MAP3K1 expression is associated with progression and poor prognosis of hormone receptor-positive, HER2-negative early-stage breast cancer

In this study, we assessed whether the overexpression of MAP3K1 promotes the proliferation, migration, and invasion of breast cancer cells, which affect the prognosis of hormone receptor (HR)–positive, human epidermal growth factor receptor 2 (HER2)–negative early stage breast cancer. Two HR-positive, HER2-negative breast cancer cell lines (MCF7 and T-47D) overexpressing MAP3K1 were transfected with two MAP3K1 short hairpin RNA plasmids (shMAP3K1 [#3] and shMAP3K1 [#5]). The proliferation, migration, and invasion of these cells were then examined. We assessed whether shMAP3K1 affects the cell cycle, levels of downstream signaling molecules (ERK, JNK, p38 MAPK, and NF-κB), and sensitivity to chemotherapeutic and hormonal agents. To assess the anti-tumor effect of MAP3K1 knockdown in the breast cancer orthotopic model, MCF7 and T-47D cells treated with or without shMAP3K1 (#3) and shMAP3K1 (#5) were inoculated into the mammary fat pads of mice. In total, 182 patients with HR-positive, HER2-negative T1 and T2 breast cancer and 0–3 nodal metastases were included. Additionally, 73 patients with T1 and T2 breast cancer and negative nodes who received adjuvant endocrine therapy alone were selected as an independent validation cohort. In both cell lines, shMAP3K1 (#3) and shMAP3K1 (#5) significantly reduced cell growth, migration, and invasion by downregulating MMP-9 and by blocking the G2/M phase of the cell cycle and its regulatory molecule cyclin B1. Moreover, both shMAP3K1 (#3) and shMAP3K1 (#5) downregulated ERK-, JNK-, p38 MAPK-, and NF-κB-dependent gene transcription and enhanced the sensitivity of both cell lines to doxorubicin, docetaxel, and tamoxifen. We observed that both shMAP3K1 (#3) and shMAP3K1 (#5) inhibited tumor growth compared with that in the scrambled group of MCF7 and T-47D cell orthotopic tumors. Patients with MAP3K1 overexpression exhibited significantly poorer 10-year disease-free survival (DFS) (70.4% vs. 88.6%, p = 0.003) and overall survival (OS) (81.9% vs. 96.3%, p = 0.001) than those without MAP3K1 overexpression. Furthermore, phospho-ERK (p < 0.001) and phospho-JNK (p < 0.001) expressions were significantly associated with MAP3K1 expression, and both phospho-ERK and phospho-JNK expressions were significantly correlated with poor 10-year DFS and OS. These biological findings, including a significant association between DFS and OS, and the expressions of MAP3K1, phospho-ERK, and phospho-JNK were further validated in an independent cohort. Multivariate analysis identified MAP3K1 expression as an independent poor prognostic factor for DFS and OS. Our results indicate that the overexpression of MAP3K1 plays a major role in the poor prognosis of HR-positive, HER2-negative early stage breast cancer.


