Abstract

Section 1

Section 2

Section 3

Section 4

Section 5

Discussion
Board

The functional relationship between the pineal gland and the immune system was initially suggested by the disorganization of thymic structures observed after pinealectomy (Px) of newborn rats. Px or any other experimental procedure that inhibits melatonin synthesis and secretion induces a state of immunodepression, counteracted by melatonin in several species (for ref. see [8-10]). In vivo, melatonin displays an immunoenhancing effect, particularly apparent in immunodepressive states. Melatonin administration augments antibody responses, such as plate-forming cell activity, cytotoxic T-cell responses and antibody dependent cellular cytotoxicity, corrects immunodeficiency that followed stress, immunosuppressant drugs or viral infections, and prevents apoptosis in hematopoietic cells and thymocytes.

Melatonin modulates in vitro proliferation of stem and precursor cells for granulocytes and macrophages in bone marrow. It also augments T-helper cell and natural killer (NK) activity, interleukin-2 and gamma-interferon production, antigen presentation and interleukin 1 mRNA production in human monocytes, as well as the expression of the proopiomelanocortin gene in lymph nodes and bone marrow. In vitro, inhibitory influences of melatonin on immune function are also found, e.g., inhibition of NK activity or of gamma-interferon synthesis [8-10].

Melatonin, in physiological concentrations, stimulated activated CD4⁺ T cells in vitro to release opioid agonist(s) with immunoenhancing and antistress properties [8]. As an indication of a local feedback loop, macrophages and monocytes appeared to synthesize melatonin upon gamma-IFN activation. Melatonin potentiated the activity of interleukin-2 in host antitumor responses in humans and was able to rescue bone marrow cells from apoptosis induced by antitumoral chemotherapeutic drugs in vitro and in vivo. These effects are probably on bone marrow T helper cells, by stimulating the production of interleukin-4 like substances [8-10].

The nature of the mechanisms involved in the immunomodulatory activity of melatonin remains unsettled. There is evidence suggesting the existence of membrane specific binding sites for melatonin in immune cells [11]. Nuclear melatonin receptors may also mediate melatonin immunomodulation, since drugs that bind to RZR/ROR receptors are active in experimental models of autoimmune diseases [12]. Melatonin is also a potent antioxidant, acting by itself rather than through specific binding sites [13]. In addition, melatonin may affect centrally hypothalamic hormone release [14].

During the last years we have examined the regulation of circadian rhythmicity of cell proliferation in rat submaxillary lymph nodes and spleen. In both immunized and non-immunized rats we reported a significant diurnal variation of submaxillary lymph node ornithine decarboxylase (ODC) activity (an index of cell proliferation in immune tissue), displaying maximal activity at early afternoon [4] and coinciding with maxima lymph node mitotic response to lipopolysaccharide and concanavalin A [5]. We also reported daily rhythms in splenic NK activity and lipopolysaccharide-induced cell proliferation which exhibited a maximum at midnight and at early morning, respectively, while concanavalin A-induced splenic proliferation peaked at midday [5].

Splenic ODC showed maxima at morning hours, which coincided temporally with changes in some related immune parameters (i.e., mitogenic activity) [6]. In our studies we observed that lymph node and splenic cell proliferation of Px rats decreased by about half still exhibiting a low amplitude, significant diurnal variations with a maximum at the afternoon. Administration of melatonin at late evening restored ODC levels and amplitude of diurnal rhythmicity [2,6]. The results were compatible with the view that the pineal gland plays a role in circadian changes of immune responsiveness in lymph nodes and spleen via an immunopotentiating effect of melatonin on cell proliferation. The interaction between the CNS and the immune system is a bi-directional process. Immunocompetent cells affect local neural processes by paracrine means through the release of cytokines and by endocrine means on the CNS via the bloodstream [15]. During the immune reaction there is an increase of sympathetic and parasympathetic activity in local nerves. Submaxillary lymph node and splenic sympathetic activity attained their maximum at early night, while cholinergic activity in lymph nodes peaked during the afternoon. After pineal removal a significant decrease in amplitude of rhythm of presynaptic activity in lymphoid organs was found. These sequela of Px were also counteracted by evening melatonin administration [2,6]. Therefore, melatonin may act in part by modifying circadian rhythmicity of neural signals conveyed to the immunocompetent organs via the autonomic nerves.

Back to the top.

Click to enlarge

Fig. 1: Effect of melatonin on 24-hour changes in TH activity in sympathetic ganglia of old rats. Groups of 6-8, 18 months old rats were studied after receiving 17 daily s.c. injections of 10 or 100 micrograms of melatonin or its vehicle 1 h before lights off. Rats were killed at 6 different times throughout a 24-hour cycle (i.e., at 0900, 1300, 1700, 2100, 0100 or 0500 h). Also an additional group of rats was killed at 0900 h the next day. Shown are the means ± SEM. Melatonin at both doses restored amplitude of 24 hour rhythms. P values shown are those obtained in a factorial ANOVA. For further details see [16].

