Plasma samples from patients with NSCLC (n=40) and healthy volunteers (n=40) were collected from the First Affiliated Hospital of Nanjing Medical University (Nanjing, China). Patients were eligible if they were diagnosed with NSCLC by biopsy and had not been treated with chemotherapy or radiotherapy prior to sample collection. Patients were excluded if they suffered from any other disease (trauma, infections, allergies, autoimmune diseases or other inflammatory diseases and cancers). Informed consent was obtained from all patients participating in this research prior to the experiment. This study was approved by the Ethics Committee of Nanjing Medical University and conformed to the guidelines outlined by the Declaration of Helsinki. Specifically, venous blood samples were collected into K3EDTA tubes (Greiner Bio-One; Frickenhausen, Germany) and were then fractionated by centrifugation (10 min, 3,000 × g). The plasma was aliquoted and stored at −80°C prior to analysis. NSCLC tissue arrays (n=60, paired) were provided by the National Engineering Center for BioChips (Shanghai, China).

The plasma IL-17 concentration was measured using an anti-IL-17A coated ELISA kit (BMS2017; Thermo Fisher Scientific). Briefly, standard, control and plasma samples were added to the IL-17A antibody coated wells. Meantime, a biotinylated IL-17A antibody was added, incubating for 2 h at room temperature. The streptavidin-HRP was then added to all washed wells and incubated for 1 h, followed by washing and TMB substrate incubation for approximately 10 min. Finally, the enzyme reaction was terminated by stop solution (provided with the kit) and the absorbance at 450 nm (OD 450) of each well was measured on a microplate reader (ELX 800, Biotek, Winooski, VT, USA) (28).

Total RNA from the NSCLC tissues and cells was isolated using TRIzol reagent (Thermo Fisher Scientific) and the cDNA was generated using the reverse transcription reagent kit (Vazyme Biotech). The PCR assay was performed with a 50 µl volume reaction containing 25 µl 2X Taq master mix, 1 µl forward primer (10 µM), 1 µl reverse primer (10 µM) and 1 µl cDNA on an ABI 2720 thermal cycler with the condition of 94°C/5 min, 35 cycles of 94°C/30 sec, 55°C/30 sec and 72°C/30 sec, 72°C/7 min. The amplification products were then analyzed by agarose gel electrophoresis. The qPCR experiment was performed with a 20 µl volume reaction containing 10 µl 2X qPCR SYBR-Green master mix, 0.4 µl forward primer (10 µM), 0.4 µl reverse primer (10 µM), 1 µl cDNA and 0.4 µl 50X ROX reference dye 1 on an ABI 7300 system in triplicate under the following conditions: 95°C/5 min, 40 cycles of 95°C/5 sec and 60°C/30 sec. The primers used are shown in Table I. The results were normalized to β-actin expression and analyzed using the 2^−ΔΔCq method (29,30).

The slides from the NSCLC tissue array were incubated with the antibodies against IL-17R (dilution 1:100), HMGA1 (dilution 1:250) and cyclin D1 (dilution 1:250), followed by incubation with secondary antibodies (dilution 1:500). The reaction was visualized with a DAB HRP substrate kit. Finally, sections were viewed under a light microscope (Eclipse 90i; Nikon, Tokyo, Japan). The tissues subjected to IHC were scored according to the staining intensity and stained area, and 5 randomly selected fields were analyzed. The scoring of the staining intensity was as follows: negative, 0; weak, 1; moderate, 2; and strong, 3. The scoring of the stained area was as follows: 0%, 0; 1–25%, 1; 26–50%, 2; 51–75%, 3; and 76–100%, 4. These scores were multiplied to produce the final score: 0–1, negative expression; 2–4, weak positive expression; and 6–12, strong positive expression (31,32).

