Brain Research, 126 (1977) 441--453
441
© Elsevier/North-HollandBiomedicalPress, Amsterdam- Printed in The Netherlands
NUCLEUS RAPHE MAGNUS INHIBITION OF SPINAL CORD DORSAL HORN NEURONS
HOWARD L. FIELDS, ALLAN I. BASBAUM, CHARLES H. CLANTON and STUART D. ANDERSON Departments of Neurology, Physiology and Anatomy, University of California, San Francisco, Calif. 94143 (U.S.A.)
(Accepted September 9th, 1976)
SUMMARY In decerebrate cats, electrical stimulation of nucleus raphe magnus (NRM) of the medulla produced marked inhibition of spinal neurons in lumbosacral dorsal horn. Only neurons with high threshold inputs were inhibited. These cells were located in lamina I and in or near laminae V and VI. The duration of inhibition produced was related to the stimulus train length. An ipsilateral lesion of the dorsolateral funiculus at L1 markedly reduced the inhibition of neurons caudal to the lesion. Although NRM stimulation was the most effective, inhibition from more lateral sites could be obtained at higher stimulus intensities. NRM induced inhibition is probably mediated by a direct projection via the dorsolateral funiculus to spinal dorsal horn laminae I, II, V and VI. The results are discussed in relation to proposed mechanisms underlying the analgesia produced by NRM stimulation.
INTRODUCTION Although it is well established that dorsal horn neurons of the spinal cord receive powerful inhibitory inputs from the brain stem, neither the precise brain stem origin of this inhibition nor its spinal mechanism have been completely characterized. Medullary sites exert powerful inhibition of reflexes elicited by group II and III peripheral afferent inputs (flexor reflex afferents)14. A pathway descending bilaterally in the dorsolateral funiculus (DLF) mediates this inhibition in the decerebrate cat 14,~1. Taub showed that spinocervical tract neurons could be inhibited by stimulating wide areas of pontomedullary tegmentumaT. Wall 4o and othersa,9, 20, using
442 reversible spinal cord block, further demonstrated that dorsal horn neurons receive a strong tonic inhibition from the brain stem in decerebrate cats. Recent studies have demonstrated that profound analgesia can be produced by electrical stimulation of medial brain stem regions24,'~6,z7,33, some of which are rich in serotonin-containing neurons 13. Stimulation of one of these sites, the dorsal raphe nucleus, inhibits spinal cord neurons with high threshold inputs 2~. The more caudal nucleus raphe magnus (NRM),which has direct connections to spinal cord 1,6,1~, also produces profound analgesia when electrically stimulated 27. There is evidence that palticular dorsal horn neurons are involved in pain transmission. Neurons maximally responsive to inputs from high threshold receptors have been identified in lamina 19,10,41 and deep to lamina IV 29,41. Some neurons in these locations project to regions of medial reticular formation 17 and thalamus 3a,41. Anatomical studies in this laboratory have demonstrated a medial medullary projection primarily to laminae I and II and to lamina V 1. This finding suggested that the analgesia elicited by electrical stimulation of NRM is produced by inhibition of spinal cord pain-transmission neurons. The present studies were undertaken to examine this hypothesis. METHODS Anesthesia was induced with ether in cats weighing 2-4 kg. The trachea was intubated and anesthesia was maintained with a halothane and oxygen mixture, while one carotid artery and jugular vein were cannulated. After mounting the head in a stereotaxic apparatus, an electrolytic decerebration is at the intercollicular level was performed and inhalation anesthesia was discontinued. Cats were paralyzed with gaUamine triethiodide and respirated. Expired carbon dioxide was monitored and maintained between 3.5 and 6 . 0 ~ . Systolic blood pressure was above 90 mm Hg, and rectal temperature was maintained between 35.5 and 38.0 °C. In most experiments, an array of 3 concentric bipolar stimulating electrodes was inserted stereotaxically through a burr hole in the occipital bone and through the cerebellum into the brain stem (Fig. 1). The middle electrode was first centered on the midline by visual comparison with the obex and was targeted on the NRM between and dorsal to the pyramids and just rostral to the inferior olive. The two lateral electrodes were positioned in the same transverse plane, about 1.5 mm lateral to the middle one and slightly more dorsal, i.e., within the medullary reticular formation. Stimulation through these electrodes, repeated at 1 Hz, consisted of 100 Hz trains of 10-12 square pulses of 200 or 300/~sec duration. In testing for inhibition, the intensity was varied from 100 to 500/~A. A laminectomy exposed the spinal cord from L1 to $2. The dura was opened and pinned back, and a unilateral lesion of the dorsal part of the lateral funiculus (DLF) was made at the rostral end of the exposure by pinching with fine forceps. Saline agar was poured over the spinal cord. Recording micropipette electrodes filled with 2 ~ Niagara sky blue dye and 0.5 M K ÷ acetate (impedance 5-10 M ~ at 1000 Hz) were inserted through the agar. The surface of~he cord was located by a DC shift, and the electrode
443
Fig. 1. Experimental scheme and histological controls. In a midcollicular decerebrate cat, 3 bipolar stimulating electrodes are inserted at an angle from behind into rostral medulla. One of these is placed in the midline raphe, the other two are placed laterally in the same transverse plane in the adjacent reticular formation. In the top photograph to the right, the track of one of the lateral electrodes is not seen due to the plane of section. Recording of spinal cord dorsal horn cells is with a dye-filled micropipette electrode. Two dye spots marking cells with physiological properties typical of lamina V cells are encircled in the lowest photograph of the L7 spinal segment. The middle photograph of the L1 segment shows a lesion in the DLF. The lesion is apparent on the right; the wedge of tissue removed at the end of the experiment from the anterolateral quadrant marks the left side.
was a d v a n c e d by a h y d r a u l i c m i c r o d r i v e in 1-10/zm increments t h r o u g h the dorsal h o r n , while m e c h a n i c a l stimuli were a p p l i e d to the h i n d l i m b . U n i t s e n c o u n t e r e d were isol a t e d a n d c h a r a c t e r i z e d b y t h e i r receptive field p r o p e r t i e s . Changes in the activity, either s p o n t a n e o u s o r p e r i p h e r a l l y evoked, o f r e c o r d e d cells in response to s t i m u l a t i o n by the b r a i n stem electrodes were recorded. Because o f the c o n t i n u o u s l y c h a n g i n g n e u r o n a l discharge in response t o a c o n s t a n t c u t a n e o u s stimulus (e.g., t o o t h e d clip), i n h i b i t i o n was usually tested a g a i n s t a v a r y i n g stimulus which was a d j u s t e d to keep the b a c k g r o u n d activity o f the cell a p p r o x i m a t e l y c o n s t a n t d u r i n g c o n t r o l a n d b r a i n
444 stem stimulation periods. The recording loci of one or more cells per penetration were marked by passing cathodal current of 5-10/~A for 7-20 rain through the microelectrode (Fig. 1). In all cases, the first microelectrode penetration was made ipsilateral to the intact DLF. After optimizing the depth of the N R M stimulating electrode for maximal inhibition, the characteristics of the inhibition were studied in neurons both ipsilateral and contralateral to the D L F lesion. The final penetration was made on the side of the intact D L F to insure that inhibition could still be obtained. At termination of the experiment, lesions were placed at the effective brain stem sites. The animal was perfused with heparinized saline, 10 ~ formalin, K ÷ ferro- and ferri-cyanide. The cord segments containing the DLF lesion and dye marks and the brain stem were removed. Serial frozen sections through the brain stem and spinal cord, stained with cresyl violet, provided histological controls of the experimental procedures. RESULTS A total of 151 dorsal horn neurons in 10 cats were studied adequately with respect to receptive field characteristics and the effects of N R M stimulation.
