The function of BMP4 in the endometrium

Endometrium is a regenerative organ that undergoes bleeding, proliferation and the secretory phase under the control of female hormones [Refs Reference Critchley, Maybin, Armstrong and Williams39, Reference Murphy, Campo and Kim40]. Bone morphogenetic protein superfamily has been proven to regulate multiple endometrial functions, including proliferation, apoptosis, decidualization and immune cell infiltration [Ref. Reference Wei, Su, Leung, Li and Chen41].

In bovine, BMP4 and its receptors were detected primarily in endometrial epithelial cells during the estrous cycle and early pregnancy, with higher protein levels observed during pregnancy. Exogenous BMP4 treatment in endometrial epithelial cells significantly suppressed proliferation without altering caspase-3/8 expression, while reducing prostaglandin E2 (PGE2) and PGF2α concentrations. Furthermore, the mRNA level of BMP4 was upregulated by pro-inflammatory cytokines, including tumour necrosis factor-α (TNF-α) and interferon-γ (IFN-γ), suggesting a potential regulatory link between immune signalling and endometrial BMP4 activity [Ref. Reference Koizumi, Hayashi and Sakumoto4]. Decidualization refers to the transformation of endometrial stromal cells (ESCs) into decidual stromal cells (DSCs), which is primarily driven by progesterone and intracellular cyclic adenosine monophosphate (cAMP) stimulation in the secretory phase. During this process, DSCs will secrete two specific protein markers, prolactin (PRL) and insulin-like growth factor binding protein 1 (IGFBP1), which will promote trophoblast invasion and support embryo development [Refs Reference Gellersen and Brosens42, Reference Tong, Lv, Yang, Li, Li and Zhang43]. Impaired decidualization may result in reproductive diseases such as RIF and RSA [Refs Reference Tang, Xu, Xu, Liu, He, Li, Sun, Shuai, Mao, Zhao and Jin44, Reference Zhang, Kong, Cao, Zhu, Jiang, Wang, Jiang, Liu, Zhou, Kang, Zhen, Kong, Wu, Yan and Sun45]. BMP4 has been proven to regulate the process of decidualization [Refs Reference Huang, Dai, Chen, Li, Liu and Cheng3, Reference Yang, Liang, Chen, Zhang, Li, Liu and Yan46].

Our group’s new study showed that BMP4 downregulation impaired decidualization by downregulating IGFBP1 and PRL, as well as the transcription factor forkhead box O1 (FOXO1). BMP4 impaired trophoblast migration and invasion by regulating FOXO1 in an in vitro co-culture model of stromal and trophoblast cells [Ref. Reference Huang, Dai, Chen, Li, Liu and Cheng3]. Furthermore, O-fucosylation of BMP1 increased BMP4 level and activated the BMP/Smad signalling pathway to promote endometrial decidualization [Ref. Reference Yang, Liang, Chen, Zhang, Li, Liu and Yan46].

Beyond its direct effect on decidualization, emerging evidence from a recent study suggests that BMP4 also influences endometrial immune tolerance. Another recent study of our group demonstrated that BMP4 inhibition will polarize THP-1 to M2-type macrophages through activating autophagy [Ref. Reference Tang, Dai, Zhang, Wang, Tan, Gu, Chen, Wang, Liu, Cheng and Xia5]. M1-type macrophages primarily secrete pro-inflammatory cytokines, which impair trophoblast invasion and embryo implantation. But M2-type macrophages primarily secrete anti-inflammatory cytokines, which help build immune tolerance [Refs Reference Zhao, Li, Fu, Ren, Jiang and Meng47, Reference Wang, Li, Wang, Kwak-Kim and Liao48]. However, current studies on their function in the endometrium remain limited. Moreover, while macrophages have been investigated, other immune cell populations in the decidua, such as natural killer cells (NK), T cells and other cell types, have not been fully explored. Hence, further studies are needed to comprehensively understand the immune regulatory functions of BMP4 at the maternal-foetal interface and its crosstalk with other members in the BMP superfamily.

