CEBPB promotes transformation of endometrial complex atypical hyperplasia to endometrial cancer.

Patients were enrolled at the Department of Obstetrics and Gynecology of the First People’s Hospital of Yunnan Province. The diagnosis of EC, CAH, reversed CAH or normal endometrium was made based on pathological diagnosis. Normal endometrium was collected from patients who underwent hysteroscopy owing to irregular menstrual cycle. Clinical characteristics of the 8 sequenced patients were listed in Table  1 . Another 20 patients diagnosed with CAH or normal endometrium were included for immunohistochemical staining and primary culture, whose clinical information was summarized in Table  2 . This study was conducted in accordance with the World Health Declaration of Helsinki and approved by the Ethics committee of the First People’s Hospital of Yunnan Province (No. KHLL2022-KY128). Written informed consent was received from all participants. Table 1 Characteristics of sequenced patients No. Age (years) BMI (kg/m2) Clinical diagnosis Pathological diagnosis Comorbidity CAN-1 45 20.20 abnormal uterine bleeding well-differentiated endometrioid carcinoma No CAN-2 42 26.44 abnormal uterine bleeding well-differentiated endometrioid carcinoma Diabetes mellitus CAH-1 38 31.50 abnormal uterine bleeding endometrial complex atypical hyperplasia No CAH-2 42 24.14 abnormal uterine bleeding endometrial complex atypical hyperplasia Hypertension Con-1 37 23.81 irregular menstrual cycle proliferative endometrium No Con-2 40 29.78 irregular menstrual cycle proliferative endometrium Infertility Res-1 38 23.92 abnormal uterine bleeding secretory endometrium, glandular secretion failure, stroma decidual-like transformation No Res-2 37 22.86 abnormal uterine bleeding secretory endometrium, partial glandular secretion failure, stroma decidual-like transformation No CAH, complex atypical hyperplasia; CAN, endometrial cancer; Con, normal endometrium; Res, reversed CAH Characteristics of sequenced patients CAH, complex atypical hyperplasia; CAN, endometrial cancer; Con, normal endometrium; Res, reversed CAH Table 2 Clinical characteristics of patients included in experimental analysis Clinical characteristics CAH ( n  = 10) Con ( n  = 10) t/Chi-square P Age (years) mean 59.40 ± 7.589 54.20 ± 7.613 1.5300 0.1453 > 50 8 7 0.2667 0.6056 ≤ 50 2 3 BMI (kg/m2) mean 25.79 ± 2.502 23.93 ± 3.205 1.4510 0.1639 > 24 8 5 1.9780 0.1596 ≤ 24 2 5 Comorbidity yes 5 3 0.8333 0.3613 no 5 7 CAH, complex atypical hyperplasia; Con, normal endometrium Clinical characteristics of patients included in experimental analysis CAH, complex atypical hyperplasia; Con, normal endometrium RNA sequencing of tissue samples (two biological replicates) and primary culture samples (three biological replicates) was performed by BerryGenomics (Beijing, China) and APExBIO (Shanghai, China) correspondingly. Briefly, tissues reserved in RNAsolid™ (G3019, Servicebio, China) and cells preserved in TRIzol ® Reagent (15596026, Invitrogen, ThermoFisher Scientific, China) were subjected to RNA extraction. Enough amount of RNA passed the purity check was used for library construction. RNA sequencing was performed on the Illumina PE150 platform. Raw data was filtered for clean data, which was mapped to the hg19 after quality check. DEseq2 was adopted for expression difference analysis. The sequencing data was archived in the NCBI SRA database under the accession number (PRJNA1192278). Tissue samples fixed in 4% paraformaldehyde (G1101, Servicebio) were embedded in paraffin and sliced into serial sections. After stepwise rehydration in gradient ethanol, heat-induced antigen retrieval was performed with citrate buffer (pH 6.0) (G1219, Servicebio). 3% hydrogen peroxide was used to block endogenous peroxidase followed by incubation with 5% BSA ( GC305010 , Servicebio). The primary antibody of CEBPB (ab53138, Abcam, China) was then added. After incubation overnight, HRP-conjugated goat anti-rabbit IgG (G1213, Servicebio) was used for signal detection. The slices were then subjected to DAB (G1212, Servicebio) and hematoxylin (G1004, Servicebio) for color development and counterstaining. Immunohistochemical staining results were then quantified and scored using the sum of the staining intensity multiplied by the corresponding percentage of positive cells and the staining intensity was subdivided into 0–3: 0, negative; 1, weak; 2, moderate; 3, strong. Since that CEBPB is a transcription factor, only nuclear expression has been scored. Tissues or cells were homogenized with sonication and RIPA lysis buffer (G2002, Servicebio) for protein