| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Instituto de Neurociencias, Universidad Miguel Hernández-Consejo Superior de Investigaciones, 03550, San Juan Alicante, Spain
Submitted 27 March 2003; accepted in final form 6 June 2003
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONDISCLOSURESACKNOWLEDGMENTSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONDISCLOSURESACKNOWLEDGMENTSREFERENCES |
|---|
; Spray 1986). Cold sensory fibers exhibit a rhythmic, ongoing discharge at neutral temperatures that increases in response to temperature reductions and is suppressed by warming (Brock et al. 2001; Hensel and Iggo 1971; Hensel and Zotterman 1951a; Kenshalo and Duclaux 1977). The information transmitted by these afferents in the form of spike trains is propagated along axons to central neurons, leading in humans to cooling sensations (Acosta et al. 2001; Craig et al. 2000; Yarnitsky and Ochoa 1991).
Recently, using intracellular calcium imaging, several groups were able to identify peripheral sensory neurons responding to cooling in culture (McKemy et al. 2002; Okazawa et al. 2002; Reid and Flonta 2001b; Suto and Gotoh 1999; Viana et al. 2002). These cold-sensitive neurons form a distinct subpopulation among primary sensory neurons with a characteristic set of electrophysiological properties (Viana et al. 2002). Their distinct properties include a relatively high expression of Ih channels (Mayer and Westbrook 1983) and low expression of a voltage-dependent, 4-aminopyridine (4-AP)-sensitive potassium conductance (IKD) (Storm 1987), which in other types of primary sensory neurons acts as an excitability brake and appears to be important in establishing the response specificity of cold-sensitive neurons to cooling (Viana et al. 2002). In cold-sensitive neurons, cooling causes the closure of a background K+ current (Reid and Flonta 2001a; Viana et al. 2002), leading to threshold depolarization and impulse firing (Viana et al. 2002). Cold-sensitive neurons can also be distinguished by their selective excitation by menthol, a well-known cooling agent (McKemy et al. 2002; Reid and Flonta 2001b; Viana et al. 2002). The effect of menthol is mediated by the opening of a nonselective cation channel (Reid and Flonta 2001b) of the transient receptor potential (TRP) family (McKemy et al. 2002; Peier et al. 2002). In cold-sensitive neurons, cooling seems to open the same TRP channel that is activated by menthol (McKemy et al. 2002; Okazawa et al. 2002; Peier et al. 2002; Reid and Flonta 2001b); thus it has been proposed that this channel would be responsible for their sensitivity to low temperatures.
The relative significance of these different mechanisms for the transduction of cold stimuli and the generation of propagated impulses at cold thermoreceptor endings remains unclear. Furthermore, they have only been identified in the soma of dissociated primary sensory neurons kept in culture. These cultured cells develop numerous outgrowing branches and can modify the expression of channels and receptors compared with those present in the soma of intact cells (Stebbing et al. 1998). Therefore it is not firmly established whether membrane mechanisms involved in thermal transduction at the peripheral endings (Brock et al. 2001) are equivalent to those in the soma of intact and cultured cold-sensitive neurons.
To address this question, we examined the electrophysiological responses to cooling of sensory neurons in an in vitro preparation of the excised trigeminal ganglion in which cell bodies were intact except for the effects of the acute axotomy, performed at a long distance from the soma. We found that cold stimuli were unable to induce propagated impulse discharges in the soma of intact trigeminal ganglion neurons, in contrast with their marked effects on cultured ganglion neurons or on peripheral nerve endings. Furthermore, in a specific subpopulation of neurons, blockade of K+ conductances with 4-AP induced thermal sensitivity. These data suggest that generation of propagated impulses by cold stimuli differs along the different compartments (soma vs. axon) of a sensory neuron and may reflect the differential membrane expression of specific ionic channels.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONDISCLOSURESACKNOWLEDGMENTSREFERENCES |
|---|
Experiments were performed in an in vitro preparation of the whole trigeminal ganglion (TG) obtained from young albino guinea-pigs (Hartley strain, 150–250 g). They were conducted according to EV animal use guidelines. Animals were deeply anesthetized with pentobarbital sodium (Nembutal, 90 mg/kg, ip) and perfused through the heart with cold, oxygenated physiological saline of the following composition (in mM): 128 NaCl, 5 KCl, 2.5 CaCl2, 1 MgCl2, 2.5 NaH2PO4, 26 NaHCO3, and 10 glucose (320–330 mOsm). This solution was bubbled with carbogen (95% O2-5% CO2) and maintained at 35°C. Immediately after decapitation, the head was placed in ice-cold physiological saline, and both TG, together with their central and maxillary roots, were dissected free of surrounding tissues. Thereafter, one ganglion was placed in a recording chamber (1 ml volume) and continuously perfused at high flow rate (7 ml/min) with saline solution. The recording time-window span from 1–6 h after excision. The second ganglion was stored in ice-cold oxygenated saline for 
6 h before recordings were started. No differences were observed in results obtained from either ganglion and the data were pooled together. Temperature reductions (35 to 25°C in <1 min; monitored with a thermistor placed in the immediate vicinity of the ganglion) were obtained with a Peltier device that cooled the solution flowing into the recording chamber. In some experiments, the perfusion was changed to solutions containing 4-AP (100 µM, Sigma), l-menthol (100 µM, Scharlau), or cesium (3 mM, Sigma).
