- Email: [email protected]
Distribution and chromatographic characterisation of CGRP-like immunoreactivity in the brain and gut of the rat
Regulatory Peptides, 12 (1985) 133-143 Elsevier
133
R P T 00410
Distribution and chromatographic characterisation of CGRP-like immunoreactivity in the brain and gut of the rat P.K. Mulderry, M.A. Ghatei, A.E. Bishop, Y.S. Allen, J.M. Polak and S.R. Bloom Departments of Medicine and Histochemistry, Royal Postgraduate Medical School, Du Cane Road, London, W12 OHS, U.K. (Received 27 March 1985; revised manuscript received 8 July 1985; accepted for publication 9 July 1985)
Summary Radioimmunoassay, chromatography and immunocytochemistry were used to study the occurrence of calcitonin gene-related peptide in the brain and gastrointestinal tract of the rat. In the brain, the highest concentrations of the peptide were found in the medulla oblongata (58.3 + 6.8 pmol/g) where immunocytochemistry showed the presence of immunoreactive cell bodies. Significant concentrations were also found in the pancreas and throughout the gastrointestinal tract, the highest levels occurring in the pyloric sphincter (48.0 4- 6.0 pmol/g). CGRP-like immunoreactivity in the gastrointestinal tract was restricted to nerve fibres. Chromatographic analysis of the CGRP-Iike immunoreactivity occurring in these tissues showed that at least 7 0 0 was indistinguishable from the synthetic peptide. However, there was also evidence of a number of smaller cross-reacting molecular species. CGRP; brain; gut; radioimmunoassay; chromatography; immunocytochemistry
Introduction Structural analyses of the rat calcitonin gene have led to predictions of the existence of a 37 amino acid calcitonin gene-related peptide (CGRP), which is coded on the Address for correspondence: Professor S.R. Bloom, Department of Medicine, Royal Postgraduate Medical School, Du Cane Road, London WI2 0HS, U.K. Tel. 01-743-2030 Ext. 274. 0167-0115/85/$03.30 © 1985 Elsevier Science Publishers B.V. (Biomedical Division)
134
same gene as calcitonin but which is translated from a separate mRNA produced by alternative splicing events during expression of the gene [1]. The presence of CGRP mRNA and immunoreactive CGRP has subsequently been demonstrated in neural tissue [2], hence CGRP has been cited as putative neuropeptide. Studies of human medullary thyroid carcinoma tissues have shown that CGRP mRNA and a peptide whose amino acid sequence differs from that of rat CGRP in only four positions are produced [3-5]. CGRP-like immunoreactivity has been reported in the spinal cords of the marmoset, horse, pig, cat, guinea pig, mouse, frog, rat and man [6], and both rat- and human-sequence CGRPs have been shown to be potent vasodilators [7]. CGRP-like immunoreactivities occurring in the hypothalamus [2], trigeminal ganglia [8,13] and blood vessels [9] have been characterised by chromatography. The major component of CGRP-like immunoreactivity was found to correspond in elution characteristics to the peptide synthesised according to the predicted CGRP sequence. Immunohistochemical studies have demonstrated a system of CGRP pathways in the brain and the presence of CGRP-containing fibres in the gut of the rat [2], although the distribution of the immunoreactivity in the gut has not yet been described nor has its chromatographic identity with synthetic CGRP been verified. In this study we have investigated the distribution and localisation of CGRP-like immunoreactivity in the brain and gastrointestinal tract of the rat. In addition, CGRP-Iike immunoreactivity occurring in these tissues was characterised by chromatography.
Materials and Methods
Tissues All specimens were taken from adult male Wistar rats. For radioimmunoassay, animals were killed by cervical dislocation, dissected and the tissues extracted by boiling in 0.5 M acetic acid as previously described [10]. For immunocytoebemistry, tissues were fixed either by intracardial perfusion with 4% paraformaldehyde in phosphate-buffered saline (PBS) or by direct immersion of specimens in 0.4% benzoquinone in PBS as described by Bishop et al. [11]. Radioirnmunoassay Tissue extracts were assayed for CGRP as duplicate 20-#1 aliquots in 0.8 ml 0.06 M phosphate buffer, pH 7.4 containing 0.01 M EDTA. 0.05% (w/v) sodium azide and 3% (w/v) bovine serum albumin (BSA). The CGRP antiserum was raised in a New Zealand white rabbit by injecting with synthetic rat CGRP coupled to bovine serum albumin by a glutaraldehyde reaction. The antiserum could be used at a final dilution of 1:400 000 and showed no cross-reactivity (less than 0.02%) with calcitonin, somatostatin-28, substance P, VIP or gastrin-releasing peptide. Radiolabelled CGRP was prepared by a chloramine-T iodination of the histidine residue in synthetic rat CGRP (Peninsula Laboratories) and purified by reverse-phase high pressure liquid chromatography (HPLC) as previously described [18]. The specific activity obtained was 58.0 Bq/fmol as determined by self displacement in the assay. Bound and free
135 fractions were separated after incubation by charcoal adsorption of the free fraction. The assay standard was Peninsula synthetic rat C G R P of which 1 fmol per assay tube could be detected with 95% confidence.
Chromatography The CGRP-Iike immunoreactivity was characterised by gel filtration and HPLC. For gel filtration, pooled tissue extracts were loaded onto Sephadex G-50 superfine columns (60 x 0.9 cm) eluted with 0.06 M phosphate buffer, pH 7.4, containing 0.3% BSA and 0.2 M NaCI at a flow rate of 3.6 ml/h. Before loading, each sample was spiked with blue dextran and Na125I as internal markers of void volume (Iio) and total volume (lit), respectively. Fractions were collected every 15 min. HPLC was carried out on a Techsil C-18 reverse-phase column (HPLC Technology) eluted at 1 ml/min with an acetonitrile gradient (see Fig. 2). Prior to loading, samples were purified on Waters Sep-Pak C-18 reverse-phase sample preparation cartridges. After washing with 10 ml 0.1% aqueous trifluoroacetic acid (TFA), peptides were eluted off with 0.5 ml 60% acetonitrile, 0.1% TFA and diluted to 2.5 ml with water containing 0.1% TFA. A volume of 2 ml was injected onto the HPLC column and the remainder assayed for assessment of recovery. The concentration of CGRP-like immunoreactivity in all column fractions was determined by assaying 100-/A aliquots directly.
