| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Instituto de Biologia Molecular e Celular (IBMC), 4150–180 Porto, Portugal; 2 Department of Anatomy, Histology, and Embryology, Faculty of Medicine, Medical and Health Science Centre, University of Debrecen, Debrecen, H-4012, Hungary
Submitted 9 September 2003; accepted in final form 25 September 2003
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
; Cervero 1987; Willis and Coggeshall 1991). Substantia gelatinosa (SG) is the dorsal horn region where most fine-calibre C- and A
-fibers terminate (LaMotte 1977; Light and Perl 1977; Rethelyi 1977; Sugiura et al. 1986) and is therefore considered as an important element of nociceptive processing system. Several types of SG neurons are distinguished on the basis of their intrinsic firing properties (Grudt and Perl 2002; Lopez-Garcia and King 1994; Thomson et al. 1989; Yoshimura and Jessell 1989). Among them, tonically firing neurons (TFNs) were described as exhibiting little spiking frequency adaptation during sustained depolarization. In functional terms, a majority of TFNs was shown to be either wide-dynamic-range or nociceptive-specific neurons (Lopez-Garcia and King 1994). According to their somatodendritic organization TFNs in SG belong to several distinct morphological types (Grudt and Perl 2002), suggesting that the intrinsic firing behavior might be mostly determined by the properties of ionic conductances present in the membrane. However, the knowledge about major types of ion channels specifically expressed in TFNs as well as their role in sustained firing is insufficient. Therefore we carried out this study and created a computer model to elucidate the ionic basis of tonic firing in SG neurons. |
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
; Chery et al. 2000; Edwards et al. 1989). The animals were killed in accordance with the local guidelines. After the anesthesia by intraperitoneal injection of pentobarbital sodium (30 mg/kg) the vertebral column was quickly cut out and immersed in ice-cold oxygenated artificial cerebrospinal fluid (ACSF). The 5- to 7-mm-long segment of the lumbar enlargement was dissected and glued to the stage of the tissue slicer (Leica VT 1000S). Four parasagittal slices (2 from each half of the spinal cord) were prepared and incubated for 30–60 min in ACSF at 33°C. SG (lamina II) was identified as a translucent band in the dorsal horn. ACSF contained (in mM) 115 NaCl, 5.6 KCl, 2 CaCl2, 1 MgCl2, 11 glucose, 1 NaH2PO4, and 25 NaHCO3 (pH 7.4 when bubbled with 95% O2-5% CO2). In some cases, 2 mM kynurenic acid was added to ACSF during preparation. Low-Ca2+, high-Mg2+ solution (ACSF*) was obtained from ACSF by setting [Ca2+] to 0.1 mM and [Mg2+] to 5 mM. In some experiments K+ currents were recorded in Na+-free choline–Cl solution containing (in mM) 135 choline-Cl, 1.1 KCl, 0.1 CaCl2, 5 MgCl2, 11 glucose, and 10 HEPES. The pH value was adjusted to 7.4 by KOH (final [K+] was 5.6 mM). Apamin and charybdotoxin were dissolved in ACSF with 0.05% BSA.
Standard pipette solution contained (in mM) 6 NaCl, 128 KCl, 2 MgCl2, 10 EGTA, and 10 HEPES. The solution with low Ca2+ buffering capacity contained (in mM) 6 NaCl, 145 KCl, 2 MgCl2, 1 EGTA, and 10 HEPES. The pH value in both solutions was adjusted to 7.3 with KOH (final [K+] was 160.5 mM). Pipette solution for studying Na+ channels contained (in mM) 4 NaCl, 131 CsCl, 2 MgCl2, 10 EGTA, and 10 HEPES. The pH value was adjusted to 7.3 by CsOH (final [Cs+], 153 mM) and NaOH (final [Na+], 6 mM). All chemicals were purchased from Sigma.
The patch pipettes were pulled from thick-walled borosilicate glass tubes (Modulohm, Denmark; 1.50 mm OD/0.86 mm ID) and had a resistance of 3–5 M
after fire-polishing. The EPC-9 amplifier (HEKA, Lambrecht, Germany) was used in all experiments. The effective corner frequency of the low-pass filter was 3 kHz. The frequency of digitization was 10 kHz. Transients and leakage currents were digitally subtracted using standard P/n protocol. Offset potentials were nulled directly before formation of a seal. Liquid junction potentials were calculated and corrected for in all experiments. In neurons subjected to detailed analysis the series resistance measured in the whole cell mode was 6–20 M and was compensated by 
60%. Action potentials were recorded using the fast current-clamp mode of the EPC-9 amplifier. Input resistance (RIN) was measured in voltage-clamp mode using negative 10- to 40-mV pulses from a holding level of –80 mV. Only cells with a resting potential (VR) negative to –60 mV were included into this study.
