Melatonin immunoreactivity in cerebrospinal fluid of schizophrenic patients and healthy controls

Psychiarry Research, 11, 107-I 10 Elsevier 107 Melatonin lmmunoreactivity in Cerebrospinal Fluid of Schizophrenic Patients and Healthy Controls Helmut Beckmann, Lennart Wetterberg, and Wagner F. Gattaz Received May 27, 1983; revised version received September 12, 1983; accepted October 30, 1983. Abstract. Melatonin is produced in the pineal gland. Its involvement in various psychiatric and somatic diseases has been suggested. We investigated melatonin in cerebrospinal fluid of 16 healthy controls, 15 paranoid schizophrenics being treated with neuroleptics, and 13 unmedicated paranoid schizophrenics. There were no significant differences in melatonin concentrations among these three groups. No significant correlations were found between melatonin concentrations and various other biochemical substances such as noradrenalin, cyclic adenosine 3’, 5’-monophosphate, prolactin, and cortisol. These negative results do not support the suggestion that melatonin is involved in the etiology of schizophrenia. However, other possibilities, e.g., a change of biological rhythms and its influence on other neuroendocrine functions, may be of importance. Key Words. Melatonin, cerebrospinal fluid, schizophrenic patients, healthy controls. The pineal gland has the structural and secretory features of an endocrine organ (Kappers, 1978), and mediates certain environmental factors which regulate the growth and function of reproductive organs (Reiter, 1978). Melatonin is one of several compounds that are produced in the pineal gland. It has been supposed that this substance might regulate endogenous rhythms, e.g., sleep-wake cycles and body temperature, and that its dysfunction could be associated with mental disorders (Wetterberg, 1978). Hanssen et al. (1980) reported a normal diurnal pattern with low 8 a.m. plasma melatonin and high midnight values in schizophrenic patients. Those on high doses of the /3-adrenergic receptor-blocking drug propranolol all showed a decreased output of melatonin in the urine as well as a disappearance of the nocturnal rise in plasma melatonin. In unmedicated depressed inpatients no evidence was found of a difference in basal melatonin levels in serum as compared to normal controls. In addition, melatonin levels in plasma were not altered by electroconvulsive therapy in patients with affective disorders (Wetterberg, 1978). The present study was designed to investigate melatonin levels in schizophrenic patients. We chose to analyze cerebrospinal fluid (CSF) on the grounds that it might better Helmut Beckmann. M.D., is Associate Professor, Central Institute of Mental Health, 6800 Mannheim, J 5, FRG. Lennart Wetterberg. M.D., is Professor, Karolinska Institute. Department of Psychiatry, St. GBrans Hospital. Stockholm, Sweden. Wagner F. Gattar. M.D., Department of Psychiatry, Faculdade de Medi- cina da Fundacao do ABC. Sao Paulo, Brazil. is presently at the Central Institute of Mental Health, 6800 Mannheim, J 5, FRG. (Reprint requests to Dr. fi. Beckmann.) 0165-1781:83~$03.00 @ 1984 Elsevier Science Publishers B.V 108 reflect the cerebral melatonin turnover than urinary output or serum levels. Methods Patients. Twenty-eight paranoid schizophrenic patients (all males, mean age 40.6 f SD 8.0 years) were included in this study. Patients were diagnosed according to the Research Diagnos- tic Criteria (Spitzer et al., 1978). Two experienced psychiatrists independently evaluated the patients’ psychopathological state by means of the Brief Psychiatric Rating Scale (BPRS) (Overall and Gorham, 1962). Fifteen patients had been treated with neuroleptic drugs (butyro- phenones and phenothiazines) for at least 3 weeks (mean dose * SD in chlorpromazine equivalents = 585 * 755 mg/ day). Thirteen patients had not taken any drugs for at least 4 weeks before the study. Controls were subjects with unspecific neurological symptomatology who required a lumbar puncture for diagnostic reasons. Controls were not receiving drug treatment at the time of lumbar puncture. Lumbar Puncture. CSF was obtained by lumbar puncture (LP) with the subjects in a sitting position between 9 and 10 a.m. Probands had fasted for 12 hours and been at bed rest for 10 hours before the LP was performed. Sixteen ml of CSF were removed without additions. To avoid rostra]-caudal gradient effects, samples were gently mixed and then immediately frozen on dry ice and stored in a freezer at -50°C. They were then shipped to the Karolinska Institute, Department of Psychiatry, where the melatonin assay was performed. Melatonin Assay. Melatonin was analyzed in the CSF using the method of Wetterberg et al. (1978) to measure melatonin (or melatonin-like compounds). The cross-reactions of the anti- melatonin serum to 13 chemically related substances have been reported by Arendt et al. (1977). The method has previously been compared to other radioimmunoassay (RIA) methods for melatonin (Wetterberg and Eriksson, 1981). The lower limit of detection is 0.01 nmole/l (3 pg/ ml) and the percentage variation (CVYc) is 6.8 (intraassay variation 7.4Ye and interassay variation 4.8% for human serum samples). Standard melatonin has been analyzed using gas chromatography/ mass spectrometry for comparison with our RIA and the results were identi- cal. All melatonin values in the present study were low, which indicates that there was a total or almost total absence of interfering melatonin-reacting substances besides melatonin in the CSF. Statistical Analysis. One-way analysis of variance, the Mann-Whitney U test, and Spear- man’s correlation coefficient were used for statistical analysis. Results As can be seen from Table 1, there were no significant differences in the mean values of the CSF concentrations of melatonin