Click to enlarge

Fig. 2: Effect of melatonin on 24-hour changes in ³H-choline conversion to acetylcholine in sympathetic ganglia of old rats. For details see legend to Fig. 1. Shown are the means ± SEM. Melatonin at both doses restored amplitude of 24 hour rhythms; in the case of hypogastric ganglia only the night peak was affected significantly by melatonin. P values shown are those obtained in a factorial ANOVA. For further details see [16].

Click to enlarge

Fig. 3: Effect of melatonin on 24-hour changes in tyrosine hydroxylase activity and ³H-choline conversion into acetylcholine in heart and adrenal glands of old rats. For details see legend to Fig. 1. Shown are the means ± SEM. Melatonin at both doses restored amplitude of 24 hour rhythms. P values shown are those obtained in a factorial ANOVA. For further details see [16].

Other studies also underline the occurrence of restoring effects of melatonin on autonomic nervous system organization in aged animals. An age-related reduction in the maximal acetylcholine-induced contraction in the prostatic portion of rat vas deferens during the day was observed [20]. Both the nocturnal administration of melatonin and melatonin incubation of the vas deferens from old rats brought about a significant potentiation of the acetylcholine response.

For many years the idea that the pineal gland and melatonin were components of the mammalian circadian system was entertained with little experimental support. More lately, convincing data were provided that Px of rats leads to the increase in the rate of re-entrainment to a phase shift and that in constant light pineal removal causes a major disruption of the circadian system [21,22]. In view of the results depicted in Fig. 1-3, the hypothesis that melatonin, in addtion ot act directly on immune cells, augments amplitude, and perhaps, delays or advances the phase of the underlying central oscillator thus indirectly affecting time and quality of immune response, can be entertained.

Back to the top.

Click to enlarge

Fig. 4: Effect of melatonin treatment on subjective assessment of quality of alertness during the day and of quality of sleep. Groups of patients included: (a) patients with sleep disturbances alone (n= 22); (b) patients with sleep disturbance together with signs of depression (n= 9). Patients were treated from day 1 to day 21 with 3 mg capsules of melatonin, 30 min before expected sleeping time. Shown are the means ± S.D. Results were statistically analyzed by a non-parametric repeated-measures ANOVA followed by a Friedman's test. (*) p< 0.05, (**) p< 0.01 as compared to pretreatment value (modified from [9]).

Table 1 summarizes the effect of melatonin treatment on coefficient of variation of bed time, an indirect estimation of agitated behavior at the beginning of the night. Such a coefficient averaged 58% in dementia patients as compared to 27-33% in patients not showing dementia, when estimated on days 0-2 of treatment. Only in dementia patients a significant decrease was detected on days 19-21 of treatment. In the case of dementia patients, a significant improvement of sundowning was clinically reported in 7 out of 10 patients studied. As far as the concomitant benzodiazepine treatment after completing melatonin treatment, 8 out of 13 patients with primary insomnia suppressed or reduced concomitant benzodiazepine use (by 50 to 75% of initial doses).

Table 1

Effect of melatonin treatment on variance of bed time (an indirect assessment of sundowning) in a geriatric, sleep-disturbed population of patients.

Insomnia (n=22)          Insomnia+depression (n=9) Insomnia+dementia (n=10)
---------------------------------------------------------------------
Days 0-2    Days 19-21   Days 0-2     Days 19-21   Days 0-2    Days 19-21
---------------------------------------------------------------------
32.6 ± 14.9 31.5 ± 10.8  27.1 ± 14.1  24.5 ± 13.5  58.0 ± 24.7 41.5 ± 20.9*
---------------------------------------------------------------------

Patients were treated from day 1 to day 21 with 3 mg capsules of melatonin p.o., 30 min before expected sleeping time. Shown are the means ± S.D. of variation coefficients, i.e. [(S.D./mean) 100], of bed time values, as computed on days 0-2 and on days 19-21 of treatment . Results were statistically analyzed by a paired non-parametric Wilcoxon's test. (*) p< 0.03, as compared to 0-2 days value.

Results of the second open trial are summarized in Table 2. Neuropsychological evaluation at diagnosis of the 14 AD patients examined indicated a primary impairment of mnesic function of a varying degree in all subjects. All patients had cognitive and neuroimaging alterations (cortical and bitemporal atrophy) compatible with different evolutionary stages of AD. After varying periods of time of treatment with melatonin a significant improvement of sleep quality was found in all cases, the differences with the initial assessment being highly significant (p= 0.001, Wilcoxon's test) (Table 2).

