Upregulated Expression of Nav18 in MS and Its Models

It now appears likely that, in addition to altered expression of Nav1.2 and Nav1.6 along demyelinated axons in MS, there is altered expression of at least one other sodium channel isoform within neuronal cell bodies that can give rise to distorted patterns of impulse generation in these cells. This line of research originated with a study by Black et al. (1999b) who examined neuronal sodium channel expression in the Taiep rat, a mutant model in which myelin initially ensheaths CNS axons in a normal manner but subsequently degenerates as a result of an abnormality of oligodendrocytes (Duncan et al., 1992). This study focused on expression of the Nav1.8 sodium channel [a slowly-inactivating tetrodotoxin (TTX)-resistant sodium channel which was originally termed SNS, Sensory Neuron Specific, since it is normally detectable only in DRG neurons and trigeminal neurons; (Akopian et al., 1996; Sangameswaren et al., 1996] because the transcription of Nav1.8 within these neurons changes markedly after axon transection (Dib-Hajj et al., 1996). In situ hybridization of the Taiep brain showed that, following loss of myelin within the cerebellar white matter, there is markedly enhanced expression of Nav1.8 mRNA within Purkinje cells. Immunocytochemical studies showed that the up-regulation of Nav1.8 mRNA is accompanied by the production of Nav1.8 protein in Purkinje cells (Black et al., 1999b).

A more recent study (Black et al., 2000) used in situ hybridization and immunohistochemistry to examine the expression of Nav1.8 in the brains of mice with chronic-relapsing EAE (CR-EAE) and in postmortem brain tissue from humans with MS. As shown in Fig. 8, in situ hybridization and immunocytochemistry demonstrated significantly increased expression of Nav1.8 mRNA and protein within Purkinje cells in CR-EAE (Black et al., 2000). A global up-regulation of all sodium channels could not account for up-regulated expression of Nav1.8 because expression of Nav1.9, another TTX-resistant sodium channel that is normally expressed preferentially, like Nav1.8, in DRG and trigeminal ganglion neurons, was not upregulated.

The increased expression of Nav1.8 within Purkinje cells is not limited to animal models of MS. Study of postmortem human brain tissue (Black et al., 2000) has demonstrated up-regulation of Nav1.8 mRNA and protein within Purkinje cells in patients with disabling progressive MS with cerebellar deficits on neurological examination (Fig. 9).

Figure 8 Sodium channel Nav1.8 (originally termed SNS, Sensory Neuron Specific) is abnormally expressed within cerebellar Purkinje neurons in mice with EAE. A-D: in situ hybridization showing Nav1.8 mRNA within Purkinje cells in EAE (A, B) but not in healthy control mice (C) or after hybridization with sense riboprobes (D). E-G: Immunostaining with Nav1.8-specific antibodies, showing upregulated expression of Na 1.8 protein in EAE (E) compared to controls (F, bright field; G, Nomarski; image). (A) 120; (B-D)x200; (E-G) x220. (From Black et al., 2000.)

Figure 8 Sodium channel Nav1.8 (originally termed SNS, Sensory Neuron Specific) is abnormally expressed within cerebellar Purkinje neurons in mice with EAE. A-D: in situ hybridization showing Nav1.8 mRNA within Purkinje cells in EAE (A, B) but not in healthy control mice (C) or after hybridization with sense riboprobes (D). E-G: Immunostaining with Nav1.8-specific antibodies, showing upregulated expression of Na 1.8 protein in EAE (E) compared to controls (F, bright field; G, Nomarski; image). (A) 120; (B-D)x200; (E-G) x220. (From Black et al., 2000.)

Figure 9 Expression of the Sensory Neuron Specific (SNS) sodium channel Nav1.8 is upregulated in cerebellar Purkinje neurons in patients with MS. In situ hybridization demonstrates increased Nav1.8 mRNA in Purkinje cells from two MS patients obtained at postmortem (A, B), compared to controls without neurological disease (C). No signal is present following hybridization with sense riboprobe (D). Immunocytochemistry with Nav1.8-specific antibodies demonstrates upregulation of Nav1.8 channel protein in Purkinje cells from two MS patients (E, F) compared to controls (G; arrowhead indicates Purkinje cell). (A) x120, inset x280; (B, C, D) x165; (E, F, G) x175. (From Black et al., 2000.)

