THE JOURNAL OF COMPARATIVE NEUROLOGY 322224-232 (1992)
Quantitative Autoradiographic Distribution of Calcitonin Gene-Related Peptide (hCGRPa) Binding Sites in the Rat and Monkey Spinal Cord K. YASHPAL, S. KAR, T. DENNIS, AND R. QUIRION Douglas Hospital Research Centre and Department of Psychiatry, Faculty of Medicine, McGill University, Montreal, Quebec, Canada H4H 1R3
ABSTRACT Calcitonin gene-related peptide (CGRP) has been implicated in various spinal functions on the basis of its presence in the substantia gelatinosa and motoneurons and the biological effects induced by intrathecal CGRP injections. We investigated here the comparative distribution of [1251]hCGRPabinding sites in various segments of the rat and monkey spinal cord. The immunocytochemical localization of CGRP-like material in rat spinal cord was also evaluated for comparison. In the rat spinal cord, high densities of [lz5I1hCGRPa binding sites were observed in lamina I, in a U-shaped band that included lamina X and the medial parts of laminae 111-IV and in the intermediolateral and intermediomedial nuclei. The substantia gelatinosa (lamina 11) contained relatively lower, but still significant, densities of [*251]hCGRPa binding sites, while the ventral horn showed low amounts of specific labeling. CGRP-like immunoreactive fibers, on the other hand, were heavily concentrated in laminae 1-11 and in the reticulated portion of lamina V of the dorsal horn. Immunoreactivity to CGRP antiserum was also noted in fibers around the central canal and in a number of motoneurons of the ventral horn. In the monkey spinal cord, [12511hCGRPcubinding sites were present in lamina I in a U-shaped band that included lamina X and the medial portions of laminae V-VI. Relatively low levels of ['25I]hCGRPa binding were detected in laminae I1 to IV of the dorsal horn, while the ventral horn was more enriched with specific [12511hCGRPabinding sites. Thus, it appears that the autoradiographic distribution of [1251]hCGRPasites is species dependent in the spinal cord. Additionally, some differences are observed between the localization of [lZ5I]hCGRPabinding sites and immunoreactive material in the rat spinal cord. These differences may be relevant to the purported roles of CGRP-like peptides in spinal functions such as nociception, control of 1992 Wiley-Liss, Inc. sympathetic output, and motor control. (c)
Key words: CGRP, receptor sub-types, receptor audioradiography, dorsal horn, primate
Calcitonin gene-related peptide (CGRP), a 37 amino acid peptide, is derived from alternate processing of the primary transcript of the calcitonin gene and is considered to be the major product of this gene in neural tissues (Amara et al., '82, Rosenfeld et al., '83). A second gene, which encodes a closely related peptide, has recently been isolated (Morris et al., '84, Amara et al., '85; Steinberg et al., '85). These peptides have been classified as CGRPa and CGRPP, respectively. Numerous studies have reported the extensive distribution of CGRP-like immunoreactivity (Gibson et al., '84; Kawai et al., '85; Skofitsch and Jacobowitz, '85a; Tschopp et al., '85; Carlton et al., '87; Okimura et al., '87; Kruger et al., '88a,b), its mRNAs (Rethelyi et al., '89), and its binding sites (Dawnbarn et al., '85; Henke et al., '85; Seifert et al., '85; Skofitsch and Jacobwitz, '85c; Tschopp et al., '85; O 1992 WILEY-LISS.
INC.
Inagaki et al.,'86; Sexton et al., '86; Kruger et al., '88a; Dennis et al., '91) in central and peripheral nervous systems. The distribution of receptor binding sites for a- and 6-CGRP in the central nervous system of both human and rat was found to be basically similar, except for the cerebellum and inferior colliculus where the specific binding of a-CGRP is slightly higher than that of P-CGRP (Henke et al., '87). Furthermore, the competition study showed that there is no consistent difference in the ability of a or p forms to displace either ligand, suggesting a and p forms may possibly act through the same receptor system (Henke et al., '87; Kruger et al., '88).
