Abstract
Activins and bone morphogenetic proteins (BMPs) share activin type 2 signaling receptors but utilize different type 1 receptors and Smads. We designed AB215, a potent BMP2-like Activin A/BMP2 chimera incorporating the high-affinity type 2 receptor-binding epitope of Activin A. In this study, we compare the signaling properties of AB215 and BMP2 in HEK293T cells and gonadotroph LβT2 cells in which Activin A and BMP2 synergistically induce FSHβ. In HEK293T cells, AB215 is more potent than BMP2 and competitively blocks Activin A signaling, while BMP2 has a partial blocking activity. Activin A signaling is insensitive to BMP pathway antagonism in HEK293T cells but is strongly inhibited by constitutively active (CA) BMP type 1 receptors. By contrast, the potencies of AB215 and BMP2 are indistinguishable in LβT2 cells and although AB215 blocks Activin A signaling, BMP2 has no inhibitory effect. Unlike HEK293T, Activin A signaling is strongly inhibited by BMP pathway antagonism in LβT2 cells but is largely unaffected by CA BMP type 1 receptors. BMP2 increases phospho-Smad3 levels in LβT2 cells, in both the absence and the presence of Activin A treatment, and augments Activin A-induced FSHβ. AB215 has the opposite effect and sharply decreases basal phospho-Smad3 levels and blocks Smad2 phosphorylation and FSHβ induction resulting from Activin A treatment. These findings together demonstrate that while AB215 activates the BMP pathway, it has opposing effects to those of BMP2 on FSHβ induction in LβT2 cells apparently due to its ability to block Activin A signaling.
Introduction
Follicle-stimulating hormone (FSH) is a gonadotrophin that is synthesized and secreted by gonadotroph cells of the anterior pituitary gland. Functionally, FSH is a glycoprotein heterodimer consisting of a β subunit specific to FSH, FSHβ, and an α subunit, chorionic gonadotrophin alpha (CGA, also known as α-glycoprotein subunit) that is shared with luteinizing hormone (LH), thyroid-stimulating hormone, and human chorionic gonadotrophin. FSH stimulates ovarian follicle growth and maturation in female mammals and disruption of its signaling leads to abnormal oogenesis (Dierich et al. 1998). Clinically, FSHB (for FSHβ subunit) or FSHR (for FSH receptor) gene mutations cause primary (Doherty et al. 2002, Kottler et al. 2010) or secondary (Beau et al. 1998) amenorrhea by premature follicle development at the pre-antral stage. Low FSH levels are also observed in polycystic ovarian syndrome (Banaszewska et al. 2003) and Kallmann syndrome (Bouvattier et al. 2012). Furthermore, uncontrolled FSH secretion, resulting in high FSH levels, was also observed in serious reproductive diseases such as premature menopause (Ahmed Ebbiary et al. 1994), gonadal dysgenesis (Breckwoldt et al. 1980), and infertility. FSH is also essential for male reproduction and plays a key role in regulating spermatogenesis (Elkington & Blackshaw 1974, Matsumoto et al. 1986, Kerr et al. 1992, Lerchl et al. 1993). Therefore, precise regulation of FSH secretion is fundamental for human reproduction and its dysregulation can cause serious illnesses.
FSH production is closely correlated with the production of the FSHβ subunit. There is a relative abundance of CGA compared with FSHβ (Bernard et al. 2010) and FSHB mRNA levels are synchronized with circulating FSH hormone levels (Ortolano et al. 1988, Halvorson et al. 1994). FSHβ is produced from the FSHB gene in gonadotroph cells by stimulation from gonadotrophin-releasing hormone (GnRH) and Activin A. GnRH is produced in the hypothalamus and secreted into the pituitary portal vasculature in a pulsatile fashion. GnRH triggers the secretion of both LH and FSH (Belchetz et al. 1978, Southworth et al. 1991) and the frequency of GnRH pulses has been considered to be the determinant of which, of these two hormones, is preferentially targeted for secretion (Marshall et al. 1993, Ferris & Shupnik 2006). Activin A, a member of the transforming growth factor beta (TGFβ) superfamily, was discovered for its ability to stimulate the production of FSH but not LH in rat primary pituitary cultures (Vale et al. 1988, Ying 1988, Bilezikjian et al. 2004). FSH levels surge twice during the female estrous cycle. The first surge occurs during the late follicular phase and is induced by GnRH (Ortolano et al. 1988, Halvorson et al. 1994). The second surge occurs in the luteal phase and is considered to be dependent on Activin A (DePaolo et al. 1979, Hoak & Schwartz 1980).
Activin A is a ubiquitous regulator of myriad cellular functions during development and in the adult animal. It plays roles in cell proliferation (Boitani et al. 1995, Ota et al. 2003, Mendis et al. 2011), differentiation (Seishima et al. 1999, Sulzbacher et al. 2009), apoptosis (Zhang et al. 1997, Chen et al. 2000), immune responses (Jones et al. 2007, Robson et al. 2009), and many other cellular activities. Activin A is a homodimer of βA subunits and signals by binding to types 1 and 2 transmembrane serine kinase receptors. High-affinity binding of Activin A to one of its type 2 receptors (ActR2 or ActR2b) allows for subsequent recruitment of its type 1 receptor Alk4 to form an active signaling complex. In this complex, the constitutively active (CA) type 2 receptors phosphorylate the type 1 receptors, thereby activating the type 1 receptor kinases which in turn phosphorylate and activate cytoplasmic SMAD2 and SMAD3 proteins. Upon phosphorylation, SMAD2 and SMAD3 form complexes with SMAD4 and are translocated into the nucleus where they regulate transcription of target genes in a complex and cell type-specific manner. This signaling cascade is regulated by feedback inhibitors of Activin A such as follistatin, which contributes to the precise regulation of FSH production in gonadotrophs.
Bone morphogenetic protein 2 (BMP2) is another member of the TGFβ superfamily that is well known for regulating osteogenesis (Wozney et al. 1988) and chondrogenesis (Yoon & Lyons 2004) via activation of the SMAD1/5/8 signaling pathway. BMP2 signals by binding to its type 1 receptors, BMPR1a (ALK3), BMPR1b (ALK6), or ActR1a (ALK2), and recruiting its type 2 receptors, i.e. ActR2, ActR2b, or BMPR2. BMP2 triggers type 1 receptor activation and phosphorylation of SMAD1/5/8, which, similar to SMAD2/3, binds SMAD4 and translocates into the nucleus where they regulate specific target genes. Although BMP2 and Activin A signal via different SMAD proteins, their coexistence in some tissues such as perinatal sensory neuron target tissues (Hall et al. 2002) and pituitary suggests some crosstalk and possible synergy between the two pathways. Indeed, BMP2 is involved in pituitary development (Ericson et al. 1998) and also expressed in adult murine pituitary (Lee et al. 2007). BMP2 regulates FSHβ synthesis in the murine gonadotroph cell line, LβT2, and can act synergistically with Activin A (Lee et al. 2007, Ho & Bernard 2010). The synergistic effect of BMP2 and Activin A was proposed to be due to BMP2 induction of inhibitor of DNA-binding (ID) proteins ID2 and ID3 via SMAD3 (Ho & Bernard 2010). However, it has recently been shown that BMP2 also activates the SMAD2/3 pathway noncanonically through ALK3 (Wang et al. 2014).