extraction. An equal amount of protein was analyzed using the SDS-PAGE method and transferred onto a polyvinylidene difluoride membrane. Following blocking incubation with 5% skim milk ( GC310001 , Servicebio), the primary antibody of CEBPB (ab53138, Abcam) or β-actin (ab6276, Abcam) was added. HRP-conjugated goat anti-rabbit IgG (GB23303, Servicebio) and HRP-conjugated goat anti-mouse IgG (GB23301, Servicebio) were used for signal detection. All experiments were repeated three times. RNA extraction was performed by using the RNA extraction kit (G3640, Servicebio). Equal amounts of RNA were reverse transcribed using the SweScript RT II first strand cDNA synthesis kit (G3333, Servicebio) followed by rt-qPCR amplification using 2×Universal blue SYBR green qPCR master mix (G3328, Servicebio) according to the manufacturer’s instruction. Primers used in rt-qPCR were listed in Table  3 . The mRNA expression level was normalized to β-actin expression and calculated using the 2 −∆∆ct method. All experiments had three replicates and were repeated three times. Table 3 Primers used in rt-qPCR Gene Forward Reverse CEBPB CGCAGGTCAAGAGCAAGGC TGAACAAGTTCCGCAGGGTG ACTIN ACTTAGTTGCGTTACACCCTT GTCACCTTCACCGTTCCA TGFBI TGGGATTGTAACTGTGAACTG CGAAGGGTCTCAAAGGTGT SERPINE2 GAACTGCCCTACCACG TCAGCGGCTCCTTCA NEDD4 ACACGGATTTCAACG ATTCATTATCGACCACA SEMA3A CATCTAATCCAGCAAGTC AAATCCTGCCCTCAA’ LICAM GGTGGATGGACTGGAACG TCTCATAGGGCACGAAGGT Primers used in rt-qPCR Freshly collected endometrial tissues were minced with a scalpel in a petri dish and then digested using 1 mg/ml collagenase type I ( GC305013 , Servicebio). Dissociated cells were centrifuged and cell pellets were resuspended and seeded onto a six-well plate. By primary culture, DMEM/F12 (G4612, Servicebio) supplemented with 15% fetal bovine serum (G8003, Servicebio), 1× penicillin/streptomycin/gentamycin (G4014, Servicebio), 1× insulin (41400045, Gibco™, ThermoFisher Scientific, China) and 1× MEM NEAA (11140050, Gibco™) was used. After several passages, cell clones with typical cobblestone-like shape were chosen for further experiments. siRNA targeting CEBPB was bought from Ribobio (China) and transfected into primary culture cells using optiMEM (G4551, Servicebio) and SweTransRNA (G1806, Servicebio). Stably transfected cells were used in experiment analysis. Cells were fixed with 4% paraformaldehyde (G1101, Servicebio). 0.1% Triton X-100 (G3068, Servicebio) was used in permeabilization followed by blocking with 10% BSA ( GC305010 , Servicebio). The cells were then incubated with the primary antibody of PAX8 (ab53490, Abcam), CK8/18 (ab17139, Abcam) and Vimentin (ab92547, Abcam). After incubation with the corresponding secondary antibody (Alexa Fluor ® 488-conjugated goat anti-mouse IgG, GB25301 and Alexa Fluor ® 488-conjugated goat anti-rabbit IgG, GB25303, Servicebio), the cells were counterstained with DAPI (G1012, Servicebio). Finally, the cells were sealed in anti-fluorescence quenching sealing agent (G1401, Servicebio) and photographed under a fluorescence microscope. All experiments had three replicates and were repeated three times. 4 × 10 3 cells per well were inoculated into 96-well plates and cultured for the indicated time. To assess cell proliferation, cell counting kit-8 (G4103, Servicebio) was used according to the manufacturer’s instruction and absorbance at 450 nm was measured. All experiments had six replicates and were repeated three times. 2 × 10 5 cells per well were seeded into 6-well plates and cultured for 2 weeks in the incubator. The cells were then fixed using 4% paraformaldehyde (G1101, Servicebio) and stained with crystal violet solution (G1014, Servicebio). Colonies with more than 40 cells were counted under a microscope and the colony formation rate was calculated. All experiments had three replicates and were repeated three times. Cell apoptosis was analyzed using an Annexin V-FITC/PI cell apoptosis detection kit (G1511, Servicebio). Briefly, cells were harvested using trypsin (G4011, Servicebio) and washed twice using a cold PBS buffer. After centrifugation, the cell pellets were resuspended using the binding buffer. Following the sequential addition of Annexin-V-FITC and PI, flowcytometry analysis was conducted. All experiments had three replicates and were repeated three times. Cells were seeded into 6-well plates to form monolayers. A wound line was created using a 200 µl sterile plastic tip perpendicularly and detached cells were removed. Cells were cultured in serum-free medium for 48 h and images were taken using a microscope after crystal violet staining. Cell migration ability was evaluated