The ER-negative BC cell lines BT-549, MDA-MB-231 MDA-MB-468, and HS578t, and the ER-positive BC cell lines T47D, MCF-7 and ZR-75-1 were purchased from the American Type Culture Collection and maintained in RPMI-1640 medium (Gibco; Thermo Fisher Scientific, Inc.) or Dulbecco's modified Eagle's medium/F12 with 10% fetal bovine serum (both from Gibco; Thermo Fisher Scientific, Inc.), 100 µg/ml streptomycin and 100 U/ml penicillin (Invitrogen; Thermo Fisher Scientific, Inc.) at 37°C in a 5% CO₂ standard humidified incubator. For time- and dose-dependence experiments, MDA-MB-231 and BT-549 BC cells were treated with recombinant human (rh)CXCL1 at concentrations of 0, 0.1, 1 or 10 ng/ml for 1 h at 37°C, and treated with 1 ng/ml rhCXCL1 for 0, 10, 30, 60 min at 37°C in a 5% CO₂ standard humidified incubator. MDA-MB-231 and BT-549 BC cells were treated the combinations of treatments (1 ng/ml rhCXCL1 + 200 nM SB225002, 1 ng/ml rhCXCL1 + 5 µM U0126, 1 ng/ml rhCXCL1 + 200 nM SB225002 + 5 µM U0126) at 37°C in a 5% CO₂ standard humidified incubator. rhCXCL1 (cat. no. 275-GR) was obtained from R&D Systems, Inc. The CXCR2 inhibitor SB225002 (cat. no. S7651) and the MEK1/2 inhibitor U0126 (cat. no. S1102) were purchased from Selleck Chemicals.

For the cell migration assay, 3×10⁵ MDA-MB-231 or BT-549 cells in 200 µl serum-free medium were seeded into the upper chamber of a Transwell plate (EMD Millipore), and complete medium with or without various concentrations (0.1, 1.0 and 10 ng/ml) of CXCL1, U0126 (5 µM) and SB225002 (200 nM) was added to the lower compartment. Wells without CXCL1 served as controls. The cells were incubated for 12 h at 37°C, then the Transwell inserted were removed and washed, and cells were fixed with 4% paraformaldehyde for 15 min at room temperature and stained with 0.5% crystal violet for 5 min at room temperature. The numbers of migratory cells in five randomly selected fields were counted under an inverted light microscope (magnification, ×200; TE2000-U; Nikon Corporation).

For the cell invasion assay the upper chamber was coated with Matrigel (EMD Millipore) as described previously (21). The remaining steps were the same as the migration assay. After 24 h of incubation at 37°C, the numbers of invaded cells in five randomly selected fields were counted (magnification, ×200).

For knockdown of CXCL1 in MDA-MB-231 and BT-549 cells, lentiviral expression vectors containing CXCL1 short hairpin RNA (shRNA) or control shRNA were obtained from Shanghai GenePharma Co., Ltd. The sequence of CXCL1 shRNA was 5′-GCACATCTGTTTT GTAACT-3′, and the control shRNA sequence was 5′-TTC TCCGAACGTGTCACGT-3′. Cells at a density of 30-50% in 6-well plates were transfected with sh-CXCL1 or sh-Ctrl lentivirus (1×10⁸ TU/ml). After 8-12 h of incubation, the medium was replaced with complete medium containing FBS and puromycin. Further experiments were performed after ≥2 weeks.

Total RNA was extracted from human tissue specimens and cells using TRIzol (Invitrogen; Thermo Fisher Scientific, Inc.), according to the manufacturer's protocol. RNA was reverse transcribed using the PrimeScript RT Master Mix kit (Takara Biotechnology, Co., Ltd.), according to the manufacturer's protocol. RT was conducted as follows: 15 min at 37°C for three times, followed by inactivation at 85°C for 5 sec. qPCR was performed with SYBR Pre-mix Ex Taq™ II (Takara Biotechnology, Co., Ltd.) according to the manufacturer's protocol. qPCR was conducted as follows: 2 min at 95°C, followed by 39 cycles at 95°C for 30 sec, 30 sec at 58°C and 20 sec at 72°C. The sequences of the primers for CXCL1, GAPDH, MMP2 and MMP9 are listed in Table I. Relative gene expression was normal ized to GAPDH and calculated using the 2^−ΔΔCq method (22). The experiment was independently repeated in triplicate.

