JN Journal of Applied Physiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

J Neurophysiol 79: 529-536, 1998;
0022-3077/98 $5.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Shrager, P.
Right arrow Articles by Mattson, D.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Shrager, P.
Right arrow Articles by Mattson, D.

The Journal of Neurophysiology Vol. 79 No. 2 February 1998, pp. 529-536
Copyright ©1998 by the American Physiological Society

Nerve Conduction Block by Nitric Oxide That Is Mediated by the Axonal Environment

Peter Shrager1, Andrew W. Custer2, Katia Kazarinova3, Matthew N. Rasband3, and David Mattson4

1 Department of Neurobiology and Anatomy; 2 Neuroscience Graduate Program; 3 Department of Biophysics; and 4 Department of Neurology, University of Rochester Medical Center, Rochester, New York 14642

    ABSTRACT
Abstract
Introduction
Methods
Results
Discussion
References

Shrager, Peter, Andrew W. Custer, Katia Kazarinova, Matthew N. Rasband, and David Mattson. Nerve conduction block by nitric oxide that is mediated by the axonal environment. J. Neurophysiol. 79: 529-536, 1998. Conduction in rat peripheral nerve has been monitored following the stimulated release of nitric oxide (NO) from diethylamine-NONOate (DEA-NONOate). Branches of the sciatic nerve were dissected, but left otherwise intact, and propagating signals recorded externally. At levels consistent with inflammation, NO exposure resulted in a complete loss of the compound action potential. Conduction was fully restored on removal of the drug. Most notably, this loss of excitability was dependent on the axonal environment. Removal of the connective tissue sheaths surrounding the nerve bundle, a process that normally enhances drug action, prevented block of signal propagation by nitric oxide. The epineurium seemed not to be required, and the decreased susceptibility to NO appeared to be correlated with a gradual loss of a component of the endoneurium that surrounds individual fibers. Tested on the rat vagus nerve, NO eliminated action potentials in both myelinated and unmyelinated fibers. One chemical mechanism that is consistent with the reversibility of block and the observed lack of effect of 8-Br-cGMP on conduction is the formation of a nitrosothiol through reaction of NO with a sulfhydryl group. In contrast to DEA-NONOate, S-nitrosocysteine, which can both transfer nitrosonium cation (NO+) to another thiol and also release nitric oxide, was effective on both intact and desheathed preparations. It has previously been demonstrated that chemical modification of invertebrate axons by sulfhydryl-reactive compounds induces a slow inactivation of Na+ channels. Nitric oxide block of axonal conduction may contribute to clinical deficits in inflammatory diseases of the nervous system.

    INTRODUCTION
Abstract
Introduction
Methods
Results
Discussion
References

The redox products of nitrogen monoxide, including the free radical nitric oxide (NO), are highly reactive molecules that have been implicated in a wide variety of physiological and pathological processes. Thought to play an important role in smooth muscle relaxation (Ignarro et al. 1987; Palmer et al. 1987), NO seems also to be involved in CNS function and in pathogenesis in inflammation. This compound has received much attention as a putative retrograde messenger at synapses involved in long-term potentiation (Dawson et al. 1992). Further, it has been suggested that NO may be involved in macrophage mediated cytotoxicity in the CNS (Dawson et al. 1993; Merrill et al. 1993), including glial damage in demyelinating disease (Bo et al. 1994; Lin et al. 1993; Mitrovic et al. 1994; Zielasek et al. 1995).

There has been much attention paid to the question of reactive chemical species and pathways in these processes. The identity of endothelium-derived relaxing factor (EDRF), for example, is still unclear. Both nitric oxide itself, and redox products including nitrosothiols have been implicated (Stamler 1995). NO is readily inactivated by oxygen, superoxide, and transition metals and its half life in aqueous media, and tissue fluids in particular, is just a few seconds. Its physiological roles may thus be enhanced by formation of more stable intermediates such as proteins and small thiols (Stamler 1995). Further, many biological functions of NO depend on activation of guanylate cyclase, resulting in elevated cGMP levels. In some cases, however, there is direct alteration of a target protein. The open probability of Ca2+-activated K+ channels in vascular smooth muscle is, for example, increased by NO in cell-free patches (Bolotina et al. 1994).

An interaction of NO with the process of excitation in axons has not been described. However, most studies of excitability have been carried out in brain slices or in neurons dissociated acutely or in culture. We examine here a preparation in which the axonal microenvironment is maintained closer to that found in vivo. As will be seen, NO has a significant influence on conduction, but its action suggests a requirement for an extra-axonal intermediate.

