Skip to main content
NCBI home page
British Journal of Cancer logoLink to British Journal of Cancer
. 2024 Mar 1;130(9):1453–1462. doi: 10.1038/s41416-024-02613-x

A cellular model provides insights into the pathogenicity of the oncogenic FOXL2 somatic variant p.Cys134Trp

Laetitia Herman 1, Angélique Amo 1, Berangère Legois 1, Caterina Di Carlo 1, Reiner A Veitia 1,2,3, Anne-Laure Todeschini 1,✉
PMCID: PMC11059147  PMID: 38429437

Abstract

Background

FOXL2 is a transcription factor expressed in ovarian granulosa cells. A somatic variant of FOXL2 (c.402 C > G, p.Cys134Trp) is the hallmark of adult-type granulosa cell tumours.

Methods

We generated KGN cell clones either heterozygous for this variant (MUT) or homozygous for the wild-type (WT) allele by CRISPR/Cas9 editing. They underwent RNA-Seq and bioinformatics analyses to uncover pathways impacted by deregulated genes. Cell morphology and migration were studied.

Results

The differentially expressed genes (DEGs) between WT/MUT and WT/WT KGN cells (DEGs-WT/MUT), pointed to several dysregulated pathways, like TGF-beta pathway, cell adhesion and migration. Consistently, WT/MUT cells were rounder than WT/WT cells and displayed a different distribution of stress fibres and paxillin staining. A comparison of the DEGs-WT/MUT with those found when FOXL2 was knocked down (KD) in WT/WT KGN cells showed that most DEGs-WT/MUT cells were not so in the KD experiment, supporting a gain-of-function (GOF) scenario. MUT-FOXL2 also displayed a stronger interaction with SMAD3.

Conclusions

Our work, aiming at better understanding the GOF scenario, shows that the dysregulated genes and pathways are consistent with this idea. Besides, we propose that GOF might result from an enhanced interaction with SMAD3 that could underlie an ectopic capacity of mutated FOXL2 to bind SMAD4.

Subject terms: Cancer genetics, Cancer genomics, Transcriptomics

Introduction

FOXL2 is a forkhead transcription factor that is expressed in ovarian granulosa cells, those surrounding the oocyte, from the earliest stages when they can be identified as such, up to their final stages of differentiation [1]. In mice, FOXL2 is required for the differentiation and proliferation of granulosa cells and for the maintenance of their identity [2–4]. FOXL2 pathogenic variants are responsible for the blepharophimosis–ptosis–epicanthus inversus (BPES) syndrome, which co-occurs most often with primary ovarian insufficiency [5]. A study using massive parallel sequencing identified a recurrent somatic pathogenic variant in FOXL2 (c.402 C > G) in adult-type granulosa cell tumours (AGCTs). This alteration results in the replacement of a Cystein residue at position 134 by a Tryptophan (i.e., p.Cys134Trp, called for simplicity below C134W) and was found in most AGCTs screened [6]. AGCTs are the most common type of sex cord tumours. They are usually detected at an early stage and often manifest with hyperestrogenism causing menstrual disturbances and uterine bleeding around or after menopause [7]. As noted above, the C134W pathogenic variant is not found in the healthy ovary or other tissues, which suggests that this somatic variant is at the root of tumorigenesis [6]. This is in agreement with our recent analysis of a mouse model harbouring the pathogenic variant in heterozygosity. Indeed, a mutational analysis of mouse AGCT (mAGCTs) transcriptomic data showed no additional driver mutations than the oncogenic variant itself [8]. Other studies have confirmed the prevalence of this FOXL2 pathogenic variant and its specificity to adult granulosa cell tumours [9, 10].

The C134W pathogenic variant does not appear to cause a massive loss of function (LOF) and mutated FOXL2 (MUT) can normally activate a series of targets in vitro [11]. However, MUT-FOXL2 has a lower ability to induce apoptosis than the WT counterpart, when overexpressed in cultured cells [12]. It has been suggested that deregulation of the expression of the GnRH receptor could play a role in this defect of apoptosis induction [13]. Another study showed that MUT-FOXL2 is responsible for an overactivation of CYP19A1 (Aromatase) expression, although a link with tumorigenesis could not be clearly established [14]. A transcriptomic study of cell lines overexpressing WT- or MUT-FOXL2, suggested that the pathogenic variant affects TGF-beta signalling, by promoting the expression of TGF-beta/activin/SMAD2/3 pathway genes and repressing the expression of BMP/SMAD1/5 pathway genes, thus promoting granulosa cell proliferation [15]. Similarly, a recent study has shown that MUT-FOXL2 acquires the ability to bind SMAD4. The resulting FOXL2/SMAD4/SMAD2/3 complex would bind to a new DNA motif and result in an enhancer-like chromatin state, allowing transcription of genes characteristic of the epithelial-to-mesenchymal transition (EMT) through a gain-of-function (GOF) process [16]. In a recent paper describing a mouse model heterozygous for the oncogenic mutation, we provided further evidence supporting the GOF scenario. However, we also suggested that a potential LOF and GOF on a subset of cognate FOXL2 targets might also be playing a role [8].

To gain insights into the unique FOXL2 C134W pathogenic variant and to further explore the GOF scenario, we generated CRISPR/Cas9 genome-edited KGN granulosa cells [17]. Because KGN cells are heterozygous for the c.402 C > G/C134W variant, we reversed it in order to produce WT/WT control cell lines. The latter were compared to heterozygous clones referred to as WT/MUT cells. In short, using up-to-date genomic approaches and by comparing our dataset with different available datasets, we show that the interaction between MUT-FOXL2 and SMAD3 is enhanced and provide further evidence supporting the GOF scenario affecting TGF-beta signalling. This translates into consistent morphological and biophysical alterations in the mutated cells.

Materials and methods

Cells culture

KGN (RRID:CVCL_0375) and HeLa (TKG Cat# TKG 0331, RRID:CVCL_0030) cells [11] were grown in DMEM-F12 medium supplemented with 10% FBS and 1% penicillin–streptomycin (Thermo Fisher Scientific, Waltham, MA, USA). More details are below, in specific sections below.

