Pain, 51 (1992) 163-168

0 1992 Elsevier Science Publishers B.V. All rights reserved 0304-3959/92/$05.00

163

PAIN 02132

Plasma and cerebrospinal fluid P-endorphin in chronic tension-type headache

Flemming W. Bach a,b, Michael Langemark ‘, Niels H. Secher ’ and Jes Olesen a

Departments of a Neurology and b Clinical Chemistry, University of Copenhagen, Gentofte Hospital, DK-2900 Hellerup (Denmark) and ’ Department of Anesthesiology, University of Copenhagen, Rigshospitalet, DK-2100 Copenhagen (Denmark)

(Received 3 February 1992, revision received 1 April 1992, accepted 21 May 1992)

Summary Previous studies have provided evidence of an increased sensitivity to pain, a decreased hypothala- mic opioid tone, and decreased cerebrospinal fluid (CSF) j%endorphin (P-EP) concentration in patients with primary chronic headache. We applied separate specific radioimmunoassays for P-EP in CSF and plasma on samples from age-matched controls and a group of 50 patients with chronic tension-type headache (CTH) fulfilling the diagnostic criteria set by the International Headache Society. Median CSF P-EP concentrations (95% confidence limits) were 12.8 pmol/l (11.0-14.5) in CTH patients and 11.9 pmol/l (10.9-14.2) in the control group, which is not significantly different (P = 0.28). Plasma /3-EP concentrations did not differ either, being 3.1 pmol/l (2.4-3.7) and 3.3 pmol/l (1.8-4.0) in the patients with CTH and in controls, respectively (P = 0.88). Plasma and CSF P-EP concentrations did not correlate. Reversed-phase high performance liquid chromatography (HPLC) of CSF pools from the headache patients and controls revealed similar profiles of p-EP-immunoreactivity both when C-terminally and N-terminally directed antisera were used, suggesting a normal post-translational processing of the pro-opiomelanocortin gene in patients with CTH. /3-EP is not involved in the pathogenesis of CTH, or such a role is not reflected in CSF or plasma concentrations of the neuropeptide.

Key words: Beta-endorphin; Cerebrospinal fluid; Chromatography, high performance liquid; Headache; Opioid peptides; Pain

Introduction

Chronic tension-type headache (CTH) represents the severest cases of tension-type headache, i.e., those having headache half of all days or more. The lifetime prevalence is 3% (Rasmussen et al. 1991). Pericranial myofascial tenderness is a prominent but not univcr- sally occurring phenomenon in this disorder (Lan- gemark et al. 1988; Schoenen et al. 1991). Increased sensitivity to pain induced by pressure (Schoenen et al. 1991). heat and cold (Langemark et al. 19891, not only pericranially but also in the extremities, indicates a generalized and multimodal increase in pain sensitivity suggesting that, apart from myofascial nociception, de-

Correspondence to: Flemming W. Bach, M.D., Anesthesiology

Research Laboratory 0818 UCSD, Gilman Drive, La Jolla, CA

92093-0818. USA. FAX: (619) 543-6070.

creased activity in the endogenous antinociceptive sys- tems may have a pathogenetic role in CTH (Olesen 1991; Schoenen et al. 1991). Involvement of the opioid part of the endogenous antinociceptive system has pre- viously been hypothesized (Sicuteri 1982), and support to this theory has been delivered by studies finding decreased levels of the endogenous opioid peptide P-endorphin (P-EP) in plasma and ccrebrospinal fluid (CSF) of patients with chronic headache (Facchinetti et al. 1981; Genazzani et al. 1984). However, since con- flicting results have emerged (Fettcs et al. I9851 pcr- haps due to methodological differences in the mea- surement of P-EP and differences in the selection of patients, we performed a study on a group of patients selected in accordance with the new criteria for CTH accepted by the International Headache Society (Headache Classification Commitee of the Interna- tional Headache Society lY88) using 2 separate ra- dioimmunoassays that have previously been validated

164

as specific for P-EP in plasma and CSF (Bach et al. 1986, 1987).

