- Email: [email protected]
The role of tetrodotoxin-resistant sodium channels of small primary afferent fibers
BRAIN RESEARCH ELSEVIER
Brain Research 639 (1994) 125-134
Research Report
The role of tetrodotoxin-resistant sodium channels of small primary afferent fibers Srdija Jeftinija * Department of VeterinaryAnatomy, Neuroscience Program, and Signal Transduction Training Group, Iowa State University, Ames, IA 50011, USA (Accepted 26 October 1993)
Abstract
Intracellular recordings from neurons in the dorsal root ganglion (DRG) and dorsal horn (DH), in an in vitro spinal cord-dorsal root ganglion preparation, were used to investigate the role of tetrodotoxin-resistant (q-TX-R) afferent fibers in the sensory synaptic transmission in the superficial DH. Bath application of 25-50 mM potassium to the DRG depolarized the DRG neurons, blocked action potentials in the large neurons, evoked action potentials in slow conducting neurons, and synaptically excited dorsal horn neurons. Excitatory postsynaptic potentials (EPSP) which were evoked in DH neurons by electrical stimulation of large myelinated fibers, but not those evoked by stimulation of small unmyelinated fibers, were blocked by the potassium treatment of the primary afferents. Tetrodotoxin, when applied to the sensory neurons, abolished the action potentials in fast fibers but had no effect on the action potentials in a population of slow conducting afferents. Peripheral application of T r x blocked the fast EPSPs evoked by electrical stimulation but failed to block the electrically evoked slow EPSPs and the synaptic activation of DH neurons induced by the application of high potassium to sensory neurons. Furthermore, high potassium potentiated electrically evoked, TFX-resistant EPSPs in the majority of neurons. This effect was abolished in Na +-free solution. These findings indicate that high [K+]e applied to the DRG, dorsal root and peripheral process selectively activates a primary afferent input to the DH, which is sodium-dependent and tetrodotoxin resistant.
Key words: Primary afferent neuron; Tetrodotoxin-resistant; Spinal cord neuron; Sodium channel; Excitatory transmission
1. Introduction
Classification of primary afferent neurons of the mammalian D R G is based on several properties of neurons. Large myelinated A a , /3 neurons are fast conducting while small myelinated A ~ and unmyelinated C fibers are slow conducting. The shape and properties of the somatic action potentials vary according to cell type [2-5,16] and, in general, the slower the conduction velocity of a neuron the longer the duration of the action potential and the greater the chance that the action potential will display an inflection or 'kink' or ' p l a t e a u ' during the falling phase. It follows that there must be a considerable degree of overlap between neurons that have a plateau in the downstroke of the action potential and those which respond to noxious stimuli.
* Corresponding author. Fax: (1) (515) 294-3932. 0006-8993/94/$07.00 © 1994 Elsevier Science B.V. All rights reserved
SSDI 0 0 0 6 - 8 9 9 3 ( 9 3 ) E 1 4 9 8 - R
The different properties of action potentials in different D R G cell types suggest that there may be differences in the complement or behavior of the ion channels which control m e m b r a n e excitability. The sensory neurons of mammalian D R G were found to have at least two types of voltage-dependent sodium channels: fast, TTX-sensitive (TTX-S) and slow, TI'X-resistant ( T T X - R ) [7,9-12]. Electrophysiological analysis of T F X - R and T-I'X-S channels in rat D R G neurons was first reported by Kostyuk et al. [7]. This work and that of others indicates that these channels differ in their activation and inactivation kinetics [14]. It has been shown that large sensory neurons and their axons have a T-I'X-S fast sodium current, while small neurons have a T T X - R sodium current [9]. The T T X - R current differs from the T I ' X - S sodium current in many ways. Both are strongly voltage dependent; but the q T X - R current has slower kinetics, the threshold for activation is about 10 m V more positive, and the potential at which the current is half inactivated is about 30 m V
126
S. Jeftinija / Brain Research 639 (1994) 125-134
more positive than the TTX-S current [8,12,13]. Recently, patch voltage clamp recordings identified three distinct kinetic types of Na + current differentially distributed among D R G neurons [1]. Small D R G neurons co-express a rapidly inactivating TTX-S fast current and a slowly activating and -inactivating T T X - R slow current. Large cells express a single T I ' X - S Na + current which is identified as intermediate by its inactivation rate constant. These results suggest that by manipulating the properties of the sodium ionic channels of primary afferent neurons it should be possible to study interactions between the incoming signals to the spinal cord. T T X - R spikes in D R G neurons can be evoked in the absence of calcium [8], and there is a convincing body of evidence which shows that sub-populations of D R G neurons express a T T X - R sodium current which is large enough to support the generation of an action potential in the soma [7,9,11,13]. The major problem in studying pain physiology and pharmacology has been to generate selective C-fiber input to establish an in vitro model to study the synaptic mechanisms of nociception. The in vitro spinal cord slice preparations were developed to provide the necessary stability for long lasting intracellular recordings from dorsal horn neurons. In this p a p e r we offer experimental data that the neurons expressing T T X - R sodium channels are inactive primary afferents which can be activated by an increase in the extracellular potassium. Furthermore, the data demonstrate the presence of the T I ' X - R sodium channels on the peripheral axons of the slow sensory neurons. Preliminary findings were communicated [6].
2. Materials and methods Horizontal spinal cord slices were obtained from young Sprague-Dawley rats (18-30 days old). Following laminectomy under ether anesthesia, a segment of lumbosacral spinal cord (1-1.5 cm long) with dorsal roots and an associated D R G attached was quickly excised and immersed in gassed 95% 0 2 and 5% CO 2 Ringer's solution (in mM: NaCI 124, KCI 5, KH2PO 4 1.2, CaCI 2 2.4, MgSO 4 1.3, NaHCO 3 26, glucose 10, pH = 7.4). Animals were decapitated following excision of the spinal cord. After the removal of the pia mater, the spinal lumbar segments (L2 to L6) were affixed, dorsal surface up, with cyanoacrylic glue to the bottom of a plexiglass cutting chamber of a Lancer Series 1000 vibratome. The bath of the vibratome was filled with the aerated Ringer's solution. The blade of the vibratome was positioned below the dorsal surface of the spinal cord in order to yield a horizontal dorsal horn slice (400-500 tzm thick) with intact associated dorsal rootlets and attached DRG. After incubation at 31+ I°C for 1-2 h, the slice was transferred to a recording chamber. The 'submerged-type' recording chamber consists of four tissue compartments separated by narrow partitions. The adjacent dorsal roots were placed in narrow grooves on the top of the partition so that the spinal slice containing the dorsal horn neurons was in one
compartment with the dorsal surface up, while the D R G with dorsal roots were in different compartments. One D R G compartment was further subdivided to allow the peripheral nerve trunk to be placed in a narrow groove which led to another compartment. Compartments were independently perfused. This allowed differential perfusion and exposure of only one population of afferents to a test compound while the adjacent dorsal root or segment of the same root was used as a control. To insure total electrical separation of the spinal cord and D R G compartments and to avoid mixing of solutions between compartments, vaseline was used to form a seal around the dorsal roots in the grooves. The perfusing medium (in mM: NaCI 127, KCI 1.9, KH2PO 4 1.2, CaCI 2 2.4, MgSO 4 1.3, NaHCO 3 26, glucose 10, pH = 7.4) was oxygenated with 95% 0 2 and 5% CO 2 in a separate reservoir for each compartment and then introduced into the tissue compartments through thin polyethylene tubing at a rate of 2-3 m l / m i n . The temperature of the recording solution was kept at 31 + I°C. Isolated lumbar dorsal roots were placed on platinum wire bipolar stimulating electrodes. The thresholds for activating large myelinated and unmyelinated afferents by stimulating the peripheral segment of the dorsal root with a constant current stimulus were determined by recording action potentials in D R G neurons. They were found to be 8-20 V, 0.02 ms for the most sensitive axons and over 35 V, 0.5 ms for C-fibers. The classification of D R G neurons was done solely on the basis of the conduction velocities calculated by measuring the distance between the cathode and the D R G and dividing it by the conduction latencies of action potentials recorded in the cell bodes of DRG. The average conduction distance was 5.2 mm (n = 45). Fibers conducting at velocities above 15 m / s were classified as A/3. Fibers conducting at velocities below 15 m / s and above 1 m / s were classified as A6 fibers. Fibers conducting at velocities below 1 m / s were classified as C fibers. Intracellular recordings from DH neurons and D R G neurons were performed with glass microelectrodes back-filled with 3 M potassium acetate and having a DC resistance of above 100 MS2. Micropipettes were pulled from capillary-filled borosilicate glass tubing on a Brown-Flaming puller. The micropipettes were placed under visual guidance in the DH slice and the DRG. The passive electrical properties of the neurons were determined by means of a dual-channel intracellular amplifier (Neuro Data IR-283) which allows current injections on the order of 0.05-3.0 nA through the recording electrode. The recorded signals were displayed on the screen of an oscilloscope (Nicolet Instruments M4094) for further analysis of the data. The membrane potential was constantly monitored on a pen recorder.
3. Results Intracellular recordings were made from 30 slow conducting D R G neurons. Threshold intensity was > 20 V, 0.2 ms, and the action potential latency evoked by afferent stimulation corresponded to a conduction velocity of 0.7 _+ 0.08 m / s (mean _+ S.E.M.) when peripheral branches were stimulated and 0.5 + 0.04 m / s when central branches (dorsal root) were stimulated. The average resting m e m b r a n e potential of D R G neurons in this group was - 60 + 2 m V ( _+S.D.). Electrical stimulation of peripheral and central processes of sensory neurons induced action potentials of average amplitude 87 +_ 2 mV and duration, measured at the resting m e m b r a n e potentials, of 4 _+ 0.7 ms followed by a
S. Jeftinija ~Brain Research 639 (1994) 125-134
afterhyperoplarization of 11 + 1 mV amplitude and 31 + 5 ms duration.