As BMP2 and Activin A both bind the type 2 receptors ActR2 and ActR2b in almost exactly the same spatial configuration (Greenwald et al. 2004, Allendorph et al. 2006), we hypothesized that we could replace the low-affinity type 2 binding epitope of BMP2 with the high-affinity type 2 receptor-binding epitope of Activin A. Our prediction was that such a swap would yield a chimeric ligand with enhanced type 2 receptor affinity and signaling potency that retains BMP2-like type 1 receptor utilization and SMAD signaling specificity. Indeed, one such ligand, AB215, utilizes the same signaling receptors and Smads as BMP2 but possesses signaling activity and potency that are enhanced relative to BMP2 as demonstrated in in vitro signaling assays (Allendorph et al. 2011). In this study, we sought to determine whether AB215 functions synergistically with Activin A in the regulation of the FSHβ production in a manner resembling BMP2. Our results indicate that AB215 and BMP2 have opposing effects on Activin A-induced FSHβ induction and that rather than synergizing with Activin A, AB215 competitively blocks Activin A signaling apparently via competition for activin type 2 receptor binding. Overall, our data demonstrate that the potencies of AB215 and BMP2 and their effects on the Activin pathway can vary dramatically in a cell type-specific manner.
Materials and methods
Reagents
BMP2 was purchased from joint Protein Central (http://jointproteincentral.com). Activin A was purchased from R&D systems (http://www.rndsystems.com). AB204 and AB215 were prepared as described previously (Allendorph et al. 2011). In brief, Activin A/BMP2 (AB2) chimera library is a synthetic assortment of six sequence segments originating from two parental molecules, and AB215 is a member of AB2 chimera library. AB215 were expressed in Escherichia coli and chemically refolded. After the purification steps of heparin affinity and C4 reverse-phase chromatography using HPLC, proteins were lyophilized. Before use, the lyophilized proteins were reconstituted in 1 mM hydrochloric acid (HCl) before diluting by at least a factor of 100 in a relevant final buffer including PBS. LDN193189 (Selleck Chemicals, Houston, TX, USA) was reconstituted in DMSO at 10 mM and diluted in media before use.
Cell culture
HEK293T cell line was purchased from American Type Culture Collection (Manassas, VA, USA) and transformed mouse gonadotroph cell line, LβT2, originates from Dr Pamela Mellon (University of California, San Diego, CA, USA) (Alarid et al. 1996). Both cell lines were grown in 37 °C humidified atmosphere of 5% CO2 in DMEM (Hyclone, Logan, UT, USA) supplemented with 10% fetal bovine serum (FBS) (Hyclone) and sub-cultured using 1:2 split every third day.
Luciferase assay
Luciferase reporter assay driven by Activin-responsive element (ARE-lux), inhibitor of DNA binding 1 (ID1-luc), and −1990/+1 mFshb (mFSHβ-luc) promoters were performed in a 96-well plate in triplicates (three wells per point) and were repeated four times. Cells were co-transfected with pCMV-β-galactosidase (β-Gal) plasmid as a normalizing factor for the transfection efficiency. CA ALKs used were ALK2-Q207D (CA-ALK2), ALK3-Q233D (CA-ALK3), and ALK6-Q203D (CA-ALK6). For transfection, Fugene 6 (Promega, Madison, WI, USA) was used in 1:2 dilution of (DNA in μg, Fugene 6 in μl) dilution for HEK293T cells and 2:6 dilution for LβT2 cells. Signaling transducers (FAST1 or FAST2) and external signaling component (SMADs) were not amplified to observe the outcome in the natural state. The entire promoter assay was carried out with reverse transfection method in reduced serum Opti-MEM (Invitrogen). After 18 h of transfection, cells were treated with indicated reagents for 24 h. At the end of the incubation, cells were lysed using Luciferase lysis buffer (Promega) and luminescence was measured using a plate luminometer (Berthhold, Bad Wildbad, Germany).
Western blot
Cells were plated in a 12-well plate (BD Biosciences, San Jose, CA, USA) at a density of 2×105 cells/well. Cells were treated as indicated and lysed using cell lysis buffer (Cell Signaling Technology, Danvers, MA, USA) supplemented with 1 mM phenylmethylsulphonyl fluoride and phosphatase inhibitor cocktail (Roche). Cell lysate's total protein amount was quantified using Bradford assay. Whole cell lysates were separated on SDS–polyacrylamide gel and transferred onto nitrocellulose or PVDF membrane for standard western blot analysis. P-SMAD2 (cat. #3101) and 3 (cat. #9520) antibodies were purchased from Cell Signaling Technology, and β-actin from Sigma.
Real-time PCR
Cells were plated in a 12-well plate (BD Biosciences) at a density of 1×105 cells/well and cultured in a medium supplemented with 10% FBS. After 16–24 h, cells were treated with indicated reagents for 48 h. After exposure, RNA was extracted with TRIsure (Bioline Taunton, MA, USA) according to the manufacturer's instruction. cDNA synthesis was performed using ReverTra Ace qPCR RT Master Mix with gDNA remover (Toyobo, Osaka, Japan) according to the manufacturer's instruction. Analysis of mRNA expression was determined with quantitative real-time PCR using Thunderbird SYBR qPCR mix (Toyobo) and 10 pM primers according to the manufacturer's instruction. The sequences of primers are as follows: mouse cyclophilin, sense: 5′-CAGACGCCACTGTCGCTTT-3′ and anti-sense: 5′-TGTCTTTGGAACTTTGTCTGCAA-3′; mouse Fshβ, sense: 5′-CTCTGAAGAGCGTGGAGTATTG-3′ and anti-sense: 5′-GTTACTACCTCCTGTCCTGTCT-3′. Abundance of mRNA in each sample was determined by the differences between the cycle threshold (Ct) values for each genes and cyclophilin, ΔCt. Relative ratios of mRNA expression levels were defined as 2−ΔΔCt, where ΔΔCt=ΔCtsample−ΔCtcontrol, which reflect changes in mRNA expression levels from treated cells compared with those from untreated cells. All experiments were carried out at least three times in triplicate.
