High expression of stratifin is a universal abnormality during the course of malignant progression of early-stage lung adenocarcinoma

Patients and tissue specimens

For subtraction analysis and quantitative RT-PCR, 41 frozen specimens of human lung adenocarcinoma were obtained from patients who had undergone surgical resection at the Department of Thoracic Surgery, Tsukuba University Hospital (Ibaraki, Japan), Tsukuba Medical Center Hospital (Ibaraki, Japan), and Ibaraki-higashi National Hospital (Ibaraki, Japan). A small amount of each specimen was embedded directly in Tissue-Tek OCT Compound (Sakura Finetek Japan, Tokyo, Japan) and frozen immediately in acetone and dry ice. The specimens were then stored at −80°C until analysis. Among the 41 specimens, we used five cases of type A from patients who survived, and five cases of type C from patients who had suffered lymph node metastasis or died, for cDNA microarray analysis, and the others were used for real-time RT-PCR. The backgrounds of these 10 patients are summarized in Supporting Information Table 1. For clinicopathological analysis using immunohistochemistry, 106 adenocarcinomas were also obtained from patients who had undergone surgical resection. All specimens were fixed with 10% formalin and embedded in paraffin.

Cell lines and culture conditions

Cell lines A549, PC-14, RERF-LC-KJ, and LC-2/ad were purchased from RIKEN Cell Bank (Ibaraki, Japan), and NCI-H23 and Calu-3 from the American Type Culture Collection. PL16T and PL16B were established in our laboratory from human lung AAH (atypical adenomatous hyperplasia) and human alveolar epithelium.5 A549 was maintained in D-MEM/F12 (Life Technologies Corporation, Carlsbad, CA) supplemented with 10% fetal bovine serum (FBS) (Sigma-Aldrich, St. Louis, MO), and 100 U/ml penicillin and streptomycin (Sigma-Aldrich). NCI-H23, PC-14, and RERF-LC-KJ were maintained in RPMI 1640 (Life Technologies) supplemented with 10% FBS, and 100 U/ml penicillin and streptomycin. Calu-3 was maintained in MEM (Sigma-Aldrich) supplemented with 10% FBS, and 100 U/ml penicillin and streptomycin. LC-2/ad was maintained in RPMI 1640/F12 (Sigma-Aldrich) supplemented with 15% FBS, 100 U/ml penicillin and streptomycin. PL16T and PL16B were maintained in MCDB153HAA (Wako, Osaka, Japan) supplemented with 2% FBS, 100 U/ml penicillin and streptomycin, 100 μg/ml EGF (Toyobo, Tokyo, Japan), 2.5 mg/ml Insulin (Wako), 600 μl/ml hydrocortisone (Wako), 10 mg/ml transferrin (Sigma-Aldrich), and 20 μg/ml sodium selenate (Sigma-Aldrich). All cells were cultured in a 5% CO₂ incubator at 37°C.

Extraction and amplification of RNA

For subtraction analysis, we used 10 frozen cryostat sections (8 μm thick) of types A and C. The tumor cells in each section were collected selectively using a laser capture microdissection system, LM-2000 (Arcturus Engineering, Mountain View, CA). Total RNA was isolated using a RNeasy Mini Plus Kit (Qiagen, Düsseldorf, Germany) in accordance with the manufacturer's protocol. The quality of the isolated total RNA was assessed by detection of 28S and 18S ribosomal RNA with an Agilent 2100 Bioanalyzer (Agilent Technologies, Waldbronn, Germany). Using 10 μl of sample buffer, T7 RNA polymerase promoter-attached, adaptor ligation-mediated, and PCR amplification followed by the in vitro T7-transcription (TALPAT) method was performed to amplify mRNA while maintaining the original gene expression ratio.6

cDNA microarray

Using the purified TALPAT product (ds cDNA) as a template, in vitro transcription was performed with biotin-labeled UTP, followed by conversion to biotin-labeled cRNA. The cRNA was purified using an RNeasy Mini Plus Kit (Qiagen) and the quality was assessed using an Agilent 2100 Bioanalyzer (Agilent Technologies) as described above. Fragmentation of cRNA, hybridization to a Human Whole Genome Bioarray (Applied Bioarrays, Tempe, AZ) including 54,841 probes, and washing of the array were performed in accordance with the manufacturer's protocols for the CodeLink™ iExpress Assay Reagent Kit (Applied Bioarrays). The signals on the arrays were detected with arrayWoRx^e (GE Healthcare UK, Little Chalfont, England). The data analysis was performed with Gene Spring GX7.3 (Agilent Technologies). MIAME-compliant data have been deposited in the MIAME Express database (Accession No. E-MEXP-2825).

