Evidence for the Production of NO in MS

Many lines of evidence show that the production of NO is significantly raised in MS, and within MS lesions in particular. For example, the nitrite/nitrate concentration in the urine and blood (the ultimate fate of most of the NO pro duced within the body is in the form of nitrite and nitrate) is significantly increased in patients with demyelinating disease (Giovannoni, 1998; Giovannoni et al., 1997, 1999). Furthermore, nitrotyrosine, regarded as a relatively stable marker of peroxynitrite production, is very significantly raised (approximately sixfold, P < 0.0001) in the serum of patients with MS compared with controls, particularly in patients with chronic progressive disease (Zabaleta et al.,

1998). More particular to the CNS are studies that have examined nitrite and nitrate concentrations within the cerebrospinal fluid. Although two early studies involving only a small number of patients did not find raised NOx in the cerebrospinal fluid (Ikeda et al., 1995; De Bustos et al.,

1999), most studies have found significant increases (Cross et al., 1998; Brundin et al., 1999; Drulovic et al., 2001; Giovannoni, 1998; Johnson et al., 1995; Peltola et al., 2001; Yamashita et al., 1997; Ali et al., 1996; Svenningsson et al., 1999; Speciale et al., 2000), and some concentrations that correlated with clinical disease activity (e.g., Danilov et al., 2003).

The evidence for the local production of NO in MS lesions is compelling, in that there is prominent expression of the inducible form of NOS. iNOS is not normally present within the CNS, but iNOS mRNA was "abundantly expressed" in a biopsy taken from a patient with Marburg's-type MS within 33 days of the onset of disseminated symptoms (Bitsch et al., 1999), indicating that NO production is likely to be present from very early in the disease process. iNOS mRNA was also found in all seven MS brains examined in another study (Bagasra et al., 1995), and it was the dominant isoform of NOS found in another (Broholm et al., 2004) (see also Bo et al., 1994). Expression of the iNOS protein has been demonstrated in acute lesions using immunohisto-chemical techniques (Oleszak et al., 1998; Liu et al., 2001; Broholm et al., 2004), but expression diminishes or is absent in chronic (noninflammatory) lesions (De Groot et al., 1997; Liu et al., 2001; Oleszak et al., 1998; Broholm et al., 2004). Expression has been described in cells of the macrophage/microglial lineage (De Groot et al., 1997; Oleszak et al., 1998; Hooper et al., 1997; Broholm et al., 2004) and in reactive astrocytes (Liu et al., 2001; Broholm et al., 2004). iNOS is prominently and widely expressed in white and gray matter that appears normal macroscopically, and expression was moderate to high in 67% of samples lacking the histological characteristics of MS plaques (Broholm et al., 2004). iNOS mRNA (Koprowski et al., 1993) and protein (van Dam et al., 1995; Cross et al., 1997) are also expressed within the inflammatory lesions of animals with experimental autoimmune encephalomyelitis (EAE), where the level of expression coincides temporally and quantitatively with the severity of the neurological deficit (Koprowski et al., 1993; Okuda et al., 1995). The eNOS isoform is also highly expressed in intraparenchymal vascular endothelial cells in MS tissue, unlike the normal brain (Broholm et al., 2004). At the pathological concentrations of NO likely to be present in MS lesions, the enzyme sGC is likely to be near maximally activated, in which case the advantages of being able to use this enzyme pathway to exert subtle effects on cellular metabolism will be denied.

Peroxynitrite has a half-life in tissue measured in milliseconds, so it cannot be directly detected histologically, but persistent evidence of its ephemeral presence is provided by the finding of nitrotyrosine in many active MS lesions, and labeling can be intense (Liu et al., 2001), with labeling of hypertrophic astrocytes (Cross et al., 1998; Oleszak et al., 1998), and, especially, iNOS positive macrophages/microglial cells (Cross et al., 1998; Oleszak et al., 1998; Bagasra et al., 1995; Hooper et al., 1997).

Of particular interest is the observation that iNOS-positive astrocytes and, to a lesser extent, macrophage/ microglial cells were diffusely scattered throughout the cerebral white matter in a patient with a rapid primary progressive course (Broholm et al., 2004), and strongly nNOSpositive astrocytes were also present. Furthermore, labeling was prominent and widespread in microglia in patients with primary progressive MS (Hans Lassmann, personal communication). These findings may be relevant to understanding the cause of the "smoldering" level of axonal degeneration that appears to be responsible for the progressive accumula tion of neurological deficit in this form of the disease. The role of NO in MS has been reviewed in more detail elsewhere (Smith et al., 1999; Santiago et al., 1998; Smith and Lassmann, 2002).

IV. NO Impairs Axonal Conduction

The first electrophysiological effect of NO on axons to be described was the block of conduction, and NO is very effective in this respect (Fig. 1) (Redford et al., 1997; Shrager et al.,1998; Garthwaiteet al., 2002; Kapoor et al., 2003). The block is imposed within a few minutes of exposure, maintained for the duration of exposure (at least for several hours) and relieved within minutes of washing (Redford et al., 1997; Kapoor et al., 2003). The effects of NO appear to be fully reversible, and no apparent lasting consequences are revealed during at least 10 hours after exposure (Kapoor et al., 2003). In this sense, NO appears to act like a local anesthetic.