Emphasis on assay specificity is important because of the different processing patterns of the /3-EP pre- cursor pro-opiomelanocortin (POMC) in the anterior pituitary and brain (Smith and Funder 1988). p-lipo- tropin and /3-EP are responsible for the major part of /3-EP immunoreactivity @-EP-ir) found in plasma (Laatikainen et al. 1985; Bach 1989). In some locations in the brain and brain stem, however, P-EP derivatives like a,N-acetylated /3-EP, P-EP,_,, and P-EP,_,, are more abundant than authentic /3-EPo_,,, (Zakarian and Smyth 1982; Millington and Smith 1991). Circulat- ing P-EP is considered to be of mainly anterior pitu- itary origin, whereas @-EP in CSF is derived from cells located within the CNS (Kerdelhue et al. 1982; Wood 1982; Schlachter et al. 1983). Thus, P-EP in these 2 compartments presumably reflects activity in more or less separate opioid systems (Millan 1986).

Material and methods

A pathophysiological study of CTH included 50 patients. CSF

was obtained from 47 patients (30 female, 17 male). The median age

was 43 years (range: 20-70 years). The patients were recruited from

the Copenhagen Acute Headache Clinic, a private neurology prac-

tice and a neurological hospital department. Inclusion in the study

was preceded by a detailed interview, and neurological examination

provided that symptoms and signs were in accordance with the IHS

criteria for CTH (Headache Classification Commitee of the Interna-

tional Headache Society 1988). Patients reporting migraine attacks

more than 1 day/month were excluded as well as patients with

severe or moderate cardiovascular disease, any endocrinological dis-

ease or other chronic diseases. Nineteen patients suffered from

migraine attacks 1-12 days/year. Patients taking part discontinued

any consumption of opioid, neuroleptic or vasoactive drugs at least 2

weeks prior to the investigation. Lumbar puncture was performed

between 08.00 and il.00 h, with the subject in the lateral recumbent

position after local analgesia with lidocaine. CSF was collected in a

fixed sequence in chilled polypropylene tubes. Tubes for p-EP con-

tained 1000 KIU/ml of aprotinin (Trasylol, Bayer, Leverkusen, Ger-

many). The samples were immediately placed on ice and centrifuged

at 4°C within 30 min. The supernatant was kept frozen at -20°C

until analyzed. CSF samples were concurrently obtained in the same

manner from 8 healthy paid volunteers and 36 patients who received

spinal analgesia before minor operations for diseases not associated

with pain. Premeditation was 0.2 mg/kg diazepam. Persons with

neurological, psychiatric or endocrinological diseases or use of cen-

trally acting drugs were excluded. Thus, the control material con-

sisted of 18 women and 26 men with a median age 42 years (range:

20-69).

Blood samples for plasma p-EP anafyses taken SimuItaneously with lumbar puncture were avaiiable in 41 of the patients, and the

results were compared to age- and sex-matched controls obtained

concurrently. Plasma /3-EP was determined by a specific N-termi-

nally directed radioimmunoassay previously described (Bach et al.

1987). p-EP concentrations in CSF were measured in 3-ml samples

by a separate C-terminally directed radioimmunoassay (Bach et al. 19921.

To investigate possible differences in the content of processing

products of the p-EP precursor POMC, a 34-ml pool of CSF from

the patients with CTH was submitted to Cl8 reversed-phase HPLC.

The elution profiles as they appeared after N- and C-terminally

directed p-EP radioimmunoas~ys of the fractions were compared to

previous profiles obtained in healthy subjects (Bach et al. 19X6).

For statistical evaluation. comparison of 2 samples was per-

formed with the chi-square and Mann-Whitney test. and correlation

was tested by the Spearman test. P levels less than 0.05 were

considered significant. The study was performed in accordance with

the Helsinki II declaration and approved by the local ethical commit-

tee.

Results

The median ages of lumbar-punctured CTH pa- tients (43 years) and controls (42 years) were not differ- ent (Mann-Whitney, P = 0.991, but the control group contained more men (chi square, P = 0.02). However, there was no significant difference between women and men with regard to CSF and plasma P-EP. The median CSF P-EP concentrations (95% confidence limits) were 12.8 pmol/l (11.0-14.5) and 11.9 pmol/l (10.9-14.2) in the CTH patients and the control group, respectively (Fig. 1). The median values were not sig- nificantly different (Mann-Whitney, P = 0.28). Median CSF levels of /3-EP were 13.9 pmol/l (11.0-16.8) in patients suffering from 1-12 migraine attacks yearly, and 11.8 pmol/l (10.3-14.6) in patients without mi- graine (Mann-Whitney, P = 0.13).

