THE PHOTOPERIOD TRANSDUCER MELATONIN AND THE IMMUNE-HEMATOPOIETIC SYSTEM
The work performed in Locarno has been supported by Swiss Nationalfonds grants no. 3.267.0.85; 31.25350.88; 31.36128.92, 31.45532.95 and by the Helmut Horten Foundation.
The pineal neurohormone melatonin synchronizes functionally the organism with the photoperiod. It is now well recognized that melatonin also plays an important immunoregulatory role. T-helper cells bear G-protein coupled melatonin cell membrane receptors and, perhaps, melatonin nuclear receptors. Activation of melatonin receptors enhances the release of T-helper cell type 1 (Th1) cytokines, such as g-interferon and interleukin-2, as well as of novel opioid cytokines which crossreact immunologically with both interleukin-4 and dynorphin B. Melatonin has been reported also to enhance the production of interleukin-6 from human monocytes. These mediators may counteract secondary immunodeficiences, protect mice against lethal viral and bacterial diseases, synergize with interleukin-2 in cancer patients and influence hematopoiesis. Hematopoiesis is apparently influenced by the action of the melatonin-induced-opioids on kappa-opioid receptors present on stromal bone marrow cells. Most interestingly, g-interferon and colony stimulating factors may modulate the production of melatonin in the pineal gland. A hypothetical pineal-immune-hematopoietic network is, therefore, taking shape. From the immunopharmacological point of view, a call is made for clinical studies on the effect of melatonin in viral disease including human immunodeficiency virus-infected patients and cancer patients. In conclusion, melatonin seems to be an important immunomodulatory hormone which deserves to be further studied to identify its relevance in immune-based diseases, its therapeutic indications and its adverse effects.
The three main components of the fundamental temporal organization of the circadian system are the retina, the suprachiasmatic nuclei (SCN), and the pineal gland. The endogenous rhythmicity is generated in the SCN, while the retina and pineal gland are involved in preventing desynchrony of internal rhythms. The retino-pineal system functions as a resetting system which synchronizes the organism with the photoperiod [1].The synchronizing signal is constituted by the indoleamine melatonin which is synthesized and released during the night in all species upon activation of pineal b1 and a1 adrenoceptors. [1-3] Melatonin regulates fertility in seasonally breeding animals [4], although its role in other species including human is less clear [4]. However, the synchronization of endocrine activities with the appropriate environmental situation is of vital importance for animals living under natural conditions. Without the synchronizing signal of melatonin, animals become out of phase with their environment, a situation dangerous for their survival. With obvious differences, the human pineal gland seems to serve the same synchronizing function [1]. In addition, a number of patho-physiological situations have been associated with altered melatonin production. These include aging, affective diseases, neurological disorders and cancer [1] .
Relevance of Endogenous Melatonin Before 1986, some reports claimed that absence of the pineal gland stimulated the proliferation of immunocompetent cells [5-7] , in contrast to others [8-10]. In general, however, most studies agreed that pinealectomy is associated with a precocious involution and histological disorganization of the thymus [5, 9, 11, 12]. The mechanism of this effect was postulated to depend on increased gonadal steroid hormones. In 1981, we published the first evidence of a possible involvement of endogenous melatonin on the antibody response and on spleen and thymus cellularity in mice [13]. Later, by various pharmacological interventions aimed at inhibiting melatonin synthesis, we confirmed this finding and showed that T cell immune reactions were also affected [14]. In another report we showed that pinealectomy inhibits leukemogenesis in a radiation, leukemia virus murine model and that melatonin has a promoting effect on the disease [15]. A number of other authors further extended this type of evidence. Pinealectomized mice were reported to have depressed humoral responses and disruption of circadian ryhthmicity [16]. In another report, inhibition of endogenous melatonin in hamsters decreased spleen weight and reduced T cell blastogenesis. Melatonin administration counteracted this effect [17]. Interleukin-2 (IL-2) production and antibody-dependent cellular cytotoxicity were inhibited in pinealectomized mice and exogenous melatonin restored these important functions [18, 19]. Endogenous melatonin has been also reported to influence the concentration of bone marrow granulocyte/macrophage colony-forming unity (GM-CFU) [20]. An interesting finding which might be associated with and explained by the immunoenhancing action of endogenous melatonin, is the widely documented oncostatic role of the pineal gland and of melatonin [21]. At variance with these findings, we have reported that pinealectomy delays the development of T-cell leukemia in mice while melatonin exerts a disease promoting effect 15]. An explanation for this unexpected effect might be that in this case the malignant transformation affected T-cells which are also the melatonin target. Pharmacological Effects Our previous work has shown that melatonin can augment the immune response and correct immunodeficiency states which may follow acute stress, viral diseases, or drug treatment.