by comparing the scratch closure and calculated as (S Con -S)/S Con *100%. S Con and S represented the remaining scratch area of normal endometrium and that compared with normal endometrium correspondingly. All experiments had three replicates and were repeated three times. For transwell invasion assay, the transwell chambers were firstly coated with Matrigel (G4130, Servicebio). Then 3 × 10 4 cells resuspended in serum-free medium were seeded into the upper chamber, while the bottom of the chamber contained 15% fetal bovine serum (G8003, Servicebio). After 24 h, the cells were then fixed using 4% paraformaldehyde (G1101, Servicebio) and stained with crystal violet solution (G1014, Servicebio). Invaded cells were counted under a microscope. All experiments had three replicates and were repeated three times. For statistical analysis, Prism 10 (Graphpad) was adopted. Data were presented as mean ± SD. By comparing between two groups, data conformed to normal distribution was analyzed using Student’s t-test. If the two datasets did not have equal SD, Welch’s correction was performed. Data that did not coincide with normal distribution were interpreted using non-parametric test. Statistical significance was set at the level of P  < 0.05 and denoted as * P  < 0.05, ** P  < 0.01, *** P  < 0.001, **** P  < 0.0001.

To identify genes involved in the pathogenesis of CAH, EC samples, CAH tissues, reversed CAH tissues and normal endometrium were collected. Following RNA sequencing, differentially expressed genes (DEGs) were identified (Fig.  1 A and C). By comparing DEGs between EC vs. normal endometrium with DEGs between CAH vs. normal endometrium, 138 genes arose as overlapping genes implicated in the transformation from endometrial CAH to EC including CEBPB (Fig.  1 D). We then inspected the upregulated DEGs of CAH vs. normal endometrium and found that 509 of them were also upregulated DEGs of CAH vs. reversed CAH (Fig.  1 E). When intersecting the 138 CAH-involving genes with the 509 CAH-upregulated genes, 59 overlapping genes were produced, among which five genes were identified to be transcription factor according to the AnimalTFDB database (Fig.  1 F). Detailed expression of the five transcription factors was then shown in a volcano plot (Fig.  1 G). We then searched the five transcription factors in the Human Protein Atlas and found that only CEBPB had notable expression in female tissues and exerted a related function (Table  4 ). The expression of CEBPB in normal endometrium, CAH and reversed CAH was then summarized in a heatmap (Fig.  1 H). The DEGs between CAH vs. normal endometrium were also subjected to enrichment analysis, which were significantly enriched in DISGENET pathways like Endometriosis and Endometrial Neoplasms (Fig.  1 I). GO pathways related to proliferation, epithelial to mesenchymal transformation (EMT) and apoptosis and KEGG pathways including Pathway in cancer and Transcription misregulation in cancer were also significantly affected during the pathogenesis of CAH (Fig.  1 J). Briefly, CEBPB is involved in endometrial CAH and EC and participates in the progression from endometrial CAH to EC. Fig. 1 CEBPB promotes transformation from endometrial CAH to EC. ( A ) Heatmap of DEGs between EC and normal endometrium. ( B ) Heatmap of DEGs between CAH and normal endometrium. ( C ) Heatmap of DEGs between CAH and reversed CAH. ( D ) Overlapping genes of DEGs between EC vs. normal endometrium and DEGs between CAH vs. normal endometrium. ( E ) Venn diagram of upregulated DEGs of CAH vs. normal endometrium and upregulated DEGs of CAH vs. reversed CAH. ( F ) Venn diagram of CAH-involving genes and CAH-upregulated genes. ( G ) Volcano plot of DEGs between CAH vs. normal endometrium and DEGs between CAH vs. reversed CAH. ( H ) Heatmap summarizing CEBPB expression in normal endometrium, CAH and reversed CAH. ( I ) DISGENET pathway enrichment analysis of DEGs between CAH vs. normal endometrium. ( J ) Go and KEGG pathway enrichment analysis of DEGs between CAH vs. normal endometrium. CAN, cancer; CAH, complex atypical hyperplasia; Con, normal endometrium; DEGs, differentially expressed genes; EC, endometrial cancer; FC, fold change; Res, reversed CAH CEBPB promotes transformation from endometrial CAH to EC. ( A ) Heatmap of DEGs between EC and normal endometrium. ( B ) Heatmap of DEGs between CAH and normal endometrium. ( C ) Heatmap of DEGs between CAH and reversed CAH. ( D ) Overlapping genes of DEGs between EC vs. normal endometrium and DEGs between CAH vs. normal endometrium. ( E ) Venn diagram of upregulated DEGs of CAH vs. normal endometrium and