The human NSCLC cell lines were purchased from ATCC. Beas-2b, H446, H460, A549 and H1299 cells were maintained in Dulbecco's modified Eagle's medium (DMEM). All the medium were supplemented with 10% heat-inactivated fetal bovine serum (FBS; Invitrogen, Carlsbad, CA, USA), penicillin (100 U/ml), and streptomycin (100 µg/ml) in a humidified atmosphere of 5% CO₂ at 37°C. All the cells were confirmed to be free from mycoplasma contamination.

The targeted knockdown sequence was the following: SH1: 5′-tctgagaagttgaggcaaa-3′; SH4: 5′-gcagcaactacagtgaaatcc-3′. The negative sequence was 5′-gtagcgcggtgtattatac-3′. Lentiviral vectors encoding shRNA were constructed by Hanyin Co. (Shanghai, China). The recombinant lentiviruses (SH) and the negative control (NC) lentivirus (Hanyin Co.) were prepared and titered to 10⁹ transfection units (TU)/ml. To obtain stable cell lines, cells were seeded in 6-well plates and infected with virus and polybrene the following day. Positive clones were selected with puromycin for 14 days to establish the stable cell lines. Additionally, the lentiviruses expressing the FSTL1 sequence (OE) and the negative control lentivirus (NC) were constructed by Hanyin Co. FSTL1-OE and control stable cell lines were then established. The efficiency of FSTL1 knockdown and overexpression was confirmed by qRT-PCR and western blot analysis.

Protein lysates (50 µg per lane) were separated by sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis and transferred onto nitrocellulose membranes. After blocking with 5% fat-free milk, the membranes were incubated with primary antibodies (1:500 dilution) at 4°C overnight, followed by horseradish peroxidase-conjugated secondary antibodies (1:3,000 dilution). Anti-human RBM8A (Santa Cruz, sc-32312), anti-human actin (Proteintech, HRP-60008), anti-human snail (Cell Signaling Technology, 3879), anti-human p-stat3 (Cell Signaling Technology, 9131), anti-human stat3 (Cell Signaling Technology, 12640), anti-human fibronectin (Sigma-Aldrich, F3648), anti-human vimentin (Cell Signaling Technology, 5741), anti-human Notch (Abways Technology, CY52444) and donkey anti-rabbit IgG (Cell Signaling Technology, 7074) were used in this study. Immune-reactive proteins were visualized by enhanced chemiluminescence (ECL).

Total RNA was isolated from NSCLC cancer cells using TRIzol reagent according to the manufacturer's instructions (Invitrogen). cDNA was reverse transcripted from 1 mg total RNA. qRT-PCR was performed with the SYBR Premix Ex Taq (Takara, Dalian, China). PCR primers were as follows: F: 5′-TCGCATCATCCAGTGGCTGGAA-3′, R: 5′-TCACTGGAGTCCAGGCGAGAAT-3′.

The cycling conditions were as initial denaturation at 95°C for 5 min, and then 36 cycles of denaturation at 95°C for 10 sec and annealing at 60°C for 30 sec. The relative mRNA expression was calculated using the comparative Ct (∆∆Ct) method.

Cells were seeded into 96-well culture plates (5×10³ cells/well). At 24, 48, 72 and 96 h, 10 µl CCK-8 reagent (Beyotime Biotechnology) was added to each well and incubated for 4 h. Then the absorbance values were read at a wavelength of 450 nm using a microplate reader (SpectraMax 250; GE Healthcare Life Sciences, Pittsburgh, PA, USA).

The in vitro migration ability of NSCLC cells was assessed by scratch assay. Cells were seeded in 6-well plates and the monolayer was scratched with 10-µl pipette tips. The wound areas were photographed 0 and 20 h after scratching and measured using a caliper. The wound-closure percentages were calculated using the following formula: [1-(current wound size/initial wound size)] × 100.

Cells were detached and re-suspended in a serum-free medium and seeded on the upper chamber of Matrigel-coated Transwell inserts with a pore size of 8 µm. The culture medium containing 10% FBS as a chemo-attractant was added to the lower chamber. After 24-h incubation, the cells on the upper surface of the insert were gently removed with a cotton swab. Invading cells (lower surface of the insert) were fixed with 4% paraformaldehyde (Sigma-Aldrich), stained with crystal violet, and counted under a microscope. Five random microscopic fields were examined for each insert.

