Intracellular responses of raphe magnus neurons during the jaw-opening reflex evoked by tooth pulp stimulation

Brain Research, 379 (1986) 232-241 Elsevier

232 BRE 11890

Intracellular Responses of Raphe Magnus Neurons during the Jaw-Opening Reflex Evoked by Tooth Pulp Stimulation P. MASON. A. STRASSMAN and R. MACIEWICZ Pain Physiology Laboratory. Neurology Service. Massachusetts General Hospital and the Neuroscience Program, Harvard Medical School. Boston. MA 02114 (U.S.A.) [Accepted December 17th. 1985~ Key words: pain

tooth pulp

jaw-opening reflex

raphe magnus - - permqueductal gray region

Neurons in the nucleus raphe magnus (RM) may play an important role in the modulation of nociception. To determine how RM neurons are activated during a nociceptive reflex, the intracellular responses of raphe neurons were studied during the jaw-opening reflex (JOR) elicited by tooth pulp shock in lightly anesthetized cats. Tooth pulp stimulation produces reflex EMG activation of the digastric muscle at a latency of 7-11 ms, resulting in jaw opening. Tooth pulp shock that elicits the JOR also produces an EPSP in a subset of raphe neurons. This EPSP consists of an early small depolarization that occurs at a latency of 10-15 ms followed by a larger depolarization at a latency of 20-60 ms. In all cases the latency to EPSP is longer than the latency to digastric EMG onset. Electrical stimulation of the 4 paws elicits oligosynaptic EPSPs in the same cells at a latency of 16-20 ms. Electrical train Stimulation of the midbrain periaqueductal gray region (PAG) suppresses the JOR. Single shock stimulation at the same PAG sites that suppress the JOR evokes monosynaptic EPSPs in the large majority of raphe neurons recorded. In all cases, the threshold for EPSP is below the threshold for suppression of the JOR. The EPSP amplitude is a direct function of PAG stimulus intensity and there is temporal summation ol EPSPs evoked by paired PAG shocks, At condition-test intervals of 40-90 ms. train stimulation of PAG suppresses the tooth pulpevoked EPSP in raphe neurons, The threshold for EPSP suppression occurs at a PAG stimulation intensity below that required for suppresssion of the JOR, The present findings provide evidence that RM neurons may play an important role in the modulation of the tooth pulp-evoked JOR, but only after the initial withdrawal reflex has occurred

INTRODUCTION Neurons in the nucleus raphe magnus (RM) may play an important rote in the m o d u l a t i o n of nociception. Electrical stimulation in R M suppresses nociceptive reflexes 11'17A8'43"53 and inhibits the responses to noxious stimuli of sensory trigeminal and spinal dorsal horn cells4"l°'15'2°'27"32"48; these effects are mediated by descending raphe projections that travel in the dorsolateral fasciculus5'°. Stimulation of the midbrain ventral periaqueductal gray region ( P A G ) also inhibits nociceptive reflexes 11"25'26"42'45"52. This effect of P A G stimulation is diminished by lesions of the caudal raphe and adjacent reticular formation 193944. evidence that the P A G pathway important for antinociception is relayed, at least in part. through RM. Although m a n y studies suggest a role for RM in

the modulation of nociception, little is known about the activity of R M n e u r o n s during a noctceptive reflex. Therefore in the present study intracellular recordings were made from RM n e u r o n s while also monitoring the jaw-opening reflex evoked by tooth pulp shock (JOR) (for review, see ref. 37t. Tooth pulp shock activates a discrete population of afferent fibers, most of which conduct in the A 3 range 8't3 1~.29,41. and produces reflex contraction of the digastric and lateral pterygoid muscles which results in jaw-opening3°'34. W h e n delivered at intensities sufficient to evoke the JOR. tooth pulp shock generally produces only painful sensations in humans14.24.3~.~0.4~. The J O R is suppressed by RM or P A G stimulation 1H7"43'~7'5°. Stimulation of P A G sites that suppress the J O R elicit monosynaptic EPSPs in R M neu-

Correspondence: R. Maciewicz, Pain Physiology Laboratory, Burnham 701. Massachusetts General Hospital. Boston. MA 02114. U.S.A. 0006-8993/86/$03.50© 1986 Elsevier Science Publishers B.V. (Biomedical Division ~

