Suboccipital cerebrospinal fluid and plasma concentrations of somatostatin, neuropeptide Y and beta-endorphin in patients with common migraine

Neuropeptides (1992) 22, 111-116 0 Longman GroupUK Ltd 1992

Suboccipital Cerebrospinal Fluid and Plasma Concentrations of Somatostatin, Neuropeptide Y and Beta-Endorphin in Patients with Common Migraine L. VECSEI*, E. WIDERLdVt, R. EKMANS, K. KOVACS§, I. JELENCSIKS, Gy. BOZSIKS and G. KAPGCS§ “Department of Neurology, University Medical School of P&s, P&s, Hungary. TDepartment of Psychiatry, University Hospital, University of Uppsala, Sweden. #Department of Psychiatry and Neurochemistry, University of Lund, Lund, Sweden. IDepartment of Neurology, Semmelweis Medical School, Budapest, Hungary

Abstract-The somatostatin-like (SLI), the neuropeptide Y-like (NPY-LI), and the betaendorphin-like (BE-LI) immunoreactivities of cerebrospinal fluid (CSF) obtained by suboccipital puncture, or plasma from patients suffering from common migraine or other neuropsychiatric disorders were analysed. The SLI concentration was tendentiously decreased in the migraine patients during the attack-free period compared to that of a ‘mixed neuropsychiatric group’. During the migraine attack the level of SLI was further decreased. Similar alteration was found in the CSF BE-LI, while the BE-LI in the plasma showed only a tendentious decrease in common migraine patients. The NPY-LI did not change during the attack period in the CSF or plasma. These findings may indicate the possible role of somatostatin in the pathogenesis of common migraine, and support earlier observations that beta-endorphin is involved in the development in this disorder.

Introduction The essential feature of migraine is a recurrent paroxysmal headache. The cause is still largely unknown but there is often a family history, the attacks are often related to some emotional disturbance, and sometimes precipitated by certain foods. The pathogenesis of this illness has been suggested to be a complex interaction between inherited traits Date received 1 November 1991 Date accepted 15 January 1992

and exogenous and endogenous factors (1). Neurovascular instability is thought to be one of the major inherited traits among the migraine sufferers. Several authors indicate that the hypothalamus is the anatomical place of this functional anomaly (2, 3). The hypothalamus contains high amounts of the tetradecapeptide somatostatin (4) and the concentration of this neuropeptide in brain tissues and cerebrospinal fluid (CSF) is markedly changed in several neurological and psychiatric illnesses (see review: 5). 111

112 The neurovascular system is influenced by several peptides, including neuropeptide Y (NPY). The peptide is a member of the pancreatic polypeptide family containing 36 amino acid residues, with a Cterminal tyrosine amide (6). The preclinical and clinical importance of this peptide have been summarized in recent overviews (7, 8). The presence of NPY in areas crucial for blood pressure control, as well as the coexistence and possible interactions of NPY and central catecholamines and somatostatin are well documented (7). Microinjections of NPY into the nucleus of the solitary tract produces a biphasic blood pressure response, with low doses inducing a slight hypertension, and higher doses causing hypotension (9). Furthermore, peripherally administered NPY induces hypertension (9, 10). Beta-endorphin is a 3 1 amino acid peptide from the COOH-terminal segment of the beta-lipotropin molecule (see review 11). The role ofthe opioidpeptides, including beta-endorphin, in pain perception and in modulation of endocrine and autonomic activity is well established (12). There is a CSF bulk flow from the central cerebroventricles into the cisterna magna (cerebellomedullary cistern) and the lateral pontine cisterns. About one-fifth then passes into the perimedullary subarachnoid spaces, while the rest of the fluid directly traverses the basal cisterns (13). Its composition is supposed to be the result of material exchange with the blood and adjacent brain tissues. Therefore, the suboccipital CSF sample gives more specific information about the brain metabolism than that ofthe lumbar CSF sample, which have been exchanged with materials also from the spinal cord. On the basis of these considerations the aim of the present study was to investigate the somatostatinlike immunoreactivity (SLI), the neuropeptide Ylike immunoreactivity (NPY-LI) and the betaendorphin-like immunoreactivity (BE-LI) in CSF obtained by suboccipital puncture, and in plasma samples from patients suffering from common migraine in two phases of the illness, during an attack and during an attack-free period. These patients were compared with a ‘mixed neuropsychiatric group’ comprising patients with various other neurological or psychiatric disorders (epilepsy, neurosis).

