Neuropeptides in the cerebrospinal fluid (CSF) in psychiatric disorders

Pros. Neuro-Psychopharmocol. 6 Biol. Psychiat. Printed in Great Britain. All rights reserved.

1965, Vol. 9, pp. 533-536 Copyright

NEUROPEPTIDES IN THE CEREBROSPINAL IN PSYCHIATRIC DISORDERS

OLE J. RAFAELSEN

0

0276-5646/65 $0.00 + SO 1985 Pergamon Press Ltd.

FLUID (CSF)

and ANNETTE GJERRIS

Department of Psychiatry, Rigshospitalet, Copenhagen, Denmark Abstract Rafaelsen, Ole J. and Annette Gjerris: Neuropeptides in the cerebrospinal fluid (CSF) in psychiatric disorders. Prog. Neuro-Psychopharmacol. & Biol. Psychiatr. 1985 I 2 (5/6):533-538. Concentrations of vasoactive intestinal polypeptide (VIP), cholecystokinin (CCK) and gastrin in the cerebrospinal fluid (CSF) was studied in patients with endogenous depression , non-endogenous depression, mania, schizophrenia and a control group. All patients were classified according to various diagnostic systems. In the group of non-endogenously depressed patients CSF-VIP levels (median 16 pmol/l) were found significantly lowered compared to controls (median = 32 pmol/l) and endogenous depression (26 pmol/l). Going through the non-endogenous group it appeared that the low CSF-VIP was due to a group of patients with a former diagnosis of endogenous depression or a present diagnosis of possible endogenous depression. Moreover, this group was clinically characterized by 'dysphoric/hysterical features', 'reversed diurnal variation* (i.e. worst in the evening), and 'lack of clearly circumscribed episode*. In many aspects this group seems similar to the atypical depressions described as monoamineoxidase responders. Concerning CSF-CCK and CSF-gastrin no significant differences between the examined groups were demonstrated. Keywords: cerebraspinal fluid, cholecystokinin, neuropeptides, psychiatry, vasoactlve intestinal polypeptide Abbreviations: Beth-Rafaelsen Melancholia Scale (BRMES), Brief Psychiatric Rating Scale (BPRS) cerebrospinal fluid (CSF) cholecystokinin (CCK), Newcastle Rating Scale for Depression (N-I), thyrotropin (TRH), vasoactive intestinal polypeptide (VIP), vasopressin (AVP) Introduction Many neuropeptides have been found in brain areas considered important in the pathogenesis of psychoses. Moreover, the coexistence with 'classical neurotransmitters' anticipated to be involved in psychoses (Hafelt et al. 1980a, Krieger 1983), underlines the importance of peptides in psychiatric research. Recently, there has been particular interest in the possible involvement of CCK in schizophrenia (HSkfelt et al. 1980b) and AVP, TRH and VIP in depression (Gold et al. 1978, 1979, 1981, Kirkegaard et al. 1979, Gjerris et al. 1984a,b). In a former study (Gjerris et al. 1984a) we found no differences in the concentrations of CCK in CSF in schizophrenic and depressed patients. In that study we used radioimmunoassays developed several years ago (Rehfeld 1978a,b, Rehfeld and Kruse-Larsen 1978). Lately, we have, however, obtained CCK-antisera that bind both larger molecular forms as well as smaller C-terminal fragments with significantly higher potency that the previous antisera. Since CCK is highly heterogenous also in CSF, it might be useful to apply the new antisera to &F-studies. In the present study we have measured CCK with the antisera G160 and 278, in CSF from patients with various psychiatric disorders. Furthermore we give a summary of our previous findings of CSF-VIP, TRH and AVP. 533

