Research ArticleExperimental Studies
Open Access

Estrogen Effects Differ Between Medium Maintenance and Replacement from Transcriptional and Clinical Perspectives in T47D Breast Cancer Cells

SEOK-HOON JANG, SE HYUN PAEK, JONG-KYU KIM, JE KYUNG SEONG and WOOSUNG LIM
Anticancer Research October 2023, 43 (10) 4447-4469; DOI: https://doi.org/10.21873/anticanres.16640

Abstract

Background/Aim: Our most recent study revealed that the responsiveness of hormone receptor-positive breast cancer (HR+ BC) cells to estrogen or endocrine therapy can be altered by certain cell culture or ambient environmental conditions. Nevertheless, we were unable to investigate the relevant molecular mechanism and clinical relevance. Therefore, this study was planned as a follow-up. Materials and Methods: RNA sequencing was mainly used with T47D cells treated with or without 17β-estradiol (E2) under medium maintenance (MTN; conventional culture method) and medium exchange (EXC; daily replacing the existing medium with fresh medium). Results: The role of E2 in transcription differed between MTN and EXC, and E2 played more important roles in transcription in terms of cancer development under EXC than under MTN, consistent with the previous functional effects of EXC. The novel concept of the “estrogen-responsive and proliferation-related gene (ERPG)” was introduced. The expression of ERPGs, which were distinguished from typical estrogen-responsive genes, was correlated with that of prognostic and predictive factors for HR+ BC. The transcriptional induction of ERPGs and typical estrogen-responsive genes regardless of E2 treatment under MTN was reminiscent of constitutive estrogen receptor (ER) activation. Additionally, phosphatidylinositol 3-kinase (PI3K)/protein kinase B (AKT)/mammalian target of rapamycin (mTOR) inhibitors were more effective under EXC than under MTN. Conclusion: This study, demonstrating the more important roles of estrogen in terms of cancer development under EXC than under MTN, supports the use of our research model in future studies to overcome endocrine resistance in HR+ BC.

Key Words:

Breast cancer (BC) is the most threatening cancer for women worldwide, accounting for 24.5% of cancer cases and 15.5% of cancer deaths (1). Despite its favorable therapeutic outcomes (5-year relative survival rate: 93.2%), there is still a risk of progression and recurrence (10-year relative survival rate: 87.7%) (2, 3). In comparison with the human epidermal growth factor receptor 2-positive (HER2+) and triple-negative breast cancer (TNBC) subtypes of BC, the hormone receptor-positive (HR+) or luminal subtype is much more common, accounting for approximately 80% of BCs (4). HR indicates estrogen receptor (ER; particularly ERα) and/or progesterone receptor (PR), and ER plays a key role in the development of not only the normal breast but also breast cancer as a type of transcription factor (5-7). Endocrine therapy (ET) or hormone therapy tailored to HR+ BC is generally effective, but ET resistance (ET-R) is much more common (30%-50%) than expected, suggesting that ET-R is a major cause of the recurrence and dissemination of HR+ BC (4, 7-10). ET drugs include selective ER modulators (SERMs) such as tamoxifen and its derivatives, selective ER down-regulators (SERDs) such as fulvestrant (Fulv), and aromatase inhibitors (AIs), which are effective only for postmenopausal women with HR+ BC, such as letrozole, anastrozole, and exemestane (8). Particularly, the aromatase enzyme uniquely converts androgenic precursors into estrogens in vertebrates (8).

A shared characteristic among HR+ BC cell lines is that they are inherently dependent on estrogen for their proliferation and survival. However, our most recent study revealed that such dependence can change under certain culture conditions, and that those cells can be intrinsically resistant to estrogen or ET (11). More specifically, a low cell density or medium exchange (EXC) increased cellular responsiveness to estrogen, and cell culture supernatants had the opposite effect. Thus, secreted substances became an important part of estrogen research. Those findings suggest that long-term artificial interventions, including long-term estrogen deprivation (LTED) and long-term ET, which have potential collateral effects, are not necessary for ET-R (12-17). Nonetheless, we were previously unable to investigate the molecular mechanism and clinical relevance related to EXC (i.e., certain autocrine factors) in depth.

Therefore, this study was planned as a follow-up to the previous study and mainly used RNA sequencing (RNA-seq) in T47D HR+ BC cells. The present findings demonstrate not only the different effects of estrogen on transcription between EXC and medium maintenance (MTN) conditions but also the possible clinical applications of our research model, which could contribute to efforts to overcome ET-R in HR+ BC.

Materials and Methods

Cell culture. T47D cells were obtained from the Korean Cell Line Bank (KCLB; Seoul, Republic of Korea) and cultured in RPMI 1640 (Cat. No. LM 011-01) supplemented with 10% fetal bovine serum (Cat. No. S 001-01) and 0.5× penicillin-streptomycin (Cat. No. LS 202-02) at 37°C in a humidified atmosphere with 5% CO2. These culture reagents and the cultureware (Cat. Nos. 20101, 30012, and 30096) were purchased from WELGENE (Gyeongsan-si, Republic of Korea) and SPL Life Sciences (Pocheon-si, Republic of Korea), respectively. Under MTN, medium used to seed cells was maintained until an experiment for analysis as in conventional cell culture experiments, while under EXC, the medium was replaced with fresh medium every day until the experiment (a total of three times).