Immunocytochemistry After thorough washing in PBS containing 15% sucrose and 0.01% NAN3, cryostat blocks were prepared and 10-#m sections cut at - 2 0 ° C and picked up onto glass slides coated with poly-L-lysine [12]. After being allowed to dry in the air at room temperature for at least 1 h, the sections were immunostained using the technique of indirect immunofluorescence (gastrointestinal tracO or peroxidase-anti-peroxidase (brain). The primary antiserum was the same as that used for radioimmunoassay, but applied at a dilution of 1:200 for immunofluorescence and 1:2000 for peroxidase-anti-peroxidase. Controls for immunostaining included omission of the first layer antiserum or its replacement with non-immune rabbit serum. Neither of these gave positive immunostaining. In addition, immunostaining with the anti-CGRP primary serum was successfully quenched by the addition of as little as 0.1 nmol of synthetic CGRP per ml of diluted antiserum.
Experimental results
1. Distributlon of CGRP-Iike immunoreactivity CGRP concentrations in the tissues studied, as determined by radioimmunoassay, are given in Table I. Significant levels of CGRP-Iike immunoreactivity were found in all regions of the brain apart from the cerebellum. The highest concentration occurred in the medulla oblongata with relatively high concentrations also in the amygdala, hypothalamus, septal area and pituitary. CGRP-like immunoreactivity
136 TABLE I Concentrations of CGRP found in regions of rat brain and gastrointestinal tract Brain region
pmol/g
Frontal cortex Parietal cortex Occipital cortex Caudate-putamen Nucleus accumbens Olfactory bulb Septum Hippocampus Thalamus Hypothalamus Pituitary Globus pallidus Amygdala Mid brain Periaqueductal grey Cerebellum Pons Medulla oblongata
0.9 1.4 2.1 1.4 5.9 2.6 10.7 0.8 6.0 9.6 10.7 7.7 19.7 4.4 6.0 < 0.5 6,4 58.3
+ 4+ 4444444± 444+
0.1 0.6 0.5 0.4 1.2 0.8 0.9 0.2 0.7 1.3 1.8 2.3 2.3 0.4 0.8
Gastrointestinal tract
pmol/g
Lower oesophageal sphincter Forestomach Corpus Antrum Pyloric sphincter Duodenum Jejunum Ileum Caecum Ascending colon Mid-colon Descending colon Rectum Pancreas
16.8 35.2 12.5 39,1 48.0 21.5 17.5 11.5 9.8 9.4 11.6 28.1 35.1 5.6
444± 444± + 444± 4-
1.9 4.1 2,8 4,9 6.0 2.1 2.1 0.9 1.8 0.8 2.6 2.5 3.6 0.8
4- 0.9 4- 6.8
Values given are mean 4- S.E.M. (n = 5).
CGRP-LI fmol/fraction
CGRP
CGRP
25O
100
125"
50
2S0
CGRP 150
~
l
Medulla Oblongata
75
Ioo
150
Colon
Pancreas
125
othalamus
75
......... Jla ......... i --=-
0
0.5
1.0
0
._--J
..... ::::~ 0.5
1.0
o (I
0.5
1.0
Kay Fig. 1. Gel permeation profiles of CGRP-Iike immunoreactivity (CGRP-LI) from extracts of rat gastrointestinal and brain tissues. In each case aliquots from five extracts were pooled and centrifuged before loading onto the column. K,, was calculated as ( V ¢ - V o ) / ( V t - Vo) where II. = elution volume, Vo = void volume and cf2 Vt ffi total volume. Synthetic CGRP was eluted in the position shown with a K,, of 0.37.
137
was found throughout the gastrointestinal tract with particularly high concentrations occurring in the pyloric sphincter, descending colon and rectum. A moderate level of CGRP-Iike immunoreactivity was found in the pancreas.
2. Characterisation of CGRP-Iike immunoreactivity Gel filtration profiles of pooled extracts from various regions of gastrointestinal tract and brain (Fig. 1) all showed a peak of CGRP-Iike immunoreactivity which was eluted in the position corresponding to synthetic CGRP, having a K,~ of 0.37 4- 0.03 (n = 5) on Sephadex G-50 superfine. In the gut and pancreas two smaller peaks with K,~ values of 0.55 + 0.03 and 0.70 4- 0.02 (n = 5) were also evident, particularly in the pancreas where they constituted approximately 30% of the total CGRP-LI (fmol/frac~on}
%Acetonitrile
a
b
c
CGRP
700-
Colon
600-
60 /
500-
/
/i
50
/J
400300-
.80
200-
.20
A 700-
-10
Medulla Oblongata
60D-
1 1 ~ -ri0
//
500-
I/ I
4(10-
~
/
-50
/
-40
"
1t10.
-30
2C0-
-20
i.
"10
~
m 20
[]i ~
~
so
6o
7b
~o
Retention Time {minutes) Fig. 2. HPLC profiles of CGRP-like immunoreactivity from rat colon and medulla oblongata. In each case aliquots from five extracts were pooled and prepared for injection onto the column as described in Methods. The column was eluted with an acetonitrilegradient, (as shown) between two solvent systems. A: distilledwater containing 0.1% (v/v) trifluoroacticacid (TFA); B: acetonitrilecontaining 0.1% (v/v) TFA. Synthetic CGRP was eluted in the position shown with 59 min retention time. Colonic extract HPLC fractions containing CGRP-Iike immunoreactivity from the peaks marked (a): retention time 2023 rain, (b): retention time 25-29 min and (c):retention time 49-57 re.inwere evaporated under vacuum to approximately half their original volume. Fractions from each peak were then pooled and loaded onto Scphadex G-50 superfine columns as described in Methods. The immunoreactive material from the various peaks was cluted with the following K~v values: (a) 0.70, (b) 0.$5, (c) 0.37.