The ganglion was illuminated tangentially with a fine fiber optic light source (F-O-Lite, WPI, Sarasota, FL) that produced a Nomarski-like image of the ganglion surface, viewed through the optics of an upright microscope (Optiphot-2, Nikon, Tokyo). The cut ends of the central and peripheral (maxillary) roots of the TG nerve were inserted into tight suction electrodes for electrical stimulation, using a constant current stimulator (model S48) and a stimulus isolation unit (model PSIU6, both from Grass Instruments, Quincy, MA). To calculate central and peripheral conduction velocities (CV), the distance between the tip of the suction electrode attached to the central or maxillary roots and the microelectrode tip was measured, and this value divided by the latency of the orthodromic or antidromic action potential (AP). The mean length of the central and peripheral segments of TG nerve were 4.8 ± 0.2 (n = 20) and 9.3 ± 0.4 mm (n = 20), respectively. Distances were measured with a calibrated ruler attached to the ocular of the microscope.
Electrical recording
Intracellular current-clamp recordings were obtained from neurons using glass microelectrodes (1.0 mm OD and 0.5 mm ID, Harvard Apparatus, Kent, UK) filled with 3 M KCl (tip resistance, 20–70 M
). All recorded neurons were located on the surface of the ganglion and impaled under visual guidance. Recordings were obtained using an Axoclamp-1A amplifier (Axon Instruments, Union City, CA). Voltage and current records were stored either on video-tape or digitized with a 1401 A-D converter (Cambridge Electronic Design, Cambridge, UK) and transferred to a PC. Cells were accepted only if their resting membrane potential was more negative than–50 mV and the evoked AP had an amplitude of 
60 mV.
Experimental protocols and data analysis
Membrane and firing properties were investigated at the physiological temperature of 35°C and during cooling. The following basic electrical properties were measured: resting membrane potential (Vm), input resistance (Rin), and membrane time constant (
). Rin was calculated from the slope (regression line) of the peak current-voltage (I/V) relationship, following injection of hyperpolarizing current pulses (duration 200 or 300 ms). The value of was obtained from the single exponential fit to the onset phase of a small hyperpolarizing voltage response. An inward rectification index in the voltage response to hyperpolarizing pulses was calculated according to the relationship: [(Rin at peak)–(Rin at steady state)/(Rin at peak)] x 100. Neurons showing an inward rectification index larger that 5% were classified as "rectifying" neurons. The following parameters of the AP were analyzed: AP amplitude, AP duration at 50% amplitude, AP maximum rate of rise and AP maximum rate of fall, amplitude of the afterhyperpolarization (AHP), and duration of AHP at 75% of maximal amplitude. Neurons were classified into F-type and S-type based on the absence or presence of a hump on the falling phase of the AP. Neurons with a maximum repolarizing rate of the AP <120 V/s were classified as S and those with a repolarizing rate >120 V/s were classified as F. The pattern of AP discharge to long (200 or 300 ms) depolarizing current pulses was also analyzed. Neurons were tested at a current intensity value twice the AP threshold. The injection of pulses was computer-controlled with the Strathclyde Whole Cell Program V 2.3. (J. Dempster, University of Strathclyde). To compare the AP parameters and Rin during cooling with those obtained at 35°C, a DC current was injected to correct any Vm change induced by temperature.
The effects of temperature on spontaneous activity in control conditions or during pharmacological manipulations were investigated and analyzed using Spike 2 (Cambridge Electronic Design). A change in membrane potential during cooling was considered significant when it was reversible and exceeded 1 mV.
Data are presented as mean ± SE. Statistical comparisons were made using Student's t-test (paired t-test or unpaired). The Z-test was used for comparison of proportions (Sigmastat 2.0, Jandel Scientific Software, Erkrath, Germany).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONDISCLOSURESACKNOWLEDGMENTSREFERENCES |
|---|
Electrophysiological types of TG neurons
Intracellular recordings confirmed that the electrophysiological properties of guinea pig TG neurons are not homogeneous (Hsiung and Puil 1990). In a random sample of recordings, two groups of neurons were distinguished according to the shape of their AP (Gallego and Eyzaguirre 1978; López de Armentia et al. 2000; Waddell and Lawson 1990): F-type (fast) neurons (55%, 60/110) with small-amplitude, narrow APs (Fig. 1A), and S-type (slow) neurons (45%, 50/110), characterized by longer duration APs with a "hump" on the repolarization phase (Fig. 1B). Electrophysiological differences between guinea pig F- and S-type neurons are summarized in Table 1 and are similar to those reported previously in the mouse (Cabanes et al. 2002; López de Armentia et al. 2000).
|
|
Figure 1, C and D, shows the relation between AP duration and peripheral or central CV for individual S-type (closed triangles) and F-type (open circles) neurons. Axonal CVs of F- and S-type neurons were clearly segregated. F-type neurons had faster peripheral and central mean CV (Table 1), suggesting that the majority were myelinated while most S-type neurons had CVs typical of unmyelinated fibers (CV < 1.5 m/s at 35°C). An additional difference between F- and S-type neurons was the incidence and magnitude of the time-dependent inward rectification (IR) in response to hyperpolarizing pulses. F-type neurons had a higher incidence of IR (92%) compared with S-type neurons (38%; P < 0.001, Z-test). Furthermore, the mean value of the IR index (see METHODS) was larger in F-type (31 ± 1.9%) than in S-type neurons (17 ± 2.6%, P < 0.001, t-test). As shown in the inset of Fig. 1A, IR was blocked reversibly (data not shown) by application of 3 mM extracellular Cs+ (Mayer and Westbrook 1983).