Results
BMP2, AB215, and CA BMP type 1 receptors inhibit Activin A signaling via distinct mechanisms in HEK293T cells
We have previously created a library of Activin A/BMP2 (AB2) chimeras denoted by the code (BXXXXX), where X is either A (Activin A) or B (BMP2) (Allendorph et al. 2011). AB215 (BABBBA) was designed to have the majority of the type 2 receptor-binding epitope of Activin A (segments 2 and 6) and the majority of type 1 receptor-binding epitope of BMP2 (segments 1, 3, 4, and 5) (Fig. 1A). Consistent with our previous findings, using C2C12 cells (Allendorph et al. 2011), AB215 has a higher potency than BMP2 in activating a Smad1/5/8-responsive ID1-luciferase reporter in HEK293T cells (Fig. 1B). We employed the potent kinase inhibitor, LDN193189, to investigate a potential crosstalk between BMP and Activin signaling pathways. LDN193189 is an ATP-competitive inhibitor, which binds on the kinase hinge region of BMP type 1 receptors. We show that LDN193189 (150 nM) effectively blocks BMP2 and AB215 induction of the ID1-luciferase reporter in HEK293T cells (Fig. 1B) and we used this dose in all subsequent experiments where LDN193189 was used. Activin A induction of the Smad2/3-responsive ARE-Lux reporter was unaffected by LDN193189 in HEK293T cells (Fig. 1C), indicating that the Activin A response does not require basal activation of the BMP pathway. However, BMP2 partially diminishes Activin A signaling in HEK293T cells in a manner that is rescued to a modest extent by the BMP inhibitor LDN193189 (Fig. 1D). On the other hand, AB215 completely blocks Activin A signaling in HEK293T cells and this does not require BMP pathway activation as it is unaffected by LDN193189 treatment (Fig. 1E). To investigate the effects of downstream BMP pathway activation on Activin A signaling, we transfected CA forms of ALK2, ALK3, and ALK6 and monitored SMAD1/5/8 and SMAD2/3 signaling using ID1-luc and ARE-lux reporter assays respectively. Each of these CA BMP type 1 receptors significantly amplifies SMAD1/5/8 signaling as expected (Fig. 1F) but, at the same time, dramatically attenuates SMAD2/3 signaling (Fig. 1G).
AB215 and BMP2 inhibit Activin A signaling to varying degrees in HEK293T cells. (A) The sequences of Activin A, BMP2, and AB215 are shown illustrating the segmental composition of AB215 and the location of the type 1 receptor- and type 2 receptor-binding epitopes. HEK293T cells were transfected with ID1-Luc (B and F), ARE-Lux (C, D, E and G), and constitutively active BMP type 1 receptors (CA-ALK2, CA-ALK3, or CA-ALK6) (F and G). Cells were treated with the indicated doses of BMP2, AB215, or Activin A (B, D, E and G) in the absence or presence of the ALK2/3/6 inhibitor LDN193189 (150 nM, dotted lines). All the assays were carried out in triplicate with five independent experiments and transfection differences were normalized using β-galactosidase. Curves were fitted using the Prism Software (GraphPad, San Diego, CA, USA) and shown in mean±s.d. RLU, relative light units.
Citation: Journal of Endocrinology 223, 1; 10.1530/JOE-14-0317
AB215 and BMP2 inhibit Activin A signaling to varying degrees in HEK293T cells. (A) The sequences of Activin A, BMP2, and AB215 are shown illustrating the segmental composition of AB215 and the location of the type 1 receptor- and type 2 receptor-binding epitopes. HEK293T cells were transfected with ID1-Luc (B and F), ARE-Lux (C, D, E and G), and constitutively active BMP type 1 receptors (CA-ALK2, CA-ALK3, or CA-ALK6) (F and G). Cells were treated with the indicated doses of BMP2, AB215, or Activin A (B, D, E and G) in the absence or presence of the ALK2/3/6 inhibitor LDN193189 (150 nM, dotted lines). All the assays were carried out in triplicate with five independent experiments and transfection differences were normalized using β-galactosidase. Curves were fitted using the Prism Software (GraphPad, San Diego, CA, USA) and shown in mean±s.d. RLU, relative light units.
Citation: Journal of Endocrinology 223, 1; 10.1530/JOE-14-0317
AB215 and BMP2 inhibit Activin A signaling to varying degrees in HEK293T cells. (A) The sequences of Activin A, BMP2, and AB215 are shown illustrating the segmental composition of AB215 and the location of the type 1 receptor- and type 2 receptor-binding epitopes. HEK293T cells were transfected with ID1-Luc (B and F), ARE-Lux (C, D, E and G), and constitutively active BMP type 1 receptors (CA-ALK2, CA-ALK3, or CA-ALK6) (F and G). Cells were treated with the indicated doses of BMP2, AB215, or Activin A (B, D, E and G) in the absence or presence of the ALK2/3/6 inhibitor LDN193189 (150 nM, dotted lines). All the assays were carried out in triplicate with five independent experiments and transfection differences were normalized using β-galactosidase. Curves were fitted using the Prism Software (GraphPad, San Diego, CA, USA) and shown in mean±s.d. RLU, relative light units.
Citation: Journal of Endocrinology 223, 1; 10.1530/JOE-14-0317
AB215 inhibits Activin A signaling in LβT2 cells, while BMP2 and Smad1/5/8 activation do not
In contrast to HEK293T cells (Fig. 1) and C2C12 cells (Allendorph et al. 2011), BMP2 and AB215 have similar potencies in LβT2 cells as measured by their ability to activate the ID1-luc reporter (Fig. 2A). As predicted, LDN193189 effectively blocks the signaling of both ligands in these cells (Fig. 2A). Also unlike what was observed in HEK293T cells, Activin A induction of the ARE-lux reporter is highly dependent on basal BMP signaling in LβT2 cells as it is substantially blocked by LDN193189 (Fig. 2B). Furthermore, and also differing from what we observe in HEK293T cells, BMP2 treatment over a range of doses has no inhibitory effect on Activin A induction of the ARE-lux reporter in LβT2 cells while blocking SMAD1/5/8 signaling again diminishes the Activin A response (Fig. 2C). However, AB215 completely and dose dependently blocks Activin A signaling in LβT2 cells, as in HEK293T cells, apparently by directly competing with Activin A for type 2 receptor binding (Fig. 2D). Finally, and in further contrast to what we observe in HEK293T cells, transfection of the CA BMP type 1 receptors had only a modest effect (CA-ALK2) or no effect (CA-ALK3 and CA-ALK6) on Activin A induction of ARE-lux in LβT2 cells (Fig. 2F).
