Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

High GPER expression in triple-negative breast cancer is linked to pro-metastatic pathways and predicts poor patient outcomes

High GPER expression in triple-negative breast cancer is linked to pro-metastatic pathways and... www.nature.com/npjbcancer ARTICLE OPEN High GPER expression in triple-negative breast cancer is linked to pro-metastatic pathways and predicts poor patient outcomes 1,8 2,8 2 3 4 5 6 1 7 Ting Xu , Ding Ma , Sheng Chen , Rui Tang , Jianling Yang , Chunhui Meng , Yang Feng , Li Liu , Jiangfen Wang , 6 2 ✉ ✉ Haojun Luo and Keda Yu Triple-negative breast cancer (TNBC) is a particularly aggressive and heterogeneous disease with few effective targeted therapies and precision therapeutic options over a long period. It is generally considered that TNBC is an estrogen-independent breast cancer, while a new estrogen receptor, namely G protein-coupled estrogen receptor (GPER), is demonstrated to mediate estrogenic actions in TNBC. Based on our transcriptomic analysis, expression of GPER was correlated with clinicopathological variables and survival of 360 TNBC patients. GPER expression at mRNA level was significantly correlated with immunohistochemistry scoring in 12 randomly chosen samples. According to the cutoff value, 26.4% (95/360) of patients showed high GPER expression and significant correlation with the mRNA subtype of TNBC (P = 0.001), total metastatic events (P = 0.019) and liver metastasis (P = 0.011). In quantitative comparison, GPER abundance is correlated with the high-risk subtype of TNBC. At a median follow-up interval of 67.1 months, a significant trend towards reduced distant metastasis-free survival (DMFS) (P = 0.014) was found by Kaplan–Meier analysis in patients with high GPER expression. Furthermore, univariate analysis confirmed that GPER was a significant prognostic factor for DMFS in TNBC patients. Besides, high GPER expression was significantly linked to the worse survival in patients with lymph node metastasis, TNM stage III as well as nuclear grade G3 tumors. Transcriptome-based bioinformatics analysis revealed that GPER was linked to pro-metastatic pathways in our cohort. These results may supply new insights into GPER-mediated estrogen carcinogenesis in TNBC, thus providing a potential strategy for endocrine therapy of TNBC. npj Breast Cancer (2022) 8:100 ; https://doi.org/10.1038/s41523-022-00472-4 INTRODUCTION either by suppressing ER-α activity or by inhibiting estrogen production, is central to the multidisciplinary management of Triple-negative breast cancer (TNBC) accounts for 10–15% of all patients with breast cancer, based not only on the significant breast cancers and is characterized by the lack of expression of the effectiveness but also on the convenience and safety of these estrogen receptor α (ER-α), the progesterone receptor (PR), and 6,7 1,2 agents . However, estrogen carcinogenesis, as well as endocrine the human epidermal growth factor receptor 2 (HER2) . When therapy, were long neglected in TNBC, logically due to the compared with other subtypes of breast cancer, TNBC exhibits the absence of ER-α. It has been recently shown that alternative ERs most aggressive course and the highest rate of early distant including G protein-coupled estrogen receptor (GPER), ER-β, and recurrence, especially in the lungs and brains, both of which 3,4 ER-α36 (a variant of ER-α) can trigger estrogen-responsivity in predict death in the short term . Notably, TNBCs occur more 8,9 TNBC , which leads to a growing concern. frequently in younger patients with its percentage increasing to 1,2 The identification of GPER, also known as GPR30, which was 25–30% in patients under 50 . Concerning systemic therapy, recognized as a membrane-associated receptor binding E2 with chemotherapy remains the standard management for TNBC while high affinity to mediate rapid and nongenomic estrogenic effects, targeting therapy is still at its early stage. Theoretically, the most including transactivation of epidermal growth factor receptor and effective strategy to improve patient outcomes may be by production of second messengers such as cAMP, calcium and blocking the metastatic process. inositol triphosphate, has challenged the traditional concept Estrogen, predominantly 17β-estradiol (E2), is a critical driver of 10–13 stating that TNBC was estrogen-independent . GPER is a mammary development and an essential etiological factor for breast cancer. As biologic mediators of estrogenic effects, ERs are member of G protein-coupled receptors (GPCRs), whose biological widely distributed in breast cancer. Specifically, ER-α, which is activity is dictated by posttranscriptional modifications (such as detected in about 70% of breast cancers, is used as not only a phosphorylation and ubiquitination) that control receptor con- centrations at the plasma membrane . In our earlier reports and powerful prognostic factor but also an efficient target for patients . The endocrine therapy that blocks estrogen signaling, others, GPER was detected in more than 60% of primary TNBC Department of Endocrine and Breast Surgery, The First Affiliated Hospital of Chongqing Medical University, 1 Youyi Road, Chongqing 400010, People’s Republic of China. Department of Breast Surgery, Precision Cancer Medicine Center, Fudan University Shanghai Cancer Center, 270 Dong’an Road, Shanghai 200032, People’s Republic of China. 3 4 Key Laboratory of Laboratory Medical Diagnostics, Chinese Ministry of Education, Chongqing Medical University, Chongqing 400016, People’s Republic of China. Department of Thyroid and Breast Surgery, Weihai Municipal Hospital, 70 Heping Road, Huancui District, Weihai, Shandong 264200, People’s Republic of China. Department of Thyroid and Breast Surgery, Heze Municipal Hospital, 2888 Caozhou West Road, Heze, Shandong 274031, People’s Republic of China. Department of Thyroid and Breast Surgery, The Second Affiliated Hospital of Chongqing Medical University, 74 Linjiang Road, Chongqing 400010, People’s Republic of China. Department of General Surgery, Shanxi Provincial People’s Hospital, Taiyuan, Shanxi 030000, People’s Republic of China. These authors contributed equally: Ting Xu, Ding Ma. email: luohaojun@hospital.cqmu.edu.cn; Yukeda@fudan.edu.cn Published in partnership with the Breast Cancer Research Foundation 1234567890():,; T. Xu et al. 15–20 samples and several TNBC cell lines . Importantly, GPER was Table 1. Tumor characteristics and GPER distribution. linked to the metastatic behaviors of TNBC cells in vitro and Variables No. (%)/Mean ± SD in vivo. Ligands including E2 and bisphenol A were claimed to trigger GPER to promote migration and metastasis of TNBC Total GPER-Low GPER-High P 21–23 cells , although the designation of GPER as a cognate ER is still (N = 360) (N = 265) (N = 95) 24–27 debated sometimes . In a recent bioinformatics analysis, GPER was also correlated with pro-metastatic genes and pathways in ER- Age 53.3 ± 11.4 53.4 ± 11.3 52.8 ± 11.5 0.624 α negative breast cancer . Considering the advantages of Menopause status 0.832 endocrine therapy, GPER has been included as a candidate Pre-menopause 132 (36.7) 98 (74.2) 34 (25.8) biomarker and a potential therapeutic target for TNBC . However, Post-menopause 224 (62.2) 164 (73.2) 60 (26.8) a controversial report concludes that GPER activation could inhibit Tumor size (pT) 2.64 ± 1.17 2.64 ± 1.25 2.66 ± 0.92 0.891 the in vivo invasive potential of TNBC via suppression of epithelial- mesenchymal transformation . Thus, the role of GPER in TNBC pT1 131 (36.4) 100 (76.3) 31 (23.7) metastasis needs further confirmation. pT2 219 (60.8) 156 (71.2) 63 (28.8) As an alternative ER, GPER has caught increasing attention in pT3 9 (2.5) 8 (88.9) 1 (11.1) breast cancer research, and the relationship between GPER and LNM (pN) 0.426 breast cancer outcomes has been addressed in multiple 30–34 pN0 209 (58.1) 158 (75.6) 51 (24.4) studies . Controversial findings on the prognostic role of GPER as well as the association between GPER expression and pN1 97 (26.9) 70 (72.2) 27 (27.8) clinicopathological determinants of breast cancer have been pN2 32 (8.9) 22 (68.8) 10 (31.2) reported. For instance, GPER was linked to worse relapse-free pN3 17 (4.7) 10 (58.8) 7 (41.2) survival (RFS) in breast cancer patients treated with tamoxifen . TNM stage 0.513 Meanwhile, GPER was also correlated with an increased distant I 89 (24.7) 70 (78.7) 19 (21.3) disease-free survival (DDFS) of ER-α positive breast cancer . The IIA 144 (40.0) 105 (72.9) 39 (27.1) biological functions of GPER largely depend on the cellular 12,13 background in vitro . In patients with TNBC, an early report IIB 76 (21.1) 56 (73.7) 20 (26.3) provided a clue that GPER might be associated with a poorer IIIA 34 (9.4) 24 (70.6) 10 (29.4) prognosis with a trend towards increased recurrences (without IIIC 17 (4.7) 10 (58.8) 7 (41.2) statistical significance, n = 18) . Recently, high expression of Histological 0.182 GPER was related to decreased outcomes in TNBC, including the IDC 330 (91.7) 246 (74.5) 84 (25.5) local RFS, DDFS, overall survival (OS), and progression-free survival (PFS) in a Chinese cohort (n = 249) . A bioinformatics analysis also Others 30 (8.3) 19 (63.3) 11 (36.7) associated the high expression of GPER with the decreased Nuclear grade 0.304 disease-free interval in ER-α negative breast cancer, although the 2 93 (25.8) 65 (69.9) 28 (30.1) scale of the cohort was limited (n = 120) . Thus, the prognostic 3 232 (64.4) 175 (75.4) 57 (24.6) significance of GPER in TNBC needs to be evaluated in larger unknown 35 (9.7) cohorts. Necrosis 0.027* We have reported the largest single-center study concerning the multi-omics profiling of TNBC, delineating the genomic and No 151 (41.9) 122 (80.8) 29 (19.2) transcriptomic landscape of Chinese TNBC patients . In this Yes 113 (31.4) 78 (69.0) 35 (31.0) cohort, 360 cases had RNA sequencing data on primary tumor Ki67 52.8 ± 25.3 51.7 ± 25.2 55.8 ± 25.5 0.180 tissue. To further evaluate the prognostic role of GPER, especially Intrinsic subtype 0.755 on metastatic manifestations in these patients, we analyzed the Basal like 277 (76.9) 205 (74.0) 72 (26.0) correlation between GPER expression and clinicopathological determinants of TNBC progression and long-term survival herein. Others 83 (23.1) 60 (72.3) 23 (27.7) We also show a bioinformatics analysis based on the transcrip- mRNA subtype 0.001* tomic profiles of our cohort to better understand the estrogenic IM 87 (24.2) 77 (88.5) 10 (11.5) carcinogenesis mediated by GPER in this aggressive breast cancer LAR 81 (22.5) 59 (72.8) 22 (27.2) subtype. BLIS 139 (38.6) 97 (69.8) 42 (30.2) MES 53 (14.7) 32 (60.4) 21 (39.6) RESULTS Recurrence 0.097 Patient characteristics No 300 (83.3) 226 (75.3) 74 (24.7) 360 patients who underwent surgery at Fudan University Yes 60 (16.7) 39 (65.0) 21 (35.0) Shanghai Cancer Center (FUSCC) between 2007 and 2014 were Metastatic events 0.019* included in this study (Table 1). The mean age of all patients was No 310 (86.1) 235 (75.8) 75 (24.2) 53.3 ± 11.4 years old (range, 25–84 years old) and 62.2% of them Yes 50 (13.9) 30 (60.0) 20 (40.0) were post-menopause. Most tumors were pT2 (60.8%), without lymph node metastasis (LNM) (58.1%), TNM stage II (61.1%), Lung 1.000 invasive ductal carcinoma (91.7%), nuclear grade 3 (64.4%), and No 344 (95.6) 253 (73.5) 91 (26.5) within the basal-like intrinsic subtype (76.9%). As for this cohort, Yes 16 (4.4) 12 (75.0) 4 (25.0) we classified the tumors into four transcriptome-based subtypes: Bone 0.210 luminal androgen receptor (LAR) subtype (22.5%), immunomodu- No 339 (94.2) 252 (74.3) 87 (25.7) latory (IM) subtype (24.2%), basal-like immune-suppressed (BLIS) Yes 21 (5.8) 13 (61.9) 8 (38.1) subtype (38.6%) and mesenchymal-like (MES) subtype (14.7%). Distant metastases were excluded present at the time of surgery. No patients received any systemic adjuvant therapy besides npj Breast Cancer (2022) 100 Published in partnership with the Breast Cancer Research Foundation 1234567890():,; T. Xu et al. staining. In some nests, weak staining was localized in the core Table 1 continued area while strong staining was observed at the margin (Fig. 1d). Variables No. (%)/Mean ± SD Association between GPER and clinicopathological variables Total GPER-Low GPER-High P (N = 360) (N = 265) (N = 95) of TNBC According to the cutoff value of RNA sequencing results, low and Brain 0.697 high GPER expression levels were detected in 73.6% (265/360) and No 354 (98.3) 261 (73.7) 93 (26.3) 26.4% (95/360) of patients, respectively (Table 1). The association Yes 6 (1.7) 4 (66.7) 2 (33.3) of GPER expression with clinicopathological variables was assessed. High GPER expression was significantly correlated with Liver 0.011* necrosis in the cancer nest (P = 0.027) and mRNA subtype No 348 (96.7) 260 (74.7) 88 (25.3) (P = 0.001) of TNBC according to our classification . Compared Yes 12 (3.3) 5 (41.7) 7 (58.3) to the GPER-low group, the GPER-high group demonstrated Contralateral supraclavicular lymph node metastasis 0.447 increased incidences of total metastatic events (21.1% vs. 11.3%; No 358 (99.4) 264 (73.7) 94 (26.3) P = 0.019) and liver metastasis (7.4% vs. 1.9%; P = 0.011) in the follow-up. Other clinicopathological variables, such as age, Yes 2 (0.6) 1 (50.0) 1 (50.0) menopausal status, and tumor size, had no significant correlations Death 0.190 with GPER expression. No 320 (88.9) 239 (74.7) 81 (25.3) Yes 40 (11.1) 26 (65.0) 14 (35.0) The distribution of GPER is different among subtypes of TNBC Chemotherapy 0.919 As mentioned, tumors in this cohort were classified into four No 10 (2.7) 8 (80.0) 2 (20.0) transcriptome-based subtypes and the distribution of GPER Yes 350 (97.2) 257 (73.4) 93 (26.6) 0.644 expression was significantly different among subtypes of TNBC Anthracycline and 214 (59.4) 159 (74.3) 55 (25.7) (Table 1). Thus, we quantitatively compared the abundance of taxane-based GPER among subtypes of TNBC by ANOVA analysis. Intriguingly, regimens the abundance of GPER was the lowest in the IM subtype which Others 136 (37.7) 98 (72.1) 38 (27.9) presented the best prognosis. Respectively, GPER abundance was highest in the MES subtype which presented the worst prognosis Bone 0.210 among all TNBC subtypes (Fig. 2). A significant difference was No 339 (94.2) 252 (74.3) 87 (25.7) found between LAR and IM (P = 0.042), BILS and IM (P = 0.002), Yes 21 (5.8) 13 (61.9) 8 (38.1) MES and IM (P < 0.001) as well as MES and LAR (P = 0.028). This Brain 0.697 may suggest that GPER is correlated with a higher risk subtype of No 354 (98.3) 261 (73.7) 93 (26.3) TNBC. Yes 6 (1.7) 4 (66.7) 2 (33.3) High expression of GPER predicted worse DMFS in TNBC Liver 0.011* patients No 348 (96.7) 260 (74.7) 88 (25.3) Survival outcomes were analyzed to explore the potential of GPER Yes 12 (3.3) 5 (41.7) 7 (58.3) as a survival predictor. Kaplan–Meier analysis of this cohort Data are expressed as the patient number (%) or mean ± SD. Statistically revealed a trend towards reduced RFS (Log Rank P = 0.078), DMFS significant differences were defined as P < 0.05. (Log Rank P = 0.014), and OS (Log Rank P = 0.136) in patients with LNM lymph node metastasis, IDC invasive ductal carcinoma, IM immuno- high GPER expression TNBCs. Note, that statistical significance is modulatory subtype, LAR luminal androgen receptor subtype, BLIS basal- only presented in DMFS (Fig. 3). like immune-suppressed subtype, MES mesenchymal-like subtype. Additionally, a Cox proportional hazard regression model was Includes 25 cases who received additional platinum regimens. b used to identify biomarkers and clinicopathological factors Consists of the following: single-agent taxane (n = 22), single-agent affecting the prognosis of patients with TNBC (Table 2). The anthracycline (n = 39), single-agent platinum (n = 2), combination of taxane and platinum (n = 40), and unknown agents (n = 33). univariate analysis confirmed that GPER was a significant prognostic factor for DMFS (hazard ratios (HR) = 2.01; 95% confidence interval (CI), 1.14–3.54; P = 0.016) in TNBC patients. chemotherapy. The median follow-up interval was 67.1 months (range However, the further multivariate analysis failed to provide 0.3–144.2 months). At the time of analysis, 60 patients underwent evidence for GPER (HR = 1.642; 95% CI, 0.92–2.92; P = 0.091) as recurrence, 50 patients had metastatic events and 40 patients died; an independent prognostic factor. Additionally, the prognostic the RFS was 83.3%, DMFS was 86.1% and OS was 88.9%. value of GPER was not significant in RFS and OS neither by univariate nor multivariate analysis. Referring to the other The expression level of GPER was correlated with the clinicopathological variables, LNM status, and TNM stage were immunohistochemistry (IHC) score in TNBC tissues identified as prognostic indicators for RFS, DMFS, and OS in the univariate model, while mRNA subtype was only associated with In our cohort, GPER expression at the mRNA level, shown as log2 DMFS. Multivariate analysis proved that TNM stage was a (FPKM+ 1) expression value, was of normal distribution in TNBC significant independent prognostic factor for RFS, DMFS, and OS tissues (Supplementary Fig. 1). To verify the expression of GPER in TNBC patients. Besides, LNM status was also suggested as an detected by RNA sequencing, samples from 12 patients were independent prognostic factor for RFS and DMFS, but not OS. randomly selected for IHC staining using an antibody against GPER. As expected, we found varying staining intensities of GPER in these tissues (Fig. 1a) and the IHC scoring was significantly correlated High GPER expression predicted worse survival in high-risk with the log2 expression value of GPER (Fig. 1b). Although both TNBC patients cytoplasmic and membrane staining of GPER were reported, we Intriguingly, the prognostic value of GPER, not only for DMFS but observed, even by oil lens, only cytoplasmic patterns in these also for RFS and OS, dramatically increased when stratifying for samples (Fig. 1c). Interestingly, we observed heterogeneity in GPER known risk factors. Revealed by Kaplan–Meier analysis, high Published in partnership with the Breast Cancer Research Foundation npj Breast Cancer (2022) 100 T. Xu et al. Fig. 1 Immunohistochemical staining for GPER. a Immunohistochemical staining for GPER with different intensity. The staining is graded as negative (−), weak (+), moderate (++), and strong (+++). Scale bar, 100 μm. b Scatter plots show that the IHC score is linearly correlated with the log2 expression value of GPER in the results of Pearson correlation coefficient calculation, which is statistically significant. Pearson correlation coefficient R and p-value are given in the scatter plot. c Image showing IHC staining of GPER is only observed in cytoplasmic. Scale bar, 50 μm. d Representative image of weak GPER staining in the core area and strong GPER staining in the corresponding margin. Scale bar, 100 μm. GPER was linked to pro-metastatic pathways in the transcriptomic landscape of Chinese TNBC patients To better understand the estrogenic carcinogenesis mediated by GPER in TNBC, we applied bioinformatics analysis based on the transcriptomic profiles of 360 TNBC patients in our cohort. The gene set enrichment analysis (GSEA) and gene set variation analysis (GSVA) were performed between the GPER-high and GPER-low groups of TNBC patients. Based on the results of GSEA analysis, the dot plot shows the significantly enriched GPER- related pathways (Fig. 5a). Of note, the enriched pathways with pro-metastatic characteristics included Focal adhesion, WNT signaling pathway, ECM receptor interaction, NOTCH signaling pathway, Hedgehog signaling pathway, Adherens junction path- way, and TGF beta signaling pathway, as indicated by their respective adjusted p-values and GSEA-plots (Fig. 5a, b). Addi- tionally, the GSVA analysis was conducted using KEGG gene sets. Firstly, 186 KEGG pathways were quantified using the GSVA package. Then, differential analysis was conducted to find specific pathways for GPER-high and GPER-low groups. Similar to the GSEA results, the GSVA results also showed that the pro-metastasis Fig. 2 Violin diagram showing the distribution of GPER expres- pathways that are significantly enriched in the GPER-high group sion among subtypes of TNBC. Statistically significant differences include NOTCH signaling pathway, Hedgehog signaling pathway, were defined as P < 0.05. Abbreviations: IM immunomodulatory WNT signaling pathway, and Adherens junction pathway (Fig. 5c). subtype; LAR luminal androgen receptor subtype; BILS basal-like immune-suppressed subtype; MES mesenchymal-like subtype. DISCUSSION expression of GPER was linked significantly to the worse RFS (Log TNBC has the worst prognosis of all breast cancer subtypes and Rank P = 0.012), DMFS (Log Rank P = 0.003), and OS (Log Rank the lack of well-defined molecular targets is the main challenge to P = 0.012) in LNM (+) patients while no difference was found in 1,2 treat TNBC patients . Although estrogens largely contribute to LNM (−) patients (Fig. 4a and Supplementary Fig. 2). Referring to the development and progression of breast cancer, estrogen stage III patients, high GPER expression correlated with lower RFS carcinogenesis was long disregarded in TNBC. In the present (Log Rank P = 0.013) and DMFS (Log Rank P = 0.014), and a trend study, we revealed that GPER expression was associated with the toward correlating with OS (Log Rank P = 0.132) (Fig. 4b and aggressive subtype of TNBC. High GPER expression predicted Supplementary Fig. 2). Similarly, high GPER expression is also reduced DMFS in our cohort. Especially in high-risk patients with associated with reduced RFS (Log Rank P = 0.045) and DMFS (Log G3 tumors, LNM (+) or stage III, the prognostic significance was Rank P = 0.008), and a trend toward OS (Log Rank P = 0.074) in increased. Transcriptome-based bioinformatics analysis revealed patients with G3 tumors (Fig. 4c and Supplementary Fig. 2). Thus, that GPER was linked to pro-metastatic pathways in the Chinese the predictive value of GPER seems to be increased in TNBC cohort of TNBC. patients with additional risk factors, including LNM positivity, Theoretically, GPER, as a membrane receptor belonging to the higher nuclear grade and later TNM stage. GPCR superfamily, is significantly different from ER-α, a nuclear npj Breast Cancer (2022) 100 Published in partnership with the Breast Cancer Research Foundation T. Xu et al. Fig. 3 Kaplan–Meier survival curves of RFS, DMFS, and OS between GPER-low and GPER-high groups in patients with TNBC. Statistically significant differences were defined as P < 0.05. Abbreviations: RFS relapse-free survival; DMFS distant metastasis-free survival; OS overall survival. steroid hormone receptor. In biology, the differences between development and progression remains elusive. In an early study, GPER and ER-α include their subcellular distribution, structure, GPER was detected in every single breast sample from 12 healthy affinity to E2, ligands pattern, the process and effects in response donors . The GPER expression in normal breast tissue was at a to E2. These differences have attracted a surge of interest and are medium level in the Human Protein Atlas database .In 12,13,24 well-reviewed in the literature . Meanwhile, GPER is not comparison, GPER expression was reported to decrease in tumor 24–27 52 totally accepted as a cognate ER in related debates . GPER was tissues ,while inflammatory breast cancer, an aggressive type of reported to coimmunoprecipitate with ER-α in MCF-7 cells and to breast cancer, exhibited stronger intensity in staining against repeatedly correlate with ER-α positivity in primary breast GPER . Interestingly, GPER expression was correlated with the 30,32,37 cancers , thus it may contribute to estrogenic responses as tumor subtype in a large cohort and strong staining was a collaborator. GPER didn’t even respond to stimulation of E2 and significantly more prevalent among TNBCs .Notably, whenTNBC G1 in some cell models . However, GPER was detected in tissues subtypes were classified by transcriptomic profile, the expression of and cell lines of not only breast cancer but also other organs GPER correlated with the subtype, and the highest level was found 12,13 lacking ER-α expression . Furthermore, GPER was observed to in the MES subtype, which presented the worst prognosis in all 10,39,40 bind E2 directly in several cell lines lacking ER-α . In clinical TNBC subtypes. In addition, weak positivity was localized in the core series, inverse correlation or non-significant association was also area while strong positivity was observed at the margin in some found between GPER and ER-α . Due to the lack of ER-α, it was nests by IHC staining, implying that GPER abundance may be assumed that TNBC is estrogen-insensitive. However, increasing increased during the invasion of cancer cells. It also implied a link circulating estrogen levels were sufficient to promote the between GPER and the aggressiveness of the breast cancer, that is, formation and progression of ER-α negative cancers including high expression of GPER was associated with more frequent TNBC and pharmacological inhibition of estrogen synthesis after necrosis in tumor sections. Besides, metastatic or recurrent cancer pregnancy prevented the formation of ER-α negative tumors . tumors also showed higher levels of GPER expression than 19,31,53 Furthermore, we found that the mRNA expression level of GPER, corresponding primary tumor .Meanwhile, the aggressive which is positively correlated with the staining score of the GPER cell lines of uterine and ovarian cancer (JEG and Hec50) expressed a protein (R² = 0.7603), is widely distributed in this large Chinese much higher level of GPER than their associated normal cell lines cohort of TNBC tissues, in line with earlier detection of GPER in (HTR8 and H, respectively) . In general, GPER expression likely 15–20 tissues and cell lines of TNBC . Taken together, these results increases with the development and progression while it indicates indicate a functional role of GPER as an alternative ER and the the natural aggressiveness of breast cancer. potential as a mediator of estrogen carcinogenesis in TNBC. The prognostic value of GPER for breast cancer patients has 30–34 However, further identification, especially based on GPER protein been addressed in multiple studies . However, controversial expression, is needed. findings have been reported, even in meta-analysis . The There are some disputes about the subcellular localization of biological function of GPER largely depends on the cellular 12,13 GPER. Known as a GPCR, GPER can be detected on the cell surface background in vitro , thus the prognostic role of GPER should 2+ and is involved in signal transduction events, such as Ca be evaluated in a specific subtype of breast cancer. In this respect, 10 43 44 mobilization , NO generation , ERK activation , and growth the contribution of GPER should be genuine in TNBC. In this large factor release . GPER was also observed to locate in the cohort of patients, high GPER expression was linked to poor DMFS endoplasmic reticulum and to act as an endoplasmic reticulum and exhibited prognostic significance, especially in high-risk stressor that induces growth inhibition and apoptotic cell patients with G3 tumors, LNM (+) and stage III. These results death . GPER distribution in mitochondria, Golgi apparatus, and were consistent with earlier reports as follows: In an earlier report nucleus also reported . The different subcellular locations of the with a small cohort of TNBC (n = 18), GPER was thought to have a GPER may have different biological implications. Importantly, our poorer prognosis with a trend towards increased recurrences . group has found that GPER exists in the nucleus and is High GPER expression was also related to decreased outcomes, translocated from the nucleus to the cytoplasm under including the local RFS, DDFS, OS, and PFS in a Chinese cohort 49 17 E2 stimulation . In this study, GPER staining only exhibited in (n = 249) . In a retrospective TNBC study (n = 199), GPER and the cytoplasm may be due to the small sample size, different estrogen-related receptor α (ERR-α) synergistically predicted poor functional status of cells, and less sensitive detection methods. patient outcomes . In summary, studies seem to agree that GPER Physiologically, the expression pattern of GPER is likely to be can mediate estrogen carcinogenesis and promote the progres- tissue-dependent and developmentally regulated. In the mam- sion of TNBC. Hence, GPER could be of significant prognostic mary ductal epithelia, GPER abundance was varied with the value. However, a larger cohort and longer follow-up are needed estrous cycle . Yet, the trend of GPER expression in breast cancer to address this issue. Published in partnership with the Breast Cancer Research Foundation npj Breast Cancer (2022) 100 T. Xu et al. Table 2. Univariate and multivariate survival analyses of clinicopathological factors and GPER expression. Univariate model Multivariate model P HR β 95% CI for HR P HR β 95% CI for HR RFS Age (<50 vs. >=50) 0.106 0.658 −0.418 0.396–1.093 Menopause status (pre-menopause