98
Neuroscience Letters, 134 (1991) 98 102 © 1991 ElsevierScientific Publishers Ireland Ltd. All rights reserved 0304-3940/91/$ 03.50
NSL 08280
Spinal afferent projections to subnucleus reticularis dorsalis in the rat L. V i l l a n u e v a , J. de P o m m e r y , D. M e n 6 t r e y a n d D. Le B a r s INSERM, U-161, Paris (France)
(Received 22 July 199I; Revised version received 18 September 1991; Accepted 20 September 1991) Key words." Brainstem;Pain; Reticular formation; Subnucleus reticularis dorsalis; Spinal cord; Spino-reticular pathway
Small amounts of the retrograde tracer WGA-apoHRP-Au complex were injected in the caudal medulla to study the spinal afferentsto the subnucleus reticularis dorsalis (SRD). Labelled neurones were found at all levels of the spinal cord: the highest numbers were in the ipsilateral cervical spinal cord (mainly laminae I, V, VI, VII, VIII and X), the lowest were at the thoracic and lumbar levelsbilaterally, while an intermediate density was found bilaterally at the sacral level. When injection sites were located in the underlying subnucleus reticularis ventralis (SRV), labelling was bilateral and mainly in the deep layers of the cervical spinal cord. Together with our previous electrophysiologicaland anatomical data, this study suggests that the SRD provides a link in spino-reticulo-spinalloops implicated in the processing of pain.
Both anatomical and physiological data suggest that the brainstem reticular formation (BRF) plays an important role in nociception [7]. However, the relative contributions of the various regions of the BRF to the processing of pain have not yet been determined. We recently reported that in the rat [27], neurones within a restricted area of the medulla, the Subnucleus Reticularis Dorsalis (SRD) [19, 25], appear to have a role in processing specifically nociceptive information. SRD is an area of the caudal BRF, located ventral to the cuneate nucleus, between trigeminal nucleus caudalis and the nucleus of the solitary tract (Fig. 1, see also refs. 19, 25). With respect to cutaneous inputs, SRD neurones are activated exclusively by volleys in A6- or AO- and C- fibres from any part of the body, and either exclusively or preferentially by noxious stimulation of the skin [27]; they encode the strength of electrical, mechanical and thermal cutaneous stimuli within a noxious range [26]; they respond to noxious chemical stimulation of the viscera, and encode mechanical visceral stimuli, again within a noxious range [22]; their A& and C-fibre-evoked activities are depressed by morphine in a dose-related and naloxone-reversible fashion [5]; finally the spinal pathways that activate SRD neurones ascend in the ventro-lateral funiculi [6]. Neurones presenting similar features have been recorded from a corresponding area in the monkey [28]. Correspondence: L. Villanueva, INSERM, U-161, 2, Rue d'Al6sia, 75014, Paris, France.
Fig. 1A represents the location of 120 nociceptive units recorded in the rat during the electrophysiological experiments cited above. It can he observed that the neurones were largely confined to the SRD region, with no nociceptive units being found within the underlying subnucleus reticularis ventralis (SRV). Spinal afferents to the SRD were compared to those projecting to adjacent areas, including the cuneate nucleus, trigeminal nucleus caudalis and the Subnucleus Reticularis Ventralis (SRV) (see Fig. 1A). Surgical procedures were carried out on male Sprague-Dawley rats (270-300 g) under chloral hydrate anaesthesia (400 mg/kg, i.p.). The animals were mounted in a ventro-flexed position in a stereotaxic device. Pressure injections were made through micropipettes (35-40 /~m diameter) which were at an angle of 60 ° to the horizontal plane. The rostro-caudal locations of the injection sites were chosen between 5.6 and 5.1 m m from the interaural line [20], to make the results comparable with our previous electrophysiological data (Fig. 1A). The tracer consisted of wheat germ agglutinin apohorseradish peroxidase conjugate ( W G A - a p o H R P ) coupled to colloidal gold [2]. Small volumes (0.15-0.20 pl) were injected in order to obtain minimal spread. The technique was essentially the same as previously reported [2], and the delineation of the structures was based upon observation of the cytoarchitecture and the nomenclature was adopted from the relevant literature [19, 20]. Injection sites were selected on the basis of their clear
99
J
ft.
