- Email: [email protected]
Lamina I-periaqueductal gray (PAG) projections represent only a limited part of the total spinal and caudal medullary input to the PAG in the cat
Brain Research Bulletin, Vol. 54, No. 2, pp. 167–174, 2001 Copyright © 2001 Elsevier Science Inc. Printed in the USA. All rights reserved 0361-9230/01/$–see front matter
PII S0361-9230(00)00442-1
Lamina I-periaqueductal gray (PAG) projections represent only a limited part of the total spinal and caudal medullary input to the PAG in the cat Leonora J. Mouton,* Esther-Marije Klop and Gert Holstege Department of Anatomy and Embryology, Faculty of Medical Sciences, Rijksuniversiteit Groningen, Groningen, The Netherlands [Received 29 May 2000; Revised 15 September 2000; Accepted 19 October 2000] ABSTRACT: The periaqueductal gray is well known for its involvement in nociception control, but it also plays an important role in the emotional motor system. To accomplish these functions the periaqueductal gray receives input from the limbic system and from the caudal brainstem and spinal cord. Earlier studies gave the impression that the majority of the periaqueductal gray projecting cells in caudal brainstem and spinal cord are located in the contralateral lamina I, which is involved in nociception. The present study in the cat, however, demonstrates that of all periaqueductal gray projecting neurons in the contralateral caudal medulla less than 7% was located in lamina I. Of the spinal periaqueductal gray projecting neurons less than 29% was located in lamina I. However, within the spinal cord large segmental differences exist: in few segments of the enlargements the lamina I-periaqueductal gray projecting neurons represent a majority. In conclusion, although the lamina I-periaqueductal gray projection is a very important nociceptive pathway, it constitutes only a limited part of the total projection from the caudal medulla and spinal cord to the periaqueductal gray. These results suggest that a large portion of the medulloand spino-periaqueductal gray pathways conveys information other than nociception. © 2001 Elsevier Science Inc.
PAG receives a major afferent input from more rostrally located limbic structures such as the hypothalamus, the central nucleus of the amygdala and the medial preoptic area [15,38]. Besides this, it has been demonstrated that the PAG receives direct input from the caudal medulla and spinal cord [13,27,30,32,43]. Concerning this spinal and medullary projections to the PAG, still no insight exists about its precise organization. Most attention until now has been given to the many cells located in lamina I of the caudal medulla [12] and lamina I of the spinal cord [36] that project to the ventral and ventrolateral PAG [6,11,21,24,25,28,32, 33,40 – 42]. This ascending lamina I-PAG projection is most likely involved in informing the PAG about the pain situation of the body, as lamina I is known for its almost exclusive involvement in nociceptive functions [5]. However, besides these many lamina I afferents probably involved in nociception, the PAG also receives other caudal medullary and spinal afferents; many cells located in other laminae than lamina I in both the caudal medulla [4,13,31] and spinal cord [20,21,28,32,34,39,43] were found to project to the PAG. It is most likely that these other afferents relay other information to the PAG than nociception and might be important for the PAG to control the emotional motor behaviors. Until now relatively little attention has been given to these other spinal and medullary projections to the PAG. The question is why so little attention has been paid to these other medullary and spinal PAG projecting cells. It might be that it derives from the fact that some well-known studies [6,41,42] on the caudal medullary and spinal afferents to the PAG in cat and monkey concluded that tracer injections in the dorsal part of the midbrain, including the PAG, resulted in an overwhelming majority of labeled spino-mesencephalic cells in lamina I. Such results might suggest a smaller importance of the other PAG projections. However, doubts have arisen as to whether the lamina I-PAG projections really form the majority of the medullary and spinal PAG projecting neurons, considering the results of some more recent studies [20,21]. Keay and Bandler [20], in a study on the upper cervical cord projections to the PAG in the cat, found that the majority of labeled neurons was located in lamina I, but only when the tracer injections were placed in the lateral PAG. When
KEY WORDS: Spinomesencephalic, Trigeminal, Nucleus caudalis, Marginal layer, Nociception.
INTRODUCTION The periaqueductal gray (PAG) is well known for its role in nociception control. Stimulation of the PAG results in a strong decrease of nociception in all parts of the body [5,23,29,35]. However, the PAG also plays an important role in the control of emotional motor behaviors necessary for the survival of the individual and species [16]. PAG-controlled emotional motor behaviors are, e.g., defence behaviors, such as fight, threat display, flight, and immobility [1,2,3,10], but also mating behavior [17,37], micturition [7,8,17], vocalization [2,14,18,19], and maternal behavior [25,26]. To control nociception and the emotional motor behaviors, the PAG needs to be informed about the internal and external environment of the individual. Anatomical studies have shown that the
* Address for correspondence: Leonora J. Mouton, Department of Anatomy and Embryology, Faculty of Medical Sciences, Rijksuniversiteit Groningen, Groningen, The Netherlands. Fax: 31-50-3632461; E-mail: [email protected]
167
168
MOUTON, KLOP AND HOLSTEGE
FIG. 1. Photographs and drawings showing the wheat germ agglutinin-conjugated horseradish peroxidase injection sites of three cases. On top: drawings of the injection sites in gray. The core of the injections is indicated in black. Bottom: brightfield photomicrographs of the injection areas in 40 m thick sections, visualized with the diamino-benzidine method. Scale bar: 1000 m.
FIG. 2. Schematic drawings of the rostrocaudal location of the injection sites in the periaqueductal gray. An indication of what is considered the core of the injections is shown as the black area.
LAMINA I-PAG REPRESENTS ONLY A LIMITED PART
169
FIG. 3. Darkfield polarized photomicrographs (Zeis Axiocam, extended focus) of labeled lamina I cells in the caudal medulla and various spinal segments, after wheat germ agglutinin-conjugated horseradish peroxidase injections in the dorsolateral, lateral, and ventrolateral parts of the intermediate and caudal periaqueductal gray (case 2385). Scale bar: 100 m.
the tracer was injected in the ventrolateral PAG the majority of labeled neurons were located in laminae VII and VIII. Furthermore, Keay et al. [21], in a study on all spino-PAG projections in the rat, reported that only 26 –28% of all spino-PAG projecting neurons were located in lamina I after injections in the lateral or ventrolateral PAG, respectively. So, in can be concluded that there is no good insight on which portion of the caudal medullary and spinal PAG projections originates in lamina I. The present retrograde tracing study in the cat is an attempt to solve these problems. Injections were made in different parts of the lateral and ventrolateral PAG, and the number of labeled cells in lamina I in both the medulla caudal to the obex and in each of the C1-Coc2 spinal cord segments was determined. These numbers were compared with the total numbers of labeled cells in all other laminae of the caudal medulla and in each of the spinal segments, resulting in a better insight in the percentages of lamina I-PAG projecting cells in the caudal medulla and spinal cord. A preliminary report of these results has been published as an abstract [22].
