Vasopressin-oxytocin in cerebrospinal fluid of schizophrenic patients and normal controls

Ps),choneuroendocrmoloRy, Vo| I0.No 2,pp 18?- 191,|985

0306- 4530/85$3 OO + 0.00 © 1985Pergamon PressLid

Pnntcd m GreatBrltam

V A S O P R E S S I N - O X Y T O C I N IN C E R E B R O S P I N A L FLUID O F S C H I Z O P H R E N I C PATIENTS A N D N O R M A L C O N T R O L S

HELMUT BECKMANN,* RUDOLF E. LANGt and WAGNER F. GATTAZ* *Central Institute of Mental Health, Mannheim, Federal Republic of Germany. )'Institute of Pharmacology of the University Heidelberg, Federal Republic of Germany. (Received 18 November 1983; in final form 2 August 1984)

SUMMARY Vasopressin and oxytocin seem to be involved in the processes of learning and memory in animals and probably in man. These peptides appear to have opposite effects in that vasopressin improves m e m o r y processes and oxytocin produces amnestic effects. W e measured these neuropeptides in the cerebrospmal fluid of schizophrenic patients with and without neuroleptic treatment, psychiatrically healthy controls and drug-free patients before and after three weeks' neuroleptic treatment. There were no significant differences in vasopressin concentrations between schizophrenics and controls. N o influence of neuroleptic treatment on vasopressin concentrations was detected. In contrast, concentrations of oxytocin were increased in all schizophrenic patients and were higher in those receiving neurolepic treatment. In addition, oxytocin concentrations increased after three weeks' neuroleptic treatment. Drug-induced increase of oxytocin concentrations m a y be of significance in the clinically observed amnestic syndromes and debilitation in schizophrenics treated with neuroleptics.

Key Words~Vasopressin; oxytocin; cerebrospinal fluid; neuroleptics; schizophrenic patients. INTRODUCTION THE

NEUROHYPOPHYSEAL

peptides vasopressin (VP) and oxytocin (OT) have been

demonstrated to modulate several types of behavior. V P has been shown to delay the extinction of active avoidance behavior and of behavior which is positivelyreinforced. It improves passive avoidance behavior in the laboratory rat (Ader & de Wied, 19"/2;de Wied, 19"/I;Walter et of., 1978). Therefore, it has been suggested that this peptide may facilitatememory consolidation (de Wied, 19-/I)and possibly the retrievalprocess as well (Bohus et al., 19-/8). These effects on memory also have been reported for humans (Legros et of., 19-/8;Oliveros el al., 19"/8). Conversely, O T has been shown to facilitatethe extinction of the active avoidance response (Bohus e/of., 19"/8)and to impair passive avoidance behavior (Thompson & de Wied, 19"/3).Thus, ithas been suggested that V P improves memory function, whereas O T has amnestic properties. V P appears to act as a C N S neuromodulator, influencing catecholaminergic nerve impulse flow (Versteeg et al., 19"/9; Tanaka et al., 19"/7a). Both noradrenaline and dopamine turnover are enhanced in various brain regions after V P treatment (Kovacs et al., 19"/"/;Tanaka el al., 1977b; Versteeg et al., 19-/9). The hypothesis has been put forward that VP's modulating influences on memory processes depend on the alteration of catecholaminergic neurotransmission (Versteeg et of., 19"/9).Indeed Telegdy & Kovacs 187

