Detailed information on the materials is listed in Supplementary Table S1.

A549 cells, H1299 cells, and H460 cells were purchased from ATCC. The cells were cultured in RPMI 1640 medium supplemented with 10% fetal bovine serum (FBS) at 37°C in a humidified atmosphere at 5% CO₂, and treated with PF-573228 (TOCRIS, Bristol, UK) at concentrations of 0, 0.1, 1, or 10 μM.

The cells were harvested and lysed in 1x RIPA buffer (Merck, Darmstadt, Germany) containing protease and phosphatase inhibitors. The protein concentration was determined using a Bio-Rad DC protein assay kit (Bio-Rad, California, USA). For Western blot analysis, 30 μg of total protein was applied to SDS-PAGE and transferred to a PVDF membrane. The membranes were blocked in 5% skim milk for 2 h in TBST buffer (20 mM Tris-Cl, 150 mM NaCl, 0.1% Tween 20, pH 7.4). After blocking, the membranes were probed with the primary antibody overnight. Antibodies against FAK, p-FAK, cyclin B1, p53, and lamin A/C were used in immunoblotting. The given protein bands were identified by horseradish peroxidase-conjugated secondary antibodies and developed with an enhanced chemiluminescence solution.

The cells were harvested and washed with PBS buffer, and then fixed in 70% (v/v) ethanol. The fixed cells were stained with propidium iodide solution and injected into an Attune NxT Flow Cytometer (Life Tech, California, USA) to analyze the cell cycle profile.

A Leica DMi8S epifluorescence microscope (Wetzlars, Germany) equipped with an X-Cite XCT10A (Lumen Dynamics, Wiesbaden, Germany) light source, filters and objectives (10x, 20x, 40x, and 63x) was used to observe fluorescent signals in the cells. In addition to epifluorescence, confocal images were captured using an OLYMPUS FV1000 confocal laser scanning microscope equipped with a light source, filters and objectives (10x, 20x, 40x, 63x, and 100x). Cells were seeded on 12-mm coverslips in a 24-well culture plate. The cells were harvested and fixed in 4% paraformaldehyde in PBS for 10 min, and permeabilized in 0.5% Triton in PBS for 5 min. After fixation, the cells were subjected to immunofluorescent staining with antibodies recognizing FAK and emerin. Phalloidin-TRITC was used as an additional reagent. Cell nuclei were stained with 0.2 μg/mL 4′, 6-diamidino-2-phenylindole (DAPI).

The cells were fixed with 4% paraformaldehyde for 15 min. After fixation, acidic β-galactosidase (SA-β-gal) was assayed in senescence assay buffer (1 mg/mL 5-bromo-4-chloro-3-indolyl β D-galactopyranoside (X-gal), 5 mM K₃Fe(CN)₆, 5 mM K₄Fe(CN)₆, 2 mM MgCl₂, 150 mM NaCl, 40 mM citric acid, and 40 mM Na₂HPO₄ at pH 6.0) in the dark at 37°C for 16 h. SA-β-gal activity was detected based on SA-β-gal-hydrolyzed X-gal, which produces a blue color. All chemical reagents were purchased from Sigma-Aldrich.

The cells were trypsinized, resuspended in 1xPBS, and stained with trypan blue (Sigma-Aldrich). The number of cells was counted with a hemocytometer.

Lentivirus-associated plasmids encoding luciferase, and FAK short hairpin RNA (shRNA) were obtained from the National RNAi Core Facility of Academia Sinica, Taiwan. The production and infection of lentiviruses were performed according to the guidelines of the National RNAi Core Facility.

The experimental data were digitized and analyzed. Data are presented as the mean ± the standard error of the mean (SEM). One-way ANOVA was used to compare digitized data and measurements from independent experiments in two or more groups, and the Student's T-test was used to compare two independent samples. A p < 0.05 was considered to indicate a statistically significant difference.

Focal adhesion signaling is involved in cell proliferation, and FAK plays a key role in the focal adhesion complex that relays focal adhesion signals to the cell proliferation program (9, 15). Given the role of FAK signaling in tumor growth and metastasis, we hypothesized that inhibiting the catalytic activity of FAK may disrupt FAK signaling and blunt tumor cell proliferation. Therefore, we treated three distinct non-small cell lung cancer cell lines (A549 lung adenocarcinoma cells and H460 and H1299 large cell carcinoma cells) with PF-573228, an enzymatic inhibitor of FAK. PF-573228 was administered to the lung cancer cells for 4 days at three doses: 0.1, 1, or 10 μM. The growth curves showed that 10 μM PF-573228 effectively induced cessation of cell growth (Figures 1A–C).

We then examined the expression level of the cell cycle regulator cyclin B1, which has been reported to be a downstream effector of FAK signaling. Western blot analysis showed that cyclin B1 expression levels were much lower after the cells were exposed to 10 μM PF-573228 (Figures 1D–F). To further characterize the effect of PF-573228 treatment on cell cycle progression, we analyzed the cell cycle distribution using flow cytometry analysis. The results showed that a low PF-573228 concentration had little influence on cell cycle progression (Figures 1G–I), whereas a high PF-573228 concentration (10 μM) halted cell cycle progression at the G2/M transition (Figures 1J,K). This showed that PF-573228 treatment effectively suppressed multiplication of lung cancer cells.

