Abstract

Metastasis is the most common pathway of cancer death. The lack of effective predictors of breast cancer metastasis is a pressing issue in clinical practice. Therefore, exploring the mechanism of breast cancer metastasis to uncover reliable predictors is very important for the clinical treatment of breast cancer patients. In this study, tandem mass tag quantitative proteomics technology was used to detect protein content in primary breast tumor tissue samples from patients with metastatic and nonmetastatic breast cancer at diagnosis. We found that the high expression of yin-yang 1(YY1) is strongly associated with poor prognosis in high-grade breast cancer. YY1 expression was detected in both clinical tumor tissue samples and tumor tissue samples from mammary-specific polyomavirus middle T antigen overexpression mouse model mice. We demonstrated that upregulation of YY1 expression was closely associated with breast cancer metastasis and that high YY1 expression could promote the migratory invasive ability of breast cancer cells. Mechanistically, YY1 directly binds to the UGT2B7 mRNA initiation sequence ATTCAT, thereby transcriptionally regulating the inhibition of UGT2B7 expression. UGT2B7 can regulate the development of breast cancer by regulating estrogen homeostasis in the breast, and the abnormal accumulation of estrogen, especially 4-OHE2, promotes the migration and invasion of breast cancer cells, ultimately causing the development of breast cancer metastasis. In conclusion, YY1 can regulate the UGT2B7-estrogen metabolic axis and induce disturbances in estrogen metabolism in breast tumors, ultimately leading to breast cancer metastasis. Disturbances in estrogen metabolism in the breast tissue may be an important risk factor for breast tumor progression and metastasis

SIGNIFICANCE STATEMENT In this study, we propose for the first time a regulatory relationship between YY1 and the UGT2B7/estrogen metabolism axis and explore the molecular mechanism. Our study shows that the YY1/UGT2B7/estrogen axis plays an important role in the development and metastasis of breast cancer. This study further elucidates the potential mechanisms of YY1-mediated breast cancer metastasis and the possibility and promise of YY1 as a predictor of cancer metastasis.

Introduction

Breast cancer is the most common cancer diagnosed worldwide and is one of the major causes of cancer deaths (Sung et al., 2021). Studies have reported that over 400,000 people die from breast cancer each year and that 90% of deaths are due to the development of metastases and related complications (Cancer Genome Atlas Network, 2012; Siegel et al., 2023). In light of significant advances in diagnosis, surgery and the development of anticancer drugs, the survival rate for primary breast cancer is now close to 100%. However, once distant metastases occur, the survival rate drops to just 25% (Torre et al., 2016). The lack of effective therapeutic targets and predictors is one of the key reasons for the high lethality of metastatic breast cancer.

Estrogen homeostasis is vital for various physiological processes such as energy metabolism regulation and sexual development. Emerging research indicates that the disruption of estrogen balance and subsequent abnormal accumulation of estrogen are critical factors in breast cancer development (Parida and Sharma, 2019; Zhuang et al., 2022). Estrogens have been implicated in the development of breast cancer due to their ability to stimulate cell growth and proliferation via receptor-mediated processes and the presence of toxic metabolites. The parent estrogens, estrone (E1) and estradiol (E2), are hypothesized to directly promote tumorigenesis by activating the estrogen receptor (ER) and initiating downstream promitogenic transcriptional programs (Santen et al., 2015).

The endogenous conversion of estrogen into genotoxic metabolites has been identified as a potential mechanism, independent of ER signaling, contributing to estrogen-dependent breast tumorigenesis (Santen et al., 2015). After hydroxylation, estrogen forms catechol estrogens, specifically 2-hydroxyestrogens (2-OHE1/E2) and 4-hydroxyestrogens (4-OHE1/E2). These catechol estrogens can be converted into methoxyestrogens [2-methoxyestradiol/estrone (2-MeOE1/E2), 4-methoxyestradiol/estrone (4-MeOE1/E2)] through methylation by COMT enzyme or further oxidized by cytochrome P450 (CYP) enzymes to form estrogen quinones. Estrogen quinones can chemically react with the guanine and adenine bases in DNA, leading to the formation of depurinating estrogen-DNA adducts (Dallal et al., 2014; Rajan et al., 2021). Studies have reported that concentrations of 4-OHE2 in breast cancer biopsies can be up to three times higher than those observed in normal breast tissue. Our previous studies have also found that estrogen homeostasis is significantly dysregulated during the development of breast cancer (Zhou et al., 2017). In particular, concentrations of 4-hydroxyestradiol/estrone (4-OHE2/1) and 2-hydroxyestradiol/estrone (2-OHE2/1) were significantly elevated in the mammary tissue of breast cancer rats (Zhou et al., 2018). Therefore, it is reasonable to suggest that restoring estrogen homeostasis could serve as a valuable therapeutic approach for treating and potentially preventing breast cancer.

