Convergence of heterotopic nociceptive information onto subnucleus reticularis dorsa - lis neurons...

10
JOURNALOFNEUROPHYSIOLOGY Vol. 63, No. 5, May 1990. Printd in U.S.A. Convergence of Heterotopic Nociceptive Information Onto Neurons of Caudal Medullary Reticular Formation in Monkey (Macaca fascicdaris) LUIS VILLANUEVA, KENNETH D. CLIFFER, LINDA S. SORKIN, DANIEL LE BARS, AND WILLIAM D. WILLIS, JR. Marine Biomedical Institute and Department of Anatomy and Neurosciences, The University of Texas Medical Branch, Galveston, Texas 77550-2772; and Unite de Recherches de Neurophysiologie Pharmacologique de 1 ‘Institut National de la Sante et de la Recherche Medicale, U. 161, 75014 Paris, France SUMMARY AND CONCLUSIONS 1. Recordings were made in anesthetized monkeys from neurons in the medullary reticular formation (MRF) caudal to the obex. Responses of 19 MRF neurons to mechanical, thermal, and/or electrical stimulation were examined. MRF neurons ex- hibited convergence of nociceptive cutaneous inputs from wide- spread areas of the body and face. 2. MRF neurons exhibited low levels of background activity. Background activity increased after periods of intense cutaneous mechanical or thermal stimulation. Nearly all MRF neurons tested failed to respond to heterosensory stimuli (flashes, whistle sounds), and none responded to joint movements. 3. MRF neurons were excited by and encoded the intensity of noxious mechanical stimulation. Responses to stimuli on contra- lateral limbs were greater than those to stimuli on ipsilateral limbs. Responses were greater to stimuli on the forelimbs than to stimuli on the hindlimbs. 4. MRF neurons responded to noxious thermal stimulation (5 1“C) of widespread areas of the body. Mean responses from stimulation at different locations were generally parallel to those for noxious mechanical stimulation. Responses increased with intensity of noxious thermal stimulation (4%50°C). 5. MRF neurons responded with one or two peaks of activa- tion to percutaneous electrical stimulation applied to the limbs, the face, or the tail. The differences in latency of responses to stimulating two locations along the tail suggested that activity was elicited by activation of peripheral fibers with a mean conduction velocity in the A6 range. Stimulation of the contralateral hind- limb elicited greater responses, with lower thresholds and shorter latencies, than did stimulation of the ipsilateral hindlimb. 6. Electrophysiological properties of monkey MRF neurons resembled those of neurons in the medullary subnucleus reticu- laris dorsalis (SRD) in the rat. Neurons in the caudal medullary reticular formation could play a role in processing nociceptive information. Convergence of nociceptive cutaneous input from widespread areas of the body suggests that MRF neurons may contribute to autonomic, affective, attentional, and/or sensory- motor processes related to pain. INTRODUCTION Several lines of evidence indicate that the brain stem reticular formation plays an important role in nociception (see references in Bowsher 1976; Gebhart 1982). Electro- physiological studies have demonstrated that the brain stem reticular formation contains neurons that respond preferentially to noxious stimulation (Benjamin 1970; Blair 1985; Burton 1968; Casey 1969; Gokin et al. 1977; Goldman et al. 1972; Guilbaud et al. 1973; LeBlanc and Gatipon 1974; Mayer and Hill 1978; Nord and Kyler 1968; Nord and Ross 1973; Pearl and Anderson 1978; Rose 1975, 1979; Yokota 1985). Behavioral studies have shown that stimulation of this area can elicit escape behavior (Casey 197 1). Anatomic studies also support the idea of a role for the brain stem reticular formation in nociception, by its connection with areas known to receive nociceptive information (Bowsher 1957, 1962; Breazile and Kitchell 1968; Kerr and Lippman 1974; Nord and Ross 1973; Rossi and Brodal 1957; Zemlan et al. 1978). The ventrolateral quadrant of the spinal cord contains important ascending nociceptive pathways (see references in Vierck et al. 1986); many spinal cord fibers that ascend to the brain in this quadrant terminate within the brain stem reticular forma- tion (Mehler et al. 1960). The particular roles of brain stem reticular neurons in processing nociceptive information have not been deter- mined. Neurons in some brain stem reticular nuclei exhibit heterosensory convergence (Blair 1985; Bowsher et al. 1968; Rose 1975; Scheibel et al. 1955), and many neurons there are inhibited by noxious stimulation (Benjamin 1970; Blair 1985; Burton 1968; Casey 1969; LeBlanc and Gatipon 1974; Mayer and Hill 1978; Pearl and Anderson 1978). Thus reticular nociceptive systems have been con- sidered nonspecific. We have recently reported (Villanueva et al. 1988) that neurons in the subnucleus reticularis dorsalis (SRD) (Val- Verde 1962) of the rat medulla are activated exclusively by volleys in cutaneous A& or A& and C-fibers from any part of the body, and either exclusively or preferentially by noxious thermal or mechanical stimulation of the skin. The majority of SRD neurons are also activated by noxious visceral stimulation. In addition, SRD neurons encode the strength of electrical and noxious mechanical and thermal stimulation (Villanueva et al. 1989). Their A& and C- fiber-evoked activities are depressed by morphine in a dose-dependent and naloxone-reversible fashion (Bing et al. 1989). The spinal pathways that activate SRD neurons ascend predominantly in the ventral lateral funiculi contra- lateral to effective sites of stimulation (Bing et al. 1990). 1118 0022-3077/90 $1 SO Copyright 0 1990 The American Physiological Society

Transcript of Convergence of heterotopic nociceptive information onto subnucleus reticularis dorsa - lis neurons...

JOURNALOFNEUROPHYSIOLOGY Vol. 63, No. 5, May 1990. Printd in U.S.A.

Convergence of Heterotopic Nociceptive Information Onto Neurons of Caudal Medullary Reticular Formation in Monkey (Macaca fascicdaris)

LUIS VILLANUEVA, KENNETH D. CLIFFER, LINDA S. SORKIN, DANIEL LE BARS, AND WILLIAM D. WILLIS, JR. Marine Biomedical Institute and Department of Anatomy and Neurosciences, The University of Texas Medical Branch, Galveston, Texas 77550-2772; and Unite de Recherches de Neurophysiologie Pharmacologique de 1 ‘Institut National de la Sante et de la Recherche Medicale, U. 161, 75014 Paris, France

SUMMARY AND CONCLUSIONS

1. Recordings were made in anesthetized monkeys from neurons in the medullary reticular formation (MRF) caudal to the obex. Responses of 19 MRF neurons to mechanical, thermal, and/or electrical stimulation were examined. MRF neurons ex- hibited convergence of nociceptive cutaneous inputs from wide- spread areas of the body and face.

