Differences between nerve terminal impulses of polymodal nociceptors and cold sensory receptors of the guinea-pig cornea

Abstract

1. Extracellular recording techniques were used to study nerve terminal impulses (NTIs) recorded from single polymodal nociceptors and cold-sensitive receptors in guinea-pig cornea isolated in vitro.

2. The amplitude and time course of NTIs recorded from polymodal nociceptors was different from those of cold-sensitive receptors.

3. Bath application of tetrodotoxin (1 μm) changed the time course of spontaneous NTIs recorded from both polymodal and cold-sensitive receptors.

4. Bath application of lignocaine (lidocaine; 1–5 mm) abolished all electrical activity.

5. Local application of lignocaine (2.5 and 20 mm) through the recording electrode changed the time course of the NTIs recorded from polymodal nociceptors but not that of NTIs recorded from cold-sensitive nerve endings.

6. It is concluded that action potentials propagate actively in the sensory nerve endings of polymodal nociceptors. In contrast, cold-sensitive receptor nerve endings appear to be passively invaded from a point more proximal in the axon where the action potential can fail or be initiated.

Activation of the nociceptive nerve terminals in tissues such as skin and joints generates action potentials that propagate both centrally to cause painful sensations and locally, in the nerve terminal axons, to trigger the release of neuropeptides producing neurogenic inflammation. The mechanisms that control the excitability of nociceptor terminals are largely a matter of speculation because of their small size (< 0.5 μm diameter) and their inaccessibility in intact tissues like skin (see Belmonte, 1996). What is known has been inferred indirectly from recordings from afferent axons when the environment of the receptors is pharmacologically manipulated (e.g. see Kress & Reeh, 1996). To investigate directly the mechanisms controlling the excitability nociceptors, we recently developed an extracellular recording technique that allows electrical activity to be recorded from identified sensory nerve terminals in guinea-pig cornea (Brock et al. 1998).

The cornea is very densely supplied by small-diameter sensory nerve endings, which terminate in the most superficial layer of the corneal epithelium (Belmonte et al. 1997). Three types of sensory receptor (polymodal, mechano-sensory and cold-sensitive) are found in the cornea (Gallar et al. 1993). Using a small-diameter suction electrode (≈50 μm) applied to the surface of the cornea, nerve impulses can be recorded and identified as originating in single sensory nerve terminals (Brock et al. 1998). Activation of these terminals by local stimuli enables them to be classified as belonging to one of the three functional classes. The estimated conduction velocities of the neurones range from 0.3 to 2.7 m s⁻¹, consistent with the recordings arising from both C fibres and thin myelinated Aδ fibres.

In the guinea-pig, the majority of sensory axons terminate abruptly as they approach the surface of the corneal epithelium and the nerve terminal impulses (NTIs) recorded at the surface of the cornea are diphasic (positive-negative) with a prominent positive component (Brock et al. 1998). Similarly shaped signals have been recorded extracellularly from the terminals of motor nerves where their configuration can be explained either by active invasion of the nerve terminal (Katz & Miledi, 1965) or by passive invasion from a point more proximal where the action potential fails (Dudel, 1963). It has been generally assumed that sensory nerve terminals are not able to support action potentials and that their activation by the sensory stimulus produces a generator potential which spreads proximally along the axon to a point that is excitable (e.g. Block, 1992). However, while this appears to be true for the specialized sensory endings of low-threshold mechano-sensory neurones (Mendelson & Loewenstein, 1965) and olfactory-sensory neurones (Firestein et al. 1990), at present nothing is known about the excitability of the free sensory endings of C and Aδ fibres.

Many of the nerve terminals in the cornea, identified using the extracellular recording approach, have an ongoing discharge of NTIs. This spontaneous activity is presumably generated at some site within the nerve terminal arbor. If action potentials are generated at a point close to the site of recording, it might be expected that the configuration of the spontaneous NTIs would differ from that of the electrically evoked NTIs. We have previously reported that, in a small percentage of recordings, the spontaneous NTIs had a discrete pre-potential, which might reflect a generator potential (Brock et al. 1998). However, in most recordings the spontaneous NTIs do not have a pre-potential, which suggests that the site of action potential initiation is electrotonically distant from the site of recording. In this case it would be predicted that the electrically evoked and spontaneous NTIs would be closely matched in both amplitude and time course.