534

0. J. Rafaelsenand A. Gjerris Material and methods

76 psychiatric patients and controls admitted to the Department of Psychiatry, Rigshospitalet, Copenhagen, entered the study. For 62 of these, proper allocation to relevant diagnostic groups and a complete set of analysis were obtained. None of the patients had symptoms or signs indicating central nervous system lesions or endocrine disorders. Diagnostic classification and severity of illness. Using the ninth Edition of the International Classification of Diseases (ICD-9) patients were classified into schizophrenia (295), endogenous depression (= melancholia) (296.1), reactive depression (298.0), neurotic depression (300.4), affective personality disorder (301.1). Moreover the depressed patients were subdivided into endogenous and non-endogenous groups according to the N-I (Carney 1965). The ICD-9 classification was performed at the time of discharge from the hospital by a psychiatrist who did not otherwise take part in the study. The N-I classification was applied on the day of the lumbar puncture by the psychiatrist who performed the puncture. The rater is a member of the rating group of the department for whom satisfactory inter-rater reliability has been obtained in several studies (Beth et al. 1983, Gjerris et al. 1983, 1984c). To measure severity of depression, the BRMES, a modification of the Hamilton Depression Scale , was used (Beth and Rafaelsen 1980). The lower entrance limit to the study was a total score >15. Severity of schizophrenia was measured by the BPRS (Overall and Gorham 1962). There was here no lower limit for entering the study. All control patients were rated according to BRMES in order to avoid inclusion of patients with a depressive symptomatology. The upper limit of total score on BRMES for the control group was 5. Sampling and methods. CSF- and blood samples were taken between 8 and 9 a.m. All patients were fasting overnight and stayed in bed 12 hours prior to the lumbar puncture. The patients stayed in their beds during the puncture, which was performed with the patients in the lateral recumbent position. The CSF-pressure was measured before removal of CSF. All patients had normal CSF-pressure, protein concentration, and cell count. Venous blood samples were obtained just before the lumbar puncture. All CSF samples were collected in fractions so that the same analysis was always performed on the same sample. Informed consent satisfying the requirements of the Helsinki Declaration was obtained from all patients. Analytical methods. CSF for analysis was sampled in chilled tubes and placed on ice. The radioimmunoassay for measuring CSF-CCK was performed as previously described (Rehfeld & Kruse-Larsen 1978). The new antisera used in this study have the following characteristics: Antiserum 278 was raised by us in rabbit No. 278 against non-sulfated CCK-8 covalently coupled to bovine serum albumin. The antiserum is highly specific for the C-terminal tetrapeptide amine, i.e. it measures CCK-4 and -5 with almost the same molar potency as CCK-8. Hence, this antiserum detects small bioactive C-terminal fragments of CCK. Antiserum G160 (a generous gift from A. Schafmayer, GBttingen)was raised in rabbit G160 against sulfated CCK-33 covalentlv coupled to bovine serum albumin. This antiserum binds the sulfated N-terminal Sequence of CCK-8. Hence, this antiserum binds all the larger molecular forms of CCK. Statistics. Non-parametric statistic was used thoughout the study (Siegel 1956). Central tendency and dispersion was expressed bv medians and range. When-comparing more than two groups of patients the Kruskall-Wallis test was applied to test a posssible significant difference, and when only two groups were compared the Mann-Whitney test or Wilcoxon paired ranksum test were used. Correlations were tested by means of Spearman correlation coefficient.

Neuropeptides in cerebrospinal fluid

535

Results Finally 62 of the total of 76 patients, originally included in the study, could be classified according to ICD-9. The groups comprised 26 patients with endogenous depression, 14 with non-endogenous depression, 7 schizophrenic patients and 15 controls. Diagnostic groups, number of patients, age, sex and number of patients who.had a repuncture after clinical recovery are seen in Table 1. Table 1 CCK-study. Diagnostic categories, number of patients, who had a complete set of analysis, age, sex and number of patients who had a repuncture after clinical recovery.