Classification of neurons According to their response to different types of mechanical stimulation, neurons were divided into 3 broad categories. (1) Neurons having inputs which presumably derive from high threshold peripheral mechanoreceptors: cells in this category included those with low threshold inputs, as well as those with relatively high thresholds (i.e., approaching or above noxious levels). Most of the neurons with high mechanical thresholds were situated in or near lamina I. In contrast, dorsal horn neurons with low threshold mechanoreceptor inputs (hairs, light touch) and wide dynamic ranges were, for the most part, located in or near laminae V and VI. (2) Neurons with only low threshold cutaneous mechanoreceptor input: the action potentials of these narrow dynamic range cells had maximal amplitude in or near lamina IV. (3) Neurons responding to joint movement: included in this last category were cells both with and without cutaneous input. This type of neuron was typically located in or near lamina VI. In addition to those neurons described above, a few others with spatially extensive receptive fields or with complex, bilateral inputs were found in the deeper parts of the dorsal horn but were not studied systematically. This distribution within the dorsal horn of functional cell types is consistent with that reported by others, e.g. refs. 28, 40, 41. On a typical penetration, a clear physiological lamination of the dorsal horn was found. Using the criteria of Bishop et al. 5 for identifying cell somata, the first recordings of cell bodies were made in lamina I. In addition to the few neurons with large amplitude spikes ( > 200/,V), numerous smaller amplitude spikes were observed. These may represent the abundant small neurons of lamina I, among which are interspersed the larger Waldeyer (marginal) neurons. Although lamina t neurons were
445 TABLE I Effect o f nucleus raphe magnus stimulation on dorsal horn neurons contra- and ipsilateral to a lesion o f one dorsolateral funiculus
The symbols - - , 0, and + indicate inhibition, no effect, and excitation, respectively. Input
Approximate laminar locus
High threshold cutaneous Only low threshold cutaneous Proprioceptive and cutaneous
I V IV VI
Totals
lntact side --
0
5 36 0 6
0 3 32 4
47
39
Lesioned side +
Totals
--
0
+
0 1 2 7
1 8 0 0
3 20 13 3
0 2 0 5
9 70 47 25
10
9
39
7
151
f r e q u e n t l y e n c o u n t e r e d , less t h a n h a l f were suitable to test for inhibition. Difficulty in testing i n h i b i t i o n was due either to r a p i d a n d p r o l o n g e d h a b i t u a t i o n o f their response to all f o r m s o f s t i m u l a t i o n , o r to a low signal-to-noise ratio. Just ventral to l a m i n a I, no n e u r o n a l activity was recorded. In the v e n t r a l p a r t o f the s u b s t a n t i a g e l a t i n o s a , low a m p l i t u d e m u l t i u n i t responses to light c u t a n e o u s stimuli were o b served. D e e p e r p e n e t r a t i o n revealed t h a t this activity was usually generated by large cells in l a m i n a IV, which r e s p o n d e d over a n a r r o w d y n a m i c r a n g e o f light c u t a n e o u s stimuli. V e n t r a l to l a m i n a IV, cells with a w i d e r d y n a m i c range o f c u t a n e o u s m e c h a n o r e c e p t o r i n p u t a n d l a r g e r receptive fields were recorded. Still d e e p e r n e u r o n s with j o i n t i n p u t s were found. A f t e r a unit with j o i n t i n p u t h a d been studied, the p e n e t r a tion was usually t e r m i n a t e d .
Intact
Lesioned
0
O Fig. 2. Locations of 111 spinal cord dorsal horn recording sites. Sites ipsilateral to a DLF lesion are illustrated on the right; sites on the intact side are seen on the left. Physiological properties of cells are indicated by the shapes of the symbols: 0, represent cells with high threshold cutaneous inputs; [], cells w~th only low threshold cutaneous inputs; ~>, cells with inputs from joints. Filled symbols indicate inhibition of that cell by medullary raphe stimulation. Open symbols indicate no effect. Excitation is indicated by lower half-filled symbols, and a mixed excitatory-inhibitory effect by upper half-filled symbols.
446
Distribution of inhibition Stimulation of NRM produced marked inhibition in certain categories of dorsal horn neurons. Table I summarizes the effect of NRM stimulation on defined types of dorsal horn neurons. Fig. 2 diagrammatically represents the anatomical distribution within the dorsal horn of the 3 categories of neurons and the effect of N R M stimulation on them. Ninety-six dorsal horn neurons ipsilateral to an intact dorsolateral funiculus were studied. Most neurons with high threshold cutaneous inputs were inhibited by N R M stimulation (41/45). These cells were located within or adjacent to laminae ! and V. By contrast, none of the 34 cells with only low threshold inputs, located predominantly in lamina IV, were inhibited by such stimulation. The effects of N R M stimulation on neurons with joint inputs were less consistent than those seen on the more dorsally located neurons, which have only ipsilateral cutaneous inputs. Inhibition, excitation and mixed effects were observed in different cells responding to joint inputs. Since spontaneous activity was minimal, inhibition was tested against a background of activity evoked by peripheral cutaneous stimulation. This was easily main-
P6 ., ~ ' ~ . :~.~. ' ~
.....
P8
Fig. 3. Locations of 10 m~ullary raphe stimulation sites from which minimum threahold inhibition of dorsal horn cells is produced. Sections traced from the atlas of Snider and Nicmer~5.
447 t a i n e d at a c o n s t a n t level for all cells except those o f l a m i n a I. Thus, the 9 l a m i n a I n e u r o n s r e p o r t e d here represent less t h a n h a l f the n u m b e r actually encountered. I n one experiment, b l o o d pressure fell well below the p h y s i o l o g i c a l limit and, a l t h o u g h units with high t h r e s h o l d inputs were found, n o n e were i n h i b i t e d by N R M s t i m u l a t i o n . N o n e u r o n a l d a t a f r o m this e x p e r i m e n t are included in this report.
Brain stem site of inhibition I n each experiment, when the first l a m i n a V-type n e u r o n was e n c o u n t e r e d ipsi-
Fig. 4. Peripherally evoked activity of a lamina V-type cell and its inhibition by stimulation of NRM. In this raster display, each recorded action potential of the unit is represented by a dot, the sweep from left to right is reset every second, the horizontal width of the display representing just less than 1 sec activity. This cell was not spontaneously active but responded over a wide dynamic range of mechanical stimulation of medial toe webbing. The peripheral stimulus was applied continuously during this record. Very strong inhibition is observed following stimulation of NRM by 100 Hz trains of twelve 300/zA square pulses, indicated by the closely spaced dots which appear at the beginning of each horizontal trace.
448 lateral to an intact D L F and found to be inhibited by stimulation from the midline electrode, the m i n i m u m threshold for inhibition was established by adjusting the depth of the electrode in 1 m m increments. The tip positions of the midline electrode at these points of m i n i m u m threshold for inhibition are shown in Fig. 3. In all cases, the m i n i m u m threshold for inhibition was less than 300 #A, averaging about 200/~A. In 7 experiments, a 3-electrode transverse array was used. In 16 of 19 units with high threshold inputs, the midline electrode elicited inhibition at lower stimulus intensities than either lateral electrode. The 3 instances of units in which a lateral electrode was more effective occurred in a single experiment. The nucleus magnocellularis o f the medial reticular formation, in which this and all other lateral electrodes lay, has been shown to inhibit lamina V-type neurons z4. O u r data, however, demonstrate that midline electrodes are generally more effective than laterally placed electrodes in generating inhibition o f dorsal h o r n neurons with high threshold inputs. Since the inhibition was weaker at more ventral sites, it is not due to current spread to the pyramidal tract. The optimal site for inhibition is thus nucleus raphe magnus. The rostrocaudal extent of the inhibitory region has not been systematically explored.
