The present study was approved by the Medical Ethics Committee of Kunming University of Science and Technology (Kunming, China). Human samples were used in accordance with the requirements of Medical Ethics Committee of Kunming University of Science and Technology, under the guidelines of the World Medical Assembly (Declaration of Helsinki). Written informed consent was obtained from the patients' families.

Lung specimens (n=50) were obtained from the tumor and an adjacent non-cancerous area ≥6 cm from the tumor tissues of 50 patients with lung cancer from Yunnan Province at the First People's Hospital of Yunnan Province (Kunming, China) between April 2014 and January 2015, as previously described (16,17). The non-neoplastic tissue was confirmed to lack tumor cell infiltration using histological analysis. The tissues were immediately placed in liquid nitrogen and stored at −80°C until use.

RNA extraction and first-strand cDNA synthesis was assessed as previously described (18). For quantitative PCR (qPCR) of HSF2, the following primers were used: HSF2 forward, 5′-AAGTTCAGGCAGTGATGGCA-3′ and reverse, 5′-TGCACAGAACTAGTGAAAAGATCA-3′; glyceraldehyde 3-phospahte dehydrogenase forward, 5′-GAAGGTCGGAGTCAACGGAT-3′ and reverse, 5′-GAGGGATCTCGCTCCTGGAAG-3′. qPCR was assessed using a continuous fluorescence detector (Opticon Monitor; Bio-Rad Laboratories Inc., Hercules, CA, USA) and PCR was assessed using an SYBR Green qPCR kit according to the manufacturer's protocol (Tiangen Bio, Inc., Beijing, China) with the following reaction conditions: Initial denaturation at 95°C for 1 min followed by 40 cycles at 95°C for 15 sec, 60°C for 15 sec and 72°C for 20 sec. Fluorescence curve analysis was assessed using Opticon Monitor software. The identities of qPCR products were confirmed by DNA sequencing. The relative gene expression in 2 different samples were compared using the 2^−ΔΔCq method (19).

Western blot analysis of the tissues and the cells were performed as previously described (16,20,21). Tissue and cell samples were homogenized in radioimmunoprecipitation assay buffer containing a protease inhibitor cocktail (Roche Applied Science, Penzburg, Germany). Samples (containing 50 µg of protein) were loaded onto sodium dodecyl sulfate-polyacrylamide electrophoresis gel, electrophoresed and then electro-transferred onto a polyvinylidene fluoride membrane. The membrane was subsequently blocked with 3% bovine serum albumin and incubated with a rabbit anti-human HSF2 polyclonal antibody (dilution, 1:1,000; cat. no. SC-13056; Santa Cruz Biotechnology, Inc., Dallas, TX, USA) and a horseradish peroxidase-conjugated goat anti-rabbit IgG secondary antibody (dilution, 1:5,000; cat. no. SC-2004; Santa Cruz Biotechnology, Inc.). Protein bands were visualized using Super Signal reagents (Thermo Fisher Scientific, Inc., Waltham, MA, USA). For detection of HSP27, HSP47, HSP70 and HSP90, the same methods were used. The antibodies were purchased from Santa Cruz Biotechnology, Inc and listed as follows: Rabbit anti-human Hsp27 polyclonal antibody (dilution, 1:1,000; cat. no. SC-9012); rabbit anti-human Hsp47 polyclonal antibody (dilution, 1:1,000; cat. no. SC-8352); rabbit anti-human Hsp70 polyclonal antibody (dilution, 1:1,000; cat. no. SC-25837); and rabbit anti-human Hsp90 polyclonal antibody (dilution, 1:1,000; cat. no. SC-7947).

The cDNA of HSF2 was amplified from normal lung tissue using an RNA extraction kit (Tiangen Bio, Inc.) and the first strand cDNA sythesis kit (Takara Biotechnology Co., Ltd., Dalian, China). The PCR primers used were as follows: Forward, 5′-CGCGTTCGGGTGTAGAATTT-3′; and reverse, 5′-CATCCCACCCCCGATCTTTC-3′. The cDNA with XhoI and BamHI restriction sites and flag tag at N-terminal were amplified from the HSF2 cDNA with the following primers: Forward, 5′-AGATCTCGAGCCTGCGCCGCGTTAACAATGA-3′; and reverse, 5′-AGATGGATCCTTACTTATCGTCGTCATCCTTGTAATCTTAGCTATCAATAAGTGGCAT-3′ (bold and underlined nucleotides are the restriction site of the enzymes). The PCR products and the plasmid pIRES2-EGFP were digested with XhoI and BamHI (Takara Biotechnology Co., Ltd.) and ligated with T4 ligase (Takara Biotechnology Co., Ltd.) and transformed into Escherichia coli DH5a competent cells (Tiangen Bio, Inc.). The positive clones were confirmed by sequencing, then the plasmid was amplified and purified for cell transfection.