[14, 22-25, 26 , 27, 28]. These findings have been confirmed and extended either in mice or in humans, to a variety of immune parameters [17-19, 29-33]. In this regard, a very significant biological effect of melatonin is the protection of mice against encephalitis viruses and lethal bacterial infections [34-36]. Recently, we also found that melatonin may act as a therapeutic agent in experimental gram-negative septic shock [37]. This effect was apparently due to inhibition of nitric oxide production but did not involve immune mechanisms [37]. In general, the immunoenhancing action of melatonin seems restricted to T-dependent antigens and to be most pronounced in immunodepressed situations. For example, melatonin may completely counteract thymus involution and the immunological depression induced by stress events or glucocorticoid treatment [25]. Melatonin is active only when injected in the afternoon or in the evening, i.e. with a schedule consonant with its physiological rhythm [14, 22]. In addition, melatonin is most active on antigen or cytokine activated immunocompetent cells [14, 22]. Consistent with these requirements, a recent report shows that melatonin may also restore depressed immunological functions after soft-tissue trauma and hemorrhagic shock [38]. Beside acquired immunity, natural immune parameters seem also influenced by melatonin. Natural killer activity was reported to be either stimulated or depressed by melatonin [18, 39-40]. However, in another study we did not find any effect of melatonin treatment on the natural killer activity of peripheral blood mononuclear cells from healthy volunteers [41]. In a tumor model of established lung metastases we found that melatonin could synergize with the anti-cancer effect of IL-2 [26]. More recently, we reported that melatonin may rescue hematopoiesis in mice transplanted with Lewis Lung Carcinoma (LLC) and treated with cancer chemotherapeutic compounds [42, 43]. However, as it is reported in detail hereunder, when the mice were tumor-free, melatonin augmented the chemotherapy-induced myelotoxicity. This indicates that melatonin does not have only beneficial or therapeutic effects. As a matter of fact, melatonin has been reported to exaggerate collagen-induced arthritis [44] and to promote T-cell leukemia [15]. Effect on Cytokines The immunopharmacological actions of melatonin seem to be mediated, at least in part, by activated T-helper (Th) cells which upon melatonin stimulation show an enhanced synthesis and/or release of cytokines such as IL-2 and g-interferon ( g-IFN) and opioid peptides [18, 25, 30, 32, 45, 46]. Melatonin activates human monocytes and stimulates interleukin-1 production [33]. However, T lymphocytes seem to be the main target of melatonin in mice [25,46] and humans in which physiological concentrations of melatonin stimulate IL-2 production [47]. We have reported that the immunoenhancing and anti-stress effect of melatonin is neutralized by the opioid antagonist naltrexone [23, 25, 41]. Known opioid peptides could mimick the effects of melatonin, with the kappa-agonist dynorphin being the most potent agent [24]. The hematopoietic protection involved the release of granulocyte/ macrophage colony-stimulating factor (GM-CSF) from bone marrow stroma upon stimulation by a Th cell factor induced by melatonin [42]. This factor was immunologically and biologically indistinguishable from interleukin-4 (IL-4) [43]. Nevertheless, further investigations aimed at verifying the melatonin-IL4 connection failed to confirm this finding. Instead, we found that this Th cell factor was constituted by 2 cytokines of 15 and 67 kDa MW with the common opioid sequence (Tyr-Gly-Gly-Phe) at their amino terminal and a carboxy-terminal extension which resembles both IL-4 and dynorphin B [48]. Both activated lymph node Th cells and bone marrow Th cells released these opioid-related cytokines which were named melatonin-induced-opioids (MIO) [48]. Due to their size and unusual immunological characterisation, the MIO might represent novel opioid cytokines. The lower molecular weight MIO (MIO-15) seems to mediate both the anti-stress and hematopoietic effects of melatonin [48]. This finding is consistent with our previous result concerning the ability of the kappa-opioid antagonist dynorphin to mimick the effects of melatonin [24]. Interestingly enough, in contrast with peripheral Th cells, bone marrow Th cells do not seem to require any antigenic activation to respond to melatonin. This may reflect an inherent difference of bone marrow Th cells from peripheral Th cells and a physiological requirement for sustained melatonin regulation of hematopoiesis. A finding which may support this is that endogenous melatonin may stimulate propiomelanocortin gene expression in rat bone marrow [49].