CRISPR-Cas9 strategy and genetic characterisation of the clones

The KGN line used is heterozygous for the c.402 C > G variant. In order to obtain homozygous WT/WT cell lines, we used the CRISPR/Cas9 methodology associated with homologous recombination repair (GeneArt® CRISPR Nuclease Vector with OFP reporter, Life Technologies). We designed a guide RNA, targeting the c.402 C > G allele in order to reverse the mutation (5’-CTTCTCGAACATGTCTTCCC-3’). Upon transfection with the guide RNA and Cas9, cells were also put in the presence of a 100 bp single-stranded DNA template (or ssODN) centred on the c.402 C > G pathogenic variant with the WT sequence (5’-TCTTCATGCGGCGGCGGCGCCGGTAGTTGCCCTTCTCGAACATGTCTTCGCAGGCCGGGTCCAGCGTCCAGTAGTTGCCCTTGCGCTCGCCGCCGCCCTC-3’), allowing repair by homologous recombination. Cells were then sorted and transfected cells (expressing the orange fluorescent protein encoded in the plasmid allowing expression of Cas9 and the guide RNA), were recovered and diluted to obtain one cell per well (in 96-well plates). Transfected cells were cultured in medium partially processed by KGN cells to provide the necessary growth factors. After 2–3 weeks, the clones were transferred to 24-well plates for amplification and genomic DNA extraction. An allele-specific PCR strategy was developed to sort out WT/WT from heterozygous cells (FOXL2-F: 5’-GAATAAGAAGGGCTGGCAAAAT-3’, FOXL2 WT-R: 5’-CCTTCTCGAACATGTCTTCG-3’ FOXL2 C134W-R: 5’-CCTTCTCGAACATGTCTTCC-3’). DNA from KGN cells (WT/MUT) and COV434 cells (RRID:CVCL_2010) (WT/WT) were used as controls. Only clones showing a single band corresponding to the presence of the WT allele were retained (for Sanger sequencing). Two FOXL2 WT/WT clones were obtained. We also produced WT/MUT clones (transfected with the CRISPR plasmid but not having undergone any modification).

Morphological studies

To study cell morphology, cells from FOXL2 WT/WT or WT/MUT clones were seeded independently on coverslips and cultured for 24 h, up to 50% confluence. DAPI (at 1/2000th) and Phalloidin (at 1/1000th) labelling were performed after fixing the cells with 4% Formaldehyde (Pierce Thermo Scientific Cat# 28908). Cells of each genotype were scored according to three different phenotypes (elongated, spread and round cells). Statistical significance was assessed using Student’s t tests. Error bars represent the standard deviation between replicates (n = 3).

RNA extraction and sequencing

RNA from WT/WT and WT/MUT KGN granulosa cells was extracted using the TRI-reagent, according to the supplier’s instructions (MRC). In total, 1 µg of total RNAs underwent high-throughput sequencing at Genom’IC, France). Specifically, total RNA quality was assessed on an Agilent Bioanalyzer 2100, using RNA 6000 pico kit (Agilent Technologies). Directional RNA-Seq Libraries were constructed using the TruSeq mRNA Stranded library prep kit (Illumina), following the manufacturer’s instructions. Final library quality was assessed on an Agilent Bioanalyzer 2100, using an Agilent High Sensitivity DNA Kit. Libraries were pooled in equimolar proportions and sequenced on Single Read 75pb runs, on an Illumina NextSeq500 instrument, using NextSeq500 High Output 75 cycles kits. Fastq files were then aligned using STAR algorithm (version 2.7.6a). Quality control of the alignment realised with Picard tools (version 2.8.1) and reads were then count using RSEM (v1.3.1). The statistical analyses on the read counts were performed with R (version 3.6.3) and the DESeq2 package (DESeq2_1.26.0, RRID:SCR_000154). Thus, the P values were adjusted for multiple testing using the Benjamini and Hochberg method, and those with an adjusted P value <0.05 were considered to be significant. Data were deposited in Gene Expression Omnibus (GEO) (RRID:SCR_005012) (GSE225781 for CRISPR clones and GSE227040 for the KD experiment).

Mammalian two-hybrid assay

To assess the interaction between FOXL2 (WT or MUT) with SMAD3, we used a mammalian two-hybrid system (CheckMate™ Mammalian Two-Hybrid System, Cat# E2440, Promega). SMAD3 and FOXL2 genes were cloned into pBind and pAct vectors to generate fusion proteins with the DNA-binding domain of GAL4 and the activation domain of VP16, respectively. HeLa cells were then transfected with FOXL2 (WT or MUT)-pACT, SMAD-pBIND or FOXL2 (WT or MUT)-pBIND, SMAD-pACT and a GAL4-responsive luciferase vector. In total, 4 × 104 cells were seeded in 96-well plates 16 h before transfection to be at confluence at the time of transfection and transfected with 180 ng of total DNA per well using the calcium phosphate method and rinsed 24 h after transfection. Forty-eight hours after transfection, cells were washed with phosphate-buffered saline solution before lysis, and luciferase measurements were performed using the Dual-Luciferase Reporter Assay System (Promega, Cat# E1960) in a TriStar LB 941 luminometer (Berthold Technologies, Bad Wildbad, Germany). The luciferase activity was normalised with respect to Renilla luciferase activity expressed from the pBIND Vector. Statistical significance was assessed using Student’s t tests. Error bars represent the standard deviation between replicates (n = 6).

Immunohistochemistry

Each KGN cell line (WT/WT or WT/MUT) was plated in 24-well culture plates containing coverslips and allowed to adhere at 37 °C overnight. After 24 h, cells were fixed with 4% paraformaldehyde (PFA) for 8 min and processed for immunostaining. The mouse anti-Paxillin (A-5) (Santa-Cruz Biotechnology, Cat# sc-390738) antibody was diluted 1/100 in PBS, 1% BSA, 0.1% triton. The secondary antibody was a donkey anti-mouse IgG coupled to Alexa Fluorophore A488 (Molecular Probes Cat# A-21202, RRID:AB_141607), diluted 1:500. F-actin was marked by adding Rhodamine-Phalloidin (Thermo Fisher Scientific, Cat# R415, 1/500) to the secondary mix. Immunofluorescence microscopy was carried out using a EVOS® FL (Thermo Fisher Scientific) or a ZEISS LSM700 confocal microscope. All the images were processed with Image J (RRID:SCR_003070) (v.1.52d) image analysis software (NIH).

Over-representation analysis (ORA) and bubble plots

ORA were performed using the Enrichr website (https://maayanlab.cloud/Enrichr/) [18, 19]. The full ORA results are provided in relevant Additional tables. The enriched pathways highlighted in the text and/or figures were selected as follows: (i) pathways were sorted based on their adjusted P values, (ii) when pathways/terms from different databases (e.g., GO BP, Bioplanet, etc.) occurred several times, the one with the smallest adjusted P value was kept and (iii) when related terms involving the similar gene subsets were present, only the most general term/pathway was retained. Terms/pathways not linked with ovarian function were not retained for the bubble plots.

Data accession

Sequence datasets generated as a part of this study have been deposited to Gene Expression Omnibus (GEO) (RRID:SCR_005012) under the accession number GSE225781 for CRISPR clones and GSE227040 for KD experiment.

