Involvement of the Orexinergic System in Cancer: Antitumor Strategies and Future Perspectives

Abstract

Peptides promote the mitogenesis and migration of tumor cells, and cancer cells overexpress peptide receptors. The involvement of the orexinergic system in cancer is reviewed here, including thirteen cancer types (e.g., adrenocortical adenoma, breast, colon, gastric, liver, neuroblastoma, pancreas, prostate). An upregulation of the orexinergic system has been reported in many tumors, and orexin receptors (OXRs) mediate a dual effect: apoptosis in some tumors and a proliferative action in others. OXR antagonists or agonists are potential antitumor agents against tumors expressing OXRs. The complexities of the biological processes associated with the orexigenic system are also described in the review, as they may provide the basis for the development of new therapies: OXR dimerization/oligomerization, epigenetic mechanisms controlling the orexinergic system, possible biomarkers of this system for tumor risk/prognosis, protective effects mediated by orexins against chemotherapeutic drugs, the combination therapy of OXR antagonists/agonists with radiotherapy or chemotherapy, and the anti-inflammatory effects mediated by orexins. Taking these data into account, future therapeutic applications as well as research lines to be developed are also mentioned and discussed. This knowledge will allow for the development of antitumor strategies in the future.

1. Introduction

Cancer is a crucial worldwide health problem due to the many deaths that this disease provokes despite the development of new anticancer therapies such as gene therapy, immunotherapy, and hormonotherapy, in addition to classical treatments such as chemotherapy, radiotherapy, and surgery [1,2,3]. An inverse comorbidity between neurodegenerative diseases (Alzheimer’s, Parkinson’s) and cancer has been suggested; thus, one disease can decrease the risk of the other, and it seems that this could be mediated, among others, by the orexinergic system [4]. This system is involved in the control of the sleep/waking cycle, and it has been reported that cancer and cancer-related inflammatory mechanisms are associated with fatigue and sleep disruption; moreover, sleep apnea is linked to cancer development [5,6,7], and a chronic sleep disruption promoted tumor growth/progression in experimental animal models [8]. A bidirectional relationship between cancer and sleep occurs, and circadian rhythm disorders are a risk factor for some cancers (e.g., breast, thyroid, squamous cell carcinoma, gastrointestinal) [9,10]. Thus, drugs targeting orexin receptors and administered for the treatment of sleep disorders could also be used against certain tumors, since these receptors mediate apoptosis in cancer cells [9,11]. It has been observed that orexins induce cellular apoptosis [12] and this means that this system (orexin peptides and their receptors) could be a useful therapeutic option for cancer treatment [13]. Several reviews have been published on the orexinergic system and its involvement in several pathologies, including cancer [1,14,15,16,17]. These reviews have been mainly focused on certain cancer types, such as colorectal cancer and pancreatic ductal adenocarcinoma. Here, our aim is to update the knowledge of the involvement of the orexinergic system in any cancer type.

2. The Orexinergic System: Precursor, Peptides, Receptors, Signaling Pathways, Physiological and Pathophysiological Actions

2.1. Precursor and Peptides

The orexinergic system includes orexins A and B and the orexin receptors type 1 (OXR1) and 2 (OXR2) [14], and integrated with other systems, such as leptin and ghrelin systems, controls the energy homeostasis [13], and is essential for the normal consolidation of sleep and waking [18]. Orexin A and orexin B (also known as hypocretin 1 and 2, respectively) are encoded by the same gene (hcrt) on chromosome 17 [1] and are constituted of two exons [16] being the enzymatic cleavage products of a single 130-residue precursor, prepro-orexin, and share a 46% amino acid identity [1,19,20]. Both peptides have an amidated C-terminal end and include two α-helices domains linked by a small flexible domain [15,21]. Orexin A includes residues 28–66 of the prepro-orexin (33-residue peptide) and shows two intramolecular disulfide bridges within the N-terminal that are not essential for its full activity [22]. The C-terminal domain (sequence 20 to 33) is essential for its activity, and the deletion of the central domain between residues 15 and 19 strongly reduced the peptide activity [22]. Orexin B is a linear 28-residue peptide, corresponding to prepro-orexin residues 69–97 [1,19,20], without disulfide bridges [1]. The C-terminal domain of orexin B is crucial for its activity, but the deletion of the first six residues of the peptide have no impact on it [22,23]. It has been reported that some single-nucleotide polymorphisms corresponding to missense mutation, mainly located in the C-terminal domain of both peptides, may have a negative impact on the peptide activity [24].

In an adult brain, in situ hybridization displayed, via cRNA probe, neurons containing prepro-orexin distributed in hypothalamic (lateral and posterior hypothalamic areas and perifornical nucleus) and subthalamic (zona incerta, subincertal, and subthalamic nuclei) areas [25,26]. In addition, neurons containing the peptides were detected by immunohistochemistry in the dorsal lateral hypothalamic area, and immunoreactive fibers were observed in the arcuate hypothalamic nucleus [25]. These neurons display extensive and wide projections to numerous regions of the central nervous system [25,26], and thus, orexins can regulate many physiological functions, although the heaviest projections are related to regions that control arousal and wakefulness [13]. The hypothalamic neurons producing orexins can be segregated into distinct populations (anatomically, genetically, and functionally) which are involved in the different functions regulated by orexins, such as reward, stress, emotions, feeding, sleep/wakefulness cycle, or arousal [27]. In addition, orexin has also been detected in the enteric nervous system and in the enteroendocrine gut cells [28], as well as in the reproductive tract, pituitary, adipocytes, thyroid, pancreas, adrenal gland, testis, and ovary [29].

2.2. Receptors

Orexins A and B exert their functions throughout receptors type 1 (OXR1) and 2 (OXR2), which are G-protein-coupled receptors with seven transmembrane domains [13,18,26] belonging to the large class A rhodopsin-like subfamily of G-protein-coupled receptors [30]. OXR1 is encoded by the hcrtr1 gene located on chromosome 1, and OXR2 is encoded by the hcrtr2 gene located on chromosome 6 [25,26]. The two receptors shared 64% of identity sequence [31] and can bind orexin A with the same affinity, and orexin B shows a better affinity for OXR2 than for OXR1 [3]. The identification of OXR2, but not OXR1, in non-mammals suggests that OXR2 could be the ancestral receptor form [21,32].

The two orexin receptors lead to signal transduction by activation and/or inhibition of different intracellular signaling pathways to obtain final cellular responses [33]. The binding of ligands to their respective receptors induces a structural conformational change and the activation of G proteins [34]. At least three G-protein families can couple to both orexin receptors: G_q, G_i/o, and G_s [18]. Like other receptors coupled to G proteins, orexin receptors (OXRs) may be engaged in dimerization/oligomerization processes, leading to homomeric and heteromeric complexes; these latter might affect receptor trafficking, signaling, or pharmacology [35,36]. For example, heterodimers between human OXR1 and the CB₁ cannabinoid receptor have been verified, but the physiological significance of these interactions has not been completely dilucidated [18].

OXRs did not show significant affinity for other peptides. The transmembrane domains 1, 3, and 5 and the amino terminus of the receptors account for the interaction with the orexin peptides, and the transmembrane domain 3 is critical for receptor interactions with small molecule antagonists [37]. OXRs and the families of G proteins coupled with them regulate non-selective cation channels, phospholipases, and adenylyl cyclase, as well as protein and lipid kinases, and plastic effects are observed in some cell types [18]. Both OXRs exhibit slow kinetics in their response to orexin binding [37] and are widely distributed throughout the central nervous system but show a more restricted expression in the peripheral nervous system. Although a certain degree of overlap has been observed, the two receptors display different distributions [13]. It is well known that OXRs regulate homeostatic processes [38,39,40,41,42]. The most significant effect of orexin upon cells is the depolarization of neurons leading to increasing excitability and firing rate [43,44,45,46], and this depolarization is produced by the activation of a non-selective cation current [18]. Although the mechanisms underlying OXR signaling in the central nervous system are mostly unknown, orexin responses in neurons include the modulation of presynaptic transmitter release and changes in synaptic plasticity [47,48,49,50]. In experimental animals it has been described that the absence of one or both OXRs affects the markers of cholinergic transmission [51,52]. In addition, the participation of the orexigenic system in human diseases (e.g., narcolepsy, obesity, drug addiction) has been reported [38,39,40,41].