2. MRF neurons exhibited low levels of background activity. Background activity increased after periods of intense cutaneous mechanical or thermal stimulation. Nearly all MRF neurons tested failed to respond to heterosensory stimuli (flashes, whistle sounds), and none responded to joint movements.

3. MRF neurons were excited by and encoded the intensity of noxious mechanical stimulation. Responses to stimuli on contra- lateral limbs were greater than those to stimuli on ipsilateral limbs. Responses were greater to stimuli on the forelimbs than to stimuli on the hindlimbs.

4. MRF neurons responded to noxious thermal stimulation (5 1 “C) of widespread areas of the body. Mean responses from stimulation at different locations were generally parallel to those for noxious mechanical stimulation. Responses increased with intensity of noxious thermal stimulation (4%50°C).

5. MRF neurons responded with one or two peaks of activa- tion to percutaneous electrical stimulation applied to the limbs, the face, or the tail. The differences in latency of responses to stimulating two locations along the tail suggested that activity was elicited by activation of peripheral fibers with a mean conduction velocity in the A6 range. Stimulation of the contralateral hind- limb elicited greater responses, with lower thresholds and shorter latencies, than did stimulation of the ipsilateral hindlimb.

6. Electrophysiological properties of monkey MRF neurons resembled those of neurons in the medullary subnucleus reticu- laris dorsalis (SRD) in the rat. Neurons in the caudal medullary reticular formation could play a role in processing nociceptive information. Convergence of nociceptive cutaneous input from widespread areas of the body suggests that MRF neurons may contribute to autonomic, affective, attentional, and/or sensory- motor processes related to pain.

INTRODUCTION

Several lines of evidence indicate that the brain stem reticular formation plays an important role in nociception (see references in Bowsher 1976; Gebhart 1982). Electro- physiological studies have demonstrated that the brain stem reticular formation contains neurons that respond

preferentially to noxious stimulation (Benjamin 1970; Blair 1985; Burton 1968; Casey 1969; Gokin et al. 1977; Goldman et al. 1972; Guilbaud et al. 1973; LeBlanc and Gatipon 1974; Mayer and Hill 1978; Nord and Kyler 1968; Nord and Ross 1973; Pearl and Anderson 1978; Rose 1975, 1979; Yokota 1985). Behavioral studies have shown that stimulation of this area can elicit escape behavior (Casey 197 1). Anatomic studies also support the idea of a role for the brain stem reticular formation in nociception, by its connection with areas known to receive nociceptive information (Bowsher 1957, 1962; Breazile and Kitchell 1968; Kerr and Lippman 1974; Nord and Ross 1973; Rossi and Brodal 1957; Zemlan et al. 1978). The ventrolateral quadrant of the spinal cord contains important ascending nociceptive pathways (see references in Vierck et al. 1986); many spinal cord fibers that ascend to the brain in this quadrant terminate within the brain stem reticular forma- tion (Mehler et al. 1960).

The particular roles of brain stem reticular neurons in processing nociceptive information have not been deter- mined. Neurons in some brain stem reticular nuclei exhibit heterosensory convergence (Blair 1985; Bowsher et al. 1968; Rose 1975; Scheibel et al. 1955), and many neurons there are inhibited by noxious stimulation (Benjamin 1970; Blair 1985; Burton 1968; Casey 1969; LeBlanc and Gatipon 1974; Mayer and Hill 1978; Pearl and Anderson 1978). Thus reticular nociceptive systems have been con- sidered nonspecific.

We have recently reported (Villanueva et al. 1988) that neurons in the subnucleus reticularis dorsalis (SRD) (Val- Verde 1962) of the rat medulla are activated exclusively by volleys in cutaneous A& or A& and C-fibers from any part of the body, and either exclusively or preferentially by noxious thermal or mechanical stimulation of the skin. The majority of SRD neurons are also activated by noxious visceral stimulation. In addition, SRD neurons encode the strength of electrical and noxious mechanical and thermal stimulation (Villanueva et al. 1989). Their A& and C- fiber-evoked activities are depressed by morphine in a dose-dependent and naloxone-reversible fashion (Bing et al. 1989). The spinal pathways that activate SRD neurons ascend predominantly in the ventral lateral funiculi contra- lateral to effective sites of stimulation (Bing et al. 1990).

1118 0022-3077/90 $1 SO Copyright 0 1990 The American Physiological Society

NOCICEPTIVE MRF NEURONS 1119

The present study was initiated to extend these findings to the medullary reticular formation (MRF) in monkeys. We have found a population of MRF neurons in monkeys that resemble the previously described SRD neurons in rats. A preliminary report of this work has appeared (Cliffer et al. 1989).

METHODS

Animal preparation

Six monkeys (Macaca fascicularis) of either sex were sedated with ketamine (10 mg/kg‘im), then anesthetized with 2-4s halo- thane in 30% O,-70% N20. In each animal, the left external jugu- lar vein was cannulated, cw-chloralose (60 mg/kg) was injected, and gaseous anesthesia was discontinued. A stable level of anes- thesia, as assessed by pupillary constriction, was maintained by an infusion of pentobarbital sodium (5 mg/kg in 11.5 ml normal saline/h). After a tracheostomy, the animal was artificially venti- lated and paralyzed with gallamine triethiodide (20 mg), which was also added to the infusion (4.6-6.1 mg. kg-’ l h-l) to main- tain paralysis. End-tidal CO2 and rectal temperature were kept within physiological limits (3.5-4.5s and 37 t 1OC).

After dissection of the soft tissue overlying the back of the skull and the cisterna magna, a piece of occipital bone over the cerebel- lum was removed. A small portion of the vermis was aspirated to improve visualization of the obex. The head was fixed in a stereo- taxic frame and ventroflexed. The vertebral column was attached to a rigid frame at C, and L 4, and a bilateral pneumothorax was performed to decrease respiratory movements. The dura was then reflected from the exposed brain stem.

Recordings

Paired stainless steel electrodes, made from 22-gauge hypoder- mic needles, were placed in the ipsilateral hand and the contralat- era1 foot for electrical stimulation. A glass hook was placed on the surface of the brain stem to improve stability. After an opening was made in the pia and arachnoid membranes, a carbon-fiber microelectrode (Anderson and Cushman 198 1; Armstrong-James and Millar 1979) (4.0-7.0 MQ) was inserted O-3.5 mm caudal to the obex and 2.0-3.0 mm lateral to the midline. After the record- ing electrode entered the tissue, warmed agar (2%) was poured over both the electrode and the glass hook. Initially, as a recording electrode was advanced through the cuneate nucleus, electrical stimulation of the hand evoked a large field potential as well as activity in many individual units. When the evoked potential had diminished (at a depth of - 1,800 pm), the search stimulus was changed to percutaneous stimulation of the contralateral foot.