Fig. 5. Inhibition of the evoked activity of cell by NRM stimulus trains of variable length. This lamina V-type cell was not spontaneously active but responded to pinching a fold of skin on the foot, which was maintained throughout this record. The periods indicated by the stimulus artifacts, from top to bottom, consist of approximately 4, 5, 8, 10 and 12 pulses at 100 Hz. Inhibition, which is clearly seen with 12 or 10 pulses, is barely apparent with 8 pulses and cannot be discerned with fewer pulses. Each sweep represents 1 sec. Stimulus intensity is constant at 300/~A.
449
Time course of inhibition The time course o f inhibition was variable f r o m cell to cell and f r o m preparation to preparation. The majority o f cells were studied in the raster display mode. Fig. 4 shows an example o f especially powerful inhibition by stimulation o f N R M o f the peripherally evoked activity o f a cell with high threshold input. Complete inhibition lasting up to 600 msec with a 120 msec train length was observed. Fig. 5 demonstrates the effect o f varying the stimulus train length with constant stimulus intensity. Short trains did not produce obvious inhibition, and increasing the train length resulted in increased duration o f inhibition.
Spinal pathway for inhibition We have demonstrated recently that the N R M projection to the spinal cord is via the D L F 1. F o r this reason, a lesion was usually placed in one D L F prior to recording. N R M - i n d u c e d inhibition o f dorsal horn neurons on the lesioned side o f the cord was c o m p a r e d with that on the intact side. Inhibition o f cells by N R M stimula-
Fig. 6. Inhibition of the peripherally evoked activity of a lamina V-type cell and its abolition by lesion of the ipsilateral DLF. In the left-hand record, made before any cord lesion, the cell was not spontaneously active but responded to stimulation of its receptive field (long vertical bar). This activity was strongly inhibited by stimulation of NRM (short vertical bars) with a 100 Hz train of 10 square pulses of 300 psec duration and 250 pA intensity. In the right-hand record, the same neuron was recorded 10 min after a lesion had been made in the ipsilateral DLF rostral to the recording site. The cell became spontaneously active and responded to lighter cutaneous stimuli. The descending inhibition from NRM stimulation (short vertical bar) is not apparent. After the period of peripheral stimulation, indicated by the long vertical bar, there was strong postexcitatory inhibition. In both records, the horizontal sweep duration is 1 sec.
450 tion caudal and ipsilateral to DLF lesions was far less common than on the intact side (Fig. 2). Further, when inhibition was observed on the side with the lesion, it was of shorter duration. Threshold currents were also higher than on the side with the intact DLF, though this was not systematically studied. Four lamina I and 30 lamina V neurons ipsilateral to a D L F lesion were studied. One of the lamina 1 and 8 of the lamina V neurons were inhibited (Table 1). One cell was observed continuously before and after an ipsilateral D L F lesion was made (Fig. 6). With the cord intact, inhibition was powerful. After DLF section, NRM stimulation with the same stimulus parameters produced no inhibition. In contrast to the striking effects of D L F lesions, a control lesion of the ventrolateral funiculus did not interfere with the inhibition. Furthermore, inhibition was present caudal to an extensive cord lesion which spared only the ipsilateral DLF. In one case where inhibition survived an ipsilateral D L F lesion, a contralateral D L F lesion abolished the residual inhibition. Thus, the lesion data indicate that the D L F is the major, if not the sole, pathway mediating the inhibitory action of N R M on spinal cord dorsal horn neurons which respond to noxious stimulation. The effect is predominantly ipsilateral, but weak contralateral effects are observed. DISCUSSION These findings demonstrate a consistent inhibition of spinal dorsal horn neurons by electrical stimulation of NRM. This inhibition is mediated by a pathway descending in the DLF. Although the inhibition produced with only one D L F intact affects units on both sides of the cord, those on the intact side are more powerfully inhibited. The inhibition is observed in lamina I neurons with small cutaneous receptive fields and high intensity mechanical thresholds, and in neurons, primarily in lamina V, responding to a wide range of cutaneous stimulus intensities. Since neurons with these properties in these locations have been implicated in pain transmission ~8,~8,4°,41, the present results support the hypothesis that N R M is crucial in modulating pain transmission at the spinal cord level. These results are consistent with the anatomical observation that after injection of tritiated leucine into NRM, labeled fibers are found in the D L F 1. The teiminal regions of this projection coincide with the locations of inhibited neurons, namely, laminae I and V. In contrast, the projection from N R M spares most of lamina IV, the cells of which characteristically have only low threshold inputs and are not inhibited by N R M stimulation. Although it is known that stimulation of the sensorimotor cortex 1~ and the pyramidal tract 15 exerts an inhibitory effect on dorsal horn neurons, we feel that current spread to the pyramidal tract is unlikely to account for the observed inhibition in the present exper,ments. First, inhibition at stimulus intensities less than 150 #A could be obtained from sites more than 2 mm distant from the pyramidal tract. According to the data of Ranck 3~, the pyramidal tract is, therefore, beyond the range of current spread. Second, brain stem sites for maximal inhibition were invariably dorsal to the
451 pyramidal tract. Finally, pyramidal tract stimulation was found by Fetz 15 to be more consisteotly inhibitory to lamina IV than to lamina V neurons. Just the opposite was the case in the present experiments. The recent findings of stereospecific opiate binding 36 and of an endogenous compound with morphine-like activitya, 2~ in the mammalian central nervous system lend support to the concept of an endogenous pain-suppression system. The opiate receptor has a relatively high concentration in medial thalamic and periaqueductal gray regions. Both microinjection of morphine and electrical stimulation in these regions produce profound analgesia. This analgesia requires a descending DLF pathway to spinal cord 1,z which is probably serotoninergiclg,aL Consistent with the anatomical studies of Ruda a4, preliminary physiological studies in this laboratory have demonstrated short latency driving of NRM neurons by single electrical stimuli to periaqueductal gray 16. Since NRM is the major source of descending serotoninergic fibers in the dorsolateral funiculus 18,~6 and since NRM lesions interfere with morphine analgesia a°, it is likely that the inhibitory NRM projection to spinal dorsal horn neurons with high intensity inputs mediates both opiate and stimulus-produced analgesia from mesencephalic sites. A recent report that iontophoresed serotonin inhibits lamina I neurons al and that LSD blocks midbrain inhibition of dorsal horn neurons with high intensity inputs 19 is consistent with this hypothesis. Engberg el al. 14 found, in decerebrate cats, that spinal interneurons with inputs from group II and III afferents are under tonic inhibitory control mediated, in part, by serotoninergic neurons in the ventromedial medulla. Electrolytic lesions of the raphe nuclei eliminated most, but not all, of the tonic inhibition. The fact that some inhibition persists despite complete raphe lesions implies that adjacent magnocellular reticular formation participates in this inhibitory control. These data are consistent with our finding that midline, i.e. raphe, electrode positions are the most effective for inhibition of spinal nociceptive neuions. However, some contribution to the inhibition by neurons outside of NRM in the adjacent reticular formation cannot be excluded. The present results provide one substrate for the repeatedly demonstrated brain stem control of spinal dorsal horn neurons. There are direct connections to NRM from sensorimotor cortex, spinal cord, cerebeUum7, periaqueductal gray 34 and adjacent reticular formation ~1. The widespread convergence onto this system, which modulates high threshold spinal dorsal horn neurons, indicates that multiple factors influence the flow of nociceptive input along central somatosensoly pathways. ACKNOWLEDGEMENTS We thank Ms. Mary Liu for histological assistance and Ms. Sherry Davis for editorial work on this manuscript. This research was supported by P.H.S. Grants NS 70777 and NS 11529 to H. L. Fields, MH 7082 to S. D. Anderson, and NS 05272 to A. I. Basbaum.