AB215 but not BMP2 inhibits Activin A signaling in LβT2 cells. HEK293T cells were transfected with ID1-Luc (A and E), ARE-Lux (B, C, D and F) and constitutively active BMP type 1 receptors (CA-ALK2, CA-ALK3, or CA-ALK6) (E and F), and treated with the indicated doses of BMP2, AB215, or Activin A (A, C, D and F) in the absence or presence of the ALK2/3/6 inhibitor LDN193189 (150 nM, dotted lines). All the assays were done in triplicate with five independent experiments and transfection differences were normalized using β-galactosidase. Curves were fitted using the Prism Software (GraphPad) and shown in mean±s.d. RLU, relative light units. The T-test was employed in (B) and one-way ANOVA with Dunnett's post hoc test was employed in (E and F).
Citation: Journal of Endocrinology 223, 1; 10.1530/JOE-14-0317
AB215 but not BMP2 inhibits Activin A signaling in LβT2 cells. HEK293T cells were transfected with ID1-Luc (A and E), ARE-Lux (B, C, D and F) and constitutively active BMP type 1 receptors (CA-ALK2, CA-ALK3, or CA-ALK6) (E and F), and treated with the indicated doses of BMP2, AB215, or Activin A (A, C, D and F) in the absence or presence of the ALK2/3/6 inhibitor LDN193189 (150 nM, dotted lines). All the assays were done in triplicate with five independent experiments and transfection differences were normalized using β-galactosidase. Curves were fitted using the Prism Software (GraphPad) and shown in mean±s.d. RLU, relative light units. The T-test was employed in (B) and one-way ANOVA with Dunnett's post hoc test was employed in (E and F).
Citation: Journal of Endocrinology 223, 1; 10.1530/JOE-14-0317
AB215 but not BMP2 inhibits Activin A signaling in LβT2 cells. HEK293T cells were transfected with ID1-Luc (A and E), ARE-Lux (B, C, D and F) and constitutively active BMP type 1 receptors (CA-ALK2, CA-ALK3, or CA-ALK6) (E and F), and treated with the indicated doses of BMP2, AB215, or Activin A (A, C, D and F) in the absence or presence of the ALK2/3/6 inhibitor LDN193189 (150 nM, dotted lines). All the assays were done in triplicate with five independent experiments and transfection differences were normalized using β-galactosidase. Curves were fitted using the Prism Software (GraphPad) and shown in mean±s.d. RLU, relative light units. The T-test was employed in (B) and one-way ANOVA with Dunnett's post hoc test was employed in (E and F).
Citation: Journal of Endocrinology 223, 1; 10.1530/JOE-14-0317
BMP2 and AB215 have opposing effects on Activin A induction of the FSHB promoter in LβT2 gonadotrophs
BMP2 and Activin A synergistically induce expression of FSHβ in gonadotroph LβT2 cells (Lee et al. 2007, Ho & Bernard 2010, Wang et al. 2014) and we compared the effects of BMP2 and AB215 on FSHB promoter activity. LβT2 cells express BMPs endogenously (Huang et al. 2001) and basal BMP signaling is required for full Activin A induction of FSHB promoter activity in these cells as indicated by the dramatic inhibitory effect of LDN193189 over a range of Activin A doses (Fig. 3A). We observe that while BMP2 dose dependently augments Activin A induction of FSHB promoter activity (Fig. 3B), AB215 has the opposite effect and inhibits the Activin A induction of FSHB (Fig. 3C). Moreover, blocking SMAD1/5/8 signaling with LDN193189 results in loss of BMP2's augmentation of the Activin A induction of FSHB promoter activity but has little or no effect on AB215 inhibition of the Activin A response (Fig. 3B and C).
AB215 and BMP2 have opposing effects on Activin A induction of FSHβ. LβT2 cells were transfected with the −1900/+1 mFSHB-Luc construct and treated with the indicated doses of AB215, BMP2, and Activin A in the absence or presence of LDN193189 (150 nM, dotted lines; A, B and C). All the assays were carried out in triplicate with five independent experiments, and transfection differences were normalized using β-galactosidase. Curves were fitted using the Prism Software and shown in mean±s.d. RLU, relative light unites. Alternatively, FSHB mRNA levels were measured by RT-PCR after transfecting LβT2 cells with CA-ALK2, CA-ALK3, or CA-ALK6 and treating them as indicated with Activin A (D) or after treating untransfected cells with Activin A, BMP2, and/or AB215 as indicated (E). LβT2 cells were also treated with Activin A, BMP2, and/or AB215 as indicated and the resulting cell lysates were analyzed for the presence of phospho-Smad2 and phospho-Smad3 (F). Significance has been calculated using one-way ANOVA with Dunnett's (D) and Tukey's (E) post hoc tests and data are shown in means±s.d.
Citation: Journal of Endocrinology 223, 1; 10.1530/JOE-14-0317
AB215 and BMP2 have opposing effects on Activin A induction of FSHβ. LβT2 cells were transfected with the −1900/+1 mFSHB-Luc construct and treated with the indicated doses of AB215, BMP2, and Activin A in the absence or presence of LDN193189 (150 nM, dotted lines; A, B and C). All the assays were carried out in triplicate with five independent experiments, and transfection differences were normalized using β-galactosidase. Curves were fitted using the Prism Software and shown in mean±s.d. RLU, relative light unites. Alternatively, FSHB mRNA levels were measured by RT-PCR after transfecting LβT2 cells with CA-ALK2, CA-ALK3, or CA-ALK6 and treating them as indicated with Activin A (D) or after treating untransfected cells with Activin A, BMP2, and/or AB215 as indicated (E). LβT2 cells were also treated with Activin A, BMP2, and/or AB215 as indicated and the resulting cell lysates were analyzed for the presence of phospho-Smad2 and phospho-Smad3 (F). Significance has been calculated using one-way ANOVA with Dunnett's (D) and Tukey's (E) post hoc tests and data are shown in means±s.d.
Citation: Journal of Endocrinology 223, 1; 10.1530/JOE-14-0317
AB215 and BMP2 have opposing effects on Activin A induction of FSHβ. LβT2 cells were transfected with the −1900/+1 mFSHB-Luc construct and treated with the indicated doses of AB215, BMP2, and Activin A in the absence or presence of LDN193189 (150 nM, dotted lines; A, B and C). All the assays were carried out in triplicate with five independent experiments, and transfection differences were normalized using β-galactosidase. Curves were fitted using the Prism Software and shown in mean±s.d. RLU, relative light unites. Alternatively, FSHB mRNA levels were measured by RT-PCR after transfecting LβT2 cells with CA-ALK2, CA-ALK3, or CA-ALK6 and treating them as indicated with Activin A (D) or after treating untransfected cells with Activin A, BMP2, and/or AB215 as indicated (E). LβT2 cells were also treated with Activin A, BMP2, and/or AB215 as indicated and the resulting cell lysates were analyzed for the presence of phospho-Smad2 and phospho-Smad3 (F). Significance has been calculated using one-way ANOVA with Dunnett's (D) and Tukey's (E) post hoc tests and data are shown in means±s.d.