vs. post-menopause) 0.547 0.853 −0.159 0.509–1.430 Nuclear grade (G2 vs. G3) 0.109 0.633 −0.457 0.362–1.107 Tumor size (pT1 vs. pT2/3) 0.332 1.309 0.270 0.760–2.256 LNM (positive vs. negative) <0.001 3.474 1.245 2.015–5.987 0.018 2.185 0.782 1.147–4.165 TNM stage (III vs. I/II) <0.001 4.763 1.561 2.828–8.022 0.001 2.963 1.086 1.599–5.493 Necrosis (positive vs. negative) 0.434 1.270 0.239 0.698–2.309 mRNA subtype (IM vs. others) 0.088 0.540 −0.616 0.266–1.097 Intrinsic subtype (Basal vs. others) 0.565 1.183 0.168 0.667–2.096 Histology (IDC vs. others) -- - - Ki67 (>=30% vs. <30%) 0.425 1.305 0.266 0.678–2.514 GPER (high vs. low) 0.081 1.604 0.473 0.944–2.727 DMFS Age (<50 vs. >=50) 0.178 0.682 −0.383 0.391–1.190 Menopause status (pre-menopause vs. post-menopause) 0.388 0.781 −0.247 0.445–1.370 Nuclear grade (G2 vs. G3) 0.392 0.760 −0.275 0.405–1.425 Tumor size (pT1 vs. pT2/3) 0.395 1.294 0.258 0.714–2.344 LNM (positive vs. negative) <0.001 5.020 1.613 2.622–9.610 0.005 3.005 1.100 1.403–6.432 TNM stage (III vs. I/II) <0.001 6.490 1.870 3.717–11.334 0.001 3.132 1.142 1.622–6.046 Necrosis (positive vs. negative) 0.116 1.721 0.543 0.874–3.388 mRNA subtype (IM vs. others) 0.022 0.340 −1.080 0.135–0.856 0.082 0.432 −0.840 0.167–1.113 Intrinsic subtype (Basal vs. others) 0.684 1.140 0.131 0.606–2.145 Histology (IDC vs. others) 0.993 0.995 −0.005 0.358–2.765 Ki67 (>=30% vs. <30%) 0.271 1.530 0.425 0.717–3.264 GPER (high vs. low) 0.016 2.009 0.698 1.141–3.538 0.091 1.642 0.496 0.923–2.920 OS Age (<50 vs. >=50) 0.489 0.802 −0.221 0.428–1.501 Menopause status (pre-menopause vs. post-menopause) 0.828 1.075 0.072 0.561–2.058 Nuclear grade (G2 vs. G3) 0.804 0.913 −0.091 0.447–1.866 Tumor size (pT1 vs. pT2/3) 0.775 1.100 0.095 0.574–2.106 LNM (positive vs. negative) 0.001 3.225 1.171 1.663–6.252 0.197 1.720 0.542 0.754–3.924 TNM stage (III vs. I/II) <0.001 5.445 1.695 2.907–10.200 0.001 3.871 1.354 1.772–8.457 Necrosis (positive vs. negative) 0.192 1.658 0.506 0.776–3.543 mRNA subtype (IM vs. others) 0.204 0.570 −0.562 0.239–1.358 Intrinsic subtype (Basal vs. others) 0.569 0.798 −0.225 0.368–1.732 Histology (IDC vs. others) 0.500 1.632 0.490 0.394–6.764 Ki67 (>=30% vs. <30%) 0.780 1.117 0.111 0.515–2.425 GPER (high vs. low) -- - - Covariates with P < 0.05 in the univariate model were included in the further multivariate model. Statistically significant differences were defined as P < 0.05. Covariates that did not satisfy the proportional hazards assumption with Schoenfeld residuals test. As aforementioned, early metastasis is underlying the poor Focal adhesion and ECM receptor interaction pathways, both of 2,3 prognosis of patients with TNBC . Interestingly, GPER was linked which were deeply involved in cancer invasion and metastasis, to the metastatic behaviors of TNBC cells in vitro and were also the most significant GPER-related pro-metastatic path- 16,18,19,21,56 in vivo . Ligands including E2 and bisphenol A were ways in the aforementioned study. Actually, estrogenic GPER claimed to trigger GPER to promote migration and metastasis of signaling was ascertained to trigger focal adhesion kinase 21–23 TNBC cells . E2 also induced the up-regulation of estrogen- phosphorylation to increase focal adhesion points and cellular related receptor α expression via GPER activation, enhanced the migration in TNBC cells . Other pro-metastatic pathways includ- migration and invasion of TNBC cells . Increased GPER expression ing NOTCH signaling, Hedgehog signaling, WNT signaling, and was observed at the invasive margin in some nests, implying that Adherens junction were also correlated with GPER in our analysis. GPER activation may contribute to the invasion of cancer cells. A However, the GPER-related pro-metastatic pathways identified by recent bioinformatics analysis showed that GPER was correlated Maggiolini et al. were somewhat different with those in this study, with pro-metastatic pathways in ER-α negative breast cancer by maybe due to the differences of tumor subtype (ER-α negative vs. Maggiolini et al. . We conducted a similar analysis in our large TNBC) and race of patients (mixed vs. Chinese). To our knowledge, single-center study and, as expected, GPER was correlated with performing the analysis in homogeneous TNBC cohort is multiple pro-metastatic pathways. Among these pathways, the significant since heterogeneity of TNBC is high enough and npj Breast Cancer (2022) 100 Published in partnership with the Breast Cancer Research Foundation T. Xu et al. Fig. 4 Kaplan–Meier curves between GPER-low and GPER-high groups in high-risk patients. a Kaplan–Meier curves of RFS, DMFS and OS in LNM (+) patients. b Kaplan–Meier curves of RFS, DMFS and OS in stage III patients. c Kaplan–Meier curves of RFS, DMFS and OS in patients with G3 tumors. Statistically significant differences were defined as P < 0.05. excluding the effect of crosstalk between estrogenic GPER GPER expression and pro-metastatic pathways was verified, signaling and HER2 signaling is necessary. Anyway, given that suggesting that GPER has significant functional roles in TNBC GPER mediates the transcriptional regulation of estrogen in TNBC metastasis. Taken together, this may provide evidence that GPER 19,57 cells and other breast cancer cell lines , GPER may trigger mediates metastatic estrogen carcinogenesis in TNBC. Considering variable signal transduction events to enhance the multi-steps of the great urgency for clinicians and researchers to develop metastasis. However, to better understand the functional role of efficient molecular targets and biomarkers, GPER could be a GPER, the mechanisms by which GPER contributes to the promising candidate for TNBC therapy and diagnosis. progression of TNBC need further clarification. Accordingly, new endocrine therapy by blocking GPER-related signaling may be an METHODS effective strategy for TNBC. Since several antagonists of GPER have been synthesized, employing them as endocrine therapy agents is Patient recruitment accessible, while more basic and translational research is needed From January 1, 2007 to December 31, 2014, primary tumor tissue and to confirm this potency. blood samples were obtained from 504 consecutive female Chinese patients with TNBC treated at Fudan University Shanghai Cancer Center In summary, in a unique and large Chinese cohort of TNBC with (FUSCC). Among these patients, 279 had whole exome sequencing (WES) long-term follow-up, we evaluated the expression of GPER and data on primary tumor tissue and paired blood samples, 401 had copy- showed that GPER expression correlates with the subtype of TNBC, number alteration (CNA) data and 360 had RNA sequencing data on with a trend to increase the aggressiveness of tumors. We primary tumor tissue. 360 patients with RNA sequencing data were concluded that GPER has significant prognostic value in TNBC and enrolled in this study according to the following defined criteria: (1) female is significantly linked to the worse survival, especially in high-risk patients diagnosed with unilateral disease; (2) histologically confirmed the patients with LNM (+), G3 or stage III tumors. Furthermore, ER-α (−), PR (−), and HER2 (−) phenotype; (3) no evidence of distant bioinformatics analysis was performed based on the transcrip- metastasis at diagnosis; (4) sufficient frozen tissues available for further tomic profile of the TNBC cohort and the correlation between high research. The examination results for chest computed tomography (CT), Published in partnership with the Breast Cancer Research Foundation npj Breast Cancer (2022) 100 T. Xu et al. Fig. 5 The GSEA and GSVA results of GPER correlated pathways based on the transcriptomic data of 360 TNBC patients. a Dotplot showing the twelve most significantly upregulated pathways in GPER-high and GPER-low groups from GSEA results. In the group of high GPER expression, seven of the twelve upregulated pathways are correlated with promoting tumor metastasis. b GSEA-plots showing the upregulated pro-metastatic pathways in the GPER-high group. c Heatmap for the eight most significantly upregulated pathways in GPER-high group by GSVA. Statistically significant differences were defined as adjusted P-value < 0.05. bone scans, abdominal ultrasound, bilateral mammography, breast deposited in NCBI Sequence Read Archive, with accession number ultrasound, and/or magnetic resonance imaging (MRI) were collected to SRP157974. Four stable clusters, IM, LAR, MES, and BLIS were identified after ascertain no metastasis beyond breasts and axillary lymph nodes metastasis before the surgery. Ethical review and approval were waived analyzing the robustness of the classification using k-means clustering with the details available in our previous study . Our classification system, for this study, due to the data reported in this paper have been described named FUSCC, correlated well with the Lehmann/Pietenpol classification in our published article and deposited in the NCBI Sequence Read Archive (SRA: SRP157974). All patients provided written informed consent system. for data and tissue use. Clinicopathological data RNA sequencing and transcriptomic profiling The ER-α, PR, and HER2 status of the breast tumor samples were confirmed RNA sequencing data and transcriptomic profiling of 360 patients with by two experienced pathologists based on immunochemical analysis and TNBC from FUSCC were used in the current study. Detailed sample in situ hybridization. ER-α and PR status were classified as negative using a preparation, library preparation, sequencing, and raw data processing were cutoff of 1%, according to the American Society of Clinical Oncology/ 36 58 described in our earlier publication . The RNA sequencing data have been College of American Pathologists (ASCO/CAP) guidelines . HER2 status npj Breast Cancer (2022) 100 Published in partnership with the Breast Cancer Research Foundation T. Xu et al. was defined as negative with 0, 1+ as well as 2+ on immunohistochem- multivariate model. The proportional hazards assumption was tested by istry without HER2 gene amplification on fluorescence in situ hybridization using Schoenfeld residual tests. If the assumption of proportional hazards (FISH) . TNBC was defined as ER-α, PR, and HER2 negative in accordance was not valid, time-dependent covariates were introduced. All tests were with the St. Gallen International Expert Consensus . Clinicopathological two-sided and P < 0.05 was deemed statistically significant. features, including age at diagnosis, menopausal status, tumor histologic type, tumor size, LNM, histologic grade, TNM stage and ER-α, PR, HER2, and Reporting summary Ki67 status, were analyzed. The tumor stage based on the TNM stage was Further information on research design is available in the Nature Research assessed according to the criteria established by the 8th edition American Reporting Summary linked to this article. Joint Committee on Cancer (AJCC 8th) staging manual of breast cancer. Patient follow-up DATA AVAILABILITY Follow-up of all patients in this cohort was completed on June 11, 2019. RNA sequencing data that support the findings of this study have been deposited in The median length of follow-up was 67.1 months with an interquartile NCBI Sequence Read Archive with the accession codes SRP157974. All other relevant range of 53.9–79.9 months. RFS was defined as the time from diagnosis to data are available from the corresponding author on request. first recurrence or a diagnosis of contralateral breast cancer. DMFS was defined as the time from diagnosis to first distant metastasis. OS was defined as the time from diagnosis to death. Patients without events were CODE AVAILABILITY censored from the time point of the last follow-up. The code used to process and analyze the RNA sequencing data is available from the corresponding author upon reasonable request. Immunohistochemistry staining and scoring Immunohistochemistry staining was performed using an SP900 Kit Received: 8 December 2021; Accepted: 9 August 2022; (Zhongshan Golden Bridge) according to the manufacturer’s protocol. Briefly, deparaffinized tissue sections of 4 μm thickness were heated for antigen retrieval at 95 °C for 15 min in 10 mM citric acid buffer (pH 6.0). After treatment with 3% H O for 10 min to quench endogenous 2 2 peroxidase activity, the sections were blocked using goat serum and then REFERENCES incubated with the primary antibody targeting GPER (1:250, ab39742, 1. Al-Mahmood, S., Sapiezynski, J., Garbuzenko, O. B. & Minko, T. Metastatic and Abcam, USA) at a 1:200 dilution at 4 °C for 16 h. The section treated with triple-negative breast cancer: challenges and treatment options. Drug Deliv. PBS worked as a negative control. Following treatment of horseradish Transl. Res. 8, 1483–1507 (2018). peroxidase-conjugated goat anti-rabbit IgG for 30 min at 37 °C, sections 2. Garrido-Castro, A. C., Lin, N. U. & Polyak, K. Insights into molecular classifications were developed using diaminobenzidine (DAB) (Zhongshan Golden of triple-negative breast cancer: Improving patient selection for treatment. Bridge) and nuclei were counterstained with Mayer’s modified hematox- Cancer Discov. 9, 176–198 (2019). ylin. As indicated by Filardo et al. , reduction mammoplasty tissue was 3. Denkert, C., Liedtke, C., Tutt, A. & von Minckwitz, G. Molecular alterations in triple- used as a positive control. negative breast cancer—the road to new treatment strategies. Lancet 389, Two observers microscopically evaluated the intensity, extent and 2430–2442 (2017). subcellular distribution of GPER using a modified semi-quantitative scoring 4. Dent, R. et al. Triple-negative breast cancer: Clinical features and patterns of system. GPER scores were assigned as follows: the percentage of positive recurrence. Clin. Cancer Res. 13, 4429–4434 (2007). cells was categorized as 0 (negative staining in all cells), 1 (<1% cells 5. Dunnwald, L. K., Rossing, M. A. & Li, C. I. Hormone receptor status, tumor char- stained), 2 (1–10% of cells stained), 3 (11–30% cells stained), 4 (31–70% acteristics, and prognosis: a prospective cohort of breast cancer patients. Breast cells stained) or 5 (71–100% cells stained), and staining intensity was Cancer Res. 9, R6 (2007). categorized as 0 (negative), 1 (weak), 2 (moderate) and 3 (strong). Adding 6. Yuan, J. et al. Acquisition of epithelial-mesenchymal transition phenotype in the the two scores together yielded a maximum score of 8. The final scores tamoxifen-resistant breast cancer cell: A new role for G protein-coupled estrogen were grouped into GPER negative (score 0−4) and positive (score 5−8) receptor in mediating tamoxifen resistance through cancer-associated fibroblast- categories for statistical analyses to reduce inter-observer differences. derived fibronectin and β1-integrin signaling pathway in tumor cells. Breast Cancer Res. 17, 69 (2015). Bioinformatics analysis 7. Davies, C. et al. Relevance of breast cancer hormone receptors and other factors to the efficacy of adjuvant tamoxifen: Patient-level meta-analysis of randomised The 360 patients with TNBC were divided into 26.4% (95/360) patients with trials. Lancet 378, 771–784 (2011). high GPER expression and 73.6% (265/360) patients with low GPER 8. Treeck, O., Schüler-Toprak, S. & Ortmann, O. Estrogen actions in triple-negative expression. RNA sequencing data of 360 TNBC patients were used to breast cancer. Cells 9, 2358 (2020). determine the differences between groups through the R software limma 9. Wang, Z.