0306-4522/92 $5.00 + 0.00 Pergamon Press Ltd 0 1992 IBRO

Neuroscience Vol. 49, No. 4, pp. 913-923, 1992 Printed in Great Britain

DIVERSE ACTIONS OF 5HYDROXYTRYPTAMINE ON FROG SPINAL DORSAL HORN NEURONS IN VITRO H. TAN* and V. MKLETIC~ Departments of Comparative Biosciences and Ne~ophys~olo~, University of Wi~ns~-Madi~n, Madison, WI 53706, U.S.A. Ahatract-The effects of 5-hydroxytryptamine on the membrane potential and input resistance of 86 dorsal horn neurons were studied using intracellular recordings in isolated, hemisected spinal cords of adult frogs (Ranu pipirns). Bath application of serotonin (5-100 PM) caused membrane depolarizations in 58 (67%) neurons, hyperpolarixations in 12 (14%) cells, biphasic responses in nine (11%) neurons, and no detectable change in seven (8%) cells. In some neurons depolarized by serotonin, the amine‘s responses could be mimicked by the sekctive 5-HTz agonist ( f jl(2,5~me~yoxy4i~opheny1)-2-~no~o~ne hydrochloride and the S-HT,, agonist ~p~-me~yl-5-hydrox~~~~ne, and blocked by the 5-HT,,, antagonists ketanserin and mianserin. In other neurons depolarized by serotonin, the S-HT, agonist 2-methyl&hydroxytryptamine mimicked, and the 5-HT, antagonist, 3-tropanyl-3,5dichlorobenxoate, blocked the serotonin-induced responses. Depolarizing responses due to activation of S-HT,,.,, receptors were generally accompanied by increases in the membrane input resistance, whereas depolarizations mediated by 5-HT, receptors were associated with a decreased membrane input resistance. Superfusion with tetrodotoxin or low-Ca2+~~-M~+~n~ning media abolished about half of the depolarizing responses. H~~la~~tions caused by serotonin were associated with a decrease in membrane input resistance, and might have been due to activation of a potassium conductance. These responses persisted in bathing solutions containing tetrodotoxin or low-Ca2+/high-Mgs+. The S-HT,, agonist 8-hydroxy-2(di-N-propylamine)tetralin hydrobromide mimicked, whereas the 5-HT,A antagonist spiroxatrine blocked, these hyperpohuixing responses. Other antagonists selective for 5-HT,,, or 5-HT, receptors were without effect. Serotonin-produced biphasic responses consisted of either an initial depolarization followed by a hyperpolarixation or the reverse. The selective S-I-IT, agonist ( rf: ~1(2,5-d~ethyoxy4i~ophenyl)-2~nopro~ne hy~~~o~de could only mimic the d~~~~tions, whereas the 5-NT,, agonist 8hydroxy-2-(di-~-propyl~ne)tetmlin hydrobromide produced only the hyperpolarixations. Spiroxatrine, a 5-IiT,,, antagonist, blocked only the hyperpolarixations without affecting the depolarizations, and methysergide, a non-specific 5-HT receptor antagonist, depressed both the depolarizations and hyperpolarizations. Serotonin also appeared to affect spinal dorsal horn neurons indirectly because it produced excitatory postsynaptic potentials, inhibitory postsynaptic potentials, and a mixture of both. In addition, the amine appeared to suppress sensory transmission in the dorsal horn because it decreased the size of dorsal root-evoked excitatory postsynaptic potentials in some cells, and increased the stimulation threshold for dorsal root-evoked action potentials in other neurons. The diverse actions of serotonin on frog dorsal horn neurons are dependent upon the activation of different 5-HT receptor types which may all contribute to the amine’s inhibition of nociceptive signal processing in the spinal dorsal horn.

Stimulation of certain discrete brainstem sites can produce profound analgesia. This blockade of pain transmission is thought to result from activation of descending pathways, and is postulated to occur in the dorsal horn of the spinal cord. Many anatomical, *Present address: Department of Anatomy, Medical College

of Ohio, P.O. Box 10008, Toledo, OH 43699, U.S.A. tTo whom correspondence should be addressed at: Department of Comparative Biosciences, University of Wisconsin-Madison, 2015 Linden Drive West, Madison, WI 53706, U.S.A. Abbreuiutions: DOI, ( &-)-l-(2,5dimethoxy4iodophenyl)2-aminopropane hydr~~o~~, EPSP, excitatory postsynaptic potentiat, 5-HT, 5-hydrox~~ne, serotonin; alpha-me-5-HT, a-methyl-5-hydroxytryptamine; 2-me-S-HT, 2-methyl-5-hydroxytryptamine; IPSP, inhibitory postsynaptic potential; MDL 72222, 3-tropanyl-3,5dichlorobenzoam, MS 222, tricaine methanesulfonate; &OH-DPAT,( f )-&hydroxy-2-(diN-propylamino)-tetralin hydrobromide; TTX, tetrodotoxin. 913

el~trophysiolo~~l, pha~a~lo~~l and behavioral studies suggest that serotonin (5hydroxytryptamine, S-HT) plays an important role in this descending modulation of the dorsal horn neurons.5*6v’1,38 Nevertheless, virtually nothing is known about the cellular mechanism(s) underlying S-HT modulation of dorsal horn ne~ona1 activity. Previous studies of 5-HT action on individual spinal dorsal horn neurons have been limited to extracellular recordings and iontophoretic applications. Administration of 5-HT in r&o typically results in inhibition of nociceptive neurons, although excitatory effects on deep dorsal horn neurons and on the s~n~neous activity of some neurons were also observed.5~~“~3* In order to examine 5-HT’s action on the neuronal membrane, and describe the amine’s overall circuitry of action in the dorsal horn, we have begun using an in vitro preparation of the spinal cord of the frog. The presence of a dense plexus of S-HT-

