In total, 102 patients were selected between January 2014 and May 2016. All tissues were collected with informed consent. The majority (90 out of 102) of lung cancer patients did not have metastasis. Overall, 52.0% (53/102) of the patients were female and 48.0% (49/102) were male. Patients at late stage or early stages were all included. All experiments were approved by the Institutional Review Board of Central Hospital of Wuhan (Wuhan, China).

All tissues were selected by an experienced pathologist and embedded in paraffin. The immunohistochemistry assay was performed following standard protocols (12,13). Briefly, all the embedded tissues were cut into 3-µm slices. The slices were deparaffinized with xylene and graded ethanol. The tissues were treated with 0.3% hydrogen peroxidase, incubated with a primary rabbit anti-IGFBP-4 antibody (dilution, 1:100; cat. no. ab4253; Abcam, Cambridge, UK) overnight at 4°C followed by incubation with anti-rabbit IgG-horseradish peroxidase (HRP) secondary antibody (dilution, 1:1,000; cat. no. MBS435036; MyBioSource, San Diego, CA, USA) at room temperature for 1 h. The tissues were washed three times with PBS between antibody incubations. The tissues were observed under microscopes (BX4; Olympus Corporation, Tokyo, Japan; magnification, ×40) and the expression levels of IGFBP-4 were considered to be extremely strong, strong, median, weak and none according to the density of signals.

The lung cancer A549 and lentivirus packaging 293T cell lines were purchased from Cell Bank of Chinese Academy of Sciences (Shanghai, China). The cells were maintained in Dulbecco's modified Eagle's medium (DMEM, Thermo Fisher Scientific, Waltham, MA, USA) supplemented with 10% fetal bovine serum (FBS, Sigma-Aldrich; Merck KGaA, Darmstadt, Germany) and passaged at 1:3 when cells had reached confluence. The cells were incubated in humid atmosphere at 37°C with 5% CO₂.

The shRNA construct to target IGFBP-4 was cloned into lentivirus plasmid pLKO1 (Addgene, Inc., Cambridge, MA, USA). The targeting sequence of IGFBP-4 was 5′-GCCACTTGCGCCCTGGGCTTG-3′. The oligo 1 (5′-AATTCGCCACTTGCGCCCTGGGCTTGATGCCAAGCCCAGGGCGCAAGTGGCG-3′) and oligo 2 (5′-AATTCGCCACTTGCGCCCTGGGCTTGCATCAAGCCCAGGGCGCAAGTGGCGG-3′) sequences were diluted to 100 µM for annealing. They were mixed together with T4 DNA ligase buffer, heated at 95°C for 10 min and cooled slowly at a rate of 1°C/min. The pLKO1 plasmid was cut using the EcoRI restriction enzyme and ligated using annealed oligos to generate the pLKO-IGFBP-4 shRNA-1 plasmid. Another IGFBP-4 shRNA was generated using the same method and its targeting sequence was 5′-GACAAGGACGAGGGTGACCA-3′.

pLKO-IGFBP-4 shRNA-1 and pLKO-IGFBP-4 shRNA-2 were co-transfected with vesicular stomatitis virus-G and delta 8.2 (Addgene, Inc.) into 293T cells to package lentivirus. The medium was changed 12 h post-transfection. The 293T medium was collected 48, 72 and 96 h post transfection. The media were centrifuged at 8,000 × g for 20 min to remove the cell debris. The medium contained the lenti-IGFBP-4 shRNA-1 or lenti-IGFBP-4 shRNA-2 viruses. The A549 cells were infected with viruses and selected with 2 µg/ml puromycin 48 h later. The stable cell lines were passaged and maintained in DMEM 10% FBS with 2 µg/ml puromycin.