    METHODS
Abstract
Introduction
Methods
Results
Discussion
References

Electrophysiology

Adult female Lewis rats were killed by CO2 asphyxiation, and sciatic nerves or vagus nerves were dissected. Some nerves were left intact, and others were mechanically desheathed. If the latter were to be further dissociated, they were incubated in collagenase/dispase (Boehringer, 3.5 mg/ml) for 15-50 min at room temperature. The desired branch was placed in a chamber fitted with two or three pairs of Pt wires for external stimulation and recording. Nerves were sealed to the electrodes with petroleum jelly (Vaseline), insulating this region from the bath. The external Locke's solution contained (in mM) 154 NaCl, 5.6 KCl, 2 CaCl2, 5.6 D-glucose, 10 N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid (HEPES), pH 7.4. The temperature was measured with a thermistor and was controlled with a proportional current passed through glass-insulated nichrome wires. The bath was stirred continuously and was partially covered by a glass coverslip. Compound action potentials (CAPs) were amplified, digitized, and stored in a computer for later analysis. The preparation was monitored for 5-10 min and was then cycled through 37°C to test its temperature dependence. Aside from speeding conduction, normal axons are little affected by this temperature change, and this allowed us to regulate the release of NO from diethylamine-NONOate (DEA-NONOate) as noted below. Control and drug protocols were identical, except that when inactive control compounds were tested, the nerve was generally monitored for a period of time longer than that known to be required for drug action. The absolute amplitude of the CAP, given in each figure, is dependent on factors not under experimental control such as the extracellular shunting conductance at the Pt wires when sealed with Vaseline. This amplitude thus varies among different preparations. However, within each experiment this level was required to be stable at constant temperature unless perturbed by a drug. Nerves that did not fulfill this criterion were discarded. Forty-one sciatic nerves (25 intact and 16 desheathed) and five vagus nerves were successfully examined with nitric oxide generating drugs.

NO release

DEA-NONOate was obtained from Cayman Chemical (Ann Arbor, MI) and was dissolved in Locke's immediately before use. The breakdown rate of the drug was measured in several experiments by extracting 50-µl aliquots from the chamber and analyzing NO-2 with the Greiss reagent (Green et al. 1982). Measured values agreed with published data (Cayman Chemical) (Hrabie et al. 1993). S-nitrosocysteine was synthesized by the method of Lei et al. (1992). 8-Br-cGMP was from Sigma Chemical (St. Louis, MO). S-nitrosocysteine and 8-Br-cGMP were each tested in three experiments with consistent results.

    RESULTS
Abstract
Introduction
Methods
Results
Discussion
References

Since the lifetime of NO in aqueous media is short, we required a means of NO generation that was controllable and sufficiently rapid to reach levels consistent with pathology. DEA-NONOate dissociates to release NO and free amine with a half-time of 8 min at 22°C and pH 7.4. The half-time is reduced to 2.1 min at 37°C (Cayman Chemical) (Hrabie et al. 1993), and this temperature dependence was used as a means of accelerating NO release at specific times, thereby minimizing dependence of kinetics on solution exchange times. CAPs were measured from branches of rat sciatic nerves. The records in Fig. 1A were taken in sequence and were at 25°C except for sweeps b and g. Traces a-c were recorded before adding the drug. Raising the temperature to 37°C in the absence of drug (up to >= 30 min) speeds conduction and reduces the amplitude by ~15% but has little further effect. After returning the preparation to 25°C, 0.62 mM DEA-NONOate was added to the bath. After an exposure of ~8 min, there was a small reduction in the CAP (d). Transiently raising the temperature to 37°C for just 2 min to increase [NO] resulted in an almost complete loss of signal (e). Washing out the DEA-NONOate restored the CAP (f-h). Since the brief rise in temperature in Fig. 1A during drug exposure increases breakdown of DEA-NONOate to DEA and NO and also speeds loss of the CAP, the conduction block is not likely to result from a direct action of intact DEA-NONOate. Several experiments were performed to examine further the participation of NO. DEA is both precursor and breakdown product of DEA-NONOate, but exposure to the free amine under a protocol similar to that above had no effect on the CAP (Fig. 1B). Subsequent introduction of DEA-NONOate blocked conduction reversibly, as seen in the third and fourth sweeps of Fig. 1B. In the experiment of Fig. 1C, after taking a control record, the bath was switched to Locke's plus 0.62 mM DEA-NONOate that had been prepared the previous day, to allow full release of NO and ultimate conversion to (primarily) NO-2. The 25-min period of drug exposure included 9 min at 37°C, but there was no decline in the CAP. Fresh DEA-NONOate was then added, and after 14 min (including just 3 min at 37°C) conduction was completely blocked. The CAP was partially restored on washing. In intact nerves exposed to 0.62 mM DEA-NONOate, the CAP was reduced 100% in 15 ± 1 (SE) min including 4.2 ± 0.6 min at 37°C (n = 10).