914

H. TAN and V. MILETIC

immunoreactive fibers has been demonstrated in the frog dorsal horn, 1 and we have shown that bulbospinal serotoninergic pathways in the frog are homologous to those of mammals. 33 Previously we have briefly described the electrophysiologic characteristics of frog spinal dorsal horn neurons, and have reported that 5-HT has varied effects on their membrane potential. 34 The purpose of the present study was to determine the specific receptor types mediating these varied effects of 5-HT in the frog spinal dorsal horn. A preliminary account of this work has appeared in abstract f o r m Y EXPERIMENTAL PROCEDURES

Commercially available adult frogs (Rana pipiens) of either sex were used. All animals were fed with live crickets, and housed in the laboratory to allow acclimation to room temperature (20-22°C) for at least two weeks prior to use. Animals were anesthetized by injecting 0.2 mg/g of tricaine methanesulfonate (MS 222) in the lymph sac in the back of the animal. The spinal cord was exposed at lumbar levels by removing the overlying bone with fine rongeurs. The VIIIth or IXth dorsal roots were also exposed proximal to the ganglion. Following complete exposure in situ, the spinal cord (with its dorsal roots attached) was removed to a dissecting dish where the cord was hemisected sagittally in pre-oxygenated Ringer's solution at 4°C. The hemisected cord was then transferred to the recording chamber (volume 1.1 ml) and pinned (cut surface up) to the Sylgard-coated bottom of the chamber. The tissue was continually perfused at 4~5 ml/min with a solution that was bubbled with 95% 02-5% CO2, and that contained (in raM): NaC1 112, KC1 2.5, CaC12 2.0, MgC12 1.2, NaHCO3 17, NaH2PO 4' 2H20 0.1, and glucose 5.6, at pH 7.4. In low-Ca2+/high-Mg2+-containing media, the Ca 2+ concentration was lowered to 0.2 mM and the Mg 2+ concentration raised to 10 mM. All experiments were performed at room temperature (20-22°C), and recordings commenced after a 1-h "equilibration" period. Intracellular recordings were performed with micropipettes filled with 4 M potassium acetate having d.c. resistances of 100-150MfL Membrane potential was altered by passage of current through the recording electrode using the bridge circuit of a high-input impedance amplifier (Axoclamp 2A, Axon Instruments). During the experiments, 0.1q).2-nA hyperpolarizing current pulses (200ms in duration) were continuously passed through the electrode to permit observation of changes in membrane input resistance. Membrane potential and current injected were continuously monitored on a digital oscilloscope and a chart recorder, and were stored on tape for later analysis. Recording electrodes were inserted into the dorsal horn with the aid of a dissecting microscope. A suction or bipolar concentric electrode was placed on the lumbar dorsal roots for synaptic activation of dorsal horn neurons. Single, constant-current (50-150 #A), square-wave pulses (0.05 ms) were used for dorsal root stimulation. Drugs used were: 5-hydroxytryptamine creatinine sulfate (5-HT, Sigma), tetrodotoxin (TTX, Sigma), methysergide maleate (a generous gift from Sandoz), ketanserin tartrate, mianserin, ( __+)- 1- (2, 5- dimethoxy-4-iodophenyl)- 2aminopropane hydrochloride (DOI), (_)-8-hydroxy-2(di-N-propylamino)tetralin hydrobromide (8-OH-DPAT), 2-methyl-5-hydroxytryptamine (2-me-5-HT), e-methyl-5hydroxytryptamine (alpha-me-5-HT), MDL 72222 (3-tropanyl-3,5-dichlorobenzoate) and spiroxatrine (all from Research Biochemicals, Inc.). All drugs were applied by changing the perfusion solution to one which contained known concentrations of drug(s)

with three-way taps so that the perfusion rate did not change. The time required for the changed solution to flow from the tap to the chamber was about 15 s. To minimize the difference between drug concentration in the bath and at receptor sites (due to uptake or degradation) the flow rate was set at 6 ml/min (to increase turnover per chamber volume), and final drug concentrations from stock were made just before use. Changes of solutions were performed while continually monitoring the intracellular potential to check for drift. RESULTS

At r o o m temperature (20-22°C) the spinal cord preparation was viable for more than 24 h. Stable intracellular recordings that routinely lasted more than 3 h were obtained from a total of 251 dorsal horn neurons. The average value for the resting membrane potential was - 6 7 . 5 + 11.7 mV (n = 118, range=-50 to - 9 8 m V ) , and that for the input resistance was 117.2_+ 67.6 Mf~ (n = 61, range = 20-340 Mf~). The effects of 5-HT, and its agonists and antagonists, were tested on 86 neurons. Because we could visualize the area of electrode placement through the dissecting microscope, all of these recordings were limited to the superficial dorsal horn. Previous studies have shown that cutaneous primary afferents, serotoninergic terminals and descending bulbospinal pathways are all concentrated in these regions of the frog spinal c o r d J '33 Superfusion of the cord with 5-HT ( 5 - 1 0 0 # M ) produced depolarizations in 58 cells (67%), hyperpolarizations in 12 neurons (14%), biphasic responses in nine neurons (11%), and no detectable change in seven cells (8%). Examples of each type of response are illustrated in Fig. 1. These three types of responses to 5-HT could be differentiated in terms of their pharmacology.

5HT (40~M)

85HT

(30#M)

C5HT

(40#M)

20 mv 1 rain

Fig. 1. Examples of the diversity of 5-HT actions on frog spinal dorsal horn neurons. (A) Bath application of 5-HT (40 p M, bar) produced a typical depolarization in this neuron. Downward deflections are electrotonic potentials in response to hyperpolarizing current pulses (0.1 nA, 200 ms). (B) In this cell 5-HT (30 #M, bar) produced a hyperpolarization. (C) 5-HT (40 pM, bar) first depolarized then hyperpolarized this neuron. Resting potentials were --64 mV (A), - 7 6 m V (B), and - 5 6 m V (C).