Figure 9 Expression of the Sensory Neuron Specific (SNS) sodium channel Nav1.8 is upregulated in cerebellar Purkinje neurons in patients with MS. In situ hybridization demonstrates increased Nav1.8 mRNA in Purkinje cells from two MS patients obtained at postmortem (A, B), compared to controls without neurological disease (C). No signal is present following hybridization with sense riboprobe (D). Immunocytochemistry with Nav1.8-specific antibodies demonstrates upregulation of Nav1.8 channel protein in Purkinje cells from two MS patients (E, F) compared to controls (G; arrowhead indicates Purkinje cell). (A) x120, inset x280; (B, C, D) x165; (E, F, G) x175. (From Black et al., 2000.)

These observations in two animal models of MS, and in humans with MS, demonstrate that expression of the Nav1.8 sodium channel is up-regulated at the transcriptional level in Purkinje neurons, producing Nav1.8 sodium channel protein which is not normally present in these cells.

The trigger of up-regulated expression of Nav1.8 is currently under study. The available evidence (Damarjian et al., 2004) demonstrates up-regulated expression of TrkA and p75 receptors for nerve growth factor (NGF, which is known to be present at elevated levels within the CNS in EAE and MS; Laudiero et al., 1991; De Simone et al., 1996), within Nav1.8-immunopositive Purkinje neurons in EAE, and suggests that activation of p75, in particular, may trigger the expression of Nav 1.8.

The Nav1.8 sodium channel, like other sodium channels, binds to accessory proteins and at least one of these is up-regulated in a coordinate manner with Nav1.8 in EAE and MS (Craner et al., 2003b). AnnexinII/p11 binds to the N-terminus of Nav1.8 and facilitates the insertion of functional channels into the neuronal cell membrane (Okuse et al., 2002). As shown in Fig. 10 the expression of annexinII/p11 is up-regulated within Purkinje cells, and it is co-localized with Nav1.8, both in EAE and in MS (Craner et al., 2003b).

While a full mapping of the CNS has not yet been completed, other neuronal cell types may show similar abnormalities in MS and its models. Expression of Nav1.8 and annexin11/p11 is also up-regulated in retinal ganglion cells

(which give rise to axons that travel in the optic nerve) in EAE (Craner, Black, and Waxman, unpublished). Nav1.8 protein is not detectable, however, along Purkinje cell axons within the cerebellar white matter (Craner, Black and Waxman, unpublished) or retinal ganglion cell axons within the optic nerve (Craner et al., 2003a).

Although the functional consequences of up-regulated Nav1.8 expression in Purkinje cells have not yet been definitively determined in humans, several studies suggest that it interferes with neuronal signaling. In theory it is possible is that Nav1.8 up-regulation might be an adaptive change, which could support the restoration of action potential conduction along demyelinated Purkinje cell axons. Renganathan et al. (2001) demonstrated that Nav1.8 channels produce a substantial fraction of the inward current that underlies the depolarizing phase of the action potential in the DRG neurons in which these channels are normally present. Thus, if Nav1.8 channels were inserted into the demyeli-nated axon membrane, they might be expected to contribute to restoration of action potential conduction in demyelinated axons. However, while Nav1.8 is up-regulated in the cell bodies of Purkinje cells and retinal ganglion cells in EAE, Nav1.8 protein has not been detected along demyelinated Purkinje cell or optic nerve axons, in contrast to Nav1.2 and Nav1.6 protein which, as shown in Fig. 2, can be clearly detected (Craner et al., 2003a). An adaptive role of Nav1.8 in restoration of conduction along demyelinated axons thus seems unlikely.

Figure 10 AnnexinII/p11, which facilitates insertion of functional Nav1.8 channels into the cell membrane, is upregulated and co-expressed with Nav1.8 in Purkinje cells in EAE and MS. (Left panels) Nav1.8 (A) and annexinII/p11 (C) are upregulated in Purkinje neurons in EAE as compared to control (B and D, respectively). Note the co-localization of annexinII/P11 in Nav1.8 in the same neurons (compare panels A and C). AnnexinII/p11 immunostaining extends along the proximal portion of the dendritic tree in EAE (E) but not in control (F). (Right) Nav1.8 is upregulated in neurons in postmortem MS tissue (G) vs. control (H). Upregulated expression of annexinII/p11 is shown for two MS cases (I, K) in comparison to controls (J, L). (Modified from Craner et al., 2003b.)

