Transcription factors of the SOX family were first discovered in mammals in 1990. The family is based on the conserved high migration group (HMG) box genes of the mammalian testis determinant Sry. Generally, proteins that contain an HGM domain and have 50% or higher amino acid similarity to the HMG are called SOX proteins (1). This family is subdivided into eight subgroups, from A-H, each subgroup contains 1–3 members ( Table 1 ) (2). The SOX gene family encodes transcription factors that are conserved across species and participate in important developmental processes (3). The sex-determining region Y box 9 (SOX9) belongs to this family (4). According to the amino acid sequence of the HMG domain and transactivation and dimerization domain, SOX9, SOX8, and SOX10 are grouped into the E subgroup (5). SOX9 protein contains an HMG box DNA binding domain that recognizes (A/T) CAA (T/A) G DNA sequences and controls the expression of target genes (6). It also contains a transcription activation domain located at the C-terminus (7) and plays an important role in inducing tissue and cell morphogenesis, survival (8), and regulation of many developmental processes (9); for example, embryonic development, lineage commitment, and stem cell maintenance (10). SOX9 expression is elevated in numerous types of cancer, including lung, prostate, skin, brain, colorectal, pancreatic, and breast cancer (11–17). These studies indicate that SOX9 acts as an oncogene in many cancers. Several additional studies have shown that SOX9 is involved in the formation of cancer because an increase in its level is conducive to the transformation of stem cells. Furthermore, high levels of SOX9 are related to the tumor grade, poor prognosis, and poor survival of some types of cancer (18). In rodents, the expression of Sry initiates the downstream signal cascade by directly regulating SOX9 ( 19). However, key SOX9 regulatory genes in most human tissues and cancers have yet to be established; it maybe that they are cell type and developmental stage specific (9). It has been noted that SOX9 has the opposite function of promoting and inhibiting proliferation, which indicates that its functions in proliferation are diverse and vary according to different environments (20). Therefore, the expression and function of SOX9 alters in different human cancers mainly by regulating the activity of cancer stem cells (CSCs), and as a tumor suppressor under certain circumstances (18).

Gynecological malignancies are tumors that occur in the female reproductive system and seriously threaten the lives of patients (21). Common gynecological malignancies include: ovarian cancer (OC), cervical cancer (CC), and endometrial cancer (EC). To date, the molecular mechanisms related to the incidence and development of gynecological malignancies are still unclear. This makes it particularly pertinent to discover their common causative molecule and to provide an effective therapeutic target. In recent years, studies have found that multiple mechanisms are involved in regulating the expression of the SOX9 gene, leading to the occurrence and development of gynecological malignancies. In this review, we discuss the prognostic value of SOX9 expression and the potential of targeting SOX9 for gynecological malignancy treatment. We also discuss progress regarding the role of SOX9 in gynecological malignancy pathogenesis through its mediation of important mechanisms, including tumor initiation and proliferation, apoptosis, migration, invasion, chemoresistance, and stem cell maintenance.

SOX9 is located at 17q24.3~q25.1 in humans. It is 3934 bps in length, has three exons, two introns, and an open reading frame (ORF). Its coding product is a polypeptide containing 509 amino acids (22). A total of 79 amino acids from positions 104 to 182 of the polypeptide chain constitute the HMG box, and it has 71% homology with the HMG box of the Sry gene (23). Its HMG domain contains two nuclear localization signal (NLS) sequences (24) and one leucine-rich nuclear export signal (NES) sequence (25). These NLS sequences and the NES sequence can be activated by different pathways, enabling SOX9 transcription factors to be present in both the nucleus and cytoplasm, and allowing regulation of the expression of related genes. NLS located at the N-terminus promotes nuclear translocation of SOX9 by binding to calmodulin activated by calcium ions, which can be inhibited by calmodulin specific antagonists (26). NLS located at the C-terminus interacts with importin β to form a complex (27), which then, via the RAN GTP-dependent pathway, mediates SOX9 to complete nuclear transport through the nuclear pore (28). Whereas, the NES sequence, located between the two NLS sequences, can interact with chromosome region maintenance 1 to mediate the nuclear export of SOX9 (25). SOX9 is a nuclear protein (24), and it is generally believed that SOX9 nuclear expression has its function (29). The structure and nuclear transport and export mechanisms of SOX9 protein are shown in Figure 1 .