View larger version (14K):
[in this window]
[in a new window]
  		FIG. 1. Block of conduction after release of nitric oxide. A: sweeps representing compound action potentials (CAPs) from an intact rat tibial nerve recorded in sequence from left to right. The connective tissue sheath surrounding this nerve was left intact after dissection. Sweeps a, c, d, e, f, and h were at 25°C; b and g were at 37°C. Diethylamine-NONOate (DEA-NONOate; 0.62 mM) was present in the Locke's solution when sweeps d and e were recorded. Between traces d and e the temperature was transiently raised to 37°C for ~2 min to speed release of nitric oxide (NO). The stimulus artifact has been eliminated from the records (brief gap preceding the CAP). B and C: controls for the responsible chemical species. B: traces from an intact tibial nerve. Records are shown before adding any drug, 40 min (including 9 min at 37°C) after adding 1.24 mM DEA-HCl (the maximum concentration after breakdown of 0.62 mM DEA-NONOate), 29 min (4 min at 37°C) after switching to 0.62 mM DEA-NONOate, and after 22 min of washing with Locke's. All CAPs were at 21-22°C. C: CAP records were from an intact sural nerve, and all sweeps were at 23-25°C. The 1st record was taken before addition of drug. The 2nd sweep was recorded after 25 min (9 min at 37°C) in Locke's plus 0.62 mM DEA-NONOate that had been prepared the previous day. Traces were then taken after 14 min (including 3 min at 37°C) in freshly made 0.62 mM DEA-NONOate, and after 22 min of washing with Locke's.

Block by DEA-NONOate was dose dependent; 0.062 mM reduced the CAP by 24% after 20-30 min, including 10 min at 37°C. Block by 0.12 mM drug under similar conditions was 54%. At 0.3 mM (and above) the CAP was reduced 100%. We chose 0.62 mM as a standard test concentration because it produced reversible and complete block of conduction in just a few minutes at 37°C. It should be recognized that, although these data reflect a dependence on drug levels, the quantitation is only approximate, since, as will be shown below, the active chemical species is at least one and possibly two steps removed from DEA-NONOate itself.

The experiments thus far all pointed to NO or its redox products as the cause of conduction block. We attempted one further test, which produced an unexpected result. Nitric oxide forms complexes with heme proteins, and hemoglobin (Hb) has been used as a scavenger in NO-dependent processes (Blatter and Wier 1994; Boulton et al. 1994; Ericinska et al. 1995). To permit access of Hb to fibers, the surrounding connective tissue sheaths must be removed. In the experiments described thus far, these sheaths were left intact (except for small disruptions at cut branches) to minimize direct contact of the NO generating drug with axons. The outermost peripheral nerve sheath, the epineurium, is composed largely of collagen, scattered fibroblasts, mast cells, and blood vessels. The perineurium, which surrounds each fascicle, consists of several concentric layers of flattened cells linked by tight junctions to form an effective permeability barrier. The endoneurium surrounds the individual Schwann cells that envelop myelinated and unmyelinated axons and contains collagen, fibroblasts, mast cells, macrophages, and capillaries (Peters et al. 1991). The epineurium and perineurium can largely be removed mechanically, and the endoneurium dispersed with collagenase. Nerves prepared in this way were exposed to DEA-NONOate (0.25-0.62 mM) plus Hb (20-30 mg/ml), and the CAP amplitude declined very little after 30 min, including 14 min at 37°C (Fig. 2, black-square). However, an ensuing positive control consisting of DEA-NONOate alone also had little effect: the CAP dropped ~35% after 24 min including 11 min at 37°C (Fig. 2, bullet ). Part of this decline was due to temperature and occurred in the absence of NO (Fig. 2, black-triangle). In contrast, the CAP amplitude in intact nerves was zero after 12 min in DEA-NONOate, including just 4 min at 37°C (Fig. 2, black-down-triangle ). Preparations with intact sheaths were thus significantly more sensitive to NO than dissociated nerves, despite the presumed more rapid access to axons in the latter.