Fig. 1. A: schematic representation (adapted from refs. 19 and 20) of a coronal section of the medulla, 5.6 mm caudal to the interaural line. Each dot represents the recording site of a unit with total or partial nociceptive convergence (from refs. 5, 6, 22, 26, 27). Note that the population is largely confined within the subnucleus reticularis dorsalis. B: bright-field photomicrograph of a coronal section from the left medulla showing an injection site of WGA-apoHRP-Au complex within the SRD (same site as in Fig. 2A). Note that the extent of the injection corresponds roughly to the area covered by the recording sites. Cu, cuneate nucleus; Pyr., pyramidal decussation; Sol, nucleus of the solitary tract; SRD, subnucleus reticularis dorsalis; SRV, subnucleus reticularis ventralis; 5, trigeminal nucleus caudalis.
delimitations, i.e. minimal spread of tracer within the location defined by the previous electrophysiological data [27], namely SRD (n = 4), SRV (n = 2), the cuneate nucleus (n = 3), and trigeminal nucleus caudalis (n = 2). A representative example of the results obtained following an injection in the SRD is shown in Fig. 2A. Labelled neurones were found at all levels of the spinal cord, but with major differences in their densities: the highest density being in the ipsilateral cervical grey matter, the lowest at the thoracic and lumbar levels bilaterally, with an intermediate density being found bilaterally at the sacral level. In the cervical cord, the highest density of labelled cells was observed ipsilaterally: mainly in laminae I, IV, V, VI but also in laminae VII, VIII and in the dorsolateral funiculus close to the neck of the dorsal horn. On the contralateral side, the highest densities of labelled cells were observed in laminae I and VII. In addition, large number of labelled cells were found bilaterally in lamina X. In the thoracic and lumbar cord, a small amount of labelling was observed bilaterally in laminae, I, V, VII, X, and in the dorsolateral funiculus close to the neck of
the dorsal horn. Labelling was also observed bilaterally in laminae I, V, VI, X, and the lateral spinal nucleus of the sacral cord. When injections were centered in the SRV (Fig. 2B), the highest density of labelled cells was again observed at the cervical level, but in this case was bilateral and mainly in the deep laminae, i.e. V, VII, VIII, and X. Although present, labelling was less marked in the dorsolateral funiculus close to the neck of the dorsal horn. Very little labelling was found as one moved caudally towards the sacral cord, This agrees with data obtained by Men6trey et al. [18]. Interestingly, their study also demonstrated differences between the spinal afferent inputs to the SRV and those to the lateral reticular nucleus (LRN), the latter receiving dense projections from all spinal levels, inluding from the most superficial layers of the dorsal horn. In contrast to the neurones which project to the SRD, those which project to the LRN are located mainly contralaterally. When injection sites were located in an area lateral to the SRD, which included the dorsal part of trigeminal nucleus caudalis, labelled cells were observed only in the upper segments of the cervical spinal cord. This labelling was bilateral, sparse, and located mainly in laminae III and X. Following injections in the cuneate nucleus, labelled cells were located in the ipsilateral cervical spinal cord, notably in lamina IV, as previously reported [10, 11]. The data reported herein show particular distributions of neurones within spinal cord which project to the SRD. The differences between the origins of spinal projections to the SRD and SRV could account partly for the finding that neurones in these two structures do not have identical electrophysiological properties. However spinal afferents to these nuclei originate from laminae that receive noxious inputs. Therefore the anatomical results cannot totally explain the physiological differences. Indeed, in contrast to SRD neurones, those in the SRV are either unresponsive to or inhibited by somatic stimuli [27] and thus may have a role in other functions (e.g. autonomic, see refs. in ref. 1). Interestingly, following noxious stimulation, a significant increase in metabolic activity is seen in the SRD, but not in the SRV [21]. The SRD receives afferents from all levels of the spinal cord. This is consistent with electrophysiological data which show that this structure contains neurones which can be activated by stimuli applied to any part of the body [27]. Many spinal afferents to the SRD come from laminae I and V-VI, which contain populations of neurones which are involved in the transmission of nociceptive information (see refs. in ref. 4). Neurones located in deeper laminae of the spinal cord (VII-VIII and X), also project to the SRD; these laminae also contain nocicep-
100
rive units [8, 14]. Many spinal neurones exhibit cutaneovisceral convergence [8, 14] and a subpopulation of SRD units encode both cutaneous and visceral noxious stimuli [221 In contrast with our results, a previous study of spinal
afferents to the caudal BRF described retrogradely labelled cells predominantly in laminae I and X [16]. However, the injections were made more rostrally (4.35.3 mm caudal to the interaural line) [20]. In addition, it is possible that some of the injection sites would have
cervical (C3)
cervical (C5)
)
mid. thoracic
;°
( lumbar (L3)
J
upp. sacral
Fig. 2. Series of camera lucida drawings of coronal sections in a case where the W G A - a p o H R P Au complex was injected within the SRD (A) or the SRV (B). Injection sites consisted of a dense core (black) surrounded by a halo filled with neuronal labelled elements (stipple). Diagrams of the spinal cord illustrate the total number of cells (dots) contained in 5 consecutive 4 0 / t m thick sections. In both cases, the largest number of labelled neurones was found in the cervical cord. However, there was one principal difference in respect of their distributions within the grey matter: the largest number of labelled cells was in the dorsal horn for injections in the SRD and in the ventral horn for injections in the SRV (see text).
