Forkhead box O1 is a repressor of basal and GnRH-induced Fshb transcription in gonadotropes

Abstract

Synthesis of the gonadotropin β-subunits is tightly controlled by a complex network of hormonal signaling pathways that may be modulated by metabolic cues. Recently, we reported that insulin regulates FOXO1 phosphorylation and cellular localization in pituitary gonadotropes and that FOXO1 overexpression inhibits Lhb transcription. In the current study, we investigated whether FOXO1 modulates Fshb synthesis. Here, we demonstrate that FOXO1 represses basal and GnRH-induced Fshb transcription in LβT2 cells. In addition, we show that PI3K inhibition, which increases FOXO1 nuclear localization, results in decreased Fshb mRNA levels in murine primary pituitary cells. FOXO1 also decreases transcription from the human FSHB promoter, suggesting that FOXO1 regulation of FSHB transcription may be conserved between rodents and humans. Although the FOXO1 DNA-binding domain is necessary for suppression of Fshb, we do not observe direct binding of FOXO1 to the Fshb promoter, suggesting that FOXO1 exerts its effect through protein-protein interactions with transcription factors required for Fshb synthesis. FOXO1 suppression of basal Fshb transcription may involve PITX1 because PITX1 interacts with FOXO1, FOXO1 repression maps to the proximal Fshb promoter containing a PITX1-binding site, PITX1 induction of Fshb or a PITX1 binding element in CV-1 cells is decreased by FOXO1, and FOXO1 suppresses Pitx1 mRNA and protein levels. GnRH induction of an Fshb promoter containing a deletion at -50/-41 or -30/-21 is not repressed by FOXO1, suggesting that these two regions may be involved in FOXO1 suppression of GnRH-induced Fshb synthesis. In summary, our data demonstrate that FOXO1 can negatively regulate Fshb transcription and suggest that FOXO1 may relay metabolic hormonal signals to modulate gonadotropin production.

Figure 1.

Constitutively active FOXO1 and inhibition of PI3K reduces basal transcription and GnRH induction of Fshb mRNA in gonadotropes. (A) Diagram illustrating wild-type FOXO1 and FOXO1-CA (T24A/S256A/S319A). NES, nuclear export signal; NLS, nuclear localization signal. (B) LβT2 cells were transduced with Ad-GFP or Ad-FOXO1-CA for 6 hours, then switched to serum-free media. Twenty-four hours after adenoviral infection, cells were treated with 0.1% BSA vehicle (Veh) or 10 nM GnRH for 6 hours, as indicated. The results represent the mean ± SEM of three experiments performed in triplicate and are presented as amount of Fshb mRNA relative to Gapdh. *, Fshb transcription is significantly repressed by FOXO1-CA compared to GFP using Student's t test. (C) After overnight incubation in serum-free media, dispersed murine primary pituitary cells were pretreated with DMSO or 50 μM LY294002 for 1 hour, then treated with Veh or 30 nM GnRH along with DMSO or LY294002 for an additional 5 hours. The results represent the mean ± SEM of four experiments performed in triplicate and are presented as amount of Fshb mRNA relative to Gapdh. #, GnRH significantly increased Fshb mRNA levels compared to Veh using Student's t test; *, Fshb transcription is significantly repressed by LY294002 compared to DMSO using Student's t test.

Figure 2.

FOXO1 suppresses basal transcription and GnRH induction of the murine Fshb-luc in LβT2 cells. (A) The pcDNA3 Flag human FOXO1 and the pcDNA3 Flag human FOXO1-CA expression vectors were transfected into LβT2 cells and then incubated overnight in serum-free media. Immunofluorescence was performed with a mouse anti-Flag primary antibody and an Alexa488-conjugated goat antimouse secondary antibody. 4′,6-diamidino-2-phenylindole (DAPI) was used as a nuclear marker. Representative images were obtained using a Nikon Eclipse TE2000-U inverted fluorescence microscope at 60× magnification. (B) The 4xFBE-luc reporter was transiently transfected into LβT2 cells along 200 ng of pcDNA3 empty vector (EV) or FOXO1 expression vectors, as indicated. After overnight incubation in serum-free media, cells were treated for 6 hours with 0.1% BSA vehicle (Veh) or 10 nM GnRH. The results represent the mean ± SEM of three experiments performed in triplicate and are presented as luc/β-gal. *, induction by FOXO1 is significantly different from the empty vector using Student's t test. (C–E) The −1000 murine Fshb-luc reporter was transfected into LβT2 cells along with EV, FOXO1, or FOXO1-CA (CA), as indicated. After overnight incubation in serum-free media, cells were treated for 6 hours with Veh or 10 nM GnRH. The results represent the mean ± SEM of three experiments performed in triplicate and are presented as luc/β-gal (C), basal transcription relative to EV (D), or fold GnRH induction relative to the vehicle control (E). The different lowercase letters indicate that Fshb-luc transcription is significantly repressed by FOXO1 or FOXO1-CA compared to EV using one-way ANOVA followed by Tukey's HSD post-hoc test.