Quantitative real time RT-PCR

Expression of genes that showed relative overexpression in type C tumors was evaluated again by the quantitative real-time RT-PCR method. Total RNA was prepared from another 31 frozen lung adenocarcinoma specimens using an RNeasy Mini Plus Kit (Qiagen) and the quality was evaluated using an Agilent 2100 Bioanalyzer (Agilent Technologies) as described above. One microgram of total RNA per 20 μl of reaction mixture was converted to cDNA using a High-capacity cDNA Reverse Transcription Kit (Life Technologies, Carlsbad, CA). Quantitative real-time PCR was performed with SYBR® Premix Ex Taq™ (Perfect Real Time; Takara Bio, Tokyo, Japan) on a GeneAmp® 7300 Sequence Detection System (Life Technologies) in accordance with the manufacturer's protocol. Primers used in this study are listed in Supporting Information Table 2.

Immunohistochemistry

Four-micrometer-thick sections were cut from 10% formalin-fixed, paraffin-embedded blocks. The deparaffinized and rehydrated sections were autoclaved in 10 mM citrate buffer (pH 6.0) at 121°C for 10 min for antigen retrieval, then incubated with monoclonal anti-SFN antibody diluted 1:40 (Immuno-Biological Laboratories, Gunma, Japan) for 30 min at room temperature. Subsequently, the sections were incubated with peroxidase-labeled polymer conjugated to goat anti-mouse IgG (DAKO, Carpinteria, CA) for 30 min at room temperature. Immunoreactivity was detected with a DAB substrate kit (Dako Japan, Kyoto, Japan), and the sections were counterstained with hematoxylin. The immunoreactivity was evaluated using a three-tier grading: negative (not stained), borderline (partially stained), positive (diffusely positive).

Pathway analysis

The PI3K-specific inhibitor LY294002 was purchased from Sigma (St. Louis, MO). The MEK1/2-specific inhibitor U0126 was from Merck (Whitehouse Station, NJ), and the JNK1/2-specific inhibitor SP600125 was from Biomol Research Laboratories. (Plymouth Meeting, MA). A549 cells carrying wild-type p53 and NCI-H23 cells harboring a p53 point mutation were cultured in 60-mm dishes (5 × 10⁵ cells/dish). Serum-starved cells were pretreated or untreated with LY294002, U0126, or SP600125 for 1 h. The phosphorylation states of PI3K, ERK and JNK were determined by Western blotting with phosphate-specific antibodies. Then, SFN expression was also determined by Western blotting.

Western blotting

Total cell lysates were prepared on ice with M-PER™ Mammalian Protein Extraction Reagent (Pierce) containing protease inhibitor cocktail (Sigma) and phosphatase inhibitor cocktail (Sigma). The lysates were centrifuged for 10 min at 4°C, and the insoluble fraction was discarded. The total protein in the soluble lysates was measured using a BCA protein assay kit (Pierce). Total protein aliquots (20 μg) were mixed with Laemmli sample buffer, denatured at 95°C for 5 min, and electrophoresed on 10% Tris-HCl gel (Bio-Rad Laboratories, Hercules, CA). Proteins were transferred to polyvinylidene difluoride membranes using an iBlot™ gel transfer system (Life Technologies Corporation). The blots were blocked and probed with the various antibodies. Antibodies used for Western blotting were obtained from various commercial sources as follows: SFN, GeneTex, (San Antonio, TX); phosphor-Akt (Ser473), Akt, phosphor-ERK1/2 (Thr202/Tyr204), and ERK1/2, Cell Signaling Technology, (Beverly, MA); phosphor-JNK1/2(Thr183/Tyr185) and JNK1/2, R&D Systems; β-actin, Pierce (Rockford, IL). After extensive washing, immunoreactivity was detected with specific secondary antibodies conjugated to horseradish peroxidase. Protein bands were visualized using SuperSignal West Dura® extended duration substrate (Pierce) and X-ray film (BioMax XAR film; Kodak, Rochester, NY).