Citation: Journal of Endocrinology 223, 1; 10.1530/JOE-14-0317
The mechanistic basis for the synergy between BMP2 and Activin A in LβT2 cells is not fully understood but was reported to involve Smad2 and Smad3 and more recently BMP signaling via ALK3 (Wang et al. 2014). Consistently, we observe that CA-ALK3, but not CA-ALK2 or CA-ALK6, increases FSHB mRNA levels in LβT2 cells (Fig. 3D). We further find that, while BMP2 treatment alone has no detectable effect on FSHβ expression, BMP2 and Activin A together result in nearly twice as much FSHB mRNA production as that produced by Activin A alone (Fig. 3E). By contrast, AB215 alone has little effect on FSHB mRNA levels but inhibits Activin A induction of FSHβ (Fig. 3E). Western blot analysis shows that Activin A treatment triggers phosphorylation of Smad2 and Smad3 in LβT2 cells as expected (Fig. 3F). We further find that BMP2 treatment alone triggers Smad3 phosphorylation and augments Smad3 phosphorylation resulting from Activin A treatment, but that it does not affect Smad2 phosphorylation levels in the absence or presence of Activin A (Fig. 3F). By contrast, AB215 reduces basal phospho-Smad3 levels and blocks phosphorylation of Smad2 and Smad3 caused by Activin A treatment (Fig. 3F). Thus, BMP2 and AB215 have opposing effects on Activin A signaling via Smad2 and Smad3, providing an explanation for the opposing effects of these ligands on FSHβ regulation in LβT2 gonadotrophs.
Discussion
BMPs and Activins are both expressed endogenously in gonadotrophs. However, modulation of FSHB transcription by BMPs varies between species (Visser & Themmen 2014). In murine primary pituitary cells, various BMPs including BMP15, BMP7, BMP6, and BMP4 stimulate FSHB transcription (Huang et al. 2001, Otsuka & Shimasaki 2002, Nicol et al. 2008), whereas BMP6 and BMP4 inhibit FSHB transcription in ovine cells (Faure et al. 2005). In accordance with the stimulatory effect of BMPs in murine primary pituitary cells, it has been shown that BMP2 acts synergistically with Activin A to regulate FSHβ expression in LβT2 cells (Lee et al. 2007, Wang et al. 2014). BMP2 has been previously reported to induce non-canonical phosphorylation of Smad3 (Holtzhausen et al. 2014, Wang et al. 2014) consistent with our results showing that BMP2 triggers Smad3 phosphorylation in LβT2 cells (Fig. 3F). However, such non-canonical BMP2 signaling is clearly cell type specific, as the ALK2/3/6 inhibitor, LDB193189, severely inhibits Activin A signaling in LβT2 cells (Fig. 2B) but has no effect on Activin A signaling in HEK293T cells (Fig. 1C). While it has also been reported that BMP2 induces SMAD2 phosphorylation in LβT2 cells (Wang et al. 2014), we did not observe this effect (Fig. 3F). However, our results agree with previous results showing that ALK3, and not ALK2 or ALK6, is responsible for BMP2-induced Smad3 phosphorylation and subsequent augmentation of Activin-induced FSH production (Fig. 3D and F). Our results confirm other reports showing that SMAD3 is the predominant SMAD mediating Activin A induction factor of FSHβ synthesis (Wang et al. 2014). Additionally, the results of SMAD depletion studies (Bernard et al. 2010) have shown that SMAD3 but not SMAD2 is the main regulator of FSHβ expression. However, our data suggest that SMAD3 phosphorylation by itself is insufficient to fully induce FSHβ expression, as BMP2 alone causes SMAD3 phosphorylation but is unable to up-regulate FSHβ expression noticeably (Fig. 3E).
In an effort to find new methods of regulating FSHβ expression, we tested a BMP2/Activin A chimera, AB215, with type 1 receptor-binding properties and SMAD signaling specificity of BMP2, but an enhanced potency resulting from high-affinity type 2 receptor binding resembling that of Activin A (Allendorph et al. 2011). We have previously demonstrated that AB215 has a higher potency than BMP2 in preosteoblastic C2C12 cells (Allendorph et al. 2011) and herein we demonstrate that the same holds true in HEK293T cells (Fig. 1B). It is therefore intriguing to observe that the potencies of AB215 and BMP2 are indistinguishable in LβT2 cells (Fig. 1C). Generally, the potency of TGFβ superfamily ligands corresponds to their affinity for their higher affinity receptor type (Qian et al. 1996). AB215 has a higher affinity for activin type 2 receptors than for BMP type 1 receptors, while BMP2 has the opposite receptor preference (Allendorph et al. 2011). As AB215 also has a higher affinity for activin type 2 receptors than BMP2 has for its type 1 receptors (Allendorph et al. 2011), it follows that AB215 should have a higher overall potency than BMP2, and this is indeed what we observe in HEK293T and C2C12 cells. However, this argument is predicted to be true only when the number and distribution of type 1 and type 2 receptors are similar. If BMP type 1 receptors are sparse relative to activin type 2 receptors, then the potencies of BMP2 and AB215 can be similar as is observed in LβT2 cells. In such a situation, rapid binding of AB215 to activin type 2 receptors would be followed by slow binding to rare BMP type 1 receptors, while rapid binding of BMP2 to limiting type 1 receptors would be followed by relative rapid recruitment of abundant type 2 receptors (Fig. 4).
AB215 and BMP2 signaling in HEK293T cells and LβT2 cells. Diagram illustrating the signaling and crosstalk of Activin A, AB215, and BMP2 in HEK293T cells and LβT2 cells. BMP pathway activation inhibits Activin A signaling in HEK293T cells but either has no inhibitory effect (ARE-Lux induction) or enhances (FSHβ induction) Activin A signaling in LβT2 cells. AB215 inhibits Activin A signaling in both cell lines while BMP2 partially inhibits Activin A signaling in HEK293T cells but not in LβT2 cells. Full colour version of this figure available via http://dx.doi.org/10.1530/JOE-14-0317.