-Y. & Yin, L. Estrogen receptor alpha-36 (ER-α36): A new player in human package and GSEA was applied to conduct gene enrichment analysis by breast cancer. Mol. Cell Endocrinol. 418, 193–206 (2015). using clusterProfiler package . GSVA analysis was performed on log2 10. Revankar, C. M., Cimino, D. F., Sklar, L. A., Arterburn, J. B. & Prossnitz, E. R. A (FPKM+ 1) expression values by using GSVA package . The “c2.cp.kegg.- transmembrane intracellular estrogen receptor mediates rapid cell signaling. v7.4.entrez.gmt” and “c2.cp.kegg.v7.2.symbols.gmt” gene sets were down- Science 307, 1625–1630 (2005). loaded from the Molecular Signatures Database . 11. Filardo, E. J. Epidermal growth factor receptor (EGFR) transactivation by estrogen via the G-protein-coupled receptor, GPR30: A novel signaling pathway with potential Statistical analysis significance for breast cancer. J. Steroid Biochem. Mol. Biol. 80,231–238 (2002). All statistical analyses were done by using the SPSS standard version 12. Hsu, L.-H., Chu, N.-M., Lin, Y.-F. & Kao, S.-H. G-Protein Coupled Estrogen Receptor 25 software and Stata version 13.0 software. For the division of high and in Breast Cancer. Int. J. Mol. Sci. 20, 306 (2019). low expression groups of GPER, the X-tile software was used to generate 13. Prossnitz, E. R. & Barton, M. The G-protein-coupled estrogen receptor GPER in the optimal cut-off value . Continuous quantitative data were expressed health and disease. Nat. Rev. Endocrinol. 7, 715–726 (2011). as mean ± standard deviation (SD) and categorical qualitative data as 14. Nelson, C. P. & Challiss, R. A. J. “Phenotypic” pharmacology: The influence of percentage. Data from the two patient groups were statistically compared cellular environment on G protein-coupled receptor antagonist and inverse using the chi-square test or t test as appropriate. The Pearson correlation agonist pharmacology. Biochem. Pharm. 73, 737–751 (2007). coefficient was used to analyze the correlation between IHC scoring and 15. Chen, Z. J. et al. Activation of GPER suppresses epithelial mesenchymal transition the log2 expression value of GPER. ANOVA test was used to determine the of triple negative breast cancer cells via NF-kappaB signals. Mol. Oncol. 10, differences of GPER expression among TNBC subtypes. RFS, DMFS, and OS 775–788 (2016). curves were drawn using the Kaplan–Meier methods and were compared 16. Yu, T. et al. GPER mediates enhanced cell viability and motility via non-genomic using Log-rank tests. Univariate and multivariate analyses of the patients’ signaling induced by 17beta-estradiol in triple-negative breast cancer cells. J. survival were performed using the Cox proportional hazards regression Steroid Biochem. Mol. Biol. 143, 392–403 (2014). model and the HRs with 95% CIs were calculated. Statistics of P < 0.05 in 17. Ye, S. et al. Prognostic role of GPER/Ezrin in triple-negative breast cancer is univariate analysis were used as inclusion criteria for covariates in the final associated with menopausal status. Endocr. Connect. 8, 661–671 (2019). Published in partnership with the Breast Cancer Research Foundation npj Breast Cancer (2022) 100 T. Xu et al. 18. Deng, Q. et al. GPER/Hippo-YAP signal is involved in Bisphenol S induced migration 46. Ahola, T. M., Manninen, T., Alkio, N. & Ylikomi, T. G protein-coupled receptor 30 is of triple negative breast cancer (TNBC) cells. J. Hazard. Mater. 355,1–9 (2018). critical for a progestin-induced growth inhibition in MCF-7 breast cancer cells. 19. Yin, J. et al. GPER-regulated lncRNA-Glu promotes glutamate secretion to Endocrinology 143, 3376–3384 (2002). enhance cellular invasion and metastasis in triple-negative breast cancer. FASEB J. 47. Song, L., De Sarno, P. & Jope, R. S. Central role of glycogen synthase kinase-3beta 34, 4557–4572 (2020). in endoplasmic reticulum stress-induced caspase-3 activation. J. Biol. Chem. 277, 20. Wang, Y. et al. NHERF1 inhibits proliferation of triple-negative breast cancer cells 44701–44708 (2002). by suppressing GPER signaling. Oncol. Rep. 38, 221–228 (2017). 48. Irannejad, R. et al. Functional selectivity of GPCR-directed drug action through 21. Rigiracciolo, D. C. et al. Focal adhesion kinase (FAK) activation by estrogens location bias. Nat. Chem. Biol. 13, 799–806 (2017). involves GPER in triple-negative breast cancer cells. J. Exp. Clin. Cancer Res. 38,58 49. Yu, T. et al. Cytoplasmic GPER translocation in cancer-associated fibroblasts (2019). mediates cAMP/PKA/CREB/glycolytic axis to confer tumor cells with multidrug 22. Xu, F. et al. Bisphenol A induces proliferative effects on both breast cancer cells resistance. Oncogene 36, 2131–2145 (2017). and vascular endothelial cells through a shared GPER-dependent pathway in 50. Wang, C., Prossnitz, E. R. & Roy, S. K. Expression of G protein-coupled receptor 30 hypoxia. Environ. Pollut. 231, 1609–1620 (2017). in the hamster ovary: differential regulation by gonadotropins and steroid hor- 23. Castillo-Sanchez, R. et al. Bisphenol A induces focal adhesions assembly and mones. Endocrinology 148, 4853–4864 (2007). activation of FAK, Src, and ERK2 via GPER in MDA-MB-231 breast cancer cells. 51. Uhlén, M. et al. Proteomics. Tissue-based map of the human proteome. Science Toxicol. Vitr. 66, 104871 (2020). 347, 1260419 (2015). 24. Luo, J. & Liu, D. Does GPER really function as a G protein-coupled estrogen 52. Ignatov, T. et al. GPER-1 expression decreases during breast cancer tumorigen- receptor? Front. Endocrinol. 11, 148 (2020). esis. Cancer Investig. 31, 309–315 (2013). 25. Levin, E. R. G Protein-coupled receptor 30: Estrogen receptor or collaborator. 53. Mo, Z. et al. GPR30 as an initiator of tamoxifen resistance in hormone-dependent Endocrinology 150, 1563–1565 (2009). breast cancer. Breast Cancer Res. 15, R114 (2013). 26. Olde, B. & Leeb-Lundberg, L. M. F. GPR30/GPER1: Searching for a role in estrogen 54. Yang, F. & Shao, Z. M. Double-edged role of G protein-coupled estrogen receptor physiology. Trends Endocrinol. Metab. 20, 409–416 (2009). 1 in breast cancer prognosis: An analysis of 167 breast cancer samples and online 27. Langer, G. et al. A critical review of fundamental controversies in the field of data sets. OncoTargets Ther. 9, 6407–6415 (2016). GPR30 research. Steroids 75, 603–610 (2010). 55. Ye, S. et al. Estrogen-related receptor α (ERRα) and G protein-coupled estrogen 28. Talia, M. et al. The G Protein-coupled estrogen receptor (GPER) expression cor- receptor (GPER) synergistically indicate poor prognosis in patients with triple- relates with pro-metastatic pathways in ER-negative breast cancer: A bioinfor- negative breast cancer. OncoTargets Ther. 13, 8887–8899 (2020). matics analysis. Cells 9, 622 (2020). 56. Marjon, N. A., Hu, C., Hathaway, H. J. & Prossnitz, E. R. G protein-coupled estrogen 29. Lappano, R., Pisano, A. & Maggiolini, M. GPER function in breast cancer: An receptor regulates mammary tumorigenesis and metastasis. Mol. Cancer Res. 12, overview. Front. Endocrinol. 5, 66 (2014). 1644–1654 (2014). 30. Filardo, E. J. et al. Distribution of GPR30, a seven membrane-spanning estrogen 57. Pandey, D. P. et al. Estrogenic GPR30 signalling induces proliferation and receptor, in primary breast cancer and its association with clinicopathologic migration of breast cancer cells through CTGF. EMBO J. 28, 523–532 (2009). determinants of tumor progression. Clin. Cancer Res. 12, 6359–6366 (2006). 58. Hammond, M. E. H. et al. American Society of Clinical Oncology/College of 31. Ignatov, A. et al. G-protein-coupled estrogen receptor GPR30 and tamoxifen American Pathologists Guideline recommendations for immunohistochemical resistance in breast cancer. Breast Cancer Res. Treat. 128, 457–466 (2011). testing of estrogen and progesterone receptors in breast cancer. J. Clin. Oncol. 28, 32. Broselid, S. et al. G protein-coupled estrogen receptor is apoptotic and correlates 2784–2795 (2010). with increased distant disease-free survival of estrogen receptor-positive breast 59. Wolff, A. C. et al. Human epidermal growth factor receptor 2 testing in breast cancer patients. Clin. Cancer Res. 19,1681–1692 (2013). cancer: American Society of Clinical Oncology/College of American Pathologists 33. Tutzauer, J. & Sjostrom, M. Plasma membrane expression of G protein-coupled Clinical Practice Guideline focused update. J. Clin. Oncol. 36, 2105–2122 (2018). estrogen receptor (GPER)/G protein-coupled receptor 30 (GPR30) is associated 60. Goldhirsch, A. et al. Personalizing the treatment of women with early breast with worse outcome in metachronous contralateral breast cancer. PLoS One 15, cancer: Highlights of the St Gallen International Expert Consensus on the Primary e0231786 (2020). Therapy of Early Breast Cancer 2013. Ann. Oncol. 24, 2206–2223 (2013). 34. Sjöström, M. et al. Lack of G protein-coupled estrogen receptor (GPER) in the 61. Ritchie, M. et al. limma powers differential expression analyses for RNA- plasma membrane is associated with excellent long-term prognosis in breast sequencing and microarray studies. Nucleic Acids Res. 43, e47 (2015). cancer. Breast Cancer Res. Treat. 145,61–71 (2014). 62. Yu, G., Wang, L., Han, Y. & He, Q. clusterProfiler: An R package for comparing 35. Steiman, J., Peralta, E. A., Louis, S. & Kamel, O. Biology of the estrogen receptor, biological themes among gene clusters. Omics: J. Integr. Biol. 16, 284–287 (2012). GPR30, in triple negative breast cancer. Am. J. Surg. 206, 698–703 (2013). 63. Hänzelmann, S., Castelo, R. & Guinney, J. GSVA: Gene set variation analysis for 36. Jiang, Y. Z. et al. Genomic and transcriptomic landscape of triple-negative breast microarray and RNA-Seq data. BMC Bioinform. 14, 7 (2013). cancers: Subtypes and treatment strategies. Cancer Cell 35, 428–440 (2019). 64. Liberzon, A. et al. Molecular signatures database (MSigDB) 3.0. Bioinformatics 27, 37. Vivacqua, A. et al. G protein-coupled receptor 30 expression is up-regulated by 1739–1740 (2011). EGF and TGF alpha in estrogen receptor alpha-positive cancer cells. Mol. Endo- 65. Camp, R. L., Dolled-Filhart, M. & Rimm, D. L. X-tile: A new bio-informatics tool for crinol. 23, 1815–1826 (2009). biomarker assessment and outcome-based cut-point optimization. Clin. Cancer 38. Tutzauer, J. et al. Ligand-independent G protein-coupled estrogen receptor/G Res. 10, 7252–7259 (2004). protein-coupled receptor 30 activity: Lack of receptor-dependent effects of G-1 and 17-estradiol. Mol. Pharm. 100, 271–282 (2021). 39. Thomas, P., Pang, Y., Filardo, E. J. & Dong, J. Identity of an estrogen membrane ACKNOWLEDGEMENTS receptor coupled to a G protein in human breast cancer cells. Endocrinology 146, This research was funded by the National Natural Science Foundation of China 624–632 (2005). 40. Mauvais-Jarvis, F., Lange, C. A. & Levin, E. R. Membrane-initiated estrogen, (82072938) for H.L., (82002802) for D.M., and (81722032 and 82072916) for K.Y. androgen, and progesterone receptor signaling in health and disease. Endocr. Rev. 43, 720–742 (2022). 41. Arias-Pulido, H. et al. GPR30 and estrogen receptor expression: New insights into AUTHOR CONTRIBUTIONS hormone dependence of inflammatory breast cancer. Breast Cancer Res. Treat. Conceptualization, T.X. and H.L.; data curation, T.X. and H.L.; formal analysis, J.W.; 123,51–58 (2010). funding acquisition, D.M., H.L., and K.Y.; methodology, D.M., Y.F., and L.L.; project 42. Gupta, P. B. & Kuperwasser, C. Contributions of estrogen to ER-negative breast administration, H.L. and K.Y.; resources, K.Y.; software, T.X. and D.M.; supervision, H.L. tumor growth. J. Steroid Biochem. Mol. Biol. 102,71–78 (2006). 43. Haynes, M. P. et al. Membrane estrogen receptor engagement activates endo- and K.Y.; validation, S.C., R.T., and J.Y.; visualization, T.X. and D.M.; writing – original thelial nitric oxide synthase via the PI3-kinase-Akt pathway in human endothelial draft, T.X. and H.L.; writing – review & editing, C.M. T.X., and D.M. contributed equally cells. Circ. Res. 87, 677–682 (2000). to this work. 44. Maggiolini, M. et al. The G protein-coupled receptor GPR30 mediates c-fos up- regulation by 17beta-estradiol and phytoestrogens in breast cancer cells. J. Biol. Chem. 279, 27008–27016 (2004). COMPETING INTERESTS 45. Filardo, E. J. & Thomas, P. GPR30: A seven-transmembrane-spanning estrogen receptor that triggers EGF release. Trends Endocrinol. Metab. 16, 362–367 (2005). The authors declare no competing interests. npj Breast Cancer (2022) 100 Published in partnership with the Breast Cancer Research Foundation T. Xu et al. ADDITIONAL INFORMATION Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, Supplementary information The online version contains supplementary material adaptation, distribution and reproduction in any medium or format, as long as you give available at https://doi.org/10.1038/s41523-022-00472-4. appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party Correspondence and requests for materials should be addressed to Haojun Luo or material in this article are included in the article’s Creative Commons license, unless Keda Yu. indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory Reprints and permission information is available at http://www.nature.com/ regulation or exceeds the permitted use, you will need to obtain permission directly reprints from the copyright holder. To view a copy of this license, visit http:// creativecommons.org/licenses/by/4.0/. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. © The Author(s) 2022 Published in partnership with the Breast Cancer Research Foundation npj Breast Cancer (2022) 100 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png npj Breast Cancer Springer Journals

High GPER expression in triple-negative breast cancer is linked to pro-metastatic pathways and predicts poor patient outcomes

Download PDF
11 pages

Loading next page...