Figure 3.

FOXO1 suppresses basal transcription and GnRH induction of human FSHB-luc. (A–C) The −1028/+7 human FSHB-luc reporter was transfected into LβT2 cells along with pcDNA3 empty vector (EV) or FOXO1, as indicated. After overnight incubation in serum-free media, cells were treated for 6 hours with 0.1% BSA vehicle (Veh) or 10 nM GnRH. The results represent the mean ± SEM of three experiments performed in triplicate and are presented as luc/β-gal (A), basal transcription relative to empty vector (B), or fold GnRH induction relative to the vehicle control (C). *, FSHB-luc transcription is significantly repressed by FOXO1 compared to EV using Student's t test.

Figure 4.

FOXO1 suppression maps to the proximal Fshb promoter. (A) Diagram illustrating the location of binding elements for NFY, AP1, and PITX1 transcription factors on the murine Fshb Promoter. (B–D) The −1000, −500, −304, −95, and −64 murine Fshb-luc reporters were transiently transfected into LβT2 cells along with pcDNA3 empty vector (EV) or FOXO1, as indicated. After overnight incubation in serum-free media, cells were treated for 6 hours with 0.1% BSA vehicle (Veh) or 10 nM GnRH. The results represent the mean ± SEM of three experiments performed in triplicate and are presented as luc/β-gal (B), basal transcription relative to empty vector (C), or fold GnRH induction relative to the vehicle control (D). *, Fshb-luc transcription is significantly repressed by FOXO1 compared to EV using one-way ANOVA followed by Tukey's HSD post-hoc test.

Figure 5.

DNA binding domain of FOXO1 is required to suppress basal and GnRH-induced Fshb gene expression. (A) Diagram illustrating FOXO1-ΔDBD (Δ208–220). (B) LβT2 cells were transfected with pcDNA3 empty vector (EV), pcDNA3-FOXO1 (WT), or pcDNA3-FOXO1-ΔDBD for 6 hours, then switched to serum-free media. Twenty-four hours after transfection, the cells were harvested. Western blot analysis was performed on whole cell extracts using FOXO1 and β-Tubulin primary antibodies and a horseradish peroxidase–linked secondary antibody. A representative image is shown. (C–D) The −1000 murine Fshb-luc reporter was transfected into LβT2 cells along with EV, FOXO1, or FOXO1-ΔDBD, as indicated. After overnight incubation in serum-free media, cells were treated for 6 hours with 0.1% BSA or 10 nM GnRH. The results represent the mean ± SEM of three experiments performed in triplicate and are presented as basal transcription relative to empty vector (C) or fold GnRH induction relative to the vehicle control (D). *, Fshb-luc transcription is significantly repressed by FOXO1 compared to EV or FOXO1-ΔDBD using one-way ANOVA followed by Tukey's HSD post-hoc test.

Figure 6.

FOXO1 does not bind directly to the proximal Fshb promoter. TnT Flag-FOXO1-CA was incubated with a consensus FBE, −95/−61, −65/31, or −35/−1 Fshb probes and tested for complex formation in EMSA. (A) FOXO1-CA-DNA complex on the FBE is shown in lane 1, while anti-Flag supershift is shown in lane 3 and IgG control in lane 2. The FOXO1-CA-DNA complex (arrow), antibody supershift (ss), and nonspecific binding of proteins (ns) are indicated on the left of the gel. (B) Self-competition with excess cold FBE is shown in lane 14, lack of competition with mutant FBE, −95/−61, −65/31, or −35/−1 Fshb oligos in lanes 15–18.

Figure 7.

