Background
Janus kinase 3 (JAK3) belongs to a family of membrane-associated intracellular non-receptor tyrosine kinase proteins that mediate signals initiated by cytokine and growth factor receptors through the JAK/STAT pathway [
1,
2]. Other members of the Janus family include JAK1, JAK2 and tyrosine kinase 2 (TYK2). In contrast to the ubiquitous expression of JAK1, JAK2 and TYK2 [
3,
4], JAK3 is predominantly expressed in immune cells and is involved in signal transduction by interleukin (IL) receptors that share the common gamma chain (γc) of the type I cytokine receptor family [
5‐
8]. Non-receptor tyrosine kinases are involved in diverse cellular processes including proliferation, differentiation, cell migration and survival [
9‐
13]. In various systems, JAK signal transduction occurs via a well-characterized JAK/STAT pathway. JAKs activate this pathway by binding to cytokine receptors leading to the recruitment of STAT proteins, which are then phosphorylated. After undergoing JAK-mediated phosphorylation, the STAT transcription factors dimerize and translocate to the nucleus where they regulate the transcription of specific genes [
14,
15].
In mammals, the JAK/STAT pathway is the principal signaling mechanism for a wide array of cytokines and growth factors. Different JAKs and STATs are recruited based on the specific tissue and the receptors engaged in the signaling event [
16]. Although the canonical JAK/STAT pathway is simple and direct, the pathway components can regulate or be regulated by members of other signaling pathways, including those involving the ERK MAP kinase, phosphatidylinositide 3-kinases (PI3K), and receptor tyrosine kinase (RTK)/Ras/MAPK pathways [
17]. Effector proteins that contribute to JAK/STAT events include the signal-transducing adapter molecules (STAMs), which facilitate the transcriptional activation of specific target genes, and the stat-interacting proteins (StIP) that can associate with JAKs and unphosphorylated STATs to facilitate JAK/STAT pathway [
18‐
20]. Conversely, the suppressors of cytokine signaling (SOCS), the protein inhibitors of activated stats (PIAS) and the protein tyrosine phosphatases (PTP) are major negative regulators of the JAK/STAT pathway [
21‐
23]. These data indicate that various proteins outside the basic pathway machinery could influence the JAK/STAT signaling.
JAK3 encodes for a 125-kDa protein made of seven JAK homology (JH) domains. The carboxyl JH1 region of JAK3 contains the activation loop, a region that includes the tyrosine kinase catalytic domain [
15]. The kinase activity of JAK proteins depends on their phosphorylation at tyrosine residues in the activation loop of the kinase domain. Multiple sites of autophosphorylation have indeed been identified in this enzymatically active JH1 domain [
24‐
26]. The JH2 domain contains the catalytically inactive pseudo-kinase domain, which is tandemly linked to the N site of the JH1 domain and represents a unique feature of JAK proteins in contrast to other tyrosine kinases. Despite the lack of catalytic activity, the pseudo-kinase domain is required for suppression of basal activity of tyrosine kinases and for cytokine-inducible activation of signal transduction [
27]. The amino terminus region of JAK3 is composed of a SH2-like domain (JH3 and JH4 domains) and a 4.1, ezrin, radixin, moesin (FERM) homology domain (JH6 and JH7 domains). The FERM domain interacts with the kinase domain, positively regulates the catalytic activity and is critical for receptor binding, signal transduction and maintenance of kinase integrity [
28].
JAK3 is primarily expressed in hematopoietic cells and the JAK/STAT pathway has been widely investigated in immune cells, but JAK3 has also been found in a wide range of tissues of both hematopoietic and non-hematopoietic origin [
29]. We previously identified
JAK3 as a differentially expressed gene in granulosa cells of bovine dominant follicles using a gene expression profiling approach [
30].
JAK3 was identified among a list of other genes that were down-regulated in granulosa cells of bovine ovulatory follicles following human chorionic gonadotropin (hCG) injection as compared to growing dominant preovulatory follicles during the estrous cycle. It is well documented that the cyclic ovarian activity results in profound modifications that require spatio-temporal coordination of proliferation, apoptosis and differentiation of various cell types within the follicle leading to changes in gene expression. Of interest, granulosa cells play a critical role in these reproductive functions as they contribute to steroid hormone synthesis [
31], oocyte maturation [
32], and corpus luteum formation after ovulation [
33]. Many factors such as follicle-stimulating hormone receptor (FSHr) in small and growing follicles, luteinizing hormone receptor (LHr) in ovulatory follicles, steroid hormones (estradiol and progesterone) and growth factors are produced by GC and affect follicular growth, ovulation and differentiation into a functional corpus luteum. Consequently, the regulation of granulosa cell proliferation and function is complex and depends on the precise regulation and activation of specific target genes. This regulation is essential for normal follicular development and timely production of paracrine factors as it affects the physiological state of the dominant preovulatory follicle. For instance, the transcription of specific genes that control the growth of a bovine dominant preovulatory follicle is rapidly downregulated or silenced in granulosa cells as a result of LH-mediated increases in intracellular signaling [
30]. These observations demonstrate the critical importance of gene regulation studies during the final stages of follicular development as well as their interactions and mode of action. In this regard, we identified JAK3 as a candidate gene associated with follicular growth and dominance. We report JAK3 differential regulation and binding partners in bovine granulosa cells as well as its effects in cell proliferation.