138 immunoreactivity (estimated by measuring the area under each peak). In the stomach, duodenum and colon they were not so prominent as in the pancreas and in the brain they were not readily discernable. Reverse-phase HPLC also showed that the major component of CGRP-Iike immunoreactivity in both the gut and brain was indistinguishable from the synthetic peptide (Fig. 2). HPLC analysis of gut extracts revealed at least two additional peaks which were eluted earlier, indicating the presence of less hydrophobic cross-reacting species. These same peaks were also observed on HPLC profiles of extracts of medulla oblongata but were not so prominent in comparison with the major peak. The co-identity of peaks on HPLC and gel-permeation profiles was investigated by loading immunoreactive material from each HPLC peak separately onto G-50 columns. The material from the earliest HPLC peak was eluted with a Kay of 0.70 and from the second HPLC peak with a K~v of 0.55 (see Fig. 2). In the assay, dilution curves of both components were parallel to that for the synthetic CGRP standard. Recovery of CGRP added to tissue extracts and reboiled was 95.0 4- 2.5% for stomach and 85.4 + 3.1% for brain stem (mean + S.E.M., n = 3 each). As previously reported [9] no alternative molecular forms were produced during boiling.
Fig. 3. A dense innervation of the central nucleus of the amygdala by CGRP-immunoreactive fibres. No positively stained cell bodies are apparent, but several cells are closely surrounded by CGRP-containing fibres (examples arrowed) ( × 350).
Fig. 4. (Top) The muscularis fibres are seen mainly
externa
of the rat small intestine
in the myenteric
immunostained
plexus (MP) and in the submucosa
for CGRP.
Immunoreactive
(SM) ( x 150).
(Bottom) Higher power micrograph of the circular muscle of rat small intestine noreactive nerves intermingling with muscle fibres ( x 250).
showing
CGRP-immu-
140
3. Immunocytochemistry Several immunoreactive cell bodies were visualised in the brain stem, particularly in the large motoneurones of the facial nucleus and in smaller neurones of the nucleus ambiguus and parabrachial nucleus. In more rostral regions CGRP-immunoreactive fibres were observed, particularly in the vicinity of the central nucleus of the amygdala (Fig. 3) and the lateral regions of the caudate-putamen. Throughout the gut, CGRP-immunoreactivity was restricted to nerve fibres and could not be detected in neuronal cell bodies of either of the main ganglionated plexuses. Immunoreactive fibres were scattered in all layers of the gut wall but showed a particularly frequent association with blood vessels. The distribution pattern was fairly consistent at each level of the gut. In the myenteric plexus, loose meshes of CGRP-immunoreactive fibres were seen around non-immunoreactive ganglion cells. In the submucosa, immunoreactive nerves were seen mainly around blood vessels and in the submucous plexus, where they surrounded non-immunoreactive ganglion cells (Fig. 4 top). The circular muscle appeared to contain the largest number of immunostained fibres (Fig. 4 bottom). A few fibres were scattered in the muscularis mucosae and mucosa, especially in the stomach. In the pancreas, CGRP immunoreactive nerves were present around blood vessels, scattered in the exocrine parenchyma and were frequently found in the islets (Fig. 5 and 6).
Fig. 5. CGRP immunoreactive nerve fibres (arrows) in close association with a vessel in pancreatic connective tissue ( × 350).
141
Fig. 6. A pancreatic islet containing CGRP-immunoreaetive nerves (arrows) ( × 450).
Discussion
The results of this study provide the first quantitative information on the occurrence of CGRP in the brain and gut of the rat and confirm previous reports that the major component of CGRP-like immunoreactivity found in the tissues corresponds to the predicted structure of the CGRP molecule [2]. In addition we have demonstrated in gut and pancreatic extracts, and at much lower levels in brain extracts the presence of cross-reacting materials of smaller molecular weight and lesser hydrophobicity. The findings of the immunocytochemical study are in agreement with the observations of Rosenfeld et al. [2]. We have already reported the concentrations of CGRP in the trigeminal ganglion as 44.0 + 8.1 pmol/g [13] and in the oesophagus as 15.8 + 2.4 pmol/g in the epithelium and 11.2 4- 1.8 pmol/g in the muscle layer [14]. The distribution of CGRP in the brain as measured by radioimmunoassay in this study is consistent with the arrangement of CGRP-containing neural systems described on the basis of immunostaining results obtained both in this study and by Rosenfeld et al. [2] who reported CGRP-stained cell groups in several nuclei of the brain stem and pathways originating in the parabrachial and peripenduncular nuclei and projecting to the hypothalamus, amygdala, caudate-putamen, globus pallidus, septal nuclei and thalamic taste nucleus. This group also described CGRP-containing
142
fibres in the anterior pituitary which did not originate in the brain. The levels of CGRP-Iike immunoreactivity measured in the pituitary by RIA in the present study suggest that this innervation may be of importance. Similar concentrations of CGRP-Iike immunoreactivity have also been reported in the human pituitary [20]. The possible physiological function of CGRP in the gut is not known but it has been shown to inhibit gastric acid secretion in rats, dogs and man [15-18], to cause a dose-dependent contraction of guinea pig ileal and colonic smooth muscle [19]. Intravenous infusion of the peptide in rats was found to stimulate colonic water and electrolyte secretion (R.K. Rolston, unpublished observations). The close association of CGRP-containing nerve fibres with blood vessels in the gut, as reported here and previously [2], and the vasodilatory actions of the peptide [7] suggest a possible role in gastrointestinal blood flow regulation. Hence it is possible that CGRP-containing neurones in the gut could be important in regulating gastrointestinal function. The absence of CGRP immunoreactive endocrine cells suggests that CGRP is more likely to be of importance as a neuropeptide than as a circulating gut hormone. To date, significant concentrations of CGRP circulating in plasma have only been demonstrated in patients with CGRP-producing tumours [3]. Our observations on the characterisation of CGRP-Iike immunoreactivity in brain extracts are in agreement with the findings of Rosenfeld et al. [2]. It is not yet clear whether the small molecular forms of CGRP-like immunoreactivity are fragments of the whole CGRP molecule or are the result of antibody cross-reactivity with independent molecular species. Their significance is completely unknown but the fact that they are relatively more abundant in some tissues than others indicates that further study may be worthwhile. It is unlikely that these peaks are artefacts of the extraction procedure as prolonged boiling of synthetic CGRP in acetic acid did not lead to the appearance of any alternative molecular forms [9]. Mason et al. [8] have demonstrated the release of CGRP, corresponding to the 37 amino acid synthetic standard from cultured rat trigeminal ganglion cells. This finding seems to favour a neurotransmitter role for the large peptide and one might hypothesise that it serves the same function in the gut and other peripheral tissues as in the brain.