NEUROPEPTIDES

Methods

Subjects The study was approved by the Human Ethics Committee of the Semmelweis Medical School, Budapest, Hungary. The diagnoses of migraine was made in accordance with the criteria of the Ad Hoc Committee on Classification of Headache (14). Group I. Common migraine (during attack). The average age of 9 female patients was 39.3 years (range 34-54 years) and the average duration of their illness was 15.2 years (range 4-26 years). Group II. Common migraine (attack free period). The average age of 13 female patients was 37.6 years (range 23-50 years) and the average duration of the illness was 9.1 years (range 4- 14 years). There is no overlap between the subjects in group I and group II. Group III. Mixed neuropsychiatric group. A. The average age of 5 male and 2 female patients suffering from epilepsy was 43.6 (range 3 l-63) years. The suboccipital puncture was performed 24 (range 1248) h after the first epileptic seizure in their life. B. The average age of 2 male and 2 female patients suffering from neurosis was 38.4 (range 31-44) years with a mean duration of the illness of 2-3 years. The patients had no signs or symptoms of neurological illness by physical examination at the time of the suboccipital puncture and the computerized tomography scan picture was negative. All subjects had been drug-free for at least 14 days prior to the sampling occasion. Ideally a group of age- and sexmatched healthy volunteers would be preferable to a control group. However, for ethical reasons it was not possible to perform a suboccipital puncture in healthy subjects. Suboccipital CSF sampling puncture All CSF samples were drawn with the subjects in a sitting position. Before the procedure, the patients stayed on the ward overnight where a low physical activity was allowed. They all had received a standard breakfast and smoking was not allowed before the puncture. A 4 ml sample of CSF was obtained from each subject. The CSF was immediately cen-

SUBOCCIPITAL

CEREBROSPINAL

FLUID AND PLASMA

CONCENTRATIONS

trifuged at 1000 x g for 10 min. The supematant was divided into 1 ml aliquots in plastic vials and stored at -80°C until assayed. Blood sampling puncture For measurement of NPY and BE-L1 concentrations in plasma, blood was collected from the anterocubital vein into tubes containing EDTA immediately after the suboccipital puncture. The samples were centrifuged at +4”C and 1000 x g for 15 min and stored at -80°C until analysis. Samples were coded and all peptide determinations were performed blind to the investigator. For methodological reasons SLI was not determined in the plasma. Biochemical measurements Determination of CSF concentrations of SLI. SLI was quantified using a previously described radioimmunoassay (15). The somatostatin antiserum (Milab, Malmii, Sweden) was used in a fmal dilution of 1: 25 000. It does not cross-react with any other known neuropeptide besides cyclic somatostatin- 14 ( 1OO%), linear somatostatin- 14 (50%), (tyrl)-somatostatin-14 (100%) and (tyrl l)-somatostatin(38%). Samples and synthetic somatostatin- 14 as standard were incubated with the antiserum for 24 h at +4’C. Bound and free tracer were separated using dextran-coated charcoal (0.5% activated charcoal, 0.1% Dextran T-70 in phosphate buffer 0.05 M, pH 7.5, containing 0.25% human serum albumin). The detection limit was 10 pg/ml. The intra- and interassay variations were 5% and 15.9%, respectively. Radioimmunoassay of NPY. The method has previously been described in detail by Widerlov et al (16). For the radioimmunoassay of NPY a rabbit antiserum raised against synthetic porcine NPY (a generous gift from P. C. Emson, Cambridge, England) conjugated to bovine serum albumin with carbodiimide was used. 1251-NPY, used as tracer, was purified by high performance liquid chromatography (HPLC). The antiserum cross-reacted with peptide YY to an extent of 33% but not with C-terminal fragments of NPY and PYY (NPY 13-36 and PYY 13-36) nor with bovine pancreatic polypeptide, gastrin inhibiting peptide, vasoactive intestinal peptide, peptide histidine isoleucin (PHI) and secretin.