The present study compared the configurations of electrically evoked and spontaneous NTIs recorded from polymodal nociceptors and cold-sensitive receptors and the effects of tetrodotoxin (TTX) and lignocaine on these NTIs. It is concluded that action potentials actively invade the terminals of polymodal nociceptors whereas those of cold-sensitive receptors are passively invaded.

METHODS

All experimental procedures conformed to the Australian National Health and Medical Research Council guidelines and were approved by the University of New South Wales Animal Care and Ethics Committee.

Electrophysiology

Eyes from guinea-pigs (150-300 g, killed with 100 mg kg⁻¹ pentobarbitone sodium i.p.) were mounted in a recording chamber and superfused with physiological saline of the following composition (mm): Na⁺, 151; K⁺, 4.7; Ca²⁺, 2; Mg²⁺, 1.2; Cl⁻, 144.5; H₂PO₃⁻, 1.3; HCO₃⁻,16.3; glucose, 7.8. This solution was gassed with 95 % O₂/5 % CO₂ (to pH 7.4) and maintained at 31-32 °C. The optic nerve and associated ciliary nerves were drawn into a suction stimulating electrode. The stimulus parameters were modified as required during the experiment (pulse width 0.1-0.5 ms, 5-30 V). A glass recording electrode (tip diameter ≈50 μm) filled with physiological saline was applied to the surface of the corneal epithelium with slight suction (Brock et al. 1998). Electrical activity was recorded through an AC amplifier (Neurolog NL104, Digitimer Ltd, Welwyn Garden City, UK; gain ×2000, high pass filter set at 0.1 Hz), digitized at 44 kHz and stored on magnetic tape using a PCM recorder (A. R. Vetter Co. Inc., Rebersburg, PA, USA). Recordings were only made from sites where the nerve impulses were readily distinguished from the noise (≈10 μV peak-to-peak when low-pass filtered at 3-5 kHz). At many sites on the corneal surface, no evoked or spontaneous electrical activity was recorded or the signals were too small to be analysed. It should be noted that this extracellular recording technique only records the fast currents occuring at the nerve terminal and will not detect the slower currents generated by the sustained application of agents like capsaicin.

The data presented were collected at recording sites where the electrical activity originated from a single nerve terminal. At these sites electrical stimulation of the ciliary nerves evoked a single all-or-none NTI at the site of recording and the spontaneously occurring orthodromic NTIs collided with the antidromically propagating electrically evoked NTIs (see Brock et al. 1998). In the present study, only NTIs that were defined as originating in either polymodal nociceptors or cold-sensitive receptors were analysed (see Brock et al. 1998). Polymodal nociceptors typically had low levels of ongoing NTI discharge (< 1 Hz) and were excited by bath application of a low concentration of capsaicin (0.05-0.2 μm). Some polymodal nociceptors had spontaneous NTIs with a clear pre-potential (see Brock et al. 1998) and, as these had clearly different time courses from the electrically evoked NTIs recorded from the same nerve terminal, they were excluded from the present analysis. The cold-sensitive receptors had relatively high levels of ongoing NTI discharge (2-15 Hz) and this activity was increased by cooling (to 28 °C) and decreased by warming (to 37 °C) the solution superfusing the cornea.

All drugs were supplied by Sigma (Castle Hill, NSW, Australia) and were applied by their addition to the superfusion solution or by internal perfusion of the recording electrode with the drug dissolved in Hepes-buffered saline of the following composition (mm): Hepes, 10; Na⁺, 146; K⁺, 4.7; Ca²⁺, 2; Mg²⁺, 1.2; Cl⁻, 157.1; glucose, 7.8. The pH of this solution was adjusted to 7.4 using NaOH. Internal perfusion of the recording electrode was achieved by inserting a fine plastic tube to within 200 μm of the electrode tip (see Brock & Cunnane, 1995).