233 rons 3~, consistent with the hypothesis that PAG effects are due, in part, to the direct excitation of caudal raphe cells 7, Suppression of the JOR by PAG stimulation is attenuated by caudal raphe lesions 16, further supporting this hypothesis. In the present study, the intracellular activity of RM neurons and the EMG activation of the digastric muscle were recorded simultaneously during the tooth pulp-evoked JOR. To determine how raphe neurons may participate in control of the tooth pulpevoked JOR, two questions are addressed: (1) what is the activity of RM neurons during the JOR evoked by tooth pulp shock?; and (2) what are the responses of RM neurons during suppression of the JOR by PAG stimulation'? The observations reported below demonstrate that: (1) tooth pulp shock, at an intensity above the digastric EMG threshold, evokes an EPSP in raphe neurons that occurs after the onset of the digastric EMG response; and (2) PAG stimulation, at an intensity below that necessary for suppression of the digastric EMG response, suppresses this tooth pulp-evoked EPSP. MATERIALS AND METHODS Experiments were performed on 30 adult cats (2.0-5.0 kg). The animals were anesthetized with halothane and mounted in a stereotaxic frame. In some animals intraperitoneal doses of a-chloralose (10-30 mg/kg) were given to supplement anesthesia. The level of inhalational halothane was adjusted throughout each experiment to maintain anesthesia without suppressing either respirations or the JOR. Silver wire EMG electrodes were inserted in the anterior digastric muscle on one side to measure latency and intensity of the motor component of the JOR. In order to determine the latency between EMG onset and jaw movement, in two experiments a screw was placed in the mandible; a rigid lead was positioned against the screw so that slight opening movements of the jaw broke contact between the screw and the lead, resulting in a measurable resistance change. Concentric bipolar or silver wire stimulating electrodes were implanted in tooth pulp through small boles drilled into a premolar or canine tooth; the holes were sealed with acrylic cement. Single or double shock stimuli (300/*s, 500 Hz) were presented

through the tooth pulp electrodes at intensities that elicited the JOR without other observable facial movement. Pairs of needle electrodes were also inserted into the footpads of all 4 paws. After positioning the peripheral stimulating electrodes, an anterior and posterior craniotomy was performed. The cerebellar vermis was aspirated, exposing the floor of the fourth ventricle. Concentric bipolar electrodes were positioned in the ventral medullary reticular formation (F -10.0, L 2.0, H -7.0), the ventrobasal thalamus (L 5.5, H 1.0), and PAG (F 0.0 to -2.0, L 1.5, H -0,0 to -2.0). The position of the PAG stimulating electrodes and the timing of the PAG stimuli relative to tooth pulp shock was adjusted to produce depression of the JOR at low intensity. Glass micropipettes were filled with a solution of 4.0% horseradish peroxidase (HRP) (Sigma type V I ) 12 or 1.0% ethidium bromide (EB) 3 and bevelled 31 to a final tip resistance of 20-40 MQ. Under direct visual guidance electrodes were advanced into the pontomedullary raphe at a 30° angle relative to the frontal plane. Recordings were made from neurons in caudal RM and rostral raphe obscurus 4~. During a recording session raphe neurons were first characterized by their response to single shock stimulation of PAG; some cells were further characterized by their responses to electrical stimulation in the medullary reticular formation and in the ventrobasal thalamus. Single shock or short train stimulation of tooth pulp and the 4 paws was then tested. Cells responding to tooth pulp shock were studied during sequential trials in which the JOR was evoked by tooth pulp shock and then suppressed by a conditioning PAG train. Electrophysiological responses were stored on magnetic tape and analyzed off-line using an analogto-digital converter and waveform analysis software (Modular Instruments Inc). The digastric EMG was measured as peak-to-peak amplitude. Following electrophysiological characterization, neurons selected for anatomical study were intracellularly labeled with HRP or EB by injection of constant depolarizing current at 20 nA for approximately 4 min 3.35. Following the recording session, the animals were sacrificed by pentobarbital overdose. Animals with HRP-labeled cells were perfused with saline fol-

234 lowed by 1.0% paraformaldehyde and 1.25% gtutaraldehyde in 0.1 M phosphate buffer. Serial 120/~m frozen sections were cut from the brains and reacted with diaminobenzidine-cobalt chloride to localize the intracellular H R P label 2. Animals with EB labeled cells were perfused with saline followed by 4.0% paraformaldehyde in 0.1 M phosphate buffer. Serial 40/zm sections were cut using a vibratome and examined with a fluorescence microscope (excitation 400 nm (Zeiss BG12 filter)).

stimulus intensity, jaw movement occurred earlier and with less variability; at 80 times threshold, the latency to jaw movement was 17.6 ms (standard error = 0.48). In all trials jaw movement followed the digastric E M G onset by at least 9.0 ms. PAG stimulating electrodes were advanced into the midbrain at symmetrical sites on both sides of the midline to test for suppression of the JOR. At 0.5 mm intervals, a short conditioning train to PAG (3-7 pulses, 300/~s, 500 Hz, 50-500/~A) that preceded the tooth pulp shock was used to test ~or suppression of the digastric E M G response. The final position of the PAG electrodes was adjusted to produce digastric EMG suppression at low P A G stimulation intensities. In most animals stimulation sites in PAG were found that could completely abolish the digastric EMG response to tooth pulp shock with stimulus intensities of 50-150 ~A. The J O R was always suppressed by stimulation through either PAG electrode, although one stimulation position often had a lower threshold for suppression than the other. Under the conditions of stimulation, the maximum J O R suppression occurred at a condition-test interval of 40-90 ms, although partial J O R suppression was of_ ten still evident 200-300 ms following PAG stimuta-

RESULTS

The jaw-opening reflex Single or double shock stimulation of tooth pulp produced a readily reproducible activation of the digastric muscle with an E M G latency of 7-11 ms (Fig. 1). The amplitude of the E M G response was directly related to the stimulus intensity; the E M G latency was inversely related to the stimulus intensity. The threshold for jaw movement was 3-5 times the tooth pulp shock intensity needed to evoke a digastric EMG. At 6 times threshold for the digastric E M G , jaw-opening occurred in every trial with a latency of 19.8 ms (standard error = 1.76). With increasing

,~

300uA I ,,

ImA

t o o t h pulp

D I G A S T R t C EMG

400uA

ir

;

it F~

600uA I

liiI jl~ll~, i/J

tOms

A Fig. 1. Digastric E M G (upper trace) and jaw movement artifact (arrow in lower trace) in response to tooth pulp shock (arrowhead below each trace). The intensity of tooth pulp stimulation is noted above each record. The time calibration for all traces is shown above the lower right hand trace.

