Phosphodiesterase 3 and 5 and cyclic nucleotide-gated ion channel expression in rat trigeminovascular system

Neuroscience Letters 404 (2006) 202–207

Phosphodiesterase 3 and 5 and cyclic nucleotide-gated ion channel expression in rat trigeminovascular system Lars S. Kruse a,b,∗ , Nicolai T.H. Sandholdt a , Steen Gammeltoft b , Jes Olesen a , Christina Kruuse a a

b

Department of Neurology, Glostrup Hospital, Nordre Ringvej 57, DK-2600 Glostrup, Denmark Department of Clinical Biochemistry, Glostrup Hospital, Nordre Ringvej 57, DK-2600 Glostrup, Denmark Received 5 May 2006; received in revised form 23 May 2006; accepted 25 May 2006

Abstract Activation of the trigeminovascular pain signalling system appears involved in migraine pathophysiology. However, the molecular mechanisms are only partially known. Stimulation of cAMP and cGMP production as well as inhibition of their breakdown induce migraine-like headache. Additionally, migraine may be associated with mutations in ion channels. The aim of the present study was to describe the expression of phosphodiesterase 3 (PDE3) and 5 (PDE5) and cyclic nucleotide-gated ion channels (CNG) in cerebral arteries, meninges, and the trigeminal ganglion. mRNA for PDE and CNG was determined in the rat middle cerebral artery, basilar artery, trigeminal ganglion, and dura mater using real-time PCR. PDE and CNG proteins were identified using Western blot. For comparison, rat aorta and mesenteric artery were analysed. PDE3A, PDE3B, and PDE5A mRNA were detected in all tissues examined except for PDE3A mRNA in dura mater and the trigeminal ganglion. PDE5A and PDE3A protein expression was present in both cerebral and peripheral arteries, whereas PDE3B protein was present only in the cerebral arteries. The CNGA4 and B1 subunit mRNAs were detected in cerebral arteries and CNGA2 also in the mesenteric artery. CNGA2 and A3 proteins were found in cerebral arteries and dura and CNGA1, CNGA2 and CNGA3 in the trigeminal ganglion. In conclusion, PDE3A, PDE3B, PDE5A, and five CNG subunits were expressed in several components of the trigeminovascular system of the rat. This suggests that modulation of cAMP and cGMP levels by PDE and activation of CNG may play a role in trigeminovascular pain signalling leading to migraine headache. © 2006 Elsevier Ireland Ltd. All rights reserved. Keywords: Migraine; Phosphodiesterase; Cyclic nucleotide; Cerebral arteries; Trigeminovascular system; Ion channel

Brain tissue is generally insensitive to pain stimuli, but afferent projections of the trigeminal nerve containing pain-mediating fibres, are located in the cerebral meninges, venous sinuses and around the large cerebral arteries including the basilar and middle cerebral arteries [22]. Local release of neurotransmitters after dilatation of the arteries may activate these fibres [21]. The signal is transmitted via the trigeminal ganglion and the trigeminal nucleus to the thalamus and cerebral cortex. This pain signalling pathway is known as the trigeminovascular pathway [8]. In migraine patients infusion of calcitonin gene-related peptide (CGRP) and nitric oxide (NO) donors induce headache, presumably via production of cAMP and cGMP [17,25,30]. Thus, it is suggested that increased levels of cAMP and cGMP in the trigeminovascular system are essential in migraine pathophysiology [4,16]. Phosphodiesterases (PDE) hydrolyse cAMP and cGMP and 11 families have been described [18]. PDE iso-

∗

Corresponding author. Tel.: +45 43 23 24 71; fax: +45 43 23 39 29. E-mail address: [email protected] (L.S. Kruse).

