Received 8 July 1999; received in revised form 20 August 1999; accepted 21 September 1999

Transcription

1 Pain 84 (2000) 389±395 Simultaneous depletion of neurokinin A, substance P and calcitonin generelated peptide from the caudal trigeminal nucleus of the rat during electrical stimulation of the trigeminal ganglion M. Samsam a,b, R. CovenÄas a, *, R. Ahangari a, b, J. Yajeya c, J.A. NarvaÂez d, G. Tramu e a Instituto de Neurociencias de Castilla y LeoÂn, Salamanca, Spain b Departamento de BiologõÂa Celular y PatologõÂa, Laboratorio de NeuroanatomõÂa de los Sistemas PeptideÂrgicos, Salamanca, Spain c Departamento de FisiologõÂa y FarmacologõÂa, Universidad de Salamanca, Facultad de Medicina, Salamanca, Spain d Departamento de FisiologõÂa, Facultad de Medicina, Universidad de MaÂlaga, Malaga, Spain e Universite de Bordeaux I, Laboratoire de Neurocytochimie Fonctionnelle, C.N.R.S., URA 339, Talence, France Received 8 July 1999; received in revised form 20 August 1999; accepted 21 September 1999 Abstract The central terminals of the primary sensory trigeminal ganglion (TG) neurons projecting into the caudal trigeminal nucleus (CTN) of the rat exhibit neurokinin A (NKA)-, substance P (SP)-, and calcitonin gene-related peptide (CGRP)-immunoreactivities (IRs). We stimulated the TG in the rat to induce some of the alterations which might occur during migraine (neurogenic in ammation). Under a stereotaxic apparatus and by means of a bipolar electrode, one-side TG of the animals were electrically stimulated (7.5 Hz, 5 ms, 0.8±1.4 ma) with square pulses for 5 min. Then, using immunohistochemical methods, the lower medulla of each rat was studied for NKA-, SP- and CGRP- IRs. Light microscopic examination of brain-stem sequencial sections revealed a simultaneous decrease in the immunoreactivities of all neuropeptides (NKA, SP and CGRP) in the CTN ipsilateral to TG stimulation in comparison with the other (not stimulated) side CTN. It is suggested that this decrease in immunoreactivity would be due to the co-release of neuropeptides following noxious stimuli and that NKA, SP and CGRP might therefore act as co-transmitters or co-modulators at the rst central synapses of the trigeminal sensory pathway. q 2000 International Association for the Study of Pain. Published by Elsevier Science B.V. Keywords: Neurokinin A; Substance P; Calcitonin gene-related peptide; Co-release; Migraine; Caudal trigeminal nucleus 1. Introduction Sensory innervation of the cerebral dura mater, an important structure involved in vascular head pain (Pen eld and McNaughton, 1983; Moskowitz, 1984), is mainly supplied by the primary sensory neurons of the trigeminal ganglion (TG). These cells contain neuropeptides and the available evidence suggests that substance P (SP) and calcitonin generelated peptide (CGRP) co-exist within these cells (Uddman et al., 1985; O'Connor and Van der Kooy, 1988). In agreement with this, neurokinin A (NKA) has also been reported to co-exist (Edvinsson et al., 1988) with CGRP and SP in TG nerve cells and their processes. There is also evidence that the plasma levels of CGRP (Goadsby et al., 1998, 1990) * Corresponding author. Facultad de Medicina, Universidad de Salamanca, Instituto de Neurociencia de Castilla y LeoÂn, Laboratorio de NeuroanatomõÂa de los Sistemas PeptideÂrgicos, y Dpto. de BiologõÂa Celular y PatologõÂa, Avda. del Campo Charro s/n 37007, Salamanca, Spain. Tel ext. 1856; fax: address: covenas@gugu.usal.es (R. CovenÄas) and CGRP/NKA (Gallai et al., 1995) in the blood of human subjects during migraine attacks are increased. In this regard, neurogenic in ammation within the blood vessels of the dura mater has been suggested as a model to account for vascular head pains, including migraine (Moskowitz, 1984). Consistent with this, electrical stimulation of the TG promotes plasma extravasation from the dural blood vessels (Markowitz et al., 1987) of the rat, probably due to the release of neuropeptide. Also, such stimulation increases CGRP levels in the superior sagittal sinus of the rat (Buzzi et al., 1991) within minutes. These events, which indicate the release of neuropeptides from the peripheral axons of TG nerve cells, may re ect some alterations that might occur during migraine headaches. Moreover, the release of CGRP from the central processes of TG cells following electrical stimulation of the TG in the ipsilateral caudal trigeminal nucleus (CTN) of the rat (KnyihaÂr Csillik et al., 1998) has been reported. In view of the co-existence of CGRP, SP and NKA in TG neurons, we were prompted to investigate the immunoreactivity (IR) of these neuropep /00/$20.00 q 2000 International Association for the Study of Pain. Published by Elsevier Science B.V. PII: S (99)