The BEAS-2B and A549 cells were cultured in DMEM/F12 with 10% fetal bovine serum (FBS). The cells were grown in a humidified atmosphere containing 5% CO₂ at 37°C. Transfection of cells was performed using Lipofectamine 2000 (Invitrogen; Thermo Fisher Scientific, Inc.) as previously described (22).

The cells were seeded onto a 96-well plate (1,000 cells/well) 24 h following transfection. Fresh medium supplemented with 10% FBS was added into the plates and changed every 24 h. The number of viable cells was estimated by the methyl thiazolyl tetrazolium (MTT) method every 24 h.

Cell migration was tested with a Boyden chamber assay, which utilized 24-well plates (Corning Incorporated, Corning, NY, USA) and Transwell plates (8 µm pore; EMD Millipore, Billerica, MA, USA). The bottom side of the Transwell plates were coated with collagen type 1 (10 µg/ml; Sigma-Aldrich; Merck Millipore, Darmstadt, Germany). Subsequent to being starved for 12 h with starvation medium (DMEM/F12) containing 0.25% bovine serum albumin, the cells were detached with cell dissociation buffer, then 1×10⁵ cells were planted into the top of each chamber. Medium containing 1% FBS was added into the bottom of the chamber for 16 h. Migrated cells were dyed with crystal violet. Bound crystal violet was eluted with 1 ml 10% acetic acid and the migration activity was expressed as the value monitored at 586 nm of extraction.

Differences in the numerical data between the cancer and adjacent normal tissues were evaluated using the paired Wilcoxon test. The in vitro data were analyzed with Student's t-test for variance. Experimental values were expressed as the mean ± standard deviation. A value of P<0.05 was considered to indicate a statistically significant difference.

A total of 50 matched normal lung and lung cancer tissues were analyzed with RT-qPCR to detect the expression level of HSF2 (Fig. 1A). The results showed that 76% (38/50) of the HSF2 expression levels in cancer tissues were upregulated compared with the normal lung tissues (P=0.0003; Fig. 1A). The protein levels of HSF2 were also detected by western blotting, and the results showed that the protein levels of HSF2 were upregulated, which is consistent with the results from RT-qPCR (Fig. 1B).

Since HSF2 is upregulated in lung cancer, it is reasonable to infer that it may affect the basic routine of lung epithelial cells and cancer cells, including cell proliferation and cell migration. Thus, a plasmid encoding HSF2 was constructed and transfected into human lung epithelia cell line BEAS-2B (Fig. 2A). The growth of HSF2-overexpressing BEAS-2B cells was monitored by MTT assay. The results showed that the HSF2 overexpression markedly enhanced the proliferation ability of BEAS-2B cells (P=0.0006; Fig. 2C). The effect of HSF2 overexpression on the lung cancer A549 cell line was also investigated (Fig. 2B), and similar results were acquired (P=0.0021; Fig. 2D). Apart from cell proliferation, enhanced cell migration was another characteristic of tumorigenesis. Thus, cell migration was tested using a Boyden chamber assay. The results showed that HSF2 overexpression significantly promoted the cell migration of BEAS-2B cells and A549 cells (Fig. 2E and F).

As the upstream regulator of HSP, HSF2 may regulate the expression level of HSP27, HSP47, HSP70 and HSP90. The present study additionally tested whether HSF2 overexpression can affect the expression of these HSPs. Subsequent to overexpression of HSF2 in BEAS-2B cells, the expression level of HSPs was semi-quantified by western blot analysis. All 4 HSPs, consisting of HSP27, HSP47, HSP70 and HSP90, were upregulated following HSF2 overexpression. Among them, HSP47 and HSP70 were only slightly upregulated, but the expression level of HSP27 and HSP90 were increased (Fig. 3A). However, HSF2 overexpression upregulated all 4 HSPs in A549 cells (Fig. 3B). These inconsistent results suggest that the regulation mechanism through which HSF2 affects the expression of HSPs may be different in normal epithelial cells and cancer cells.

Since HSF2 overexpression can promote lung epithelial cell and lung cancer cell proliferation and migration, and at the same time upregulate HSP, particularly HSP27 and HSP90, it is reasonable to infer that HSP27 and HSP90 is pivotal for HSF2 functioning in promoting cell proliferation and cell migration. Thus, in the present study, HSP27 or HSP90 siRNA (or a combination of the two) were co-transfected with HSF2-overexpressing plasmids into BEAS-2B and A549 cells to assess the role of HSP27 and HSP90 in HSF2 function (Fig. 4A). The results showed that HSP27 and HSP90 RNA interference significantly attenuated HSF2 enhanced cell growth (Fig. 4B) and cell migration (Fig. 4D) on BEAS-2B cells. Similar results were obtained in A549 cells (Fig. 4C and E).