Accepted March 27, 1992
225
SPINAL CGRP RECEPTORS In the spinal cord, a dense network of CGRP-immunoreactive fibers is present in the dorsal horn, a site of the first synapse in sensory pathways (Gibson et al., '84; Chung et al., '88; McNeill et al., '88a,b). It has been shown that dorsal root ganglion cells are the source of these CGRP immunoreactive fibers (Chung et al., '88). Functional studies have provided evidence for roles of CGRP in the modulation of thermal (Wiesenfeld-Hallin et al., '84; Gamse and Saria, '86; Cridland and Henry, '88) and mechanical (Oku et al., '87) nociception as well as in the control of sympathetic output (Rochford et al., '90). The co-existence of CGRP and substance P in a subpopulation of sensory ganglion neurones (Wiesenfeld-Hallin et al., '84; Gibson et al., '84, Gibbins et al., '85; Lee et al., '85; Skofitsch and Jacobowitz, '85b) and their depletion following neonatal capsaicin treatment (Skofitsch and Jacobowitz, '85b) suggest possible interaction between these peptides in the regulation of spinal nociceptive information. Furthermore, Hokfelt and Terenius ('87) have proposed that upon their simultaneous release from primary afferent fibers, CGRP could indirectly prolong the effect of substance P by saturating the proteolytic enzymes involved in the metabolism of these peptides. In spite of possible important roles played by CGRP in the spinal cord, relatively little information is currently available on the presence and distribution of CGRP receptors at various segmental levels. Most, if not all, studies have focused on the cervical segment of the rat spinal cord (Seifert et al., '85; Skofitsch and Jacobowitz, '85c; Tschopp et al., '85; Inagaki et al., '86; Sexton et al., '86; Kruger et al., '88a). At that level, [1251]CGRPreceptor sites appear to be concentrated in lamina I and around the central canal with an apparent paucity of sites in the substantia gelatinosa (Sexton et al., '86; Kruger et al., '88a; but see Seifert et al., '85; Skofitsch and Jacobowitz, '85c; Inagaki et al., '86). In view of our interest in the chemical neuroanatomy of the spinal cord especially in relation to peptide receptors (Gouarderes et al., '85, '91; Yasphal et al., '90, '91a,b), the present study was undertaken to examine the distribution of [1251]hCGRPareceptor sites at various segmental levels of the rat and monkey spinal cord. The immunocytochemical localization of CGRP-like material was also evaluated for comparison in the rat spinal cord. Data reveal that the distribution of ['25I]hCGRPa receptor sites may be speciesdependent and without evident rostrocaudal gradients. Preliminary results were presented elsewhere in abstract form (Yasphal et al., '89).
supplied by Dr. Alain Fournier (I.N.R.S. Sante, Pointe Claire, Quebec). Bacitracin, bovine serum albumin (BSA), goat anti-rabbit I&, and peroxidase-antiperoxidase(PAP) complex were obtained from Sigma Chemical Company (St. Louis, MO). All other chemicals were of the highest analytical grade and purchased from Fisher Scientific (Montreal, Quebec).
Receptor autoradiography Rats were sacrificed by decapitation and monkeys by the administration of an overdose of pentobarbital. Spinal cords were removed and segments identified before being rapidly snap-frozen in 2-methyl-butane at -40°C. Tissue blocks were fixed to microtome chucks and serial transverse sections (20 wm) were cut at -18°C using a cryostat microtome (Bright, Cambridge, U.K.), thaw-mounted onto gelatin-coated microscope slides, dried overnight under vacuum, and stored at -80°C until use. Binding experiments were carried out as previously described (Dennis et al., '91). Briefly, spinal cord sections at various segmental levels were preincubated at 25°C for 15 minutes in 50 mM Tris-HC1 buffer at pH 7.4, containing 100 mM NaCl, 0.2% BSA, and 0.4 mM bacitracin. Following the preincubation period, sections were incubated for 1 hour in fresh buffer containing 50 pM [1251]hCGRPa. Specific binding was determined by incubating adjacent sections in the presence of 1pM unlabelled hCGRPa. At the end of the incubation period, sections were washed (4 x 1 minutes) in ice-cold buffer followed by a rapid rinse in ice-cold deionized water to remove salts and then dried under a stream of cold air. Autoradiograms were generated from the apposition of the labelled tissue sections along with slide-mounted [12511-microscalesto tritium-sensitive films in light-proof X-ray cassettes for 4-6 days. Films were then developed under standard conditions (Dennis et al., '9 1).Film autoradiograms were quantified by densitometry with the aid of a computerized image analysis system (MCID; Imaging Research, Inc., St. Catherines, Ontario). Arbitrary optical density units, obtained from autoradiograms, were converted into radioactivities by reference to the [lz5I]microscales coexposed with radiolabelled sections. Reported values refer to specific binding determined as the difference in radioactivity bound in the presence and absence of 1pM hCGRPa. To facilitate anatomical identification (Grant, '851, sections used for quantification and photomicrographs were stained with cresyl violet.