The function of BMP4 in the ovary

Our group has previously summarized the expression of BMP4 in the ovary, as well as its roles in ovarian development and ovarian steroidogenesis in 2021 [Ref. Reference Yang, Yang, Dai, Wang, Yang, Hu and Cheng2]. This review pointed out that the BMP4 expression pattern in ovarian tissue exhibits spatiotemporal and species-specific characteristics and is closely associated with ovarian physiology. Specifically, BMP4 plays an important role in steroidogenesis and follicular development. Given the importance in this field, we present an updated synthesis of the latest findings to extend and complement our earlier review. Folliculogenesis is a highly coordinated process involving the growth and maturation of ovarian follicles from primordial to preovulatory follicles [Ref. Reference Ezz and Balboula49]. During this process, granulosa cells (GCs) play a central role in supporting oocyte development, regulating follicular growth and producing steroid hormones that are essential for ovarian function [Ref. Reference Wang50]. GCs are tightly regulated by multiple local growth factors, including BMP4. BMP4 enhances GCs proliferation while suppressing the synthesis of estradiol and progesterone induced by follicle-stimulating hormone (FSH) in bovine [Ref. Reference Spicer, Schutz and Aad51], suggesting a complex role for BMP4 in the regulation of ovarian steroidogenesis. BMP4 suppresses the expression of steroidogenic acute regulatory protein (StAR), CYP11A1, as well as the NOTCH target gene hairy and enhancer of split 1 (HES1) in bovine GCs, highlighting the complex crosstalk between the NOTCH and BMP/Smad signalling pathways in regulating GCs function [Ref. Reference Li, Jing, Dang, Jia, Guo, Kebreab, Lyu and Zhao52]. In addition, BMP4 overexpression promoted GCs proliferation, whereas BMP4 knockdown increased apoptosis. BMP4 could induce Smad8 phosphorylation to regulate activin membrane binding inhibitor (BAMBI) expression, thereby promoting lipid synthesis in Wanxi White goose GCs [Ref. Reference Wei, Chen, Peng, Yu, Chang, Jin and Geng53]. An earlier review pointed out the potential role of BMP4 in cumulus cell expansion [Ref. Reference Yang, Yang, Dai, Wang, Yang, Hu and Cheng2]; subsequent studies also provide support. BMP4 is expressed in cumulus cells, whose protein is localized in both the cytoplasm and nucleus in bovine. BMP4 knockdown increases apoptosis and inhibits proliferation by upregulating cysteine-aspartic acid protease 3 (Caspase-3) and bcl-2 associated x protein (BAX), while downregulating B-cell lymphoma 2 (Bcl-2) and cell cycle arrest at the G0/G1 phase [Ref. Reference Tian, Li, Wang, Hao, Zou, Pang, Zhao, Zhu and du54].

The function of BMP4 in ovarian steroidogenesis

The development and maturation of ovarian follicles are tightly regulated by FSH and luteinizing hormone (LH), whose secretion is governed by hypothalamic gonadotropin-releasing hormone (GnRH) [Refs Reference Conn and Crowley55, Reference Kaiser, Jakubowiak, Steinberger and Chin56]. Moreover, the secretion of FSH and LH is regulated by ovarian-derived steroid hormones, activin and inhibin through feedback mechanisms [Refs Reference Ling, Ying, Ueno, Shimasaki, Esch, Hotta and Guillemin57, Reference Santoro, Filicori and Crowley58]. Our previous review has summarized the role of BMP4 in ovarian steroidogenesis. Overall, BMP4 could regulate FSH receptor (FSHR) expression in both theca and GCs, although these effects vary across species. In addition, BMP4 is also demonstrated to regulate the expression of oestrogen, progesterone, as well as anti-Müllerian hormone (AMH) [Ref. Reference Yang, Yang, Dai, Wang, Yang, Hu and Cheng2]. A new study found that BMP4 inhibited FSH- and IGF1-induced progesterone and oestrogen production in primary GCs isolated from bovine small and large follicles. Notably, BMP4 exerted a more pronounced inhibitory effect on oestrogen production in large follicles [Ref. Reference Spicer, Schutz and Aad51]. This indicates BMP4 may exert a direct effect on ovarian steroidogenesis. In addition, BMP4 stimulates Noggin4 expression, and Noggin4 knockdown reduces the expression of the steroidogenic genes StAR and CYP11A1 in chicken GCs [Ref. Reference Gong, Li, Sun, Kang and Jiang59]. This finding highlights a potential paracrine regulatory mechanism of BMP4-mediated ovarian steroidogenesis.