upregulated DEGs of CAH vs. reversed CAH. ( F ) Venn diagram of CAH-involving genes and CAH-upregulated genes. ( G ) Volcano plot of DEGs between CAH vs. normal endometrium and DEGs between CAH vs. reversed CAH. ( H ) Heatmap summarizing CEBPB expression in normal endometrium, CAH and reversed CAH. ( I ) DISGENET pathway enrichment analysis of DEGs between CAH vs. normal endometrium. ( J ) Go and KEGG pathway enrichment analysis of DEGs between CAH vs. normal endometrium. CAN, cancer; CAH, complex atypical hyperplasia; Con, normal endometrium; DEGs, differentially expressed genes; EC, endometrial cancer; FC, fold change; Res, reversed CAH Table 4 Detailed information of overlapped transcription factors Gene Description HPA protein function Expression in female tissues Function-related BARX1 BARX homeobox 1 Craniofacial development, odontogenesis and stomach organogenesis. N/A No CEBPB CCAAT enhancer binding protein beta Immune and inflammatory responses, adipogenesis, gluconeogenic pathway, liver regeneration, hematopoiesis and female reproduction. Yes Yes EGR4 Early growth response 4 Mitogenesis and differentiation. N/A Yes HLX H2.0 like homeobox TBX21/T-bet-dependent maturation of Th1 cells, maintenance of Th1-specific gene expression, embryogenesis and hematopoiesis. Yes No NR4A2 Nuclear receptor subfamily 4 group A member 2 Meso-diencephalic dopaminergic neurons during development. Yes No HPA, the Human Protein Atlas; N/A, not applicable Detailed information of overlapped transcription factors HPA, the Human Protein Atlas; N/A, not applicable We then assessed CEBPB expression in clinical samples of CAH and normal endometrium immunohistochemically and found that CAH tissues had pronounced CEBPB expression (Fig.  2 A and B). Consistently, CAH samples had overexpressed CEBPB compared with normal endometrium in protein and mRNA levels (Fig.  2 C and D). The tissues were then subjected to primary culture and cell clones with typical cobblestone-like shape were chosen for further experiment (Fig.  3 A). To confirm origin, PAX-8 expression was detected by using immunofluorescent staining. All 20 primary cultures of CAH and normal endometrium had a near 80% positive rate of PAX8 expression (Fig.  3 B). In western blotting and rt-qPCR analysis, similar results were obtained and primary culture of CAH tissues had enhanced CEBPB expression compared with primary culture of normal endometrium (Fig.  3 C and D). Namely, CEBPB expression is upregulated in endometrial CAH. Fig. 2 Endometrial CAH tissues have overexpressed CEBPB expression. ( A ) Immunohistochemical staining of CEBPB in endometrial CAH and normal endometrium. ( B ) Quantitative results of immunohistochemical staining. ( C ) Western blotting analysis of CEBPB expression in endometrial CAH and normal endometrium. ( D ) rt-qPCR analysis of CEBPB expression in endometrial CAH and normal endometrium. CAH, complex atypical hyperplasia; Con, normal endometrium. P  value was denoted as denoted as * P  < 0.05, ** P  < 0.01, *** P  < 0.001, **** P  < 0.0001 Endometrial CAH tissues have overexpressed CEBPB expression. ( A ) Immunohistochemical staining of CEBPB in endometrial CAH and normal endometrium. ( B ) Quantitative results of immunohistochemical staining. ( C ) Western blotting analysis of CEBPB expression in endometrial CAH and normal endometrium. ( D ) rt-qPCR analysis of CEBPB expression in endometrial CAH and normal endometrium. CAH, complex atypical hyperplasia; Con, normal endometrium. P  value was denoted as denoted as * P  < 0.05, ** P  < 0.01, *** P  < 0.001, **** P  < 0.0001 Fig. 3 Endometrial CAH primary cultures have overexpressed CEBPB expression. ( A ) Morphology of primary cultures. ( B ) Immunofluorescent staining of PAX8 in primary cultures. ( C ) Protein expression of CEBPB in primary cultures. ( D ) mRNA expression of CEBPB in primary cultures. CAH, complex atypical hyperplasia; Con, normal endometrium. P  value was denoted as denoted as * P  < 0.05, ** P  < 0.01, *** P  < 0.001, **** P  < 0.0001 Endometrial CAH primary cultures have overexpressed CEBPB expression. ( A ) Morphology of primary cultures. ( B ) Immunofluorescent staining of PAX8 in primary cultures. ( C ) Protein expression of CEBPB in primary cultures. ( D ) mRNA expression of CEBPB in primary cultures. CAH, complex atypical hyperplasia; Con, normal endometrium. P  value was denoted as denoted as * P  < 0.05, ** P  < 0.01, *** P  < 0.001, **** P  < 0.0001 Expression of CEBPB was then measured in all CAH primary cultures and CAH (1) with relative higher CEBPB expression was chosen for subsequent experiments (Fig.  4 A). siRNA