Introduction
Breast cancer is the most common cancer in women worldwide [1].In Taiwan, there has been a continuous rise in the incidence of breast cancer [2].Based on gene expression profiles, breast cancers are classified into different molecular subtypes [3,4].Using immunohistochemistry (IHC) to assess the expression patterns of estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER2) and using fluorescence in situ hybridization (FISH) technique to assess the amplification of HER2 encoding gene in HER2 IHC (score: 2+) breast cancer, its subtypes can be categorized as follows according to the 2017 St. Gallen Consensus: HR-positive and HER2negative; HR-positive and HER2-positive, HR-negative and HER2-positive, and triple-negative [5,6].Systemic adjuvant chemotherapy is routinely administered for providing the long-term benefits of decreasing recurrence and metastases and prolonging survival outcomes in patients with HER2-enriched or triple-negative subtypes of breast cancer [6][7][8][9].For both HR2-positive and HER2-negative patients, systemic adjuvant chemotherapy and endocrine treatment are routinely administered for those with high-risk factors, such as high-grade, larger tumor size, and positive axillary lymph nodes (LNs), whereas endocrine treatments are dispensed for those without high-risk factors [6][7][8][9].
In addition to clinicopathological features, several multiple gene assays, including MammaPrint®, Oncotype DX®, PAM-50 (Prosigna®), and EndoPredict® have been demonstrated to predict the survival outcomes of patients with HR-positive, HER2-negative, and LN-negative breast cancer to help physicians and patients in opting either adjuvant chemotherapy combined with endocrine therapy or endocrine therapy alone [10][11][12][13].Considering that certain lowrisk and high-risk patients still develop local recurrence and distant metastases despite receiving endocrine treatment or chemotherapy followed by endocrine therapy, respectively, the identification of novel genes that can act as alternative prognostic markers as well as targeted genes for HR-positive and HER2-negative patients is warranted.
Based on aforementioned evidences [19][20][21][22], we hypothesized that MAP3K1 can mediate the cell proliferation of HR-positive, HER2-negative breast cancer cells, and is possibly related to the resistance against the adjuvant tamoxifen and other chemotherapeutic agents, and thus, contributes to the early recurrence and metastasis of early-stage HR-positive, HER2-negative breast cancer.To prove this hypothesis, we investigated whether downregulation of MAP3K1 could inhibit cell proliferation, migration, and invasion, and affect ERK1/2, JNK, p38 MAPK, and NF-κB activity in the in vitro HR-positive, HER2-negative breast cancer cell lines.We further used MCF7 and T-47D breast cancer orthotopic models to evaluate whether knockdown of MAP3K1 can inhibit the tumor growth.To further elucidate the biological functions of MAP3K1, we assessed the relationship between MAP3K1 expression in tumor cells and clinical outcomes in patients with early-stage HR-positive, HER2-negative breast cancers.We also assessed whether the downstream effector molecule of MAP3K1, phospho (p)-ERK, and p-JNK correlated with the expression of MAP3K1 and the clinical outcome of the same group of tumors.The prognostic values of the aforementioned MAP3K1 and MAP3K1-related molecules, p-ERK and p-JNK, were further validated with an independent validation cohort.
The short hairpin RNA (shRNA)-expression vectors and MAP3K1 shRNA constructs were obtained from the National RNAi Core Facility (Taipei, Taiwan).For lentivirus packaging, psPAX2, pMD2.G, and shRNAexpression vectors were co-transfected into 293T cells.The pGIPZ vector, containing a TurboGFP cassette, was used as a scrambled control and a tool for determining viral titer.The supernatant was harvested at 48 h after transfection, and then the lentiviral particles in supernatant were concentrated with PEG-it™ Virus Precipitation Solution (System Biosciences, CA, USA).We determined the titers of lentivirus by infecting 293T cells with varying concentrations of pGIPZ lentivirus.TurboGFP expression was assessed by flow cytometry, and the lentivirus titer was approximately 1 × 10 7 infectious units per mL (IFU/mL).The breast cancer cell lines, MCF7 and T-47D, were infected with virus at a multiplicity of infection (MOI) of 5.At 24 h after infection, these infected cells were screened with 2 µg/mL puromycin (InvivoGen, ant-pr-1).

Clonogenic survival assay, viability, migration, and invasion assay
The detailed information of clonogenic survival assay, viability, migration, and invasion assay are listed in the Supplementary Materials and methods [23].

Cell cycle analysis, apoptosis analysis, and luciferase assay
The detailed information of cell cycle analysis, apoptosis analysis, and luciferase assay are listed in the Supplementary Materials and methods [24,25].
Pathological and clinical information about treatment (including type of surgery, receipt or non-receipt of adjuvant systemic therapy, and type and dose of adjuvant systemic therapy) and follow-up information (including recurrence and distant metastasis) were obtained from pathology reports and clinical records.Patients with high-risk factors, such as grade III cancers, large tumors, lymphovascular invasion (LVI), and lymph node (LN) positivity (N1), received standard adjuvant chemotherapy, such as CMF (cyclophosphamide, methotrexate, fluorouracil), CEF (cyclophosphamide, epirubicin, fluorouracil), CAF (cyclophosphamide, adriamycin [doxorubicin], and fluorouracil), AC/EC, or AC/EC followed by paclitaxel/docetaxel regimens as defined in our previous study [8].All enrolled patients received adjuvant endocrine therapy, with drugs such as tamoxifen [8].Adjuvant radiotherapy was administered to all patients after breast conservation surgery [8].After surgery and adjuvant therapy, the patients were regularly followed-up in our clinic.If patients were lost during follow-up, information on their disease status and survival was obtained from the patients' charts, hospital cancer registry records, and the National Death Registry.
The immunohistochemical analyses of MAP3K1 protein expression and detailed demographic information were obtained from the patients and their medical charts with their written informed consent.The pathologic review and immunohistochemical studies were approved by the National Taiwan University Hospital (NTUH) ethics committee (Institutional Review Board [IRB] Number: 201804056RINB).The patients' medical data were anonymized before access and analysis.