Normal, tumor-free or Lewis lung carcinoma (LLC)-bearing mice were treated with cyclophosphamide (CY) ± melatonin (ME) and naltrexone (NA). ME ± NA were injected s.c. once a day at 16.00 hour, at a dose of 1 mg / kg body weight (b.w.), from day 8 through day 12 after tumor inoculation . Control mice were injected with saline (PBS). CY was injected i.p. once a day at 12.00 hour at a dose of 160 mg / kg b.w. on the same days. The mice were then left untreated for 3 days and then sacrificed for GM-CFU assay. . The values represent the mean number of GM-CFU ± standard deviation. In brackets is reported the number of mice per experimental group.
This finding is germane to our studies on the hematopoietic action of the melatonin-MIO network [42, 43, 50]. Recently, we performed experiments in which we compared the ability of melatonin to protect hematopoiesis in LLC-bearing mice and in tumor-free normal mice treated with the cytotoxic drug cyclophosphamide. This experiment was suggested by the fact that melatonin added in granulocyte/macrophage colony-forming units (GM-CFU) cultures could directly enhance the number of GM-CFU but only in presence of suboptimal concentration of colony stimulating factors (CSF), i.e. in presence of activated bone marrow adherent cells [42, 43]. In addition, LLC is known to produce CSF and exert myelopoietic activity in vivo [50]. Melatonin did not exert any hematopoietic protection in tumor-free mice. Rather, the myelotoxicity of cyclophosphamide was increased by melatonin treatment (Table 1). However, both in tumor-free and LLC-bearing mice the effect of melatonin was neutralized by naltrexone (table 1) which suggested the involvement of MIO. We found that this dual effect of melatonin seems to depend on kappa-opioid receptors expressed by bone marrow stromal cells. In fact , the order of potency of opioid agonists in increasing the number of GM-CFU when added directly in cultures was consistent with the presence of a kappa-1 opioid receptor ( Fig. 1). The presence of GM-CSF seems to be needed for the kappa-opioid agonists (or the MIO) to exert their colony stimulating activity (Table 2). In absence of GM-CSF the MIO seems to increase the chemotherapy-induced myelotoxicity [52]. These surprising effects reveal the existence of complex regulatory mechanisms and need further studies.
