Calcitonin gene-related peptide and substance P in afferents to the upper gastrointestinal tract in the rat

Neuros('iencc Letter,v, 76 (1987) 151 156 Elsevier Scientitic Publishers Ireland Ltd.

15 I

NSL 04559

Calcitonin gene-related peptide and substance P in afferents to the upper gastrointestinal tract in the rat T. G r e e n and G.J. D o c k r a y .4,1R(',~'ecr('lorl' ('ontrol Group, Ph.vsiolo~zical l.ahoratory, Univ(,r.~ity ~!1 Liverpool, l.iveu?ool , {. K. (Received 29 December 1986" Accepted 16 January 1987) Key words:

Calcitonin gene-related peptide (CGRP); Stomach: Visceral afferent: Oesophagus: Rat

Combined immunohistochemistry and relrograde tracing was used to investigate the origin ¢~t"the sensory calcitonm gene-related peptide (CGRP) innervation of the rat stomach. Up to 85?;, of spinal gastric affcrenls (Te, Lt) contained CGRP immunoreactivity compared with <6% of vagal gastric affcrents. B5 comparison substance P immunoreactivity occurred in about 50% of spinal gastric afferents and less than 2% of vagal aflerents. The vagal afferents to the oesophagus were 14 and 26% substance P- and ('(}RPimmunoreactive respectively. The results suggest an important role t\~r CGRP in gastric spinal affcrents.

Calcitonin gene-related peptide (CGRP) is a 37 amino acid polypeptide derived from the calcitonin gene [1, 14]. It is widely distributed in the nervous system, including primary sensory neurones: most, possibly all, substance P-containing afferents also contain C G R P immunoreactivity, but there are other CGRP-containing afterents that do not contain substance P [4, 5, 9, 20]. Previous studies using combined immunohistochemistry and retrograde tracing have shown that about 50% of the spinal afferents projecting to the stomach contain substance P, compared with less than 1% of vagal afferents [3, 15]. In the present study we have used the same method to examine the CGRP-afferent innervation of the stomach and oesophagus. The results suggest that up to 85% of gastric spinal afferents contain CGRP. Adult male Sprague Dawley rats (180 250 g) were anaesthetised with sodium pentobarbitone (60 mg/kg, i.p,; Sagatal, May and Baker). Using aseptic techniques 5 15 Ill of a suspension of True blue in distilled water (5%, w/v) was injected at one (oesophagus, n = 6) or multiple (stomach, n = 7) sites using a Hamilton microsyringe [15, 16]. Care was taken to prevent leakage of the dye by leaving the needle in place tk)r up to I rain following injection and by swabbing the area thoroughly with saline. In control experiments, unilateral cervical vagotomy was per[k~rmed at the time of True blue injection (oesophagus, n = 2 : stomach, , = 2 ) , or the coeliac ganglion (through which spinal afferents pass to the stomach) was removed ( s t o m a c h , , = 2). ('orrcv~omh'm'c: G.J. Dockray, Department of Physiology, University of Liverpool, P.O. Box 147, Lixcrpool L69 3BX, U.K. 031)4-3940 87 S I)3.50 @ 1987 Elsevier Scientilic Publishers Ireland Ltd.