Ion channels were studied in two types of isolated structures: nucleated patches excised from somatic membrane (Sather et al. 1992) or entire somata (Safronov et al. 1997). Nucleated patches were usually obtained from deep neurons or larger superficial neurons and had diameters ranging from 5 to 8 µm. From the majority of superficial neurons, however, the entire soma with a diameter of 8–10 µm was isolated. We did not distinguish between those two structures and in the following text will refer them to as nucleated patches. For estimation of the current densities in isolated patches their mean diameter was assumed to be 7.5 µm.
The current–voltage (I-V) relationship for Na+ channels was fitted with equation: G0 /[1 + exp((V 50 – V)/k)] x (V – VREV), where G0 is the maximum conductance, V50 is the potential of half-maximum channel activation, k is a steepness factor, and VREV is the reversal potential. VREV obtained by fitting for each I-V curve was 10–20 mV more negative than a theoretical VNa (+79 mV) due to appearance of outward Cs+ current through delayed-rectifier K+ channels incompletely blocked by 1 mM TEA. K+ conductances were calculated assuming VREV equal to VK of –84 mV, in agreement with the data shown in Fig. 4D. The activation and steady-state inactivation characteristics were fitted with Boltzmann function: 1/(1 + exp[(V50 – V)/k)]. The time course of Na+ channel recovery from inactivation was fitted with a two-exponential function: 1 – A x exp(–t/
F) – (1 – A) x exp(–t/S), where F and S are the fast and slow time constants and A is the relative amplitude of the fast component.
|
In some experiments, 0.5% biocytin (Sigma) was included in the pipette solution for later cell reconstruction. Following the recording session, the slices with biocytin-filled TFNs were transferred into a fixative containing 4% paraformaldehyde, 1.25% glutaraldehyde, and 0.2% picric acid in 0.1 M phosphate buffer (pH 7.4) for 7–10 days. After resectioning at 60 µm, slices were treated according to the avidin-biotinylated horseradish peroxidase method (Extravidin, diluted 1:1000, Vector Labs, Burlingame, CA) and the reaction was completed with a diaminobenzidine (Sigma, St. Louis, MO) chromogen reaction. Sections were counterstained with toluidine blue, dehydrated, and mounted with DPX (Fluka, Buchs, Switzerland). The somata as well as the dendritic and axonal arbors of labeled neurons were reconstructed from serial sections using a camera lucida with a x100 oil immersion objective.
Computer simulations were done using NEURON software (Hines 1993; Hines and Carnevale 1997) with an integration step of 50 µs. Model consisted of the axon initial segment (AIS) and soma connected to an equivalent dendrite. This allowed us to construct a model of TFN independent of its belonging to one of three morphological groups described below. Parameters of the equivalent dendrite were calculated using standard procedure (Rall 1959, 1969; Dodge and Cooley 1973) on the basis of our recordings from six TFNs with RIN = 1.7 ± 0.3 G. Electrotonic length of the cylinder was calculated from the equation L = 
/(0/1 – 1)0.5, where 0 = 91 ± 10 ms (n = 6) and 1 = 4.1 ± 0.3 ms (n = 6) were the two slowest membrane time constants obtained by exponential "peeling" from the averaged (500 episodes) low-amplitude (<2 mV) passive decay transients evoked in current-clamp mode by short (1 ms) hyperpolarizing current pulses (not shown). Calculated L value was 0.68, indicating a compact electrotonic structure of TFNs. Assuming a uniform membrane capacitance (Cm) of 1 µF/cm2, the specific membrane resistivity, Rm = 0/Cm, was estimated to be 91 kcm2. Using standard equations describing a passive cable, 
2 = DRmRi–1/4, RIN = 4Ri–1D–2cothL and l = L, where is a characteristic length, D is a diameter of the equivalent dendrite, Ri is an axial resistance, l is a length of the equivalent cylinder, and, assuming Ri of 80 cm (Barrett and Crill 1974; Thurbon et al., 1998) and RIN of 1.7 G, one could estimate the dimensions of the equivalent dendrite, l = 1371 µm and D = 1.4 µm. The soma was considered a 10-µm-long cylinder with a 10-µm diameter. AIS was 30 µm long and linearly tapered in diameter from 1.0 µm at base to 0.5 µm at its distal end (Gobel et al. 1980). The Cm, Rm, and Ri values in the soma and AIS were the same as in the dendrite. The soma, AIS, and equivalent dendrite consisted of 10, 30, and 50 compartments, respectively. VR in the model was –70 mV.