Signals were amplified, displayed, and stored with standard physiological equipment. Activity from a single unit to be studied was isolated with the use of a window discriminator. Pulses were fed to a laboratory computer for construction of peristimulus time histograms during mechanical and thermal stimulation (bin size, 0.5 s) and poststimulus time histograms during percutaneous electrical stimulation (bin size, 0.1 ms). For analysis of responses, background activity was subtracted from activity during a stimu- lus to obtain the response frequency.

Mechanical stimulation

The skin electrodes were removed after the discharge of a cell was isolated. Background activity was recorded, and a series of graded mechanical stimuli was applied to each distal extremity, to each side of the muzzle or lip, and to the tail. The stimuli were brush (repetitive, with a soft-bristled brush), pressure (with a large

arterial clip), pinch (with a small arterial clip), and squeeze (with serrated forceps). When applied to the experimenter’s skin, the first was not painful, the second was near the pain threshold, and the last two were frankly painful. Each was applied for 10 s, with 10 s between stimuli. On the muzzle, it was frequently impossible to apply the small clip (pinch); thus responses to this stimulus were omitted from analyses.

In some instances, corneas were stimulated with a moistened cotton swab, and the effects of light (photographic flash) and sound (referee’s whistle) were also tested.

Thermal stimulation

Responses to thermal stimuli were elicited by dipping each distal extremity or a IO-cm length of the tail into heated (5 1 “C) water for at least 10 s. Responses during the first 10 s of immer- sion were analyzed. The temperature of the water, measured be- fore and after immersion, changed by < 1.5OC. Ability to encode thermal intensity was examined for some cells by immersing the contralateral foot into water at various temperatures from 30 to 52OC.

Electrical stimulation

Pairs of electrodes were inserted into the skin of each distal extremity, the muzzle, and at two sites 100 mm apart on the tail. Threshold for activation was determined for each site. Responses to repeated single square-wave pulses [50- 100 trials, 30 mA (=6-22 times threshold), 2-ms duration, 1 .O pulse/s] were ana- lyzed. Frequently, pulse trains (3 pulses, 2-ms duration, 20 ms apart) were applied to the site giving the most robust response (generally the contralateral foot) to determine if there was a clear response to volleys in C fibers. Conduction velocities of primary afferent fibers contributing to the response were the differences in latencies from stimulating the tail.

talc two

ulated from sites on the

FIG. 1. Summary of recording sites in the medullary reticular forma- tion (MRF). L&: filled symbols, locations marked with lesions; open symbols, sites in the same tracks as lesions; square, recording site of neuron with responses illustrated in Figs. 2 and 7; triangle, recording site of neuron with responses illustrated in Figs. 3, 8, and 10. Righl: large hatched circles, locations plotted solely on the basis of stereotaxic mea- surements. Cu, cuneate nucleus; ECU, external cuneate nucleus; Gr, gra- tile nucleus; IOn, inferior olivary nucleus; LRn, lateral reticular nucleus; V, spinal trigeminal nucleus. Numbers at k:ft indicate approximate dis- tances caudal to obex in millimeters.

1120 VILLANUEVA ET AL.

Histology CONTRA

In each animal, up to three recording sites were marked by small lesions made by passing current through the electrode. Some animals were perfused with saline, followed by mixed alde- hydes (2.5% glutaraldehyde and 0.5% paraformaldehyde) in phosphate buffer. In others the tissue containing the recording sites was removed and placed in 10% Formalin for 36 h. After postfixation in 10% Formalin with 30% sucrose, the tissue was sectioned at 50 pm and stained with neutral red. Sections with marking lesions were projected and traced. Recording sites in the same tracks as lesions were inferred from relative depth measure- ments; others were inferred from stereotaxic coordinates only. A summary of recording locations was plotted on drawings of transverse sections.

Statistics

Data on responses to stimuli, thresholds to electrical stimuli, and latencies of responses to electrical stimuli were compared for adjacent areas of the body or for progressively increasing me- chanical stimuli on a given area of the body with the use of a two-tailed Wilcoxon signed-rank test. The same test was used to determine whether weak responses to innocuous stimuli were sig- nificant. Results with P values ~0.05 were considered significant.

BR PR so -- - BR PR SQ

z: : a : 2 : Y

h : 0 :

BR PR PI SQ

200.

TIME (s)

z: co

2 Y h cn

0;. *.a. * , - + 160

TIME k-i) RESULTS

Locations of recorded units FIG. 3. Responses of 1 MRF neuron to graded mechanical stimulation of the skin on indicated areas of the body. Presentation and symbols as in Fig. 2. Responses from ipsilateral hindlimb and from tail could not be elicited by any stimuli. Recording site is indicated in Fig. 1 (triangle).

Cells belonging to the c uneate nucleu .s were encountered penetrations between 2.0 and 3.0 mm lateral to the in

midline. Deep to these were cells within the MRF that responded to ma1 stimulati

electrical and noxious on of widespread areas

mechanical of the body

ings were made from 19 such neurons. tions, typical trigeminal units respond

In

and ther- . Record-

sonic only

penetra- to facial

IPSI CONTRA

1ng these discharged-during in- vigorously, during noxious

mechanical stimulation. Their general features have been described for several species, including monkeys (Hu et al. 198 1; Price et al. 1976; Yokota 1985). For the two such

stimulation were encountered; nocuous stimulation and, more

neurons that the receptive

were field

tested, activity evoked by stimulation of (RF) on the face was strongly inhibited

by noxious stimulation of a limb, a phenomenon pre- viously described in the rat as diffuse noxious inhibitory controls (DNIC) (Le Bars et al. 1979a; Dickenson et al. 1980).

Figure 1 is a summary of recording sites of nociceptive MRF neurons with large RFs. Locations of recordings from two cells whose responses are illustrated in subse- quent figures are represented by a square (Figs. 2 and 7) and a triangle (Figs. 3, 8, and 10). Although some units recorded for this study were adjacent to the cuneate nu- cleus or the trigeminal nucleus caudalis, most were well within the MRF. Recording sites of cells with responses characteristic of these three structures did not appear to overlap.

< t5 $

1 (Joy +ml-+--“y JlhL .-.1-.

-100 i,. L I

TIME (9 I TIME kd

200:

0 : ki :

FIG. 2. Responses of 1 MRF neuron to graded mechanical stimulation of the skin on indicated areas of the body (dorsal view). Stimuli included brushing with a soft-bristled brush (BR), pressure from a large arterial clip (PR), pinch from a small arterial clip (PI), and squeezing with forceps (SQ). Each stimulus lasted 10 s (bars). Recording site is indicated in Fig. 1 (square).

