Department of Basic Medical Sciences, University of Arizona College of Medicine – Phoenix, Phoenix, Arizona, USA
Address correspondence and reprint requests to Robert J. Handa, Department of Basic Medical Sciences, University of Arizona College of Medicine – Phoenix, Building ABC1, Room 422, 425 N. 5th St, Phoenix, AZ 85004, USA. E-mail: firstname.lastname@example.org
In the dorsal raphe nucleus, 17β-estradiol (E2) increases the expression of the brain-specific, rate-limiting enzyme for serotonin biosynthesis, tryptophan hydroxylase-2 (Tph2). Although estrogen receptor beta (ERβ) has been localized to Tph2 neurons, little is known about the transcriptional regulation of the Tph2 gene by estrogen. Since the ERβ agonist, diarylpropionitrile (DPN) also increases Tph2 expression, we tested the hypothesis that E2 regulates the Tph2 promoter through direct interactions with ERβ. A serotonergic cell line, B14, which endogenously expresses ERβ was transiently transfected with a fragment of the human TPH2 5′-untranslated region (5′-UTR) cloned into a luciferase reporter vector (TPH2-luc). Treatment with E2 or DPN caused a dose-dependent increase of TPH2-luc activity. In contrast, E2 conjugated to bovine serum albumin, which is cell membrane impermeable, had no effect on TPH2-luc activity. An estrogen receptor (ER) antagonist blocked E2 or DPN-induced TPH2-luc activity suggesting a classical ER mechanism. In silico analysis revealed an estrogen-response element (ERE) half-site located within the TPH2 5′-UTR. Deletion and site-directed mutation of this site abolished ligand-induced TPH2-luc activity. These results support the concept that there is a direct and functional interaction between E2:ERβ and the ERE half-site of the TPH2 promoter to regulate Tph2 expression.
We illustrate a direct regulation of the TPH2 transcription by estradiol and ERβ via a newly identified ERE half-site within the TPH2 promoter: (i) Estradiol- or an ERβ agonist-induced TPH2 transcription was blocked by an ER antagonist, while (ii) membrane impermeable form of estradiol did not induce transcription. (iii) Deletion or mutation of the ERE half-site abolished ligand-induced TPH2 transcription.
17β-estradiol conjugated with bovine serum albumin
half maximal effective concentration
estrogen response elements
estrogen response element driven luciferase reporter
relative light units
tryptophan hydroxylase-2 promoter cloned into the luciferase reporter vector
transcription start site
Anxiety disorders are prevalent in today's society and women are twice as likely to suffer from affective disorders than men (Palanza 2001; Pigott 2003; Steiner et al. 2003). Fluctuations in circulating estrogens either across the menstrual cycle, or across their lifetime have been linked to this disproportionate rate (Best et al. 1992; Sichel et al. 1995; Arpels 1996; Gregoire et al. 1996). Reduced or changing levels of estrogen in perimenopausal women are also associated with anxiety and depression, and these may be effectively treated by hormone therapies (Best et al. 1992; Sichel et al. 1995; Arpels 1996; Gregoire et al. 1996). Correspondingly, preclinical studies in rodent models also show that 17β-estradiol (E2) can reduce anxiety- and depressive-like behaviors. Underlying these observations are differential roles for the two main forms of estrogen receptor (ER) in controlling anxiety- and depressive-like behaviors. In rats, selective agonists for ERβ are anxiolytic- and antidepressant-like, whereas activation of ERα increases anxiety- and depressive-like behaviors (Walf et al. 2004; Lund et al. 2005; Walf and Frye 2005). Thus, the possibility exists that ERβ may be an effective target for modulating affective disorders in humans.