among healthy controls, patients receiving neuroleptics, and unmedicated patients. Application of one-way analysis of variance showed no heterogeneity among the three subgroups @ = 0.40). There were no correlations between the catecholamines dopamine and noradre- nalin, their metabolites homovanillic acid and 3-methoxy-4-hydroxyphenylglycol, cyclic guanosine monophosphate (GMP), cyclic adenosine 3’, 5’-monophosphate (AMP), the electrolytes, and CSF melatonin concentrations except for magnesium in controls (rs = 72, p < 0.00 I) and in unmedicated patients (rs q 64, p < 0.0 1) (Gattaz et al., 1982; Gattaz et al., 1983). There was some correlation between CSF melatonin and the dose of neuroleptics (in chlorpromazine equivalents) for the day the LP was performed (rs = 47, p < 0.05). The total BPRS score was correlated with the CSF concentration of melatonin (rs = 34. p < 0.05). 109 Table 1. Mean (k SD) concentrations of melatonin in CSF Patients with Patients without Melatonin in CSF Controls (n = 16) neuroleptics (n = 15) neuroleptics (n = 13) nmole/l 0.064 ?Z 0.108 0.066 + 0.071 0.065 * 0.060 pg/mt 15 + 25 15 2 16 15 * 14 Discussion The levels of melatonin in CSF found in this study are in the range of those reported by Fyro and Wetterberg (unpublished data). Melatonin concentrations in CSF of schi- zophrenic patients had not been measured previously. Our finding that there were no significant differences between healthy controls and paranoid schizophrenic patients, with and without neuroleptics, is somewhat in accordance with the findings of Hanssen et al. (1980) who reported equal mean resting morning levels of melatonin in serum for three schizophrenic females and their six nonschizophrenic siblings. As there were no differences between medicated and unmedicated schizophrenics, a major influence of neuroleptics on melatonin production in this sample may be excluded. In the serum high levels of melatonin have been reported in psychiatric patients treated with chlorpromazine but not in schizophrenic patients following fluphenazine or flupenthixol (Smith et al., 1979). Several studies indicate that diurnal as well as annual rhythms are of importance for the activity of the melatonin-forming enzyme as well as for melatonin (Weitzman et al., 1978). We tried to avoid diurnal variations by performing the LPs at the same time of the day. Melatonin formation is under noradrenergic control, and we therefore investigated its possible correlations with CSF noradrenalin and cyclic AMP. However, no con- sistent association could be detected. The only correlation found, with the electrolyte magnesium, must not be overemphasized as many correlations with various parame- ters have been calculated and this was the only significant one. Similarly, the negative correlation between the global psychopathological score on the BPRS might be a chance finding. The significance of melatonin metabolism for psychiatric disorders in general or for schizophrenic psychoses is not ruled out by these largely negative results. It is unknown whether the diurnal rhythms in schizophrenic patients are disturbed in comparison to those described in healthy controls by Smith et al. (1981) and Greiner and Chan (1978). The occurrence of disturbed circadian rhythms and their possible influence on various biochemical parameters should be further investigated. References Arendt, J., Wetterberg, L., Heyden, T., Sizonenko, P.C., and Paunier, L. Radioimmunoas- say of melatonin: Human serum and cerebrospinal fluid. Hormone Research, 8,65 (1977). Gattaz, W.F., Kattermann, R., Gattaz, D., and Beckmann, H. Magnesium and calcium in the CSF of schizophrenic patients. Biological Psychiarry, 18,935 (1983). Gattaz, W.F., Waldmeier, P., and Beckmann, H. CSF monoamine metabolite in schizo- phrenic patients. Acta Psychiarrica Scandinavica, 66, 350 (1982). 110 Greiner, A.C., and Chart, S.C. Melatonin content of the human pineal gland. Science, 199,83 (1978). Hanssen, T., Heyden, T., Sundberg, J., Alfredsson, G., Nybdck, H., and Wetterberg, L. Propranolol in schizophrenia: Clinical, metabolic and pharmacological findings. Archives of General Psychiarry, 37,685 ( 1980). Kappers, J.A. Localization of indoleamine and protein synthesis in the mammalian pineal gland. Journal of Neural Transmission, 13, I3 (I 978). Overall, J.E.. and Gorham, D. R. The Brief Psychiatric Rating Scale. P.~ychological Reports, 10, 799 (1962). Reiter, R.J. Evidence for an endocrine function of the human pineal gland. Journal ofNeural Transmission, 13, 247 (1978). Smith, J.A., Barnes, J.L.. and Mee, T.J.M. The effect of neuroleptic drugs on serum and cerebrospinal fluid melatonin concentration in psychiatric subjects. Journal of Pharmac.v and Pharmacology, 31,246 (1979). Smith, J.A., Mee, T.J.M., and Padwick, D.J. Human post-mortem pineal enzyme activity. Clinical Endocrinology, 14, 75 (198 1). Spitzer, R.L., Endicott, J., and Robins, E. Research Diagnostic Criteria: Rationale and reliability. Archives of General Psychiatry, 35, 773 (1978). Weitzman, E.D., Weinsberg, U., D’Eletto, R.. Lynch, H., Wartmann, R.J., Czeisler, C., and Erlich, S. Studies of the 24 hour rhythm of melatonin in man. Journalof Neural Transmission, 13, 325 (1978). Wetterberg, L. Melatonin in humans: Physiological and clinical studies. Journal of Neural Transmission, 13, 289 (1978). Wetterberg, L., and Eriksson, 0. Melatonin in human serum-A collaborative study of current radioimmunoassays. In: Birau, N.. and Schloot, W., eds. Melutonin-Current Snafus and Perspeclives. Pergamon Press, Oxford. p. 15 (1981). Wetterberg, L., Eriksson, O., Friberg, Y.. and Vangbo, B. A simplified radioimmunoassay for melatonin and its application to biological fluids: Preliminary observations on the half-life of plasma melatonin in man. Clinica Chimica Acta, 86, 169 (1978).