Results

In vitro insights into the transcriptomic impact of FOXL2 C134W

Here, we have used KGN cells, derived from a human granulosa cell tumour [17] and thus heterozygous for the c.402 C > G/C134W pathogenic variant. To obtain suitable control cells, we modified KGN cells by CRISPR/Cas9 to reverse the mutation and to obtain WT/WT cells. Once the relevant clones were obtained, WT-FOXL2 expression was verified by cDNA sequencing.

In order to explore the transcriptional effects of MUT-FOXL2, we compared the transcriptomes of WT/WT and WT/MUT cells. We considered as differentially expressed genes (DEGs) those whose expression levels varied by a ratio of 1.5 in any direction between WT/MUT and WT/WT KGN cells (which corresponds to a log2FoldChange (logFC) of 0.585), with adjusted P values < 0.05. Such genes are called hereafter DEGs-WT/MUT. This comparison pointed to 1133 DEGs-WT/MUT, out of which 664 genes were downregulated and 469 genes were upregulated (i.e., in the WT/MUT condition compared to the WT/WT cells, Additional Table S1). We also compared the DEGs-WT/MUT with those found when FOXL2 was KD in WT/WT KGN cells (that is, with the 360 DEGs in cells transfected with siRNAs targeting FOXL2 versus those transfected with scramble/control siRNA) (Additional Table S2 from ref. [20]). This experiment is supposed to uncover direct and indirect targets of FOXL2. We observed only a modest overlap between the DEGs in both experiments (only 35 DEGs in common, Fig. 1a and Additional Table S2). A subset of 1098 genes were not DE in the KD experiment, supporting a GOF scenario.

Fig. 1. Profound transcriptomic changes due to the presence of the C134W pathogenic variant point to a perturbation of TGFb signalling.

Fig. 1

Open in a new tab

a Venn diagram of DEGs when FOXL2 is KD [20] and DEGs-WT/MUT (KGN cells). b Bubble plot of ORA of the 1113 DEGs-WT/MUT. If a GO term occurred several times in different databases, only the one with the smallest adjusted P value was kept (valid for all of the bubble plots in this paper). c Mammalian 2H experiments, performed in HeLa cells, show an increased interaction between MUT-FOXL2 and SMAD3. d Luciferase activity of the reporter promoter 3XGRAS-luc in HeLa cells. The reporter vector was co-transfected with those encoding either FOXL2 WT or C134W, and TGFBR1-CA, SMAD3 or pcDNA3.1 control vector. Error bars represent the standard deviation of six replicates. Asterisks represent statistical significance according to a Student’s t test. P values: **P < 0.01 and ***P < 0.001.

We then used the list of DEGs-WT/MUT to search for enriched annotations (ORA), using the EnrichR server (https://maayanlab.cloud/Enrichr/). The enrichments found highlighted several signalling pathways such as TGF-beta regulation of extracellular matrix (ECM), BDNF and PDGF signalling, collagen biosynthesis, interleukin regulation of various processes and focal adhesion. They also pointed to a potential dysregulation of cell migration and to EMT (Fig. 1b and Additional Table S3).

A further analysis based on the entries available in “TF Perturbations Followed by Expression” of EnrichR pointed to various TFs whose targets are potentially dysregulated such as SOX9 (gene overexpression experiment, Gene Expression Omnibus entry GSE34060 [21], P-adj = 10−5), involved in testis determination/differentiation, which is consistent with its upregulation (2.4-fold) in WT/MUT cells. We observed this upregulation also in mAGCTs [8]. Targets of TP53 also appeared dysregulated (KD experiment, GSE68248, P-adj = 10−4 [22] as well as those of the pluripotency factor POU5F1 (KD experiment, GSE21135 [23], P-adj = 10−9). Consistent with previous reports pointing to dysregulation of TGF-beta signalling, the 1133 DEGs-WT/MUT were enriched in known direct targets of SMAD3 and 4. More specifically, they were enriched in potential targets SMAD3 (TRRUST Transcription Factors 2019 [24], P-adj  = 0.03) and in DEGs in the Smad3 knocked-out mouse (GSE28598 [25], P-adj = 10−9). This is further supported by our results using a mammalian two-hybrid assay. Specifically, we expressed SMAD3 and FOXL2 fused to the DNA-binding domain of GAL4 or to the activation domain of VP16 (and vice versa). A protein-protein interaction is revealed by an increased luciferase activity. Interestingly, MUT-FOXL2 displayed a significantly increased luciferase activity compared to WT-FOXL2 (Fig. 1c), suggesting that the mutation increases the strength of the interaction between MUT-FOXL2 and its partner. We have previously reported that the C134W mutation reduces the transactivation capacity of MUT-FOXL2 along with SMAD3 when stimulated by a constitutively active version of the TGFBR1 on the 3XGRAS artificial reporter promoter in COV434 cells. Here, we confirm these results in HeLa cells (Fig. 1d).

Given that MUT-FOXL2 is known to ectopically interact with SMAD4, we compared the DEGs-WT/MUT with those found in a SMAD4 KD experiment using porcine granulosa cells (gene list obtained from Supplementary Table S1 of ref. [26]). Sixty-five genes were common to both lists, out of which 52 were consistently deregulated (down/down or up/up) (Additional Table S4). Not surprisingly, these genes were enriched in players of ECM organisation and its regulation by TGF-beta (DPP4, ITGA11, CASP1, FN1, TIMP3, NEDD9, SLIT3, TGFBI, SMAD7, FBN1) as well as integrin-cell surface interactions and actin cytoskeleton regulation, which suggests a causal link with the different morphological phenotypes of the WT/MUT and WT/WT cells (see below). The significant overlap along with the concordance of gene dysregulation are consistent with MUT-FOXL2 hijacking SMAD4.

A previous study has shown that loss of the FOXL2 interactor TRIM28 [27, 28] leads to transdifferentiation of ovarian granulosa into Sertoli cells. In addition, TRIM28 sits on the chromatin in close proximity to FOXL2 peaks to repress testis-specific genes and maintain ovarian identity. Thus, we compared the DEGs-WT/MUT with those in the conditional Trim28 KO (vs. WT) in the somatic compartment of the XX gonad at 13.5 days post coitum (gene list obtained from Supplemental Data 1 of ref. [27], GSE166385)). We found 207 common DEGs (Fig. 2a), the ORA of which showed that they are involved in TGF-beta regulation of ECM, proteolysis, various interleukin signalling pathways, TNF-alpha and BDNF signalling, regulation of angiogenesis, phosphatidylinositol 3-kinase signalling, cell proliferation and motility (Fig. 2b and Additional Table S5). Given this rather unexpected result, we also analysed the overlap between DEGs found in our murine model of AGCTs (mAGCTs) obtained by comparing WT and C134W mutated ovaries (GSE202242 and supplemental Table S2 from ref. [8]). We found 675 common DEGs, which represent more than 50% of the DEGs in mAGCTs (Fig. 2c). Consistently, they point to dysregulation of TGF-beta-related processes among others (Fig. 2d and Additional Table S6) and to a novel potential player (TRIM28) in the oncogenic process.