In addition to methylation, endogenous estrogens and catechol estrogens can undergo glucuronidation in the liver through the action of UDP-glucuronosyltransferases (UGTs), resulting in their conjugation with glucuronic acid. UGTs are phase II enzymes that catalyze the covalent addition of glucuronide to facilitate the elimination and metabolism of estrogen (Lu et al., 2018). In contrast to other metabolic pathways of estrogen, the UGT-mediated process leads to the formation of glucuronides, which lack biological activity and are easily excreted from tissues into the circulation (Mitra et al., 2009). In previous studies, we have established that UGTs play a significant role in the metabolism and elimination of estrogen. The glucuronidation capacity of UGTs has been shown to impact the regulation of estrogen signaling pathways and the pathogenesis of breast cancer (Hao et al., 2022). It has been reported that UGT1A1, UGT1A8, and UGT1A9 of the UGT1A family and UGT2B7 of the UGT2B family are the main metabolic enzymes that metabolize estrogen and related substances, with UGT2B7 having high expression specific to breast tissue (Zhou et al., 2018; Zhao et al., 2020). The literature reports that UGT2B7 is expressed at low levels or even absent in breast tumor tissues, leading to the accumulation of estrogen in breast tissue, while supraphysiological concentrations of estrogen can mediate the overexpression of various growth factors and promote cell growth and tumorigenesis (Guillemette et al., 2004; Zhou et al., 2017). In our previous study, we verified that UGT enzymes contribute to estrogen elimination. The glucuronidation capacity of UGT enzymes influences the estrogen signaling pathway and the pathogenesis of breast cancer. Among them, UGT2B7 may play a key role in the elimination of estrogen (Hao et al., 2022). Thus, it is clear that UGT2B7 efficiently metabolizes estrogen and its toxic metabolites in the mammary tissue of rats with breast cancer and that its functional activity will directly affect the in vivo exposure, homeostatic balance, and physiopathological activity of estrogen.

Yin-yang 1 (YY1) is a zinc finger protein that was first identified as a member of the YY family (Shi et al., 1991). As a transcription factor, its downstream target genes are involved in a range of cellular processes involved in tumor progression, including cell proliferation, invasion, metastasis, and angiogenesis (Khachigian, 2018). Structurally, YY1 has an activation domain in the N-terminus along with a repression domain in the C-terminus (Khachigian, 2018). Therefore, the transcriptional regulatory role of YY1 can either activate or suppress the transcription of its downstream genes, depending on the interaction between YY1 and its associated cofactors and the binding status of these target gene promoters (Thomas and Seto, 1999). There is growing evidence that YY1 can promote the development and progression of many cancers (Sarvagalla et al., 2019). However, its functional role in breast cancer progression is controversial (Wottrich et al., 2017). YY1 can promote the invasion of Erb-B2 receptor tyrosine kinase 2 subtype breast cancers by upregulating the expression of ERBB2 and its transcriptional coactivator protein 2 (AP-2) (Hickish et al., 2022). Furthermore, YY1 has the ability to suppress cell proliferation by upregulating the expression of breast cancer type 1 susceptibility protein (BRCA1). Moreover, YY1 binds to the BRCA1 promoter, exerting positive regulatory effects (Lee et al., 2012).

However, the exact role of YY1 in high-grade breast cancer, especially in metastatic breast cancer, needs to be further elucidated. Therefore, this study will investigate whether YY1 can directly transcribe and regulate the expression of UGT2B7, thereby inducing abnormal accumulation of estrogen in the tumor site and ultimately promoting breast cancer metastasis.

MethodsClinical Sample Collection

The clinical trial in this study was a case-control study conducted at the Affiliated Hospital of Xuzhou Medical University from October 2017 to March 2023. The tumor tissues and serum samples of the breast cancer patients came from the affiliated hospital of Xuzhou Medical University, Xuzhou, China. During surgery, Venous blood samples were collected during the follicular phase, which was determined by a questionnaire survey. Additionally, premenopausal female patients with malignant breast cancer were also recruited to collect samples of cancerous tissues and paracancerous tissues. We collected a series of fresh-frozen tissues and clinical data from breast cancer patients. Tumor tissues were obtained from patients who underwent surgery and promptly stored at –80°C. None of the female participants included in our study had undergone systemic hormone therapy. The inclusion and exclusion criteria of this study were consistent with those previously reported (Sampson et al., 2017). The experiment was approved by the Ethics Committee of the Affiliated Hospital of Xuzhou Medical University (No. XYFY2017-KL008-01), and this study was registered in the Chinese Clinical Trial Register on May 13, 2017 (No. ChiCTR-DOD-17011393) and was performed in accordance with the Declaration of Helsinki. Each patient provided written informed consent.