2 390 M. Samsam et al. / Pain 84 (2000) 389±395 tides in the rat CTN following electrical stimulation of the TG to gain further insight into the possible co-release and interactions of SP, NKA and CGRP in the brain-stem of the rat in an experimental migraine model. 2. Materials and methods Twelve young adult Wistar albino rats of both sexes (weight 200±250 g) were used in this study, two of which were used as controls without any stimulation. Following an injection (i.p.) of chloral hydrate (4%), the animals were deeply anesthetized and were placed in a stereotaxic apparatus on a horizontal plane. One-side trigeminal ganglia of the animals were electrically stimulated by means of a concentric bipolar electrode (0.5 mm in diameter). The coordinates of the stimulation were according to Schneider et al. (1981): bregma was thus considered as the zero point, at 3.2±3.4 mm posteriorly and 2.8±3.2 mm lateral to the bregma and at 9.3 mm depth from the external surface of the dura the ganglion was reached. The stimulating parameters were: square pulses with a frequency of 7.5 Hz, 5 ms duration, and 0.8±1.4 ma intensity for 5 min. Immediately following electrical stimulation of the TG, the animals were subjected to transcardial ushing with 100 ml of cold physiologic saline (NaCl 9%), followed by 500 ml of cold xative solution (4% paraformaldehyde). The head and neck regions were dissected and the brain-stem was removed for further immunohistochemical investigations. The site of the puncture on the ganglion induced by the stimulating electrode was always inspected grossly in successful animals. The medulla was post xed overnight and processed in 25% sucrose solution for 24 h at 48C. Sections (50 mm thick) from frozen blocks (1.5 mm) of the lower medulla were obtained using a Reichert cryostat. The sections were divided into three series, each containing selected sections from the upper, middle, and lower parts. Thus, each series was investigated for one of the 3 neuropeptides (NKA, SP or CGRP). Starting from either side of the lower medulla, when one section had been chosen for CGRP-IR, the next section was chosen for NKA and the next one for SP and the whole process was repeated for the next sequential sections. Using the free- oating technique, the sections were treated with a mixture of methanol and H 2 O 2 (30%) for 30 min. The sections were then preincubated in 1% normal horse serum in 0.15 M phosphate buffer saline (PBS) containing 0.3% Triton X-100 for 30 min. Then, the sections were incubated in the same buffer containing the anti-nka or - SP or -CGRP antisera at dilutions of 1/2500 for NKA and 1/ 1500 for SP and CGRP overnight at 48C. The next day, after a 30 min wash with PBS, the sections were incubated with sheep anti-rabbit immunoglobulin coupled to horseradish peroxidase as the second antiserum, diluted 1/250 in PBS for 1h. This was followed by washing in PBS (30 min) and also in Tris-HCL buffer (ph 7.4) for 10 min. Tissue bound peroxidase was developed using the 3,3 0 -diaminobenzidine method. Finally, the sections were mounted and coverslipped. Antisera were raised in rabbits against immunogens prepared by coupling the peptides, synthetic human NKA or SP, or CGRP to a carrier protein (human serum albumin) with glutaraldehyde. The immunological characteristics of the antisera used as well as the speci city of the immunostaining have been reported previously (Burgos et al., 1988; Marcos et al., 1993; 1999). A densitometric analysis was performed to measure the intensity of the immunoreaction in the CTN. Optical density measurements were made from the same selected areas of the ipsilateral and contralateral (to the electrical stimulation of the TG) CTN in the immunostained sections. Grey values of the selected areas were obtained using the Scion NIH image analysis software. Using a 5 objective lens, images were captured directly from microscope slides by means of a black and white camera (Cohu CCD) and displayed on a computer monitor. Portions of the CTN were outlined on the computer screen and the software automatically assigned the average grey value of the screen pixels con ned to the outlined area (a value of 0 indicated a white pixel and 255, a black pixel). 