Fig. 2. Comparison of DEGs in WT/MUT KGN cells or mAGCTs and DEGs in the ovary of the Trim28 KO mouse.

Fig. 2

Open in a new tab

a Venn diagram of the overlap between the DEGs-WT/MUT (KGN cells) and DEGs in the ovary of the Trim28 KO mouse. b Bubble plot of an ORA of the 207 DEGs in the overlap described in (a). c Venn diagram showing 675 DEGs in mAGCTs (mouse model) also deregulated in the Trim28 KO mouse. d Bubble plot of the ORA of the 675 DEGs in the overlap described in (c).

Finally, we assessed the overlap between the DEGs-WT/MUT with our previous transcriptomic profiling of ten hAGCTs [29]. The latter dataset was derived from a reanalysis of our previous transcriptomic results (Table S4 from ref. [8]) and involved 1394 DEGs (1340 downregulated and 54 upregulated genes) in hAGCTs versus control granulosa cells, with FC > 1.5 in any direction (with Bonferroni-corrected t test-based P values < 0.05, raw data can be found at https://www.ebi.ac.uk/biostudies/ reference E-MTAB-483). Out of the 86 DEGs in both experiments, we focused on the 52 genes whose deregulation was in the same direction (in this case, the 52 genes were downregulated in both conditions) (Fig. 3a). Such genes were enriched in players of TGF-beta regulation of ECM, BDNF signalling, p53 and p63 transcriptional networks, Nuclear Receptors Pathway, TNF-alpha effects on cell motility and apoptosis and many others that make sens in the context of cancer (Fig. 3b and Additional Table S7). A similar comparison with our transcriptomic data for mAGCTs [8] showed that, although the overlap of DEGs-WT/MUT and mAGCTs was small, the dysregulated processes were similar (Fig. 3c, d).

Fig. 3. Comparison of DEGs-WT/MUT (KGN cells) versus hAGCTs or mAGCTs.

Fig. 3

Open in a new tab

a Venn diagram of DEGs-WT/MUT (KGN cells) and those found in our previous transcriptomic profiling of ten hAGCTs [29]. b Bubble plot of a ORA of the 86 common DEGs shown in (a). c Venn diagram of DEGs-WT/MUT (KGN cells) and those found in our previous transcriptomic profiling of mAGCTs [8]. d Bubble plot of a ORA of the 72 common DEGs shown in (c).

On genomic grounds, we also explored the interplay of MUT-FOXL2-dependent changes of gene expression observed in our RNA-seq data with chromatin structure. For this, we used previously published data on topologically associating domains (TADs) [30], known to be conserved across different cell types [31, 32]. TADs are genomic regions where chromatin segments physically interact with each other more frequently than with segments outside these regions [33]. Specifically, we combined our RNA-seq data with KGN ChIPseq data from ref. [16] and TADs data obtained from the Topologically Associating Domain Knowledge Base (http://dna.cs.miami.edu/TADKB/). First, we retrieved the location of the 839 genes (out of 1133), which were unambiguously assigned to one TAD using the Ensembl Biomart tool. Then, we matched this information with the TAD coordinates and the locations of FOXL2 ChIP peaks found in KGN cells (heterozygous for C134W variant, GSE138496). This analysis pointed to 682 TADs in which at least one DEG and one ChIPseq peak were present (i.e., 839 DEGs-WT/MUT and 3880 ChIP peaks). More specifically, 81% of the TADs (555/682) contained only one DEG (and at least one peak). Of the 682 TADs, 359 contained only downregulated genes, while 273 contained only upregulated genes. In most of the TADs containing several DEGs-WT/MUT, they were deregulated in the same direction. Only 50 TADs contained both down- and upregulated genes (grey bars) (Fig. 4).

Fig. 4. Interplay of DEGs-WT/MUT (KGN cells) with TADs and ChIPseq peaks.

Fig. 4

Open in a new tab

a Analysis uncovered 682 TADs containing at least one DEG and one ChIPseq peak. b Distribution of the TADs according to the number of DEGs mapping to them. c Distribution of the TADs according to the direction of the dysregulation of DEGs mapping to them.

The presence of MUT-FOXL2 affects cell morphology and migration

During the screening of the WT/MUT and WT/WT cell clones, we noticed morphological differences differentially enriched according to the two genotypes. Three different phenotypes (elongated, spread and round cells) were observed and scored (Fig. 5a, b). WT/WT cells were on average more elongated than WT/MUT cells. As expected, clones with the same genotypes had similar distributions of the three phenotypes. To obtain further insights, we looked at the status of stress fibres (actin staining using phalloidin) and focal adhesions (paxillin staining) in cells of each genotype (i.e., WT/MUT or WT/WT). Stress fibres were clearly sharper and brighter in WT/WT compared to WT/MUT cells, which is consistent with the downregulation of ACTB (stress fibres), VASP and Zyxin (focal adhesion) in the latter (Fig. 6a). However, no objective differences could be found in the number, intensity or distribution of the focal adhesions between the two genotypes. That said, we were able to detect clear differences in the distribution of the fluorescent signals within the cells of each genotype. Indeed, the WT/WT nuclei could be easily spotted in cells stained for paxillin, whereas this was not the case in WT/MUT cells. In addition, confocal microscopy images showed that stress fibres in WT/MUT cells are less well structured and paxillin staining appeared to cover a larger area around the nuclei (Fig. 6b).

Fig. 5. Morphological differences of WT/MUT vs WT/WT KGN cells.

Fig. 5

Open in a new tab

a Representative images of the three distinct morphological phenotypes observed in the different CRISPR clones (elongated, spread and round cells). KGN cells labelled with DAPI (1/2000th) and phalloidin (1/1000th) for F-actin labelling. b Distribution of cells according to the three phenotypes for each CRISPR clone obtained. Asterisks represent statistical significance according to a Student’s t test. P values: **P < 0.01 and ***P < 0.001.

Fig. 6. Stress fibres and focal adhesions in WT/MUT and WT/WT KGN cells.

Fig. 6

Open in a new tab

a Representative epifluorescence images for each genotype: paxillin (1/100, GFP) for focal adhesion and phalloidin (1/500, rhodamine) for F-actin labelling. Scale bar = 100 µm. b Confocal microscopy analysis (same labelling). Scale bar = 40 µm. Arrowhead: presence (white) or absence (yellow) of clearly defined stress fibres (rhodamine labelling) or paxillin around the nucleus.