Animals and Treatments

All the experiments that involved animals were approved and conducted under the oversight of the Animal Ethics Committee of Xuzhou Medical University (No. 202208S040). Four-week-old female mammary-specific polyomavirus middle T antigen overexpression mouse model (MMTV-PyMT) transgenic mice were obtained from the Shanghai Southern Model Animal Centre (Shanghai, China), and nontransgenic FVB/n female siblings served as negative controls. Mice in each group were randomly assigned. This paper follows the growth characteristics of this model, which includes a proliferative phase at 4 to 5 weeks, an early carcinogenic phase at 9 to 10 weeks, and a late carcinogenic phase at 12 to 13 weeks. Therefore, MMTV-PyMT mice were divided into three groups, namely, the hyperplasia group, the early cancer group, and the late cancer group, with six mice in each group.

For the ectopic implant trial, 4-week-old female BALB/C mice were injected with 2 × 106 cancer cells that were transduced with shCtrl or shYY1 lentiviral vector mixed with Matrigel (1:1) via subcutaneous injection. The humane endpoints were when the largest tumor size was >15 mm in diameter. None of the mice reached the endpoints of the present study. The laboratory mice were housed under controlled conditions, including a temperature range of 22 to 24°C, humidity levels between 40% and 60%, a 12-h light/dark cycle, and continuous access to food and water to ensure their well-being during research. Animal health and behavior were monitored for animals every 3 days during the trial. Tumor volumes were evaluated every 3 days after injection. The tumor volume was calculated using the following formula: V = (L × W2) × 0.5 (V, volume of tumor; L, length of tumor; W, width of tumor). After 2 months, mice were humanely sacrificed under 5% isoflurane for 10 to 20 min. Mice were placed into a chamber filled with vapor of the anesthetic isoflurane until respiration ceased and continued to be exposed to isoflurane until 2 min after respiratory arrest. The tumors were weighed, and volumes were measured for tumorigenesis evaluation.

For the colonization assay, BALB/C mice were injected with 2 × 105 4T1-GFP/LUC cells that were transduced with shCtrl or shYY1 lentiviral vector mixed with Matrigel. The humane endpoints were when the mice showed hind limb weakness or paralysis. None of the mice reached the endpoints of the present study. An IVIS kinetics imaging system (Caliper LifeSciences) was applied to monitor tumor metastasis via tail vein injection to assess pulmonary metastasis in a nude mouse model, and it was monitored every 3 days by the IVIS kinetics imaging system. After 4 weeks, the mice were humanely sacrificed under overdosed isoflurane and placed into a chamber filled with vapor of the anesthetic isoflurane until respiration ceased. The lungs were collected, fixed, and sectioned. The number of metastatic lung nodules was counted using H&E staining.

Bioinformatic AnalysisCluster Analysis

Hierarchical clustering analysis was performed using Cluster 3.0 and Java Treeview software. Definitions for technical terms were provided upon their initial mention. Similarities were evaluated using the Euclidean distance algorithm, and the average linkage clustering algorithm—leveraging observation centroids—was employed for clustering. Alongside the dendrogram, a heatmap was frequently included as a visual tool. Furthermore, standard academic formatting was adhered to, ensuring the text was devoid of grammatical or punctuation inaccuracies.

Subcellular Localization

CELLO (http://cello.life.nctu.edu.tw/), which is a multiclass support vector machine classification system, was used to predict protein subcellular localization.

Domain Annotation

Protein sequences are analyzed using InterProScan to identify specific protein domain signatures contained within the Pfam member database of InterPro.

Gene Ontology Annotation

The protein sequences of the selected differentially expressed proteins underwent a localized exploration through NCBI BLAST+ client software (ncbi-blast-2.2.28+-win32.exe) and InterProScan to locate homologous sequences. Following this, the discovered sequences were annotated, and gene ontology (GO) terms were associated using Blast2GO software. The results of the GO annotations were presented visually utilizing R scripts.

Enrichment Analysis

Enrichment analysis was carried out utilizing Fisher’s exact test, with the entire set of quantified proteins employed as the reference dataset. The Benjamini–Hochberg correction method was then utilized to rectify the obtained P values for multiple comparisons. Only functional categories and pathways displaying P values lower than 0.05 were considered statistically significant.

Cell Culture and Cell Treatment

The breast cancer cell line MCF-7 was provided by Dr. Yanyan Yu. The breast cancer cell line MDA-MB-231, the breast cancer cell line 4T1, and HEK 293T cells were provided by Dr. Zhao Liu. MCF-7 cells and MDA-MB-231 cells were maintained in Dulbecco’s modified Eagle’s medium (DMEM) (KGM12800-500, Jiangsu Kaiji Biotechnology Co., Ltd.) supplemented with 10% FBS (FB25015, Clark) at 37°C in a 5% CO2 atmosphere. 4T1 cells were maintained in DMEM/F12 (C11875500BT, Gibco) supplemented with 10% FBS (FB25015, Clark). All the cell lines tested negative for mycoplasma.