Effects of cooling on membrane potential and impulse firing
Under control conditions, cooling did not induce AP firing in any of the TG neurons tested (n = 108). This result was unexpected from rough estimates of cold-sensitive afferents (between 10 and 19%) present in the trigeminal nerve (Brock et al. 1998; Wang et al. 1993) and in cultured TG neurons (McKemy et al. 2002; Viana et al. 2002). A sampling bias due to the small size of cold-sensitive TG neurons (Viana et al. 2002) is unlikely to explain this result because about 30% (17/58) of the recorded cells, in which CV was measured, were unmyelinated and probably of small size.
The effects of temperature on the resting membrane potential (Vm) and impulse firing were examined in 64 F-type neurons and in 44 S-type neurons during cooling to 25 ± 1°C. Cooling induced a sustained reversible Vm depolarization in 64% of F-type (41/64) and in 59% of S-type (26/44), the magnitude of which was proportional to the degree of cooling (Fig. 2, A and B). However, in contrast with observations made in mouse cultured neurons, none of the cells reached firing threshold during cooling. All the cells tested were able to fire AP with intracellular injection of depolarizing current steps. In all remaining F-type neurons (36%, 23/64) and in a subset of S-type neurons (23%, 10/44), Vm did not change during cooling (Fig. 2, A and B). Finally, the remaining S-type neurons (18%, 8/44) hyperpolarized reversibly during cooling (Fig. 2B).
|
Common effects of cooling on electrophysiological properties of S- and F-type neurons
Temperature reductions also affected some electrophysiological properties of both S- and F-type neurons. As expected from previous studies (Bolton et al. 1981; Franz and Iggo 1968), axonal CV was reduced by cooling (Table 1; Fig. 3, A and D). Also, cooling from 35 to 25 ± 1°C caused a marked increase in the duration of the somatic AP (Table 1; Fig. 3, A and D). The prolongation of the AP was due to a slowing of both depolarization and repolarization rates (Table 1). The AHP duration of F- and S-type neurons did not change during cooling (Table 1).
|
Effects of cooling specific to F-type neurons
In contrast to the general effects of cooling on AP duration, the amplitude of the AP increased only in F-type neurons (Table 1; Fig. 3A). As a result, the marked differences in AP amplitude observed between F- and S-type neurons at 35°C were nearly abolished at 25°C (Table 1; Fig. 3, A and D).
Cooling also augmented significantly the input resistance (Rin) of F-type neurons (Table 1; Fig. 3B). Moreover, in F-type neurons, the IR in response to hyperpolarizing pulses decreased reversibly during cooling to 25°C (Fig. 3C). This change is reflected in a significant reduction of the IR index (Table 1). In contrast, Rin and IR were not significantly modified by cold in S-type neurons (Table 1; Fig. 3, E and F).
Changes in excitability and firing pattern induced by cooling were examined with injection of long depolarizing current pulses. During cooling, the rheobase current decreased markedly in F-type neurons (Table 1; Fig. 4A). In a few F-type neurons (4/60), the decrease in the threshold current was accompanied by a change in the firing pattern, changing from single spiking to burst firing (Fig. 4B). In contrast, rheobase and firing pattern did not change significantly in S-type neurons (Table 1; Fig. 4C).
|
Effects of 4-AP on membrane properties
4-AP blocks various types of voltage-gated K+ channels (Hille 2001; Storm 1987). We tested the effects of low concentrations of 4-AP on the electrophysiological properties and the thermosensitivity (see below) of the two subtypes of TG neurons. Perfusion with 4-AP (100 µM) caused an increase in AP and AHP duration in F- and S-type neurons (Table 2; Fig. 5, A and D, respectively). The longer AP duration was due primarily to a reduced rate of AP repolarization. Also, in both types of neurons, the rheobase current was significantly reduced by 4-AP (Puil et al. 1989) (Table 2; Fig. 5, B and E, respectively). In contrast, Rin, assessed with hyperpolarizing current pulses, was not modified by this low concentration of 4-AP (Table 2; Fig. 5, C and F, respectively), suggesting that 4-AP was acting primarily on depolarization-activated K+ conductances.
|
|
Induction of thermal sensitivity by 4-AP
In cultured mice TG neurons, modulation of cold sensitivity is critically determined by a 4-AP–sensitive slow transient K+ current (Viana et al. 2002). Moreover, application of 4-AP modifies the sensitivity to temperature of peripheral endings of guinea pig cold thermoreceptors (unpublished observations). Therefore we tested the effects of low concentrations of 4-AP on the sensitivity to cooling of TG neurons in intact guinea pig ganglia. Changes in excitability during cooling occurred in 61% of the TG neurons tested, and depending on the effects of 4-AP, neurons could be classified into three groups.