Background activit v -

Some cells (10 of 13 measured) discharged without in- tentional stimulation of their RFs, but generally at low rates (overall mean t SD: 3.4 -t 5.3 spikes/s). There was a

NOCICEPTIVE MRF NEURONS 1121

30 i

Oh BR PR BR PR

FIG. 4. Individual stimulus-response functions for 10 neurons, from graded cutaneous mechanical stimulation of either the ipsilateral (A) or the contralateral (B) foot. Ordinate, mean firing frequency during stimulation period ( 10 s); abscissa, different intensities of mechanical stimulation (same abbreviations as in Fig. 2).

highly variable but significant increase in background ac- ing from 19% to development of activity > 10 spikes/s from tivity measured 5- 14 min after initiation of trials with no background activity at all. Three cells with no back- noxious mechanical or thermal stimulation (Wilcoxon ground activity did not develop any, and two others signed-rank test); 7 of 12 cells showed such increases, rang- showed small decreases (~7%). After intense percutaneous

A

STIMULUS LOCATION

‘RA

SQUEEZE I

FIG. 5. Mean responses of MRF neurons to graded mechanical cutaneous stimulation. A: data arranged to show differences among stimuli for each location stimulated. B-E: data arranged to show differences for given stimuli between ad- jacent areas; B, brush; C, pressure; I>, pinch; E, squeeze. Dotted lines span bars for which values were tested statistically for differences; filled cir- cles between bars indicate values were signifi- cantly different; error bars, standard deviations; open circles on short bars indicate values that were not significantly different from zero. Num- bers of cells from which means were calculated are at the bases of bars. Abbreviations: I, ipsilat- eral; C, contralateral; M, muzzle; F, forelimb; H, hindlimb; T, tail. Other abbreviations as in Fig. 2.

TAL or - I

MUZZLE FORiiLW HNDLIhB TAIL

1122 VILLANUEVA ET AL.

A

0 20 40 60 80 100

IPSI CONTRA

TIME (SEC)

B 40 1

0 ’ I I PSI CONTRA

FIG. 6. A: responses of 1 MRF neuron to mechanical stimulation of either cornea (repetitive gentle rubbing with a cotton swab). B: mean responses of all neurons tested to stimulation of the corneas. Filled squares with dashed lines. overall means + 1 SD.

electrical stimulation, increases in background activity were also noticed but not measured.

Responses to mechanical stimulation

Figure 2 illustrates responses of an MRF neuron to stim- ulation of widely separated parts of the body. The record- ing site for this neuron is represented by a square in Fig. 1. Although responses were elicited by innocuous stimulation (brush) on the face, noxious stimulation (squeeze) of the same area elicited the greatest response from that location. As was often the case for MRF neurons, responses elicited by noxious stimuli were frequently followed by afterdis- charges.

Several neurons exhibited relatively weak responses to noxious stimulation of one or more areas of the body tested. Responses from a neuron responsive to mechanical stimulation of most, but not all, of the parts of the body tested are illustrated in Fig. 3. The recording location of this neuron is represented by the triangle in Fig. 1. As was the case for other neurons, the greatest responses were ob- tained by stimulation of locations on the side of the body contralateral to the recording. In this case, no responses were elicited by mechanical stimulation of the ipsilateral foot or of the tail.

The preferential activation of MRF neurons by intense mechanical stimulation on the contralateral side of the body is summarized for the hindlimbs in Fig. 4. Each neuron was more responsive to squeezing the skin on the contralateral foot than on the ipsilateral foot. Increases in the responses of these neurons to greater intensities of me- chanical stimulation, mainly within the noxious range, were also apparent.

Figure 5A shows mean responses of MRF neurons to increasing intensities of mechanical stimuli ( 10-s duration) to various parts of the body. In general, MRF neurons responded weakly or not at all to innocuous mechanical stimuli (brushing, pressure, rubbing, or stroking) applied to the distal extremities. Substantial responses to brushing were seen only when the face was stimulated (Fig. 5, A and B). Nine of 10 neurons tested (90%) responded to brushing of either contralateral or ipsilateral face. Of the six respon- sive neurons that were tested bilaterally, five (83%) re- sponded to stimulation of either side. In contrast, only 3 of 13, 4 of 11, and 3 of 9 cells responded to brushing on a forelimb, a hindlimb, and the tail, respectively. Large re- sponses to pressure could be evoked either from the face or the tail (Fig. 5, A and C). Although small, the responses to brushing or pressure on the contralateral hindlimb and to brushing the tail were significantly positive. Responses to brushing or pressure on the forelimbs and ipsilateral hind- limb were not significant.

In contrast, MRF neurons responded preferentially to noxious mechanical (pinch, squeeze) stimulation of all or most parts of the body tested, with greater responses to increasingly intense stimuli (Fig. 54). Responses to nox- ious stimulation of contralateral limbs were greater than

IPSI CONTRA

, l

0 : # _ 3; : 2 -

iii5 ooL 50

TIME (s) TIME (s)

TIME (s)

FIG. 7. Responses of 1 MRF neuron (unit with responses illustrated in Fig. 2) to noxious (5 1 “C) thermal stimulation of indicated areas of the body. Stimulus Period is indicated bv the bar at the bottom of each nlot.

NOCICEPTIVE MRF NEURONS 1123

IPSI CONTRA

TEMP (“Cl

All MRF neurons responded to immersion of distal limbs or the tail in water at 51°C but not to innocuous warming. Mean responses to noxious heat for the various parts of the body are illustrated in Fig. 9B. Although signifi- cant differences were not found between responses to heat- ing of adjacent body areas, relative responses to heating various locations paralleled those to pinch and squeeze (compare Fig. 9B to Fig. 5, D and E).

38.0

40.0

42.0

43.5

L . iI . 11, I

. 45.5 Responses to noncutaneous stimulation

-AL 47.5 None of the MRF neurons tested responded consistently to auditory stimulation or to movement ofjoints. Only one of the seven tested MRF neurons responded consistently to visual stimuli (peak, 4 spikes in a 0. l-s bin).

Responses to percutaneous electrical stimulation o- ' --* 3 : . . .** . . . . 8

0 50

TIME (s) MRF neurons responded with one or two peaks of acti- vation to suprathreshold percutaneous electrical stimula- tion of any of the locations tested. Figure 10 illustrates an example of the responses to electrical stimulation of the various parts of the body (same cell illustrated in Figs. 3 and 8).

TIME (s) 4 +. :.: I&. __ 0 - 50

TIME (8)

FIG. 8. Responses of 1 MRF neuron (unit with responses illustrated in Fig. 3) to noxious thermal stimulation (5 1 “C) of indicated areas of the body and to a series of stimuli of increasing temperatures (right) applied to the contralateral hindlimb. Discharges at the beginning and end of innocu- ous stimuli were due to mechanical stimulation during immersion in and removal from the waterbath.

A 50

1

those to stimulation of ipsilateral limbs, and responses to noxious stimulation of the forelimbs were greater than those to stimulation of the hindlimbs (Fig. 5, D and E).

Repetitive rubbing of the ipsilateral or contralateral cor- nea elicited clear responses from all MRF neurons tested (n = 7). The most striking example is illustrated in Fig. 6A. Note the continuation of each response after the end of the stimulus. Figure 6B illustrates mean responses to stimula- tion of the corneas. For all but one neuron, the response to stimulation of the contralateral cornea was greater than that to stimulation of the ipsilateral cornea.