Discussion
We report for the first time JAK3 regulation and protein interactions in the reproductive system using granulosa cells from ovarian follicles. We also demonstrated that a functional JAK3 affects phosphorylation of STAT proteins suggesting that JAK/STAT signaling is operating in these bovine reproductive cells. The major findings of this study are that: 1) the greatest expression of JAK3 is associated with the growing, estrogen-active follicle; 2) JAK3 expression is hormonally-regulated as it is significantly reduced by the endogenous LH surge and by hCG injection in a time-dependent manner, and in the corpus luteum after granulosa cells differentiation into luteal cells; and 3) JAK3 phosphorylates STAT3 in endometrial cells and increases cell viability since inhibition of JAK3 with JANEX-1 significantly reduces both the phosphorylation of STAT3 and cell viability. These observations indicate that JAK3 is associated with increased cell proliferation. It is known that JAK activation and the JAK/STAT pathway regulate cell proliferation, differentiation, migration and apoptosis depending on the signal, the tissue, and the cellular context [
10,
34,
35]. Because JAK3 expression in the OF and CL is considerably reduced compared to the growing follicle, it is conceivable that JAK3 might be involved in granulosa cell proliferation rather than in their differentiation.
Members of the JAK and STAT families constitute a crucial signaling system that has been the focus of extensive studies, notably in the function of immune cells. JAK3 is primarly expressed in leukocytes and is required for their development and function [
36]. JAK3 typically associates with the common γc subunit of cytokine receptors, becomes activated following cytokine binding to the receptor [
6], and phosphorylates other proteins, including STAT proteins, that mediate gene transcription [
37]. The range of proteins activated specifically by JAK3 as well as the protein-protein interactions in which JAK3 is involved could mediate different mechanisms in various systems. JAK3 has been shown to phosphorylate insulin receptor substrate-1 (IRS-1), IRS-2, PI3K/Akt and focal adhesion kinase (FAK) [
38], which contain SH2 or other phospho-tyrosine-binding domains. Other studies have used Drosophila to investigate the JAK/STAT pathway in ovarian cell migration, sex determination and stem-cell maintenance [
39,
40], and in macrophages infected with
F. tularensis, JAK3 is involved in the phosphorylation of p38MAPK [
41]. Herein, we observed that JAK3 strongest expression was found in the dominant follicle, which is active and growing while in the ovulatory follicle, following the endogenous LH surge or hCG injection, JAK3 expression was dramatically reduced. These results suggest a specific role for JAK3 in granulosa cells during follicular growth between the stages of small antral follicles to preovulatory follicles, prior to the LH surge, such as in granulosa cells proliferation through the activation or inhibition of target proteins.
Using a yeast two-hybrid screening, we identified and confirmed novel JAK3-interacting proteins in granulosa cells that can act as downstream target proteins for JAK3 signaling within the follicle. Some of these partners are known to be expressed in dominant follicles and participate actively in follicular development. Of interest, we showed that inhibin beta A (INHBA) interacts with JAK3 in granulosa cells of dominant follicles. Previous spatiotemporal expression studies showed greatest mRNA expression for
INHBA in estrogen active follicles [
42‐
44] and an increase in activin-A protein secreted in follicular fluid [
45]. Activin promotes granulosa cells proliferation and steroidogenesis and potentiates FSH actions on granulosa cells by increasing
FSH receptor expression [
46,
47], which underscores a key role for activin-A in the dominant follicle’s development. Our data confirmed expression of
INHBA mRNA and protein in dominant follicles in agreement with already published data. The interaction between JAK3 and INHBA could increase granulosa cells proliferation and participate in the growth of the dominant follicle into the ovulatory stage, prior to the LH surge.
In addition to INHBA, physical interactions were confirmed by co-IP between JAK3 and potential partners including LEPROTL1 and CDKN1B. Leptin receptor overlapping transcript-like 1 (LEPROTL1), also referred to as endospanin 2 [
48], is a transmembrane protein involved in the regulation of intracellular protein trafficking [
49]. It belongs to the OB-RGRP/VPS55 family and is widely expressed [
50]. LEPROTL1 was isolated from a human fetal brain cDNA library and was shown to possess a JAK binding site (Pro(46)-Ile-Pro(48)) [
50]. LEPROTL1 negatively regulates growth hormone (GH) receptor cell surface expression in liver and may play a role in liver resistance to GH during periods of reduced nutrient availability [
51]. Functionally, it has been shown that transgenic mice overexpressing LEPROTL1 displayed growth retardation and an impairment of GH-induced STAT5 phosphorylation in the liver [
51]. These studies indicate that LEPROTL1 expression decreased GH signaling while LEPROTL1 silencing increased GH signaling. Furthermore, recent studies demonstrated that fibroblast growth factor 21 (FGF21) inhibition of GH stimulatory actions on IGF-1 expression in chondrocytes depends on the intracellular activity of LEPROTL1 and LEPROT, another member of the family [
49]. Based on the available data and our findings, one could assume that JAK3 binding to LEPROTL1 may participate in the follicular growth process by reducing the potential negative feedback of LEPROTL1 on the GH signaling and increasing IGF-1 availability in the growing follicle. In this regard, it has been shown that GH increased IGF-1 secretion by ovine granulosa cells
in vitro [
52] supporting the idea that an increased GH signaling may regulate the ovarian function through IGF-1 production by granulosa cells.