MATERIALS AND METHODS Surgical Procedures A total of five cats was used. The surgical procedures, pre- and postoperative care, handling, and housing of the animals were in accordance with protocols approved by the Committee of Animal Experiments of the Faculty of Medicine of the University of Groningen. For surgery the animals were initially anesthetized with ketamine (Nimatek, 0.1 ml/kg, intramuscular [i.m.]) and xylazine (Sedamun, 0.1 ml/kg i.m.), after which they were kept anesthetized by ventilation with a mixture of O2, N2O, and halothane. During surgery, heart rate and body temperature were monitored. Following a survival time of 3 days the animals were initially anesthetized with ketamine and xylazine i.m., followed by an overdose of 6% pentobarbital sodium (Nembutal; intraperitoneal [i.p.]). The cats were perfused transcardially with 2 l of heparinized saline at 37°C, directly followed by 2 l of 0.1 M phosphate buffer (pH 7.4), containing 4% sucrose, 1% paraformaldehyde, and 2% glutaraldehyde.
170
MOUTON, KLOP AND HOLSTEGE
FIG. 4. Schematic drawings of the labeled neurons in the caudal medulla and various spinal segments after a wheat germ agglutinin-conjugated horseradish peroxidase injection in the lateral intermediate and caudal periaqueductal gray (case 2300). Note that each drawing of a segment represents the labeled neurons in six 40-m thick sections.
Experimental Procedures In two female cats, three injections, each containing approximately 30 nl 2.5% wheat germ agglutinin-conjugated horseradish peroxidase (WGA-HRP), were placed in the lateral and ventrolateral parts of the rostral (case 2367) or intermediate and caudal PAG (case 2385). In three other female cats, one injection of approximately 20 –50 nl 2.5% WGA-HRP was placed in different parts of the lateral and/or ventrolateral intermediate and caudal PAG (case 2401: lateral part of the ventrolateral and lateral PAG, case 2390: lateral PAG, case 2300: central part of the lateral PAG). After perfusion, the brain and spinal cord were removed, postfixed for 2 h and stored overnight in 20% sucrose in phosphate buffer at 4°C. Subsequently, in each case the brainstem and all spinal segments (C1-Coc2) were cut into 40-m frozen transverse sections. Every fourth section was incubated according to the tetramethyl-benzidine method (TMB), dehydrated, and coverslipped. Of the area containing the injection site an extra series of sections was processed with diamino-benzidine (DAB). In each TMB processed transverse section of the spinal cord and of the medulla caudal to the obex, the retrogradely labeled profiles were plotted and counted. For counting the labeled cells in lamina I, lamina I was defined according to Rexed for the spinal cord [36] and according to Gobel for the caudal medulla [12]. It was assumed that each labeled profile consisting of at least part of a labeled cell body and part of a labeled axon or dendrite represents one labeled neuron. It must be emphasized that the labeled neurons in the dorsal column nuclei, solitary nucleus and lateral cervical nucleus were not included in the counts.
RESULTS Figure 1 contains photomicrographs containing the injection sites in cases 2385, 2401, and 2300, as well as the corresponding schematic drawings. In the drawings, the core of the injections, as observed with brightfield illumination of the DAB stained sections, is shown in black. A schematic representation of the rostrocaudal extent of all injection sites is given in Fig. 2. In case 2367, the large injection site involves the dorsolateral, lateral, and ventrolateral parts of the rostral PAG. In case 2385, the large injection site involves the dorsolateral, lateral, and ventrolateral parts of the intermediate and caudal PAG. The other, smaller injection sites each involve the lateral intermediate and caudal PAG. Figure 3 contains photomicrographs of lamina I cells in the caudal medulla and spinal cord (case 2385). Figure 4 shows schematic drawings of the labeled lamina I cells found in the caudal medulla and various spinal segments, as well as the labeled cells found in the other laminae (case 2300). Figure 5 shows the numbers of labeled lamina I cells and the numbers of all other labeled cells per segment per case. Surprisingly, the results show that of all counted labeled cells in both the caudal medulla and entire spinal cord in the five different cases only 1–9% of the ipsilaterally, and 12–24% of the contralaterally located labeled cells were located in lamina I (Table 1 and Fig. 6). Also, when taking the results of the caudal medulla and spinal cord separate, low percentages of lamina I cells were found; in the caudal medulla in all cases less than 4% of all labeled cells located ipsilaterally to the injection sites was found in lamina I, while for the contralaterally located cells this was less than 7% (Table 1). In the spinal cord in all cases less
LAMINA I-PAG REPRESENTS ONLY A LIMITED PART
171
FIG. 5. Histograms of the numbers of labeled cells in the contralateral caudal medulla and spinal cord (counted in a 1:4 series of 40-m thick sections) after wheat germ agglutinin-conjugated horseradish peroxidase injections in different parts of the lateral periaqueductal gray. The numbers of labeled cells in lamina I of the medulla caudal to the obex and in each C1-Coc2 spinal cord segment are shown in black. The total numbers of labeled cells in the caudal medulla and all spinal cord segments are shown in gray. Note that the labeled cells in the dorsal column nuclei, the solitary nucleus, and the lateral cervical nucleus were not included. CMed, medulla caudal to the obex.
than 14% of all ipsilaterally labeled cells was located in lamina I, and of all contralaterally located labeled cells this was less than 29% (Table 1). Within the spinal cord different percentages of labeled lamina I were found per segment and per case; among the segments the percentages of labeled cells in lamina I varied ipsilaterally from 0 – 67% per segment, and contralaterally from 0 –78% per segment (Table 1). Per case, in only one or two segments more than 50% of the labeled cells were found in lamina I, mainly on the contralateral side. These segments were always in the middle of the enlargements, i.e., C7, C8, T1, or L6. In case 2300,
with an injection involving the central portion of the lateral part of the intermediate and caudal PAG, the number of segments in which a majority of labeled cells was located in lamina I was slightly higher than in the other cases. In this case a majority of lamina I cells was found in C7 and C8 contralaterally, L7 bilaterally, and S1 and S2 ipsilaterally. Averaged percentages of labeled lamina I cells over the five cases show that only in C7 more than 50% of all labeled cells were present in lamina I, while in most other segments less than 20% of the ipsilaterally, and less than 30% of the contralaterally located labeled cells was found in lamina I (Fig. 7).