113

Antiserum (200 j.0, diluted 1:40 000) was incubated first with 100 pl of standard (synthetic NPY; Peninsula, Belmont, Ca, USA) or sample for 24 h at +4”C and 200 pl (about 2500 cpm) of the HPLC purified tracer for another24 h. Bound and free 1251NPY were separated using dextran-coated charcoal. Each sample was assayed in serial dilution and corrected for unspecific binding. The detection limit was 11.7 pmol/l. Intra-assay variation was 6.5% while the interassay variation was 7.0%. Because of the cross-reactivity with PYY, each sample was also assayed for PYY-LI and 33% of the values obtained was subtracted from the corresponding values for NPY-LI (16). Radioimmunoassay ofBE-LI Beta-endorphin concentrations were determined by radioimmunoassay using a BE antiserum with a detection limit of 12 pgrnl and with a negligible cross-reactivity against betalipotropin. Recently the original method ( 17) has been modified by the elimination of the extraction step (18). The inter- and intra-assay coefficients of variance were 7.2% and 7.1% respectively. Statistical analysis The neurochemical data were evaluated by analysis of variance (ANOVA) followed by Tukey’s a posteriori test.

Results The concentrations of the peptides in the CSF and plasma did not significantly differ between the male and female subgroups ofthe mixed neuropsychiatric group. The CSF SLI concentration of common migraine patients during the attack-free period was tendentiously lower than that of the ‘mixed neuropsychiatric group’. However, during the attack-period the CSF SLI was significantly decreased (F(2,32) = 4.65, p < 0.02 ANOVA, p c 0.05 Tukey, vs migraine attack free patients; p < 0.01 Tukey, vs ‘mixed neuropsychiatic group’ (Fig. 1). The NPY concentrations in CSF and plasma of the migraine patients during the attack and attackfree period did not differ significantly from each other, or from the ‘mixed neuropsychiatric group’ (Fig. 2).

114

NEUROPEPTIDES

SOMATOSTATIN

pmol

(SLI)

BETA-ENDORPHINE

CSF

(BE-LI) PLASMA

CSF pmoll 100.

f r,r+ 50

f,, 1 5 m

11)

I

0-k 4, I Common II

Common

migraine

(attack)

mlgralne

(free

tf,, ?J common mlprslne

(attack)

Common mi~ralne

III

Mixed

neuropaychlatrlc

period) DrWP

(tree period1

lU Mlxed neuropsychlalrlc

group

Fig. 1 Suboccipital CSF somatostatin concentrations in migraine during attack and attack-free period. x = p < 0.05 (vs migraine attack-free group), xx = p < 0.01 (vs mixed neuropsychiatric group), (Tukey test). Vertical lines represent the standard error of the mean. Number in bars represents the number of patients in the different groups.

Fig. 3 Suboccipital CSF and plasma beta-endorphin concentrations in common migraine patients during attack and attackfree period. xx = p < 0.01 (vs ‘mixed neuropsychiatric group’; ANOVA followed by Tukey test). Vertical lines represent the standard error of the mean. Number in bars represents the number of patients in the different groups.

Discussion NEUROPEPTIDE

PLASMA

CSF

I

pt7lOlll

Y (NPY-LI)

100

2,

II I

Common

11 Common m

Mixed

mlgralne

(attack)

mlgralne

(tree

neuropsychlatrio

period) group

Fig. 2 Suboccipital CSF and plasma NF’Y concentrations in common migraine patients during attack and attack-free period. Vertical lines represent the standard error of the mean. Number in bars represents the number of patients in the different groups.

The beta-endorphin level in CSF of common migraine patients during the attack period was significantly decreased (F(2,32) = 5.89, p < 0.01, ANOVA; p < 0.01 Tukey test) compared to that of the ‘mixed neuropsychiatric group’. Both the CSF and plasma concentrations of beta-endorphin in migraine patients during the attack-free period was tendentiously lower compared to the ‘mixed neuropsychiatric group’ (Fig. 3).

CSF somatostatin deficiencies have been found in several neurological (parkinsonism, multiple sclerosis etc.) and psychiatric (depression, Alzheimer’s disease etc.) disorders (see reviews: 5, 19), suggesting that this peptide has several functions in the organization of brain processes. Preclinical studies have shown several behavioural effects of somatostatin i.e. influence on the locomotor activity in a dose-related manner (20), antiamnesic action (21), and inhibition of the extinction of active avoidance behaviour (22). In the present clinical study we found a reduced CSF SLI concentration in migraine patients, particularly during a migraine attack, when compared to a mixed neuropsychiatric group. Neurovascular instability seems to be one of the key factors in the pathogenesis of a migraine attack. Caleri et al (23) demonstrated a venoconstrictive action of somatostatin on dorsal hand vein in vivo. Moreover, data are emerging about a stimulation of the adrenergic system innervating the arterial and venous cerebral vessels by somatostatin. In fact, somatostatin increased the activity of tyrosine hydroxylase, the rate limiting enzyme in the catecholamine synthesis, in sympathetic superior cervical ganglia (24), where somatostatin coexists with noradrenaline. A direct vasoconstrictive effect of this peptide on arte-