The models of Na+ and KDR currents were developed on the basis of our recordings. Na+ current was described by a Hodgkin–Huxley style equation gNam3h(V – VNa), where gNa was the Na+ conductance, m and h were the variables of activation and inactivation, respectively, and VNa was +60 mV. The steady-state activation variable (m
) and the time constant of activation (m) were determined as m = 
m/(m + 
m) and m = 1/(m + m), where m = 0.182(V + 33)/{1 – exp[–(V + 33)/9]} and m = –0.124(V + 33)/{1 – exp[(V + 33)/9]} were reaction rates. The time constant of inactivation (h) was determined as h = 1/(h + h) with h = –0.0018(V + 82)/{1 – exp[(V + 82)/18]} and h = 0.061(V + 46)/{1 – exp[–(V + 46)/3]} + 0.0166. The steady-state inactivation variable was given as h = 1/{1 + exp[(V + 75)/9]}. KDR current was described as gKn4h(V – VK), where gK was the KDR conductance, n was the variable of activation, and VK was –84 mV. The parameters were n = n/(n + n), n = 1/(n + n), n = 0.035(V + 15)/{1 – exp[–(V + 15)/9]}, n = 0.014exp[–(V – 12)/46], h = h/(h + h), h = 1/(h + h), h = 0.0083{1 + 1/[exp((V + 20)/10) + 1]} and h = 0.0083/{exp[–(V + 20)/10] + 1}.
KDR current density in the soma was adjusted in agreement with our present measurements. Those in the dendrite and AIS were determined by simulating the experiments from Wolff et al. (1998), which showed that the ratio of somatic/dendritic/AIS components of KDR current recorded in voltage-clamp mode with electrode placed on the soma was 15/47/38. These conditions were satisfied if gK was 34, 4.3 and 76 mmho/cm2 for the dendrite, soma, and AIS, respectively. Inactivating Na+ channels were only inserted in the soma and AIS (Safronov 1999). On the basis of our present results gNa was set to 8 mmho/cm2 for the soma. In AIS gNa was set to 1800 mmho/cm2 to reach the maximum velocity of spike depolarization of 229 V/s (mean value for 70 TFNs). At a VR of –70 mV, the density of available Na+ channels with a conductance of 11.6 pS (Safronov et al. 1997) was 0.55 µm–2 in the soma and 124 µm–2 in AIS. Since the KA current in TFNs was small and almost completely inactivated at VR, it was not included into the model.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
(n = 117). In all TFNs the firing frequency progressively increased with stimulation intensity (Fig. 1A). Frequency-current (f-I) curves for the first interspike interval (instantaneous) and the last few intervals (steady-state) as well as the instantaneous firing frequency as a function of time (f-t plot) are shown in Fig. 1B. The steady-state f-I characteristic was nonlinear with an initial slope of 0.61 ± 0.08 Hz/pA (n = 8).
|
Blockers of Ca2+ and KCA channels were tested to study the role of Ca2+-dependent conductances. In these experiments intracellular solution containing 1 mM EGTA was used. After equimolar substitution of 2 mM Ca2+ in ACSF by inorganic Ca2+ channel blockers Co2+ (n = 8) or Mn2+ (n = 5), several modifications of firing pattern were observed. At low stimulation intensities, the firing frequencies increased (Fig. 1, A and B, shown for Co2+), so that the initial slope of the steady-state f-I curve reached 1.14 ± 0.12 Hz/pA (n = 8). At strong stimulation, however, the steady-state frequency was similar to control value, but pronounced spike attenuation within the train appeared in Co2+ or Mn2+. Thus the block of Ca2+ influx into the neuron led to a left-shift in the f-I characteristics and a reduction of firing stability. The effects of Co2+ or Mn2+ on TFN firing resulted from a reduction of a slow afterhyperpolarization (Fig. 1C). Similar effects were also seen in ACSF after addition of 500 nM apamin (Fig. 1D; n = 5) but not 100 nM charybdotoxin (Fig. 1E; n = 4), indicating the involvement of small conductance apamin-sensitive, rather than big conductance KCA channels. It could be therefore concluded that 1) Ca2+-dependent conductances are involved in regulation of firing frequency in TFNs, but 2) the basic pattern of tonic firing is generated by voltage-gated Na+ and K+ channels.
The following study of Na+ and K+ channels was carried out in ACSF* (0.1 mM Ca2+–5 mM Mg2+) with 10 mM EGTA in pipette solution to minimize the contribution of Ca2+-dependent conductances. Under these conditions, TFNs showed sustained firing, but addition of 2 mM Co2+ did not shorten intervals between spikes (Fig. 1F, n = 5). Further substitution of internal EGTA with fast Ca2+ chelator BAPTA (10 mM) did not change the pattern recorded in ACSF* (n = 5; not shown). Thus a combination of external ACSF* and internal 10 mM EGTA, used in all of the following experiments, was adequate for minimizing Ca2+-dependent conductances.
An appearance of tonic firing did not depend on VR in the range between –80 and –60 mV (Fig. 1G). Hyper- or depolarization of TFN by injection of sustained current did not prevent generation of tonic firing evoked by depolarizing current pulses (n = 25). Therefore in current-clamp experiments all neurons were uniformly kept at –70 mV by injecting the holding current, which did not exceed 45 pA.