188

HELMUT BECKMANN,RUDOLFE, LANG and WAGNERF. GATTAZ

(1979) have reported that OT changed catecholaminergic neurotransmission in some brain regions in a direction opposite to that induced by VP. We have measured VP and OT in cerebrospinal fluid (CSF) of schizophrenic patients with and without neuroleptics and psychiatrically healthy controls. We report similar VP concentrations in all three groups but increased OT levels in patients treated with neuroleptics. PATIENTS AND METHODS This study is one part of a larger biological investigation on this sample which has already been described elsewhere (Gattaz et al., 1982; Beckmann et aL, 1982). It includes 28 consecutively admitted paranoid schizophrenic patients (all male, Caucasian, mean age ± SD = 30.6 ± 8.0 yr) and 15 controls (13 male and two female, mean age ± SD = 35.0 ± 15.7 yr). Since the results obtained from the two women were very similar to those of the men, the data were considered together. Patients were diagnosed according to the Research Diagnostic Cnterm (Sp~tzer et al., 1978). Two experienced psychiatrists evaluated independently tbelr psychopathological state by means of the Brief Psychiatric Rating Scale (BPRS, Overall & Gotham, 1962). Fifteen patients were under treatment with neuroleptic drugs (butyrophenones and phenothiazines) for at least three weeks (mean doses ± S D in chlorpromazine equivalents = 585 ± 755 rag/day). Thirteen patients did not take any drug for a period of at least four weeks prior to the study. The controls were subjects with nonspecific neurological symptoms (headaches, dizziness, etc.) who necessitated a lumbar puncture for diagnostic reasons. They did not take either analgesics or psychotropic drugs for at least two weeks prior to the lumbar puncture. Lumbar puncture was performed in a sitting position between 0900 hr and I000 hr after the subjects bad fasted for 12 hr and had had bed rest for I0 hr. The C S F samples (16 ml) were removed without additions. To avoid rostral-caudal gradient effects, samples were gently mixed, and frozen on dry ice and stored untd analysis. The second part of the study comprised nine male schlzophremc patlents (mean age ± S D = 30.4 ± 5.9 yr) classified according to the Research Diagnostic Criteria (Spttzer et aL, 1978). All patients were of the chronic paranoid type with acute exacerbations. The average duration of the disease was 7.2 ± 4 yr, the number of hospitalizations was 10.6 ± 8 and the average age at onset of disease was 23. I ± 6 yr. After a lumbar puncture was performed for baseline determinations, the patients were treated for 21 days w~th haloperidol (mean dose ± S D = 25.7 ± 3.3 rag/day), at which time a second lumbar puncture was performed as described above. Informed consent was obtained from all of the subjects or their first-degree relatives after the nature of the study and its possible complications bad fully been explained. Arginine-vasopressin (AVP) was determined by radioimmunoassay using an antlserum (No. 10169, Drs. J. A. Dtlrr and M. D. Linheimer, University of Chicago), allowing a high sensitivityof the assay. Cross-reaction of this antibody with lysine vasopressin was 4.3~0, with arginine vasotocin 3.3070, and with oxytocin less than 0.01~0, A V P was extracted from C S F (0.5 ml) by the method of Robertson et aL (1973). Each tube was preincubated with A V P antiserum at a finaldilution of I :1,200,000. After 48 hr of preincubatlon of the standard solution or of the redissolved freeze-dried extract wlth antiserum (total volume 0.3 ml), ('aq) A V P (2500 cpm/0. I ml) was added. The assay mixture was then incubated for a further period of 48 hr at 4 ° C The sensitivityof the assay was 0.2 pg/tube. Fifty percent displacement occurred with 3.5 pg. The recovery of synthetic A V P (2- 8 pg/ml) added to AVP-free plasma from Brattleboro rats, homozygous for diabetes msipidus, was 64.4 ± 6.8~0 (SD) (n = 12). The intra-assay coefficient of variance was 13 1070 A V P values are indicated as fmol/ml (I fmol = 1.09 pg) (Rascher et al., 1981). Oxytocin (OT) also was measured by radioimmunoassay, developed m our laboratory (Lang et aL, 1981). Synthetic O T (Ferring, MalmO) was conjugated to thyroglobuhn and injected into white N e w Zealand rabbits. The antiserum was used m a final dilution of 1:600,000. Oxytocin was labelled with 125-I by the chloramine-T procedure and purified on a cation exchange column of CM-Sephadex C 50 m 0.6 M acetate buffer, p H 4.0. The labelled hormone, antiserum, and standards of samples were incubated for 48 hr. Separation of bound from free O T was achieved by charcoal. The assay was sensitiveto 1.05 fmol OT; 5007o of labelled O T were displaced by 26 fmol/tube. Synthetic A V P showed less than 0.5070 cross-reaction. No cross-reactivity was found with neuropbysin, alpha-MSH, A C T H , somatostatin and angiotensin up to I000 ng. The mira-assay coefficient of variation was 7. I~/o and the inter-assay coefficient of variation was 18.1070. All samples were run in one single assay Non-parametric tests were used for the statisticalevaluation of the data.