It has been reported that the regulatory phosphorylation site (Ser-262) on the OXR1 did not affect its interaction with beta-arrestin1/2, G proteins or GRK (G-protein-coupled receptor kinase) 5 but abolished its interaction with GRK2 [53]. This means that Ser-262 is involved in the internalization of OXR1 and in promoting a GRK2-dependent biased signaling through orexin A. Moreover, the biased signaling between beta-arrestin/G protein is a promising strategy to improve drug efficacy [53]. It has been suggested that OXR1-selective antagonists might be potential anti-addiction drugs, but the role of OXR2 in drug seeking has not been established. The blockade of this latter receptor as pharmacological treatment for addiction should be considered cautiously because of the predominant physiological consequence of OXR2 in wakefulness regulation [13]. The determination of the structures of both OXRs allows for analyzing interactions between receptors and antagonists, with the aim of designing new antagonist molecules [54].

The activation of OXRs may also produce long-term plastic effects. For example, it has been reported that placentally produced orexins may contribute to the prenatal development of brown adipose tissue in mice via OXR1 [50,55]. It has been reported that orexin A and OXR1 are present in low amounts in blood, and they do not follow a circadian pattern [42]. The serum orexin A level was lower in normal patients than that found in patients with cancer cachexia [56], and a lower level of this peptide was accompanied by elevated expression of OXR2 in benign prostatic hyperplasia [57]. In addition, the epigenetic silencing of OXR2 was in association with endometrial cancer [58]. A decrease in food consumption and body weight is frequently observed in patients with cancer cachexia, as well as anorexia and fatigue [59]: the Japanese herbal medicine Ninjinyoeito (NYT, contains twelve herbal crude drugs) is prescribed to attenuate these symptoms, since it promotes hyperphagia. It has been reported that one of the tested NYT formulations without Citrus unshiu peel failed to activate OXR1; by contrast, Citrus unshiu peel activated this receptor, which was blocked with SB-674,042, a selective OXR1 antagonist [59]. Thus, Citrus unshiu peel, after activating the hypothalamic neurons expressing OXR1, augmented food intake; this suggests the potential use of NYT in cancer patients with anorexia. In addition, an abnormal expression of OXR1 was observed in human peripheral organs in pathologic conditions [14]; for instance, primary colorectal tumors express OXR1 [60]. This observation emphasizes the potential therapeutic importance of OXRs. More specifically, OXR activation has been suggested to be a promising treatment for chemotherapy-resistant carcinoma [14,60].

2.3. Signaling Pathways

The two major signal transduction pathways associated with ligand–receptor binding are the cAMP pathway (through the adenylyl cyclase effector) and the phosphatidylinositol signal pathway (through the phospholipase C effector) [61]. OXR signaling is highly diverse depending on the milieu in which they are operating, and different responses have been observed at different sites [18,62]. Orexins, via OXRs, activate phospholipase A2, C and D, diacylglycerol lipase, PI3K (phosphatidylinositol 3-kinase)-Akt (protein kinase B), adenylyl cyclase/cAMP, and MAPK (mitogen-activated protein kinases)-ERK (extracellular-signal-regulated kinases) 1/2 and JNK (jun amino-terminal kinases) signaling cascades [18,63]. The two OXRs may couple to several G-protein species; thus, although the Gq phospholipase C (PLC) pathway plays an important role in many cases [62], the idea that OXR1 couples only to Gq and OXR2 couples to Gq and G_i/o needs to be reconsidered [18], and no evidence shows significantly different signaling of the two OXR subtypes [62]. In general, OXRs coupled with Gq proteins stimulate intracellular calcium via PLC [13,18,62]. They are also capable of regulating adenylyl cyclase, but this action seems less prominent than coupling to the PLC and calcium cascades [18].

In general, changes in the structure of the receptor after ligands binding [37] trigger a protein kinase C (PKC)-mediated influx of calcium across the plasma membrane via L-type calcium channels [13]. The activation of the calcium channels is related to many other signaling pathways, including the activation of MAPK, especially the ERK p38, cAMP-response element binding protein (CREB), adenylyl cyclase, and PLC [13]. The main responses of activation of OXRs in neurons are the inhibition of K⁺ channels and the activation of the Na⁺/Ca²⁺ exchanger and non-selective cation channels of unknown identity, although the G protein implicated as the main signal transducer in these cells is still unknown [62]. Both OXRs strongly activate PLC [18], and intracellular calcium stores are released from the endoplasmic reticulum by a PLC-mediated pathway, producing sustained excitation of related neurons [64,65,66,67] and inducing cell responses mainly via Ca^2+- and diacylglycerol-mediated pathways [18]. Orexin B induced ERK1/2 activation through OXR2, predominantly through the Gq/PLC/PKC pathway [68], although this receptor can differentially couple to G_q, G_i/o, and G_s proteins in different tissues [69,70]. Anyway, PLC activation may be a very central OXR cascade, and the role of PLC might be to produce Ca²⁺ elevation or to elevate diacylglycerol for possible PKC activation [18]. Under certain circumstances, the Gq-PLC pathway may take the route to diacylglycerol lipase, leading to the production of the endocannabinoid 2-arachidonoyl glycerol and connecting orexins with endocannabinoid signaling [62]. This interaction between orexins and endocannabinoids has been reported, with indirect evidence, for neuronal circuits regulating antinociception and reward seeking, as well as an autocrine mechanism in orexigenic neurons [71]. Orexin activates reward pathways by upregulating the addiction-related neurotransmitters and receptor activity [13]. It has been reported that OXR1 signaling is a part of the dopamine-mediated reward circuit, and that the increased orexin signaling (application of orexin A or enhancement of OXR2) induced a glutamate-stimulated effect on alcohol consumption, an activation that promoted drug seeking via a PLC/PKC pathway [13]. Orexins increase the GH-releasing hormone that stimulated GH release through an L-type Ca²⁺ current and PKC-mediated signaling pathways [72,73]. In addition, other authors report that mechanisms involved in the central control of metabolism, such as AMPK, unfolded protein response, and endoplasmic reticulum stress, are related to the control of energy homeostasis by the orexinergic system [42].

Orexin signaling induces apoptosis in some cells, whereas in others, orexin signaling raises proliferative activity [13]. Persistent stimulation of OXRs induces programmed cell death [18]; moreover, the pro-apoptotic property of orexin signaling is highly cell type dependent, and this may be due to the pathways induced mainly by the activation of OXR1. Activation of this receptor drives apoptosis via the Gq protein but independent of classical Gαq activation of PLC [13]. The signal cascade was suggested to involve (in addition of Gq) Src or a related kinase, phosphorylation of OXR1 and the recruitment of the protein phosphatase SHP-2. It has been reported that p38 MAPK could be the carrier of the cell death response [18], and the neuroprotective actions of orexin A might be mediated by the PKC and PI3K signaling pathways [34]. On the other hand, the activation of OXR1 promotes cell proliferation in some cancer cell types (pancreatic PANC1 cancer cells) through inhibiting cell apoptosis. It has been reported that the Akt/mammalian target of rapamycin (mTOR) signaling pathway is involved in this mechanism, leading to cell proliferation by the inhibition of Bcl-2/caspase-9/c-myc-mediated apoptosis [42]. Figure 1 shows the main signaling pathways mediated by orexins.