Estrogen receptors are expressed in neurons located in many areas of the central nervous system (Shughrue et al. 1997; Alves et al. 1998; Lu et al. 2001), including the serotonin (5-HT) neurons in the dorsal raphe nucleus (DRN) in rats (Lu et al. 2001). These neurons are the major source of 5-HT in the forebrain and disruption of their function has been implicated in the etiology of affective disorders (Maes and Meltzer 1995; Sun et al. 2004; Nash et al. 2005; Zhang et al. 2005). Serotonin is synthesized in brain through the actions of a brain specific rate-limiting enzyme, tryptophan hydroxylase-2 (Tph2) and polymorphisms of the Tph2 have been associated with increased vulnerability to suicide and affective disorders (Sun et al. 2004; Nash et al. 2005; You et al. 2005; Zhang et al. 2005). Recent studies have shown that 17β-estradiol (E2) or an ERβ agonist, diarylpropionitrile (DPN) increases Tph2 mRNA in the DRN and this increase is important for the anxiolytic and antidepressant effects of E2 in rats (Hiroi et al. 2006; Donner and Handa 2009). Collectively, these studies suggest that E2 increases Tph2 mRNA in the rat DRN via ERβ; however, the molecular mechanisms for this effect remain unknown.
Little is known about the transcriptional regulation of the Tph2 gene. Previous studies have identified select regions of the 5′-untranslated region (UTR) of the human TPH2 promoter that play a critical role in regulating Tph2 expression (Patel et al. 2007). Moreover, single nucleotide polymorphisms (SNPs) in TPH2 have been shown to have significant association with major depressive disorder (Zill et al. 2004). However, to date, there have been no reports examining the TPH2 promoter sequence for functional estrogen response elements (ERE) or other regulatory regions that may be sites of transcriptional control by ERs. Consequently, this study tested the hypothesis that E2 regulates TPH2 expression through interaction with ERβ. To this end, we have identified a regulatory site in the TPH2 promoter region that may be important for ER-induced TPH2 transcriptional activity.
A serotonergic cell line, RN46A-B14 (B14) that was derived from embryonic rat medullary raphe cells (Eaton and Whittemore 1996) (kindly provided by Dr. John Neumaier, University of Washington) was used for the transfection studies. The B-14 cell line was chosen because its phenotypic endocrine profile resembles that of 5-HT neurons in vivo, in that they express ERβ, but not ERα (Bethea et al. 2003). We confirmed this expression pattern in our B14 cells using RT-PCR to amplify ERβ mRNA. Undifferentiated B14 cells were maintained in Neurobasal-A (Invitrogen Inc., Carlsbad, CA, USA) with phenol red supplemented with 10% fetal bovine serum (FBS; Gemini Bioproducts, Woodland, CA, USA), 100 U/mL penicillin, 100 μg/mL streptomycin, 250 μg/mL G418 (to select for large T antigens) and 10 μg/mL hygromycin (to select for brain-derived neurotrophic factor) in a 33°C incubator with 5% CO2 at physiological pH 7.4. In addition, a mouse-derived hippocampal cell line, HT-22 (generously provided by Dr. Dave Schubert, Salk Institute, San Diego, CA, USA), was maintained as previously described (Pak et al. 2005). No differences were found when transfections were performed in media with or without phenol red supplements. Cells were grown to 70–80% confluency, and all transient transfection experiments were performed within 10 passages.
Reporter constructs and expression vectors
A 1036 bp fragment of the human TPH2 promoter cloned into the luciferase reporter vector (TPH2-luc) was purchased from Switchgear Genomics (Switchgear Genomics, Menlo Park, CA, USA). For some studies, constructs were made with successive deletions of the 5′-end by restriction enzyme digestion upstream from the transcription start site (TSS) at positions −637 and −158 yielding constructs with TPH2 fragment spanning nucleotides +111 to −637 (−637 TPH2), +111 to −158 (−158 TPH2) from TSS, respectively. All constructs were ligated by T4 ligase, transformed and amplified. The deletion constructs were confirmed by DNA sequencing (Operon, Huntsville, AL, USA). The pRL-CMV renilla luciferase reporter construct (Promega, Madison, WI, USA) was used as an internal control for plasmid transfection efficiency. Because we used human TPH2-luc construct to examine interactions with ERβ expressed in a rodent cell line, to obviate issues concerning species difference, we also examined the responses to human ERβ. A full-length human ERβ cDNA, cloned into a pCMV6-XL4 vector, was purchased from Origene (Rockville, MD, USA). The ERE-luciferase reporter (generously provided by Dr. Paul Budworth, Case Western Reserve University, Cleveland, OH, USA) contained two copies of vitellogenin ERE sequence coupled to the minimal tk-luciferase promoter and subcloned into pGL2-Basic plasmid (Promega Corp.).