 
/lp/springer-journals/high-gper-expression-in-triple-negative-breast-cancer-is-linked-to-pro-PWN15C0AAN

References (68)

Publisher
Springer Journals
Copyright
Copyright © The Author(s) 2022
eISSN
2374-4677
DOI
10.1038/s41523-022-00472-4
Publisher site
See Article on Publisher Site

Abstract

www.nature.com/npjbcancer ARTICLE OPEN High GPER expression in triple-negative breast cancer is linked to pro-metastatic pathways and predicts poor patient outcomes 1,8 2,8 2 3 4 5 6 1 7 Ting Xu , Ding Ma , Sheng Chen , Rui Tang , Jianling Yang , Chunhui Meng , Yang Feng , Li Liu , Jiangfen Wang , 6 2 ✉ ✉ Haojun Luo and Keda Yu Triple-negative breast cancer (TNBC) is a particularly aggressive and heterogeneous disease with few effective targeted therapies and precision therapeutic options over a long period. It is generally considered that TNBC is an estrogen-independent breast cancer, while a new estrogen receptor, namely G protein-coupled estrogen receptor (GPER), is demonstrated to mediate estrogenic actions in TNBC. Based on our transcriptomic analysis, expression of GPER was correlated with clinicopathological variables and survival of 360 TNBC patients. GPER expression at mRNA level was significantly correlated with immunohistochemistry scoring in 12 randomly chosen samples. According to the cutoff value, 26.4% (95/360) of patients showed high GPER expression and significant correlation with the mRNA subtype of TNBC (P = 0.001), total metastatic events (P = 0.019) and liver metastasis (P = 0.011). In quantitative comparison, GPER abundance is correlated with the high-risk subtype of TNBC. At a median follow-up interval of 67.1 months, a significant trend towards reduced distant metastasis-free survival (DMFS) (P = 0.014) was found by Kaplan–Meier analysis in patients with high GPER expression. Furthermore, univariate analysis confirmed that GPER was a significant prognostic factor for DMFS in TNBC patients. Besides, high GPER expression was significantly linked to the worse survival in patients with lymph node metastasis, TNM stage III as well as nuclear grade G3 tumors. Transcriptome-based bioinformatics analysis revealed that GPER was linked to pro-metastatic pathways in our cohort. These results may supply new insights into GPER-mediated estrogen carcinogenesis in TNBC, thus providing a potential strategy for endocrine therapy of TNBC. npj Breast Cancer (2022) 8:100 ; https://doi.org/10.1038/s41523-022-00472-4 INTRODUCTION either by suppressing ER-α activity or by inhibiting estrogen production, is central to the multidisciplinary management of Triple-negative breast cancer (TNBC) accounts for 10–15% of all patients with breast cancer, based not only on the significant breast cancers and is characterized by the lack of expression of the effectiveness but also on the convenience and safety of these estrogen receptor α (ER-α), the progesterone receptor (PR), and 6,7 1,2 agents . However, estrogen carcinogenesis, as well as endocrine the human epidermal growth factor receptor 2 (HER2) . When therapy, were long neglected in TNBC, logically due to the compared with other subtypes of breast cancer, TNBC exhibits the absence of ER-α. It has been recently shown that alternative ERs most aggressive course and the highest rate of early distant including G protein-coupled estrogen receptor (GPER), ER-β, and recurrence, especially in the lungs and brains, both of which 3,4 ER-α36 (a variant of ER-α) can trigger estrogen-responsivity in predict death in the short term . Notably, TNBCs occur more 8,9 TNBC , which leads to a growing concern. frequently in younger patients with its percentage increasing to 1,2 The identification of GPER, also known as GPR30, which was 25–30% in patients under 50 . Concerning systemic therapy, recognized as a membrane-associated receptor binding E2 with chemotherapy remains the standard management for TNBC while high affinity to mediate rapid and nongenomic estrogenic effects, targeting therapy is still at its early stage. Theoretically, the most including transactivation of epidermal growth factor receptor and effective strategy to improve patient outcomes may be by production of second messengers such as cAMP, calcium and blocking the metastatic process. inositol triphosphate, has challenged the traditional concept Estrogen, predominantly 17β-estradiol (E2), is a critical driver of 10–13 stating that TNBC was estrogen-independent . GPER is a mammary development and an essential etiological factor for breast cancer. As biologic mediators of estrogenic effects, ERs are member of G protein-coupled receptors (GPCRs), whose biological widely distributed in breast cancer. Specifically, ER-α, which is activity is dictated by posttranscriptional modifications (such as detected in about 70% of breast cancers, is used as not only a phosphorylation and ubiquitination) that control receptor con- centrations at the plasma membrane . In our earlier reports and powerful prognostic factor but also an efficient target for patients . The endocrine therapy that blocks estrogen signaling, others, GPER was detected in more than 60% of primary TNBC Department of Endocrine and Breast Surgery, The First Affiliated Hospital of Chongqing Medical University, 1 Youyi Road, Chongqing 400010, People’s Republic of China. Department of Breast Surgery, Precision Cancer Medicine Center, Fudan University Shanghai Cancer Center, 270 Dong’an Road, Shanghai 200032, People’s Republic of China. 3 4 Key Laboratory of Laboratory Medical Diagnostics, Chinese Ministry of Education, Chongqing Medical University, Chongqing 400016, People’s Republic of China. Department of Thyroid and Breast Surgery, Weihai Municipal Hospital, 70 Heping Road, Huancui District, Weihai, Shandong 264200, People’s Republic of China. Department of Thyroid and Breast Surgery, Heze Municipal Hospital, 2888 Caozhou West Road, Heze, Shandong 274031, People’s Republic of China. Department of Thyroid and Breast Surgery, The Second Affiliated Hospital of Chongqing Medical University, 74 Linjiang Road, Chongqing 400010, People’s Republic of China. Department of General Surgery, Shanxi Provincial People’s Hospital, Taiyuan, Shanxi 030000, People’s Republic of China. These authors contributed equally: Ting Xu, Ding Ma. email: luohaojun@hospital.cqmu.edu.cn; Yukeda@fudan.edu.cn Published in partnership with the Breast Cancer Research Foundation 1234567890():,; T. Xu et al. 15–20 samples and several TNBC cell lines . Importantly, GPER was Table 1. Tumor characteristics and GPER distribution. linked to the metastatic behaviors of TNBC cells in vitro and Variables No. (%)/Mean ± SD in vivo. Ligands including E2 and bisphenol A were claimed to trigger GPER to promote migration and metastasis of TNBC Total GPER-Low GPER-High P 21–23 cells , although the designation of GPER as a cognate ER is still (N = 360) (N = 265) (N = 95) 24–27 debated sometimes . In a recent bioinformatics analysis, GPER was also correlated with pro-metastatic genes and pathways in ER- Age 53.3 ± 11.4 53.4 ± 11.3 52.8 ± 11.5 0.624 α negative breast cancer . Considering the advantages of Menopause status 0.832 endocrine therapy, GPER has been included as a candidate Pre-menopause 132 (36.7) 98 (74.2) 34 (25.8) biomarker and a potential therapeutic target for TNBC . However, Post-menopause 224 (62.2) 164 (73.2) 60 (26.8) a controversial report concludes that GPER activation could inhibit Tumor size (pT) 2.64 ± 1.17 2.64 ± 1.25 2.66 ± 0.92 0.891 the in vivo invasive potential of TNBC via suppression of epithelial- mesenchymal transformation . Thus, the role of GPER in TNBC pT1 131 (36.4) 100 (76.3) 31 (23.7) metastasis needs further confirmation. pT2 219 (60.8) 156 (71.2) 63 (28.8) As an alternative ER, GPER has caught increasing attention in pT3 9 (2.5) 8 (88.9) 1 (11.1) breast cancer research, and the relationship between GPER and LNM (pN) 0.426 breast cancer outcomes has been addressed in multiple 30–34 pN0 209 (58.1) 158 (75.6) 51 (24.4) studies . Controversial findings on the prognostic role of GPER as well as the association between GPER expression and pN1 97 (26.9) 70 (72.2) 27 (27.8) clinicopathological determinants of breast cancer have been pN2 32 (8.9) 22 (68.8) 10 (31.2) reported. For instance, GPER was linked to worse relapse-free pN3 17 (4.7) 10 (58.8) 7 (41.2) survival (RFS) in breast cancer patients treated with tamoxifen . TNM stage 0.513 Meanwhile, GPER was also correlated with an increased distant I 89 (24.7) 70 (78.7) 19 (21.3) disease-free survival (DDFS) of ER-α positive breast cancer . The IIA 144 (40.0) 105 (72.9) 39 (27.1) biological functions of GPER largely depend on the cellular 12,13 background in vitro . In patients with TNBC, an early report IIB 76 (21.1) 56 (73.7) 20 (26.3) provided a clue that GPER might be associated with a poorer IIIA 34 (9.4) 24 (70.6) 10 (29.4) prognosis with a trend towards increased recurrences (without IIIC 17 (4.7) 10 (58.8) 7 (41.2) statistical significance, n = 18) . Recently, high expression of Histological 0.182 GPER was related to decreased outcomes in TNBC, including the IDC 330 (91.7) 246 (74.5) 84 (25.5) local RFS, DDFS, overall survival (OS), and progression-free survival (PFS) in a Chinese cohort (n = 249) . A bioinformatics analysis also Others 30 (8.3) 19 (63.3) 11 (36.7) associated the high expression of GPER with the decreased Nuclear grade 0.304 disease-free interval in ER-α negative breast cancer, although the 2 93 (25.8) 65 (69.9) 28 (30.1) scale of the cohort was limited (n = 120) . Thus, the prognostic 3 232 (64.4) 175 (75.4) 57 (24.6) significance of GPER in TNBC needs to be evaluated in larger unknown 35 (9.7) cohorts. Necrosis 0.027* We have reported the largest single-center study concerning the multi-omics profiling of TNBC, delineating the genomic and No 151 (41.9) 122 (80.8) 29 (19.2) transcriptomic landscape of Chinese TNBC patients . In this Yes 113 (31.4) 78 (69.0) 35 (31.0) cohort, 360 cases had RNA sequencing data on primary tumor Ki67 52.8 ± 25.3 51.7 ± 25.2 55.8 ± 25.5 0.180 tissue. To further evaluate the prognostic role of GPER, especially Intrinsic subtype 0.755 on metastatic manifestations in these patients, we analyzed the Basal like 277 (76.9) 205 (74.0) 72 (26.0) correlation between GPER expression and clinicopathological determinants of TNBC progression and long-term survival herein. Others 83 (23.1) 60 (72.3) 23 (27.7) We also show a bioinformatics analysis based on the transcrip- mRNA subtype 0.001* tomic profiles of our cohort to better understand the estrogenic IM 87 (24.2) 77 (88.5) 10 (11.5) carcinogenesis mediated by GPER in this aggressive breast cancer LAR 81 (22.5) 59 (72.8) 22 (27.2) subtype. BLIS 139 (38.6) 97 (69.8) 42 (30.2) MES 53 (14.7) 32 (60.4) 21 (39.6) RESULTS Recurrence 0.097 Patient characteristics No 300 (83.3) 226 (75.3) 74 (24.7) 360 patients who underwent surgery at Fudan University Yes 60 (16.7) 39 (65.0) 21 (35.0) Shanghai Cancer Center (FUSCC) between 2007 and 2014 were Metastatic events 0.019* included in this study (Table 1). The mean age of all patients was No 310 (86.1) 235 (75.8) 75 (24.2) 53.3 ± 11.4 years old (range, 25–84 years old) and 62.2% of them Yes 50 (13.9) 30 (60.0) 20 (40.0) were post-menopause. Most tumors were pT2 (60.8%), without lymph node metastasis (LNM) (58.1%), TNM stage II (61.1%), Lung 1.000 invasive ductal carcinoma (91.7%), nuclear grade 3 (64.4%), and No 344 (95.6) 253 (73.5) 91 (26.5) within the basal-like intrinsic subtype (76.9%). As for this cohort, Yes 16 (4.4) 12 (75.0) 4 (25.0) we classified the tumors into four transcriptome-based subtypes: Bone 0.210 luminal androgen receptor (LAR) subtype (22.5%), immunomodu- No 339 (94.2) 252 (74.3) 87 (25.7) latory (IM) subtype (24.2%), basal-like immune-suppressed (BLIS) Yes 21 (5.8) 13 (61.9) 8 (38.1) subtype (38.6%) and mesenchymal-like (MES) subtype (14.7%). Distant metastases were excluded present at the time of surgery. No patients received any systemic adjuvant therapy besides npj Breast Cancer (2022) 100 Published in partnership with the Breast Cancer Research Foundation 1234567890():,; T. Xu et al. staining. In some nests, weak staining was localized in the core Table 1 continued area while strong staining was observed at the margin (Fig. 1d). Variables No. (%)/Mean ± SD Association between GPER and clinicopathological variables Total GPER-Low GPER-High P (N = 360) (N = 265) (N = 95) of TNBC According to the cutoff value of RNA sequencing results, low and Brain 0.697 high GPER expression levels were detected in 73.6% (265/360) and No 354 (98.3) 261 (73.7) 93 (26.3) 26.4% (95/360) of patients, respectively (Table 1). The association Yes 6 (1.7) 4 (66.7) 2 (33.3) of GPER expression with clinicopathological variables was assessed. High GPER expression was significantly correlated with Liver 0.011* necrosis in the cancer nest (P = 0.027) and mRNA subtype No 348 (96.7) 260 (74.7) 88 (25.3) (P = 0.001) of TNBC according to our classification . Compared Yes 12 (3.3) 5 (41.7) 7 (58.3) to the GPER-low group, the GPER-high group demonstrated Contralateral supraclavicular lymph node metastasis 0.447 increased incidences of total metastatic events (21.1% vs. 11.3%; No 358 (99.4) 264 (73.7) 94 (26.3) P = 0.019) and liver metastasis (7.4% vs. 1.9%; P = 0.011) in the follow-up. Other clinicopathological variables, such as age, Yes 2 (0.6) 1 (50.0) 1 (50.0) menopausal status, and tumor size, had no significant correlations Death 0.190 with GPER expression. No 320 (88.9) 239 (74.7) 81 (25.3) Yes 40 (11.1) 26 (65.0) 14 (35.0) The distribution of GPER is different among subtypes of TNBC Chemotherapy 0.919 As mentioned, tumors in this cohort were classified into four No 10 (2.7) 8 (80.0) 2 (20.0) transcriptome-based subtypes and the distribution of GPER Yes 350 (97.2) 257 (73.4) 93 (26.6) 0.644 expression was significantly different among subtypes of TNBC Anthracycline and 214 (59.4) 159 (74.3) 55 (25.7) (Table 1). Thus, we quantitatively compared the abundance of taxane-based GPER among subtypes of TNBC by ANOVA analysis. Intriguingly, regimens the abundance of GPER was the lowest in the IM subtype which Others 136 (37.7) 98 (72.1) 38 (27.9) presented the best prognosis. Respectively, GPER abundance was highest in the MES subtype which presented the worst prognosis Bone 0.210 among all TNBC subtypes (Fig. 2). A significant difference was No 339 (94.2) 252 (74.3) 87 (25.7) found between LAR and IM (P = 0.042), BILS and IM (P = 0.002), Yes 21 (5.8) 13 (61.9) 8 (38.1) MES and IM (P < 0.001) as well as MES and LAR (P = 0.028). This Brain 0.697 may suggest that GPER is correlated with a higher risk subtype of No 