101
included areas containing neuronal populations which are functionally different from SRD neurones: Lima [16] reported that lesions of the dorsal columns strongly reduced the number of labelled cells, especially in the superficial dorsal horn, whereas we have shown electrophysiologically that lesions of the dorsal columns do not affect the spinal input to the SRD [6]. In fact the ascending spinal pathways which are responsible for activating SRD neurones are crossed and confined to the lateral parts of the ventral quadrant [6]. It has been demonstrated using degeneration techniques following ventrolateral cordotomy in the rat [24] that a large number of fibres travelling in the ventrolateral quadrant terminate in the SRD. One intriguing question arises from the finding that the largest number of afferents to the SRD originates from the ipsilateral cervical cord: to date, all neurones recorded in this structure have shown a 'whole body receptive field' with a contralateral dominance [27]. This apparent discrepancy could be due to the transport of tracer to the adjacent cervical cord being easier than to more caudal areas. However, this possibility seems unlikely since in every case, the potency of labeling following injections within the SRD was: cervical > sacral > thoracic = lumbar segments. Interestingly, the largest numbers of retrogradely labelled cells in the spino-thalamic (STT) and spino-mesencephalic tracts (SMT) in the rat were also found in the upper cervical cord [12, 30]. Is this a common functional organisation of ascending somatosensory pathways? One could imagine the upper cervical cord acting as a somatosensory relay between caudal areas of the spinal cord and higher centres. Recent electrophysiological reports of neurones in the upper cervical spinal cord with widespread receptive fields including the oro-facial region, the hindpaws and the tail support such a possibility in the cat and the monkey [13, 23, 29]. As previously suggested [29], inputs to the cervical enlargement can originate from different sources, including from collaterals of ascending axons. Within the framework of this hypothesis, one could envisage that at least some inputs to SRD neurones have relays in the cervical cord. Together with the fact that other tracts involved in the transmission of nociceptive information may have a similar organisation, this could explain the widespread relief of pain, including pain from caudal segments of the body, following commissural myelotomies of the upper cervical spinal cord in humans (see refs. in ref. 9). Another interesting finding is that most of the laminae which contain spinal afferents to the SRD have been shown by the PHA-L method [3] to receive dense efferent inputs from the same region. This is especially true for laminae V, VI, VII and X. In terms of density of
efferent projections from the SRD, an identical rank of potencies was found for the different levels of the spinal cord, i.e. is very dense at the cervical level, moderate in the thoracic and lumbar cords and intermediate at the sacral level [3]. Such reciprocal connections strengthen the suggestion that SRD neurones may belong to spinoreticulo-spinal loops implicated in the processing of nociceptice information via feed-back or more diffuse mechanisms [15, 27]. The authors are grateful to Dr. S.W. Cadden for advice in the preparation of the manuscript. This work was supported by l'Institut National de la Sant6 et de la Recherche Mrdicale (INSERM) and la Direction de Recherches et Etudes Techniques (DRET). 1 Aicher, S.A. and Randich, A., Antinociception and cardiovascular responses produced by electrical stimulation in the nucleus tractus solitarius, nucleus reticularis ventralis, and the caudal medulla, Pain, 42 (1990) 103-119. 2 Basbaum, A. and Menrtrey, D., Wheat germ agglutinin-apoHRP gold: a new retrograde tracer for light- and electron-microscopic single- and double-label studies, J. Comp. Neurol., 261 (1987) 306318. 3 Bernard, J.F., Villanueva, L., Carrour, J. and Le Bars, D., Efferent projections from the subnucleus reticularis dorsalis (SRD): a Phaseolus vulgaris leucoagglutinin study in the rat, Neurosci. Lett., 116 (1990) 257-262. 4 Besson, J.M. and Chaouch, A., Peripheral and spinal mechanisms of nociception, Physiol. Rev., 67 (1987) 67-186. 5 Bing, Z., Villanueva, L. and Le Bars, D., Effects of systemic morphine upon A~- and C-fibre evoked activities of subnucleus reticularis dorsalis neurones in the rat medulla, Eur. J. Pharmacol., 164 (1989) 85-92. 6 Bing, Z., Villanueva, L. and Le Bars, D., Ascending pathways in the spinal cord involved in the activation of subnucleus reticularis dorsalis neurons in the medulla of the rat, J. Neurophysiol., 63 (1990) 424-438. 7 Bowsher, D., Role of the reticular formation in responses to noxious stimulation, Pain, 2 (1976) 361 378. 8 Cervero, F. and Tattersall, J.E.H., Somatic and visceral sensory integration in the thoracic spinal cord. In F. Cervero and J.F.B. Morrison (Eds.), Visceral sensation, Progress in Brain Research, Vol. 67, Elsevier, 1986, pp. 189-205. 9 Cook, A.W., Nathan, P.W. and Smith, M.C., Sensory consequences of commissural myelotomy, Brain, 107 (1984) 547-568. 10 De Pommery, J., Roudier, F. and Men&rey, D., Postsynaptic fibers reaching the dorsal column nuclei in the rat, Neurosci. Lett., 50 (1984) 319-323. I1 Giesler, G.J,, Nahin, R.L. and Madsen, A., Postsynaptic dorsal column pathway of the rat. I. Anatomical studies, J. Neurophysiol., 51 (1984) 260-275. 12 Granum, S.L., The spinothalamic system of the rat, I. Locations of cells of origin, J. Comp. Neurol., 247 (1986) 159-180. 13 Hodge, C.J., Apakarian, V.A., Gingold, S. and Stevens, R.T., Spinothalamic tract cells of the high cervical spinal cord of primate, Pain, Suppl. 5 (1990) $98. 14 Honda, C., Visceral and somatic afferent convergence onto neurones near the central canal in the sacral spinal cord of the cat, J. Neurophysiol., 53 (1985) 105%1078.