Reagents and treatment. A type of estrogen, 17β-estradiol (E2; Cat. No. E8875), Fulv (Cat. No. F1144), alpelisib (Cat. No. A8346) and everolimus (Cat. No. A8169), and MK-2206 (Cat. No. S1078) were purchased from Sigma-Aldrich (St. Louis, MO, USA), Tokyo Chemical Industry (Tokyo, Japan), ApexBio Technology (Houston, TX, USA), and Selleck Chemicals (Houston, TX, USA), respectively. These reagents were dissolved in dimethyl sulfoxide (DMSO; Cat. No. D1370) (Duchefa Biochemie, Haarlem, the Netherlands), which was also used for the vehicle control (CTL), to obtain 3,000× working solutions. Except for E2 (300 pg/ml), all other reagents were administered at a final concentration of 0.1 μM. One day after the cells were seeded at 4×104 cells/ml for RNA-seq or 2×104 cells/ml for 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay, they were treated with a reagent by adding a 20 μl mixture of reagent-RPMI 1640 to the medium. RNA isolation or MTT assay was performed three days after the treatment. For EXC, the existing medium was replaced with fresh medium daily for three days, and a fresh reagent was added each time.

RNA-seq and transcriptome analysis. RNA-seq was performed in T47D cells treated with and without E2 treatment under MTN and EXC. After total RNA was extracted using TRIzol Reagent (Thermo-Fisher Scientific, Waltham, MA, USA), RNA-seq and analysis were conducted by Macrogen, Inc. (Seoul, Republic of Korea). Libraries were prepared using the TruSeq Stranded mRNA Library Prep Kit (Illumina, San Diego, CA, USA) according to the manufacturer’s instructions and sequenced as paired-end reads using NovaSeq 6000 (Illumina). The quality of the RNA and library was assessed using TapeStation D1000 ScreenTape (Agilent Technologies, Santa Clara, CA, USA). Sequenced raw reads were quality-checked and preprocessed using FastQC (v0.11.7) and Trimmomatic (v0.38) (18). Then, the trimmed reads were aligned to the human reference genome of GRCh38 using HISAT2 (v2.1.0) (19). Using StringTie (v2.1.3b), the reads were further assembled into known transcripts, and the total number of assembled reads in each transcript was measured as the read count (20). For comparison between two samples, the Bioconductor package of edgeR was used to identify differentially expressed genes (DEGs) based on the fold change and raw p-value, using the trimmed mean of M-value (TMM) normalization and the exactTest method after filtering out low-quality genes (21). The N-value for the gene expression level was calculated by TMM normalization of the read count, taking the size factor into account. Genes with a read count of zero in at least one sample were excluded, and only protein-coding genes were analyzed in this study. Hierarchical clustering analysis (HCA) based on the Euclidean distance and complete linkage was performed using Morpheus software (Broad Institute, Cambridge, MA, USA). For statistical comparison, the relative expression levels of the genes were used, with each N-value normalized to the mean of the corresponding values of all four samples. To identify specific signaling pathways induced by E2, Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis was conducted with DEGs with ≥1.5-fold change due to E2 treatment using the KEGG PATHWAY database (Kanehisa Laboratories, Kyoto, Japan) (22).

Cell proliferation assay. MTT assay was performed as described previously (11). In brief, after the cells were exposed to solution containing 0.25 mg/ml MTT (Cat. No. M5655; Sigma-Aldrich), they were incubated at 37°C for 30 min. Then, the solution was removed, and the precipitate was dissolved in DMSO. Finally, the absorbance was measured at 570 nm with an appropriate volume of the mixtures without reaching absorbance saturation using a microplate reader (Molecular Devices, San Jose, CA, USA). The result value was obtained by subtracting the blank (DMSO) value from the measured value. The relative value, calculated by dividing the result value of the reagent group by that of the CTL group, was used for comparison.

Statistical analysis. Data used for statistical analysis are presented as the mean±standard deviation (SD). For comparison between sample groups, p-values were calculated by repeated measures or two-way analysis of variance (ANOVA) with Bonferroni post hoc analysis or Student’s t-test using GraphPad Prism 5 (GraphPad Software Inc., San Diego, CA, USA). A p-value of less than 0.05 was considered significant, and p<0.05, p<0.01, and p<0.001 are shown as *, **, and ***, respectively.

Results

Greater response of “typical” E2-responsive genes to E2 under EXC than under MTN. RNA-seq results in T47D cells were verified based on well-known E2-responsive genes, which have been well-summarized in a previous meta-analysis (23). Only genes up-regulated by E2, as covered in the meta-analysis, were examined. Genes changed by ≥1.2-fold due to E2 treatment were considered DEGs. Of the top 65 genes up-regulated by E2 in the early or late phase in the meta-analysis, 35 (53.8%) genes were up-regulated by E2 under an MTN or EXC condition, and the average fold increase was 1.93 (Table I). Among the 35 validated genes, GREB1, PGR, IGFBP4, EGR3, TFF1, and MYC are known direct target genes of ERα (24-29). In particular, 20 (30.8%) genes were up-regulated by E2 under both MTN and EXC, and two (3.1%) and 13 (20.0%) genes were exclusively up-regulated by E2 under MTN and EXC, respectively (Table I). However, only two (3.1%) and four (6.2%) genes were rather down-regulated by E2 under MTN and EXC, respectively (Table I). The validation results demonstrate that our experimental conditions and RNA-seq analyses are reliable, and that “typical” estrogen-responsive genes respond more to E2 under EXC than under MTN.

View this table:
Table I.

Expression levels of 65 well-known E2-responsive (E2-up-regulated) genes based on E2 treatment and EXC.