Discussion

As noted above, we observed morphological differences between WT/WT and WT/MUT cells. The former tend to be more elongated than those containing the MUT allele. The presence of the C134W allele is associated with the acquisition of a round cell shape that points to a perturbation of the cytoskeleton and cell adhesion properties. This is in agreement with a TGF-beta-mediated dysregulation of ECM organisation, and a perturbation of actin cytoskeleton and focal adhesion (Fig. 1). At the transcriptomic level, DEGs-WT/MUT bear the hallmarks of cancer. As noted, there is a poor overlap between DEGs-WT/MUT and those reported in the FOXL2 KD experiment performed in WT/WT cells [20]. Although the results of the KD experiment are not exhaustive, this comparison suggests that the MUT-FOXL2 exerts an impact on a limited subset of direct and indirect FOXL2 targets and supports the GOF scenario suggested by previous ChIPseq and transcriptome experiments in engineered cell lines [34].

We have previously reported an upregulation of Sox9 in a mouse tumour model (mAGCTs) but its expression was much weaker than that of Foxl2 in terms of FPKMs [8]. In the KGN cell model, there is a clear overexpression of SOX9, which is more strongly expressed than FOXL2 (in terms of FPKMs). Not surprisingly, DEGs in a SOX9 overexpression model (GSE34060) such as SEMA5A, GATA6, NEDD4L, PDGFD, EGR1, EGR2, TGFB2, NGF, PIK3AP1, etc., are also DEGs in WT/MUT KGN cells. Thus, we cannot exclude a contribution of some degree of transdifferentiation (from a granulosa to a Sertoli-like state) to tumorigenesis. Interestingly, there is a connection between TGF-beta signalling and SOX9 in the context of the production of ECM and in cancer. Indeed, it is known that TGF-beta-dependent SMAD3 signalling enhances the transcriptional activity of SOX9 and the expression of COL2A1 through the formation of SMAD-SOX9 transcriptional complexes on the enhancer region of COL2A1 [35]. In addition, the TGF-beta/SOX9 axis has been shown at play in cancer progression by promoting EMT [36].

Interestingly, we have shown that C134W reduces the transactivation capacity of FOXL2 along with SMAD3 on the 3XGRAS reporter promoter in two different cell lines COV434 ([37]) and HeLa (this paper). These results and the one described by Nonis et al. [38] on GDF9 regulation of follistatin promoter activity (in the presence of WT or MUT-FOXL2) suggest that the interaction between MUT-FOXL2 with SMAD3 is also perturbed. In line with this, our mammalian 2H results show that the strength of the interaction between MUT-FOXL2 and SMAD3 is increased, suggesting that the interaction strength at the protein-protein level is not necessarily correlated with the transcriptional effect at least on the specific promoter studied. Previous ChIPseq and transcriptome profiling experiments in engineered cell lines have shown that MUT-FOXL2 has an altered genomic binding [34] supporting a GOF scenario. This might result from an ectopic capacity of mutated FOXL2 to bind SMAD4 and, according to our results, to an enhanced interaction with SMAD3. Such perturbed interactions can drive the complex MUT-FOXL2/SMAD4/SMAD3 to recognise novel targets. In line with this is the comparison of the DEGs-WT/MUT with those found in a SMAD4 KD in porcine granulosa cells. The vast majority of the DEGs common to both experiments were consistently deregulated (down/down or up/up) which is in line with a hijack of SMAD4 by MUT-FOXL2 (thus mimicking a KD of the latter).

Because of the transdifferentiation observed in the conditional Trim28 KO in the somatic compartment of XX embryonic gonads, we also explored the overlap between the DEGs-WT/MUT and DEGs found in the Trim28 KO. Interestingly, we found that about 18% of the DEGs-WT/MUT were also DEGs in the Trim28 KO, and impact in the first place TGF-beta regulation of ECM. The overlap was even more striking when using data from our murine model of AGCTs (mAGCTs) with more than 50% of the DEGs in mAGCTs also being DEGs in the Trim28 KO [8]. Both comparisons point to a consistent dysregulation of TGF-beta-related processes among other processes. These results point to TRIM28 as a potential player in the oncogenic process driven by MUT-FOXL2. However, further studies are required to explore the interplay between MUT-FOXL2, SMAD3/4 and TRIM28.

In order to link transcriptomic perturbations to genomic features, we matched DEGs-WT/MUT with TADs and ChIPseq peaks. This analysis uncovered 682 TADs containing at least one DEG and one ChIPseq peak (Fig. 4a, b). As expected, most of them had only one DEG (and at least one peak). In addition, in 61% of the TADs containing several DEGs the deregulation was in the same direction for the relevant genes (down–down or up–up, Fig. 4c). This points to a causal link between the co-occurrence of FOXL2 ChIP peaks and dysregulated genes within the same TADs lending credence to the GOF of MUT-FOXL2.

Finally, consistent with the transcriptomic analysis pointing to altered interactions between the mutated cells and the ECM, driven by an altered TGF-beta signalling, we found that WT/MUT cells had specific morphological characteristics. Indeed, they tended to be round, which correlates with the presence of less bright and more diffuse stress fibres. Although we could not find obvious differences in the number, intensity or distribution of the focal adhesions, we did find that the nuclei of WT/WT cells could be better spotted than in WT/MUT cells (paxillin staining) and that paxillin staining covered a larger area around the nuclei (most probably the endoplasmic reticulum). The biological significance of these differences deserves further exploration.

Several mouse models of granulosa cell tumours have been identified or produced. For instance, mice KO for the Inha gene develop GCTs with full (100%) penetrance [39]. Inactivation of Smad1 and Smad5 genes specifically in granulosa cells leads to a similar phenotype, suggesting that the BMP/SMAD pathway is required to control granulosa cell proliferation [40]. In another model, a constitutively active mutant of beta-catenin was expressed specifically in granulosa cells. Females developed abnormal follicles with the appearance of tumours in about 60% of them [41]. Stabilisation of circulating LH also leads to the appearance of GCTs [42]. Subsequent inactivation of candidate genes in these models pointed to factors involved GCT progression. For example, the inactivation of Esr2 and/or Esr1 in Inha−/− mice leads to earlier and more aggressive tumour development [43]. Along similar lines, inactivation of Pten in mice where beta-catenin is constitutively active also leads to a worsening of the tumour phenotype, suggesting that the PI3K/AKT pathway can indeed be mobilised in GCTs. More surprisingly, deregulation of the ERK pathway in these same mice, via the use of a Kras mutant, also leads to a worsening of the phenotype, whereas the introduction of this mutant into WT mice leads to cell cycle arrest in granulosa cells [44]. Another work addressing GCT initiation and progression using mice with targeted expression of SV40 large T-antigen in granulosa cells (AT mouse) showed that tumorigenesis was associated with the combined inactivation of p53 and Rb pathways. They conclude that “this model challenges the current paradigm that impaired FOXL2 signalling is a major switch of granulosa cell tumorigenesis”[45]. We have shown that this assertion does not hold in the context of mAGCTs. Thus, the FOXL2 mutation remains the main marker of AGCTs, most of which express the INHA and do not display constitutive beta-catenin activation [46, 47].