Cells were seeded into 6-well plates at a density of 1×10⁶ cells/well for 24 h. Subsequently, the cells were collected and stained with the ANXA5 (Annexin V)-PE apoptosis detection kit (4A Biotech Co. Ltd., FXP018-100) according to the manufacturer's instructions and analyzed by flow cytometry (FACSCalibur, BD Bioscience, San Jose, CA, USA).

Unless stated otherwise, data are presented as mean ± SD in the figures. A Student's t-test was performed to compare the two groups of in vitro data. For more than two groups, we analyzed with one-way ANOVA followed by Tukey's multiple comparison test. All statistical tests were two-sided.

In order to explore the function of FSTL1 in NSCLC, we collected an array of lung cancer cells and lung normal epithelial cell line, BEAS-2B. Expression of FSTL1 was examined by qRT-PCR and western blot analysis. As shown in Fig. 1A, the mRNA levels of FSTL1 in NSCLC cells were much lower than normal BEAS-2B cells. Consistently, the protein level of FSTL1 in BEAS-2B was higher than NSCLC cells (Fig. 1B). These results suggest that FSTL1 is downregulated in NSCLC cells.

We then constructed FSTL1 overexpression in H446 cells. Both RT-PCR and western blot analysis revealed the successful establishment of FSTL1 overexpression (Fig. 1C and D). Then FSTL1 expression was knocked down in A549 cells. The results of qRT-PCR and western blot analysis shown, FSTL1 was effectively suppressed by SH1 and SH4 (Fig. 1E and F).

To analyze the function of FSTL1 in NSCLC cells, we examined the cell proliferation ability using CCK8. The results showed that A549 cells with FSTL1 knockdown proliferated faster than control cells (Fig. 2A). On the contrary, H446 cells with FSTL1 overexpression proliferated slower than control cells (Fig. 2B). In order to further clarify the function of FSTL1 in NSCLC cells, we examined the cell cycle of A549 cells with FSTL1 knockdown. As shown in Fig. 2C and D, the percentage of G1 phase in FSTL1 suppressed A549 cells was reduced while the percentage of G2 phase was elevated. However, the percentage of G1 phase in FSTL1 overexpressed H446 cells was increased while the percentage of G2 phase was reduced (Fig. 2E and F).

We also examined cell apoptosis after changing FSTL1 expression. As shown in Fig. 3A and B, the percentage of apoptotic cells in FSTL1-suppressed A549 cells was reduced significantly. However, the percentage of apoptotic cells in FSTL1 overexpressed H446 cells was increased (Fig. 3C and D).

Taken together, these results indicate that FSTL1 reduced NSCLC cell proliferation with altered cell cycle and elevated apoptosis.

To investigate the effect of FSTL1 on NSCLC cell migration and invasion, control cells, FSTL1 knockdown cells were submitted to scratch assay. Our results showed that the migration ability of A549 cells was significantly increased upon FSTL1 knockdown (Fig. 4A and B). However, the FSTL1 overexpression strongly reduced the migration ability of H446 cells.

Further, the Transwell assay showed that the invasion ability of A549 cells was significantly increased after FSTL1 knockdown (Fig. 5A and B). On the contrary, ectopic expression of FSTL1 reduced the invasion of H446 cells (Fig. 5C and D). Thus, our results demonstrated that FSTL1 suppressed NSCLC cell migration and invasion.

In order to elucidate the underlying mechanisms of FSTL1 mediated NSCLC cell apoptosis, we investigated the critical factors in cell apoptosis. We found a significant reduction of FASL, cleaved caspase-3 and −7, and upregulation of FAS and cleaved caspase-9 in FSTL1 knockdown A549 cells (Fig. 6A). On the contrary, in FSTL1 over expressed H446 cells, we observed an increase of FASL, cleaved caspase-3 and −7 (Fig. 6B) compared with control cells.

The important factors in cell migration and invasion were analyzed. MMP2, MMP3 and MMP9 were increased in FSTL1 knockdown A549 cells (Fig. 6C). However, MMP2, MMP3 and MMP9 were reduced in FSTL1 overexpressing H446 cells (Fig. 6D).

Taken together, our results demonstrated that FSTL1 might functioned as a tumor suppressor in NSCLC, including suppressing tumor cell proliferation, invasion and survival.