LeeLemonoil
Member
- Joined
- Sep 24, 2016
- Messages
- 4,265
A Testosterone Metabolite 19-Hydroxyandrostenedione Induces Neuroendocrine Trans-Differentiation of Prostate Cancer Cells via an Ectopic Olfactory Receptor
OR51E2 activation by agonists seem to promote cancer - Retinoic acid seems to be the only identified antagonis so far.
As the number of ectopic olfactory receptors associated with diverse pathological states continues to increase, the implications and significance of these receptors will be greatly enhanced by receptor “deorphanization” (i.e., defining the ligands). Previously, we successfully identified novel ligands for mouse OR using a similar in silico approach with VLS (38). Here, we present a highly successful approach of combining in silico and in vitro analyses to identify novel biologically relevant ligands for the human ectopic OR, OR51E2. This method can be used to elucidate ligand specificities of other ectopic ORs. Once identified, these new ligands can help define the role and function of ORs in cancer and other diseases.
Among the newly discovered metabolites identified as OR51E2 agonists, several were previously reported to be associated with PC, including bradykinin, kojibiose, glycylglycine, N-acetylglutamic acid, and d-alanyl-d-alanine (47). Thus, our results indicate that these metabolites are likely endogenous agonists. New agonists with known biological roles were also discovered: epitestosterone, known to be a major metabolite of androstenedione and testosterone (65) and androstanedione (also known as 5α-androstane-3,17-dione), an intermediate in steroid synthesis (66).
In addition to these agonists, we also identified a previously under-reported metabolite of the complex steroid biosynthetic network, 19-hydroxyandrost-4-ene-3,17-dione (19-OH AD) (53, 67, 68). It is produced by aromatase P450 (CYP19A1), which catalyzes the irreversible aromatization of the androgens androst-4-ene-3,17-dione and testosterone and their consequent conversion to estrogens.4 We detected this testosterone metabolite in agonist-stimulated prostate cancer cells. These results demonstrate that 19-OH AD is actively produced by cancer cells when the OR51E2 receptor is activated. Thus, we demonstrate that 19-OH AD is an endogenous agonist produced by activation of OR51E2 in prostate cancer cells.
Aromatase is increased 30-fold in metastatic PC (59), and aromatase-knockout mice have a reduced incidence of PC following exposure to testosterone and estrogen, indicating that aromatase metabolites, mainly 19-OH AD and estradiol, are likely involved in prostate carcinogenesis. Results from our study demonstrate that 19-OH AD is a potent OR51E2 agonist (EC50 = 1.5−10 M) and support the notion that increased in situ estrogen production via 19-OH AD is an important factor in PC (58).
Acetyl-N-formyl-5-methoxykynurenamine is a metabolite of melatonin (69). Previous studies demonstrated that melatonin reduces proliferation of LNCaP cells, leading to NEtD, and the phenotype was not reversed by melatonin receptor antagonists, suggesting that additional receptors may be mediating this process (60). An additional source of AFMK might be a tryptophan metabolic pathway (70). Tumors produce high levels of tryptophan and kynurenic acid metabolites (71). Significant amplification of tryptophan-2,3-dioxygenase (EC 1.13.11.11) TDO2, which catalyzes the oxidation of l-tryptophan to N-formyl-l-kynurenamine, was observed in NEPC and PC (72, 73). Thus, in more advanced stages of PC, AFMK production may be increased via this tryptophan metabolic pathway.
We also identified bradykinin as an agonist for OR51E2. Prostatic secretions of PC patients have elevated levels of human kallikrein 2 (74), which produces bradykinin and thus stimulates proliferation of androgen-independent PC cells in later stages of PC (75).
The OR51E2 antagonist 13-cis RA is an endogenous component of human serum, and many of its actions can be explained by isomerization to all-trans RA and 9-cis RA, which both act via retinoid receptors. However, since 13-cis RA does not have potent gene regulatory activities, additional pathways via membrane receptors have been proposed to explain its pharmacological and anti-inflammatory actions (76). Our results demonstrate that 13-cis RA acts via the OR51E2 receptor when expressed heterologously.