targeting CEBPB was then transfected into the primary culture of CAH (1) to knockdown CEBPB expression and siCEBPB (2) and siCEBPB (3) with high depletion efficiency were selected for further usage (Fig.  4 B and C). In cell proliferation assay, primary culture cells of CAH exhibited significantly higher proliferation rate than primary culture cells of normal endometrium, while knockdown of CEBPB significantly suppressed the proliferation of CAH primary culture cells as shown by reduced OD450 value (Fig.  4 D). Similar results were observed in colony formation assay, wherein CAH primary culture cells had significantly higher colony formation rate than primary culture cells of normal endometrium whereas siRNA targeting CEBPB significantly inhibited the colony formation rate of CAH primary culture cells (Fig.  4 E). The effect of CEBPB on apoptosis of CAH was also evaluated. CAH primary culture cells had a lower apoptosis rate than the primary culture of normal endometrium, but knocking down CEBPB could remarkably induce apoptosis of CAH primary culture cells as indicated in flowcytometry analysis (Fig.  4 F). CEBPB could promote EMT of CAH as well. By immunofluorescent staining, CAH primary culture had fewer CK8/18 positive cells but a higher Vimentin positive rate compared with primary culture of normal endometrium, but the positive rate of CK8/18 increased while the positive rate of Vimentin decreased after CEBPB knockdown (Fig.  5 A and B). Primary culture of CAH exhibited increased migration capacity compared with normal endometrium, but depletion of CEBPB could reverse the phenotype (Fig.  5 C). As for cell invasion, primary culture cells of CAH had enhanced invasion but knocking CEBPB down could inhibit the invasion of CAH primary culture cells (Fig.  5 D). Fig. 4 CEBPB enhances proliferation while inhibits apoptosis of endometrial CAH. ( A ) Summary of CEBPB protein expression in all CAH primary cultures. siRNA targeting CEBPB was transfected into the primary culture of CAH (1). ( B ) Protein expression of CEBPB after siRNA transfection. ( C ) mRNA expression of CEBPB after siRNA transfection. siCEBPB (2) and (3) were chosen for subsequent experiments. ( D ) Cell proliferation analysis of primary cultures after siRNA transfection. ( E ) Colony formation assay of primary cultures after siRNA transfection. ( F ) Apoptosis analysis of primary cultures after siRNA transfection. CAH, complex atypical hyperplasia; Con, normal endometrium. P  value was denoted as denoted as * P  < 0.05, ** P  < 0.01, *** P  < 0.001, **** P  < 0.0001 CEBPB enhances proliferation while inhibits apoptosis of endometrial CAH. ( A ) Summary of CEBPB protein expression in all CAH primary cultures. siRNA targeting CEBPB was transfected into the primary culture of CAH (1). ( B ) Protein expression of CEBPB after siRNA transfection. ( C ) mRNA expression of CEBPB after siRNA transfection. siCEBPB (2) and (3) were chosen for subsequent experiments. ( D ) Cell proliferation analysis of primary cultures after siRNA transfection. ( E ) Colony formation assay of primary cultures after siRNA transfection. ( F ) Apoptosis analysis of primary cultures after siRNA transfection. CAH, complex atypical hyperplasia; Con, normal endometrium. P  value was denoted as denoted as * P  < 0.05, ** P  < 0.01, *** P  < 0.001, **** P  < 0.0001 Fig. 5 CEBPB promotes EMT, migration and invasion of endometrial CAH. ( A ) Immunofluorescent staining of CK8/18 in primary cultures after siRNA transfection. ( B ) Immunofluorescent staining of Vimentin in primary cultures after siRNA transfection. ( C ) Wound healing assay of primary cultures after siRNA transfection. ( D ) Transwell invasion assay of primary cultures after siRNA transfection. CAH, complex atypical hyperplasia; Con, normal endometrium. P  value was denoted as denoted as * P  < 0.05, ** P  < 0.01, *** P  < 0.001, **** P  < 0.0001 CEBPB promotes EMT, migration and invasion of endometrial CAH. ( A ) Immunofluorescent staining of CK8/18 in primary cultures after siRNA transfection. ( B ) Immunofluorescent staining of Vimentin in primary cultures after siRNA transfection. ( C ) Wound healing assay of primary cultures after siRNA transfection. ( D ) Transwell invasion assay of primary cultures after siRNA transfection. CAH, complex atypical hyperplasia; Con, normal endometrium. P  value was denoted as denoted as * P  < 0.05, ** P  < 0.01, *** P  < 0.001, **** P  < 0.0001 RNA sequencing was also performed to further explore the biological role of CEBPB in CAH. After siRNA transfection, CAH primary cultures were sequenced. A total of 2312 genes were