What Forms of Activity Arise in Axons

When mammalian axons are injured, they can generate spontaneous activity. The characteristics of such ectopic activity in isolated nerve were studied by Adrian (1930). He described both high frequency (>150 Hz) regular and low frequency irregular activity, and a third type consisting of bursts of impulses repeating at a lower frequency (<10 Hz). However, these forms of injury-induced discharge differ from the low frequency, regular discharge seen in experimental demyelination.

Adrian's third type of activity is similar to that known to arise in human nerve fibers in situ after a period of ischemia (Culp et al., 1983). Kapoor et al. (1993) reported very similar burst discharges induced by loading myelin with a solution containing a high concentration of K+ ions., This finding provides a clue as to how postischemic activity is generated. It seems that to generate these burst discharges, the axonal membrane potential must be able to acquire two stable values (i.e.. it must become bistable) and an internode loaded with K+ is necessary to allow a bistable membrane potential to exist. David et al. (1992) recorded membrane bistability in lizard axons with K+-loaded internodes and demonstrated that regenerative K+ currents were responsible for producing long-lasting depolarizing events. Before these findings were published, Bostock deduced that the explanation for muscle fasciculation after peripheral nerve ischemia depends on the membrane potential of myelinated motor axons becoming bistable in the postischemic state. A depolarized value of membrane potential is associated with raised external K+ (expected to result from ischemia) and a hyperpolarized value independent of the K+ equilibrium potential (EK) where the K+ channels are closed and a high Na+/K+ ATPase current operates (Bostock et al., 1991). Any movement between the two values from the hyperpolarized to depolarized state would be relatively rapid (i.e., a flip) and associated with ectopic impulse generation. Such a flip could be triggered by changes in Na+/K+ ATPase current, or by the arrival of an independently generated propagated action potential. Journeying back to the hyperpolarized state (i.e., a flop) silences activity and potentially reprimes the axon to generate more impulses.

With transection of peripheral nerve, regenerative sprouting occurs that fails to achieve reconnection with the peripheral target and may result in neuroma formation. The afferent endings trapped within the nerve end are a source of spontaneous activity (e.g., Lisney and Devor, 1987) and are pathologically mechanosensitive (e.g., Welk et al., 1990). The latter is also a characteristic of demyelinating lesions (Smith and McDonald, 1980; Calvin et al., 1982). Neuromas in humans can be a source of chronic pain, and surgical excision runs the risk of further neuroma formation. The hyperexcitability and ectopic impulse generation at diseased terminals are plausibly related to the accumulation of Na+ channels, which has been demonstrated by using specific Na+ channel antibodies in peripheral nerve of fish, Apteronotus (Devor et al., 1989) and in rat (Devor et al., 1993). Consistent with Na+ -channel involvement, ectopic activity can be attenuated and inhibited by agents that block Na+ channels, such as the marine toxin tetrodotoxin (TTX) (Matzner and Devor, 1994). It is also silenced by low concentrations of local anesthetic (Devor et al., 1992), insufficient to prevent normal impulse transmission but consistent with the block of low-threshold Na+ channels that do not undergo fast inactivation (Baker, 2000b). (Low concentrations of local anesthetic also increase the fraction of inactivated transient Na+ channels at the resting potential [for example, Hille, 1977], and will tend to quench high frequency burst discharges, by promoting transient channel inactivation.) The underlying mechanisms of such spontaneous activity in neuromas may be related to those involved in promoting ectopic activity in demyelinated or partially demyelinated axons, because abnormal Na+ currents are probably responsible for initiating and maintaining the activity in both cases.

IV The Importance of Persistent Na+ Current

Although ectopic discharges after experimental demyeli-nation had been recorded previously (Smith and McDonald, 1980; Burchiel, 1980, 1981; Calvin et al., 1982), I assisted Hugh Bostock to make the first and only membrane potential and membrane current recordings from the site of impulse initiation in ectopically active demyelinated axons (Baker and Bostock, 1992). The active fibers were spinal root axons that had undergone diphtheritic demyelination. The importance of these recordings were twofold. First, they showed that ectopic impulses were generated at heminodes, where the myelin was missing or loosened on one side of a node of Ranvier (Fig. 2), although the node maintained a more focused population of Na+ channels and was the site at which regenerative inward current was initiated. Second, our records clearly indicated that a pacemaker potential was responsible for initiating and maintaining highly rhythmic discharge, where the pacemaker was generated by a sustained inward current operating within the subthreshold potential range. At the time, we did not know what channel