3.1. Effect of high potassium and tetrodotoxin on slow primary afferents Brief exposure of the soma of D R G neurons to 50 mM potassium produced a rapid depolarization of the soma membrane (32.8 + 2.5 mV, n = 30). Once the peak of the depolarization was reached, continued exposure to potassium maintained a steady plateau. No
re
DRG ~_.~orsal root
A
50 mM K on DRG T r x 0.SpM on
1 mln
DRG
I ~ I I I I N U l I I N N I N I I I H M I I I U I I N m I I I I I H I l t H I I I I I I I I I H I I I U g l I I m m NHmmln~mmlOlmllWm#limnlNNllll
m
~ M K on DRG B
? 1mV
1
1 /
":,':,-'.-JlllllJ -
'
. . . . . .
....... U T r r =-'
.......
r .......
, r .......
1 rain
1 l~lSV !
Fig. 1. Excitatory action of high potassium on sensory neurons as recorded intracellularly from the soma of a D R G neuron (inset, re = recording electrode). A: bath application of 50 m M K ÷ to the D R G resulted in a 25 m V depolarization accompanied by firing of action potentials (upper trace, upward deflections on chart recording
are action potentials). Action potentials are truncated by the fiequency response of the chart recorder. Perfusion of TTX for 10 min failed to block the stimulatory effect of high potassium (lower trace). B: bath application of high potassium to the dorsal root resulted in firing of action potentials in the soma of this slow sensory neuron (conduction velocity 0.35 m/s). Oscilloscope tracings in upper row were obtained simultaneously(at faster time scale) with chart recordings, as indicated by arrows. In the presence of high potassium, the conduction velocityof the sensory neuron decreased from 0.35 m/s to 0.26 m/s. The electrical artifacts in the oscilloscopetraces are due to dorsal root stimulation. Resting membrane potential (Em) was - 58 mV.
127
desensitization was observed following repeated applications at 10-15 min intervals. About half of the tested D R G neurons (n = 16) did not fire action potentials, but the rest of the neurons fired action potentials on the rising phase of the 50 mM K+-evoked depolarization or throughout the potassium-induced depolarization (Fig. 1A, upper trace). The potassium-evoked depolarization in all neurons and the firing of action potentials in three neurons were not blocked by tetrodotoxin ('I-TX, 0.5 to 1/zM) (Fig. 1A, lower trace). In order to study the mechanism of conduction of the potassium-evoked activation of D R G neurons, potassium was applied to the central or peripheral processes of sensory neurons while recording from the soma (Fig. 1 inset, n = 15). In no experiment was a change in resting membrane potential of the soma recorded when the high potassium was applied to the processes of sensory neurons. In 12/15 D R G neurons, application of potassium to the processes resulted in a block of action potentials induced by electrical stimulation of the same segment. In three neurons, however, application of high potassium to the processes of the sensory neurons resulted in firing of action potentials recorded from the soma (Fig. 1B). In addition to the firing of 'spontaneous' action potentials, the action potentials evoked by electrical stimulation of the dorsal root stimulation were not blocked. An increase in the response latency from 50 ms to 65 ms was recorded indicating a decrease in conduction velocity from 0.35 m / s to 0.26 m / s (Fig. 1B, oscillograms). The number of long-lasting, stable intracellular recordings from small D R G neurons that fired tetrodotoxin-resistant ( T T X - R ) action potentials throughout potassium application was too low (10%) to study the role of T T X - R sensory neurons in primary afferent transmission. Therefore, we indirectly determined the functional role of T T X - R afferents by recording intracellularly from D H neurons in laminae I - I I I that are activated synaptically by primary afferents. This experimental approach amplified the signal coming from slow conducting afferent fibers because most of superficial D H neurons receive input from C-fibers.
3.2. Synaptic response of dorsal horn neurons to high potassium applied to sensory neurons." role of tetrodotoxin-resistant afferents Intracellular recordings were made from 82 D H neurons. The depth of the neurons ranged from 9 to 433 /zm (147 + 139, mean + S.D.). The mean resting membrane potential (Em) was - 7 2 + 2 mV ( + S.D.) and the input resistance ranged from 38 to 168 MO. All of the recordings were done in neurons located lateral to the dorsal root entry zone, and most of them were localized between roots L4 and L5.
S. Jeftinija / Brain Research 639 (1994) 125-134
128
All DH neurons were excited synaptically by electrical stimulation of the dorsal root or the peripheral nerve trunk. Single electrical shocks were calibrated to produce activation either of large fibers (10-20 V, 0.02 ms) or the whole fiber population, including unmyelinated afferents (supramaximal stimulus: > 35 V, 0.5 ms). Low intensity stimulation of afferent fibers (10-20
A
control
50 mM K on L5
V, 0.02 ms) induced fast EPSP. The mean amplitude was 12 + 1 mV, and the mean duration was 43 + 10 ms. The latencies suggested conduction velocities from 2 to 14 m / s . Fifteen neurons did not respond to low intensity stimulation of primary afferents. In an additional 15/82 neurons, the fast EPSP was followed by an inhibitory postsynaptic response (IPSP). High stimulus
50raM K on L4
II- 9 j , ~
wash
'~'J
"
L4
, mv 50
mm
L5
.la
,lO.V
L4 L5
B
L4 L5
L4 L5
peripheral
central
20V,0.02ms
20V, 0.02ms
1 mln
Ill
50V,0.5ms
IJl control
50mM K L4 peripheral
~
~
~
4
L_ 50raM K L4central
~-~,,,,,.~.~.. ,
~
, 50raM K both
I
~
Fig. 2. Effect of high potassium on excitability of primary afferent fibers. Records were obtained from two different neurons, A and B. A: slow chart recordings (bottom set of recordings), obtained by intracellular recording from a D H neuron, revealed synaptic activation when high potassium was applied to either the L4 or L5 D R G (inset). The afferent fibers in the adjacent roots were stimulated electrically and the effects of high potassium were recorded on an oscilloscope at a much faster rate (two rows of oscilloscope tracings). Application of high potassium to either dorsal root depressed the fast potentials from that root. Note that the oscilloscope tracings were taken at a time when the m e m b r a n e potential had returned to the resting level. B: application of high potassium to different segments of the afferent fibers (see inset) demonstrated that the potassium-induced potentiation of excitatory transmission conducted by small fibers does not require the action on the soma of the sensory neurons. Application of potassium to the peripheral processes inhibited 30 ms of the EPSP produced by stimulation of the same segment, and it was without effect on the EPSP produced by stimulation of the dorsal root (central). W h e n applied to the central compartment (compartment with L4 DRG), high potassium abolished low intensity-evoked responses from both stimulating sites, and the slow component of the peripherally evoked EPSP was much smaller (about 15%) than the response obtained during peripherally applied potassium. W h e n potassium was applied to both compartments, the slow EPSP in response to peripheral stimulation reappeared. A: Em = - 6 8 mV, depth of the neuron was 28 /xm. B: Em = - 7 2 mV, depth = 80/xm.
s. Jeftinija/ Brain Research 639 (1994) 125-134 intensity synaptically affected all of the DH neurons. In the majority of DH neurons (52/82), the slow component of the EPSP evoked by activation of C-fibers was a continuation of the fast EPSP. In the 15/82 neurons that did not respond to low intensity stimulation of primary afferents, high intensity stimulation produced a long delayed EPSP corresponding to a conduction velocity of 0.8 + 0.05 m/s. In the 15 neurons that responded with an EPSP-IPSP to low intensity stimulation, high intensity stimulation produced a complex response composed of the fast EPSP (16 + 4 mV, 136 5:47 ms) followed by the IPSP (6 + 1 mV, 237 + 84 ms), and then followed by a slow EPSP (5 + 0.5 mV, 28 + 7 s). This slow EPSP was well separated from the fast EPSP and in 6 neurons was accompanied by the firing of action potentials. The application of high K ÷ (25 to 50 mM) to the DRG produced a concentration-dependent excitation in the majority of DH neurons. One minute applications of 50 mM potassium to DRG produced an average depolarization of dorsal horn neurons of 7.6 + 2.1 mV. This high potassium effect was associated with an increase in the frequency of EPSP and firing of action potentials. The type of synaptic excitation in DH neurons varied from short bursts of EPSPs and action potentials to prolonged excitation that lasted for several minutes. To further study the mechanism of the potassiumevoked stimulatory action on primary afferents, we investigated the effect of high K ÷ applied to the sensory neurons on the synaptic responses evoked in DH neurons by electrical activation of primary afferents. Perfusion of 50 mM K ÷ to one of the DRG compartments induced an excitatory postsynaptic depolarization in the DH neuron (Fig. 2A, slow chart recordings) and completely abolished the EPSPs evoked by low intensity stimulation of the corresponding root. When the roots were activated with high intensity stimulation, the initial phase of the prolonged excitatory postsynaptic response was selectively and reversibly abolished from the root exposed to high potassium while the slow component of the EPSP appeared to be enhanced (Fig. 2A, oscilloscope tracings). The blocking effect of high potassium on transmission from fast fibers outlasted the period of potassium-induced excitatory depolarization for several minutes, and EPSP measurements were done at the resting membrane potential to exclude the membrane potential effect. The EPSP evoked by stimulation of the adjacent dorsal root at the same stimulus strength was unchanged (Fig. 2A, oscilloscope tracings). Bath application of high potassium to the nerve processes peripheral to the DRG (Fig. 2, L4-p inset) produced synaptic excitation of the DH neurons (n = 12) and abolished both the fast EPSP evoked by the low intensity stimulus (20 V, 0.02 ms) and the fast component of the postsynaptic response evoked by the
129