0304-3940/$ – see front matter © 2006 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.neulet.2006.05.045

forms display different modes of regulation as well as tissue specific expression [1,27]. The PDE3 inhibitor cilostazol and the PDE5 inhibitors dipyridamole and sildenafil are able to induce migraine-like headache [3,12,16]. However, sildenafil induces migraine without a concomitant dilation of the cerebral arteries suggesting a non-vascular mechanism. Both PDE3 and PDE5 are present in the large cerebral arteries [2,15] but their presence in meninges and the trigeminal nerve has not been investigated. Recently, familial hemiplegic migraine, a rare dominantly inherited subform of migraine with aura, was found to be associated with various mutations in a calcium channel (CACNA1A), a sodium channel (SCN1A) and a Na/K-ATPase (ATP1A2) [24]. cAMP and cGMP directly affect cyclic nucleotide-gated (CNG) ion channels that belong to the voltage-gated ion channel super family [9]. Accordingly, modulation of ion channels in the trigeminovascular system may be involved in migraine pathophysiology. The distribution of CNG ion channels in the cerebral arteries, meninges and trigeminal nerve has not been described. Our hypothesis was that induction of migraine involves increased cyclic nucleotide levels and excessive activation of

L.S. Kruse et al. / Neuroscience Letters 404 (2006) 202–207

the trigeminovascular pain signalling system, perhaps due to dysfunction of PDE enzymes or CNG ion channels. The present study aimed to investigate the distribution of PDE3A, PDE3B, and PDE5A, and six subunits of CNG ion channels in the trigeminovascular system represented by the middle cerebral artery, basilar artery, dura mater and trigeminal ganglion. For comparison, the presence of PDE and CNG ion channels was studied in two peripheral arteries including the mesenteric artery and aorta. Since the rat is a common animal model in studies of migraine pathophysiology [5,20,23,29], this species was chosen for our study. Male Spraque–Dawley rats, 3-months-old, 300–350 g, were sacrificed after sedation with 0.7 ml 50 mg/ml Nebumal© (sodium pentobarbital). The basilar artery, middle cerebral artery, aorta, mesenteric artery, dura and trigeminal ganglia were collected. For PCR, tissues were stored at 4 ◦ C in RNAlater. For Western blotting the tissue was homogenised immediately in lysis buffer (150 mM NaCl, 50 mM Tris–HCl (pH 7.4), 0.5% Triton-X100, 5 mM EDTA, 10 ␮M leupeptin, 1 mM PMSF, 10 nM calyculin A, 4.000 KIU Trazylol, 5 mM pepstatin, 50 mM NaF, 10 mM ␤-glycerolphosphate and 1 mM ortho-vanadate), centrifuged and the supernatant stored at −80 ◦ C. All animal experiments were performed in accordance with the Danish Experimental Animal Act and the European Communities Council Directive of 24 November 1986 (86/609/EEC). RNeasy kits were from Qiagen Nordic (West Sussex, UK). dNTP mix was from Advanced Biotechnologies Ltd (Surrey, UK), RNAlater from Ambion (Huntingdon, UK) and RNasefree DNase and SYBR© Green Taq ReadyMixTM for Quanti-