Beyond its well-established roles in regulating FSH, oestrogen and progesterone synthesis, recent five-year studies have highlighted that BMP4 could also modulate androgen synthesis, which was not fully addressed in previous reviews. BMP4 has been shown to negatively regulate androgen synthesis by suppressing CYP17A1 expression, thereby inhibiting LH-induced androgen production in bovine theca cells [Refs Reference Glister, Regan, Samir and Knight60, Reference Glister, Satchell and Knight61]. And exogenous BMP4 treatment suppressed androgen synthesis in theca cells and promoted oestrogen production in GCs in a dehydroepiandrosterone (DHEA)-induced PCOS mouse model [Refs Reference Liu, du, Ding, Dou, Zhang, Wu, Qian, Zhang, Tang and Xu62]. Our new study comfirmed that exogenous dihydrotestosterone (DHT) was found to induce BMP4 expression in a vitro human granulosa tumour cell line (KGN) in a dose-dependent manner. Overexpression of BMP4 mitigated the DHT-induced glycolysis and autophagy reduction, suggesting its protective role in ovary function [Ref. Reference Gu, Dai, Xiang, Chen, Yang, Tan, Wang, Liu and Cheng63]. Moreover, intratesticular administration of BMP4 to rats did not affect testicular and epididymal weights, but serum testosterone levels in rats were downregulated in a dose-dependent manner. Smad1/4/5 was activated in BMP4-treated Leydig cells, and ERK1/2 and AMPL pathways were also activated, suggesting that BMP4 inhibits androgen synthesis by activating the above pathways [Ref. Reference Li, Fang, Chen, Quan, Wang and Ge64]. Evidence from other tissues also supports BMP4’s role in androgen synthesis. In the H295R adrenocortical cell line, BMP4 reduced DHEA production by downregulating CYP17A1 via SMAD/1/5/8 activation [Ref. Reference Rege, Nishimoto, Nishimoto, Rodgers, Auchus and Rainey65]. Additionally, mRNA expression levels of BMP4 and BMP type I receptor (Bmpr1a) were upregulated in intestinal stem cells of orchiectomized male mice, and supplementation with DHT suppressed BMP4 expression [Ref. Reference Yu, Li, Xu, Zhang, Ma, Liang, Lu, Ji, Liu, Chen and Li66]. These findings further suggest that BMP4 may broadly influence androgen synthesis, reinforcing its potential role in regulating ovarian and androgen production.

The function of BMP4 in adipogenesis

There are two main types of adipose tissue: WAT and BAT. WAT serves mainly as an energy storage in response to nutritional need. While BAT mediates non-shivering thermogenesis by uncoupling mitochondrial respiration from ATP production [Refs Reference Altinova68–Reference van Marken Lichtenbelt71]. Adipogenesis, a process from MSCs to fully mature adipocytes, is divided into two main stages. During the first stage, MSCs commit to the adipogenic lineage under the control of critical transcription factors, especially peroxisome proliferator-activated receptor γ (PPARγ) and C/EBPα. The second stage is the terminal differentiation of the adipogenic lineage into fully mature adipocytes [Refs Reference Al Haj, Rey, Giallongo, Colli, Marzani, Giuliani, Gorio, Zuccotti, Di Giulio and Carelli72, Reference Lai, Wang, Tang, Bian, Lu, He, Huang, Liu, Zhou, Liu, Tao and Tang73, Reference Tang and Lane74]. BMP4 plays a crucial role in directing MSCs toward the adipocyte lineage [Refs Reference Bowers, Kim, Otto and Lane6, Reference Huang, Song, Li, Hu, He, Liu, Lane and Tang7]. But the role of BMP4 in inducing pre-adipocytes toward BAT or WAT remains controversial. One study demonstrated that BMP4 promotes white-to-brown adipocyte transition by increasing the expression of the BAT marker uncoupling protein 1 (UCP1) and decreasing the WAT marker transcription factor 21 (TCF21) in human subcutaneous adipose stem cells [Ref. Reference Elsen, Raschke, Tennagels, Schwahn, Jelenik, Roden, Romacho and Eckel75]. BMP4 promotes pre-adipocytes toward beige or brown adipose phenotypes and increases PGC1α expression through activating PPARγ transcriptional activity. But this process can be inhibited by the BMP antagonist Gremlin 1 (GREM1) [Refs Reference Huang, Song, Li, Hu, He, Liu, Lane and Tang7, Reference Gustafson, Hammarstedt, Hedjazifar, Hoffmann, Svensson, Grimsby, Rondinone and Smith8]. Another study also demonstrated that BMP4 promotes the browning or beiging program by inducing UCP1 expression in mouse-derived C3H10T1/2 cells [Ref. Reference Xue, Wan, Zhang, Zhang, Ye and Li76]. In addition, BMP4, secreted by adipocyte cells, could activate M2 type macrophage and inhibit M1 type macrophage to induce beige adipocyte shift [Ref. Reference Qian, Tang, Li, Liu, Zhang, Huang, Xue, Yu, Guo, Gao, Liu, Sun, Li, Jia and Tang10]. Another study found that BMP4 increased the WAT marker (Resistin) but decreased the BAT marker UCP1, suggesting that BMP4 promotes a shift from a brown to a white-like adipocyte phenotype. In addition, BMP4 impaired BAT bioenergetics, including decreasing the mitochondrial number, maximal fatty acid oxidation capacity and tricarboxylic acid (TCA) cycle activity. This may be due to activating SMAD signalling pathway or suppressing lipolysis ability [Ref. Reference Modica, Straub, Balaz, Sun, Varga, Stefanicka, Profant, Simon, Neubauer, Ukropcova, Ukropec and Wolfrum9]. The contradictory results may be due to differences in the origins of adipocytes across studies, highlighting the need for further research to clarify the mechanisms.