3. Results Under light microscopic examination, the immunoreactivity of the central trigeminal terminals in the CTN of adult rats (normal controls) appeared as a curved laminar structure, close to the dorsal surface of the medulla (Fig. 1A). The central terminals of the pseudounipolar trigeminal ganglion cells projecting into this region showed a characteristic dense and massive neuropeptide immunoreactivity (Fig. 1A, arrows). No differences between the two sexes were found. Five min electrical stimulation (7.5 Hz, 5 ms, 0.8±1.4 ma) of the TG caused a simultaneous and signi cant depletion of SP-, CGRP- and NKA-IR (Figs. 1B and 2A,B) of the central trigeminal terminals in the rat brainstem ipsilateral CTN in comparison with the other (notstimulated) side. At higher magni cation (Fig. 2C,D), the central terminals of the TG neurons could be seen more clearly and the depletion was more evident. We measured the immunoreactivity of the CTN in both sides, as described before, and the results were subjected to statistical analysis using Student's t-test. Accordingly, the difference between the immunoreactivity of the stimulated (ipsilateral side) and not- stimulated (contralateral side) CTN at P, 0:05 was signi cant (Table 1). In control animals, however, the image analysis study did not reveal any differences in the immunoreactivity of either side CTN (Table 2). 4. Discussion This study provides strong evidence that tachykinins

3 M. Samsam et al. / Pain 84 (2000) 389± Fig. 1. (A) Substance P (SP) immunoreactivity (IR) in the caudal trigeminal nucleus (CTN) of a control rat (two animals). Arrows denote the immunoreactive central trigeminal terminals which project into the CTN close to the dorsal surface of the medulla. Following a unilateral electrical stimulation (B) of the trigeminal ganglion (TG), a marked decrease in the SP-IR of the CTN on the ipsilateral or stimulated side (S) is evident in comparison with the non-stimulated (NS) side where a much stronger SP-IR is seen. Scale bar: 100 mm. (NKA, SP) and CGRP are co-released from the central terminals of TG neurons upon electrical stimulation of the ganglion. We used different sets of stimulating parameters (1±7.5 Hz, 5 ms, 0.1±1.4 ma) for 5±30 min. However, clear results (depletion of the neuropeptide content of the central trigeminal terminals in the ipsilateral CTN) were obtained within 5 min stimulation under the stimulating parameters (7.5 Hz, 5 ms, 0.8±1.4 ma) used in this study. Our results are in agreement with earlier studies (Hua et al., 1986; Saria et al., 1986), in which the simultaneous release of several tachykinins and CGRP from the spinal cord was described. According to the present results, the co-release of NKA, SP, and CGRP following activation of the trigeminal system would further enhance their role as co-transmitters of nociception in the rst central synapses of the trigeminal pathway. This suggestion is in agreement with earlier studies (Woolf and Wiesenfeld-Hallin, 1986; Oku et al., 1987) reporting the co-transmittory role of CGRP and SP in nociception. Both co-release and co-transmission have also been reported (Holzer, 1988). The role of SP as a transmitter or modulator in a population of primary sensory neurons has been con rmed elsewhere (HoÈkfelt et al., 1975; Lembeck and Gamse, 1977) already. In this sense, Otsuka and Konishi (1976) have suggested that SP would act as an excitatory transmitter of primary afferent bers and ensuing studies (Otsuka et al., 1982) revealed that the release of SP from the primary afferent terminals of the spinal cord following electrical stimulation of the dorsal root strongly excites spinal neurons. These studies are consistent with the results (Davis and Dostrovsky, 1986) reported for the neuronal responses in the trigeminal subnucleus caudalis in the brain-stem evoked by electrical stimulation of the middle meningeal artery, indicating that brainstem