172
MOUTON, KLOP AND HOLSTEGE TABLE 1 PERCENTAGES OF LABELED CELLS IN LAMINA I
Caudal medulla and spinal cord Caudal medulla Spinal cord In total Per segment
Ipsilateral (%)
Contralateral (%)
1–9 1–3
12–24 1–6
1–14 0–67
15–28 0–78
DISCUSSION Total Numbers of Spino-PAG and Caudal Medulla-PAG Neurons A reliable comparison between the numbers of labeled cells observed in the present study and the numbers found in studies of other authors [6,20,25,28,32,40 – 42] is difficult, because in all previous studies the numbers are much less precise. First, in these studies the numbers of cells are often the result of large injections, including the PAG, but also involving very large parts of the adjoining mesencephalic regions. Second, these studies only present numbers that are an average of many different segments and/or of several cases, each with different injection sites. Percentages of Lamina I-PAG Neurons The present results show that, averaged over the entire caudal medulla and spinal cord, only a limited part of the PAG projecting neurons is located in lamina I. However, as the present study, for the first time, studied the caudal medulla and each spinal segment separately, results also give evidence that large differences exist between the different segments (Figs. 5 and 7). These large segmental differences are most probably the reason that until now no good insight existed on which portion of the spinal projections to the PAG originates in lamina I. In the earlier studies on the spino-PAG afferents [6,20,21,42] the numbers of lamina I cells
were only counted in a few spinal segments and/or were derived from countings in spinal blocks consisting of several different segments. When counting lamina I-PAG cells in only a few segments, the choice of segments is crucial for the result: taking a segment of the enlargement might give the conclusion of a majority of lamina-I PAG cells, while, e.g., taking a segment of the upper cervical cord might not lead to this conclusion. When counting lamina I-PAG cells in larger blocks consisting of several segments, it is crucial which segments are selected to form one block: a block consisting of for instance L6, L7 and S1 might give the conclusion of a majority of lamina I-PAG cells, while a block of L7, S1 and S2 might not (Fig. 7). Percentages of Lamina I-PAG Neurons in the Lumbosacral Cord In case 2300, with an injection involving the central portion of the lateral part of the intermediate and caudal PAG, the number of segments in which a majority of labeled cells was located in lamina I was slightly higher than in the other cases. Regarding the relatively many labeled cells in lamina I of the lumbosacral cord in this case, it has to be emphasized that part of these lumbosacral lamina I cells probably belong to a distinct cell group located in the L6-S3 segments. An earlier study of our laboratory [39] showed the existence of such a distinct cell group located in the lateral part of lamina I of L6 –S2 and in laminae V–VII of S2. These cells project to the central part of the lateral and ventrolateral caudal and intermediate PAG [9,39]. In all likelihood, this cell group is involved in the relay to the PAG of information concerning micturition and mating behavior. It implies that the lamina I cells located in L6, L7 and S1 that belong to this specific group are probably not involved in nociception. Conclusion There is no doubt that the lamina I-PAG projection is a very important pathway involved in nociception, also because it involves more lamina I neurons than the spinothalamic tract [33]. The most important conclusion is, however, that lamina I neurons are only a minority of all spinal and caudal medullary neurons
FIG. 6. Histograms showing the total numbers of ipsi- and contralaterally located labeled cells as well as the numbers of labeled cells in lamina I in the caudal medulla and entire spinal cord, after injections in different parts of the lateral periaqueductal gray. The percentages of labeled lamina I cells ipsi- and contralaterally are indicated. Labeled cells were counted in a 1:4 series of 40-m thick sections. The labeled cells in the dorsal column nuclei, the solitary nucleus, and the lateral cervical nucleus were not included.
LAMINA I-PAG REPRESENTS ONLY A LIMITED PART
173
FIG. 7. Averaged results over the five cases: averaged percentages (and SD) of the labeled cells in the ipsi- and contralateral lamina I in the medulla caudal to the obex and in each C1-Coc2 spinal cord segment, after wheat germ agglutinin-conjugated horseradish peroxidase injections in different parts of the lateral periaqueductal gray. Labeled cells were counted in a 1:4 series of 40-m thick sections. Note that the labeled cells in the dorsal column nuclei, the solitary nucleus, and the lateral cervical nucleus were not included. CMed, medulla caudal to the obex.
projecting to the PAG. Although part of the non lamina I-PAG projecting cells are probably involved in the relay of nociceptive information also, clearly many of them relay other information to the PAG. Examples are bladder filling and information related to mating behavior, but there are many other possible, yet unknown, sorts of information the PAG needs for its strong control of emotional behavior.
10.
11.
ACKNOWLEDGEMENTS
The authors thank Mr. K. van Linschoten and Mrs. E. Meyer for their histotechnical help.
REFERENCES 1. Bandler, R.; Carrive, P.; Zhang, S. P. Integration of somatic and autonomic reactions within the midbrain periaqueductal grey: Viscerotopic, somatopic and functional organization. Brain Res. 87:269 – 305; 1991. 2. Bandler, R.; Depaulis, A. Midbrain periaqueductal gray control of defensive behavior in the cat and the rat. In: Depaulis, A.; Bandler, R., eds. The midbrain periaqueductal gray matter: Functional, anatomical, and neurochemical organization. New York: Plenum Press; 1991:175– 198. 3. Bandler, R.; Keay, K. A. Columnar organization in the midbrain periaqueductal gray and the integration of emotional expression. Prog. Brain Res. 107:285–300; 1996. 4. Beitz, A. J. The organization of afferent projections to the midbrain periaqueductal gray of the rat. Neuroscience 7:133–159; 1982. 5. Besson, J. M.; Chaouch, A. Peripheral and spinal mechanisms of nociception. Physiol. Rev. 67:67–186; 1987. 6. Blomqvist, A.; Craig, A. D. Organization of spinal and trigeminal input to the PAG. In: Depaulis, A.; Bandler, R., eds. The midbrain periaqueductal gray matter. New York: Plenum Press; 1991:345–363. 7. Blok, B. F. M.; Holstege, G. The central nervous system control of micturition in cats and humans. Behav. Brain Res. 92:119 –125; 1998. 8. Blok, B. F. M.; Sturms, L.; Holstege, G. Brain activation during micturition in women. Brain 121:2033–2042; 1998. 9. Blok, B. F.; De Weerd, H.; Holstege, G. Ultrastructural evidence for a
12.
13.
14. 15.
16.
17.
18. 19. 20.