OR51E2 receptor activation by 19-OH AD, AFMK, and PA induced pronounced metabolic reprograming of LNCaP cells, with the most significant changes being decreased intracellular serine and threonine levels. Because metabolism of these amino acids includes one-carbon metabolism, which provides cofactors for biosynthetic reactions in highly proliferating cells, intracellular depletion may indicate a general decrease in anabolic reactions (77). Furthermore, an intracellular decrease in aspartate, which is normally required for protein, purine, and pyrimidine synthesis, and an increase in the CM indicate that agonist-activated LNCaP cells are not preparing for proliferation. We also detected decreased intermediates of glycolysis (glucose-6-phosphate and fructose-6-phosphate) in activated cells. Agonist treatment decreased intracellular lactate, suggesting a slower rate of glycolysis. An intriguing result was the increased intracellular level of PEP. We also found increased NSE transcription for the glycolytic enzyme enolase, which catalyzes the formation of PEP, indicating predominance of the PEP-forming reaction.
Neuron-specific enolase is not only a marker of neuronal differentiation and maturation characteristic of neurons and neuroendocrine cells (78), but it also has an important role in synaptogenesis (79) and is reported to be stable in the high-chloride environment characteristic of neurons, in which it reaches a concentration of 2% to 3% of the soluble protein (80). Taken together, these results indicate that receptor activation results in a neuronal-like phenotype of LNCaP cells.
Cystine was increased in the medium after 19-OH AD and AFMK treatments (Tables S12 and S13 in Supplementary Material). In healthy cells, cystine is transported into cells and reduced to cysteine, which can then be utilized for synthesis of gluthatione, a protective antioxidant. As a consequence of rapid cell growth during tumorigenesis, the production of reactive oxygen species increases, providing a proliferative signal for glutamine to enter the cell and, after deamidation, condense with cysteine to form a precursor of glutathione. However, in our experiments, the medium, but not the cells, showed increased levels of glutamine and cysteine, indicating a reduction in protective oxidative and proliferative signals in agonist-stimulated cells. The alanine/aspartate/glutamine pathway is the most affected biochemical pathway during NEtD of LNCaP cells induced by steroid-reduced medium, which corroborates our results (Figure 4F) (81). In cancer-associated fibroblasts, asparagine and aspartate are involved in glutamine synthesis (82), and our experiments showed decreased intracellular levels of these amino acids, suggesting increased use for intracellular synthesis of glutamine. These results might also indicate a decreased cellular influx of asparagine, since it is abundant in CM. Flux studies will be necessary to determine the exact relationship between glutamine synthesis and transport in PC cells upon receptor activation.
To explain the role of OR51E2 in PC, we propose the following model: agonist stimulation generates new cells through asymmetric division and gradually increases the subpopulation of terminally differentiated cells expressing neuroendocrine markers.
NE-like cells from PC are characterized by increased expression of NSE and AMACR and decreased expression of K18 and AR (83). Increased expression of NSE and AMARC and decreased expression of AR and K18 following 19-OH AD and AFMK treatment demonstrate that these OR51E2 agonists induce a neuroendocrine phenotype. We confirmed that this effect is indeed receptor-mediated, as treatment of OR51E2-knockout LNCaP cells significantly reduced the NSE and AMACR transcript levels (Figure 5; Figure S13 in Supplementary Material).
Cell proliferation and differentiation have an inverse relationship, and terminal differentiation coincides with proliferation arrest and exit from the division cycle. Our results demonstrate that agonist treatment during the first 3 days induces cell proliferation at a rate similar to control cells, but after 4 days the viability of these cells, as measured by ATP content, was significantly reduced. Our results also demonstrate that receptor activation results in a new subpopulation of cells that undergoes G0/G1 cell cycle arrest and has decreased DNA synthesis, which is concordant with the results from our metabolomics analysis. Cellular senescence is an irreversible growth arrest, and senescent cells actively suppress apoptosis. We found that agonist treatment decreases the fraction of apoptotic cells, indicating that growth arrest likely induces cellular senescence. Future studies are needed to confirm the irreversibility of this process.
Furthermore, recent whole-exome sequencing of NEPC and CRPC showed an overlap in genomic alterations, and in both demonstrated increased amplification of the OR51E2 gene, supporting our hypothesis that this receptor contributes to the NE-phenotype of PC (72, 73).
Prostatic adenocarcinomas typically contain foci of non-proliferating NE-like cells that increase in number as cancer progresses (84). Although these cells are non-mitotic, proliferating carcinoma cells have been found in their proximity, suggesting that the non-proliferating NE-like cells likely provide paracrine stimuli for growth of the surrounding carcinoma cells. Our results demonstrate that activation of OR51E2 by newly discovered PC-relevant agonists induces and/or facilitates cellular transformation, resulting in NEtD, a characteristic phenotype of CRCP. This indicates that activation of OR51E2 in PC might contribute to development of non-proliferating foci. Our data demonstrate that activation of OR51E2 results in NEtD and establish this GPCR as a novel and therapeutic target for NEPC and CRPC.