0

1

----

-101 1 , , , , , , , , , , ,

28 32 36 40 44 48 52

TEMPERATURE (“C) Responses to thermal stimulation

An example of responses to noxious heat is illustrated in Fig. 7 (same cell as illustrated in Fig. 2). The responses exhibited rapid onsets, sometimes with initial phasic com- ponents. Tonic discharges occurred throughout the periods of stimulation. Some responses continued beyond the du- ration of the stimulus.

Figure 8 illustrates responses to noxious heat for another cell (that illustrated in Fig. 3). In this case, noxious heating of either forelimb or of the contralateral hindlimb was ef- fective, whereas stimulation of the ipsilateral hindlimb or the tail was not. Thus, in this case, the responses paralleled those to noxious mechanical stimuli. At the right are illus- trated the effects of graded thermal stimuli applied to the contralateral foot. Innocuous warming had no effect, whereas stimuli of increasingly intense noxious heat elic- ited increasing responses. This encoding of the intensity of the noxious heat stimuli is apparent in the summary in Fig. 9A of responses of cells tested with stimuli of various tem- peratures applied to the contralateral foot.

B HEAT I

TRA

OY I I 1 FORELW

1 HNDLIhB TAIL

SllMULUS LOCATlON

FIG. 9. Summaries of responses of MRF neurons to thermal stimuli applied to various parts of the body. A: responses of 6 neurons to various temperatures applied to the contralateral hindlimb; 2 triangles are not connected because intermediate temperatures were not applied for that case. B: mean responses to noxious (5 1 “C) thermal stimulation of indi- cated areas of the body. Abbreviations, sample sizes, and dotted lines as in Fig. 5.

1124 VILLANUl?\rA ET AL.

Means of the ~unsber of spikes evoked per stimul.us, of the thresholds for activation, and of the latencies of activa- tion are illustrated in Fig. I I, A-C, for the earliest peaks evoked from various parts of the body and face, The rela- tive magnitudes of responses paralleled those for noxious mechanical and t ermal stimuli. Thresholds and lateneies each generally showed an inverse pattern to that of the response magnitudes. Locations from which relatively weak responses were elicited tended to show higher thresh- olds and longer latencies for activating the cells. This is especially evident for the ipsilateral hindlimb. Stimulation there tended to elicit weak responses, with high thresholds and long latencies. Later&s of the early responses ranged from 4 to 12 ms for the ipsilateral cheek, from 5 to 25 ms for the contralatera1 cheek, from X to 16 ms for the fore- limbs, from I5 to 26 ms for the hindlimbs, and from I I to 27 ms for the tail.

Later components of the responses to the electrical stim- uli were more variable in their presence, magnitude, and timing. They were present between 55 and 80% of the time after stimulation of the various locations. Latencies of these later responses were 1 X- 170 ‘ms for the cheeks. 26-65 ms for the forelimbs, 34-76 ms for the hindlimbs, and 28-66 ms forthe tail.

The latency of activation was always shorter from the proximal R from the distal location on the tail. The difference ween the latencies from the two sites ofstim- ulatiDn was 5.X f 2.8 ms (mean ?C SD; n = 12: range, 3.6-I 3.0). Assuming that this diFerence was due solely to the distance of 100 mm between the sets of electrodes, the

IPSI CdNfRA

STtMUUS LOCATDN

FIG, I I. Summary of data from carly (first wak) responses of MRF ntrurnng LO electrical stimulation of various pans of the body and fax. A: mean responses. U: thresholds. C’: latcneies. Rbbrcviations, sample sizes, dotrcd lines, and 61led virc1c-s 2s in Fig. 5.

era1 conduction velocity averaged 17.2 m/s (range, 7.7-27,s) within the A6 range (Burgess and Per1 1973; Gasser and Erlanger 1927). The latencies of the responses to electrical stimuli from all parts of the body were consis- tent with activation of AS-aRerent fibers.

The present study demonstrates that a population of neurons in the caudal ponion of the MR tosensory, particularly nociceptive, in much of the surface of the body. Our results are in general agreement with those from studies of SRD neurons in the rat (Villanueva ct al. 1988. 1989), showing that a group of MRF neurons with heterO%gmentdl nociceptive conver- gence is located ventrolateral to the nucleus cuneatus, with some its at the ventral edge of this nucleus. The latter could long to a population of neurons in the ventral portion of the dorsa column nuclei (DCN) that projecl to the spinal cord. Such neurons have been reported in rats (Rurton and Loewy 1977: M&Cellar amd Loewy 1982; Ber-

NOCICEPTIVE MRF NEURONS 1125

nard et al. 1989) monkeys (Carlton et al. 1985), and cats (Holstege 1988; for a review see Berkley et al. 1986).

In several species, (Dickenson et al. 1979; Hu et al. 198 1; Shigenaga et al. 1976; Yokota 1976, 1985; Yokota and Nishikawa 1980) including monkeys (Hoffman et al. 198 1; Nord and Ross 1973; Price et al. 1976), the medullary reticular formation has been shown to receive nociceptive information related to the trigeminal distribution. These studies may have classified MRF neurons such as those described here as wide dynamic-range neurons, based on their responses to stimulation of the face. Segundo et al. ( 1967) tested cells in the caudal medullary reticular forma- tion of cats for responses to innocuous stimulation of wide- spread areas of the body. Only 1 of 26 had a widespread RF, but intense noxious stimuli such as those used in the present study were not used. Alternatively, the neurons in the present study, with widely convergent nociceptive input, may comprise a separate population, possibly spa- tially segregated, from those previously studied; our search stimulus, once the electrode passed through the cuneate nucleus, was electrical stimulation to the foot, so neurons with RFs confined to the trigeminal distribution would not have been selected. Further study will be necessary to de- termine anatomic and physiological relationships among MRF neurons with widespread RFs, MRF neurons with more restricted RFs, and trigeminal neurons.

The widespread RFs, nociceptive properties, and deep locations of the neurons described in the present study resemble those features of spinothalamic tract (STT) neurons that project to the medial thalamus (Giesler et al. 198 1). Indeed, MRF neurons with descending axons (Carl- ton et al. 1985) may contribute to the excitation of me- dially projecting STT neurons that require brain stem loops for major portions of their RFs (Giesler et al. 198 1). Conversely, STT neurons (not necessarily projecting only to medial thalamus) that send collaterals to the MRF (Giesler et al. 198 1) may contribute to the excitation of MRF neurons such as those reported here. Axons carrying input essential to widespread RFs of SRD neurons in rats ascend in the same area of the white matter as most STT axons (Bing et al. 1990).

