Human A549 NSCLC cells were purchased from the American Type Culture Collection (Rockville, MD, USA). Cells were maintained in RPMI-1640 supplemented with 10% fetal bovine serum and 1% antibiotic-antimycotic solution (Gibco; Thermo Fisher Scientific, Inc., Grand Island, NY, USA). For semi-solid Matrigel-embedded cell culture, a cell suspension (3×10⁵ cells in 350 µl of complete medium) was gently mixed with 150 µl of Matrigel (containing growth factors) (Corning Life Sciences, Bedford, MA, USA), with a final concentration of 30% v/v Matrigel. This was plated into a 24-well plate and incubated at 37°C for 30 min. Then, 500 µl of complete media was gently added to the well and the culture was maintained for 3 days. This diluted Matrigel polymerizes at 37°C to form a semi-solid gel that is soft enough to allow cells to spread as a monolayer beneath the gel. Cells were harvested from the semi-solid culture environment, for further analysis of cell behavior, by tapping the plate, aspirating the collapsed gel and washing with PBS. This was followed by conventional trypsinization. Cell morphology was examined by inverted phase-contrast microscopy. In all experiments, the cells were maintained at 37°C in a humidified atmosphere containing 5% CO₂.

Cell proliferation was determined using the Quant-iT™ PicoGreen^® dsDNA Assay kit (Invitrogen; Thermo Fisher Scientific, Inc., San Diego, CA, USA) to quantify DNA, according to the manufacturer's protocol. After harvesting cells from standard cell culture or semi-solid Matrigel-embedded cell culture, the cells were seeded into 96-well plates at a density of 1×10⁴ cells/well in 100 µl of culture medium. Cell proliferation was assessed on days 0, 2, 4, 7 and 9 after plating. Fluorescence-based dsDNA measurements were obtained using a SpectraMax fluorescence microplate reader (Molecular Devices, LLC, Sunnyvale, CA, USA), with an excitation wavelength of 485 nm and an emission wavelength of 520 nm.

Cell migration was assessed using a wound-healing assay. Wounds were generated by scratching confluent cultures of cells on plastic plates, or cells in semi-solid Matrigel (after removing the gel and washing with PBS), with a sterile 20-µl pipette tip. These wounds were then washed twice with PBS to remove cell debris, covered with complete medium or 30% v/v Matrigel, and incubated for a further 24 h. Wounds were marked and digitally photographed in the same field at 0 and 24 h after scratching. The wound area was assessed with ImageJ software, and the extent of healing was calculated as follows:

% Wound closure = (wound area at 0 h - wound area at 24 h)/wound area at 0 h × 100%

The invasiveness of cancer cells was evaluated using Matrigel-coated Transwell chambers (8 µm pore size). The upper chamber was seeded with 1×10⁵ cells in a 200-µl suspension, while 500 µl of conditioned medium prepared from human lung fibroblasts (MRC-5) was added to the lower chamber. After incubation for 24 h at 37°C, non-invaded cells on the upper surface of the insert membrane were removed with a cotton swab. Invaded cells on the lower surface of the membrane were fixed with 25% methanol, stained with crystal violet, and acid-extracted with 0.1 N HCl in methanol. The absorbance at 550 nm was measured.

The chemosensitivity of the cells was determined using an MTT assay for assessing metabolically active cells, as previously described (18). Briefly, the cell suspension was seeded into 96-well culture plates at a density of 1×10⁴ cells/100 µl/well. Then, 100 µl of vehicle (cell culture medium) or various concentrations of cytotoxic agents were added, and the plates were incubated for 48 h. Cell culture medium containing MTT (Sigma-Aldrich, St. Louis, MO, USA) was added to each well and incubated for 2 h. The number of viable cells was assessed by determining the absorbance at 550 nm and subtracting the absorbance at 650 nm (reference wavelength). The IC₅₀ value of each drug, defined as the concentration of the drug that results in a 50% decrease in cell viability, was extrapolated from a concentration-dependent curve of the drug. Assays were performed in quadruplicate, and data were expressed as the percentage of viability relative to the control.

Cell cycle distribution was analyzed using a Muse™ Cell Analyzer (Merck Millipore, Hayward, CA, USA). The Muse™ Cell Cycle Assay kit (Merck Millipore) was used according to the manufacturer's protocol. Briefly, cell pellets from each condition were washed three times with PBS and fixed in ice-cold 70% ethanol overnight at −20°C prior to staining. Cells were washed again with PBS, stained with 200 µl of Muse™ Cell Cycle reagent, incubated in the dark for 30 min at room temperature and then processed for cell cycle analysis. Results were expressed as a percentage of the cells in the G0/G1, S and G2/M phases based on differential DNA content. Experiments were performed in triplicate.