The function of BMP4 in insulin resistance and inflammation

Obesity is commonly characterized by chronic systemic inflammation. This inflammatory state contributes to metabolic dysregulation, including insulin resistance, and is associated with the development of various diseases such as cardiovascular disease and PCOS [Refs Reference Piche, Tchernof and Despres77, Reference Marinelli, Napoletano, Straccamore and Basile78]. A study pointed out that serum BMP4 concentrations positively correlated with adipocyte hypertrophy and whole-body insulin sensitivity [Ref. Reference Modica, Straub, Balaz, Sun, Varga, Stefanicka, Profant, Simon, Neubauer, Ukropcova, Ukropec and Wolfrum9], suggesting a potential role for BMP4 in regulating insulin signalling. Another study also found that serum BMP4 levels increased with age in a type 2 diabetes animal model. This may be due to BMP4 promoting insulin resistance by activating PKC-θ, which in turn attenuates insulin signaling [Ref. Reference Chattopadhyay, Singh, Gupta and Surolia11]. In addition, elevated BMP4 levels in diabetic animals were found to inhibit glucose-stimulated insulin secretion through upregulation of calbindin 1 [Ref. Reference Christensen, Jacobsen, Wendt, Mollet, Friberg, Frederiksen, Meyer, Bruun, Eliasson and Billestrup79]. Collectively, these studies indicate that BMP4 may represent an important modulator of insulin signalling.

Chronic low-grade inflammation is increasingly recognized as a hallmark of obesity. Emerging evidence suggests that BMP4 may participate in regulating obesity-associated inflammatory responses. Increased BMP4 expression has been reported in the myocardium and aorta of obese mice, where BMP4 was associated with elevated inflammatory cytokine levels, and BMP4 knockdown alleviated local inflammatory responses [Ref. Reference Wu, Chen, Shen, Wang and Zong80]. These findings suggest that BMP4 may exert distinct effects on inflammatory pathways depending on the tissue context. Further studies are required to clarify the role of BMP4 in obesity associated inflammation and its contribution to obesity-related metabolic complications.