Data analysis

A MacLab data acquisition system (ADInstruments Pty Ltd, Castle Hill, NSW, Australia) was used to digitize (sampling frequency 20 kHz) the electrophysiological signals previously recorded on tape. Prior to digitizing, the signals were filtered using a low-pass filter (cut-off, 3-5 kHz). Subsequent analysis was made with the computer program Igor Pro (Wavemetrics, Lake Oswego, OR, USA). All NTIs analysed were averages of 25-100 records. Prior to averaging, NTIs were aligned at their point of maximum rate of rise or fall.

To make comparisons between spontaneously occurring and electrically evoked NTIs, and between NTIs recorded from polymodal nociceptors and cold-sensitive receptors, the positive- and negative-peak amplitude of the NTI and the maximum rate of change of voltage during the initial upstroke and the downstroke of the NTI (+dV/dt max and -dV/dt max) were measured. The NTIs recorded from each type of receptor had similar time courses but their amplitudes varied considerably between experiments. In addition, drug treatment changed both NTI amplitude and time course. Therefore, to compare the time course of the NTIs, the maximum rates of change of voltage were normalized with respect to NTI amplitude by dividing them by the positive-peak amplitude. In all figures the NTIs are displayed together with their first derivative with respect to time to reveal differences in their time course.

All data are presented as means ±s.e.m. Statistical comparisons were made with Student's t tests for paired or unpaired observations as indicated. P values < 0.05 were considered significant.

RESULTS

Comparison of electrically evoked and spontaneous NTIs

Figure 1 shows examples of electrically evoked and spontaneous NTIs recorded from a polymodal nociceptor and a cold-sensitive receptor. In general, the configuration of all NTIs was similar, being diphasic (positive-negative) with a prominent positive component. Table 1 shows the mean measurements of NTI amplitude and time course (see Methods) for the electrically evoked and spontaneous NTIs recorded from 36 polymodal nociceptors and 51 cold-sensitive receptors. For both types of receptor, the mean positive-peak amplitude of the spontaneous NTIs was significantly smaller than that of the electrically evoked NTIs (Fig. 1a and B). In addition, there were small but consistent differences in the time course of the spontaneous and electrically evoked NTIs recorded from both polymodal and cold-sensitive receptors. For the polymodal nociceptors, the normalized maximum rate of change of voltage during the initial upstroke was slower for the spontaneous NTIs than for the electrically evoked NTIs. In contrast, for cold-sensitive receptors the normalized maximum rate of change of voltage during the initial upstroke and the ratios between the maximum rate of change of voltage during the initial upstroke and the downstroke of the NTI were greater for the spontaneous NTIs compared to those of the electrically evoked NTIs.

Figure 1. NTIs recorded from polymodal and cold-sensitive nerve endings.

NTIs recorded from a polymodal nociceptor (A) and a cold-sensitive receptor (B). Averages of spontaneous (a) and electrically evoked NTIs (c) and their first derivative with respect to time (b and d) are shown.

Table 1.

The amplitude and time course of evoked and spontaneous NTIs recorded from polymodal and cold-sensitive receptors

	n	Conduction velocity (m s−1)	+Peak(μV)	−Peak(μV)	+dV/dt max(normalized)	−dV/dt max(normalized)	Ratio
Polymodal
Evoked	36	1.1 ± 0.1	40 ± 3	−9 ± 1	2900 ± 127	−4119 ± 172	0.74 ± 0.05
Spontaneous	36	—	34 ± 2	−7 ± 1	2683 ± 120	−4186 ± 212	0.70 ± 0.06
Difference	—	—	5 ± 1**	−1 ± 1	211 ± 108*	65 ± 122	0.04 ± 0.02
Cold
Evoked	51	1.7 ± 0.1††	50 ± 4†	−8 ± 1	3260 ± 103††	−3449 ± 173††	0.99 ± 0.05
Spontaneous	51	—	47 ± 4††	−9 ± 1	3386 ± 107††	−3450 ± 179††	1.07 ± 0.05
Difference	—	—	3 ± 1**	1 ± 1	−146 ± 89*	−126 ± 59	−0.08 ± 0.03**