The HMGA1 expression plasmid was constructed by inserting the complete open reading frame (ORF) of the human HMGA1 gene (NM_145899.2) into the pIRES2-EGFP vector (Clontech/Takara Bio, Shiga, Japan). The specific primer sequences are shown in Table I. The pGpU6/GFP/Neo vector carrying small hairpin RNA targeting HMGA1 (shHMGA1) was provided by GenePharma (Shanghai, China). The shRNA-targeted sequences of HMGA1 gene were as follows: shHMGA1-1, 5′-CAACTCCAGGAAGGAAACCAA-3′; shHMGA1-2, 5′-CCTTGGCCTCCAAGCAGGAAA-3′; shHMGA1-3, 5′-GAAGGAGGAAGAGGAGGG CAT-3′; and scrambled shRNA control (shCTR), 5′-GTTCTCCGAACGTGTCACGT-3′. In addition, the reporter plasmids carrying cyclin D1 full-length (FL) and truncated promoter plasmids were constructed by inserting the corresponding fragments into the pGL3-basic vector (Promega). The detailed primers are listed in Table I.

The cells were maintained in DMEM with 10% FBS in an incubator containing 5% CO₂ at 37°C. For IL-17 stimulation, the cells were cultured overnight and starved for 24 h; IL-17 was then added into the medium at various concentrations (0, 0.5, 5, 50 and 500 ng/ml) for different periods of time (according to the needs of distinct experiments). For plasmid transient transfection, 3×10⁵ cells were seeded per well in a 6-well plate. A mixture of 3 µg plasmid and 6 µl transfection reagent was then added followed by incubation for 48 h. The transfection efficiency was determined by GFP expression at 24 h following transfection. For IL-17R siRNA (siIL-17R) transfection, the cells were seeded in a 6-well plate and cultured overnight, and the mixture of 100 pmol siIL-17R and 5 µl Lipofectamine 2000 was then added followed by 24 h of incubation. The siIL-17R was supplied by GenePharma and the sequences were as follows: forward, 5′-CCUGCAGCUGAACACCAAUTT-3′ and reverse, 5′-AUUGGUGUUCAGCUGCAGGTT-3′.

The cells were lysed using RIPA buffer (Beyotime, Beijing China) and the total protein concentration was measured by BCA assay. The protein samples were loaded (50 µg per well) and electrophoresed on a 10% SDS-PAGE gel and transferred onto PVDF membranes (Pall Corp., Port Washington, NY, USA). The blots were then blocked with 5% non-fat milk for 1 h at room temperature, and probed with the IL-17R, HMGA1 or cyclin D1 antibody overnight at 4°C, followed by incubation with the rabbit IgG HRP-conjuncated (dilution 1:1,000) antibody for 30 min at room temperature, and then exposed using regular X-ray film. The dilution of anti-IL-17R, anti-HMGA1 and anti-cyclin D1 antibodies was 1:500, 1:10,000 and 1:1,000 respectively. The control antibody was β-actin antibody (Cat. no. AF0003; Beyotime Beijing, China) and the dilution was 1:1,000. Autoradiograms were quantified by densitometry (Quantity One software; Bio-Rad, Hercules, CA, USA).

The HMGA1 response elements on cyclin D1 gene promoter were predicted by using the online software JASPAR (http://jaspar.genereg.net/). Briefly, the matrix model of HMGA1 binding element was found and selected, and the promoter sequence of Cyclin D1 gene was then input into the scan window, with the relative profile score threshold set at about 80%. After scanning, the putative sites were displayed.

The afore-mentioned promoter reporters were transfected into the A549 cells with a pRL-SV40 vector (Promega), separately. A dual-luciferase reporter reagent was used to measure the promoter activity according to the manufacturer's instructions and as previously described (33).

ChIP assay was performed using ChIP-grade HMGA1 antibody and ChIP-grade protein G agarose beads in accordance with the manufacturer's instructions provided with the SimpleChIP Plus Enzymatic Chromatin IP kit and as previously described (34). A non-specific rabbit IgG was used as a negative control. PCR and RT-qPCR were performed to analyze the cyclin D1 promoter in the ChIP materials. The primers for the cyclin D1 promoter fragment in the ChIP assay were as follows: forward, 5′-CCCCATAAATCATCCAGGC-3′; and reverse, 5′-CCCGAGCACCCACAATC-3′.

Cell proliferation was measured by CCK-8 and colony formation assays. For CCK-8 assay, based on the manufacturer's instructions, the optical density (OD) values at 450 nm were documented using a microplate reader (Biotek). For colony formation assay, the cells were seeded in a 6-well plate at 500 cells/per well. Following culture for 10 days, the cells were fixed and stained with 0.1% crystal violet. Visible colonies were counted.