Our results, taken together, confirm that GOF is a very plausible mechanism to explain the oncogenicity of the C134W variant. Our current and previous transcriptomic and molecular data suggest the existence of a feedback loop whereby TGF-beta mediates somehow the GOF of MUT-FOXL2, which would in turn modulate TGF-beta receptivity and response. This is captured in Fig. 7, which displays the interaction network of DEGs involved in the TGF-beta signalling in WT/MUT cells.

Fig. 7. TGF-β/Smad signalling in WT/MUT cells.

Fig. 7

Open in a new tab

DEGs-WT/MUT (KGN cells), described as belonging to the TGF-β/Smad signalling are displayed. Activated genes appear in red, repressed genes are displayed in blue and non-deregulated genes in grey.

Supplementary information

Table S1 DEGs-WTMUT (168.5KB, xlsx)
41416_2024_2613_MOESM2_ESM.xlsx (21.3KB, xlsx)

Table S2 35 common DEGs between CRISPR cells and KD experiment

Table S3 GSEA DEGs-WTMUT (24.5KB, xlsx)
41416_2024_2613_MOESM4_ESM.xlsx (60.2KB, xlsx)

Table S4 Comparison of the DEGs-WTMUT with DEGs in SMAD4 KD

41416_2024_2613_MOESM5_ESM.xlsx (100.7KB, xlsx)

Table S5 Comparison of the DEGs-WTMUT with DEGs in Trim28 KO

41416_2024_2613_MOESM6_ESM.xlsx (104.1KB, xlsx)

Table S6 Comparison of the DEGs in mAGCTs with DEGs in Trim28 KO

41416_2024_2613_MOESM7_ESM.xlsx (29.5KB, xlsx)

Table S7 Comparison of the DEGs-WTMUT with DEGs in hAGCTs and GSEA

Acknowledgements

The authors are indebted to Emma Vidal, Lakshmi Balasubramaniam, Joseph d’Alessandro and Alexandros Glentis for their help and advices.

Author contributions

LH: conceptualisation, formal analysis, investigation, methodology, validation, visualisation and writing—original draft. AA: investigation. BL: investigation. CDC: investigation. RAV: conceptualisation, formal analysis, funding acquisition, project administration, supervision, validation, writing—original draft. ALT: conceptualisation, formal analysis, funding acquisition, investigation, project administration, supervision, validation, visualisation, writing—original draft.

Funding

This work was supported by the University of Paris Cité and the Centre National de la Recherche Scientifique and by ARC (Association pour la Recherche contre le Cancer) and Les Entreprises contre le cancer.

Data availability

The datasets generated and/or analysed during the current study (RNA-seq) can be accessed at [Gene Expression Omnibus, GSE225781 and GSE227040] repository.

Competing interests

The authors declare no competing interests.

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Footnotes

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

These authors jointly supervised this work: Reiner A. Veitia, Anne-Laure Todeschini.

Supplementary information

The online version contains supplementary material available at 10.1038/s41416-024-02613-x.