Different effects of E2 on transcription between MTN and EXC, and partial mimicry of E2 treatment under MTN by EXC alone in transcription. Overall gene expression levels were compared using HCA. Unexpectedly, global expression profiles were distinct among the four samples, and the greatest distance was between the CTLs of MTN and EXC, which was reduced by E2 treatment (Figure 1A). These contrasting expression profiles were also observed in a Venn diagram summarizing the RNA-seq results into the number of DEGs that differed by ≥1.5-fold between two groups according to three criteria to be used to evaluate an E2 effect under MTN, an E2 effect under EXC, and the effect of EXC alone (Figure 1B). In particular, the effectiveness of E2 on transcription was greater under EXC (713 genes up-regulated and 579 genes down-regulated) than under MTN (605 genes up-regulated and 426 genes down-regulated), and 69.2% to 77.4% (70.9% up-regulated and 69.2% down-regulated under MTN; 75.3% up-regulated and 77.4% down-regulated under EXC) of the genes responded to E2 in a mutually exclusive manner between MTN and EXC (Figure 1B). Interestingly, the effectiveness of EXC alone was greater than expected (759 genes up-regulated and 655 genes down-regulated), and 22.7% to 32.3% (245 genes up-regulated and 149 genes down-regulated) of genes induced by EXC alone were induced by E2 under MTN in the same way, whereas relatively, only a very small number of genes (31 genes up-regulated and 24 genes down-regulated) of the EXC-induced genes were induced by E2 under EXC (Figure 1B). Taken together, these results show that the effects of estrogen on transcription differ significantly between MTN and EXC, and that EXC alone can partially mimic estrogen under MTN in transcription.

Figure 1.
Figure 1.

Effects of E2 and EXC on overall transcription based on hierarchical clustering analysis (HCA) and Venn diagram analysis. (A) HCA and heatmap results of all genes: The gene expression level is expressed as log2(1+N), where “1+N” converts a negative value to a positive one; N indicates the N-value. (B) Venn diagram analysis of differentially expressed genes (DEGs) with a fold change of ≥1.5 based on the three classifications: In the MTN-E2/CTL and EXC-E2/CTL sets, each gene expression value (N) of MTN-E2 and EXC-E2 was divided by the corresponding N-value of MTN-CTL and EXC-CTL, respectively; in the EXC-CTL/MTN-CTL set, each N-value of EXC-CTL was divided by the corresponding N-value of MTN-CTL; expression ratios of ≥1.5 and ≤0.66 were defined as up-regulated (marked by an upward arrow) and down-regulated (marked by a downward arrow), respectively; ERPGs are specifically indicated.

Importance of genes that respond to E2 under EXC in terms of cancer development. Under usual cell culture conditions, only estrogen combined with EXC can promote the proliferation of T47D cells, whereas EXC alone reduces their proliferation (11). These phenomena suggest that genes that respond to estrogen under MTN or respond to EXC alone do not play a major role in promoting the proliferation of T47D cells. Hence, we focused on genes that responded to E2 only under EXC. We identified such genes by subtracting the MTN-E2/CTL and EXC-CTL/MTN-CTL sets from the EXC-E2/CTL set in the Venn diagram and named them as “estrogen-responsive and proliferation-related genes (ERPGs)” (Figure 1B). For example, both EGR3 and PDZK1 are known as estrogen-responsive genes (27, 30). However, PDZK1 was up-regulated by E2 only under EXC and not up-regulated by EXC alone, whereas EGR3 was up-regulated by E2 under both MTN and EXC (when applying ≥2.0-fold DEGs) (Table I). Accordingly, only PDZK1 is considered an ERPG based on our criteria. Statistically significant E2-responsive genes in T47D cells under MTN and EXC are listed in Table II and Table III, respectively, and the latter includes ERPGs. Among these ERPGs, RET is notable for its frequent over-expression in HR+ BCs with ET-R (31). Additionally, the ERPGs included FLT1 (or VEGFR1; Table III), AKT3 (p=0.156), and FOS (p=0.101) (32-34). These findings indicate the important role of ERPGs in the development and treatment of BC.

View this table:
Table II.

Statistically significant differentially expressed genes (DEGs) induced by E2 treatment under MTN.

View this table:
Table III.

Statistically significant differentially expressed genes (DEGs) (including ERPGs) induced by E2 treatment under EXC.

Induction of ERPG expression regardless of E2 treatment under MTN and greater influence of E2 under EXC than under MTN in the context of cancer development. To elucidate the significance of ERPGs further, HCA was performed using ERPGs up-regulated by ≥1.5-fold due to E2 treatment. Interestingly, regardless of E2 treatment, the ERPGs remained up-regulated under MTN to a similar extent as up-regulated by E2 treatment under EXC (Figure 2A). The result was consistent when the criterion of ≥2-fold change and mean-normalized gene expression levels were employed (Figure 2B). Similar results were obtained for ERPGs down-regulated by E2 (Figure 2C and D). Taken together, these results suggest that certain pathways, such as the ERα pathway, that regulate the expression of ERPGs are constitutively activated independently of estrogen under usual culture conditions, and that certain autocrine factors induce this constitutive activation (11).

Figure 2.
Figure 2.

Expression levels of ERPGs depending on E2 treatment under MTN and EXC. HCA and heatmap results of ERPGs up-regulated (A) and down-regulated (C) by ≥1.5-fold due to E2 treatment: The gene expression level is expressed as log2(1+N), where “1+N” converts a negative value to a positive one; N indicates the N-value. Relative expression levels of ERPGs up-regulated (B) and down-regulated (D) by ≥2-fold due to E2 treatment: The totals of 188 and 168 genes were analyzed, respectively; each gene expression value was normalized to the mean of the corresponding values of all four samples; data are expressed as the mean±SD; and repeated measures ANOVA was performed (***p<0.001).