235 A

~ 1oo uv [ ~

5 .s

,, ,

CIJ 28

Oo MS

i LI~'L •

,4 ' it,,

CONTROL

•

"",1~"7 ' ~ ' - ' ~ ' ' ' 4

,r~

tOO MS

~

. ~"~'",

"

250 MS

t ooth'pu~p

,'

I ,

30 MS

t 11

,,

.o,s

, ...

r

J L

u

'f 290 MS

~'; "\'

B

B ~ 'oo uJ I-

NRO

80

°

6o

.J el ,< ee O

2o I SO

I loo

I 15o

J 200

I ~so

I 300

CONDITIONING- SHOCK INTERVAL (MS)

Fig. 2. Time course of digastric EMG suppression by prior PAG train stimulation, A: the digastric EMG response evoked by paired tooth pulp shocks (arrowheads below each trace). A conditioning PAG train precedes tooth pulp shock by a latency indicated above each trace. B: digastric EMG amplitude (percentage of control) (JOR amplitude) plotted as a function of the latency between PAG stimulation and tooth pulp shock (the conditioning-shock interval).

tion (Fig. 2). The location of effective P A G stimulation sites used in this study are shown in Fig. 3A,

Location and anatomy of electrophysiologically characterized pontomedullary raphe neurons

Fig. 3. A: locations of PAG stimulation sites (squares) shown on 3 midbrain sections. B: locations of intracellularly labeled raphe neurons (circles) plotted on a midsagittal section of the caudal brainstem. Unlabeled cells have a similar distribution but could not be as accurately localized and are therefore not included in the figure. AQ, aqueduct; NRM, nucleus raphe magnus; NRO, nucleus raphe obscurus: NRP, nucleus raphe pallidus; TB, trapezoid body.

were medium-sized cells with somatic diameters of 25-40/~m. Medium-sized raphe neurons had a fusiform (8/23) or stellate (13/23) s o m a t o d e n d r i t i c morphology. Two medium-sized cells were not adequately filled with EB to be classified.

Responses" of pontomedullary raphe neurons to central stimulation P A G stimulation. In the large m a j o r i t y of studied

Most raphe neurons were r e c o r d e d at a depth of 4 6 mm from the floor of the fourth ventricle, within 400 g m of the midline. Following electrophysiological characterization, 30 raphe neurons were intracellularly stained with H R P or EB. Twenty-seven cells were located in caudal R M and 3 in rostral nucleus raphe obscurus. The distribution of labeled neurons is shown in Fig. 3B.

raphe neurons (130/140) single shock stimulation of P A G on either side e v o k e d a monosynaptic EPSP (0.5-6.0 ms latency, 2.4 ms average latency) as shown in Fig. 5 A - H . The P A G shock intensity necessary to elicit a threshold E P S P was always well below that required to suppress the digastric E M G response to tooth pulp shock.

Cells labeled with H R P were h o m o g e n e o u s l y filled with reaction product that extended into the distal dendrites (Fig. 4A). The fluorescent labeling in EB injected neurons was far more restricted, filling only the cell body and initial processes (Fig. 4B, C). A p plying the criteria of Fox et al. 23 to adult cats, 7 of these ceils were large, multipolar neurons with average somatic diameters of 50-90/~m; the remaining 23

Medulla U reticular formation stimulation. Single shock stimulation of the medullary reticular formation on either side also e v o k e d monosynaptic EPSPs in the large m a j o r i t y of raphe neurons tested (38/50) (Fig. 5 M - P ) . The latency to E P S P following reticular shock was 0 . 5 - 2 . 0 ms (1.2 ms average latency). Although hyperpolarizations frequently followed the EPSPs e v o k e d by P A G or medullary reticular stimu-

23¢1

Fig. 4. Three intracellulary labeled pontomedullary raphe neurons. A: a medium-sized stellate neuron (4-16-1) from nucleus raphe obscurus intrace[lularly labeled with HRP. B: a large multipolar neuron (24-16-1) from RM labeled with EB. PhysiOlogical recordings from this cell are illustrated in Figs. 5 and 7. C: a medium fusiform neuron (24-11-1) from RM labeled with EB. The calibration marker refers to 10(//~m in A and 20 um in B and (7.