Transient Transfection and Lentiviral Transduction

A YY1 overexpression plasmid and negative control vector, YY1-specific short hairpin (shRNA), and noncoding shRNA were synthesized by Genechem Co., Ltd. (Shanghai, China). The vectors used for shRNA and overexpression are GV344 and GV492. The sequence of the shRNA targeting YY1 was as follows: 5′- GCTTCGAGGATCAGATTCTCA -3′. The sequence of the control shRNA was as follows: 5′- TTCTCCGAACGTGTCACGT -3′. For transient transfection, cells were seeded in 6-well plates at a density of 4 × 105 cells/well. After reaching 70% confluence, the cells were transfected with the shRNAs or plasmids. shRNA (80 pmol) or plasmids (2 µg) were transfected into the cells using Lipofectamine 2000 reagent (GenePharma Co., Ltd.) at 37°C for 8 h according to the manufacturer’s protocol. Eight hours after transfection, the cell culture medium was discarded and fresh DMEM containing 10% FBS was added to each well. After 48 h, transfection efficiency was further assessed by Western blot analysis and the following experiments were performed. For lentiviral transduction, the second-generation system was used in the lentivirus transduction experiment. Multiplicity of infection values and optimal infection conditions were determined by pretesting. Cells were seeded in 6-well plates at a density of 3 × 105 cells/well. After reaching 30% confluence, the cells were transfected with the shRNAs. The shRNAs (multiplicity of infection, 40) were transduced into cells at 37°C for 24 h according to the manufacturer’s protocol. The cell culture medium was then discarded and fresh DMEM containing 10% FBS was added to each well. After 72 h, transduction efficiency was observed using a fluorescence microscope. Stable YY1 knockdown or overexpression cell lines were selected using 2 µg/ml puromycin for 7 days. Western blot analysis was performed to detect the knockdown or overexpression efficiency of YY1.

Liquid Chromatography with Tandem Mass Spectrometry

E1, E2, estriol (E3), 16-epiestriol, 17-epiestriol, 2-OHE2/1, 4-OHE2/1, 16α-hydroxyestrone (16α-OHE1), 2-MeOE2/1, 4-MeOE2/1, d5-E2, dansylchloride, and vitamin C were purchased from Sigma-Aldrich.

The analysis was performed with the high-pressure liquid chromatography series system (SCIEX ExionLC AD, Singapore) coupled with a linear ion trap mass spectrometer as the detector (SCIEX TRIPLEQUAD 5500, AB Sciex Instruments, Singapore) under electrospray ionization conditions in positive mode. Chromatographic separation was carried out using an Agilent ZORBAX Extend-C18 column (2.1 mm inner diameter × 100 mm length with a particle size of 1.8 µm) maintained at a temperature of 40°C. The separation process involved the use of mobile phases A (consisting of 100% H2O with 0.1% formic acid) and B (comprising 100% acetonitrile with 0.1% formic acid). The gradient elution was programmed in the following manner: initiating with 70% B for 2 min, followed by a linear increase from 30% B to 90% B over 2.5 min, a wash step with 90% B for 2 min, a gradual rise from 90% B to 95% B in 1.5 min, a wash with 95% B for 4 min, returning to 70% B over 0.5 min, and re-equilibration for 1.5 min, resulting in a total runtime of 13 min. The flow rate was set at 0.3 mL/min. Multiple reaction monitoring scan mode was used for positive ion detection; detection of ion pairs, residence time, fragmentation voltage, and collision energy are listed in Supplemental Table 1. The method of liquid chromatography with tandem mass spectrometry (LC-MS/MS) was carried out according to the method previously reported (Xu et al., 2007).

Quantitative Reverse Transcriptase Polymerase Chain Reaction

RNA was extracted with TRIzol® reagent and isolated according to the manufacturer’s protocol, and cDNA was generated using reverse transcription (PrimeScript RT kit). mRNA expression was measured by quantitative reverse transcriptase polymerase chain reaction analysis using a LightCycler® 480 II. The primers were designed by Sangon Biotech, and their specificity was checked by melting curve analysis. The sequences of the primers are listed in Supplemental Table 2.

Western Blotting

After total protein extraction, the proteins were separated by SDS–PAGE, and their content was quantified by immunoblotting with specific antibodies. Primary antibodies included antibodies against YY1 (D595Z, Cell Signaling Technology), UGT2B7 (DF12140, Affinity), E-cadherin (BS1098, Bioworld), N-cadherin (22018-1-AP, Proteintech), Vimentin (BS4483, Bioworld), MMP2 (66366-Ig, Proteintech), MMP9 (10375-2-AP, Proteintech), and GAPDH (AP0063, Bioworld). The intensity of the bands was quantified using Odyssey®Sa (LICOR, USA). The expression of GAPDH was used as a control to quantify the signal intensity of the target protein.

