it is now known that some hormones are produced by different organs, and they can act through a number of pathways, including membrane and nuclear receptors. Furthermore, a series of hormone metabolites, previously considered inactive compounds, are physiologically relevant. These features are also applicable to melatonin. Besides its multiple sites of production, melatonin displays numerous physiological functions through its interaction with membrane, nuclear, and cytosolic-specific binding sites (13). An example of its multiple regulatory pathways is the interaction of melatonin with cytosolic calcium-calmodulin complex, modifying many of the calcium-calmodulin-dependent enzymes, including the nitric oxide synthase and nitric oxide production (12). Even more, melatonin is able to exert direct, nonreceptor-mediated effects, such as the scavenging of free radicals, yielding a series of metabolites with significant physiological activities (20). These findings led to what is referred to as melatonin's cascade of actions (21). Melatonin is easily formed from tryptophan, with a low energetic cost, supporting the presence of the indoleamine in primitive life forms (8). A main feature of melatonin is its persistence during the phylogeny, from one-cell organisms to humans, regulating two main aspects of cell physiology: circadian rhythms, including those related to cell division and antioxidative and anti-inflammatory cell protection (5, 8, 13). An important question regarding melatonin's role in cell physiology is now being asked: is melatonin involved in one of the main features of the phylogeny, i.e., cell differentiation?
Multipotential bone marrow mesenchymal stem cells (MSC) are a common precursor of osteocytes and adipocytes, and it is known that a number of stimuli, including hormones and metabolic signals, can direct MSC differentiation toward one of these two cell lines (7). Basically, agents inducing osteoblast differentiation inhibit adipogenesis, whereas those inducing adipocyte differentiation inhibit osteogenesis. MSC differentiation, however, is a very complex regulated phenomenon, and it involves multiple regulatory agents, such as redox signaling and peroxisome proliferator-activated receptor (PPAR-γ) modulation (7). There are some studies showing that hypoxia reduces MSC adipogenic differentiation through the hypoxia-inducible factor-1α transcription factor (20). Also, superoxide dismutase (SOD)-deficient mice show spontaneous adipogenesis (11). These changes suggest that the redox balance in bone marrow may induce differentiation of MSC cells toward osteogenesis or adipogenesis, suggesting a role for oxygen free radicals in these regulatory pathways. Nitric oxide derivatives of linoleic acid are also potent adipogenic agonists in the physiological range, and they can inhibit osteogenesis through PPAR-γ receptors (19). The free radical scavenging properties of melatonin against oxygen and nitrogen reactive species, and its ability to induce expression of antioxidative enzymes, including SOD, may underline its ability to guide the MSC differentiation toward osteogenesis (3, 6, 20, 21). The relationship among age, oxidative stress, osteopenia, and adipocyte accumulation in the marrow cavity seems clear. Senescent-accelerated mice (SAM) mice, a murine strain of accelerated senescence and oxidative stress, show osteopenia and high levels of PPAR-γ mRNA (4). Melatonin prevents the age-dependent oxidative stress and inflammation in SAM mice (15, 16), suggesting that the significant decay in melatonin production with age may favor the adipogenetic pathway of MSC differentiation.
The paper by Reiter's group (18) first confirmed previous reports pointing to a possible role of melatonin in osteogenesis by directly analyzing fatty acid accumulation in the rat osteoblast-like ROS17/2.8 cell line with oil red O staining, a highly sensitive staining for triglycerides. The main conclusion of this study is that nanomolar concentrations of melatonin inhibited oleic oil-induced triglyceride accumulation by these cells. The authors attempt to address the mechanism(s) of this important action of melatonin. Working with the melatonin membrane receptors, Sánchez-Hidalgo and coworkers (18) found that luzindole, a membrane melatonin receptor subtypes 1 and 2 melatonin (MT1 and MT2) receptor antagonist, or S20928, a specific MT1 melatonin receptor blocker, blocked the antiadipogenic effect of melatonin. MT1 membrane receptors are involved in the inhibition of cAMP levels by melatonin. In the presence of forskolin, a stimulator of cAMP accumulation, melatonin was also unable to prevent triglyceride accumulation by ROS17/2.8 cells. One of the mechanisms suggested by the authors to explain these effects of melatonin was the inhibition of fatty acid uptake by these cells. Whereas membrane receptors are important targets for melatonin action, an interplay between membrane and nuclear receptors of the indoleamine has been proposed (23). Indeed, gene regulation by melatonin has been reported (23, 24), and thus, the effects reported by Sánchez-Hidalgo and coworkers (18) may also involve a genomic effect of melatonin. Retinoid acid receptor-related orphan receptor (ROR)-α, PPAR-α, and PPAR-γ, are members of the nuclear receptor superfamily of ligand-activated transcription factors. It was proposed that melatonin is the natural ligand for ROR receptor, and the indoleamine binds to ROR with a Kd of 1 nM (24).The thiazolidinedione derivative CGP 52608 specifically binds to and activates nuclear ROR and PPAR receptors with a Kd in the low nanomolar range (1 nM) (22). Other thiazolidinedione PPAR-γ ligands, such as rosiglitazone and pioglitazone, induce MSC adipogenesis and inhibit osteogenesis. Different thiazolidinediones exert adipogenesis and/or inhibit osteogenesis to different degrees. So, rosiglitazone inhibits bone formation through a suppression of the osteogenic transcription factors Runx2/Cbfal, osterix, and Dlx5 (2), whereas netoglitazone, which exerts less osteogenic inhibition than rosiglitazone, does not reduce the levels of Runx2 or Dlx5 (10). It seems that some type of interaction between Runx2 and PPAR-γ may be related to the age-dependent increased bone marrow adipogenesis and osteopenia. Molecular experiments led to the findings that PPAR−/− embryonic stem cells spontaneously displayed osteogenesis (1), whereas PPAR+/− mice, which showed an increased mRNA levels of osteogenic genes, also increased bone mass with a concomitant reduction in adipocyte formation by MSC (9). Thus, PPAR-γ may be considered a target for osteopenia therapy. So, the relationships between melatonin, ROR, and PPAR-γ should be analyzed to clarify the nuclear pathway of MSC differentiation by melatonin.
Melatonin has been related to the adipose tissue metabolism and obesity. Given melatonin in high fat feeding rats increased weight loss (25). Several mechanisms have been proposed including increased exercise and inhibition of lipid synthesis. But melatonin's effects on fatty acid metabolism go far beyond simple regulatory effect on lipid metabolism, influencing bone marrow cell differentiation. The known effects of melatonin in redox balance (15, 16, 20, 21), its inhibition of CaCaM-dependent NO production (12), its role in the genomic regulation of antioxidative enzymes including SOD upregulation (3, 6), and its putative relation to PPAR-γ, support a more complex regulatory effect on bone marrow cell differentiation. Moreover, the decay of melatonin with age (14, 17), which reduces all of these functions of melatonin, further supports the complexity of melatonin signaling on bone marrow differentiation, favoring the adipogenetic pathway of MSC differentiation. The clear state-of-the-art model constructed by Reiter's group (18) should be the first step to follow in future investigations of these other mechanism(s) involved in the cell differentiation properties of melatonin.
Work in the author's laboratory is supported by the Instituto de Salud Carlos III (Ministerio de Sanidad), the Ministerio de Educación, and the Consejería de Innovación, Ciencia y Empresa, Junta de Andalucía (Spain).
- Copyright © 2007 the American Physiological Society