452 REFERENCES 1 Basbaum, A. I., Ctanton, C. H. and Fields, H. L., Contrasting descending projections of the nucleus gigantocellularis (NGC) and nucleus raphe magnus (NRM) of the cat, Paper presented at Cajal Club, Louisville, Ky., 1976. 2 Basbaum, A. I., Marley, N. and O'Keefe, J., Effects of spinal cord lesions on the analgesic properties of electrical brain stimulation, Advanc. Neurol., in press. 3 Belluzzi, J. D., Grant, N., Garsky, V., Sarantakis, D., Wise, C. D. and Stein, L., Analgesia induced in vivo by central administration of enkephalin in rat, Nature (Lond.), 260 (1976) 625-626. 4 Besson, J. M., Guilbaud, G. and LeBars, D., Descending inhibitory influences exerted by the brain stem upon the activities of dorsal horn lamina V cells induced by intra-arteriat injection of bradykinin into the limbs, J. Physiol. (Lond.), 248 (1975) 725-739. 5 Bishop, P. O., Burke, W. and Davis, R., The identification of single units of central visual pathways, J. Physiol. (Lond.), 162 (1962) 409-431. 6 Brodal, A., Taber, E. and Walberg, F., The raphe nuclei of the brain stem in the cat. 11. Efferent connections, J. comp. Neurol., 144 (1960) 239-260. 7 Brodal, A., Walberg, F. and Taber, E., The raphe nuclei of the brain stem in the cat. III. Afferent connections, J. comp. NeuroL, 144 (1960) 261-282. 8 Brown, A. G., Kirk, E. J. and Martin, H. F., III, Descending and segmental inhibition of transmission through the spinocervical tract, J. Physiol. (Lond.), 230 (1973) 689-705. 9 Cervero, F., lggo, A. and Ogawa, H., Nociceptor-driven dorsal horn neurones in the lumbar spinal cord of the cat, Pain, 2 (1976) 5-24. 10 Christensen, B. N. and Perl, E. R., Spinal neurons specifically excited by noxious or thermal stimuli: marginal zone of the dorsal horn, J. Neurophysiol., 33 (1970) 293-307. 11 Couch, J. R., Jr., Responses of neurons in the raphe nuclei to serotonin, norepinephrine and acetylcholine and their correlat;on with an excitatory synaptic input, Brain Research, 19 (1970) 137-150. 12 Coulter, J. D., Maunz, R. A. and Willis, W. D., Effects of stimulation of sensorimotor cortex on primate spinothalamic neurons, Brain Research, 65 (1974) 351-356. 13 Dahlstr6m, A. and Fuxe, K., Evidence for the existence of monoamine neurons in the central nervous system. II. Experimentally induced changes in the intraneuronal amine levels of bulbospinal neuron systems, Acta physiol, scand., 64, Suppl. 247 (1965) 5-36. 14 Engberg, 1., Lundberg, A. and Ryall, R. W., Reticulospinal inhibition of transmission in reflex pathways, J. Physiol. (Lond.), 194 (1968) 201-223. 15 Fetz, E. E., Pyramidal tract effects on interneurones in the cat lumbar dorsal horn, J. Neurophysiol., 31 (1968) 69-80. 16 Fields, H. L., Anderson, S. D., Clanton, C . H . and Basbaum, A. I., Nucleus raphe magnus: a common mediator of opiate and stimulus produced analgesia, Trans. Amer. neurol. Ass., in press. 17 Fields, H. L., Clanton, C. H. and Anderson, S. D., Somatosensory properties of spinoreticular neurons, Brain Research, 120 (1977) 49-66. 18 Fields, H. L., Wagner, G. M. and Anderson, S. D., Some properties of spinal neurons projecting to the medial brain-stem reticular formation, Exp. NeuroL, 47 (1975) 118-134. 19 Guilbaud, G., Besson, J. M., Oliveras, J. L. and Liebeskind, J. C., Suppression by LSD of the inhibitory effect exerted by dorsal raphe stimulation on certain spinal cord interneurons in the cat, Brain Research, 61 (1973) 417-422. 20 Handwerker, H. O., Iggo, A., and Zimmermann, M., Segmental and supraspinal actions on dorsal horn neurons responding to noxious and non-noxious skin stimuli, Pain, 1 (1975) 147-166. 21 Holmqvist, B. and Lundberg, A., On the organization of the supraspinat inhibitory control of interneurones of various spinal reflexes, Arch. ital. Biol., 97 (1959) 340-356. 22 Hughes, J., Isolation of an endogenous compound from the brain with pharmacological properties similar to morphine, Brain Research, 88 (1975) 295-308. 23 Mayer, D. J., Price, D. D. and Becker, D. P., Neurophysiological characterization of the anterolateral spinal cord neurons contributing to pain perception in man, Pain, 1 (1975) 51-58. 24 Mayer, D. J. and Liebeskind, J. C., Pain reduction by focal electrical stimulation of the brain: an anatomical and behavioral analysis, Brain Research, 68 (1974) 73-93. 25 McCreery, D.B. and Bloedel, J. R., Reduction of the response of cat spinothalamic neurons to graded mechanical stimuli by electrical stimulation of the lower brainstem, Brain Research, 97 (1975) 151-156.
453 26 Oliveras, J. L., Besson, J. M., Guilbaud, G. and Liebeskind, J. C., Behavioral and electrophysiological evidence of pain inhibitionfrom midbrain stimulation in the cat, Exp. Brain Res., 20 (1974) 32-44. 27 Oliveras, J. L., Redjemi, F., Guilbaud, 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. 28 Price, D. D. and Mayer, D. J., Neurophysiological characterization of the anterolateral quadrant neurons subserving pain in M. mulatta, Pain, 1 (1975) 59-72. 29 Price, D. D. and Browe, A. C., Spinal cord coding of graded non-noxious and noxious temperature increases, Exp. Neurol., 48 (1975) 201-221. 30 Proudfit, H. K. and Anderson, E. G., Morphine analgesia: blockade by raphe magnus lesions, Brain Research, 98 (1975) 612-618. 31 Randic, M. and Yu, H. H., Micro-electrophoretic study of cat dorsal horn neurones activated by noxious stimuli, J. Physiol. (Lond.), 252 (1975) 23-24. 32 Ranck, J.B., Jr., Which elements are excited in electrical stimulation of mammalian central nervous system: a review, Brain Research, 98 (1975) 417--440. 33 Reynolds, D. V., Surgery in the rat during electrical analgesia induced by focal brain stimulation, Science, 164 (1969) 444--445. 34 Ruda, M. A., Autoradiographic Examination of the Efferent Projections of the Midbrain Central Grey in the Cat, Ph.D. Thesis, University of Pennsylvania, 1976. 35 Snider, R. S. and Niemer, W. T., ,4 Stereotaxic Atlas of the Cat Brain, University of Chicago Press, Chicago, II1., 1961. 36 Snyder, S. H., Opiate receptor in normal and drug altered brain function, Nature (Lond.), 257 (1975) 185-189. 37 Taub, A., Local, segmental and supraspinal interaction with a dorsolateral spinal cutaneous afferent system, Exp. NeuroL, 10 (1964) 357-374. 38 Trevino, D. L. and Carstens, E., Confirmation of the location of spinothalamic neurons in the cat and monkey by the retrograde transport of horseradish peroxidase, Brain Research, 98 (1975) 177-182. 39 Vogt, M., The effect of lowering the 5-hydroxytryptamine content of the rat spinal cord on analgesia produced by morphine, J. PhysioL (Lond.), 236 (1974) 483-498. 40 Wall, P. D., The laminar organization of dorsal horn and effects of descending influences, J. Physiol. (Lond.), 188 (1967) 403-424. 41 Willis, W. D., Trevino, D. L., Coulter, J. D. and Maunz, R. A., Responses of primate spinothalamic tract neurons to natural stimulation of hindlimb, J. NeurophysioL, 37 (1974) 358-372.