Involvement gf’ MRF nel4rons in nociwption

The features of MRF neurons indicate that their func- tion involves processing nociceptive information from the whole body. They were preferentially or exclusively acti- vated by noxious mechanical and thermal stimulation from most or all of the locations stimulated. The cells were also activated by mechanical stimulation of the cornea, which is richly supplied by nociceptive afferents (Hoyes and Barber 1977). Other kinds of stimuli, either heterosen- sory or proprioceptive, were nearly always ineffective. These neurons appeared to be driven by activity originat- ing in A&fibers, as has been observed in other studies of reticular units (Casey 1969; Collins and Randt 1958, 1960; Goldman et al. 1972). Background activity of MRF neurons tended to increase after intense stimulation. Af- terdischarges were sometimes observed after noxious natu- ral stimulation. The neurons encoded the intensity of nox- ious mechanical and thermal stimuli. The ability of AG-no-

ciceptors to encode intensity of noxious mechanical and thermal stimulation (Adriansen et al. 1983; Handwerker et al. 1987) suggests that such fibers were among the A&fibers that contribute input to MRF neurons. Responses to stim- ulation of contralateral limbs were greater than those to stimulation of ipsilateral limbs. Latencies and thresholds of responses to electrical stimulation of the contralateral hindlimb were less than those to stimulation of the ipsilat- era1 hindlimb. These physiological properties, in conjunc- tion with the locations of the cells in the medulla, support the idea that the MRF neurons we recorded in the monkey were comparable to nociceptive SRD neurons in the rat (Villanueva et al. 1988, 1989).

In contrast to the results from SRD neurons in the rat, long-lasting, long-latency evoked responses, indicating C- fiber inputs, were only rarely seen after suprathreshold trains of percutaneous electrical stimulation. This differ- ence could be due to differences between the species or between the specific cell populations studied. However, the MRF could be particularly sensitive to the depressant effect of the anesthetic (a-chloralose plus pentobarbital) admin- istered to the monkeys. This could depress activity evoked by C-fibers. The faster, more synchronized responses to A&fiber activation may be more resistant to anesthetics than activation evoked by C fibers. That DNIC are also less easily demonstrated in monkeys than in rats (compare Gerhart et al. 198 1; Le Bars et al. 1979a,b) may reflect the effect of the anesthetic on reticular nociceptive systems. DNIC has been shown to depend on spinoreticular circuits in the rat (Villanueva et al. 1986) and has been studied under halothane and N20-O3 anesthesia; the strength of DNIC in rats depends on the anesthetic regime (Alarcon and Cervero 1989).

The scarcity of C-fiber-evoked activity did not allow us to distinguish unequivocally between MRF neurons with partial versus total nociceptive convergence, as previously described in rats (Villanueva et al. 1988). Nevertheless, some MRF neurons had whole-body RFs, whereas others could not be activated easily with natural stimuli from some ipsilateral areas of the body or from the tail. This difference resembles that seen between the totally and par- tially convergent cells in rats.

F24nct ion al considwut ions

Our present data, together with the results previously described in the rat, reinforce the idea of an important role of MRF neurons in processing nociceptive information. Their most intriguing feature in this regard is their ability to encode the intensity of noxious stimuli from wide areas of the body. Compared with neurons recorded within other reticular structures (Blair 1985; Bowsher et al. 1968; Rose 1975; Scheibel et al. 1955) MRF neurons exhibited higher levels of functional specificity, had responses that were large and easy to reproduce, and usually did not present heterosensory or proprioceptive convergence. These neurons, therefore, do not fit the original concept of a non- specific reticular nociceptive system.

Because some medullary neurons project to the spinal cord (Bernard et al. 1989; Burton and Loewy 1977; Carlton et al. 1985; Holstege 1988; McKellar and Loewy 1982),

1126 VILLANUEVA ET AL.

MRF neurons could be involved in spinoreticulospinal cir- era1 quadrant of the spinal cord to the subdiencephalic brain stem in

cuits. Such loops are involved in DNIC (Le Bars et al. man. Psychiatr. Neurol. 143: 75-99, 1962.

1979a,b, 1986). Spinoreticulospinal circuits, as discussed BOWSHER, D. Role of the reticular formation in responses to noxious

stimulation. Pain 2: 36 l-378. 1976. above, have also been proposed to play a role in positive feedback loops for the spinal transmission of nociceptive information (Cervero and Wolstencroft 1984; Giesler et al. 1981).

of the body suggests that innocuous facial stimuli may par- ticipate in the function of these cells. Further anatomic and electrophysiological experiments are required to under-

Alternatively or additionally, convergence of nociceptive inputs from wide areas of the body and face onto MRF neurons might reflect involvement of MRF neurons in au-

stand more clearly the hodology and functional potential-

tonomic, affective, attentional, and/or sensory-motor inte- gration related to pain. Responsiveness to low-intensity stimuli on the face as well as to noxious stimuli from most

BOWSHER, D., MALLART, A., PETIT, D., AND ALBE-FESSARD, D. A bulbar relay to the centre median. J. Neurophysiol. 2 1: 288-300, -1968.

BREAZILE, J. E. AND KITCHELL, R. L. Ventrolateral spinal cord afferents to the brain stem in the domestic pig. J. Comp. Neural. 133: 363-372, 1968.

BURGESS, P. R. AND PERL, E. R. Cutaneous mechanoreceptors and noci- ceptors. In: Handbook QfSensory Physiology, edited by A. Iggo. Berlin: Springer-Verlag, 1973, pp. 29-78.

1977. CARLTON, S. M., CHUNG, J. M., LEONARD, R. B., AND WILLIS, W. D.

Funicular trajectories of brainstem neurons projecting to the lumbar

BURTON, H. Somatic sensory properties of caudal bulbar reticular

spinal cord in the monkey (Macaca.fhscicularis): a retrograde labeling

neurons in the cat (Felis domestica). Brain Res. 11: 357-372, 1968.

study. J. Camp. Neurol. 24 1: 382-404, 1985.

BURTON, H. AND LOEWY, A. D. Projections to the spinal cord from medullary somatosensory relay nuclei. J. Comp. Neurol. 173: 773-792,

ity of MRF neurons. Such studies might yield new insights CASEY; K. L. Somatic stimuli, spinal pathways, and size of cutaneous

into the role of the MRF in the processing of nociceptive fibers influencing unit activity in the medial medullary reticular forma-

information. tion. Exp. Neural. 25: 35-56, 1969. CASEY, K. L. Somatosensory responses of bulboreticular units in awake

cat: relation to escape producing stimuli. Science Wash. DC 173: 77-80, The authors are grateful to K. Gondesen for technical assistance, G.

Gonzales for drawings and photography, and M. Cayla and M. Watson for secretarial help.

This work was supported by North Atlantic Treaty Organization Grant 0384/88, by National Institute of Neurological and Communicative Dis- orders and Stroke Grants NS-09743 and NS- 1 1255 and by National Insti- tutes of Health National Research Service Award NS-08 15 1 (to K. D. Cliffer).