203

tative PCR, Capillary Formulation from Sigma-Aldrich Denmark (Brøndby, Denmark). Reverse transcriptase (M-MuLV) and buffer were from Finnzymes (Espoo, Finland), RNasin from Promega (Fitchburg, WI, USA), oligo(dT)16 from MWG Biotech (Ebersberg, Germany). Hybond ECL nitrocellulose membranes and ECL Advance Western blotting Detection Kit was from GE Healthcare (Hillerød, Denmark) and the RC DC Protein Assay Kit was from Bio-Rad Laboratories (Herlev, Denmark). Specific antibodies to PDE or CNG ion channel subunits were either a gift: PDE5A from Joseph A. Beavo, University of Washington, Seattle, WA, USA, PDE3B from Eva Degerman, Lund University, Lund, Sweden, PDE3A from Vincent Manganiello, NIH, Bethesda, MD, USA, and CNGA1 from Robert S. Molday, University of British Colombia, Vancouver, Canada, or purchased: CNGA2 and CNGA3 from Alomone Laboratories, Jerusalem, Israel, and CNGA1 from Alpha Diagnostics, San Antonio, TX, USA. Polyclonal HRP-conjugated anti-rabbit IgG was from GE Helthcare, Hillerød, Denmark and HRPconjugated anti-mouse IgG from Pierce Biotechnology, Inc, Rockford, IL, USA. Purified total RNA was obtained using the Qiagen RNeasy kit including a DNAse treatment. Reverse transcription was performed on 0.82 ␮g purified RNA using 200 U M-MuLV reverse transcriptase, and 1 ␮l oligo(dT)16 (800 ␮g/ml) as first-strand primer. All primers were designed based on the most recently published rat sequences in GenBank (Table 1). Where possible, primers were designed to encompass all known splice variants and to be intron-spanning. The functionality and specificity of

Table 1 Primers used in this study Protein PDE3A, 134 bp PDE3B, 163 bp PDE5Aall, 193 bp CNGA1, 127 bp CNGA2, 132 bp CNGA3, 134 bp CNGA4, 128 bp CNGB1, 171 bp CNGB1b, 131/210 bp CNGB3, 111 bp Beta-actin, 158 bp

Accession no.