Transfection, hormone treatment, and luciferase assay
B-14 cells were transfected with TPH2-luc using a lipid-mediated reagent (Fugene6; Promega) according to manufacturer's protocol and allowed to express TPH-2-luc for 24 h. 0.3 μg of pCMV-renilla plasmid was used as an internal control for plasmid transfection efficiency and for normalization of data. Empty vector controls were used with each plasmid for negative control. Constructs were transfected in triplicate. Maintenance media were then replaced with treatment media containing hormones at a final concentration ranging from 10−15 to 10−6 M. Treatment media contained one of the following: vehicle, E2 (Steraloids, Newport, RI, USA), E2 conjugated with bovine serum albumin (E2BSA; Sigma, St. Louis, MO, USA), DPN or an ER antagonist, ICI182,780 (Tocris, Ellisville, MO, USA). The E2BSA is a membrane impermeable form of E2 that was used to test for membrane-associated effects of E2. It was filtered prior to use according to a previous protocol (Han et al. 2006) to remove unconjugated E2.
The ERβ agonist used in this study is the biologically active form of DPN, an ERβ agonist that is commonly used in its racemic form. Previously, it was separated into its two enantiomers using chiral chromatography (Weiser et al. 2009) and the more selective and potent isomer was designated S-DPN based on the modeling studies of Sun and colleagues (Sun et al. 2003). Recently, it was found using enantiospecific synthesis that the more active and potent isomer is actually R-DPN (Carroll et al. 2012). Therefore, we now designate the more active form, R-DPN [previously labeled S-DPN, (Weiser et al. 2009; Oyola et al. 2012)] to be consistent with the current nomenclature. In these studies, we used the non-selective ER antagonist, ICI182,780 to block the actions of E2 and R-DPN and have not utilized an ERβ selective antagonist in our studies because we have not identified an appropriate antagonist that is selective and of sufficiently high affinity. Fetal Bovine Serum was charcoal stripped with dextran-coated charcoal to remove small molecules such as steroid hormones. All drugs were first diluted in 100% ethanol and then further diluted with media to achieve a final concentration of ethanol in the media of ≤ 0.001%. Control cells were treated with equivalent amounts of ethanol vehicle. One, six, twenty-four, or forty-eight hours later, cells were lysed and lysates were assayed for luciferase expression, using the Dual Luciferase assay according to manufacturer's protocol (Promega). Briefly, cells were lysed in passive lysis buffer for at least 15 min and luciferase activity of 10 μL cell lysate was measured by adding 100 μL luciferin substrate. Then, renilla activity was measured by adding 100 μL Stop and Glo solution. A 20/20 TD luminometer (Turner Designs, Sunnyvale, CA, USA) was used to measure the relative light units (RLUs).
We scanned the TPH2 promoter region for a classic consensus ERE using the nucleotide sequences, GGTCA or TGACC, and identified a putative ERE half-site located at −792 nucleotides upstream of the TSS of the human TPH2 promoter. This site was then mutated using the QuikChange II XL kit according to manufacturer's instructions (Stratagene, La Jolla, CA, USA). The antisense primer sequence 5′-GAAAAAGCTTATTAACATAAAATGGAGTTGAGCCATGAGTAAAAAAAATATGCTGATGGGAGGG-3′ was designed to mutate three specific sites of the ERE half-site (GGTCA → GAGCC, with bold nucleotides denoting the mutation). The presence of the mutation was confirmed by DNA sequencing (Operon).
The RLUs for each treatment were normalized to the respective empty expression vector control and data were expressed as percent change compared with vehicle-treated, empty expression vector controls. A half maximal effective concentration (EC50) for each hormone was calculated from the dose-response curves to allow comparison between hormones. One-way (Hormone treatment) or Two-way (hormone treatment vs. treatment time, hormone treatment vs. antagonist, hormone treatment vs. deletion construct, or hormone treatment vs. mutation) anova and Bonferroni post hoc test were used where appropriate. Differences were considered significant when p <0.05.