^*The maximum rates (V s⁻¹) of rise and fall were normalized with respect to NTI amplitude by dividing them by the peak-positive amplitude (see Methods). The ratio was calculated by dividing +dV/dt max by –dV/dt max. The difference was calculated by subtracting the values for the spontaneous NTIs from those for the evoked NTIs. The significance of the difference was calculated using one-sided, paired t tests (P < 0.05)

^**The maximum rates (V s⁻¹) of rise and fall were normalized with respect to NTI amplitude by dividing them by the peak-positive amplitude (see Methods). The ratio was calculated by dividing +dV/dt max by −dV/dt max. The difference was calculated by subtracting the values for the spontaneous NTIs from those for the evoked NTIs. The significance of the difference was calculated using one-sided, paired t tests (P < 0.01).

^†Statistical comparisons between the evoked or spontaneous NTIs from the polymodal and cold-sensitive receptors were made using unpaired t tests (P < 0.05)

^††Statistical comparisons between the evoked or spontaneous NTIs from the polymodal and cold-sensitive receptors were made using unpaired t tests (P < 0.01).

Comparison between polymodal and cold-sensitive receptor NTIs

Statistical comparisons between NTIs recorded from polymodal nociceptors and cold-sensitive receptors revealed a number of differences (see Table 1). For both the spontaneous and electrically evoked NTIs, the positive-peak amplitude of the NTIs recorded from polymodal nociceptors was smaller than that of the cold-sensitive receptors. In addition, the normalized maximum rate of change of voltage during the initial upstroke of both the spontaneous and electrically evoked NTIs was slower for the polymodal nociceptors than for the cold-sensitive receptors. Conversely, the normalized maximum rate of change of voltage during the downstroke of the NTI was faster for the polymodal nociceptors than for the cold-sensitive receptors. These differences in time course are clearly revealed by the ratios between the maximum rate of change of voltage during the initial upstroke and the downstroke of the NTI (see Table 1). These ratios show that in polymodal nociceptors the maximum rate of change of voltage during the downstroke of the NTI was faster than during the initial upstroke (Fig. 1a), whereas in cold-sensitive receptors the maximum rate of change of voltage during the initial upstroke and the downstroke of the NTI were very similar (Fig. 1B). The estimated conduction velocity of the electrically evoked NTIs recorded from polymodal nociceptors was significantly slower than that for the cold-sensitive receptors (Table 1).

Effects of bath-applied TTX

Bath application of TTX (1 μm) for periods greater than 30 min has previously been reported to abolish electrically evoked NTIs in polymodal nociceptors and cold-sensitive receptors but not to prevent the ongoing NTIs or those evoked by chemical, mechanical or thermal stimulation (see Brock et al. 1998). This finding indicates a major role for TTX-resistant Na⁺ channels in activation of the sensory nerve terminals. TTX did change the shape of the ongoing NTIs recorded from polymodal nociceptors and cold-sensitive receptors (Fig. 2a and B), indicating that TTX-sensitive Na⁺ channels are present and normally contribute to the action potential. Table 2 shows the relative changes in the spontaneous NTI amplitude and time course produced by TTX (1 μm). For the NTIs recorded from both polymodal nociceptors and cold-sensitive receptors, application of TTX reduced the positive-peak amplitude of the NTI and slowed the normalized maximum rate of change of voltage during the downstroke of the NTI (see Fig. 2a and B). TTX also reduced the negative-peak amplitude of the NTIs recorded from polymodal receptors but that for the cold-sensitive receptors was unchanged. In both polymodal nociceptors and cold-sensitive receptors, the change in time course produced an increase in the ratio between the maximum rate of change of voltage during the initial upstroke and the downstroke of the NTI.

Figure 2. Effects of bath application of TTX (1 μm) on the configuration of spontaneous NTIs.

NTIs recorded from a polymodal nociceptor (A) and a cold-sensitive receptor (B). A and B show averages of NTIs recorded before (a) and 30 min after application of TTX (c) and their first derivatives with respect to time (b and d).