Acknowledgements This work was supported by the Medical Research Council (U.K.) and the British Diabetic Association. We wish to thank Dr. H. Suzuki for assistance with the brain dissection and J.P. Chatterton for typing the manuscript.
References 1 Amara, S.G., Jonas, V., Rosenfeld, M.G., Ong, E.S. and Evans, R.M., Alternative RNA processing in calcitonin gene expression generates mRNAs encoding different polypeptide products, Nature, 298 (1982) 240-244. 2 Rosenfeld, M.G., Mermod, J.J., Amara, S.G., Swanson, L.W., Sawchenko, P.E., Rivier, J., Vale, W.W. and Evans, R.M., Production of novel neuropeptide encoded by the calcitonin gene via tissue specific RNA processing, Nature, 304 (1983) 129-135.
143 3 Morris, H.R., Panico, M., Etienne, T., Tippins, J., Girgls, S.I. and McIntyre, I., Isolation and characterisation of human calcitonin gene-related peptide, Nature, 308 (1984) 746-748. 4 Steenbergh, P.H., Hoppener, J.W.M., Zandbcrg, J., Van de Ven, W.J.M., Jansz H.S. and Lips C.J.M., Calcitonin gene-related peptide coding sequence is conserved in the human genome and is expressed in medullary thyroid carcinoma, J. Clin. Endocrinol. Metab., 59 (1984) 358-360. 5 Nelkin, B.D., Rosenfeld, K.I., de Bustros, A., Leong, S.S., Roos, B.A. and Baylin, S.B., Structure and expression of a gene encoding human calcitonin and calcitonin gene-related peptide, Biochem. Biophys. Res. Commun., 123 (1984) 648-655. 6 Gibson, S.J., Polak, J.M., Bloom, S.R., Sabate, I.M., Mulderry, P.K., Ghatei, M.A., McGregor, G.P., Morrison, J.F.B., Kelly, J.S. and Rosenfdd, M.G., Calcitonin gene-related peptide in the spinal cord of man and eight other species, J. Neurosci., 4 (1984) 3101-3111. 7 Brain, S.D., Williams, T.J., Tippins, J.R., Morris, H.R. and MacIntyre, I., Calcitonin gene-related peptide is a potent vasodilator, Nature, 313 (1984) 54-56. 8 Mason, R.T., Peterfreund, R.A., Sawchenko, P.E., Corrigan, A.Z., Rivier, J.E. and Vale, W.W., Release of the predicted calcitonin gene-related peptide from cultured rat trigeminal ganglion cells, Nature, 308 (1984) 653-655. 9 Mulderry, P.K., Ghatei, M.A., Rodrigo, J., Allen, J.M., Rosenfeld, M.G., Polak, J.M. and Bloom, S.R., Calcitonin gene-related peptide in cardiovascular tissues of the rat, Neuroscience, 14 (1985) 947-954. 10 Bryant, M.G. and Bloom, S.R., Measurement in tissue. In S.R. Bloom and R.G. Long (eds.), Radioimmunoassay of Gut Regulatory Peptides, W.B. Saunders, London, 1982, pp. 36-41. 11 Bishop, A.E., Polak, J.M., Bloom, S.R., Pearse, A.G.E., A new universal technique for the immunocytochemical localisation of peptiderglc innervation, J. Endocrinol., 77 (1978) 25-26. 12 Huang, W.M., Gibson, S.J., Facer, P., Gu, J., Polak, J.M., Improved section adhesion for immunocytochemistry using high molecular weight polymers of L-lysineas a slide coating, Histochemistry, 77 (1983) 275-279. 13 Terenghi, G., Polak, J.M., Ghatei, M.A., Mulderry, P.K., Butler, J.M., Unger, W.G. and Bloom, S.R., Distribution and origin of calcitonin gene-related peptide (CGRP) immunoreactivity in the sensory innervation of the mammalian eye, J. Comp. Neurol., 233 (1985) 506-516. 14 Rodrigo, J., Polak, J.M., Fernandez, L., Ghatei, M.A., Mulderry, P.K., Bloom, S.R. and Rosenfeld, M.G., CGRP-immunoreactive sensory and motor nerves of the mammalian oesophagus, Gastroenterology, 88 (1985) 444-451. 15 Tache, Y., Pappas, T., Lauffenburger, M., Goto, Y., Walsh, J.H. and Debas, H., Calcitonin generelated peptide: potent peripheral inhibitor of gastric acid secretion in rats and dogs, Gastroenterology, 87 (1984) 344-9. 16 Hughes, J.J., Levine, A.S., Morley, J.E., Gosnell, B.A. and Silvis, S.E., Intraventricular calcitonin gene-related peptide inhibits gastric acid secretion, Peptides, 5 (1984) 665-667. 17 Lenz,H.J., Mortrud, M.T., Yale, W.W., Rivier, J.E. and Brown, M.R., Calcitonin gene-related peptide acts within the central nervous system to inhibit gastric acid secretion, Regul. Peptides, 9 (1984) 271-277. 18 Kraenzlin, M.E., Ch'ng, J.L.C., Mulderry, P.K., Ghatei, M.A. and Bloom, S.R., Infusion of a novel peptide, calcitonin gene-related peptide (CGRP) in man. Pharmacokinetics and effects on gastric acid secretion and on gastrointestinal hormones, Regul. Peptides, 10 (1985) 189-197. 19 Ghatei, M.A., Christofides, N.D., Bishop, A.E., Mulderry, P.K., Bennett, A., Polak, J.M. and Bloom, S.R., Distribution and effect of calcitonin gene-related peptide in the gastrointestinal tract of the guinea pig, Regul. Peptides, 9 (1984) 330 (Abstract). 20 Tschopp, F.A., Tobler, P.H. and Fischer, J.A., Calcitonin gene-related peptide in the human thyroid, pituitary and brain, Mol. Cell. Endocrinol., 36 (1984) 53-57.
