Snapshot: implications for melatonin in endoplasmic reticulum homeostasis

Tables of Links

Introduction

Melatonin, known as the ‘hormone of darkness’, is important in both animals (Brzezinski, 1997; Tosches et al., 2014) and plants (Arnao and Hernandez-Ruiz, 2014). In mammals, melatonin is primarily produced by the pineal gland, displays characteristic daily and seasonal patterns of secretion and plays an important role in sleep regulation (Tosches et al., 2014; Gandhi et al., 2015). Melatonin and the melatonin receptor have been demonstrated to be associated with major depression (Hickie and Rogers, 2011), jet lag (Waterhouse et al., 2007; Sack, 2010), multiple sclerosis (Farez et al., 2015), diabetes mellitus (DM) (McMullan et al., 2013), cancer (Dauchy et al., 2014; Sigurdardottir et al., 2015; Ma et al., 2016), fibrosis in multiple organs (Hu et al., 2016) and drug addiction (Feng et al., 2013) and might represent novel treatment targets. In particular, the administration of melatonin is able to normalize the increased pro-inflammatory profile observed during metabolic syndrome (Cano Barquilla et al., 2014), atherosclerosis (Hu et al., 2013) and sepsis (Galley et al., 2014). Furthermore, the presence of numerous transmitters originating from various sources, particularly noradrenaline, has been reported to regulate the production and secretion of melatonin in the pineal gland (Simonneaux and Ribelayga, 2003; Gonzalez et al., 2012). The melatonin produced by the pineal gland further influences the rest of the body (Borjigin et al., 1999), and it has been reported that the reproductive system is controlled by the daily rhythm in melatonin production (Tamarkin et al., 1985).

The endoplasmic reticulum (ER) has a remarkably complex structure consisting of a single bilayer with a network of sheets and dynamic tubules (Westrate et al., 2015). Numerous genetic and environmental insults have been demonstrated to disturb the ultrastructure and volume of the ER, leading to a buildup of misfolded proteins in this organelle (a condition called ER stress) and initiation of the unfolded protein response (UPR) to remove the misfolded proteins (Walter and Ron, 2011; Brodsky, 2012; Hetz, 2013; Mehnert et al., 2014; Yang et al., 2015). An abnormal ER and ER stress are emerging as key contributors to a growing list of human diseases (Tabas and Ron, 2011; Walter and Ron, 2011; Oakes and Papa, 2015), including cancer (Mahoney et al., 2011; Clarke et al., 2014), neurological disease (Roussel et al., 2013; Hetz and Mollereau, 2014) and metabolic diseases (Fu et al., 2011; Arruda et al., 2014; Yang et al., 2015). Notably, there is close relationship between ER structure and function, and defects in ER structure are associated with diseases (Westrate et al., 2015). The structure and morphology of the ER can be regulated by various factors (Chen et al., 2012; English and Voeltz, 2013), and melatonin has attracted increasing attention.

This review focuses on the relationship between melatonin and the ER. Firstly, we briefly summarize the normal structure and functions of the ER. We then discuss the existing implications for melatonin in ER homeostasis, and the regulation of ER stress by melatonin in various pathologies is subsequently introduced. The information compiled in this review should help scientists understand the involvement of melatonin in ER homeostasis.

Summary of the ER

The ER is an important intracellular membranous organelle in eukaryotic organisms that consists of an interconnected network of flattened, membrane-enclosed sacs or tubes known as cisternae (Hu et al., 2011; Smith et al., 2011). The ER is located throughout much of the cytoplasm, and the membranes of the ER are continuous with the plasma and nuclear membranes (Giordano et al., 2013; Chung et al., 2015; Westrate et al., 2015). A close association exists between the ER and the mitochondria (Rowland and Voeltz, 2012; Arruda et al., 2014), Golgi complex (Zanetti et al., 2012; He et al., 2013) and peroxisome (van der Zand et al., 2012; Tabak et al., 2013). The ER is found in most types of eukaryotic cells, including the most primitive Giardia (Soltys et al., 1996), and can be classified into two types, namely, rough ER (RER) and smooth ER (SER), depending on the presence of protein-manufacturing ribosomes on the surface (Reid and Nicchitta, 2015). The RER is mainly associated with protein synthesis, and the SER is mainly responsible for lipid metabolism, carbohydrate metabolism and detoxification (Friedman and Voeltz, 2011). In general, the ER is responsible for protein folding and trafficking (Rowland and Voeltz, 2012; Zanetti et al., 2012; Reid and Nicchitta, 2015), lipid synthesis (Rowland and Voeltz, 2012; Prinz, 2014) and the maintenance of calcium homeostasis (Smith et al., 2011; Stutzmann and Mattson, 2011; Rowland and Voeltz, 2012).