VASOPRESSIN- OXYTOCIN IN SCHIZOPHRENIC PATIENTS

189

RESULTS Vasopressin

There were no differences in CSF AVP levels among the controls, the schizophrenics treated with haioperidol, and the schizophrenics without neuroleptics (Table I). Furthermore, there was no statistically significant influence of three weeks' neuroleptic treatment on CSF AVP concentrations in the group of nine male paranoid schizophrenics (Table If). There was a negative correlation between the BPRS factor 'anxiety-depression' and C~F AVP concentrations before treatment (r = 0.82; p < 0.05) and during neuroleptic treatment (r = 0.82; p < 0.01). There were positive correlations between CSF AVP and CSF homovanillic acid both before (rs = 0.79; p < 0.01) and during treatment (rs = 0.59; p < 0.01). Oxytocin

There was a difference in CSF OT concentrations between both the healthy controls and schizophrenics without neuroleptics Co < 0.05) and the healthy controls and schizophrenics with neuroleptic treatment (p < 0.01) (Table I). There was no significant change in CSF OT levels after three weeks of haloperidol treatment. DISCUSSION The finding that there was no difference in CSF AVP concentrations between schizophrenics with and without neuroleptics and healthy controls disagrees with results reported by van Kammen et a[. (1981) who reported significantly decreased CSF AVP concentrations in schizophrenics compared to healthy controls. As neuroleptic treatment status cannot account for this discrepancy, we suggest that differences in schizophrenic subgroups may be responsible, as these researchers investigated a broader range of schizophrenic patients including those with schizoaffective disorders. Differences in water intake and smoking status also could account, at least in part, for the heterogeneity in the results. However, this possibility could be obviated by the finding that, whereas water deprivation and nicotine increased AVP levels in the plasma of rats, they did not influence its concentration in the CSF (Mens et al., 1980). The positive correlation between CSF AVP concentration and CSF homovaniUic acid before and after neuroleptic treatment is in agreement with the finding that dopamine turnover is increased after AVP treatment in terminal areas of the nigrostriata] and the tuberoinfundibular dopamine systems (Kovacs et al., 1977; Tanaka et aL, 1977b). Of further interest is the negative correlation between AVP concentration and the BPRS factor 'anxiety-depression'. Gold et al. (1978) found that depressed patients had lower levels of CSF AVP compared with normal volunteers. The increased concentration of CSF OT in the group of schizophrenics without neuroleptic treatment, the even higher concentrations after treatment and the (non significant) increase of OT during neuroleptic treatment may well be of functional significance. OT has been described as an amnestic peptide (Bohus et al., 1978). It has been reported to decrease catecholamine turnover in one study (Kovacs, 1977) but not in another study (van Wimersma Greidanus eta[., 1980).

190

HELMUT BECKMANN, RUDOLF E. LANG and WAGNER F. GATTAZ TABLE I. MEAN CSF VASOPRESSINAND OXYTOCINCONCENTRATIONS(± SD) IN THREE GROUPSOF SUBJECTS

Schizophrenic patients Without treatment With haloperidoi

Controls (n = 16)

(n = 13)

( n - - 15)

Vasopressin (pg/ml)

1.34 ± 0.76

1.15 ± 0.38

1.34 ± 0.38

Oxytocin (pg/ml)

7.11 ± 4.03

10.03 ± 4.03

13.46 ± 5.96

Vasoprcssin - no significant differences among the three groups. Oxytocin - significantly higher in patients with neuroleptics compared to controls ( p < 0.01) and in patients without neuroleptics compared to controls ( p < 0.05).