Figure 2 Membrane current contour maps recorded for two spontaneously active rat spinal root axons, previously demyelinated after exposure to diphtheria toxin. Regular discharges were induced by 4-aminopyridine (1 mM). Continuous lines represent inward current, and broken lines outward current. The outward currents are largely capacitative, the magnitude of late active outward current in the demyelinated internode reduced by the presence of the K+ channel blocker. (A) Sensory fiber 6 days after exposure to the toxin. (B) Motor fiber 7 days after exposure. Arrow in (A) indicates the point along the axon at which inward current is first recorded, and represents the ectopic site. Inward current in (B) can be seen to precede the initiation of the action potential. In both recordings, action potentials propagate away from the ectopic site, in a saltatory fashion caudally, but much more slowly and continuously in the rostral direction over a whole demyelinated internode. The ectopic sites are heminodes, where the myelin is functional on only one side. (Figure reproduced, with permission, from Baker and Bostock, 1992.)

type provided the sustained inward current, although more recent experiments are consistent with a persistent Na+ current (Baker and Bostock, 1997, 1998; Bostock and Rothwell, 1997). Action-potential firing frequency was low, on the order of 10 to 20 Hz, and the firing frequency was increased by exposure to tetraethyl-ammonium ions, indicating that axonal slow K+ channels control the discharge frequency (Baker and Bostock, 1992) by modifying the slope of the pacemaker potential. Subsequently, subthreshold and barely suprathreshold membrane potential oscillations at a similar frequency were recorded by Kapoor et al. (1997), using intra-axonal electrodes in dorsal column axons, focally demyelinated by exposure to ethidium bromide. These oscillations could occur without action potential generation and could thus be studied alone. The oscillations were eliminated by exposure to TTX, providing strong evidence that a TTX-sensitive (TTX-s) form of persistent Na+ current was responsible for the depolarizing phase of the cycle, and that, in at least some diseased axons, the current is large enough to drive a membrane potential oscillation (reviewed by Baker, 2000a).

Usually, when activated by a suprathreshold depolarization, Na+ channels generate a brief inward current that underlies the upswing of an action potential. The channels that contribute to membrane excitability in this way are referred to as transient Na+ channels. Eight of the nine functionally described mammalian Na+ channel subtypes generate a transient current, although the current kinetics are variable. Of these, the TTX-resistant (TTX-r) sensory neuron-specific channel (NaV1.8) has activation and inactiva-tion kinetics four to five times slower than those of TTX-s channels, and this may be related to its function in small diameter and unmyelinated afferents. The transient nature of a Na+ current is brought about by an inactivation mechanism that plugs the channel and prevents persistent opening (reviewed by Catterall, 2000). The inactivation mechanism appears voltage-dependent and becomes more complete with increasing depolarization. A highly conserved motif of three amino acids (IFM; isoleucine-phenylalanine-methio-nine), located on the intracellular loop between the third and fourth membrane-spanning domains of the protein, is associated with the inactivation gate; these residues are known to be involved in blocking the channel pore from the inside.

We now know, however, that Na+ channel gating is more complicated than this, and that individual Na+ channels can exhibit a range of gating behaviors. Thus Na+ channel gating, even of a supposedly uniform population of channels, can be nonhomogeneous. Unitary-current studies of muscle Na+ channels by Patlak and Ortiz (1985, 1986) indicated that apparently normal cardiac and frog skeletal muscle Na+ channels could sporadically lose their ability to inactivate and thus produce sustained Na+ currents. Some of the most convincing recordings of apparently normal cardiac Na+ channels undergoing a series of sporadic behavioral changes or "modal-gating switches" were provided by the studies of Bohle and Bendorff (1995). A range of Na+ channel behaviors were recorded by Baker and Bostock (1998) in primary sensory neurons, including brief, sporadic, openings that were increased in number (but unaffected in duration) by depolarization (Fig. 3). (One possibility is that these openings correspond with a second Na+ channel open-state; see for example, Keynes and Elinder, 1998.) Such openings (present even at -80 mV) and other types of openings appearing throughout long duration voltage-clamp protocols might be expected to contribute to persistent Na+ current in sensory neurons over a wide range of membrane potentials.