SUBOCCIPITAL

CEREBROSPINAL

FLUID AND PLASMA

CONCENTRATIONS

rial or venous cerebral vessels, or an increased cerebral vascular tone mediated by a stimulation of sympathetic superior cervical ganglia, might be involved in the pathogenesis of migraine. It has also been suggested that the pain of the migraine patients is of central origin, and the vasodilatation may be considered as a concomitant vegetative symptom along with nausea, vomiting, shivering, lacrimation etc. (25). Rezek et al (26) described an opiate-like naloxone reversible effect of somatostatin. Therefore, the reduced concentration of CSF SLI during the migraine attack might be involved in the pathogenesis of pain in the migraine patients. The concentration of NPY in the CSF and plasma of common migraine patients did not differ during the attack, or attack-free period, from that of the ‘mixed neuropsychiatric group’. Recently, Jansen et al (27) found that NPY administered to human temporal arteries never caused any response in the majority of the arteries. Generally, it appears that NPY does not induce contraction per se in human temporal arteries. This observation, and our present data, suggest that NPY is not a major factor in the pathogenesis of the migraine attack. It is well established that the endogenous opioid peptides, considered to be physiological neurotransmitters and/or neuromodulators, play a role in the pathogenesis of hyperociceptive syndromes (28). In agreement with our suboccipital CSF findings Baldi et al (29) have previously found reduced BE concentrations in CSF (obtained by lumbar puncture) of migraine patients during the attack, when compared to the levels in the migraine free period. Furthermore, they did not detect significant difference in the plasma BE levels of healthy subjects and of patients during the migraine free period. Our findings suggest that somatostatin but not neuropeptide Y is involved in the pathogenesis of common migraine attack. The concentration of neuropeptide Y in CSF of patients suffering from migraine did not change compared to that of a heterogeneous mixed neuropsychiatric group.

Acknowledgements This study was supported by grants from H. Lundbeck A/S Copenhagen, Denmark, the Bank of Sweden Tercentenary Foundation (85/77), the Faculty of Medicine, University of Lund,

115

and from the Swedish Medical Research Council (07517). L.V. held a visiting scientist fellowship at the Department of Psychiatry and Neurochemistry at Lund University.

References 1. Diamond, S. and Medina, J. L. (1980). Review article: Current thoughts on migraine. Headache 20: 208-212. 2. Pearce, J. (1969). Migraine: clinical features, mechanisms and management. Thomas, Springfield. 3. Herberg, L. J. (1967). The hypothalamus and the aetiology of migraine. In: Smith, R. ed. Background to migraine. Heinemamr, London, pp. 96-105. 4. Brownstein, M., Arimura, A., Sato, H., S&ally, A. V. and Kizer, J. S. (1975). The regional distribution of somatostatin in the rat. Endocrinology 96: 1456-1461. 5. Vecsei, L. and Widerliiv, E. (1988). Brain and CSF somatostatin concentrations in patients with psychiatric and neurological illness. An overview. Acta Psychiatr. Stand. 78: 657-667. 6. Tatemoto, K. ( 1982). Neuropeptide Y: Complete amino acid sequence of the brain peptide. Proc. Natl. Acad. Sci. USA 79: 5485-5489. 7. Wahlestedt, C., Ekman, R. and Widerlov, E. (1989). Neuropeptide Y (NPY) and the central nervous system: Distribution, effects and possible relationship to neurological and psychiatric disorders. Prog. Neuro-Psychopharmacol. Biol. Psychiat.13: 31-54. 8. Heilig, M. and Widerlov, E. (1990). Neuropeptide Y. An overview of central distribution, functional aspects, and possible involvement in neuropsychiatric illnesses. Acta Psychiatr. Stand. 82: 95-l 14. 9. Carter, D. A., Vallejo, M. and Lightman, S. L. (1985). Cardiovascular effects of neuropeptide Y in the nucleus tractus solitarius of rats: relationship with noradrenaline and vasopressin. Peptides 6: 42 l-425 10. Petty, M. A., Dietrich, R. and Lang, R. E. (1984A). The cardiovascular effects of neuropeptide Y (NPY). Clin. Exp. Hypertens. 6: 1889-1892. 11. Li, C. H. (1984). Beta-endorphins: some aspects of structurefunctional relationship. In: Muller E. E., Genazanl A. R. (eds) Central and Peripheral Endotphins: Basic and Clinical Aspects. Raven Press, New York, pp 17-26. 12. Nappi, G., Facchinetti, F., Martignoni, E., Petraglia, F., Bono, G., Volpe, A., Sinforiani, E., Manzonl, G. C. and Genazzani, A. R. (1984). Central and peripheral endorphins as biological markers in headache. In: Muller, E. E. and Genazzani, A. R. (eds) Central and Peripheral Endorphins: Basic and Clinical Aspects. Raven Press, New York, pp 301307. 13. Mumenthaler, M. (1983). Neurology. Georg Thieme Verlag, Stuttgart. 14. Ad Hoc Committee on Classification of Headache (1962) JAMA 179: 717-718. 15. Wallengren, J., Ekman, R. and Sundler, F. (1987). Occurrence and distribution of neuropeptides in human skin. Acta Dermatol. Venerol. (Stockh.) 67: 185-192. 16. Wide&%, E., Lindstrom, L. H., Wahlestedt, C. and Ekman, R. (1988). Neuropeptide Y and peptide YY as possible cerebrospinal fluid markers for major depression and schizophrenia, respectively. J. psychiat. Res. 22: 69-79. 17. Bramnert, M., Ekman, R., Larsson, I. and Thorell, J. L. (1982). Characterization and application of a radioimmuno-