Table 2.

Effects of bath applied 1 μM TTX on the amplitude and time course of the NTIs recorded from polymodal and cold-sensitive receptors

	n	+Peak	−Peak	+dV/dt max (normalized)	−dV/dt max (normalized)	Ratio
Polymodal	11	0.67 ± 0.04**	0.72 ± 0.12*	1.20 ± 0.13	0.73 ± 0.09**	2.13 ± 0.50*
Cold	14	0.84 ± 0.06**	1.01 ± 0.18	1.35 ± 0.27	0.79 ± 0.11*	1.72 ± 0.17**

All values are the mean ±s.e.m. of the test (1 μM TTX) values expressed relative to the control values. Prior to calculating the relative change in the maximum rates (V s⁻¹) of rise and fall, the values were normalized with respect to NTI amplitude by dividing them by the peak-positive amplitude (see Methods). The ratio was calculated by dividing +dV/dt max by −dV/dt max. Statistical comparisons were made between control and test values using one-sided, paired t tests

^*(P < 0.05)

^**(P < 0.01).

Effects of bath applied lignocaine

The local anaesthetic lignocaine blocks both TTX-sensitive and TTX-resistant Na⁺ channels (Roy & Narahashi, 1992). Bath application of lignocaine (1-5 mm) abolished both the electrically evoked and ongoing NTIs recorded from both polymodal nociceptors (n = 6) and cold-sensitive receptors (n = 6) within 5-10 min. In the presence of lignocaine, polymodal nociceptors and cold-sensitive receptors failed to respond to capsaicin and thermal stimuli, respectively.

Effects of locally applied lignocaine

To investigate whether the nerve terminals of polymodal nociceptors and cold-sensitive receptors are able to support regenerative action potentials or are passively invaded by signals propagated electrotonically from a point more proximal in the axon where the action potential fails or is initiated, the effects of locally applying lignocaine by internal perfusion of the recording electrode were investigated. In actively invaded nerve terminals, it would be expected that the rate of change of potential during the downstroke of the NTI would be increased by the activation of the Na⁺ channels (see Discussion) and that local application of lignocaine would produce a slowing of this component of the signal. If the nerve terminal is passively invaded by the electrotonic spread of potential from a point more proximal in the axon where the action potential fails, local application of lignocaine would be expected to have little effect on the time course of the NTI.

For the electrically evoked NTIs recorded from polymodal nociceptors, local application of lignocaine (2.5 and 20 mm) for 10-20 min reduced the positive-peak amplitude and markedly slowed the normalized maximum rate of change of voltage during the downstroke of the NTI (Table 3, Fig. 3a). In 2.5 mm lignocaine, there was a reduction in the normalized maximum rate of change of voltage during the initial upstroke of the NTI. In 20 mm lignocaine, the negative-peak amplitude was reduced and the ratio between the maximum rate of change of voltage during the initial upstroke and the downstroke of the NTI was increased. Locally applied lignocaine also had similar effects on the spontaneous NTIs (Fig. 3a) but, in most experiments investigating the effects of locally applied lignocaine on polymodal nociceptors, the frequency of spontaneous NTIs under control conditions and in the presence of lignocaine was too low to analyse the effects on their amplitude and time course. For the electrically evoked NTIs, the effect of 20 mm lignocaine on the normalized maximum rate of change of voltage during the downstroke of the NTI was significantly greater than that of 2.5 mm lignocaine (Table 3).

Table 3.

Effects of locally applied lignocaine on the amplitude and time course of the NTIs recorded from polymodal and cold-sensitive receptors