102 15 Le Bars D. and Villanueva, L., Electrophysiological evidence for the activation of descending inhibitory controls by nociceptive afferent pathways. In H.L. Fields and J.M. Besson (Eds.), Pain Modulation, Progress in Brain Research, Vol. 77, Elsevier, 1988, pp. 275-299. 16 Lima, D., A spinomedullary projection terminating in the dorsal reticular nucleus of the rat, Neuroscience, 34 (1990) 577-590. 17 Men6trey, D. and Basbaum, A., Spinal and trigeminal projections to the nucleus of the solitary tract: a possible substrate for somatovisceral and viscerovisceral reflex activation, J. Comp. Neurol., 255 (1987) 439-450. 18 Men6trey, D., Roudier, F. and Besson, J.M., Spinal neurons reaching the lateral reticular nucleus as studied in the rat by retrograde transport of horseradish peroxidase, J. Comp. Neurol., 220 (1983) 439-452. 19 Newman, D.B., Distinguishing rat brainstem reticulospinal nuclei by their neuronal morphology. I. Medullary nuclei, J. Hirnforsch., 26 (1985) 187-226. 20 Paxinos, G. and Watson, C., The Rat Brain in Stereotaxic Coordinates, Academic Press, San Diego, 1986. 21 Porro, C.A., Cavazzuti, M., Galetti, A. and Sassatelli, L., Functional activity mapping of the rat brainstem during formalininduced noxious stimulation, Neuroscience, 41 (1991) 667q580. 22 Roy, J.C., Bing, Z., Villanueva, L. and Le Bars, D., Activation of Subnucleus Reticularis Dorsalis neurones (SRD) of the rat medulla by colorectal distention, Pain, Suppl. 5 (1990) 166. 23 Smith, M.V., Apkarian, A.V. and Hodge, C.J. Jr., Somatosensory response properties of contralaterally projecting spinothalamic and
24
25
26
27
28
29
30
nonspinothalamic neurons in the second cervical segment of the cat, J. Neurophysiol., 66 (1991) 83-102. Torvik, A., Afferent connections to the sensory trigeminal nuclei, the nucleus of the solitary tract and adjacent structures. An experimental study in the rat, J. Comp. Neurol., 106 (1956) 51 141. Valverde, F., Reticular formation of the albino rat brain stem cytoarchitecture and corticofugal connections, J. Comp. Neurol., 119 (1962) 25-49. Villanueva, L., Bing, Z., Bouhassira, D. and Le Bars, D., Encoding of electrical, thermal, and mechanical noxious stimuli by subnucleus reticularis dorsalis neurons in the rat, J. Neurophysiol., 61 (1989) 391-402. Villanueva, L., Bouhassira, D., Bing, Z. and Le Bars, D., Convergence of heterotopic nociceptive information onto subnucleus reticularis dorsalis neurons in the rat medulla, J. Neurophysiol., 60 (1988) 980-1009. Villanueva, L., Cliffer, K.D., Sorkin, L., Le Bars. D and Willis, W.D., Convergence of heterotopic nociceptive information onto neurons of the caudal medullary reticular formation in the monkey (Macacafascicularis), J. Neurophysiol., 63 (1990) 1118-1127. Yeziersky, R.P. and Broton, J.G., Functional properties of spinomesencephalic tract (SMT) cells in the upper cervical spinal cord of the cat, Pain, 45 (1991) 187-196. Yeziersky, R.P. and Mendez, C.M., Spinal distribution and collateral projections of rat spinomesencephalic tract cells, Neuroscience, 44 (1991) 113-130.