The third JAK3 partner that was confirmed by co-IP was the cyclin-dependent kinase inhibitor 1B (CDKN1B or p27(KIP1)). CDKN1B is a cyclin-dependent kinase inhibitor that binds to and prevents the activation of cyclin E-CDK2 or cyclin D-CDK4 complexes, and thus controls the cell cycle progression at G1 [
53,
54]. The degradation of CDKN1B, which is triggered by its CDK-dependent phosphorylation and subsequent ubiquitination, is required for the cellular transition from quiescence to the proliferative state [
55]. CDKN1B acts either as an inhibitor or an activator of cyclin type D-CDK4 complexes depending on its phosphorylation state and/or stoichiometry. The phosphorylation of CDKN1B occurs on serine, threonine and tyrosine residues. Phosphorylation on Ser-10 is the major site of phosphorylation in resting cells and takes place at the G(0)-G(1) phase leading to protein stability [
56]. Phosphorylation on other sites is greatly enhanced by mitogens, growth factors, cMYC and in certain cancer cell lines [
57,
58]. As a result, the phosphorylated CDKN1B form found in the cytoplasm is inactive. Based on these reports, and in light of our current findings, it may be possible that CDKN1B binding to JAK3 leads to its phosphorylation and inactivation in granulosa cells of dominant follicle, thus allowing the progression of granulosa cell division and proliferation. Indeed, the greatest expression of CDKN1B in the corpus luteum suggests a role for CDKN1B in establishing the nonproliferative state, which is required for differentiation or for proper functioning of the differentiated luteal cells. Our findings are consistent with previous results showing that the luteinization process is associated with up-regulation of
CDKN1B (p27) that accumulated during initial phases of luteinization and remained elevated until termination of the luteal function [
59]. Our data are also consistent with the report that, in the bovine ovary, the expression of
CDKN1B (p27) mRNA was significantly reduced in granulosa cells from oestrogen-active dominant follicles as compared to oestrogen-inactive follicles [
60], meaning that cyclins and their inhibitors are associated with granulosa cell proliferation at specific follicular developmental stages.
Because of the difficulty of obtaining and culturing bovine primary granulosa cells from dominant follicles, we used a bovine endometrial cell line to demonstrate that JAK3 increased phosphorylation of STAT3 and that addition of JANEX-1, a specific JAK3 inhibitor, reduced the amount of JAK3-phosphorylated STAT3. These results suggest that the phosphorylation of STAT3 is directly linked to the presence of JAK3 in these cells. Moreover, the presence of JAK3 was also associated with increased cell viability since JANEX-1 significantly reduced the proportion of viable cells while overexpression of JAK3 delayed the decrease in cell viability induced by JANEX-1. Together, these data suggest the presence of a functional JAK3 and possibly JAK/STAT pathway in endometrial cells that may be involved in cell proliferation. During follicular development, JAK3 is predominantly expressed in the growing dominant follicle, suggesting that JAK3 may impact granulosa cell proliferation through the phosphorylation of target proteins including STAT transcription factors. We have shown that levels of phosphorylated STAT3 was stronger in small and dominant follicles as compared to ovulatory follicles and even stronger in small compared to dominant follicles. These observations could mean that STAT3 phosphorylation by JAK3 in granulosa cells may be associated with follicular growth. However, it was recently proposed that the stronger amount of pSTAT3 in small follicles was associated with granulosa cells death and follicular atresia [
61]. It would be relevant to verify whether JAK3 increases STAT3 phosphorylation in bovine granulosa cells of atretic small follicles as compared to healthy growing follicles. Although we have demonstrated the presence of pSTAT3 in granulosa cells and an increased STAT3 phosphorylation by JAK3 in endometrial cells, further functional studies should clarify the roles of pSTAT3 in granulosa cell death in small follicles and in granulosa cell proliferation in the growing dominant follicles following a potential activation by JAK3. As for STAT5, surprisingly, JAK3 overexpression seemed to reduce the amount of pSTAT5 in endometrial cells transfected with JAK3 although pSTAT5 is detected in non-transfected cells and that JANEX-1 reduced pSTAT5 amounts. These findings could mean that basal expression of JAK3 is required for STAT5 phosphorylation while overexpression of JAK3 leads to the inhibition of STAT5 phosphorylation.