2.4. Physiological and Pathophysiological Actions

The major biological action of orexins is the regulation of sleep/wakefulness state [74] but they are also involved in drug addiction, motivation, food consumption, homeostasis, hormone secretion, reproductive function, lipolysis, and blood pressure regulation [1,15]. In particular, orexinergic neurons perceive rapidly the body’s nutritional status and respond to metabolic signals; for instance, the hypothalamic expression of the orexin precursor is increased during fasting, and as a part of the survival strategy elevated orexin levels have been detected during food deprivation periods [75,76,77,78]. Moreover, the orexinergic system exerts neuroprotective, immunoregulatory and anti-inflammatory effects and has been involved in high fat diet-induced obesity, intestinal bowel diseases, neuroinflammation, multiple sclerosis, Alzheimer’s disease, and septic shock [34]. Orexins are also related to pathologies such as narcolepsy, metabolic syndrome, and cancer [15]. In this sense, the orexinergic system can trigger opposite mechanisms depending on the environment. Thus, it seems that the cellular signaling after the activation of OXRs is more complex than that of other receptors.

The orexigenic system is very sensitive to the action of certain drugs and is closely related to addiction-reward processes because it inhibits the activity of the insula, a brain region involved in these functions by acting on the reward circuit [13]. In addition, the role of orexins on the regulation of central cardiovascular mechanisms has been demonstrated since the action of orexin A on some cardiovascular centers of the brainstem leads to the inhibition of vagal activity and the increase of sympathetic activity to the heart, as well as the alteration of the excitability of central cardiovascular circuits [13]. It has been proposed that the actions of orexins on the regulation of blood pressure and heart rate could be mediated throughout α- or β-adrenoreceptors [13].

The orexinergic system also regulates endocrine axes [13] and may operate as putative neuroendocrine and autocrine/paracrine regulators of gonadal function [79]. The detection of OXRs in peripheral tissues suggests a paracrine action of orexins; in addition, the circulating levels of orexins in blood in healthy individuals is very low [34]. The abnormal expression of OXRs in some human pathologies may lead to new therapeutic approaches, and it has been suggested that orexins or the orexin neural circuitry represent potential targets for the treatment of multiple pathologies related to inflammation including multiple sclerosis, obesity, septic shock, or several types of cancer [1,14].

3. Orexins and Cancer

Orexins, acting at OXR1 or OXR2, induce apoptosis resulting in massive reduction in cell growth in various cancer lines [13]. The antitumor effect of the orexigenic system is mediated by a signaling pathway involving the presence of two immunoreceptor tyrosine-based inhibitory motifs in both receptor subtypes, and this mechanism could be the basis for the development of a possible therapeutic target [42]. The expression of OXRs has been demonstrated in many cancer types such as prostate, cholangiocarcinoma (biliary tract), neuroblastoma, esophagus, stomach, colorectal, liver, pancreatic ductal adenocarcinoma, pheochromocytoma, cortical adenoma, endometrial carcinoma and in lung and liver metastases from colon [1,14,15,58,80,81,82]. Moreover, a study has shown the expression of OXR2 in many human cancer types: small intestine (100% of the samples studied), maxilla (100%), mandible (100%), gallbladder (60%), larynx (53%), stomach (53%), pancreas (50%), salivary gland (47%), colon (35%), nasopharynx (33%), rectum (33%), liver (11%) and lung (11%), but OXR2 was not observed in other cancers (esophagus, tongue, bile duct, lower lip, tonsil, anus) [83]. Although in some of these cancers the number of patients studied was small and, hence a larger number of patients must be studied, the data show the widespread expression of OXRs in different human tumors and that these receptors could be antitumor therapeutic targets.

3.1. Acute Lymphoblastic Leukemia

Dexamethasone chemotherapy did not block the orexinergic signaling in children with acute lymphoblastic leukemia; the same finding was observed in rodents [84]. Thus, a high dose of dexamethasone did not disrupt the human orexin physiology and this system was well preserved (e.g., OXR expression and neural output). Plasma orexin levels were significantly increased in females that received chemoradiotherapy (as an acute lymphoblastic leukemia treatment) compared to those that did not receive this treatment [85].

3.2. Adrenocortical Adenoma

Orexin A, through OXR1, promoted the release of cortisol from scattered human adrenocortical cells; orexins A/B and prepro-orexin were not located in the normal adrenal cortex. In addition, prepro-orexin mRNA and orexin A, but not orexin B, were observed in cortisol-secreting adenomas (a benign neoplasm of the adrenal cortex); these adenomas expressed OXR1/OXR2 mRNAs which were upregulated compared with the levels observed in normal adrenal cortices [80]. Orexin A, but not orexin B, increased the basal cortical release from adrenocortical adenomas and normal cells and both orexins augmented normal/adrenocortical adenoma cell proliferation [80]. Another study reported a high level of OXR1 mRNA and a low level of OXR2 mRNA in the human adrenal medulla and cortex, whereas only OXR2 has been detected in benign secreting pheochromocytomas [81]. OXR1 and OXR2 have been detected in human normal glands, adrenocortical adenomas and pheochromocytomas [86]. OXR1 was expressed in the cortex (reticular, fasciculate and glomerulosa regions) and OXR2 in the medulla (norepinephrine and epinephrine cells) of normal glands, whereas adrenocortical adenomas (cortical tumor) expressed OXR1, but not OXR2, and pheochromocytomas (medullar tumor) only expressed OXR2 [86]. Orexins A and B also decreased basal pituitary adenylate cyclase activating polypeptide (PACAP)-induced tyrosine hydroxylase mRNA level and blocked the PACAP-induced cAMP augmentation level in the rat pheochromocytoma PC12 cell line; this means that the suppressive action on tyrosine hydroxylase mRNA was, at least in part, mediated by the cAMP/protein kinase A pathway [87].

Orexin A decreased OXR2 expression and increased cortisol release in the NCI-H295R cell line [88,89]. Akt signaling is involved in the survival and proliferation of NCI-H295R human adrenocortical cells mediated by orexin A [89]. The peptide also increased OXR1 protein/mRNA expression, cell proliferation, 3β-hydroxysteroid dehydrogenase protein/mRNA expression, Akt phosphorylation and cortisol synthesis in these cells [89]. Orexins A and B promoted the release of catecholamines and increased inositol trisphosphate synthesis (not cAMP production) from pheochromocytoma slices; the latter effect was inhibited with U-73122 (a PLC blocker) and the former with U-73122 or calphostin (a PKC antagonist), but it was not blocked with SQ-22536 (an adenylate cyclase inhibitor) or H-89 (a protein kinase A blocker) [81]. This means that orexins favor catecholamine release from human pheochromocytomas via OXR2 coupled to the PLC-PKC signaling cascade. Moreover, both orexins also blocked basal and PACAP-induced dopamine release from PC12 cells; OXR2 was strongly expressed in the rat adrenal gland but not OXR1, and OXR1/OXR2 were not expressed in PC12 cells [87]. However, both orexins decreased the level of tyrosine hydroxylase in PC12 cells, suggesting that orexins exert this action through non-orexin receptors [87].

3.3. Breast Cancer

Orexin plays an important role in sleep and metabolic abnormalities in a mouse experimental non-metastatic breast cancer model [90]. This model displayed an altered activity in the population of hypothalamic neurons containing orexin; however, using a dual OXR antagonist, metabolic abnormalities were rescued, and sleep quality enhanced in mice with tumors [90]. Moreover, tumor-induced increase in serum glucose was prevented after peripheral sympathetic denervation [90]. Thus, orexinergic hypothalamic neurons are involved in tumor-induced changes in metabolism.

3.4. Cervical Cancer

A high occurrence rate of cervical cancer occurs in Chinese Uyghur women; a study has been focused on OXR1 and OXR2 expressions in these women with cervical cancer and in patients with cervicitis [91]. The study showed that OXR2 was overexpressed in women with cervical cancer when compared with those showing cervicitis; however, no difference was found in both groups regarding OXR1 expression [91]. Moreover, the study showed that normal human placentas expressed OXR2 but not OXR1. The authors suggest that OXR2 expression could be a marker for the invasive capacity of cervical cancer cells.