354 (98.3) 261 (73.7) 93 (26.3) TNBC. Yes 6 (1.7) 4 (66.7) 2 (33.3) High expression of GPER predicted worse DMFS in TNBC Liver 0.011* patients No 348 (96.7) 260 (74.7) 88 (25.3) Survival outcomes were analyzed to explore the potential of GPER Yes 12 (3.3) 5 (41.7) 7 (58.3) as a survival predictor. Kaplan–Meier analysis of this cohort Data are expressed as the patient number (%) or mean ± SD. Statistically revealed a trend towards reduced RFS (Log Rank P = 0.078), DMFS significant differences were defined as P < 0.05. (Log Rank P = 0.014), and OS (Log Rank P = 0.136) in patients with LNM lymph node metastasis, IDC invasive ductal carcinoma, IM immuno- high GPER expression TNBCs. Note, that statistical significance is modulatory subtype, LAR luminal androgen receptor subtype, BLIS basal- only presented in DMFS (Fig. 3). like immune-suppressed subtype, MES mesenchymal-like subtype. Additionally, a Cox proportional hazard regression model was Includes 25 cases who received additional platinum regimens. b used to identify biomarkers and clinicopathological factors Consists of the following: single-agent taxane (n = 22), single-agent affecting the prognosis of patients with TNBC (Table 2). The anthracycline (n = 39), single-agent platinum (n = 2), combination of taxane and platinum (n = 40), and unknown agents (n = 33). univariate analysis confirmed that GPER was a significant prognostic factor for DMFS (hazard ratios (HR) = 2.01; 95% confidence interval (CI), 1.14–3.54; P = 0.016) in TNBC patients. chemotherapy. The median follow-up interval was 67.1 months (range However, the further multivariate analysis failed to provide 0.3–144.2 months). At the time of analysis, 60 patients underwent evidence for GPER (HR = 1.642; 95% CI, 0.92–2.92; P = 0.091) as recurrence, 50 patients had metastatic events and 40 patients died; an independent prognostic factor. Additionally, the prognostic the RFS was 83.3%, DMFS was 86.1% and OS was 88.9%. value of GPER was not significant in RFS and OS neither by univariate nor multivariate analysis. Referring to the other The expression level of GPER was correlated with the clinicopathological variables, LNM status, and TNM stage were immunohistochemistry (IHC) score in TNBC tissues identified as prognostic indicators for RFS, DMFS, and OS in the univariate model, while mRNA subtype was only associated with In our cohort, GPER expression at the mRNA level, shown as log2 DMFS. Multivariate analysis proved that TNM stage was a (FPKM+ 1) expression value, was of normal distribution in TNBC significant independent prognostic factor for RFS, DMFS, and OS tissues (Supplementary Fig. 1). To verify the expression of GPER in TNBC patients. Besides, LNM status was also suggested as an detected by RNA sequencing, samples from 12 patients were independent prognostic factor for RFS and DMFS, but not OS. randomly selected for IHC staining using an antibody against GPER. As expected, we found varying staining intensities of GPER in these tissues (Fig. 1a) and the IHC scoring was significantly correlated High GPER expression predicted worse survival in high-risk with the log2 expression value of GPER (Fig. 1b). Although both TNBC patients cytoplasmic and membrane staining of GPER were reported, we Intriguingly, the prognostic value of GPER, not only for DMFS but observed, even by oil lens, only cytoplasmic patterns in these also for RFS and OS, dramatically increased when stratifying for samples (Fig. 1c). Interestingly, we observed heterogeneity in GPER known risk factors. Revealed by Kaplan–Meier analysis, high Published in partnership with the Breast Cancer Research Foundation npj Breast Cancer (2022) 100 T. Xu et al. Fig. 1 Immunohistochemical staining for GPER. a Immunohistochemical staining for GPER with different intensity. The staining is graded as negative (−), weak (+), moderate (++), and strong (+++). Scale bar, 100 μm. b Scatter plots show that the IHC score is linearly correlated with the log2 expression value of GPER in the results of Pearson correlation coefficient calculation, which is statistically significant. Pearson correlation coefficient R and p-value are given in the scatter plot. c Image showing IHC staining of GPER is only observed in cytoplasmic. Scale bar, 50 μm. d Representative image of weak GPER staining in the core area and strong GPER staining in the corresponding margin. Scale bar, 100 μm. GPER was linked to pro-metastatic pathways in the transcriptomic landscape of Chinese TNBC patients To better understand the estrogenic carcinogenesis mediated by GPER in TNBC, we applied bioinformatics analysis based on the transcriptomic profiles of 360 TNBC patients in our cohort. The gene set enrichment analysis (GSEA) and gene set variation analysis (GSVA) were performed between the GPER-high and GPER-low groups of TNBC patients. Based on the results of GSEA analysis, the dot plot shows the significantly enriched GPER- related pathways (Fig. 5a). Of note, the enriched pathways with pro-metastatic characteristics included Focal adhesion, WNT signaling pathway, ECM receptor interaction, NOTCH signaling pathway, Hedgehog signaling pathway, Adherens junction path- way, and TGF beta signaling pathway, as indicated by their respective adjusted p-values and GSEA-plots (Fig. 5a, b). Addi- tionally, the GSVA analysis was conducted using KEGG gene sets. Firstly, 186 KEGG pathways were quantified using the GSVA package. Then, differential analysis was conducted to find specific pathways for GPER-high and GPER-low groups. Similar to the GSEA results, the GSVA results also showed that the pro-metastasis Fig. 2 Violin diagram showing the distribution of GPER expres- pathways that are significantly enriched in the GPER-high group sion among subtypes of TNBC. Statistically significant differences include NOTCH signaling pathway, Hedgehog signaling pathway, were defined as P < 0.05. Abbreviations: IM immunomodulatory WNT signaling pathway, and Adherens junction pathway (Fig. 5c). subtype; LAR luminal androgen receptor subtype; BILS basal-like immune-suppressed subtype; MES mesenchymal-like subtype. DISCUSSION expression of GPER was linked significantly to the worse RFS (Log TNBC has the worst prognosis of all breast cancer subtypes and Rank P = 0.012), DMFS (Log Rank P = 0.003), and OS (Log Rank the lack of well-defined molecular targets is the main challenge to P = 0.012) in LNM (+) patients while no difference was found in 1,2 treat TNBC patients . Although estrogens largely contribute to LNM (−) patients (Fig. 4a and Supplementary Fig. 2). Referring to the development and progression of breast cancer, estrogen stage III patients, high GPER expression correlated with lower RFS carcinogenesis was long disregarded in TNBC. In the present (Log Rank P = 0.013) and DMFS (Log Rank P = 0.014), and a trend study, we revealed that GPER expression was associated with the toward correlating with OS (Log Rank P = 0.132) (Fig. 4b and aggressive subtype of TNBC. High GPER expression predicted Supplementary Fig. 2). Similarly, high GPER expression is also reduced DMFS in our cohort. Especially in high-risk patients with associated with reduced RFS (Log Rank P = 0.045) and DMFS (Log G3 tumors, LNM (+) or stage III, the prognostic significance was Rank P = 0.008), and a trend toward OS (Log Rank P = 0.074) in increased. Transcriptome-based bioinformatics analysis revealed patients with G3 tumors (Fig. 4c and Supplementary Fig. 2). Thus, that GPER was linked to pro-metastatic pathways in the Chinese the predictive value of GPER seems to be increased in TNBC cohort of TNBC. patients with additional risk factors, including LNM positivity, Theoretically, GPER, as a membrane receptor belonging to the higher nuclear grade and later TNM stage. GPCR superfamily, is significantly different from ER-α, a nuclear npj Breast Cancer (2022) 100 Published in partnership with the Breast Cancer Research Foundation T. Xu et al. Fig. 3 Kaplan–Meier survival curves of RFS, DMFS, and OS between GPER-low and GPER-high groups in patients with TNBC. Statistically significant differences were defined as P < 0.05. Abbreviations: RFS relapse-free survival; DMFS distant metastasis-free survival; OS overall survival. steroid hormone receptor. In biology, the differences between development and progression remains elusive. In an early study, GPER and ER-α include their subcellular distribution, structure, GPER was detected in every single breast sample from 12 healthy affinity to E2, ligands pattern, the process and effects in response donors . The GPER expression in normal breast tissue was at a to E2. These differences have attracted a surge of interest and are medium level in the Human Protein Atlas database .In 12,13,24 well-reviewed in the literature . Meanwhile, GPER is not comparison, GPER expression was reported to decrease in tumor 24–27 52 totally accepted as a cognate ER in related debates . GPER was tissues ,while inflammatory breast cancer, an aggressive type of reported to coimmunoprecipitate with ER-α in MCF-7 cells and to breast cancer, exhibited stronger intensity in staining against repeatedly correlate with ER-α positivity in primary breast GPER . Interestingly, GPER expression was correlated with the 30,32,37 cancers , thus it may contribute to estrogenic responses as tumor subtype in a large cohort and strong staining was a collaborator. GPER didn’t even respond to stimulation of E2 and significantly more prevalent among TNBCs .Notably, whenTNBC G1 in some cell models . However, GPER was detected in tissues subtypes were classified by transcriptomic profile, the expression of and cell lines of not only breast cancer but also other organs GPER correlated with the subtype, and the highest level was found 12,13 lacking ER-α expression . Furthermore, GPER was observed to in the MES subtype, which presented the worst prognosis in all 10,39,40 bind E2 directly in several cell lines lacking ER-α . In clinical TNBC subtypes. In addition, weak positivity was localized in the core series, inverse correlation or non-significant association was also area while strong positivity was observed at the margin in some found between GPER and ER-α . Due to the lack of ER-α, it was nests by IHC staining, implying that GPER abundance may be assumed that TNBC is estrogen-insensitive. However, increasing increased during the invasion of cancer cells. It also implied a link circulating estrogen levels were sufficient to promote the between GPER and the aggressiveness of the breast cancer, that is, formation and progression of ER-α negative cancers including high expression of GPER was associated with more frequent TNBC and pharmacological inhibition of estrogen synthesis after necrosis in tumor sections. Besides, metastatic or recurrent cancer pregnancy prevented the formation of ER-α negative tumors . tumors also showed higher levels of GPER expression than 19,31,53 Furthermore, we found that the mRNA expression level of GPER, corresponding primary tumor .Meanwhile, the aggressive which is positively correlated with the staining score of the GPER cell lines of uterine and ovarian cancer (JEG and Hec50) expressed a protein (R² = 0.7603), is widely distributed in this large Chinese much higher level of GPER than their associated normal cell lines cohort of TNBC tissues, in line with earlier detection of GPER in (HTR8 and H, respectively) . In general, GPER expression likely 15–20 tissues and cell lines of TNBC . Taken together, these results increases with the development and progression while it indicates indicate a functional role of GPER as an alternative ER and the the natural aggressiveness of breast cancer. potential as a mediator of estrogen carcinogenesis in TNBC. The prognostic value of GPER for breast cancer patients has 30–34 However, further identification, especially based on GPER protein been addressed in multiple studies . However, controversial expression, is needed. findings have been reported, even in meta-analysis . The There are some disputes about the subcellular localization of biological function of GPER largely depends on the cellular 12,13 GPER. Known as a GPCR, GPER can be detected on the cell surface background in vitro , thus the prognostic role of GPER should 2+ and is involved in signal transduction events, such as Ca be evaluated in a specific subtype of breast cancer. In this respect, 10 43 44 mobilization , NO generation , ERK activation , and growth the contribution of GPER should be genuine in TNBC. In this large factor release . GPER was also observed to locate in the cohort of patients, high GPER expression was linked to poor DMFS endoplasmic reticulum and to act as an endoplasmic reticulum and exhibited prognostic significance, especially in high-risk stressor that induces growth inhibition and apoptotic cell patients with G3 tumors, LNM (+) and stage III. These results death . GPER distribution in mitochondria, Golgi apparatus, and were consistent with earlier reports as follows: In an earlier report nucleus also reported . The different subcellular locations of the with a small cohort of TNBC (n = 18), GPER was thought to have a GPER may have different biological implications. Importantly, our poorer prognosis with a trend towards increased recurrences . group has found that GPER exists in the nucleus and is High GPER expression was also related to decreased outcomes, translocated from the nucleus to the cytoplasm under including the local RFS, DDFS, OS, and PFS in a Chinese cohort 49 17 E2 stimulation . In this study, GPER staining only exhibited in (n = 249) . In a retrospective TNBC study (n = 199), GPER and the cytoplasm may be due to the small sample size, different estrogen-related receptor α (ERR-α) synergistically predicted poor functional status of cells, and less sensitive detection methods. patient outcomes . In summary, studies seem to agree that GPER Physiologically, the expression pattern of GPER is likely to be can mediate estrogen carcinogenesis and promote the progres- tissue-dependent and developmentally regulated. In the mam- sion of TNBC. Hence, GPER could be of significant prognostic mary ductal epithelia, GPER abundance was varied with the value. However, a larger cohort and longer follow-up are needed estrous cycle . Yet, the trend of GPER expression in breast cancer to address this issue. Published in partnership with the Breast Cancer Research Foundation npj Breast Cancer (2022) 100 T. Xu et al. Table 2. Univariate and multivariate survival analyses of clinicopathological factors and GPER expression. Univariate model Multivariate model P HR β 95% CI for HR P HR β 95% CI for HR