Figure 10 AnnexinII/p11, which facilitates insertion of functional Nav1.8 channels into the cell membrane, is upregulated and co-expressed with Nav1.8 in Purkinje cells in EAE and MS. (Left panels) Nav1.8 (A) and annexinII/p11 (C) are upregulated in Purkinje neurons in EAE as compared to control (B and D, respectively). Note the co-localization of annexinII/P11 in Nav1.8 in the same neurons (compare panels A and C). AnnexinII/p11 immunostaining extends along the proximal portion of the dendritic tree in EAE (E) but not in control (F). (Right) Nav1.8 is upregulated in neurons in postmortem MS tissue (G) vs. control (H). Upregulated expression of annexinII/p11 is shown for two MS cases (I, K) in comparison to controls (J, L). (Modified from Craner et al., 2003b.)

A number of observations strongly suggest that upregulated expression of Nav1.8 in EAE and MS is maladaptive. Electrogenesis within Purkinje cells depends, in part, on sodium channel activity (Llinas and Sugimori, 1980; Stuart and Hausser, 1994; Raman and Bean, 1997) and mutations of sodium channels that are expressed in Purkinje cells can produce substantial changes in the patterns of impulse generation in these cells that can result in clinical signs of cere-bellar dysfunction such as ataxia (Raman et al., 1997; Kohrman et al., 1996). Nav1.8 displays unique physiological properties including a depolarized voltage-dependence of inactivation, slow development of inactivation (Akopian et al., 1996; Sangameswaren et al., 1996) and rapid recovery from inactivation (Elliott and Elliott, 1993; Dib-Hajj et al., 1997). As a result of these properties Nav1.8 channels are available over a wider range of dynamic activity and membrane potential than other sodium channels (S child and Kunze, 1997), so that cells expressing Nav1.8 are predicted to be more slowly adapting than cells lacking Nav1.8 (Elliott and Elliott, 1993).

Electrophysiological observations confirm this prediction and demonstrate that Nav1.8 expression can, in fact, substantially alter the temporal pattern of electrical activity. Renganathan et al. (2001) used current-clamp and voltage-clamp recording to examine the pattern of action potential generation in DRG neurons from transgenic Nav1.8 -/mice in which functional Nav1.8 channels are absent (Akopian et al., 1999) and compared it with electrogenesis in Nav1.8 +/+ neurons. This study showed that the presence of Nav1.8 channels within DRG neurons markedly influences both the configuration of the action potentials (a change that can effect activation of N-type channels and thus transmitter release [Scroggs and Fox, 1992] if Nav1.8 is deployed to axon terminals) and the temporal pattern of firing in response to depolarizing stimuli. Consistent with the suggestion that cells expressing Nav1.8 should be slowly adapting , Nav1.8 +/+ DRG neurons produce sustained pacemaker-like trains of action potentials in response to depolarizing stimuli, which are not present within Nav1.8 -/neurons (Renganathan et al., 2001).

In a more recent study, Renganathan et al. (2003) trans-fected Purkinje cells in vitro with Nav1.8. Voltage-clamp recordings demonstrated that functional Nav1.8 channels can be expressed within Purkinje cells where they produce slowly inactivating TTX-resistant Nav1.8 currents at physiological levels (with current amplitudes of approximately 4 to 5 nA, similar to those observed within DRG neurons where Nav1.8 is normally expressed). Current-clamp recordings demonstrated that expression of Nav1.8 produces several physiological changes within Purkinje neurons. First, expression of Nav1.8 increases action potential amplitude and duration (compare Fig. 11B and 11A). Second, Nav1.8 expression decreases the proportion of action potentials that display the conglomerate configuration characteristically observed in Purkinje cells (15% in Nav1.8-transfected neurons compared to 62% in controls) and the number of spikes per conglomerate action potential, which was 2.13 in Nav1.8-transfected neurons compared to 3.38 in controls

Figure I I Expression of Nav1.8 alters action potential electrogenesis within Purkinje cells. These current clamp recordings show spontaneous action potentials recorded in Purkinje neurons two days after biolistic expression of GFP that provided a marker of transfection (without Nav1.8) (Aj - A4) or of Nav1.8/GFP (B; - B3). Aj-A4: Action potentials in control neurons lacking Nav1.8 (Aj - A4) show little if any overshoot (dotted lines indicate 0 mV) and tend to be conglomerate (62%; A2 -A4). Bj-B3: Action potentials in neurons expressing Nav1.8 display larger overshoot. Conglomerate action potentials are less common after expression of Nav1.8 (15%) and, when present, tend to consist of doublets (B2), only rarely consisting of more than 2 spikes (B3). Time calibration in A2 applies to A;; time calibration in B; applies to Bj-B3. The mV calibration in A2 applies to all panels. A5, B4: Percentage of action potentials that were single, or conglomerate with 2, 3, 4, 5-8 or more spikes, in Purkinje cells lacking Nav1.8 and with Nav1.8, respectively. There is a lower percentage of conglomerate action potentials and a smaller number of spikes per conglomerate action potential in Purkinje neurons expressing Nav1.8. (Modified from Renganathan et al., 2003.)