A

! ~ :, !~ i,o.. ',

,';

,,

~

'1

,

E

5 ms

, 4.20-

1

!IL I

', =

_

1 M:[

' ", r

t~! li

Pag

,

B

}~i

r, m r f

5 ms

I

F

j

N

%

Fig, 5. EPSPs evoked in raphe neuron 4-20-1 by central shock stimulation. A-C: paired shocks (small arrowheads) to right PAG (shock interval 6 ms) elicit action potentials. Intracellular iniection of increasing hyperpolarizing current (0;5, 1.0, 3,0 nA) blocks the action potentials and enhances the underlying EPSPs. With moderate hyperpolarization, the EPSPs summate and result in an action potential (arrow in B). D: the extracellular field response to the same PAG stimulus. E - H : EPSPs ev0ked by left:PAG Stimulation as described for A - D . I-K: EPSPs evoked by left thalamic stimulation as described for A - C . L: right thalamic stimulation has noeffect on this neuron. M-P: EPSPs evoked by stimulation Of the right medullary reticular formation as described for A>D, The calibration in A applies to A - H ; the time calibration in I applies to I-P.

237 plitude of the tooth-pulp EPSP exhibited marked variability. Tooth pulp EPSPs had two components (Fig. 6G); the first consisted of a small depolarization with a latency of approximately 10-15 ms. The second, larger phase of depolarization began at 20-60 ms. The amplitudes of the first and second components varied independently in individual trials, although both augmented with intracellular injection of hyperpolarizing current. Most neurons that responded to tooth pulp shock also responded with an EPSP to electrical stimulation of each of the 4 paws (Fig. 6E-G). In all cases the tooth pulp-evoked EPSP in raphe neurons occurred after the onset of the digastric E M G response; no raphe neuron responded prior to the digastric EMG onset. When tested, the tooth pulp shock intensity required to evoke an EPSP in a raphe neuron was greater than that needed to elicit a threshold digastric E M G activation. In several studied neurons, the effect of motor activity on neuron response was determined by paralyzing the cat with

lation, single shock stimulation to these sites invariably elicited an initial EPSP. Thalamic stimulation. In 10 of 22 cells tested, single shock stimulation to the ventrobasal complex also elicited an EPSP (Fig. 5 I - L ) . The latency to this EPSP was 2.0-5.0 ms (3.2 ms average latency). When present, both the amplitude and latency of the EPSP evoked by thalamic stimulation were more variable than the EPSPs evoked by PAG or medullary reticular stimulation.

Responses of pontomedullary raphe neurons during the tooth pulp-evoked jaw-opening reflex The intracellular responses of 83 raphe neurons were recorded during multiple trials of the tooth pulp-evoked JOR. Single or double-shock stimulation of tooth pulp, at an intensity above that needed to evoke a threshold digastric EMG response, elicited an EPSP in 24 of 83 raphe neurons (29%) (Fig. 6 A - D ) . No raphe neuron responded with an initial IPSP to tooth pulp shock. Both the latency and am-

A

i~

CIJ 2 4 - 1 6 - 1 !,~ SOurI

lOOuv I

E

d

dlgalltiic EMG

AA

o

, r~-~j..~ 5 m y

2Omv

TOOth Pulp

•

,~: /~,. ,/

~'.--r~,~...--,---,,--~'--"....

l'

20mse¢

4[~ t f -._~ Tooth Pulp ..~.~.' "".,.~--,._o....-..~. ,.,-,......

B

Y

~

L

,II Rt Xindpaw /I

lOmllec

D I,

TOoth Pulp

~,~

,..°., .,%

- ~"~"':"~'~'~" **

•

/

Rt Forepaw

Lt HJndpaw

Fig. 6. Intracellular responses of cell 24-16-1 to peripheral electrical stimulation. The spike amplitudes are truncated by the digitizer. A: tooth pulp shock (small arrowheads below traces in A - C ) evokes a digastric EMG response (upper trace) and an intraccllular EPSP (lower trace). B: the upper two traces show two examples of tooth pulp-evoked EPSPs shown at higher gain. The bottom trace shows the extracellular field potential for the same tooth pulp stimulus. C: average of 4 intracellular (lower trace) and E M G (upper trace) responses to tooth pulp shock. Hollow arrows show early and late components of the tooth pulp-evoked EPSP; both components follow the onset of the digastric E M G response. D: short train stimulation (6 pulses, 500 Hz, represented by a bar beneath the upper trace in D - F ) to the left paw evokes an EPSP (lower trace) and an occasional action potential (large arrowhead in upper traces). E and F: right hindpaw and right forepaw stimulation elicit similar responses. The calibration in A applies to the intracellular records in A and D - F . The calibration in C applies to the intracellular records in B and C.