Immunohistochemistry analysis
The immunohistochemistry staining for MAP3K1 (MEKK1, #PA5-15085, Thermo Fisher Scientific Inc. MA, USA) [26], p-ERK [28], and p-JNK [30] were performed on paraffinembedded sections of surgical specimens using an indirect immunoperoxidase method, according to manufacturer's instructions.To confirm the specificity of MAP3K1, staining was performed on paraffin-embedded sections in the absence of the first, the second, or both the primary antibodies as negative controls.All sections were observed under a light microscope.The percentages of MAP3K1-positive cells (tumor cells with readily visible brown staining distinctly marking the tumor cell nucleus) were averaged to yield an immunohistological score ranging from 0 to 100%.The results were classified into two groups according to the intensity and extent of staining: in the MAP3K1-negative were incubated with Alexa Fluor 488-conjugated secondary antibody (#A11034, Life Technologies, Carlsbad, CA) for 1 h at room temperature.Finally, the cell nuclei were stained with DAPI (#40,043, Biotium, Hayward, CA, USA) and mounted with fluorescent mounting medium (FMH030; ScyTek Laboratories, Logan, UT, USA).The stained cells were analyzed using a fluorescence microscope (IX71; Olympus, Tokyo, Japan).

The animal experiment of orthotopic breast cancer model
Athymic female BALB/c nude mice (five to six weeks old) were purchased from the National Laboratory Animal Center, Taiwan.All experiment procedures on animal were revised and approved by the College of Medicine, College of Public Health, and Institutional Animal Care and Use Committee (No. 20170503) of National Taiwan University.The 1 × 10 7 MCF7 cells stably expressing scramble shRNA, shMAP3K1 (#3), and shMAP3K1 (#5) and the 1 × 10 7 T-47D cells stably expressing scramble shRNA, shMAP3K1 (#3), and shMAP3K1 (#5) were mixed with Matrigel® (354,248; Corning, NY, USA) and then orthotopically injected into the right mammary fat pad 7 days after the 17β-estradiol pellets (NE-131; Innovative Research of America, Sarasota, FL, USA) were implanted subcutaneously on the left side of the neck.Body weight and tumor volume were measured once per week.We generated three groups of five mice for each group in the MCF7 cell orthotopic model and generated three groups of five mice for each group in the T-47D cell orthotopic model.Tumor volumes were calculated using the Eq.0.5 × L × W 2 , where L is the length, and W is the width.At the end of the experiment, the mice were sacrificed, and individual tumor weights were determined.

Characteristics, treatments, and tissue samples of HR-positive, HER2-negative early breast cancer patients
Patients diagnosed with stage I or II (AJCC 2007) HR-positive, HER2-negative early breast cancers at the National Taiwan University Hospital between January 1, 1994, and June 30, 2006, were enrolled for the study.Patients were considered HR-positive if the percentage of ER-or PRpositive epithelial cells was ≥ 1% [6,17].HER2 expression was measured using the universal iView-Dab detection kit.Scores of 0 and 1 + were considered negative, and a score of 3 + was considered positive.Gene amplification through FISH-based PathVysion assay (Vysis Inc., Des Plaines, IL, USA) was performed for tumors with a score of 2+.For the HER2 gene: chromosome 17 ratio of ≥ 2.0, tumors were considered positive based on the American Society of log-rank test.In this study, all prognostic variables that were investigated in the univariate analysis, including tumor stage, tumor grade, axillary LN status, LVI, and MAP3K1, were also included in the multivariate analysis by using the Cox proportional hazards regression model.The clinical characteristics were compared using the chi-square test and Fisher's exact tests.The p < 0.05 was considered statistically significant.

Downregulation of MAP3K1 attenuates cellular proliferation, migration, and invasion of HRpositive, HER2-negative breast cancer cell lines
As shown in Supplementary Figs.S1 and S2, we used the target mRNA sequences and corresponding amino acid sequences of five shMAP3K1 variants (shMAP3K1 , and shMAP3K1 [#5]) in this study to assess whether the efficiency of shMAP3K1 mRNA inhibition in both MCF7 and T-47D cells was prominent.Thereafter, we assessed the mRNA levels of MAP3K1 in control MCF7 and T-47D cells and in cells transfected with five shMAP3K1 variants by using RT-qPCR.We found that the mRNA expression levels of MAP3K1 were significantly downregulated in shMAP3K1 (#3)-and shMAP3K1 (#5)-transfected MCF7 cells compared with MCF7 cells transfected with shMAP3K1 (#1), shMAP3K1 (#2), and shMAP3K1 (#4) (Supplementary Fig. S3).Similarly, the mRNA levels of shMAPK3-transfected T-47D cells and the mRNA expression of MAP3K1 were significantly inhibited in shMAP3K1 (#3)-and shMAP3K1 (#5)-transfected T-47D cells (Supplementary Fig. S3).To approve the specificity of MAP3K1, we used immunofluorescence, immunohistochemical analysis, and western blotting to detect MAP3K1 expression in control MCF7 and control T-47D cells (primary antibody alone, secondary antibody alone, and combination of primary and secondary antibodies), and in shMAP3K1 (#3)-transfected MCF7 cells and in shMAP3K1 (#3)-transfected T-47D cells (combination of primary and secondary antibodies).The results are illustrated in Supplementary Fig. S3.
Breast cancer cell lines expressing higher levels of MAP3K1 were transfected with shMAP3K1 to downregulate MAP3K1 protein expression (Fig. 1A).shMAP3K1 (#3) treatment significantly reduced the expression level of MAP3K1 by 90% in MCF7 cells (p < 0.001) and by 85% in T-47D cells (p < 0.001) compared with the scrambled group (Fig. 1A).In addition, shMAP3K1 (#5) treatment significantly reduced the expression level of MAP3K1 by 40% in group, either no staining was present (staining intensity score: 0) or mild immunostaining or positive staining was detected in < 20% of the cells (staining intensity score: 1), and in the MAP3K1-positive group, moderate or strong immunostaining was present in 20-40% (staining intensity score: 2) or more than 40% of the cells (staining intensity score: 3).For the p-ERK marker and p-JNK, positive expression was defined as positive nuclear staining of p-ERK or p-JNK in ≥ 20% of tumor cells [30,31].