To examine the role of E2 under MTN and EXC, KEGG pathway analysis was conducted. As a result, 36 and 59 E2-responsive cancer-related pathways, which are involved in proliferation, growth, survival, metabolism, stemness, cell-to-cell interaction, inflammation, and the microenvironment, were identified as statistically significant under MTN and EXC, respectively (data not shown). Of them, 35 pathways overlapped between MTN (97.2%) and EXC (59.3%), leaving only one pathway (“apoptosis-multiple species”) exclusive to MTN (data not shown). In addition, EXC specifically induced 24 pathways (Table IV). Furthermore, more E2-driven perturbations in the cancer-related pathways were observed under EXC than under MTN (p=0.007 when using Student’s t-test) (Figure 3). These functional analysis results show that estrogen could be more favorable for the development of BC corresponding to the EXC model than to the MTN model.

View this table:
Table IV.

Statistically significant cancer-related KEGG pathways induced by E2 treatment exclusively under EXC.

Figure 3.
Figure 3.

Numbers of differentially expressed genes (DEGs) induced by E2 treatment in the top 10 significant cancer-related Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways under MTN and EXC. Among the 35 overlapping E2-responsive cancer-related pathways, the top 10 pathways with the lowest mean p-values for MTN and EXC were selected. The p-value between the MTN and EXC groups was 0.007 when paired Student’s t-test was simply used by considering each pathway value as each sample value.

Differentiation of ERPG expression from “typical” estrogen-responsive gene expression and correlation of ERPG expression with the expression of prognostic factors for HR+ BC. Several groups of biomarkers and associated algorithms for predicting prognosis, which is related to recurrence risk, chemotherapy efficacy, or ET duration, have been developed and used in clinical practice for early HR+ BC (35). We used EndoPredict (EP), a multigene test, to verify the clinical relevance of ERPGs (36). Interestingly, the expression of eight EP genes was similar to that of ERPGs up-regulated by E2, depending on E2 treatment and EXC, and the EP risk scores calculated by the EP algorithm were correlated with the expression levels of the ERPGs (Figure 4A and Figure 2B). The expression pattern of the “typical” E2-responsive genes appeared largely similar to that of the EP genes or ERPGs, but the expression level of only the “typical” genes was higher in MTN-E2 than in MTN-CTL (Figure 4A, B and Figure 2B). The expression pattern of genes up-regulated by E2 under EXC—including those up-regulated by E2 under MTN or up-regulated by EXC alone (i.e., those unrelated to proliferation)—was similar to that of the “typical” genes (Figure 4B and C). Therefore, genes that show a response to E2 under EXC could be regarded as “typical” estrogen-responsive genes. Notably, the mean expression levels of the EP genes (p=0.045 when using Student’s t-test), “typical” E2-responsive genes, and genes up-regulated by E2 under EXC as well as ERPGs up-regulated by E2 were all higher under MTN than under EXC without E2 treatment, as indicated above (Figure 4A-C and Figure 2B). However, this feature did not appear in genes up-regulated by E2 under MTN (Figure 4A-D and Figure 2B). In addition, these genes were not up-regulated by E2 under EXC, and many of them were up-regulated inversely by EXC alone (Figure 4D). Taken together, these results suggest that ERPGs are distinguished from typical estrogen-responsive genes and more closely correlated with prognostic or predictive factors for HR+ BC, and that many atypical estrogen-responsive genes respond to estrogen under usual culture conditions in T47D cells.

Figure 4.
Figure 4.

Relative gene expression levels of EndoPredict (EP)’s prognostic genes, well-known E2-responsive genes, and genes up-regulated by E2 under EXC and MTN depending on E2 treatment and EXC. (A) Expression levels of eight EP genes: Each EP risk score shown below the graph was calculated using a previously established formula (36); the p-value between MTN-CTL and EXC-CTL is 0.045 when paired Student’s t-test was simply used. (B) Expression levels of 65 “typical” E2-responsive genes. (C, D) Expression levels of genes up-regulated by ≥2-fold due to E2 treatment under EXC and MTN: The totals of 293 and 232 genes were analyzed, respectively. Each gene expression value was normalized to the mean of the corresponding values of all four samples. Data are expressed as the mean±SD, and repeated measures ANOVA was performed (**p<0.01; ***p<0.001).

Greater efficacy of phosphatidylinositol 3-kinase (PI3K)/AKT (or protein kinase B)/mammalian target of rapamycin (mTOR) inhibitors under EXC than under MTN. We previously found that the antiproliferative efficacy of Fulv was reduced by EXC (11). Therefore, an alternative drug to SERDs or SERMs may be needed for premenopausal women with HR+ BC corresponding to the EXC model. We also found that phosphorylated (or activated) AKT showed a positive correlation to some extent with the protein level of ERα and EXC (11). In clinical practice, alpelisib (PI3K inhibitor) and everolimus (mTOR inhibitor) are currently used to overcome ET-R (37). Therefore, the inhibitors of the PI3K-AKT-mTOR pathway were examined. Interestingly, in contrast to Fulv, alpelisib, everolimus, and MK-2206 (AKT inhibitor) were all more effective under EXC than under MTN (Figure 5A-C). However, these results were not observed with the combination of Fulv and one of the inhibitors, which had only an additive effect, but not a synergistic effect, except for the combination of Fulv and MK-2206 (Figure 5A-C). These results suggest not only that EXC alone makes HR+ BC cells more dependent on the PI3K-AKT-mTOR pathway for their proliferation, but also that PI3K/AKT/mTOR inhibitors are effective for treating HR+ BC corresponding to the EXC model.

Figure 5.
Figure 5.