3.5. Colon Cancer

All colon tumor samples studied expressed OXR1 but not OXR2 and, importantly, this was independent of the genetic alterations, grade state and location; OXR1 was also observed in dysplastic polyps which are precancer lesions leading adenocarcinomas [16,60]. OXR1 mRNA has been detected in many colon cancer cell lines (SW48, SW480, SW620, Colo205, LS174T, T84, LoVo, Caco-2, HT29-D4, HT29-FU, HT-29) but not in HCT-116 cells; in addition, orexins A and B promoted apoptosis in LS174T, T84, Colo205, LoVo, Caco-2, SW480, SW620 and SW48 cell lines [82]. Moreover, orexins promoted apoptosis and blocked tumor growth in experimental animals [23]. OXR1 mediated mitochondrial apoptosis in colon cancer Caco-2, LoVo and HT-29 cells, a mechanism that was blocked using NSC-87877 (a SHP-2 inhibitor) or PD-169,316 (a p38 inhibitor). Orexins A and B favored apoptosis but did not alter cell proliferation in human HT29-D4 colon cancer cells, in which OXR1 but not OXR2 was expressed [92]. In HT29-D4 colon cancer cells, OXR1 mediated orexin-induced calcium transients and apoptotic mechanisms have been detected since caspases 3 and 7 were activated, cytochrome c was released into the cytoplasm, chromatin condensation/DNA fragmentation were observed, and cell shape was changed [92]. However, these mechanisms were not observed in normal colonic epithelial cells and orexins promoted a pro-apoptotic effect in LoVo, SW480 and Caco-2 colon tumor cells expressing OXR1, but this was not observed in normal colonic epithelial cells [92]. All the primary colorectal tumors and hepatic metastasis studied showed OXR1 expression, whereas adjacent normal colonocytes/hepatocytes were devoid of such expression [60]. Moreover, human colon cancer cell lines established from metastases or primary tumors also expressed OXR1 mRNA and suffered apoptotic mechanisms (through the activation of caspase 3) after treatment with orexin A [60]. Orexin A also showed an antitumor effect in xenografted human colon cancer cells by decreasing tumor size [60]. Normal colonic epithelial cells do not express OXR1, but this receptor is aberrantly expressed in colorectal tumors and after local (lymph node) or distant (lung, liver) metastasis [93]. Orexins promoted apoptosis not only in colorectal tumor cells but also in those cells resistant to the treatment with the chemotherapy drug 5-fluorouracil. The administration of orexin A decreased tumor volume in in vivo experiments, slowed the tumor grow and reverted the development of well-established tumors (xenografted LoVo or HT-29 colon cancer cells) by promoting apoptotic mechanisms [60,93]. Importantly, endogenous orexin A did not exert an intrinsic effect on tumor development; thus, the orexin A circulating level was very low and neither orexin A nor orexin B were observed in the tumor environment. Previous findings suggest that the orexinergic system (OXR1, orexins) is involved in metastasis and chemoresistance and hence OXR1 is a promising antitumor target and orexins could be used as antitumor agents against tumor cells showing chemoresistance.

OXR1, via G(q) protein but not via Gα(q) activation of PLC, promoted apoptosis; thus, the βγ dimer of G(q) stimulated src-tyrosine kinase leading to the phosphorylation of the immunoreceptor tyrosine-based inhibitory motif in OXR1 and subsequent recruitment by this receptor of the phosphotyrosine phosphatase SHP-2, which is then activated, followed by the activation of p38 mitogen-stress protein kinase and the entry of Bax protein into the mitochondria [17,93,94,95]. The immunoreceptor tyrosine-based switch motif presented in OXR1 is required for OXR1-medited apoptosis [17,95]. An OXR1 structural model has been developed and it has been shown that the spatial localization of phosphotyrosines in the immunoreceptor tyrosine-based switch/immunoreceptor tyrosine-based inhibitory motifs in OXR1 was compatible with their interaction with the two SH₂ domains of SHP-2 [95]. The C-terminus of orexin B favored the induction of OXR1-mediated apoptosis in colon tumor cells, and this means that orexin B interacted with some residues belonging to the OXR1 [23]. Thus, OXR1 residues (K (120, 321), P (123), Y (124), N (318), F (340), T(341), H (344), W (345)) are crucial for orexin B recognition and orexin B/OXR1-induced apoptosis; alanine substitution of orexin B residues (L (11, 15, 26), A (22), G (24), I (25), M (28)) altered the binding affinity of the peptide, and substitution of Q (16), A(17), S (18), N (20) and T (27) residues blocked apoptosis in OXR1 expressing cells [23]. This study shows the importance of knowing in-depth the structure-function relationships between orexins and their receptors for the developing of new antitumor compounds.

It has been observed that orexin A inhibited exosomal PD-L1 expression in colon cancer and blocked the JAK2/STAT3 signaling pathway promoting T cell activity [96]. Moreover, cholecystokinin A receptor and OXR1 heterodimerize in human HT-29 colon cancer and HEK293 cells; the stimulation of OXR1/cholecystokinin A receptor heterodimers with both orexin A and cholecystokinin decreased the migration and activation of Gαq/Gαi2/Gα12/Gα13 in HT-29 cells when compared to that observed after stimulation with orexin A or cholecystokinin alone; however, the interaction of both receptors with β-arrestin was not altered [97]. This is important, since OXR1/cholecystokinin heterodimerization mediates antimigratory and antimetastatic effects.

Orexin A, via the ERK signaling pathway, promoted autophagy in human HCT-116 colon cancer cells, inhibited the viability of these cells, and upregulated ERK phosphorylation; both apoptosis and autophagy (an accumulation of punctuate microtubule-associated protein-1 light chain 3 occurs) were activated by the peptide [98]. However, U0126 (an ERK inhibitor) or chloroquine (an autophagy inhibitor) decreased autophagy and blocked ERK phosphorylation in colon cancer cells [98]. Because autophagy can promote tumorigenesis by promoting tumor cell proliferation, the blockade of autophagy could be a useful antitumor strategy.

Finally, ulcerative colitis, an inflammatory bowel disease, affects both colon and rectum mucosa and is a risk factor for colon cancer development [99]. A high OXR1 expression has been reported in colon cancer cells as well as in the colonic epithelium of most patients suffering from ulcerative colitis, but this was not observed in the normal mucosa [99]. The administration of orexin A exerted an anti-inflammatory effect in colitis experimental murine models, and this was due to a decrease in the level of pro-inflammatory cytokines in immune cells (e.g., T cells); moreover, orexin A blocked the release of cytokines (e.g., interleukins 1α, 1β, 6, 8, and 17, tumor necrosis factor α) and the canonical NF-κB activation in immune cells [99]. The anti-inflammatory effect promoted by orexin A was also due to PLC activation, which promoted the release of intracellular calcium, and NF-κB inactivation, which regulated the release of inflammatory cytokines [100]. Thus, orexin A could be a promising treatment against ulcerative colitis and other intestinal bowel diseases such as Crohn’s disease in which OXR1 is expressed [16] and, in addition to the anti-inflammatory effect mediated by orexin A, the peptide could also prevent the development of tumors by promoting apoptosis in tumor cells.

3.6. Endometrial Cancer

OXR2 expression has been studied in human endometrial endometrioid carcinoma (EEC) and in normal endometrium; OXR2 was detected in normal endometrial epithelia but frequently lacking in EEC, and the latter observation was related to OXR2 hypermethylation [58]. This suggests that an OXR2 epigenetic silencing occurs in EEC. Moreover, this hypermethylation was correlated with a weak OXR2 expression in endometrial cell lines (ECC-1, MFE-280, Ishikawa), in which neither orexin A nor orexin B changed the proliferative activity.