133
R P T 00410
Distribution and chromatographic characterisation of CGRP-like immunoreactivity in the brain and gut of the rat P.K. Mulderry, M.A. Ghatei, A.E. Bishop, Y.S. Allen, J.M. Polak and S.R. Bloom Departments of Medicine and Histochemistry, Royal Postgraduate Medical School, Du Cane Road, London, W12 OHS, U.K. (Received 27 March 1985; revised manuscript received 8 July 1985; accepted for publication 9 July 1985)
Summary Radioimmunoassay, chromatography and immunocytochemistry were used to study the occurrence of calcitonin gene-related peptide in the brain and gastrointestinal tract of the rat. In the brain, the highest concentrations of the peptide were found in the medulla oblongata (58.3 + 6.8 pmol/g) where immunocytochemistry showed the presence of immunoreactive cell bodies. Significant concentrations were also found in the pancreas and throughout the gastrointestinal tract, the highest levels occurring in the pyloric sphincter (48.0 4- 6.0 pmol/g). CGRP-like immunoreactivity in the gastrointestinal tract was restricted to nerve fibres. Chromatographic analysis of the CGRP-Iike immunoreactivity occurring in these tissues showed that at least 7 0 0 was indistinguishable from the synthetic peptide. However, there was also evidence of a number of smaller cross-reacting molecular species. CGRP; brain; gut; radioimmunoassay; chromatography; immunocytochemistry
Introduction Structural analyses of the rat calcitonin gene have led to predictions of the existence of a 37 amino acid calcitonin gene-related peptide (CGRP), which is coded on the Address for correspondence: Professor S.R. Bloom, Department of Medicine, Royal Postgraduate Medical School, Du Cane Road, London WI2 0HS, U.K. Tel. 01-743-2030 Ext. 274. 0167-0115/85/$03.30 © 1985 Elsevier Science Publishers B.V. (Biomedical Division)
134
same gene as calcitonin but which is translated from a separate mRNA produced by alternative splicing events during expression of the gene [1]. The presence of CGRP mRNA and immunoreactive CGRP has subsequently been demonstrated in neural tissue [2], hence CGRP has been cited as putative neuropeptide. Studies of human medullary thyroid carcinoma tissues have shown that CGRP mRNA and a peptide whose amino acid sequence differs from that of rat CGRP in only four positions are produced [3-5]. CGRP-like immunoreactivity has been reported in the spinal cords of the marmoset, horse, pig, cat, guinea pig, mouse, frog, rat and man [6], and both rat- and human-sequence CGRPs have been shown to be potent vasodilators [7]. CGRP-like immunoreactivities occurring in the hypothalamus [2], trigeminal ganglia [8,13] and blood vessels [9] have been characterised by chromatography. The major component of CGRP-like immunoreactivity was found to correspond in elution characteristics to the peptide synthesised according to the predicted CGRP sequence. Immunohistochemical studies have demonstrated a system of CGRP pathways in the brain and the presence of CGRP-containing fibres in the gut of the rat [2], although the distribution of the immunoreactivity in the gut has not yet been described nor has its chromatographic identity with synthetic CGRP been verified. In this study we have investigated the distribution and localisation of CGRP-like immunoreactivity in the brain and gastrointestinal tract of the rat. In addition, CGRP-Iike immunoreactivity occurring in these tissues was characterised by chromatography.
Materials and Methods
Tissues All specimens were taken from adult male Wistar rats. For radioimmunoassay, animals were killed by cervical dislocation, dissected and the tissues extracted by boiling in 0.5 M acetic acid as previously described [10]. For immunocytoebemistry, tissues were fixed either by intracardial perfusion with 4% paraformaldehyde in phosphate-buffered saline (PBS) or by direct immersion of specimens in 0.4% benzoquinone in PBS as described by Bishop et al. [11]. Radioirnmunoassay Tissue extracts were assayed for CGRP as duplicate 20-#1 aliquots in 0.8 ml 0.06 M phosphate buffer, pH 7.4 containing 0.01 M EDTA. 0.05% (w/v) sodium azide and 3% (w/v) bovine serum albumin (BSA). The CGRP antiserum was raised in a New Zealand white rabbit by injecting with synthetic rat CGRP coupled to bovine serum albumin by a glutaraldehyde reaction. The antiserum could be used at a final dilution of 1:400 000 and showed no cross-reactivity (less than 0.02%) with calcitonin, somatostatin-28, substance P, VIP or gastrin-releasing peptide. Radiolabelled CGRP was prepared by a chloramine-T iodination of the histidine residue in synthetic rat CGRP (Peninsula Laboratories) and purified by reverse-phase high pressure liquid chromatography (HPLC) as previously described [18]. The specific activity obtained was 58.0 Bq/fmol as determined by self displacement in the assay. Bound and free
135 fractions were separated after incubation by charcoal adsorption of the free fraction. The assay standard was Peninsula synthetic rat C G R P of which 1 fmol per assay tube could be detected with 95% confidence.
Chromatography The CGRP-Iike immunoreactivity was characterised by gel filtration and HPLC. For gel filtration, pooled tissue extracts were loaded onto Sephadex G-50 superfine columns (60 x 0.9 cm) eluted with 0.06 M phosphate buffer, pH 7.4, containing 0.3% BSA and 0.2 M NaCI at a flow rate of 3.6 ml/h. Before loading, each sample was spiked with blue dextran and Na125I as internal markers of void volume (Iio) and total volume (lit), respectively. Fractions were collected every 15 min. HPLC was carried out on a Techsil C-18 reverse-phase column (HPLC Technology) eluted at 1 ml/min with an acetonitrile gradient (see Fig. 2). Prior to loading, samples were purified on Waters Sep-Pak C-18 reverse-phase sample preparation cartridges. After washing with 10 ml 0.1% aqueous trifluoroacetic acid (TFA), peptides were eluted off with 0.5 ml 60% acetonitrile, 0.1% TFA and diluted to 2.5 ml with water containing 0.1% TFA. A volume of 2 ml was injected onto the HPLC column and the remainder assayed for assessment of recovery. The concentration of CGRP-like immunoreactivity in all column fractions was determined by assaying 100-/A aliquots directly.