238 intravenous gallamine triethiodide, Paralysis eliminated the digastric E M G response but had no effect on the tooth pulp-evoked EPSP.

lntracellular responses of raphe neurons during suppression of the tooth pulp-evoked jaw-opening reflex by PA G stimulation PA G stimulation. Single shock stimulation of P A G evoked a monosynaptic E P S P in the large majority of raphe neurons, as stated above, including 22 of 23 neurons (96%) that responded to tooth pulp shock (Fig. 7 E - G ) . With single shocks or short pulse trains, the intensity of P A G stimulation necessary to evoke an EPSP was always less than that required to suppress the digastric E M G response to tooth pulp shock. This relation was true in all raphe neurons (130 tested), including those that did not respond to tooth pulp shock. Electrical train stimulation of P A G evoked a complex, prolonged postsynaptic response in raphe neurons that consisted primarily of depolarization. The train-evoked depolarization augmented as the stimulus intensity was increased to levels which were sufficient to suppress the J O R . FAG suppression of the tooth pulp EPSP. When train stimulation of P A G preceded tooth pulp shock by 40-90 ms. the tooth pulp-evoked EPSP was usually completely suppressed. The intensity of P A G stim-

Tooth i~llp

Tooth

A

ulation necessary to block the tooth pulp E P S P was usually well below that required to suppress the J O R , The responses of a neuron exhibiting this supwession are shown in Fig. 7. in Fig. 7A, B tooth pulp stimulation evoked an EPSP and an occasional action potential. The EPSP followed the o n s e t of the digastric E M G response in each trial. In Fig. 7C, D a conditioning P A G shock train p r e c e d e d the t o o t h pulp stimulus with a condition-test interval of 60 ms. The P A G stimulus evoked a complex depolarization and the subsequent tooth pulp EPSp was suppressed. The tooth pulp-evoked digastric E M G response was only slightly reduced at this intensity o f P A G stimulation. In order to suppress the tooth pulp-evoked EPSP, condition-test intervals of 40-60 ms were used, as this interval was optimal for J O R Suppression; other condition-test intervals were not studied. DISCUSSION The location, appearance and electrophysiological responses of the neurons recorded in the present study are similar to cells previously described using deeply anesthetized, paralyzed cats ~5. Most studied neurons are large- and medium-~zed cells in RM. However. the percentage of raphe neurons responding to tooth pulp shock is smaller than previously re-

pulp

PAG

c I

I

E IOmv

,

.

[ PAG

--~""v'~'~"~--~("~'~

/~" ~ v ~ ' ~ - /

" ~ u ~~ ...... "-

B

T..,h,.,."

'~'~r ~ I

" ';"~ ~ " ' ~ ' ~ ' * " ~

m

~

~

~"

'~" "

PAGTrain ...........

'~~ ~..~.,.,.-,,~ ..~,.~--,-

/,~

iD "-.,_,.._. ~

'L~'

Too~..,, , ',j

~,oo..

......

"~'~"~--"~'~""~"

G

"'°,

t r!;

..

-"

. . . . . . . .

-- -_._.... -_

Fig. 7. Suppression of the tooth pulp-evoked EPSP by prior PAG stimulation in neuron 24-16-1. A and B: tooth pulp shock t small arrowheads) evokes an EPSP (large arrowheads) and an occasional action potential, The digastric EMG response is shownin the upper traces in A-D. C and D: PAG train stimulation (6 pulses. 500 Hz, barbetow tracesl ¢yokes a complex postsynap.tic response .and suppresses the subsequent tooth pulp-evoked EPSP, At this intensity of PAG stimulation, the tooth pulp EPSP is siappressed biat the digastric EMG response is unaffected. E-G: monosynaptic EPSPs to single, double and triple shock of PAG. The calibration in C applies to the intracellular records in A-D. The calibration in G applies tc E-(3.

239 ported; this difference is likely due to the use of a more restricted, lower intensity tooth pulp stimulus in the present experiment. In a study of the tail-flick reflex elicited by noxious heat, Fields et al. 2~ demonstrated a population of raphe cells that increase or decrease their discharge prior to actual tail movement. They proposed that such cells gate the flow of activity through the spinal reflex loop that mediates the tail flick 21. In the present study, tooth pulp shock elicits an initial EPSP in RM neurons that often occurs just prior to jaw movement. However, this EPSP always occurs after the initial digastric EMG: this EMG response precedes actual jaw movement by 9-14 ms. In addition, the onset of the tooth pulp response in all raphe neurons is later than the reported latency of digastric motoneuron activation by tooth pulp shock 28 by at least 3-4 ms. Therefore, RM neurons are activated too late to directly affect the circuitry which mediates the JOR during any single trial of tooth pulp stimulation. The JOR is a flexion withdrawal reflex 34 that consists of sudden jaw opening in response to a noxious stimulus. This reflex movement prevents further injury from potentially tissue-damaging stimulation in the oral cavity. It is therefore not surprising that RM neurons are activated too late to participate in any single trial of the tooth pulp-evoked JOR. Raphe neurons that may be important for reflex inhibition are activated by tooth pulp shock only after the initial withdrawal reflex has occurred; at that time they may play an important antinociceptive role in the modulation of sustained or repeated tooth pulp inputs. The tooth pulp-evoked JOR overrides low-threshold orofacial reflexes and voluntary movements during mastication or a controlled bite 33. As discussed above, initially the JOR provides reflex withdrawal from noxious oral or perioral stimulation. Following this initial withdrawal, more complex movements may be necessary to avoid further injury. Inhibition of the JOR may be required to allow these movements to occur, In this regard, the tooth pulp-evoked JOR may be comparable to the spinal nociceptive flexion withdrawal reflex, where noxious paw shock produces a large amplitude flexion response that can disrupt the normal step cycle22. Reflex limb withdrawal can be an important rapid response to avoid injury. After the initial withdrawal, however, inhibition of limb flexor responses may be required to