21.
paucity of projections from the lumbosacral cord to the pontine micturition center or M-region in the cat: A new concept for the organization of the micturition reflex with the periaqueductal gray as central relay. J. Comp. Neurol. 359:300 –309; 1995. Carrive, P. The periaqueductal gray and defensive behavior: Functional representation and neuronal organization. Behav. Brain Res. 58:27– 47; 1993. Craig, A. D. Distribution of brainstem projections from the spinal lamina I neurons in the cat and the monkey. J. Comp. Neurol. 361: 225–248; 1995. Gobel, S.; Hockfield, S. The division of the dorsal and ventral horns of the mammalian caudal medulla into eight layers using anatomical criteria. In: Anderson, D. J.; Matthews, B., eds. Pain in the trigeminal region. Amsterdam: Elsevier; 1977:443– 453. Herbert, H.; Saper, C. B. Organization of medullary adrenergic and noradrenergic projections to the periaqueductal matter in the rat. J. Comp. Neurol. 315:23–52; 1992. Holstege, G. Anatomical study of the final common pathway for vocalization in the cat. J. Comp. Neurol. 284:242–252; 1989. Holstege, G. Descending motor pathways and the spinal motor system: Limbic and non-limbic components. Prog. Brain Res. 87:307– 412; 1991. Holstege, G. Emotional motor system. In: Dulbeco R., ed. Encyclopedia of human biology, 2nd ed., vol. 3. New York: Academic Press; 1997:643– 660. Holstege, G. The emotional motor system in relation to the supraspinal control of micturition and mating behavior. Behav. Brain Res. 92:103– 109; 1998. Ju¨rgens, U.; Pratt, R. Role of the periaqueductal grey in vocal expression of emotion. Brain Res. 167:367–378; 1979. Kanai, T.; Wang, S. C. Localization of the central vocalization mechanism in the brainstem of the cat. Exp. Neurol. 6:426 – 434; 1962. Keay, K. A.; Bandler, R. Anatomical evidence for segregated input from the upper cervical spinal cord to functionally distinct regions of the periaqueductal gray region of the cat. Neurosci. Lett. 139:143–148; 1992. Keay, K. A.; Feil, K.; Gordon, B. D.; Herbert, H.; Bandler, R. Spinal afferents to functionally distinct periaqueductal gray columns in the rat: An anterograde and retrograde tracing study. J. Comp. Neurol. 385:207–229; 1997.
174 22. Klop, E. M.; Mouton, L. J.; Holstege, G. Only a small portion of the caudal medullary and spinal projections to the PAG originate in lamina I. Soc. Neurosci. 25:146; 1999. 23. Liebeskind, J. C.; Guilbaud, G.; Besson, J. M.; Oliveras, J. L. Analgesia from electrical stimulation of the periaqueductal gray matter in the cat: Behavioral observations and inhibitory effects on spinal cord interneurons. Brain Res. 50:441– 446; 1973. 24. Lima, D.; Coimbra, A. Morphological types of spinomesencephalic neurons in the marginal zone lamina I. of the rat spinal cord, as shown after retrograde labeling with cholera toxin subunit B. J. Comp. Neurol. 279:327–339; 1989. 25. Liu, R. P. Laminar origins of spinal projection neurons to the periaqueductal gray of the rat. Brain Res. 264:118 –122; 1983. 26. Lonstein, J. L.; Stern, J. M. Role of the midbrain periaqueductal gray in maturnal nurturance and aggression: c-fos and electrolytic lesion studies in lactating rats. J. Neurosci. 17:3364 –3378; 1997. 27. Lonstein, J. S.; Simmons, D. A.; Stern, J. N. Functions of the caudal periaqueductal gray in lactating rats: Kyphosis, lordosis, maternal aggression, and fearfulness. Behav. Neurosci. 112:1502–1518; 1998. 28. Mantyh, P. W. The ascending input to the midbrain periaqueductal gray of the primate. J. Comp. Neurol. 211:50 – 64; 1982. 29. Mayer, J. D.; Wolfte, T. L.; Akil, H.; Carder, B.; Liebeskind, J. C. Analgesia from electrical stimulation in the brainstem of the cat. Science 174:1351–1354; 1971. 30. Mehler, W. Some neurological species differences—A posteriori. Ann. N.Y. Acad. Sci. 167:424 – 468; 1969. 31. Meller, S. T.; Dennis, B. J. Afferent projections to the periaqueductal gray in the rabbit. Neuroscience 19:927–964; 1986. 32. Menetrey, D.; Chaouch, A.; Binder, D.; Besson, J. M. The origin of the spinomesencephalic tract in the rat: An anatomical study using the retrograde transport of horseradish peroxidase. J. Comp. Neurol. 206: 193–207; 1982. 33. Mouton, L. J.; Holstege, G. Three times as many lamina I neurons
MOUTON, KLOP AND HOLSTEGE
34.
35. 36. 37. 38.
39. 40. 41. 42. 43.
project to the periaqueductal gray than to the thalamus: A retrograde tracing study in the cat. Neurosci. Lett. 255:107–110; 1998. Mouton, L. J.; Holstege, G. The segmental and laminar organization of the spinal neurons projecting to the periaqueductal gray (PAG) in the cat suggests the existence of at least five separate clusters of spinoPAG neurons. J. Comp. Neurol. 428:389 – 410; 2000. Reynolds, E. S. The use of lead citrate at high pH as an electronopaque stain in electron-microscopy. J. Cell. Biol. 17:208 –212; 1963. Rexed, B. A cytoarchitectonic atlas of the spinal cord in the cat. J. Comp. Neurol. 100:297–380; 1954. Sakuma, Y.; Pfaff, D. W. Facilitation of female reproductive behavior from mesencephalic central gray in the rat. Am. J. Physiol. 237:R278 – R284; 1979. Shipley, M. T.; Ennis, M.; Rizvi, T. A.; Behbehani, M. M. Topographical specificity of forebrain inputs to the midbrain periaqueductal gray: Evidence for discrete longitudinally organized input columns. In: Depaulis, A.; Bandler, R., eds. The midbrain periaqueductal gray matter: Functional, anatomical, and neurochemical organization. New York: Plenum Press; 1991:417– 448. VanderHorst, V. G. J. M.; Mouton, L. J.; Blok, B. F. M.; Holstege, G. Distinct cell groups in the lumbosacral cord of the cat project to different areas in the periaqueductal gray. J. Comp. Neurol. 376:361–385; 1996. Wiberg, M.; Blomqvist, A. The spinomesencephalic tract in the cat: Its cells of origin and termination pattern as demonstrated by the intraaxonal transport method. Brain Res. 291:1–18; 1984. Wiberg, M.; Westman, J.; Blomqvist, A. The projection to the mesencephalon from the sensory trigeminal nuclei. An anatomical study in the cat. Brain Res. 399:51– 68; 1986. Wiberg, M.; Westman, J.; Blomqvist, A. Somatosensory projection to the mesencephalon: An anatomical study in the monkey. J. Comp. Neurol. 264:92–117; 1987. Yezierski, R. P.; Mendez, C. M. Spinal distribution and collateral projections of rat spinomesencephalic tract cells. Neuroscience 44: 113–130; 1991.