Melatonin Receptors A major advance in understanding melatonin's circadian action has been the recent cloning of a family of G-protein-coupled receptors for melatonin. The melatonin receptors subtypes so far characterized are three, Mel1a, Mel1b and Mel1c with a Kd ranging from 20 to 160 pM [53]. As far as it concerns melatonin receptors in immunocompetent cells, high affinity binding sites for melatonin have been described in the membrane homogenates of thymus, bursa of Fabricius and spleen of a number of birds and mammals [54]. We have described a high affinity binding site in murine bone marrow Th cells [55]. Another study showed that melatonin binds to human lymphoid cells modulating their proliferative response. Consistent with our findings, T cell activation significantly increased melatonin binding [56]. Melatonin binding sites and melatonin receptor mRNA was mostly found in human Th cells, but also in CD8+ T cells and B cells [47]. Beside membrane receptors, nuclear receptors for melatonin have been described in human myeloid cells. Melatonin seems to be the natural ligand for nuclear orphan receptors RZR/ROR. It appears that melatonin down regulates the expression of the RZR/ROR responding gene which encodes for 5-lipoxygenase, a key enzyme in allergic and inflammatory disease [57]. It is noteworthy that this gene is not expressed in the brain and is not involved in circadian rhythmicity. Clinical Studies On the basis of our animal studies in which we showed that melatonin may synergize with IL-2 in controlling tumor growth [26], the group of Dr. Lissoni in Italy has conducted an impressive series of clinical studies in cancer patients. Over 200 patients with advanced solid tumor in which the standard anticancer therapies were not tolerated or not effective were treated with IL-2 and melatonin. The results obtained show that this neuroimmunotherapeutic strategy may amplify the anti-tumoral activity of low dose IL-2, induces objective tumor regression, prolongs progression-free time and overall survival and, moreover, the treatment was very well tolerated. It should be stressed that melatonin seems to be required for the effectiveness of low dose IL-2 in those neoplasias that are generally resistant to IL-2 alone (reviewed in 58 ). Similar findings were obtained in a smaller study in which melatonin was combined with g-IFN in metastatic renal cell carcinoma [58] . In addition, melatonin in combination with low-dose IL-2 was able to neutralize the surgery-induced lymphocytopenia in cancer patients [60]. On the contrary, a most recent double blind study investigating the myeloprotective effect of melatonin given in combination with carboplatin and etoposide to lung cancer patients shows that melatonin seems to increase the time of chemotherapy induced neutropenia (M. Ghielmini, in preparation). This would confirm the effect of melatonin in tumor-free mice and constitutes an important evidence of a potentially dangerous adverse effect of melatonin.
Mechanism of Action The mechanism of most immunopharmacological effects of melatonin seems straightforward. Melatonin binds to specific melatonin receptors on the membrane of Th cells stimulating the production of g-IFN, IL-2 and MIO which in turn upregulate the immune response. Second messengers are not completely understood but include G-proteins and inhibition of cAMP production [47]. The immunotherapeutic effect of melatonin against encephalitis viruses or bacterial infections [35-36] might be explained by the increased production of g-IFN and/or IL-2 as well as by an increased myelopoiesis due to the hematopoietic action of the MIO. A mechanism involving Th type 1 cytokines might also account for the capacity of melatonin to restore immunodeficiency states secondary to aging [30, 32], trauma-hemorrhage [38] or to synergize with IL-2 in cancer patients [58, 60]. In regard to the ability of melatonin to counteract the thymus involution and immunodepression caused by stress or corticosteroid treatment, the major mediators seems to be the MIO. We do not yet know whether the MIO action is exerted on peripheral immunocompetent cells and in the thymus or wether the hematopoietic effects of these novel opioid cytokines are also involved. On the contrary, it seems clear that the unexpected myelotoxicity induced by melatonin when administered together with cancer chemotherapeutic compounds or, viceversa, the hematopoietic rescue observed in LLC-bearing mice, depend on a complex series of events which involve the hematopoietic effects of the MIO. The finding that adherent bone marrow cells express kappa-opioid receptors [52] seems of considerable relevance. for understanding the melatonin effects and, in general, the physiology of hematopoiesis. MIO might belong to a new family of endogenous kappa-opioid agonists which, in the case of hematopoietic protection, seem to synergize with GM-CSF on stromal cell kappa-receptors [52]. This would explain why in LLC-bearing mice, MIO rescues hematopoiesis against the toxic action of cancer chemotherapy. LLC is, in fact, known to release GM-CSF [51]. In absence of a substantial concentration of CSF, melatonin seems to induce the opposite effect. This might depend on the presence of other opioid receptor types or on the involvement of g-IFN or of other hematopoietic inhibitors. The synergistic effect of melatonin with IL-2 probably depends on the fact that IL-2 per se produces an activation of peripheral Th cells and this seems to increase the expression of melatonin receptors [56]. Activated Th cells may produce also GM-CSF.This might explain the therapeutic and positive hematopoietic effects of melatonin when administered together with IL-2 in cancer patients [58, 60]. The cytokines involved in the immune-hematopoietic action of melatonin may exert an influence on the production of melatonin by the pineal gland. The pineal gland is, in fact, located outside the blood-brain barrier and some reports show that g-IFN may directly affect the synthesis of melatonin in the pineal gland [61]. Another report shows that also hematopoietic cytokines such as CSF may influence melatonin production [62]. Figure 2 shows the pineal melatonin-immune-hematopoietic network. Moreover, besides acting on membrane receptors, melatonin might affect the immune system via nuclear receptors [47, 57] or even via pituitary hormones such as growth hormone [63]. Finally, it has been also proposed that melatonin affects the immune system via the zinc-pool and the thymic hormone thymulin [64].