Type

Sequence

Location in rat

TM

NM 017337

Forward Reverse

CTGGACAAACCAATTCTTGCTCC AGAATACGGCCACATTTTCTTCCTA

1999–2021 2132–2108

60.6 59.7

NM 017229

Forward Reverse

GACCGTCGTTGCCTTGTATTTCTCG GGGTCAATCAGAAGGTCTGACACCA

963–987 1125–1101

64.6 64.6

NM 133584

Forward Reverse

CCCTGGCCTATTCAACAACGG ACGTGGGTCAGGGCCTCATA

2659–2679 2852–2833

61.8 61.4

NM 053497

Forward Reverse

CATCCCAGAGGGAGCATTACTTGCC TCATCGGCCTTGCTCTTCTTTTCCT

200–224 326–302

66.3 63.0

NM 012928

Forward Reverse

ACCATCTGACTGGTGAGAGCCCTGG GGTCATCTTTGCCATTGGCCTTGA

305–329 436–413

67.9 62.7

NM 053495

Forward Reverse

AAGACCCGACGCCTGACTCCTTTT CTTCTTCCTGCCTTTCCCTCCCTCT

111–134 244–220

64.4 66.3

NM 053496

Forward Reverse

GCCCCAACCAAAGCCAGGAAGT AAGCAGGCCCTGCATACAACGATG

154–175 281–258

64.0 64.4

NM 031809

Forward Reverse

GGAACCAGATCCTCCAGGAACCCCT AGCCACTCGATCAGCCTGACAGGGT

712–736 882–858

67.9 67.9

AF068572 RNCNG43

Forward Reverse

GGACCTTCCTACCAGCAGAGCC TGGTAACTCCTGGATCGATGCTGTC

161–182/142–163 370–346/272–248

66.3 64.6

AY564232

Forward Reverse

TCAGCCATCCATAGCACAGGGAGAA TCCCCAGTGCTGTTGTCTTCTTTGC

30–54 140–116

64.6 64.6

NM 031144

Forward Reverse

GCCACCAGTTCGCCATGGATGA ACCCATACCCACCATCACACCCTGG

68–89 225–201

64.0 67.9

204

L.S. Kruse et al. / Neuroscience Letters 404 (2006) 202–207

the primers was tested using cDNA from control tissues: olfactory bulb and retina for the CNG ion channel subunits, heart, and fat tissue for PDE3A and 3B, respectively, and cerebellum for PDE5 followed by DNA sequencing. Real-time PCR was performed using 1 ␮l cDNA sample, 10 ␮l SYBR Green Taq ReadyMix and 300 nM primer on a Roche LightCycler at the following settings: Preheat: 94 ◦ C for 30 s, PCR (40 cycles): 94 ◦ C for 0 s, 60 ◦ C for 5 s and 72 ◦ C for 7 s, post-PCR heating to 94 ◦ C for 0 s then step-wise melting from 63 to 95 ◦ C with 0.1 ◦ C/s increment. PCR reactions were run in duplicates with beta-actin as control. The PCR products were fractionated by 2% agarose gel electrophoresis containing 0.5 ␮g/ml of ethidium bromide. Images were taken using a UV light board and a polaroid camera. Tissue lysates were fractionated by 9% SDS–PAGE and proteins transferred to Hybond ECL nitrocellulose membranes using semi-dry electro blotting. Membranes were blocked in 2% ECL Advance Blocking Agent, incubated overnight at 4 ◦ C with primary antibody, followed by incubation for 1 h at room temperature with secondary antibody. Membranes were developed using ECL Advance and scanned using a FujiLas 1000 (Fujifilm A/S). The expression of PDE3A, PDE3B, and PDE5A mRNAs in the trigeminovascular system was investigated using real-time PCR. Fig. 1A is representative of three experiments. PDE3A mRNA is present in the middle cerebral artery and basilar artery, mesenteric artery and aorta, but is absent in dura and trigeminal ganglion. PDE3B and PDE5A mRNAs are present in the cerebral and peripheral arteries as well as dura and trigeminal ganglion. Next, the presence of PDE3A, PDE3B, and PDE5A proteins was investigated using Western blot analysis. Fig. 1B is representative of three experiments. PDE3A is present in cerebral and peripheral vessels, but is absent from dura and trigeminal ganglion. PDE3B is present in middle cerebral and basilar arteries although the signal is weak, and could not be detected in the dura, trigeminal ganglion, mesenteric artery, or aorta. PDE5A is present in cerebral and peripheral arteries and is highly expressed in dura, but is only weakly detected in the trigeminal ganglion. PDE mRNA and protein expression data is summarized in Table 2.

Table 2 Overview of PDE mRNA and protein expression Expression

mRNAa

Proteina

PDE

3A

3B

5A

3A

3B

5A

Middle cerebral artery Basilar artery Dura Trigeminal ganglion Mesenteric artery Aorta

+ + − − + +

+ + + + + +

+ + + + + +

+ + − − + +

+ + − − − (+)

+ + + (+) + +

Results based on three separate experiments. “+” strong signals in three experiments; “(+)” minor signal in one to three experiments; “−” no signal in three experiments. a n = 9 for middle cerebral artery and basilar artery. n = 6 for PCR and n = 3 for Western blot for remaining tissues.

Fig. 1. PDE mRNA and protein expression; PDE3A, PDE3B and PDE5 mRNA and protein expression was measured in the middle cerebral artery, basilar artery, dura, trigeminal ganglion, mesenteric artery, and aorta: (A) real-time PCR analysis of mRNA of PDE3A, PDE3B, and PDE5 in duplicates; (B) Western blot analysis of PDE3A, PDE3B, and PDE5A protein in duplicates. Antibodies were used in the following dilutions: anti-PDE3A (1:500), anti-PDE3B (1:1000), antiPDE5A (1:500), and anti-␤-actin (1:10,000). Approximate sizes of proteins are 124 kDa (PDE3A), 123 kDa (PDE3B) and 94 kDa (PDE5A). Results in (A) and (B) are representative of three experiments.