neurons are involved in vascular head pain. Moreover, NKA has also been recognized as a transmitter of nociceptive primary afferent neurons (Gamse and Saria, 1986). In this respect, a previous study (Saria et al., 1985) indicated that NKA evokes a slow membrane depolarization in neurons of the inferior mesenteric ganglion in vitro, suggesting that the actions of NKA and SP on prevertebral neurons are similar and that these two peptides may act on the same receptors or share similar ionic channels. Similarly, SP and NKA released from the central trigeminal terminals may have an excitatory effect on the trigeminal nucleus. In this regard, expression of the c-fos oncoprotein, which is increased in the medulla by noxious stimuli (Nozaki et al., 1992; Kaube et al., 1993), is also increased in the interneurons of the rat CTN following electrical stimulation of the TG. The release of SP and NKA from the trigeminal central terminals, exciting the neuronal network in the CTN, is in agreement with an orthodromic conduction of painful stimuli (Moskowitz, 1989) to the brain-stem and higher brain centers to register the pain. In attempts to examine the effects of several antimigraine drugs, NK1 receptor antagonists such as RP67580 (Shepheard et al., 1993) have been reported to be far more potent than 5-HT1B/D receptor agonists (CP-93,129 and sumatriptan) in blocking neurogenic plasma extravasation following TG stimulation (Saxena, 1994). Consistent with this, GR (a potent NK1 receptor

4 392 M. Samsam et al. / Pain 84 (2000) 389±395 Fig. 2. (A) Upper half of a section of the medulla from the same experiment as shown in Fig. 1B. CGRP-IR in the CTN on the stimulated side (S) is signi cantly depleted in comparison with the non-stimulated (NS) side (arrows). (B) The lower parts of the medulla from the same experiment were examined for NKA-IR; note the marked difference in the intensity of immunoreaction of the overall CTN (arrows) between the non-stimulated (NS) and the stimulated (S) side. (C) and (D) (shown in rectangles) are the corresponding parts of the CTN from (B). (C) A large number of NKA- immunoreactive central trigeminal nerve terminals (arrowheads) can be observed in the non-stimulated (NS) side at higher magni cation. In (D) the number and intensity of NKA- immunoreactive nerve terminals (arrows) are signi cantly decreased as compared to (C) which is the non-stimulated (NS) side at higher magni cation. Scale bar: 100 mm. antagonist) inhibits the expression of c-fos oncoprotein in the CTN by up to 60% following electrical stimulation of the TG (Polley et al., 1997). By contrast, a recent report (Goadsby et al., 1998) has indicated that the NK1 receptor Table 1 Results of the statistical analysis (Stat. View TM) showing the mean values of grey level measurements obtained from similar areas of each side caudal trigeminal nucleus in three sections of the lower medulla of a stimulated rat (six areas on each side of a section are compared). Six such stimulated animals were studied (three sections per each animal) No. area Mean grey level SD a SE SP stimul. Side SP non-stimul. side CGRP stimul. side CGRP non-stimul. Side NKA stimul. side NKA non-stimul. side a Using Student's t-test, the differences between stimulated and nonstimulated side, in all cases at P, 0:05, are signi cant. antagonist GR does not affect central trigeminal activity following e1ectrical stimulation of the superior sagittal sinus (it did not inhibit or decrease the expression of c-fos). Moreover, sumatriptan, an effective drug in aborting acute migraine attacks, decreased the expression of c-fos in the rat medulla (Hoskin et al., 1996). However, the same drug was reported to be ineffective in reducing c-fos activity in the rat CTN (KnyihaÂr Csillik et al., 1998). The differences in the results of the above authors could be due to the Table 2 Statistical analysis (Stat. View TM) of the mean grey level values measured in similar areas of each side caudal trigeminal nucleus in one section of a control rat. Two sections (from the lower medulla) from each of the two control animals were analyzed in this study a No. area Mean grey level SD SE SP control side SP control side a Using Student's t-test, the differences between two control sides at P, 0:05 are not signi cant.

