Paired human breast cancer clinical samples and adjacent normal samples used in the present study were obtained from 39 female patients, aged from 30 to 82 (median age, 52.9) at the Affiliated Hospital of Southwest Medical University (Luzhou, China) between December 2018 and March 2019. The present study was approved by The Medical Ethics Committee of the Institutional Review Board of the Affiliated Hospital of Southwest Medical University (Luzhou, China). Written informed consent was provided by all patients. All types of breast cancer were invasive ductal carcinoma as confirmed by pathologists. Inclusion criteria were histological diagnosis of invasive ductal carcinoma of the breast before operation. None of the patients received chemotherapy or radiotherapy prior to surgical excision Detailed clinicopathological information were obtained, including sex, age, tumor depth, tumor size, differentiation, lymph node invasion and distant metastasis (data not shown).

Three human cell lines, including breast cancer cell lines MCF-7 and MDA-MB-231, and breast non-tumor cell line MCF-10a, were used in the present study. All cell lines were frozen in liquid nitrogen and cultured in MEM supplemented with 10% fetal bovine serum (FBS, Gibco; Thermo Fisher Scientific, Inc.), 100 µ/ml penicillin and 100 µ/ml streptomycin (Amresco, LLC) in a 5% CO₂ incubator at 37°C. Cells were passaged once every three days. To inhibit autophagy, 2 mM 3-methyladenine (3-MA, Sigma-Aldrich; Merck KGaA) was added into culturing medium for 24 h.

Total RNA was extracted from BC clinical samples and adjacent normal samples using TRIzol^® (Thermo Fisher Scientific, Inc). A total of 5 mg of each sample was suspended in 1 ml RNA EasyOut kit (Chengdu Daosheng Biological Technology Co., Ltd.) and sonicated at room temperature for 5 sec using a SoniConvert™ system (Chengdu Daosheng Biological Technology Co., Ltd.), both according to the manufacturer's instructions. For RT-qPCR analysis, 1 µg total RNA was reverse-transcribed into cDNA using an Superscript III RT-qPCR kit (Thermo Fisher Scientific, Inc.). Then the cDNA was used as template under the following conditions: 40 cycles at 95°C for 30 sec and 60°C for 1 min. An ABI 7500 Fast qPCR system (Applied Biosystems; Thermo Fisher Scientific, Inc.) was used for qPCR. For amplification, the following procedure was performed: Predenaturation at 98°C for 2 min for each cycle, denaturation at 98°C for 10 sec, annealing and extension at 60°C for 60 sec, 40 cycles; final extension at 60°C for 5 min. The primers used for qPCR were as follows: GAS5 forward, 5′-GAGCAAGCCTAACTCAAGCC-3′ and reverse, 5′-ACACAGTGTAGTCAAGCCGA-3′; ULK1 forward, 5′-GGCAAGTTCGAGTTCTCCCG-3′ and reverse, 5′-CGACCTCCAAATCGTGCTTCT-3′; ULK2 forward, 5′-TGGAGACCTCGCAGATTATTTGC-3′ and reverse, 5′-CTGTGCAGGATTCGCATGG-3′; β-actin forward, 5′-CATGTACGTTGCTATCCAGGC-3′ and reverse, 5′-CTCCTTAATGTCACGCACGAT-3′. The qPCR results were analyzed and expressed relative to the Cq (threshold cycle) values (25) and then converted to fold change values. A fold-change of 2.0 was considered significant.

Cells (1×10⁶) were suspended in 1 ml of RIPA buffer (Sigma-Aldrich; Merck KGaA) and lysed using SoniConvert^® Sonicator (Chengdu Daosheng Biological Technology Co., Ltd.). Protein concentration was determined via BCA assay (Sigma-Aldrich; Merck KGaA) according to the manufacturer's instructions. In each lane, 20 µg total protein was applied to a 5–15% SDS-PAGE gel. Fractionated proteins were transferred onto a nitrocellulose membrane and blocked using 5% BSA containing PBS at room temperature for 1 h. Membranes were then incubated with primary antibodies overnight at 4°C. The primary antibodies used were obtained from Abcam and are listed as follows: Rabbit monoclonal anti-ULK1 antibody (1:2,000; cat. no. ab133766), rabbit polyclonal anti-ULK2 antibody (1:3,000; cat. no. ab97695), rabbit monoclonal anti-Atg3 antibody (1:500; cat. no. ab108251), rabbit polyclonal anti-Atg5 antibody (1:2,000; cat. no. ab228668), rabbit monoclonal anti-Atg7 antibody (1:2,000; cat. no. ab52472), rabbit monoclonal anti-Beclin 1 antibody (1:2,000; cat. no. ab207612), rabbit polyclonal anti-LC3B antibody (1:500; cat. no. ab48394) and mouse monoclonal anti-β-actin antibody (1:5,000; cat. no. ab8226). Following three washes with TBST, secondary antibody was incubated with the membrane at room temperature for 1 h. Goat anti-rabbit IgG H&L antibody (HRP-labelled; 1:10,000; cat. no. ab7090) and goat anti-mouse IgG H&L antibody (HRP-labelled; 1:10,000; cat. no. ab97040) were used as the secondary antibodies. Blot bands were quantified via densitometry using ImageJ software (v2.0; National Institutes of Health). β-actin was used as an internal reference.