RNA was extracted from A549 cells and A549 cells stably transfected with IGFBP-4 shRNA-1 and IGFBP-4 shRNA-2 with TRIzol (Invitrogen; Thermo Fisher Scientific, Inc., Waltham, MA, USA), according to the manufacturer's protocol. In total, 1 µg RNA was reverse transcribed into cDNA by Superscript II Reverse Transcription kit (Invitrogen; Thermo Fisher Scientific, Inc.). The RT-qPCR were performed with SYBR-Green master mix (Thermo Fisher Scientific, Inc.), as follows: Preheating at 94°C for 5 min, then 40 cycles of 94°C denaturation for 30 sec, 55°C annealing for 30 sec and 72°C elongation and data collection for 1 min. The primers for IGFBP-4 were as follows: Forward, 5′-AGGGTGACCACCCCAACAAC-3′; and reverse, 5′-TCCAGCGCCCGGTGCAGCTC-3′. The primers for the internal control gene GAPDH were as follows: Forward, 5′-GCGAGATCCCTCCAAAATCAA-3′; and reverse 5′-GTTCACACCCATGACGAACAT-3′. Cq values were processed using 2^−ΔΔCq method to calculate the relative expression level (14).

Total protein was extracted from A549 cells, A549 IGFBP-4 shRNA-1 cells and A549 IGFBP-4 shRNA-2 cells using NP40 lysis buffer containing protease inhibitors and phosphatase inhibitors cocktail (Thermo Fisher Scientific, Inc.). The protein concentration was measured with bicinchoninic acid assay (Bio-Rad Laboratories, Inc., Hercules, CA, USA). Denatured protein (40 µg) was loaded and separated 10% SDS-PAGE. The proteins were transferred onto polyvinylidene fluoride membranes (Merck KGaA, Darmstadt, Germany), blocked with 5% non-fat milk, primary rabbit anti-IGFBP-4 antibody (dilution, 1:100; cat. no. ab4253; Abcam) and anti-rabbit IgG-HRP secondary antibody (dilution, 1:1,000; cat. no. MBS435036; MyBioSource). The membranes were exposed to the film for band development. GAPDH blotted by primary rabbit anti-IGFBP-4 antibody (dilution, 1:100; cat. no. ab8245; Abcam). GAPDH (Cell Signaling Technology, Inc., Danvers, MA, USA) was used as an endogenous control in this assay.

The A549 cells, A549 IGFBP-4 shRNA-1 cells and A549 IGFBP-4 shRNA-2 cells were plated onto 96-well dishes at 2,000 cells per well. The cell viability was measured using an MTT assay (Thermo Fisher Scientific, Inc.) at the indicated days, according to the manufacturer's protocol.

The Kaplan-Meier Plotter online database (8) was used to analyze the association between IGFBP-4 and lung cancer.

The data were expressed as the mean + standard deviation (SD) or mean ± SD. The differences between groups were analyzed by one-way analysis of variance and least significant difference analysis. The survival curve was generated using the Kaplan-Meier method. The median survival comparison between groups was calculated using the log-rank test. P<0.05 was considered to indicate a statistically significant difference.

To assess the expression level of IGFBP-4 in lung cancer tissues and SNCTs, the tissues were stained with IGFBP-4 by immunohistochemistry. Lung cancer tissues showed increased IGFBP-4 expression compared with SNCT (Fig. 1). In 53 of the 102 paired tissues, cancer tissues expressed an increased level of IGFBP-4 compared with SNCTs. This percentage (51.96%) is larger than the proportion of patients with similar (11.76%) or decreased (36.27%) levels of IGFBP-4 expression in cancer tissues (Fig. 2).

To investigate whether IGFBP-4 expression is associated with lung cancer prognosis, the present study compared IGFBP-4 expression with the survival of lung cancer patients. It was shown that patients with increased IGFBP-4 expression have a shorter median survival rate (Fig. 3). It was also found that IGFBP-4 expression was associated with metastasis, TNM stages and tumor malignancy, but not with age, sex or tumor size (Table I).

To assess whether blocking the function of IGFBP-4 could inhibit lung cancer progression, the present study knocked down IGFBP-4 expression by shRNAs. Lentivirus-mediated shRNA sufficiently knocked down IGFBP-4 expression at the mRNA (Fig. 4A) and protein level (Fig. 4B). With IGFBP-4 knockdown, A549 cells showed a slower proliferation rate (Fig. 5). This indicated that IGFBP-4 promoted lung cancer cell proliferation.