5 M. Samsam et al. / Pain 84 (2000) 389± different methods of experimentation and/or the different stimulation sites and parameters. Clinically, however, an orally -administered NK1 receptor antagonist (Lanepitant) was not effective in alleviating the pain in acute migraine attacks (Goldstein et al., 1997), although this might be due to the poor bioavailability of this drug during such attacks. Consistent with this, the endothelin receptor antagonist bosentan, which is able to block neurogenic plasma extravasation in the dura mater (Brandli et al., 1996), is not effective in aborting headache in migraine patients and this might be attributed to its lack of vasoconstrictor activity (May et al., 1996). However, in this respect, sumatriptan inhibits the release of CGRP from the peripheral perivascular trigeminal terminals in the rat dura mater (KnyihaÂr Csillik et al., 1997). There is evidence that CGRP increases the plasma protein leakage induced by SP, NKA and NKB (Gamse and Saria, 1985). This enhancement of the effect of these neuropeptide has also been reported to be important in myofacial pain induced by injection of CGRP and SP or NKA in the human temporal muscle (Pedersen-Bjergaard et al., 1991). Also in this regard, Uddman et al. (1985) suggested that the co-release of NKA, SP and CGRP from cerebrovascular nerve bers following activation of the trigeminovascular system cooperates in the mediation of vascular reactions. Consistent with this, many studies carried out on the central nervous system (CNS) report concerning the cooperation of these neuropeptides. CGRP has been reported to enhance the action of SP when these peptides are coadministered in the CNS (Wiesenfeld-Halin et al., 1984; Goodman and Iversen, 1986). Thus, SP-induced effects enhanced by CGRP have been suggested to be due to the ability of CGRP to inhibit an enzyme involved in SP degradation (Greves et al., 1985). In agreement with this, in the CTN there might be an interaction between CGRP and SP or NKA in the brain-stem when these substances are released from central trigeminal terminals following electrical stimulation of TG neurons. Accordingly, when SP is released from the central trigeminal terminals it would excite the neuronal network in the CTN, and CGRP probably controls its degradation there. Hence, central release of CGRP may be correlated with the intensity and/or duration of painful stimuli. It is known that after its synthesis in the cell bodies SP is transported to both peripheral and central axons; in view of the co-existence of SP with NKA and CGRP in the TG neurons, the same process of transport could hold for NKA and/or CGRP. In addition to this, assuming the co-release of CGRP, SP and NKA from central trigeminal terminals during experimental migraine, it may be postulated that such co-release would also occur from the peripheral axons of the TG sensory neurons innervating dural blood vessels during the activation of the trigeminovascular system. Consistent with this idea, a signi cant increase in NKA and CGRP levels has been detected during migraine attacks in young migraine sufferers (Gallai et al., 1995). Also in this respect, SP and CGRP levels have been reported to be increased in the salivary secretions of migraine and cluster headache patients (Nikolodi, 1990). However, the controversy in reports concerning the absence of these neuropeptides (Friberg et al., 1994) in the blood of human subjects during the onset of migraine with aura may indeed derive from the time when the blood was withdrawn (VeÂcsei et al., 1995). Thus, the time for the potential buildup of neuropeptide blood concentrations may not have been suf cient and precise, or the peptides may already have been degraded when the blood was withdrawn. Consistent with this, since the released CGRP is rapidly metabolized (VeÂcsei et al., 1995), its inhibitory effect on the enzymatic degradation of SP is lost and therefore SP degradation is accelerated. Time is an important factor in real and experimental models of migraine and it is possible that the co-release of SP, NKA and CGRP from peripheral trigeminal axons could be detected at a speci c time during migraine. However, there is a lack of SP in the blood of human subjects during migraine (Goadsby et al., 1990; Friberg et al., 1994). Consistent with this, the SP antagonist GR is unable to inhibit or decrease c-fos expression in trigeminal nucleus of the cat following electrical stimulation of the superior sagittal sinus (Goadsby et al., 1998). This is in agreement with the ineffectiveness of SP antagonists to abort the acute migraine attacks clinically (Diener et al., 1996; Goldstein et al., 1997). According to all these negative data concerning the existence of SP during migraine attacks, it does not seem very unreasonable to assume that increases in SP levels simply may not occur in the blood during migraine. However, centrally in the CTN, our current study supports the involvement of SP, CGRP and NKA during activation of the trigeminovascular system. The link of these neuropeptides with migraine is related to the possible pathomechanism of migraine postulated by Moskowitz which holds that: activation of the trigeminovascular system (e.g. through injury to the blood vessel wall) would produce and synthesize or transport nociceptive mediators such as prostaglandins, histamine, serotonin and bradykinins from the circulation (which lower the threshold or depolarize the sensory perivascular nerve bers through binding to speci c receptors on these bers). Such depolarization may induce a local release of vasoactive neurotransmitters such as SP from peripheral trigeminal bers into the walls of blood vessels to promote neurogenic in ammation. Depolarization is also accompanied by antidromic and orthodromic conduction of the sensory stimulus. While antidromic conduction is responsible for spreading the in ammatory response to adjacent tissues, the orthodromic axis transmits nocicepive information towards the CTN and higher brain centers for the registration of pain (Moskowitz et al., 1989). Hence, peripheral activation of the trigeminovascular system would cause the co-release of CGRP, SP and NKA centrally to activate the nociceptive network in the CTN and higher brain centers.