Total proteins were extracted using RIPA lysis buffer with PMSF (both from Beyotime Institute of Biotechnology). Protein concentrations were assessed using a BCA Protein Assay kit (Beyotime Institute of Biotechnology). A total of 40 µg protein was separated by 8-10% SDS-PAGE and transferred onto PVDF membranes. After blocking with 5% skim milk for 1 h at room temperature, the membranes were incubated at 4°C overnight with the following primary antibodies: p-ERK1/2 (1:1,000; cat. no. AF1891; Beyotime Institute of Biotechnology); ERK1/2 (1:5,000; cat. no. ab184699); p-AKT (1:1,000; cat. no. ab38449) (both from Abcam); AKT (1:1,000; cat. no. 9272S); STAT3 (1:1,000; cat. no. 12640S) (both from Cell Signaling Technology, Inc.); p-STAT3 (1:5,000; cat. no. ab76315); p-ribosomal S6 kinase P90 (p-RSK1P90; 1:5,000; cat. no. ab32203); RSK1P90 (1:5,000; cat. no. ab32114); MMP9 (1:5,000; cat. no. ab76003); MMP2 (1:2,000; cat. no. ab92536) (all from Abcam); and GAPDH (1:1,000; cat. no. 5174S; Cell Signaling Technology, Inc.). Subsequently, appropriate horseradish peroxidase (HRP)-conjugated secondary antibodies (1:1,000; cat. nos. 7074S and 7076S; Cell Signaling Technology, Inc.) were applied for 1 h at 37°C. The immunoreactive bands were detected using an enhanced chemiluminescence reagent (EMD Millipore).

An IF assay was performed as described previously (21). Cells were grown on glass coverslips for 24 h, fixed with 4% paraformaldehyde for 20 min at room temperature, permeabilized with 0.1% Triton X-100 for 15 min and then blocked with 10% normal goat serum (cat. no. C0265; Beyotime Institute of Biotechnology) for 30 min at room temperature. The cells were incubated overnight at 4°C with specific primary antibodies against CXCL1 (1:200; cat. no. ab89318; Abcam). After washing three times with PBS, the cells were stained with FITC-conjugated goat anti-rabbit secondary antibody (1:200; cat. no. TA130022; OriGene Technologies, Inc.) for 1 h at room temperature. The cell nucleus was stained with DAPI for 5 min at room temperature. IF images were obtained with a Nikon Eclipse 80i microscope (magnification, ×400; Nikon Corporation)

A total of 87 paired human breast tissue specimens, including tumor and adjacent non-tumor tissue, were obtained from the First Affiliated Hospital of Chongqing Medical University. All patients (20-72 years old) underwent surgery for BC at the First Affiliated Hospital of Chongqing Medical University between November 2015 and June 2016. All patients had their primary site in the breast and were diagnosed specifically with BC for the first time by the Clinical Diagnostic Pathology Center of Chongqing Medical University. The ER status of the patient was determined according to the results of immunohistochemistry by the Clinical Diagnostic Pathology Center of Chongqing Medical University. The study was approved by the Ethics Committee of Chongqing Medical University. Written informed consent was obtained from all patients.

IHC staining was performed as described previously (21). The human tissues were fixed with 4% formaldehyde buffer for 12-24 h at room temperature. Deparaffinized specimens were then sectioned (4-µm thick slices). The slices were autoclaved at 115°C for 5 min for antigen retrieval in citric acid buffer (pH 6.0), quenched for endogenous peroxidase activity with 0.3% H₂O₂ solution for 10-15 min, blocked for non-specific binding with 10% normal goat serum for 10-15 min at room temperature, and incubated with specific rabbit primary antibodies against CXCL1 (1:400; cat. no. ab89318; Abcam) and p-ERK1/2 (1:200; cat. no. AF1891; Abcam) overnight at 4°C. Subsequently, the sections were treated with HRP-conjugated goat anti-rabbit IgG secondary antibody (1:200; cat. no. TA140003; OriGene Technologies, Inc.) for 30 min at room temperature. After staining with diaminobenzidine (OriGene Technologies, Inc.) and hematoxylin for 5 sec at room temperature, images were captured using a Nikon Eclipse 80i microscope (magnification, ×200; Nikon Corporation). CXCL1 and p-ERK1/2 staining intensities (I) were scored as: 0, 1, 2, 3. The percentage of the stained area (A) was scored as: 1 (0-25%), 2 (26-50%), 3 (51-75%) and 4 (76-100%). The sum of the intensity and percentage scores (I + A) was used as the final IHC score. Expression was analyzed using Image-Pro Plus 6.0 software (Media Cybernetics, Inc.).