Immunocytochemistry After thorough washing in PBS containing 15% sucrose and 0.01% NAN3, cryostat blocks were prepared and 10-#m sections cut at - 2 0 ° C and picked up onto glass slides coated with poly-L-lysine [12]. After being allowed to dry in the air at room temperature for at least 1 h, the sections were immunostained using the technique of indirect immunofluorescence (gastrointestinal tracO or peroxidase-anti-peroxidase (brain). The primary antiserum was the same as that used for radioimmunoassay, but applied at a dilution of 1:200 for immunofluorescence and 1:2000 for peroxidase-anti-peroxidase. Controls for immunostaining included omission of the first layer antiserum or its replacement with non-immune rabbit serum. Neither of these gave positive immunostaining. In addition, immunostaining with the anti-CGRP primary serum was successfully quenched by the addition of as little as 0.1 nmol of synthetic CGRP per ml of diluted antiserum.
Experimental results
1. Distributlon of CGRP-Iike immunoreactivity CGRP concentrations in the tissues studied, as determined by radioimmunoassay, are given in Table I. Significant levels of CGRP-Iike immunoreactivity were found in all regions of the brain apart from the cerebellum. The highest concentration occurred in the medulla oblongata with relatively high concentrations also in the amygdala, hypothalamus, septal area and pituitary. CGRP-like immunoreactivity
136 TABLE I Concentrations of CGRP found in regions of rat brain and gastrointestinal tract Brain region
pmol/g
Frontal cortex Parietal cortex Occipital cortex Caudate-putamen Nucleus accumbens Olfactory bulb Septum Hippocampus Thalamus Hypothalamus Pituitary Globus pallidus Amygdala Mid brain Periaqueductal grey Cerebellum Pons Medulla oblongata
0.9 1.4 2.1 1.4 5.9 2.6 10.7 0.8 6.0 9.6 10.7 7.7 19.7 4.4 6.0 < 0.5 6,4 58.3
+ 4+ 4444444± 444+
0.1 0.6 0.5 0.4 1.2 0.8 0.9 0.2 0.7 1.3 1.8 2.3 2.3 0.4 0.8
Gastrointestinal tract
pmol/g
Lower oesophageal sphincter Forestomach Corpus Antrum Pyloric sphincter Duodenum Jejunum Ileum Caecum Ascending colon Mid-colon Descending colon Rectum Pancreas
16.8 35.2 12.5 39,1 48.0 21.5 17.5 11.5 9.8 9.4 11.6 28.1 35.1 5.6
444± 444± + 444± 4-
1.9 4.1 2,8 4,9 6.0 2.1 2.1 0.9 1.8 0.8 2.6 2.5 3.6 0.8
4- 0.9 4- 6.8
Values given are mean 4- S.E.M. (n = 5).
CGRP-LI fmol/fraction
CGRP
CGRP
25O
100
125"
50
2S0
CGRP 150
~
l
Medulla Oblongata
75
Ioo
150
Colon
Pancreas
125
othalamus
75
......... Jla ......... i --=-
0
0.5
1.0
0
._--J
..... ::::~ 0.5
1.0
o (I
0.5
1.0
Kay Fig. 1. Gel permeation profiles of CGRP-Iike immunoreactivity (CGRP-LI) from extracts of rat gastrointestinal and brain tissues. In each case aliquots from five extracts were pooled and centrifuged before loading onto the column. K,, was calculated as ( V ¢ - V o ) / ( V t - Vo) where II. = elution volume, Vo = void volume and cf2 Vt ffi total volume. Synthetic CGRP was eluted in the position shown with a K,, of 0.37.
137
was found throughout the gastrointestinal tract with particularly high concentrations occurring in the pyloric sphincter, descending colon and rectum. A moderate level of CGRP-Iike immunoreactivity was found in the pancreas.
2. Characterisation of CGRP-Iike immunoreactivity Gel filtration profiles of pooled extracts from various regions of gastrointestinal tract and brain (Fig. 1) all showed a peak of CGRP-Iike immunoreactivity which was eluted in the position corresponding to synthetic CGRP, having a K,~ of 0.37 4- 0.03 (n = 5) on Sephadex G-50 superfine. In the gut and pancreas two smaller peaks with K,~ values of 0.55 + 0.03 and 0.70 4- 0.02 (n = 5) were also evident, particularly in the pancreas where they constituted approximately 30% of the total CGRP-LI (fmol/frac~on}
%Acetonitrile
a
b
c
CGRP
700-
Colon
600-
60 /
500-
/
/i
50
/J
400300-
.80
200-
.20
A 700-
-10
Medulla Oblongata
60D-
1 1 ~ -ri0
//
500-
I/ I
4(10-
~
/
-50
/
-40
"
1t10.
-30
2C0-
-20
i.
"10
~
m 20
[]i ~
~
so
6o
7b
~o
Retention Time {minutes) Fig. 2. HPLC profiles of CGRP-like immunoreactivity from rat colon and medulla oblongata. In each case aliquots from five extracts were pooled and prepared for injection onto the column as described in Methods. The column was eluted with an acetonitrilegradient, (as shown) between two solvent systems. A: distilledwater containing 0.1% (v/v) trifluoroacticacid (TFA); B: acetonitrilecontaining 0.1% (v/v) TFA. Synthetic CGRP was eluted in the position shown with 59 min retention time. Colonic extract HPLC fractions containing CGRP-Iike immunoreactivity from the peaks marked (a): retention time 2023 rain, (b): retention time 25-29 min and (c):retention time 49-57 re.inwere evaporated under vacuum to approximately half their original volume. Fractions from each peak were then pooled and loaded onto Scphadex G-50 superfine columns as described in Methods. The immunoreactive material from the various peaks was cluted with the following K~v values: (a) 0.70, (b) 0.$5, (c) 0.37.