(compare Fig. 11B and 11A). Third, Nav1.8 expression supports sustained, pacemaker-like impulse trains in response to depolarization, which are not seen in the absence of Nav1.8 (Fig. 12). More recently, Saab et al. (2004) observed similar changes in the firing patterns of Purkinje cells in vivo in mice with EAE (Fig. 13). The results of these studies indicate that expression of Nav1.8 within Purkinje neurons in vitro distorts the pattern of activity of these cells, a change that could interfere with cerebellar function, and show that similar changes occur in vivo in EAE. Whether the upregu-lated Nav1.8 expression and resultant change in excitability lead to degeneration of Purkinje cells is not known.

Biopsy of the cerebellum is only infrequently performed in MS, and it is therefore difficult to directly establish whether similar physiological changes occur in Purkinje cells in humans with MS. Support for this suggestion nonetheless is provided by the observations on postmortem MS tissue described previously, and by several observations

Figure 12 Purkinje neurons transfected with Nav1.8 show sustained repetitive firing, not present in the absence of Nav1.8, on injection of depolarizing current. (A) Control Purkinje neuron lacking Nav1.8 produces a conglomerate action potential consisting of five spikes, but no sustained firing, in response to a sustained depolarizing stimulus (80 pA, 1 second). (B) Purkinje neuron expressing Nav1.8 produces larger-amplitude action potentials and shows sustained pacemaker-like activity in response to identical stimulus. The current pulse protocol is shown in C. (Modified from Renganathan et al., 2003.)

Figure 12 Purkinje neurons transfected with Nav1.8 show sustained repetitive firing, not present in the absence of Nav1.8, on injection of depolarizing current. (A) Control Purkinje neuron lacking Nav1.8 produces a conglomerate action potential consisting of five spikes, but no sustained firing, in response to a sustained depolarizing stimulus (80 pA, 1 second). (B) Purkinje neuron expressing Nav1.8 produces larger-amplitude action potentials and shows sustained pacemaker-like activity in response to identical stimulus. The current pulse protocol is shown in C. (Modified from Renganathan et al., 2003.)

from the clinic. First, most clinicians have observed occasional patients of MS with cerebellar deficits on clinical examination, but without apparent cerebellar lesions in neu-roimaging studies; this observation is consistent with a contribution of molecular changes, too subtle to be detected by currently available imaging techniques, to cerebellar dysfunction. Second, cerebellar signs in MS are often non-remitting even if developing early in the course of the disease (see, for example, Matthews et al., 1991), and this is not readily explained by inflammation or demyelination per se. Craner et al. (2003c) observed that, in relapsing-remitting EAE, the level of Na 1.8 expression within Purkinje cells is correlated with the severity of nonremitting deficits (which include cerebellar dysfunction). Finally, paroxysmal ataxia has been well described in MS and, in many cases, responds to treatment with sodium channel blockers such as carbamazepine (Andermann et al., 1959; Espir et al., 1966). These sudden and brief episodes of ataxia in MS are similar to the paroxysmal attacks that occur in the episodic ataxias (which are known to be inherited chan-nelopathies) and are not easily explained by demyelination or axonal degeneration. The therapeutic response to carba-mazepine suggests that sodium channels may participate in their pathogenesis.

The observation of upregulated Nav1.8 expression within retinal ganglion neurons in EAE (Craner, Black, and Waxman, unpublished) may also have implications for MS, as sodium channels contribute significantly to electrogenesis in retinal ganglion cells (Lipton and Tauck, 1986). The electroretino-gram provides some evidence for retinal dysfunction in MS, suggestive of an abnormality of retinal neurons even in patients without a history of optic neuritis (Plant et al., 1986; see also Chapter 16). Study of suitably preserved human retinal tissue should make it possible to determine whether Na 1.8 is in fact expressed in retinal ganglion neurons in MS.

The hypothesis that aberrant expression of Nav1.8 channels contributes to symptomatology in MS would be strengthened if it could be shown that Nav1.8 blocking drugs ameliorate ataxia or other clinical abnormalities. Nav1.8-specific channel blocking drugs are not yet available, but the high level of expression of Nav1.8 within nociceptive DRG neurons has made this channel an attractive target for drug screening. When Nav1.8-specific blocking drugs are developed, it will be possible to examine the effect of these drugs on animal models of MS such as EAE and, if indicated, in humans with MS.