6 394 M. Samsam et al. / Pain 84 (2000) 389±395 Acknowledgements M. Samsam has been supported by the FundacioÂn Mapfre Medicina (Spain). This work has been supported by the D.G.I.C.Y.T. (PB 96/1467), Spain. The authors are indepted to Professor M. MerchaÂn for his kind advice, to Dra. R. Riquelme for her tremendous work in the image analysis study. The assistance of Dr. P. Marcos and the technical work of Dr. A. de la Fuente and Mr. Jose Antonio Romo are gratefully acknowledged. The work of N. Skinner in revising the language of the manuscript is acknowledged. References Brandli P, Lof er B-M, Breu V, Osterwalder R, Maire J-P, Clozel M. Role of endothelin in mediating neurogenic plasma extravasation in rat dura mater. Pain 1996;64:315±322. Burgos C, Aguirre JA, Alonso JR, CovenÄas R. Immunocytochemical study of substance P-like bers and cell bodies in the rat diencephalon. J Hirnforsch 1988;29:651±657. Buzzi MG, Carter WB, Shimizu T, Heath III H, Moskowitz MA. Dihydroergotamine and Sumatriptan attenuate levels of CGRP in plasma in rat superior sagittal sinus during electrical stimulation of the trigeminal ganglion. Neuropharmacology 1991;30:1193±1200. Davis KD, Dostrovsky J. Activation of trigeminal brain-stem nociceptive neurons by dural artery stimulation. Pain 1986;25:395±401. Diener HC. Substance-P antagonist RPR is not effective in human migraine attacks. In: Olesen J, Tfelt-Hansen P, editors. Proceedings of the VIth international headache seminar, New York: Lippincott- Raven, Edvinsson L, Brodin E, Jansen I, Uddman R. Neurokinin A in cerebral vessels: characterization, localization and effects in vitro. Regul Pept 1988;20:181±197. Friberg L, Olesen J, Olsen TS, Karle A, Ekman R, Fahrenkrug J. Absence of vasoactive peptide release from brain to cerebral circulation during onset of migraine with aura. Cephalalgia 1994;14:47±54. Gallai V, Sarchielli P, Floridi A, Francceschini M, Codini M, Trequattrini A, Palumbo R. Vasoactive peptide levels in the plasma of young migraine patients with and without aura assessed both interictally and ictally. Cephalalgia 1995;15(384):390. Gamse R, Saria A. Potentiation of tachykinin-induced plasma protein extravasation by calcitonin gene-related peptide. Eur J Pharmacol 1985;114:61±66. Gamse R, Saria A. Nociceptive behavior after intrathecal injection of substance P, neurokinin A and calcitonin gene-related peptide in mice. Neurosci Lett 1986;70:143±147. Goadsby PJ, Edvinsson L, Ekman R. Release of vasoactive peptides in the extracerebral circulation of humans and the cat during activation of the trigeminovascular system. Ann Neurol 1988;23:193±196. Goadsby PJ, Edvinsson L, Ekman R. Vasoactive peptide release in the extracerebral circulation of human during migraine headache. Ann Neurol 1990;28:183±187. Goadsby PJ, Hoskin KL, Knight YE. Substance P blockade with the potent and centrally acting antagonist GR does not effect central trigeminal activity with superior sagittal sinus stimulation. Neuroscience 1998;86:337±343. Goldstein DJ, Wang O, Saper JR, Stoltz R, Silberstein SD, Mathew NT. Ineffectiveness of neurokinin-1 antagonist in acute migraine: a crossover study. Cephalalgia 1997;17:785±790. Goodman