MCF-7 cells were seeded at 2×10⁵ cells/well in a 12-well plate and attached overnight at 37°C in a 5% CO₂ incubator. Cells were transfected using pre-made lentiviral particles expressing a fusion target of GFP-RFP-LC3 (cat. no. GM-1314L204H-S, Genomeditech Co., Ltd.) according to the manufacturer's instructions. After 2 h incubation at 37°C in the dark, the medium was refreshed and cells were incubated in MEM supplemented with 10% FBS at 37°C for 48 h. Cells were visualized under an X71 (U-RFL-T) fluorescence microscope (magnification, ×100; Olympus Corporation) and analyzed with ImageJ software (v2.0; National Institutes of Health).

MCF-7 cells were seeded at 2×10⁵ cells/well in a 12-well plate and incubated with Lysotracker green and mitotracker sox red according to the manufacturer's instructions. After 1 h incubation at 37°C, 1 µg/ml 4′,6-diamidino-2-phenylindole (DAPI) was used to stain nuclei for 5 min at room temperature in the dark. Cellular fluorescence was observed using an X71 (U-RFL-T) fluorescence microscope (magnification, ×100; Olympus Corporation).

A total of 5×10³ cells/well were seeded in 96-well plates and allowed to attach overnight. After addition of 1, 2, 3, 4 and 5 µmol/L of cisplatin (Thermo Fisher Scientific, Inc.), for 1–5 days, the cell proliferation assay was performed by adding 10 µl CCK-8 solution (Guangzhou RiboBio Co., Ltd.) according to the manufacturer's protocol. After 1 h incubation at 37°C, the absorbance of each well was measured at a wavelength of 450–620 nm using a Multiskan spectrum microplate reader (Thermo Fisher Scientific, Inc.).

A total of 1×10⁶ MCF-7 cells were collected and washed with ice-cold PBS twice, and fixed using 1 ml 70% ethanol and stored at 4°C overnight. After washing with ice-cold PBS three times, the staining solution containing 100 ng/ml RNase A and 5 µg/ml propidium iodide (PI) (Sigma-Aldrich; Merck KGaA) was added to cells and incubated in the dark for 30 min at 4°C. Cells were the screened by a three laser Navios flow cytometer (Beckman Coulter, Inc.) and analyzed by ModFit LT software (v2.0, Verity Software House).

MCF-7 Cells were treated with 2 µM cisplatin for 24 h, and untreated cells were considered as control group. Cells were resuspended in trypsin and washed twice with ice-cold PBS. Then, Annexin V-FITC/PI double staining was performed according to the instructions of the Annexin V-FITC/PI apoptosis detection kit (Thermo Fisher Scientific, Inc.). For each sample, ~2×10⁴ cells were collected and scanned by a three laser Navios flow cytometer (Beckman Coulter, Inc.). Data files was analyzed using FlowJo v9.7.6 (FlowJo LLC).

Transwell membranes were precoated with 60 µl Matrigel (8%) in MEM and incubated at room temperature for 2 h. For each well, 1×10⁴ cells were added into upper chamber. Subsequently, 600 µl of MEM supplemented with 10% FBS was added to the lower chamber. After 24 h, medium was removed and 4% paraformaldehyde was used for fixation at room temperature for 10 min. After washing with PBS twice, cells were stained with 0.1% crystal violet (in PBS) at room temperature for 30 min. After staining, the crystal violet solution was removed and the cells were washed with PBS five times. Then the cells in five random views were counted under an X71 fluorescence microscope (magnification, ×100; Olympus Corporation).

A total of 0.8 µg GAS5-expressing plasmid or empty vector (Addgene, Inc.) was mixed with 4 µl Lipofectamine™ 2000 transfection reagent (Thermo Fisher Scientific, Inc.) in 0.5 ml OptiMEM medium (Thermo Fisher Scientific, Inc.), and this was added to MCF-7 cells for 4 h followed by replacing the medium. After 24 h, the culture media was refreshed containing 500 µg/ml geneticin sulfate 418 (G418, Sigma-Aldrich; Merck KGaA) for antibiotic selection. Two weeks later, the transfected cells that had survived were collected and maintained in 4,000 µg/ml G418. Individual clones (GAS5 clone-1 and clone-2) were detected, separately.

Low-melting agarose (2 ml 0.6%, Sigma-Aldrich; Merck KGaA) diluted in MEM supplemented with 10% FBS was added to 6-well plates. For each well, a total of 5×10³ cells were added in 1 ml of 0.3% (weight/volume) low-melting agarose diluted in MEM supplemented with 10% FBS. After 2 weeks of incubation at 37°C, cells were stained with 10 ng/ml nitro blue tetrazolium at 37°C and colony number and diameter were quantified under an X71 fluorescence microscope (magnification, ×100; Olympus Corporation).

