SMAD-FOXL2 Regulation of FSHB: A Game of Human and Mouse

The pituitary gonadotropins, luteinizing hormone (LH) and follicle-stimulating hormone (FSH), are tightly regulated glycoproteins that are produced in a cyclic manner to orchestrate follicular maturation and ovulation (females) and spermatogenesis (males). These hormones share a constitutively produced α-subunits and distinct β-subunits. Thus, the production of the β-subunit is considered the rate-limiting step in gonadotropin production and secretion. A rise in FSH secretion occurs during the luteal–follicular period (or in the early morning hours of estrus in mice) and is a key determinant in the maturation of follicles and selection of the subsequent dominant follicle (1, 2). Notably, studies have elucidated an important “FSH threshold” requirement for individual follicles below which a given follicle will not develop (1, 3). FSH concentrations must surpass this threshold to promote follicular recruitment, growth, selection and dominance. Follicles exhibit different degrees of FSH sensitivity as they mature, and those follicles with the highest sensitivity benefit most from the increasing FSH levels. The exact mechanisms behind this rise in FSH β secretion have yet to be described. Hall et al. reported that the rise coincides with the increase in frequency of gonadotropin-releasing hormone (GnRH) pulses during the luteal–follicular transition (from every 4 hours to every 90 minutes) (1). This implicated GnRH as a regulator. However, this group also noted that the rise can be correlated with a decline in inhibin, a decrease in steroid negative feedback, and/or paracrine stimulation by pituitary activin.

Originally discovered as a component of ovarian follicular fluid, activin is now believed to be a paracrine factor that regulates FSH synthesis and secretion in rodents (4-6). Activin signals through the activin type IIA and IB or IC receptors on gonadotropes to transphosphorylate SMAD3 (homolog of Drosophila mothers against decapentaplegic), which complexes with SMAD4 in the cytoplasm. The SMAD complexes enter the nucleus and bind the Fshb promoter in combination with the transcription factor forkhead box L2 (FOXL2) (5, 6). The evidence for activin regulation of FSH in rodents is based on reductions in FSH in mice lacking activin receptors, or following Foxl2 or Smad4 deletion in gonadotropes (5-7).

The regulation of FSHB by activin in humans is less well characterized, as is highlighted in the current study by Ongaro et al. (7). This group, which has extensively characterized the regulation of Fshb in rodents, addresses the issues surrounding human FSHB studies. These include the lack of both a human gonadotrope cell line and a truly activin-responsive human FSHB promoter construct. The present study employs a clever approach using a “humanized” mouse model bearing a human FSHB transgene that contains all exons and introns along with sufficient 3′ and 5′ flanking sequences. This human FSHB construct has previously rescued fertility in murine Fshb knockouts and is known to depend on functional activin type IIA receptors.

Using male and female primary pituitary cultures expressing human FSHB and murine Fshb, the authors show that both human and mouse Fshb transcripts are stimulated by activin (A and B). However, when endogenous activin signaling through the Type I receptor is blocked, human FSHB expression is more stable, not significantly decreasing until 12 hours post-treatment (10 hours later than the decline in murine transcripts). Additionally, when transcription is blocked, human FSHB expression is maintained. The investigators conclude that this is due to the higher stability of human FSHB mRNA, which is logical given the significantly longer cycles in human females compared to rodents. Thus, while both murine and human transcripts appear to be dependent on activin signaling, murine Fshb stability may require continuous activin stimulation.

Deletion of Foxl2 in these human FSHB-expressing male and female cultures (depleting Foxl2 mRNA levels by 60-70%) results in significant decreases in basal and activin-stimulated murine Fshb and human FSHB expression. However, when both Foxl2 and Smad4 are ablated, expression of both murine Fshb and human FSHB are severely reduced. These in vitro studies suggest not only that activin regulates human FSHB, but also that human FSHB regulation is indeed dependent on SMAD-FOXl2 transactivation pathways.

To demonstrate this dependence in vivo, the authors crossed the floxed Foxl2/Smad4 mice expressing human FSHB with GRIC mice (co-expressing Cre-recombinase driven by gonadotropin-releasing hormone receptor to target gonadotropes). The design is based on the knowledge that the hFSHB transgene rescues the reproductive phenotype of Fshb knockout mice. They reason that if human FSHB expression depends on SMAD4 and FOXL2, then hFSHB should not be able to rescue the reproductive phenotype of these deletion mutants (which are Fshb deficient). Their hypothesis is confirmed as hFSHB fails to rescue the pituitary and gonadal deficiencies in mutant males and females. Furthermore, expression of both murine Fshb and human FSHB is significantly reduced in these deletion mutant mice.

This novel approach provides new insights as to potential differences in the activin signaling pathways. Their in silico analyses identify differences in the composite hFSHB versus Fshb SMAD/FOXL2 cis-elements indicating that the elements critical for Fshb promoter activity in the mouse are not perfectly conserved in humans. This may explain the reduced responsiveness to activin in the previous assays that tested human FSHB promoter reporters. They also suggest that there may be alternative or additional sites for binding to the human FSHB gene. Perhaps the most intriguing conclusion drawn by the authors is that there may be different mechanisms underlying murine and human FSHB mRNA stability. The investigators suggest that this could involve miRNAs, a hypothesis that has been previously tested in rats. This hypothesis makes sense in view of differential timing of the estrous cycle (4-5 days) versus the menstrual cycle (28 days). The stability of human FSHB mRNA clearly needs to be maintained for a longer period of time. Importantly, the current study by Ongaro et al. further implicates the mouse as a viable model for studies of both mouse and human FSHB regulation. Given the critical role of FSH in the maintenance of gonadal function, future work building on the current study exploring the regulation of human FSHB will define the pathways that can be targeted in the treatment of reproductive axis disorders.

Additional Information

Disclosure Summary: The authors have no conflicts.

Data Availability: All data generated or analyzed during this study are included in this published article or in the data repositories listed in References.