Patients, treatment, and tissue samples for the independent validation cohort
Patients diagnosed with T1-T2 disease with negative axillary lymph nodes (AJCC 2007) and HR-positive, HER2negative early-stage breast cancer who received adjuvant hormone therapy but no adjuvant chemotherapy at our institute between January 1, 2008, and June 30, 2012, who had events including locoregional recurrence, distant metastases, and death, and matched the clinicopathological features of patients without events, were selected as the independent validation cohort.This independent validation cohort comprised 73 patients, and the adjuvant endocrine therapy consisted of mostly tamoxifen and few patients receiving gonadotropin-releasing hormone agonists plus tamoxifen, or aromatase inhibitor (AI) for premenopausal women and tamoxifen or an AI for postmenopausal women.We further assessed the expression patterns of MAP3K1, p-ERK, and p-JNK in tumor specimens from this independent validation cohort to emphasize the biological significance of MAP3K1 in early-stage HR-positive and HER2-negative breast cancer.

Statistical analysis
In vitro experiments of the proliferation, migration, invasion, cell cycle, apoptosis assay, and luciferase assay, were repeated at least three times; the data of aforementioned assays were presented as the mean ± standard deviation (SD).The p values of the aforementioned experiments were determined using the student's t test, and statistical significance was defined for a p value < 0.05.The association between MAP3K1, p-ERK, and p-JNK was analyzed by Spearman's correlation.