identified as DEGs regulated by CEBPB in CAH (Fig.  6 A and B). The sequencing results were then validated by using rt-qPCR (Fig.  6 C). Consistent with previous enrichment analysis results of DEGs between CAH vs. normal endometrium, genes regulated by CEBPB were accumulated in DISGENET pathways including Carcinomatosis, Carcinoma and Adenocarcinoma (Fig.  6 D). In CAH, GO pathways related to proliferation, EMT and apoptosis and KEGG pathways including Pathway in cancer and Transcription misregulation in cancer were also significantly affected by CEBPB (Fig.  6 E). Taken together, CEBPB promotes progression of endometrial CAH, enhancing proliferation, EMT, migration and invasion while inhibiting apoptosis. Fig. 6 RNA sequencing of CAH primary culture after CEBPB knockdown. ( A ) Heatmap of DEGs after siRNA transfection. ( B ) Volcano plot of DEGs. ( C ) Verification of sequencing results using rt-qPCR. ( D ) DISGENET pathway enrichment analysis of DEGs. ( E ) Go and KEGG pathway enrichment analysis of DEGs. CAH, complex atypical hyperplasia; DEGs, differentially expressed genes; FC, fold change. P  value was denoted as denoted as * P  < 0.05, ** P  < 0.01, *** P  < 0.001, **** P  < 0.0001 RNA sequencing of CAH primary culture after CEBPB knockdown. ( A ) Heatmap of DEGs after siRNA transfection. ( B ) Volcano plot of DEGs. ( C ) Verification of sequencing results using rt-qPCR. ( D ) DISGENET pathway enrichment analysis of DEGs. ( E ) Go and KEGG pathway enrichment analysis of DEGs. CAH, complex atypical hyperplasia; DEGs, differentially expressed genes; FC, fold change. P  value was denoted as denoted as * P  < 0.05, ** P  < 0.01, *** P  < 0.001, **** P  < 0.0001

Endometrial complex atypical hyperplasia (CAH) is the precancerous lesion of endometrial cancer (EC), which represents the only cancer with decreased survival in the past 4 decades and whose incidence rate has increased annually by about 1.0% [ 1 , 2 ]. Characterized by a marked increase in the gland-to-stroma ratio of more than 3:1 and complex architecture, CAH is the prominent type of endometrial hyperplasia [ 2 , 3 ]. Almost half of CAH patients will eventually progress to EC and up to 85% of EC mostly the endometrioid subtype arise from CAH [ 4 ]. Moreover, approximately 40% of CAH patients have concurrent EC [ 5 ]. Chronic unopposed estrogen exposure is the most common risk factor for the development of CAH [ 4 ]. Specific conditions include obesity, early menarche, late menopause, chronic anovulation, estrogen-secreting tumors and estrogen treatment regimens without counterbalanced progestin [ 2 ]. CAH can lead to irregular menstrual cycle, abnormal uterine bleeding, abnormal vaginal discharge, abdominal pain and infertility [ 2 , 4 ]. Although the treatment strategy for CAH including oral progestins, intrauterine devices, fertility-sparing surgery and even hysterectomy is evolving, the underlying mechanism by which CAH progresses to EC remains elusive. By binding to promoters or enhancers, transcription factors play crucial roles in regulating gene expression and account for 20% of all identified oncogenes so far [ 6 ]. Dysregulated transcription factors mediate aberrant gene expression and induce tumor progression, representing hallmarks of cancers [ 7 ]. Playing a critical role in cell proliferation, differentiation, apoptosis, invasion and migration, transcription factor CEBPB is implicated in the tumorigenesis of various cancers such as esophageal squamous cell carcinoma, cervical cancer, colorectal cancer, glioma, pancreatic cancer and breast cancer [ 8 – 13 ]. Downregulation of CEBPB was reported to inhibit endometrial decidualization, wherein CEBPB regulated endometrial cell proliferation and differentiation [ 14 ]. By inducing maternal-fetal immune tolerance, CEBPB could boost macrophage towards M2-like polarization [ 15 ]. During the mid-secretory phase of the menstrual cycle, CEBPB was found to be intensively expressed in the endometrium [ 16 ]. Abnormal CEBPB expression was also found to be associated with adenomyosis [ 17 ]. Namely, CEBPB affects the physiological status of the endometrium and might possibly be involved in CAH. In the present study, we sequenced EC, CAH, reversed CAH and normal endometrium samples to identify genes involved in the transformation from CAH to EC. CEBPB expression was then compared between CAH and normal endometrium. By using primary culture, the effects of CEBPB on the progression of CAH were evaluated. Finally, RNA sequencing of CAH primary cultures after CEBPB knockdown was performed to further confirm the effects of CEBPB on CAH transformation.