TABLE I1. MEAN CSF VASOPRESSINAND OXYTOCINCONCENTRATIONS('*" SD) IN NINE SCHIZOPHRENICPATIENTSBEFOREAND DURINGHALOPERIDOLTREATMENT

Before treatment Vasopressin

During treatment

1.16 ± 0.45

0.94 ± 0 16

7.72 ± 6.81

10.10 ± 5.91

(pg/ml) Oxytocin (pg/ml)

Vasoprcssin and oxytocin - no significant changes after three weeks of neuroleptic treatment.

It may be speculated that increased OT may be of etiological relevance for the occurrence of amnestic syndromes or even for Korsakow psychoses which may develop during chronic neuroleptic treatment (Kalinowsky et ai., 1982). However, we did not administer specific memory tests with our probands in this study. OT enhances and VP decreases intracranial self-stimulation (Schwarzberg et ai., 1980). As a decrease in self-stimulation has been hypothesized to be an animal correlate of certain forms of schizophrenia (Stein & Wise, 1972), amelioration of the psychotic state might be associated with the increased CSF OT concentrations resulting from neuroleptic drug therapy. Controversy remains concerning both the origin and the central function of AVP and OT found in the CSF. Thus it would be premature to speculate further from our results before experimental and clinical studies have established a better understanding of the many variables that influence these peptides. REFERENCES ADI k, R &. DE WJED, D. (1972) Effect of lysine vasopressm on passive avoidance learning. Psychonorrn. So. 29.46 - 48. BI ( KMANN, H., REYNOLDS, G. P., WALDMEIER,P. el al. (1982) Phenylethylamme and phenylacetlc acid Jn CSF of schizophrenics and healthy controls. Arch. Psychtatr. Nervenkr. 232, 463 - 4 7 1 . BOHI,~, B , KovAcs, G. L. & DE WIED, D. (1978) Oxytocin, vasopressin and memory: opposite effects on consolidauon and retrieval processes. Brain Res. 157, 4 1 4 - 4 1 7 .

VASOPRESSIN- OXYTOCININ SCHIZOPHRENIC PATIENTS

191

DE W IED, D. (1971) Long term effect of vasopressin on the maintenance of a conditioned avoidance response