PII S0361-9230(00)00442-1
Lamina I-periaqueductal gray (PAG) projections represent only a limited part of the total spinal and caudal medullary input to the PAG in the cat Leonora J. Mouton,* Esther-Marije Klop and Gert Holstege Department of Anatomy and Embryology, Faculty of Medical Sciences, Rijksuniversiteit Groningen, Groningen, The Netherlands [Received 29 May 2000; Revised 15 September 2000; Accepted 19 October 2000] ABSTRACT: The periaqueductal gray is well known for its involvement in nociception control, but it also plays an important role in the emotional motor system. To accomplish these functions the periaqueductal gray receives input from the limbic system and from the caudal brainstem and spinal cord. Earlier studies gave the impression that the majority of the periaqueductal gray projecting cells in caudal brainstem and spinal cord are located in the contralateral lamina I, which is involved in nociception. The present study in the cat, however, demonstrates that of all periaqueductal gray projecting neurons in the contralateral caudal medulla less than 7% was located in lamina I. Of the spinal periaqueductal gray projecting neurons less than 29% was located in lamina I. However, within the spinal cord large segmental differences exist: in few segments of the enlargements the lamina I-periaqueductal gray projecting neurons represent a majority. In conclusion, although the lamina I-periaqueductal gray projection is a very important nociceptive pathway, it constitutes only a limited part of the total projection from the caudal medulla and spinal cord to the periaqueductal gray. These results suggest that a large portion of the medulloand spino-periaqueductal gray pathways conveys information other than nociception. © 2001 Elsevier Science Inc.
PAG receives a major afferent input from more rostrally located limbic structures such as the hypothalamus, the central nucleus of the amygdala and the medial preoptic area [15,38]. Besides this, it has been demonstrated that the PAG receives direct input from the caudal medulla and spinal cord [13,27,30,32,43]. Concerning this spinal and medullary projections to the PAG, still no insight exists about its precise organization. Most attention until now has been given to the many cells located in lamina I of the caudal medulla [12] and lamina I of the spinal cord [36] that project to the ventral and ventrolateral PAG [6,11,21,24,25,28,32, 33,40 – 42]. This ascending lamina I-PAG projection is most likely involved in informing the PAG about the pain situation of the body, as lamina I is known for its almost exclusive involvement in nociceptive functions [5]. However, besides these many lamina I afferents probably involved in nociception, the PAG also receives other caudal medullary and spinal afferents; many cells located in other laminae than lamina I in both the caudal medulla [4,13,31] and spinal cord [20,21,28,32,34,39,43] were found to project to the PAG. It is most likely that these other afferents relay other information to the PAG than nociception and might be important for the PAG to control the emotional motor behaviors. Until now relatively little attention has been given to these other spinal and medullary projections to the PAG. The question is why so little attention has been paid to these other medullary and spinal PAG projecting cells. It might be that it derives from the fact that some well-known studies [6,41,42] on the caudal medullary and spinal afferents to the PAG in cat and monkey concluded that tracer injections in the dorsal part of the midbrain, including the PAG, resulted in an overwhelming majority of labeled spino-mesencephalic cells in lamina I. Such results might suggest a smaller importance of the other PAG projections. However, doubts have arisen as to whether the lamina I-PAG projections really form the majority of the medullary and spinal PAG projecting neurons, considering the results of some more recent studies [20,21]. Keay and Bandler [20], in a study on the upper cervical cord projections to the PAG in the cat, found that the majority of labeled neurons was located in lamina I, but only when the tracer injections were placed in the lateral PAG. When
KEY WORDS: Spinomesencephalic, Trigeminal, Nucleus caudalis, Marginal layer, Nociception.
INTRODUCTION The periaqueductal gray (PAG) is well known for its role in nociception control. Stimulation of the PAG results in a strong decrease of nociception in all parts of the body [5,23,29,35]. However, the PAG also plays an important role in the control of emotional motor behaviors necessary for the survival of the individual and species [16]. PAG-controlled emotional motor behaviors are, e.g., defence behaviors, such as fight, threat display, flight, and immobility [1,2,3,10], but also mating behavior [17,37], micturition [7,8,17], vocalization [2,14,18,19], and maternal behavior [25,26]. To control nociception and the emotional motor behaviors, the PAG needs to be informed about the internal and external environment of the individual. Anatomical studies have shown that the
* Address for correspondence: Leonora J. Mouton, Department of Anatomy and Embryology, Faculty of Medical Sciences, Rijksuniversiteit Groningen, Groningen, The Netherlands. Fax: 31-50-3632461; E-mail: [email protected]
167
168
MOUTON, KLOP AND HOLSTEGE
FIG. 1. Photographs and drawings showing the wheat germ agglutinin-conjugated horseradish peroxidase injection sites of three cases. On top: drawings of the injection sites in gray. The core of the injections is indicated in black. Bottom: brightfield photomicrographs of the injection areas in 40 m thick sections, visualized with the diamino-benzidine method. Scale bar: 1000 m.
FIG. 2. Schematic drawings of the rostrocaudal location of the injection sites in the periaqueductal gray. An indication of what is considered the core of the injections is shown as the black area.
LAMINA I-PAG REPRESENTS ONLY A LIMITED PART
169
FIG. 3. Darkfield polarized photomicrographs (Zeis Axiocam, extended focus) of labeled lamina I cells in the caudal medulla and various spinal segments, after wheat germ agglutinin-conjugated horseradish peroxidase injections in the dorsolateral, lateral, and ventrolateral parts of the intermediate and caudal periaqueductal gray (case 2385). Scale bar: 100 m.
the tracer was injected in the ventrolateral PAG the majority of labeled neurons were located in laminae VII and VIII. Furthermore, Keay et al. [21], in a study on all spino-PAG projections in the rat, reported that only 26 –28% of all spino-PAG projecting neurons were located in lamina I after injections in the lateral or ventrolateral PAG, respectively. So, in can be concluded that there is no good insight on which portion of the caudal medullary and spinal PAG projections originates in lamina I. The present retrograde tracing study in the cat is an attempt to solve these problems. Injections were made in different parts of the lateral and ventrolateral PAG, and the number of labeled cells in lamina I in both the medulla caudal to the obex and in each of the C1-Coc2 spinal cord segments was determined. These numbers were compared with the total numbers of labeled cells in all other laminae of the caudal medulla and in each of the spinal segments, resulting in a better insight in the percentages of lamina I-PAG projecting cells in the caudal medulla and spinal cord. A preliminary report of these results has been published as an abstract [22].