Time course of TPH2-luc activity with E2 and R-DPN treatments
The time course for TPH2-luc induction following treatment with 100 nM E2 or R-DPN for 1, 6, 24, or 48 h showed that optimal treatment duration was at 24 h (Fig. 1). Significant increases in TPH2-luc activity were found after 24 h of treatment in both treatment groups. Two-way anova showed significant main effects for both Hormone Treatment: F(2, 139) = 3.084, p = 0.0489 and Treatment Time: F(3,139) = 4.578, p = 0.0301. Post hoc analysis revealed significant differences between vehicle versus E2 (p < 0.05) and R-DPN (p < 0.05) at 24 h, but not at any other time points. Undifferentiated B14 cells endogenously expressed ERβ mRNA (Bethea et al. 2003); however, we also examined the effects of over-expression of ERβ by transiently co-transfecting B14 cells with TPH2-luc and human ERβ and found similar results (data not shown).
Dose response for E2- and R-DPN- induced TPH2-luc activity
Using the optimal 24-h treatment duration, we next constructed a dose-response curve for E2 and R-DPN induced TPH2-luc activity to determine the EC50. Dose-response curves of the hormone treatments ranged from 1−15 to 1−6 M and revealed an EC50 for E2- and R-DPN-induced TPH2-luc activation of 5.15 × 10−15 and 1.53 × 10−12 M, respectively, when B14 cells were transfected with TPH2-luc (Fig. 1b). We also examined the effects of over-expression of ERβ by transiently co-transfecting TPH2-luc with human ERβ and found similar results to cells transfected with TPH2-luc alone (data not shown). Similar E2 and R-DPN induced luciferase activities were found when B14 cells were transfected with an ERE-driven luciferase reporter (ERE-luc), with calculated EC50s of 3.69 × 10−17 M for E2 and 3.31 × 10−13 M for R-DPN (Fig. 1c).
E2 and R-DPN induced increases in TPH2-luc activity are cell line specific
To assess whether E2-induced TPH2 activity was cell-type specific, we compared the effects of E2 and R-DPN-induced TPH-luc activity in the B14 and the hippocampal derived cell line, HT22, which is a non-serotonergic cell line that reportedly lacks ER expression (Behl et al. 1995; Green et al. 1998; Kim et al. 2001; Pak et al. 2005). Using the highest dose of E2 and R-DPN from the previous study (100 nM) to induce maximal TPH2-luc activation with both ligands, we found significantly increased TPH2-luc activity when TPH2-luc alone or TPH2-luc and ERβ were co-transfected into the B14 cell line (Fig. 2a and b, respectively). For TPH2-luc alone (Fig. 2a), there was a significant main effect of treatment using one-way anova: F(2,49) = 6.012, p = 0.0046. Post hoc test revealed significant differences between vehicle versus E2 and R-DPN. For TPH2-luc and ERβ (Fig. 2b), there was a significant main effect of treatment using one-way anova: F(2,39) = 12.51, p < 0.0001. Post hoc test revealed significant differences between vehicle versus E2 and R-DPN. In contrast, there was no significant effect of E2 or R-DPN to induce TPH2-luc activity when TPH2-luc alone or TPH2-luc and ERβ were co-transfected into the HT-22 cell line (Fig. 2c and d, respectively).
E2 and R-DPN increase TPH2-luc activity via the classical ERβ signaling pathway
To determine whether ER mediates the E2-induced effects on TPH2-luc activity, cells were treated with ICI 182,780, an ER antagonist, at 100x the concentration of the hormone treatment. ICI 182,780 (10 μM) was sufficient to significantly block both 100 nM E2- and R-DPN-induced increases of TPH2-luc (Fig. 3a). Two-way anova showed a significant main effect of antagonist: F(1, 36) = 8.44, p = 0.0062. Post hoc test showed significant differences between E2 versus E2 + ICI: p = 0.0426; and R-DPN versus R-DPN+ICI: p = 0.0381; Vehicle versus E2: p = 0.0081; and Vehicle versus R-DPN: p = 0.003. We also examined the effects of over-expression of ERβ by transiently co-transfecting TPH2-luc with human ERβ and found similar results (data not shown). To determine whether classical ERβ signaling pathways, or other membrane associated receptors are involved in the E2-induced TPH2 activity, the effects of E2BSA, a cell membrane impermeable conjugate of E2, was compared to that of E2 and R-DPN (Fig. 3b). Despite robust and significant activation of TPH2-luc activity following E2 and R-DPN treatment, E2BSA had no significant effects on TPH2-luc. One-way anova showed a significant main effect of treatment: F(3,15) = 9.957, p = 0.0007. Post hoc analysis revealed significant differences between vehicle versus E2 and R-DPN. These results support the hypothesis that a classical ERβ signaling pathway mediates the hormone-induced TPH2 promoter activation.