References

  • 1.Cocquet J, Pailhoux E, Jaubert F, Servel N, Xia X, Pannetier M, et al. Evolution and expression of FOXL2. J Med Genet. 2002;39:916–21. doi: 10.1136/jmg.39.12.916. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Uda M, Ottolenghi C, Crisponi L, Garcia JE, Deiana M, Kimber W, et al. Foxl2 disruption causes mouse ovarian failure by pervasive blockage of follicle development. Hum Mol Genet. 2004;13:1171–81. doi: 10.1093/hmg/ddh124. [DOI] [PubMed] [Google Scholar]
  • 3.Schmidt D, Ovitt CE, Anlag K, Fehsenfeld S, Gredsted L, Treier A-C, et al. The murine winged-helix transcription factor Foxl2 is required for granulosa cell differentiation and ovary maintenance. Development. 2004;131:933–42. doi: 10.1242/dev.00969. [DOI] [PubMed] [Google Scholar]
  • 4.Uhlenhaut NH, Jakob S, Anlag K, Eisenberger T, Sekido R, Kress J, et al. Somatic sex reprogramming of adult ovaries to testes by FOXL2 ablation. Cell. 2009;139:1130–42. doi: 10.1016/j.cell.2009.11.021. [DOI] [PubMed] [Google Scholar]
  • 5.Crisponi L, Deiana M, Loi A, Chiappe F, Uda M, Amati P, et al. The putative forkhead transcription factor FOXL2 is mutated in blepharophimosis/ptosis/epicanthus inversus syndrome. Nat Genet. 2001;27:159–66. doi: 10.1038/84781. [DOI] [PubMed] [Google Scholar]
  • 6.Shah SP, Köbel M, Senz J, Morin RD, Clarke BA, Wiegand KC, et al. Mutation of FOXL2 in granulosa-cell tumors of the ovary. New Engl J Med. 2009;360:2719–29. doi: 10.1056/NEJMoa0902542. [DOI] [PubMed] [Google Scholar]
  • 7.Hanby, A.M., walker, C. Tavassoli FA, Devilee P: Pathology and Genetics: Tumours of the Breast and Female Genital Organs. WHO Classification of Tumours series - volume IV. Lyon, France: IARC Press. Breast Cancer Res 2004;6:133. 10.1186/bcr788
  • 8.Llano E, Todeschini AL, Felipe-Medina N, Corte-Torres MD, Condezo YB, Sanchez-Martin M, et al. The oncogenic FOXL2 C134W mutation is a key driver of granulosa cell tumors. Cancer Res. 2023;83:239–50. doi: 10.1158/0008-5472.CAN-22-1880. [DOI] [PubMed] [Google Scholar]
  • 9.Jamieson S, Fuller PJ. Molecular pathogenesis of granulosa cell tumors of the ovary. Endocr Rev. 2012;33:109–44. doi: 10.1210/er.2011-0014. [DOI] [PubMed] [Google Scholar]
  • 10.Jamieson S, Butzow R, Andersson N, Alexiadis M, Unkila-Kallio L, Heikinheimo M, et al. The FOXL2 C134W mutation is characteristic of adult granulosa cell tumors of the ovary. Mod Pathol. 2010;23:1477–85. doi: 10.1038/modpathol.2010.145. [DOI] [PubMed] [Google Scholar]
  • 11.Benayoun BA, Caburet S, Dipietromaria A, Georges A, D’Haene B, Pandaranayaka PJE, et al. Functional exploration of the adult ovarian granulosa cell tumor-associated somatic FOXL2 mutation p.Cys134Trp (c.402C>G) PLoS ONE. 2010;5:e8789. doi: 10.1371/journal.pone.0008789. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Kim J-H, Yoon S, Park M, Park H-O, Ko J-J, Lee K, et al. Differential apoptotic activities of wild-type FOXL2 and the adult-type granulosa cell tumor-associated mutant FOXL2 (C134W) Oncogene. 2011;30:1653–63. doi: 10.1038/onc.2010.541. [DOI] [PubMed] [Google Scholar]
  • 13.Cheng J-C, Klausen C, Leung PCK. Overexpression of wild-type but not C134W mutant FOXL2 enhances GnRH-induced cell apoptosis by increasing GnRH receptor expression in human granulosa cell tumors. PLoS ONE. 2013;8:e55099. [DOI] [PMC free article] [PubMed]
  • 14.Fleming NI, Knower KC, Lazarus KA, Fuller PJ, Simpson ER, Clyne CD. Aromatase is a direct target of FOXL2: C134W in granulosa cell tumors via a single highly conserved binding site in the ovarian specific promoter. PLoS ONE. 2010;5:e14389. doi: 10.1371/journal.pone.0014389. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Rosario R, Araki H, Print CG, Shelling AN. The transcriptional targets of mutant FOXL2 in granulosa cell tumours. PLoS ONE. 2012;7:e46270. doi: 10.1371/journal.pone.0046270. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Weis-Banke SE, Lerdrup M, Kleine-Kohlbrecher D, Mohammad F, Sidoli S, Jensen ON, et al. Mutant FOXL2C134W hijacks SMAD4 and SMAD2/3 to drive adult granulosa cell tumors. Cancer Res. 2020;80:3466–79. doi: 10.1158/0008-5472.CAN-20-0259. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Nishi Y, Yanase T, Mu Y, Oba K, Ichino I, Saito M, et al. Establishment and characterization of a steroidogenic human granulosa-like tumor cell line, KGN, that expresses functional follicle-stimulating hormone receptor. Endocrinology. 2001;142:437–45. doi: 10.1210/endo.142.1.7862. [DOI] [PubMed] [Google Scholar]
  • 18.Chen EY, Tan CM, Kou Y, Duan Q, Wang Z, Meirelles GV, et al. Enrichr: interactive and collaborative HTML5 gene list enrichment analysis tool. BMC Bioinforma. 2013;14:128. doi: 10.1186/1471-2105-14-128. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Kuleshov MV, Jones MR, Rouillard AD, Fernandez NF, Duan Q, Wang Z, et al. Enrichr: a comprehensive gene set enrichment analysis web server 2016 update. Nucleic Acids Res. 2016;44:W90–97. doi: 10.1093/nar/gkw377. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Zutterling C, Todeschini A-L, Fourmy D, Busso D, Veaute X, Ducongé F, et al. The forkhead DNA-binding domain binds specific G2-rich RNA sequences. Nucleic Acids Res. 2023;51:12367–80. doi: 10.1093/nar/gkad994. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Shih HP, Kopp JL, Sandhu M, Dubois CL, Seymour PA, Grapin-Botton A, et al. A Notch-dependent molecular circuitry initiates pancreatic endocrine and ductal cell differentiation. Development. 2012;139:2488–99. doi: 10.1242/dev.078634. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Walerych D, Lisek K, Sommaggio R, Piazza S, Ciani Y, Dalla E, et al. Proteasome machinery is instrumental in a common gain-of-function program of the p53 missense mutants in cancer. Nat Cell Biol. 2016;18:897–909. doi: 10.1038/ncb3380. [DOI] [PubMed] [Google Scholar]
  • 23.Kunarso G, Chia N-Y, Jeyakani J, Hwang C, Lu X, Chan Y-S, et al. Transposable elements have rewired the core regulatory network of human embryonic stem cells. Nat Genet. 2010;42:631–4. doi: 10.1038/ng.600. [DOI] [PubMed] [Google Scholar]
  • 24.Han H, Cho J-W, Lee S, Yun A, Kim H, Bae D, et al. TRRUST v2: an expanded reference database of human and mouse transcriptional regulatory interactions. Nucleic Acids Res. 2018;46:D380–6. doi: 10.1093/nar/gkx1013. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Yadav H, Quijano C, Kamaraju AK, Gavrilova O, Malek R, Chen W, et al. Protection from obesity and diabetes by blockade of TGF-β/Smad3 signaling. Cell Metab. 2011;14:67–79. doi: 10.1016/j.cmet.2011.04.013. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Liu L, Li Q, Yang L, Li Q, Du X. SMAD4 feedback activates the canonical TGF-β family signaling pathways. Int J Mol Sci. 2021;22:10024. doi: 10.3390/ijms221810024. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Rossitto M, Déjardin S, Rands CM, Le Gras S, Migale R, Rafiee M-R, et al. TRIM28-dependent SUMOylation protects the adult ovary from activation of the testicular pathway. Nat Commun. 2022;13:4412. doi: 10.1038/s41467-022-32061-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Penrad-Mobayed M, Perrin C, Herman L, Todeschini A-L, Nigon F, Cosson B, et al. Conventional and unconventional interactions of the transcription factor FOXL2 uncovered by a proteome-wide analysis. FASEB J. 2020;34:571–87. doi: 10.1096/fj.201901573R. [DOI] [PubMed] [Google Scholar]