3.7. Gastric Cancer

OXR1, but not OXR2, has been detected in SGC-7901 and BGC-823 gastric cancer cells and it has been reported that orexin A, through the Akt signaling pathway, increased OXR1 expression, decreased caspase 3 apoptotic activity, and augmented both viability and cell proliferation in the latter cell type [82,101]. Orexin A, through the ERK1/2 signaling pathway, augmented SGC-7901 cell viability/proliferation and exerted an anti-apoptotic effect [102]. The orexinergic system has been studied in patients with gastric cancer and in those with chronic atrophic gastritis (CAG) [103]. Orexin A was upregulated in cancer patients compared with normal and CAG groups, and its expression was higher in CAG patients than in normal individuals [103]. On the contrary, OXR1/OXR2 expressions were downregulated in patients with gastric cancer compared with the normal and CAG groups [103]. The expression of both receptors was higher in CAG individuals than in cancer patients and tumor tissues were depleted of prepro-orexin compared to that found in tissues from normal or CAG individuals [103]. Finally, inflammation was higher in gastric cancer tissues than in those taken from normal and CAG individuals; moreover, the expression of orexin A was not related to differential grades, age, and gender [103]. Thus, it seems that orexin A interacts with OXR1 and OXR2 and activates prepro-orexin, leading to inflammation in gastric tumor tissues.

OXR1 overexpression, via the activation of MAPK-dependent caspases and src-tyrosine, promoted tumor regression in gastric adenocarcinoma [104]. However, another study has shown that orexin A, through OXR1 and the Akt signaling pathway, blocked apoptosis in BGC-823 gastric cancer cells [101]. The peptide increased the expression of OXR1 protein in these cells, improving their viability and proliferation and protecting them from apoptosis [101]. Moreover, orexin A increased the phosphorylated Akt protein and decreased the pro-apoptotic activity of caspase 3 [101]. These actions, mediated by orexin A, were blocked with PF-04691502 (an Akt antagonist) or SB-334,867 (an OXR1 antagonist). Orexin A, through ERK1/2, upregulates OXR1 expression and increases the proliferation of human SGC-7901 gastric cancer cells [102]. Thus, the peptide increased cell proliferation and viability, protected cells from apoptosis, decreased the pro-apoptotic activity of caspase 9, and increased ERK1/2 phosphorylation [102]. These actions were inhibited, to a certain extent, with the administration of U0126 (an ERK1/2 antagonist) or SB-334,867 (an OXR1 antagonist). Finally, GBC-823 gastric tumor cells expressed OXR1 that, after activation, blocked apoptosis through the Akt signaling cascade [101].

3.8. Glioma

It has been reported that orexin A, through a caspase-dependent mechanism, decreased the viability of rat C6 glioma cells [105]. C6 cells expressed OXR1 and OXR2, which did not mediate the proliferation of these cells, but orexin A (at a high concentration) decreased cell viability by increasing the activity of caspase 3, leading to the death of C6 cells [105]. However, orexin A did not affect cell viability at a physiological level (at a lower concentration) [105]. Moreover, compared with that found in normal individuals, a low level of orexin A has been observed in the cerebrospinal fluid of a patient with a hypothalamic tumor [106].

3.9. Head and Neck Cancer

The methylation status of the orexin precursor gene promoter has been studied in head and neck squamous cell carcinoma; it has been related with cell survival and recurrence and it has been suggested that it be used as an epigenetic marker for risk/prognosis in this cancer type [107].

3.10. Hepatocellular Carcinoma

It has been described that the hepatocellular Hep3B cell line expressed OXR1 mRNA, but not OXR2 mRNA, and that orexin A increased glucose uptake in these cells by promoting pyruvate shunting into the tricarboxylic acid cycle and oxidative phosphorylation, thus decreasing glycolysis [108]. Human hepatocellular samples also expressed OXR1 but not OXR2 [108]. Stress and depression increased the risk of developing liver cancer via the epigenetic downregulation of orexin [109]. The tumor load and the number of liver tumors in depressed rats exposed to chronic unpredictable mild stress were higher than that found in animals without a chronic stimulus [109]. The orexin gene was downregulated in the hippocampi of rats with chronic unpredictable mild stress compared with non-stressed animals, and it has been found that the promoter for orexin was hypermethylated and the orexin mRNA expression downregulated in those animals with depression [109]. Thus, it seems that chronic psychological stressors, through an epigenetic orexin downregulation, play an important role in cancer progression and that this downregulation could be the link between depression and cancer.

The pharmacogenetic activation of hypothalamic neurons containing orexin or the intracerebroventricular administration of orexin A activated the hepatic mTOR-sXbp1 signaling cascade and attenuated the endoplasmic reticulum stress and inflammation in orexin-deficient mice fed with a high-fat diet; these actions were inhibited with autonomic ganglionic blockers [110]. Moreover, orexin also prevented associated hepatocellular carcinoma under obesity [110]. Orexin A regulated glucose metabolism through HIF-1α-independent and -dependent processes in HepG2 human hepatocellular carcinoma cells; in fact, orexin A favored a glucose flux into the mitochondrial oxidative metabolism rather than glycolysis [111]. This study also showed OXR1 in the nuclei of cancer cells, and orexin A favored both mRNA and protein expression of HIF-1α and its nuclear accumulation; this expression was related to the activation of the PI3K/Akt/mTOR signaling pathway [111]. Orexin A promoted glucose uptake, glucose transporter 1 expression, and ATP generation acting via the HIF-1α pathway through OXR1 [111]. Orexin A blocked, independently of HIF-1α, lactate production and lactate dehydrogenase/pyruvate dehydrogenase kinase 1 expression and favored pyruvate dehydrogenase B expression and mitochondrial pyruvate dehydrogenase activity [111]. A similar study was performed in the human Hep3B hepatocellular carcinoma cell line; orexin A, through PI3K/Akt/mTOR-dependent and -independent mechanisms, also regulated glucose metabolism in these cells [108]. Thus, orexin A increased glucose uptake and glucose transporter 1 expression, which was associated with the activation of the PI3K/Akt/mTOR pathway, but the peptide also increased the expression of pyruvate dehydrogenase B/mitochondrial pyruvate dehydrogenase activity and decreased lactate dehydrogenase/pyruvate dehydrogenase kinase 1 mRNA levels and lactate generation independently of the PI3K/Akt/mTOR pathway [108]. Thus, orexin A regulates the cellular metabolism towards mitochondrial glucose oxidation rather than glycolysis.

3.11. Neuroblastoma

Orexin A protein/mRNA has been detected in neuroblastoma and ganglioneuroblastoma [112]. Orexins A/B promoted growth suppression and apoptotic mechanisms in the neuroblastoma SK-N-MC cell line, which expresses OXR1, and previous effects were also promoted by orexins in Chinese hamster ovary cells transfected with OXR1 cDNA [92]. Cell death occurred via p38 MAPK activation, but independently of p53 and caspase activation in the latter cells [113]. Orexin A, through PI3K and PCK signaling cascades, protected human SH-SY5Y neuroblastoma cells against 6-hydroxydopamine, which decreased cell viability and increased mitochondrial membrane potential as well as the levels of COX-2 and intracellular Ca⁺⁺ [114]. It has been reported that orexin A prevented the effects mediated by 6- hydroxydopamine and also exerted anti-apoptotic and antioxidant actions and decreased biochemical markers of cell death, whereas the OXR1 antagonist SB-3344867 or PKC/PI3K inhibitors suppressed the beneficial actions mediated by orexin A on neuroblastoma cells [114,115]. Orexin A decreased the levels of intracellular reactive oxygen species and cytochrome c, the Bax/Bcl-2 ratio, and caspase 3 activity; induced the expression of HIF-1α and activated the downstream targets of this factor (e.g., erythropoietin, vascular endothelial growth factor) in human SH-SY5Y neuroblastoma cells [115,116].