Effects of an endocrine therapy drug and PI3K/mTOR/AKT inhibitors on cell proliferation under MTN and EXC. MTT assay results of alpelisib (A), everolimus (B), and MK-2206 (C) with or without Fulv. Fulv, alpelisib, everolimus, and MK-2206 were used as representatives of ER, PI3K, mTOR, and AKT inhibitors, respectively. Data from triplicate experiments are expressed as the mean ± SD and normalized to each CTL in MTN and EXC. Basic comparisons, indicated by asterisks above the graphs, were performed using Student’s t-test based on each CTL in MTN and EXC. Additional comparisons, indicated by asterisks above the guide lines, were performed using two-way ANOVA between MTN and EXC. *p<0.05; **p<0.01; ***p<0.001; Fulv: fulvestrant; Combi: combination; ER: estrogen receptor; PI3K: phosphatidylinositol 3-kinase; mTOR: mammalian target of rapamycin; and AKT: protein kinase B.

Discussion

This study, based on RNA-seq in T47D cells, was a follow-up to our most recent study, and the findings have three major implications. First, the present results showing the different effects of estrogen on transcription between MTN and EXC are consistent with previous findings on the functional effects of EXC. Second, we gained some understanding of the EXC model (or certain autocrine factors that influence cellular responsiveness to estrogen or ET) at the molecular level as follows: 1) Estrogen induces the expression of more genes, including typical estrogen-responsive genes, in the EXC model than in the MTN model; 2) likewise, estrogen has a more favorable effect on transcription in the EXC model in terms of cancer development; and 3) typical estrogen-responsive genes are constitutively activated in the MTN model (or under usual culture conditions). The last one suggests that certain autocrine factors inhibit estrogen response in HR+ BC cells by inducing the constitutive activation of the estrogen-ER pathway. Third, we confirmed the clinical significance of the EXC model as follows: 1) As our ERPGs include several prognostic or predictive genes that are currently used for treating HR+ BC patients, some of ERPGs may be further utilized clinically in the future; 2) AKT pathway inhibitors may be more aggressively used to treat HR+ BC.

Although our simple RNA-seq results were sufficient to demonstrate a difference in the role of estrogen in transcription between the MTN and EXC models, we introduced the novel concept of ERPG for the clinical application of these models. However, for convenience, we approached ERPGs themselves from a somewhat qualitative perspective. Therefore, further studies on ERPGs are necessary to consider quantitative, relative, variable, multifaceted, and functional aspects. We think that basal estrogen levels in normal FBS media may not be sufficient for EXC alone to increase ER activity, and that the AKT pathway activation and the increased dependence of HR+ BC cells on the AKT pathway may occur conditionally only under the EXC condition with a low level of estrogen. Therefore, a combination of an AKT pathway inhibitor and an AI would generally be much more effective than either monotherapy for HR+ BC patients. However, we were unable to verify this hypothesis because we did not have an AI model system (38).

Certain autocrine factors also appear to manipulate ERPGs independently of estrogen in the MTN model, while mimicking estrogen in the EXC model. This phenomenon is reminiscent of the constitutive or ligand-independent activation of ERα (10). Constitutive ERα activation is often observed in HR+ BC cells exhibiting ET-R and is directly induced by an alteration in the amino acid sequence or post-translational modification of ERα, such as phosphorylation of serine 118 and serine 167 (8, 10, 39). The growth factor receptor (GFR), mitogen-activated protein kinase (MAPK), and PI3K/AKT/mTOR signaling pathways can induce such modifications, resulting in ET-R (8, 10, 39). Further studies are needed to clarify the specific mechanism of action that a certain autocrine factor or the MTN condition causes the constitutive activation of the estrogen-ER pathway by.

In addition, we confirmed that EXC alone can partially mimic E2 treatment. However, it was correlated more closely with E2 treatment under MTN than under EXC. Furthermore, it altered the expression of a large number of specific genes, suggesting that causative autocrine factors play an independent role in transcription. Similarly, E2 played an unusual role in transcription under MTN (or usual culture conditions). Nevertheless, E2 also induced the expression of some typical E2-responsive genes under MTN. It is unclear whether estrogen plays a separate, unknown role in the development of HR+ BC cells under the MTN condition, or whether the unusual changes are simply a type of biological “noise.” All these issues remain challenges.

Conclusion

RNA-seq in T47D HR+ BC cells showed that the role of E2 in transcription differed between MTN and EXC conditions, and that E2 played more important roles in transcription in the context of cancer development (e.g., promoting cell proliferation) under EXC than under MTN. Our findings support not only the functional effects of EXC that we previously demonstrated but also the use of our research model in developing new therapeutic drugs or predictive biomarkers for HR+ BC.

Acknowledgements

This research was supported by a private grant funded by Dalim Corporation (Seoul, Republic of Korea) (No. 2-2017-1967-001-1).

Footnotes

  • Authors’ Contributions

    Conceptualization: Jang SH, Seong JK, & Lim W; Data curation: Jang SH & Paek SH; Formal analysis: Jang SH & Paek SH; Funding acquisition: Lim W; Methodology: Jang SH; Project administration: Lim W; Supervision: Lim W & Seong JK; Writing - original draft: Jang SH, Paek SH, & Kim JK; Writing - review & editing: Jang SH, Paek SH, & Kim JK.

  • Conflicts of Interest

    The Authors declare no conflicts of interest regarding this study.

  • Received July 28, 2023.
  • Revision received September 3, 2023.
  • Accepted September 4, 2023.

This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).