Gene expression levels were determined by RT-qPCR. Cells were harvested and total RNA isolation was performed using the RNeasy Mini kit (Qiagen, Inc., Valencia, CA, USA), following the manufacturer's instructions. Approximately 2 µg of total RNA was reverse transcribed to cDNA with SuperScript^® III Reverse Transcriptase (Invitrogen; Thermo Fisher Scientific, Inc.). RT-qPCR was performed using KAPA SYBR^® FAST qPCR kits (Kapa Biosystems, Inc., Woburn, MA, USA) and a StepOnePlus™ Real-Time PCR system (Applied Biosystems; Thermo Fisher Scientific, Inc., Foster City, CA, USA) with the following steps: 95°C for 10 min, followed by 40 cycles at 95°C for 15 sec, 60°C for 30 sec and 72°C for 30 sec. All primers used to amplify genes are listed in Table I. The level of gene expression relative to an internal control, RPS13 (ribosomal protein S13), was determined using the 2^−ΔΔCq method (19).

The expression levels of the signaling proteins were determined by western blot analysis. Cells cultured on plastic plates or embedded in semi-solid Matrigel were harvested, and then lysed in cell signaling lysis buffer (Merck Millipore) containing protease inhibitors and phosphatase inhibitors (1 mM Na₃VO₄, 10 mM NaF and 20 mM β-glycerophosphate). Cell extracts were resolved by 10% SDS-PAGE and transferred to Immobilon-P Transfer Membranes (EMD Millipore, Bedford, MA, USA). After blocking with 3% bovine serum albumin in Tris-buffered saline containing 0.1% Tween-20 at room temperature for 1 h, the membranes were probed with the following primary antibodies overnight at 4°C: Akt (1:1,000; rabbit, polyclonal; cat. no 9272), phospho-Akt (1:500; rabbit, polyclonal; cat. no. 9271), ERK (1:4,000; rabbit, polyclonal; cat. no. 9102), phospho-ERK (1:10,000; rabbit, polyclonal; cat. no. 9101), FAK (1:2,000; rabbit, polyclonal; cat. no. 3285), phospho-FAK (1:1,000; rabbit, monoclonal; cat. no. 8556), STAT3 (1:3,000; rabbit, polyclonal; cat. no. 9132), phospho-STAT3 (1:500; rabbit, monolonal; cat. no. 9145), p38 (1:1,000; rabbit, polyclonal; cat. no. 9212), phospho-p38 (1:500; rabbit, polyclonal; cat. no. 9211), p65 (1:1,000; rabbit, polyclonal; cat. no. 3034), phospho-p65 (1:500; rabbit, polyclonal; cat. no. 3031), cyclin D1 (1:1,000; rabbit, polyclonal; cat. no. 2922), p21 (1:2,000; rabbit, monoclonal; cat. no. 2947), Bcl-xL (1:10,000; rabbit, monoclonal; cat. no. 2764), Bax (1:5,000; rabbit, polyclonal; cat. no. 2772), IκB (1:1,000; rabbit, monoclonal; cat. no. 4812) (all from Cell Signaling Technology, Danvers, CA, USA), uPAR (1:500; rabbit, polyclonal; cat. no. sc-10815; Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA), MYLK (1:500; mouse, monoclonal; cat. no. sc-365352; Santa Cruz Biotechnology, Inc.) and GAPDH (1:50,000; rabbit, monoclonal; cat. no. ab190480; Abcam, Cambridge, MA, USA). Following incubation with the appropriate horseradish peroxidase-conjugated secondary antibody (1:5,000; goat, monoclonal; cat. no. 7074; Cell Signaling Technology, Danvers, CA, USA; or 1:5,000; rabbit, polyclonal; cat. no. P0260; Dako; Agilent Technologies, Inc., Santa Clara, CA, USA), the protein bands were visualized using enhanced chemiluminescence reagents.

All experiments were performed in triplicate, and the data are presented as the mean ± standard deviation. Statistical analysis was performed using the Student's t-test. P<0.05 was considered to indicate a statistically significant difference.

When A549 cells plated on plastic dishes or embedded in semi-solid Matrigel were examined by inverted phase-contrast microscopy, the cells grown within the Matrigel matrix had undergone morphological changes by 24 h after plating (Fig. 1), transforming from their typical shape to branched cells. This altered morphology could be reversed by replating the semi-solid Matrigel embedded cells back into plastic plates (data not shown).