We can reasonably conclude that melatonin may be an important endogenous neuroimmunomodulator and a potential immunotherapeutic agent.. However, we are still far from a complete understanding of the mechanism underlying such properties. For example, it is not clear whether melatonin acts on Th1 or Th2 cells or on both. In addition, it is not known whether melatonin may induce cytokine gene expression or whether its action is posttranslational only.These seem rather important points as the Th1/Th2 balance and the resulting cytokine production are crucial for a successful immune response and may be relevant in immune-based pathologies [65]. The stimulatory effect of melatonin on IL-2 and g-IFN and the lack of influence on IL-4 suggests the involvement of Th1 cells. Perhaps, the same Th cell type may also produce MIO which are radically different from the enkephalin-containing molecules reported to be produced by Th2 cells [66]. On the other hand, the dramatic protection exerted by melatonin in experimental models of viral encephalitis and lethal bacterial infections as well as its capacity to restore depressed immune functions is consonant with Th1 cell involvement [67].
Physiologically, it seems possible to distinguish two different roles for melatonin. The first one occurs in acute conditions during a viral or bacterial infection which produces a substantial activation of the immune system. In that condition, endogenous and/or exogenous melatonin may optimize the immune response by sustaining Th cell functions and production of cytokines, part of which (MIO) have also myelopoietic activity. A second, more general role may be exerted at the hematopoietic-immune level by a chronic circadian resetting of the immunological machinery to maintain immune homeostasis. This is suggested by the observation that in healthy mice, i.e. in absence of any infection and immunological activation, only the Th cells which sit in the bone marrow are sensitive to melatonin [27, 50]. Products of this melatonin-bone marrow Th cell interaction are the MIO which may affect hematopoiesis and thymocyte proliferation [27, 50]. Both the acute and chronic mechanisms might be exploited in the use of melatonin as an immunotherapeutic agent to correct secondary immunodeficiency or fight viral diseases. As we already stated in a preceding review [27], we would like to stress the need for a large double blind study in human immunodeficiency virus (HIV)-positive patients. In the presence of normal Th cell counts, the apparent ability of melatonin to sustain Th cell functions and IL-2 and g-IFN production might result in delayed development or occurrence of AIDS. Reduction of plasma viremia was associated with an increased IL-2 mRNA expression in lymph nodes of HIV-infected patients [66]. IL-2 is the most potent cytokine capable of inducing the CD8+ T cell -mediated inhibition of HIV replication which seems to override the ability of IL-2 to stimulate HIV expression [68]. If effective, melatonin administration would be a relatively cheap and safe prevention of this devastating disease. Alternatively, melatonin treatment might be combined with low-dose IL-2 which seems to be beneficial in HIV-associated malignancies [69] or alternatively with HIV protease inhibitors. The use of melatonin in combination with IL-2 in cancer neuroimmunotherapy might prolong survival and improve the patient's quality of life [58], These encouraging results obtained by Lissoni and coworkers deserve, therefore, to be expanded and confirmed in other studies. In regard to the use of melatonin in combination with cancer chemotherapeutic drugs, the results obtained so far are disappointing. Melatonin seems to worsen the bone marrow toxicity of common cancer chemotherapeutic regimens. This fact calls for further studies to understand the role of melatonin in hematopoiesis and indicates that, in certain conditions, melatonin may have serious adverse effects.
|