View larger version (13K):
[in this window]
[in a new window]
  		FIG. 2. Rate of decline of the CAP in dissociated vs. intact nerves. The ordinate represents the CAP amplitude relative to that at the start of each 37°C period. black-square, dissociated sural nerve in Hb, 30 mg/ml plus DEA-NONOate, 0.62 mM. bullet , same nerve as above, but in DEA-NONOate, 0.62 mM alone. black-triangle, intact tibial nerve in Locke's plus diethylamine HCl, 1.25 mM. black-down-triangle , intact sural nerve in DEA-NONOate, 0.62 mM.

Experiments comparing intact and desheathed nerves were repeated with consistent results in individual trials. However, because it was difficult to control the NO concentration with precision, a test was designed that ensured equal exposure to both preparations. The chamber was modified to accept two nerves. Two pairs of Pt electrodes allowed independent stimulation of each nerve, and a third pair served as a common recording site for both bundles. In the experiment illustrated in Fig. 3, the sural nerve from each leg of one animal was dissected. One nerve was left intact (Fig. 3A, stimulus S1) and the other treated as above to remove the sheaths and dissociate the bundle (S2). After recording control CAPs, 0.62 mM DEA-NONOate was added. After 14 min, the CAP was almost entirely absent in the intact nerve (top set of sweeps), but was only partially reduced in the dissociated preparation (bottom set). Both recovered on washing with Locke's. The full sequence of events is shown in Fig. 3B, which plots in the top graph the CAP amplitude relative to that immediately following the period of stabilization for intact and dissociated nerves. The temperature is plotted over the same time period in the bottom graph. The CAP amplitude in the intact nerve (bullet ) fell almost to zero on adding DEA-NONOate (), particularly after the temperature reached 37°C. The dissociated nerve was much less affected by the NO releasing compound (open circle ). After washing and recovery, a subsequent period at 37°C in the absence of drug had little effect on either nerve.


View larger version (23K):
[in this window]
[in a new window]
  		FIG. 3. Differential response of intact and dissociated nerves to NO release. The sural nerve was dissected from each leg of one animal. A: intact nerve (sheaths indicated by the dashed lines) was stimulated at S1, while the desheathed and enzymatically dissociated bundle was stimulated independently at S2. There was a common recording electrode (R). CAP records are shown from the intact (top set) and dissociated (bottom set) nerves. Sweeps were taken before adding the drug, after 14 min in 0.62 mM DEA-NONOate, and 42 min after washing with Locke's. All records were at 27-28°C. B: top graph plots CAP amplitude relative to that just before adding DEA-NONOate vs. time. bullet  ------ bullet , intact nerve. open circle  - - - open circle , results for the dissociated nerve. Bottom graph: temperature is plotted over the same time period (black-down-triangle ). The horizontal bar indicates the time of drug application.

During mechanical desheathing the epineurium was always completely removed. However, in several instances the perineurium broke and left at least one branch covered either partially or completely. This allowed a partial test of the connective tissue elements involved in NO action. In the experiment of Fig. 4, the tibial nerve was stripped of its epineurium, but remained covered by the perineurium (confirmed later in semithin sections). Both outer sheaths were removed from the sural nerve, but the endoneurium was left undisturbed (the collagenase exposure was omitted). The branches were stimulated individually in retrograde fashion (tibial, S1; sural, S2). Sweeps a and e illustrate signals before NO exposure. Traces b and f were taken 10 min after adding 0.62 mM DEA-NONOate, holding at 25°C. The amplitude is reduced ~35% in the tibial nerve (b) and is virtually unchanged in the desheathed sural nerve (f). Records c and g show the CAP after raising the temperature to 37°C for three additional minutes to speed release of NO. Conduction is blocked almost completely in the nerve with a remaining perineurium, but is much less affected in the desheathed nerve. Access to fibers in the tibial nerve is impeded by both the perineurium and the larger diameter, yet NO action was significantly stronger than in the case of the thinner, desheathed sural nerve. Both signals are restored on washing for 20 min (d and h). The results show first that the epineurium is not required for NO block. Further, effects in the sural nerve were significantly smaller despite the lack of disruption of the endoneurium by collagenase.