Na+ channels
Na+ channels were studied in nucleated patches using pipette solution in which K+ was substituted with Cs+. Since neurons could not keep VR and tonic firing without internal K+, their characterization was done during the first 10–15 s after membrane was broken, before pipette Cs+ substituted intracellular K+ (Fig. 2A). For Na+ current recording in patches, 1 mM TEA was added to ACSF* to reduce outward K+ current. Na+ channels began to activate at –50 mV and had fast opening kinetics (Fig. 2, B and D). Their activation characteristic fitted with Boltzmann equation had V50 = –35.7 ± 0.6 mV and k = 7.5 ± 0.5 mV (Fig. 2C, n = 6). The steady-state inactivation of Na+ channels, studied with 50-ms conditioning prepulses, revealed a half-maximum inactivation at –75.5 ± 0.1 mV and k = –9.1 ± 0.1 mV (Fig. 2C, n = 5).
|
in) of Na+ current. The in changed from 6.1 ± 2.1 ms at –50 mV to 0.39 ± 0.04 ms at +10 mV (Fig. 2D, n = 6). The time course of Na+ channel recovery from inactivation at potentials close to VR was studied using a standard two-pulse protocol (Fig. 3A). The membrane was held at –80 mV and two 25-ms voltage pulses to –30 mV with varying intervals were applied. Recovery of channels from inactivation followed a double-exponential time course. The time constants, fast and slow, were F = 21.8 ± 1.7 ms (63%) and S = 793 ± 92 ms (37%), respectively (Fig. 3B, n = 7).
|
Delayed-rectifier K+ (KDR) channels
The major voltage-gated K+ current found in TFNs was a slowly inactivating KDR current (Fig. 4A). Patch was held at –80 mV and depolarizing voltage pulses were applied after a 150-ms prepulse to –60 mV inactivating transient A-type K+ (KA) currents. The threshold of KDR current was –40 mV and conductance reached saturation at +20 mV (Fig. 4B). The fitting of the activation characteristic with Boltzmann equation gave V50 = –19.8 ± 0.4 mV and k = 9.9 ± 0.4 mV (n = 13). To describe the activation kinetics we measured the rise time of a half-maximum current (0.5). At potentials positive to +30 mV, the 0.5 became shorter than 1 ms (Fig. 4B, n = 9). Thus gating kinetics of KDR current was sufficiently fast for its involvement in spike repolarization.
Starting from –20 mV, a partial inactivation of KDR current with a monoexponential time course developed. The in was weakly voltage dependent, changing from 79.8 ± 12.5 ms at –10 mV to 59.0 ± 3.2 ms at +60 mV (Fig. 4C, n = 9). VREV for KDR conductance was estimated from tail currents evoked by voltage return from +40 mV to different levels (Fig. 4D). The tail current changed almost linearly with a voltage and reversed its polarity near to VK of –84 mV (n = 5).
Tail currents were also used to measure the closing rate of KDR channel at –60 mV. They were recorded after short and long depolarizing pulses to +60 mV (Fig. 4E). The short pulse of 3–5 ms was adjusted to terminate at a peak of the current giving the tails corresponding to the total (noninactivated) KDR current. This tail current decayed monoexponentially with a time constant of 5.9 ± 1.3 ms (n = 6). Similar measurement was also done after a 200-ms pulse, for a partially inactivated KDR current. The tail currents became smaller but decayed monoexponentially with a similar time constant of 5.3 ± 1.1 ms. The observed fast closing rate of KDR channel correlated well with high frequencies of tonic firing seen in current-clamp experiments.
In 1 mM TEA, KDR current was blocked to 20.3 ± 1.5% (Fig. 4F, n = 4). The kinetics of control and remaining currents were very similar, which was better seen when the traces were normalized and superimposed (Fig. 4F, bottom). In 10 mM TEA, the current was reduced to 8.4 ± 1.3% (n = 5) but the current kinetics remained unchanged (not shown).
KA current
In somatic patches from TFNs, a transient KA current was much smaller than KDR current. KA current could not be separated using a standard procedure with two prepulses (to –120 and –60 mV), since the difference trace was always dominated by the slowly inactivating KDR current, the amplitude of which also depended on the prepulse. Therefore we recorded K+ currents elicited by a voltage step from –120 to +60 mV before and after addition of 10 mM TEA (Fig. 5A). In control solution no fast inactivating component could be revealed and a monoexponential fitting with a slow in typical for KDR current adequately described inactivation. In 10 mM TEA, KDR current decreased (Fig. 5A, left) and the fast inactivating component could be resolved in the remaining current (right, the same trace is shown amplified). A double-exponential fitting revealed the fast component of inactivation, A = 6.6 ± 0.8 ms at +60 mV (n = 8).
|
Densities of Na+ and K+ currents
For estimation of the current densities the mean diameter of nucleated patch was assumed to be 7.5 µm giving a membrane area of 177 µm2. The maximum amplitude of Na+ current activated after a 50-ms prepulse to –120 mV was 208 ± 31 pA (n = 15), corresponding to a density of 1.18 pA µm–2. The amplitude of KDR current measured at a voltage step from –60 to +60 mV was 1049 ± 193 pA (n = 8), giving a mean current density of 5.9 pA µm–2. It should be noted the KDR channels were inactivated to some degree at –60 mV. The total K+ current seen after a –120 mV prepulse was larger by a factor of 1.23 ± 0.05 (n = 11), but it represented a mixture of the KDR and KA currents that could not be clearly separated.