S-Hydroxytryptamine and dorsal horn neurons Depolarizations

The amplitude of S-HT-produced depolarizations in a given cell was do~de~dent, although the peak value varied from cell to cell (range = 2-23 mV). The average size of the depolarization produced by an initial dose of 4OpM 5-HT was 9.0 + 5.1 mV (mean f S.D., n = 28). The depolarization began soon after the S-HT reached the spinal cord (15-20 s), and the membrane potential returned to baseline 2-Smin after the perfusion with 5-HT was discontinued. We employed several selective 5-HT,,, or 5-HT, agonists and antagonists to determine which serotonin receptor types mediate the amine’s depolarizing responses in spinal dorsal horn neurons. Of 38 neurons depolarized by 5-HT and tested with these selective agents, 18 (47%) were sensitive to 5-HT,,,,, and 15 (39%) to 5-HT, drugs. Three of the 38 neurons could be depolarized by either 5-HT,c,* or 5-HT, agonists, but could not be blocked by any 5-HT receptor antagonist. The remaining two cells, although depolarized by 5-HT, did not respond to any 5-HT agonist or antagonist we tried. To examine whether the 5-HT-produced depolarizations were exerted directly onto the neurons recorded from, we compared these responses in normal Ringer’s with those during synaptic transmission blockade, i.e. in solutions containing TTX (1 PM) or low-Cat+ (0.2 mM)/high-Mg*+ (10 mM). The efficacy of synaptic transmission blockade was determined by failure to record dorsal root-evoked potentials. In four cells, the depolarizations were blocked during perfusion with these solutions, whereas in six cells the responses persisted. There was no correlation between receptor subtype (5 HT,c,* or 5-NT,) and persistence of the depolarizing response in TTX or low-Ca*+/high-Mg2+-containing media. The 5-HT, agonist DOP3 (10 PM) and the 5-HT,c,* agonist alpha-me-5-HT2’ (10 fi M) mimicked serotonin’s depolarizing response in 11 neurons. In five of these cells the depolarizing responses were accompanied by an increase in membrane input resistance which became especially evident when the membrane potential was manually clamped to its pretreatment level. However, in the remaining six neurons, no obvious change in the membrane input resistance was apparent even when the membrane potentials were brought back to pretreatment levels. The 5-HT,,,, antagonists ketanserin (1 PM) and mianserin (15 hM) blocked the 5-HT-induced depolarizations in nine tested neurons (Fig. 2C, D). Neither of these two drugs produced changes in the resting membrane potentials when administered alone and at the above concentrations. At higher concentrations (5 and 10 fi M), however, ketanserin depolarized the membrane, and in two neurons the depolarizations were sufficient to cause firing. The

915

5-HT, receptor antagonist, MDL 7222226 (10 FM), had no effect on two tested cells that were affected by ketanserin and mianserin. The 5-HT, receptor agonist, 2-me-5-HT,2’ applied at 15 ,uM, mimicked 5-HT’s depolarizing responses in 15 dorsal horn neurons (Fig. 3A). These depolarizations were usually associated with a decrease in membrane resistance (Fig. 3B). However, as with the 5-HT,,,-produced depolarizations, in some instances the 5-HT, receptor mediated depola~zations were not accompanied by any obvious changes in the membrane input resistance. The depolarizations could be consistently evoked with repeated applications (two to five times), and our visual inspection revealed no obvious desensitization of the response. MDL 72222 (10 FM), an antagonist at 5-HT, receptors, produced no effect of its own, and blocked the depolarizations induced by 5-HT in all five cells tried (Fig. 3B). Ketanserin or mianserin failed to block the depolarizations antagonized by MDL 72222 (n = 3).

Hyperpolarizations were observed in 12 of 86 neurons tested. The average amplitude of the hyperpolarizations produced by an initial 40 pM dose of 5-HT was 4.1 + 1.9 mV (mean + SD., n = 10). As with the depola~~tions, there was va~ability in the peak amplitude of the response from cell to cell (range = 2-7 mV). In all instances tried (n = 8), the 5-HT-produced hyperpolarizations persisted during superfusion with solutions containing TTX or low-Ca*+/high-Mg*+, indicting a direct effect of the amine on the dorsal horn neuron affected. To determine which 5-HT receptor type was involved in mediating the hyperpolarizations, we used the highly selective 5-HT,, analog, 8-OH-DPAT.4 Administration of 8-OH-DPAT (10pM) was found to mimic the hy~r~larizing effect of 5-HT in all five neurons tested (Fig. 4A). This effect of 8-OH-DPAT did not change in the presence of TTX (1 FM). Spiroxatrine (10 PM), a reported selective 5-HT,, receptor antagonist,23 blocked the hyperpolarizing responses produced by 5-HT (Fig. 4A, n = 4). In contrast, ketanserin and MDL 72222 were without effect. Decreases in membrane input resistance generally accompanied 5-HT-produced hyperpolarizing responses, and these could again best be seen when the membrane potential was brought back to pretreatment values. In four cells, the S-HT-induced change in membrane resistance was calculated from the slope of current-voltage curves before, during and after 5-HT treatment. As illustrated in Fig. 4B, the amine decreased the membrane resistance as compared to the control. The reversal potential, taken as the point of intersection of the curves, was about -98 mV.

H. TAN and V. MILETIC

916

B

A

5HT (100pM)

5HT (40~M)

DOI (10~M) J

ALPHA-ME-5HT (10#M)

i0 mv

1 min

C

,

D

5HT (5~M)

i~,a/,

I "~

t'"L

'az'"

- ""

,.~.~l,,.Im

.,L .......

,m, * ~ / . ,

.L

i .-.

~L .