All data are presented as the means ± SE. Data analysis was carried out using GraphPad Prism 6.01 (GraphPad Software, San Diego, CA, USA). The significant difference between 2 groups was determined by a Student's t-test. Multi-group comparisons were carried out by one-way ANOVA with Dunnett's post hoc test. Using a Chi-square test, the association of the clinical parameters of the patients with NSCLC with the expression of proteins was analyzed. The correlation between the IHC scores was determined by computing Pearson's correlation coefficient. A P-value <0.05 was considered to indicate a statistically significant difference.

At the beginning of the present study, we examined the concentration of IL-17 in plasma and the expression levels of IL-17R, HMGA1 and cyclin D1 in the tumor tissues from patients with NSCLC. Using ELISA, we found that the plasma IL-17 levels were significantly increased in the 40 cases of NSCLC (the clinical characteristics of the patients are summarized in Table II) compared with the 40 healthy donors (Fig. 1A). Moreover, a marked upregulation in the expression of IL-17R, HMGA1 and cyclin D1 in the NSCLC tissues in comparison with the adjacent normal tissues was demonstrated by IHC staining in the NSCLC tissue microarrays (n=60, Fig. 1B and C). Furthermore, we also discovered that the expression of IL-17R, HMGA1 and cyclin D1 was associated with the tumor size, lymph node metastasis and TNM stage in these patients with NSCLC (Table III). Notably, positive correlations were also found between the expression of the above-mentioned 3 genes in the NSCLC tissues (Fig. 1D). These data thus suggest that the production of IL-17, IL-17R, HMGA1 and cyclin D1 may probably contribute to the development of NSCLC.

Given that IL-17, IL-17R, HMGA1 and cyclin D1 expression levels were all elevated in the patients with NSCLC, and the expression of IL-17R positively correlated with HMGA1 or cyclin D1 expression, we assumed that IL-17, as an extracellular stimulus, may trigger NSCLC cell proliferation and upregulate HMGA1 and cyclin D1 by binding to IL-17R. Firstly, to determine the most susceptible NSCLC cell line to IL-17 stimulation, we assessed IL-17R expression in different NSCLC cell lines. The results revealed that IL-17R was markedly overexpressed in the PC9, A549 and SPC-A1 cells in comparison with the human bronchial epithelial 16HBE cells, particularly in the A549 cells (Fig. 2A). Subsequently, we selected the A549 cells and treated the cells with human recombinant IL-17 (i.e., IL-17A) and found that A549 cell proliferation (examined by CCK-8 assay and colony formation assay) was prominently upregulated by IL-17 stimulation in a dose-dependent manner, particularly when the IL-17 concentration reached 50 and 500 ng/ml (Fig. 2B–D). In addition, to ascertain whether IL-17 increases HMGA1 and cyclin D1 expression as well as the optimal time-point for the IL-17-induced production of HMGA1 and cyclin D1, we stimulated the A549 cells with IL-17 (50 ng/ml) for different periods of time and detected the transcription and expression of the afore-mentioned two genes by RT-qPCR and western blot analysis. The results revealed that the mRNA and protein expression levels of HMGA1 and cyclin D1 were significantly increased at 2 h (HMGA1) or 3 h (cyclin D1) and both peaked at 6 h (Fig. 2E–G). These results indicate that IL-17 not only induces A549 cell proliferation, but also increases the expression of HMGA1 and cyclin D1.

To further confirm that IL-17 stimulation promotes A549 cell proliferation and increases the expression of HMGA1 or cyclin D1 by binding to IL-17R, we subjected the A549 cells to siIL-17R transfection or IL-17 antibody incubation, prior to treatment with IL-17 and then examined cell proliferation and the levels of the two above-mentioned proteins. The results of CCK-8 assay and colony formation assay revealed that the viability (OD value) and the colony numbers of A549 cells were greatly multiplied following treatment with IL-17; however, these effects were notably reversed by the silencing of IL-17R with siIL-17R or by the neutralization of IL-17 with anti-IL-17 antibody (Fig. 3A–C). Similarly, transfection with siIL-17 or incubation with anti-IL-17 also significantly decreased the IL-17-induced mRNA and protein expression of HMGA1 and cyclin D1 in the A549 cells (Fig. 3D–F). These results further denote that IL-17 effectively promotes the proliferation, and increases the expression of HMGA1 and cyclin D1 in A549 cells through its interaction with IL-17R.