Wound Healing Assay

The scratch assay was performed with 105 cells in a 6-well plate. After the cell density reached 80% to 90% confluence, a scratch was created under sterile conditions using a 200 μl sterile pipette tip, and the migration of cells toward the notch was observed. Then, the cells were washed twice with PBS to remove dead cells and cultured in FBS-free medium. Three replicate wells were performed in 6-well plates for each experimental condition.

Coimmunoprecipitation

The interaction between YY1 and UGT2B7 was analyzed by coimmunoprecipitation (Co-IP) using YY1 and UGT2B7 antibodies. For CoIP assays, a total of approximately 500 μg of protein extract was incubated with 1 mg of normal IgG (Beyotime Biotech Inc., Jiangsu, China) and 10 μl of fully mixed Protein A/G Agarose (Beyotime Biotech Inc.) before being gently shaken at 4°C for 2 h to prevent nonspecific binding. The supernatant was then collected and used for immunoprecipitation following centrifugation at 3000 g for 5 min. Next, 2 μg of either YY1 or UGT2B7 primary antibody was added to the supernatant and gently shaken overnight at 4°C. IgG was employed as a negative control. The following day, 20 µL of protein A/G Agarose in a completely suspended state was introduced into the reaction mixture and left at 4°C for 4 h. The beads containing bound protein were then precipitated by centrifugation and thoroughly washed five times with Co-IP buffer. They were subsequently denatured by boiling in SDS-sample buffer before the resulting immunoprecipitate was subjected to immunoblot analysis. The method of Co-IP was carried out according to the method previously reported.

Chromatin Immunoprecipitation Assays

Chromatin immunoprecipitation (ChIP) assays were used to measure the binding of YY1 to the CpG-rich region of the UGT2B7 promoter, which we detected using the ChIP kit (Epigentek, p-2001). The method of Co-IP was carried out according to the method previously reported. Quantitative PCR was performed to detect protein-associated promoter regions using the primers listed in Supplemental Table 2.

Promoter Luciferase Assay

For 293T cells, 1 × 105 cells were seeded in a 12-well plate and cotransfected with wild-type or mutant YY1 and Renilla. Luc (for normalization) and UGT2B7.The method of Co-IP was carried out according to the method previously reported.

Immunohistochemistry

Formalin-fixed paraffin-embedded sections of primary and xenograft tumor tissue were used for immunohistochemistry (IHC). IHC assays were performed as previously described (Rodriguez et al., 2020). The tissue slides were subjected to deparaffinization, rehydration, and epitope retrieval and were incubated at 4 °C overnight with primary antibodies against YY1(66281-1-Ig, Proteintech) and UGT2B7(DF12140, Affinity). Slides were examined with a digital section scanning system (Olympus VS120 microscope). Linearity measurements were performed with Image-Pro Plus (Media Cybernetics, Silver Spring, MD).

Statistical Analysis

Statistical analyses were performed using Prism8 software. Kaplan–Meier analysis was used to calculate the survival differences between divided groups, and a log-rank test was used to compare differences. All differences between groups were examined for statistical significance using a two-tailed Student’s t- test or one-way ANOVA to compare multiple groups; the results are presented as the mean ± S.E.M. A P value of < 0.05 was considered statistically significant for all the data sets. All in vitro and in vivo experiments were repeated at least three times.

ResultsTandem Mass Tag Quantitative Proteomics Screening for Differentially Expressed Proteins

To identify key factors that promote breast cancer metastasis, we performed tandem mass tag (TMT) quantitative proteomics analysis of six metastatic breast cancer tumor tissues and six nonmetastatic breast cancer tumor tissues. The results showed that the proteomics analysis yielded a total of 849,433 secondary spectra (total number of database matched spectra 49,449), 23,092 peptides identified (total number of unique peptides 21,098) and 4320 proteins identified (4313 quantifiable proteins) (Fig. 1A). The proteins in the comparison groups were plotted on a volcano plot (Fig. 1B) using the two factors of fold change and P value (T test) as criteria, with fold change > 1.2-fold (upregulation > 1.2-fold or downregulation < 0.83-fold), and P value < 0.05 (T test or other) as the criteria; 767 upregulated proteins and 798 downregulated proteins were identified between the comparison groups. Then, we performed subcellular localization analysis of all differentially expressed proteins using the subcellular structure prediction software CELLO, and the results showed that most of the differentially expressed proteins were localized in the nucleus (Fig. 1C). The structural domain prediction software InterProScan was used to predict the structural domains of the differentially expressed proteins, and the results showed that the nucleic acid recognition region was a significantly enriched structural domain for the differentially expressed proteins (Fig. 1D). GO functional annotation of all differentially expressed proteins was performed using Blast2Go (https://www.blast2go.com/) software, and the number of differentially expressed proteins was counted at the GO secondary functional annotation level (Fig. 1E). To comprehensively figure out the functions, localization, and biological pathways related to differential proteins, protein annotation was performed through the GO analysis. The top 20 items in cellular components, biological processes, and molecular functions are listed in Fig. 1, F–H. When referring to the biological process category, nuclear-transcribed mRNA catabolic process was significantly altered, and YY1 was involved in this process as a nuclear transcription factor (Fig. 1F). For the molecular function, structural molecule activity was changed significantly, and YY1 was involved in this function as it has a regulatory function in the activity of DNA-binding transcription factors (Fig. 1G). Thus, it is reasonable to believe that this protein is relevant to the enriched GO terms.