152 Two further animals were given unilateral injection of True blue into the connective tissue sheath surrounding the carotid artery and cervical vagus. Five days after injection, animals were perfused-tixed with 4% paraformaldehyde in phosphate-buffered saline (PBS). Dorsal root ganglia from levels CI Tj (oesophageal True blue) or T(, l_j (stomach), were removed, together with the nodose ganglia and tissue f'rom either oesophagus or stomach, and processed for indirect immunofluorescence. Sections were stained alternately with either antiserum L83 to substance P (1:500, C-terminal specific) or the C G R P antibody L271 (1:500). The latter was raised by immunizing with rat Tyr 27 36 C G R P conjugated to thyroglobulin with glutaraldehyde. The specificities of the antisera were established by preabsorption of immune sera with the original hapten (1 and 10 nmol/ml), and in the case of L271 with 10 and 100 nmol/ml of CCKS, substance P, F M R F amide and L P L R F amide. True blue-containing cells in each section taken through the ganglion were counted, and labelled cells that also had substance P-like or CGRP-like immunoreactivity were recorded. When True blue was injected into the ventral surface of the stomach labelled cells were found mainly in the left nodose ganglia (ratio left to right, 5:1) and were concentrated in the body and caudal (peripheral) pole of the ganglion. Labelling on that side was abolished by left unilateral cervical vagotomy. Substance P-like and CGRPlike immunoreactive cell bodies tended to be localized at the rostral pole. Not surprisingly therefore, few True blue-labelled cells contained substance P (I.8 _+0.9% (n = 5)) or CGRP-like immunoreactivity (5.7_+1.4% (n=4)) (Fig. I). True blue-labelled cells were abundant in spinal ganglia T~, La (Table I) with the greatest numbers in T~ IOf these, 55.5_+2.1% 0 I = 7 ) contained SP-LI and 82.1_+1.9% ( n = 5 ) contained CGRP-LI (Fig. 2). Removal of the coeliac ganglion abolished blue labelling in spinal ganglia, but labelling in the nodose ganglia was unaffected. Consistent with previous findings [2] CGRP-immunoreactive fibres in the stomach were mainly found in the submucosa, often in association with blood vessels. Two weeks after coeliac ganglionectomy there was a complete loss of C G R P immunoreactivity, indicating the extrinsic origin of the material. In contrast, coeliac ganglionectomy decreased substance P immunoreactivity in the mucosa and submucosa, with a complete loss of perivascular fibres, but had little effect on that in the muscle layers and myenteric plexus. Since there are substance P-immunoreactive cell bodies in the myenteric plexus of the stomach it is clear that most substance P is of intrinsic origin and only mucosal and submucosal fibres are extrinsic [15]. After injection of True blue into the oesophagus, labelled cells were seen bilaterally in the nodose ganglia where they were distributed throughout the caudal (peripheral) pole and mid region (Fig. 1). A higher proportion of the True blue cells contained CGRP-Iike immunoreactivity 26_+ 2% (n = 4), and substance P-like immunoreactivity 14+ 3% (n = 5) compared with vagal afferents to the stomach. Scattered True bluelabelled cells were also found in dorsal root ganglia. The highest numbers of cells were found in C4 and of these 39_+ 2% (n = 6) contained substance P immunoreactivity, and 70_+ 13% (n = 5) contained C G R P immunoreactivity. After unilateral cervical vagotomy, True blue-labelled cells in the nodose ganglia on the lesioned side were reduced but not abolished, consistent with the idea that both the cervical vagus and

153

C

,29i"~ Fig. 1. Photomontages of the left nodose ganglion after injection of True blue inlo tile ventral wall of lhe stomach. (a) True blue-labelledcells occur in the mid region and caudal pole, {b) C(}RP-imnmnorcacrive cells in tile same section and (c) substance P-immunoreactivc cells in an adjacent section. The pcptides occur predominantly in lhe rostral pole. Bar= 100 #m. Rostral, left: caudal, right.

superior laryngeal nerves serve the cervical oesophagus [7, 12], spinal ganglia labelling was unaffected. Injection of tracer into the connective tissue sheath s u r r o u n d i n g the vagus resulted in heavy labelling t h r o u g h o u t the nodose ganglion on that side, i.e. both caudal a n d rostral poles, p r e s u m a b l y due to non-specific uptake. The m a j o r finding o f this study is that C G R P occurs in a high p r o p o r t i o n (up to 85%) of spinal gastric afferents. Oesophageal C G R P has previously been shown by

154 FABLE 1 THE CGRP AND SUBSTANCE P AFFERENT INNERVATION OF THE STOMACH AND CERVICAL OESOPHAGUS IN THE RAT Values represent the proportion of True blue labelled cells in spinal or vagal ganglia reacting with substance P or CGRP antibodies (mean + S.E.M.), and are expressed as % for each ganglion for the number of rats indicated (n). The total number of True blue cells counted for each ganglia are indicated in the last column. % Substance P

% CGRP

No. blue cells counted per ganglion

Stomach Ganglion T~, T7 T~ To Tjo T~E TI2 T~ LI

35.8__+ 2.8(n=7) 43.1_+ 2.5 57.8_+ 4.3 53.6__+ 3.7 55.4__+ 3.6 55.5-+ 2.1 45.6-+ 4.6 37.8_+10.7 43.4_+ 18.8

35.5-+ 2.5(n=5) 46.6_+ 4.2 50.4-+ 6.8 79.5_+ 4.8 79,1-+ 2.2 82.1_+ 1.9 67.7-+ 6.1 84.4_+ 6.6 66.7_+33.3

12.1-+ 2.0 21.3_+ 2.3 26.6_+ 5.4 61.7_+ 7.0 90.0_+ 13.9 99.1-+14.0 64,0_+10.2 27.3_+ 6.9 1.6_+ 0.6