The World Health Organization (WHO) categorizes ovarian tumors according to their histological differentiation, defined as epithelial, sex-cord stromal, and germ cell tumors, of which epithelial ovarian tumors constitute the most common type of OC (30). According to the latest data, OC ranks eighth in the world for female cancers, with 313 959 new cases and 207 252 deaths in 2020 (31).

CC is one of the most common malignant tumors in female patients. Although early detection of CC improves the prognosis, it is still the fourth most common cause of death in women and ranks first in gynecological malignancies, with 604 127 new cases and 41 831 deaths in 2020 (31). Therefore, it is necessary to explore the detailed mechanism of CC progression and to identify new effective treatment targets.

EC is a malignant tumor of the endometrial epithelium and one of the most common tumors of the female reproductive system (75), with 417 367 new cases and 97 370 deaths in 2020 (31). It is divided into type I, which is estrogen-related and accounts for 75%–90% of EC, and type II, which is not estrogen-related (76). Type I tumors have a good prognosis and include grade 1–2 EC, while type II malignancies include grade 3 EC, uterine serous carcinoma (USC), and clear cell carcinoma, which have a poor prognosis (77). Most EC studies suggest a role for SOX9 as an oncogene. In EC (G1, G2, G3 tumors), atypical hyperplasia and biopsy of normal endometrial specimen tissue (proliferative and secretory stages) samples and nuclear expression of SOX9: (1) was usually found in the epithelial component, but not in the stromal component; (2) was significantly higher in the proliferative phase than in the secretory phase; (3) was significantly higher in the late secretory phase than the early secretory phase; (4) was significantly higher in atypical proliferative lesions than in normal and endometrial cancer tissues; (5) was significantly higher in EC tissue than normal endometrial tissue; (6) was increased significantly and progressively from G1 to G2 to G3 tumors; (7) NF-κB as well as AKT transcriptionally upregulated SOX9 expression in EC cells, then activated the p14^ARF/p53/p21^WAF1 pathway, resulting in inhibition of EC cell proliferation (78). Although these findings suggest an important role for SOX9 in endometrial cell proliferation and carcinogenesis, it cannot explain why the expression of SOX9 in EC is higher than in normal endometrial tissue. Over-expression of SOX9 in a mouse model induced alterations in the tissue structure of the reproductive tract in female mice and played a role in the development of histological lesions similar to endometrial polyps and hyperplasia in humans (79); endometrial hyperplasia is a precancerous lesion of EC (80). This suggests that SOX9 is involved in the pathogenesis of endometrial diseases and may contribute to the formation of EC. SOX9 was significantly more expressed in the UCS than in normal control endometrial tissue in USC of TCGA and GETX datasets. Meanwhile, a 4-gene (KRT23, CXCL1, SOX9, and ABCA10) signature robustly predicts OS and recurrence-free survival(RFS) of USC (81). Furthermore, the expression of both mRNA and protein of SOX9 was significantly higher in tumor tissues than in paired adjacent tissues. The mechanism was that Circ_0109046 increased SOX9 expression by binding miR-105, which further activated the Wnt/β-catenin signaling pathway and promoted EC cell proliferation and metastasis (82). In conclusion, upregulation of SOX9 can promote endometrial cell proliferation, induce endometrial precancerous lesions, and predict the prognosis of EC patients in combination with the expression of other genes.

However, in the study by Opławski et al., the expression of SOX9 was downregulated in endometrioid endometrial adenocarcinomas compared to normal endometrium (83). The possible reason for this result is the different role of SOX9 in different EC subtypes; therefore, these roles need to be further investigated. In the same study, they used sequencing to identify three differentially expressed miRNAs, using the mirTARtool to suggest that miR-144 and miR-30d could target SOX9 ( 83). This further suggests that miRNA regulation of SOX9 expression leads to epithelial mesenchymal transition (EMT) development and ultimately to endometrioid adenocarcinogenesis. The regulation mechanism of SOX9 in EC is shown in Figure 4 .