View larger version (13K):
[in this window]
[in a new window]
  		FIG. 4. NO susceptibility with partial removal of the sheath. The tibial (stimulus S1) and sural (S2) branches of the same sciatic nerve were mounted for retrograde transmission as shown in the sketch. After mechanical desheathing, the perineurium remained on the tibial nerve, but the sural nerve retained only the endoneurium (see text). a-d: CAP records after stimulation of the tibial nerve at S1. e-h: recorded from the sural nerve by stimulating at S2. a and e: before application of the drug. b and f: 10 min after adding 0.62 mM DEA-NONOate at 25°C. c and g: after 3 min at 37°C in the presence of drug. d and h: after 20 min of washing with Locke's. All records were at 25°C except for c and g, which were at 35-36°C.

Comparison with tetrodotoxin (TTX)

The connective tissue sheaths normally provide an effective permeability barrier. We used TTX, a small, hydrophilic molecule (MW 319) that blocks Na+ channels with a Kd of 3-5 nM (Colquhoun and Ritchie 1972) as a test. In the experiment of Fig. 5A, the sural nerves from each leg of one animal were dissected, with one nerve left intact (S1) and the other desheathed and dissociated with collagenase (S2). After taking control records, DEA-NONOate (0.62 mM) was applied. Following a brief period at 37°C, the CAP in the intact nerve had decreased much more than that in the dissociated nerve. The NO-generating drug was washed out, and 20 nM TTX was added to the bath. The CAP in the desheathed nerve decreased to zero while that of the intact preparation was unaffected. After wash out and reversal, 200 nM TTX produced an abrupt loss of conduction in the desheathed nerve, but again, no detectable decline in the nerve with intact sheaths. TTX action, like that of virtually all known drugs and toxins, is thus enhanced by increased access to axons. NO, on the other hand, has an entirely opposite dependence.


View larger version (11K):
[in this window]
[in a new window]
  		FIG. 5. Reciprocal sensitivity of intact and dissociated nerves to NO and tetrodotoxin (TTX). A: sketch shows 2 sural nerves, with one (stimulus S1) intact and the other (S2) desheathed and dissociated with collagenase. Records were taken sequentially from the intact (top set) and dissociated (bottom set) nerves. Sweeps in DEA-NONOate were recorded after 16 min, including 4 min at 37°C. Other solution changes are indicated above each record. B: lack of NO block after an extended desheathed period. CAP sweeps from a sural nerve that was mechanically desheathed and enzymatically dissociated, and then allowed to stand in Locke's for 4 h at room temperature before recording. Traces are shown before adding drug, 34 min (11 min at 37°C) after adding 0.62 mM DEA-NONOate, and after 16 min of washing with Locke's. C: lack of effect of 8-Br-cGMP. Records from a desheathed and dissociated tibial nerve before drug, after 38 min (19 min at 37°C) with 1 mM 8-Br-cGMP added to the bath, and 38 min after returning to Locke's.

As seen in Figs. 2-5 above, in most desheathed and dissociated nerves there was some decrease in the CAP amplitude by NO, but the reduction was always significantly less than that of an intact branch serving as a positive control. This remaining susceptibility to NO may reflect an incomplete removal of some endoneurial component since normally measurements began very shortly after collagenase exposure. When a nerve was mechanically desheathed, enzymatically dissociated, and allowed to stand in Locke's for 4 h before recording, DEA-NONOate had no effect on the CAP. Figure 5B shows records before drug, after 34 min in 0.62 mM DEA-NONOate, including 11 min at 37°C, and after 16 min of washing with Locke's. In intact sural nerves, 0.62 mM DEA-NONOate reduced the CAP 100% in 14 ± 1 min (including 4 min at 37°C, n = 6). In acutely dissociated preparations the CAP was reduced at this time point by an average of 22 ± 12% (n = 4).

In many systems, nitric oxide functions through the activation of soluble guanylate cyclase to raise levels of the second-messenger cyclic GMP (Garthwaite 1991). We have made one test for participation of this mechanism in conduction block. 8-Bromoguanosine 3':5'-cyclic monophosphate (8-Br-cGMP), a membrane-permeant analogue of cGMP, was applied (1 mM) to one desheathed and two intact nerves. Conditions, including temperature protocols, were identical to those used for DEA-NONOate, but no significant effect on the CAP was observed (Fig. 5C).