Interfering with canonical transient receptor potential 3/6 channels, through the sodium–calcium exchanger, disrupted OXR1 signaling in human IMR32 neuroblastoma cells [117]. This means that the canonical transient receptor potential 3/6-sodium calcium exchanger channel interaction plays an important role in receptor-mediated calcium influx. Moreover, OXR1 controlled, via diacylglycerol-activated channels, the entry of calcium into IMR-32 neuroblastoma cells [118].

3.12. Pancreatic Cancer

Orexin A, through OXR1, promotes the proliferation of pancreatic cancer PANC1 cells and protects these cells from apoptosis [119]. The peptide activates the Akt/mTOR signaling pathway, which favors cancer cell proliferation by blocking c-myc/caspase 9/Bcl-2-mediated apoptosis [119]. However, the blockade of OXR1 promoted apoptosis by regulating c-myc, caspase 9, and Bcl-2 levels in pancreatic cancer cells. Thus, OXR1 is a potential antitumor target.

Pancreatic ductal adenocarcinoma (PDAC) expressed OXR1, but not OXR2, and this expression was independent of age, gender, and tumor grade; OXR1 was also observed in pancreatic intraepithelial neoplasia, a PDAC pre-neoplastic lesion [120]. Orexin A, through the SHP2 signaling cascade, promoted mitochondrial apoptosis in AsPC-1 cells (a PDAC cell line) and in PDAC cells isolated from cancer patients [120]. Orexin A also decreased tumor volume (due to apoptosis) in those animals xenografted with AsPC-1 or PDAC cells, but the peptide did not exert an antitumor action against HPAF-II cells in which OXR1 was absent [120]. PDAC often develops chemoresistance to treatments with abraxane (Nab-paclitaxel) or gemcitabine. Orexin A, through OXR1, promoted an antitumor action by activating mitochondrial pro-apoptotic processes [19]. This peptide blocked the growth of pancreatic cancer cells, inhibited tumor volume in preclinical experimental models, and its co-administration with abraxane or gemcitabine was additive regarding the blockade of cancer cells’ growth and tumor development [19]. Orexin A promoted a strong blockade against tumor development in chemoresistant xenograft tumors with abraxane or gemcitabine and promoted the death of pancreatic cancer cells resistant to these compounds [19]. Moreover, ex vivo, in vivo, and in vitro studies have demonstrated that orexin A and almorexant (ACT-078573, a dual OXR1 and OXR2 antagonist) promoted apoptosis in PDAC cells via OXR1 [120]. Most PDAC tissues showed OXR1, but this was not observed in adjacent normal exocrine pancreas; OXR1 expression was also observed in pre-cancerous lesions and orexin A or almorexant promoted apoptosis in AsPC-1 cells expressing OXR1, which was inhibited with NSC-87877, a tyrosine phosphatase SHP2 inhibitor [120]. Compared to control samples, a higher number of cancer cells expressing caspase 3 was observed after treatment with orexin A, but cell proliferation was not changed and, in in vivo experiments, orexin A or almorexant significantly decreased tumor growth [120]. The data suggest the potential use of both orexin A and almorexant to treat PDAC. Finally, orexins promoted growth inhibition and apoptosis in the rat pancreatic acinar AR [119,120] cell line, which expresses OXR2 but not OXR1, and in Chinese hamster ovary cells transfected with OXR2 cDNA [12].

Orexin A, via OXR1, increased the release of insulin by rat insulinoma INS-1 cells by activating Akt and its downstream target mTOR [121]. These mechanisms were blocked with SB-674,042, a selective OXR1 antagonist, or with PF-04691502, a PI3k/mTOR antagonist. This study also showed that orexin A increased mTOR phosphorylation and OXR1 expression in insulinoma cells [121].

3.13. Prostate Cancer

The expression of OXRs has been detected in patients with prostate cancer (90%), with benign prostate hyperplasia (53.3%), or with chronic prostatitis (26.7%) [122]. Only five patients belonging to the first group showed a strong staining, whereas the rest of the patients showed a weak immunoreactivity. In addition, the distribution of OXRs was more widespread in patients with prostate cancer than that found in those with benign prostate hyperplasia or with chronic prostatitis [122]. The authors suggest that OXR expression can be an indicator of poor prognosis in prostate cancer. No expression of OXR/prepro-orexin genes has been reported in human normal (PrSmC, PrSc, PrEC) and prostate cancer (PC3, LNCaP, Du145) cells [123]. However, the presence of OXR1 and prepro-orexin has been described in human normal and hyperplastic prostate tissues, as well as the expression of orexin A, OXR1, and prepro-orexin in human prostate cancer tissues [124]. Moreover, orexin A upregulated the expression of OXR1 and decreased the survival of human LNCaP cells (androgen-dependent prostate carcinoma cells) [124]. The peptide also prevented the testosterone-induced nuclear translocation of the androgen receptor in LNCaP cells; this was counteracted with SB-408,124, an OXR1 antagonist. This is important since orexin A, via OXR1, interferes with the activity of the androgen receptor which controls prostate cancer onset and progression [124].

OXR2 mRNA/protein expressions, but not OXR1/prepro-orexin, have been observed in normal prostate glands (apical region of epithelial cells, smooth muscle cells) and benign prostatic hyperplasia [57]. An OXR2 upregulation associated with a low serum orexin A level was found in benign prostatic hyperplasia; the serum orexin A level was not changed in prostate cancer compared to control samples, whereas the level of orexin B was similar in all groups studied [57]. The results suggest that orexin A/OXR2 are involved in the maintenance and/or pathogenesis of benign prostatic hyperplasia.

OXR1 mediated apoptosis, and it has been reported that this receptor was overexpressed in advanced prostate cancer, but its expression was lower in low-grade prostate tumors [125]. The number of cells expressing this receptor augmented according to the prostate cancer grade, whereas OXR2 was observed in a few malignant cells in high-grade prostate cancer [125]. Moreover, orexin A was not observed in normal prostate epithelium and, in addition, the peptide was not observed in tumor tissues [125,126]. Thus, it seems that normal/cancer cells located inside and surrounding the tumor do not express orexin A. OXR1 was found in androgen-dependent LNCaP cells and in androgen-independent Du145 cells; the acquisition of a neuroendocrine phenotype by the latter cells has been related to OXR1 overexpression [125]. Both orexins increased non-transformed Du145 cell growth but, when these cells suffered neuroendocrine transformation, apoptosis was observed when they were treated with orexin A [125]. Thus, it seems that orexin could exert a beneficial action against prostate cancer at a certain period. Orexin A, via OXR1, blocked cell growth in LNCaP and Du145 cells, and the peptide decreased the tumor volume of xenografted Du145 cells [124,125,126,127].

3.14. Circadian Rhythm and Cancer

Many of the metabolic processes dysregulated in cancer (inflammation, increased oxidative stress, etc.) are under the regulation of the circadian rhythm. This phenomenon is the basis of the bidirectional relationship that has been found between sleep disorders and certain types of cancer [9]. Furthermore, this influence of the circadian rhythm on cellular processes supports studies describing that the time of day at which anticancer treatment (radio or chemotherapy) is administered is important for optimizing the effects of such treatment, a phenomenon known as chronotherapy [128]. Some studies in animal models have shown that circadian sensitivity to the cellular metabolism of pharmaceutical compounds influences the effect of treatments, which improves their efficacy and minimizes adverse effects if the optimal time for their administration is considered [10]. Although in many cases, a clear relationship between sleep disorders and an increase in the incidence of cancer cannot be assured, the studies carried out so far in humans provide promising results as to the most favorable time of day for the application of chemotherapy [128]. For this purpose, it is essential to know in detail the circadian rhythms of the different cell types.