An ERE half-site on the promoter region is important for the E2 and R-DPN-induced TPH2 activity
To identify the regulatory region important for the ligand-induced TPH2 transcriptional activity, two deletion constructs, −637 TPH2 and −158 TPH2, of the promoter fragment were made and transfected into the B14 cells (Fig. 4a). Both E2 and R-DPN treatment significantly increased luciferase activity of the full length TPH2-luc. In contrast, E2- and R-DPN-induced luciferase activity was abolished by the deletion constructs (Fig. 4b). Two-way anova showed a significant interaction effect: F(4, 43) =3.251, p = 0.0204; treatment effect: F(2,43) = 4.118, p = 0.0231; and deletion construct effect: F(2,43) = 9.622, p = 0.0004. Post hoc test revealed significant differences between vehicle versus E2 or R-DPN when using full length TPH2. Next, we scanned the TPH2 5′UTR for classical ERE sites and identified an ERE half site beginning at −792 nt upstream of the TSS. Because the putative ERE half-site was eliminated in both of the deletion constructs, we performed site-directed mutagenesis of the ERE half-site in the full length TPH2 promoter to determine whether this region of the promoter was required for the ERβ-induced TPH2 transcription activity (Fig. 4c). Mutation of the ERE half-site abolished the effect of E2 and dramatically reduced the effect of R-DPN on TPH2-luc activity. Two-way anova showed a significant main effect of interaction: F(2,34) = 5.842, p = 0.0066; mutation: F(1,34) = 8.170, p = 0.0072. Post hoc test revealed significant differences between vehicle versus E2 in Full length TPH2. To our knowledge, this is the first report identifying a functional ERE half-site on the TPH2 promoter region.
In this study, we examined the functional regulation of the human TPH2 gene by E2 to identify the region within the TPH2 promoter that ERβ may use for this interaction. We found that both E2 and R-DPN increased TPH2 promoter activity, as measured by increases in TPH2-luc activity in the serotonergic B14 cell line, in vitro. This E2-induced TPH2-luc activity was blocked by co-treatment of the B14 cells with the ER antagonist, ICI182,780. While E2 and R-DPN treatment produced robust TPH2-luc activity, the cell membrane impermeable conjugate, E2BSA, was unable to affect luciferase activity. As B14 cells solely expressed ERβ, and not ERα (Bethea et al. 2003) these data support the concept that E2 induced TPH2 transcriptional activity is through classical ERβ signaling pathways. Moreover, our studies identified a classical ERE half-site located at nucleotides −792 to −787 from the TSS on the TPH2 promoter, and showed that this site is functional in that the deletion and mutation of this sequence blocked the E2-induced TPH2-luc activity. This finding confirms that the ERE half-site plays an important role in the ER mediated regulation of TPH2 transcriptional activity.
Previous studies examining the regulation of the TPH2 promoter have identified select regions of the 5′UTR that play a critical role in gene expression via interaction with multiple transcriptional factors. Such studies have also identified numerous putative DNA elements in the 5′-UTR of the TPH2 promoter, including a cAMP-response element (CRE), Sp1, AP-1, AP-2, CCAAT/enhancer binding protein (C/EBP), and TATA box, GREF/PRE, STRE, Sox-5, SBPF, GFI1, GATA, NEUROD1, Brn-2, IA-1, Myf-3, RU49, VDR/RXR (Remes Lenicov et al. 2007; Chen et al. 2008). In addition, Patel and colleagues confirmed a binding site for the bipartite neural restrictive silencing element (NRSE) (Patel et al. 2007). Examination of human TPH2 polymorphisms also revealed 5′-UTR regions of TPH2 that modulate gene expression (Scheuch et al. 2007; Chen and Miller 2009). The results of these studies suggest that there are a number of distinct transcriptional regulatory elements within the TPH2 5′-UTR that regulate transcriptional activity. However, to our knowledge, this study is the first to describe a functional site in the TPH2 promoter that provides E2 sensitivity.