5nT

-.lla..l.j.ma.j,.,Al,.l_j

(4o.M)

L ~ ,2e.~

5HT (5~M) + Ketanserin (I~M)

|

1

20 mv

5HT (40~M) + Mianserin (15~M)

1 rain Fig. 2. Some depolarizing responses were mediated by 5-HTuc/2 receptors. (A) 5-HT (bar) depolarized the cell (top panel) and the effect was mimicked by bath application of the selective 5-HT 2 receptor agonist DOI (bar, bottom panel). (B) 5-HT (100 # M) depolarized another dorsal horn neuron (bar, top panel). Note that an increase in membrane input resistance is clearly seen when the membrane potential is manually clamped to pretreatment levels. The 5-HT-produced depolarization in this neuron was mimicked by the 5-HT~c/2 agonist alpha-me-5-HT (bar, bottom panel). (C) In a solution containing the 5-HT~c/2 receptor antagonist ketanserin (bottom panel), the depolarization produced by application of 5-HT is blocked. (D) Similarly, in a solution containing the 5-HT~c/2 receptor antagonist mianserin (bottom panel), the depolarizing effect of 5-HT is also abolished. Resting potentials were --74 mV (A), - 7 0 mV (B), -- 80 mV (C) and -- 72 mV (D).

A

5HT

(40#M)

2-ME-5HT

(15~M)

85HT

J 1 min

(40#M)

5HT (40#M) + MDL 72222

(10#M)

m

Fig. 3. Examples of 5-HT 3 receptor-mediated depolarizations. (A) The depolarization caused by 5-HT (bar, top panel) could be mimicked by the 5-HT~ agonist 2-methyl-5-HT (bar, bottom panel). (B) This depolarizing response of 5-HT (bar, top panel) was blocked in a solution containing the 5-HT 3 antagonist MDL 72222 (bottom, panel, 5-HT was again applied at bar). The break in the trace in the bottom panel is a perfusion artifact. Resting potentials were --64 mV (A) and - 5 2 mV (B).

SHydroxytryptamine

A

5HT

917

and dorsal horn neurons 5HT

(4OpM)

(40/M) +

Spiroxatrine

(10m)

Control

I)-OH-DPAT

5IiT (401rM)

(10@4) Wash

Agonist

6

-o-

Control

-0-

-75

J

5-HT i

20

mv

min

-83 5 E

-91

a P >”

-99

-107

Fig. 4. The amine’s hyperpolarizing response was mediated by 5-HT,, receptors. (A) S-HT (bar, left top panel) produced a long-lasting hyperpolarization. This effect was mimicked by the SHT,, receptor agonist 8-OH-DPAT (bar, bottom left panel), and blocked in a solution containing the SHT,, receptor antagonist spiroxatrine (right top panel; 5-HT application is again denoted by a bar). The long-lasting 5-HT-produced hyperpolarixation returned after spiroxatrine was removed from the bath (right, bottom panel). Resting membrane potential was - 74 mV (B) Current-voltage relationship for another neuron hyperpolarized by bath application of 5-HT (40 PM). The reversal potential, taken as the point of intersection of the curves, was about - 98 mV. Resting membrane potential was - 76 mV.

Biphasic responses Biphasic responses were seen in nine of 86 neurons tested. The frequency of their occurrence was not

dependent on the concentration of 5-HT applied. The responses were composed of either an initial depolarization followed by a hyperpolarization (Figs 1C and 5B, n = 5) or the reverse (Fig. 5A, n = 4). The average amplitudes of the depolarizations and hyperpolarizations were 5.1 f 1.7 mV and 5.3 f 3.2 mV (mean f S.D., n = 6), respectively. When the second part of the response was a hyperpolarization, the response was usually long-lasting (from 7 up to 30 min). In TTX-containing media (n = 3), the hyperpolarizing part of a biphasic response to 5-HT persisted, whereas the depolarizing component was attenuated (Fig. 5A, bottom panel). We have also examined the 5-HT receptor types mediating the biphasic responses in three cells. As illustrated in Fig. 5, administration of the selective 5-HTz agonist DO1 (10 PM) mimicked only the depolarizing part of the response in these neurons (Fig. SA, B, third panels). On the other hand, the

5-HT,, receptor agonist I-OH-DPAT (10 PM) produced only the hyperpolarizing part of the response in these same three cells (Fig. 5A, second panel). The 5-HT,A receptor antagonist spiroxatrine blocked only the hyperpolarization without affecting the depolarization (Fig. 5B, bottom panel). The non-specific 5-HT antagonist methysergide= (2pM) reduced both the depolarization and the hyperpolarization (Fig. 5B, panel). These data suggest that second the depolarizations were due to activation of 5-HT,, and the hyperpolarizations to 5-HTIA receptor types. Indirect efects of 5-hydroxytryptamine application In addition to the effects on the membrane potential, 5-HT also affected the synaptic activity in some dorsal horn neurons. Bath application of 5-HT could produce excitatory postsynaptic potentials (EPSPs, n = 14, Fig. 6A), inhibitory postsynaptic potentials (IPSPs, n = 5, Fig. 6B), or a mixture of EPSPs and IPSPs (n = 8, Fig. 6C). The 5-HT-induced activity could be abolished in low-Ca*+, high-M$+ or TTXcontaining solutions (not shown). This suggests an

H. TAN and V. MILETIC

918

A

5HT

6

(60f.0.l)

5HT

(40j~M)

5HT (4OfiM) + Methysergide

(lOfiN)

8-OH-DPAT

DO1

DO1

(lO@M)

5HT

(60wM)+TTX

5HT

(21rM)

(lO@M)

(40pM) +

(NM) Spiroxatrine

_I

-I

1 min

1 min

(lOtiM)

20 mv

10 mv

Fig. 5. Biphasic responses were mediated by S-HI’,, and 5-HT, receptors. (A) Bath application of 5-HT (bar, top panel) first hyperpolarized then depolarized the membrane. 8-OH-DPAT mimicked only the hyperpolarizing part of the responses (bar, second panel), and DO1 produced only the depolarization (which was accompanied by a slight increase in the membrane input resistance; bar, third panel). Application of J-HT in a ‘ITX-containing solution (bar, bottom panel), indicated that the hyperpolarization was direct, but the depolarization was indirect (i.e. it was abolished during synaptic transmission blockade). (B) Another example of the amine’s biphasic response. 5-HT (bar, top panel) first depolarized the membrane, and then caused a long-lasting hyperpolarization. Upward deflections are EPSPs evoked by repetitive stimulation of the dorsal roots. In a solution containing methysergide (second panel), both the depolarization and hyperpolarjzation are attenuated (5-HT applied at bar). DO1 (bar, third panel) only mimicked the depolarizing response, whereas a solution containing spiroxatrine only blocked the hyperpolarizing response (again, S-HT applied at bar). Resting potentials were - 69 mV (A) and - 72 mV (B).