Immunocytochemistry Rats were anaesthetized with sodium pentobarbital and MATERIALS AND METHODS fixed for immunocytochemistry by intracardial perfusion Animals and chemicals with Bouin's solution. Following this, segments of spinal Male Sprague-Dawley rats weighing 250-300 g were cord from cervical, thoracic, lumbar, and sacral levels were obtained from Canadian Breeding Farms (St. Constant, dissected and postfixed in the same fixative for 2 hours. Quebec). Animals were housed in polyethylene cages, main- Samples were then washed and stored in phosphatetained on a 12 hour light-dark cycle and given free access to buffered saline (PBS, 0.01 M, pH 7.21, containing 15% food and water. Two male green monkeys (Cercopethicus sucrose and 0.1% sodium azide for at least 24 hours before aethiops)were obtained from Dr. R. Palmour, Department processing to cryostat blocks. Serial sections (20 pm) were of Psychiatry, McGill University, following behavioral stud- cut in the transverse plane and processed for immunocyies not involving any treatments which could affect the tochemistry with the glucose oxidase-DAB-nickelenhancepresent study. Animals were kept according to rules and ment technique described in detail elsewhere (Kar et al., regulations of the Canadian Council on Animal Care and '89). In brief, sections were incubated with a wellcharacterized (for specificity, see Kar et al., '89) polyclonal McGill University policies. L12511hCGRPa(specific activity: 2,000 Ci/mmol), Hyper- hCGRPa antiserum (dil. 1:2,000) at 4°C for 48 hours, film, and [lZ5I]microscales were obtained from Amersham washed three times in PBS (0.01 M, pH 7.21, and incubated Canada (Oakville, Ontario). Synthetic hCGRPa was kindly for 45 minutes with goat anti-rabbit IgG (1:25). After
K. YASHPAL ET AL.
226 TABLE 1. Quantitative Autoradiographic Analysis of Specific ['"IlhCGRPol Binding Sites at Various Segments of the Adult Rat Spinal Cord'
Rat spinal cord: Immunocytochemistry
['"SIIhCGRPol (fmolimgtissue equivalent,wet weight) Cervical Dorsal Horn Lamina I Lamina I t Laminae 111-V Lamina X ILN and IMML Ventral horn White matter Dorsalfuniculus Ventral funiculus
16.7 i 1.4* 10.2 i- 0.9 18.8 i 1.3* 22.8 i 2.2
-
8.2
2
0.7
10.5 2 0.8 18.6 i- 1.1
Thoracic 15.1 5 10.6 5 17.2 5 24.4 5 22.3 2 9.8 5
1.3' 1.2 14* 1.1 1.6 1.2
Lumbar
Sacral
16.2? 1.4' 11.6 i 1.3 18.4 + 0.,9* 25.1 -+ 0.8
18.2 i 1.6* 12.5 5 1.8 22.1 i 1.2* 25.7 5 2.1
-
8.8i-1.3 8.8 2 0.8 18.2 ? 0.5
10.2 2 0.7 16.3 i 0.9
-
9.3
5
1.2
9.4 2 1.1 18.6 2 0.8
'Values represent means i S.E.M. of data obtained from 5-10 sections at each segmental level. Non-specific binding in the presence of 1 pM unlabelled hCGRPa was subtracted from all readinrs. "LN, Intermediolateral nucleus; IMM, Intermediomedlal nucleus. * P
Quantitative Autoradiographic Distribution of Calcitonin Gene-Related Peptide (hCGRPa) Binding Sites in the Rat and Monkey Spinal Cord K. YASHPAL, S. KAR, T. DENNIS, AND R. QUIRION Douglas Hospital Research Centre and Department of Psychiatry, Faculty of Medicine, McGill University, Montreal, Quebec, Canada H4H 1R3