Expression of six CNG ion channel subunit mRNAs in the trigeminovascular system was investigated using real-time PCR. Fig. 2A is representative of three experiments. No CNGA1, CNGA3 and CNGB3 mRNAs are observed in cerebral and peripheral vessels, dura, or trigeminal ganglion. CNG ion channel subunit CNGA4 and CNGB1 mRNA are expressed in middle cerebral and basilar artery as well as mesenteric artery. CNGA2 is also weakly found in the latter. As positive control CNGA2, CNGA4 and CNGB1b mRNAs were found in the olfactoric bulb and CNGA1, CNGA3, CNGB1 and CNGB3 in the retina (Fig. 2B). The presence of CNG ion channel subunit proteins was analyzed by Western blot. Fig. 2C is representative of three experiments. CNGA1 is strongly expressed in trigeminal ganglion and weakly in dura, but not in the cerebral or peripheral vessels. CNGA2 and CNGA3 are present in the middle cerebral and basilar arteries, and trigeminal ganglion, but not in dura or peripheral arteries. The presence of CNGA4, CNGB1, and CNGB3 was

L.S. Kruse et al. / Neuroscience Letters 404 (2006) 202–207

Fig. 2. CNG ion channel subunit mRNA and protein expression; CNGA1–4, CNGB1 and CNGB3 mRNA and protein expression was measured in the middle cerebral artery, basilar artery, dura, trigeminal ganglion, mesenteric artery and aorta: (A) real-time PCR analysis of mRNA of CNG ion channel subunits A1–A4, B1, and B3 in duplicates; (B) real-time PCR analysis of all CNG channels in control tissues: olfactory bulb (O) for CNGA2, CNGA4 and CNGB1b and retina (R) for CNGA1, CNGA3, CNGB1 and CNGB3; (C) Western blot analysis of CNG ion channel subunits A1–A3. Antibodies were used in the following dilutions: anti-CNGA1 (1:100), anti-CNGA2 (1:250), anti-CNGA3 (1:250) and anti-␤-actin (1:10,000). Approximate sizes of proteins are: 79 kDa (CNGA1), 122 kDa (CNGA2) and 70 kDa (CNGA3). Results in (A) and (B) are representative of three experiments.

not analyzed because no good commercial antibodies against these subunits were available at the time of analysis. The CNG ion channel mRNA and protein expression is summarized in Table 3. In this study we have analysed for the first time the distribution of PDE isoforms, PDE3A, PDE3B, and PDE5, and six CNG ion channel subunits, CNGA1, CNGA2, CNGA3, CNGA4,

205

CNGB1 and CNGB3 in the rat trigeminovascular pain signalling system. We used real-time PCR and Western blot analysis on extracts of middle cerebral artery, basilar artery, dura and trigeminal ganglion to measure mRNA and protein expression. We found that the middle cerebral artery and basilar artery express cGMP-inhibited and cAMP-degrading PDE3A and PDE3B and cGMP-degrading PDE5 as well as four subunits of the CNG ion channel: CNGA2, CNGA3, CNGA4, and CNGB1. The trigeminal ganglion expresses PDE3B, PDE5, CNGA1, CNGA2, and CNGA3, and dura expresses PDE3B and PDE5 and, weakly, CNGA1. These findings show that signalling molecules, which may be involved in induction of migraine, are expressed in the trigeminovascular system. In comparison, the mesenteric artery and aorta also expressed PDE3A, PDE3B, and PDE5 but only weakly CNGA4 and CNGB1 subunits. Accordingly, cerebral and peripheral arteries appear similar regarding PDE expression, but slightly different in CNG expression. The discrepancies observed in this study between mRNA and protein expression are not uncommon [31] and may be due to various factors. Technical issues such as sensitivity of the assays or antibodies used as well as low-abundance targets may explain the differences. Several studies have demonstrated effects of PDE inhibitors on cerebral arteries, but only few have described the presence of PDE mRNA or protein in isolated cerebral arteries of various animal species including humans [2,26,32]. We have previously reported expression of PDE1A, PDE1B as well as PDE5A in the cerebral arteries of guinea pigs [15] and humans [13]. It is not known if there is a difference in the PDE expression between the basilar or middle cerebral artery, which may affect the response to inhibitors in the posterior and anterior cerebral vascular bed, respectively. In the present study, the Western blot analysis was performed in a semiquantitative manner, and our results show that PDE3A and PDE5A are expressed at similar levels in the basilar artery and the middle cerebral artery. Furthermore, the level of expression is significantly higher in cerebral than in the peripheral vessels. CNG ion channels were first described in the retina and olfactory neurons [7], but recent findings suggest that they are widely expressed in the organism including the brain [19]. The distribution of CNG ion channel mRNA has been described in the rat brain [10,28], but not in cerebral arteries, dura or trigeminal ganglion. Only the CNGA2 subunit has been detected in a human cerebral artery using in situ hybridisation and immunohistochemistry, but the location of the artery was not specified [6]. In the present study we demonstrated strong expression of several CNG ion channel subunits including CNGA2, CNGA3, CNGA4 and CNGB1 in the middle cerebral and basilar arteries as well as the trigeminal ganglion. Furthermore, the expression level is significantly higher in cerebral than in peripheral vessels. The co-expression of four CNG ion channel subunits and three PDE isoforms in rat cerebral arteries may suggest that both CNG and PDE are involved in modulation of cyclic nucleotide signalling in cerebral arteries. We have previously shown that PDE3 and PDE5 inhibitors induce migraine-like headache in healthy subjects [4,16] and migraine in migraine patients [11,14] and hypothesized that