BMP4 is involved in polycystic ovary syndrome (PCOS)

PCOS is a common female reproductive disorder that affects approximately 10%–30% of women worldwide [Ref. Reference Teede, Tay, Laven, Dokras, Moran, Piltonen, Costello, Boivin, Redman, Boyle, Norman, Mousa, Joham, Arlt, Azziz, Balen, Bedson, Berry, Boivin, Boyle, Brennan, Brown, Burgert, Busby, Ee, Garad, Gibson-Helm, Harrison, Hart, Hopkins, Hirschberg, Ho, Hoeger, Jordan, Legro, Li, Lujan, Ma, Mansfield, Marsh, Mocanu, Mol, Mormon, Norman, Oberfield, Patel, Pattuwage, Peña, Redman, Rombauts, Romualdi, Shah, Spritzer, Stener-Victorin, Tehrani, Thangaratinam, Thondan, Vanky, Wijeyaratne, Witchel, Yang, Yildiz, Alesi, Alur-Gupta, Avery, Khomami, Benham, Bidstrup, Chua, Cooney, Coster, Ee, Fitz, Flanagan, Forslund, Jiskoot, Kazemi, Kempegowda, Louwers, Lujan, Melin, Melson, Mengistu, Naderpoor, Neven, Pastoor, Rocha, Sabag, Subramanian and Tan94]. The commonly used clinical diagnostic criteria are still the Rotterdam criteria, which require that patients meet at least two of the following symptoms: 1) oligo- or anovulation, 2) clinical and/or biochemical signs of hyperandrogenism, 3) polycystic ovaries, and exclusion of other etiologies (congenital adrenal hyperplasia, androgen-secreting tumours and Cushing’s syndrome) [Ref. 95]. However, the exact mechanism of PCOS is still obscure. Current studies have investigated the role of BMP4 in PCOS using both human samples and androgen-induced PCOS animal models (Table 1).

In humans, BMP4 expression has been examined in peripheral blood or GCs by using ELISA assay, RT-PCR or immunohistochemistry. A study including 8 PCOS patients and 6 healthy controls found no significant difference in BMP4 mRNA expression in GCs between PCOS and control groups [Ref. Reference Khalaf, Morera, Bourret, Reznik, Denoual, Herlicoviez, Mittre and Benhaim87]. In contrast, another study that involved 40 PCOS and 20 healthy controls found BMP4 mRNA expression downregulated in GCs of PCOS patients when compared to healthy controls [Ref. Reference Wu, Wang, Xu, Liu, Bi, Wang and Zhang90]. To date, only two studies have examined BMP4 expression in GCs from PCOS patients, and they reported differently. These discrepancies may be due to differences in ethnicity and the small sample sizes. Therefore, additional studies with larger sample sizes and a diverse population are needed. Circulating BMP4 levels have also been investigated in PCOS patients. The paper published in 2013 examined serum BMP4 levels in 20 PCOS patients and found that BMP4 expression was undetectable across all samples. This finding may be attributed to the lower detection limit of the ELISA assay used (31.2 pg/mL) [Ref. Reference van Houten, Laven, Louwers, McLuskey, Themmen and Visser88]. However, a recent 2025 paper compared serum BMP4 levels between 40 PCOS patients and 40 healthy controls. When stratified by BMI, BMP4 levels were significantly upregulated in normal-weight controls compared with normal-weight PCOS patients. However, no significant difference was observed between PCOS patients and healthy controls when BMI was not considered [Ref. Reference Bakar, Arpa, Kuruca and Şentürk92]. Hence, these findings suggest that BMP4 may be more closely associated with metabolic status, such as obesity. In addition, circulating BMP4 levels appear to vary with androgen status. One study reported that serum BMP4 was upregulated in hyperandrogenic PCOS when compared with the non-hyperandrogenic PCOS group [Ref. Reference Gu, Dai, Xiang, Chen, Yang, Tan, Wang, Liu and Cheng63].