Since phosphorylation of FAK at Tyr-576 and Tyr-577 (p-FAK) represents enzymatic activation of FAK (37), an antibody against p-FAK was used to confirm the kinase activity of FAK and verify the effect of PF-573228 on FAK inactivation. FAK activity was practically blocked by 10 μM PF-573228 in A549 cells (Figures 2A,C). To further confirm the inactivation of FAK by PF-573228 treatment, we also examined the phosphorylation of tyrosine 397 in FAK (pTyr-397). The results showed that the intensity of pTyr-397 was decreased after PF-573228 treatment (Figure S1). FAK is a key regulator of integrin signaling for focal adhesion assembly (38, 39). In addition to FAK inhibition, PF-573228 has been shown to perturb integrin-based signaling for focal adhesion maturation (36).

Treatment of the lung cancer cells with PF-573228 resulted in failure of FAK activation, and translocation to focal adhesion was observed in immunofluorescent imaging (Figure 2B). When the cells were cultured in PF-573228-free medium, more FAK translocated to focal adhesions, which appeared as plaque-like patterns in the cell periphery that formed at the tips of stress fibers, as visualized in cells stained with phalloidin-labeled F-actin and an antibody against FAK (Figure 2B). By contrast, in the A549 cells treated with PF-573228, only a few FAK molecules translocated to focal adhesions, and tiny FAK-based focal adhesions formed at the tips of F-actin bundles, indicating failure of focal adhesion maturation (Figure 2B). The sizes of focal adhesions were measured and the areas of FAK at the tips of the actin stress fibers were digitized using Image Pro software. The sizes of focal adhesions ranged from 12 to 28 μm² in the cells without PF-573228 treatment, whereas the extent of FAK-based focal adhesion was approximately 3–12 μm² after PF-573228 treatment (Figure 2D).

In the absence of PF-573228, most cells contained oval or round nuclei, as visualized by DAPI staining (Figure 2B). Interestingly, a distorted nuclear morphology was observed in the A549 cells treated with PF-573228 (Figure 2B). As DAPI staining was insufficient to clearly visualize the nuclear appearance in detail, an antibody against emerin (40), a nuclear inner membrane protein, was used to visualize the nuclear shape in the PF-573228-treated A549 cells. Emerin staining revealed that the cells without PF-573228 treatment harbored oval-like nuclei. By contrast, most nuclei were larger and had irregular shapes with invagination in the cells upon exposure to 10 μM PF-573228 (Figure 3A).

Lamin A/C is the nuclear skeleton responsible for maintaining and stabilizing the nuclear architecture (41–44). Because interrupting FAK signaling resulted in nuclear deformity (Figure 2B), changes in lamin A/C expression levels were assessed. The effects of FAK signaling on the expressions of lamin A/C and other nuclear skeletal proteins inferred that the nuclear deformity caused by PF-573228 was attributable to changes in lamin A/C expression.

To investigate changes in lamin A/C expressions in cells exposed to PF-573228, PF-573228-treated cells were harvested and subjected to Western blot analysis. Inhibition of FAK activity led to lower p-FAK levels (Figures 3B,C) and deformed nuclei in the lung cancer cells (Figure 3A). To quantify the expressions of lamin A and C, the intensities of their protein bands were normalized to β-actin (Figure 3B). The expressions of lamin A and C were much lower in the A549, H460, and H1299 cells treated with PF-573228 compared to those without PF-573228 treatment (Figure 3B and Figures S2A,B). The lamin A and C band intensities were quantified and plotted in bar charts. The ratios of lamin A/β-actin and lamin C/β-actin in A549 cells exposed to PF-573228 were reduced by one third and one half, respectively, compared to A549 cells without PF-573228 treatment (Figures 3D,E). Similar trends of downregulation of lamin A and lamin C by PF-573228 treatment were also detected in the two other lung cancer cell lines (Figures S2A,B).

Mutant LMNA, mutations that affect lamin A/C expression, and lamin A/C depletion in cells have been associated with premature aging and cellular senescence (8, 30, 32, 42, 45). Based on the concurrent lamin A/C downregulation and nuclear deformity observed in lung cancer cells exposed to PF-573228 (Figures 3A,B), we examined the development of cellular senescence in lung cancer cells treated with PF-573228. The SA-β-gal activity in cells was assayed by in situ staining using the chromogenic substrate X-gal, which colored SA-β-gal-positive cells blue. As noted in Figure 4A, blue cells were clearly visible in the cells treated with PF-573228 (Figure 4A), whereas a sporadic distribution of blue-colored cells was observed in the cells without PF-573228 treatment (Figure 4A). The bar chart in Figure 4B shows that nearly 90% of the cells exposed to a higher dose of PF-573228 were positive for SA-β-gal, compared to ~20% of the cells exposed to a lower dose of PF-573228, and ~1% of the cells without PF-573228 treatment.