Based on our whole cell recordings, it could be assumed that KA currents in patches were not substantially reduced in 10 mM TEA. At the voltage step from –120 to +60 mV, the amplitude of KA current estimated by the fitting was 123.1 ± 34.6 pA (n = 8), giving a density of 0.7 pA µm–2. It should be noted that the KA current formed only 8.6 ± 1.6% (n = 8) of a total (KDR + KA) current when recorded after a –120-mV prepulse and became negligible when depolarization was applied after prepulses close to the VR level.
Role of KDR current in tonic firing
An involvement of KDR channels in tonic firing was tested in current-clamp mode by comparing the patterns recorded in the presence and absence of 1 mM TEA, which partially blocked KDR, but not KA, current. The firing frequency and stability, at a given stimulation strength, were reduced in 1 mM TEA (Fig. 5D, n = 6). These effects resulted from slowing down the spike repolarization and disappearance of fast afterhyperpolarization (Fig. 5D, right). Thus a reduction of KDR rather than KA current was critical for generation of repetitive firing.
Morphology of TFNs
Twelve biocytin-filled TFNs were reconstructed and six of them are shown in Fig. 6. On the basis of their dendritic arborization and appearance, TFNs could be divided into three major groups. The cells belonging to the first group (Fig. 6A, n = 5) had triangular/pyramidal somata with three to four main dendrites passing the border between lamina II (SG) and III. The axon, when reconstructed, stayed at the level of the cell body and was extensive in a rostrocaudal orientation. Neurons with a similar appearance were classified as stalked cells (Eckert et al. 2003; Gobel 1978; Todd 1988). The second group of neurons (Fig. 6B, n = 3) was characterized by smaller fusiform somata, dendrites bearing large beads, and axon branching around the cell body. The cells from the final group (Fig. 6C, n = 4) had rounded somata with multiple extensively branching dendrites either in a multipolar or a bipolar organization. The axons were recovered mainly in lamina II (SG). Neurons with similar features were referred to as islet cells (Eckert et al. 2003; Gobel 1978; Todd 1988). The neurons from all three groups had similar electrophysiological parameters: VR, –70 ± 2, –73 ± 3, and –71 ± 4 mV; RIN, 1.25 ± 0.30, 1.16 ± 0.26, and 1.17 ± 0.15 G; 0, 94.4 ± 9.3, 98.3 ± 23.4, and 102.7 ± 7.4 ms for the first (n = 5), second (n = 3), and third (n = 4) group TFNs, respectively.
|
To test our assumption about the role of Na+ and KDR currents in the generation of tonic firing, we built a computer model of TFN. Inclusion of these conductances was sufficient to provide firing in a broad range of frequencies (Fig. 7A). The steady-state firing f-I characteristic of the model with the initial slope of 1.5 Hz/pA (Fig. 7B) was very similar to that recorded in TFNs. At high-frequency firing the degree of interspike recovery of Na+ channels from inactivation was about one-fifth of the resting level seen before the first spike (Fig. 7C), indicating that a high safety factor for the spike generation is important for the appearance of high-frequency firing.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
; Lopez-Garcia and King 1994; Prescott and De Koninck 2002; Thomson et al. 1989), but conflict with Ruscheweyh and Sandkuhler (2002) who have not found TFNs in rat lamina II (SG). According to cutaneous afferent input, most TFNs were shown to be either wide-dynamic-range or nociceptive-specific neurons (Lopez-Garcia and King 1994). The intrinsic tonic firing seems to be an important property of this class of neurons, allowing them to fire bursts of spikes in response to synaptic stimulation (Thomson et al. 1989).
The present study determined the role of major conductances in excitability of TFNs. Both Ca2+ and KCA currents are involved in modulation of firing frequency. The effect is based on activation of apamin-sensitive KCA channels, regulating the slow component of afterhyperpolarization as was shown for several types of neurons (Barrett and Barrett 1976; Nishimura et al. 1989; Savic et al. 2001; Schwindt et al. 1988; Smith et al. 2002; Viana et al. 1993). Activation of KCA channels by entering Ca2+ reduced the firing at a given depolarization strength and therefore modified the input–output characteristic of the neuron. Thus Ca2+-dependent conductances play an important modulatory and stabilizing role in firing of SG neurons, but they are unlikely to be responsible for the appearance of the basic form of tonic firing. In similar manner, the variation of VR does not change the pattern of tonic firing but regulates the strength of stimulation needed to evoke it. Therefore the voltage-gated Na+ and K+ channels are mostly responsible for the appearance of tonic firing.
