study of light/dark rhythm of melatonin in relation to cortisol
and prolactin secretion in schizophrenia
ViganÚ, Paolo Lissoni, Franco Rovelli, Maria Grazia Roselli,
Fabio Malugani, Carlo Gavazzeni, Ario Conti & George Maestroni
cortisol, melatonin, pineal gland, prolactin,
April 10, 2001
Accepted: April 12, 2001
Recent studies have suggested the involvement of the pineal
gland and its main hormone melatonin (MLT) in the pathogenesis
of psychiatric disturbances, namely the depressive syndrome.
In contrast, the behavior of MLT secretion in schizophrenia
is still controversial.
MATERIAL & METHODS: The present study was carried out
to analyze light/dark rhythm of MLT secretion in relation
to that of cortisol and prolactin (PRL) in schizophrenic patients.
The study included 13 schizophrenic patients, 8 of whom were
untreated, while the other 5 patients were on neuroleptic
therapy. Serum levels of MLT, PRL and cortisol were measured
by RIA on venous blood samples collected at 8 A.M., 12 A.M.,
8 P.M. and 1 A.M. The control group consisted of 20 age-matched
RESULTS: A physiological nocturnal increase in MLT levels
occurred in 6/13 patients, whereas the other 7 patients showed
an abnormally low MLT peak during the night. Moreover, both
light and night mean levels of MLT were significantly lower
in patients than in controls. In addition, mean nocturnal
levels of MLT were significantly lower in chronic patients
than in those evaluated at the onset of disease. Cortisol
rhythm was normal in 11/13 patients, whereas PRL levels were
abnormally high in 10/13 patients.
CONCLUSIONS: This preliminary study would suggest that schizophrenia
may be associated with a diminished secretion of MLT from
the pineal gland, and pineal deficiency would be more evident
in the chronic disease. Finally, pineal alterations have appeared
to be associated with an altered secretion of PRL and cortisol,
by suggesting that the schizophrenic disease may be characterized
by marked neuroendocrine disturbances, whose physio-pathological
and prognostic significance needs to be established by successive
the historical mistake of the interpretation of the pineal gland
as a non-functioning organ, the recent advances in the knowledge
of the psychoneuroendocrinology have shown that the pineal gland
is a multifunctional neuroendocrine organ, able to modulate
almost all biological systems, in relation to environmental
and emotional stimulations through the circadian release of
indole hormones, among them melatonin (MLT) would represent
the most investigated pineal substance [1, 2]. MLT has been
proven to influence psychoneuroendocrine, immune and vascular
activities by acting on specific cell surface and nuclear melatoninergic
receptors , as well as by modulating the expression of other
receptors for neurotransmitters and neuropeptides.
In physiological conditions, MLT secretion increases during
the dark period of the day, resulting in a well defined circadian
rhythm [1-4], which plays a fundamental role in maintaining
other circadiian biological variations, including brain opioid
tone and neurotransmitter contents, as well as the immune functions
An altered pineal rhythm of MLT secretion, mainly consisting
of a lack of the physiological night increase, has been observed
in several severe human illnesses, including advanced cancer
, ischemic stroke  and depression . Therefore, because
of its importance as a regulator of the biological systems,
the evidence of a lack of MLT circadian rhythm would constitute
a non-specific neurobiological marker of severe systemic human
As far as psychiatric disorders are concerned, even though controversial
results exist, it is commonly accepted that the depressive syndrome
may at least be associated with a reduced MLT night secretion
[8, 9], which could play a role in determining depressionassociated
enhanced activity of the pituitary-adrenal axis. On the contrary,
the possible association between pineal alterations and schizophrenia
has been less investigated. Preliminary evidences would suggest
that the pineal might be involved in determining at least some
schizophrenia-related symptoms, such as hallucinations .
In fact, pineal obtained from deceased schizophrenic patients
have been shown to be able to transform MLT into 10-methoxy-armalane
, which may induce important psychedelic effects, because
of its chemical structure, similar to that of the hallucinogenic
agent armaline. Moreover, several hallucinogens, e.g. mescaline,
armaline, dimethyl-tryptamine  and cannabinoid compounds
 have been proven to stimulate MLT secretion by activating
the pineal enzyme hydroxyindoleOmethyltransferase (HIOXT), whiKeych
is the last enzyme involved in the biosynthesis of MLT from
Nacetylserotonin. In addition, the acute intravenous injection
of highly pharmacological doses of MLT has appeared to re-induce
the hallucinatory sy;mptomatology for some days in schizophrenic
patients clinically free from psychotic symptoms . These
experimental evidences would suggest an enhanced MLT release
in schizophrenia, since MLT and other less known psychomimetic
pineal substances are probably involved in the biochemistry
of the status of consciousness . Unfortunately, preliminary
clinical studies would seem to exclude a pineal hyperfunction
in schizophrenic patients, who in contrast seem to be characterized
by a reduced pineal function, at least in terms of MLT circadian
secretion . In fact, abnormally low nocturnal blood concentrations
of MLT have been observed in chronic schizophrenic patients
. This evidence has not been confirmed by other authors
, who described normal blood levels of MLT in schizophrenia.