Susceptibility of unmyelinated axons

The nerve segments tested thus far all have a high proportion of myelinated axons. At the stimulus strengths used, the CAPs reflect entirely the fast-conducting myelinated population. Does NO action require normal myelin structure? In contrast to the sciatic branches used above, the percentage of myelinated fibers in the rat vagus nerve is small. With relatively strong stimulating currents, it is possible to record from the unmyelinated population. Sweep a in Fig. 6 shows the resulting CAP. The first peak has a high conduction velocity and represents myelinated axons. The second peak (arrow) is much slower and appears only with much stronger stimuli. It is therefore likely to be derived from unmyelinated fibers. Record b was taken ~16 min after adding 0.62 mM DEA-NONOate at room temperature. Both peaks are partially reduced. Sweep c was recorded after ~1 min at 37°C. The second peak is absent, while the myelinated fibers continue to conduct. Four minutes later, conduction in both families of fibers was blocked (d). Signals in both myelinated and unmyelinated axons recovered after washing with Locke's (e and f). The higher sensitivity of unmyelinated axons to NO was a consistent finding in the five vagus nerves tested.


View larger version (10K):
[in this window]
[in a new window]
  		FIG. 6. Loss of signals in unmyelinated and myelinated axons in the vagus nerve. Sweep a, before drug. Arrow indicates the slowly conducting signal originating from unmyelinated axons; 23°C. b: 16 min after adding 0.62 mM DEA-NONOate; 23°C. c: after ~1 min at 37°C. d: after 3 more min at 37°C and 1 min of cooling; 30°C. e: after 12 min of washing with Locke's; 23°C. f: after 9 additional minutes of washing; 37°C.

S-nitrosothiols

The reversibility of NO action places constraints on the likely chemical interactions involved to produce block. Reactions with both heme proteins and thiols have been shown to be highly significant in biological systems. As noted earlier, we could not demonstrate an involvement of soluble guanylate cyclase, an example of the former. With respect to the latter, nitrosothiols represent possible intermediates, and S-nitrosocysteine has been suggested as a reactive species in control of the vasculature (Jia et al. 1996; Myers et al. 1990). This compound can transfer NO directly (as NO+) to other sulfhydryl groups, e.g., within proteins. Additionally, however, in experimental physiological solutions that typically contain traces of Fe or Cu, there is rapid cleavage of the S-N bond to release NO rapidly (seconds to minutes) (Myers et al. 1990; Stamler 1995). Figure 7 shows results from both intact (a-e) and desheathed (f-j) nerves. In the former case, S-nitrosocysteine acted in a manner very similar to that of DEA-NONOate. All records were at 37°C. Sweep b was recorded after introduction of 0.5 mM S-nitrosocysteine and just after raising the temperature to 37°C. After an additional 4 min, the CAP was reduced ~90% (c) and was almost completely blocked after 2 additional min (d). Conduction recovered on washing with Locke's (e). The same drug applied to a desheathed and dissociated nerve also blocked action potentials, but with slower kinetics. Trace g was recorded just after raising the temperature to 37°C in the presence of 0.5 mM S-nitrosocysteine. Four minutes later the CAP was reduced ~40% (h), and after 10 more minutes there was 90% block (i). There was partial recovery after washing (j). Thus, in contrast to DEA-NONOate, S-nitrosocysteine was highly effective in both preparations, but sensitivity was greater in the intact nerve.


View larger version (10K):
[in this window]
[in a new window]
  		FIG. 7. Block of action potentials by S-nitrosocysteine. a-e: CAPs from an intact sural nerve. a: Locke's. b: 0.5 mM S-nitrosocysteine added to the bath and temperature raised to 37°C. c and d: 4 and 6 min, respectively, later than b. e: after removal of the drug. f-j: records from a tibial nerve desheathed and dissociated in collagenase for 50 min. f: Locke's. g: 0.5 mM S-nitrosocysteine added to the bath and temperature raised to 37°C. h and i: 4 and 14 min, respectively, later than g. j: after removal of the drug. All records are at 37°C.