Neuroscience Letters, 134 (1991) 98 102 © 1991 ElsevierScientific Publishers Ireland Ltd. All rights reserved 0304-3940/91/$ 03.50
NSL 08280
Spinal afferent projections to subnucleus reticularis dorsalis in the rat L. V i l l a n u e v a , J. de P o m m e r y , D. M e n 6 t r e y a n d D. Le B a r s INSERM, U-161, Paris (France)
(Received 22 July 199I; Revised version received 18 September 1991; Accepted 20 September 1991) Key words." Brainstem;Pain; Reticular formation; Subnucleus reticularis dorsalis; Spinal cord; Spino-reticular pathway
Small amounts of the retrograde tracer WGA-apoHRP-Au complex were injected in the caudal medulla to study the spinal afferentsto the subnucleus reticularis dorsalis (SRD). Labelled neurones were found at all levels of the spinal cord: the highest numbers were in the ipsilateral cervical spinal cord (mainly laminae I, V, VI, VII, VIII and X), the lowest were at the thoracic and lumbar levelsbilaterally, while an intermediate density was found bilaterally at the sacral level. When injection sites were located in the underlying subnucleus reticularis ventralis (SRV), labelling was bilateral and mainly in the deep layers of the cervical spinal cord. Together with our previous electrophysiologicaland anatomical data, this study suggests that the SRD provides a link in spino-reticulo-spinalloops implicated in the processing of pain.
Both anatomical and physiological data suggest that the brainstem reticular formation (BRF) plays an important role in nociception [7]. However, the relative contributions of the various regions of the BRF to the processing of pain have not yet been determined. We recently reported that in the rat [27], neurones within a restricted area of the medulla, the Subnucleus Reticularis Dorsalis (SRD) [19, 25], appear to have a role in processing specifically nociceptive information. SRD is an area of the caudal BRF, located ventral to the cuneate nucleus, between trigeminal nucleus caudalis and the nucleus of the solitary tract (Fig. 1, see also refs. 19, 25). With respect to cutaneous inputs, SRD neurones are activated exclusively by volleys in A6- or AO- and C- fibres from any part of the body, and either exclusively or preferentially by noxious stimulation of the skin [27]; they encode the strength of electrical, mechanical and thermal cutaneous stimuli within a noxious range [26]; they respond to noxious chemical stimulation of the viscera, and encode mechanical visceral stimuli, again within a noxious range [22]; their A& and C-fibre-evoked activities are depressed by morphine in a dose-related and naloxone-reversible fashion [5]; finally the spinal pathways that activate SRD neurones ascend in the ventro-lateral funiculi [6]. Neurones presenting similar features have been recorded from a corresponding area in the monkey [28]. Correspondence: L. Villanueva, INSERM, U-161, 2, Rue d'Al6sia, 75014, Paris, France.
Fig. 1A represents the location of 120 nociceptive units recorded in the rat during the electrophysiological experiments cited above. It can he observed that the neurones were largely confined to the SRD region, with no nociceptive units being found within the underlying subnucleus reticularis ventralis (SRV). Spinal afferents to the SRD were compared to those projecting to adjacent areas, including the cuneate nucleus, trigeminal nucleus caudalis and the Subnucleus Reticularis Ventralis (SRV) (see Fig. 1A). Surgical procedures were carried out on male Sprague-Dawley rats (270-300 g) under chloral hydrate anaesthesia (400 mg/kg, i.p.). The animals were mounted in a ventro-flexed position in a stereotaxic device. Pressure injections were made through micropipettes (35-40 /~m diameter) which were at an angle of 60 ° to the horizontal plane. The rostro-caudal locations of the injection sites were chosen between 5.6 and 5.1 m m from the interaural line [20], to make the results comparable with our previous electrophysiological data (Fig. 1A). The tracer consisted of wheat germ agglutinin apohorseradish peroxidase conjugate ( W G A - a p o H R P ) coupled to colloidal gold [2]. Small volumes (0.15-0.20 pl) were injected in order to obtain minimal spread. The technique was essentially the same as previously reported [2], and the delineation of the structures was based upon observation of the cytoarchitecture and the nomenclature was adopted from the relevant literature [19, 20]. Injection sites were selected on the basis of their clear
99
J
ft.