supramaximal stimulus applied at the L4-p stimulation site. It was without effect on responses evoked by stimulation al L4-c of the same afferents in the central compartment (Fig. 2B, second row of oscillograms). Application of high potassium to the central compartment (n = 8) in the same preparation abolished the fast components of postsynaptic responses from both stimulation sites and almost completely abolished the slow component evoked by peripheral stimulation (Fig. 2B, fourth row, second tracing from left). To demonstrate that the slow component from the peripheral segment is present only when primary afferents are depolarized, 50 mM potassium was applied to both compartments (n = 8), and the slow EPSPs evoked by stimulation at either site were recovered (Fig. 2B, bottom row of oscillograms). 3.3. The effects of tetrodotoxin on electrical and chemical activation of the primary afferent fibers After establishing, by recording from DRG cells, that there are TrX-R primary afferents that fire action potentials when depolarized by high potassium, studies on the effect of high potassium were continued by recording from postsynaptic DH neurons. Independent of the presence (T-FX applied to DRG) or absence (TTX applied to dorsal root or peripheral nerve trunk) of neuronal cell bodies, TTX abolished the fast component of the EPSP (n = 52), but in 38 of 52 neurons failed to block the slow component. Although it was possible to completely block the excitatory transmission in DH neurons evoked by electrical stimulation of high threshold primary afferents, high concentrations of TTX only decreased, but did not abolish, the excitation of DH neurons evoked by high potassium applied to the sensory neurons. Fig. 3 illustrates the time course of the effect of appliying TTX to a DRG while recording from a dorsal horn neuron. A 2 min perfusion application of TI'X (0.5 ~M) almost abolished the fast EPSP evoked by low intensity stimulus (left oscillogram in row b; 20 V, 0.02 ms) and abolished 30 ms of EPSP evoked by high intensity stimulus (right oscillogram in row b; 35 V, 0.5 ms). With time TTX's blocking effect was more pronounced, and after 15 min of perfusion of 0.5/xM TTX almost all EPSPs were abolished (row c and d). Since A-fibers were now blocked by TFX, it was possible to study the effect of high potassium on the slow C-fibers. Potassium applied to primary afferents treated with TTX produced synaptic activation of DH neurons (slow chart recordings) and potentiated the TTX-R component of electrically evoked EPSP (Fig. 3, oscillograms e and f). This kind of facilitation of excitatory transmission was recorded in 15 independent experiments (100%) and was measured as an increase of 304 + 64% in the size (area under the curve) of the slow EPSPs evoked by T-I'X-R fibers and
130 A
S. Jeftinija / Brain Research 639 (1994) 125-134
2OV, 0.02rns
35V, 0.5ms
20V, 0.02ms
report). W h e n these differences in r e s p o n s e latency were c o n v e r t e d to c o n d u c t i o n velocity, the high potassium-resistant E P S P ' s latency c o r r e s p o n d e d to a n average c o n d u c t i o n velocity of 0.39 + 0.04 m / s ( n = 8), a n d the T I ' X - R E P S P ' s latency c o r r e s p o n d e d to 0.27 + 0.03 m / s ( n = 8; P < 0.025). T h e response latency of the T I ' X - R E P S P decreased following application of high potassium to p r i m a r y afferents r e p r e s e n t i n g a significant increase in c o n d u c t i o n velocity from 0.27 + 0.03
35V, 0.5ms
control
2 rain 1"1"X 35 rain
t l l TTX 10 rain 1-1"X+ 25raM K on
ORG d]J
A
control
T r x 0.5ixM on DRG
l"rX + 50raM K otl DRG
TT'X 15 rain
I"TX + 2SrnM K on DRG
TTX 40 rain -
~
" r~~ 0.5
on DRO ~
/
f~lnS
,
rJI/IdIJ/Jdf/111//
25mV
50ms
B TTX 0.5pM on DRG -:::::--::::::-:-::::::::::::::::::::::::lllLql ~ liraUlmillUlMIIIIIlUlUlMImlmlmlm,ram u m mmamm 25rnM K on DRG
10 mV
mm lm mim m 1 mln
~
J
20mY
1 rain
Fig. 3. Blocking effect of TTX is time dependent. When applied to a sensory neuron, TTX (0.5 /.LM) eliminated the fast EPSP within 2 min and within 15 min almost completely abolished the synaptic response of this DH neuron. The application of high potassium, after 15 min of TTX, synaptically activated the DH neuron (B, slow chart recording) and induced the appearance of a slow EPSP (A, oscillograms e and f). Prolonged perfusion of afferents with TTX for 35 min failed to block this stimulatory effect of high potassium (A, oscillogram i). Em = -75 mV, depth 18 ~m.
an increase in the n u m b e r of action p o t e n t i a l s (from 1.2 + 0.4 to 4.5 + 1) fired d u r i n g these slow EPSP. A t the same time that these obvious c h a n g e s were i n d u c e d by T r x a n d high K ÷ applied to o n e D R G , transmission p r o d u c e d by s t i m u l a t i n g the a d j a c e n t dorsal root was completely u n c h a n g e d . F u r t h e r perfusion of T I ' X (35 m i n ) failed to p r e v e n t the facilitatory effect of high potassium o n the p r i m a r y afferent n e u r o n s (Fig. 3, oscillogram i). T h e facilitatory effect of high p o t a s s i u m o n p r i m a r y afferent n e u r o n s was not uniform, a n d Fig. 4 illustrates three types of responses o b t a i n e d by the application of T I ' X a n d t h e n high p o t a s s i u m to the p r i m a r y afferent n e u r o n s . I n n e u r o n s A a n d B, a q ' T X - R slow E P S P was p r e s e n t a n d in A this c o m p o n e n t was p o t e n t i a t e d w h e n potassium was applied ( n = 12) while in B p o t e n t i a t i o n was not r e c o r d e d ( n = 8). I n d e p e n d e n t of the effect of high p o t a s s i u m o n the size of the T-I'X-R EPSP, the response delay of the T T X - R E P S P was noticeably longer t h a n the response latency of the slow E P S P preserved following application of high potassium to afferent fibers (high p o t a s s i u m - r e s i s t a n t E P S P in this
" r ' ~ 0,5 gM ~ ORG ~ 1 1 1 1 1 1
f
.
~
-
/
l
/
/
/
-
/
i
.
s
.
J
.
-
i
.
o
'
/
,
J
~
ck_ll ~
~
/
i
/
/
/
,
~
/
/
/
/
/
/
/
50 msec
50mMK
Fig. 4. The effect of 0.5 ~M TTX (applied to the DRG compartment) on electrically and 50 mM K+-evoked membrane potential changes recorded in three different DH neurons. In all three neurons, TTX pretreatment of the sensory neurons failed to block the synaptic excitatory depolarization of the dorsal horn neurons by high potassium applied to the sensory neurons (slow chart recordings) A: TI'X abolished the first 50 ms of the EPSP evoked by electrical stimulation (middle oscillogram in the top row). High potassium applied in the presence of TTX-induced synaptic activation of the DH neuron and augmented the size of the EPSP evoked by electrical activation of C fibers (top row far right oscillogram). B: bath application of TTX abolished 40 ms of the EPSP evoked by the electrical stimulation of primary afferents but failed to block the synaptic excitation of DH neuron evoked by high potassium applied to the same DRG. In this neuron, the TTX-R component of the EPSP was not potentiated during potassium application. C: application of TTX to the DRG eliminated the electrically evoked EPSP but failed to block the excitatory effect of high potassium applied to the same DRG. In the presence of high potassium, a long latency EPSP was evoked by electrical stimulation of afferent neurons.
S. Jeftinija /Brain Research 639 (1994) 125-134 control
131
TTX O.S~M
2 mln later
_
_ _ _
_
~
S~V
S0ms
T r x 0.,S~M on DRG IINIHINIIHHIHIIHNNNHIflIflll
SOmM K on DRG
~
0 mV
,'5 eec 50V, 0.Sins
Fig. 5. Facilitation of excitatory transmission induced by high potassium applied to sensory neurons does not depend on membrane potential. Bath application of high potassium to sensory neurons in the presence of TTX produced a synaptic excitation of this DH neuron followed by a hyperpolarizing response. This hyperpolarization was associated with a block of spontaneous firing while EPSPs evoked by electrical stimulation of exposed afferents were potentiated. The doted lines represent membrane potentials when individual EPSPs were recorded. The dashed line marks the resting membrane potential. EPSPs at b and e were recorded at about the same hyperpolarizing potential but the EPSP in b was about two-fold bigger than the EPSP in e. The EPSP in c was bigger than the EPSP in d although they were recorded at about the same membrane potential. Em = - 72 mV, depth = 96/.Lm.
m / s to 0.33 _+ 0.03 m / s ( n = 8; P < 0.1). T h e n e u r o n in C d i d n o t h a v e a T T X - R E P S P at t h e r e s t i n g p o t e n t i a l , a n d o n l y w h e n p o t a s s i u m w a s a p p l i e d to t h e s e n s o r y n e u r o n s was a T T X - R E P S P r e c o r d e d ( n = 10). T h e s e t h r e e n e u r o n s a n d t h e r e m a i n i n g 79 D H n e u r o n s w e r e s y n a p t i c a l l y a c t i v a t e d w h e n 50 m M p o t a s s i u m w a s a p p l i e d to t h e s e n s o r y n e u r o n s (Fig. 4, slow chart recordings). The facilitatory effect of high potassium on TTX-R transmission outlasted the period of
p o t a s s i u m - i n d u c e d e x c i t a t o r y d e p o l a r i z a t i o n by s e v e r a l minutes, and EPSP measurements were done when the cell r e t u r n e d to t h e r e s t i n g m e m b r a n e p o t e n t i a l f o l l o w ing s y n a p t i c d e p o l a r i z a t i o n . In four neurons, the application of high potassium r e s u l t e d in a d e p o l a r i z a t i o n f o l l o w e d by a h y p e r p o l a r i z a t i o n . T h e s e n e u r o n s w e r e u s e d to e s t a b l i s h t h e r e l a tionship between membrane polarization and potentiation of excitatory transmission. Both the depolarizing
A control
L4-p 35V, 0.5rns
"l-rX on L4-p
L4-p 35V, 0.5ms
TI'X+SOmM K on L4-p
L4-p 35V, 0.Sms
100 ms
B L4-p S0mM K 90 sec
C L ~ TrX 0.rmU :7~7~7:~ ~ : ~ z T ~ : ~ : : : : ~ : : ] m m . m u m u m m 0 m m l t 0 m . m a m m M m a m
.mn0..mmmm.mumm0m,mlm.nNmlmuwm#mnm
t . ~ r~mM K
l,o.v 3 0 ~
Fig. 6. Peripheral processes of sensory neurons express TTX-R sodium channels. Recordings were made from one DH neuron. Application of high potassium to the peripheral processes synaptically activated the DH neuron (B,C) and potentiated the TTX-R component of the EPSP evoked by electrical stimulation of the same processes (A). Em = - 7 4 mV, depth = 96/zm, Rm = 151 MO.