References

  1. ↵
    1. Sung H,
    2. Ferlay J,
    3. Siegel RL,
    4. Laversanne M,
    5. Soerjomataram I,
    6. Jemal A,
    7. Bray F
    : Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 71(3): 209-249, 2021. DOI: 10.3322/caac.21660
    OpenUrlCrossRefPubMed
  2. ↵
    1. Freelander A,
    2. Brown LJ,
    3. Parker A,
    4. Segara D,
    5. Portman N,
    6. Lau B,
    7. Lim E
    : Molecular biomarkers for contemporary therapies in hormone receptor-positive breast cancer. Genes (Basel) 12(2): 285, 2021. DOI: 10.3390/genes12020285
    OpenUrlCrossRefPubMed
  3. ↵
    1. Pan H,
    2. Gray R,
    3. Braybrooke J,
    4. Davies C,
    5. Taylor C,
    6. McGale P,
    7. Peto R,
    8. Pritchard KI,
    9. Bergh J,
    10. Dowsett M,
    11. Hayes DF, EBCTCG
    : 20-year risks of breast-cancer recurrence after stopping endocrine therapy at 5 years. N Engl J Med 377(19): 1836-1846, 2017. DOI: 10.1056/NEJMoa1701830
    OpenUrlCrossRefPubMed
  4. ↵
    1. Dai X,
    2. Li T,
    3. Bai Z,
    4. Yang Y,
    5. Liu X,
    6. Zhan J,
    7. Shi B
    : Breast cancer intrinsic subtype classification, clinical use and future trends. Am J Cancer Res 5(10): 2929-2943, 2015.
    OpenUrlCrossRefPubMed
  5. ↵
    1. Cheang MC,
    2. Chia SK,
    3. Voduc D,
    4. Gao D,
    5. Leung S,
    6. Snider J,
    7. Watson M,
    8. Davies S,
    9. Bernard PS,
    10. Parker JS,
    11. Perou CM,
    12. Ellis MJ,
    13. Nielsen TO
    : Ki67 index, HER2 status, and prognosis of patients with luminal B breast cancer. J Natl Cancer Inst 101(10): 736-750, 2009. DOI: 10.1093/jnci/djp082
    OpenUrlCrossRefPubMed
    1. Yaşar P,
    2. Ayaz G,
    3. User SD,
    4. Güpür G,
    5. Muyan M
    : Molecular mechanism of estrogen-estrogen receptor signaling. Reprod Med Biol 16(1): 4-20, 2016. DOI: 10.1002/rmb2.12006
    OpenUrlCrossRefPubMed
  7. ↵
    1. Porras L,
    2. Ismail H,
    3. Mader S
    : Positive regulation of estrogen receptor Alpha in breast tumorigenesis. Cells 10(11): 2966, 2021. DOI: 10.3390/cells10112966
    OpenUrlCrossRef
  8. ↵
    1. Augusto TV,
    2. Correia-da-Silva G,
    3. Rodrigues CMP,
    4. Teixeira N,
    5. Amaral C
    : Acquired resistance to aromatase inhibitors: where we stand! Endocr Relat Cancer 25(5): R283-R301, 2018. DOI: 10.1530/ERC-17-0425
    OpenUrlAbstract/FREE Full Text
    1. Collin LJ,
    2. Cronin-Fenton DP,
    3. Ahern TP,
    4. Goodman M,
    5. McCullough LE,
    6. Waller LA,
    7. Kjærsgaard A,
    8. Damkier P,
    9. Christiansen PM,
    10. Ejlertsen B,
    11. Jensen MB,
    12. Sørensen HT,
    13. Lash TL
    : Early discontinuation of endocrine therapy and recurrence of breast cancer among premenopausal women. Clin Cancer Res 27(5): 1421-1428, 2021. DOI: 10.1158/1078-0432.CCR-20-3974
    OpenUrlAbstract/FREE Full Text
  10. ↵
    1. Belachew EB,
    2. Sewasew DT
    : Molecular mechanisms of endocrine resistance in estrogen-positive breast cancer. Front Endocrinol (Lausanne) 12: 599586, 2021. DOI: 10.3389/fendo.2021.599586
    OpenUrlCrossRefPubMed
  11. ↵
    1. Jang SH,
    2. Paek SH,
    3. Kim JK,
    4. Seong JK,
    5. Lim W
    : A new culture model for enhancing estrogen responsiveness in HR+ breast cancer cells through medium replacement: Presumed involvement of autocrine factors in estrogen resistance. Int J Mol Sci 24(11): 9474, 2023. DOI: 10.3390/ijms24119474
    OpenUrlCrossRefPubMed
  12. ↵
    1. Masri S,
    2. Phung S,
    3. Wang X,
    4. Wu X,
    5. Yuan YC,
    6. Wagman L,
    7. Chen S
    : Genome-wide analysis of aromatase inhibitor-resistant, tamoxifen-resistant, and long-term estrogen-deprived cells reveals a role for estrogen receptor. Cancer Res 68(12): 4910-4918, 2008. DOI: 10.1158/0008-5472.CAN-08-0303
    OpenUrlAbstract/FREE Full Text
    1. Ochi Y,
    2. Shiomi K,
    3. Hachiya T,
    4. Yoshimura M,
    5. Miyazaki T
    : Dextrancoated charcoal technique to make the hormone-free serum as a diluent for standard curve of radioimmunoassay. Endocrinol Jpn 20(1): 1-7, 1973. DOI: 10.1507/endocrj1954.20.1
    OpenUrlCrossRefPubMed
    1. Vanetti C,
    2. Bifari F,
    3. Vicentini LM,
    4. Cattaneo MG
    : Fatty acids rather than hormones restore in vitro angiogenesis in human male and female endothelial cells cultured in charcoal-stripped serum. PLoS One 12(12): e0189528, 2017. DOI: 10.1371/journal.pone.0189528