In 29% (19/65), cooling in the presence of 100 µM 4-AP elicited a reversible impulse discharge that did not occur during cooling under control conditions (Fig. 6, A and B). The neurons were silenced on rewarming. The effects of 4-AP on thermosensitivity were reversible on washout of the drug (data not shown). Average temperature threshold for inducing firing in 4-AP was 28 ± 1°C. Notably, firing occurred before any clear somatic membrane depolarization took place (Fig. 6B). Furthermore, somatic current injections at 35°C that produced depolarizations similar to those induced by cooling did not mimic the excitatory effects of the temperature drop. These results suggest that either the AP firing threshold at the soma was reduced by cooling or that the change in excitability during cooling occurred at a site remote from the soma.
|
In an additional 32% of TG neurons (21/65), cells were quiescent and unresponsive to cooling in control conditions (Fig. 7A) but developed ongoing activity following application of 100 µM 4-AP at 35°C (Fig. 7B, middle). In most of these neurons (90%, 19/21), the firing frequency decreased significantly with temperature reductions (Fig. 7B, top), and in all of them, firing was silenced by warming to 38–40°C (Fig. 7B). Their mean firing rate was 12 ± 14 Hz at 35°C and 7 ± 0.7 Hz at 25°C (P < 0.001, paired t-test). Interestingly, neurons of the first group, those that were silent at 35°C and fired only during cooling, had a higher Rin (65 ± 9.0 vs. 36 ± 12.0 M, P < 0.001, t-test) and lower rheobase current (1.0 ± 0.19 vs. 2.1 ± 0.29, P < 0.001, t-test) than those neurons excited by 4-AP at 35°C.
|
In the remaining 39% of neurons, extracellular application of 100 µM 4-AP did not induce thermosensitivity (Fig. 6, D and E), and cooling produced only modest effects in resting membrane potential.
Remarkably, the presence of strong time-dependent IR was a good predictor for development of thermosensitivity in the presence of 4-AP. All neurons that became thermosensitive with 4-AP had strong IR (Figs. 6C and 7D), and the majority (18/23) were F-type, conducting in the myelinated A
range (CV >1.5 m/s). Furthermore, the few (n = 3) F-type neurons lacking IR failed to develop thermosensitivity (Fig. 8). A similar correlation between presence of inward rectification and thermosensitivity in 4-AP was observed for S-type neurons. Those S-type neurons lacking IR were not affected by temperature changes during perfusion with 100 µM 4-AP (Figs. 6F and 8). In this group of S-type neurons, increasing the concentration of 4-AP to 2 mM (n = 2) or applying 2 mM 4-AP plus 10 mM TEA (n = 2) also failed to evoke activity during cooling (data not shown). In contrast, 55% of S-type neurons with IR developed thermosensitivity. A summary of the response to cooling, in the presence of 100 µM 4-AP, in the different subtypes of TG neurons is presented in Fig. 8, in the form of a cumulative histogram, showing the percentage of cells that reached firing threshold as a function of temperature.
|
We analyzed the relation between conduction velocity and cold sensitivity after 4-AP application within the group of F-type neurons. Forty percent (n = 10) of F-type neurons with a CV <5 m/s became cold-sensitive after 4-AP. In the group with CV >5 m/s, 28% (n = 18) became cold-sensitive, and all fibers with a CV >9.2 m/s (n = 7) remained insensitive. There is a trend in the data supporting that cold-sensitivity among F-type neurons is higher in the slower conducting subgroup, but the results were not statistically significant (Z-test).
Site of induction of thermal sensitivity by 4-AP
Intrasomatic injection of hyperpolarizing DC current stopped AP firing induced by 4-AP in only one of the seven neurons tested. In the remaining six neurons, this lack of effect on firing was observed even with strong somatic hyperpolarizations (to approximately–100 mV) that reduced the AHP amplitude significantly (Fig. 9A). The mean firing rate was reduced from 8.6 ± 2.1 Hz in control (no DC current injection) to 6.7 ± 2.3 Hz with–2 nA DC current (P = 0.08, n = 7). In contrast, the AHP mean amplitude was reduced from–11.0 ± 1.3 to–1.7 ± 0.6 mV (P < 0.001, n = 7). The differential sensitivity of firing frequency and AHP amplitude to intrasomatic current injection is plotted for individual cells in Fig. 9B. Notably, the discharge evoked during cooling was also insensitive to strong somatic hyperpolarization. Traces in Fig. 9, C and D, show an example of the maintained response to cold in a neuron in which the membrane potential was brought to–90 mV by continuous current injection. These results suggest that spontaneous activity induced by 4-AP and modulated by temperature was not initiated at the soma, but at some distant point along the cut axon.