206

L.S. Kruse et al. / Neuroscience Letters 404 (2006) 202–207

Table 3 Overview of CNG mRNA and protein expression Expression

mRNAa

Proteina

CNG

A1

A2

A3

A4

B1

B3

A1

A2

A3

A4

B1

B3

Middle cerebral artery Basilar artery Dura Trigeminal ganglion Mesenteric artery Aorta

− − − − − −

− − − − (+) −

− − − − − −

+ + − − (+) −

+ + − − + −

− − − − − −

− − (+) + − −

+ + − + − −

+ + (+) + − (+)

n/a n/a n/a n/a n/a n/a

n/a n/a n/a n/a n/a n/a

n/a n/a n/a n/a n/a n/a

Results based on based on three separate experiments. “+” strong signals in three experiments; “(+)” minor signal in one to three experiments; “−” no signal in three experiments; “n/a” no antibodies were available. a n = 9 for middle cerebral artery and basilar artery. n = 6 for PCR and n = 3 for Western blot for remaining tissues.

the targets may be the perivascular pain-sensitive nerve fibres, the trigeminal ganglion or more centrally in the pain pathway. In addition, we now propose that CNG ion channels could be involved in pain signalling mediated by cyclic nucleotides leading to migraine. In conclusion, the presence of PDE3 and PDE5 in cerebral arteries indicates that inhibitors of these enzymes may act in or around the large cerebral arteries in rat. However, inhibition of the enzymes may not only affect vasodilation of the arteries, since the selective PDE5 inhibitor sildenafil is a poor vasodilator both in vitro as well as in vivo where migraine is induced without concomitant cerebral vasodilatation. Thus, based on the present results from rats, PDE3 and PDE5 inhibitors might also modulate cyclic nucleotide responses and CNG ion channel activation in the trigeminal ganglion and the dura in humans, but a species difference cannot be excluded. The colocalisation of PDE3A, PDE3B, and PDE5A and CNG ion channel subunits observed in the trigeminovascular system of the rat should be confirmed in humans. Acknowledgements We gratefully acknowledge Joseph A. Beavo, Vincent Manganiello and Eva Degerman for the generous gift of PDE3 and PDE5 antibodies and professor Beavo for useful comments and suggestions for the experimental work. Robert S. Molday is thanked for a CNGA1 ion channel antibody. Dr. Inger JansenOlesen and Kenneth Beri Ploug are thanked for technical advice. The study was funded by the Hørslev Foundation, The A.P. Møller Foundation for the Advancement of Medical Science, Christian and Ottilia Brorson’s travel grant for young scientists, Cool Sorption Foundation of 1988 and the Danish Medical Research Council. References [1] J.A. Beavo, Cyclic nucleotide phosphodiesterases: functional implications of multiple isoforms, Physiol. Rev. 75 (1995) 725–748. [2] S. Birk, L. Edvinsson, J. Olesen, C. Kruuse, Analysis of the effects of phosphodiesterase type 3 and 4 inhibitors in cerebral arteries, Eur. J. Pharmacol. 489 (2004) 93–100. [3] S. Birk, C. Kruuse, K.A. Petersen, O. Jonassen, P. Tfelt-Hansen, J. Olesen, The phosphodiesterase 3 inhibitor cilostazol dilates large cerebral arteries in humans without affecting regional cerebral blood flow, J. Cereb. Blood Flow Metab. 24 (2004) 1352–1358.