    DISCUSSION
Abstract
Introduction
Methods
Results
Discussion
References

These experiments describe an apparently unique mode of conduction block. We know of no other drug or toxin whose effectiveness is reduced with increased accessibility to axons. The results suggest an involvement of the connective tissue sheaths in nitric oxide action. This could involve a chemical intermediate, or an increase in reactivity resulting from a change in the pH or reduction-oxidation potential of the environment (Lipton et al. 1993; Stamler et al. 1992). The epineurium seems not to be required, but it is difficult to distinguish between the perineurium and endoneurium. Removal of the former allows soluble components in the latter to diffuse away, even if not enzymatically dissociated. Because the endoneurium surrounds all nerve fibers within each branch, it is the more likely source of an intermediate or environmental factor. From the period required for loss of NO action following enzymatic dissociation of the endoneurium (22% remaining after ~1 h and zero after 4 h), it can be judged that the half-time for removal of the environmental factor is on the order of 30 min. This suggests that diffusion of this component within the connective tissue may normally be restricted. Alternatively, the differential sensitivity between intact and dissociated nerves may result from diffusion of a trace substance, O2, or pH from bath to axons that increases the rate of inactivation of NO. However, small molecules should reach axons quickly. The half-time of 20 nM TTX, which has a slow blocking action even after reaching axons, was just a few minutes. Thus the interacting component is more likely to reside in the nerve sheath. It is of interest that a similar intermediate has been postulated with regard to the function of EDRF (Stamler 1996; Wei and Kontos 1990).

The chemistry of NO provides some clues regarding mechanisms that might be active in the loss of conduction. S-nitrosothiol formation is reversible and is known to play a role in several biological processes. This can result from reaction of thiols with NO or with peroxynitrite (ONOO-, a product of NO· and O-2). ONOO- can also oxidize sulfhydryl groups to form disulfide bonds (Radi et al. 1991; Wu et al. 1994). Physiologically, the rapidity, thoroughness, and reversibility of NO block of the CAP suggest an effect on voltage-dependent Na+ channels. Many years ago we demonstrated that sulfhydryl blockade by N-ethylmaleimide (NEM) induced a slow inactivation of Na+ channels in crayfish axons (Shrager 1977; Starkus and Shrager 1978) and Roed (1989) subsequently found that NEM blocks action potentials in rat peripheral nerve. Our results with S-nitrosocysteine support this idea. This drug can transfer the NO moiety directly to protein sulfhydryl groups, and this may underlie its action on dissociated nerves that have lost an endogenous intermediate. On the other hand, block of intact axons by S-nitrosocysteine could result from the known rapid release of NO in saline followed by the same reaction path as for NO released from DEA-NONOate. S-nitrosothiols are more stable in vivo due to maintenance of exceptionally low levels of free Fe and Cu, enabling their function as intermediates in NO pathways (Stamler 1995).

There are reports of the modulation of ion channel function in other cells by nitric oxide, in some cases involving sulfhydryl groups. Chen and Schofield (1995) tested NO donors on superior cervical ganglion neurons. NO exposure enhanced Ca2+ current amplitude through a cGMP-dependent mechanism. On the other hand, Bolotina et al. (1994) found that NO activated Ca2+-dependent K+ channels from vascular smooth muscle in isolated patches, independent of cGMP. NEM interfered with this activation, suggesting that NO action was dependent on channel sulfhydryl groups. Further, S-nitrosocysteine has been shown to block glutamate receptors by a cGMP-independent pathway that is thought to involve nitrosylation of sulfhydryl groups on the receptor (Lei et al. 1992).

Estimation of effective concentrations of NO

Are our results consistent quantitatively with a neuropathogenic role for NO during inflammation? We first estimate the concentration of NO required for axonal conduction block, using some typical values for kinetic parameters. In the bath the half-time for release of NO from DEA-NONOate is 2.1 min at 37°C, and the half-life of NO is ~30 s (Palmer et al. 1987). The solution to the kinetic equation (Moore 1962) shows that in the time from dissolving the DEA-NONOate to block of the CAP (12-18 min at 20°C and 2-5 min at 37°C), [NO] reaches a maximum value of 6% of the initial concentration of DEA-NONOate. In tissue the half-life of NO is ~4 s (Palmer et al. 1987), and including this in solving the diffusion equation for a cylinder with a diffusion coefficient of 3.3 × 10-5 cm2/s (Malinski et al. 1993) demonstrates that the maximum [NO] at the center of a 0.5-mm-diam nerve branch is about one-half the bath value (APPENDIX ). Thus, with a prepared solution of 0.3 mM DEA-NONOate (which blocked the CAP 100%), [NO] at the center of the branch was estimated to be 4-10 µM. A prepared solution of 62 µM DEA-NONOate, which blocked the CAP by 24%, resulted in [NO] within the branch of 1-2 µM.