EC, Iversen LL. Calcitonin gene-related peptide: Novel neuropeptide. Life Sci. 1986;38:2169±2178. Greves LE, Nyberg F, Terenius L, HoÈkfelt T. Calcitonin gene-related peptide is a potent inhibitor of substance P degradation. Eur J Pharmacol 1985;115:309±311. HoÈkfelt T, Keller JO, Nilson G, Pernow B. Experimental immunohistochemical studies on the localization and distribution of substance P in cat primary sensory neurons. Brain Res 1975;19:235±252. Holzer P. Local effector functions of capsaicin-sensitive sensory nerve endings: involvement of tachykinins, calcitonin gene-related peptide and other neuropeptides. Neuroscience 1988;24:739±768. Hoskin KL, Kaube H, Goadsby PJ. Sumatriptan can inhibit trigeminal afferents by an exclusively neural mechanism. Brain 1996;119:1419± Hua XY, Saria A, Gamse R, Theodorsson-Norheim E, Brodin E, Lundberg JM. Capsaicin-induced release of multiple tachykinins (substance P and eledosin-like material) from guinea-pig spinal cord and ureter. Neuroscience 1986;19:313±319. Kaube H, Keay KA, Hoskin KL, Bandler R, Goadsby PJ. Expression of c- fos like immunoreactivity in the caudal medulla and upper cervical spinal cord following stimulation of the superior sagittal sinus in cat. Brain Res 1993;629:95±102. KnyihaÂr Csillik E, Tajti J, Samsam M, SaÂry G, Slezak S, VeÂcsei L. Effect of a serotonin agonist (sumatriptan) on the peptidergic innervation of the rat cerebral dura mater and on the expression of the c-fos in the caudal trigeminal nucleus in an experimental migraine model. J Neurosci Res 1997;48:449±464. KnyihaÂr Csillik E, Tajti J, Samsam M, SaÂry G, BuzaÂs P, VeÂcsei L. Depletion of calcitonin gene-related peptide from the caudal trigeminal nucleus of the rat after electrical stimulation of the Gasserian ganglion. Exp Brain Res 1998;118:111±114. Lembeck F, Gamse R. Lack of algesia effect of substance P on paravascular pain receptors. Naunyn Schmiedebergs Arch Pharmacol 1977;299:295± 303. Marcos P, CovenÄas R, de LeoÂn M, NarvaÂez JA, Tramu G, Aguirre JA, GonzaÂlez-BaroÂn S. Neurokinin A-like immunoreactivity in the cat brain-stem. Neuropeptides 1993;25:105±114. Marcos P, CovenÄas R, NarvaÂez JA, Aguirre JA, Tramu G, GonzaÂlez-BaroÂn S. Mapping of enkephalins-, neuropeptide Y-, calcitonin gene-related peptide-, and gastrin-releasing peptide/bombesin-like immunoreactivities in the cat amygdala. Peptides 1999;20:635±644. Markowitz S, Saito K, Moskowitz MA. Neurogenically mediated leakage of plasma protein occurs from blood vessels in dura mater but not brain. J Neurosci 1987;7:4129±4136. May A, Gijsman HJ, Wallnoefer A, Jones R, Diener HC, Ferrari MD. Endothelin antagonist bosentan blocks neurogenic in ammation, but is not effective in aborting migraine attacks. Pain 1996;67:375±378. Moskowitz MA. The neurobiology of vascular head pain. Ann Neurol 1984;16:157±168. Moskowitz MA, Buzzi MG, Sakas DE, Linnik MD. Pain mechanisms underlying vascular headaches. Rev Neurol 1989;145:181±193. Nikolodi M. Del Bianco. E. Sensory neuropeptides (substance P, calcitonin gene-related peptide) and vasoactive intestinal polypeptide in human saliva: their pattern in migraine and cluster headache. Cephalalgia 