S. Jeftinija / Brain Research 639 (1994) 125-134
132
and hyperpolarizing components of the high potassium response were resistant to the application of T T X (Fig. 5, slow chart recording). Although the hyperpolarizing component of the high potassium-evoked response was associated with a decrease in 'spontaneous' synaptic firing, the amplitude of electrical stimulus-evoked q T X - R slow EPSP and the firing of action potentials were increased during this phase (Fig. 5, oscillograms). 3.4. Axons of sensory neurons express TTX-R sodium channels Our results from the application of high potassium to axons of sensory neurons while recording from soma of D R G neurons or D H neurons suggested the presence of slow sodium channels on processes of sensory neurons. These channels were previously demonstrated on soma of cultured neurons in whole-cell patch-clamp experiments [7]. To provide supporting data for the presence of T T X - R sodium channels on the processes of sensory neurons, recordings were made from the soma of D H neurons while stimulating the peripheral processes of sensory neurons in the stump of the spinal control
nerve. In 12 (100%) neurons the application of T T X to peripheral processes abolished the fast EPSP produced by stimulating the peripheral proces, but this had no effect on responses induced by stimulating afferents in the central compartment (not shown). In all 12 neurons, the application of high potassium to the peripheral processes following the TTX pretreatment resuited in an excitatory depolarization comparable to the one recorded in the control (Fig. 6B,C). In addition, potentiation of the q'-I'X-R slow EPSP was recorded in 6 / 1 2 experiments (Fig. 6 oscillograms). 3.5. Effect of Na +-free environment on TTX-R responses Differential effects of divalent cations on TTX-R and TTX-S currents were reported previously [7,14]. To demonstrate the involvement of sodium channels in the excitation of afferent fibers in the presence of TTX, we bathed afferent fibers in a sodium-free solution. Replacing the sodium ion with TRIS (n = 6) abolished all effects of high potassium and electrical stimulation applied to the same segment of the primary afferents (Fig. 7). These effects of TRIS were reversible after washing for 2-5 min with the Na+-con taining recording solution.
mM K on DRG 1
4. Discussion
I lomV 1 rain TTX 0.5~M on DRG I
Trx + TRISon
DRG
~ / / / / / / / , / / / ' / / / . / / / / / / h
/
TTX post TRIS
x
Fig. 7. Stimulatory effect of high potassium applied to sensory neurons and recorded in DH neurons was TTX-R but was sodium dependent. Bath application of high potassium to sensory neurons synaptically activated the D H neuron. This effect was resistant to the application of TTX but was reversibly eliminated by incubating the sensory neurons in a sodium-free solution. Sodium was replaced with TRIS (tris[hydroxymethyl]aminomethane) in equimolar concentrations. Em = - 6 6 mV, Rm =61 MI2, depth 45/zm.
Our results reveal the ability of primary afferent neurons to respond differently to changes in extracellular potassium. Using a combination of electrical and chemical activation of primary afferent fibers under in vitro conditions, we found that the presence of fast and slow sodium channels provides a sufficient substrate for the blocking effect of high potassium on low threshold afferents and its facilitatory effect on high threshold afferent fibers. Our approach was to block the action potential generation in large fibers by manipulating the Na+-channels with a depolarizing agent (high K +), or with the sodium channel blocker tetrodotoxin, or with a combination of these. The later approach revealed the presence of primary afferent neurons that are unexcitable by electrical square pulses at the resting membrane potential and that become active only when depolarized by elevated extracellular potassium. 4.1. TTX-resistant sodium channels and small primary afferent fibers The shapes and properties of the somatic action potentials vary according to the D R G cell type [2,4,5,16]. These findings may reflect differences in the distribution a n d / o r behavior of the ion channels which control membrane excitability in different D R G cell groups. The finding that action potentials of some cell
S. Jeftinija / Brain Research 639 (1994) 125-134
types are blocked by T T X while action potentials in slow conducting cells are resistant to T T X [1,15,16] is in agreement with this. Slow conducting cells include most, if not all, nociceptive neurons [4]. T T X - R action potentials, at least in C-cells, usually contain a Ca 2+dependent component which contributes to the inflection on the downstroke of the action potential [15,16]. Although both T F X - R and TTX-S currents are voltage dependent, the T T X - R current has slower kinetics, and the potential at which the sodium inward current is half inactivated is about 30 mV more positive than for the T r x - s current [7,11,12]. This significant difference in the levels for inactivation was utilized to selectively block spike generation in fibers with TTX-S Na+-chan nels by depolarizing them with high K ÷. Our data further confirm that action potential conduction in large afferent fibers was blocked during the exposure to high K ÷, but C-fibers remained active. In addition, these results suggest that K+-evoked depolarization facilitates action potential generation in small primary afferent fibers that have a threshold for sodium current activation more positive than A fibers [1,7,11,12]. Intracellular recordings from slow conducting primary afferents provide direct evidence of the effect of high potassium on small fibers, and we assume that the consequent synaptic activation of D H neurons was due to the activation of these small afferent fibers. Preservation of the postsynaptic activity of the D H neurons is consistent with this explanation. It is interesting that while the fast fiber input was blocked (by T T X and high potassium), the synaptic activation of the D H neurons by high intensity electrical stimulation of the dorsal root was often enhanced during potassium perfusion. This enhancement was related to the increase in excitability of the primary afferent fibers rather than to the depolarization of the postsynaptic neuron because EPSPs evoked by the identical stimulation of the adjacent D R remained unchanged or smaller. Since an increase in the excitability of fibers was observed in the presence of high K ÷, one may assume that during sustained firing of nociceptors (hyperalgesia and pain) the local accumulation of K + at the injury site may further enhance small fiber firing and, on the other hand, block large fibers. This could, at least in part, explain why hypoesthesia may occur during hyperalgesia. Our inability to completely block action potentials conduction in small unmyelinated fibers even with 10 /xM TI'X is not in agreement with the results obtained by Yoshimura and Jessell [17] in a transverse spinal cord slice preparation. One possible explanation is that in our preparation T T X is applied to the D R G and the portion of afferent fibers in the D R G compartment of the recording chamber, but in the experiments with the transverse slice, T T X was applied to dorsal root and spinal cord. Under those conditions, TI~X is acting on
133
D H neurons as well as terminals of primary afferents. Possibly, it blocks the spread of action potentials to presynaptic terminals.
Acknowledgements. I thank Dr. Mary Helen Greer for helpful comments on the manuscript. This work was supported by National Institutes of Health Grant NS 27751.
5. References [1] Caffrey, B., Eng, D.L., Black, J.A., Waxman, S.G. and Kocsis, J.D., Three types of sodium channels in adult rat dorsal root ganglion neurons, Brain Res., 592 (1992) 283-297. [2] Fulton, B., Postnatal changes in conduction velocity and some action potential parameters of rat dorsal root ganglion neurons, Neurosci. Lett., 73 (1987) 125-130. [3] Gorke, K. and Pierau, F.-K., Spike potentials and membrane properties of dorsal root ganglion cells in pigeons, Pflugers Arch., 386 (1980) 21-28. [4] Harper, A.A. and Lawson, S.N., Conduction velocity is related to morphological cell type in rat dorsal root ganglion neurones, J. Physiol., 359 (1985) 31-46. [5] Harper, A.A. and Lawson, S.N., Electrical properties of rat dorsal root ganglion neurons with different peripheral nerve conduction velocities, J. Physiol., 359 (1985) 47-63. [6] Jeftinija, S. and Urban, L., Anaysis of synaptic transmission in the spinal cord evoked by activation of dorsal root ganglion neurons with chemical irritants, Soc. Neurosci. Abstr., 17 (1991) 537. [7] Kostyuk, P.G., Veselovsky, N.S. and Tsyndrenko, A.Y., Ionic currents in the somatic membrane of rat dorsal root ganglion neurons. I. Sodium currents, Neuroscience, 6 (1981) 2423-2430. [8] Matsuda, Y., Yoshida, S. and Yonezawa, T., Tetrodotoxin sensitivity and Ca component of action potentials of mouse dorsal root ganglion cells cultured in vitro, Brain Res., 154 (1978) 69-82. [9] McLean, M.J., Bennett, P.B. and Thomas, R.M., Subtypes of dorsal root ganglion neurons based on different inward currents as measured by whole-cell voltage clamp, Mol. Cell. Biochem., 80 (1988) 95-107. [10] Meiri, H., Spira, G., Sammar, M., Namir, M., Schwartz, A., Komoriya, A., Kosower, E.M. and Palti, Y., Mapping a region associated with Na channel inactivation using antibodies to synthetic peptide corresponding to a part of the channel, Proc. Natl. Acad. Sci. USA, 84 (1987) 5058-5062. [11] Ogata, N. and Tatebayashi, H., Ontogenic development of the TTX-sensitive and TTX-insensitive Na + channels in neurons of the rat dorsal root ganglia, Dev. Brain Res., 65 (1992) 93-100. [12] Omri, G. and Meiri, H., Characterization of sodium currents in mammalian sensory neurons cultured in serum-free defined medium with and without nerve growth factor, J. Membrane Biol., 115 (1990) 13-29. [13] Orozco, C.B., Epstein, C.J. and Rapoport, S.I., Voltage activated sodium conductance in cultured normal, and trisomy 16 dorsal root ganglion neurons from fetal mouse, Dev. Brain Res., 3 (1988) 265-274. [14] Roy, M.L. and Narahashi, T., Differential properties of tetrodotoxin-sensitive and tetrodotoxin-resistant sodium channels in rat dorsal root ganglion neurons, J. Neurosci., 12 (1992) 2104-2111.
134
S. Jeftinija /Brain Research 639 (1994) 125-134
[15] Waddell, P.J. and Lawson, S.N., Electrophysiological properties of subpopulation of rat dorsal root ganglion neurons in vitro, Neuroscience, 36 (1990) 811-822. [16] Yoshida, S., Matsuda, Y. and Samejima, A., Tetrodotoxin-resistant sodium and calcium components of action potentials in
dorsal root ganglion cells of the adult mouse, J. Neurophysiol., 41 (1978) 1096-1106. [17] Yoshimura, M. and Jessell, T., Amino acid-mediated EPSPs at primary afferent synapses with substantia gelatinosa neurones in the rat spinal cord, J. PhysioL, 430 (1990) 315-335.