facilitate complex escape movements. During suppression of the JOR by PAG stimulation, the PAG train stimulus has two effects on raphe neurons: it elicits a short latency PSP sequence and it blocks the subsequent tooth pulp-evoked EPSP. Vanegas et al. 51 used PAG stimulation to suppress the rat tail-flick response evoked by noxious heat and reported that the threshold for extracellular raphe neuron activation is the same as the threshold for tailflick suppression. In contrast, in the present study monosynaptic EPSPs can be elicited in raphe neurons with single, low intensity PAG shocks although brief train stimulation is usually required to suppress the JOR. Consistent with a previous report 3s, the shock intensity necessary to elicit a threshold EPSP is uniformly below that required to suppress the digastric EMG response. These findings are evidence that summated activity in pontomedullary raphe neurons may be necessary for suppression of the JOR. When PAG train stimulation precedes tooth pulp shock by 40-90 ms in a condition-test paradigm, the tooth pulp-evoked EPSP is suppressed. This suppression may be mediated by the PAG excitation of RM neurons themselves. PAG shock excites raphe neurons, which in turn inhibit trigeminal tooth pulp responsive cells. This inhibition of trigeminal neurons then blocks the tooth pulp input to raphe cells, and may account for the suppression of the tooth pulpevoked EPSP seen in RM neurons. However, other pathways that link PAG and the trigeminal complex may also be important mediators of this EPSP suppression. Tooth pulp shock activates neurons at multiple brainstem levels, including PAG, the ventrobasal thalamus and the medullary reticular formation. Electrical stimulation of these nuclei evokes EPSPs in RM neurons, evidence that these regions contribute to the excitatory drive on raphe cells. The activation of RM neurons leads to JOR suppression. However, once raphe neurons are activated and the JOR is suppressed, the excitatory drive on RM cells is reduced when the tooth pulp EPSP is blocked. This reduced excitatory drive may provide a mechanism that then releases the JOR from inhibition. ACKNOWLEDGEMENTS The authors wish to thank Dr. Joseph B. Martin

240 for his c o n t i n u e d e n c o u r a g e m e n t and s u p p o r t . This

tained in a c c o r d a n c e w i t h t h e guidelines of the C o m -

study was s u p p o r t e d in p a r t by N I H g r a n t NS00634,

m i t t e e on A n i m a l s of t h e H a r v a r d M e d i c a l School

EY05242, D E 0 5 4 1 9 , a n d a g e n e r o u s gift f r o m t h e

and those p r e p a r e d by the C o m m i t t e e on C a r e and

M a u r i c e T. F r e e m a n family; P . M . was s u p p o r t e d by

Use of L a b o r a t o r y A n i m a l s of the Institute of L a b o -

a N S F g r a d u a t e s t u d e n t fellowship, W e t h a n k R.

ratory A n i m a l R e s o u r c e s , N a t i o n a l R e s e a r c h C o u n -

C h u n g for technical assistance t h r o u g h o u t all phases

cil ( D H E W

of this study. A n i m a l s u s e d in this study w e r e m a i n -

1978).

REFERENCES

14 Bratzlavsky, M.. De Boever. J and Van Der Eecken H., Tooth pulpal reflexes in jaw musculature m man. Arch. Oral Biol. 21 (1976)491-493. 15 Chapman. C.D., Ammons. W.S. and Foreman. R.D.. Raphe magnus inhibition of feline q'~-T~ spinoreticular tract cell responses to visceral and somatic inputs. J. Neurophysiol., 53 (1985) 773-785. 16 Chung, R.. Mason. P.. Strassman. A. and Maciew~cz. R.. Pathways mediating suppression of the jaw-opening reflex. I.U.P,S. Orofacial Sensory Function Syrrq~osium. 1986, Abstract. 17 Dostrovsky, J.O.. Hu. J,W.. Sesslc. B.J. and Sumino. R.. Stimulation sites in periaqueductat gray, nucleus raphe magnus and adjacent regions effective in suppressing oralfacial reflexes, Brain Research. 252 (1982) 287-297. 18 Engberg, I., Lundberg, A. and Ryatl. R.W., Reticutosplnal inhibition of transmission in reflex p~tthways. J Physiol. (LondonI. 194 (1968) 201-223. 19 Engstrom, H. and Ohman, A.. Studies on the innervauon of the human teeth. J. Dent. Res., 39~t 960)799-809. 20 Fields. H.L., Basbaum, A.I.. Clanton, C.H. and Anderson. S.D., Nucleus raphe magnus inhibition of spinal cord dorsal horn neurons, Brain Research, 126 ( 1977 ) 441-453. 21 Fields. H.L.. Bry, J.. Hentall, t,D, and Zorman. G., The activity of neurons in the rostral medulla of the rat durin~ withdrawal from noxious heat, 1. Neurosci. 3 11983) 2545-2552. 22 Forssberg, H.. Stumbling correcuve reaction: a phase-dependent compensatory reaction during locomotion, J ~'europhysiol., 42 [19811 936-953. 23 Fox, G.Q., Pappas. G.D. and Purpura. D,P., Morphology and fine structure of the feline neonatal medullary raphe nuclei, Brain Research. 10t [19761 385-410, 24 Fung, D.T., Hwang, J.C, and Chung S.H (orrclativc study of pain perception and masfieatorv muscle reflexes in man. Oral Surg., 45 (1978) 44-50. 25 Gebhart. G.F.. Sandkuhler, J,. Thalhammer, J.(L and Zimmermann. M.. Quantitative comparison of inhibition m the spinal cord of nociceptive information by stimulation in the periaqueductal gray and nucleus raphe magnus of the cat. J. Neurophysiol.. 50 ( 1983/ 1433--1445. 26 Giesler, G.J. and Liebeskind, J.C., Inhibition of visceral pain by electrical stimulation ol the pertaqueductal gra~ matter, Pain, 2 (1976) 43-48. 27 Gray, B.G and Dostrovsky, J.O.. Descending inhibitor~ influences from periaqueductal gray, nucleus raphe magnus, and adjacent reticular formation. I. Effects on lumbar spinal cord nociceptive and nonnoc~ceptwe neurons. ,l Neurophysiol., 49 (1983) 932-947. 28 Greenwood. L.F. and Sessle, ]3..1.. Pain, brain stem mechanLsms, and motor function. In B.J Sessle and A.G Han-