ABSTRACT Calcitonin gene-related peptide (CGRP) has been implicated in various spinal functions on the basis of its presence in the substantia gelatinosa and motoneurons and the biological effects induced by intrathecal CGRP injections. We investigated here the comparative distribution of [1251]hCGRPabinding sites in various segments of the rat and monkey spinal cord. The immunocytochemical localization of CGRP-like material in rat spinal cord was also evaluated for comparison. In the rat spinal cord, high densities of [lz5I1hCGRPa binding sites were observed in lamina I, in a U-shaped band that included lamina X and the medial parts of laminae 111-IV and in the intermediolateral and intermediomedial nuclei. The substantia gelatinosa (lamina 11) contained relatively lower, but still significant, densities of [*251]hCGRPa binding sites, while the ventral horn showed low amounts of specific labeling. CGRP-like immunoreactive fibers, on the other hand, were heavily concentrated in laminae 1-11 and in the reticulated portion of lamina V of the dorsal horn. Immunoreactivity to CGRP antiserum was also noted in fibers around the central canal and in a number of motoneurons of the ventral horn. In the monkey spinal cord, [12511hCGRPcubinding sites were present in lamina I in a U-shaped band that included lamina X and the medial portions of laminae V-VI. Relatively low levels of ['25I]hCGRPa binding were detected in laminae I1 to IV of the dorsal horn, while the ventral horn was more enriched with specific [12511hCGRPabinding sites. Thus, it appears that the autoradiographic distribution of [1251]hCGRPasites is species dependent in the spinal cord. Additionally, some differences are observed between the localization of [lZ5I]hCGRPabinding sites and immunoreactive material in the rat spinal cord. These differences may be relevant to the purported roles of CGRP-like peptides in spinal functions such as nociception, control of 1992 Wiley-Liss, Inc. sympathetic output, and motor control. (c)
Key words: CGRP, receptor sub-types, receptor audioradiography, dorsal horn, primate
Calcitonin gene-related peptide (CGRP), a 37 amino acid peptide, is derived from alternate processing of the primary transcript of the calcitonin gene and is considered to be the major product of this gene in neural tissues (Amara et al., '82, Rosenfeld et al., '83). A second gene, which encodes a closely related peptide, has recently been isolated (Morris et al., '84, Amara et al., '85; Steinberg et al., '85). These peptides have been classified as CGRPa and CGRPP, respectively. Numerous studies have reported the extensive distribution of CGRP-like immunoreactivity (Gibson et al., '84; Kawai et al., '85; Skofitsch and Jacobowitz, '85a; Tschopp et al., '85; Carlton et al., '87; Okimura et al., '87; Kruger et al., '88a,b), its mRNAs (Rethelyi et al., '89), and its binding sites (Dawnbarn et al., '85; Henke et al., '85; Seifert et al., '85; Skofitsch and Jacobwitz, '85c; Tschopp et al., '85; O 1992 WILEY-LISS.
INC.
Inagaki et al.,'86; Sexton et al., '86; Kruger et al., '88a; Dennis et al., '91) in central and peripheral nervous systems. The distribution of receptor binding sites for a- and 6-CGRP in the central nervous system of both human and rat was found to be basically similar, except for the cerebellum and inferior colliculus where the specific binding of a-CGRP is slightly higher than that of P-CGRP (Henke et al., '87). Furthermore, the competition study showed that there is no consistent difference in the ability of a or p forms to displace either ligand, suggesting a and p forms may possibly act through the same receptor system (Henke et al., '87; Kruger et al., '88).