	n	+Peak	−Peak	+dV/dt max(normalized)	−dV/dt max(normalized)	Ratio
Polymodal(electrically evoked)
Hepes	6	0.96 ± 0.04	0.94 ± 0.05	1.04 ± 0.05	1.04 ± 0.10	1.02 ± 0.16
2.5 mM lignocaine	7	0.87 ± 0.06*	0.98 ± 0.24	0.86 ± 0.04*	0.75 ± 0.04**	1.30 ± 0.15
20mM lignocaine	6	0.80 ± 0.09*	0.67 ± 0.05**†	1.21 ± 0.19	0.56 ± 0.07**†	2.28 ± 0.47*†
Cold (spontaneous)
Hepes	7	0.99 ± 0.04	0.95 ± 0.04	1.13 ± 0.07	1.16 ± 0.09	0.99 ± 0.06
2.5 mM lignocaine	9	0.92 ± 0.04*	0.81 ± 0.07*	0.93 ± 0.05	0.91 ± 0.07	1.12 ± 0.11
20mM lignocaine	7	0.77 ± 0.06**†	0.75 ± 0.10*	0.96 ± 0.05	1.09 ± 0.15	0.98 ± 0.13

^*All values are the mean ±s.e.m. of the test values expressed relative to the control values. Prior to calculating the relative change in the maximum rates (V s⁻¹) of rise and fall, the values were normalized with respect to NTI amplitude by dividing them by the peak-positive amplitude (see Methods). The ratio was calculated by dividing +dV/dt max by −dV/dt max. Statistical comparisons between control and test values were made using one-sided, paired t tests (P < 0.05)

^**All values are the mean ±s.e.m. of the test values expressed relative to the control values. Prior to calculating the relative change in the maximum rates (V s⁻¹) of rise and fall, the values were normalized with respect to NTI amplitude by dividing them by the peak-positive amplitude (see Methods). The ratio was calculated by dividing +dV/dt max by −dV/dt max. Statistical comparisons between control and test values were made using one-sided, paired t tests (P < 0.01).

^†Statistical comparisons between the effects of 2.5 and 20mM lignocaine on NTIs were made using unpaired t tests (P < 0.05).

Figure 3. Effects of local application of lignocaine (20 mm) on the configuration of NTIs.

NTIs recorded from a polymodal nociceptor (A) and a cold-sensitive receptor (B). A shows averages of electrically evoked (a and c) and spontaneous NTIs (e and g) recorded before (a and e) and during lignocaine application (c and g) and their first derivatives with respect to time (b, d, f and h). B shows averages of spontaneous NTIs recorded before (a) and during lignocaine application (c) and their first derivatives with respect to time (b and d).

Local application of lignocaine (2.5 and 20 mm for 10-20 min) to cold-sensitive receptors produced a decrease in the positive- and negative-peak amplitudes of the spontaneously occurring NTIs (Table 3, Fig. 3B). However, lignocaine at both 2.5 and 20 mm had no effect on the normalized maximum rate of change of voltage during the initial upstroke or the downstroke of the NTI. In addition, the ratio between the maximum rate of change of voltage during the initial upstroke and the downstroke of the NTI was not significantly affected by lignocaine. Similar effects were also observed for the electrically evoked NTIs (results not shown). The effects of 20 mm lignocaine on the positive-peak amplitude were significantly greater than those of 2.5 mm, but no other significant differences between the drug concentrations were detected.

Perfusion of the recording electrode with the vehicle alone (i.e. Hepes-buffered saline) had no significant effect on the amplitude or time course of NTIs recorded from polymodal nociceptors or cold-sensitive receptors (Table 3).

DISCUSSION

Differences between NTIs in polymodal and cold-sensitive receptors

All the NTIs recorded from corneal polymodal nociceptors and cold-sensitive receptor nerve terminals had a diphasic (positive-negative) shape, with a prominent positive component. Comparison between the NTIs recorded from polymodal nociceptors and cold-sensitive receptors did, however, reveal two major differences. Firstly, the average positive-peak amplitude of NTIs recorded from polymodal nociceptors was smaller than that recorded from cold-sensitive receptors. This difference in amplitude could reflect the spatial relationship between the nerve terminals and the sites of recording, with polymodal nociceptors terminating farther from the epithelial surface than the cold-sensitive receptors. Alternatively, the membrane currents generated at polymodal nociceptor terminals may be smaller than those at cold-sensitive receptor nerve terminals. This possibility may reflect structural and/or functional differences between the two types of nerve ending. Secondly, the NTIs recorded from polymodal nociceptors had faster maximum rates of change of voltage during the downstroke of the NTI than during the initial upstroke, while in cold-sensitive receptors the maximum rates of change of voltage during the initial upstroke and the downstroke of the NTI were very similar. This difference in time course suggests that the currents underlying the NTIs may differ between polymodal nociceptors and cold-sensitive receptor nerve endings. To appreciate the possible reasons for this difference in time course it is necessary to consider the factors that determine the configuration of the NTIs.