OR51E2 activation by agonists seem to promote cancer - Retinoic acid seems to be the only identified antagonis so far.
As the number of ectopic olfactory receptors associated with diverse pathological states continues to increase, the implications and significance of these receptors will be greatly enhanced by receptor “deorphanization” (i.e., defining the ligands). Previously, we successfully identified novel ligands for mouse OR using a similar in silico approach with VLS (38). Here, we present a highly successful approach of combining in silico and in vitro analyses to identify novel biologically relevant ligands for the human ectopic OR, OR51E2. This method can be used to elucidate ligand specificities of other ectopic ORs. Once identified, these new ligands can help define the role and function of ORs in cancer and other diseases.
Among the newly discovered metabolites identified as OR51E2 agonists, several were previously reported to be associated with PC, including bradykinin, kojibiose, glycylglycine, N-acetylglutamic acid, and d-alanyl-d-alanine (47). Thus, our results indicate that these metabolites are likely endogenous agonists. New agonists with known biological roles were also discovered: epitestosterone, known to be a major metabolite of androstenedione and testosterone (65) and androstanedione (also known as 5α-androstane-3,17-dione), an intermediate in steroid synthesis (66).
In addition to these agonists, we also identified a previously under-reported metabolite of the complex steroid biosynthetic network, 19-hydroxyandrost-4-ene-3,17-dione (19-OH AD) (53, 67, 68). It is produced by aromatase P450 (CYP19A1), which catalyzes the irreversible aromatization of the androgens androst-4-ene-3,17-dione and testosterone and their consequent conversion to estrogens.4 We detected this testosterone metabolite in agonist-stimulated prostate cancer cells. These results demonstrate that 19-OH AD is actively produced by cancer cells when the OR51E2 receptor is activated. Thus, we demonstrate that 19-OH AD is an endogenous agonist produced by activation of OR51E2 in prostate cancer cells.
Aromatase is increased 30-fold in metastatic PC (59), and aromatase-knockout mice have a reduced incidence of PC following exposure to testosterone and estrogen, indicating that aromatase metabolites, mainly 19-OH AD and estradiol, are likely involved in prostate carcinogenesis. Results from our study demonstrate that 19-OH AD is a potent OR51E2 agonist (EC50 = 1.5−10 M) and support the notion that increased in situ estrogen production via 19-OH AD is an important factor in PC (58).
Acetyl-N-formyl-5-methoxykynurenamine is a metabolite of melatonin (69). Previous studies demonstrated that melatonin reduces proliferation of LNCaP cells, leading to NEtD, and the phenotype was not reversed by melatonin receptor antagonists, suggesting that additional receptors may be mediating this process (60). An additional source of AFMK might be a tryptophan metabolic pathway (70). Tumors produce high levels of tryptophan and kynurenic acid metabolites (71). Significant amplification of tryptophan-2,3-dioxygenase (EC 1.13.11.11) TDO2, which catalyzes the oxidation of l-tryptophan to N-formyl-l-kynurenamine, was observed in NEPC and PC (72, 73). Thus, in more advanced stages of PC, AFMK production may be increased via this tryptophan metabolic pathway.
We also identified bradykinin as an agonist for OR51E2. Prostatic secretions of PC patients have elevated levels of human kallikrein 2 (74), which produces bradykinin and thus stimulates proliferation of androgen-independent PC cells in later stages of PC (75).
The OR51E2 antagonist 13-cis RA is an endogenous component of human serum, and many of its actions can be explained by isomerization to all-trans RA and 9-cis RA, which both act via retinoid receptors. However, since 13-cis RA does not have potent gene regulatory activities, additional pathways via membrane receptors have been proposed to explain its pharmacological and anti-inflammatory actions (76). Our results demonstrate that 13-cis RA acts via the OR51E2 receptor when expressed heterologously.
OR51E2 receptor activation by 19-OH AD, AFMK, and PA induced pronounced metabolic reprograming of LNCaP cells, with the most significant changes being decreased intracellular serine and threonine levels. Because metabolism of these amino acids includes one-carbon metabolism, which provides cofactors for biosynthetic reactions in highly proliferating cells, intracellular depletion may indicate a general decrease in anabolic reactions (77). Furthermore, an intracellular decrease in aspartate, which is normally required for protein, purine, and pyrimidine synthesis, and an increase in the CM indicate that agonist-activated LNCaP cells are not preparing for proliferation. We also detected decreased intermediates of glycolysis (glucose-6-phosphate and fructose-6-phosphate) in activated cells. Agonist treatment decreased intracellular lactate, suggesting a slower rate of glycolysis. An intriguing result was the increased intracellular level of PEP. We also found increased NSE transcription for the glycolytic enzyme enolase, which catalyzes the formation of PEP, indicating predominance of the PEP-forming reaction.