BC cells were seeded in a 6-well cultured plate at a density of 5×10⁵ cells. Following culture for 12 h, the suspension was replaced with 1 ml serum-free media. After the cells were starved for 24 h, the supernatants were harvested and centrifuged in 1,000 × g for 10 min at room temperature. Concentrations of secreted CXCL1 in the supernatants were determined using a human CXCL1/GROα Quantikine ELISA kit (cat. no. DGR00B; R&D Systems, Inc.) according to the manufacturer's protocol.

Oncomine, a cancer microarray database, was screened for breast cancer datasets where ER status was determined (wwww.oncomine.org) (23). A total of 4 independent microarrays, including Bittner (GSE2109), The Cancer Genome Atlas database, Sorlie (24) and Desmedt (25) were obtained from the Oncomine database. CXCL1 expression was analyzed in ER-negative and ER-positive BC with the R (version 3.5.1) package ggstatspot (indrajeet-patil.github.io/ggstatsplot).

SPSS 20.0 software (IBM Corp.) was used for all statistical analysis. Data of three independent experiments are presented as the mean ± standard deviation. One-way ANOVA followed by Dunnett's multiple comparisons tests was used to evaluate the significant differences among multiple groups. Fisher's exact test was used to evaluate associations between the detected protein expression levels of CXCL1 and p-ERK1/2. P<0.05 was considered to indicate a statistically significant difference.

To analyze the expression of CXCL1 in human BC tissues, the relative mRNA expression levels of CXCL1 in all 87 samples were examined. The clinical parameters of the patients with BC are presented in Table II. The CXCL1 mRNA levels in ER-negative BC tissues (n=55) were significantly upregulated compared with the ER-positive BC tissues (n=32; Fig. 1A). In addition, four independent microarrays obtained from the Oncomine public database were analyzed. The mRNA expression levels of CXCL1 were significantly upregulated in the ER-negative BC cases compared with the ER-positive BC cases in the Bittner, Sorlie and Desmedt breast databases and The Cancer Genome Atlas database (Fig. 1B-E). In summary, these results suggest that there is high expression of CXCL1 mRNA in ER-negative breast tumors.

To further verify the association between CXCL1 expression and ER-negative BC, four ER-negative BC cell lines (BT-549, MDA-MB-231, MDA-MB-468 and HS578t) and three ER-positive BC cell lines (T47D, MCF-7, ZR-75-1) were analyzed. The mRNA and protein CXCL1 expression levels in these cells were detected by RT-qPCR and ELISA. The levels of CXCL1 mRNA (Fig. 2A) and protein (Fig. 2B) were markedly upregulated in the ER-negative BC cells compared with the ER-positive cells. CXCL1 was predominantly located in the cell cytoplasm, as determined via IF assays (Fig. 2C). These data demonstrated that CXCL1 exhibits increased expression in ER-negative BC cells compared with ER-positive BC cells.

Based on the aforementioned findings, it was hypothesized that CXCL1 overexpression in ER-negative BC may be associated with the aggressive nature of ER-negative BC. To investigate the effect of CXCL1 on the invasion of ER-negative BCs, MDA-MB-231 and BT-549 cells were treated with/without rhCXCL1 (0.1, 1.0 and 10 ng/ml). A Transwell assay revealed that CXCL1 significantly increased the migration (Fig. 2D) and invasion (Fig. 2E) of MDA-MB-231 and BT-549 cells in a dose-dependent manner, compared with control treatment. Subsequently, SB225002, a specific CXCR2 antagonist, was used to determine whether the effects of CXCL1 on the migration and invasion of ER-negative cells were associated with CXCR2. The CXCL1-induced increases in cell migration (Fig. 2F) and invasion (Fig. 2G) were significantly attenuated by treatment with SB225002. In summary, these data suggested that enhanced CXCL1 in ER-negative BC promotes cell migration and invasion in a CXCR2-dependent manner.