Citation: Journal of Endocrinology 223, 1; 10.1530/JOE-14-0317
AB215 and BMP2 signaling in HEK293T cells and LβT2 cells. Diagram illustrating the signaling and crosstalk of Activin A, AB215, and BMP2 in HEK293T cells and LβT2 cells. BMP pathway activation inhibits Activin A signaling in HEK293T cells but either has no inhibitory effect (ARE-Lux induction) or enhances (FSHβ induction) Activin A signaling in LβT2 cells. AB215 inhibits Activin A signaling in both cell lines while BMP2 partially inhibits Activin A signaling in HEK293T cells but not in LβT2 cells. Full colour version of this figure available via http://dx.doi.org/10.1530/JOE-14-0317.
Citation: Journal of Endocrinology 223, 1; 10.1530/JOE-14-0317
AB215 and BMP2 signaling in HEK293T cells and LβT2 cells. Diagram illustrating the signaling and crosstalk of Activin A, AB215, and BMP2 in HEK293T cells and LβT2 cells. BMP pathway activation inhibits Activin A signaling in HEK293T cells but either has no inhibitory effect (ARE-Lux induction) or enhances (FSHβ induction) Activin A signaling in LβT2 cells. AB215 inhibits Activin A signaling in both cell lines while BMP2 partially inhibits Activin A signaling in HEK293T cells but not in LβT2 cells. Full colour version of this figure available via http://dx.doi.org/10.1530/JOE-14-0317.
Citation: Journal of Endocrinology 223, 1; 10.1530/JOE-14-0317
It was initially surprising to find that AB215 does not mimic the ability of BMP2 to synergize with Activin A in promoting FSH expression in LβT2 cells (Fig. 3B), but rather inhibits Activin A-induced FSH production (Fig. 3C). It is, however, very clear from our results that, in the context of the cooperative signaling between BMP and Activin pathways, AB215 is a potent Activin A antagonist through direct competition for activin type 2 receptor binding (Figs 1E, 2D and 3F). An inhibitory effect of BMP2 on Activin A signaling is also consistent with competition between these two ligands for type 2 receptors in HEK293T cells (Fig. 1D). Unlike AB215, however, BMP2 only partially blocks Activin A signaling in these cells even at the highest BMP2 dose (3 mg/ml), suggesting that BMP2 antagonism of Activin A signaling requires a factor that is limiting. BMP2 assembles its signaling receptors by first binding to its higher affinity type 1 receptors and then recruiting its lower affinity type 2 receptors. Therefore, BMP type 1 receptors may be limiting in HEK293T cells such that only a fraction of the type 2 receptors can be sequestered by BMP2, thereby preventing BMP2 from completely blocking Activin A signaling (Fig. 1D). We also note that blocking BMP signaling with an inhibitor of ALK2/3/6 kinases partially alleviates the inhibitory effect of BMP2 on Activin A signaling in HEK293T cells, suggesting that this effect may also be caused in part by competition between the SMAD2/3 and SMAD1/5/8 pathways for common downstream effectors such as SMAD4 (Fig. 1B). Consistently, transfecting HEK293T cells with CA BMP type 1 receptors (CA-ALK2, CA-ALK3, and CA-ALK6) dramatically suppresses Activin A signaling (Fig. 1G). In contrast to its effects in HEK293T cells, BMP2 has no inhibitory effect on Activin A signaling in LβT2 cells. This suggests that: i) the number of BMP type 1 receptors is even more limiting in LβT2 than in HEK293T cells such that sequestration of ‘spare’ type 2 receptors by BMP2 does not produce an inhibitory effect and ii) that there is no limiting downstream effector in LβT2 cells (Fig. 2C). The first point, as explained above, is also supported by the equal potencies of AB215 and BMP2 in SMAD1/5/8 signaling in LβT2 cells and the second point is additionally suggested by the fact that transfecting LβT2 cells with CA type 1 BMP receptors (CA-ALK3 and CA-ALK6) does not result in decreased Activin A signaling (Fig. 2F). In accordance with the fact that AB215 is a potent inhibitor of Activin A, AB215 blocks the Activin-induced FSH production in LβT2 cells (Fig. 3C and E).
These data together indicate the role of AB215 both as a potent BMP2-like ligand and as an Activin A antagonist. Our results further reflect the cooperativity between the Activin and BMP pathways in LβT2 cells while clearly indicating the requirement for Activin A signaling in FSHβ induction. This work highlights the importance of cellular context for determining the biological outcome of both native and chimeric TGFβ superfamily ligands. It also demonstrates that AB215 could have medical implications where potent Activin A inhibition coupled with BMP pathway activation is desirable.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
Funding
This study was supported by the World Class University Program (R32-10215) and the Incheon Free Economy Zone of Korea (jCB).
Acknowledgements
The authors thank Dr Daniel Bernard at McGill University for the Luciferase plasmid constructs. They also thank Dr Mario Kuo and his team for protein preparation and Ms Hae Jung Han and Ms Jin Sook Lim at joint Center for Biosciences for their great technical assistance.
References
Ahmed Ebbiary NA, Lenton EA & Cooke ID 1994 Hypothalamic–pituitary ageing: progressive increase in FSH and LH concentrations throughout the reproductive life in regularly menstruating women. Clinical Endocrinology 41 199–206. (doi:10.1111/j.1365-2265.1994.tb02530.x)
Alarid ET, Windle JJ, Whyte DB & Mellon PL 1996 Immortalization of pituitary cells at discrete stages of development by directed oncogenesis in transgenic mice. Development 122 3319–3329.
Allendorph GP, Vale WW & Choe S 2006 Structure of the ternary signaling complex of a TGF-β superfamily member. PNAS 103 7643–7648. (doi:10.1073/pnas.0602558103)
Allendorph GP, Read JD, Kawakami Y, Kelber JA, Isaacs MJ & Choe S 2011 Designer TGFβ superfamily ligands with diversified functionality. PLoS ONE 6 e26402. (doi:10.1371/journal.pone.0026402)
Banaszewska B, Spaczyński RZ, Pelesz M & Pawelczyk L 2003 Incidence of elevated LH/FSH ratio in polycystic ovary syndrome women with normo- and hyperinsulinemia. Roczniki Akademii Medycznej w Bialymstoku 48 131–134.