indirect effect of the amine on the recorded neurons, mediated either by action on spinal interneurons or the primary afferents themselves. Another apparent indirect effect of 5-HT was exerted on EPSPs evoked by dorsal root stimulation. As shown in Fig. 7, S-HT could either decrease the amplitude of the dorsal root-evoked EPSPs (n = 3, Fig. 7A), or increase the electrical stimulation threshold needed to evoke action potentials in the dorsal horn neurons (n = 2, Fig. 7B). These effects were seen, although there were no obvious changes in membrane potential or input resistance.

DISCUSSION Previous intracellular studies of the action of S-HT in the spinal cord only focused on lateral or ventral

horn neurons,20-32-37 and to our knowledge, this is the first comprehensive intracellular study of 5-HT action on dorsal horn neurons. Our data indicate that S-HT has diverse actions in the frog that are mediated by different receptor types, and appear to involve distinct ionic mechanisms. Ligand-binding and autoradiographic mapping studies have reported the existence of S-HT,, S-HT, and 5-HT, receptor subtypes in the mammalian spinal cord.‘*~“**’ Similar

5-Hydroxytryptamine A

Control

919

and dorsal horn neurons Wash

5liT (4Ofm -h-

-

B

Wash

5HT (4Ofm

Control

I

SHT (40/&M)

Wash

50 ms

Fig. 6. Synaptic activity produced by bath application of 5-HT. The amine generated EPSPs (A), IPSPs (B), and both EPSPs annd IPSPs (C) with no obvious changes in the membrane potential. Resting potentials were -59 mV (A), -56 mV (B), and -51 mV (C). information

is unavailable

for the amphibian

spinal

cord. Our study suggests the existence of at least the SHT,,, SHT,,,, and SHT, receptor subtypes in the frog spinal dorsal horn. Depolarizations Depolarizations

horn

neurons

in some of were apparently

the frog mediated b

-

SHT

C

a+b

Wash

-P------

*-ii= Ba

dorsal by S-

HT,c/2receptors. These depolarizations were accompanied by an increase in membrane input resistance, and could be blocked by the 5-HT,c12 antagonists ketanserin and mianserin, but not by other 5-HT receptor type antagonists. The 5HT, agonist DO1 and the 5-HT,,,, agonist alpha-me-5-HT mimicked such depolarizing responses.

f-control

I--

b

C

SHT I

-

209

-/L _-

4

4

209 PA

SHT

+--

PA

4

415 PA

d

Wash I

IL-

20 mv

220 PA

20 ms

Fig. 7. Effects of 5-HT on dorsal root-evoked EPSPs. (A) In this neuron 5-HT produced a depression in the amplitude of a dorsal root-evoked EPSP. Before any 5-HT application, stimulation of the dorsal root evokes an EPSP (a). Following perfusion with 5-HT (40 PM), the dorsal root-evoked EPSP decreases in amplitude (b). The amplitude of the evoked EPSP recovers to pretreatment levels following cessation of superfusion with 5-HT (c). The right-most panel presents traces a + b superimposed to more clearly illustrate the decrease in EPSP amplitude produced by 5-HT. These responses were not accompanied by any obvious changes in membrane potential or input resistance. (B) Another effect of 5-HT on dorsal root-evoked EPSPs was to increase the electrical stimulation threshold. The intensity of the stimulus pulse necessary to evoke an action potential from the dorsal root was 209 PA in normal Ringer solution (a). After application of 5-HT (40 PM), the same intensity of stimulation failed to evoke an action potential (b). Only when the intensity was doubled to 415 PA did an action potential occur (c). Following 5-HT removal, the threshold returned to the pre-5-HT value (d). Resting membrane potentials were -73 mV (A) and - 77 mV (B).

920

H. TAN

and

We did not attempt to further differentiate between the pharmacologically and biochemically closely related SHT,, and SHT, receptor types3’ Because all depolarizations antagonized by ketanserin and mianserin were mimicked by DO1 (an agonist at SHT, but not 5-HT,, receptors) we presume that the activation of 5-HT, receptors is responsible for these depolarizations. Our results are consistent with the findings in the rat nucleus accumbens,24 association cortex,’ and in frog I6 or neonatal rat3’ motoneurons. 5-HT-induced depolarizations in other dorsal horn neurons appeared due to activation of 5-HT, receptors. These depolarizations were accompanied by a decrease in membrane input resistance, and could be mimicked by the 5-HT, agonist 2-me-5-HT, and blocked by the 5-HT, antagonist MDL 72222 (but not other 5-HT antagonists). Depolarizations mediated by 5-HT, receptors have been reported in many areas, e.g., rabbit nodose and superior cervical ganglia,2* guinea-pig submucosal One common plexus3’ and mouse hippocampus. observation in these studies was the relatively rapid desensitization of 5-HT, receptors after repeated application of 5-HT. In our experiments no desensitization was seen. The slow drug-delivery method used in our experiments could, however, have easily obscured any partial desensitization, but similar findings have recently been reported in rat dorsal root ganglion neurons.36 In three neurons we observed that the 5-HTinduced depolarization could not be blocked by . application of any 5-HT,,, or 5-HT3 receptor antagonists alone. This may indicate the presence of both 5-HT,,,, and 5-HT, receptors on the same neuron. Recently, Araneda and Andrade3 have reported that 5-HT action in many frontal cortical neurons is mediated by both 5-HT,, and 5-HT, receptors located on the same layer V pyramidal cell. Alternatively, these depolarizing responses might have been mediated by a different 5-HT receptor subtype. Recently, Chaput et al.’ have described a novel 5-HT receptor that resembles the 5-HT,, or 5-HT, subtypes, and that is responsible for the development of 5-HT,,,2- and 5-HT,-insensitive, slowlydeveloping but prolonged depolarizations in hippocampal neurons. Whether or not this novel receptor type exists in the spinal dorsal horn remains, however, to be established. Previous studies have shown that the depolarizations mediated by 5-HT, receptors are usually accompanied by an increase in input resistance. This is thought to result from a decreased conductance to potassium.20s24,37In contrast, the activation of 5-HT3 receptor results in a decrease in input resistance, which is apparently due to the opening of a cationic Our finding that the 5(Na+, K+) channel.10~3i~39 HT ,c,Z-receptor-mediated responses are generally accompanied by increases, and the 5-HT,-receptormediated responses by decreases in the membrane