Fig. 1. A: schematic representation (adapted from refs. 19 and 20) of a coronal section of the medulla, 5.6 mm caudal to the interaural line. Each dot represents the recording site of a unit with total or partial nociceptive convergence (from refs. 5, 6, 22, 26, 27). Note that the population is largely confined within the subnucleus reticularis dorsalis. B: bright-field photomicrograph of a coronal section from the left medulla showing an injection site of WGA-apoHRP-Au complex within the SRD (same site as in Fig. 2A). Note that the extent of the injection corresponds roughly to the area covered by the recording sites. Cu, cuneate nucleus; Pyr., pyramidal decussation; Sol, nucleus of the solitary tract; SRD, subnucleus reticularis dorsalis; SRV, subnucleus reticularis ventralis; 5, trigeminal nucleus caudalis.
delimitations, i.e. minimal spread of tracer within the location defined by the previous electrophysiological data [27], namely SRD (n = 4), SRV (n = 2), the cuneate nucleus (n = 3), and trigeminal nucleus caudalis (n = 2). A representative example of the results obtained following an injection in the SRD is shown in Fig. 2A. Labelled neurones were found at all levels of the spinal cord, but with major differences in their densities: the highest density being in the ipsilateral cervical grey matter, the lowest at the thoracic and lumbar levels bilaterally, with an intermediate density being found bilaterally at the sacral level. In the cervical cord, the highest density of labelled cells was observed ipsilaterally: mainly in laminae I, IV, V, VI but also in laminae VII, VIII and in the dorsolateral funiculus close to the neck of the dorsal horn. On the contralateral side, the highest densities of labelled cells were observed in laminae I and VII. In addition, large number of labelled cells were found bilaterally in lamina X. In the thoracic and lumbar cord, a small amount of labelling was observed bilaterally in laminae, I, V, VII, X, and in the dorsolateral funiculus close to the neck of
the dorsal horn. Labelling was also observed bilaterally in laminae I, V, VI, X, and the lateral spinal nucleus of the sacral cord. When injections were centered in the SRV (Fig. 2B), the highest density of labelled cells was again observed at the cervical level, but in this case was bilateral and mainly in the deep laminae, i.e. V, VII, VIII, and X. Although present, labelling was less marked in the dorsolateral funiculus close to the neck of the dorsal horn. Very little labelling was found as one moved caudally towards the sacral cord, This agrees with data obtained by Men6trey et al. [18]. Interestingly, their study also demonstrated differences between the spinal afferent inputs to the SRV and those to the lateral reticular nucleus (LRN), the latter receiving dense projections from all spinal levels, inluding from the most superficial layers of the dorsal horn. In contrast to the neurones which project to the SRD, those which project to the LRN are located mainly contralaterally. When injection sites were located in an area lateral to the SRD, which included the dorsal part of trigeminal nucleus caudalis, labelled cells were observed only in the upper segments of the cervical spinal cord. This labelling was bilateral, sparse, and located mainly in laminae III and X. Following injections in the cuneate nucleus, labelled cells were located in the ipsilateral cervical spinal cord, notably in lamina IV, as previously reported [10, 11]. The data reported herein show particular distributions of neurones within spinal cord which project to the SRD. The differences between the origins of spinal projections to the SRD and SRV could account partly for the finding that neurones in these two structures do not have identical electrophysiological properties. However spinal afferents to these nuclei originate from laminae that receive noxious inputs. Therefore the anatomical results cannot totally explain the physiological differences. Indeed, in contrast to SRD neurones, those in the SRV are either unresponsive to or inhibited by somatic stimuli [27] and thus may have a role in other functions (e.g. autonomic, see refs. in ref. 1). Interestingly, following noxious stimulation, a significant increase in metabolic activity is seen in the SRD, but not in the SRV [21]. The SRD receives afferents from all levels of the spinal cord. This is consistent with electrophysiological data which show that this structure contains neurones which can be activated by stimuli applied to any part of the body [27]. Many spinal afferents to the SRD come from laminae I and V-VI, which contain populations of neurones which are involved in the transmission of nociceptive information (see refs. in ref. 4). Neurones located in deeper laminae of the spinal cord (VII-VIII and X), also project to the SRD; these laminae also contain nocicep-
100
rive units [8, 14]. Many spinal neurones exhibit cutaneovisceral convergence [8, 14] and a subpopulation of SRD units encode both cutaneous and visceral noxious stimuli [221 In contrast with our results, a previous study of spinal
afferents to the caudal BRF described retrogradely labelled cells predominantly in laminae I and X [16]. However, the injections were made more rostrally (4.35.3 mm caudal to the interaural line) [20]. In addition, it is possible that some of the injection sites would have
cervical (C3)
cervical (C5)
)
mid. thoracic
;°
( lumbar (L3)
J
upp. sacral
Fig. 2. Series of camera lucida drawings of coronal sections in a case where the W G A - a p o H R P Au complex was injected within the SRD (A) or the SRV (B). Injection sites consisted of a dense core (black) surrounded by a halo filled with neuronal labelled elements (stipple). Diagrams of the spinal cord illustrate the total number of cells (dots) contained in 5 consecutive 4 0 / t m thick sections. In both cases, the largest number of labelled neurones was found in the cervical cord. However, there was one principal difference in respect of their distributions within the grey matter: the largest number of labelled cells was in the dorsal horn for injections in the SRD and in the ventral horn for injections in the SRV (see text).