  • 29.Benayoun BA, Anttonen M, L’Hôte D, Bailly-Bechet M, Andersson N, Heikinheimo M, et al. Adult ovarian granulosa cell tumor transcriptomics: prevalence of FOXL2 target genes misregulation gives insights into the pathogenic mechanism of the p.Cys134Trp somatic mutation. Oncogene. 2013;32:2739–46. doi: 10.1038/onc.2012.298. [DOI] [PubMed] [Google Scholar]
  • 30.Bonev B, Mendelson Cohen N, Szabo Q, Fritsch L, Papadopoulos GL, Lubling Y, et al. Multiscale 3D genome rewiring during mouse neural development. Cell. 2017;171:557–.e24. doi: 10.1016/j.cell.2017.09.043. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Dixon JR, Selvaraj S, Yue F, Kim A, Li Y, Shen Y, et al. Topological domains in mammalian genomes identified by analysis of chromatin interactions. Nature. 2012;485:376–80. doi: 10.1038/nature11082. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Rao SSP, Huntley MH, Durand NC, Stamenova EK, Bochkov ID, Robinson JT, et al. A 3D map of the human genome at kilobase resolution reveals principles of chromatin looping. Cell. 2014;159:1665–80. doi: 10.1016/j.cell.2014.11.021. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Pombo A, Dillon N. Three-dimensional genome architecture: players and mechanisms. Nat Rev Mol Cell Biol. 2015;16:245–57. doi: 10.1038/nrm3965. [DOI] [PubMed] [Google Scholar]
  • 34.Carles A, Trigo-Gonzalez G, Cao Q, Cheng S-WG, Moksa M, Bilenky M, et al. The pathognomonic FOXL2 C134W mutation alters DNA-binding specificity. Cancer Res. 2020;80:3480–91. doi: 10.1158/0008-5472.CAN-20-0104. [DOI] [PubMed] [Google Scholar]
  • 35.Furumatsu T, Tsuda M, Taniguchi N, Tajima Y, Asahara H. Smad3 induces chondrogenesis through the activation of SOX9 via CREB-binding protein/p300 recruitment. J Biol Chem. 2005;280:8343–50. doi: 10.1074/jbc.M413913200. [DOI] [PubMed] [Google Scholar]
  • 36.Haga K, Yamazaki M, Maruyama S, Kawaharada M, Suzuki A, Hoshikawa E, et al. Crosstalk between oral squamous cell carcinoma cells and cancer-associated fibroblasts via the TGF-β/SOX9 axis in cancer progression. Transl Oncol. 2021;14:101236. doi: 10.1016/j.tranon.2021.101236. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Georges A, Auguste A, Bessière L, Vanet A, Todeschini A-L, Veitia RA. FOXL2: a central transcription factor of the ovary. J Mol Endocrinol. 2014;52:R17–33. doi: 10.1530/JME-13-0159. [DOI] [PubMed] [Google Scholar]
  • 38.Nonis D, McTavish KJ, Shimasaki S. Essential but differential role of FOXL2wt and FOXL2C134W in GDF-9 stimulation of follistatin transcription in co-operation with Smad3 in the human granulosa cell line COV434. Mol Cell Endocrinol. 2013;372:42–8. doi: 10.1016/j.mce.2013.02.020. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Matzuk MM, Finegold MJ, Su JG, Hsueh AJ, Bradley A. Alpha-inhibin is a tumour-suppressor gene with gonadal specificity in mice. Nature. 1992;360:313–9. doi: 10.1038/360313a0. [DOI] [PubMed] [Google Scholar]
  • 40.Pangas SA, Li X, Umans L, Zwijsen A, Huylebroeck D, Gutierrez C, et al. Conditional deletion of Smad1 and Smad5 in somatic cells of male and female gonads leads to metastatic tumor development in mice. Mol Cell Biol. 2008;28:248–57. doi: 10.1128/MCB.01404-07. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Boerboom D, Paquet M, Hsieh M, Liu J, Jamin SP, Behringer RR, et al. Misregulated Wnt/beta-catenin signaling leads to ovarian granulosa cell tumor development. Cancer Res. 2005;65:9206–15. doi: 10.1158/0008-5472.CAN-05-1024. [DOI] [PubMed] [Google Scholar]
  • 42.Risma KA, Clay CM, Nett TM, Wagner T, Yun J, Nilson JH. Targeted overexpression of luteinizing hormone in transgenic mice leads to infertility, polycystic ovaries, and ovarian tumors. Proc Natl Acad Sci USA. 1995;92:1322–6. doi: 10.1073/pnas.92.5.1322. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Burns KH, Agno JE, Chen L, Haupt B, Ogbonna SC, Korach KS, et al. Sexually dimorphic roles of steroid hormone receptor signaling in gonadal tumorigenesis. Mol Endocrinol. 2003;17:2039–52. doi: 10.1210/me.2003-0039. [DOI] [PubMed] [Google Scholar]
  • 44.Richards JS, Fan H-Y, Liu Z, Tsoi M, Laguë M-N, Boyer A, et al. Either Kras activation or Pten loss similarly enhance the dominant-stable CTNNB1-induced genetic program to promote granulosa cell tumor development in the ovary and testis. Oncogene. 2012;31:1504–20. doi: 10.1038/onc.2011.341. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Cluzet V, Devillers MM, Petit F, Chauvin S, François CM, Giton F, et al. Aberrant granulosa cell-fate related to inactivated p53/Rb signaling contributes to granulosa cell tumors and to FOXL2 downregulation in the mouse ovary. Oncogene. 2020;39:1875–90. doi: 10.1038/s41388-019-1109-7. [DOI] [PubMed] [Google Scholar]
  • 46.Ohishi Y, Oda Y, Kurihara S, Kaku T, Kobayashi H, Wake N, et al. Nuclear localization of E-cadherin but not beta-catenin in human ovarian granulosa cell tumours and normal ovarian follicles and ovarian stroma. Histopathology. 2011;58:423–32. doi: 10.1111/j.1365-2559.2011.03761.x. [DOI] [PubMed] [Google Scholar]
  • 47.Watson RH, Roy WJJ, Davis M, Hitchcock A, Campbell IG. Loss of heterozygosity at the alpha-inhibin locus on chromosome 2q is not a feature of human granulosa cell tumors. Gynecol Oncol. 1997;65:387–90. doi: 10.1006/gyno.1997.4682. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Table S1 DEGs-WTMUT (168.5KB, xlsx)
41416_2024_2613_MOESM2_ESM.xlsx (21.3KB, xlsx)

Table S2 35 common DEGs between CRISPR cells and KD experiment

Table S3 GSEA DEGs-WTMUT (24.5KB, xlsx)
41416_2024_2613_MOESM4_ESM.xlsx (60.2KB, xlsx)

Table S4 Comparison of the DEGs-WTMUT with DEGs in SMAD4 KD

41416_2024_2613_MOESM5_ESM.xlsx (100.7KB, xlsx)

Table S5 Comparison of the DEGs-WTMUT with DEGs in Trim28 KO

41416_2024_2613_MOESM6_ESM.xlsx (104.1KB, xlsx)

Table S6 Comparison of the DEGs in mAGCTs with DEGs in Trim28 KO

41416_2024_2613_MOESM7_ESM.xlsx (29.5KB, xlsx)

Table S7 Comparison of the DEGs-WTMUT with DEGs in hAGCTs and GSEA

Data Availability Statement

The datasets generated and/or analysed during the current study (RNA-seq) can be accessed at [Gene Expression Omnibus, GSE225781 and GSE227040] repository.

Articles from British Journal of Cancer are provided here courtesy of Cancer Research UK

RESOURCES