OoZl-972x/94/7805-1075$03.00/0 Journal of Clinical Endocrinology and Metabolism Copyright 0 1994 by The Endocrine Society
Vol. 73, No. 5
Printed
in U.S.A.
Progressive Disruption of the Circadian Melatonin in Fatal Familial Insomnia*
Rhythm
of
FRANCE%0 PORTALUPPI, PIETRO CORTELLI, PATRIZIA AVONI, LUCIANA VERGNANI, PAOLO MALTONI, ANNA PAVANI, EMILIA SFORZA, ETTORE C. DEGLI UBERTI, PIERLUIGI GAMBETTI, AND ELI0 LUGARESI
Institute of Internal Medicine, University of Ferrara, Ferrara; and the NeurologicalInstitute (P.C., P.A., ES., E.L.) and the Central Laboratory of 5’. OrsolaHospital (P.M., A.P.), University of Bologna,Bologna, Ztaly; and the Division of Neuropathology,Institute of Pathology, CaseWesternReserveUniversity (P.G.), Cleveland,Ohio 44106
ABSTRACT Fatal familial insomnia (FFI) is a disease characterized by loss of sleep activity due to selective thalamic degeneration. To assess the secretory pattern of melatonin (MT) in FFI, we studied two cases of overt disease under standardized conditions and polysomnographic control. Each patient underwent repeated 24-h study sessions, and MT was assayed at IO-min intervals. Six healthy volunteers were used as controls. Slow wave sleep was never recorded, whereas occasional episodes of enacted dreaming accompanied by rapid ocular movements and complex muscular activities were documented, with no detectable rhythm. Plasma MT concentrations gradually decreased as the disease progressed. A significant circadian rhythm was detected in the earlier recordings, with decreasing amplitudes with disease progression. Complete rhythm obliteration was achieved in the most advanced stage. Normally placed nocturnal acrophases were detected in the earlier stages, but then a shift toward the daytime hours was observed. Thalamic lesions of FFI appear to determine a progressive disruption of the sleep/wake cycle accompanied by decreased circulating levels of MT, with progressive alterations in the circadian rhythm of this hormone. On the other hand, decreased secretion of MT may contribute to the sleep disturbances of FFI. (J Clin Endocrirwl Metab 78: 10751078,1994)
(MT) has been proposed as a putative sleep-inducing factor (1) on the basis of some pharmacological and pathophysiological evidence, although no relationship between MT plasma variations and sleepphases has been demonstrated (2). MT is also known to influence the electrical activity of the central nervous system, causing its reduction in animals(3) and the acute decreasein electrical seizure activity of epileptic patients (4). The present report focuses on changes in the diurnal profiles of plasma MT levels in patients with a rare, genetically determined disease,fatal familial insomnia (FFI). The diseaseis characterized by a progressive reduction then loss of physiological sleep and impaired autonomic and motor functions due to selective degeneration of the anterior and mediodorsal thalamic nuclei (5-12).
ELATONIN
M
Subjects and Methods
We studied two affected members from the original family in which FFI was first described. Patient 1 was a 58-yr-old woman at the onset of the disease, which lasted 35 months. She underwent three 24-h study sessions in the early stages, at 6,9, and 14 months. Patient 2 was a 35yr-old woman at disease onset and was studied in more advanced stages Received October 28, 1993. Accepted January 21, 1993. Address all correspondence and requests for reprints to: Francesco Portaluppi, M.D., Institute of Internal Medicine, University of Ferrara, Via Savonarola 9, I-44100 Ferrara, Italy. * This work was supported by grants from the Italian National Research Council (CNR, Rome, Italy), Prevention and Control of Disease Factors, Subproject Stress (no. 93.0071.PF41), and from the Italian Ministry of University and Scientific and Technological Research
(40-60%).
(17, 20, and 23 months). Death occurred 40 days after the last session, at 25 months. In addition, six normal volunteers (four females, 35-59 yr old) underwent a 24-h study session with the same protocol and were used as controls. The protocol was approved by the local ethics committee. The 24-h study sessions were preceded by at least 2 nights in the hospital for habituation and were performed under continuous recumbency, except for meal taking, at 0800-0900,1200-1300, and 1730-1830 h, when the head of the bed was elevated 30°. Lights were on between 0700-2200 h. A sound-proofed room at constant temperature (24-25 C) was used. Two hours before the beginning of each session, an iv cannula was inserted in an antecubital vein and kept open with a slow infusion of 0.9% saline (10 mL/h). Venous samples were drawn every 30 min, beginning at 2000 h. Polygraphic tracings were obtained to simultaneously monitor electroencephalograph, ocular movements, muscle activity, and other vital parameters. Blood samples were drawn into precooled Vacutainers (Becton Dickinson, Rutherford, NJ) containing 1.5 mg/mL EDTA disodium salt. They were promptly centrifuged at 3000 X g for 15 min at 4 C, then the plasma was frozen at -80 C until the assay was performed. All samples for each hormone were processed in duplicate in the same assay. Plasma MT was measured by RIA after diethyl ether extraction from plasma, using a commercial kit supplied by Nichols Institute Diagnostics (San Juan Capistrano, CA). The assay sensitivity, i.e. the smallest single value that can be distinguished from zero at the 99% confidence limit, was 13 pmol/L. The precision (intraassay variance) was 9%, and the reproducibility (interassay variance) was 16%. The cross-reactivity of the antiserum is, on a molar basis, 100% with human MT. The interference with the assay from 6-sulfatoxymelatonin, N-acetyltryptophan, 5-methoxytryptophan, tryptamine, L-tryptophan, serotonin, and serotonin creatine sulfate is always less than 0.1%. The interferences with N-acetylserotonin, 5-methoxytryptamine, and 6-hydroxymelatonin are 0.02%, 0.05%, and l%, respectively. Analysis of rhythmicity was performed on the individual data of each recording session. The time series for MT were analyzed according to the single cosinor method, using a computerized procedure that we developed (13). With this procedure it is possible to determine whether
1075
Downloaded from jcem.endojournals.org by on May 7, 2008
1076
PORTALUPPI
ET AL.