Beau I, Touraine P, Meduri G, Gougeon A, Desroches A, Matuchansky C, Milgrom E, Kuttenn F & Misrahi M 1998 A novel phenotype related to partial loss of function mutations of the follicle stimulating hormone receptor. Journal of Clinical Investigation 102 1352–1359. (doi:10.1172/JCI3795)
Belchetz PE, Plant TM, Nakai Y, Keogh EJ & Knobil E 1978 Hypophysial responses to continuous and intermittent delivery of hypopthalamic gonadotropin-releasing hormone. Science 202 631–633. (doi:10.1126/science.100883)
Bernard DJ, Fortin J, Wang Y & Lamba P 2010 Mechanisms of FSH synthesis: what we know, what we don't, and why you should care. Fertility and Sterility 93 2465–2485. (doi:10.1016/j.fertnstert.2010.03.034)
Bilezikjian LM, Blount AL, Leal AM, Donaldson CJ, Fischer WH & Vale WW 2004 Autocrine/paracrine regulation of pituitary function by activin, inhibin and follistatin. Molecular and Cellular Endocrinology 225 29–36. (doi:10.1016/j.mce.2004.02.010)
Boitani C, Stefanini M, Fragale A & Morena AR 1995 Activin stimulates Sertoli cell proliferation in a defined period of rat testis development. Endocrinology 136 5438–5444. (doi:10.1210/endo.136.12.7588293)
Bouvattier C, Maione L, Bouligand J, Dodé C, Guiochon-Mantel A & Young J 2012 Neonatal gonadotropin therapy in male congenital hypogonadotropic hypogonadism. Nature Reviews. Endocrinology 8 172–182. (doi:10.1038/nrendo.2011.164)
Breckwoldt M, Roll H, Zahradnik HP, Amann K, Reck G & Peters F 1980 Plasma levels of FSH and LH in patients with gonadal dysgenesis during sequential estrogen and progestogen therapy. Archives of Gynecology 230 159–165. (doi:10.1007/BF02108271)
Chen W, Woodruff TK & Mayo KE 2000 Activin A-induced HepG2 liver cell apoptosis: involvement of activin receptors and smad proteins. Endocrinology 141 1263–1272. (doi:10.1210/endo.141.3.7361)
DePaolo LV, Hirshfield AN, Anderson LD, Barraclough CA & Channing CP 1979 Suppression of pituitary secretion of follicle-stimulating hormone by porcine follicular fluid during pro-oestrus and oestrus in the rat: effects on gonadotrophin and steroid secretion, follicular development and ovulation during the following cycle. Journal of Endocrinology 83 355–368. (doi:10.1677/joe.0.0830355)
Dierich A, Sairam MR, Monaco L, Fimia GM, Gansmuller A, LeMeur M & Sassone-Corsi P 1998 Impairing follicle-stimulating hormone (FSH) signaling in vivo: targeted disruption of the FSH receptor leads to aberrant gametogenesis and hormonal imbalance. PNAS 95 13612–13617. (doi:10.1073/pnas.95.23.13612)
Doherty E, Pakarinen P, Tiitinen A, Kiilavuori A, Huhtaniemi I, Forrest S & Aittomäki K 2002 Novel mutation in the FSH receptor inhibiting signal transduction and causing primary ovarian failure. Journal of Clinical Endocrinology and Metabolism 87 1151–1155. (doi:10.1210/jcem.87.3.8319)
Elkington JS & Blackshaw AW 1974 Studies in testicular function. I. Quantitative effects of FSH, LH, testosterone and dihydrotestosterone on restoration and maintenance of spermatogenesis in the hypophysectomized rat. Australian Journal of Biological Sciences 27 47–57.
Ericson J, Norlin S, Jessell TM & Edlund T 1998 Integrated FGF and BMP signaling controls the progression of progenitor cell differentiation and the emergence of pattern in the embryonic anterior pituitary. Development 125 1005–1015.
Faure MO, Nicol L, Fabre S, Fontaine J, Mohoric N, McNeilly A & Taragnat C 2005 BMP-4 inhibits follicle-stimulating hormone secretion in ewe pituitary. Journal of Endocrinology 186 109–121. (doi:10.1677/joe.1.05988)
Ferris HA & Shupnik MA 2006 Mechanisms for pulsatile regulation of the gonadotropin subunit genes by GNRH1. Biology of Reproduction 74 993–998. (doi:10.1095/biolreprod.105.049049)
Greenwald J, Vega ME, Allendorph GP, Fischer WH, Vale W & Choe S 2004 A flexible activin explains the membrane-dependent cooperative assembly of TGF-β family receptors. Molecular Cell 15 485–489. (doi:10.1016/j.molcel.2004.07.011)
Hall AK, Burke RM, Anand M & Dinsio KJ 2002 Activin and bone morphogenetic proteins are present in perinatal sensory neuron target tissues that induce neuropeptides. Journal of Neurobiology 52 52–60. (doi:10.1002/neu.10068)
Halvorson LM, Weiss J, Bauer-Dantoin AC & Jameson JL 1994 Dynamic regulation of pituitary follistatin messenger ribonucleic acids during the rat estrous cycle. Endocrinology 134 1247–1253.
Ho CC & Bernard DJ 2010 Bone morphogenetic protein 2 acts via inhibitor of DNA binding proteins to synergistically regulate follicle-stimulating hormone β transcription with activin A. Endocrinology 151 3445–3453. (doi:10.1210/en.2010-0071)
Hoak DC & Schwartz NB 1980 Blockade of recruitment of ovarian follicles by suppression of the secondary surge of follicle-stimulating hormone with porcine follicular field. PNAS 77 4953–4956. (doi:10.1073/pnas.77.8.4953)
Holtzhausen A, Golzio C, How T, Lee Y-H, Schiemann WP, Katsanis N & Blobe GC 2014 Novel bone morphogenetic protein signaling through Smad2 and Smad3 to regulate cancer progression and development. FASEB Journal 28 1248–1267. (doi:10.1096/fj.13-239178)
Huang HJ, Wu JC, Su P, Zhirnov O & Miller WL 2001 A novel role for bone morphogenetic proteins in the synthesis of follicle-stimulating hormone. Endocrinology 142 2275–2283. (doi:10.1210/endo.142.6.8159)
Jones KL, Mansell A, Patella S, Scott BJ, Hedger MP, de Kretser DM & Phillips DJ 2007 Activin A is a critical component of the inflammatory response, and its binding protein, follistatin, reduces mortality in endotoxemia. PNAS 104 16239–16244. (doi:10.1073/pnas.0705971104)
Kerr JB, Maddocks S & Sharpe RM 1992 Testosterone and FSH have independent, synergistic and stage-dependent effects upon spermatogenesis in the rat testis. Cell and Tissue Research 268 179–189. (doi:10.1007/BF00338067)
Kottler ML, Chou YY, Chabre O, Richard N, Polge C, Brailly-Tabard S, Chanson P, Guiochon-Mantel A, Huhtaniemi I & Young J 2010 A new FSHβ mutation in a 29-year-old woman with primary amenorrhea and isolated FSH deficiency: functional characterization and ovarian response to human recombinant FSH. European Journal of Endocrinology 162 633–641. (doi:10.1530/EJE-09-0648)
Lee KB, Khivansara V, Santos MM, Lamba P, Yuen T, Sealfon SC & Bernard DJ 2007 Bone morphogenetic protein 2 and activin A synergistically stimulate follicle-stimulating hormone β subunit transcription. Journal of Molecular Endocrinology 38 315–330. (doi:10.1677/jme.1.02196)
Lerchl A, Sotiriadou S, Behre HM, Pierce J, Weinbauer GF, Kliesch S & Nieschlag E 1993 Restoration of spermatogenesis by follicle-stimulating hormone despite low intratesticular testosterone in photoinhibited hypogonadotropic Djungarian hamsters (Phodopus sungorus). Biology of Reproduction 49 1108–1116. (doi:10.1095/biolreprod49.5.1108)
Marshall JC, Dalkin AC, Haisenleder DJ, Griffin ML & Kelch RP 1993 GnRH pulses – the regulators of human reproduction. Transactions of the American Clinical and Climatological Association 104 31–46.