m rats. Nature, Lond. 232, 5 8 - 6 0 . GATTAZ, W. F., WALDMEIER, P. & BECKMANN, H. (1982) CSF monoamine metabolites in schizophrenic patients. Acta Psychtat. Scand. 66, 350 - 360. GOLD, P. W., GODDWIN, F. K. & REUS, V. I. (1978) Vasopressin in affective illness. Lancet i, 1233 - 1236. KALINOWSKY, L. B., HIPPIUS, H. & KLEIN, H. E. (Eds.) (1982) Biological Treatments in Psychiatry. Grune & Stratton, New York. KOVACS, G. L., VECSEI, L., SZABO, G. et al. (1977) The involvement of catecholaminergic mechanisms in the behavioral effects of vasopressin. Neurosci Letts $, 337 - 344. LAN¢, R. E., RASCHER, W., HELL, J. et al. (1981) Angiotensin stimulates oxytocin release. Lore Sci. 29, 1425- 1429. LEGROS, J. J., GILOT, P., JERON, X. et al. (1978) Influence of vasopressm on learning and memory. Lancet i, 41 - 4 2 . MENS, W. B. J., BOffMAN, H. J., BAKKER, E. A. D. et al. (1980) Differential effects of various stimuli on AVP levels m blood and cerebrospinal fluid. Fur. J. Pharmac. 68, 8 9 - 92. OLIVEROS, J. C., YANDALI, M. K., TIMSITT-BERTHIER, M. et al. (1978) Vasopressin in amnesia. Lancet i, 42. OVERALL, J. E. & GORHAM, D. R. (1962) The brief psychiatric rating scale. Psychol. Rep. 10, 799-812. RASCHER, W., WEIDMANN, E. & GROSS, F. (1981) Vasopressin in the plasma of stroke-prone spontaneously hypertensive rats. Chn. Scl. 6 1 , 2 9 5 - 299. ROBERTSON, G. L., MAHR, E. A., ARTHUR, S. et al. (1973) Development and clinical application of a new method for the radioimmunoassay of arginine vasopressin m human plasma. J. Clin. Invest. $2, 2 3 4 0 - 2352. SCHWARZnERG, H., BETSCHEN, K. & UNGER, H. (1980) Action of neuropeptides on self-stimulation behavior in rats. In Neuropeptldes and Neural Transmisston, Marsan, C. A. & Traczyk, W. Z. (Eds.), pp. 339 - 341. Raven Press, New York. SPITZER, R. L., ENDICOTt, J. & ROBINS, E. (1978) Research diagnostic criteria: rationale and reliability. Arch. Gen. Psychiat. 35, 773 - 782. S'rEIN, L. & WISE, C. D. (1972) Possible etiology 6f schizophrenia: progressive damage to the noradrenergic reward system by endogenons 6-OH dopamine. In Neurotransmttters, Kopin, I. J. (Ed.), pp. 298-314. Williams & Wilkins, Baltimore. TANAKA, M., VERSTEEG, D. H. G. & DE WIED, D. (1977a) Regional effects of vasopressin on rat brain catecholamme metabolism. Neurosci. Lefts 4, 321 - 325. TANAKA, M., DE KLOET, E. R., DE WIED, D. et al. (1977b) Arginine-vasopressin affects catecholamine metabolism in specific brain nuclei. Ltfe Sci. 20, 1799- 1808. TELEGDY, G. & KOVACS, G. L. (1979) Role of monoamines in mediating the action of hormones on learning and memory. In Brain Mechanisms in Memory and Learning: From Style Neuron to Man, Brazier, M. A. B. (Ed.), pp. 2 4 9 - 268. Raven Press, New York. THOMPSON, E. A. & DE WIED, D. (1973) The relationship between the antidiuretic activity of rat eye plexus blood and passive avoidance behavior. Physzol. Behav. 11, 377 - 380. VAN KAMMEN, D. P., WATERS, R. N., GOLD, P. et al. (1981) Spinal fluid vasopressin, angiotensin I and It, beta endorphin and opioid activity in schizophrenia: a preliminary evaluation. In Biological Psychiatry, Perris, C., Struwe, G. & Jansson, B. (Eds.), pp. 3 3 9 - 344. Elsevier - North Holland, Amsterdam. VA~ WIMERSMAGREIDANUS, T. B., VAN REE, J. M. & VERSTEEG, D. H. G. (1980) Neurohypophyseal peptides and avoidance behavior: the involvement of vasopressin and oxytocin in memory process. In Neuropeptides and Neural Transmissaon, Marson, C. A., Traczyk, W. Z. (Eds.), pp. 293 - 300. Raven Press, New York. VERSTEEG, D. H. G., DE KLOET, E. R., VAN WIMERSMAGREXDANUS,T. B. et al. (1979) Vasopressin modulates the activity of catecholamine containing neurons in specific brain regions. Neurasci. Letts 11, 69 - 73. WALTER, R., VAN REE, J. M. & DE WIED, D. (1978) Modification of conditioned behavior of rats by neurohypophyseal hormones and analogues. Proc. Natl. Acad. Sc:. U.S.A.7$, 2 4 9 3 - 2496.