The different subtypes of schizophrenia and stages of disease,
as well as the possible influence of treatments, could be the
variables responsible for the controversial results referred
by the various authors in literature. Therefore, a dysregulation
of the pineal gland rather than a reduced endocrine activity
could be involved in the pathogenesis of at least some schizophrenia-related
symptoms. The present study was performed, as an attempt to
better investigate MLT circadian rhythm in a group of schizophrenic
patients, in relation to that of two other hormones, whose secretion
is often altered in psychiatric diseases, e.g. cortisol and
study included 13 consecutive schizophrenic patients (M/F: 6/7;
median age: 26 years, range 20-37 years), who were admitted
at Psychiatric Division of the Hospital of Monza. The diagnosis
of schizophrenia was made according to DSM-III-R. Eight patients
were evaluated at the onset of disease and they were free from
therapy. The other five were chronic patients and they were
on treatment with neuroleptic drugs at the time of the study.
The control group consisted of 20 age- and sex-matched healthy
subjects. To evaluate MLT circadian secretion, venous blood
samples were collected at 8 A.M., 12 A.M., 8 P.M. and at 1 A.M.
Cortisol and PRL were also evaluated in the same blood samples.
Both patients and healthy controls followed a normal wake-sleep
rhythm during the study. Samples were collected through an indwelling
catheter inserted in an antecubital vein. The night samples
were drawn, after at least 3 hours of complete darkness, by
using a red light during the blood collection, representing
the only light that does not inhibit MLT secretion.
Serum samples were obtained by centrifugation and stored at
-70°C until assayed. Samples were measured in duplicate
within 30 days from the blood sampling. Serum levels of MLT,
PRL and cortisol were measured with the double antibody RIA
method by using commercial kits, and results were statistically
evaluated by the chisquare test, the Student's T test and the
analysis of variance at multiple time points, as appropriate.
expected, night mean serum levels of MLT observed in the healthy
controls were significantly higher than those seen during the
light period of the day (p < 0.001). Moreover, all healthy
subjects showed night concentrations of MLT at least higher
than 30 pg/ml. A normal nocturnal increase in MLT levels occurred
in 6 patients only, whereas the remaining 7/13 (54%) patients
showed an abnormally low MLT night increase. Moreover, MLT mean
serum levels observed in schizophrenic patients were lower than
in control subjects over the whole 24-hour daily period, and
this difference was statistically significant at 1.00 A.M. (p
< 0.001) and at 8.00 A.M. (p < 0.05). In addition, schizophrenic
patients evaluated at the onset of the psychotic symptomatology
in a therapy-free period showed mean nocturnal levels of MLT
significantly higher with respect to those seen in the chronic
patients (p < 0.05). Daily mean concentrations of MLT observed
in healthy subjects and in schizophrenic patients are illustrated
in Fig. 1, whereas Fig. 2 shows those found in chronic patients
and in patients evaluated at the onset of disease.
normal cortisol circadian secretion, with the highest concentrations
during the morning and the lowest ones during the night, was
observed in 11/13 patients, whereas 2 patients only (15%) had
an altered cortisol rhythmicity. In addition, as illustrated
in Fig. 3, mean levels of cortisol observed at 8.00 A.M. and
at noon were significantly higher in schizophrenic patients
who had no MLT rhythm than in those showing a normal MLT nocturnal
increase (p < 0.05). Finally, PRL serum concentrations were
within the normal range only in 3 patients, whereas they were
abnormally high in 10/13 (76%) patients. More in details, PRL
was elevated in all 5/5 chronic patients and in 5/8 acute patients.
This difference was not significant. In addition, as reported
in Table 1, mean levels of PRL were higher in the group of patients
with a lack of MLT rhythm than in those showing a normal light/dark
MLT circadian secretion, without, however, statistically significant
limited to a small number of patients, this study shows that
schizophrenia is associated with a diminished production of
the pineal hormone MLT, particularly during the night, which
represents the daily period of its physiological increase [1-4].