The correct folding of newly synthesized proteins is made possible by the involvement of several ER chaperone proteins, including protein disulfide isomerase, the heat shock protein 70 (Hsp70) family member that binds immunoglobulin protein (BIP or glucose-regulated protein 78), calnexin, calreticulin and members of the peptidylpropyl isomerase family. Only properly folded proteins are transported from the RER to the Golgi apparatus. Numerous environmental, physiological and pathological stressors, nutrient fluctuations and chemical triggers disturb the ER protein folding environment (Zaouali et al., 2010; Brown et al., 2014; Carloni et al., 2014; Chan et al., 2015; Garcia-Marques et al., 2015); these factors cause protein misfolding, resulting in the accumulation of misfolded or unfolded proteins, a condition defined as ER stress (Wang and Kaufman, 2014).

The UPR is a collection of signalling pathways that have evolved for the maintenance of a productive ER protein folding environment (Wang and Kaufman, 2014). The UPR comprises three parallel signalling branches: PRKR-like ER kinase (PERK)–eukaryotic translation initiation factor 2α (eIF2α), inositol-requiring protein 1 (IRE1)–X-box binding protein 1 (XBP1) and activating transcription factor 6 (ATF6) (Santos et al., 2014; Wang and Kaufman, 2014). Under non-stress conditions, BIP binds to the domains of PERK, IRE1 and ATF6 to stabilize them and prevent their activation (Ozcan et al., 2004; Cnop et al., 2012). Stress can promote the separation of the BIP from PERK, IRE1 and ATF6 and the subsequent activation of these three molecules (Figure 1). Briefly, PERK autophosphorylation promotes the phosphorylation of eIF2α to strongly inhibit mRNA translation, shuts down global protein synthesis and induces an increase in activating transcription factor 4 (ATF4). IRE1 works as an endoribonuclease by slicing the mRNA of XBP1. ATF6 is processed by membrane-bound transcription factor site-1 (also known as site-1 protease, SP1) and SP2 in the Golgi complex to generate cleaved ATF6. ATF4, XBP1 and cleaved ATF6 promote genes encoding ER chaperones. Taken together, these results indicate that the triggering of ER stress and activation of the UPR might be associated with a pro-survival or pro-apoptotic outcome via different mechanisms, which ultimately determine the fate of cells (Santos et al., 2014). BIP, PERK, eIF2α, ATF4, growth arrest- and DNA damage-inducible genes 153 (GADD153 or CHOP, whose transcription is activated by ATF4; Harding et al., 2000; Palam et al., 2011; Wang and Kaufman, 2014), IRE1, XBP1, ATF6, BIP and glucose-regulated protein 94 (GRP94) are often used as ER stress markers.

Melatonin and the ER in the pineal gland

Seasonal variation in the volume densities of the ER of at least some tissues is distinct and synchronized by changes in the photoperiod regulated by the pineal gland and the melatonin rhythm (Munoz et al., 2001). This finding indicates that melatonin has implications in ER homeostasis. As an important type of indoleamine, the synthesis of melatonin and associated key enzymes in the pineal gland may be closely associated with the state of the ER, particularly the RER (Kappers, 1978; Tan et al., 2015). Higher melatonin secretion appears to be partly related to the existence of developmental changes in the morphology and quantity of pinealocytes (the functional units of the pineal gland), which are characterized by a relatively large volume of RER (Redondo et al., 2003). Due to the diurnal rhythm of pineal melatonin production, the plasma melatonin concentrations are significantly higher at 02:00 h (at night) than at 09:00 h, 14:00 h (during the day) and 21:00 h (early night), but no significant differences in the RER are discernible (Lewczuk et al., 2004). Correlated with the well-known nocturnal enhancement in the secretion of melatonin from the pineal gland, higher relative volumes of granular ER are found at night than during the daytime in pinealocytes (Karasek et al., 1990; Swietoslawski and Karasek, 1993). This finding indicates that the ER might be closely related to the secretion of melatonin by the pineal gland.