Nodose

1.8_+ 0.9(n=5)

5.7_+ 1.4(n=4)

33_+ 6 ( n = 6 ) 17+ 12 33-+ 2 39+ 2 46_+15

69_+16(n=5) 90_+ 10 56+ 6 70_+13 61_+20

C. C~ C~ TI

46-+ 5 38+ 12 38_+ 5

75_+14 67-+33 54_+ 4

Nodose

14-+ 3 ( n = 5 )

26+_ 2 ( n = 4 )

Oesophagus Ganglion Cj C2 C~ C4 C,

315.3-+40.0(left) 66.3_+ 5.4 (right)

9_+ 3+ 8+ 10_+ 34-

4 1 3 3 2

3+ 1 3-+ 1 4+ 1 [72±21 (left and right)

surgical methods to originate from cell bodies o f vagal motor neurones, and afferents in the nodose ganglion [13]. The latter is confirmed in the present study. In addition, we have shown cervical spinal ganglia to provide a C G R P innervation to the oesophagus. Previous studies using retrograde tracing and immunohistochemistry have suggested that between 10 and 50% of the afferents to different visceral structures contain substance P [3, 15, 16]. In contrast, the present study and similar ones on other organs, e.g. kidney, bladder, reproductive tract suggest that C G R P occurs in up to 85% of afferent neurons [8, 17, 18]. Our results therefore contribute directly to the idea that C G R P is a marker for a major sub-set o f visceral afferent neurones.

155

t ~B

kjG~,P Fig. 2. Fluorescence micrographs (la) ot" retrogradely labelled Truc blue cells in a section of rat dorsal root ganglion (Ca) after injection of True blue into the cervical oesophagus; (lb) CGP, P-like immunoreactive cells in the same section. (2a) and (3a) True blue-labelled cells in a section of rat dorsal root ganglion (T~L) after injection into the ventral stomach wall; (2b) and (3b) CGRP-like immunoreactivitx and substance P-like inamunoreactivily in the same sections, Note cells containing both True blue and pcplide immunoreactivity (closed arrow), cell containing True blue bul not peptide immunoreactivity Iopcn aTrow), and cells containing peptide immunoreactivily but not True blue (arrowhead). Bar - 50/tin.

In the case of the stomach, CGRP, unlike substance P, appears to occur exclusively in afferent fibres. It is known that C G R P has potent central and peripheral effects as an inhibitor of gastric acid secretion [6, 19], it also releases somatostatin from the isolated perfused rat stomach [21], relaxes gastric smooth muscle cells [11], and acts on the mesenteric vasculature [10]. Release of C G R P at the peripheral ends of gastric alferent fibres may therefore be responsible for a variety of reflex responses. Since it appears that C G R P occurs predominantly in spinal afferents to the stomach, this system should provide a good model for examining the mechanisms of action of neuropeptides released at peripheral afferent terminals in the gut. I Amara, S.G., Jonas, V., Rosenfeld, M.G., Ong, E.S. and Evans, R.M., Alternative R N A processing in calcitonin gene expression generates m R N A s encoding different polypeptide products, Nature (l~on don), 298 (1982) 240 244. 2 Claguc, J.R., Slernini, C. and Brecha, N.C.. Localization ofcalcitonin gene-related peptide-like immunoreactivily in neurons of the rat gastrointestinal tract, Neurosci. Len., 56 (1985) 63 68. 3 Dockray. G.J. and Sharkey, K.A.. The neurochemistry of visceral afferenls, Prog. Brain Res.. 67 ([9S6) 133-148. 4 Gibbins. l i . . Furness, J.B., Costa. M., Maclnlyre, 1., Hillyard, ('.]. and Girgis, S., ('o-localization of calcitonin gene-related peptide-like immunoreactivity with substance P m cutalleous vasctllar and xisceral sensory neurons of guinea pigs, Ncurosci. Len., 57 (1985) 125 130. 5 Gibson, S.J., Polak, J.M., Bloom, S.R., Sabate, I.M., Mulderry, P.M., Ghalei, M.A., McGregor, G P . . Morrison, J.[:.g., Kelly, J.S., Ewtns. R.M. and Rosenfeld, M.(}., Calcitonin genc-rclatcd pcptide immunorcactivity in the spinal cord of m a n and of eight other species. J. Ncurosci., 4 (1984) 3 I I) 31 I. 6 Hughes, J.J., Levir~e, A.S., Morley. J.l.i., Gosnell, B.A. and Silvis, S.E., lntraventrieular calcitonin gene related peptide inhibits gastric acid secretion. Peptides. 5 (1984) 310 31 I. 7 llwang, K., Grossman, M.I. and Yves, A.C., Nervous control of the cervical portion of tile ocsophagus, Am, J. Physiol., 154(1948)343 357.