To characterize the effects of E2-induced activation of the TPH2 gene, we first examined the time course of the effects of hormone treatment on the TPH2 promoter activity. The maximal increase in TPH2-luc activity occurred after 24 h of E2 treatment. Luciferase activity was also significantly increased after 24 h of R-DPN, and although further elevations occurred after 48 h, these were not statistically significant. At this point, the ½ life of R-DPN or E2 in B14 cells has not been explored. An examination of a later time point beyond 48 h may reveal distinct kinetic profiles between E2 and R-DPN in their interaction with ERβ to induce TPH2 promoter activity.
In addition to differences in the kinetic profile of TPH2 induction, E2 and R-DPN also differ in their relative potencies to activate the TPH2 promoter. Estradiol had a much greater potency than R-DPN in activating TPH2-luc, as measured by a lower EC50 value for E2 compared to R-DPN. This higher potency of E2 in TPH2-luc induction may be in part because of the greater binding affinity of E2 over R-DPN for the ERβ (Meyers et al. 2001; Weiser et al. 2009; Carroll et al. 2012), but this approximately 300 fold difference does not completely explain the much higher potency that E2 possesses for transcriptional regulation of the TPH2 promoter.
In these studies, we observed that the EC50 for both E2 and R-DPN in B14 cells is well below the typical range for binding the receptor (Paech et al. 1997; An et al. 1999; Kulakosky et al. 2002). This extremely high potency of E2 in driving the TPH2 promoter may be because of one of several factors including the artificial condition created by the in vitro system herein, or a unique property of the B14 cell line. Despite the robust E2 and R-DPN-induced TPH2-luc activation in B14 cells, the same hormone treatment, when tested in a separate cell line, HT22, transfected with TPH2-luc did not have any effect on the TPH2 promoter activity. As HT22 cells originated from embryonic mouse hippocampal cells and are non-serotonergic (Behl et al. 1995; Green et al. 1998; Kim et al. 2001; Pak et al. 2005), this cell line may not possess the appropriate cellular machinery for E2-induced activation of tph2. On the other hand, B14 cells are derived from embryonic rat serotonergic cells and also express Tph1 and 5-HT, despite the low levels of expression in the undifferentiated state (White et al. 1994; Eaton et al. 1995; Bethea et al. 2003). Therefore, these cells are most likely equipped with the necessary constellation of co-regulators and transcription factors involved in TPH2 transcription.
Cell-type specific induction of human (and rat) TPH2 transcription has been previously reported. Remes Lenicov and colleagues (Remes Lenicov et al. 2007) have shown that calcium mobilization induced TPH2 transcription in the B14 cells, but not in the Tph2-negative non-neuronal L6 cell line, or in the Tph2-expressing pituitary GH4C1 cell line. Thus, these results suggest that there is a distinct milieu of transcriptional factors and co-regulators that uniquely impact TPH2 transcription, even within cell lines that are capable of producing 5-HT. These studies underscore the importance of cell-type specific effects on TPH2 transcription and suggest that ERβ-bound estrogens may require recruitment of a cell-specific set of co-regulators and transcriptional factors for TPH2 gene activation.
Indeed, ERβ plays an important role in estrogen-induced TPH2 promoter activity. We found that E2 and R-DPN-induced TPH2-luc activity was blocked when cells were co-treated with a non-selective ER antagonist. Given that our B14 cell line solely express ERβ, and not ERα, these results support the hypothesis that ERβ mediates the estrogen-induced TPH2 promoter activity. Moreover, despite a robust induction of TPH2 promoter following E2 and R-DPN, the membrane impermeable form of E2, E2BSA, was not able to induce TPH2-luc activity. Collectively, these studies suggest that there is likely a direct interaction of E2 with the TPH2 promoter utilizing classical ERβ signaling pathways to increase transcriptional activity.