Proliferation of A549 cells cultured on plastic plates or embedded in semi-solid Matrigel was examined on days 0, 2, 4, 7 and 9 via DNA quantification using a PicoGreen DNA assay. The data revealed that the growth dynamics differed between the two cell populations; proliferative and quiescent growth characteristics were observed. Proliferative growth characteristics were observed in cells cultured on plastic plates, while quiescent growth was observed in cells cultured in semi-solid Matrigel (Fig. 2). A549 cells grown within the Matrigel matrix exhibited a small increase in the number of cells during the 9 days of culture, while the number of the cells grown on plastic plates had increased 4-fold by day 4.

To assess possible mechanisms underlying Matrigel-mediated cell growth inhibition, the cell cycle distribution of A549 cells cultured on plastic plates and in semi-solid Matrigel was investigated. The results demonstrated that cells embedded in semi-solid Matrigel exhibited a higher G0/G1 population compared with cells cultured on plastic plates (80.1±2.2 vs. 60.1±1.5%; Fig. 3A and B). Consistent with this result, Matrigel embedding caused a decrease in the proportion of cells in the S phase (6.9±0.8%) and G2/M phase (12.7±1.7%) of the cell cycle compared with cells grown on plastic plates (17.5±0.4 and 22.1±1.0%, respectively). These data revealed that Matrigel embedding induced cell cycle arrest (cellular dormancy) at the G0/G1 phase, resulting in growth inhibition of A549 cells.

Drug sensitivities of A549 cells cultured on plastic plates and embedded in semi-solid Matrigel were explored. Multiple anticancer drugs commonly used to treat a variety of cancers were used in this study. The harvested cells from both culture conditions were treated with various concentrations of the drugs for 48 h, then MTT assays were performed to determine the survival rates. The drug concentration causing a 50% decrease in the number of cells (IC₅₀) was extrapolated. As presented in Table II, the results demonstrated that cell culture in semi-solid Matrigel caused a marked increase in IC₅₀ for anticancer drugs that target actively dividing cells [etoposide (10.6-fold), paclitaxel (5.7-fold), vinblastine (4.4-fold), doxorubicin (4.1-fold) and 2-deoxy-D-glucose (3.9-fold)], as compared with cells cultured on plastic plates. This indicated a higher degree of chemoresistance in cells cultured in semi-solid Matrigel. However, the responses of the two cell populations to cytotoxic agents that are not related to cell proliferation rate were similar, such as curcumin (1.5-fold increase in IC₅₀ in cells cultured in Matrigel) and Bay 11–7085 (0.9-fold increase; Table II). These results indicated that the chemoresistance of semi-solid Matrigel-embedded cells may be the result of a decreased rate of cell proliferation.

A Transwell invasion assay was performed to assess the effect of semi-solid Matrigel embedding on cell invasiveness. A549 cells were cultured on plastic plates or in semi-solid Matrigel for 72 h, then the cells were harvested and subjected to an invasion assay. Photography and quantitative analysis of the results revealed a significant decrease in the invasion rate of cells cultured in semi-solid Matrigel (Fig. 4). The number of invading cells was 41% lower among cells cultured in semi-solid Matrigel, as compared with those grown on plastic plates (P<0.001).

A wound-healing assay was performed to investigate the effect of semi-solid Matrigel embedding on cell motility. Cell migration was expressed as a percentage of wound closure. The results demonstrated that, at 24 h, ~50% of the gap had closed for A549 cells grown on plastic plates and covered with culture medium, while only 30% of the initial gap had closed for cells grown on plastic plates and covered with semi-solid Matrigel (Fig. 5A). For A549 cells grown within semi-solid Matrigel, only 25 and 10% of the gap had closed when covered with culture medium and semi-solid Matrigel, respectively (Fig. 5B). These results revealed that A549 cells cultured on plastic plates exhibited higher motility compared with cells grown within semi-solid Matrigel in both conditions. In addition, cell motility was decreased in the presence of Matrigel.

It has previously been reported that gene expression is altered upon cell contact with ECM proteins (4). Therefore, we investigated the consequences of semi-solid Matrigel embedding on the expression patterns of genes associated with invasion (MMP2, MMP7, MMP9, MMP11, MMP13, HGF and CXCR4) and drug resistance (MDR-1 and MRP3). In addition, genes that are associated with tumor progression and metastasis [uPA and uPA receptor (uPAR)] were analyzed. RT-qPCR was performed to determine the level of gene expression. Gene expression was normalized against RPS13 expression. Using a threshold value of 3-fold expression change, the results revealed that A549 cells cultured in semi-solid Matrigel exhibited decreased expression levels of uPA, uPAR and multiple genes associated with invasion (MMP2, MMP7, MMP9 and CXCR4), as compared with cells cultured on plastic plates. However, no notable differences were observed in the expression levels of drug resistance genes, MDR-1 and MRP3 (Fig. 6).