138 immunoreactivity (estimated by measuring the area under each peak). In the stomach, duodenum and colon they were not so prominent as in the pancreas and in the brain they were not readily discernable. Reverse-phase HPLC also showed that the major component of CGRP-Iike immunoreactivity in both the gut and brain was indistinguishable from the synthetic peptide (Fig. 2). HPLC analysis of gut extracts revealed at least two additional peaks which were eluted earlier, indicating the presence of less hydrophobic cross-reacting species. These same peaks were also observed on HPLC profiles of extracts of medulla oblongata but were not so prominent in comparison with the major peak. The co-identity of peaks on HPLC and gel-permeation profiles was investigated by loading immunoreactive material from each HPLC peak separately onto G-50 columns. The material from the earliest HPLC peak was eluted with a Kay of 0.70 and from the second HPLC peak with a K~v of 0.55 (see Fig. 2). In the assay, dilution curves of both components were parallel to that for the synthetic CGRP standard. Recovery of CGRP added to tissue extracts and reboiled was 95.0 4- 2.5% for stomach and 85.4 + 3.1% for brain stem (mean + S.E.M., n = 3 each). As previously reported [9] no alternative molecular forms were produced during boiling.
Fig. 3. A dense innervation of the central nucleus of the amygdala by CGRP-immunoreactive fibres. No positively stained cell bodies are apparent, but several cells are closely surrounded by CGRP-containing fibres (examples arrowed) ( × 350).
Fig. 4. (Top) The muscularis fibres are seen mainly
externa
of the rat small intestine
in the myenteric
immunostained
plexus (MP) and in the submucosa
for CGRP.
Immunoreactive
(SM) ( x 150).
(Bottom) Higher power micrograph of the circular muscle of rat small intestine noreactive nerves intermingling with muscle fibres ( x 250).
showing
CGRP-immu-
140
3. Immunocytochemistry Several immunoreactive cell bodies were visualised in the brain stem, particularly in the large motoneurones of the facial nucleus and in smaller neurones of the nucleus ambiguus and parabrachial nucleus. In more rostral regions CGRP-immunoreactive fibres were observed, particularly in the vicinity of the central nucleus of the amygdala (Fig. 3) and the lateral regions of the caudate-putamen. Throughout the gut, CGRP-immunoreactivity was restricted to nerve fibres and could not be detected in neuronal cell bodies of either of the main ganglionated plexuses. Immunoreactive fibres were scattered in all layers of the gut wall but showed a particularly frequent association with blood vessels. The distribution pattern was fairly consistent at each level of the gut. In the myenteric plexus, loose meshes of CGRP-immunoreactive fibres were seen around non-immunoreactive ganglion cells. In the submucosa, immunoreactive nerves were seen mainly around blood vessels and in the submucous plexus, where they surrounded non-immunoreactive ganglion cells (Fig. 4 top). The circular muscle appeared to contain the largest number of immunostained fibres (Fig. 4 bottom). A few fibres were scattered in the muscularis mucosae and mucosa, especially in the stomach. In the pancreas, CGRP immunoreactive nerves were present around blood vessels, scattered in the exocrine parenchyma and were frequently found in the islets (Fig. 5 and 6).
Fig. 5. CGRP immunoreactive nerve fibres (arrows) in close association with a vessel in pancreatic connective tissue ( × 350).
141
Fig. 6. A pancreatic islet containing CGRP-immunoreaetive nerves (arrows) ( × 450).
Discussion
The results of this study provide the first quantitative information on the occurrence of CGRP in the brain and gut of the rat and confirm previous reports that the major component of CGRP-like immunoreactivity found in the tissues corresponds to the predicted structure of the CGRP molecule [2]. In addition we have demonstrated in gut and pancreatic extracts, and at much lower levels in brain extracts the presence of cross-reacting materials of smaller molecular weight and lesser hydrophobicity. The findings of the immunocytochemical study are in agreement with the observations of Rosenfeld et al. [2]. We have already reported the concentrations of CGRP in the trigeminal ganglion as 44.0 + 8.1 pmol/g [13] and in the oesophagus as 15.8 + 2.4 pmol/g in the epithelium and 11.2 4- 1.8 pmol/g in the muscle layer [14]. The distribution of CGRP in the brain as measured by radioimmunoassay in this study is consistent with the arrangement of CGRP-containing neural systems described on the basis of immunostaining results obtained both in this study and by Rosenfeld et al. [2] who reported CGRP-stained cell groups in several nuclei of the brain stem and pathways originating in the parabrachial and peripenduncular nuclei and projecting to the hypothalamus, amygdala, caudate-putamen, globus pallidus, septal nuclei and thalamic taste nucleus. This group also described CGRP-containing
142
fibres in the anterior pituitary which did not originate in the brain. The levels of CGRP-Iike immunoreactivity measured in the pituitary by RIA in the present study suggest that this innervation may be of importance. Similar concentrations of CGRP-Iike immunoreactivity have also been reported in the human pituitary [20]. The possible physiological function of CGRP in the gut is not known but it has been shown to inhibit gastric acid secretion in rats, dogs and man [15-18], to cause a dose-dependent contraction of guinea pig ileal and colonic smooth muscle [19]. Intravenous infusion of the peptide in rats was found to stimulate colonic water and electrolyte secretion (R.K. Rolston, unpublished observations). The close association of CGRP-containing nerve fibres with blood vessels in the gut, as reported here and previously [2], and the vasodilatory actions of the peptide [7] suggest a possible role in gastrointestinal blood flow regulation. Hence it is possible that CGRP-containing neurones in the gut could be important in regulating gastrointestinal function. The absence of CGRP immunoreactive endocrine cells suggests that CGRP is more likely to be of importance as a neuropeptide than as a circulating gut hormone. To date, significant concentrations of CGRP circulating in plasma have only been demonstrated in patients with CGRP-producing tumours [3]. Our observations on the characterisation of CGRP-Iike immunoreactivity in brain extracts are in agreement with the findings of Rosenfeld et al. [2]. It is not yet clear whether the small molecular forms of CGRP-like immunoreactivity are fragments of the whole CGRP molecule or are the result of antibody cross-reactivity with independent molecular species. Their significance is completely unknown but the fact that they are relatively more abundant in some tissues than others indicates that further study may be worthwhile. It is unlikely that these peaks are artefacts of the extraction procedure as prolonged boiling of synthetic CGRP in acetic acid did not lead to the appearance of any alternative molecular forms [9]. Mason et al. [8] have demonstrated the release of CGRP, corresponding to the 37 amino acid synthetic standard from cultured rat trigeminal ganglion cells. This finding seems to favour a neurotransmitter role for the large peptide and one might hypothesise that it serves the same function in the gut and other peripheral tissues as in the brain.