The activation kinetics of Na+ channels was sufficiently fast to provide spike depolarization. A low channel density found in the soma supports previous observations that the major part of Na+ conductance necessary for the spike generation in spinal neurons is located in AIS (Alessandri-Haber et al. 1999; Safronov et al. 1999a,b). In addition, our simulations showed that removal of somatic channels did not change the spike overshoot. Rapid inactivation of Na+ channels at positive potentials (in < 0.4 ms at +10 mV), in combination with fast KDR channel opening (0.5 < 0.9 ms at +40 mV), is critical for the spike repolarization. Double-exponential recovery of Na+ channels from inactivation was similar to that observed in rat motoneurons (Safronov and Vogel 1995) and hippocampal neurons (Martina and Jonas 1997). About one-fifth of the channels reprimed within the first 10 ms, which corresponds to interspike intervals at 100 Hz firing. However, our recordings and model showed that this amount of Na+ channels is sufficient to support firing. Thus a high safety factor for the spike generation (see also Eckert 1978) plays a critical role in maintaining the tonic firing under conditions in which most Na+ channels are inactivated.
KDR channel underlies the major type of K+ conductance in TFNs. Delayed activation and a high sensitivity to TEA allowed us to attribute it to a family of delayed-rectifier K+ channels. Its inactivation was by an order of magnitude slower than that of KA channels. Moreover, in contrast to the KA component, KDR current did not show inactivation kinetics at potentials below –10 mV. KDR current was proportionally reduced in 1 and 10 mM TEA, indicating homogeneity of the channel population with respect to its sensitivity to the blocker. The study of the tail currents at –60 mV did not provide any evidence for existence of a channel subpopulation, which was active at the beginning of the pulse but inactivated at the end of 200 ms depolarization. Thus, based on analysis of tail currents and sensitivity to TEA, which are used as standard tools for separating delayed-rectifier current components (Dubois 1981; Safronov et al. 1996; Stühmer et al. 1989), we assume that KDR current is carried through a one-channel type highly permeable for K+. Somatic density of KDR current of 5.9 pA µm–2 found here is much higher than one calculated from Wolff et al. (1998) for nonidentified dorsal horn neurons of 3- to 7-day-old rats (240 pA for the 10 µm soma, corresponding to 0.76 pA µm–2). This difference can be explained if it is assumed that the neuronal development during the first postnatal month is accompanied by somatic expression of KDR channels, in contrast to the Na+ channels that are exclusively expressed in AIS (Safronov et al. 1999b).
The activation of KDR current was sufficiently rapid to provide membrane repolarization during an action potential. The channel closing with a time constant of <6 ms, in turn, was important for sustained firing at high frequencies. An involvement of KDR channels in both processes was confirmed by a spike prolongation and reduction of firing frequency seen in 1 mM TEA (see also Olschewski et al. 2001).
KA current was very small in somatic patches and strong inactivation at VR would further reduce it, making its participation in tonic firing unlikely. It appears that lack of KA channels is critical for appearance of sustained firing, since a pronounced KA current in dorsal horn neurons was shown to result in delayed firing onset, irregular bust-like firing, or frequency adaptation (Grudt and Perl 2002; Ruscheweyh and Sandkuhler 2002; Yoshimura and Jessell 1989).
Our model has confirmed that a combination of Na+ and KDR channels is sufficient for appearance of a basic pattern of tonic firing. The model could reproduce the firing in a broad range of frequencies. The maximum firing rates were determined by biophysical properties of the channels in such a way that at a given stimulus strength the closing of KDR channels determined the length of the interspike interval, whereas the recovery of Na+ channels from inactivation determined the spike amplitude. Our model was not based on the specific anatomy of any particular type of neurons and, therefore, it can be useful for studying the mechanisms of firing adaptation or appearance of delayed-firing patterns in other types of SG neurons.
The population of TFNs was morphologically inhomogeneous and at least three groups of neurons were distinguished on the basis of their somatodendritic organization. Our results support the observations of others that tonic firing can be generated by several anatomical groups of SG neurons (Grudt and Perl 2002). Moreover, neurons belonging to one group could show firing patterns with differing degrees of adaptation (Grudt and Perl 2002), implying the absence of strict correlation between the firing pattern and cell morphology in SG. It can be therefore suggested that the balanced expression of ion channels described here is responsible for appearance of tonic firing in morphologically distinct types of SG neurons. We could not combine a nucleated patch recording with staining in the same neuron, because the isolation of the nucleus usually resulted in a deterioration of relatively small SG neuron and did not allow us to obtain a sufficiently good staining. Nevertheless, in more than 60 nucleated patches from TFNs, we recorded ion channels with similar properties, suggesting their presence in cells of all three subtypes.
In conclusion, a balanced system of ionic conductances underlies tonic firing in SG neurons. Voltage-gated Na+ current in combination with a pronounced KDR but small KA currents generate a basic firing pattern, while Ca2+-dependent conductances stabilize tonic firing, efficiently regulate discharge frequency, and modulate an input–output characteristic in a neuron.