441
© Elsevier/North-HollandBiomedicalPress, Amsterdam- Printed in The Netherlands
NUCLEUS RAPHE MAGNUS INHIBITION OF SPINAL CORD DORSAL HORN NEURONS
HOWARD L. FIELDS, ALLAN I. BASBAUM, CHARLES H. CLANTON and STUART D. ANDERSON Departments of Neurology, Physiology and Anatomy, University of California, San Francisco, Calif. 94143 (U.S.A.)
(Accepted September 9th, 1976)
SUMMARY In decerebrate cats, electrical stimulation of nucleus raphe magnus (NRM) of the medulla produced marked inhibition of spinal neurons in lumbosacral dorsal horn. Only neurons with high threshold inputs were inhibited. These cells were located in lamina I and in or near laminae V and VI. The duration of inhibition produced was related to the stimulus train length. An ipsilateral lesion of the dorsolateral funiculus at L1 markedly reduced the inhibition of neurons caudal to the lesion. Although NRM stimulation was the most effective, inhibition from more lateral sites could be obtained at higher stimulus intensities. NRM induced inhibition is probably mediated by a direct projection via the dorsolateral funiculus to spinal dorsal horn laminae I, II, V and VI. The results are discussed in relation to proposed mechanisms underlying the analgesia produced by NRM stimulation.
INTRODUCTION Although it is well established that dorsal horn neurons of the spinal cord receive powerful inhibitory inputs from the brain stem, neither the precise brain stem origin of this inhibition nor its spinal mechanism have been completely characterized. Medullary sites exert powerful inhibition of reflexes elicited by group II and III peripheral afferent inputs (flexor reflex afferents)14. A pathway descending bilaterally in the dorsolateral funiculus (DLF) mediates this inhibition in the decerebrate cat 14,~1. Taub showed that spinocervical tract neurons could be inhibited by stimulating wide areas of pontomedullary tegmentumaT. Wall 4o and othersa,9, 20, using
442 reversible spinal cord block, further demonstrated that dorsal horn neurons receive a strong tonic inhibition from the brain stem in decerebrate cats. Recent studies have demonstrated that profound analgesia can be produced by electrical stimulation of medial brain stem regions24,'~6,z7,33, some of which are rich in serotonin-containing neurons 13. Stimulation of one of these sites, the dorsal raphe nucleus, inhibits spinal cord neurons with high threshold inputs 2~. The more caudal nucleus raphe magnus (NRM),which has direct connections to spinal cord 1,6,1~, also produces profound analgesia when electrically stimulated 27. There is evidence that palticular dorsal horn neurons are involved in pain transmission. Neurons maximally responsive to inputs from high threshold receptors have been identified in lamina 19,10,41 and deep to lamina IV 29,41. Some neurons in these locations project to regions of medial reticular formation 17 and thalamus 3a,41. Anatomical studies in this laboratory have demonstrated a medial medullary projection primarily to laminae I and II and to lamina V 1. This finding suggested that the analgesia elicited by electrical stimulation of NRM is produced by inhibition of spinal cord pain-transmission neurons. The present studies were undertaken to examine this hypothesis. METHODS Anesthesia was induced with ether in cats weighing 2-4 kg. The trachea was intubated and anesthesia was maintained with a halothane and oxygen mixture, while one carotid artery and jugular vein were cannulated. After mounting the head in a stereotaxic apparatus, an electrolytic decerebration is at the intercollicular level was performed and inhalation anesthesia was discontinued. Cats were paralyzed with gaUamine triethiodide and respirated. Expired carbon dioxide was monitored and maintained between 3.5 and 6 . 0 ~ . Systolic blood pressure was above 90 mm Hg, and rectal temperature was maintained between 35.5 and 38.0 °C. In most experiments, an array of 3 concentric bipolar stimulating electrodes was inserted stereotaxically through a burr hole in the occipital bone and through the cerebellum into the brain stem (Fig. 1). The middle electrode was first centered on the midline by visual comparison with the obex and was targeted on the NRM between and dorsal to the pyramids and just rostral to the inferior olive. The two lateral electrodes were positioned in the same transverse plane, about 1.5 mm lateral to the middle one and slightly more dorsal, i.e., within the medullary reticular formation. Stimulation through these electrodes, repeated at 1 Hz, consisted of 100 Hz trains of 10-12 square pulses of 200 or 300/~sec duration. In testing for inhibition, the intensity was varied from 100 to 500/~A. A laminectomy exposed the spinal cord from L1 to $2. The dura was opened and pinned back, and a unilateral lesion of the dorsal part of the lateral funiculus (DLF) was made at the rostral end of the exposure by pinching with fine forceps. Saline agar was poured over the spinal cord. Recording micropipette electrodes filled with 2 ~ Niagara sky blue dye and 0.5 M K ÷ acetate (impedance 5-10 M ~ at 1000 Hz) were inserted through the agar. The surface of~he cord was located by a DC shift, and the electrode
443
Fig. 1. Experimental scheme and histological controls. In a midcollicular decerebrate cat, 3 bipolar stimulating electrodes are inserted at an angle from behind into rostral medulla. One of these is placed in the midline raphe, the other two are placed laterally in the same transverse plane in the adjacent reticular formation. In the top photograph to the right, the track of one of the lateral electrodes is not seen due to the plane of section. Recording of spinal cord dorsal horn cells is with a dye-filled micropipette electrode. Two dye spots marking cells with physiological properties typical of lamina V cells are encircled in the lowest photograph of the L7 spinal segment. The middle photograph of the L1 segment shows a lesion in the DLF. The lesion is apparent on the right; the wedge of tissue removed at the end of the experiment from the anterolateral quadrant marks the left side.
was a d v a n c e d by a h y d r a u l i c m i c r o d r i v e in 1-10/zm increments t h r o u g h the dorsal h o r n , while m e c h a n i c a l stimuli were a p p l i e d to the h i n d l i m b . U n i t s e n c o u n t e r e d were isol a t e d a n d c h a r a c t e r i z e d b y t h e i r receptive field p r o p e r t i e s . Changes in the activity, either s p o n t a n e o u s o r p e r i p h e r a l l y evoked, o f r e c o r d e d cells in response to s t i m u l a t i o n by the b r a i n stem electrodes were recorded. Because o f the c o n t i n u o u s l y c h a n g i n g n e u r o n a l discharge in response t o a c o n s t a n t c u t a n e o u s stimulus (e.g., t o o t h e d clip), i n h i b i t i o n was usually tested a g a i n s t a v a r y i n g stimulus which was a d j u s t e d to keep the b a c k g r o u n d activity o f the cell a p p r o x i m a t e l y c o n s t a n t d u r i n g c o n t r o l a n d b r a i n
444 stem stimulation periods. The recording loci of one or more cells per penetration were marked by passing cathodal current of 5-10/~A for 7-20 rain through the microelectrode (Fig. 1). In all cases, the first microelectrode penetration was made ipsilateral to the intact DLF. After optimizing the depth of the N R M stimulating electrode for maximal inhibition, the characteristics of the inhibition were studied in neurons both ipsilateral and contralateral to the D L F lesion. The final penetration was made on the side of the intact D L F to insure that inhibition could still be obtained. At termination of the experiment, lesions were placed at the effective brain stem sites. The animal was perfused with heparinized saline, 10 ~ formalin, K ÷ ferro- and ferri-cyanide. The cord segments containing the DLF lesion and dye marks and the brain stem were removed. Serial frozen sections through the brain stem and spinal cord, stained with cresyl violet, provided histological controls of the experimental procedures. RESULTS A total of 151 dorsal horn neurons in 10 cats were studied adequately with respect to receptive field characteristics and the effects of N R M stimulation.