Brain Research 639 (1994) 125-134
Research Report
The role of tetrodotoxin-resistant sodium channels of small primary afferent fibers Srdija Jeftinija * Department of VeterinaryAnatomy, Neuroscience Program, and Signal Transduction Training Group, Iowa State University, Ames, IA 50011, USA (Accepted 26 October 1993)
Abstract
Intracellular recordings from neurons in the dorsal root ganglion (DRG) and dorsal horn (DH), in an in vitro spinal cord-dorsal root ganglion preparation, were used to investigate the role of tetrodotoxin-resistant (q-TX-R) afferent fibers in the sensory synaptic transmission in the superficial DH. Bath application of 25-50 mM potassium to the DRG depolarized the DRG neurons, blocked action potentials in the large neurons, evoked action potentials in slow conducting neurons, and synaptically excited dorsal horn neurons. Excitatory postsynaptic potentials (EPSP) which were evoked in DH neurons by electrical stimulation of large myelinated fibers, but not those evoked by stimulation of small unmyelinated fibers, were blocked by the potassium treatment of the primary afferents. Tetrodotoxin, when applied to the sensory neurons, abolished the action potentials in fast fibers but had no effect on the action potentials in a population of slow conducting afferents. Peripheral application of T r x blocked the fast EPSPs evoked by electrical stimulation but failed to block the electrically evoked slow EPSPs and the synaptic activation of DH neurons induced by the application of high potassium to sensory neurons. Furthermore, high potassium potentiated electrically evoked, TFX-resistant EPSPs in the majority of neurons. This effect was abolished in Na +-free solution. These findings indicate that high [K+]e applied to the DRG, dorsal root and peripheral process selectively activates a primary afferent input to the DH, which is sodium-dependent and tetrodotoxin resistant.
Key words: Primary afferent neuron; Tetrodotoxin-resistant; Spinal cord neuron; Sodium channel; Excitatory transmission
1. Introduction
Classification of primary afferent neurons of the mammalian D R G is based on several properties of neurons. Large myelinated A a , /3 neurons are fast conducting while small myelinated A ~ and unmyelinated C fibers are slow conducting. The shape and properties of the somatic action potentials vary according to cell type [2-5,16] and, in general, the slower the conduction velocity of a neuron the longer the duration of the action potential and the greater the chance that the action potential will display an inflection or 'kink' or ' p l a t e a u ' during the falling phase. It follows that there must be a considerable degree of overlap between neurons that have a plateau in the downstroke of the action potential and those which respond to noxious stimuli.
* Corresponding author. Fax: (1) (515) 294-3932. 0006-8993/94/$07.00 © 1994 Elsevier Science B.V. All rights reserved
SSDI 0 0 0 6 - 8 9 9 3 ( 9 3 ) E 1 4 9 8 - R
The different properties of action potentials in different D R G cell types suggest that there may be differences in the complement or behavior of the ion channels which control m e m b r a n e excitability. The sensory neurons of mammalian D R G were found to have at least two types of voltage-dependent sodium channels: fast, TTX-sensitive (TTX-S) and slow, TI'X-resistant ( T T X - R ) [7,9-12]. Electrophysiological analysis of T F X - R and T-I'X-S channels in rat D R G neurons was first reported by Kostyuk et al. [7]. This work and that of others indicates that these channels differ in their activation and inactivation kinetics [14]. It has been shown that large sensory neurons and their axons have a T-I'X-S fast sodium current, while small neurons have a T T X - R sodium current [9]. The T T X - R current differs from the T I ' X - S sodium current in many ways. Both are strongly voltage dependent; but the q T X - R current has slower kinetics, the threshold for activation is about 10 m V more positive, and the potential at which the current is half inactivated is about 30 m V
126
S. Jeftinija / Brain Research 639 (1994) 125-134
more positive than the TTX-S current [8,12,13]. Recently, patch voltage clamp recordings identified three distinct kinetic types of Na + current differentially distributed among D R G neurons [1]. Small D R G neurons co-express a rapidly inactivating TTX-S fast current and a slowly activating and -inactivating T T X - R slow current. Large cells express a single T I ' X - S Na + current which is identified as intermediate by its inactivation rate constant. These results suggest that by manipulating the properties of the sodium ionic channels of primary afferent neurons it should be possible to study interactions between the incoming signals to the spinal cord. T T X - R spikes in D R G neurons can be evoked in the absence of calcium [8], and there is a convincing body of evidence which shows that sub-populations of D R G neurons express a T T X - R sodium current which is large enough to support the generation of an action potential in the soma [7,9,11,13]. The major problem in studying pain physiology and pharmacology has been to generate selective C-fiber input to establish an in vitro model to study the synaptic mechanisms of nociception. The in vitro spinal cord slice preparations were developed to provide the necessary stability for long lasting intracellular recordings from dorsal horn neurons. In this p a p e r we offer experimental data that the neurons expressing T T X - R sodium channels are inactive primary afferents which can be activated by an increase in the extracellular potassium. Furthermore, the data demonstrate the presence of the T I ' X - R sodium channels on the peripheral axons of the slow sensory neurons. Preliminary findings were communicated [6].
2. Materials and methods Horizontal spinal cord slices were obtained from young Sprague-Dawley rats (18-30 days old). Following laminectomy under ether anesthesia, a segment of lumbosacral spinal cord (1-1.5 cm long) with dorsal roots and an associated D R G attached was quickly excised and immersed in gassed 95% 0 2 and 5% CO 2 Ringer's solution (in mM: NaCI 124, KCI 5, KH2PO 4 1.2, CaCI 2 2.4, MgSO 4 1.3, NaHCO 3 26, glucose 10, pH = 7.4). Animals were decapitated following excision of the spinal cord. After the removal of the pia mater, the spinal lumbar segments (L2 to L6) were affixed, dorsal surface up, with cyanoacrylic glue to the bottom of a plexiglass cutting chamber of a Lancer Series 1000 vibratome. The bath of the vibratome was filled with the aerated Ringer's solution. The blade of the vibratome was positioned below the dorsal surface of the spinal cord in order to yield a horizontal dorsal horn slice (400-500 tzm thick) with intact associated dorsal rootlets and attached DRG. After incubation at 31+ I°C for 1-2 h, the slice was transferred to a recording chamber. The 'submerged-type' recording chamber consists of four tissue compartments separated by narrow partitions. The adjacent dorsal roots were placed in narrow grooves on the top of the partition so that the spinal slice containing the dorsal horn neurons was in one
compartment with the dorsal surface up, while the D R G with dorsal roots were in different compartments. One D R G compartment was further subdivided to allow the peripheral nerve trunk to be placed in a narrow groove which led to another compartment. Compartments were independently perfused. This allowed differential perfusion and exposure of only one population of afferents to a test compound while the adjacent dorsal root or segment of the same root was used as a control. To insure total electrical separation of the spinal cord and D R G compartments and to avoid mixing of solutions between compartments, vaseline was used to form a seal around the dorsal roots in the grooves. The perfusing medium (in mM: NaCI 127, KCI 1.9, KH2PO 4 1.2, CaCI 2 2.4, MgSO 4 1.3, NaHCO 3 26, glucose 10, pH = 7.4) was oxygenated with 95% 0 2 and 5% CO 2 in a separate reservoir for each compartment and then introduced into the tissue compartments through thin polyethylene tubing at a rate of 2-3 m l / m i n . The temperature of the recording solution was kept at 31 + I°C. Isolated lumbar dorsal roots were placed on platinum wire bipolar stimulating electrodes. The thresholds for activating large myelinated and unmyelinated afferents by stimulating the peripheral segment of the dorsal root with a constant current stimulus were determined by recording action potentials in D R G neurons. They were found to be 8-20 V, 0.02 ms for the most sensitive axons and over 35 V, 0.5 ms for C-fibers. The classification of D R G neurons was done solely on the basis of the conduction velocities calculated by measuring the distance between the cathode and the D R G and dividing it by the conduction latencies of action potentials recorded in the cell bodes of DRG. The average conduction distance was 5.2 mm (n = 45). Fibers conducting at velocities above 15 m / s were classified as A/3. Fibers conducting at velocities below 15 m / s and above 1 m / s were classified as A6 fibers. Fibers conducting at velocities below 1 m / s were classified as C fibers. Intracellular recordings from DH neurons and D R G neurons were performed with glass microelectrodes back-filled with 3 M potassium acetate and having a DC resistance of above 100 MS2. Micropipettes were pulled from capillary-filled borosilicate glass tubing on a Brown-Flaming puller. The micropipettes were placed under visual guidance in the DH slice and the DRG. The passive electrical properties of the neurons were determined by means of a dual-channel intracellular amplifier (Neuro Data IR-283) which allows current injections on the order of 0.05-3.0 nA through the recording electrode. The recorded signals were displayed on the screen of an oscilloscope (Nicolet Instruments M4094) for further analysis of the data. The membrane potential was constantly monitored on a pen recorder.
3. Results Intracellular recordings were made from 30 slow conducting D R G neurons. Threshold intensity was > 20 V, 0.2 ms, and the action potential latency evoked by afferent stimulation corresponded to a conduction velocity of 0.7 _+ 0.08 m / s (mean _+ S.E.M.) when peripheral branches were stimulated and 0.5 + 0.04 m / s when central branches (dorsal root) were stimulated. The average resting m e m b r a n e potential of D R G neurons in this group was - 60 + 2 m V ( _+S.D.). Electrical stimulation of peripheral and central processes of sensory neurons induced action potentials of average amplitude 87 +_ 2 mV and duration, measured at the resting m e m b r a n e potentials, of 4 _+ 0.7 ms followed by a
S. Jeftinija ~Brain Research 639 (1994) 125-134
afterhyperoplarization of 11 + 1 mV amplitude and 31 + 5 ms duration.