Downregulation of MAP3K1 induces G2/M phase arrest and apoptosis and enhances drug sensitivity in HR-positive, HER2-negative breast cancer cell lines
Hu et al. reported that transfection with MAP3K1 small interfering RNA leads to the downregulation of expression of CDC25C and cyclin B1 (key molecules for G2/M transition during the cell cycle) in MCF7 cells and MCF-12 F cells (normal mammary epithelial cell line) [33].In this study, we sought to assess whether shMAP3K1 could inhibit the cell number in both breast cancer cell lines by blocking programmed G2/M phase and downregulating cyclin B1.
After 48 h, we determined the distribution of the cell cycle phases in each cell line.As shown in Fig. 2A, in shMAP3K1 (#3)-and shMAP3K1 (#5)-transfected MCF7 cells, there was a significant increase in the number of cells in the G2/M phase of the cell cycle and, concomitantly, a significant decrease in the number of cells in the G0/G1 phase.By contrast, in T-47D cells, shMAP3K1 (#3) and shMAP3K1 (#5) treatments led to the arrest of a significant number of cells in the G2/M phase of the cell cycle.These findings indicate that shMAP3K1 blocked cell cycle progression in the G2/M (#3) transfection downregulated Bcl-2 expression in MCF7 cells, and Bcl-xL expression in both MCF7 and T-47D cells, whereas c-Myc expression remained unaltered (Fig. 3B).Transfection with shMAP3K1 (#3) also led to increased expression of PARP and decreased MMP-9 expression (Fig. 3B).These results indicate that using shMAP3K1 to inhibit MAP3K1 can promote apoptosis and attenuate migration and invasion in these breast cancer cells.
NF-κB, which acts as a downstream factor in the MAP3K1 signaling pathway, is involved in the pathogenesis of HR-positive breast cancer cells [40].As shown in Fig. 3C, transfection with shMAP3K1(#3) inhibited the expression of p-IκBα, which is an essential regulator of NF-κB, and p-NF-κB (p65) in both breast cancer cell lines.Transfection with shMAP3K1 (#3) downregulated the nuclear expression Previous studies have revealed that the deletion of MAP3K1 results in apoptosis when mouse embryonic stem cells are subjected to hyperosmolarity and microtubule disruption or cardiomyocytes are subjected to oxidative stress [34,35].Zang et al. reported that MAP3K1 silencing inhibits cell proliferation and increases apoptosis of esophageal squamous cell carcinoma cells [36].However, the role of MAP3K1 in regulating anti-apoptotic function in breast cancer cells remains unclear.The members of Bcl-2-related anti-apoptotic protein family, including Bcl-2 and Bcl-xL, play critical roles in the pathogenesis of ERpositive breast cancer cells [37][38][39].Therefore, we assessed whether shMAP3K1 transfection led to the downregulation of expression of anti-apoptotic proteins, Bcl-2 and Bcl-xL, in both MCF7 and T-47D cells.We found that shMAP3K1 breast cancer cell lines (Supplementary Fig. S4).Furthermore, fluorescence imaging showed that shMAP3K1 (#3) and shMAP3K1 (#5) transfection significantly decreased the expression of p-NF-κB in both MCF7 and T-47D breast cancer cells compared with breast cancer cells transfected of NF-κB (p65) in both breast cancer cell lines.Furthermore, the NF-κB-dependent expression of genes such as nuclear p52 and BCL3 was downregulated in MCF7 and T-47D cells after transfection with shMAP3K1 (#3) (Fig. 3C).We found that shMAP3K1 (#5) downregulated p-NF-κB (p65) in both We measured the tumor volume from the appearance of the initial tumor burden to the sacrifice phase each week.We found that there were no significant differences in body weight between the scrambled MCF7 group and the two shMAP3K1-transfected MCF7 groups (Fig. 4A).Similarly, we observed that there were no significant differences in body weight between the scrambled T-47D group and the two shMAP3K1-transfected T-47D groups.
Taken together, our findings indicated that MAP3K1 might play a role in promoting cell proliferation, antiapoptotic function, migration, invasion, and NF-κB transcriptional activity, and thus, contributes to malignant progression and poor prognosis of HR-positive, HER2-negative breast cancer patients.

Inhibition of MAP3K1 expression reduces breast tumor growth in an orthotopic model
To assess the anti-tumor effect of MAP3K1 knockdown in the breast orthotopic model, MCF7 and T-47D cells treated LN-negative, whereas 44 patients (24.2%) had 1 to 3 LN metastases.Eight-two patients (45.1%) did not receive any chemotherapy, and 100 patients (54.9%) received standard adjuvant chemotherapy (Table 1).Furthermore, 138 patients (75.8%) were positive for both ER and PR.
We detected MAP3K1 expression (56 cases, score 2; 17 cases, score 3) in tumor cells of 73 patients (40.1%), whereas 109 patients exhibited negative MAP3K1 expression (101 cases, score 0; eight cases, score 1) (Fig. 5).Table 1 displays the demographic characteristics of the two groups of patients (MAP3K1-positive vs. MAP3K1-negative) and their clinicopathological features.Tumor size, histological grade, axillary LN, LVI, ER and PR status, and adjuvant chemotherapy were not significantly different between the two groups, except older age was closely associated with the expression of MAP3K1 (p = 0.038).

Correlation between the expression of MAP3K1, p-ERK, and p-JNK and clinical outcomes based on an independent validation cohort
In an independent validation cohort, 16 patients had events, including locoregional recurrence, distant metastases, and death from breast cancer or other etiologies.