Fig. 1.

Screening of the most differential protein YY1 by proteomic analysis. (A) Statistical histogram of identification and quantitative results. (B) Differential protein volcano plot. The horizontal coordinate is the difference in fold; the vertical coordinate is the significance of the difference in P value. (C) Pie chart showing the number and proportion of proteins in each subcellular organelle. (D) Bar chart showing the number of proteins in domain (top 20). (E) The GO annotation statistics of differentially expressed proteins (level 2) indicate the secondary function annotation information. (F) GO functional enrichment bubble diagram under biological process classification. (G) Bubble diagram of GO functional enrichment under cell component classification. (H) Molecular functional classification under GO functional enrichment bubble map. (I) Heat map for the top 10 upregulated proteins and top 10 downregulated proteins.

Finally, we comprehensively screened 207 differentially expressed target proteins according to the screening principles (including t-test, difference ploidy greater than 2, expression abundance analysis, etc.), including 106 proteins whose functions and structures were included in the National Center for Biotechnology Information and UniProt protein databases. We selected the top 10 proteins with upregulated and downregulated ploidy for analysis and found that the most differentially expressed protein in the proteomics assay was YY1 (Fig. 1I, Supplemental Table 3). YY1 was selected as the candidate protein that is located in the nucleus, has a regulatory function in DNA binding transcription factor activity, and has the highest differential ploidy.

High Expression of YY1 in Breast Cancer Is Associated with Metastasis

A plethora of evidence suggests that YY1 plays a crucial role in cancer cell proliferation and tumor growth. However, the involvement of YY1 in cancer metastasis remains unclear. To address this issue, we initially employed an IHC technique to examine the expression of YY1 in human breast cancer samples. As shown in Fig. 2A, YY1 protein levels were dramatically increased in metastasized breast cancer samples. In addition, the protein levels of MMP9, which is a common indicator protein in cancer metastasis, were detected in clinical samples. IHC analyses showed that MMP9 protein levels were significantly increased in metastasized breast cancer samples when compared with the primary breast cancer samples and the adjacent tissues (Fig. 2B). Notably, the expression of the YY1 and MMP9 proteins exhibited a clear correlation (R2 = 0.2568 and P = 0.003, Supplemental Fig. 1A). Furthermore, patients with breast cancer with either high YY1 and MMP9 mRNA levels had decreased recurrence-free survival (Fig. 2, C–D). In addition, YY1 protein expression was significantly increased in a higher degree of breast cancer specimens (Fig. 2E). These results indicated that elevated levels of YY1 expression are closely associated with the development and metastasis of breast cancer.

Fig. 2.

Confirmation of the correlation between YY1 expression and breast cancer progression and metastasis. (A) Expression levels of YY1 during breast disease progression. The data are presented as mean ± S.E.M. and significant differences detected using T test. (B) Expression levels of MMP9 during breast disease progression. The data are presented as mean ± S.E.M. and significant differences detected using T test. (C) Overall survival curve. K-M Plotter was used to analyze the role of YY1 in breast cancer prognosis. (D) Overall survival curve. K-M Plotter was used to analyze the role of MMP9 in breast cancer prognosis. (E) YY1 expression in tumors of breast cancer samples at different stages was analyzed by IHC. The data are presented as mean ± S.E.M. and significant differences detected using T test. (F) The protein expression levels of YY1 in MMTV-PyMT mice at different stages was analyzed by Western blot. The data are presented as mean ± S.E.M. and significant differences detected using T test, n = 6. (G) YY1 expression in tumors of MMTV-PyMT mice at different stages was analyzed by IHC, n = 6. The data are presented as mean ± S.E.M. ***P < 0.001, **P < 0.01, *P < 0.05.

In an effort to characterize the dynamic changes in estrogen during breast cancer metastasis, we analyzed the changes in estrogen content at different stages, utilizing the MMTV-PyMT of breast cancer. MMTV-PyMT mice are characterized by multistep carcinogenesis, including hyperplasia at 4 to 5 weeks, carcinoma at 9 to 10 weeks, and advanced carcinoma at 12 to 13 weeks (Ershaid et al., 2019).