[4] S. Birk, C. Kruuse, K.A. Petersen, P. Tfelt-Hansen, J. Olesen, The headache inducing effect of cilostazol in human volunteers, Cephalalgia (2006) (accepted). [5] S. Bolton, C.T. O’Shaughnessy, P.J. Goadsby, Properties of neurons in the trigeminal nucleus caudalis responding to noxious dural and facial stimulation, Brain Res. 1046 (2005) 122–129. [6] K.T. Cheng, F.L. Chan, Y. Huang, W.Y. Chan, X. Yao, Expression of olfactory-type cyclic nucleotide-gated channel (CNGA2) in vascular tissues, Histochem. Cell Biol. 120 (2003) 475–481. [7] J.T. Finn, M.E. Grunwald, K.W. Yau, Cyclic nucleotide-gated ion channels: an extended family with diverse functions, Annu. Rev. Physiol. 58 (1996) 395–426. [8] P.J. Goadsby, Migraine pathophysiology, Headache 45 (Suppl. 1) (2005) S14–S24. [9] U.B. Kaupp, R. Seifert, Cyclic nucleotide-gated ion channels, Physiol. Rev. 82 (2002) 769–824. [10] P.A. Kingston, F. Zufall, C.J. Barnstable, Widespread expression of olfactory cyclic nucleotide-gated channel genes in rat brain: implications for neuronal signalling, Synapse 32 (1999) 1–12. [11] C. Kruuse, E. Frandsen, S. Schifter, L.L. Thomsen, S. Birk, J. Olesen, Plasma levels of cAMP, cGMP and CGRP in sildenafil-induced headache, Cephalalgia 24 (2004) 547–553. [12] C. Kruuse, T.B. Jacobsen, L.H. Lassen, L.L. Thomsen, S.G. Hasselbalch, H. Dige-Petersen, J. Olesen, Dipyridamole dilates large cerebral arteries concomitant to headache induction in healthy subjects, J. Cereb. Blood Flow Metab. 20 (2000) 1372–1379. [13] C. Kruuse, T.S. Khurana, S.D. Rybalkin, S. Birk, U. Engel, L. Edvinsson, J. Olesen, Phosphodiesterase 5 and effects of sildenafil on cerebral arteries of man and guinea pig, Eur. J. Pharmacol. 521 (2005) 105– 114. [14] C. Kruuse, L.H. Lassen, S. Oestergaard, H.K. Iversen, J. Olesen, Dipyridamole may cause migraine in patients with migraine without aura, Cephalalgia (in press). [15] C. Kruuse, S.D. Rybalkin, T.S. Khurana, I. Jansen-Olesen, J. Olesen, L. Edvinsson, The role of cGMP hydrolysing phosphodiesterases 1 and 5 in cerebral artery dilatation, Eur. J. Pharmacol. 420 (2001) 55–65. [16] C. Kruuse, L.L. Thomsen, S. Birk, J. Olesen, Migraine can be induced by sildenafil without changes in middle cerebral artery diameter, Brain 126 (2003) 241–247. [17] L.H. Lassen, P.A. Haderslev, V.B. Jacobsen, H.K. Iversen, B. Sperling, J. Olesen, CGRP may play a causative role in migraine, Cephalalgia 22 (2002) 54–61. [18] C. Lugnier, Cyclic nucleotide phosphodiesterase (PDE) superfamily: a new target for the development of specific therapeutic agents, Pharmacol. Ther. 109 (2006) 366–398. [19] K. Matulef, W.N. Zagotta, Cyclic nucleotide-gated ion channels, Annu. Rev. Cell Dev. Biol. 19 (2003) 23–44. [20] K.A. Petersen, S. Birk, H. Doods, L. Edvinsson, J. Olesen, Inhibitory effect of BIBN4096BS on cephalic vasodilatation induced by CGRP or transcranial electrical stimulation in the rat, Br. J. Pharmacol. 143 (2004) 697–704.