Address for reprint requests: W. D. Willis, Jr., Marine Biomedical Insti- tute, The University of Texas Medical Branch, 200 University Blvd., Gal- veston, TX 77550-2772.

Received 18 July 1989; accepted in final form 30 December 1989.

REFERENCES

ADKIANSEN, H., GYBELS, J., HANDWERKER, H. O., AND VAN HEES, J. Response properties of thin myelinated (Ah) fibres in human skin nerves. J. N~~l~ro~lh-v..riol. 49: 1 1 1- 122, 1983.

ALARCON, G. AND CERVERO, F. Effects of two anaesthetic regimes on the heterotopic inhibition of rat dorsal horn neurones (Abstract). J. Physiol. Land. 416: 19P, 1989.

ANDERSON, C. W. AND CUSHMAN, M. R. A simple and rapid method for making carbon fiber microelectrodes. J. Neurosci. Methods 4: 435-436, 1981.

ARMSTRONG-JAMES, M. AND MILLAR, J. Carbon fibre microelectrodes. J.

199-225, 1986.

Neurosci. Methods 1: 279-287, 1979.

BERNARD, J. F., VII,LANUEVA, L., CARROUE, J., AND LE BARS, D. Effer-

BENJAMIN, R. M. Single neurons in the rat medulla responsive to noci- ceptive stimulation. Brain Rex 24: 525-529, 1970.

BERKI-EY, K. J., BUDDELL, R. J., BLOMQVIST, A., AND BULL, M. Output

ent projections from the subnucleus reticularis dorsalis (SRD) of the rat

systems of the dorsal column nuclei in the cat. Brain Res. Rev. 11:

1971. CERVERO, F. AND WOLSTENCROFT, J. H. A positive feedback loop be-

tween spinal cord nociceptive pathways and antinociceptive areas of the cat’s brain stem. Pain 20: 125- 138, 1984.

CLIFFER, K. D., LE BARS, D., SORKIN, L., VILLANUEVA, L., AND WILLIS, W. D. Convergence of heterotopic nociceptive information onto medul- lary reticular formation (MRF) neurones in the anesthetized monkey (Macaca.fa.scicularis) (Abstract). J. Ph-vsiol. Lond. 4 15: 69P, 1989.

COLLINS, W. F. AND RANDT, C. T. Evoked central nervous system activity relating to peripheral unmyelinated or “C” fibers in cat. J. Neurophys- iol. 2 1: 345-352, 1958.

COLLINS, W. F. AND RANDT, C. T. Midbrain evoked responses relating to peripheral unmyelinated or “C” fibers in rat. J. Neurophysiol. 23: 47-53, 1960.

DICKENSON, A. H., HELLON, R. F., AND TAYLOR, D. C. M. Facial ther- mal input to the trigeminal spinal nucleus of rabbits and rats. J. Comp. Neural. 185: 203-2 10, 1979.

DICKENSON, A. H., LE BARS, D., AND BESSON, J. M. Diffuse noxious inhibitory controls (DNIC). Effects on trigeminal nucleus caudalis neu- rones in the rat. Brain Rex 200: 293-305, 1980.

GASSER, H. S. AND ERLANGER, J. The role played by the sizes of the constituent fibers of a nerve trunk in determining the form of its action potential wave. Am. J. Physiol. 80: 522-547, 1927.

GEBHART, G. F. Opiate and opioid peptide effects on brain stem neurons: relevance to nociception and antinociceptive mechanisms. Pain 12: 93-140, 1982.

GERHART, K. D., YEZIERSKI, R. P., GIESLER, G. J., JR., AND WILLIS, W. D. Inhibitory receptive fields of primate spinothalamic tract cells. J. Neurophysiol. 46: 1309- 1325, 198 1.

thalamic nuclei: evidence for a physiologically novel population of spi- nal cord neurons. J. Neurophysiol. 46: 1285- 1308, 198 1.

GOKIN, A. P., KOSTYUK, P. G., AND PREOBRAZHENSKY, N. N. Neuronal mechanisms of interactions of high-threshold visceral and somatic af- ferent influences in spinal cord and medulla. J. Physiol. Paris 73: 319-333, 1977.

GIESLER, G. J., JR., YEZIERSKI, R. P., GERHART, K. D., AND WILLIS, W. D. Spinothalamic tract neurons that project to medial and/or lateral

medulla: studies with the Phaseolus vulgaris leuco-agglutinin (PHA-L) method. Abstr. 12th Meeting Eur. Neurosci. Assoc., 1989, p. 167.

BING, Z., VILLANUEVA, L., AND LE BARS, D. Effects of morphine upon A& and C-fibre evoked activities of subnucleus reticularis dorsalis (SRD) neurones in the rat medulla. Eur. J. Pharmacol. 164: 85-92, 1989.

BING, Z., VUANUEVA, L., AND LE BARS, D. Ascending pathways in the spinal cord involved in the activation of subnucleus reticularis dorsalis

GOLDMAN, P. L., COLLINS, W. F., TAUB, A., AND FITZMARTIN, J. Evoked bulbar reticular unit activity following delta fiber stimulation of periph- eral somatosensory nerve in cat. Exp. Neurol. 37: 597-606, 1972.

GUILBAUD, G., BESSON, J.-M., OLIVERAS, J. L., AND WYON-MAILLARD, M. C. Modifications of the firing rate of bulbar reticular units (nucleus

neurons in the medulla of the rat. J. Neurophysiol. 63: 424-438, 1990. gigantocellularis) after intra-arterial injection of bradykinin into the BLAIR, R. W. Noxious cardiac input onto neurons in medullary reticular limbs. Brain Res. 63: 13 1- 140, 1973.

formation. Brain Res. 326: 335-346, 1985. HANDWERKER, H. O., ANTON, F., AND REEH, P. W. Discharge patterns of BOWSHER, D. Termination of the central pain pathway in man: the con- afferent cutaneous nerve fibers from the rat’s tail during prolonged

scious appreciation of pain. Brain 80: 606-622, 1957. noxious mechanical stimulation. Exp. Brain Res. 65: 493-504, 1987. BOWSHER, D. The topographical projection of fibres from the anterolat- HOFFMAN, D. S., DUBNER, R., HAYES, R. L., AND MEDLIN, T. P. Neuro-

NOCICEPTIVE MRF NEURONS 1127

nal activity in medullary dorsal horn of awake monkeys trained in a thermal discrimination task. 1. Responses to innocuous and noxious thermal stimuli. J. Neurophysiol. 46: 409-427, 198 1.

HOLSTEGE, G. Direct and indirect pathways to lamina I in the medulla oblongata and spinal cord of the cat. In: Progress in Brain Research. Pain Modulation, edited by H. L. Fields and J.-M. Besson. Amsterdam: Elsevier, 1988, vol. 77, p. 47-94.