156 8 lnyama, C.O., Wharton, J., Su, H.C. and Polak, J.M., CGRP-immunoreactive nerves in the genitalia of the female rat originate from dorsal root ganglia THL3 and L6 S~: a combined immunocytochemical and retrograde tracing study, Neurosci. Lett., 69 (1986) 13 18. 9 Lee, Y., Takami, K., Kawai, Y., Girgis, S., Hillyard, C.J., MacIntyre, I., Emson, P.C. and Tohyama, H., Distribution of calcitonin gene-related peptide in the rat peripheral nervous system with reference to its coexistence with substance P, Neuroscience, 15 (1985) 1227 1237 l0 Marshall, I., AI-Kazwini, SJ., Holman, J.J. and Craig, R.K., Human and rat-CGRP but not catcitonin cause mesenteric vasodilation in rats, Eur. J. Pharmacol., 123 (1986) 217 222. I 1 Maton, P.N., Sutcliff, V.E., Zhou, Z-C., Gardner, J.D. and Jensen, R.T., Characterization of receptors for calcitonin-gene related peptide (CGRP) on gastric smooth muscle cells, Gastroenterology, 90 (1986) 1537. 12 Rodrigo, J., de Felipe, J., Robles-Chillida, E.M., Perez Anton, J.A, Mayo, I. and Gomez, A., Sensory vagal nature and anatomical access paths to esophagus laminar nerve endings in myenteric ganglia. Determination by surgical degeneration methods. Acta Anat., 112 (1982) 47-57. 13 Rodrigo, J., Polak, J.M., Fernandez, L., Ghatei, M.A., Mulderry, P. and Bloom, S.R., Calcilonin gene-related peptide immunoreactive sensory and motor nerves of the rat, cat and monkey esophagus, Gastroenterology, 88 (1985)444 451. 14 Rosenfeld, M.G.. Mermod, J-J., Amara, S.G., Swanson, L.W., Sawchenko, P.E., Rivier, J., Vale, W.W, and Evans, R,H., Production of a novel neuropeptide encoded by the calcitonin gene via tissuespecitic RNA processing, Nature (London), 304 (1983) 129 135. 15 Sharkey, K.A., Williams, R.G. and Dockray, G.J., Sensory substance P innervation of the stomach and pancreas: demonstration of capsaicin-sensitive sensory neurons in the rat by combined immunohistochcmistry and retrograde tracing, Gastroenterology, 87 (19841 914 921, 16 Sharkey, K.A.. Williams, R.G., Schultzberg, M. and Dockray, G.J., Sensory substance P-innervation of the urinary bladder: possible site of action of capsaicin in causing urine retention in rats, Neuroscience, 10 (1983) 861 868. 17 Su, H.C., Bishop, A.E., Gibson, S.J. and Potak, J.M., The origins and nature of the CGRP- and NPYimlnunoreactive nerves of the rat stomach determined by retrograde tracing and denervation procedures combined with immunocytochemistry, Regul. Peptides, 13 (1985) 66. 18 Su, H.C., Wharton, J., Polak, J.M.. Mulderry, P.K., Ghatei, M.A., Gibson, S.J., Terenghi, G., Morrison, J.F.B.. Ballesla. J. and Bloom, S.R., Calcitonin gene-related peptide immunoreactivity in afferent neurons supplying the urinary tract: combined retrograde tracing and immunohistochemistry, Neuroscience, 18 (1986) 727 747. 19 Taehe, Y., Pappas. T., Lauffenburger, H., 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) 344 349. 20 Wiesenfeld-Hallin. Z., H6kfelt, T., Lundberg, J.M., Forssmann, W.G., Reinecke, M., Tschopp, F.A. and Fischer, J.A., lmmunoreaetive calcitonin gene-related peptide and substance P coexist in sensory neurons to the spinal cord and interact in spinal behavioural responses of the rat, Neurosci. Lett., 52 (1984) 199 204. 21 Yamatani, Y., Kadowaki, S., Chiba, T., Abe, H., Chihara, K., Fukase, M. and Fujita, 1"., Calcitonin gene-related peptide stimulates somatostatin release from isolated perfused rat stomach, Endocrinology, 118 (1986) 2144 2145.