There is not yet any proof that NO is involved in causing neurological deficits in patients, but several observations are consistent with this possibility. Perhaps the clearest has arisen from some surprising observations in patients treated with an experimental MS therapy, the administration of the antibody CAMPATH-1H (Moreau et al., 1996). Indeed, it

Figure 1 A series of records showing that NO causes a reversible block of conduction in central demyelinated axons when injected into an experimental lesion situated in the rat dorsal columns (inset). The records are plotted with three-dimensional perspective, with the earliest records shown at the front; the records were taken every 2 minutes and show about 5 hours of recording time. All the axons contributing to the compound action potentials were known to be affected by the demyelinating lesion on the basis of other recordings (not shown) and confirmed by histological examination. Conduction along the axons was stable until NO was applied by the injection of the NO donor spermine NONOate, when conduction in approximately half the axons was promptly blocked. The block reversed gradually over the next 30 minutes as the production of NO diminished. A second injection of a lower concentration of donor reinstated the block, in a smaller number of axons. (Modified from Redford et al., 1997.)

Figure 1 A series of records showing that NO causes a reversible block of conduction in central demyelinated axons when injected into an experimental lesion situated in the rat dorsal columns (inset). The records are plotted with three-dimensional perspective, with the earliest records shown at the front; the records were taken every 2 minutes and show about 5 hours of recording time. All the axons contributing to the compound action potentials were known to be affected by the demyelinating lesion on the basis of other recordings (not shown) and confirmed by histological examination. Conduction along the axons was stable until NO was applied by the injection of the NO donor spermine NONOate, when conduction in approximately half the axons was promptly blocked. The block reversed gradually over the next 30 minutes as the production of NO diminished. A second injection of a lower concentration of donor reinstated the block, in a smaller number of axons. (Modified from Redford et al., 1997.)

was a search for a mechanism to explain these observations that prompted the initial experiments with NO in the author's laboratory. The CAMPATH-1H antibody is directed against the CD52 antigen expressed on all lymphocytes, and its administration was intended to deplete patients of these lymphocytes as they are believed to be responsible for disease activity. Although the therapy had been administered with no ill effects to individuals without MS, when administered to patients with MS, the antibody quite promptly (2 to 4 hours) elicited a surprising reappearance of signs and symptoms that had previously been expressed, but from which recovery had occurred. New symptoms did not appear, indicating that the antibody administration had somehow reimposed conduction block in axons that had previously been damaged by the disease process. The reenactment of the symptoms occurred in conjunction with a surge in the level of circulating pro-inflammatory cytokines, but the initial suspicion that these were responsible for the exacerbation was not confirmed by experimentation (unpublished observations). Realization that the pro-inflammatory cytokines would result in the appearance of iNOS, however, directed attention to the potential role of NO and the findings described previously. In this regard, it is notable that demyelinated axons are particularly vulnerable to the effects of NO (Redford et al., 1997), providing a plausible explanation for the observations upon CAMPATH-1H administration.

Apart from the CAMPATH observations, other findings also indicate that inflammation may be capable of imposing severe conduction and neurological deficits in patients with inflammatory demyelinating disease (Youl et al., 1991; Coles et al., 1999), even in the absence of demyeli-nation (Bitsch et al., 1999). Whether NO is involved in causing the deficits remains unproven, although it may be worth noting that iNOS was abundantly expressed in the inflammatory lesions believed to cause neurological deficits (Bitsch et al., 1999).

The mechanisms underlying NO-mediated conduction block are not known. However, the discoveries that NO can impair mitochondrial respiration, affect sodium channel properties, and cause axonal depolarization have provided several reasonable candidates to explain the block of conduction. These effects of NO are discussed later.

It is interesting to speculate on whether NO may play a role in Uhthoff's phenomenon (Smith and McDonald, 1999) (see Chapter 6), namely the tendency for the expression of some symptoms to be modulated by changes in body temperature. Body warming is deleterious, and cooling beneficial, such that in extreme cases a patient might experience an improvement in vision upon drinking a glass of cold water (Hopper et al., 1972; McDonald, 1986). Such effects arise from the modulation of axonal conduction block in affected pathways, and although the effect is conventionally explained by reference to the effects of temperature on action potential duration (Smith and McDonald, 1999) (see

Chapter 6), NO might play a role (Beenakker et al., 2001). Patients with MS who are temperature-sensitive and who wore a cooling garment achieved a clinical improvement, as expected, but the effect was associated in this study with a substantial reduction in leukocytic NO production. Thus, although changes in action potential duration can explain the clinical observations, changes in the rate of NO production may also play a role.

V. NO and Axonal Depolarization

One effect of NO that could contribute to the conduction abnormalities described previously is that it causes a prompt depolarization of axons and neurons, respectively, when applied to rat optic nerve (Garthwaite et al., 2002) and neurons in the hypothalamic paraventricular region (Bains and Ferguson, 1997). In the neurons at least, the depolarization could be reproduced by a cGMP analog (Bains and Ferguson, 1997). In the optic nerve, the depolarization commenced almost immediately on exposure to NO and consisted of two stages (Fig. 2). The initial stage resulted in only

B

Figure 2 Depolarization of optic nerve axons in response to exposure to the NO donor PAPA NONOate (solid bar). (A) A small, prompt depolarization was followed by a larger depolarization that reversed rapidly on washout. (B) The sodium channel blocking agent tetrodotoxin (TTX) (open bar) caused a small hyperpolarization and abolished the delayed depolarization. (Reproduced from Garthwaite et al., 2002.)

a few millivolts of depolarization, but it progressed to a second, larger depolarization after 7 to 10 minutes. Depolarization was reversed on washing and was followed by a period of hyperpolarization. Although conduction persisted during the first stage of depolarization, it was blocked during the second. The cause of the first stage of depolarization is not certain, but the second stage appears to involve sodium entry via axolemmal sodium channels in conjunction with a reduction in adenosine triphosphate (ATP). The concentration of ATP fell by 44% after 10 minutes of NO exposure (Garthwaite et al., 2002). A potential cause of the reduction in ATP is described in the next section.

0 0

Post a comment