RFS Age (<50 vs. >=50) 0.106 0.658 −0.418 0.396–1.093 Menopause status (pre-menopause vs. post-menopause) 0.547 0.853 −0.159 0.509–1.430 Nuclear grade (G2 vs. G3) 0.109 0.633 −0.457 0.362–1.107 Tumor size (pT1 vs. pT2/3) 0.332 1.309 0.270 0.760–2.256 LNM (positive vs. negative) <0.001 3.474 1.245 2.015–5.987 0.018 2.185 0.782 1.147–4.165 TNM stage (III vs. I/II) <0.001 4.763 1.561 2.828–8.022 0.001 2.963 1.086 1.599–5.493 Necrosis (positive vs. negative) 0.434 1.270 0.239 0.698–2.309 mRNA subtype (IM vs. others) 0.088 0.540 −0.616 0.266–1.097 Intrinsic subtype (Basal vs. others) 0.565 1.183 0.168 0.667–2.096 Histology (IDC vs. others) -- - - Ki67 (>=30% vs. <30%) 0.425 1.305 0.266 0.678–2.514 GPER (high vs. low) 0.081 1.604 0.473 0.944–2.727 DMFS Age (<50 vs. >=50) 0.178 0.682 −0.383 0.391–1.190 Menopause status (pre-menopause vs. post-menopause) 0.388 0.781 −0.247 0.445–1.370 Nuclear grade (G2 vs. G3) 0.392 0.760 −0.275 0.405–1.425 Tumor size (pT1 vs. pT2/3) 0.395 1.294 0.258 0.714–2.344 LNM (positive vs. negative) <0.001 5.020 1.613 2.622–9.610 0.005 3.005 1.100 1.403–6.432 TNM stage (III vs. I/II) <0.001 6.490 1.870 3.717–11.334 0.001 3.132 1.142 1.622–6.046 Necrosis (positive vs. negative) 0.116 1.721 0.543 0.874–3.388 mRNA subtype (IM vs. others) 0.022 0.340 −1.080 0.135–0.856 0.082 0.432 −0.840 0.167–1.113 Intrinsic subtype (Basal vs. others) 0.684 1.140 0.131 0.606–2.145 Histology (IDC vs. others) 0.993 0.995 −0.005 0.358–2.765 Ki67 (>=30% vs. <30%) 0.271 1.530 0.425 0.717–3.264 GPER (high vs. low) 0.016 2.009 0.698 1.141–3.538 0.091 1.642 0.496 0.923–2.920 OS Age (<50 vs. >=50) 0.489 0.802 −0.221 0.428–1.501 Menopause status (pre-menopause vs. post-menopause) 0.828 1.075 0.072 0.561–2.058 Nuclear grade (G2 vs. G3) 0.804 0.913 −0.091 0.447–1.866 Tumor size (pT1 vs. pT2/3) 0.775 1.100 0.095 0.574–2.106 LNM (positive vs. negative) 0.001 3.225 1.171 1.663–6.252 0.197 1.720 0.542 0.754–3.924 TNM stage (III vs. I/II) <0.001 5.445 1.695 2.907–10.200 0.001 3.871 1.354 1.772–8.457 Necrosis (positive vs. negative) 0.192 1.658 0.506 0.776–3.543 mRNA subtype (IM vs. others) 0.204 0.570 −0.562 0.239–1.358 Intrinsic subtype (Basal vs. others) 0.569 0.798 −0.225 0.368–1.732 Histology (IDC vs. others) 0.500 1.632 0.490 0.394–6.764 Ki67 (>=30% vs. <30%) 0.780 1.117 0.111 0.515–2.425 GPER (high vs. low) -- - - Covariates with P < 0.05 in the univariate model were included in the further multivariate model. Statistically significant differences were defined as P < 0.05. Covariates that did not satisfy the proportional hazards assumption with Schoenfeld residuals test. As aforementioned, early metastasis is underlying the poor Focal adhesion and ECM receptor interaction pathways, both of 2,3 prognosis of patients with TNBC . Interestingly, GPER was linked which were deeply involved in cancer invasion and metastasis, to the metastatic behaviors of TNBC cells in vitro and were also the most significant GPER-related pro-metastatic path- 16,18,19,21,56 in vivo . Ligands including E2 and bisphenol A were ways in the aforementioned study. Actually, estrogenic GPER claimed to trigger GPER to promote migration and metastasis of signaling was ascertained to trigger focal adhesion kinase 21–23 TNBC cells . E2 also induced the up-regulation of estrogen- phosphorylation to increase focal adhesion points and cellular related receptor α expression via GPER activation, enhanced the migration in TNBC cells . Other pro-metastatic pathways includ- migration and invasion of TNBC cells . Increased GPER expression ing NOTCH signaling, Hedgehog signaling, WNT signaling, and was observed at the invasive margin in some nests, implying that Adherens junction were also correlated with GPER in our analysis. GPER activation may contribute to the invasion of cancer cells. A However, the GPER-related pro-metastatic pathways identified by recent bioinformatics analysis showed that GPER was correlated Maggiolini et al. were somewhat different with those in this study, with pro-metastatic pathways in ER-α negative breast cancer by maybe due to the differences of tumor subtype (ER-α negative vs. Maggiolini et al. . We conducted a similar analysis in our large TNBC) and race of patients (mixed vs. Chinese). To our knowledge, single-center study and, as expected, GPER was correlated with performing the analysis in homogeneous TNBC cohort is multiple pro-metastatic pathways. Among these pathways, the significant since heterogeneity of TNBC is high enough and npj Breast Cancer (2022) 100 Published in partnership with the Breast Cancer Research Foundation T. Xu et al. Fig. 4 Kaplan–Meier curves between GPER-low and GPER-high groups in high-risk patients. a Kaplan–Meier curves of RFS, DMFS and OS in LNM (+) patients. b Kaplan–Meier curves of RFS, DMFS and OS in stage III patients. c Kaplan–Meier curves of RFS, DMFS and OS in patients with G3 tumors. Statistically significant differences were defined as P < 0.05. excluding the effect of crosstalk between estrogenic GPER GPER expression and pro-metastatic pathways was verified, signaling and HER2 signaling is necessary. Anyway, given that suggesting that GPER has significant functional roles in TNBC GPER mediates the transcriptional regulation of estrogen in TNBC metastasis. Taken together, this may provide evidence that GPER 19,57 cells and other breast cancer cell lines , GPER may trigger mediates metastatic estrogen carcinogenesis in TNBC. Considering variable signal transduction events to enhance the multi-steps of the great urgency for clinicians and researchers to develop metastasis. However, to better understand the functional role of efficient molecular targets and biomarkers, GPER could be a GPER, the mechanisms by which GPER contributes to the promising candidate for TNBC therapy and diagnosis. progression of TNBC need further clarification. Accordingly, new endocrine therapy by blocking GPER-related signaling may be an METHODS effective strategy for TNBC. Since several antagonists of GPER have been synthesized, employing them as endocrine therapy agents is Patient recruitment accessible, while more basic and translational research is needed From January 1, 2007 to December 31, 2014, primary tumor tissue and to confirm this potency. blood samples were obtained from 504 consecutive female Chinese patients with TNBC treated at Fudan University Shanghai Cancer Center In summary, in a unique and large Chinese cohort of TNBC with (FUSCC). Among these patients, 279 had whole exome sequencing (WES) long-term follow-up, we evaluated the expression of GPER and data on primary tumor tissue and paired blood samples, 401 had copy- showed that GPER expression correlates with the subtype of TNBC, number alteration (CNA) data and 360 had RNA sequencing data on with a trend to increase the aggressiveness of tumors. We primary tumor tissue. 360 patients with RNA sequencing data were concluded that GPER has significant prognostic value in TNBC and enrolled in this study according to the following defined criteria: (1) female is significantly linked to the worse survival, especially in high-risk patients diagnosed with unilateral disease; (2) histologically confirmed the patients with LNM (+), G3 or stage III tumors. Furthermore, ER-α (−), PR (−), and HER2 (−) phenotype; (3) no evidence of distant bioinformatics analysis was performed based on the transcrip- metastasis at diagnosis; (4) sufficient frozen tissues available for further tomic profile of the TNBC cohort and the correlation between high research. The examination results for chest computed tomography (CT), Published in partnership with the Breast Cancer Research Foundation npj Breast Cancer (2022) 100 T. Xu et al. Fig. 5 The GSEA and GSVA results of GPER correlated pathways based on the transcriptomic data of 360 TNBC patients. a Dotplot showing the twelve most significantly upregulated pathways in GPER-high and GPER-low groups from GSEA results. In the group of high GPER expression, seven of the twelve upregulated pathways are correlated with promoting tumor metastasis. b GSEA-plots showing the upregulated pro-metastatic pathways in the GPER-high group. c Heatmap for the eight most significantly upregulated pathways in GPER-high group by GSVA. Statistically significant differences were defined as adjusted P-value < 0.05. bone scans, abdominal ultrasound, bilateral mammography, breast deposited in NCBI Sequence Read Archive, with accession number ultrasound, and/or magnetic resonance imaging (MRI) were collected to SRP157974. Four stable clusters, IM, LAR, MES, and BLIS were identified after ascertain no metastasis beyond breasts and axillary lymph nodes metastasis before the surgery. Ethical review and approval were waived analyzing the robustness of the classification using k-means clustering with the details available in our previous study . Our classification system, for this study, due to the data reported in this paper have been described named FUSCC, correlated well with the Lehmann/Pietenpol classification in our published article and deposited in the NCBI Sequence Read Archive (SRA: SRP157974). All patients provided written informed consent system. for data and tissue use. Clinicopathological data RNA sequencing and transcriptomic profiling The ER-α, PR, and HER2 status of the breast tumor samples were confirmed RNA sequencing data and transcriptomic profiling of 360 patients with by two experienced pathologists based on immunochemical analysis and TNBC from FUSCC were used in the current study. Detailed sample in situ hybridization. ER-α and PR status were classified as negative using a preparation, library preparation, sequencing, and raw data processing were cutoff of 1%, according to the American Society of Clinical Oncology/ 36 58 described in our earlier publication . The RNA sequencing data have been College of American Pathologists (ASCO/CAP) guidelines . HER2 status npj Breast Cancer (2022) 100 Published in partnership with the Breast Cancer Research Foundation T. Xu et al. was defined as negative with 0, 1+ as well as 2+ on immunohistochem- multivariate model. The proportional hazards assumption was tested by istry without HER2 gene amplification on fluorescence in situ hybridization using Schoenfeld residual tests. If the assumption of proportional hazards (FISH) . TNBC was defined as ER-α, PR, and HER2 negative in accordance was not valid, time-dependent covariates were introduced. All tests were with the St. Gallen International Expert Consensus . Clinicopathological two-sided and P < 0.05 was deemed statistically significant. features, including age at diagnosis, menopausal status, tumor histologic type, tumor size, LNM, histologic grade, TNM stage and ER-α, PR, HER2, and Reporting summary Ki67 status, were analyzed. The tumor stage based on the TNM stage was Further information on research design is available in the Nature Research assessed according to the criteria established by the 8th edition American Reporting Summary linked to this article. Joint Committee on Cancer (AJCC 8th) staging manual of breast cancer. Patient follow-up DATA AVAILABILITY Follow-up of all patients in this cohort was completed on June 11, 2019. RNA sequencing data that support the findings of this study have been deposited in The median length of follow-up was 67.1 months with an interquartile NCBI Sequence Read Archive with the accession codes SRP157974. All other relevant range of 53.9–79.9 months. RFS was defined as the time from diagnosis to data are available from the corresponding author on request. first recurrence or a diagnosis of contralateral breast cancer. DMFS was defined as the time from diagnosis to first distant metastasis. OS was defined as the time from diagnosis to death. Patients without events were CODE AVAILABILITY censored from the time point of the last follow-up. The code used to process and analyze the RNA sequencing data is available from the corresponding author upon reasonable request. Immunohistochemistry staining and scoring Immunohistochemistry staining was performed using an SP900 Kit Received: 8 December 2021; Accepted: 9 August 2022; (Zhongshan Golden Bridge) according to the manufacturer’s protocol. Briefly, deparaffinized tissue sections of 4 μm thickness were heated for antigen retrieval at 95 °C for 15 min in 10 mM citric acid buffer (pH 6.0). After treatment with 3% H O for 10 min to quench endogenous 2 2 peroxidase activity, the sections were blocked using goat serum and then REFERENCES incubated with the primary antibody targeting GPER (1:250, ab39742, 1. Al-Mahmood, S., Sapiezynski, J., Garbuzenko, O. B. & Minko, T. Metastatic and Abcam, USA) at a 1:200 dilution at 4 °C for 16 h. The section treated with triple-negative breast cancer: challenges and treatment options. Drug Deliv. PBS worked as a negative control. Following treatment of horseradish Transl. Res. 8, 1483–1507 (2018). peroxidase-conjugated goat anti-rabbit IgG for 30 min at 37 °C, sections 2. Garrido-Castro, A. C., Lin, N. U. & Polyak, K. Insights into molecular classifications were developed using diaminobenzidine (DAB) (Zhongshan Golden of triple-negative breast cancer: Improving patient selection for treatment. Bridge) and nuclei were counterstained with Mayer’s modified hematox- Cancer Discov. 9, 176–198 (2019). ylin. As indicated by Filardo et al. , reduction mammoplasty tissue was 3. Denkert, C., Liedtke, C., Tutt, A. & von Minckwitz, G. Molecular alterations in triple- used as a positive control. negative breast cancer—the road to new treatment strategies. Lancet 389, Two observers microscopically evaluated the intensity, extent and 2430–2442 (2017). subcellular distribution of GPER using a modified semi-quantitative scoring 4. Dent, R. et al. Triple-negative breast cancer: Clinical features and patterns of system. GPER scores were assigned as follows: the percentage of positive recurrence. Clin. Cancer Res. 13, 4429–4434 (2007). cells was categorized as 0 (negative staining in all cells), 1 (<1% cells 5. Dunnwald, L. K., Rossing, M. A. & Li, C. I. Hormone receptor status, tumor char- stained), 2 (1–10% of cells stained), 3 (11–30% cells stained), 4 (31–70% acteristics, and prognosis: a prospective cohort of breast cancer patients. Breast cells stained) or 5 (71–100% cells stained), and staining intensity was Cancer Res. 9, R6 (2007). categorized as 0 (negative), 1 (weak), 2 (moderate) and 3 (strong). Adding 6. Yuan, J. et al. Acquisition of epithelial-mesenchymal transition phenotype in the the two scores together yielded a maximum score of 8. The final scores tamoxifen-resistant breast cancer cell: A new role for G protein-coupled estrogen were grouped into GPER negative (score 0−4) and positive (score 5−8) receptor in mediating tamoxifen resistance through cancer-associated fibroblast- categories for statistical analyses to reduce inter-observer differences. derived fibronectin and β1-integrin signaling pathway in tumor cells. Breast Cancer Res. 17, 69 (2015). Bioinformatics analysis 7. Davies, C. et al. Relevance of breast cancer hormone receptors and other factors to the efficacy of adjuvant tamoxifen: Patient-level meta-analysis of randomised The 360 patients with TNBC were divided into 26.4% (95/360) patients with trials. Lancet 378, 771–784 (2011). high GPER expression and 73.6% (265/360) patients with low GPER 8. Treeck, O., Schüler-Toprak, S. & Ortmann, O. Estrogen actions in triple-negative expression. RNA sequencing data of 360 TNBC patients were used to breast cancer. Cells 9, 2358 (2020). determine the differences between groups through the R software limma 9. Wang, Z.