Matsumoto AM, Karpas AE & Bremner WJ 1986 Chronic human chorionic gonadotropin administration in normal men: evidence that follicle-stimulating hormone is necessary for the maintenance of quantitatively normal spermatogenesis in man. Journal of Clinical Endocrinology and Metabolism 62 1184–1192. (doi:10.1210/jcem-62-6-1184)
Mendis SH, Meachem SJ, Sarraj MA & Loveland KL 2011 Activin A balances Sertoli and germ cell proliferation in the fetal mouse testis. Biology of Reproduction 84 379–391. (doi:10.1095/biolreprod.110.086231)
Nicol L, Faure M-O, McNeilly JR, Fontaine J, Taragnat C & McNeilly AS 2008 Bone morphogenetic protein-4 interacts with activin and GnRH to modulate gonadotrophin secretion in LβT2 gonadotrophs. Journal of Endocrinology 196 497–507. (doi:10.1677/JOE-07-0542)
Ortolano GA, Haisenleder DJ, Dalkin AC, Iliff-Sizemore SA, Landefeld TD, Maurer RA & Marshall JC 1988 Follicle-stimulating hormone β subunit messenger ribonucleic acid concentrations during the rat estrous cycle. Endocrinology 123 2149–2151. (doi:10.1210/endo-123-4-2149)
Ota F, Maeshima A, Yamashita S, Ikeuchi H, Kaneko Y, Kuroiwa T, Hiromura K, Ueki K, Kojima I & Nojima Y 2003 Activin A induces cell proliferation of fibroblast-like synoviocytes in rheumatoid arthritis. Arthritis and Rheumatism 48 2442–2449. (doi:10.1002/art.11249)
Otsuka F & Shimasaki S 2002 A novel function of bone morphogenetic protein-15 in the pituitary: selective synthesis and secretion of FSH by gonadotropes. Endocrinology 143 4938–4941. (doi:10.1210/en.2002-220929)
Qian SW, Burmester JK, Tsang ML, Weatherbee JA, Hinck AP, Ohlsen DJ, Sporn MB & Roberts AB 1996 Binding affinity of transforming growth factor-beta for its type II receptor is determined by the C-terminal region of the molecule. Journal of Biological Chemistry 271 30656–30662.
Robson NC, Wei H, McAlpine T, Kirkpatrick N, Cebon J & Maraskovsky E 2009 Activin-A attenuates several human natural killer cell functions. Blood 113 3218–3225. (doi:10.1182/blood-2008-07-166926)
Seishima M, Nojiri M, Esaki C, Yoneda K, Eto Y & Kitajima Y 1999 Activin A induces terminal differentiation of cultured human keratinocytes. Journal of Investigative Dermatology 112 432–436. (doi:10.1046/j.1523-1747.1999.00558.x)
Southworth MB, Matsumoto AM, Gross KM, Soules MR & Bremner WJ 1991 The importance of signal pattern in the transmission of endocrine information: pituitary gonadotropin responses to continuous and pulsatile gonadotropin-releasing hormone. Journal of Clinical Endocrinology and Metabolism 72 1286–1289. (doi:10.1210/jcem-72-6-1286)
Sulzbacher S, Schroeder IS, Truong TT & Wobus AM 2009 Activin A-induced differentiation of embryonic stem cells into endoderm and pancreatic progenitors-the influence of differentiation factors and culture conditions. Stem Cell Reviews 5 159–173. (doi:10.1007/s12015-009-9061-5)
Vale W, Rivier C, Hsueh A, Campen C, Meunier H, Bicsak T, Vaughan J, Corrigan A, Bardin W & Sawchenko P 1988 Chemical and biological characterization of the inhibin family of protein hormones. Recent Progress in Hormone Research 44 1–34.
Visser JA & Themmen APN 2014 Role of anti-Müllerian hormone and bone morphogenetic proteins in the regulation of FSH sensitivity. Molecular and Cellular Endocrinology 382 460–465. (doi:10.1016/j.mce.2013.08.012)
Wang Y, Ho CC, Bang E, Rejon CA, Libasci V, Pertchenko P, Hébert TE & Bernard DJ 2014 Bone morphogenetic protein 2 stimulates non-canonical SMAD2/3 signaling via the BMP type 1A receptor in gonadotrope-like cells: implications for follicle-stimulating hormone synthesis. Endocrinology 155 1970–1981. (doi:10.1210/en.2013-1741)
Wozney JM, Rosen V, Celeste AJ, Mitsock LM, Whitters MJ, Kriz RW, Hewick RM & Wang EA 1988 Novel regulators of bone formation: molecular clones and activities. Science 242 1528–1534.
Ying SY 1988 Inhibins, activins, and follistatins: gonadal proteins modulating the secretion of follicle-stimulating hormone. Endocrine Reviews 9 267–293. (doi:10.1210/edrv-9-2-267)
Yoon BS & Lyons KM 2004 Multiple functions of BMPs in chondrogenesis. Journal of Cellular Biochemistry 93 93–103. (doi:10.1002/jcb.20211)
Zhang Z, Zheng J, Zhao Y, Li G, Batres Y, Luo M, Wan M & Ying S 1997 Overexpression of activin A inhibits growth, induces apoptosis, and suppresses tumorigenicity in an androgen-sensitive human prostate cancer cell line LNCaP. International Journal of Oncology 11 727–736.
J W Jung is now at Interdisciplinary Graduate Program in Genetic Engineering, Seoul National University, 1 Gwanak-ro, Gwanak-gu, Seoul, Korea