Statistical analyses were carried out using SPSS v19.0 for Windows (IMB Corp). Student's paired t-tests and ANOVA were used to compare two or more groups for statistical analysis followed by Tukey's post hoc test, respectively. Pearson's correlation analysis was used for the correlation between GAS5, ULK1 and ULK2 levels. Data are presented as the mean ± standard deviation. P<0.05 was considered to indicate a statistically significant difference. All experiments were repeated at least three times.

The present study detected the expression levels of GAS5 in paired BC clinical samples and adjacent normal tissues by performing RT-qPCR. As revealed in Fig. 1A, among the 39 cases of BC, ~66.7% exhibited lower levels of GAS5 compared with matched controls, and GAS5 was significantly downregulated in BC tissue samples compared with those in adjacent normal tissues (1.95±0.36 vs. 0.40±0.22, P=0.002, Fig. 1B). To confirm whether similar expression patterns of GAS5 were presented in BC cell lines, GAS5 expression levels were analyzed in MCF-7 and MDA-MB-231 cells compared with the non-tumor cell line MCF-10a. As anticipated, the expression levels of GAS5 in both MCF-7 and MDA-MB-231 was significantly lower compared with those in MCF-10a cells (*P=0.037; **P 0.006, Fig. 1C), indicating that GAS5 may exert antitumor effects in BC.

The present study also assessed the expression levels of ULK1 and ULK2 in BC clinical samples and adjacent samples. Results revealed that GAS5 expression in BC clinical samples was positively and significantly correlated with ULK1 and ULK2 levels (Fig. 2A). To confirm whether the expression levels of GAS5 positively regulated ULK1, ULK2 and other autophagy-associated proteins, including AuTophaGy-related (Atg)3, Atg5, Atg7, Beclin-1 and microtubule-associated proteins 1A/1B light chain (LC)3B, cells stably overexpressing GAS5 were selected and analyzed for the expression levels of the aforementioned proteins. As revealed in Fig. 2B, ULK1 and ULK2 protein levels were both significantly upregulated in GAS5 clone-1 and clone-2 (*P<0.05), which was consistent with the tendency observed in clinical samples. In addition, GAS5 overexpression induced an increase in Beclin-1 and LC3B levels with no effect on Atg3, Atg5 and Atg7, indicating that GAS5 potentially initiates autophagy by upregulating ULK1 and ULK2, but not Atgs (data not shown).

To further confirm whether autophagosome formation was promoted by overexpressed GAS5, images of MCF-7 cells stably overexpressing GAS5 and transiently expressed LC3-RFP-GFP were captured using confocal microscopy to view the autophagosomes. As revealed in Fig. 3A, without autophagy promoter treatment, overexpression of GAS5 promoted the formation of autophagosomes. Double staining of lysotracker green and mitotracker sox red also demonstrated that overexpression of GAS5 promoted colocalization of mitochondria and lysosomes, indicating the promoting effect of GAS5 on autophagy/mitophagy (Fig. 3B).

To evaluate the effects of overexpressed GAS5 on the malignant behaviors of MCF-7 cells, cell viability was analyzed. As revealed in Fig. 4A, mixed clone-1 and clone-2 (GAS5-1/2) inhibited cell viability compared with the vector-transfected clone (P<0.05). Pre-treatment of 10 µM 3-MA was performed to demonstrate whether autophagy induced by GAS5 affected cell viability. 3-MA treatment notably decreased cell viability in the vector group, indicating the inhibitory effect of 3-MA on proliferation alone, and this inhibitory effect on cell viability was reversed by GAS5 overexpression (Fig. 4A). By performing cell cycle analysis, it was observed that GAS5-induced block of the cells at G₁/G₀ was reversed by 3-MA treatment. The effects of GAS5 on invasion and colony formation in MCF-7 cells was then investigated, and results demonstrated that invasion (Fig. 4C) and colony formation (Fig. 4D) of MCF-7 cells were significantly inhibited by GAS5 overexpression compared with vector group and partially reversed by inhibiting autophagy compared with GAS5-1/2 group, indicating that GAS5 may exert antitumor effects by partially inducing autophagy in MCF-7 cells.

Since autophagy is a critical physiological process for chemosensitivity in cancer cells, it was then evaluated whether GAS5 affects chemosensitivity by regulating autophagy in MCF-7 cells. By performing chemosensitivity on cisplatin, as anticipated, both 3-MA treatment and overexpression of GAS5 desensitized MCF-7 cells to cisplatin (Fig. 5A, *P<0.05, vs. GAS5-1/2 group; ^#P<0.05, vs. Vector group). However, 3-MA treatment decreased chemosensitivity in MCF-7 cells overexpressing GAS5 instead of reversing the promoting effect of GAS5 on chemosensitivity. To analyze the apoptotic rate induced by cisplatin, Annexin V-FITC/PI double staining was performed after 24-h treatment with 2 µM cisplatin. Autophagy inhibition significantly increased apoptotic cell death compared with mock group, and GAS5-promoted apoptosis was not notably affected by autophagy inhibition, indicating that the effect of GAS5 on chemosensitivity may be autophagy-independent (Fig. 5B).