It has been reported that HMGA1 is a transcription factor which can activate the transcription of downstream target genes and finally increase their expression (35,36). Since our experiments verified that HMGA1 and cyclin D1 expression levels were increased in the process of IL-17-induced A549 cell proliferation, and the expression phase of HMGA1 was earlier than that of cyclin D1, we hypothesized that HMGA1 itself may enhance cyclin D1 transcription and expression, leading to an increment in cell proliferation, and that the absence of HMGA1 may impede the afore-mentioned phenomena mediated by IL-17. To confirm our hypothesis, we first constructed the HMGA1 overexpression plasmid (pIRES2-HMGA1) and the plasmid carrying small hairpin RNA targeting HMGA1 (shHMGA1) and examined the transfection efficiency of these plasmids (Fig. 4A–C). Subsequently, CCK-8 and colony formation assays were performed. As was expected, A549 cell proliferation was markedly elevated following HMGA1 overexpression, whereas the knockdown of HMGA1 markedly suppressed cell proliferation upon IL-17 stimulation (Fig. 4D–F). In addition, the mRNA and protein levels of cyclin D1 were eminently upregulated or downregulated when HMGA1 was overexpressed or silenced with IL-17 treatment, respectively (Fig. 4G–I), indicating that IL-17-induced HMGA1 expression exerts a promoting effect on the proliferation of and cyclin D1 expression in A549 cells mediated by IL-17.

As mentioned earlier, IL-17 stimulation increased cyclin D1 expression via the transcription factor, HMGA1; hence, we wished to determine whether IL-17 activates the cyclin D1 gene promoter through HMGA1. For this purpose, we carried out luciferase reporter assay, and found that IL-17 markedly increased cyclin D1 promoter activity at 3 h (peaked at 6 h) after the A549 cells were stimulated (Fig. 5A). In addition, the overexpression of HMGA1 markedly activated the full length of the cyclin D1 promoter (cyclin D1-FL, −2,099 to +86 nt), whereas the activity of cyclin D1-FL markedly decreased in accordance with HMGA1 knockdown (Fig. 5B). Subsequently, to locate the region in which HMGA1 binds to on the cyclin D1 promoter, we first predicted three potential HMGA1 response elements (−1,700 to −1,691 nt, −1,026 to −1,017 nt, and −139 to −130 nt) using the online software JASPAR (http://jaspar.genereg.net/), and then constructed three truncated promoter reporters based on the prediction (Fig. 5C), which were truncate 1 (T1, −1,672 to +86 nt), truncate 2 (T2, −1,026 to +86 nt) and truncate 3 (T3, −139 to +86 nt). By luciferase assay, we confirmed that in the A549 cells treated with IL-17 or transfected with the HMGA1 overexpression plasmid, the activity of all cyclin D1 truncated promoters was markedly decreased compared to that of the cyclin D1-FL. However, no statistically significant differences were observed between the luciferase activity in these three truncated reporters (Fig. 5D and E), suggesting that the most effective HMGA1 binding element may be located in the region of −2,099 to −1,673 nt on the cyclin D1 promoter (probably −1,700 to −1,691 nt). Finally, to determine the exact binding of HMGA1 to the above-mentioned element, a ChIP assay was performed using antibody against HMGA1, and the region −1,752 to −1,622 nt (containing −1,700 to −1,691 nt) was then amplified by RT-qPCR. The results revealed that the binding of HMGA1 to the indicated fragment of the cyclin D1 promoter was prominently increased both in the A549 cells subjected to IL-17 stimulation or in those transfected with the HMGA1 overexpression vector; however, this binding was diminished when HMGA1 was silenced (Fig. 5F and G). These findings imply that IL-17 enhances cyclin D1 promoter activity via HMGA1, directly binding to its response element in the region −1,752 to −1,622 nt, which eventually results in an increased cyclin D1 expression and A549 cell proliferation.