Thus far, the regulatory regions involved in estrogen-induced TPH2 promoter activity have not been identified. Therefore, we scanned the promoter region for a consensus palindromic ERE sequence and found that the TPH2 promoter does not contain a classical ERE; however, a putative ERE half-site was found at −792 nucleotides upstream of the TSS of the human TPH2 promoter. Our studies now demonstrate that this ERE half-site on the TPH2 promoter is important for the E2 and R-DPN-induced TPH2 promoter activity, as deletion and site-directed mutation of the ERE half-site blocked E2 and R-DPN-induced TPH2-luc activity. In addition, E2 and R-DPN treatment of these B14 cells also induced luciferase activity of transiently transfected ERE-luc, suggesting that the environment of the cells is also sufficient for hormone interaction with the ERE to induce transcriptional activation. Future studies are warranted to determine the specific environment essential for the ERβ-bound E2 to interact with this ERE half-site on the TPH2 promoter, as the identification of co-regulatory proteins and alternative transcription factors involved in controlling Tph2 expression is of important biological relevance.
The regulation of the Tph2 gene by estrogens has been shown to have important roles in neurobiology, notably related to the regulation of anxiety-like and depressive-like behaviors. Recent studies have shown that E2 treatment of ovariectomized rats increases Tph2 mRNA in the DRN (Hiroi et al. 2006), a major source of 5-HT neurotransmitter in the forebrain. This increase in Tph2 is important for the anxiolytic effects of E2, as over-expression or knockdown of Tph2 mRNA in the DRN of the ovariectomized rats mimicked and reversed the anxiolytic effects of E2, respectively (Hiroi et al. 2011). Furthermore, the direct bilateral stereotaxic implantation of racemic DPN-containing wax pellets flanking the DRN in ovariectomized rats also increased Tph2 mRNA expression in the DRN and induced anti-depressant like effects (Donner and Handa 2009). Collectively, these studies suggest that the anxiolytic and antidepressant effects of E2 may, in part, be mediated by increases in Tph2 expression via the ERβ selective activation of Tph2 expression within neurons of the DRN. This in turn may result in changes in the forebrain 5-HT neurotransmission that have important ramifications for regulating behaviors.
Although it is feasible that ERα may have effects on the TPH2 promoter, we did not test it here as B14 cells used in this study were derived from rats and do not express ERα. However, it would be important to test the effects of both ERα and ERβ on Tph2 promoters of different species, as distribution of ERα and ERβ on 5-HT neurons differs in distinct species and there is a marked species specific regulation of Tph2 expression by E2. Chronic E2 treatment increases Tph2 mRNA in the DRN in rats (Hiroi et al. 2006; Donner and Handa 2009) and macaques (Sanchez et al., 2005), but not in mice (Clark et al., 2005). This may, in part, be because of the differential distribution of the two ERs within the DRN of each species. For instance, like rats, non-human primates express ERβ, but not ERα, in 5-HT neurons (Bethea 2002), but ERα and ERβ are co-expressed in some 5-HT neurons of mice (Mitra et al., 2003). Therefore, it is possible that species specific regulation of the brain serotonergic system by the estrogens may result in distinct biological functions, allowing adaptive responses to differential environment encountered by each species.
The results from this study add to the growing body of evidence showing that ERβ mediates the E2-induced increases in Tph2 activity. Specifically, a newly identified ERE half-site located within the TPH2 promoter region plays an essential role in activating E2 induced TPH2 activity. Further analysis of the co-regulators and transcription factors involved in mediating the interaction among ERβ, E2, and TPH2 may reveal novel pharmacological targets that could be used alone or in conjunction with current treatment options available for women suffering from anxiety and affective disorders.
Acknowledgements/Conflicts of interest disclosure
This study was supported by R01-NS039951 (to RJH) and F32-MH093145 (to RH). We thank Dr. Jessica Healy and Ms. Laura Hinds for technical assistance. The authors have no potential conflict of interest to disclose.