A bidirectional relationship has been found between the incidence of some types of cancer and various sleep disorders (insomnia, hypersomnia, etc.) [9]. The expression of several genes, some of them related to cancer, is also subject to this circadian rhythm, coordinated in the hypothalamus. Orexigenic neurons are located in this brain region and are involved in, among other functions, the regulation of sleep–wake cycles [18]. In fact, it has been proposed that the orexigenic system may be a promising therapeutic target in some types of tumors [9] (see Section 4).

Table 1. Involvement of the orexinergic system in cancer.

Tumor	Orexigenic System	References
Acute lymphoblastic leukemia	Plasma orexin levels increased in patients receiving chemoradiotherapy. Dexamethasone chemotherapy did not block the orexinergic signaling.	[84,85]
Adrenocortical adenoma	Prepro-orexin mRNA, orexin A (not orexin B) and OXR1/OXR2 upregulation. Orexin A, but not orexin B, augmented normal/adrenocortical adenoma cell proliferation. NCI-H295R cell line: orexin A decreased OXR2 expression and increased OXR1 expression, cell proliferation, and cortisol release. Pheochromocytoma: OXR1 expression but not OXR2. Orexin A/B, via OXR2, favor catecholamine release from pheochromocytomas. PC12 cells: no OXR1/OXR2 expression, but orexin A/B decreased the level of tyrosine hydroxylase; effect exerted via non-orexin receptors.	[80,81,86,87,88,89]
Breast cancer	Orexinergic hypothalamic neurons involved in tumor-induced changes in metabolism.	[90]
Cervical cancer	OXR2 overexpression.	[91]
Colon cancer	OXR1 expression. Orexins promoted apoptosis and decreased tumor volume. Apoptosis observed in tumor cells resistant to chemotherapy drugs. Orexin A promoted autophagy in cancer cells. High OXR1 expression in colon cancer cells and colonic epithelium of patients with ulcerative colitis.	[60,82,93,98,99]
Endometrial cancer	OXR2, frequently lacking, related to OXR2 hypermethylation.	[58]
Gastric cancer	OXR1 expression. Orexin A increased OXR1 expression, blocked apoptosis, and augmented viability and cell proliferation. OXR1 overexpression, via MAPK-dependent caspases and src-tyrosine, promoted tumor regression in gastric adenocarcinoma. Orexin A upregulation, OXR1/OXR2 expressions downregulated and high inflammation.	[82,101,102,103,104]
Glioma	Orexin A, via a caspase-dependent mechanism, decreased tumor cell viability. C6 cells express OXR1/OXR2, which do not mediate cell proliferation.	[105]
Head and neck cancer	Methylation status of the orexin precursor gene promoter related to cell survival and recurrence. An epigenetic marker for risk/prognosis.	[107]
Hepatocellular carcinoma	OXR1 mRNA but not OXR2 mRNA expression. OXR1 located in the nuclei of cancer cells. Orexin A favored HIF-1α expression and its nuclear accumulation. Orexin A increased glucose uptake and glucose transporter 1 expression in tumor cells. Orexin A regulated cellular metabolism towards mitochondrial glucose oxidation rather than glycolysis. Stress and depression increased, via orexin downregulation, the risk of tumor development.	[108,109,111]
Neuroblastoma	Orexin A expression. Orexins A/B promoted growth suppression and apoptotic mechanisms. Orexin A exerted an anti-apoptotic and antioxidant actions and decreased biochemical markers of cell death. Orexin A decreased the levels of intracellular reactive oxygen species and cytochrome c, Bax/Bcl-2 ration, and caspase 3 activity; HIF-1α induction. OXR1 regulated, via diacylglycerol-activated channels, the entry of calcium into tumor cells.	[92,112,114,115,116,118]
Pancreatic cancer	Orexin A, via OXR1, promoted tumor cell proliferation and protected these cells from apoptosis. OXR1 blockade promoted apoptosis. Orexin A, via the SHP2 signaling cascade, promoted mitochondrial apoptosis in AsPC-1 cells. Orexin A decreased tumor volume by apoptosis. Orexin A blocked tumor development in tumors chemoresistant to abraxane or gemcitabine. Orexins blocked tumor growth and promoted apoptosis in the AR42J pancreatic acinar cell line expressing OXR2.	[19,119,120]
Prostate cancer	OXRs expression: more widespread in prostate cancer than in benign prostate hyperplasia or chronic prostatitis. OXR expression indicator of poor prognosis. OXR1 promoted apoptosis and overexpressed in advanced tumors, but its expression was lower in low-grade tumors. Orexin A/OXR2 involved in maintenance and/or pathogenesis of benign prostatic hyperplasia. Orexin A upregulated OXR1 expression and decreased LNCaP cell survival. Orexin A, via OXR1, interfered with androgen receptor activity, which controls cancer onset and progression. Orexins increased non-transformed DU145cell growth, but, after neuroendocrine transformation, apoptosis was observed after treatment with orexin A. Orexin A, via OXR1, blocked cell growth in LNCaP and Du145 cells, and the peptide decreased the tumor volume of xenografted Du145 cells. Orexin A increased OXR1 expression in LNCaP cells.	[57,122,124,125,127]

and Figure 2 summarize the main mechanisms mediated by the orexinergic system in tumor cells.

4. Therapeutic Strategies and Future Directions

Despite all the progress that has been made in the fight against cancer, this disease continues to be a major health problem. According to data provided by the GLOBOCAN database (https://gco.iarc.fr/ accessed on 27 May 2023) [129], it is estimated that in 2020 there were more than 19 million new cases of cancer, of which almost 50% were found in Asia and almost 23% in Europe. A lesser percentage was detected in North America (13.3%), Latin America, the Caribbean (7.6%), Africa (5.7%), and Oceania (1.3%). Concerning incidence and mortality rates, the highest incidence rate corresponded to breast cancer (47.8%), with a mortality rate of 13.6%. The highest mortality rate was detected in lung cancer (18%), although its incidence was 22.4%, much lower than breast cancer. These data reflect that, although rates are continually improving, there is still much work to be done in terms of early detection and treatment. Knowledge of the orexinergic system may provide an additional resource as a potential therapeutic target in the fight against cancer.

4.1. OXR Agonists and OXR Antagonists

Signaling mediated through OXRs promoted a dual effect: orexins, through OXRs, induced apoptosis in some tumor cells, but in others promoted a proliferative activity. This means that OXR antagonists or agonists are potential antitumor agents against tumors expressing OXRs, since OXR antagonists block the proliferative action mediated by orexins and, in addition, promote the apoptosis of tumor cells. In some tumors, OXR agonists favor the apoptosis of cancer cells. Thus, the use of these compounds is a promising antitumor research line that must be fully developed.