VIII. Is Expression of Other Channels Altered in MS?

Although this chapter has focused on sodium channels, expression of other types of channels may also be altered in MS. Increased expression of K 1.1 and K 1.2 potassium

Figure 13 The firing patterns of Purkinje cells within their native cerebellar environment in vivo are altered in EAE. (A) Conglomerate action potentials in Purkinje cells from control (upper traces) and EAE (lower traces). Superimposed conglomerate action potentials show regularity of secondary spikes in controls and irregularity in EAE (note the irregularity in latencies of secondary spikes for each cell; downward arrows on right). (B, left panel) Percentage of Purkinje cells with irregular temporal organization of conglomerate action potentials is higher in EAE (*P < 0.001). (B, right panel) Average number of secondary spikes per conglomerate action potential is lower in EAE (*P < 0.005). (C, D, E) Abnormal high-frequency bursting in Purkinje cells in EAE. (C) Purkinje cells in EAE produced brief bursts (Sb) of repetitive single spikes consisting of doublets, triplets, or quadruplets, not seen in controls, interspersed with isolated single spikes (SS) and conglomerate action potentials. (C, top). Second and third rows in C show a continuous recording; isolated action potentials, plotted using template matching techniques, were used to plot instantaneous frequency (C—E, bottom). In another Purkinje cell from a mouse with EAE (D), sustained high-frequency SS bursts (up to 60 Hz) lasting for >10 seconds are followed by an extended period (more than 3 min) of no recorded SS activity (conglomerate action potentials are indicated by dots; individual conglomerate action potentials and SS are shown to the right). (E) These activity patterns were not observed in Purkinje cells recorded in control mice. (Modified from Saab et al., 2004.)

Figure 13 The firing patterns of Purkinje cells within their native cerebellar environment in vivo are altered in EAE. (A) Conglomerate action potentials in Purkinje cells from control (upper traces) and EAE (lower traces). Superimposed conglomerate action potentials show regularity of secondary spikes in controls and irregularity in EAE (note the irregularity in latencies of secondary spikes for each cell; downward arrows on right). (B, left panel) Percentage of Purkinje cells with irregular temporal organization of conglomerate action potentials is higher in EAE (*P < 0.001). (B, right panel) Average number of secondary spikes per conglomerate action potential is lower in EAE (*P < 0.005). (C, D, E) Abnormal high-frequency bursting in Purkinje cells in EAE. (C) Purkinje cells in EAE produced brief bursts (Sb) of repetitive single spikes consisting of doublets, triplets, or quadruplets, not seen in controls, interspersed with isolated single spikes (SS) and conglomerate action potentials. (C, top). Second and third rows in C show a continuous recording; isolated action potentials, plotted using template matching techniques, were used to plot instantaneous frequency (C—E, bottom). In another Purkinje cell from a mouse with EAE (D), sustained high-frequency SS bursts (up to 60 Hz) lasting for >10 seconds are followed by an extended period (more than 3 min) of no recorded SS activity (conglomerate action potentials are indicated by dots; individual conglomerate action potentials and SS are shown to the right). (E) These activity patterns were not observed in Purkinje cells recorded in control mice. (Modified from Saab et al., 2004.)

channels (Wang et al., 1999), for example, has been reported along dysmyelinated axons within the brain of the Shiverer mouse. A recent study has suggested the ectopic expression of the a1B pore-forming subunit of N-type calcium channels within dystrophic (presumably degenerating) axons in EAE and MS (Kornek et al., 2001). Another study suggests the presence of mGlu1a-glutamate receptor immunoreactivity along axons in MS (Geurts et al., 2003), although, again, it is not clear whether the receptors are inserted into the axon membrane. If they are, there is the possibility of activation of signaling cascades that activate injurious intra-axonal processes. Ouardouz et al. (2003) demonstrated, in rat dorsal column axons, that ischemia can trigger a rise in intra-axonal calcium due to ryanodine-sensitive intracellular calcium release that is triggered by L-type calcium channels. This observation raises the possibility that activation of ectopic ion channels may trigger release of calcium from intracellular stores within axons. In summary, there may be changes in expression of multiple channel and receptor mRNAs within neurons whose axons have been demyeli-

nated or subjected to inflammation, and it will be important to determine the full ensemble of channels and receptors that are expressed and their precise pattern of distribution.

0 0

Post a comment