Nonhomogeneous Na+ channel gating behavior in squid axons was reported by Chandler and Meves (1970), who observed persistent Na+ currents when F- ions were introduced to the inside of the axon. The persistent Na+ current activated with a more negative voltage-dependence than the transient current and had the same voltage dependence as steady-state inactivation, leading the authors to suggest that a second open state could be reached once the channel had undergone fast inactivation. The channel openings attained either with loss of fast inactivation or the attainment of a second open channel state are limited in duration by other, slower inactivation processes (Patlak and Ortiz, 1986). The second open state is incorporated in the model of the squid axon Na+ channel subsequently developed by Keynes and Elinder (1998). One important characteristic predicted by a second open state exhibiting the same voltage dependence as fast inactivation is that persistent openings will occur at more negative membrane potentials than transient openings, and this is the case for persistent Na+ currents in large-diameter dorsal root ganglion (DRG) neurons, recorded without F- (Baker and Bostock, 1997, 1998). This is also a necessary characteristic of the currents driving spontaneous activity in demyelinated nerve. Thus, the ability of a single subtype of Na+ channel to generate a variety of Na+ currents has been considered to be caused by the loss of fast inacti-vation or the attainment of a second open state, through the fast-inactivated state. While the molecular mechanisms underlying these behaviors or gating modes are not fully understood, the reduction/oxidation state of axonal Na+ channels (Mitrovic et al., 1993) modifiable by glutathione, and whether or not neuronal Na+ channels are co-associated with Py G-protein subunits (for NaV1.2, Ma et al., 1997) are two possible contributing factors. Normally, Na+ channel a-subunits are complexed with accessory P-subunits and heterologous expression studies have shown that the co-expression of P-subunits can substantially speed inactivation (i.e., reduce channel open-times) (for example, for NaV1.6 [Smith et al., 1998]). Channel gating behavior may thus also be affected by the gain (and perhaps loss) of P-subunits.

Persistent Na+ current almost certainly operates in normal human peripheral nerves. Data obtained on sensory nerves in normal subjects using the technique of threshold-tracking have shown that current threshold responds to the application of a brief hyperpolarizing current as though a persistent inward membrane current is switched off by the resulting change in membrane potential (Bostock and

Figure 3 TTX-s Na+ channels in primary sensory neurons exhibit a variety of gating behaviors. (A) Brief sporadic channel openings recorded from an outside-out patch, pulled from a large diameter DRG neuron. (B) In this patch, channel open-time does not depend on membrane potential (open-time distribution at -80 and -85 mV shown as solid bars, distribution at -35 mV shown as open bars). Right hand panel, frequency histogram of pooled measurements made between -55 and -35 mV is well described by a single exponential, indicating a mean open-time of 100 ^s. (C) Example openings recorded at -60 mV (D) The number of channel openings increases with depolarization, revealing that channel activation is voltage-dependent. (E) Recordings from outside-out patches also exhibit burst-opening Na+ channels. Example bursts lasting up to a few 10s of milliseconds. (F) In the same patch as shown in (E), much longer burst openings were recorded, at -55 (upper trace) and -35 mV (lower trace). At -35 mV the intraburst flicker is less apparent. (G) Estimates of burst duration from single best-fit exponentials to burst duration histograms are 5.8 ms at -55 mV and 4.2 ms at -45 mV None of these behaviors can be explained by transient channel activation-inactivation gating overlap. (Figure reproduced, with permission, from Baker and Bostock, 1998.)

Figure 3 TTX-s Na+ channels in primary sensory neurons exhibit a variety of gating behaviors. (A) Brief sporadic channel openings recorded from an outside-out patch, pulled from a large diameter DRG neuron. (B) In this patch, channel open-time does not depend on membrane potential (open-time distribution at -80 and -85 mV shown as solid bars, distribution at -35 mV shown as open bars). Right hand panel, frequency histogram of pooled measurements made between -55 and -35 mV is well described by a single exponential, indicating a mean open-time of 100 ^s. (C) Example openings recorded at -60 mV (D) The number of channel openings increases with depolarization, revealing that channel activation is voltage-dependent. (E) Recordings from outside-out patches also exhibit burst-opening Na+ channels. Example bursts lasting up to a few 10s of milliseconds. (F) In the same patch as shown in (E), much longer burst openings were recorded, at -55 (upper trace) and -35 mV (lower trace). At -35 mV the intraburst flicker is less apparent. (G) Estimates of burst duration from single best-fit exponentials to burst duration histograms are 5.8 ms at -55 mV and 4.2 ms at -45 mV None of these behaviors can be explained by transient channel activation-inactivation gating overlap. (Figure reproduced, with permission, from Baker and Bostock, 1998.)

Rothwell, 1997). Consistent with this interpretation, Tokuno et al. (2003) have presented direct evidence for a steady-state depolarization of rat axons in vitro in both sensory and motor nerve, near rest, by a TTX-s Na+ current. In addition, optic nerve axons are known to have a steady-state TTX-s Na+ conductance that must contribute to the pathological influx of Na+ that precipitates white matter damage during ischemia (Stys et al., 1993).

Peripheral Neuropathy Natural Treatment Options

Peripheral Neuropathy Natural Treatment Options

This guide will help millions of people understand this condition so that they can take control of their lives and make informed decisions. The ebook covers information on a vast number of different types of neuropathy. In addition, it will be a useful resource for their families, caregivers, and health care providers.

Get My Free Ebook


Post a comment