The profiles of p-EP-ir after Cl8 reversed-phase HPLC and radioimmunoassay of P-EP using a N- terminally directed antiserum (Figs, 2A and 3A) and a C-te~inally directed antiserum (Figs. 2B and 3BI on pools of CSF from the headache patients (Fig. 2) and a group of healthy persons (Fig. 3) showed striking simi- larity. The C-terminally directed antiserum only de- tected 1 peak corresponding to authentic @EP,,_,,, in both groups. As in controf persons, several peaks ap- peared when the ~-terminally directed antiserum was applied. The main peaks appear in both groups. C-

CSF

”

01 I 1 I CTB controls

Fig. 1. CSF p-EP concentrations in 47 patients suffering from chronic

tension-type headache (CTH) and in 44 controls. The median PEP levels (horizontal lines) were not different.

lmoltfrac 1-31

1 A

r?-_______

__-r B -40 ,_r/

__-’

__-- a, __fl- z

.,---------- -325 ;

;i;

70 Retention time

(mmutesl

Fig. 2. Reversed-phase HPLC of a CSF pool from patients with CTH. Eluted fractions were assayed for /3-EP using the N-terminally directed

(A), and the C-terminally directed (B) antiserum. The column was calibrated with authentic p-EP (@-EP,_3,). The acetonitrile gradient is

indicated in B. The apparent peaks are numbered for comparison with Fig. 3.

terminally truncated P-EP fragments such as P-EP,_2, The median plasma concentration of P-EP was 3.1 and PEP,_,, or cY,N-acetylated P-EP were not de- pmol/l (2.4-3.7) and 3.3 pmol/l (1.8-4.0) in the pa- tected. tients with CTH and in controls, respectively (Fig. 4).

fmol/frac

30

I

Ndrminal ab

20

Ii 6

)_

a 30 40 50 60 70 Retentton time (minutes)

Fig. 3. Reversed-phase HPLC of a CSF pool from healthy controls. Eluted fractions were assayed for p-EP using the N-terminally directed (A),

and the C-terminally directed (B) antiserum. The column was calibrated with human p-EP _ methionyl-5-p-EP,_,,, /34ipotropin, &EP,_,,, /3-EP,_

, 17 (y-endorphin), a,N-acetylated P-EP,_,,, 27, P-EP,_,, and cr,N-acetylated P-EP,_,,. The acetonitrile gradient is indicated in B.

Reproduced by permission from Bach et al. (1986).

Plasma

0

0

0 0

#l =b CTli controls

Fig. 3. Plasma p-EP in 41 patients suffering from CTH and in 31

age- and gender-matched controls. The median p-EP levels (hori-

zontal lines) were not different. # denotes the detection limit of the

plasma p-EP assay.

plasma 0-endorphin (pmol/l)

Fig. 5. Scatterplot of plasma and CSF p-EP concentrations in pa-

tients with CTH.

These values are not significantly different (Mann- Whitney, P = 0.88). Patients with and without migraine had similar plasma /3-EP levels (Mann-Whitney, P = 0.58). There was no correlation between paired P-EP values from plasma and CSF in the headache patients (Spearman Rank, P = 0.6) (Fig. 5).

Discussion

The /3-EP level did not differ between patients with CTH and controls in CSF or plasma. The over-repre- sentation of men in our CSF control group is not considered to influence the results, since previous stud- ies (like this) showed that gender lacked influence on CSF P-EP concentrations (Neuser et al. 1984; Scheller et al. 1985; Jolkkonen et al. 1987; Bach et al. 1992).