Classification of neurons According to their response to different types of mechanical stimulation, neurons were divided into 3 broad categories. (1) Neurons having inputs which presumably derive from high threshold peripheral mechanoreceptors: cells in this category included those with low threshold inputs, as well as those with relatively high thresholds (i.e., approaching or above noxious levels). Most of the neurons with high mechanical thresholds were situated in or near lamina I. In contrast, dorsal horn neurons with low threshold mechanoreceptor inputs (hairs, light touch) and wide dynamic ranges were, for the most part, located in or near laminae V and VI. (2) Neurons with only low threshold cutaneous mechanoreceptor input: the action potentials of these narrow dynamic range cells had maximal amplitude in or near lamina IV. (3) Neurons responding to joint movement: included in this last category were cells both with and without cutaneous input. This type of neuron was typically located in or near lamina VI. In addition to those neurons described above, a few others with spatially extensive receptive fields or with complex, bilateral inputs were found in the deeper parts of the dorsal horn but were not studied systematically. This distribution within the dorsal horn of functional cell types is consistent with that reported by others, e.g. refs. 28, 40, 41. On a typical penetration, a clear physiological lamination of the dorsal horn was found. Using the criteria of Bishop et al. 5 for identifying cell somata, the first recordings of cell bodies were made in lamina I. In addition to the few neurons with large amplitude spikes ( > 200/,V), numerous smaller amplitude spikes were observed. These may represent the abundant small neurons of lamina I, among which are interspersed the larger Waldeyer (marginal) neurons. Although lamina t neurons were
445 TABLE I Effect o f nucleus raphe magnus stimulation on dorsal horn neurons contra- and ipsilateral to a lesion o f one dorsolateral funiculus
The symbols - - , 0, and + indicate inhibition, no effect, and excitation, respectively. Input
Approximate laminar locus
High threshold cutaneous Only low threshold cutaneous Proprioceptive and cutaneous
I V IV VI
Totals
lntact side --
0
5 36 0 6
0 3 32 4
47
39
Lesioned side +
Totals
--
0
+
0 1 2 7
1 8 0 0
3 20 13 3
0 2 0 5
9 70 47 25
10
9
39
7
151
f r e q u e n t l y e n c o u n t e r e d , less t h a n h a l f were suitable to test for inhibition. Difficulty in testing i n h i b i t i o n was due either to r a p i d a n d p r o l o n g e d h a b i t u a t i o n o f their response to all f o r m s o f s t i m u l a t i o n , o r to a low signal-to-noise ratio. Just ventral to l a m i n a I, no n e u r o n a l activity was recorded. In the v e n t r a l p a r t o f the s u b s t a n t i a g e l a t i n o s a , low a m p l i t u d e m u l t i u n i t responses to light c u t a n e o u s stimuli were o b served. D e e p e r p e n e t r a t i o n revealed t h a t this activity was usually generated by large cells in l a m i n a IV, which r e s p o n d e d over a n a r r o w d y n a m i c r a n g e o f light c u t a n e o u s stimuli. V e n t r a l to l a m i n a IV, cells with a w i d e r d y n a m i c range o f c u t a n e o u s m e c h a n o r e c e p t o r i n p u t a n d l a r g e r receptive fields were recorded. Still d e e p e r n e u r o n s with j o i n t i n p u t s were found. A f t e r a unit with j o i n t i n p u t h a d been studied, the p e n e t r a tion was usually t e r m i n a t e d .
Intact
Lesioned
0
O Fig. 2. Locations of 111 spinal cord dorsal horn recording sites. Sites ipsilateral to a DLF lesion are illustrated on the right; sites on the intact side are seen on the left. Physiological properties of cells are indicated by the shapes of the symbols: 0, represent cells with high threshold cutaneous inputs; [], cells w~th only low threshold cutaneous inputs; ~>, cells with inputs from joints. Filled symbols indicate inhibition of that cell by medullary raphe stimulation. Open symbols indicate no effect. Excitation is indicated by lower half-filled symbols, and a mixed excitatory-inhibitory effect by upper half-filled symbols.
446
Distribution of inhibition Stimulation of NRM produced marked inhibition in certain categories of dorsal horn neurons. Table I summarizes the effect of NRM stimulation on defined types of dorsal horn neurons. Fig. 2 diagrammatically represents the anatomical distribution within the dorsal horn of the 3 categories of neurons and the effect of N R M stimulation on them. Ninety-six dorsal horn neurons ipsilateral to an intact dorsolateral funiculus were studied. Most neurons with high threshold cutaneous inputs were inhibited by N R M stimulation (41/45). These cells were located within or adjacent to laminae ! and V. By contrast, none of the 34 cells with only low threshold inputs, located predominantly in lamina IV, were inhibited by such stimulation. The effects of N R M stimulation on neurons with joint inputs were less consistent than those seen on the more dorsally located neurons, which have only ipsilateral cutaneous inputs. Inhibition, excitation and mixed effects were observed in different cells responding to joint inputs. Since spontaneous activity was minimal, inhibition was tested against a background of activity evoked by peripheral cutaneous stimulation. This was easily main-
P6 ., ~ ' ~ . :~.~. ' ~
.....
P8
Fig. 3. Locations of 10 m~ullary raphe stimulation sites from which minimum threahold inhibition of dorsal horn cells is produced. Sections traced from the atlas of Snider and Nicmer~5.
447 t a i n e d at a c o n s t a n t level for all cells except those o f l a m i n a I. Thus, the 9 l a m i n a I n e u r o n s r e p o r t e d here represent less t h a n h a l f the n u m b e r actually encountered. I n one experiment, b l o o d pressure fell well below the p h y s i o l o g i c a l limit and, a l t h o u g h units with high t h r e s h o l d inputs were found, n o n e were i n h i b i t e d by N R M s t i m u l a t i o n . N o n e u r o n a l d a t a f r o m this e x p e r i m e n t are included in this report.
Brain stem site of inhibition I n each experiment, when the first l a m i n a V-type n e u r o n was e n c o u n t e r e d ipsi-
Fig. 4. Peripherally evoked activity of a lamina V-type cell and its inhibition by stimulation of NRM. In this raster display, each recorded action potential of the unit is represented by a dot, the sweep from left to right is reset every second, the horizontal width of the display representing just less than 1 sec activity. This cell was not spontaneously active but responded over a wide dynamic range of mechanical stimulation of medial toe webbing. The peripheral stimulus was applied continuously during this record. Very strong inhibition is observed following stimulation of NRM by 100 Hz trains of twelve 300/zA square pulses, indicated by the closely spaced dots which appear at the beginning of each horizontal trace.
448 lateral to an intact D L F and found to be inhibited by stimulation from the midline electrode, the m i n i m u m threshold for inhibition was established by adjusting the depth of the electrode in 1 m m increments. The tip positions of the midline electrode at these points of m i n i m u m threshold for inhibition are shown in Fig. 3. In all cases, the m i n i m u m threshold for inhibition was less than 300 #A, averaging about 200/~A. In 7 experiments, a 3-electrode transverse array was used. In 16 of 19 units with high threshold inputs, the midline electrode elicited inhibition at lower stimulus intensities than either lateral electrode. The 3 instances of units in which a lateral electrode was more effective occurred in a single experiment. The nucleus magnocellularis o f the medial reticular formation, in which this and all other lateral electrodes lay, has been shown to inhibit lamina V-type neurons z4. O u r data, however, demonstrate that midline electrodes are generally more effective than laterally placed electrodes in generating inhibition o f dorsal h o r n neurons with high threshold inputs. Since the inhibition was weaker at more ventral sites, it is not due to current spread to the pyramidal tract. The optimal site for inhibition is thus nucleus raphe magnus. The rostrocaudal extent of the inhibitory region has not been systematically explored.