3.1. Effect of high potassium and tetrodotoxin on slow primary afferents Brief exposure of the soma of D R G neurons to 50 mM potassium produced a rapid depolarization of the soma membrane (32.8 + 2.5 mV, n = 30). Once the peak of the depolarization was reached, continued exposure to potassium maintained a steady plateau. No
re
DRG ~_.~orsal root
A
50 mM K on DRG T r x 0.SpM on
1 mln
DRG
I ~ I I I I N U l I I N N I N I I I H M I I I U I I N m I I I I I H I l t H I I I I I I I I I H I I I U g l I I m m NHmmln~mmlOlmllWm#limnlNNllll
m
~ M K on DRG B
? 1mV
1
1 /
":,':,-'.-JlllllJ -
'
. . . . . .
....... U T r r =-'
.......
r .......
, r .......
1 rain
1 l~lSV !
Fig. 1. Excitatory action of high potassium on sensory neurons as recorded intracellularly from the soma of a D R G neuron (inset, re = recording electrode). A: bath application of 50 m M K ÷ to the D R G resulted in a 25 m V depolarization accompanied by firing of action potentials (upper trace, upward deflections on chart recording
are action potentials). Action potentials are truncated by the fiequency response of the chart recorder. Perfusion of TTX for 10 min failed to block the stimulatory effect of high potassium (lower trace). B: bath application of high potassium to the dorsal root resulted in firing of action potentials in the soma of this slow sensory neuron (conduction velocity 0.35 m/s). Oscilloscope tracings in upper row were obtained simultaneously(at faster time scale) with chart recordings, as indicated by arrows. In the presence of high potassium, the conduction velocityof the sensory neuron decreased from 0.35 m/s to 0.26 m/s. The electrical artifacts in the oscilloscopetraces are due to dorsal root stimulation. Resting membrane potential (Em) was - 58 mV.
127
desensitization was observed following repeated applications at 10-15 min intervals. About half of the tested D R G neurons (n = 16) did not fire action potentials, but the rest of the neurons fired action potentials on the rising phase of the 50 mM K+-evoked depolarization or throughout the potassium-induced depolarization (Fig. 1A, upper trace). The potassium-evoked depolarization in all neurons and the firing of action potentials in three neurons were not blocked by tetrodotoxin ('I-TX, 0.5 to 1/zM) (Fig. 1A, lower trace). In order to study the mechanism of conduction of the potassium-evoked activation of D R G neurons, potassium was applied to the central or peripheral processes of sensory neurons while recording from the soma (Fig. 1 inset, n = 15). In no experiment was a change in resting membrane potential of the soma recorded when the high potassium was applied to the processes of sensory neurons. In 12/15 D R G neurons, application of potassium to the processes resulted in a block of action potentials induced by electrical stimulation of the same segment. In three neurons, however, application of high potassium to the processes of the sensory neurons resulted in firing of action potentials recorded from the soma (Fig. 1B). In addition to the firing of 'spontaneous' action potentials, the action potentials evoked by electrical stimulation of the dorsal root stimulation were not blocked. An increase in the response latency from 50 ms to 65 ms was recorded indicating a decrease in conduction velocity from 0.35 m / s to 0.26 m / s (Fig. 1B, oscillograms). The number of long-lasting, stable intracellular recordings from small D R G neurons that fired tetrodotoxin-resistant ( T T X - R ) action potentials throughout potassium application was too low (10%) to study the role of T T X - R sensory neurons in primary afferent transmission. Therefore, we indirectly determined the functional role of T T X - R afferents by recording intracellularly from D H neurons in laminae I - I I I that are activated synaptically by primary afferents. This experimental approach amplified the signal coming from slow conducting afferent fibers because most of superficial D H neurons receive input from C-fibers.
3.2. Synaptic response of dorsal horn neurons to high potassium applied to sensory neurons." role of tetrodotoxin-resistant afferents Intracellular recordings were made from 82 D H neurons. The depth of the neurons ranged from 9 to 433 /zm (147 + 139, mean + S.D.). The mean resting membrane potential (Em) was - 7 2 + 2 mV ( + S.D.) and the input resistance ranged from 38 to 168 MO. All of the recordings were done in neurons located lateral to the dorsal root entry zone, and most of them were localized between roots L4 and L5.
S. Jeftinija / Brain Research 639 (1994) 125-134
128
All DH neurons were excited synaptically by electrical stimulation of the dorsal root or the peripheral nerve trunk. Single electrical shocks were calibrated to produce activation either of large fibers (10-20 V, 0.02 ms) or the whole fiber population, including unmyelinated afferents (supramaximal stimulus: > 35 V, 0.5 ms). Low intensity stimulation of afferent fibers (10-20
A
control
50 mM K on L5
V, 0.02 ms) induced fast EPSP. The mean amplitude was 12 + 1 mV, and the mean duration was 43 + 10 ms. The latencies suggested conduction velocities from 2 to 14 m / s . Fifteen neurons did not respond to low intensity stimulation of primary afferents. In an additional 15/82 neurons, the fast EPSP was followed by an inhibitory postsynaptic response (IPSP). High stimulus
50raM K on L4
II- 9 j , ~
wash
'~'J
"
L4
, mv 50
mm
L5
.la
,lO.V
L4 L5
B
L4 L5
L4 L5
peripheral
central
20V,0.02ms
20V, 0.02ms
1 mln
Ill
50V,0.5ms
IJl control
50mM K L4 peripheral
~
~
~
4
L_ 50raM K L4central
~-~,,,,,.~.~.. ,
~
, 50raM K both
I
~
Fig. 2. Effect of high potassium on excitability of primary afferent fibers. Records were obtained from two different neurons, A and B. A: slow chart recordings (bottom set of recordings), obtained by intracellular recording from a D H neuron, revealed synaptic activation when high potassium was applied to either the L4 or L5 D R G (inset). The afferent fibers in the adjacent roots were stimulated electrically and the effects of high potassium were recorded on an oscilloscope at a much faster rate (two rows of oscilloscope tracings). Application of high potassium to either dorsal root depressed the fast potentials from that root. Note that the oscilloscope tracings were taken at a time when the m e m b r a n e potential had returned to the resting level. B: application of high potassium to different segments of the afferent fibers (see inset) demonstrated that the potassium-induced potentiation of excitatory transmission conducted by small fibers does not require the action on the soma of the sensory neurons. Application of potassium to the peripheral processes inhibited 30 ms of the EPSP produced by stimulation of the same segment, and it was without effect on the EPSP produced by stimulation of the dorsal root (central). W h e n applied to the central compartment (compartment with L4 DRG), high potassium abolished low intensity-evoked responses from both stimulating sites, and the slow component of the peripherally evoked EPSP was much smaller (about 15%) than the response obtained during peripherally applied potassium. W h e n potassium was applied to both compartments, the slow EPSP in response to peripheral stimulation reappeared. A: Em = - 6 8 mV, depth of the neuron was 28 /xm. B: Em = - 7 2 mV, depth = 80/xm.
s. Jeftinija/ Brain Research 639 (1994) 125-134 intensity synaptically affected all of the DH neurons. In the majority of DH neurons (52/82), the slow component of the EPSP evoked by activation of C-fibers was a continuation of the fast EPSP. In the 15/82 neurons that did not respond to low intensity stimulation of primary afferents, high intensity stimulation produced a long delayed EPSP corresponding to a conduction velocity of 0.8 + 0.05 m/s. In the 15 neurons that responded with an EPSP-IPSP to low intensity stimulation, high intensity stimulation produced a complex response composed of the fast EPSP (16 + 4 mV, 136 5:47 ms) followed by the IPSP (6 + 1 mV, 237 + 84 ms), and then followed by a slow EPSP (5 + 0.5 mV, 28 + 7 s). This slow EPSP was well separated from the fast EPSP and in 6 neurons was accompanied by the firing of action potentials. The application of high K ÷ (25 to 50 mM) to the DRG produced a concentration-dependent excitation in the majority of DH neurons. One minute applications of 50 mM potassium to DRG produced an average depolarization of dorsal horn neurons of 7.6 + 2.1 mV. This high potassium effect was associated with an increase in the frequency of EPSP and firing of action potentials. The type of synaptic excitation in DH neurons varied from short bursts of EPSPs and action potentials to prolonged excitation that lasted for several minutes. To further study the mechanism of the potassiumevoked stimulatory action on primary afferents, we investigated the effect of high K ÷ applied to the sensory neurons on the synaptic responses evoked in DH neurons by electrical activation of primary afferents. Perfusion of 50 mM K ÷ to one of the DRG compartments induced an excitatory postsynaptic depolarization in the DH neuron (Fig. 2A, slow chart recordings) and completely abolished the EPSPs evoked by low intensity stimulation of the corresponding root. When the roots were activated with high intensity stimulation, the initial phase of the prolonged excitatory postsynaptic response was selectively and reversibly abolished from the root exposed to high potassium while the slow component of the EPSP appeared to be enhanced (Fig. 2A, oscilloscope tracings). The blocking effect of high potassium on transmission from fast fibers outlasted the period of potassium-induced excitatory depolarization for several minutes, and EPSP measurements were done at the resting membrane potential to exclude the membrane potential effect. The EPSP evoked by stimulation of the adjacent dorsal root at the same stimulus strength was unchanged (Fig. 2A, oscilloscope tracings). Bath application of high potassium to the nerve processes peripheral to the DRG (Fig. 2, L4-p inset) produced synaptic excitation of the DH neurons (n = 12) and abolished both the fast EPSP evoked by the low intensity stimulus (20 V, 0.02 ms) and the fast component of the postsynaptic response evoked by the
129