Hyperandrogen-induced PCOS mouse models enable direct examination of ovarian tissues, which is difficult to achieve in human studies because of ethical constraints. Two studies each included 10 PCOS mouse models and 10 healthy control mouse models to investigate the ovarian BMP4 expression. However, the findings were inconsistent. One study reported decreased BMP4 mRNA and protein level in the PCOS mouse model ovary by using RT-PCR and western blot [Ref. Reference Gu, Dai, Xiang, Chen, Yang, Tan, Wang, Liu and Cheng63]. Whereas another study observed increased BMP4 mRNA expression in the PCOS mouse model ovary by RT-PCR [Ref. Reference Tang, Huang, Wang, Ren, Chen, Xiao, Wang, Pan, Tsang, Zabel, Ma, Zhao and Zhang89]. Liu et al. found that the BMP4-Smad pathway could regulate ovarian steroidogenesis by inhibiting androgen synthesis in theca cells and promoting estrogen production [Ref. Reference Liu, du, Ding, Dou, Zhang, Wu, Qian, Zhang, Tang and Xu62]. In addition, a 2024 study combined transcriptomic and DNA methylation analyses of ovarian GCs from four DHEA-induced PCOS-like mice and four control mice. BMP4 was identified as a hypomethylated and upregulated gene, suggesting that aberrant DNA hypomethylation may contribute to its increased expression in PCOS through an epigenetic mechanism [Ref. Reference Cao, Li, Wang, Zhang, Guo, Zhao, Guo, Li and Nashun91]. It is known that PCOS is always associated with obesity and insulin resistance [Ref. Reference Zeng, Xie, Liu, Long and Mo96]. A recent study indicates that in a high-fat diet-induced mouse obese model, BMP4 expression is upregulated in ovarian follicles, along with higher levels of inflammatory cytokines, oxidative stress (ROS), DNA damage and lipid accumulation [Ref. Reference Wang, Chen, Wei, Guo, Li, Han, Wang, Zhu, Zhang, Li and Dai83].

BMP4 is Involved in recurrent spontaneous abortion (RSA)

RSA is defined as two or more consecutive miscarriages before 20 weeks of gestation with the same sexual partner. The incidence of RSA is difficult to estimate because the criteria of RSA are still somewhat controversial. Some women also conceal their history of abortion for social or cultural reasons [Ref. Reference Dimitriadis, Menkhorst, Saito, Kutteh and Brosens97]. However, there are still several clinical studies with large samples which allow us to get some approximate idea of the incidence of miscarriages. In Europe, RSA affects about 1%–2% of couples, according to guidelines from the European Society for Human Reproduction and Embryology [Ref. Reference Bender Atik, Christiansen, Elson, Kolte, Lewis, Middeldorp, Mcheik, Peramo, Quenby, Nielsen, van der Hoorn, Vermeulen and Goddijn98]. Therefore, it is crucial to clarify the pathogenesis and treatment of RSA.

Although studies on BMP4 in RSA remain limited, accumulating evidence in recent years has highlighted its potential importance in RSA pathogenesis. Wei et al. reviewed the important role of the BMP family in the menstrual cycle and pregnancy [Ref. Reference Wei, Su, Leung, Li and Chen41]. In our group’s previous studies, BMP4 expression was found to be reduced in both decidual and villous tissues from RSA patients, highlighting its complex regulatory roles at the maternal–foetal interface. Specifically, decreased BMP4 impaired decidualization by modulating FOXO1 expression, suppressed trophoblast invasion partly via runt-related transcription factor 2 (RUNX2) and promoted macrophage M1 polarization through inhibition of autophagy [Refs Reference Huang, Dai, Chen, Li, Liu and Cheng3, Reference Tang, Dai, Zhang, Wang, Tan, Gu, Chen, Wang, Liu, Cheng and Xia5]. Furthermore, a case–control study analysing 70 RSA and 100 healthy controls in Iran reported that the AA genotype of BMP4 rs121912765 polymorphism was significantly enriched in RSA patients [Ref. Reference Mazdapour, Ashkezari and Seifati93]. A previous study has also found BMP4 to be associated with embryonic development, suggesting the importance of BMP4 in pregnancy. scRNA-seq analysis and spatiotemporal analysis of mouse extraembryonic ectoderm (ExE) revealed a steep increase in the expression level of Bmp4 at day 7 of embryonic development, when Bmp4 is present in the early neonatal mesoderm and has the highest level of expression in the extraembryonic mesoderm (ExM), amniotic membrane and urothelium. Furthermore, germline Bmp4-KO embryos exhibited developmental delay and had no urothelium visible, indicating impaired developmental competence [Ref. Reference Hadas, Rubinstein, Mittnenzweig, Mayshar, Ben-Yair, Cheng, Aguilera-Castrejon, Reines, Orenbuch, Lifshitz, Chen, Elowitz, Zernicka-Goetz, Hanna, Tanay and Stelzer99]. Although the role of BMP4 in RSA remains unclear, its importance in embryonic development and our current findings suggest that it plays a key role in pregnancy and warrants further investigation.