Protein expression levels of signaling molecules associated with cell proliferation, metastasis, chemoresistance and apoptosis were assessed. Western blot analysis of proteins extracted from A549 cells indicated that cells grown within semi-solid Matrigel had reduced expression of p-ERK, p-Akt and p-STAT3 (positive regulators of cell growth, proliferation and survival), compared with cells cultured on plastic plates (Fig. 7). Bcl-xL, an anti-apoptotic protein, was up-regulated in cells cultured in semi-solid Matrigel, while a decreased level of the pro-apoptotic protein, Bax, was detected (Fig. 7). An increased level of p-p65 (NF-κB) protein and decreased levels of IκB were observed in cells cultured in semi-solid Matrigel (Fig. 7), suggested that NF-κB signaling was activated in the Matrigel-embedded cells. Furthermore, the upregulation of p-p38 and p21 was observed in cells cultured in semi-solid Matrigel, while the expression of p-FAK, cyclin D1, uPAR and myosin light-chain kinase (MYLK) was decreased, compared with cells cultured on plastic plates (Fig. 7). These findings revealed that changes in the expression of signaling molecules in cells cultured in semi-solid Matrigel shifted the cells from a highly proliferative to a less proliferative state, and also enhanced apoptotic resistance.

Specific kinase inhibitors of MEK1/2 (U0126), PI3K (LY-294002) and JAK2 (AG-490) were used to investigate whether these signaling pathways are associated with Matrigel-mediated dormancy and chemoresistance. First, the effects of U0126, LY-294002 and AG-490 on cell viability were evaluated. A549 cells were treated with various concentrations of U0126, LY-294002 and AG-490, after which an MTT assay was performed to determine cytotoxicity following treatment. Inhibitor concentrations that resulted in >80% cell viability were used. Pretreatment for 24 h with 30 µM of U0126, 20 µM of LY-294002 and 30 µM of AG-490 did not significantly affect A549 cell viability (data not shown). Western blot analysis revealed decreased phosphorylation of ERK1/2, Akt and STAT3 following treatment with 30 µM of U0126, 20 µM of LY-294002 and 30 µM of AG-490, respectively, for 24 h (data not shown). Cell cycle analysis revealed that blockade of ERK1/2 and Akt with U0126 and LY-294002 resulted in G0/G1 phase cell cycle arrest, similar to that observed in cells embedded in semi-solid Matrigel (Table III). Treatment with AG-490 had no notable effect on the cell cycle distribution of A549 cells (Table III).

To determine whether the decrease in phosphorylation of ERK1/2, Akt and STAT3 mediates chemoresistance of semi-solid Matrigel-embedded cells, chemosensitivity to etoposide and Bay 11–7085 was evaluated in the presence of the specific kinase inhibitors. Etoposide and Bay 11–7085 were used as representative examples of drugs that are associated and not associated with, respectively, the cell proliferation rate. The cells were pre-incubated with the kinase inhibitors for 24 h prior to and during the chemosensitivity assay. Comparing the effects on chemosensitivity, U0126 and LY-294002 were determined to increase etoposide resistance, similar to that observed in cells cultured in semi-solid Matrigel (the IC₅₀ value of cells treated with U0126 and LY-294002 was 299±23.3 µM and 243.3±16.5 µM, respectively; Fig. 8). However, the chemoresistance to etoposide was unaffected by STAT3 inhibition with AG-490 (the IC₅₀ value of cells pre-treated with AG-490 was 30.8±1.2 µM). For Bay 11–7085, treatment with U0126, LY-294002 and AG-490 exhibited no effect on chemosensitivity (the IC₅₀ value of cells treated with U0126, LY-294002 and AG-490 were 19.0±0.5, 19.4±0.1 and 17.0±1.1 µM, respectively; Fig. 8). These findings were in agreement with those of the cell cycle distribution experiment, suggesting that inhibition of ERK1/2 and Akt with U0126 and LY-294002 results in cell dormancy (Table III), which leads to enhanced chemoresistance for the drugs that target actively proliferating cells. These data revealed that Matrigel-mediated growth inhibition of A549 cells and protection from chemotherapy-induced apoptosis occurred predominantly via the MAPK and Akt pathways.