V. MILETIC

input resistance, suggest that similar ionic mechanisms are involved in 5-HT’s action in the frog dorsal horn. Many neurons recorded in our study did not, however, show any change in membrane input resistance during 5-HT applications. It is unlikely that this was due to voltage-sensitive membrane rectification, because after the membrane potential was manually clamped to pretreatment levels, the change in membrane resistance was still unclear. This failure to detect input resistance changes might result from simultaneous activation of multiple 5-HT receptors on the same cell (as suggested above). Consequently, the effects of altering the membrane conductance by different receptors might cancel each other out. An alternative explanation may be that the recording in the soma was not sensitive enough to detect membrane conductance changes occurring at distal dendritic sites. We do not know the distribution pattern of 5-HT terminals on the dendritic trees of frog spinal dorsal horn neurons. It is known in the cat, however, where the 5-HT terminal distribution is proximal rather than distal.22 By using superfusion solutions containing TTX or low-Ca*+/high-Mg*+ to block synaptic transmission, we were able to determine that some of the depolarizing responses were due to a direct action of 5-HT on the recorded neurons. Other depolarizing responses, however, were indirect, and possibly mediated by the release of transmitters from spinal interneurons or primary afferents. These results are consistent with the recent report in frog motoneurons.‘6 Hyperpolarizations

5-HT also produced membrane hyperpolarizations in frog dorsal horn neurons. In all instances tried, the hyperpolarizing responses persisted in TTX-containing superfusion solutions, suggesting a direct action of 5-HT on the recorded dorsal horn neurons. This confirms the most recent studies of 5-HT actions in the ventral horn of the frogi and neonatal rat.)’ Unlike the observations in frog motoneurons, however,16 we saw no correlation between the occurrence of hyperpolarizing responses and the concentration of 5-HT applied. We have no explanation for this discrepancy. 5-HT’s hyperpolarizing responses in the frog dorsal horn were mediated by 5-HT,, receptors, since they could be mimicked by the 5-HT,, agonist &OHDPAT and blocked by the 5-HT,, antagonist spiroxatrine, but not by antagonists for S-HT,,,, or 5-HT, receptors. These results are again in agreement with previous findings.2~‘5*37 The 5-HT-induced hyperpolarizations in the frog dorsal horn were usually associated with a decrease in membrane input resistance. This is in agreement with previous reports on the ionic mechanism of 5-HT actions which indicate that 5-HT,,-produced

S-Hydroxytryptamine and dorsal horn neurons hyperpolarizations ium channels.2*‘8

are due to the opening of potass-

Biphasic responses Although receptor subtypes mediating 5-HTproduced depolarizations and h~~la~~tions have been well studied, ph~~lo~~ characteristics of these components in a biphasic response are less well known. In the frog spinal dorsal horn, the depolarizing portions of biphasic responses were mediated by S-HT, receptors, whereas the hyperpolarizations were due to activation of the 5-HTu receptors. We employed TTX-containing solutions to determine whether either portion of the biphasic responses would be modified during synaptic transmission blockade. The results showed that the depolarizing responses were attenuated or completely abolished, but the hyperpolarizing responses persisted. This suggests that the specific receptor subtypes mediating these responses were not located on the same dorsal horn neuron. Due to the small number of cells tested (n = 3), the possibility that multiple receptor types are present on other dorsal horn neurons cannot be excluded. Serotonin’s inhibitionof nociceptivetransmissionin the spinal dorsal horn The dorsal horn of the spinal cord is the site where the initial stage of the central processing of nociceptive signals occurs. The primary afferent fibers carrying nociceptive info~ation enter the spinal dorsal horn, and make synapses on the processes and somata of projection neurons and interneurons. The projection neurons then relay the sensory information to the thalamus and other brain areas. It has been well documented that dorsal horn neurons receive dense serotoninergic inputs from the brainstem (especially the raphe nuclei), and that these serotoninergic terminals form at least one component of a descending system exerting profound modulatory actions on nociceptive neurons in the dorsal hom.5,6*“*38Little is known, however, about the details of the neural circuitry involved in serotonin’s action. Based on previous studies and our present results, we envision at least three possible complementary ways by which the descending fibers could release 5-HT to inhibit nociceptive information signalling in the spinal dorsal horn. First, 5-HT could directly inhibit nocieeptive projection neurons by activating 5-HTiI, receptors to cause hyperpolarizations. Second, 5-HT may activate 5-HT,,,, and S-NT, receptors to depolarize, and excite, inhibitory interneurons. These interneurons in turn would inhibit nociception in the dorsal horn either directly by inhibition of projection neurons, or indirectly by presynaptic action on the nociceptive primary afferent fibers. Third, S-HT may itself activate as yet unknown receptor types to presynaptically depress transmitter release from primary afferent fibers.

