The human breast cancer cell line SK-Br-3 was purchased from the cell library of the Chinese Academy of Sciences. SK-Br-3 cells were cultured in DMEM (HyClone; Cytiva), supplemented with 10% FBS (Natocor) under 5% CO₂/95% in a humidified incubator for 72 h.

A total of 25 female patients with breast cancer who were hospitalized at the West China Hospital Affiliated with Sichuan University between December 2018 and April 2019 were included in the patient group for the present study (Table SI). The healthy controls included 19 women who underwent health examinations in the same hospital during the same time period. The TNM system was used to determine breast cancer stage in these patients. The inclusion criteria for patients with breast cancer were as follows: i) breast IDC confirmed by pathology, without distant metastasis. The exclusion criteria were as follows: i) non-IDC of the breast confirmed by pathology; ii) previous radiotherapy, chemotherapy or surgical treatment; iii) a family history of breast cancer; and iv) use of oral estrogen receptor (ER) drugs. For all patients, fasting venous blood was collected in the morning, and blood samples were allowed to naturally coagulate prior to centrifugation. Serum samples were collected and stored at −80°C. All patients and/or their families provided written informed consent prior to participation in the present study, and were informed in detail regarding the purpose and methods involved.

IHC was performed using the SPlink Detection kit (cat. no. SP9000; ZSGB-BIO; OriGene Technologies, Inc.), according to the manufacturers protocol. The primary antibody used was mouse anti-human FASN (1:300; cat. no. sc-48357; Santa Cruz Biotechnology, Inc.). For the negative control, the primary antibody was replaced with PBS solution.

A total of five random fields of view were assessed under a light microscope (Nikon ECLIPSE Ti-U; Nikon Corporation) at a magnification of ×400. The staining intensity score and the proportion of FASN-positive cells were evaluated by a pathologist as follows: Staining intensity: i) 0, negative; ii) 1, weakly stained; iii) 2, moderately stained; and iv) 3, strongly stained. Staining extent (percentage of stained cells): i) 0, none; ii) 1, 1–20%; iii) 2, 21–40%; iv) 3, 41–60%; v) 4, 61–80%; and vi) 5, 81–100%. The final immunoreactive score (IRS) of the FASN expression level was calculated by multiplying the staining intensity score with the staining extent score. IRS was dichotomised using X-tile software (version 3.4.7; Yale School of Medicine), which is a useful bioinformatics tool for outcome-based cut-point optimization. IRS ≤7.5 was designated as low expression, while IRS >7.5 was designated as high expression (17).

Cell sample and serum sample preparation. Cells (1×10⁶ per plate) in 10-cm plates were incubated in growth media for 1 day, collected and washed with 0.9% NaCl. Lipids were extracted from the cells using procedures similar to the Folch method and evaporated under nitrogen gas, as previously described (18). After adding 0.75 ml methanol/water solvent (v/v=4:1), the cells were harvested, pelleted and snap-frozen in liquid nitrogen. The serum sample was dissolved at room temperature. Methanol (0.7 ml) and saturated heptadecanoic acid/methanol solution (1 mg/ml, 20 µl) that served as an internal standard was added to the samples. The samples were suspended via vortex for 30 sec and then left to stand for 10 min. Following centrifugation at 13,000 × g at 4°C for 10 min, the supernatant was collected. A total of 2 ml 5% methanol sulfate solution was added, swirled for 60 sec, bathed at 62°C for 2 h, cooled to room temperature, and centrifuged (13,000 × g/min at 4°C) for 10 min. The supernatant was placed in the sample bottle and 2 ml n-hexane was added. After being swirled for 60 sec, the solution was stratified for 10 min and sampled from the n-hexane layer for testing.

Metabolomics analysis was performed on a SHIMADZU instrument (Shimadzu Corporation). The experiments were performed following a previously described protocol (19). Raw data were obtained in a full scan mode. The samples were run at random, and blank samples and quality control samples were inserted during sample analysis.

Cells were seeded in a 6-well plate at a density of 6×10⁵ cells per well. The FASN siRNA sequences were as follows: FASN siRNA#1: 5′-GGAGCGTATCTGTGAGAAA-3′; and FASN siRNA#2: 5′-CCGTGGACCTGATCATCAA-3′. The negative control siRNA sequences were as follows: 5′-TTCTCCGAACGTGTCACGTTT-3′ (20). The Lipofectamine™ RNAi MAX reagent (Invitrogen; Thermo Fisher Scientific, Inc.) was incubated with 60 pmol siRNA in OPTI-MEM (Gibco; Thermo Fisher Scientific, Inc.) for 5 min at room temperature. The siRNA-reagent mixture was then added to the culture media and incubated for 48 h prior to cell use in downstream experiments.