Table 2

Effect of melatonin treatment (9 mg, p.o.) on subjective assessment of sleep quality and scores of FAST and Mini-Mental in 14 patients with AD.

#   Age  Sex  Evolut.   Treatm. Sleep Qual.  FAST       Mini-Mental
              (months) (months) I     F      I     F    I     F
-------------------------------------------------------------------
1   58   F    34        25      2     7      5     4    18    18
2   80   F    50        23      3     7      6a    5    15    15
3   76   F    70        30      2     8      7     6c   24    23
4   63   M    42        30      2     6      6a    5    24    23
5   77   M    48        23      2     5      5     4    20    19
6   78   F    65        30      4     6      3     2    19    19
7   68   M    38        25      2     5      4     3    24    25
8   68   F    46        22      3     6      4     4    19    19
9   82   F    70        30      3     6      6c    6c    5     5
10  53   M    80        24      2     5      6c    7a    2     2
11  77   F    30        22      3     6      2     2    24    25
12  66   M    82        24      2     5      6a    6c    9     9
13  69   F    72        30      3     5      4     5     9     9
14  80   M    70        35      2     8      6b    6b   10    10
-------------------------------------------------------------------

Sundowning, diagnosed clinically in all 14 patients examined, was not longer detectable in 12 of them, and persisted (although attenuated) in patients # 8 and # 9. Neuropsychological evaluation by FAST and Mini-Mental indicated absence of significant differences between initial and present state of evolution of the disease (Table 2). This should be contrasted with the significant deterioration of clinical conditions of the disease expected from patients after 1-3 years of evolution of AD (see e.g., [32]. No side effects of melatonin were detected in any of patients' population examined.

Changes in sleep-wake patterns are among the hallmarks of aging. During sleep there is an increase in the number and duration of waking episodes, a reduction in the nondreaming phases of sleep, the first REM phase occurs earlier in the night, and the tendency to fall asleep is increased during the day. Evidence indicates that since impaired melatonin secretion may be associated with sleep disorders in old age, the administration of melatonin can be helpful to treat sleep disorders in the elderly [25,26]. In a recent study we were able to demonstrated that the type of sleep augmented by melatonin in the elderly was mainly the non-REM sleep [33].

In the first study herein reported, an open, short-term, pilot examination of the sleep-promoting action of melatonin (3 mg p.o. for 21 days) was conducted in a small non-homogenous group of elderly patients with primary insomnia and with insomnia associated with depression or dementia [31]. The results indicated that melatonin therapy can be beneficial in the initiation and maintenance of sleep in patients suffering from insomnia alone or associated with depression symptoms. Melatonin efficacy to improve sleep did not correlate with a similar beneficial effect on mood in depressed patients. Melatonin was also hepful to reduce or suppress benzodiazepine comsumption, mainly in patients with primary insomnia.

Sundowning agitation is found in about half of the patients with dementia and is the most common cause of institutionalization of the demented geriatric patient [34]. In view that there was a clinical amelioration of sundowning in demented patients with the 3 mg melatonin dose used, but with relatively few effects on sleep quality, we decided to administer a higher melatonin dose (9 mg daily) to AD patients showing vesperal agitation. The retrospective account of these group of AD patients indicate that all patients in the present series treated with 9 mg melatonin daily did improve sleep quality and reduced or eliminated sundowning. So far, management of sundowning includes restriction of daytime sleep, exposure to bright light and social interaction schedules during the day [34]. The present results are strong evidence in favor of a beneficial effect of melatonin to treat sundowning.

Another important observation in the subjects studied was the halted evolution of the cognitive and mnesic alterations expected in comparable populations of patients not receiving melatonin. Although it cannot be stated at the present time whether this was a consequence of melatonin improvement in sleep-wake rhythms or an effect of melatonin treatment on any primary mechanism of AD, a putative interferring effect of melatonin in AD pathophysiology is not without experimental basis [35]. Indeed, in vitro data indicated that melatonin protects neurons against beta-amyloid toxicity [36,37], prevents beta-amyloid-induced lipid peroxidation [38] and alters the metabolism of the beta-amyloid precursor protein [39]. In a case report of two monozygotic twins suffering from AD of 8 years of evolution (treated one of them with melatonin, 6 mg p.o. for 36 months), evolution of the disease was halted in the melatonin-treated subject, as indicated by a stable impairment of mnesic function and a substantial improvement of sleep quality and reduction of sundowning [40]. Therefore, these results suggest that melatonin can be very helpful to treat AD patients.

This study was a part of a project supported by the University of Buenos Aires (TM 07), the Consejo Nacional de Investigaciones Científicas y Técnicas, Argentina (PIP 4156), the Agencia Nacional de Promoción Científica y Tecnológica, Argentina (A97B01 and PICT 2350) and Elisium S.A., Buenos Aires.

Back to the top.

Back to the top.

Back to the top.