supramaximal stimulus applied at the L4-p stimulation site. It was without effect on responses evoked by stimulation al L4-c of the same afferents in the central compartment (Fig. 2B, second row of oscillograms). Application of high potassium to the central compartment (n = 8) in the same preparation abolished the fast components of postsynaptic responses from both stimulation sites and almost completely abolished the slow component evoked by peripheral stimulation (Fig. 2B, fourth row, second tracing from left). To demonstrate that the slow component from the peripheral segment is present only when primary afferents are depolarized, 50 mM potassium was applied to both compartments (n = 8), and the slow EPSPs evoked by stimulation at either site were recovered (Fig. 2B, bottom row of oscillograms). 3.3. The effects of tetrodotoxin on electrical and chemical activation of the primary afferent fibers After establishing, by recording from DRG cells, that there are TrX-R primary afferents that fire action potentials when depolarized by high potassium, studies on the effect of high potassium were continued by recording from postsynaptic DH neurons. Independent of the presence (T-FX applied to DRG) or absence (TTX applied to dorsal root or peripheral nerve trunk) of neuronal cell bodies, TTX abolished the fast component of the EPSP (n = 52), but in 38 of 52 neurons failed to block the slow component. Although it was possible to completely block the excitatory transmission in DH neurons evoked by electrical stimulation of high threshold primary afferents, high concentrations of TTX only decreased, but did not abolish, the excitation of DH neurons evoked by high potassium applied to the sensory neurons. Fig. 3 illustrates the time course of the effect of appliying TTX to a DRG while recording from a dorsal horn neuron. A 2 min perfusion application of TI'X (0.5 ~M) almost abolished the fast EPSP evoked by low intensity stimulus (left oscillogram in row b; 20 V, 0.02 ms) and abolished 30 ms of EPSP evoked by high intensity stimulus (right oscillogram in row b; 35 V, 0.5 ms). With time TTX's blocking effect was more pronounced, and after 15 min of perfusion of 0.5/xM TTX almost all EPSPs were abolished (row c and d). Since A-fibers were now blocked by TFX, it was possible to study the effect of high potassium on the slow C-fibers. Potassium applied to primary afferents treated with TTX produced synaptic activation of DH neurons (slow chart recordings) and potentiated the TTX-R component of electrically evoked EPSP (Fig. 3, oscillograms e and f). This kind of facilitation of excitatory transmission was recorded in 15 independent experiments (100%) and was measured as an increase of 304 + 64% in the size (area under the curve) of the slow EPSPs evoked by T-I'X-R fibers and
130 A
S. Jeftinija / Brain Research 639 (1994) 125-134
2OV, 0.02rns
35V, 0.5ms
20V, 0.02ms
report). W h e n these differences in r e s p o n s e latency were c o n v e r t e d to c o n d u c t i o n velocity, the high potassium-resistant E P S P ' s latency c o r r e s p o n d e d to a n average c o n d u c t i o n velocity of 0.39 + 0.04 m / s ( n = 8), a n d the T I ' X - R E P S P ' s latency c o r r e s p o n d e d to 0.27 + 0.03 m / s ( n = 8; P < 0.025). T h e response latency of the T I ' X - R E P S P decreased following application of high potassium to p r i m a r y afferents r e p r e s e n t i n g a significant increase in c o n d u c t i o n velocity from 0.27 + 0.03
35V, 0.5ms
control
2 rain 1"1"X 35 rain
t l l TTX 10 rain 1-1"X+ 25raM K on
ORG d]J
A
control
T r x 0.5ixM on DRG
l"rX + 50raM K otl DRG
TT'X 15 rain
I"TX + 2SrnM K on DRG
TTX 40 rain -
~
" r~~ 0.5
on DRO ~
/
f~lnS
,
rJI/IdIJ/Jdf/111//
25mV
50ms
B TTX 0.5pM on DRG -:::::--::::::-:-::::::::::::::::::::::::lllLql ~ liraUlmillUlMIIIIIlUlUlMImlmlmlm,ram u m mmamm 25rnM K on DRG
10 mV
mm lm mim m 1 mln
~
J
20mY
1 rain
Fig. 3. Blocking effect of TTX is time dependent. When applied to a sensory neuron, TTX (0.5 /.LM) eliminated the fast EPSP within 2 min and within 15 min almost completely abolished the synaptic response of this DH neuron. The application of high potassium, after 15 min of TTX, synaptically activated the DH neuron (B, slow chart recording) and induced the appearance of a slow EPSP (A, oscillograms e and f). Prolonged perfusion of afferents with TTX for 35 min failed to block this stimulatory effect of high potassium (A, oscillogram i). Em = -75 mV, depth 18 ~m.
an increase in the n u m b e r of action p o t e n t i a l s (from 1.2 + 0.4 to 4.5 + 1) fired d u r i n g these slow EPSP. A t the same time that these obvious c h a n g e s were i n d u c e d by T r x a n d high K ÷ applied to o n e D R G , transmission p r o d u c e d by s t i m u l a t i n g the a d j a c e n t dorsal root was completely u n c h a n g e d . F u r t h e r perfusion of T I ' X (35 m i n ) failed to p r e v e n t the facilitatory effect of high potassium o n the p r i m a r y afferent n e u r o n s (Fig. 3, oscillogram i). T h e facilitatory effect of high p o t a s s i u m o n p r i m a r y afferent n e u r o n s was not uniform, a n d Fig. 4 illustrates three types of responses o b t a i n e d by the application of T I ' X a n d t h e n high p o t a s s i u m to the p r i m a r y afferent n e u r o n s . I n n e u r o n s A a n d B, a q ' T X - R slow E P S P was p r e s e n t a n d in A this c o m p o n e n t was p o t e n t i a t e d w h e n potassium was applied ( n = 12) while in B p o t e n t i a t i o n was not r e c o r d e d ( n = 8). I n d e p e n d e n t of the effect of high p o t a s s i u m o n the size of the T-I'X-R EPSP, the response delay of the T T X - R E P S P was noticeably longer t h a n the response latency of the slow E P S P preserved following application of high potassium to afferent fibers (high p o t a s s i u m - r e s i s t a n t E P S P in this
" r ' ~ 0,5 gM ~ ORG ~ 1 1 1 1 1 1
f
.
~
-
/
l
/
/
/
-
/
i
.
s
.
J
.
-
i
.
o
'
/
,
J
~
ck_ll ~
~
/
i
/
/
/
,
~
/
/
/
/
/
/
/
50 msec
50mMK
Fig. 4. The effect of 0.5 ~M TTX (applied to the DRG compartment) on electrically and 50 mM K+-evoked membrane potential changes recorded in three different DH neurons. In all three neurons, TTX pretreatment of the sensory neurons failed to block the synaptic excitatory depolarization of the dorsal horn neurons by high potassium applied to the sensory neurons (slow chart recordings) A: TI'X abolished the first 50 ms of the EPSP evoked by electrical stimulation (middle oscillogram in the top row). High potassium applied in the presence of TTX-induced synaptic activation of the DH neuron and augmented the size of the EPSP evoked by electrical activation of C fibers (top row far right oscillogram). B: bath application of TTX abolished 40 ms of the EPSP evoked by the electrical stimulation of primary afferents but failed to block the synaptic excitation of DH neuron evoked by high potassium applied to the same DRG. In this neuron, the TTX-R component of the EPSP was not potentiated during potassium application. C: application of TTX to the DRG eliminated the electrically evoked EPSP but failed to block the excitatory effect of high potassium applied to the same DRG. In the presence of high potassium, a long latency EPSP was evoked by electrical stimulation of afferent neurons.
S. Jeftinija /Brain Research 639 (1994) 125-134 control
131
TTX O.S~M
2 mln later
_
_ _ _
_
~
S~V
S0ms
T r x 0.,S~M on DRG IINIHINIIHHIHIIHNNNHIflIflll
SOmM K on DRG
~
0 mV
,'5 eec 50V, 0.Sins
Fig. 5. Facilitation of excitatory transmission induced by high potassium applied to sensory neurons does not depend on membrane potential. Bath application of high potassium to sensory neurons in the presence of TTX produced a synaptic excitation of this DH neuron followed by a hyperpolarizing response. This hyperpolarization was associated with a block of spontaneous firing while EPSPs evoked by electrical stimulation of exposed afferents were potentiated. The doted lines represent membrane potentials when individual EPSPs were recorded. The dashed line marks the resting membrane potential. EPSPs at b and e were recorded at about the same hyperpolarizing potential but the EPSP in b was about two-fold bigger than the EPSP in e. The EPSP in c was bigger than the EPSP in d although they were recorded at about the same membrane potential. Em = - 72 mV, depth = 96/.Lm.
m / s to 0.33 _+ 0.03 m / s ( n = 8; P < 0.1). T h e n e u r o n in C d i d n o t h a v e a T T X - R E P S P at t h e r e s t i n g p o t e n t i a l , a n d o n l y w h e n p o t a s s i u m w a s a p p l i e d to t h e s e n s o r y n e u r o n s was a T T X - R E P S P r e c o r d e d ( n = 10). T h e s e t h r e e n e u r o n s a n d t h e r e m a i n i n g 79 D H n e u r o n s w e r e s y n a p t i c a l l y a c t i v a t e d w h e n 50 m M p o t a s s i u m w a s a p p l i e d to t h e s e n s o r y n e u r o n s (Fig. 4, slow chart recordings). The facilitatory effect of high potassium on TTX-R transmission outlasted the period of
p o t a s s i u m - i n d u c e d e x c i t a t o r y d e p o l a r i z a t i o n by s e v e r a l minutes, and EPSP measurements were done when the cell r e t u r n e d to t h e r e s t i n g m e m b r a n e p o t e n t i a l f o l l o w ing s y n a p t i c d e p o l a r i z a t i o n . In four neurons, the application of high potassium r e s u l t e d in a d e p o l a r i z a t i o n f o l l o w e d by a h y p e r p o l a r i z a t i o n . T h e s e n e u r o n s w e r e u s e d to e s t a b l i s h t h e r e l a tionship between membrane polarization and potentiation of excitatory transmission. Both the depolarizing
A control
L4-p 35V, 0.5rns
"l-rX on L4-p
L4-p 35V, 0.5ms
TI'X+SOmM K on L4-p
L4-p 35V, 0.Sms
100 ms
B L4-p S0mM K 90 sec
C L ~ TrX 0.rmU :7~7~7:~ ~ : ~ z T ~ : ~ : : : : ~ : : ] m m . m u m u m m 0 m m l t 0 m . m a m m M m a m
.mn0..mmmm.mumm0m,mlm.nNmlmuwm#mnm
t . ~ r~mM K
l,o.v 3 0 ~
Fig. 6. Peripheral processes of sensory neurons express TTX-R sodium channels. Recordings were made from one DH neuron. Application of high potassium to the peripheral processes synaptically activated the DH neuron (B,C) and potentiated the TTX-R component of the EPSP evoked by electrical stimulation of the same processes (A). Em = - 7 4 mV, depth = 96/zm, Rm = 151 MO.