In the pineal gland of old shrews, the parenchyma undergoes alterations that mainly affect the ER cisternae, which show increased numbers of dense bodies, the formation of many concretions and a depletion of presumed secretory products (Dekar-Madoui et al., 2012). These changes indicate that the gradual changes in the ER cisternae that occur during aging in Crocidura russula represent a reduced pinealocyte metabolism. Various stressors, including sleep deprivation and weaning of offspring, result in changes in the physiological function and morphological character of the ER, and these changes are reversed by melatonin supplementation. Thus, in goat kids subjected to premature weaning, melatonin might contribute to improvements in the histophysiological function of the ER (Redondo et al., 2010). The mean concentrations of melatonin in the plasma of weaned goat kids are significantly lower than those in non-weaned offspring. A quantitative ultrastructural analysis of pinealocytes showed that the relative volume of the RER in pineal glands is significantly lower in weaned goat kids than in non-weaned goat kids; however, treatment with melatonin significantly increased the RER volume in weaned kids (Redondo et al., 2010). Additionally, as a result of sleep deprivation, many pinealocytes exhibit dilation of the cisternae of the RER and SER, and these changes are attenuated by a single injection of melatonin (Lan et al., 2001). Collectively, these findings indicate that exogenous melatonin might promote a positive feedback in its pineal production through ER regulation.

Melatonin and ER function in target glands

Exogenously administered melatonin not only affects the ER morphology of the pineal gland but also regulates that of numerous organ systems, including the reproductive organs, the parathyroid gland and cancer cells. Melatonin directly or indirectly alters the ultrastructural appearance of mouse Leydig cells and influences their secretory activity by inhibiting their capacity to secrete steroids (Reiter et al., 2009). Melatonin treatment induces an obvious reduction in the volume of the SER, RER mitochondria and Golgi complex (Redins et al., 2002). Similarly, in adult Syrian hamsters, melatonin treatment alters the morphology of the seminal vesicles and ventral prostate, which presents conspicuous secretory granules and parallel arrays of granular ER with narrow cisternae (Chow and Pang, 1989). Furthermore, the cytoplasm of the coagulating glands and dorsal prostate of non-melatonin-treated animals are dominated by distended cisternae of granular ER and apical blebbing of endothelial cells. The administration of melatonin induces structural modifications, reflecting a loss of functional activities or even death of accessory sex gland secretory cells and thus confirming the suppressive effects of melatonin on the reproductive structures of rodents. Although the actions of melatonin on the reproductive morphology of the hamster might be indirect due to inhibition of gonadotropin secretion (Reiter et al., 2009), melatonin also exerts direct inhibitory effects on the peripheral reproductive organs (Reiter et al., 2013a).

Melatonin might affect the secretory activity of the parathyroid gland. The in vitro melatonin treatment of these glands significantly reduces the cisternae of the RER compared with those of control parathyroid glands (Shoumura et al., 1992). These changes are hypothesized to be induced by melatonin suppressing the synthesis of parathyroid hormone in cells (Shoumura et al., 1992). In human breast cancer cells, specifically MCF-7 cells, 4 days of exposure to melatonin resulted in the disruption of mitochondrial cristae and vesiculation of the SER (Hill and Blask, 1988). Thus, under normal conditions, the ultrastructure and physiological function of the ER in target organs are changed as a consequence of melatonin treatment.