Accepted March 27, 1992
225
SPINAL CGRP RECEPTORS In the spinal cord, a dense network of CGRP-immunoreactive fibers is present in the dorsal horn, a site of the first synapse in sensory pathways (Gibson et al., '84; Chung et al., '88; McNeill et al., '88a,b). It has been shown that dorsal root ganglion cells are the source of these CGRP immunoreactive fibers (Chung et al., '88). Functional studies have provided evidence for roles of CGRP in the modulation of thermal (Wiesenfeld-Hallin et al., '84; Gamse and Saria, '86; Cridland and Henry, '88) and mechanical (Oku et al., '87) nociception as well as in the control of sympathetic output (Rochford et al., '90). The co-existence of CGRP and substance P in a subpopulation of sensory ganglion neurones (Wiesenfeld-Hallin et al., '84; Gibson et al., '84, Gibbins et al., '85; Lee et al., '85; Skofitsch and Jacobowitz, '85b) and their depletion following neonatal capsaicin treatment (Skofitsch and Jacobowitz, '85b) suggest possible interaction between these peptides in the regulation of spinal nociceptive information. Furthermore, Hokfelt and Terenius ('87) have proposed that upon their simultaneous release from primary afferent fibers, CGRP could indirectly prolong the effect of substance P by saturating the proteolytic enzymes involved in the metabolism of these peptides. In spite of possible important roles played by CGRP in the spinal cord, relatively little information is currently available on the presence and distribution of CGRP receptors at various segmental levels. Most, if not all, studies have focused on the cervical segment of the rat spinal cord (Seifert et al., '85; Skofitsch and Jacobowitz, '85c; Tschopp et al., '85; Inagaki et al., '86; Sexton et al., '86; Kruger et al., '88a). At that level, [1251]CGRPreceptor sites appear to be concentrated in lamina I and around the central canal with an apparent paucity of sites in the substantia gelatinosa (Sexton et al., '86; Kruger et al., '88a; but see Seifert et al., '85; Skofitsch and Jacobowitz, '85c; Inagaki et al., '86). In view of our interest in the chemical neuroanatomy of the spinal cord especially in relation to peptide receptors (Gouarderes et al., '85, '91; Yasphal et al., '90, '91a,b), the present study was undertaken to examine the distribution of [1251]hCGRPareceptor sites at various segmental levels of the rat and monkey spinal cord. The immunocytochemical localization of CGRP-like material was also evaluated for comparison in the rat spinal cord. Data reveal that the distribution of ['25I]hCGRPa receptor sites may be speciesdependent and without evident rostrocaudal gradients. Preliminary results were presented elsewhere in abstract form (Yasphal et al., '89).
supplied by Dr. Alain Fournier (I.N.R.S. Sante, Pointe Claire, Quebec). Bacitracin, bovine serum albumin (BSA), goat anti-rabbit I&, and peroxidase-antiperoxidase(PAP) complex were obtained from Sigma Chemical Company (St. Louis, MO). All other chemicals were of the highest analytical grade and purchased from Fisher Scientific (Montreal, Quebec).
Receptor autoradiography Rats were sacrificed by decapitation and monkeys by the administration of an overdose of pentobarbital. Spinal cords were removed and segments identified before being rapidly snap-frozen in 2-methyl-butane at -40°C. Tissue blocks were fixed to microtome chucks and serial transverse sections (20 wm) were cut at -18°C using a cryostat microtome (Bright, Cambridge, U.K.), thaw-mounted onto gelatin-coated microscope slides, dried overnight under vacuum, and stored at -80°C until use. Binding experiments were carried out as previously described (Dennis et al., '91). Briefly, spinal cord sections at various segmental levels were preincubated at 25°C for 15 minutes in 50 mM Tris-HC1 buffer at pH 7.4, containing 100 mM NaCl, 0.2% BSA, and 0.4 mM bacitracin. Following the preincubation period, sections were incubated for 1 hour in fresh buffer containing 50 pM [1251]hCGRPa. Specific binding was determined by incubating adjacent sections in the presence of 1pM unlabelled hCGRPa. At the end of the incubation period, sections were washed (4 x 1 minutes) in ice-cold buffer followed by a rapid rinse in ice-cold deionized water to remove salts and then dried under a stream of cold air. Autoradiograms were generated from the apposition of the labelled tissue sections along with slide-mounted [12511-microscalesto tritium-sensitive films in light-proof X-ray cassettes for 4-6 days. Films were then developed under standard conditions (Dennis et al., '9 1).Film autoradiograms were quantified by densitometry with the aid of a computerized image analysis system (MCID; Imaging Research, Inc., St. Catherines, Ontario). Arbitrary optical density units, obtained from autoradiograms, were converted into radioactivities by reference to the [lz5I]microscales coexposed with radiolabelled sections. Reported values refer to specific binding determined as the difference in radioactivity bound in the presence and absence of 1pM hCGRPa. To facilitate anatomical identification (Grant, '851, sections used for quantification and photomicrographs were stained with cresyl violet.