1 Abort, F.V. and Melzack, R., Dissociation in the mechanisms of stimulation-produced analgesia in tests of tonic and phasic pain, Adv. Pain Res. Ther,, 5 (1983) 401-409. 2 Adams, J,C., Technical consideration on the use of horseradish peroxidase as a neuronal marker, Neuroseience, 2 (1977) 141-151. 3 Aghajanian, G.K. and Vandermaelen, C.P., Intracellular identification of central noradrenergic and serotonergic neurons by a new double labeling procedure, J. Neurosci,, 2 (1982) 1786-1792. 4 Andersen, R.K., Lund, J.P. and Puil, E., Excitation and inhibition of neurons in the trigeminal nucleus caudatis following periaqueductal gray stimulation, Can. J. Physiol. Pharmacol., 56 (1978) 157-161, 5 ]3asbaum, A.I., Clanton, C.H. and Fields, H.L., Three bulbospinal pathways from the rostral medulla of the cat: an autoradiographic study of pain modulating systems, J. Comp, Neurol., 178 (1978) 209-224. 6 Basbaum, A.1. and Fields, H.L., The origin of descending pathways in the dorsolateral funiculus of the spinal Cord of the cat and rat: further studies on the anatomy of pain modulation, J. Comp. Neurol., 187 (1979)513-531. 7 Basbaum, A.I. and Fields, H.L., Endogeneous pain control systems: brainstem spinal pathways and endorphin circuitry, Annu. Rev. Neurosci., 7 (1984) 309-338. 8 Beasley, W.L. and Holland, G.R., A quantitative analysis of the innervation of the pulp of the cat's canine tooth, J. Comp. Neurol., 178 (1978) 487-494. 9 Behbehani, M.M. and Fields, H.L., Evidence that an excitatory connection between the periaqueductal gray and nucleus raphe magnus mediates Stimulation produced analgesia, Brain Research, 170 (1979) 85-93. 10 Belcher, G., Ryall, R.W. and Schaffner, R., The differential effects of 5-hydroxytryptamine, noradrenaline and raphe stimulation on nociceptive and non-nociceptive dorsal horn interneurones in the cat. Brain Research, 151 (1978) 307-321. ll Besson, J,-M. and Oliveras, J.-L., Analgesia induced by electrical stimulation of the brainstem in animals: involvement of serotonergic mechanisms, Acta Neurochir., Suppl., 30(1980) 201-217. 12 Bishop, M.A. and King, J,S., Intracellular horseradish peroxidase injections for tracing neural connections. In M:-M. Mesulam led.), Tracing Neural Connections with Horseradish Peroxidase, John Wiley and Sons, New York, 1982, pp. 185-247. 13 Bishop, M.A., A fine-structural survey of the pulpal innerration in the rat mandibular incisor, Am. J. Anat,, 160 (1981) 213-229.