425, ..
JCE & M .1994 Vol’lS.No5
there is a rhythm within a 24-h period (P < 0.05) and to evaluate the following parameters with their 95% confidence limits: 1) mesor (midline estimating statistic of rhythm), rhythm-adjusted 24-h average; 2) amplitude, the difference between the maximum value measured at acrophase and the mesor in the cosine curve; and 3) acrophase, lag between reference time (midnight) and time of highest value of the cosine function used to approximate the rhythm.
CASE 1
, ,I,
I I
CASE 2
Month 17
425 340
..
Month
6
Results The polysomnographic tracings revealed the disappearance of any physiological sleep activity in all recording sessions both patients. Instead, short episodes electroenof of cephalograph desynchronization were recorded, associated with muscular jerks of the upper limbs intermixed with more organized and complex motor activities in the electromyograph of limb muscles. These episodeswere interspersed throughout the 24 h, so that no temporal pattern in their occurrence was detectable. In the control group, plasma MT levels showed the expected 24-h pattern of variation (Fig. 1). In FFI, lower MT concentrations with reduced day-night differences were already present in the earlier recordings, and a decreasing trend was apparent with the advancement of the disease (Fig. 2). A significant circadian rhythm in plasma MT levels was detected at rhythm analysis of all recording sessions except the last of month 23 in patient 2 (Table 1). The circadian mesor and amplitude of MT were reduced. As the diseaseprogressed, these parameters showed a decreasing trend, causing complete obliteration of the circadian rhythm in the last sessionof patient 2. Normal nocturnal acrophases were calculated in the earlier recordings, but a later shift toward the daytime hours was observed. Discussion The major finding of the present study is that the circadian rhythm of plasma MT concentrations is progressively disrupted in patients with FFI. The mean 24-h levels and the
El o--o04 08 2 00 425’1
340
12
16 14
20
COCKl 04 11425
OR
12
16 23
20
00 1340
1
Month
1 (
Month
o00
04
08
12
16
20
0000
I,,,,,,,,,;,b 04 08
12
16
20
CO0
TIME (clock hour)
FIG.
2. Circadian profile of plasma MT levels in different stages of
right)
FFI. Increasing severity (top to bottom left to and
of autonomic
and motor impairment was present, whereas simultaneous electroencephalogram demonstrated the total loss of sleep activity. When significant, the 24-h cosine fit is given by the solid curve. The rhythmadjusted 24-h mean is represented by the horizontal line.
Orlllliilllili
TIME (clock hour)
FIG.
00
1. Mean circadian profile of plasma MT levels in six healthy (+SEM) over 30-min intervals. The 24-h cosine fit is given by the solid curve. The rhythm-adjusted 24-h mean is represented by the horizontal line.
subjects. Datapoints arethemean values
circadian variation (amplitude) of this hormone decrease gradually with the advancement of the diseaseand parallel a progressive phase shift toward the daytime hours. As a result, complete rhythm obliteration is achieved in the more advanced stagesof the disease. The progressive changes in MT circadian rhythm were assessed controlled and reproducible conditions and canin not be attributed to external interfering factors. A circadian variation in blood MT levels is well established in normal subjects, with barely detectable concentrations during the day and a clear nocturnal rise, peaking at lOO300 nmol/L between midnight and 0300 h (14-20). The rhythm in the production and secretion of MT is synchronized by the light-dark cycle. In particular, MT production is controlled by electrical signals received by the pineal gland from the suprachiasmaticnucleus (SCN) (21), which, in turn, receives modulatory inputs from the periventricular nucleus of the thalamus (22). The CAMP-dependent signal transduction pathway serves as a relay to stimulate MT synthesis (23), and rhythmic adrenergic signals are essentialin determining the circadian fluctuation in CAMP-induced transcription (24). At night, postganglionic fibers originating from the superior cervical ganglia release noradrenaline (25, 26), which, in turn, regulates MT synthesis through P-adrenergic receptors (24). Both chronic denervation of pineal glandsand injection of the P-adrenergic antagonist propranolol before the onset of darknessblock CAMP-induced transcription and MT synthesis (24). On the other hand, the /3-adrenergic agonist isoproterenol induces expressionof the specific transcription factors in ganglionectomized animalsconcomitantly with MT synthesis (24, 27). In patients affected by FFI, a
Downloaded from jcem.endojournals.org by on May 7, 2008
MT IN FFI
TABLE 1. Circadian
Months after onset Patient 6 9 14 Patient 17 20 23 Controls 1