What levels of [NO] may be reached in an inflammatory lesion? In their Fig. 2, Wood and Garthwaite (1994) calculated the spread of NO from a point source into a spherical zone and showed that despite the short half-life, [NO] at a distance of 50 µm would be ~70% of its value at the source. These authors computed [NO] at the surface of the releasing (endothelial) cell to be ~1 µM. We calculate from published results (Alleva et al. 1994; Hoffman et al. 1993) that [NO] in a single activated macrophage in vitro reaches 1 µM in ~10 s, and production can continue for many hours. The dense population of macrophages in perivascular zones would provide multiple sources in which [NO] summates and could reach pathological levels (Wood and Garthwaite 1994). These calculations are necessarily approximate but suggest that nitric oxide may contribute to block of axonal conduction under pathological conditions.

    ACKNOWLEDGEMENTS

  We are grateful to Dr. S. Neil Rasband for assistance in diffusion calculations. We thank P. Guthikonda and E. Brunschweiger for excellent technical assistance.

  This work was supported by National Multiple Sclerosis Society (NMSS) Grant RG 2687 and National Institute of Neurological Disorders and Stroke Grant NS-17965 to P. Shrager. D. Mattson is a Harry Weaver Neuroscience Scholar of the NMSS.

    APPENDIX

The diffusion equation in cylindrical coordinates, with a constant rate of chemical consumption (Eq. A1), was solved by separation of variables
∂<IT>C</IT>/∂<IT>t</IT>= (1/<IT>r</IT>)[(∂/∂<IT>r</IT>)(<IT>rD</IT>∂<IT>C</IT>/∂<IT>r</IT>)] − kC (A1)
The solution was subject to the initial condition that C(r, 0) = 0, where 0 < r < a. The boundary condition was that at t >=  0, C(a, t) = Co, where Co is the bath concentration of nitric oxide and a is the radius of the nerve branch. The solution is then given by Eq. A2.
<IT>C</IT>(<IT>r</IT>, <IT>t</IT>) = <IT>C</IT><SUB>o</SUB><IT>I</IT><SUB>o</SUB>(<IT>Kr</IT>)/<IT>I</IT><SUB>o</SUB>(<IT>Ka</IT>) + <LIM><OP>∑</OP><LL><SUB><IT>j</IT>=1</SUB></LL><UL>∞</UL></LIM>(−2<IT>C</IT><SUB>o</SUB>/<IT>W</IT>)
× &cjs0362;<LIM><OP>∫</OP></LIM><SUP><IT>a</IT></SUP><SUB>0</SUB><IT>Qdr′</IT>&cjs0363;<IT>J</IT><SUB>o</SUB>(<IT>Lr</IT>) exp(−<IT>Zt</IT>) (A2)
where K = (k/D)1/2, L = xj/a, W = a2Io(Ka)J21(xj), Q =r'Io(Kr')Jo(Lr'), Z = D(L2 + K2), Io is the modified Bessel function, Jo is the Bessel function of the zeroth kind, J1 is the Bessel function of the first kind, k is the rate of NO chemical consumption, D is the NO diffusion constant, and xj is the zeroes of Jo(x).

Equation A2 was solved numerically, using the 1st 13 terms of the infinite sum since the latter converges rapidly.

    FOOTNOTES

  Address for reprint requests: P. Shrager, Dept. of Neurobiology and Anatomy, Box 603, University of Rochester Medical Center, 601 Elmwood Ave., Rochester, NY 14642.

  Received 29 April 1997; accepted in final form 25 September 1997.

    REFERENCES
Abstract
Introduction
Methods
Results
Discussion
References
0022-3077/98 $5.00 Copyright ©1998 The American Physiological Society



This article has been cited by other articles:


Home page
 BrainHome page
M. J. Craner, B. C. Hains, A. C. Lo, J. A. Black, and S. G. Waxman
Co-localization of sodium channel Nav1.6 and the sodium-calcium exchanger at sites of axonal injury in the spinal cord in EAE
Brain, February 1, 2004; 127(2): 294 - 303.
[Abstract] [Full Text] [PDF]

Home page
 J. Neurophysiol.Home page
M. Renganathan, T. R. Cummins, W. N. Hormuzdiar, J. A. Black, and S. G. Waxman
Nitric Oxide Is an Autocrine Regulator of Na+ Currents in Axotomized C-Type DRG Neurons
J Neurophysiol, April 1, 2000; 83(4): 2431 - 2442.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Shrager, P.
Right arrow Articles by Mattson, D.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Shrager, P.
Right arrow Articles by Mattson, D.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Visit Other APS Journals Online