|
To investigate further the anatomical site in the neuron where cooling triggers propagated impulses, an additional set of experiments was conducted where 4-AP was applied intracellularly using microelectrodes filled with high concentrations of the drug (100 mM). In none of the cells tested (n = 13) did application of 4-AP into the soma caused ongoing or cold-induced activity (Fig. 9E). This lack of thermosensitivity was not due to inadequate diffusion of 4-AP from the microelectrode into the soma, because membrane properties and AP duration showed the typical modifications induced by 4-AP in sensory neurons (Puil et al. 1989). As shown in the inset of Fig. 9E, the AP was prolonged, and the neuron developed rebound spiking (see also Fig. 5C) and fired in a burst during depolarizations. Notably, 71% (5/7) of neurons injected with 4-AP intracellularly developed spontaneous activity that was modulated by temperature when they were perfused extracellularly with 100 µM 4-AP, (Fig. 9F). All of them had strong time-dependent IR (Fig. 9E, inset). These results further suggest that the ongoing activity was generated at a site remote from the soma.
Effects of menthol on cold sensitivity
Menthol is a specific activator of cold-sensitive neurons (Hensel and Zotterman 1951a; McKemy et al. 2002; Reid and Flonta 2001b; Viana et al. 2002). We examined the effects of 100 µM menthol on 14 F-type neurons and 3 S-type neurons, all showing IR. None of them developed spontaneous activity at 35°C; neither were they activated by temperature reductions to 25°C. Furthermore, 100 µM menthol did not change the mean Vm at 35°C (–59 ± 6.5 mV in control vs.–58 ± 6.4 mV in menthol, n = 17) or the rheobase current (2.5 ± 1.2 nA in control vs. 2.7 ± 1.4 nA in menthol, n = 17).
However, application of menthol affected the response to cold in most neurons that became thermosensitive in the presence of 100 µM 4-AP. Thus in three of four neurons, 100 µM menthol caused a marked decrease in temperature threshold during cooling. A typical example is shown in Fig. 10. In 100 µM 4-AP, this neuron became active only during cooling (Fig. 9A). After addition of menthol (100 µM), the cell developed ongoing activity at 35°C (Fig. 10B), and this activity was silenced on slight warming. The excitatory effects of menthol were readily reversible (Fig. 10C). Finally, in five neurons that were insensitive to 4-AP, application of menthol had no direct excitatory effects neither did this drug induce thermal sensitivity.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONDISCLOSURESACKNOWLEDGMENTSREFERENCES |
|---|
Absence of cold-induced activity in the soma of trigeminal neurons
We found that in the guinea pig TG superfused in vitro, cooling never evoked a firing response in the soma or the cut axon of a large sample of neurons. This sample presumably included neurons possessing cold-sensitive peripheral nerve terminals in vivo. Likewise, menthol failed to activate or sensitize to cold any of the explored TG neurons in control solution. The lack of excitatory effect of these stimuli contrasts with the robust impulse discharge induced by temperature reductions or menthol application in peripheral cold nerve endings of TG neurons (Brock et al. 2001; Schäfer et al. 1986) and in mice cold-sensitive trigeminal neurons maintained in culture (McKemy et al. 2002; Viana et al. 2002). Recent studies have demonstrated the specific expression in cultured cold-sensitive primary sensory neurons of an inward current that is activated by menthol and cooling (McKemy et al. 2002; Reid and Flonta 2001b). This inward current is small and flows through a nonselective cation channel of the TRP family known as CMR1 (McKemy et al. 2002) or TRPM8 (Peier et al. 2002). Our findings suggest that the density of these channels must be low at the level of the soma. A similar discrepancy in the response to purinergic agonists between intact DRG neurons and those obtained after dissociation has been noted by Stebbing et al. (1998).
Excitatory effects of cooling are specific to thinly myelinated (F) neurons
Nonspecific influences of cooling on membrane properties of excitable cells, including primary sensory neurons and their axons, have been reported previously and were confirmed in this study (Franz and Iggo 1968; Kiyosue et al. 1993; Li et al. 2002). However, in addition, marked effects of cooling were noticed in TG neurons with strong IR (e.g., high expression levels of Ih) and short-duration spikes (F-type), the majority of them with thin myelinated axons (A). Most significant was the decrease in rheobase current (e.g., increased excitability) during cooling. With the exception of a low Rin, most membrane properties of this subpopulation of F-type neurons resembled those of cold- and menthol-sensitive mouse TG neurons in culture (Viana et al. 2002). However, they also exhibited many of the functional characteristics of primary sensory neurons identified as mechanosensitive in experiments in vivo (Djouhri and Lawson 2001; Koerber et al. 1988) and in vitro (Viana et al. 2001). The majority of cold thermoreceptor fibers are thinly myelinated in most species, including the guinea pig (Brock et al. 2001; reviewed by Hensel 1976). Also, indirect evidence from microneurographic recordings in humans suggests that cold-sensitive fibers have a prominent hyperpolarization-activated cation current (Ih) (Serra et al. 1999). The role of Ih in thermal transduction is presently unclear. The density of Ih channels is higher in cold-sensitive neurons compared with cold-insensitive neurons of similar size (Viana et al. 2002). However, pharmacological blockade of Ih does not prevent excitation during cooling (Robert et al. 2002).