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
GRANTS
The work was supported by a grant from the Portuguese Foundation for Science and Technology (Fundação para a Ciência e a Tecnologia).
|
FOOTNOTES |
|---|
Address for reprint requests and other correspondence: B. V. Safronov, Instituto de Biologia Molecular e Celular (IBMC), Rua do Campo Alegre 823, 4150–180 Porto, Portugal (E-mail: safronov{at}ibmc.up.pt).
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
Barrett EF and Barrett JN. Separation of two voltage-sensitive potassium currents, and demonstration of a tetrodotoxin-resistant calcium current in frog motoneurones. J Physiol 255: 737–774, 1976.
Barrett JN and Crill WE. Specific membrane properties of cat motoneurones. J Physiol 239: 301–324, 1974.
Bentley GN and Gent JP. Electrophysiological properties of substantia gelatinosa neurones in a novel adult spinal slice preparation. J Neurosci Methods 53: 157–162, 1994.[CrossRef][ISI][Medline]
Brown AG. Organization in the Spinal Cord. Berlin: Springer-Verlag, 1981.
Cervero F. Dorsal horn neurones and their sensory inputs. In: Spinal Afferent Processing, edited by TL Yaksh. New York: Plenum, 1987, p. 197–216.
Chery N, Yu XH, and De Konink Y. Visualization of lamina I of the dorsal horn in live adult rat spinal cord slices. J Neurosci Methods 96: 133–142, 2000.[CrossRef][ISI][Medline]
Dodge FA and Cooley JW. Action potential of motoneuron. IBM J Res Dev 17: 219–229, 1973.
Dubois J-M. Evidence for the existence of three types of potassium channels in the frog Ranvier node membrane. J Physiol 318: 297–316, 1981.
Eckert R. Nerve cells and signals. In: Animal Physiology, edited by R Eckert and D Randall. San Francisco, CA: Freeman, 1978, p. 148–192.
Eckert WA, McNaughton KK, and Light AR. Morphology and axonal arborization of rat spinal inner lamina II neurons hyperpolarized by µ-opioid-selective agonists. J Comp Neurol 458: 240–256, 2003.[CrossRef][ISI][Medline]
Edwards FA, Konnerth A, Sakmann B, and Takahashi T. A thin slice preparation for patch clamp recordings from neurones of the mammalian central nervous system. Pflügers Arch 414: 600–612, 1989.[CrossRef][ISI][Medline]
Gobel S. Golgi studies of the neurons in layer II of the dorsal horn of the medulla (trigeminal nucleus caudalis). J Comp Neurol 180: 395–414, 1978.[CrossRef][ISI][Medline]
Gobel S, Falls WM, Bennet GJ, Abdelmoumene M, Hayashi H, and Humphrey E. An EM analysis of the synaptic connections of horseradish peroxidase-filled stalked cells and islet cells in the Substantia Gelatinosa of adult cat spinal cord. J Comp Neurol 194: 781–807, 1980.[CrossRef][ISI][Medline]
Grudt TJ and Perl ER. Correlations between neuronal morphology and electrophysiological features in the rodent superficial dorsal horn. J Physiol 540: 189–207, 2002.
Hines ML. Neuron: a program for simulation of nerve equation. In: Neural Systems: Analysis and Modeling, edited by FH Eeckman. Boston, MA: Kluwer, 1993, p. 127–136.
Hines ML and Carnevale NT. The NEURON simulation environment. Neural Comput 9: 1179–1209, 1997.[Abstract]
LaMotte C. Distribution of the tract of lissauer and the dorsal root fibers in the primate spinal cord. J Comp Neurol 172: 529–561, 1977.[CrossRef][ISI][Medline]
Light AR and Perl ER. Differential termination of large-diameter and small-diameter primary afferent fibers in the spinal dorsal gray matter as indicated by labelling with horse-radish peroxidase. Neurosci Lett 6: 59–63, 1977.[CrossRef][ISI]
Lopez-Garcia JA and King AE. Membrane properties of physiologically classified rat dorsal horn neurons in vitro: correlation with cutaneous sensory afferent input. Eur J Neurosci 6: 998–1007, 1994.[CrossRef][ISI][Medline]
Martina M and Jonas P. Functional differences in Na+ channel gating between fast-spiking interneurones and principal neurones of rat hippocampus. J Physiol 505: 593–603, 1997.
Nishimura Y, Schwindt PC, and Crill WE. Electrical properties of facial motoneurons in brainstem slices from guinea pig. Brain Res 502: 127–142, 1989.[CrossRef][ISI][Medline]
Olschewski A, Hempelmann G, Vogel W, and Safronov BV. Suppression of potassium conductance by droperidol has influence on excitability of spinal sensory neurons. Anesthesiology 94: 280–289, 2001.[CrossRef][ISI][Medline]
Prescott SA and De Koninck Y. Four cell types with distinctive membrane properties and morphologies in lamina I of the spinal dorsal horn of the adult rat. J Physiol 539: 817–836, 2002.