ICD-9

Endogenous

depression

Non-endogenous depression Schizophrenia Controls

Age median (range)

Sex female/ male

Repuncture

26

51(36-69)

13/13

5

14

51(24-69)

7/7

3

7

35(28-41)

Q/4

15

46(24-62)

8/7

For both measurements of CSF-CCK a significant correlation to age was demonstrated for the total group of patients. R, for CCK measured with antiserum G160 (n = 73) was 0.25, P < 0.05, and for CCK measured with antiserum 278 (n = 76) R, was 0.31, P 6 0.01. Furthermore a significant correlation was demonstrated between CCK (Ab. G160) and CCK (Ab. 278) (n = 73), R, being 0.27, P 5 0.05. Using the Kruskall-Wallis test a comparison of CSF-CCK (Ab. G160) and CSF-CCK (Ab. 278) between the different diagnostic groups classified according to ICD-9 did not reach significant levels (Table 2). Neither did a comparison between endogenous and non-endogenous groups classified according to the N-I. The Wilcoxon paired test was used to compare the two set of CSF-CCK measurements in eight patients studied during illness and after clinical recovery. No differences between first and second puncture could be demonstrated. Former results on CSF-VIP, -TRH and -AVP. In a previous study CSF-VIP was found significantly lowered in non-endogenous depression characterized by dysphoric/hysteric features, reversed diurnal variation, and lack of clearly circumscribed episodes (Table 3). CSF-VIP stayed low also after clinical recovery (Gjerris et al. 1984a). Concerning CSF-TRH no differences were found between the groups. CSF-AVP was found significanly lowered in a well endogenous as non-endogenous depression (Table 3) (Gjerris et al. 1984b).

536

0.

J. Kafaelsen

and

A. Gjerris

Table 2 CSF-CCK

(Ab. 6160) and CCK (Ab. 278) medians (range) in different diagnostic groups

ICD-9

Ab. G160 pm01 eqv. CCK-8/l

Ab. 278 pm01 eqv. CCK-8/l

Endogenous depression

7.0 (5.3-10.9) (n=26)

4.4 (1.8-8.8) (n=26)

Non-endogenous depression

7.8 (6.1-9.2) (n=l4)

4.8 (2.2-7.0) (n=14)

Schizophrenia

6.6 (6.1-8.3) (n=7)

4.4 (3.1-6.6) (n=7)

Controls

6.8 (0.4-10.9) (n=lS)

3.9 (0.9-5.7) (n=lS)

Ab. G160 recognizes large, sulfated molecular forms of CCK. Ab. 278 recognizes small C-terminal fragments of CCK Table 3 CSF-VIP, TRH and AVP medians (range) in different diagnostic groups ICD-9

VIP pmol/l

TRH rig/l

AVP rig/l

Endogenous depression

36 (5-67) (n=28)

6.9 (2.7-15.2) (n=21)

1.1 (0.9-2.2)X (n=32)

Non-endogenous depression

16 (1-42)xxz (n=24)

5.7 (2.4-10.0) (n=13)

1.2 (0.8-1.8)X (n=14)

Mania

45 (25-59) (n=E)

5.3 (2.3-6.9) (n=E)

2.0 (0.9-2.8) (n-7)

Schizophrenia

30 (9-58) (n=14)

8.6 (4.8-11.6) (n=9)

1.6 (1.0-2.1) (n=9)

Controls

32 (9-67) (n=17)

4.2 (3.0-12.4) (n=16)

1.8 (0.9-2.4) (n=lO)

P = 0.05 xl xxx) P = 0.001 Discussion The main purpose of this paper has been to present the results when CCK in CSF was measured by use of two new and different antisera in different diagnostic groups of psychiatric patients. Secondarily to give a summary of our previous data on peptides in CSF. Based on the occurrence of CCK in mesolimbit dopamine neurons and the inhibitory effect of CCK on dopamine (Hakfelt et al. 1980, Widerlbv et al. 1983), the CCK function has been hypothesized to