Uterine carcinosarcoma is a rare gynecological malignancy, accounting for 1%–3% of malignant tumors of the female reproductive tract (84). It is a highly aggressive, biphasic malignancy with carcinomatous and sarcomatous components (85). The former is usually glandular and includes endometrioid, clear cell, or papillary plasmacytoma, while the latter is divided into two categories, homologous (normal uterine tissue resembling endometrial mesenchymal sarcoma, smooth muscle sarcoma or fibrosarcoma) and heterologous (most commonly malignant cartilage or skeletal muscle) (86). UCS with heterologous mesenchymal stroma usually behaves more aggressively and therefore has a worse prognosis than those with homologous features. According to transformation theory, it is believed that the sarcoma component is derived from the carcinoma component through EMT (87). EMT is the process by which epithelial cells transform into an aggressive mesenchymal cell phenotype involved in invasion and metastasis of various cancer types (88). In the study of Yoshida et al., 32 UCSs were investigated. Among them, six patients with non-endometrioid carcinoma showed morphological changes in the heterologous component of the sarcoma component toward the chondrocyte phenotype. It was found that SOX9 expression was significantly different between the two groups, and the mechanism may involve the NF-κB/p65 signaling pathway, as well as SOX9, contributing to the change in UCS cell morphology toward the chondrocyte phenotype by regulating COL2A1 transcription (89). However, they did not compare the overall survival and PFS between the two groups. Therefore, the role of SOX9 on their prognosis is unknown. SOX9 can also cooperate with other members of its family. Increased expression of SOX7 and SOX9, as well as cooperation between SOX7 and SOX4, are involved in the process in which upregulation of slug by SOX4, β-catenin/p300 complexes induces EMT and related CSC properties, which in turn promote homologous and heterologous sarcoma components in UCS (90). In sum, upregulation of SOX9 induces EMT, which leads to the formation of UCS, while contributing to the change of UCS cell morphology to a heterologous mesenchymal stroma, resulting in poor patient prognosis. The regulation mechanism of SOX9 in UCS is shown in Figure 4 .

(1) In OC, EC, and UCS, most studies suggest that the expression of SOX9 in cancer tissues is higher than normal control tissues ( Table 2 ) and has a certain relationship with the prognosis. This suggests that it plays an important role in these cancers and may become a new target for treatment. SOX9 is either a tumor suppressor or an oncogene in CC, the current research conclusions are inconclusive, so it remains to be further studied. (2) SOX9 plays an important role in gynecological precancerous lesions, carcinogenesis, development, EMT, chemotherapy resistance, maintenance of cancer stem cell properties, anti-apoptosis, migration, and invasion of cancer cells. (3) There are many mechanisms for regulating SOX9, including epigenetics (methylation of the promoter region); the regulation of non-coding RNA (lncRNA, miRNA, and cirRNA) ( Table 3 ); and several genes. All these provide new ideas for treatment. (4) The WNT/β-catenin pathway can be activated by SOX9 upregulation to promote carcinogenesis, chemoresistance, cell proliferation, and metastasis. (5) Whether nuclear expression or plasma expression plays a role varies by tumor subtype, but further studies are needed. (6) There is currently no study on the expression of this gene in choriocarcinoma, but the expression of its family genes has been found in germ cell tumors (91) and dissecting gonadoblastoma (92). Both tumors have choriocarcinoma components, so it can be speculated that SOX9 is also expressed in choriocarcinoma; however, confirmation needs further study.

All authors listed have made a substantial, direct, and intellectual contribution to the work and approved it for publication.

This work was supported by Effect of neoadjuvant chemotherapy with paclitaxel and carboplatin combined with intraperitoneal perfusion of bevacizumab on prognosis of advanced ovarian cancer [grant numbers S2021SFYLJS0026].

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.