Disruption of FAK signaling by PF-573228 caused cellular senescence. However, the mechanisms by which inhibition of FAK signaling affects senescence programming remain unclear. Cellular senescence in chemotherapy-affected cancer cells has been observed in several studies (24, 29, 46). In addition, clinical studies have reported that p53 plays a role in the development of cellular senescence in chemotherapy-affected cancer cells (46, 47). p53 is known to be a transcription factor in programed senescence and cell cycle arrest (48), and it may play a similar role in the cellular senescence program in lung cancer cells exposed to PF-573228 as in cells in which FAK signaling is interrupted.

To investigate whether or not p53 plays a role in PF-573228-induced cellular senescence, p53 expression levels were examined in PF-573228-treated cells. Western blot analysis showed that p53 expression levels increased significantly by more than 3-fold compared to cells without PF-573228 treatment and cells treated with a low concentration of PF-573228 (Figures 4C,D, and Figure S3).

FAK is not the only molecule targeted by PF-573228 (36). Although FAK enzymatic activity was blocked by PF-573228 administration, off-target effects could also have turned off other kinases, for example, cyclin-dependent kinases. Therefore, signaling perturbations of other kinases may have caused the pathogenic senescence in the lung cancer cells.

If FAK has an anti-senescence effect, FAK depletion would cause anti-senescence to fail and escalate senescence programing. To clarify the role of loss of FAK signaling in the development of cellular senescence and nuclear deformity with changes in lamin A and lamin C expressions, we used an shRNA targeting FAK to deplete the expression of FAK in lung cancer cells. After introducing shFAK into lung cancer cells, cells harboring shFAK were selected. To assess FAK knockdown, two shFAK clones were selected for Western blot analysis and senescence assays, which showed that shFAK successfully caused FAK depletion in A549, H460, and H1299 cells (Figure 5A and Figure S4A). In addition, the impact of FAK depletion on the downregulation of lamin C and cyclin B1 and upregulation of p53 was validated (Figure 5A and Figure S4A). We also assessed nuclear appearance using emerin staining, which revealed that A549 cells without FAK depletion harbored oval-like nuclei. By contrast, the ratio of nuclei harboring slightly larger and irregular shapes was increased in the cells with FAK depletion (Figure 5B). Similar results were obtained in the H460 and H1299 cells (Figures S4B,C). In addition, lung cancer cells harboring shLuc or shFAK were subjected to senescence assays, which revealed more SA-β-gal-positive A549 cells harboring shFAK (Figure 5C). By contrast, few SA-β-gal-positive cells were visible in those harboring shLuc. The SA-β-gal-positive cells represented ~0.7% of the shLuc A549 population. In the two shFAK cell clones, SA-β-gal-positive cells represented ~11 and 15%, respectively (Figure 5D). Similar results were observed in H460 and H1299 cells (Figures S4D–F). However, H460 cells grew in single and multiple layers (Figure S4D), and it was difficult to measure the ratio of SA-β-gal-positive cells.

Aging and cellular senescence are often present in replicative failure, oncogenic induction, and telomere shortening (8, 34, 49). In clinical cases, chemotherapy or radiation has been shown to induce cellular senescence, and DNA damage or genomic instability is thought to be the pathological cause. Therapy-induced senescence can be classified as replicative senescence (24, 27, 33, 48), and replicative senescence is able to cease tumor growth (47, 50). However, this cellular senescence is reversible (31, 51, 52). Senescent cells expressing low levels of p16 have been shown to reversibly exit senescence when p53 expression levels fall (52).

In this study, FAK signaling downregulated the expression of p53, and inhibition of FAK signaling upregulated the expression of p53 in A549 cells (Figures 4C,D, 5A). The A549 cells that entered senescence due to PF-573228 administration exhibited regrowth and return to a non-senescent state after withdrawal of PF-573228 from the culture (Figures 6A–C). The proliferative activity of A549 cells was based on cyclin B1 expression after the cells were incubated in PF-573228-free medium (Figure 6D).

We then tested whether senescence in lung cancer cells with FAK inhibition was reversible or irreversible. The three lung cancer cells lines were cultured in medium containing 10 μM PF-573228 to induce cell senescence in a 5-day induction course, after which the expressions of lamin A and lamin C decreased (Figure 7A). After the induction of senescence, PF-573228 withdrawal was scheduled over 6 days. Lamin A and lamin C expressions in A549 cells and H1299 cells gradually recovered after PF-573228 withdrawal (Figures 7A,B). By contrast, lamin A and lamin C levels in H460 cells remained lower when senescent H460 cells were incubated in PF-573228-free medium (Figures 7A,B). After PF-573228 withdrawal, senescent A549 cells escaped from senescence, as the SA-β-gal-positive A549 cell population declined to nearly half by the fifth day of PF-73228 withdrawal (Figure 8A).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.