101
included areas containing neuronal populations which are functionally different from SRD neurones: Lima [16] reported that lesions of the dorsal columns strongly reduced the number of labelled cells, especially in the superficial dorsal horn, whereas we have shown electrophysiologically that lesions of the dorsal columns do not affect the spinal input to the SRD [6]. In fact the ascending spinal pathways which are responsible for activating SRD neurones are crossed and confined to the lateral parts of the ventral quadrant [6]. It has been demonstrated using degeneration techniques following ventrolateral cordotomy in the rat [24] that a large number of fibres travelling in the ventrolateral quadrant terminate in the SRD. One intriguing question arises from the finding that the largest number of afferents to the SRD originates from the ipsilateral cervical cord: to date, all neurones recorded in this structure have shown a 'whole body receptive field' with a contralateral dominance [27]. This apparent discrepancy could be due to the transport of tracer to the adjacent cervical cord being easier than to more caudal areas. However, this possibility seems unlikely since in every case, the potency of labeling following injections within the SRD was: cervical > sacral > thoracic = lumbar segments. Interestingly, the largest numbers of retrogradely labelled cells in the spino-thalamic (STT) and spino-mesencephalic tracts (SMT) in the rat were also found in the upper cervical cord [12, 30]. Is this a common functional organisation of ascending somatosensory pathways? One could imagine the upper cervical cord acting as a somatosensory relay between caudal areas of the spinal cord and higher centres. Recent electrophysiological reports of neurones in the upper cervical spinal cord with widespread receptive fields including the oro-facial region, the hindpaws and the tail support such a possibility in the cat and the monkey [13, 23, 29]. As previously suggested [29], inputs to the cervical enlargement can originate from different sources, including from collaterals of ascending axons. Within the framework of this hypothesis, one could envisage that at least some inputs to SRD neurones have relays in the cervical cord. Together with the fact that other tracts involved in the transmission of nociceptive information may have a similar organisation, this could explain the widespread relief of pain, including pain from caudal segments of the body, following commissural myelotomies of the upper cervical spinal cord in humans (see refs. in ref. 9). Another interesting finding is that most of the laminae which contain spinal afferents to the SRD have been shown by the PHA-L method [3] to receive dense efferent inputs from the same region. This is especially true for laminae V, VI, VII and X. In terms of density of
efferent projections from the SRD, an identical rank of potencies was found for the different levels of the spinal cord, i.e. is very dense at the cervical level, moderate in the thoracic and lumbar cords and intermediate at the sacral level [3]. Such reciprocal connections strengthen the suggestion that SRD neurones may belong to spinoreticulo-spinal loops implicated in the processing of nociceptice information via feed-back or more diffuse mechanisms [15, 27]. The authors are grateful to Dr. S.W. Cadden for advice in the preparation of the manuscript. This work was supported by l'Institut National de la Sant6 et de la Recherche Mrdicale (INSERM) and la Direction de Recherches et Etudes Techniques (DRET). 1 Aicher, S.A. and Randich, A., Antinociception and cardiovascular responses produced by electrical stimulation in the nucleus tractus solitarius, nucleus reticularis ventralis, and the caudal medulla, Pain, 42 (1990) 103-119. 2 Basbaum, A. and Menrtrey, D., Wheat germ agglutinin-apoHRP gold: a new retrograde tracer for light- and electron-microscopic single- and double-label studies, J. Comp. Neurol., 261 (1987) 306318. 3 Bernard, J.F., Villanueva, L., Carrour, J. and Le Bars, D., Efferent projections from the subnucleus reticularis dorsalis (SRD): a Phaseolus vulgaris leucoagglutinin study in the rat, Neurosci. Lett., 116 (1990) 257-262. 4 Besson, J.M. and Chaouch, A., Peripheral and spinal mechanisms of nociception, Physiol. Rev., 67 (1987) 67-186. 5 Bing, Z., Villanueva, L. and Le Bars, D., Effects of systemic morphine upon A~- and C-fibre evoked activities of subnucleus reticularis dorsalis neurones in the rat medulla, Eur. J. Pharmacol., 164 (1989) 85-92. 6 Bing, Z., Villanueva, L. and Le Bars, D., Ascending pathways in the spinal cord involved in the activation of subnucleus reticularis dorsalis neurons in the medulla of the rat, J. Neurophysiol., 63 (1990) 424-438. 7 Bowsher, D., Role of the reticular formation in responses to noxious stimulation, Pain, 2 (1976) 361 378. 8 Cervero, F. and Tattersall, J.E.H., Somatic and visceral sensory integration in the thoracic spinal cord. In F. Cervero and J.F.B. Morrison (Eds.), Visceral sensation, Progress in Brain Research, Vol. 67, Elsevier, 1986, pp. 189-205. 9 Cook, A.W., Nathan, P.W. and Smith, M.C., Sensory consequences of commissural myelotomy, Brain, 107 (1984) 547-568. 10 De Pommery, J., Roudier, F. and Men&rey, D., Postsynaptic fibers reaching the dorsal column nuclei in the rat, Neurosci. Lett., 50 (1984) 319-323. I1 Giesler, G.J,, Nahin, R.L. and Madsen, A., Postsynaptic dorsal column pathway of the rat. I. Anatomical studies, J. Neurophysiol., 51 (1984) 260-275. 12 Granum, S.L., The spinothalamic system of the rat, I. Locations of cells of origin, J. Comp. Neurol., 247 (1986) 159-180. 13 Hodge, C.J., Apakarian, V.A., Gingold, S. and Stevens, R.T., Spinothalamic tract cells of the high cervical spinal cord of primate, Pain, Suppl. 5 (1990) $98. 14 Honda, C., Visceral and somatic afferent convergence onto neurones near the central canal in the sacral spinal cord of the cat, J. Neurophysiol., 53 (1985) 105%1078.