S. Jeftinija / Brain Research 639 (1994) 125-134
132
and hyperpolarizing components of the high potassium response were resistant to the application of T T X (Fig. 5, slow chart recording). Although the hyperpolarizing component of the high potassium-evoked response was associated with a decrease in 'spontaneous' synaptic firing, the amplitude of electrical stimulus-evoked q T X - R slow EPSP and the firing of action potentials were increased during this phase (Fig. 5, oscillograms). 3.4. Axons of sensory neurons express TTX-R sodium channels Our results from the application of high potassium to axons of sensory neurons while recording from soma of D R G neurons or D H neurons suggested the presence of slow sodium channels on processes of sensory neurons. These channels were previously demonstrated on soma of cultured neurons in whole-cell patch-clamp experiments [7]. To provide supporting data for the presence of T T X - R sodium channels on the processes of sensory neurons, recordings were made from the soma of D H neurons while stimulating the peripheral processes of sensory neurons in the stump of the spinal control
nerve. In 12 (100%) neurons the application of T T X to peripheral processes abolished the fast EPSP produced by stimulating the peripheral proces, but this had no effect on responses induced by stimulating afferents in the central compartment (not shown). In all 12 neurons, the application of high potassium to the peripheral processes following the TTX pretreatment resuited in an excitatory depolarization comparable to the one recorded in the control (Fig. 6B,C). In addition, potentiation of the q'-I'X-R slow EPSP was recorded in 6 / 1 2 experiments (Fig. 6 oscillograms). 3.5. Effect of Na +-free environment on TTX-R responses Differential effects of divalent cations on TTX-R and TTX-S currents were reported previously [7,14]. To demonstrate the involvement of sodium channels in the excitation of afferent fibers in the presence of TTX, we bathed afferent fibers in a sodium-free solution. Replacing the sodium ion with TRIS (n = 6) abolished all effects of high potassium and electrical stimulation applied to the same segment of the primary afferents (Fig. 7). These effects of TRIS were reversible after washing for 2-5 min with the Na+-con taining recording solution.
mM K on DRG 1
4. Discussion
I lomV 1 rain TTX 0.5~M on DRG I
Trx + TRISon
DRG
~ / / / / / / / , / / / ' / / / . / / / / / / h
/
TTX post TRIS
x
Fig. 7. Stimulatory effect of high potassium applied to sensory neurons and recorded in DH neurons was TTX-R but was sodium dependent. Bath application of high potassium to sensory neurons synaptically activated the D H neuron. This effect was resistant to the application of TTX but was reversibly eliminated by incubating the sensory neurons in a sodium-free solution. Sodium was replaced with TRIS (tris[hydroxymethyl]aminomethane) in equimolar concentrations. Em = - 6 6 mV, Rm =61 MI2, depth 45/zm.
Our results reveal the ability of primary afferent neurons to respond differently to changes in extracellular potassium. Using a combination of electrical and chemical activation of primary afferent fibers under in vitro conditions, we found that the presence of fast and slow sodium channels provides a sufficient substrate for the blocking effect of high potassium on low threshold afferents and its facilitatory effect on high threshold afferent fibers. Our approach was to block the action potential generation in large fibers by manipulating the Na+-channels with a depolarizing agent (high K +), or with the sodium channel blocker tetrodotoxin, or with a combination of these. The later approach revealed the presence of primary afferent neurons that are unexcitable by electrical square pulses at the resting membrane potential and that become active only when depolarized by elevated extracellular potassium. 4.1. TTX-resistant sodium channels and small primary afferent fibers The shapes and properties of the somatic action potentials vary according to the D R G cell type [2,4,5,16]. These findings may reflect differences in the distribution a n d / o r behavior of the ion channels which control membrane excitability in different D R G cell groups. The finding that action potentials of some cell
S. Jeftinija / Brain Research 639 (1994) 125-134
types are blocked by T T X while action potentials in slow conducting cells are resistant to T T X [1,15,16] is in agreement with this. Slow conducting cells include most, if not all, nociceptive neurons [4]. T T X - R action potentials, at least in C-cells, usually contain a Ca 2+dependent component which contributes to the inflection on the downstroke of the action potential [15,16]. Although both T F X - R and TTX-S currents are voltage dependent, the T T X - R current has slower kinetics, and the potential at which the sodium inward current is half inactivated is about 30 mV more positive than for the T r x - s current [7,11,12]. This significant difference in the levels for inactivation was utilized to selectively block spike generation in fibers with TTX-S Na+-chan nels by depolarizing them with high K ÷. Our data further confirm that action potential conduction in large afferent fibers was blocked during the exposure to high K ÷, but C-fibers remained active. In addition, these results suggest that K+-evoked depolarization facilitates action potential generation in small primary afferent fibers that have a threshold for sodium current activation more positive than A fibers [1,7,11,12]. Intracellular recordings from slow conducting primary afferents provide direct evidence of the effect of high potassium on small fibers, and we assume that the consequent synaptic activation of D H neurons was due to the activation of these small afferent fibers. Preservation of the postsynaptic activity of the D H neurons is consistent with this explanation. It is interesting that while the fast fiber input was blocked (by T T X and high potassium), the synaptic activation of the D H neurons by high intensity electrical stimulation of the dorsal root was often enhanced during potassium perfusion. This enhancement was related to the increase in excitability of the primary afferent fibers rather than to the depolarization of the postsynaptic neuron because EPSPs evoked by the identical stimulation of the adjacent D R remained unchanged or smaller. Since an increase in the excitability of fibers was observed in the presence of high K ÷, one may assume that during sustained firing of nociceptors (hyperalgesia and pain) the local accumulation of K + at the injury site may further enhance small fiber firing and, on the other hand, block large fibers. This could, at least in part, explain why hypoesthesia may occur during hyperalgesia. Our inability to completely block action potentials conduction in small unmyelinated fibers even with 10 /xM TI'X is not in agreement with the results obtained by Yoshimura and Jessell [17] in a transverse spinal cord slice preparation. One possible explanation is that in our preparation T T X is applied to the D R G and the portion of afferent fibers in the D R G compartment of the recording chamber, but in the experiments with the transverse slice, T T X was applied to dorsal root and spinal cord. Under those conditions, TI~X is acting on
133
D H neurons as well as terminals of primary afferents. Possibly, it blocks the spread of action potentials to presynaptic terminals.
Acknowledgements. I thank Dr. Mary Helen Greer for helpful comments on the manuscript. This work was supported by National Institutes of Health Grant NS 27751.
5. References [1] Caffrey, B., Eng, D.L., Black, J.A., Waxman, S.G. and Kocsis, J.D., Three types of sodium channels in adult rat dorsal root ganglion neurons, Brain Res., 592 (1992) 283-297. [2] Fulton, B., Postnatal changes in conduction velocity and some action potential parameters of rat dorsal root ganglion neurons, Neurosci. Lett., 73 (1987) 125-130. [3] Gorke, K. and Pierau, F.-K., Spike potentials and membrane properties of dorsal root ganglion cells in pigeons, Pflugers Arch., 386 (1980) 21-28. [4] Harper, A.A. and Lawson, S.N., Conduction velocity is related to morphological cell type in rat dorsal root ganglion neurones, J. Physiol., 359 (1985) 31-46. [5] Harper, A.A. and Lawson, S.N., Electrical properties of rat dorsal root ganglion neurons with different peripheral nerve conduction velocities, J. Physiol., 359 (1985) 47-63. [6] Jeftinija, S. and Urban, L., Anaysis of synaptic transmission in the spinal cord evoked by activation of dorsal root ganglion neurons with chemical irritants, Soc. Neurosci. Abstr., 17 (1991) 537. [7] Kostyuk, P.G., Veselovsky, N.S. and Tsyndrenko, A.Y., Ionic currents in the somatic membrane of rat dorsal root ganglion neurons. I. Sodium currents, Neuroscience, 6 (1981) 2423-2430. [8] Matsuda, Y., Yoshida, S. and Yonezawa, T., Tetrodotoxin sensitivity and Ca component of action potentials of mouse dorsal root ganglion cells cultured in vitro, Brain Res., 154 (1978) 69-82. [9] McLean, M.J., Bennett, P.B. and Thomas, R.M., Subtypes of dorsal root ganglion neurons based on different inward currents as measured by whole-cell voltage clamp, Mol. Cell. Biochem., 80 (1988) 95-107. [10] Meiri, H., Spira, G., Sammar, M., Namir, M., Schwartz, A., Komoriya, A., Kosower, E.M. and Palti, Y., Mapping a region associated with Na channel inactivation using antibodies to synthetic peptide corresponding to a part of the channel, Proc. Natl. Acad. Sci. USA, 84 (1987) 5058-5062. [11] Ogata, N. and Tatebayashi, H., Ontogenic development of the TTX-sensitive and TTX-insensitive Na + channels in neurons of the rat dorsal root ganglia, Dev. Brain Res., 65 (1992) 93-100. [12] Omri, G. and Meiri, H., Characterization of sodium currents in mammalian sensory neurons cultured in serum-free defined medium with and without nerve growth factor, J. Membrane Biol., 115 (1990) 13-29. [13] Orozco, C.B., Epstein, C.J. and Rapoport, S.I., Voltage activated sodium conductance in cultured normal, and trisomy 16 dorsal root ganglion neurons from fetal mouse, Dev. Brain Res., 3 (1988) 265-274. [14] Roy, M.L. and Narahashi, T., Differential properties of tetrodotoxin-sensitive and tetrodotoxin-resistant sodium channels in rat dorsal root ganglion neurons, J. Neurosci., 12 (1992) 2104-2111.
134
S. Jeftinija /Brain Research 639 (1994) 125-134
[15] Waddell, P.J. and Lawson, S.N., Electrophysiological properties of subpopulation of rat dorsal root ganglion neurons in vitro, Neuroscience, 36 (1990) 811-822. [16] Yoshida, S., Matsuda, Y. and Samejima, A., Tetrodotoxin-resistant sodium and calcium components of action potentials in
dorsal root ganglion cells of the adult mouse, J. Neurophysiol., 41 (1978) 1096-1106. [17] Yoshimura, M. and Jessell, T., Amino acid-mediated EPSPs at primary afferent synapses with substantia gelatinosa neurones in the rat spinal cord, J. PhysioL, 430 (1990) 315-335.