-Y. & Yin, L. Estrogen receptor alpha-36 (ER-α36): A new player in human package and GSEA was applied to conduct gene enrichment analysis by breast cancer. Mol. Cell Endocrinol. 418, 193–206 (2015). using clusterProfiler package . GSVA analysis was performed on log2 10. Revankar, C. M., Cimino, D. F., Sklar, L. A., Arterburn, J. B. & Prossnitz, E. R. A (FPKM+ 1) expression values by using GSVA package . The “c2.cp.kegg.- transmembrane intracellular estrogen receptor mediates rapid cell signaling. v7.4.entrez.gmt” and “c2.cp.kegg.v7.2.symbols.gmt” gene sets were down- Science 307, 1625–1630 (2005). loaded from the Molecular Signatures Database . 11. Filardo, E. J. Epidermal growth factor receptor (EGFR) transactivation by estrogen via the G-protein-coupled receptor, GPR30: A novel signaling pathway with potential Statistical analysis significance for breast cancer. J. Steroid Biochem. Mol. Biol. 80,231–238 (2002). All statistical analyses were done by using the SPSS standard version 12. Hsu, L.-H., Chu, N.-M., Lin, Y.-F. & Kao, S.-H. G-Protein Coupled Estrogen Receptor 25 software and Stata version 13.0 software. For the division of high and in Breast Cancer. Int. J. Mol. Sci. 20, 306 (2019). low expression groups of GPER, the X-tile software was used to generate 13. Prossnitz, E. R. & Barton, M. The G-protein-coupled estrogen receptor GPER in the optimal cut-off value . Continuous quantitative data were expressed health and disease. Nat. Rev. Endocrinol. 7, 715–726 (2011). as mean ± standard deviation (SD) and categorical qualitative data as 14. Nelson, C. P. & Challiss, R. A. J. “Phenotypic” pharmacology: The influence of percentage. Data from the two patient groups were statistically compared cellular environment on G protein-coupled receptor antagonist and inverse using the chi-square test or t test as appropriate. The Pearson correlation agonist pharmacology. Biochem. Pharm. 73, 737–751 (2007). coefficient was used to analyze the correlation between IHC scoring and 15. Chen, Z. J. et al. Activation of GPER suppresses epithelial mesenchymal transition the log2 expression value of GPER. ANOVA test was used to determine the of triple negative breast cancer cells via NF-kappaB signals. Mol. Oncol. 10, differences of GPER expression among TNBC subtypes. RFS, DMFS, and OS 775–788 (2016). curves were drawn using the Kaplan–Meier methods and were compared 16. Yu, T. et al. GPER mediates enhanced cell viability and motility via non-genomic using Log-rank tests. Univariate and multivariate analyses of the patients’ signaling induced by 17beta-estradiol in triple-negative breast cancer cells. J. survival were performed using the Cox proportional hazards regression Steroid Biochem. Mol. Biol. 143, 392–403 (2014). model and the HRs with 95% CIs were calculated. Statistics of P < 0.05 in 17. Ye, S. et al. Prognostic role of GPER/Ezrin in triple-negative breast cancer is univariate analysis were used as inclusion criteria for covariates in the final associated with menopausal status. Endocr. Connect. 8, 661–671 (2019). Published in partnership with the Breast Cancer Research Foundation npj Breast Cancer (2022) 100 T. Xu et al. 18. Deng, Q. et al. GPER/Hippo-YAP signal is involved in Bisphenol S induced migration 46. Ahola, T. M., Manninen, T., Alkio, N. & Ylikomi, T. G protein-coupled receptor 30 is of triple negative breast cancer (TNBC) cells. J. Hazard. Mater. 355,1–9 (2018). critical for a progestin-induced growth inhibition in MCF-7 breast cancer cells. 19. Yin, J. et al. GPER-regulated lncRNA-Glu promotes glutamate secretion to Endocrinology 143, 3376–3384 (2002). enhance cellular invasion and metastasis in triple-negative breast cancer. FASEB J. 47. Song, L., De Sarno, P. & Jope, R. S. Central role of glycogen synthase kinase-3beta 34, 4557–4572 (2020). in endoplasmic reticulum stress-induced caspase-3 activation. J. Biol. Chem. 277, 20. Wang, Y. et al. NHERF1 inhibits proliferation of triple-negative breast cancer cells 44701–44708 (2002). by suppressing GPER signaling. Oncol. Rep. 38, 221–228 (2017). 48. Irannejad, R. et al. Functional selectivity of GPCR-directed drug action through 21. Rigiracciolo, D. C. et al. Focal adhesion kinase (FAK) activation by estrogens location bias. Nat. Chem. Biol. 13, 799–806 (2017). involves GPER in triple-negative breast cancer cells. J. Exp. Clin. Cancer Res. 38,58 49. Yu, T. et al. Cytoplasmic GPER translocation in cancer-associated fibroblasts (2019). mediates cAMP/PKA/CREB/glycolytic axis to confer tumor cells with multidrug 22. Xu, F. et al. Bisphenol A induces proliferative effects on both breast cancer cells resistance. Oncogene 36, 2131–2145 (2017). and vascular endothelial cells through a shared GPER-dependent pathway in 50. Wang, C., Prossnitz, E. R. & Roy, S. K. Expression of G protein-coupled receptor 30 hypoxia. Environ. Pollut. 231, 1609–1620 (2017). in the hamster ovary: differential regulation by gonadotropins and steroid hor- 23. Castillo-Sanchez, R. et al. Bisphenol A induces focal adhesions assembly and mones. Endocrinology 148, 4853–4864 (2007). activation of FAK, Src, and ERK2 via GPER in MDA-MB-231 breast cancer cells. 51. Uhlén, M. et al. Proteomics. Tissue-based map of the human proteome. Science Toxicol. Vitr. 66, 104871 (2020). 347, 1260419 (2015). 24. Luo, J. & Liu, D. Does GPER really function as a G protein-coupled estrogen 52. Ignatov, T. et al. GPER-1 expression decreases during breast cancer tumorigen- receptor? Front. Endocrinol. 11, 148 (2020). esis. Cancer Investig. 31, 309–315 (2013). 25. Levin, E. R. G Protein-coupled receptor 30: Estrogen receptor or collaborator. 53. Mo, Z. et al. GPR30 as an initiator of tamoxifen resistance in hormone-dependent Endocrinology 150, 1563–1565 (2009). breast cancer. Breast Cancer Res. 15, R114 (2013). 26. Olde, B. & Leeb-Lundberg, L. M. F. GPR30/GPER1: Searching for a role in estrogen 54. Yang, F. & Shao, Z. M. Double-edged role of G protein-coupled estrogen receptor physiology. Trends Endocrinol. Metab. 20, 409–416 (2009). 1 in breast cancer prognosis: An analysis of 167 breast cancer samples and online 27. Langer, G. et al. A critical review of fundamental controversies in the field of data sets. OncoTargets Ther. 9, 6407–6415 (2016). GPR30 research. Steroids 75, 603–610 (2010). 55. Ye, S. et al. Estrogen-related receptor α (ERRα) and G protein-coupled estrogen 28. Talia, M. et al. The G Protein-coupled estrogen receptor (GPER) expression cor- receptor (GPER) synergistically indicate poor prognosis in patients with triple- relates with pro-metastatic pathways in ER-negative breast cancer: A bioinfor- negative breast cancer. OncoTargets Ther. 13, 8887–8899 (2020). matics analysis. Cells 9, 622 (2020). 56. Marjon, N. A., Hu, C., Hathaway, H. J. & Prossnitz, E. R. G protein-coupled estrogen 29. Lappano, R., Pisano, A. & Maggiolini, M. GPER function in breast cancer: An receptor regulates mammary tumorigenesis and metastasis. Mol. Cancer Res. 12, overview. Front. Endocrinol. 5, 66 (2014). 1644–1654 (2014). 30. Filardo, E. J. et al. Distribution of GPR30, a seven membrane-spanning estrogen 57. Pandey, D. P. et al. Estrogenic GPR30 signalling induces proliferation and receptor, in primary breast cancer and its association with clinicopathologic migration of breast cancer cells through CTGF. EMBO J. 28, 523–532 (2009). determinants of tumor progression. Clin. Cancer Res. 12, 6359–6366 (2006). 58. Hammond, M. E. H. et al. American Society of Clinical Oncology/College of 31. Ignatov, A. et al. G-protein-coupled estrogen receptor GPR30 and tamoxifen American Pathologists Guideline recommendations for immunohistochemical resistance in breast cancer. Breast Cancer Res. Treat. 128, 457–466 (2011). testing of estrogen and progesterone receptors in breast cancer. J. Clin. Oncol. 28, 32. Broselid, S. et al. G protein-coupled estrogen receptor is apoptotic and correlates 2784–2795 (2010). with increased distant disease-free survival of estrogen receptor-positive breast 59. Wolff, A. C. et al. Human epidermal growth factor receptor 2 testing in breast cancer patients. Clin. Cancer Res. 19,1681–1692 (2013). cancer: American Society of Clinical Oncology/College of American Pathologists 33. Tutzauer, J. & Sjostrom, M. Plasma membrane expression of G protein-coupled Clinical Practice Guideline focused update. J. Clin. Oncol. 36, 2105–2122 (2018). estrogen receptor (GPER)/G protein-coupled receptor 30 (GPR30) is associated 60. Goldhirsch, A. et al. Personalizing the treatment of women with early breast with worse outcome in metachronous contralateral breast cancer. PLoS One 15, cancer: Highlights of the St Gallen International Expert Consensus on the Primary e0231786 (2020). Therapy of Early Breast Cancer 2013. Ann. Oncol. 24, 2206–2223 (2013). 34. Sjöström, M. et al. Lack of G protein-coupled estrogen receptor (GPER) in the 61. Ritchie, M. et al. limma powers differential expression analyses for RNA- plasma membrane is associated with excellent long-term prognosis in breast sequencing and microarray studies. Nucleic Acids Res. 43, e47 (2015). cancer. Breast Cancer Res. Treat. 145,61–71 (2014). 62. Yu, G., Wang, L., Han, Y. & He, Q. clusterProfiler: An R package for comparing 35. Steiman, J., Peralta, E. A., Louis, S. & Kamel, O. Biology of the estrogen receptor, biological themes among gene clusters. Omics: J. Integr. Biol. 16, 284–287 (2012). GPR30, in triple negative breast cancer. Am. J. Surg. 206, 698–703 (2013). 63. Hänzelmann, S., Castelo, R. & Guinney, J. GSVA: Gene set variation analysis for 36. Jiang, Y. Z. et al. Genomic and transcriptomic landscape of triple-negative breast microarray and RNA-Seq data. BMC Bioinform. 14, 7 (2013). cancers: Subtypes and treatment strategies. Cancer Cell 35, 428–440 (2019). 64. Liberzon, A. et al. Molecular signatures database (MSigDB) 3.0. Bioinformatics 27, 37. Vivacqua, A. et al. G protein-coupled receptor 30 expression is up-regulated by 1739–1740 (2011). EGF and TGF alpha in estrogen receptor alpha-positive cancer cells. Mol. Endo- 65. Camp, R. L., Dolled-Filhart, M. & Rimm, D. L. X-tile: A new bio-informatics tool for crinol. 23, 1815–1826 (2009). biomarker assessment and outcome-based cut-point optimization. Clin. Cancer 38. Tutzauer, J. et al. Ligand-independent G protein-coupled estrogen receptor/G Res. 10, 7252–7259 (2004). protein-coupled receptor 30 activity: Lack of receptor-dependent effects of G-1 and 17-estradiol. Mol. Pharm. 100, 271–282 (2021). 39. Thomas, P., Pang, Y., Filardo, E. J. & Dong, J. Identity of an estrogen membrane ACKNOWLEDGEMENTS receptor coupled to a G protein in human breast cancer cells. Endocrinology 146, This research was funded by the National Natural Science Foundation of China 624–632 (2005). 40. Mauvais-Jarvis, F., Lange, C. A. & Levin, E. R. Membrane-initiated estrogen, (82072938) for H.L., (82002802) for D.M., and (81722032 and 82072916) for K.Y. androgen, and progesterone receptor signaling in health and disease. Endocr. Rev. 43, 720–742 (2022). 41. Arias-Pulido, H. et al. GPR30 and estrogen receptor expression: New insights into AUTHOR CONTRIBUTIONS hormone dependence of inflammatory breast cancer. Breast Cancer Res. Treat. Conceptualization, T.X. and H.L.; data curation, T.X. and H.L.; formal analysis, J.W.; 123,51–58 (2010). funding acquisition, D.M., H.L., and K.Y.; methodology, D.M., Y.F., and L.L.; project 42. Gupta, P. B. & Kuperwasser, C. Contributions of estrogen to ER-negative breast administration, H.L. and K.Y.; resources, K.Y.; software, T.X. and D.M.; supervision, H.L. tumor growth. J. Steroid Biochem. Mol. Biol. 102,71–78 (2006). 43. Haynes, M. P. et al. Membrane estrogen receptor engagement activates endo- and K.Y.; validation, S.C., R.T., and J.Y.; visualization, T.X. and D.M.; writing – original thelial nitric oxide synthase via the PI3-kinase-Akt pathway in human endothelial draft, T.X. and H.L.; writing – review & editing, C.M. T.X., and D.M. contributed equally cells. Circ. Res. 87, 677–682 (2000). to this work. 44. Maggiolini, M. et al. The G protein-coupled receptor GPR30 mediates c-fos up- regulation by 17beta-estradiol and phytoestrogens in breast cancer cells. J. Biol. Chem. 279, 27008–27016 (2004). COMPETING INTERESTS 45. Filardo, E. J. & Thomas, P. GPR30: A seven-transmembrane-spanning estrogen receptor that triggers EGF release. Trends Endocrinol. Metab. 16, 362–367 (2005). The authors declare no competing interests. npj Breast Cancer (2022) 100 Published in partnership with the Breast Cancer Research Foundation T. Xu et al. ADDITIONAL INFORMATION Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, Supplementary information The online version contains supplementary material adaptation, distribution and reproduction in any medium or format, as long as you give available at https://doi.org/10.1038/s41523-022-00472-4. appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party Correspondence and requests for materials should be addressed to Haojun Luo or material in this article are included in the article’s Creative Commons license, unless Keda Yu. indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory Reprints and permission information is available at http://www.nature.com/ regulation or exceeds the permitted use, you will need to obtain permission directly reprints from the copyright holder. To view a copy of this license, visit http:// creativecommons.org/licenses/by/4.0/. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. © The Author(s) 2022 Published in partnership with the Breast Cancer Research Foundation npj Breast Cancer (2022) 100

Journal

npj Breast Cancer – Springer Journals

Published: Aug 30, 2022

There are no references for this article.