Transfection using small interfering RNA (siRNA)

A549 cells were seeded at a density of 1.0–1.5 × 10⁶/mL and incubated overnight in antibiotics-free D-MEM/F12 supplemented with 10% FBS. On the following day, the cells were washed with phosphate-buffered saline (PBS), and then OPTI-MEM® reduced serum medium (Life Technologies) was added to the cells. SFN-specific siRNA (forward, 5′-ACCACGUU CUUAUAGGCUACUGAGA-3′; reverse,5′-UCUCAGUAGC CUAUAAGAACGUGGU-3′; Life Technologies) and a nucleic acid transferring agent, lipofectamine RNAiMAX (Life Technologies), were incubated together in OPTI-MEM® reduced serum medium for 15 min at room temperature to form an siRNA-lipofectamine complex. The medium containing the siRNA-lipofectamine complex was added to the cells to give a final siRNA concentration of 5 nM. Six hours later, the complex-containing medium was exchanged for antibiotics-free D-MEM/F12 supplemented with 10% FBS. The cells were incubated at 37°C in a CO₂ incubator for 24–120 h, and then further analysis was carried out.

Transfection of expression vectors

An entry clone encoding SFN (Ultimate Human ORF CLONE, IOH2986) was purchased from Life Technologies (Carlsbad, CA). With the entry clone and pcDNA-DEST47 Gateway Vectors (Life Technologies Corporation), we cloned SFN destination vectors (pDEST-SFN) using Gateway LR Clonase II Enzyme Mix (Life Technologies) in accordance with the manufacturer's protocol. The plasmid DNA was purified using an EndoFree Plasmid Maxi kit (Qiagen).

The day before transfection, A549 or PL16T cells were trypsinized and counted, and plated at 2 × 10⁵ cells per well into 6-well plates, so that they were 50–80% confluent on the day of transfection. Cells were plated in 2 mL of complete growth medium. For each well of cells to be transfected, 2.5 μg pDEST-SFN or pDEST-GFP DNA was diluted in 500 μL of OPTI-MEM® (Life Technologies) and incubated for 5 min at room temperature. Then, 6.25 μL of Lipofectamine LTX Reagent (Life Technologies) was added, followed by incubation for 30 min at room temperature. We added 0.5 mL of DNA-Lipofectamine LTX complex directly to each well, and performed gentle mixing by rocking the plate back and forth. The cells were incubated at 37°C in a CO₂ incubator for 24 h, after which we performed further analysis.

Cell proliferation assay

To analyze cell proliferation activity, a cell proliferation ELISA, BrdU (colorimetric) kit (Roche Diagnostics) was used in accordance with the manufacturer's protocol.

Invasion assay

Cell invasion potential was assessed using a cell invasion assay kit (Millipore) in accordance with the manufacturer's protocol. The kit had a system for evaluating the invasion of tumor cells through a basement membrane model. The assays were performed 48 h after siRNA transfection.

cDNA microarray

From the cDNA microarray data, we first eliminated the genes whose quality flags were L (near background; poor signal quality) in arrayWoRx analysis. Then we calculated the expression ratio of type C and type A, using the expression average of each type. Finally we selected 25 genes whose expression in type C tumors was more than two times higher than that in type A (type C/type A, >2, Supporting Information Table 3). These included some well-known cancer-related genes such as survivin, BIRC5, and MDK. These candidate genes were then classified by their functions, such as binding, transcriptional regulation, transport, and signal transduction. Among the functions of the various genes, most (12 genes) were classified as having a binding function. We therefore focused on those genes that played specific roles in the cell by binding to various molecules (e.g., nucleic acids, protein, and ion). With the aim of finding genes that were expressed specifically in type C tumors, we selected six genes whose expression in all five type A cases was lower than the expression average in type C cases. The final candidates were SCGB1A1, HTATIP2, IL1F7, SFN, HIST2H2AA and CP, as shown in Table 1.