HOYES, A. D. AND BARBER, P. Ultrastructure of cornea1 receptors. In: Pain in the Trigeminal Region, edited by D. J. Anderson and B. Matthews. Amsterdam: Elsevier, 1977, p. 1-12.

Hu, J. W., DOSTROVSKY, J. O., AND SESSLE, B. J. Functional properties of neurons in cat trigeminal subnucleus caudalis (medullary dorsal horn). I. Responses to oral-facial noxious and nonnoxious stimuli and projec- tions to thalamus and subnucleus oralis. J. Neurophysiol. 45: 173- 192, 1981.

KERR, F. W. L. AND LIPPMAN, H. H. The primate spinothalamic tract as demonstrated by anterolateral cordotomy and commissural myelot- omy. In: Advances in Neurology. International Symposium on Pain, edited by J. J. Bonica. New York: Raven, 1974, vol. 4, p. 147-156.

LE BARS, D., DICKENSON, A. H., AND BESSON, J.-M. Diffuse noxious inhibitory controls (DNIC). I. Effects on dorsal horn convergent neu- rones in the rat. Pain 6: 283-304, 1979a.

LE BARS, D., DICKENSON, A. H., AND BESSON, J.-M. Diffuse noxious inhibitory controls (DNIC). II. Lack of effect on non-convergent neu- rones, supraspinal involvement and theoretical implications. Pain 6: 305-327, 1979b.

LE BARS, D., DICKENSON, A. H., BESSON, J.-M., AND VILLANUEVA, L. Aspects of sensory processing through convergent neurons. In: Spinal Afirent Processing, edited by T. L. Yaksh. New York: Plenum, 1986, p. 467-504.

LEBLANC, H. J. AND GATIPON, G. B. Medial bulboreticular response to peripherally applied noxious stimuli. Exp. Neurol. 42: 264-273, 1974.

MCKELLAR, S. AND LOEWY, A. D. Efferent projections of the Al cate- cholamine cell group in the rat: an autoradiographic study. Brain Res. 241: 11-29, 1982.

MAYER, M. L. AND HILL, R. G. The effects of intravenous fentanyl, morphine and naloxone on nociceptive responses of neurones in the rat caudal medulla. Neuropharmacology. 17: 533-539, 1978.

MEHLER, W. R., FEFERMAN, M. E., AND NAUTA, W. J. H. Ascending axon degeneration following anterolateral cordotomy. An experimental study in the monkey. Brain 83: 7 18-75 1, 1960.

NORD, S. G. AND KYLER, H. J. A single unit analysis of trigeminal projec- tions to bulbar reticular nuclei of the rat. J. Comp. Neurol. 134: 485-494, 1968.

NORD, S. G. AND ROSS, G. S. Responses of trigeminal units in the monkey bulbar lateral reticular formation to noxious and non-noxious stimula- tion of the face: experimental and theoretical considerations. Brain Res. 58: 385-399, 1973.

PEARL, G. S. AND ANDERSON, K. V. Response patterns of cells in the feline caudal nucleus reticular-is gigantocellularis after noxious trigemi- nal and spinal stimulation. Exp. Neurol. 58: 23 l-24 1, 1978.

PRICE, D. D., DUBNER, R., AND Hu, J. W. Trigeminothalamic neurons in

nucleus caudalis responsive to tactile, thermal, and nociceptive stimula- tion of monkey’s face. J. Neurophysiol. 39: 936-953, 1976.

ROSE, J. D. Response properties and anatomical organization of pontine and medullary units responsive to vaginal stimulation in the cat. Brain Res. 97: 79-93, 1975.

ROSE, J. D. Anatomical distribution and sensory properties of brain stem and posterior diencephalic neurons responding to genital, somatosen- sory, and nociceptive stimulation in the squirrel monkey. Exp. Neurol. 66: 169-185, 1979.

ROSSI, G. F. AND BRODAL, A. Terminal distribution of spinoreticular fibers in the cat. Arch. Neural. Psychiatry 78: 439-453, 1957.

SCHEIBEL, M. E., SCHEIBEL, A. B., MOLLICA, A., AND MORUZZI, G. Convergence and interaction of afferent impulses on single units of reticular formation. J. Neurophysiol. 18: 309-33 1, 1955.

SEGUNDO, J. P., TAKENAKA, T., AND ENCABO, H. Somatic sensory prop- erties of bulbar reticular neurons. J. Neurophysiol. 30: 122 1-1238, 1967.

SHIGENAGA, Y., SAKAI, A., AND OKADA, K. Effects of tooth pulp stimula- tion in trigeminal nucleus caudalis and adjacent reticular formation in rat. Brain Res. 103: 400-406, 1976.

VALVERDE, F. Reticular formation of the albino rat’s brain stem cytoar- chitecture and corticofugal connections. J. Comp. Neural. 1 19: 25-49, 1962.

VIERCK, C. L., JR., GREENSPAN, J. D., RITZ, L. A., AND YEOMANS, D. C. The spinal pathways contributing to the ascending conduction and the descending modulation of pain sensations and reactions. In: Spinal Afferent Processing, edited by T. L. Yaksh. New York: Plenum, 1986, p. 275-329.

VILLANUEVA, L., PESCHANSKI, M., CALVINO, B., AND LE BARS, D. As- cending pathways in the spinal cord involved in triggering of diffuse noxious inhibitory controls in the rat. J. Neurophysiol. 55: 34-55, 1986.

VILLANUEVA, L., BING, Z., BOUHASSIRA, D., AND LE BARS, D. Encoding of electrical, thermal and mechanical noxious stimuli by subnucleus reticular-is dorsalis neurons in the rat medulla. J. Neurophwvsiol. 6 1: 391-402, 1989.

VILLANUEVA, L., BOUHASSIRA, D., BING, Z., AND LE BARS, D. Conver- gence of heterotopic nociceptive information onto subnucleus reticu- laris dorsalis (SRD) neurons in the rat medulla. J. Neurophysiol. 60: 980-1009, 1988.

YOKOTA, T. Two types of tooth pulp units in the bulbar lateral reticular formation. Brain Res. 104: 325-329, 1976.

YOKOTA, T. Neural mechanisms of trigeminal pain. In: Advances in Pain Research and Therapy, edited by H. L. Fields, R. Dubner, and F. Cer- vero. New York: Raven, 1985, vol. 9, p. 2 1 l-232.

YOKOTA, T. AND NISHIKAWA, N. Reappraisal of somatotopic tactile rep- resentation within trigeminal subnucleus caudalis. J. Neurophysiol. 43: 700-712, 1980.

ZEMLAN, F. P., LEONARD, C. M., Kow, L. M., AND PFAFF, D. W. Ascend- ing tracts of the lateral columns of the rat spinal cord: a study using the silver impregnation and horseradish peroxidase techniques. Exp. Neurol. 62: 298-334, 1978.