921

The net result of the amine’s action in the dorsal horn would thus be a depression of firing of the recorded neuron, and (if that neuron is a nociceptive cell) a net inhibition of nociceptive processing in the dorsal horn. Even without directly inhibiting a given nociceptive neuron, 5-HT would, thus, inhibit nociceptive ~ans~ssion in the spinal cord. This may explain the preponderance of excitatory responses recorded in our study in the light of the known overall inhibitory action of 5-HT in both in v&o recordings and in behavioral experiments.5~6*“~38 Serotonin’s action on primary aflerent fibers Evidence exists to support a direct inhibitory action of 5-HT on nociceptive projection neurons.5,6J1,38 5-I-K’s interactions with primary afferents are less well understood. There is little anatomical evidence to suggest that 5-HT fibers participate in axo-axonic synapses in the dorsal hom.‘92’n” 5-HT,A receptors are, nonetheless, present on the primary a&rent fibers of rats? and it has been previously shown that iontophoretic application of 5-HT increases the threshold for antidromic activation of primary atTerent fibers in the at.7 Bath application of the amine similarly increases the excitability of neonatal rat primary afferents in vitro.” More recently, Holohean et al.15have reported that S-HT causes both hyperpolarizations and depolarizations in sucrose-gap recordings from frog primary afferents, and Wang and Dun3’ have observed that 5-NT can significantly alter synaptic activity, and depress the ~plitude of EPSPs, in neonatal rat motoneurons. Our data similarly provide support for an additional indirect component in the amine’s action in the dorsal horn. 5-HT could either decrease the amplitude of primary-afferent-evoked EPSPs or increase the electrical stimulation threshold for primary-afferent-evoked action potentials. This suggests that 5-HT was acting on the primary afferents themselves. Diverse types of S-H?: receptors appear to mediate the amine’s actions on primary afferents as well. Recent experiments suggest the involvement of SHT,~‘5*37annd 5-HT, receptor subtypesi The former might cause primary @rent h~~la~~tion, the iatter primary afTerent depolarization.‘5 In addition, 5-HT, and S-HI-, receptors have been implicated in the amine’s depolarization of dorsal root ganglion cell.” Additional studies are needed to clarify the mechanism(s) of 5-HT action on primary afferent fibers in the spinal dorsal horn. CONCLUSIONS

In this study 5-HT could activate 5-HT,,, or 5-HT, receptors to produce memb~ne depolarizations in a large number of frog dorsal horn neurons. The 5-HT,, receptor-mediated responses could be associated with a decrease in membrane conductance,

922

H.

TAN

and V. MILETIC

whereas the 5-HT, ~ceptor-mediated responses were associated with an increase in membrane conductance. In a smaller number of neurons, 5-HT could bind to 5-HT,, receptors and hyperpolarize the membrane. This was shown to be a direct postsynapti~ effect of 5-HT. Biphasic responses could also be observed after perfusion with 5-HT, and these responses appeared to be mediated by 5-FIT,, and S-NT, receptors. .5-HT also appeared to suppress sensory transmission in the dorsal horn by reducing the size of dorsal root-evoked EPSPs in some neurons, and

increasing the stimulating

threshold for dorsal rootevoked action potentials in other neurons. The diverse actions of 5-HT on frog dorsal horn neurons are dependent upon the activation of different S-HT receptor types, which may all contribute to the amine’s inhibition of nociceptive signal processing in the spinal dorsal horn. ~~~~o~~e~~e~e~r~-We wish to thank Drs M. Behan, L. Haberly, W. Rhode and L. Stanford for helpful discussions and critically reading the manuscript. This study was supported in part by USPHS National Institutes of Health grant NS21278 to V. Miletie.

REFERENCES

1. Adli D. S. H., Rosenthal B. M., Yuen G. L., Ho R. H. and Cruce W. L. R. (1988) hnmunohistochemical localization of substance P, somatos~tin, enkephalin, and serotonin in the spinal cord of the northern leopard frog, KaM pipiens. J. camp. New& 275, 106-116. 2. Andrade R. and Nicoll R. (1987) Pharmacologically distinct actions of serotonin on single pyramidal neurons of the rat hippocampus recorded in vitro. J. Physiol., bnd. 394, 99-124. 3. Araneda R. and Andrade R. (1991) %Hydroxytryptamine, and 5-hydroxytryptamine,. receptors mediate opposing responses on membrane excitability in rat association cortex. Neuroscience 40, 399-412. 4. Arvidsson L.-E., Hacksell U., Nilsson J. L. G., Hjorth S., Carlsson A., Lindberg P., Sanchez D. and Wikstrom H. (1981) I-OH-DPAT, a new centrally acting 5-hydroxyt~tamine receptor agonist. J. med. Chem. 24, 921-923. 5. Basbaum A. 1. and Fields H. L. (1978) Endogenous pain control mechanisms: review and hypothesis. Ann. Neuml. 4, 451-462. 6. Besson J.-M. and Chaouch A. (1987) Peripheral and spinal mechanisms of nociception. Physic& Rev. 67, 67-186. 7. Carstens E., Klumpp D., Randic M. and ~mme~ann M. (1981) Effect of iontophoret~~ly appli~-S-hydroxytr~ tamine on the excitability of single primary afferent C- and A-fibers in the cat spinal cord. Brain Res. 220, 151-158. 8. Chaput Y., Araneda R. C. and Andrade R. (1990) Pharmacological and functional analysis of a novel serotonin receptor in the rat hippocampus. Eur. J. Pharmac. 182, 441-456. 9. Daval G., Verge D., Basbaum A. I., Bourgoin S. and Hamon M. (1987) Autoradiograph~c evidence of serotonin, binding sites on primary afferent fibers in the dorsal horn of the rat spinal cord. ~earosei. Left, 83, 71-76. 10. Derkach V., Surprenant A. and North R. A. (1989) S-HT, receptors are membrane ion channels. Nature 339,706709. Il. Gebhart G. F. (1986) Modulatory effects of descending systems on spinal dorsal horn neurons, In Spinal Afirent Processing (ed. Yaksh T. L.), pp. 391-416. Plenum Press, New York. 12. Glaum S. R. and Anderson E. G. (1988) Identification of 5-HT, binding sites in rat spinal cord synaptosomal membranes. Eur. J. Phurmac. 156, 287-290. 13. Glennon R. G., Seggei M. R., Solne W. H., Herrick-Davis K., Lyon R. A. and Titeler M. (1988) [‘2sI~-I-(2,~-Dimethoxy4-iodophenyl)-2-aminopropane: an iodinated radioligand that specifically labels the agonist high affinity state of 5-HT, serotonin receptors. J. med. Chem. 31, 5-7. 14. Hentall I. D. and Fields H. L. (1983) Actions of opiates, substance P, and serotonin on the excitability of primary afferent terminals and observations on interneuronal activity in the neonatal rat’s dorsal horn in uitro. ~~rase~e~ee 9, 521-528. 1.5. Holohean A. M.. Hackman J. C. and Davidoff R. A. (1990) An in vitro study of the effects of serotonin on frog primary afferent terminals. Neurosci. Lett. 113, 1X5-180. 16. Holohean A. M., Hackman J, C. and Davidoff R. A. (1990) Changes in membrane potential of frog motoneurones induced by activation of serotonin receptor subtypes. neuroscience 34, 555-W. 17. Nuang J. C. and Peroutka S. J. (1987) Identification of 5”hydroxytryptamine, subtypes in rat spinal cord. Brain Res. 436, 173-176. 18. Joels M., Shinnick-Gallagher P. and Gallagher J. P. (1987) Effect of serotonin and serotonin analogues on passive membrane properties of lateral septal neurons in oitro. Bruin Res. 417, 99-107. 19. Light A. R., Kavookjian A. M. and Petrusz P. (1983) The ultrastructure and synaptic connections ofserotonin”immunoreactive terminals in spinal laminae I and II. So~fose~sory Res. 1, 35-S. 20. Ma R. C. and Dun N. J. (1986) Excitation of lateral horn neurons of the neonatal rat spinal cord by 5-hydroxytryptamine. Devl Brain Res. 24, 89-98. 21. Maxwell D. J., Leranth C. and Verhofstad A. A. J. (1983) Fine structure of serotonin-containing axons in the marginal zone of the rat spinal cord. Brain Res. 266, 253-259. 22. Miletic V., Hoffert M. J., Ruda M. A., Dubner R. and Shigenaga Y. (1984) Serotoninergic axonal contacts on identified cat spinal dorsal horn neurons and their correlation with nucleus raphe magnus stimulation. J. camp. Neural. 228, 129-141.