Immunocytochemistry Rats were anaesthetized with sodium pentobarbital and MATERIALS AND METHODS fixed for immunocytochemistry by intracardial perfusion Animals and chemicals with Bouin's solution. Following this, segments of spinal Male Sprague-Dawley rats weighing 250-300 g were cord from cervical, thoracic, lumbar, and sacral levels were obtained from Canadian Breeding Farms (St. Constant, dissected and postfixed in the same fixative for 2 hours. Quebec). Animals were housed in polyethylene cages, main- Samples were then washed and stored in phosphatetained on a 12 hour light-dark cycle and given free access to buffered saline (PBS, 0.01 M, pH 7.21, containing 15% food and water. Two male green monkeys (Cercopethicus sucrose and 0.1% sodium azide for at least 24 hours before aethiops)were obtained from Dr. R. Palmour, Department processing to cryostat blocks. Serial sections (20 pm) were of Psychiatry, McGill University, following behavioral stud- cut in the transverse plane and processed for immunocyies not involving any treatments which could affect the tochemistry with the glucose oxidase-DAB-nickelenhancepresent study. Animals were kept according to rules and ment technique described in detail elsewhere (Kar et al., regulations of the Canadian Council on Animal Care and '89). In brief, sections were incubated with a wellcharacterized (for specificity, see Kar et al., '89) polyclonal McGill University policies. L12511hCGRPa(specific activity: 2,000 Ci/mmol), Hyper- hCGRPa antiserum (dil. 1:2,000) at 4°C for 48 hours, film, and [lZ5I]microscales were obtained from Amersham washed three times in PBS (0.01 M, pH 7.21, and incubated Canada (Oakville, Ontario). Synthetic hCGRPa was kindly for 45 minutes with goat anti-rabbit IgG (1:25). After
K. YASHPAL ET AL.
226 TABLE 1. Quantitative Autoradiographic Analysis of Specific ['"IlhCGRPol Binding Sites at Various Segments of the Adult Rat Spinal Cord'
Rat spinal cord: Immunocytochemistry
['"SIIhCGRPol (fmolimgtissue equivalent,wet weight) Cervical Dorsal Horn Lamina I Lamina I t Laminae 111-V Lamina X ILN and IMML Ventral horn White matter Dorsalfuniculus Ventral funiculus
16.7 i 1.4* 10.2 i- 0.9 18.8 i 1.3* 22.8 i 2.2
-
8.2
2
0.7
10.5 2 0.8 18.6 i- 1.1
Thoracic 15.1 5 10.6 5 17.2 5 24.4 5 22.3 2 9.8 5
1.3' 1.2 14* 1.1 1.6 1.2
Lumbar
Sacral
16.2? 1.4' 11.6 i 1.3 18.4 + 0.,9* 25.1 -+ 0.8
18.2 i 1.6* 12.5 5 1.8 22.1 i 1.2* 25.7 5 2.1
-
8.8i-1.3 8.8 2 0.8 18.2 ? 0.5
10.2 2 0.7 16.3 i 0.9
-
9.3
5
1.2
9.4 2 1.1 18.6 2 0.8
'Values represent means i S.E.M. of data obtained from 5-10 sections at each segmental level. Non-specific binding in the presence of 1 pM unlabelled hCGRPa was subtracted from all readinrs. "LN, Intermediolateral nucleus; IMM, Intermediomedlal nucleus. * P