Diphasic NTIs similar in shape to those recorded in the present study have been recorded extracellularly from motor nerve terminals (Dudel, 1963; Katz & Miledi, 1965) and synaptic terminations in the CNS (Brooks & Eccles, 1947), where their configuration can be explained by either active or passive invasion of the nerve terminal (see Smith, 1988). In both cases the recorded signal is proportional to the membrane current at the site of recording, positive and negative signals being produced by net outward and inward current, respectively (see Fig. 4). The membrane current consists of both capacitive and ionic components. If the nerve terminal is passively invaded, the NTI is produced by the electrotonic spread of potential from a point more proximal in the axon where the action potential fails. In this case, where it is assumed there is no excitable ionic current, the membrane current will be composed of a capacitive and a resistive ionic component and the NTI will be positive during depolarization of the nerve terminal and negative during repolarization (Fig. 4a). If the nerve terminal is actively invaded there is an additional excitable ionic current and this will be reflected primarily in a speeding of the rate of the downstroke of the NTI, which reflects the inward Na⁺ current at the site of recording (Fig. 4B, t₂). The later currents (i.e. K⁺ and Ca²⁺ currents) in the action potential are poorly reflected in the NTI because they are slower in activation and because, for much of the decaying phase of the action potential, the magnitudes of the inward and outward ionic currents are closely matched, leading to little net current across the membrane (Fig. 4B, t₃).

Figure 4. Schematic diagram illustrating the differences between NTIs generated by passive and active invasion of a nerve terminal.

In A, the action potential fails at a point more proximal in the axon and the recorded NTI is produced by the electrotonic spread of potential (a). In this case the NTI (b) will be positive during depolarization of the nerve terminal (t₁) and negative during repolarization (t₂). The + and - symbols in panels t₁ and t₂ indicate the relative potential difference between the site of action potential failure (hatched area) and the nerve terminal. In B the nerve terminal is actively invaded by an action potential (a). The initial part of the upstroke of the NTI (b) reflects an outward current produced by the electrotonic spread of potential from the more proximal arriving action potential (t₁). The remaining positive part of the NTI reflects depolarization of the membrane potential produced by the rapid activation of Na⁺ channels at the site of recording (t₂). During t₂ the net current is initially outward because the capacitive current is dominant. However, as the rate of change of membrane potential slows, the inward Na⁺ current produces the rapid downstroke of the NTI. During repolarization (t₃) the Na⁺ and K⁺ currents are closely matched leading to little net movement of current across the nerve terminal membrane.

The findings with bath-applied TTX and lignocaine confirm that action potentials propagating in the pre-terminal axons are supported by Na⁺ influx and that TTX-resistant Na⁺ channels alone are able to support action potential generation in the sensory nerve terminals (see also Brock et al. 1998). However, when bath applied, the Na⁺ channel-blocking agents will affect the entire nerve axon, so it is not possible to discern whether the observed effects on NTI configuration are due to local blockade of Na⁺ influx at nerve endings or to blockade at a point more proximal in the axon. The possibility that NTIs in polymodal nociceptors and cold-sensitive receptors are generated by voltage-dependent Ca²⁺ influx can be excluded as, in the presence of TTX, blockade of Ca²⁺ entry with Cd²⁺ did not prevent their occurrence (Brock et al. 1998).