Discussion
In the present study, we demonstrated that MAP3K1 plays an important role in the pathogenesis of HR-positive and HER2-negative breast cancer.We identified a close link between MAP3K1 expression and the expression of MAP3K1-regulated molecules, p-ERK, and p-JNK in tumor samples from the experimental cohort (T1-T2 and N0-1, adjuvant endocrine therapy with and without chemotherapy) and an independent validated cohort (T1-T2 and N0, adjuvant endocrine therapy alone).Importantly, the overexpression of each of MAP3K1, p-ERK, or p-JNK significantly correlated with poor DFS and poor OS for both the experimental cohort and the independent validation cohort of patients with early-stage HR-positive and HER2-negative breast cancer.
Growing evidence suggests that MAP3K1 participates in cell proliferation, invasion, and migration of human pancreatic cancer cell lines and the cell migration of ovarian cancer cell lines [41][42][43].Cuevas et al. reported that in the polyoma middle T antigen-driving mammary gland tumor, knockdown of MAP3K1 delays the dissemination and metastases of tumor cells [44].Rangaswami et al. also found that the activation of MAP3K1-dependent MMP-9 signaling contributes to the osteopontin-triggered tumor growth and pulmonary metastases of melanoma [45].These results are supported by our current findings that shMAP3K1 downregulated MMP-9 expression and attenuated the cellular associated with p-JNK expression (Spearman's correlation coefficient, R = 0.528; p < 0.001).
Our findings showing that shMAP3K1 caused the G2/M phase arrest in cells via downregulation of cyclin B1 further supported the results of Hu et al. that transfection with MAP3K1 small interfering RNA significantly enhances the paclitaxel-mediated cell proliferation inhibition via G2/M phase arrest and downregulation of cyclin B1 expression [33].The G2/M phase arrest might be a potential underlying mechanism of the enhanced cytotoxicity of docetaxel and doxorubicin in MCF7 and T-47D cells transfected with shMAP3K1 [46,47].In addition to the cell cycle arrest, we demonstrated that shMAP3K1 transfection led to induction of apoptosis via downregulation of Bcl-2 in MCF7 cells, Overall survival (OS) for all patients associated with the expression of MAP3K1 (C).DFS for all patients associated with the expression of p-ERK (D).OS for all patients associated with the expression of p-ERK (E).DFS for all patients associated with the expression of p-JNK (F).OS for all patients associated with the expression of p-JNK (G).MAP3K1 promotes HR-positive, HER2-negative breast cancer cell proliferation, migration, invasion, and resistance to the drugs (docetaxel, doxorubicin, and tamoxifen) through activating ERK, JNK, and NF-κB signaling; and inhibition of MAP3K1 attenuates tumor growth, migration and invasion, and enhances drugs sensitivities of this subtype of breast cancer through downregulation of ERK, JNK, and NF-κB signaling, and thus contributes to the better survival of this subtype of breast cancer Fig. 7 Association between MAP3K1 expression, p-ERK expression, or p-JNK in tumor cells and survival based on an independent validation cohort.(A).Disease-free survival (DFS) for all patients associated with the expression of MAP3K1 (B).Overall survival (OS) for all patients associated with the expression of MAP3K1 (C).DFS for all patients associated with the expression of p-ERK (D).OS for all patients associated with the expression of p-ERK (E).DFS for all patients associated with the expression of p-JNK (F).OS for all patients associated with the expression of p-JNK.
might promote cell proliferation, enhance drug resistance, and contribute to the poor prognosis of this subtype of HRpositive, HER2-negative breast cancer.
In a study aimed at identifying somatic copy number changes and mutations in the exons of protein-coding genes in 100 breast cancer tumors, Stephens et al. identified several new cancer-related genes, including AKT2, ARID1B, CASP8, CDKN1B, MAP3K1, MAP3K13, NCOR1, SMARCD1, and TBX, which may be involved in the tumorigenesis of breast cancer [66].Among these driver mutations, somatic mutations of MAP3K1 are observed in 6% of breast cancers, predominantly in ER-positive breast cancer [66].Furthermore, during the analyses of molecular heterogeneity of primary breast cancers through comprehensive molecular portraits, the authors showed that mutations of MAP3K1 are enriched in the HR-positive and HER2-negative breast cancer [67].Further assessment of the association between MAP3K1 mutation and overexpression of MAP3K1 in patients with HR-positive and HER2-negative breast cancer is merited.
In addition to the expression of MAP3K1 in MCF7 cells, Liu et al. reported the mRNA and protein expression of MEKK1 (MAP3K1) in two triple-negative breast cancer cell lines, in human MDA-MB-231 breast cancer cells, and another murine 4T1 breast cancer cells [68].Furthermore, Liu et al. showed that downregulation of MAP3K1 by MAP3K1targeting therapeutic artificial microRNA (amiRNA), attenuated the proliferation and inhibited the migration and invasion of murine 4T1 breast cancer cells [68].Further study to explore the biological function of MAP3K1 expression in human triple-negative breast cancer cell lines and patients with triple-negative breast cancer are warranted.
Although the cut-off points for the positive expression of MAP3K1 have not been described in previous reports, previous reports defined the cut-off points for the positive expression of p-ERK and p-JNK, which are the downstream signaling molecules of MAP3K1, as the positive nuclear staining of p-ERK or p-JNK in ≥ 20% of tumor cells [30,31].Therefore, in the current study, the positive expression of MAP3K1 was defined as a moderate or strong immunostaining of MAP3K1 in ≥ 20% of nuclei of tumor cells.We found that in the experimental cohort, a positive MAP3K1 expression was detected in the tumor cells of 73 of 182 patients (40.1%), whereas MAP3K1 expression was found in 32 of 73 patients (43.8%) in an independent validation cohort.However, in the cohort of breast cancer patients obtained from the Human Protein Atlas, most immunostainings of MAP3K1 proteins were located in the cytoplasmic and membranous cells of tumor cells.These findings may explain why the frequency (41.2%) of positive MAP3K1 expression in our cohort was lower than nearly 100% of adjuvant chemotherapy [52].In this study, we found that shMAP3K1 downregulated p-ERK and sensitized MCF7 and T-47D cells to doxorubicin.In tumor samples, our results further showed that p-ERK expression significantly correlated with MAP3K1 expression and with the poor DFS and OS of patients.These findings suggest that MAP3K1/ ERK-signaling might be involved in the pathogenesis of HR-positive, HER2-negative breast cancer.
In addition to ERK signaling pathway, MAP3K1 also activates the JNK signaling and promotes cell proliferations and cell migrations [53].In the MCF7 cells-xenograft models, the expression level of p-JNK and p-Jun were increased in tamoxifen-resistant tumors when compared with tamoxifen-sensitive tumors [54].Gutierrez et al. revealed that expression of p-ERK and p-p38 MAPK (the upstream molecule of JNK/Jun signaling) was upregulated in relapsed tumor samples of patients who had ER-positive, HER2-negative breast cancer and received adjuvant tamoxifen when compared with paired pre-treatment specimens of the same patients [55].These findings are supported by our current results showing that inhibition of MAP3K1 downregulated the expression of p-JNK and p-p38 MAPK in both breast cancer cell lines, and the expression of p-JNK was significantly associated with MAP3K1 expression and the poor DFS and OS of patients with HR-positive, HER2-negative early-stage breast cancer.
Previous studies have suggested that MAP3K1 may phosphorylate and activate IκBα and IκBβ kinase complexes and thus activate NF-κB in response to extracellular cytokines and stress [56,57].Other studies have revealed that NF-κB is involved in the epithelial-mesenchymal transition and metastasis of breast cancer cells [58,59].Following the assessment of the role of NF-κB expression in breast tumor samples, one study showed that the frequency of p65 expression is higher in tumor cells of HER2 and basal-like subtypes compared to its expression in luminal A subtype cancer cells [60].Another study, however, reported an association between ER and NF-κB expression, where the NF-κB expression correlates with higher tumor grade, stage III-IV, and lymph node metastasis [61].Furthermore, Oida et al. reported that NF-κB participated in the tamoxifen resistance in breast cancer cell line [62].Previous studies have suggested that BCL3 not only activates NF-κB signaling via interaction with nuclear NF-κB p50 but also regulates the transcription of NF-κB-dependent genes [63][64][65].Our findings demonstrated that inhibition of MAP3K1 downregulated NF-κB signaling pathway and BCL3 expression, and thus, suppressed cell growth of both HR-positive, HER2-negative breast cancer cell lines.Besides, inactivated NF-κB signaling may increase the sensitivity to tamoxifen in MCF7 and T-47D cells via downregulation of MAP3K1.