MATERIALS AND METHODS Surgical Procedures A total of five cats was used. The surgical procedures, pre- and postoperative care, handling, and housing of the animals were in accordance with protocols approved by the Committee of Animal Experiments of the Faculty of Medicine of the University of Groningen. For surgery the animals were initially anesthetized with ketamine (Nimatek, 0.1 ml/kg, intramuscular [i.m.]) and xylazine (Sedamun, 0.1 ml/kg i.m.), after which they were kept anesthetized by ventilation with a mixture of O2, N2O, and halothane. During surgery, heart rate and body temperature were monitored. Following a survival time of 3 days the animals were initially anesthetized with ketamine and xylazine i.m., followed by an overdose of 6% pentobarbital sodium (Nembutal; intraperitoneal [i.p.]). The cats were perfused transcardially with 2 l of heparinized saline at 37°C, directly followed by 2 l of 0.1 M phosphate buffer (pH 7.4), containing 4% sucrose, 1% paraformaldehyde, and 2% glutaraldehyde.
170
MOUTON, KLOP AND HOLSTEGE
FIG. 4. Schematic drawings of the labeled neurons in the caudal medulla and various spinal segments after a wheat germ agglutinin-conjugated horseradish peroxidase injection in the lateral intermediate and caudal periaqueductal gray (case 2300). Note that each drawing of a segment represents the labeled neurons in six 40-m thick sections.
Experimental Procedures In two female cats, three injections, each containing approximately 30 nl 2.5% wheat germ agglutinin-conjugated horseradish peroxidase (WGA-HRP), were placed in the lateral and ventrolateral parts of the rostral (case 2367) or intermediate and caudal PAG (case 2385). In three other female cats, one injection of approximately 20 –50 nl 2.5% WGA-HRP was placed in different parts of the lateral and/or ventrolateral intermediate and caudal PAG (case 2401: lateral part of the ventrolateral and lateral PAG, case 2390: lateral PAG, case 2300: central part of the lateral PAG). After perfusion, the brain and spinal cord were removed, postfixed for 2 h and stored overnight in 20% sucrose in phosphate buffer at 4°C. Subsequently, in each case the brainstem and all spinal segments (C1-Coc2) were cut into 40-m frozen transverse sections. Every fourth section was incubated according to the tetramethyl-benzidine method (TMB), dehydrated, and coverslipped. Of the area containing the injection site an extra series of sections was processed with diamino-benzidine (DAB). In each TMB processed transverse section of the spinal cord and of the medulla caudal to the obex, the retrogradely labeled profiles were plotted and counted. For counting the labeled cells in lamina I, lamina I was defined according to Rexed for the spinal cord [36] and according to Gobel for the caudal medulla [12]. It was assumed that each labeled profile consisting of at least part of a labeled cell body and part of a labeled axon or dendrite represents one labeled neuron. It must be emphasized that the labeled neurons in the dorsal column nuclei, solitary nucleus and lateral cervical nucleus were not included in the counts.
RESULTS Figure 1 contains photomicrographs containing the injection sites in cases 2385, 2401, and 2300, as well as the corresponding schematic drawings. In the drawings, the core of the injections, as observed with brightfield illumination of the DAB stained sections, is shown in black. A schematic representation of the rostrocaudal extent of all injection sites is given in Fig. 2. In case 2367, the large injection site involves the dorsolateral, lateral, and ventrolateral parts of the rostral PAG. In case 2385, the large injection site involves the dorsolateral, lateral, and ventrolateral parts of the intermediate and caudal PAG. The other, smaller injection sites each involve the lateral intermediate and caudal PAG. Figure 3 contains photomicrographs of lamina I cells in the caudal medulla and spinal cord (case 2385). Figure 4 shows schematic drawings of the labeled lamina I cells found in the caudal medulla and various spinal segments, as well as the labeled cells found in the other laminae (case 2300). Figure 5 shows the numbers of labeled lamina I cells and the numbers of all other labeled cells per segment per case. Surprisingly, the results show that of all counted labeled cells in both the caudal medulla and entire spinal cord in the five different cases only 1–9% of the ipsilaterally, and 12–24% of the contralaterally located labeled cells were located in lamina I (Table 1 and Fig. 6). Also, when taking the results of the caudal medulla and spinal cord separate, low percentages of lamina I cells were found; in the caudal medulla in all cases less than 4% of all labeled cells located ipsilaterally to the injection sites was found in lamina I, while for the contralaterally located cells this was less than 7% (Table 1). In the spinal cord in all cases less
LAMINA I-PAG REPRESENTS ONLY A LIMITED PART
171
FIG. 5. Histograms of the numbers of labeled cells in the contralateral caudal medulla and spinal cord (counted in a 1:4 series of 40-m thick sections) after wheat germ agglutinin-conjugated horseradish peroxidase injections in different parts of the lateral periaqueductal gray. The numbers of labeled cells in lamina I of the medulla caudal to the obex and in each C1-Coc2 spinal cord segment are shown in black. The total numbers of labeled cells in the caudal medulla and all spinal cord segments are shown in gray. Note that the labeled cells in the dorsal column nuclei, the solitary nucleus, and the lateral cervical nucleus were not included. CMed, medulla caudal to the obex.
than 14% of all ipsilaterally labeled cells was located in lamina I, and of all contralaterally located labeled cells this was less than 29% (Table 1). Within the spinal cord different percentages of labeled lamina I were found per segment and per case; among the segments the percentages of labeled cells in lamina I varied ipsilaterally from 0 – 67% per segment, and contralaterally from 0 –78% per segment (Table 1). Per case, in only one or two segments more than 50% of the labeled cells were found in lamina I, mainly on the contralateral side. These segments were always in the middle of the enlargements, i.e., C7, C8, T1, or L6. In case 2300,
with an injection involving the central portion of the lateral part of the intermediate and caudal PAG, the number of segments in which a majority of labeled cells was located in lamina I was slightly higher than in the other cases. In this case a majority of lamina I cells was found in C7 and C8 contralaterally, L7 bilaterally, and S1 and S2 ipsilaterally. Averaged percentages of labeled lamina I cells over the five cases show that only in C7 more than 50% of all labeled cells were present in lamina I, while in most other segments less than 20% of the ipsilaterally, and less than 30% of the contralaterally located labeled cells was found in lamina I (Fig. 7).