p u b l i c a t i o n No. ( N I H ) 78-23. revised

241

29

30

31

32

33

34

35

36

37

38

39

40

41

nam (Eds.), Mastication and Swallowing: Biological and Clinical Correlates, Univ. of Toronto Press, Toronto, Ont., 1976, pp. 108M17. Johnsen, D.C. and Karlsson, U.L., Electron microscopic quantitations of feline primary and permanent incisor innervation, Arch. Oral Biol., 19 (1974) 671-678. Junge, D., Faermark, W. and Friedman, D., Bilaterality of the jaw-opening reflex in the cat, Arch. Oral Biol., 24 (1979) 21-25. Lederer, W.J., Spindler, A,J. and Eisner, D.A., Thick slurry bevelling: a new technique for bevelling extremely fine microelectrodes and micropipettes, Pfliigers Arch., 381 (1979) 287-288. Lovick, T.A. and Wolstencroft, J.H., Inhibitory effects of nucleus raphe magnus on neuronal responses in the spinal trigeminal nucleus to nociceptive compared with non-nociceptive inputs, Pain, 7 (1979) 135-145. Lund, J.P.. Rossignol, S. and Murakami, T., Interactions between the jaw-opening reflex and mastication, ('an. J. Physiol. Pharmacol., 59 (1981) 683-690. Luschei, E.S. and Goldberg, L.J., Neural mechanisms of mandibular control: mastication and voluntary biting. In J.M. Brookhart and V.B. Mountcastle (Eds.), Handbook of Physiology, Section l, The Nervous System, Vol, 2, Motor Control, Am. Physiol. Soc., Baltimore, MD, 1981, pp. 1237-1274. Maciewicz, R., Sandrew, B.B., Phipps, B.S., Poletti, C.E. and Foote, W.E., Pontomedullary raphe neurons: intracellular responses to central and peripheral electrical stimulation, Brain Research, 293 (1984) 17-33. Mason, P., Strassman, A. and Maciewicz. R., Pontomedullary raphe neurons: monosynaptic excitation from midbrain sites that suppress the jaw opening reflex, Brain Research, 329 (19851 384-389. Mason, P., Strassman, A. and Maciewicz, R., Is the jawopening reflex a valid model of pain?, Brain Res. Rev., 10 (1985) 137-146. Matthews, B,, Baxter, J. and Watts, S., Sensory and reflex responsed to tooth pulp stimulation in man, Brain Research, 113 (1976) 83-94. Mohrland, J.S., McManus, D.Q. and Gebhart, G.F., Lesions in nucleus reticularis gigantocellularis: effect on the antinociception produced by microinjection of morphine and focal electrical stimulation in the periaqueductal gray matter, Brain Research, 231 (1982) 143-152. Mumford. J.M. and Bowsher, D., Pain and protopathic sensibility. A review with particular reference to the teeth, Pain, 2 (1976) 223-243. Nord, S.G., Electrical stimulation of the tooth pulp in the

study of pain, Brain Res. Bull., 1 (19761 251-254. 42 Oliveras, J.L., Besson, J.-M., Guilbaud, G. and Liebeskind, J.C., Behavioral and electrophysiological evidence of pain inhibition from midbrain stimulation in the cat, Exp. Brain Res., 20 (1974) 32-44. 43 Oliveras, J.L., Redjemi, F., Guilbad, G. and Besson, J.-M., Analgesia induced by electrical stimulation of the inferior centralis nucleus of the raphe in the cat, Pain, 1 (1975) 139-145. 44 Prieto, G.J., Cannon, J.T. and Liebeskind, J.C., N. raphe magnus lesions disrupt stimulation-produced analgesia from ventral but not dorsal midbrain areas in the rat, Brain Research, 261 (1983) 53- 57. 45 Rhodes, D.L. and Liebeskind, J.C., Analgesia from rostral brain stern stimulation in the rat, Brain Research, 143 (19781 521-532. 46 Sessle, B.J., Is the tooth pulp a "pure" source of noxious input?, Adv. Pain Res. Ther., 3 (19791 245-260. 47 Sessle, B.J. and Hu, J.W., Raphe-induced suppression of the jaw-opening reflex and single neurons in trigeminal subnucleus oralis, and influence of naloxone and subnucleus caudalis, Pain, 10 (1981) 19-36. 48 Sessle. B.J.. Hu, J.W., Dubner, R. and Lucier, G.E., Functional properties of neurons in cat trigeminal subnucleus caudalis (medullary dorsal horn). II. Modulation of response to noxious and non-noxious stimulation by periaqueductal gray, raphe magnus cerebellar cortex, and afferent influences, and effect of naloxone. J. Neurophysiol., 45 (1981) 193-207. 49 Taber, E., Brodal, A. and Walberg, F., The raphe nuclei of the brainstem in the cat. I. Normal topography and cytoarchitecture, J. Cornp. NeuroL, 114 ( 19601 161 - 187. 5(1 Tanaka, H. and Toda, K., Inhibition of the tooth pulpevoked jaw opening reflex by stimulation of raphe nuclei in the rat, Exp. Neurol., 77 (1982) 102-112. 51 Vanegas, H.. Barbaro, N.M. and Fields, H,L., Midbrain stimulation inhibits tail-flick only at currents sufficient to excite rostra/ medullary neurons, Brain Research. 321 (19841 127-133. 52 Yeung, J.C., Yaksh, T.L. and Rudv, T.A., Concurrent mapping of brain sites for sensitivity to the direct application of morphine and focal electrical stimulation in the production of antinociception in the rat, Pain. 4 (1977) 23-40. 53 Zorman, G., Hentall, I.D., Adams, J.E. and Fields, H.L.. Naloxone-reversible analgesia produced by microstimulation in the rat medulla, Brain Re.search. 219 (19811 137-148.