102 15 Le Bars D. and Villanueva, L., Electrophysiological evidence for the activation of descending inhibitory controls by nociceptive afferent pathways. In H.L. Fields and J.M. Besson (Eds.), Pain Modulation, Progress in Brain Research, Vol. 77, Elsevier, 1988, pp. 275-299. 16 Lima, D., A spinomedullary projection terminating in the dorsal reticular nucleus of the rat, Neuroscience, 34 (1990) 577-590. 17 Men6trey, D. and Basbaum, A., Spinal and trigeminal projections to the nucleus of the solitary tract: a possible substrate for somatovisceral and viscerovisceral reflex activation, J. Comp. Neurol., 255 (1987) 439-450. 18 Men6trey, D., Roudier, F. and Besson, J.M., Spinal neurons reaching the lateral reticular nucleus as studied in the rat by retrograde transport of horseradish peroxidase, J. Comp. Neurol., 220 (1983) 439-452. 19 Newman, D.B., Distinguishing rat brainstem reticulospinal nuclei by their neuronal morphology. I. Medullary nuclei, J. Hirnforsch., 26 (1985) 187-226. 20 Paxinos, G. and Watson, C., The Rat Brain in Stereotaxic Coordinates, Academic Press, San Diego, 1986. 21 Porro, C.A., Cavazzuti, M., Galetti, A. and Sassatelli, L., Functional activity mapping of the rat brainstem during formalininduced noxious stimulation, Neuroscience, 41 (1991) 667q580. 22 Roy, J.C., Bing, Z., Villanueva, L. and Le Bars, D., Activation of Subnucleus Reticularis Dorsalis neurones (SRD) of the rat medulla by colorectal distention, Pain, Suppl. 5 (1990) 166. 23 Smith, M.V., Apkarian, A.V. and Hodge, C.J. Jr., Somatosensory response properties of contralaterally projecting spinothalamic and
24
25
26
27
28
29
30
nonspinothalamic neurons in the second cervical segment of the cat, J. Neurophysiol., 66 (1991) 83-102. Torvik, A., Afferent connections to the sensory trigeminal nuclei, the nucleus of the solitary tract and adjacent structures. An experimental study in the rat, J. Comp. Neurol., 106 (1956) 51 141. Valverde, F., Reticular formation of the albino rat brain stem cytoarchitecture and corticofugal connections, J. Comp. Neurol., 119 (1962) 25-49. Villanueva, L., Bing, Z., Bouhassira, D. and Le Bars, D., Encoding of electrical, thermal, and mechanical noxious stimuli by subnucleus reticularis dorsalis neurons in the rat, J. Neurophysiol., 61 (1989) 391-402. Villanueva, L., Bouhassira, D., Bing, Z. and Le Bars, D., Convergence of heterotopic nociceptive information onto subnucleus reticularis dorsalis neurons in the rat medulla, J. Neurophysiol., 60 (1988) 980-1009. Villanueva, L., Cliffer, K.D., Sorkin, L., Le Bars. D and Willis, W.D., Convergence of heterotopic nociceptive information onto neurons of the caudal medullary reticular formation in the monkey (Macacafascicularis), J. Neurophysiol., 63 (1990) 1118-1127. Yeziersky, R.P. and Broton, J.G., Functional properties of spinomesencephalic tract (SMT) cells in the upper cervical spinal cord of the cat, Pain, 45 (1991) 187-196. Yeziersky, R.P. and Mendez, C.M., Spinal distribution and collateral projections of rat spinomesencephalic tract cells, Neuroscience, 44 (1991) 113-130.