172
MOUTON, KLOP AND HOLSTEGE TABLE 1 PERCENTAGES OF LABELED CELLS IN LAMINA I
Caudal medulla and spinal cord Caudal medulla Spinal cord In total Per segment
Ipsilateral (%)
Contralateral (%)
1–9 1–3
12–24 1–6
1–14 0–67
15–28 0–78
DISCUSSION Total Numbers of Spino-PAG and Caudal Medulla-PAG Neurons A reliable comparison between the numbers of labeled cells observed in the present study and the numbers found in studies of other authors [6,20,25,28,32,40 – 42] is difficult, because in all previous studies the numbers are much less precise. First, in these studies the numbers of cells are often the result of large injections, including the PAG, but also involving very large parts of the adjoining mesencephalic regions. Second, these studies only present numbers that are an average of many different segments and/or of several cases, each with different injection sites. Percentages of Lamina I-PAG Neurons The present results show that, averaged over the entire caudal medulla and spinal cord, only a limited part of the PAG projecting neurons is located in lamina I. However, as the present study, for the first time, studied the caudal medulla and each spinal segment separately, results also give evidence that large differences exist between the different segments (Figs. 5 and 7). These large segmental differences are most probably the reason that until now no good insight existed on which portion of the spinal projections to the PAG originates in lamina I. In the earlier studies on the spino-PAG afferents [6,20,21,42] the numbers of lamina I cells
were only counted in a few spinal segments and/or were derived from countings in spinal blocks consisting of several different segments. When counting lamina I-PAG cells in only a few segments, the choice of segments is crucial for the result: taking a segment of the enlargement might give the conclusion of a majority of lamina-I PAG cells, while, e.g., taking a segment of the upper cervical cord might not lead to this conclusion. When counting lamina I-PAG cells in larger blocks consisting of several segments, it is crucial which segments are selected to form one block: a block consisting of for instance L6, L7 and S1 might give the conclusion of a majority of lamina I-PAG cells, while a block of L7, S1 and S2 might not (Fig. 7). Percentages of Lamina I-PAG Neurons in the Lumbosacral Cord In case 2300, with an injection involving the central portion of the lateral part of the intermediate and caudal PAG, the number of segments in which a majority of labeled cells was located in lamina I was slightly higher than in the other cases. Regarding the relatively many labeled cells in lamina I of the lumbosacral cord in this case, it has to be emphasized that part of these lumbosacral lamina I cells probably belong to a distinct cell group located in the L6-S3 segments. An earlier study of our laboratory [39] showed the existence of such a distinct cell group located in the lateral part of lamina I of L6 –S2 and in laminae V–VII of S2. These cells project to the central part of the lateral and ventrolateral caudal and intermediate PAG [9,39]. In all likelihood, this cell group is involved in the relay to the PAG of information concerning micturition and mating behavior. It implies that the lamina I cells located in L6, L7 and S1 that belong to this specific group are probably not involved in nociception. Conclusion There is no doubt that the lamina I-PAG projection is a very important pathway involved in nociception, also because it involves more lamina I neurons than the spinothalamic tract [33]. The most important conclusion is, however, that lamina I neurons are only a minority of all spinal and caudal medullary neurons
FIG. 6. Histograms showing the total numbers of ipsi- and contralaterally located labeled cells as well as the numbers of labeled cells in lamina I in the caudal medulla and entire spinal cord, after injections in different parts of the lateral periaqueductal gray. The percentages of labeled lamina I cells ipsi- and contralaterally are indicated. Labeled cells were counted in a 1:4 series of 40-m thick sections. The labeled cells in the dorsal column nuclei, the solitary nucleus, and the lateral cervical nucleus were not included.
LAMINA I-PAG REPRESENTS ONLY A LIMITED PART
173
FIG. 7. Averaged results over the five cases: averaged percentages (and SD) of the labeled cells in the ipsi- and contralateral lamina I in the medulla caudal to the obex and in each C1-Coc2 spinal cord segment, after wheat germ agglutinin-conjugated horseradish peroxidase injections in different parts of the lateral periaqueductal gray. Labeled cells were counted in a 1:4 series of 40-m thick sections. Note that the labeled cells in the dorsal column nuclei, the solitary nucleus, and the lateral cervical nucleus were not included. CMed, medulla caudal to the obex.
projecting to the PAG. Although part of the non lamina I-PAG projecting cells are probably involved in the relay of nociceptive information also, clearly many of them relay other information to the PAG. Examples are bladder filling and information related to mating behavior, but there are many other possible, yet unknown, sorts of information the PAG needs for its strong control of emotional behavior.
10.
11.
ACKNOWLEDGEMENTS
The authors thank Mr. K. van Linschoten and Mrs. E. Meyer for their histotechnical help.
REFERENCES 1. Bandler, R.; Carrive, P.; Zhang, S. P. Integration of somatic and autonomic reactions within the midbrain periaqueductal grey: Viscerotopic, somatopic and functional organization. Brain Res. 87:269 – 305; 1991. 2. Bandler, R.; Depaulis, A. Midbrain periaqueductal gray control of defensive behavior in the cat and the rat. In: Depaulis, A.; Bandler, R., eds. The midbrain periaqueductal gray matter: Functional, anatomical, and neurochemical organization. New York: Plenum Press; 1991:175– 198. 3. Bandler, R.; Keay, K. A. Columnar organization in the midbrain periaqueductal gray and the integration of emotional expression. Prog. Brain Res. 107:285–300; 1996. 4. Beitz, A. J. The organization of afferent projections to the midbrain periaqueductal gray of the rat. Neuroscience 7:133–159; 1982. 5. Besson, J. M.; Chaouch, A. Peripheral and spinal mechanisms of nociception. Physiol. Rev. 67:67–186; 1987. 6. Blomqvist, A.; Craig, A. D. Organization of spinal and trigeminal input to the PAG. In: Depaulis, A.; Bandler, R., eds. The midbrain periaqueductal gray matter. New York: Plenum Press; 1991:345–363. 7. Blok, B. F. M.; Holstege, G. The central nervous system control of micturition in cats and humans. Behav. Brain Res. 92:119 –125; 1998. 8. Blok, B. F. M.; Sturms, L.; Holstege, G. Brain activation during micturition in women. Brain 121:2033–2042; 1998. 9. Blok, B. F.; De Weerd, H.; Holstege, G. Ultrastructural evidence for a
12.
13.
14. 15.
16.
17.
18. 19. 20.