As a premalignant lesion, endometrial CAH represents a significant clinical concern. Elucidating the pathogenesis mechanisms and then exploiting the blockade strategy is therefore of great importance. Here, by integrating RNA sequencing data of EC, CAH, reversed CAH and normal endometrium, we found that CEBPB is involved in endometrial CAH and participates in the transformation from CAH to EC. Compared with normal endometrium, endometrial CAH tissues have overexpressed CEBPB. Using primary culture, CEBPB was found to promote proliferation, EMT, migration and invasion while inhibiting apoptosis of endometrial CAH. RNA sequencing results of CAH primary cultures after CEBPB knockdown further confirmed the promotion effects of CEBPB on endometrial CAH progression. CEBPB could potentially serve as a surrogate screening and surveillance biomarker for endometrial CAH at high cancerous risk, enabling better risk stratification and individualized treatment. Currently, standard surveillance imaging and endometrial biopsies are widely used for the screening and diagnosis of endometrial CAH and EC [ 4 ]. Even the minimally invasive microscale endometrial sampling biopsy has been developed for histopathological examination [ 18 ]. However, there is indeed a shortage of specific biomarkers [ 5 ]. Transvaginal sonography is used for screening of endometrial hyperplasia in clinical practice, but endometrium thickness varies with the menstrual cycle and the measurement of which could be biased by ultrasound doctors [ 19 ]. Adipokines especially a high level of leptin and leptin to adiponectin ratio were reported to be associated with increased risk of CAH [ 20 ]. However, the underlying molecular mechanisms were not experimentally validated [ 20 ]. Serum HE4 level was associated with progestin treatment response in women undergoing conservative therapy for CAH and EC, but the threshold for clinical use is indefinite [ 21 ]. Nomograph integrating HE4, menopausal status and BMI was also explored to stratify endometrial atypical hyperplasia patients with concurrent EC, whose discrimination efficiency yet warrants further improvement [ 22 ]. In the present study, we found that CEBPB promotes the transformation from endometrial CAH to EC and is upregulated in endometrial CAH compared to normal endometrium. Therefore, CEBPB has the potential to be exploited for screening and surveillance of endometrial CAH. Upon cancer growth, there is an imbalance between cell gain and cell loss, namely, tumor cell proliferation exceeds tumor cell death [ 9 , 14 ]. Cell proliferation without brake represents a common oncogenic property, wherein aberrant cell proliferation contributes to the malignant degree [ 14 ]. As the prominent form of programmed cell death, apoptosis functions as a common tumor-suppressing mechanism [ 9 ]. Resistance to apoptosis endows tumor cells survival advantage, inducing tumor outgrowth as well as treatment failure [ 9 ]. EMT, the trans-differentiation process characterized by reduction of epithelial markers such as CK8/18 and simultaneous promotion of mesenchymal biomarkers like vimentin, is closely related to tumorigenesis [ 23 ]. EMT was also found to promote the generation of cancer stem-like cells [ 24 ]. Cancer cells are characterized as having a migratory and invasive phenotype [ 25 , 26 ]. During tumorigenesis, cancer cells can disseminate in several strategies including collective and single-cell migration [ 25 ]. As the most important trait of cancer, invasive capacity can distinguish benign from malignant lesions [ 26 ]. CEBPB was found to regulate cell viability and apoptosis in cervical cancer [ 9 ]. In pancreatic cancer, CEBPB was reported to be associated with EMT and tumor growth [ 12 ]. In breast cancer, CEBPB was reported to affect cancer cell migration and invasion [ 26 ]. In this study, we found that CEBPB promotes proliferation, EMT, migration and invasion while inhibiting apoptosis of endometrial CAH. Compared with normal endometrium cells, endometrial CAH cells have increased proliferation, EMT, migration and invasion but decreased apoptosis, namely, the same effects mediated by CEBPB. In endometrial CAH and EC, genes regulated by CEBPB accumulate in tumor-related DISGENET pathways. Besides, in