Neuron-specific enolase is not only a marker of neuronal differentiation and maturation characteristic of neurons and neuroendocrine cells (78), but it also has an important role in synaptogenesis (79) and is reported to be stable in the high-chloride environment characteristic of neurons, in which it reaches a concentration of 2% to 3% of the soluble protein (80). Taken together, these results indicate that receptor activation results in a neuronal-like phenotype of LNCaP cells.
Cystine was increased in the medium after 19-OH AD and AFMK treatments (Tables S12 and S13 in Supplementary Material). In healthy cells, cystine is transported into cells and reduced to cysteine, which can then be utilized for synthesis of gluthatione, a protective antioxidant. As a consequence of rapid cell growth during tumorigenesis, the production of reactive oxygen species increases, providing a proliferative signal for glutamine to enter the cell and, after deamidation, condense with cysteine to form a precursor of glutathione. However, in our experiments, the medium, but not the cells, showed increased levels of glutamine and cysteine, indicating a reduction in protective oxidative and proliferative signals in agonist-stimulated cells. The alanine/aspartate/glutamine pathway is the most affected biochemical pathway during NEtD of LNCaP cells induced by steroid-reduced medium, which corroborates our results (Figure 4F) (81). In cancer-associated fibroblasts, asparagine and aspartate are involved in glutamine synthesis (82), and our experiments showed decreased intracellular levels of these amino acids, suggesting increased use for intracellular synthesis of glutamine. These results might also indicate a decreased cellular influx of asparagine, since it is abundant in CM. Flux studies will be necessary to determine the exact relationship between glutamine synthesis and transport in PC cells upon receptor activation.
To explain the role of OR51E2 in PC, we propose the following model: agonist stimulation generates new cells through asymmetric division and gradually increases the subpopulation of terminally differentiated cells expressing neuroendocrine markers.
NE-like cells from PC are characterized by increased expression of NSE and AMACR and decreased expression of K18 and AR (83). Increased expression of NSE and AMARC and decreased expression of AR and K18 following 19-OH AD and AFMK treatment demonstrate that these OR51E2 agonists induce a neuroendocrine phenotype. We confirmed that this effect is indeed receptor-mediated, as treatment of OR51E2-knockout LNCaP cells significantly reduced the NSE and AMACR transcript levels (Figure 5; Figure S13 in Supplementary Material).
Cell proliferation and differentiation have an inverse relationship, and terminal differentiation coincides with proliferation arrest and exit from the division cycle. Our results demonstrate that agonist treatment during the first 3 days induces cell proliferation at a rate similar to control cells, but after 4 days the viability of these cells, as measured by ATP content, was significantly reduced. Our results also demonstrate that receptor activation results in a new subpopulation of cells that undergoes G0/G1 cell cycle arrest and has decreased DNA synthesis, which is concordant with the results from our metabolomics analysis. Cellular senescence is an irreversible growth arrest, and senescent cells actively suppress apoptosis. We found that agonist treatment decreases the fraction of apoptotic cells, indicating that growth arrest likely induces cellular senescence. Future studies are needed to confirm the irreversibility of this process.
Furthermore, recent whole-exome sequencing of NEPC and CRPC showed an overlap in genomic alterations, and in both demonstrated increased amplification of the OR51E2 gene, supporting our hypothesis that this receptor contributes to the NE-phenotype of PC (72, 73).
Prostatic adenocarcinomas typically contain foci of non-proliferating NE-like cells that increase in number as cancer progresses (84). Although these cells are non-mitotic, proliferating carcinoma cells have been found in their proximity, suggesting that the non-proliferating NE-like cells likely provide paracrine stimuli for growth of the surrounding carcinoma cells. Our results demonstrate that activation of OR51E2 by newly discovered PC-relevant agonists induces and/or facilitates cellular transformation, resulting in NEtD, a characteristic phenotype of CRCP. This indicates that activation of OR51E2 in PC might contribute to development of non-proliferating foci. Our data demonstrate that activation of OR51E2 results in NEtD and establish this GPCR as a novel and therapeutic target for NEPC and CRPC.