Melatonin rectifies the abnormal appearance of the ER in target glands

The RER is decreased in volume in the neurons of patients with Alzheimer's disease (AD), and this effect is reversed by melatonin supplementation (Ling et al., 2009). Another study found that unglycosylated ER-bound β-APP derivatives are the predominant forms that are marginally affected by melatonin treatment (Lahiri, 1999). Under pathological conditions, melatonin rectifies not only the amount of ER, but also the ultrastructural appearance and presumed function of the ER. Similarly, dilation of the RER is observed in experimental models of acute pancreatitis (Esrefoglu et al., 2006), X-ray-irradiated intestine (Hussein et al., 2008) and water avoidance stress-induced degeneration of the liver parenchyma (Contuk et al., 2005). Dilation of the ER also occurs in the testicular damage associated with hyperlipidaemia (Zhang et al., 2012). Melatonin administration prevents these changes in the ER. Methanol intoxication-induced liver injury results in the appearance of extensive tubules of the SER, an effect that is also prevented by melatonin administration (Koksal et al., 2012).

Melatonin regulates calcium homeostasis via the ER

An elevation in the intracellular concentration of Ca²⁺ is considered a common pathological precursor (Santofimia-Castano et al., 2014), and the ER is the major inositol trisphosphate (IP3)-sensitive Ca²⁺ store and plays an important role in the homeostasis and function of Ca²⁺ (Alves et al., 2011). Cholecystokinin octapeptide (CCK-8) induces Ca²⁺ mobilization in mouse, freshly isolated pancreatic acinar cells (del Castillo-Vaquero et al., 2010). However, melatonin reduces Ca²⁺ release and modulates the pancreatic responses induced by CCK-8, which might be explained by the stimulation of Ca²⁺ transport from cells through the plasma membrane and subsequent Ca²⁺ reuptake into the ER (del Castillo-Vaquero et al., 2010; Santofimia-Castano et al., 2014). Melatonin also exerts a dual inhibitory effect on gonadotropin-releasing hormone-induced increases in intracellular Ca²⁺ and the mobilization of Ca²⁺ from the ER in the pituitary gland (Vanecek, 1999; Watanabe, 1999). These studies demonstrate that the maintenance of Ca²⁺ in the ER, including both inhibition of its mobilization and the enhancement of its uptake by the ER, is affected by melatonin.

Melatonin also regulates the Plasmodium falciparum cell cycle via a Ca²⁺-dependent pathway (Alves et al., 2011). The Ca²⁺ responses to melatonin and the uncaging of IP3 are mutually exclusive in infected red blood cells (Alves et al., 2011). Melatonin promotes the generation of IP3 and opens ER-localized IP3-sensitive Ca²⁺ channels in P. falciparum (Alves et al., 2011). These data not only support the assertion that melatonin is involved in Ca²⁺ signalling but also provide clues regarding the mechanism underlying this effect. In addition, melatonin inhibits intracellular Ca²⁺ overload-induced apoptosis. Treatments with a specific inhibitor of cytosolic Ca²⁺ reuptake, thapsigargin and/or a Ca²⁺-mobilizing agonist, N-formyl-methionyl-leucyl-phenylalanine, induce mitochondrial membrane depolarization, caspase activation, phosphatidylserine externalization and DNA fragmentation in leukocytes of both young and elderly volunteers; these effects, however, were far more evident in aged leukocytes (Espino et al., 2011). Notably, melatonin treatment substantially preserves the mitochondrial membrane potential, reverses caspase activation, reduces phosphatidylserine exposure and stops DNA fragmentation in leukocytes of both age groups, and these findings suggest that melatonin can delay Ca²⁺ overload-induced apoptosis in aged leukocytes, particularly in those of older individuals (Espino et al., 2011).

Melatonin and ER stress

By controlling ER stress, melatonin protects against multiple disorders, including DM, liver diseases, neurological disorders, reproductive system diseases, lung diseases and chemical poisoning (Espino et al., 2011; Zha et al., 2012; Zaouali et al., 2013; Carloni et al., 2014; Romero et al., 2014; Ali and Kim, 2015; Jeong and Park, 2015; Thakor et al., 2015) (Figure 2). Interestingly, melatonin can cooperate with ER stress to promote the apoptosis of cancer cells and inhibit ER stress to attenuate chemotherapy-associated side effects and chemoresistance.