Perfusing the recording electrode with lignocaine is expected to restrict the effects of the local anaesthetic to the area of the membrane where the signals recorded are being generated. Local application of lignocaine to polymodal nociceptors produced a marked slowing of the normalized maximum rate of change of voltage during the downstroke of the NTI. This change is consistent with a blockade of Na⁺ influx at the site of recording and provides clear evidence that the terminals of polymodal nociceptors are able to support regenerative, Na⁺-dependent action potentials. In contrast, local application of lignocaine to cold-sensitive receptors did not significantly change the normalized maximum rates of change of voltage during either the initial upstroke or the downstroke of the NTI. The simplest explanation for this finding is that the nerve terminals of the cold-sensitive receptors are passively invaded and that lignocaine applied through the recording electrode did not reach the point of action potential failure. Locally applied lignocaine did reduce both the positive- and negative-peak amplitudes of the cold-sensitive receptor NTIs. This finding demonstrates that locally applied lignocaine did reach the cold-sensitive receptor nerve terminals and can be explained if this agent reduces the leak conductance of the nerve terminal membrane, as has been reported for crayfish stretch receptor neurones (Lin & Rydqvist, 1999).

These observations have important implications for the functioning of the sensory nerve terminals. The present findings do not resolve the site of action potential initiation in the polymodal nociceptors, which may be located proximal to the nerve terminal where generator potentials spreading from a number of sensory endings are integrated and where there is a high safety factor for triggering action potentials. However, once initiated, action potentials can propagate both orthodromically to provide information to the CNS and antidromically to trigger the secretion of neuropeptides from the nerve terminals. Previous studies have reported that the release of neuropeptides produced by chemical activation of capsaicin-sensitive primary afferent nerve terminals is unaffected by TTX and local anaesthetics (e.g. see Maggi, 1991; Del Bianco et al. 1994; Szolcsányi et al. 1998). This neuropeptide release is presumably evoked by the chemically induced depolarization of the nerve terminals and does not depend on action potentials. Action potentials can also trigger neuropeptide release from the sensory nerve terminals and this is neccessary for axon reflexes (see Maggi, 1991). Importantly, previous studies on sympathetic nerve terminals have demonstrated that local application of TTX blocks active propagation of nerve action potentials and that electrotonically conducted nerve impulses spreading from the site of action potential failure are unable to elicit neurotransmitter release (Brock & Cunnane, 1988; Åstrand & Stjärne, 1989). If a similar situation occurs in polymodal nociceptors, the ability of action potentials to propagate actively into the nerve endings may be important for their efferent function.

The findings in cold-sensitive receptors suggest that there are not sufficient numbers of Na⁺ channels available in the nerve terminal membrane to support action potentials. This finding may indicate that the level of expression of Na⁺ channels in the nerve terminal membrane is lower than that found more proximally in the axon and may reflect a specialization of axonal membrane in the receptive region of the neurone. Alternatively, the sensory endings of cold-sensitive receptors may have a relatively low membrane potential and as a result most of the Na⁺ channels present in the nerve terminal membrane are inactivated. The spread of depolarization proximally along the axon to a point that is excitable may generate the relatively high levels of ongoing activity recorded in cold-sensitive receptors. Indeed, a low membrane potential in the receptive nerve endings may be integral to the function of these cold-sensitive neurones; i.e. if heating decreases the level of activity by hyperpolarizing the nerve terminals and cooling increases the level of activity by depolarizing the nerve terminals.

Differences between spontaneous and electrically evoked NTIs

In both polymodal nociceptors and cold-sensitive receptors there were differences between the spontaneous and electrically evoked NTIs. In both cases, the positive-peak amplitude of the spontaneous NTIs was smaller than that of the electrically evoked NTIs and there were consistent differences in their time course. If the spontaneous action potentials are generated at a point in the nerve terminal arbor electrotonically distant from the site of recording and conducted antidromically to the site of recording, no differences would be expected between the spontaneous and electrically evoked NTIs. However, if the site of spontaneous action potential initiation is electrotonically close to the site of recording, the smaller amplitude of the spontaneous NTIs can be explained if the axonal membrane is depolarized by the generator potential at the time of their arrival in the nerve terminal.

In summary, the present results indicate that free nerve endings of mammalian primary sensory neurones subserving separate stimulus modalities possess different membrane characteristics associated with the propagation of nerve impulses in their peripheral nerve terminals.