Declarations
Ethical approval and Consent to participate (Human and animal ethics statements) The immunohistochemical analyses of protein expression of MAP3K1, p-ERK, and p-JNK and detailed demographic information were obtained from the patients and their medical charts with their written informed consent.The pathologic review and immunohistochemical studies were approved by the National Taiwan University positive MAP3K1 expression in cases from the Human Protein Atlas.
In addition to our findings, Wang et al. revealed that glioma patients with a high expression of MAP3K1 (mostly located in the nucleus) in tumor cells had worse OS than those with low MAP3K1 expression [69].In their study, the high expression of MAP3K1 was defined as an immunohistochemical score > 5 on the basis of the percentage of positive cells for MAP3K1 (0, negative; 1, < 10% positive; 2, 11-50% positive; 3, 51-80% positive; and 4, > 80% positive) and the intensity staining for MAP3K1 (0, negative; 1, weakly positive; 2, moderately positive; and 3, strongly positive) [69].Wu et al. also showed that a high expression of long noncoding RNAs (transcribed RNA molecules with lengths > 200 nucleotides and without open reading frames of significant length [< 100 amino acids]) of MAP3K1-2) was significantly associated with the poor survival of patients with gastric cancer [70].

Conclusion
In summary, we demonstrated that MAP3K1 might act as an essential factor for promoting HR-positive, HER2negative breast cancer cell proliferation, migration, invasion, and resistance to the drugs, and thus, for increasing local recurrences and metastases of this subtype of breast cancer (Fig. 6G).Further studies on the possible relationship between MAP3K1 signaling and anti-apoptosis related proteins, such as Bcl-2 and Bcl-xL, are needed.Nevertheless, our results further elucidated the expression profiles of MAP3K1 in both experimental cancer cell lines, orthotopic animal model, and clinical specimens, and thus, it may facilitate the development of novel MAP3K1-related therapeutic strategies for this subtype of HR-positive, HER2-negative breast cancer.