Neuropeptides

in cerebrospinal

fluid

537

be diminished in schizophrenia. Moreover, it has been claimed, but still not proven, that CCK-like peptides have a long-lasting antipsychotic effect on schizophrenia patients (Ito et al. 1982, Moroji et al. 1982). In a former study of CSF-CCK concentrations in psychiatric patients using an older CCK antiserum (Gjerris et al. 1984a) we found in accordance with Gerner and Yamada (1982) no differences in CSF-CCK between the diagnostic groups investigated. The present study indicates that even when a more specific analytical methodology is applied, a possible imbalance between CCK and dopamine in schizophrenic or in any of the other diagnostic groups investigated is not reflected in CSF-concentrations. In summary the results of our present and previous studies on CSF-peptides in psychiatric disorders seem to show that both VIP and AVP are affected in depression, whereas we, finding no changes in CSF-TRH in depressed patients, have not beeen able to confirm the results of Kirkegaard et al. (1979). As to our results of two studies on CSF-CCK supported by the results of Gerner and Yamada (1982) it seems established that a possible impaired CCK-function in psychiatric disorders is not reflected in CSF-concentrations. The finding of low CSF-VIP in non-endogenous depression is at first glance surprising, but a further examination of the course of disease, symptoms and signs of these patients seems to bring this group of patients into the group of atypical depressions described by Davidson (1982), which apparently benefit from treatment with MAO-inhibitors and thus probably possess biological dysfunctions. Also the phase-independency of CSF-VIP fits well with the clinical feature of 'lack of clearly circumscribed episodes' characteristic for this group of patients. References BECH, P and RAFAELSEN, O.J. (1980) The use of rating scales exemplified by a comparison of the Hamilton and the Beth-Rafaelsen Melancholia Scale. Acta Psychiat. Stand. 62, Suppl. 285: 128-132. BECH, P., GJERRIS, A., ANDERSEN, J., B0JHOLM, S. KRAMP, P., BOLWIG T.G., KASTRUP, M., CLEMMESEN, L. and RAFAELSEN O.J. (1983) The Melancholia Scale and the Newcastle Scales: Item combination and interobserver-reliability. Br. J. Psychiat. 143: 58-63. CARNEY, M.W.P., ROTTM. and GARSIDE, R.F. (1965) The diagnosis of depressive syndromes and prediction of ECT response. Br. J. Psychiat. 111: 659-674. DAVIDSON, J.R.T., MILLER, R.D., TURNBULL, C.D. and SULLIVAN, G. (1982) Atypical depression. Arch. Gen. Psychiat. 39: 527-534. GERNER, R.H. and YAMADA, T. (1982) Altered neuropeptide concentrations in cerebrospinal fluid of psychiatric patients. Brain Res. 238: 298-302. GJERRIS, A., BECH, P., BBJHOLM, S., BOLWIG, T.G., KRAMP, P.,CLEMMESEN, L., ANDERSEN, J., JENSEN, E. and RAFAELSEN, O.J. (1983) The Hamilton Anxiety Scale. J. Affective Dis. 2: 163-170. GJERRIS, A., RAFAELSEN, O.J., VENDSBORG, P., FAHRENKRUG, J. and REHFELD, J.F. (1984a) Vasoactive intestinal polypeptide decreased in cerebrospinal fluid (CSF) in atypical depression. J. Affective Dis. 1: 325-337. GJERRIS A, RAFAELSEN O.J., VENDSBORG, P., HAMMER, M., REHFELD, J.F., KIRKEGAARD C. and FAHRENKRUG, J. (1984B) Brain peptides in psychiatric disorders. Clin. Neuropharm. 2, Suppl 1: 298-299. GJERRIS, A. BECH, P., ANDERSEN, J., BBJHOLM, S., KRAMP, P., BOLWIG, T.G., KASTRUP, M., CLEMMESEN, L. and RAFAELSEN, O.J. (1984c) The Melancholia Scale, the Newcastle Scales, and the Eight International Classification of Disease: Interobserver reliability and diagnostic validity. In: Biological Psychiatry: Recent Studies, G.D. Burrows (ed), pp 121-125. John Libbey & Company Limited, London. GOLD, P.W., GOODWIN, P.K. and REUS, V.I. (1978) Vasopressin in affective illness. Lancet i: 1233-1236. GOLD, P.W., WEINGAiiTNER, H., BALLENGER, J.C., GOODWIN, F.K. and POST, R.M. (1979) EffeCtS of 1-desamino-8-D-arginine vasopressin on behaviour and cognition in primary affective disorder. Lancet -ii: 992-994.