Previous studies of /3-EP in plasma showed reduced levels in 8 patients (Baldi et al. 1982) and 11 patients (Facchinetti et al. 1981) with daily chronic (tension) headache. A normal /?-EP level was found in 47 pa-

tients with so-called daily vascular headache without concomitant neurological symptoms (Fettes et al. 1985). Thus, with the present study, we were unable to pro- vide support to the hypothesis that circulating P-EP is a marker of CTH or has a role in antinociception. Low levels of plasma P-EP were reported during migraine attacks (Baldi et al. 1982), but that finding was not confirmed by us (Bach et al. 1985). During headache- free periods in migraine patients, plasma P-EP has been reported to be normal (Facchinetti et al. 1981; Baldi et al. 1982; Bach et al. 19851, although one group found decreased levels in patients suffering from mi- graine with aura (Fettes et al. 1985). Evidence in support of a role of plasma P-EP in antinociception is scarce. Recently, however, a study in rats suggested that pituitary-derived /3-EP participates in physiologi- cal modulation of pain (Hargreaves et al. 1990).

More evidence points towards a role of /3-EP lo- cated in the CNS, especially in the arcuate nucleus of the hypothalamus and in the periaqueductal gray area, in pain modulation (Besson and Chaouch 1987; Millan et al. 1987; Hamba 1988). Patients with chronic mi- graine with interparoxysmal headache secreted less luteinizing hormone to the circulation than control persons after an i.v. naloxone injection, suggesting a decreased hypothalamic opioid tone in the headache patients (Facchinetti et al. 1988). The same dose of naloxone (4 mg) did, however, not aggravate pain, evaluated as visual analog and verbal ratings, in the group of CTH patients also investigated in the present study (Langemark 1989). Our finding of similar ,%EP CSF levels in patients with CTH and in controls is suggestive of a normal activity in the central P-EP system. This is somewhat in contrast to the finding of Genazzani et al. (1984) of markedly decreased CSF concentrations in 6 patients suffering from migraine with interval headache. Patients with migraine without interval headache revealed higher P-EP concentrations although still lower than controls (Genazzani et al. 1984). Among the patients with CTH in the present study, we found a trend towards higher CSF @-EP levels in patients with occasional migraine (l-12 at- tacks/year) than in patients who never experienced migraine, but the difference was not significant. Co-ex- isting migraine and CTH is common. Patients with migraine followed by tension-type headache and pa- tients with tension-type headache followed by migraine showed remarkably small differences with regard to history and clinical findings (Langemark et al. 1988). Thus, migraine was not associated with low CSF P-EP levels in our study, but migraine with interval headache as studied by Genazzani et al. (1984) and CTH may possibly be 2 different entities of disease. Alternatively, methodological differences in the P-EP assays are re- sponsible for the divergent results with regard to P-EP in CSF. Notably, the CSF /3-EP concentrations in the

167

control group of the previous study (Genazzani et al. 1984) was strikingly higher than in the present study, and also higher than the control levels obtained by several other groups (for review, see Bach et al. 1992). A recent study found normal CSF P-EP concentrations in patients with fibromyalgia (Vaeroy et al. 1988).

It might be questioned whether CSF samples ob- tained by lumbar puncture are representative of activ- ity in /3-EP neurons localized in the brain and brain stem (Bach 1989). However, we recently showed that simultaneously obtained samples of ventricular and lumbar CSF revealed /I-EP concentrations that corre- lated significantly. Furthermore, the post-translational processing pattern of POMC seemed to be similar in ventricular and lumbar CSF pools, and we concluded that lumbar CSF samples probably are fairly good indicators of more rostra1 P-endorphinergic activity (Bach et al. 1992). Our finding of similar HPLC pro- files of /3-EP-ir in patients with CTH and controls (see Figs. 2 and 3) is suggestive of a normal POMC post- translational processing in the headache patients. Thus, there was no signs of decreased P-EP production or increased P-EP inactivation by C-terminal truncation or a,Wacetylation.

The lack of correlation of plasma and CSF /3-EP concentrations is in accordance with the literature (Facchinetti et al. 1983; Schlachter et al. 1983; Scheller et al. 1985; Myer et al. 1987), and our data adds support to the idea of 2 separately regulated /I-EP systems in the circulation and in the CNS.

Although previous studies suggested increased sen- sitivity to pain and maybe a decreased opioid hypotha- lamic tone in patients with CTH, we found in this study normal P-EP concentrations in plasma and CSF and evidence of a normal POMC post-translational pro- cessing in the CNS. Thus, either fl-EP is not involved in the pathogenesis of CTH, or such a role is not reflected in CSF or plasma concentrations of the neu- ropeptide.