Selective OXR2 agonists such as TAK-925, TAK-994, and YNT-185 have been developed, but not selective OXR1 agonists [1,130,131,132]. OXR antagonists are classified into dual OXR antagonists (DORAs (e.g., almorexant, suvorexant, daridoxerant, emborexant)), which similarly interact with OXR1 and OXR2, and single orexin receptor antagonists (SORAs) which are in turn classified into SORA1s (selective OXR1 antagonists (e.g., SB-334,867)) and SORA2s (selective OXR2 antagonists (e.g., JNJ-42847922 (seltorexant)) [1,133,134,135,136,137,138]. Currently, suvorexant, daridoxerant, and emborexant are approved by the U.S. Food & Drug Administration (FDA) for the treatment of insomnia [1]. Suvorexant and almorexant induced apoptosis in AsPC-1 cells and blocked the release of calcium promoted by orexin A; almorexant also decreased tumor volume (AsPC-1 xenografted cells) [120]. It has been suggested that almorexant acts as a partial agonist and antagonist toward Gq proteins by activating src kinases, through β/λ subunits, and by inactivating αq subunits [1]. Another important finding is that orexin A, via OXR1, promoted the proliferation of pancreatic cancer cells and protected them from apoptosis, but the blockade of OXR1 promoted apoptosis in these cancer cells [119]. This emerging antitumor strategy must be studied in depth and entirely developed. Previous data show the important role that OXR antagonists could play as antitumor agents; however, the antitumor effect exerted by more OXR agonists and antagonists must be tested in the future against more tumor types, since, for example, only a few OXR antagonists have been currently tested against some tumors. This is a research line that must be fully developed to establish new and promising antitumor strategies. But it is also crucial to develop basic research lines focused on the expression of the orexinergic system in many cancer types, because only in a few tumors has this system been studied and, in some of them, the reported data are very fragmentary. Thus, in vitro and in vivo systematic studies on the involvement of the orexinergic system in cancer must be developed in many tumors (e.g., orexin A and B expression, OXR1 and OXR2 expression, tumor progression, or blockade by orexins). Much basic research knowledge is lacking. Moreover, the molecular interactions between OXRs and OXR antagonists or agonists must be understood in depth; this knowledge will allow us to design and develop new and more potent and specific antitumor agents. In this sense, an allosteric potentiation of OXR has been reported, and that allosteric compounds bind to OXR, on a place different from that of natural ligands. This allosteric potentiation does not stimulate the function of OXR independently but increases the response of OXR to ligands [139]. It seems that these potentiators promote natural cycles of receptor activation but do not drive chronic receptor activation [140,141].

4.2. OXRs Dimerization/Oligomerization

OXRs suffer dimerization and oligomerization processes, which could explain the antitumor or tumor actions mediated by orexins, since these processes could influence pharmacology, signal pathways, and receptor trafficking. This is important, since cholecystokinin A receptor/OXR1 heterodimerize in tumor cells; the stimulation of heterodimers with both orexin A and cholecystokinin decreased the migration of tumor cells; this means that OXR1/cholecystokinin heterodimerization is involved in antimigratory and antimetastatic actions. This is also an important line of research for developing future antimigratory and antimetastatic strategies. Moreover, OXR1/CB1 cannabinoid receptor heterodimers have been reported, but the physiological significance of this finding is currently unknown. This must be elucidated.

4.3. Epigenetic Mechanisms

Another important line of research is to study the epigenetic mechanisms controlling the orexinergic system, since endometrial cancer development has been associated with the epigenetic silencing of OXR2 and the methylation status of the orexin precursor gene promoter with cell survival and recurrence in head and neck cancer. Moreover, stress and depression increase the risk of developing liver cancer via the epigenetic downregulation of orexins, and it seems that this downregulation could be the link between depression and cancer. Previous findings (e.g., methylation status), related to the orexinergic system, are examples of possible biomarkers for tumor risk/prognosis; this must be confirmed and studied in depth in different tumors. In this sense, the correlation between OXR expression and poor prognosis must be confirmed in prostate cancer.

4.4. Orexins and Chemotherapeutic Drugs

Seabuckthorn seed oil prevented cisplatin-induced vomiting in rodents; in addition, an increase in orexin A plasma and lateral hypothalamic area levels was observed after the administration of this oil [142], as well as an increased OXR1 expression in both hypothalamus and brainstem [142]. Thus, orexin A could be involved in the prevention of cisplatin-induced vomiting. This must be investigated in depth, since orexin A could act as an anti-emetic agent and it could be used against vomiting in patients with cancer who have received chemotherapy. Moreover, orexin A improved cisplatin-induced acute kidney injury, exerting anti-inflammatory, anti-apoptotic and antioxidant effects [143], and orexin A improved the gastric disorders promoted by cisplatin in rodents. It seems that the peptide acted via the peptidergic neurons containing neuropeptide Y located in the hypothalamic arcuate nucleus [144]. This study also showed that cisplatin decreased the expression of prepro-orexin mRNA in the hypothalamus and that orexin A attenuated cisplatin-induced anorexia and improved weakened gastric motility; these actions were partially blocked after the administration of a neuropeptide Y1 receptor antagonist [144]. This means that orexin A could counteract some of the harmful side effects mediated by the chemotherapeutic drug cisplatin. This must be clarified. Moreover, orexin A inhibits tumor development in chemoresistant tumors; this is a crucial point that must be studied in depth. It is also important to develop studies focused on the combination therapy of OXR antagonists/agonists exerting an antitumor action with radiotherapy or chemotherapy.

4.5. Other Research Lines

It is worthy to investigate the involvement of the orexinergic system in the inverse comorbidity observed between neurodegenerative diseases and cancer, between cancer and circadian rhythm disorders, and between cancer and inflammatory mechanisms. For example, orexin A exerts an anti-inflammatory effect, and this peptide could be a promising treatment against ulcerative colitis and other intestinal bowel diseases (e.g., Crohn’s disease) in which OXR1 is expressed. Because autophagy can induce tumorigenesis by promoting tumor cell proliferation, the induction of autophagy in tumor cells by orexins must be studied in depth to develop antitumor strategies. The latter has been reported in colon cancer cells and it is currently unknown in many types of cancer. Another important point that must be confirmed is the involvement of orexin A in glucose uptake/glycolysis in tumor cells, as well as the physiological significance of the presence of OXR1 in the nuclei of hepatocellular carcinoma cells. This must be confirmed in other tumors.

Table 2. The orexinergic system: future research lines and therapeutic strategies.

Basic research on the expression of the orexinergic system in many cancer types

OXR antagonists: against those tumors in which orexins promote the proliferation of tumor cells

OX agonists: against tumors in which orexins favor apoptotic mechanisms in cancer cells

Molecular interactions between OXRs and OXR agonists or antagonists

Design and develop of new and more potent/specific OXR agonists or antagonists

Involvement of OXR dimerization/oligomerization in antimigratory and antimetastatic strategies

Epigenetic mechanisms regulating the orexinergic system

Orexin-chemotherapeutic drug interactions

Orexins and autophagy in tumor cells

Orexins and glucose uptake/glycolysis relationship

summarizes the main research lines to be developed in the future regarding the orexinergic system, as well as some promising therapeutic strategies.

In sum, much work must be done to fully understand the involvement of the orexinergic system in cancer. Most of the studies on cancer have been focused on orexin A, and the lack of knowledge about the involvement of orexin B in cancer is currently enormous; it is a topic that must be urgently explored and developed. In this sense, a systematic study focused on the expression of orexins A/B and OXRs in tumor cells, the plasma level of orexins in patients with cancer, the antitumor and/or tumorigenic effects mediated by orexins, the use of OXR antagonists or agonists, the signaling pathways activated by orexins/OXRs involved in tumor progression and inhibition, and the involvement of orexins in tumor metastasis, angiogenesis and the tumor microenvironment must be performed in many types of cancer. This basic knowledge will allow us to develop antitumor strategies (currently, no clinical trial focused on the antitumor action exerted by OXR agonists or OXR antagonists has been reported) and to explain, for example, how orexins promote and inhibit tumor progression in the same cancer type.

5. Conclusions

The orexigenic system, which, a priori, has functions unrelated to this pathology, such as the control of homeostasis and sleep/waking cycle, exhibits characteristics that make it a very good candidate as a possible therapeutic target for the treatment of cancer. These peculiar characteristics open the possibility of adapting these treatments to specific types of cancer, thus making it possible to administer treatments that are as personalized as possible. In fact, some drugs that are OXR1 antagonists have already been developed and are administered for the treatment of other pathologies. However, the particularities of the orexigenic system known so far (for example, its ability to induce apoptosis as well as the opposite mechanism, cellular proliferation) suggest that this knowledge is only the tip of the iceberg, and that the properties of orexins and their receptors are still largely unknown. Further research on this system, including its relationships with other substances and/or receptors, is therefore very necessary, as it seems very promising as one of the possible therapies that can be used, individually or in combination with others, to alleviate the enormous worldwide problem caused by cancer.