyte types was reduced to low levels by 14 DIV, and was not detectable at 40 DIV. Except for types 1b and 1e, A2B5 staining was present on all astrocyte morphologies at some time in culture, and was generally attenuated with longer times in vitro; in contrast to cultures derived from neonatal rat optic nerve, A2B5 staining does not distinguish unequivocally between the various classes of morphologically different astrocytes derived from spinal cord. O4 immunoreactivity was consistently observed only on bipolar, elongated, and process-bearing astrocytes, though not all process-bearing astrocytes were O4+. These results demonstrate that astrocytes derived from neonatal spinal cord are morphologically and antigenically heterogeneous. Moreover, while spinal cord astrocytes express sodium channels, these astrocytes exhibit a time-course of channel expression that is different from astrocytes derived from several other CNS regions where sodium channel staining is maintained even for extended times in culture, suggesting a regional modulation of astrocyte function.

Immunocytochemical and electrophysiological methods were used to examine the effect of retinal ablation on the expression of sodium channels within optic nerve astrocytes in situ and in vitro. Enucleation was performed at postnatal day 3 (P3), and electron microscopy of the enucleated optic nerves at P28-P40 revealed complete degeneration of retinal ganglion axons, resulting in optic nerves composed predominantly of astrocytes. In contrast to control (non-enucleated) optic nerve astrocytes, which exhibited distinct sodium channel immunoreactivity following immunostaining with antibody 7493, the astrocytes in enucleated optic nerves did not display sodium channel immunoreactivity in situ. Cultures obtained from enucleated optic nerves consisted principally (greater than 90%) of glial fibrillary acidic protein (GFAP)+/A2B5- ("type-1") astrocytes, as determined by indirect immunofluorescence; GFAP+/A2B5+ ("type-2") astrocytes were not present, nor were GFAP-/A2B5+ (O-2A) progenitor cells. Sodium channel immunoreactivity was not present in GFAP+/A2B5- astrocytes obtained from enucleated optic nerves; in contrast, GFAP+/A2B5- astrocytes from control optic nerves exhibited 7493 immunostaining for the first 4-6 days in culture. Sodium current expression, studied using whole-cell patch-clamp recording, was attenuated in cultured astrocytes derived from enucleated optic nerves. Whereas 39 of 50 type-1 astrocytes cultured from intact optic nerves showed measurable sodium currents at 1-7 days in vitro, sodium currents were present in only 6 of 38 astrocytes cultured from enucleated optic nerves. Mean sodium current densities in astrocytes from the enucleated optic nerves (0.66 +/- 0.3 pA/pF) were significantly smaller than in astrocytes from control optic nerves (7.15 +/- 1.1 pA/pF). The h infinity-curves of sodium currents were similar in A2B5- astrocytes from enucleated and control rat optic nerves. These results suggest that there is neuronal modulation of sodium channel expression in type-1 optic nerve astrocytes, and that, following chronic loss of axonal association in vivo, sodium channel expression is down-regulated in this population of optic nerve astrocytes.

No abstract