Table 1. Comprehensive differential gene expression analysis using cDNA microarray

Quantitative real-time RT-PCR

To confirm previously identified differences in the expression levels of the genes selected by cDNA microarray, quantitative real-time PCR assays were carried out using the samples that had been used for microarray analysis (TALPAT products). These assays confirmed that all of the candidate genes were expressed at a higher level in type C tumors than in type A.

Moreover, quantitative real-time RT-PCR assays with 20 cases of lung adenocarcinoma mixed subtype with BAC and 11 cases of type A were performed to validate the difference in expression of these genes on a large scale. As most of the patients with type C tumors survived, we examined mixed-subtype adenocarcinomas with BAC, which were thought to be type C tumors at an advanced stage, in the assays. Among the six genes, SFN and CP showed significantly higher expression in adenocarcinoma mixed subtype than in type A (Fig. 1a) but the expression of the others was not significantly higher. CP is a copper-containing acute-phase protein primarily of hepatic origin, and has important antioxidant properties that protect lipids from peroxidation.7 We speculated that the high level of CP might reflect its property as an acute-phase reactant rather than a cancer-related function. Therefore, in this study, we focused on SFN and subjected it to further analysis.

Expression of SFN in lung adenocarcinoma cell lines and resected materials

First, we examined SFN expression in six lung adenocarcinoma cell lines and two immortalized cell lines by real time RT-PCR and Western blotting (Fig. 1b). A549 and PC14 cells showed high expression of SFN, but all of the six lung adenocarcinoma cell lines showed SFN expression to some extent. Conversely, we were unable to detect SFN expression in PL16T and PL16B, which were immortalized cell lines from AAH and normal bronchial epithelium. Secondly, we examined SFN expression in 106 small lung adenocarcinomas (<2 cm) by immunohistochemistry. Since SFN is localized mainly in the cytoplasm, we evaluated cytoplasmic immunoreactivity (Fig. 2). Normal bronchial tissue and lung parenchyma were completely negative for SFN. Among adenocarcinomas, 98% of type C, D, E and F cases showed positive staining in the cytoplasm, whereas only 14% of type A and B cases did so (Table 2, p < 0.001). It was of particular interest that in situ spreading cells of type C tumors were also immunopositive for SFN, similarly to the tumor cells in invasive areas (Fig. 2c).

Table 2. SFN immunoreactivity in small lung adenocarcinomas (less than 2 cm in diameter)

Function analysis of SFN

Proliferation assay and invasion assay were performed using A549 and PL16T cells after transfection with siRNA-SFN or SFN expression vectors. We found that transfection with siRNA-SFN significantly decreased the proliferation of A549 cells continuously for 120 h, although there was no apparent change in cell invasiveness (Figs. 3a and 3b). On the other hand, transfection with the SFN expression vector led to a significant increase in the proliferation of both A549 and PL16T cells (Fig. 3c).

Pathway analysis

Whereas SFN (14-3-3 sigma) is known to regulate cell cycle progression negatively as a p53 downstream factor, overexpression of SFN has been reported in several malignancies (colon and breast),8, 9 where the PI3K/Akt pathway has been shown to be one of the representative SFN upstream factors. To validate whether the PI3K/Akt pathway is related to SFN expression also in lung adenocarcinoma, we examined the signaling pathway of SFN in lung adenocarcinoma using pharmacological inhibitors specific to the pathways (LY294002). As SFN has been reported to be a downstream factor of the p53 gene, we used A549 that bears wild-type p53 and NCI-H23 that harbors a p53 point mutation. Serum-starved A549 or NCI-H23 cells were treated with the inhibitors (final concentration 25, 50 or 100 μM) and without them (DMSO at a final concentration of 100 μM as a negative control) for 1 h. As shown in Figure 4, LY294002 inhibited p-Akt, and we found that the PI3K inhibitor LY294002 decreased the expression of SFN in a dose-dependent manner in both cell lines (Fig. 4). However, neither U0126 (a MAPK pathway inhibitor) nor SP600125 (a JNK1/2 pathway inhibitor) affected SFN expression (data not shown). Therefore, the expression of SFN in A549 and NCI-H23 cells was shown to be mediated mainly via the PI3K/Akt pathway.