Previous studies have reported that chemokines can bind to their receptors to induce cancer progression by stimulating a series of downstream signaling pathways, including the PI3K/AKT, Janus kinase (JAK)/STAT3 and ERK1/2 pathways (26-29). Therefore, possible signaling mechanisms associated with the CXCL1/CXCR2-induced promotion of ER-negative BC cell migration and invasion were examined by western blot analysis. It was identified that only p-ERK1/2 was activated by rhCXCL1 in MDA-MB-231 and BT-549 cells in a time- and dose-dependent manner (Figs. 3A and B, and 4A and B).

The results demonstrated that CXCL1-mediated cell migration and invasion were significantly inhibited by either SB225002 or U0126 compared with rhCXCL1 treatment alone (Figs. 3C and D, and 4C and D). Similarly, the activated ERK1/2 and RSK1P90 proteins in the ERK pathway that were stimulated by CXCL1 were inhibited following treatment with SB225002 and U0126 (Figs. 3E and 4E). These findings suggested that CXCL1 regulates the migration and invasion of ER-negative cells via ERK signaling in a CXCR2-dependent manner.

To further determine the role of CXCL1 in the invasion of ER-negative BC cells, the lentivirus-mediated shCXCL1 and control vector were stably transduced into ER-negative MDA-MB-231 and BT-549 cells. The efficiency of knockdown was verified via RT-qPCR analysis and ELISAs (Fig. 5A-D). As hypothesized, reduced CXCL1 significantly attenuated the migratory abilities of MDA-MB-231 and BT-549 cells (Fig. 6A and B). Similar results were observed in the cell invasion assay (Fig. 6C and D). Subsequently, the levels and phosphorylation of ERK and RSK1P90, key proteins associated with ERK signaling activation, were detected via western blot analysis. It was identified that knockdown of CXCL1 in MDA-MB-231 and BT-549 cells inhibited ERK1/2 pathway activation (Fig. 6E and F). These data demonstrated that silencing CXCL1 in ER-negative cells prevents cell migration and invasion due to inhibition of the ERK1/2 pathway.

It has been reported that MMP2 and MMP9 are strongly associated with tumor metastasis (30-33). Thus, it was hypothesized that activated ERK1/2 signaling may contribute to CXCL1-mediated MMP2/9 expression in ER-negative cells. To verify this hypothesis, the MDA-MB-231 and BT-549 cells pretreated with SB225002, U0126 and/or rhCXCL1 were evaluated for their mRNA and protein expression levels of MMP2/9 via RT-qPCR and western blot analyses. As presented in Fig. 7A and C, rhCXCL1 treatment significantly increased the mRNA and protein levels of MMP2/9; however, the effects of CXCL1 on the activation of MMP2/9 in MDA-MB-231 cells were reversed by pretreatment with SB225002 or U0126. Similar results were observed in BT-549 cells (Fig. 7B and D). Furthermore, it was determined that knockdown of CXCL1 in MDA-MB-231 and BT-549 cells by shCXCL1 inhibited ERK//MMP2/9 signaling, and this inhibitory effect could be reversed by the treatment of these cells with rhCXCL1 (Fig. 7E and F). In summary, these data suggested that CXCL1 can stimulate MMP2/9 expression in ER-negative cells via ERK1/2 activation in a CXCR2-dependent manner.

The protein expression levels of CXCL1 and p-ERK1/2 were detected in 88 BC tissue samples via IHC. CXCL1 and p-ERK1/2 were expressed in 62.5% (55/88) and 52.3% (46/88) of these tumor cases, respectively. Representative images are presented in Fig. 8A, and quantitative analysis revealed that CXCL1 expression was significantly increased in ER-negative BC tissues compared with ER-positive tissues (P<0.05; Fig. 8B). Furthermore, a significant association between CXCL1 and p-ERK1/2 expression was observed via IHC; p-ERK1/2 expression was observed in 61.8% (34.55) of CXCL1-positive tissues, but only 36.4% (12/33) of CXCL1-negative tissues (P<0.05; Fig. 8C). These data suggested an enhanced CXCL1 protein expression in ER-negative BC, that is associated with the expression of p-ERK1/2 protein.