Stable SK-Br-3 cells overexpressing FASN containing rs17848945[G] were generated. In brief, cDNA encoding FASN was commercially synthesized (Youkang Science Instruments Co., Ltd.) and subcloned into the EcoRI and BamHI sites of the empty vector. SK-Br-3 cells were electroporated with empty vector and FASN expression vector. Positive colonies were then picked, expanded and verified by western blotting (21).

Cells (5×10⁵/ml) were plated in 6-well dishes and incubated overnight to yield confluent monolayers. On the following day, the cells were cultured in serum-free medium, and linear scratch wounds were made in the monolayer using a 10 µl sterile pipette tip. Images were captured immediately (0 h) and subsequently at 24, 48 and 72 h after wounding. At all time points, the width of the scratch area was measured at 4 points, allowing for a measurement of monolayer migration into the wounded area over time. The result was expressed as follows: Migration index = (0 h scratch width - × h healing width)/0 h scratch width ×100.

Proteins were extracted from cells using RIPA lysis buffer (Beyotime Institute of Biotechnology) containing PMSF (Amresco, LLC) at a dilution of 1:100. Protein concentrations were then measured via bicinchoninic acid protein assay kit (Beyotime Institute of Biotechnology). Proteins were next resuspended in sample loading buffer, boiled for 5 min, and electrophoresed on a polyacrylamide 10% gel. The proteins were then transferred to a nitrocellulose membrane, which was blocked with 5% low-fat dried milk at room temperature for 1 h. The blots were then probed with specific antibody against FASN (1:1,000; cat. no. sc-48357; Santa Cruz Biotechnology, Inc.) overnight at 4°C, followed by rabbit-anti-mouse secondary antibody (1:2,000; cat. no. sc-358914; Santa Cruz Biotechnology, Inc.) at room temperature for 2 h. ECL chromogenic substrate (Beyotime Institute of Biotechnology) was then used for chemiluminescence imaging to measure protein levels.

Cells were initially serum-starved for 12 h, and were then harvested and resuspended in serum-free media at 5×10⁵ cells/ml. The cells were then plated using a 24-well Transwell chambers, with 200 µl cell suspension added to the upper chamber and 700 µl culture medium containing 20% FBS added to the lower chamber. After 48 h, the Transwell chamber was removed, the original medium was discarded, and cells were washed twice with PBS, after which time the lower chamber was fixed with 100% methanol at room temperature for 30 min. The chambers were then air-dried and stained with 0.1% crystal violet solution on a shaking table for 20 min at room temperature. The wells were then washed twice with PBS; cells that did not migrate were gently removed, and the samples were allowed to dry at 37°C. The stained Transwell chamber was placed under an optical microscope (magnification, ×100), and the cells were counted in five randomly selected fields.

All data are presented as mean ± SEM. GraphPad Prism (version 8.0; GraphPad Software, Inc.) was used for all statistical analyses. Group sizes per experiment were based on a power analysis. Unpaired Students t-test or one-way ANOVA were used as appropriate to determine statistical significance. Bonferronis post hoc test was used where appropriate. P<0.05 was considered to indicate a statistically significant difference.

An increase in fatty acid synthesis is one of the most obvious metabolic changes in tumor cells. The final products of fatty acid metabolism form signaling molecules that mediate the activation of tumor signal transduction pathways, thereby promoting tumor growth, angiogenesis, survival and metastasis (22). In order to investigate which fatty acids affect tumor metastasis and invasion, the present study assessed serum free fatty acids (FFAs) in 25 patients with IDC and 19 healthy controls. It was revealed that the levels of 12 fatty acids were significantly increased in the serum of patients with IDC, including 6 saturated fatty acids (C14:0, C15:0, C16:0, C18:0, C20:0 and C22:0), 2 monounsaturated fatty acids (C18:1 and C20:1), and 4 polyunsaturated fatty acids (C18:2, C20:4, C22:4 and C22:6) (Fig. 1A and B and Table SII). According to the analysis of GC-MS metabolites in IDC patients and healthy controls, 17 metabolites were identified via a National Institute of Standards and Technology (NIST) library search. The loading graph of Orthogonal Projections to Latent Structures Discriminant Analysis (OPLS-DA) model output demonstrated that C18:1, C18:2, C16:0, C20:4 and C18:0 were furthest from the origin, and may represent potential tumor markers (Fig. 1C). The potential tumor markers in patients with IDC were obtained from the results of metabolite extraction with Variable Importance in Projection and the results of the loading graph. The results revealed that C18:1, C18:2, C16:0, C20:4 and C18:0 may be potential tumor markers for IDC (Fig. 1C and D). These data suggest that multiple fatty acids may be involved in the metastasis and invasion of breast cancer.