Acknowledgements This work was supported by the Medical Research Council (U.K.) and the British Diabetic Association. We wish to thank Dr. H. Suzuki for assistance with the brain dissection and J.P. Chatterton for typing the manuscript.
References 1 Amara, S.G., Jonas, V., Rosenfeld, M.G., Ong, E.S. and Evans, R.M., Alternative RNA processing in calcitonin gene expression generates mRNAs encoding different polypeptide products, Nature, 298 (1982) 240-244. 2 Rosenfeld, M.G., Mermod, J.J., Amara, S.G., Swanson, L.W., Sawchenko, P.E., Rivier, J., Vale, W.W. and Evans, R.M., Production of novel neuropeptide encoded by the calcitonin gene via tissue specific RNA processing, Nature, 304 (1983) 129-135.
143 3 Morris, H.R., Panico, M., Etienne, T., Tippins, J., Girgls, S.I. and McIntyre, I., Isolation and characterisation of human calcitonin gene-related peptide, Nature, 308 (1984) 746-748. 4 Steenbergh, P.H., Hoppener, J.W.M., Zandbcrg, J., Van de Ven, W.J.M., Jansz H.S. and Lips C.J.M., Calcitonin gene-related peptide coding sequence is conserved in the human genome and is expressed in medullary thyroid carcinoma, J. Clin. Endocrinol. Metab., 59 (1984) 358-360. 5 Nelkin, B.D., Rosenfeld, K.I., de Bustros, A., Leong, S.S., Roos, B.A. and Baylin, S.B., Structure and expression of a gene encoding human calcitonin and calcitonin gene-related peptide, Biochem. Biophys. Res. Commun., 123 (1984) 648-655. 6 Gibson, S.J., Polak, J.M., Bloom, S.R., Sabate, I.M., Mulderry, P.K., Ghatei, M.A., McGregor, G.P., Morrison, J.F.B., Kelly, J.S. and Rosenfdd, M.G., Calcitonin gene-related peptide in the spinal cord of man and eight other species, J. Neurosci., 4 (1984) 3101-3111. 7 Brain, S.D., Williams, T.J., Tippins, J.R., Morris, H.R. and MacIntyre, I., Calcitonin gene-related peptide is a potent vasodilator, Nature, 313 (1984) 54-56. 8 Mason, R.T., Peterfreund, R.A., Sawchenko, P.E., Corrigan, A.Z., Rivier, J.E. and Vale, W.W., Release of the predicted calcitonin gene-related peptide from cultured rat trigeminal ganglion cells, Nature, 308 (1984) 653-655. 9 Mulderry, P.K., Ghatei, M.A., Rodrigo, J., Allen, J.M., Rosenfeld, M.G., Polak, J.M. and Bloom, S.R., Calcitonin gene-related peptide in cardiovascular tissues of the rat, Neuroscience, 14 (1985) 947-954. 10 Bryant, M.G. and Bloom, S.R., Measurement in tissue. In S.R. Bloom and R.G. Long (eds.), Radioimmunoassay of Gut Regulatory Peptides, W.B. Saunders, London, 1982, pp. 36-41. 11 Bishop, A.E., Polak, J.M., Bloom, S.R., Pearse, A.G.E., A new universal technique for the immunocytochemical localisation of peptiderglc innervation, J. Endocrinol., 77 (1978) 25-26. 12 Huang, W.M., Gibson, S.J., Facer, P., Gu, J., Polak, J.M., Improved section adhesion for immunocytochemistry using high molecular weight polymers of L-lysineas a slide coating, Histochemistry, 77 (1983) 275-279. 13 Terenghi, G., Polak, J.M., Ghatei, M.A., Mulderry, P.K., Butler, J.M., Unger, W.G. and Bloom, S.R., Distribution and origin of calcitonin gene-related peptide (CGRP) immunoreactivity in the sensory innervation of the mammalian eye, J. Comp. Neurol., 233 (1985) 506-516. 14 Rodrigo, J., Polak, J.M., Fernandez, L., Ghatei, M.A., Mulderry, P.K., Bloom, S.R. and Rosenfeld, M.G., CGRP-immunoreactive sensory and motor nerves of the mammalian oesophagus, Gastroenterology, 88 (1985) 444-451. 15 Tache, Y., Pappas, T., Lauffenburger, M., Goto, Y., Walsh, J.H. and Debas, H., Calcitonin generelated peptide: potent peripheral inhibitor of gastric acid secretion in rats and dogs, Gastroenterology, 87 (1984) 344-9. 16 Hughes, J.J., Levine, A.S., Morley, J.E., Gosnell, B.A. and Silvis, S.E., Intraventricular calcitonin gene-related peptide inhibits gastric acid secretion, Peptides, 5 (1984) 665-667. 17 Lenz,H.J., Mortrud, M.T., Yale, W.W., Rivier, J.E. and Brown, M.R., Calcitonin gene-related peptide acts within the central nervous system to inhibit gastric acid secretion, Regul. Peptides, 9 (1984) 271-277. 18 Kraenzlin, M.E., Ch'ng, J.L.C., Mulderry, P.K., Ghatei, M.A. and Bloom, S.R., Infusion of a novel peptide, calcitonin gene-related peptide (CGRP) in man. Pharmacokinetics and effects on gastric acid secretion and on gastrointestinal hormones, Regul. Peptides, 10 (1985) 189-197. 19 Ghatei, M.A., Christofides, N.D., Bishop, A.E., Mulderry, P.K., Bennett, A., Polak, J.M. and Bloom, S.R., Distribution and effect of calcitonin gene-related peptide in the gastrointestinal tract of the guinea pig, Regul. Peptides, 9 (1984) 330 (Abstract). 20 Tschopp, F.A., Tobler, P.H. and Fischer, J.A., Calcitonin gene-related peptide in the human thyroid, pituitary and brain, Mol. Cell. Endocrinol., 36 (1984) 53-57.