To investigate the changes in YY1 throughout breast cancer progression, we analyzed lung nodules in MMTV-PyMT mice at different times. As shown in Supplemental Fig. 1, B–C, MMTV-PyMT mice with advanced carcinoma developed distinct lung nodules, indicating that distant metastases had developed by this time. We examined tumor tissues from MMTV-PyMT mice at distinct tumorigenic stages including a hyperplasia group, carcinoma group, and advanced carcinoma group and found that YY1 expression increased significantly as breast cancer progression advanced, with the highest levels occurring when distal metastases occurred (Fig. 2F). Moreover, as shown in Fig. 2G, IHC staining also confirmed that YY1 expression increased significantly when breast cancer had metastasized. Taken together, these results suggest that elevated expression of YY1 is closely linked to breast cancer metastasis.

Alteration of YY1 Expression Impacts Cancer Cell Invasion and Migration and the Expression of Epithelial-to-Mesenchymal Transition Genes

We found that YY1 knockdown in MCF-7 cells and MDA-MB-231 cells (Fig. 3A, Supplemental Fig. 2, A–B) significantly inhibited migration and invasion ability (Fig. 3, C and E). We also found that YY1 overexpressed in MCF-7 and MDA-MB-231 cells (Fig. 3B, Supplemental Fig. 2, C–D) significantly promoted breast cancer cell migration and invasion ability (Fig. 3, D and F).

Fig. 3.

YY1 expression can promote breast cancer metastasis. (A and B) Changes in the protein expression levels of YY1 in MCF-7 cells and MDA-MB-231cells after virus infection. Three replicate experiments were performed, and the data are presented as mean ± S.E.M. and significant differences detected using T test. (C–F) MCF-7 cells stably expressing YY1-shCON, YY1-shRNA, MDA-MB-231 cells stably expressing YY1-shCON, YY1-shRNA, MCF-7 cells stably expressing YY1-OENC, YY1-OE and MDA-MB-231 cells stably expressing YY1-OENC, YY1-OE were subjected to wound healing assay for cell migration and transwell assays for cell invasion. Three replicate experiments were performed, and the data are presented as mean ± S.E.M. and significant differences detected using T test. (G and H) The protein expression levels of E-cadherin、N-cadherin、Vimentin、MMP2, and MMP9 in YY1-shRNA and YY1-OE cell models of MCF-7 cells and MDA-MB-231 cells. The data are presented as mean ± S.E.M. n = 3. ***P < 0.001, **P < 0.01, *P < 0.05.

The initiation of cancer metastasis, including invasion, migration, and intravascular metastasis, is heavily dependent on tumor cell spread. Consistent with previous observations, reduced expression of YY1 resulted in changes in epithelial-to-mesenchymal transition (EMT) genes, including E-cadherin, N-cadherin, Vimentin, MMP2, and MMP9, all of which are essential for the mobility and dissemination of cancer cells (Luo et al., 2018; Qin et al., 2018) (Fig. 3G). Conversely, overexpression of YY1 upregulated these genes (Fig. 3H). Taken together, these results suggest that upregulation of YY1 expression promotes the migration and invasion of breast cancer cells, ultimately leading to breast cancer metastasis.

YY1 Regulates the UGT-Mediated Estrogen Metabolic Axis to Promote Breast Cancer Metastasis

First, this study has confirmed that alterations in YY1 expression can affect breast cancer metastasis. Abnormal estrogen metabolism plays a critical role in breast cancer metastasis, but it is unknown whether YY1 can regulate estrogen metabolism to affect breast cancer metastasis. Since UGT2B7 is a critical mediator of estrogen accumulation, we determined the effect of YY1 on UGT2B7 expression. As shown in Fig. 4A and Supplemental Fig. 3A, we found that UGT2B7 expression was increased after knockdown of YY1 in MCF-7 and MDA-MB-231 cells. Not surprisingly, UGT2B7 expression was significantly reduced when YY1 was overexpressed (Fig. 4B, Supplemental Fig. 3B). Other enzymes responsible for the estrogen metabolism, including COMT, CYP1A1, and CYP1B1, were also detected (Supplemental Fig. 3, C–D). These findings suggest that YY1 may regulate estrogen metabolism by modulating UGT2B7 expression.

Fig. 4.