116

18.

19. 20.

2 1.

22.

23.

assay for beta-endorphin using an antiserum with negligible cross-reactivityagainstbeta-lipotropin. Regul. Pept. 5: 65-75. Ekdahl, C., Ekman, R., Andersson, S. J., Melander, A. and Svensson, B. (in press). Dynamic training and circulating levels of corticotropin-releasing factor, beta-lipotropin and betaendorphin in rheumatoid arthritis. Pain. Rubinow, D. R. (1986). Cerebrospinal fluid somatostatin and psychiatric illness. Biol. Psychiatry 21: 341-365. Rezek, M., Havlicek, V., Leybin, L., Pinsky, C., Kroeger, E. A., Hughes, K. R. and Friesen, H. (1977). Neostriatal administration of somatostatin differential effect of small and large doses on behavior and motor control. Can. J. Physiol. Pharmacol. 55: 234-242. Vecsei, L., Bollok, I. and Telegdy, G. (1983). Intracerebroventricular somatostatin attenuates electroconvulsive shockinduced amnesia in rats. Peptides 4: 293-295. Vecsei, L., Bollok, I. and Telegdy, G. (1983). Comparative studies with cyclic and linear somatostatin on active avoidance behaviour and open-field activity in rats. Acta. Physiol. Hung. 61: 43-49. Caleri, D., Marabini, S., Panconesi, A. and Pietrini, U. (1987). A pharmacological approach to the analgesizing mechanism of somatostatin in cluster headache. La Ricerca

NEUROPEPTIDES

Clin. Lab. 17: 155-162. 24. Kessler, J. A., Adler, J. E. and Black, I. B. (1983). Substance P and somatostatin regulate sympathetic noradrenaline function. Science 221: 1059-1061. 25. Sicuteri, F. (1972). Headache as possible expression of deficiency of brain 5 hydroxytryptamine (central denervation supsensitivity). Headache 12: 69-72. 26. Rezek, M., Havlicek, V., Leybin, L., LaBella, F. S. and Friesen, H. (1978). Opiate-like naloxone-reversible actions of somatostatin given intracerebrally. Can J. Physiol. Pharmacol. 56: 227-23 1 27. Jansen, I., Uddman, R., Hocherman, M., Ekman, R., Jensen, K., Olesen, J., Stierholm, P. and Edvinsson, L. (1986). Localization and effects of neuropeptide Y, vasoactive intestinal polypeptide, substance P and calcitonin gene-related peptide in human temporal arteries. Ann. Neurol. 20: 496-501. 28. Terenius, L. (1978). Endogenous opioid peptides and analgesia. Ann. Rev. Pharmacol. Toxicol. 18: 189-205. 29. Baldi, E., Conti, P., Conti, R., Fanaciullacci, M., Michelacci, S., Salmon, S., Sicuteri, F. and Spillantini, M. G. (1986). Hypemociceptive syndromes and pharmacological inhibition of endogenous opioid degradation. Int. J. Clin. Pharmacol. Res. VL(3): 193-197.