Fig. 5. Inhibition of the evoked activity of cell by NRM stimulus trains of variable length. This lamina V-type cell was not spontaneously active but responded to pinching a fold of skin on the foot, which was maintained throughout this record. The periods indicated by the stimulus artifacts, from top to bottom, consist of approximately 4, 5, 8, 10 and 12 pulses at 100 Hz. Inhibition, which is clearly seen with 12 or 10 pulses, is barely apparent with 8 pulses and cannot be discerned with fewer pulses. Each sweep represents 1 sec. Stimulus intensity is constant at 300/~A.
449
Time course of inhibition The time course o f inhibition was variable f r o m cell to cell and f r o m preparation to preparation. The majority o f cells were studied in the raster display mode. Fig. 4 shows an example o f especially powerful inhibition by stimulation o f N R M o f the peripherally evoked activity o f a cell with high threshold input. Complete inhibition lasting up to 600 msec with a 120 msec train length was observed. Fig. 5 demonstrates the effect o f varying the stimulus train length with constant stimulus intensity. Short trains did not produce obvious inhibition, and increasing the train length resulted in increased duration o f inhibition.
Spinal pathway for inhibition We have demonstrated recently that the N R M projection to the spinal cord is via the D L F 1. F o r this reason, a lesion was usually placed in one D L F prior to recording. N R M - i n d u c e d inhibition o f dorsal horn neurons on the lesioned side o f the cord was c o m p a r e d with that on the intact side. Inhibition o f cells by N R M stimula-
Fig. 6. Inhibition of the peripherally evoked activity of a lamina V-type cell and its abolition by lesion of the ipsilateral DLF. In the left-hand record, made before any cord lesion, the cell was not spontaneously active but responded to stimulation of its receptive field (long vertical bar). This activity was strongly inhibited by stimulation of NRM (short vertical bars) with a 100 Hz train of 10 square pulses of 300 psec duration and 250 pA intensity. In the right-hand record, the same neuron was recorded 10 min after a lesion had been made in the ipsilateral DLF rostral to the recording site. The cell became spontaneously active and responded to lighter cutaneous stimuli. The descending inhibition from NRM stimulation (short vertical bar) is not apparent. After the period of peripheral stimulation, indicated by the long vertical bar, there was strong postexcitatory inhibition. In both records, the horizontal sweep duration is 1 sec.
450 tion caudal and ipsilateral to DLF lesions was far less common than on the intact side (Fig. 2). Further, when inhibition was observed on the side with the lesion, it was of shorter duration. Threshold currents were also higher than on the side with the intact DLF, though this was not systematically studied. Four lamina I and 30 lamina V neurons ipsilateral to a D L F lesion were studied. One of the lamina 1 and 8 of the lamina V neurons were inhibited (Table 1). One cell was observed continuously before and after an ipsilateral D L F lesion was made (Fig. 6). With the cord intact, inhibition was powerful. After DLF section, NRM stimulation with the same stimulus parameters produced no inhibition. In contrast to the striking effects of D L F lesions, a control lesion of the ventrolateral funiculus did not interfere with the inhibition. Furthermore, inhibition was present caudal to an extensive cord lesion which spared only the ipsilateral DLF. In one case where inhibition survived an ipsilateral D L F lesion, a contralateral D L F lesion abolished the residual inhibition. Thus, the lesion data indicate that the D L F is the major, if not the sole, pathway mediating the inhibitory action of N R M on spinal cord dorsal horn neurons which respond to noxious stimulation. The effect is predominantly ipsilateral, but weak contralateral effects are observed. DISCUSSION These findings demonstrate a consistent inhibition of spinal dorsal horn neurons by electrical stimulation of NRM. This inhibition is mediated by a pathway descending in the DLF. Although the inhibition produced with only one D L F intact affects units on both sides of the cord, those on the intact side are more powerfully inhibited. The inhibition is observed in lamina I neurons with small cutaneous receptive fields and high intensity mechanical thresholds, and in neurons, primarily in lamina V, responding to a wide range of cutaneous stimulus intensities. Since neurons with these properties in these locations have been implicated in pain transmission ~8,~8,4°,41, the present results support the hypothesis that N R M is crucial in modulating pain transmission at the spinal cord level. These results are consistent with the anatomical observation that after injection of tritiated leucine into NRM, labeled fibers are found in the D L F 1. The teiminal regions of this projection coincide with the locations of inhibited neurons, namely, laminae I and V. In contrast, the projection from N R M spares most of lamina IV, the cells of which characteristically have only low threshold inputs and are not inhibited by N R M stimulation. Although it is known that stimulation of the sensorimotor cortex 1~ and the pyramidal tract 15 exerts an inhibitory effect on dorsal horn neurons, we feel that current spread to the pyramidal tract is unlikely to account for the observed inhibition in the present exper,ments. First, inhibition at stimulus intensities less than 150 #A could be obtained from sites more than 2 mm distant from the pyramidal tract. According to the data of Ranck 3~, the pyramidal tract is, therefore, beyond the range of current spread. Second, brain stem sites for maximal inhibition were invariably dorsal to the
451 pyramidal tract. Finally, pyramidal tract stimulation was found by Fetz 15 to be more consisteotly inhibitory to lamina IV than to lamina V neurons. Just the opposite was the case in the present experiments. The recent findings of stereospecific opiate binding 36 and of an endogenous compound with morphine-like activitya, 2~ in the mammalian central nervous system lend support to the concept of an endogenous pain-suppression system. The opiate receptor has a relatively high concentration in medial thalamic and periaqueductal gray regions. Both microinjection of morphine and electrical stimulation in these regions produce profound analgesia. This analgesia requires a descending DLF pathway to spinal cord 1,z which is probably serotoninergiclg,aL Consistent with the anatomical studies of Ruda a4, preliminary physiological studies in this laboratory have demonstrated short latency driving of NRM neurons by single electrical stimuli to periaqueductal gray 16. Since NRM is the major source of descending serotoninergic fibers in the dorsolateral funiculus 18,~6 and since NRM lesions interfere with morphine analgesia a°, it is likely that the inhibitory NRM projection to spinal dorsal horn neurons with high intensity inputs mediates both opiate and stimulus-produced analgesia from mesencephalic sites. A recent report that iontophoresed serotonin inhibits lamina I neurons al and that LSD blocks midbrain inhibition of dorsal horn neurons with high intensity inputs 19 is consistent with this hypothesis. Engberg el al. 14 found, in decerebrate cats, that spinal interneurons with inputs from group II and III afferents are under tonic inhibitory control mediated, in part, by serotoninergic neurons in the ventromedial medulla. Electrolytic lesions of the raphe nuclei eliminated most, but not all, of the tonic inhibition. The fact that some inhibition persists despite complete raphe lesions implies that adjacent magnocellular reticular formation participates in this inhibitory control. These data are consistent with our finding that midline, i.e. raphe, electrode positions are the most effective for inhibition of spinal nociceptive neuions. However, some contribution to the inhibition by neurons outside of NRM in the adjacent reticular formation cannot be excluded. The present results provide one substrate for the repeatedly demonstrated brain stem control of spinal dorsal horn neurons. There are direct connections to NRM from sensorimotor cortex, spinal cord, cerebeUum7, periaqueductal gray 34 and adjacent reticular formation ~1. The widespread convergence onto this system, which modulates high threshold spinal dorsal horn neurons, indicates that multiple factors influence the flow of nociceptive input along central somatosensoly pathways. ACKNOWLEDGEMENTS We thank Ms. Mary Liu for histological assistance and Ms. Sherry Davis for editorial work on this manuscript. This research was supported by P.H.S. Grants NS 70777 and NS 11529 to H. L. Fields, MH 7082 to S. D. Anderson, and NS 05272 to A. I. Basbaum.