Therefore, in agreement with the results reported by Fanget
et al. , this study would suggest the occurrence of a pineal
hypofunction in schizophrenia. The existence of a reduced MLT
secretion as a neurobiochemical alteration, at least in part
responsible for schizophrenic symptoms, is also suggested by
the evidence that effective drugs in the treatment of the schizophrenic
symptomatology, such as the neuroleptic drug haloperidol, have
appeared to stimulate MLT secretion . Moreover, this study,
by showing lower nocturnal levels of MLT in chronic schizophrenic
patients than in those evaluated at the onset of the psychotic
symptomatology, would suggest that the clinical course of schizophrenia
may be characterized by a progressive and worsening pineal damage.
However, longitudinal studies, by monitoring patients during
the clinical course of disease, will be required to confirm
schizophrenia-related progressive pineal hypofunction, at least
in terms of MLT nocturnal production. The controversial results
described in literature about MLT production in schizophrenia
could depend on the possible changes in the pineal function,
in relation to the different stages of disease.
Finally, the evidence of higher daily levels of cortisol in
schizophrenic patients, who showed no MLT light/dark rhythm
than in those with normal circadian secretion, as observed in
the present study, could suggest a possible involvement of the
pineal in determining psychiatric disease-associated adrenal
dysfunctions, as well as the existence of different types of
pineal-adrenal interactions in the biochemistry of the schizophrenic
process. At present, it remains to be established whether there
is a relation between pinealadrenal connections and cl_inical
and prognostic characteristics of schizophrenia. In contrast,
PRL secretion does not seem to present different patterns in
relation to that of MLT.
Obviously, MLT is only one of the pineal hormones possibly involved
in the pathogenesis of schizophrenia. Other pineal indoles,
also provided by evident effects on the psychoemotional status
and by potential psychomimetic activity, such as 5-methoxytryptophol
and 5-methoxytryptamine, could also show altered secretions
and be involved in the generation of psychotic symptoms. In
any case, the evidence of a reduced MLT night secretion particularly
in chronic schizophrenic patients would justify randomized therapeutic
studies with neuroleptics alone, such as haloperidol, versus
neuroleptics plus MLT and/or other pineal indoles.
1 Axelrod J. The pineal gland: a neurochemícal transducer.
Science 1974; 184:1341-1348.
2 Arendt J. Melatonin. Clin Endocrinol 1988; 29:205-229.
3 Brzezinsky A. Melatonin in humans. N Engl J Med 1997; 336:186-195.
4 Attanasio A, Borrelli P, Gupta D. Circadian rhythms in serum
melatonin from infancy to adolescence. J Clin Endocrinol Metab
5 Régelson W, Pierpaoli W. Melatonin: a rediscovered
antitumor hormone? Cancer Invest 1987; 5:379-385.
6 Bartsch C, Bartsch H, Lippert TH. The pineal gland and cancer:
facts, hypotheses and perspectives. Cancer J 1992; 5:194-199.
7 Fiorina P, Lattuada G, Ponari O, Silvestrini C, Dall'Aglio
P. Impaired nocturnal melatonin excretion and changes of immunological
status in ischemic stroke patients. Lancet 1996; 347:692-693.
8 Branchey L, Weinberg U, Branchey M, Linkowski P, Mendelewicz
J. Simultaneous study of 24-hour pattern of melatonin and cortisol
secretion in depressed patients. Neuropsychobiology 1982; 8:225-232.
9 Grof E, Grof P, Brown GM, Arato M, Lane J. Investigations
of melatonin secretion in man. Prog Neuropsychopharmacol Biol
Psychiat 1985; 9:609-612.
10 McIsaac WM. A biochemical concept of mental disease. Postgrad
Med 1961; 30:111-118.
11 Lynch HJ, Wang P, Wurtman RJ. Increase in rat pineal melatonin
content following L-dopa administration. Life Sci 1973; 12:145-151.
12 Lissoni P, Resentini M, Mauri R, Esposti D, Esposti G, Rossi
D, Legname G, Fraschini F. Effects of tetrahydrocannabinol on
melatonin secretion in man. Horm Metab Res 1986; 18: 77-78.
13 Altschule MD. Some effects of aqueous extracts of acetone
extracts of acetone dried been pineal substance in chronic schizophrenia.
N Engl J Med 1957; 257:919-922.
14 Fanget F, Claustrat B, Dalery C, Brun J, Terra JL, Marie-Cardin
M, Guyotat J. Nocturnal plasma melatonin levels in schizophrenic
patients. Biol Psychiat 1989; 25:499-501.
15 Beckmann H, Wetterberg L, Gattaz WF. Melatonin immunoreactivity
in cerebrospinal fluid of schizophrenic patients and healthy
controls. Psychiat Res 1984; 11:107-110.
16 Gaffori O, Geffard M, Van Ree JM. Destyrgammaendorphin and
haloperidol increase pineal melatonin levels in rat. Peptides