Since the proportion of TG neurons that developed sensitivity to cold after 4-AP (61%) was higher than that of cold-sensitive fibers found in peripheral nerves (Dykes 1975; Hensel and Iggo 1971), only a fraction of the 4-AP–sensitive neurons could correspond to those possessing cold-sensitive peripheral terminals. It can be speculated that these were represented by the subpopulation of neurons with higher input resistance and lower rheobase, which were silent at 35°C and responded to cooling with a brisk discharge, all features typical of mouse TG neurons identified as cold-sensitive in tissue culture (Viana et al. 2002). The remaining neurons showing some degree of temperature sensitivity after 4-AP may be mechanosensory neurons. It has been reported that many cutaneous and muscle mechanoreceptors have a marked sensitivity to cooling (Duclaux and Kenshalo 1972; Hensel and Zotterman 1951b; Lippold et al. 1960). Activation of these mechanosensory neurons by cold would explain the psychophysical observation that objects with equal weight feel heavier in the hand when they are cold, known as Weber's illusion (Hensel and Zotterman 1951b; Weber 1846). In contrast to the specific effects of cold on F-type neurons, those neurons exhibiting broad spikes (S-type) and low conduction velocity, which correspond to polymodal nociceptor neurons (Djouhri and Lawson 2001; Koerber et al. 1988), only showed unspecific changes in membrane properties during cooling.
4-AP–sensitive channels modulate the excitatory effects of cooling
4-AP has a well-documented effect on the excitability of myelinated sensory axons (Honmou et al. 1994; Kocsis et al. 1986). Furthermore, it has been proposed (Viana et al. 2002) that cold-sensitive neurons in culture are characterized by a particular combination of ionic conductances, including the low expression of a voltage-dependent, 4-AP–sensitive potassium conductance (IKD), which in other classes of primary sensory neurons counteracts the net depolarizing effect of cold stimuli. This current is important in establishing the specificity of the response of cold-sensitive neurons to cooling. In the subpopulation of TG neurons displaying strong IR, most of them of the F type and conducting in the A range, submillimolar doses of 4-AP induced thermal sensitivity. This effect was not observed in neurons lacking IR, which are generally of the S type. Our new findings suggest that 4-AP–sensitive K+ conductances are also important determinants of cold sensitivity and temperature threshold in peripheral axons. In the presence of 4-AP, cooling may generate enough depolarization to evoke propagated APs. These findings may have special relevance to the development of altered responses to cooling in neuropathic pain states.
Excitatory effects of cooling occur at a site distant from the soma
Several observations suggest that impulse firing during cooling in 4-AP–treated neurons originated at a site distant from the soma. Thus firing took place without a noticeable depolarization of the membrane potential, and hyperpolarization of the soma with current injection did not prevent the generation of APs. Moreover, intracellular injection of 4-AP did not mimic the effects of extracellular application of the drug.
The ability of the axon to generate propagated impulses with cooling in contrast with the soma appears to be the consequence of a heterogeneous distribution and/or density of particular ion channels, which would be preferentially expressed and inserted at the peripheral terminals. There is a considerable traffic of ionic channels toward the periphery in sensory axons, which is interrupted by nerve ligation (Devor and Govrin-Lippman 1983; England et al. 1994; Novakovic et al. 1998). Also, anchoring proteins seem to contribute to the targeting of channels to presynaptic terminals (reviewed by Sheng and Wyszynski 1997). The geometry of nerve terminals may also play an important role in determining the response to cooling. Cold-sensitive terminals are thought to be small free nerve endings with a large surface to volume ratio (Hensel et al. 1974; Heppelmann et al. 2001). Thus small ionic currents may produce sufficient local depolarization to reach threshold at the terminal. The same may be true for the fine neurites that extend out of sensory neurons in culture, and this may explain why cooling or menthol can depolarize cultured mice and rat TG and DRG neurons beyond threshold (McKemy et al. 2002; Reid and Flonta 2002; Viana et al. 2002). In both circumstances, small currents may induce significant local depolarizations favoring the genesis and propagation of AP to the soma. This would not be the case for the soma of neurons that were axotomized prior to the experiment, as is the case for the excised ganglion. In these circumstances, depolarization induced by cooling may not evoke enough current flow to bring the neuron to the threshold level, although this was achieved in a portion of them after blockade of 4-AP–sensitive K+ conductances. In a study by Michaelis et al. (1999) in axotomized sural nerve C-fibers of the rat performed 2–24 h postaxotomy, 6.6% of fibers showed cold sensitivity. The authors did not report the proportion of fibers developing cold sensitivity in the first 6 h postaxotomy, the time frame of our recordings.