Acknowledgements

The Danish Migraine Society supported this study. The HPLC was performed at the Department of Psy- chiatry and Neurochemistry, University of Lund, Lund, Sweden, under the supervision of Rolf Ekman, M.D., Ph.D.

References

Bach, F.W., Beta-endorphin-related peptides in human cere-

brospinal fluid. In: A.R. Genazzani and M. Negri (Eds.), Opioid

Peptides in Biological Fluids, Parthenon, Carnforth, 1989, pp.

17-25.

Bach, F.W., Jensen, K., Blegvad, N., Fenger, M., Jordal, R. and

Olesen, J., P-endorphin and ACTH in plasma during attacks of

common and classic migraine, Cephalalgia, 5 (198.5) 177-182.

Bach, F.W., Ekman, R. and Jensen, F.M., P-endorphin-immunoreac-

tive components in human cerebrospinal fluid, Regul. Peptides,

16 (1986) 189-198.

Bach, F.W., Fahrenkrug, J., Jensen, K., Dahlstrom, G. and Ekman,

R., Plasma P-endorphin during clinical and experimental is-

chaemic pain, Stand. J. Clin. Lab. Invest., 47 (1987) 751-758.

Bach, F.W., Schmidt, J.F. and Faber, T., Radioimmunoassay of

P-endorphin in human ventricular and lumbar cerebrospinal fluid,

Clin. Chem., 38 (1992) 847-852.

Baldi, E., Salmon, S., Anselmi, B., Spillantini, M.G., Capelli, G.,

Brocchi, A. and Sicuteri, F., Intermittent hypoendorphinaemia in

migraine attack, Cephalalgia, 2 (1982) 77-81.

Besson, J.-M. and Chaouch, A., Peripheral and spinal mechanisms of

nociception, Physiol. Rev., 67 (1987) 67-186.

Facchinetti, F., Nappi, G., Savoldi, F. and Genazzani, A.R., Primary

headaches: reduced circulating P-lipotropin and P-endorphin

levels with impaired reactivity to acupuncture, Cephalalgia, 1

(1981) 195-201.

Facchinetti, F., Petraglia, F., Nappi, G., Martignoni, E., Antoni, G.,

Parrini, D. and Genazzani, A.R., Different patterns of central

and peripheral P-EP, @-LPH and ACTH throughout life, Pep-

tides, 4 (1983) 469-474.

Facchinetti, F., Martignoni, E., Gallai, V., Micieli, G., Petraglia, F.,

Nappi, G. and Genazzani, A.R., Neuroendocrine evaluation of

central opiate activity in primary headache disorders, Pain, 34

(1988) 29-33.

Fettes, I., Gawel, M., Kuzniak, S. and Edmeads, J., Endorphin levels

in headache syndromes, Headache, 25 (1985) 37-39. ’ Genazzani, A.R., Nappi, G., Facchinetti, F., Micieli, G., Petraglia,

F., Bono, G., Monittola, C. and Savoldi, F., Progressive impair-

ment of CSF p-EP levels in migraine sufferers, Pain, 18 (1984)

127-133.

Hamba, M., Effects of lesion and stimulation of rat hypothalamic

arcuate nucleus on the pain system, Brain Res. Bull., 21 (1988)

757-763.

Hargreaves, K.M., Flores, C.M., Dionne, R.A. and Mueller, G.P.,

The role of pituitary P-endorphin in mediating corticotropin-re-

leasing factor-induced antinociception, Am. J. Physiol., 258 (1990)

E235-E242.

Headache Classification Commitee of the International Headache

Society, Classification and diagnostic criteria for headache disor-

ders, cranial neuralgias and facial pain, Cephalalgia, 8, Suppl. 7

(1988) l-96.

Jolkkonen, J.T., Soininen, H.S. and Riekkinen, P.J., P-endorphin-like

immunoreactivity in cerebrospinal fluid of patients with

Alzheimer’s disease and Parkinson’s disease, J. Neural. Sci., 77

(1987) 153-159.

Kerdelhue, B., Bethea, C.L., Ling, N., Chretien, M. and Weiner,

R.I., P-endorphin concentrations in serum, hypothalamus and

central gray of hypophysectomized and mediobasal hypothalamus

lesioned rats, Brain Res., 231 (1982) 85-91.