FOXO1 suppression of basal Fshb transcription may involve PITX1. (A) The 4xHDBE-luc reporter was transiently transfected into CV-1 cells along with pcDNA3 empty vector (EV), FOXO1, and PITX1, as indicated. The results represent the mean ± SEM of three experiments performed in triplicate and are presented as fold PITX1 induction relative to pcDNA3 empty vector. *, transcription is significantly repressed by FOXO1 compared to EV using Student's t test. (B) Diagram illustrating the mutations in the PITX1 binding element (PITXBE) in the murine Fshb promoter. (C–D) The −1000 Fshb-luc reporter and the Fshb PITX1 mutant (mut) were transiently transfected into CV-1 (C) or LβT2 (D) cells along with EV, FOXO1, and PITX1, as indicated. The results represent the mean ± SEM of three experiments performed in triplicate and are presented as fold PITX1 induction relative to pcDNA empty vector (C) or luc/β-gal (D). *, Fshb-luc transcription is significantly repressed by FOXO1 compared to EV using Student's t test; #, PITX1 mut was significantly repressed compared to the wild-type Fshb promoter. (E) GST interaction assays were performed using bacterially expressed GST-fusion proteins (indicated above each lane) and ³⁵S-labeled in vitro transcribed and translated FOXO1, FOXO1-ΔDBD, and GFP (indicated on the left of the panels). GFP was used as a negative control. The GST-fusion proteins included GST alone and GST-PITX1. One-quarter of the protein used in the interaction assay was loaded in the lane marked input. The experiment was repeated several times with the same results and a representative experiment is shown.

Figure 8.

FOXO1 suppression of basal Fshb transcription may involve FOXO1 regulation of Pitx1 mRNA and protein levels. LβT2 cells were transduced with Ad-GFP or Ad-FOXO1-CA for 6 hours, then switched to serum-free media. Twenty-four hours after adenoviral infection, cells were harvested for total mRNA (A) or protein (B–C). (A) The results represent the mean ± SEM of seven experiments performed in triplicate and are presented as amount of Pitx1 mRNA relative to Gapdh. *, Pitx1 transcription is significantly repressed by FOXO1-CA compared to GFP using Student's t test. (B–C) Western blot analysis was performed on whole cell extracts using PITX1, FOXO1, and β-Tubulin primary antibodies and a horseradish peroxidase–linked secondary antibody. A representative image is shown (B) or graph representing the mean ± SEM of four experiments presented as the amount of PITX1 protein relative to β-Tubulin (C). *, PITX1 protein levels are significantly decreased by FOXO1-CA compared to GFP using Student's t test.

Figure 9.

FOXO1 suppression of GnRH-induced Fshb transcription does not involve regulation of c-Fos or c-Jun mRNA or protein levels. LβT2 cells were transduced with Ad-GFP or Ad-FOXO1-CA for 6 hours, then switched to serum-free media. Twenty-four hours after adenoviral infection, cells were treated with 0.1% BSA vehicle (Veh) or 10 nM GnRH for 1 hour, as indicated. Cells were harvested for total mRNA (A–D) or protein (E). (A–D) The results represent the mean ± SEM of three experiments performed in triplicate and are presented as amount of c-Fos mRNA relative to Gapdh (A), fold GnRH induction of c-Fos relative to Veh (B), c-Jun mRNA relative to Gapdh (C), fold GnRH induction of c-Jun relative to Veh (D). *, transcription is significantly repressed by FOXO1-CA compared to GFP using Student's t test. (E) Western blot analysis was performed on whole cell extracts using c-FOS, c-JUN, and FOXO1 primary antibodies and a horseradish peroxidase–linked secondary antibody. A representative image is shown.

Figure 10.

FOXO1 suppression of GnRH-induced Fshb transcription involves two elements in the proximal Fshb promoter. (A) Diagram illustrating the mutations in the AP1 binding element (AP1BE) in the murine Fshb promoter. (B) The −398 Fshb-luc reporter and the Fshb AP1 mutation (mut) were transiently transfected into LβT2 cells along with pcDNA3 empty vector (EV) or FOXO1, as indicated. After overnight incubation in serum-free media, cells were treated for 6 hours with 0.1% BSA vehicle or 10 nM GnRH. The results represent the mean ± SEM of three experiments performed in triplicate and are presented as fold GnRH induction relative to the vehicle control. *, Fshb-luc transcription is significantly repressed by FOXO1 compared to EV using Student's t test; #, the AP1 mut was significantly repressed compared to the wild-type Fshb promoter. (C) The −398 Fshb-luc reporter and 10-bp deletions ranging from −70/−61 to −20/−11 were transiently transfected into LβT2 cells along with EV or FOXO1, as indicated. After overnight incubation in serum-free media, cells were treated for 6 hours with 0.1% BSA or 10 nM GnRH. The results represent the mean ± SEM of three experiments performed in triplicate and are presented as fold GnRH induction relative to the vehicle control. *, Fshb-luc transcription is significantly repressed by FOXO1 compared to EV using one-way ANOVA followed by Tukey's HSD post-hoc test.