21.
paucity of projections from the lumbosacral cord to the pontine micturition center or M-region in the cat: A new concept for the organization of the micturition reflex with the periaqueductal gray as central relay. J. Comp. Neurol. 359:300 –309; 1995. Carrive, P. The periaqueductal gray and defensive behavior: Functional representation and neuronal organization. Behav. Brain Res. 58:27– 47; 1993. Craig, A. D. Distribution of brainstem projections from the spinal lamina I neurons in the cat and the monkey. J. Comp. Neurol. 361: 225–248; 1995. Gobel, S.; Hockfield, S. The division of the dorsal and ventral horns of the mammalian caudal medulla into eight layers using anatomical criteria. In: Anderson, D. J.; Matthews, B., eds. Pain in the trigeminal region. Amsterdam: Elsevier; 1977:443– 453. Herbert, H.; Saper, C. B. Organization of medullary adrenergic and noradrenergic projections to the periaqueductal matter in the rat. J. Comp. Neurol. 315:23–52; 1992. Holstege, G. Anatomical study of the final common pathway for vocalization in the cat. J. Comp. Neurol. 284:242–252; 1989. Holstege, G. Descending motor pathways and the spinal motor system: Limbic and non-limbic components. Prog. Brain Res. 87:307– 412; 1991. Holstege, G. Emotional motor system. In: Dulbeco R., ed. Encyclopedia of human biology, 2nd ed., vol. 3. New York: Academic Press; 1997:643– 660. Holstege, G. The emotional motor system in relation to the supraspinal control of micturition and mating behavior. Behav. Brain Res. 92:103– 109; 1998. Ju¨rgens, U.; Pratt, R. Role of the periaqueductal grey in vocal expression of emotion. Brain Res. 167:367–378; 1979. Kanai, T.; Wang, S. C. Localization of the central vocalization mechanism in the brainstem of the cat. Exp. Neurol. 6:426 – 434; 1962. Keay, K. A.; Bandler, R. Anatomical evidence for segregated input from the upper cervical spinal cord to functionally distinct regions of the periaqueductal gray region of the cat. Neurosci. Lett. 139:143–148; 1992. Keay, K. A.; Feil, K.; Gordon, B. D.; Herbert, H.; Bandler, R. Spinal afferents to functionally distinct periaqueductal gray columns in the rat: An anterograde and retrograde tracing study. J. Comp. Neurol. 385:207–229; 1997.
174 22. Klop, E. M.; Mouton, L. J.; Holstege, G. Only a small portion of the caudal medullary and spinal projections to the PAG originate in lamina I. Soc. Neurosci. 25:146; 1999. 23. Liebeskind, J. C.; Guilbaud, G.; Besson, J. M.; Oliveras, J. L. Analgesia from electrical stimulation of the periaqueductal gray matter in the cat: Behavioral observations and inhibitory effects on spinal cord interneurons. Brain Res. 50:441– 446; 1973. 24. Lima, D.; Coimbra, A. Morphological types of spinomesencephalic neurons in the marginal zone lamina I. of the rat spinal cord, as shown after retrograde labeling with cholera toxin subunit B. J. Comp. Neurol. 279:327–339; 1989. 25. Liu, R. P. Laminar origins of spinal projection neurons to the periaqueductal gray of the rat. Brain Res. 264:118 –122; 1983. 26. Lonstein, J. L.; Stern, J. M. Role of the midbrain periaqueductal gray in maturnal nurturance and aggression: c-fos and electrolytic lesion studies in lactating rats. J. Neurosci. 17:3364 –3378; 1997. 27. Lonstein, J. S.; Simmons, D. A.; Stern, J. N. Functions of the caudal periaqueductal gray in lactating rats: Kyphosis, lordosis, maternal aggression, and fearfulness. Behav. Neurosci. 112:1502–1518; 1998. 28. Mantyh, P. W. The ascending input to the midbrain periaqueductal gray of the primate. J. Comp. Neurol. 211:50 – 64; 1982. 29. Mayer, J. D.; Wolfte, T. L.; Akil, H.; Carder, B.; Liebeskind, J. C. Analgesia from electrical stimulation in the brainstem of the cat. Science 174:1351–1354; 1971. 30. Mehler, W. Some neurological species differences—A posteriori. Ann. N.Y. Acad. Sci. 167:424 – 468; 1969. 31. Meller, S. T.; Dennis, B. J. Afferent projections to the periaqueductal gray in the rabbit. Neuroscience 19:927–964; 1986. 32. Menetrey, D.; Chaouch, A.; Binder, D.; Besson, J. M. The origin of the spinomesencephalic tract in the rat: An anatomical study using the retrograde transport of horseradish peroxidase. J. Comp. Neurol. 206: 193–207; 1982. 33. Mouton, L. J.; Holstege, G. Three times as many lamina I neurons
MOUTON, KLOP AND HOLSTEGE
34.
35. 36. 37. 38.
39. 40. 41. 42. 43.
project to the periaqueductal gray than to the thalamus: A retrograde tracing study in the cat. Neurosci. Lett. 255:107–110; 1998. Mouton, L. J.; Holstege, G. The segmental and laminar organization of the spinal neurons projecting to the periaqueductal gray (PAG) in the cat suggests the existence of at least five separate clusters of spinoPAG neurons. J. Comp. Neurol. 428:389 – 410; 2000. Reynolds, E. S. The use of lead citrate at high pH as an electronopaque stain in electron-microscopy. J. Cell. Biol. 17:208 –212; 1963. Rexed, B. A cytoarchitectonic atlas of the spinal cord in the cat. J. Comp. Neurol. 100:297–380; 1954. Sakuma, Y.; Pfaff, D. W. Facilitation of female reproductive behavior from mesencephalic central gray in the rat. Am. J. Physiol. 237:R278 – R284; 1979. Shipley, M. T.; Ennis, M.; Rizvi, T. A.; Behbehani, M. M. Topographical specificity of forebrain inputs to the midbrain periaqueductal gray: Evidence for discrete longitudinally organized input columns. In: Depaulis, A.; Bandler, R., eds. The midbrain periaqueductal gray matter: Functional, anatomical, and neurochemical organization. New York: Plenum Press; 1991:417– 448. VanderHorst, V. G. J. M.; Mouton, L. J.; Blok, B. F. M.; Holstege, G. Distinct cell groups in the lumbosacral cord of the cat project to different areas in the periaqueductal gray. J. Comp. Neurol. 376:361–385; 1996. Wiberg, M.; Blomqvist, A. The spinomesencephalic tract in the cat: Its cells of origin and termination pattern as demonstrated by the intraaxonal transport method. Brain Res. 291:1–18; 1984. Wiberg, M.; Westman, J.; Blomqvist, A. The projection to the mesencephalon from the sensory trigeminal nuclei. An anatomical study in the cat. Brain Res. 399:51– 68; 1986. Wiberg, M.; Westman, J.; Blomqvist, A. Somatosensory projection to the mesencephalon: An anatomical study in the monkey. J. Comp. Neurol. 264:92–117; 1987. Yezierski, R. P.; Mendez, C. M. Spinal distribution and collateral projections of rat spinomesencephalic tract cells. Neuroscience 44: 113–130; 1991.