Rall W. Branching dendritic trees and motoneuron membrane resistivity. Exp Neurol 1: 491–527, 1959.[CrossRef][ISI][Medline]
Rall W. Time constants and electrotonic length of membrane cylinders and neurons. Biophys J 9: 1483–1508, 1969.
Rethelyi M. Preterminal and terminal axon arborizations in the substantia gelatinosa of cat's spinal cord. J Comp Neurol 172: 511–521, 1977.[CrossRef][ISI][Medline]
Ruscheweyh R and Sandkuhler J. Lamina-specific membrane and discharge properties of rat spinal dorsal horn neurones in vitro. J Physiol 541: 231–244, 2002.
Safronov BV. Spatial distribution of Na+ and K+ channels in spinal dorsal horn neurones: role of the soma, axon and dendrites in spike generation. Prog Neurobiol 59: 217–241, 1999a.[CrossRef][ISI][Medline]
Safronov BV, Bischoff U, and Vogel W. Single voltage-gated K+ channels and their functions in small dorsal root ganglion neurones of rat. J Physiol 493: 393–408, 1996.
Safronov BV and Vogel W. Single voltage-activated Na+ and K+ channels in the somata of rat motoneurones. J Physiol 487: 91–106, 1995.
Safronov BV, Wolff M, and Vogel W. Functional distribution of three types of Na+ channel on soma and processes of dorsal horn neurones of rat spinal cord. J Physiol 503: 371–385, 1997.
Safronov BV, Wolff M, and Vogel W. Axonal expression of sodium channels in rat spinal neurones during postnatal development. J Physiol 514: 729–734, 1999b.
Sather W, Dieudonné S, MacDonald JF, and Ascher P. Activation and desensitization of N-methyl-D-aspartate receptors in nucleated outside-out patches from mouse neurones. J Physiol 450: 643–672, 1992.
Savic N, Pedarzani P, and Sciancalepore M. Medium afterhyperpolarization and firing pattern modulation in interneurons of stratum radiatum in the CA3 hippocampal region. J Neurophysiol 85: 1986–1997, 2001.
Schwindt PC, Spain WJ, Foehring RC, Stafstrom CE, Chubb MC, and Crill WE. Multiple potassium conductances and their functions in neurons from cat sensorimotor cortex in vitro. J Neurophysiol 59: 424–449, 1988.
Smith MR, Nelson AB, and Lac S. Regulation of firing response gain by calcium-dependent mechanisms in vestibular nucleus neurons. J Neurophysiol 87: 2031–2042, 2002.
Stühmer W, Ruppersberg JP, Schröter KH, Sakmann B, Stocker M, Giese KP, Perschke A, Baumann A, and Pongs O. Molecular basis of functional diversity of voltage-gated potassium channels in mammalian brain. EMBO J 8: 3235–3244, 1989.[ISI][Medline]
Sugiura Y, Lee CL, and Perl ER. Central projections of identified, unmyelinated (C) afferent fibers innervating mammalian skin. Science 234: 358–361, 1986.
Thomson AM, West DC, and Headley PM. Membrane characteristics and synaptic responsiveness of superficial dorsal horn neurons in a slice preparation of adult rat spinal cord. Eur J Neurosci 1: 479–488, 1989.[CrossRef][ISI][Medline]
Thurbon D, Luscher HR, Hofstetter T, and Redman SJ. Passive electrical properties of ventral horn neurons in rat spinal cord slices. J Neurophysiol 80: 2485–2502, 1998.[Medline]
Todd AJ. Electron microscope study of Golgi-stained cell in lamina II of the rat spinal dorsal horn. J Comp Neurol 275: 145–157, 1988.[CrossRef][ISI][Medline]
Viana F, Bayliss DA, and Berger AJ. Multiple potassium conductances and their role in action potential repolarization and repetitive firing behavior of neonatal rat hypoglossal motoneurons. J Neurophysiol 69: 2150–2163, 1993.
Willis WD and Coggeshall RE. Sensory Mechanisms of the Spinal Cord. New York: Plenum, 1991.
Wolff M, Vogel W, and Safronov BV. Uneven distribution of K+ channels in soma, axon and dendrites of rat spinal neurones: ffunctional role of the soma in generation of action potentials. J Physiol 509: 767–776, 1998.
Yoshimura M and Jessell TM. Membrane properties of rat substantia gelatinosa neurons in vitro. J Neurophysiol 62: 109–118, 1989.
This article has been cited by other articles:
|
|
 |
|
  K. J. Dougherty and S. Hochman Spinal Cord Injury Causes Plasticity in a Subpopulation of Lamina I GABAergic Interneurons J Neurophysiol, July 1, 2008; 100(1): 212 - 223. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. A. Graham, A. M. Brichta, and R. J. Callister Recording Temperature Affects the Excitability of Mouse Superficial Dorsal Horn Neurons, In Vitro J Neurophysiol, May 1, 2008; 99(5): 2048 - 2059. [Abstract] [Full Text] [PDF] |