endometrial CAH and EC, GO and KEGG pathways associated with proliferation, EMT and apoptosis are significantly affected by CEBPB. CEBPB indeed involves the transformation of endometrial CAH to EC. CEBPB, a member of the basic leucine zipper transcription factor family, plays a pivotal role in several biological processes including tumor formation, growth, metastasis and drug resistance [ 27 , 28 ]. In colon cancer, CEBPB could mediate epinephrine-induced EMT and stemness under chronic emotional stress [ 24 ]. In glioblastoma, CEBPB promotes tumor growth by driving the formation of M2 tumor-associated macrophages [ 29 ]. In breast cancer, CEBPB inhibits O-fucosylation and regulates the migration, invasion and EMT of malignant cells [ 13 ]. In non-small cell lung cancer, CEBPB mediates the enhancement of tumor-initiating activity and drug resistance [ 30 ]. The role of CEBPB in endometrial CAH especially in the malignant transformation of CAH remains largely unknown. Our results suggested that CEBPB is involved in endometrial CAH and EC and promotes the transformation from endometrial CAH to EC. Therefore, targeting CEBPB might thus substantially hamper the premalignant condition. Combination therapy including proapoptotic drugs might further enhance the efficacy of CEBPB-targeted therapy. Conservative treatment options for endometrial CAH involving oral progestins or levonorgestrel intrauterine devices could elicit a complete remission rate approaching 80% [ 2 ]. The combination of progestin and metformin was reported to further improve the reversal rate of endometrial hyperplasia [ 31 ]. However, the recurrent rate is beyond 40% [ 2 , 4 ]. Promising molecular diagnostics and timely targeted therapies could halt and reverse the malignant progression, eventually reducing EC incidence [ 2 , 32 ]. Transcription factors account for approximately 20% of all identified oncogenes, being deeply involved in cancer development [ 33 ]. By driving transcription cascades, transcription factors dynamically modulate gene expression profiles and transit cell state after activation by a differentiation cue or an exogenous stimulus [ 33 ]. Previously, the relationship between the transcription factor GATA2 and hormone receptors was studied but GATA2 expression could not differentiate CAH from EC [ 34 ]. CEBPB, which promotes the disease progression of CAH, thus evolves as a potential surveillance biomarker and therapeutic target. The biological functions of CEBPB in CAH deserve further in vivo research. Although there was the limitation of the less druggable property of transcription factor, exploring treatment vulnerability using ubiquitin ligase activator and cell-penetrating decoys holds potential in CEBPB-targeted therapy [ 35 , 36 ]. Along with the menstrual cycle, endometrium undergoes dynamic changes [ 2 ]. Consequently, construct of long-term culture systems that could actually recapitulate endometrium function in vivo is challenging [ 37 , 38 ]. Although CAH has affected a significant number of women, currently, there is a lack of cell lines that could faithfully recapitulate the pathophysiology of this precursor disease [ 37 ]. The self-organizing 3D culture system organoid could reliably simulate the in vivo state of endometrium, but it is expensive [ 38 ]. As a cost-effective cell model, primary culture could also appropriately simulate the in vivo signature of endometrium and allow for a better understanding of the pathogenesis of CAH [ 37 ]. Here, in the present study, we verified the effects of CEBPB on endometrial CAH using primary cultures. However, the existence of fibroblasts in primary cultures could interfere with the experiments including biasing the effects of CEBPB on EMT. The pro-oncogenic effects of CEBPB in CAH warrant further validation using in vivo models.

In summary, CEBPB is involved in endometrial CAH and endometrial CAH has overexpressed CEBPB compared with normal endometrium. CEBPB promotes proliferation, EMT, migration and invasion while inhibiting apoptosis of endometrial CAH, promoting the progression of endometrial CAH to EC. CEBPB could potentially serve as a screening and surveillance biomarker for endometrial CAH at high cancerous risk.

Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 1