Laatikainen, T., Salminen, K., Stenman, U.-H. and Leppaluoto, J.,

Determination of plasma P-endorphin and P-lipotropin by

cation-exchange liquid chromatography and radioimmunoassay,

Clin. Chem., 31 (1985) 134-136.

Langemark, M., Naloxone in moderate doses does not aggravate

chronic tension headache, Pain, 39 (1989) 85-93.

Langemark, M., Olesen, J., Poulsen, D.L. and Beth, P., Clinical

characterization of patients with chronic tension headache,

Headache, 28 (1988) 590-596.

Langemark, M., Jensen, K., Jensen, T.S. and Olesen, J., Pressure

pain thresholds and thermal nociceptive thresholds in chronic

tension-type headache, Pain, 38 (1989) 203-210.

16X

Millan. M.J.. Multiple opioid systems and pain, Pain, 27 (1986)

303-347.

Millan, M.J.. Czlonkowski, A., Millan, M.H. and Hertz, A.. Activa-

tion of periaqueductal grey pools of P-endorphin by analgetic

electrical stimulation in freely moving rats, Brain Res., 407 (1987)

199-203.

Millington, W.R. and Smith, D.L., The posttranslational processing

of /3-endorphin in human hypothalamus, J. Neurochem., 57t1991)

775-781.

Myer. E.C., Morris. D.L.. Adams. M.L.. Brase. D.A. and Dewey.

W.L., Increased cerebrospinal fluid p-endorphin immunoreactiv-

ity in infants with apnea and in siblings of victims of sudden

infant death syndrome, J. Pediat., 1 I1 (1987) 660-666.

Neuser. D.. Lesch. K.P., Stasch, J.P. and Przuntek, H., Beta-en-

dorphin-. leucine enkephalin- and methionine-enkephalin-like

immunoreactivity in human cerebrospinal fluid. Simultaneous

determination and relation to neurological disorders, Eur. Neu-

rol.. 23 (1984) 73-81.

Glesen. J., Clinical and pathophysiological observations in migraine

and tension-type headache explained by integration of vascular.

supraspinal and myofascial inputs, Pain, 46 (1991) 125-132.

Rasmussen. B.K.. Jensen, R. and Olesen, J., A population-based

analysis of the diagnostic criteria of the International Headache

Society, Cephalalgia, 11 (1991) 129-132.

Scheller, M.S., Risch. S.C. and Turken, A., The relationship of

P-endorphin and ACTH in the cerebrospinal fluid and plasma of

adult surgical patients, Ann. Clin. Res., 17 (1985) 323-326.

Schlachter, L.B., Wardlaw, S.L., Tindall, G.T. and Frantz, A.G..

Persistence of P-endorphin in human cerebrospinal fluid after

hypophysectomy, J. Clin. Endocrinol. Metab., 57 (1983) 221-224.

Schoenen. J., Bottin, D., Hardy, F. and Gerard, P., Cephalic and

extracephalic pressure pain thresholds in chronic tension-type

headache, Pain, 47 (1991) 145-149.

Sicuteri, F., Natural opioids in migraine. In: M. Critchley, A. Fried-

man, S. Gorini and F. Sicuteri (Eds.), Advances in Neurology,

Vol. 33, Headache: Physiopathological and Clinical Concepts,

Raven, New York, 1982, pp. 65-74.

Smith, A.I. and Funder, J.W., Pro-opiomelanocortin processing in

the pituitary, central nervous system, and peripheral tissues.

Endocrinol. Rev., 9 (1988) 159-179.

Vaeroy, H., Helle, R.. Forre. 0.. K&s, E. and Terenius, L., Cere-

hrospinal fluid levels of P-endorphin in patients with fibromyal-

gia (fibrositis syndrome), J. Rheumatol., 15 (1988) 1804-1806.

Wood, J.H., Neuroendocrinology of cerebrospinal fluid: peptides,

steroids. and other hormones, Neurosurgery, 11 (1982) 293-305.

Zakarian, S. and Smyth, D.G., Distribution of p-endorphin-related

peptides in rat pituitary and brain, Biochem. J.. 202 (1982)

561-571.