Taken together, our observations stress the complexities involved in studying the mechanism of cold transduction in intact neurons. Our findings are difficult to explain invoking just the opening by temperature decreases of a single class of cold-sensitive ionic channel and provide additional support to our previous proposal (Viana et al. 2002) that responsiveness of primary sensory neurons to cooling is critically modulated by 4-AP–sensitive K+ conductances. Our new findings suggest that channels regulated by temperature may have a preferential location at peripheral sites, where temperature is normally sensed. It can be speculated that these channels, including CMR1/TRPM8, voltage-gated K+ channels, and leak K+ channels, are heterogeneously distributed in the soma and the axonal branches of primary sensory neurons. In the peripheral terminals of neurons possessing thermosensitivity, the net effect of cold would be to produce propagated nerve impulses in the axon, at a point that is apparently located close to the ending (Brock et al. 2001). This does not occur at the level of the soma or in other classes of nerve endings sharing broadly similar membrane properties, as seems to be the case of mechanoreceptors, either because the density of IKD channels in their endings is too high or the density of leak K+ channels or CMR1 channels is too low to reach threshold depolarization. In other functional types of peripheral sensory fibers like those belonging to S-type neurons, many of which are probably polymodal nociceptors, insensitivity to cold does not appear to be associated to a 4-AP–sensitive K+ conductance and may involve other K+ channels acting as brakes or result from the absence of CMR1/TRPM8 or other temperature-sensitive channels.
|
DISCLOSURES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONDISCLOSURESACKNOWLEDGMENTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONDISCLOSURESACKNOWLEDGMENTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
Address for reprint requests and other correspondence: C. Cabanes, Instituto de Neurociencias, Universidad Miguel Hernández-CSIC 03550, San Juan de Alicante, Spain (E-mail: carmen.cabanes{at}umh.es).
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONDISCLOSURESACKNOWLEDGMENTSREFERENCES |
|---|
Bolton CF, Sawa GM, and Carter K. The effects of temperature on human compound action potentials. J Neurol Neurosurg Psychiatry 44: 407–413, 1981.[Abstract]
Brock JA, McLachlan E, and Belmonte C. Tetrodotoxin-resistant impulses in single nociceptive nerve terminals in guinea-pig cornea. J Physiol 512: 211–217, 1998.
Brock JA, Pianova S, and Belmonte C. Differences between nerve terminal impulses of polymodal nociceptors and cold sensory receptors of the guinea-pig cornea. J Physiol 533: 493–501, 2001.
Cabanes C, López de Armentia M, Viana F, and Belmonte C. Postnatal changes in membrane properties of mice trigeminal ganglion neurons. J Neurophysiol 87: 2398–2407, 2002.
Craig AD, Chen K, Bandy D, and Reiman EM. Thermosensory activation of insular cortex. Nat Neurosci 3: 184–190, 2000.[ISI][Medline]
Devor M and Govrin-Lippmann R. Axoplasmic transport block reduces ectopic impulse generation in injured peripheral nerves. Pain 16: 73–85, 1983.[ISI][Medline]
Djouhri L and Lawson SN. Differences in the size of the somatic action potential overshoot between nociceptive and non-nociceptive dorsal root ganglion neurones in the guinea pig. Neuroscience 108: 479–491, 2001.[ISI][Medline]
Duclaux R and Kenshalo DR. The temperature sensitivity of type I slowly adapting mechanoreceptors in cats and monkeys. J Physiol 224: 647–665, 1972.
Dykes RW. Coding of steady and transient temperatures by cutaneous cold fibers serving the hand of monkeys. Brain Res 98: 485–500, 1975.[ISI][Medline]
England JD, Gamboni F, Ferguson MA, and Levinson SR. Sodium channels accumulate at the tips of injured axons. Muscle Nerve 17: 593–598, 1994.[ISI][Medline]
Franz DN and Iggo A. Conduction failure in myelinated and non-myelinated axons at low temperatures. J Physiol 199: 319–345, 1968.
Gallego R and Eyzaguirre C. Membrane and action potential characteristics of A and C nodose ganglion cells studied in whole ganglia and in tissue slices. J Neurophysiol 41: 1217–1232, 1978.
Hensel H. Functional and structural basis of thermoreception. Prog Brain Res 43: 105–118, 1976.[Medline]
Hensel H, Andres KH, and von During M. Structure and function of cold receptors. Pflügers Arch 352: 1–10, 1974.[ISI][Medline]
Hensel H and Iggo A. Analysis of cutaneous warm and cold fibres in primates. Pflügers Arch 329: 1–8, 1971.[ISI][Medline]
Hensel H and Zotterman Y. The response of the cold receptors to constant cooling. Acta Physiol Scand 22: 96–105, 1951a.[ISI][Medline]
Hensel H and Zotterman Y. The response of mechanoreceptors to thermal stimulation. J Physiol 115: 16–24, 1951b.
Heppelmann B, Gallar J, Trost B, Schmidt RF, and Belmonte C. Three-dimensional reconstruction of scleral cold thermoreceptors of the cat eye. J Comp Neurol 441: 148–154, 2001.[ISI][Medline]
Hille B. Potassium channels and chloride channels. In: Ionic Channels of Excitable Membranes (3rd ed.). Sunderland, MA: Sinauer, 2001, p. 131–167.
Honmou O, Utzschneider DA, Rizzo MA, Bowe CM, Waxman SG, and Kocsis JD. Delayed depolarization and slow sodium currents in cutaneous afferents. J Neurophysiol 71: 1627–1637, 1994.
Hsiung GR and Puil E. Ionic dependencies of tetrodotoxin-resistant action potentials in trigeminal root ganglion neurons. Neuroscience 37: 115–125, 1990.[ISI][Medline]
Kenshalo DR and Duclaux R. Responses characteristics of cutaneous cold receptors in monkey. J Neurophysiol 40: 319–322, 1977.