452 REFERENCES 1 Basbaum, A. I., Ctanton, C. H. and Fields, H. L., Contrasting descending projections of the nucleus gigantocellularis (NGC) and nucleus raphe magnus (NRM) of the cat, Paper presented at Cajal Club, Louisville, Ky., 1976. 2 Basbaum, A. I., Marley, N. and O'Keefe, J., Effects of spinal cord lesions on the analgesic properties of electrical brain stimulation, Advanc. Neurol., in press. 3 Belluzzi, J. D., Grant, N., Garsky, V., Sarantakis, D., Wise, C. D. and Stein, L., Analgesia induced in vivo by central administration of enkephalin in rat, Nature (Lond.), 260 (1976) 625-626. 4 Besson, J. M., Guilbaud, G. and LeBars, D., Descending inhibitory influences exerted by the brain stem upon the activities of dorsal horn lamina V cells induced by intra-arteriat injection of bradykinin into the limbs, J. Physiol. (Lond.), 248 (1975) 725-739. 5 Bishop, P. O., Burke, W. and Davis, R., The identification of single units of central visual pathways, J. Physiol. (Lond.), 162 (1962) 409-431. 6 Brodal, A., Taber, E. and Walberg, F., The raphe nuclei of the brain stem in the cat. 11. Efferent connections, J. comp. Neurol., 144 (1960) 239-260. 7 Brodal, A., Walberg, F. and Taber, E., The raphe nuclei of the brain stem in the cat. III. Afferent connections, J. comp. NeuroL, 144 (1960) 261-282. 8 Brown, A. G., Kirk, E. J. and Martin, H. F., III, Descending and segmental inhibition of transmission through the spinocervical tract, J. Physiol. (Lond.), 230 (1973) 689-705. 9 Cervero, F., lggo, A. and Ogawa, H., Nociceptor-driven dorsal horn neurones in the lumbar spinal cord of the cat, Pain, 2 (1976) 5-24. 10 Christensen, B. N. and Perl, E. R., Spinal neurons specifically excited by noxious or thermal stimuli: marginal zone of the dorsal horn, J. Neurophysiol., 33 (1970) 293-307. 11 Couch, J. R., Jr., Responses of neurons in the raphe nuclei to serotonin, norepinephrine and acetylcholine and their correlat;on with an excitatory synaptic input, Brain Research, 19 (1970) 137-150. 12 Coulter, J. D., Maunz, R. A. and Willis, W. D., Effects of stimulation of sensorimotor cortex on primate spinothalamic neurons, Brain Research, 65 (1974) 351-356. 13 Dahlstr6m, A. and Fuxe, K., Evidence for the existence of monoamine neurons in the central nervous system. II. Experimentally induced changes in the intraneuronal amine levels of bulbospinal neuron systems, Acta physiol, scand., 64, Suppl. 247 (1965) 5-36. 14 Engberg, 1., Lundberg, A. and Ryall, R. W., Reticulospinal inhibition of transmission in reflex pathways, J. Physiol. (Lond.), 194 (1968) 201-223. 15 Fetz, E. E., Pyramidal tract effects on interneurones in the cat lumbar dorsal horn, J. Neurophysiol., 31 (1968) 69-80. 16 Fields, H. L., Anderson, S. D., Clanton, C . H . and Basbaum, A. I., Nucleus raphe magnus: a common mediator of opiate and stimulus produced analgesia, Trans. Amer. neurol. Ass., in press. 17 Fields, H. L., Clanton, C. H. and Anderson, S. D., Somatosensory properties of spinoreticular neurons, Brain Research, 120 (1977) 49-66. 18 Fields, H. L., Wagner, G. M. and Anderson, S. D., Some properties of spinal neurons projecting to the medial brain-stem reticular formation, Exp. NeuroL, 47 (1975) 118-134. 19 Guilbaud, G., Besson, J. M., Oliveras, J. L. and Liebeskind, J. C., Suppression by LSD of the inhibitory effect exerted by dorsal raphe stimulation on certain spinal cord interneurons in the cat, Brain Research, 61 (1973) 417-422. 20 Handwerker, H. O., Iggo, A., and Zimmermann, M., Segmental and supraspinal actions on dorsal horn neurons responding to noxious and non-noxious skin stimuli, Pain, 1 (1975) 147-166. 21 Holmqvist, B. and Lundberg, A., On the organization of the supraspinat inhibitory control of interneurones of various spinal reflexes, Arch. ital. Biol., 97 (1959) 340-356. 22 Hughes, J., Isolation of an endogenous compound from the brain with pharmacological properties similar to morphine, Brain Research, 88 (1975) 295-308. 23 Mayer, D. J., Price, D. D. and Becker, D. P., Neurophysiological characterization of the anterolateral spinal cord neurons contributing to pain perception in man, Pain, 1 (1975) 51-58. 24 Mayer, D. J. and Liebeskind, J. C., Pain reduction by focal electrical stimulation of the brain: an anatomical and behavioral analysis, Brain Research, 68 (1974) 73-93. 25 McCreery, D.B. and Bloedel, J. R., Reduction of the response of cat spinothalamic neurons to graded mechanical stimuli by electrical stimulation of the lower brainstem, Brain Research, 97 (1975) 151-156.
453 26 Oliveras, J. L., Besson, J. M., Guilbaud, G. and Liebeskind, J. C., Behavioral and electrophysiological evidence of pain inhibitionfrom midbrain stimulation in the cat, Exp. Brain Res., 20 (1974) 32-44. 27 Oliveras, J. L., Redjemi, F., Guilbaud, 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. 28 Price, D. D. and Mayer, D. J., Neurophysiological characterization of the anterolateral quadrant neurons subserving pain in M. mulatta, Pain, 1 (1975) 59-72. 29 Price, D. D. and Browe, A. C., Spinal cord coding of graded non-noxious and noxious temperature increases, Exp. Neurol., 48 (1975) 201-221. 30 Proudfit, H. K. and Anderson, E. G., Morphine analgesia: blockade by raphe magnus lesions, Brain Research, 98 (1975) 612-618. 31 Randic, M. and Yu, H. H., Micro-electrophoretic study of cat dorsal horn neurones activated by noxious stimuli, J. Physiol. (Lond.), 252 (1975) 23-24. 32 Ranck, J.B., Jr., Which elements are excited in electrical stimulation of mammalian central nervous system: a review, Brain Research, 98 (1975) 417--440. 33 Reynolds, D. V., Surgery in the rat during electrical analgesia induced by focal brain stimulation, Science, 164 (1969) 444--445. 34 Ruda, M. A., Autoradiographic Examination of the Efferent Projections of the Midbrain Central Grey in the Cat, Ph.D. Thesis, University of Pennsylvania, 1976. 35 Snider, R. S. and Niemer, W. T., ,4 Stereotaxic Atlas of the Cat Brain, University of Chicago Press, Chicago, II1., 1961. 36 Snyder, S. H., Opiate receptor in normal and drug altered brain function, Nature (Lond.), 257 (1975) 185-189. 37 Taub, A., Local, segmental and supraspinal interaction with a dorsolateral spinal cutaneous afferent system, Exp. NeuroL, 10 (1964) 357-374. 38 Trevino, D. L. and Carstens, E., Confirmation of the location of spinothalamic neurons in the cat and monkey by the retrograde transport of horseradish peroxidase, Brain Research, 98 (1975) 177-182. 39 Vogt, M., The effect of lowering the 5-hydroxytryptamine content of the rat spinal cord on analgesia produced by morphine, J. PhysioL (Lond.), 236 (1974) 483-498. 40 Wall, P. D., The laminar organization of dorsal horn and effects of descending influences, J. Physiol. (Lond.), 188 (1967) 403-424. 41 Willis, W. D., Trevino, D. L., Coulter, J. D. and Maunz, R. A., Responses of primate spinothalamic tract neurons to natural stimulation of hindlimb, J. NeurophysioL, 37 (1974) 358-372.