    OpenUrlCrossRefPubMed
    1. Tu C,
    2. Fiandalo MV,
    3. Pop E,
    4. Stocking JJ,
    5. Azabdaftari G,
    6. Li J,
    7. Wei H,
    8. Ma D,
    9. Qu J,
    10. Mohler JL,
    11. Tang L,
    12. Wu Y
    : Proteomic analysis of charcoal-stripped fetal bovine serum reveals changes in the insulin-like growth factor signaling pathway. J Proteome Res 17(9): 2963-2977, 2018. DOI: 10.1021/acs.jproteome.8b00135
    OpenUrlCrossRefPubMed
    1. Sikora MJ,
    2. Johnson MD,
    3. Lee AV,
    4. Oesterreich S
    : Endocrine response phenotypes are altered by charcoal-stripped serum variability. Endocrinology 157(10): 3760-3766, 2016. DOI: 10.1210/en.2016-1297
    OpenUrlCrossRefPubMed
  17. ↵
    1. Cao Z,
    2. West C,
    3. Norton-Wenzel CS,
    4. Rej R,
    5. Davis FB,
    6. Davis PJ,
    7. Rej R
    : Effects of resin or charcoal treatment on fetal bovine serum and bovine calf serum. Endocr Res 34(4): 101-108, 2009. DOI: 10.3109/07435800903204082
    OpenUrlCrossRefPubMed
  18. ↵
    1. Bolger AM,
    2. Lohse M,
    3. Usadel B
    : Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics 30(15): 2114-2120, 2014. DOI: 10.1093/bioinformatics/btu170
    OpenUrlCrossRefPubMed
  19. ↵
    1. Kim D,
    2. Langmead B,
    3. Salzberg SL
    : HISAT: a fast spliced aligner with low memory requirements. Nat Methods 12(4): 357-360, 2015. DOI: 10.1038/nmeth.3317
    OpenUrlCrossRefPubMed
  20. ↵
    1. Pertea M,
    2. Kim D,
    3. Pertea GM,
    4. Leek JT,
    5. Salzberg SL
    : Transcript-level expression analysis of RNA-seq experiments with HISAT, StringTie and Ballgown. Nat Protoc 11(9): 1650-1667, 2016. DOI: 10.1038/nprot.2016.095
    OpenUrlCrossRefPubMed
  21. ↵
    1. Dillies MA,
    2. Rau A,
    3. Aubert J,
    4. Hennequet-Antier C,
    5. Jeanmougin M,
    6. Servant N,
    7. Keime C,
    8. Marot G,
    9. Castel D,
    10. Estelle J,
    11. Guernec G,
    12. Jagla B,
    13. Jouneau L,
    14. Laloe D,
    15. Le Gall C,
    16. Schaeffer B,
    17. Le Crom S,
    18. Guedj M,
    19. Jaffrezic F, French StatOmique Consortium
    : A comprehensive evaluation of normalization methods for Illumina high-throughput RNA sequencing data analysis. Brief Bioinform 14(6): 671-683, 2013. DOI: 10.1093/bib/bbs046
    OpenUrlCrossRefPubMed
  22. ↵
    1. Ogata H,
    2. Goto S,
    3. Fujibuchi W,
    4. Kanehisa M
    : Computation with the KEGG pathway database. Biosystems 47(1-2): 119-128, 1998. DOI: 10.1016/s03032647(98)00017-3
    OpenUrlCrossRefPubMed
  23. ↵
    1. Jagannathan V,
    2. Robinson-Rechavi M
    : Meta-analysis of estrogen response in MCF-7 distinguishes early target genes involved in signaling and cell proliferation from later target genes involved in cell cycle and DNA repair. BMC Syst Biol 5: 138, 2011. DOI: 10.1186/1752-0509-5-138
    OpenUrlCrossRefPubMed
  24. ↵
    1. Zhang Y,
    2. Chan HL,
    3. Garcia-Martinez L,
    4. Karl DL,
    5. Weich N,
    6. Slingerland JM,
    7. Verdun RE,
    8. Morey L
    : Estrogen induces dynamic ERα and RING1B recruitment to control gene and enhancer activities in luminal breast cancer. Sci Adv 6(23): eaaz7249, 2020. DOI: 10.1126/sciadv.aaz7249
    OpenUrlFREE Full Text
    1. Schultz JR,
    2. Petz LN,
    3. Nardulli AM
    : Estrogen receptor α and Sp1 regulate progesterone receptor gene expression. Mol Cell Endocrinol 201(1-2): 165-175, 2003. DOI: 10.1016/s0303-7207(02)00415-x
    OpenUrlCrossRefPubMed
    1. Arruabarrena-Aristorena A,
    2. Maag JLV,
    3. Kittane S,
    4. Cai Y,
    5. Karthaus WR,
    6. Ladewig E,
    7. Park J,
    8. Kannan S,
    9. Ferrando L,
    10. Cocco E,
    11. Ho SY,
    12. Tan DS,
    13. Sallaku M,
    14. Wu F,
    15. Acevedo B,
    16. Selenica P,
    17. Ross DS,
    18. Witkin M,
    19. Sawyers CL,
    20. Reis-Filho JS,
    21. Verma CS,
    22. Jauch R,
    23. Koche R,
    24. Baselga J,
    25. Razavi P,
    26. Toska E,
    27. Scaltriti M
    : FOXA1 mutations reveal distinct chromatin profiles and influence therapeutic response in breast cancer. Cancer Cell 38(4): 534-550.e9, 2020. DOI: 10.1016/j.ccell.2020.08.003
    OpenUrlCrossRefPubMed
  27. ↵
    1. Bourdeau V,
    2. Deschênes J,
    3. Laperrière D,
    4. Aid M,
    5. White JH,
    6. Mader S
    : Mechanisms of primary and secondary estrogen target gene regulation in breast cancer cells. Nucleic Acids Res 36(1): 76-93, 2008. DOI: 10.1093/nar/gkm945