L.S. Kruse et al. / Neuroscience Letters 404 (2006) 202–207 [21] D. Pietrobon, J. Striessnig, Neurobiology of migraine, Nat. Rev. Neurosci. 4 (2003) 386–398. [22] B.S. Ray, H.G. Wolff, Experimental studies on headache, Arch. Surg. 41 (1940) 813–855. [23] U. Reuter, H. Bolay, I. Jansen-Olesen, A. Chiarugi, M. Sanchez del Rio, R. Letourneau, T.C. Theoharides, C. Waeber, M.A. Moskowitz, Delayed inflammation in rat meninges: implications for migraine pathophysiology, Brain 124 (2001) 2490–2502. [24] M. Sanchez-Del-Rio, U. Reuter, M.A. Moskowitz, New insights into migraine pathophysiology, Curr. Opin. Neurol. 19 (2006) 294–298. [25] P. Sarchielli, A. Alberti, M. Codini, A. Floridi, V. Gallai, Nitric oxide metabolites, prostaglandins and trigeminal vasoactive peptides in internal jugular vein blood during spontaneous migraine attacks, Cephalalgia 20 (2000) 907–918. [26] C.G. Sobey, L. Quan, Impaired cerebral vasodilator responses to NO and PDE V inhibition after subarachnoid hemorrhage, Am. J. Physiol. 277 (1999) H1718–H1724. [27] S.H. Soderling, J.A. Beavo, Regulation of cAMP and cGMP signaling: new phosphodiesterases and new functions, Curr. Opin. Cell Biol. 12 (2000) 174–179.

207

[28] P.J. Strijbos, G.D. Pratt, S. Khan, I.G. Charles, J. Garthwaite, Molecular characterization and in situ localization of a full-length cyclic nucleotide-gated channel in rat brain, Eur. J. Neurosci. 11 (1999) 4463– 4467. [29] C. Tassorelli, F. Blandini, R. Greco, G. Nappi, Nitroglycerin enhances cGMP expression in specific neuronal and cerebrovascular structures of the rat brain, J. Chem. Neuroanat. 27 (2004) 23–32. [30] L.L. Thomsen, C. Kruuse, H.K. Iversen, J. Olesen, A nitric oxide donor (nitroglycerin) triggers genuine migraine attacks, Eur. J. Neurol. 1 (1994) 73–80. [31] A.D. Weston, L. Hood, Systems biology, proteomics, and the future of health care: toward predictive, preventative, and personalized medicine, J. Proteome. Res. 3 (2004) 179–196. [32] R.N. Willette, A.O. Shiloh, C.F. Sauermelch, A. Sulpizio, M.P. Michell, L.B. Cieslinski, T.J. Torphy, E.H. Ohlstein, Identification, characterization, and functional role of phosphodiesterase type IV in cerebral vessels: effects of selective phosphodiesterase inhibitors, J. Cereb. Blood Flow Metab. 17 (1997) 210–219.