538

0.

J.

Rafaelsenand A. Gjerris

GOLD, P.W., GOODWIN, F.K., POST, R.M. and ROBERTSON, G.L. (1981) Neuroendo_crine studies of depressive disorders. Psychopharmacol. Bull. 17: 7-9. HOKFELT, T., JOHANSSON, O., LJUNGDAHL, A., LUNDBERG, J.M. and SCELTZBERG, _M. (1980a) Peptidergic neurones. Nature 284: 515-521. HOKFELT, T., REHFELD, J-F., SKIRBOLL, L., IVERMARK, B., GOLDSTEIN, M. and MARKEY, K. (1980b) Evidence of coexistence of dopamine and CCK in mesolimbic neurones. Nature 285: 476-478. ITOH, H., TANOUE, S., YAGI,., TATEYAMA, M., KAMISADA, M., FUJII, Y., TAKAMIYA, M., and NAKAJIMA, S. (1982) Clinical study on the psychotropic effects of caerulein - and open clinical trial in chronic schizophrenic patients. Keio J. Med. 31: 71-95. KIRKEGAARD, C., FABER, J.,HUMMER, L. and RAGOWSKI, P. (1979) Increased levels of TRH in cerebrospinal fluid from patients with endogenous depression. Psychoneuroendocrinology A: 227-235. KRIEGER, D.T. (1983) Brain peptides: What, where, and why? Science 222: 975-985. MAROJI, T., WATANABEE, N., AOKI, N. and ITOH, S. (1982) Antipsychotic effects of ceruletide (caerulein) on chronic schizophrenia. Arch. Gen. Psychiat. 39: 485-486. OVEALL, J.E. and GORHAM, D.R. (1962) The brief psychiatric rating scale. Psychol. Rep. 10: 799-812. REHFELD, J.F. (lv8a) Immunochemical studies on cholecystokinin. I. Development of sequence-specific radioimmunoassays for porcine triacontatriapeptide cholecystokinin. J. Biol. Chem. 253: 4016-4021. REHFELD, J.F. (1978b) Immunochemical studies onholecystokinin. II. Distribution and molecular heterogeneity in the central nervous system and small intestine of man and hog. J. Biol. Chem. 253: 4022-4030. REHFELD, J.F. and KRUSE-LARSEN, C. (1978c) Gastrinnd cholecystokinin in human cerebrospinal fluid. Immunochemical determination of concentrations and molecular heterogeneity. Brain Res. 155: 19-26. SIEGEL, S. (1956) Non-parametric Statistics.McGraw Hill, New York. WEINGARTNER, H., GOLD, P., BALLENGER, J-C., SMALLBERG, S.A., SUMMERS, R., RUBENOW, D.R., POST, R.M., GOODWIN, F.K. (1981) Effects of vasopressin on humap memory functions. Science 211: 601-603. WIDERLOV, E., KALIVAS, P.W., LEWIS--H., PRANGE Jr, A.J. and BREESE G.R. (1983) Influence of cholecystokinin on central monoaminergic pathways. Regulatory Peptides 5: 99-109. WORLD HEALTH ORGANIZATION MENTAL DISORDERS (1978) Glossary and guide to their classification in accordance with the ninth revision of the International Classification of Disease. WHO, Geneva. Inquiriesand reprintrequestsshouldbe addressedto: Dr. Ole J. Rafaelsen Departmentof Psychiatry Rigshospitalet, Copenhagen Denmark