23. Nelson D. L. and Taylor E. W. (1986) Spiroxatrine: a selective serotonin,, receptor antagonist. Eur. j. Pharmuc. 124, 207-208. acts at 5-HT, receptors to decrease potassium conductance 24. North R. A. and Uchimura N. (1989) 5-Hydroxyt~pta~ne in rat nucleus accumbens neurons. J. Physioi., Land. 471, l-12. 25. Pazos A., Cortes R. and Palacios J. M. (1985) Quantitative autoradiographic mapping of serotonin receptors in the rat brain. II. Serotonin-2 receptors. Brain Res. 346, 231-249. 26, Richardson B. P. and Engel G. (1986) The pharmacology and function of S-HT, receptors. Trends Neurosci. 9,424-428. 27. Richardson B. P., Engel G., Donatsch P. and Stadler P. A. (1985) Identjfication of serotonin M-receptor subtypes and their specific blockade by a new class of drugs. Narure 316, 126-I 31.

5Hydroxytryptamine

and dorsal horn neurons

923

28. Round A. and Walhs D. I. (1986) The de~la~ng action of 5-hydroxyt~t~ne on rabbit vagal afferent and s~pathetic neurones in vitro and its selective btockade by ICS 20.5-930. Br. J. Pharmac. f@, 485-496. 29. Ruda M. A., Coffield J. A. and Steinbusch H. W. M. (1982) I~un~t~hemic~ analysis of serotonergic axons in laminae I and II of the hnnbar spinal cord of the cat. J. Neurosci. 2, 1660-1671. 30. Schmidt A. W. and Peroutka S. J. (1989) 5-Hydroxytryptamine receptor “families”. Fedn Am. Sot. exp. Biol. J. 3, 2242-2249. 3 1. Surprenant A. and Crist J. (1988) Electrophysiological characterization of functionally distinct S-hydroxytryptamine receptors on guinea-pig submucous plexus. Neuroscience 24, 283-295. 32. Takahashi T. and Berger A. (1990) Direct excitation of rat spinal motoneurones by serotonin. J. Physiol., Lond. 423, 63-76. 33. Tan H. and Miletic V. (1990) Bulbospinal serotoninergic pathways in the frog Runa pipiens. J. camp. Neural. 292,

291-302. 34. Tan H. and Miletic V. (1990) Electrophysiological properties of frog spinal dorsal horn neurons and their responses to serotonin: an intracellular study in the isolated hemisected spinal cord. Bruin Rex 528, 344348. 35. Tan H. and Miletic V. (1991) Diverse actions of 5-hydroxyt~tamine on frog spinal dorsal horn neurons in vitro. Abstr. 3rd IBRU World Congr. Neurosci., p. 287. 36. Todorovic S. and Anderson E. G. (1990) S-HT, and 5-HT3 receptors mediate two distinct depolarizing responses in rat dorsal root ganglion neurons. Bruin Res. 511, 71-79. 37. Wang M. Y. and Dun N. J. (1990) 5-Hydroxytryptamine responses in neonate rat motoneurones in vitro. J. Physiol., Lond. 430, 87-103. 38. Willis W. D. (1984) The raphe-spinal

system. In Brain Stem Control of Spinal Cord Function (ed. Barnes C. D.), pp. 141-214. Academic Press, New York. 39. Yakel J. L. and Jackson M. B. (1988) 5-HT, receptors mediate rapid responses in cultured hippocampus and a clonal cell line. Neuron 1, 615-621. (Accepted

13 March 1992)