Fig. 1
Fig. 1 Silencing of the MAP3K1 gene inhibits cellular proliferation and migration in HR-positive, HER2-negative breast cancer cell lines.(A) Two breast cancer cell lines, MCF7 and T-47D, were infected with pGIPZ lentiviral (scrambled) and five variant MAP3K1 shRNA lentiviral vectors.The protein expression of MAP3K1 was examined by western blotting analysis.The data showed an efficient knockdown of MAP3K1 in breast cancer cell lines transfected with shMAP3K1 (#3) and shMAP3K1 (#5), and therefore, we used these two shMAP3K1 (#3) and shMAP3K1 (#5) to conduct further experiments (B).Clonogenic survival assay showing that both the shMAP3K1 (#3)-and shMAP3K1 (#5)-transfected cell lines exhibited a significant reduction in the rate of cell survival compared with the scrambled group of the same cell lines.(C).Cell migration assay showed that knockdown of MAP3K1 by shMAP3K1 (#3) and shMAP3K1 (#5) inhibited both breast can-

Fig. 2
Fig. 2 Inhibition of MAP3K1 causes G2/M phase arrest and increases drug sensitivity in HR-positive, HER2-negative breast cancer cell lines.(A).Diagram of flow cytometric analyses of cell cycle distributions in MCF7 and T-47D cells transfected with scrambled RNA, shMAP3K1 (#3), and shMAP3K1 (#5) (left panel).Inhibition of MAP3K1 by shMAP3K1 (#3) transfection resulted in G2/M phase arrest in both MCF-7 and T-47D cells (right panel).Similarly, shMAP3K1 (#5) transfection resulted in G2/M phase arrest in both MCF-7 and T-47D cells (right panel).The results are expressed as the mean ± SD of three independent experiments from each cell line (B).Flow cytometry analysis of annexin and propidium iodine of apoptotic cells following control, scrambled RNA, shMAP3K1 (#3), and shMAP3K1 (#5) (left panel).Summary of the annexin V-FITC apoptosis assay results showing the percentages of early apoptotic, late apop-

Fig. 3
Fig. 3 Inhibition of MAP3K1 downregulates cell cycle-, anti-apoptosis-, and MAP3K1-related proteins and decreases the NF-κBdependent gene transcription.(A).shMAP3K1 (3#) transfection downregulated the expression of cyclin B1, p-ERK, p-JNK, and p-p38 MAPK; however, it did not affect cyclin D1 in total cell lysates of both MCF-7 and T-47D cancer cells (B).shMAP3K1 (#3) transfection downregulated the expression of Bcl-2, Bcl-xL, and MMP-9, and increased the levels of c-PARP; however, it did not affect c-Myc expression in total cell lysates of MCF7 cells.shMAP3K1 (#3) transfection decreased the expression of Bcl-xL and MMP-9, and increased the c-PARP levels; however, it did not affect c-Myc expression in total cell lysates of T-47D cells (C).shMAP3K1 (#3) decreased the expression of p65, p52, and BCL3 in nuclear lysates, and p-IκBα,

Fig. 5
Fig. 5 Expression of MAP3K1, p-ERK, and p-JNK in tumor cells of early-stage HR-positive, HER2-negative breast cancer patients.(A).Negative expression of MAP3K1 in tumor specimens of breast IDC (B).Representative images of moderate expression of MAP3K1 in tumor specimens of breast IDC (red arrow, MAP3K1 expression in the nucleus of tumor cells) (right upper inset 1000×) (C).Representative images of high expression of MAP3K1 in tumor specimens of breast IDC (red arrow, MAP3K1 expression in the nucleus of tumor cells) (right upper inset, 1000×) (D).Negative expression of p-ERK in tumor specimens of breast IDC (E).Representative images of moderate expression of p-ERK in tumor specimens of breast IDC (red

Fig. 6
Fig. 6 Association between MAP3K1 expression, p-ERK expression, or p-JNK in tumor cells and survival of early-stage HR-positive, HER2-negative breast cancer patients.(A).Disease-free survival (DFS) for all patients associated with the expression of MAP3K1 (B).Overall survival (OS) for all patients associated with the expression of MAP3K1 (C).DFS for all patients associated with the expression of p-ERK (D).OS for all patients associated with the expression of p-ERK (E).DFS for all patients associated with the expression of p-JNK (F).OS for all patients associated with the expression of p-JNK .2% [95% CI: 53.5 − 82.9%] vs. 88.8%[95% CI:

Table 2 Univariate and multivariate analyses of the relationship between pathological features and expression of MAP3K1 and clinical outcomes of HR-positive and HER2-negative breast cancer patients
These findings indicated that MAP3K1/NF-κB signaling