FASN is an important biological synthetase and a key enzyme of tumor lipid production. The present study investigated the association between FASN expression and fatty acid metabolism in patients with IDC. Our previous studies demonstrated an increased expression level of FASN in lymph node metastases, IDC clinical stages II and III, and larger tumors (>2 cm) (23). This means that FASN is involved in tumor growth and metastasis. FASN expression in 25 patients with IDC was next assessed by IHC staining. There were 12 cases in the high expression group and 13 cases in the low expression group (Fig. 2A and Table I). The levels of different FFAs were then compared between patients with high and low FASN expression levels, and the levels of 4 fatty acids were found to be significantly increased in patients with IDC with high FASN expression compared with those with low expression. These included two saturated fatty acids (C16:0 and C18:0), one unsaturated fatty acid (C18:1) and one polyunsaturated fatty acid (C20:4) (Fig. 2B and C and Table SIII). These data suggested that the expression of FASN may affect the FFA profiles in the serum of patients with IDC, and that FASN may be involved in tumor proliferation and migration, which is mediated through changes in fatty acid levels.

FASN is upregulated in a variety of malignant tumors and is associated with tumorigenesis and metastasis. However, to the best of our knowledge, the effect of FASN-mediated fatty acid changes on tumor cell proliferation and migration has not yet been reported. To investigate whether FASN affects cell proliferation and metastasis by regulating fatty acid metabolism in breast cancer cells, SK-Br-3 cells expressing FASN were selected to inhibit and overexpress FASN in order to detect whether changes in fatty acid metabolism occurred in intracellular and extracellular media. Western blotting revealed that the FASN protein levels were significantly decreased in SK-Br-3 cells following treatment with 20 µg/ml of the FASN inhibitor cerulenin for 24 h. The two different FASN-specific siRNA constructs also significantly inhibited FASN expression. Silencing was more efficient for the siRNA#1 construct; thus, this siRNA was used in subsequent experiments. SK-Br-3 cells with FASN overexpression were constructed (Fig. 3A-D). In SK-Br-3 cells, the FASN was knocked down and overexpressed. The content and composition of FFAs in the intracellular and extracellular medium were determined by GC-MS. It was revealed that the composition of extracellular fatty acid was lower compared with that of intracellular fatty acids (Fig. 3E and I). The expression of FASN was decreased or upregulated, and the composition of FFA remained unchanged, but the fatty acid content was altered. After FASN was decreased by cerulenin (20 µg/ml), the levels of 5 fatty acids, including C14:0, C16:1, C18:1, C18:0 and C22:4, were decreased in the cells (Fig. 3E and F). After FASN was silenced by siRNA, the levels of 7 fatty acids, including C12:0, C14:0, C16:1, C16:0, C18:1, C18:0 and C22:4, were decreased in the cells (Fig. 3E and G). Following overexpression of FASN, the levels of 8 fatty acids, including C16:1, C16:0, C18:2, C18:1, C18:0, C20:0, C22:4 and C22:0, were increased in the cells (Fig. 3E and H). After FASN was decreased by cerulenin and siRNA, the levels of 4 fatty acids, including C12:0, C14:0, C16:0 and C18:0, were decreased in the medium (Fig. 3I and J). After FASN was overexpressed, the levels of 3 fatty acids, including C14:0, C16:0 and C18:0, were increased in the medium (Fig. 3I and K). After FASN expression was decreased, the levels of FASN-regulated fatty acids inside and outside the cell were decreased; similarly, after FASN expression was increased, the levels of FASN-regulated fatty acids inside and outside the cell were increased. Therefore, FASN may promote the proliferation and migration of cancer cells by altering the metabolism of fatty acids.

The present study investigated the effect of FASN on the migration ability of breast cancer cells. A wound healing assay revealed that the migration of SK-Br-3 cells in the FASN group was inhibited by cerulenin and FASN siRNA (Fig. 4A-D), and was significantly decreased when compared with that in the controls. The Transwell experiment revealed that, after 48 h of treatment of SK-Br-3 cells with cerulenin or FASN siRNA (Fig. 4E-H), migration was significantly decreased compared with the controls, which was consistent with the results of the wound healing assay. In conclusion, when FASN in SK-BR-3 cells is decreased, cell migration is inhibited. Combined with the changes in fatty acids after FASN was decreased or increased, the findings suggest that FASN may promote tumor development by mediating changes in the levels of specific fatty acids.