YY1 can repress expression by regulating UGT2B7 transcription. (A and B) The protein expression levels of UGT2B7 in YY1-shRNA and YY1-OE cell models of MCF-7 cells and MDA-MB-231 cells. Three replicate experiments were performed, and the data are presented as mean ± S.E.M. and significant differences detected using T test. (C) Diagram shows three putative YY1 binding sites on the UGT2B7 promoter predicted using ISMAR and the mutated promoter sequences used in luciferase reporter assays. Luciferase reporter assays were conducted using 293T cells cotransfected with pcDNA. YY1 and the wild-type (Wt) or mutant UGT2B7 promoter constructs. Three replicate experiments were performed, and the data are presented as mean ± S.E.M. and significant differences detected using T test. (D) Co-IP of UGT2B7 and YY1 in MCF-7 cells and MDA-MB-231 cells. IgG was used as a control. The experiments were repeated three times. Co-IP of YY1 and UGT2B7 in MCF-7 cells and MDA-MB-231 cells. IgG was used as a control. The experiments were repeated three times. (E) ChIP-quantitative polymerase chain reaction analysis of UGT2B7 occupancy on the YY1 promoter region containing putative UGT2B7 binding site in YY1-shRNA and YY1-OE cell models of MCF-7 cells and MDA-MB-231 cells. Three replicate experiments were performed, and the data are presented as mean ± S.E.M. and significant differences detected using T test. (F-G) The protein expression levels of UGT2B7 in MMTV-PyMT mice at different stages was analyzed by Western blot and IHC. The data are presented as mean ± S.E.M. and significant differences detected using T test, n = 6. (H) The protein expression levels of UGT2B7 in breast cancer samples at different stages was analyzed by IHC. The data are presented as mean ± S.E.M. and significant differences detected using T test, n = 6. ***P < 0.001, **P < 0.01, *P < 0.05.

To further elucidate whether YY1 directly suppressed UGT2B7 transcription, we used promoter luciferase assay to identify putative YY1 binding sites on the UGT2B7 promoter and found that mutation of one of these sites (–1396∼–1391) abolished YY1 inhibition of UGT2B7 expression. We identified the putative YY1 binding site on the UGT2B7 promoter by a dual luciferase reporter gene and found that mutation of one of the sites (TCCATT) eliminated the inhibitory effect of YY1 on UGT2B7 (Fig. 4C). To understand how YY1 carried out this transcriptional suppression, we performed Co-IP assays and found that YY1 interacted directly with UGT2B7 in breast cancer cells (Fig. 4D). To further investigate whether YY1 binds to the UGT2B7 promoter and undergoes transcriptional repression, we performed ChIP assays using cells with or without YY1 knockdown. The results showed that YY1 bound to UGT2B7 at the initiation sequence and significantly suppressed UGT2B7 expression. We carried out the same experiments on cells with or without YY1 overexpression, and the conclusions were consistent (Fig. 4E).

Furthermore, animal experiments in vivo were conducted to verify that YY1 has a direct effect on UGT2B7. We found that the expression of UGT2B7 decreased significantly with cancer progression, especially when metastasis occurred and UGT2B7 was barely expressed. As shown in Fig. 4F, the expression of UGT2B7 in the tumor tissues of MMTV mice with advanced carcinoma was much lower than that of mice in other stages. In addition, we further validated this result by IHC experiments (Fig. 4G). We observed that UGT2B7 expression was lower in tumor tissues from metastatic breast cancer patients than in those from the primary group by IHC of tumor tissues from breast cancer patients; furthermore, UGT2B7 expression was highest in nontumor tissues, and the results were highly consistent with previous results (Fig. 4H). Taken together, these results indicate that YY1 is a direct transcription factor of UGT2B7 that suppresses UGT2B7 expression through transcriptional regulation.

Abnormal Accumulation of Estrogen During Breast Cancer Metastasis

Our group is working on exploring the detection of estrogen levels in serum and tissues and has utilized a well-established method for the detection of estrogen based on LC-MS/MS (Hao et al., 2022). A total of 146 premenopausal patients with metastatic breast cancer (metastatic group), 132 premenopausal patients with primary breast cancer (primary group), and 143 matched control subjects (normal group) were included in the study. An LC-MS/MS analytical method was employed to measure the concentrations of E1, E2, E3, and 11 other estrogen metabolites in serum samples. We found an increase in total estrogen accumulation as the breast cancer metastasis, and multivariate analysis (Orthogonal Projections to Latent Structures-Discriminant Analysis), which is an unsupervised statistical method, suggested that with the occurrence of breast cancer metastasis, there are changes in estrogen metabolism (Fig. 5A, Supplemental Fig. 4A). Similarly, six cases of tumor tissue from patients diagnosed with metastatic breast cancer, six cases of tumor tissues from nonmetastatic breast cancer patients, and six cases of paracancerous tissues were tested for hormones; analysis revealed estrogen accumulation in patients with metastatic breast cancer (Fig. 5B, Supplemental Fig. 4B). To better substantiate our results, we performed the same hormone assays and analyses on serum and tumor tissue from MMTV-PyMT mice at different times, and the results once again demonstrated that breast cancer metastasis is closely associated with abnormal estrogen accumulation (Fig. 5, C–D, Supplemental Fig. 4, C–D).