1990;10:39±50. Nozaki K, Boccalin P, Moskowitz MA. Expression of c-fos-like immunoreactivity in brain-stem after meningeal irritation by blood in the subarachnoid space. Neuroscience 1992;49:669±680. O'Connor TP, Van der Kooy D. Enrichment of a vasoactive neuropeptide (calcitonin gene-related peptide) in the trigeminal sensory projection to the intracranial arteries. J Neurosci 1988;8:2468±2476. Oku R, Satoh M, Fujii N, Otaka A, Yajima H, Takagi H. Calcitonin generelated peptide promotes mechanical nociception by potentiating release of substance P from the spinal dorsal horn in rats. Brain Res 1987;403:350±354. Otsuka M, Konishi S. Release of substance P-like immunoreactivity from isolated spinal cord of new born rat. Nature 1976;264:83±84. Otsuka M, Konishi S, Yanagisawa M, Tsunoo A, Akagi H. Role of substance P as a sensory transmitter in spinal cord and sympathetic ganglia. Ciba Found Symp 1982;91:13±34.

7 M. Samsam et al. / Pain 84 (2000) 389± Pedersen-Bjergaard U, Nielsen LB, Jensen K, Edvinsson L, Jansen I, Olesen J. Calcitonin gene-related peptide, neurokinin A and substance P: effects on nociception and neurogenic in ammation in human skin and temporal muscle. Peptides 1991;12:333±337. Pen eld W, McNaughton F. Dural headaches and the innervation of the dura mater. Am Neurol Assoc 1983;64:106±109. Polley JS, Gaskin PJ, Perren MJ, Connor HE, Ward P, Beattie DT. The activity of GR205171, a potent non-peptide tachykinin NK1 receptor antagonist, in the trigeminovascular system. Regul Pept 1997;68:23±29. Saria A, Ma RC, Dun NJ. Neurokinin A depolarizes neurons of the guinea pig inferior mesenteric ganglia. Neurosci Lett 1985;60:145±150. Saria A, Gamse R, Petermen J, Fischer J, Theodersson-Norheim E, Lundberg JM. Simultaneous release of several tachykinins and calcitonin gene-related peptide from rat spinal cord slices. Neurosci Lett 1986;63:310±314. Saxena PR. Progress in 5-HT receptor classi cation. Newslett Eur Headache Fed 1994;3:1±8. Schneider JS, Denaro FJ, Olazabal UE, Leard HO. Stereotaxic atlas of the trigeminal ganglion in rat, cat, and monkey. Brain Res Bull 1981;7:93± 95. Shepheard SL, Williamson DJ, Hill RG, Hargreaves RJ. The non-peptide neurokinin1 receptor antagonist RP 67580, blocks neurogenic plasma extravasation in the dura mater of rats. Br J Pharmacol 1993;108:11±12. Uddman R, Edvinsson L, Ekman R, Kingman T, McCulloch J. Innervation of the feline cerebral vasculature by nerve bers containing calcitonin gene-related peptide: trigeminal origin and co-existence with substance P. Neurosci Lett 1985;62:131±136. VeÂcsei L, Tajti J. KnyihaÂr Csillik. E., Samsam, M. Neuropeptides and migraine: the role of CGRP. Cephalalgia 1995;16:3. Wiesenfeld-Halin Z, HoÈkfelt T, Lundberg JM, Forssmann WG, Reinecke M, Tschopp FA, Fischer JA. Immunoreactive calcitonin gene-related peptide and substance P co-exist in sensory neurons to the spinal cord and interact in spinal behavioral responses of the rat. Neurosci Lett 1984;52:199±204. Woolf C, Wiesenfeld-Hallin Z. Substance P and calcitonin gene-related peptide synergistically modulate the gain of the nociceptive exor withdrawal re ex in the rat. Neurosci Lett 1986;66:226±230.