Prospects and conclusion

Melatonin is an agent that regulates protein secretion by influencing various organelles. As a functional organelle, the ER is only one of the targets of melatonin, which also include the mitochondria and Golgi complex (Redondo et al., 2010; Koksal et al., 2012; Zhang et al., 2012). Melatonin treatment could play an important role in improving the histophysiological function of the pineal gland, including the mitochondria, RER and Golgi complex, as observed in goat kids subjected to stress due to premature weaning (Redondo et al., 2010). In addition to extensive SER tubules, increased mitochondria, elevated primary lysosomes and some marked openings of the bile canaliculi are also observed during methanol intoxication, and these effects can be prevented by melatonin (Koksal et al., 2012). In the testicular damage associated with hyperlipidaemia, vacuolar degeneration of the mitochondria and dilation of the ER are observed, and the numbers of mitochondria and lipid droplets are decreased significantly in Leydig and Sertoli cells; these effects were also reversed by melatonin (Zhang et al., 2012). However, further investigations are required to determine whether melatonin specifically influences the ER.

The ER not only participates in the secretion of melatonin but can also be regulated by melatonin in the pineal gland and other tissues, including the reproductive gland, parathyroid gland, pancreas, pituitary, breast cancer cells and leukocytes. Thus, melatonin increases its activation in the pineal gland but reduces this process in other targets. The abnormal ultrastructure and volume of the ER resulting from stress can be reversed by melatonin. Furthermore, melatonin maintains calcium homeostasis via the ER (Figure 3). Thus, in addition to being a well-known participant of melatonin production and secretion, the ER is also the intracellular regulatory target of melatonin, but the underlying mechanisms require further investigation. In particular, although the protective actions of melatonin are either mediated by its receptors or receptor-independent (Cardinali et al., 1997; Reiter et al., 2007; Miettunen and Raevuori, 2012), whether the regulation of ER and ER stress by melatonin is mediated by its receptors still needs further investigation through pharmacological methods (Alexander et al., 2015a).

After understanding the protective effects of melatonin, it is important to focus on its regulation. Huang et al. (2010) found that 12-h-light:12-h-dark (LD) and continuous-light (LL) regimes have opposite effects on the plasma level of melatonin. LL regimes are characterized by increased pineal clock and decreased liver clock genes compared with LD regimes. Under LL regimes, the clock and nuclear transcription factor mRNAs, but not pineal ER stress-related genes, do not show any daily variations in the organs studied (Huang et al., 2010). Thus, the daily routine might play an important role in maintaining the plasma level of melatonin, normal levels of expression of clock and incidence of ER stress. Regulating the levels of melatonin and ER stress by controlling the day length might serve as a basic contributor to health. Sleep deprivation is one reason for decreased melatonin, and the systemic administration of melatonin appears to constitute a potential neuroprotective treatment against neuronal damage induced by sleep deprivation (Lan et al., 2001). The potential treatment options for restoring a normal sleep cycle include the use of melatonin or its analogues (Reiter et al., 1983; Reiter et al., 2013b) and cognitive behavioural therapy. In modern life, cognitive behavioural therapy might be useful for increasing the production of melatonin and ultimately for regulating ER stress and protecting against various disorders.

Melatonin is a type of indolamine that was initially found to be produced in the pineal gland and is now known to be also synthesized in a variety of other tissues. Rigorous investigations have found that the amount of melatonin produced by enterochromaffin-like cells in the gastrointestinal tract is approximately 400-fold greater than that normally produced by pinealocytes (Magierowski et al., 2013; Bertrand et al., 2014). Whether melatonin from tissues other than the pineal gland can be stimulated by any means will be determined by further research. As an endogenous compound, melatonin is safe and presents little evidence of toxicity, but additional investigations are required to determine whether it interacts with other transmitters or hormones.

Author contributions

D.W. and Y.Y. designed the manuscript. W.H., C.F. and S.J. collected the data for the review. W.H., S.D. and L.Y. wrote the paper. Z.M. and W.H. illustrated the manuscript.

Conflict of interest

The authors declare no conflicts of interest.