    OpenUrlCrossRefPubMed
    1. Madak-Erdogan Z,
    2. Lupien M,
    3. Stossi F,
    4. Brown M,
    5. Katzenellenbogen BS
    : Genomic collaboration of estrogen receptor alpha and extracellular signal-regulated kinase 2 in regulating gene and proliferation programs. Mol Cell Biol 31(1): 226-236, 2011. DOI: 10.1128/MCB.00821-10
    OpenUrlAbstract/FREE Full Text
  29. ↵
    1. Wang C,
    2. Mayer JA,
    3. Mazumdar A,
    4. Fertuck K,
    5. Kim H,
    6. Brown M,
    7. Brown PH
    : Estrogen induces c-myc gene expression via an upstream enhancer activated by the estrogen receptor and the AP-1 transcription factor. Mol Endocrinol 25(9): 1527-1538, 2011. DOI: 10.1210/me.2011-1037
    OpenUrlCrossRefPubMed
  30. ↵
    1. Ghosh MG,
    2. Thompson DA,
    3. Weigel RJ
    : PDZK1 and GREB1 are estrogen-regulated genes expressed in hormone-responsive breast cancer. Cancer Res 60(22): 6367-6375, 2000.
    OpenUrlAbstract/FREE Full Text
  31. ↵
    1. Morandi A,
    2. Plaza-Menacho I,
    3. Isacke CM
    : RET in breast cancer: functional and therapeutic implications. Trends Mol Med 17(3): 149-157, 2011. DOI: 10.1016/j.molmed.2010.12.007
    OpenUrlCrossRefPubMed
  32. ↵
    1. Hinz N,
    2. Jücker M
    : Distinct functions of AKT isoforms in breast cancer: a comprehensive review. Cell Commun Signal 17(1): 154, 2019. DOI: 10.1186/s12964-019-0450-3
    OpenUrlCrossRefPubMed
    1. Lu C,
    2. Shen Q,
    3. DuPré E,
    4. Kim H,
    5. Hilsenbeck S,
    6. Brown PH
    : cFos is critical for MCF-7 breast cancer cell growth. Oncogene 24(43): 6516-6524, 2005. DOI: 10.1038/sj.onc.1208905
    OpenUrlCrossRefPubMed
  34. ↵
    1. Ning Q,
    2. Liu C,
    3. Hou L,
    4. Meng M,
    5. Zhang X,
    6. Luo M,
    7. Shao S,
    8. Zuo X,
    9. Zhao X
    : Vascular endothelial growth factor receptor-1 activation promotes migration and invasion of breast cancer cells through epithelial-mesenchymal transition. PLoS One 8(6): e65217, 2013. DOI: 10.1371/journal.pone.0065217
    OpenUrlCrossRefPubMed
  35. ↵
    1. Saini G,
    2. Mittal K,
    3. Rida P,
    4. Janssen EAM,
    5. Gogineni K,
    6. Aneja R
    : Panoptic view of prognostic models for personalized breast cancer management. Cancers (Basel) 11(9): 1325, 2019. DOI: 10.3390/cancers11091325
    OpenUrlCrossRefPubMed
  36. ↵
    1. Filipits M,
    2. Rudas M,
    3. Jakesz R,
    4. Dubsky P,
    5. Fitzal F,
    6. Singer CF,
    7. Dietze O,
    8. Greil R,
    9. Jelen A,
    10. Sevelda P,
    11. Freibauer C,
    12. Müller V,
    13. Jänicke F,
    14. Schmidt M,
    15. Kölbl H,
    16. Rody A,
    17. Kaufmann M,
    18. Schroth W,
    19. Brauch H,
    20. Schwab M,
    21. Fritz P,
    22. Weber KE,
    23. Feder IS,
    24. Hennig G,
    25. Kronenwett R,
    26. Gehrmann M,
    27. Gnant M, EP Investigators
    : A new molecular predictor of distant recurrence in ER-positive, HER2-negative breast cancer adds independent information to conventional clinical risk factors. Clin Cancer Res 17(18): 6012-6020, 2011. DOI: 10.1158/1078-0432.CCR-11-0926
    OpenUrlAbstract/FREE Full Text
  37. ↵
    1. McAndrew NP,
    2. Finn RS
    : Clinical review on the management of hormone receptor–positive metastatic breast cancer. JCO Oncol Pract 18(5): 319-327, 2022. DOI: 10.1200/OP.21.00384
    OpenUrlCrossRefPubMed
  38. ↵
    1. Itoh T,
    2. Karlsberg K,
    3. Kijima I,
    4. Yuan YC,
    5. Smith D,
    6. Ye J,
    7. Chen S
    : Letrozole-, anastrozole-, and tamoxifen-responsive genes in MCF-7aro cells: a microarray approach. Mol Cancer Res 3(4): 203-218, 2005. DOI: 10.1158/1541-7786.MCR-04-0122
    OpenUrlAbstract/FREE Full Text
  39. ↵
    1. Rani A,
    2. Stebbing J,
    3. Giamas G,
    4. Murphy J
    : Endocrine resistance in hormone receptor positive breast cancer-from mechanism to therapy. Front Endocrinol (Lausanne) 10: 245, 2019. DOI: 10.3389/fendo.2019.00245
    OpenUrlCrossRefPubMed
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools
Reprints and Permissions
Estrogen Effects Differ Between Medium Maintenance and Replacement from Transcriptional and Clinical Perspectives in T47D Breast Cancer Cells
SEOK-HOON JANG, SE HYUN PAEK, JONG-KYU KIM, JE KYUNG SEONG, WOOSUNG LIM
Anticancer Research Oct 2023, 43 (10) 4447-4469; DOI: 10.21873/anticanres.16640
Twitter logo Facebook logo Mendeley logo

Jump to section

Keywords