Kiyosue T, Arita M, Muramatsu A, Spindler J, and Noble D. Ionic mechanisms of action potential prolongation at low temperature in guinea-pig ventricular myocytes. J Physiol 468: 85–106, 1993.
Kocsis JD, Bowe CM, and Waxman SG. Different effects of 4-aminopyridine on sensory and motor fibers: pathogenesis of paresthesias. Neurology 36: 117–120, 1986.
Koerber HR, Druzinsky RE, and Mendell LM. Properties of somata of spinal dorsal root ganglion cells differ according to peripheral receptor innervated. J Neurophysiol 60: 1584–1596, 1988.
Li H, Liu BG, Dobretsov M, Brull SJ, and Zhang JM. Thermosensitivity of large primary sensory neurons. Brain Res 926: 18–26, 2002.[ISI][Medline]
Lippold OCJ, Nicholls JG, and Redfearn. A study of the afferent discharge produced by cooling a mammalian muscle spindle. J Physiol 153: 218–231, 1960.
López de Armentia M, Cabanes C, and Belmonte C. Electrophysiological properties of identified trigeminal ganglion neurons innervating the cornea of the mouse. Neuroscience 101: 1109–1115, 2000.[ISI][Medline]
Mayer ML and Westbrook GL. A voltage-clamp analysis of inward (anomalous) rectification in mouse spinal sensory ganglion neurones. J Physiol 340: 19–45, 1983.
McKemy D, Neuhausser WM, and Julius D. Identification of a cold receptor reveals a general role for TRP channels in thermosensation. Nature 416: 52–58, 2002.[Medline]
Michaelis M, Blenk KH, Vogel C, and Jänig W. Distribution of sensory properties among axotomized cutaneous C-fibres in adult rats. Neuroscience 94: 7–10, 1999.[ISI][Medline]
Novakovic SD, Tzoumaka E, McGivern JG, Haraguchi M, Sangameswaran L, Gogas KR, Eglen RM, and Hunter JC. Distribution of the tetrodotoxin-resistant sodium channel PN3 in rat sensory neurons in normal and neuropathic conditions. J Neurosci 18: 2174–2187, 1998.
Okazawa M, Takao K, Hori A, Shiraki T, Matsumura K, and Kobayashi S. Ionic basis of cold receptors acting as thermostats. J Neurosci 22: 3994–4001, 2002.
Peier AM, Moqrich A, Hergarden AC, Reeve AJ, Andersson DA, Story GM, Earley TJ, Dragoni H, McIntyre P, Bevan S, and Patapoutian A. A TRP channel that senses cold stimuli and menthol. Cell 108: 705–715, 2002.[ISI][Medline]
Puil E, Miura RM, and Spigelman I. Consequences of 4-aminopyridine applications to trigeminal root ganglion neurons. J Neurophysiol 62: 810–820, 1989.
Reid G and Flonta ML. Cold transduction by inhibition of a background potassium conductance in rat primary sensory neurones. Neurosci Lett 297: 171–174, 2001a.[ISI][Medline]
Reid G and Flonta ML. Physiology. Cold current in thermoreceptive neurons. Nature 413: 480, 2001b.[Medline]
Reid G and Flonta ML. Ion channels activated by cold and menthol cultured rat dorsal root ganglion neurones. Neurosci Lett 324: 164–168, 2002.[ISI][Medline]
Robert C, Belmonte C, and Viana F. Properties and functional role of Ih current in cold-sensitive trigeminal neurons of the mouse. Soc Neurosci Abstr 449.2, 2002.
Schäfer K, Braun HA, and Isenberg C. Effect of menthol on cold receptor activity. Analysis of receptor processes. J Gen Physiol 88: 757–776, 1986.
Serra J, Campero M, Ochoa J, and Bostock H. Activity-dependent slowing of conduction differentiates functional subtypes of C fibres innervating human skin. J Physiol 515: 799–811, 1999.
Sheng M and Wyszynski M. Ion channel targeting in neurons. Bioessays 19: 847–853, 1997.[ISI][Medline]
Spray DC. Cutaneous temperature receptors. Annu Rev Physiol 48: 625–638, 1986.[ISI][Medline]
Stebbing MJ, McLachlan EM, and Sah P. Are there functional P2X receptors on cell bodies in intact dorsal root ganglia of rats? Neuroscience 86: 1235–1244, 1998.[ISI][Medline]
Storm JF. Action potential repolarization and a fast after-hyperpolarization in rat hippocampal pyramidal cells. J Physiol 385: 733–759, 1987.
Suto K and Gotoh H. Calcium signaling in cold cells studied in cultured dorsal root ganglion neurons. Neuroscience 92: 1131–1135, 1999.[ISI][Medline]
Viana F, de la Peña E, and Belmonte C. Specificity of cold thermotransduction is determined by differential ionic channel expression. Nat Neurosci 5: 254–260, 2002.[ISI][Medline]
Viana F, de la Peña E, Pecson B, Schmidt RI, and Belmonte C. Swelling-activated calcium
) and S-type (
) neurons. D: correlation between AP duration and central CV for F-type (
