A549 and HCC827 cells were cultured with F12K medium and RPMI-1640 (both from Sigma-Aldrich; Merck KGaA, Darmstadt, Germany) at 37°C in a humidified atmosphere (95% air, 5% CO₂), respectively. Fetal bovine serum (FBS; 10%) (JRH Biosciences, Lenexa, KS, USA), 100 µg/ml penicillin/streptomycin and 2 mM L-glutamine (Gibco; Thermo Fisher Scientific, Inc., Waltham, MA, USA) were supplemented in culture medium.

Study subjects were recruited during routine health examinations with signed consents. Twenty-nine study subjects with varying age and sex were randomly chosen including 16 lung adenocarcinoma patients and 13 healthy volunteers, enrolling for analysis. This study was a prospective study that aimed to identify potential biomarkers to predict cancer progression. The Ethics Committee of the Taipei Tzu-Chi General Hospital approved the study (Approval no. TCRD-TPE-100-43; IRB no. 98-IRB-019-X; IRB extension no. 05-XD55-107). Our study was conducted from January 2010-January 2012 and December 2017-July 2018. The inclusion criteria for lung adenocarcinoma patients were as follows: The patients were required to be >18 years of age with the pathology type confirmed by pathology or cytology. Patients were excluded from the study if they had received chronic steroid therapy, adjuvant or neoadjuvant chemotherapy, radiotherapy, or both. Patients with a history of any other type of cancer were not eligible. The inclusion criteria for healthy volunteers were as follows: >18 years of age was required with no evidence of cancers. Healthy volunteers were excluded if they had inflammatory or infectious conditions (empyema, interstitial lung disease, or active pneumonitis) or a history of any other type of cancer.

Blood samples were centrifuged at 200 × g at 8°C for 45 min. Serum was transferred into new polypropylene tubes and re-centrifuged at 8,000 × g for 20 min. The supernatant was collected and stored at −80°C. Serum DNA was extracted from 400 µl sample with QIAmp DNA Blood Mini Kit (Qiagen, Inc., Valencia, CA, USA).

DNA from patient serum was collected to performed quantitative (qPCR) on Applied Biosystems 7900 Sequence Detector (Applied Biosystems; Thermo Fisher Scientific, Inc.). The TaqMan probe sequence purchased from Applied Biosystems (Thermo Fisher Scientific, Inc.) was Mit 3153T (5′-FAM-TTCACAAAGCGCCTTCCCCCGTAAATGA-TAMRA-3′), where FAM was 6-carboxyfluorescein and TAMRA was 6-carboxytetramethylrhodamine. The thermocycling conditions included an initial denaturation at 50°C for 2 min and 95°C for 10 min, followed by 45 cycles at 95°C for 15 sec, 60°C for 1 min. The internal control primer sequences were plasmid forward primer (pls F) (5′-AATACGCAAACCGCCTCTCC-3′) and plasmid reverse primer (pls R) (5′-ACAACATACGAGCCGGAAGC-3′). The internal control for the TaqMan probe sequence was plasmid TaqMan probe (pls T) (5′-FAM-CGCAACGCAATTAATGTGAGTTAGCTCAC-TAMRA-3′). The method of quantification used was 2^−ΔΔCq (24).

A 172-bp mtDNA from genomic DNA of a healthy volunteer was cloned and used as a calibrator. Calibrators were prepared by serial dilution of the stock solution and contained 10–10⁷ mitochondrial DNA copies/µl. The results were expressed as genome-equivalent (GE)/ml of serum using a conversion factor of 6.6 pg of DNA/cell. The concentration was calculated using an equation that was previously described (25). Primers used to amplify Mt3130-3301 were Mit3130F (5′-AGGACAAGAGAAATAAGGCC-3′) and Mit 3301R (5′-TAAGAAGAGGAATTGAACCTCTGACTGTAA-3′).

Cells were washed twice with ice-cold phosphate-buffered saline (PBS) (Gibco; Thermo Fisher Scientific, Inc., Waltham, MA, USA), lysed in 1 ml/plate with ice-cold IP lysis buffer supplemented with protease inhibitors (Gibco; Thermo Fisher Scientific, Inc.) and harvested by scraping. Protein concentration was determined by BCA method. A total of 30 µg of protein sample was separated by 10% SDS-PAGE, and then transferred onto polyvinylidene fluoride (PVDF) membranes (Bio-Rad Laboratories, Inc., Hercules, CA, USA). Membrane was blocked by 5% non-fat dry milk in TBST with 0.05% Tween-20 for 1 h. Primary antibodies to TLR9 (cat. no. ab88101; Abcam, Cambridge, MA, USA), MyD88 (cat. no. GTX112986; GeneTex International Corporation, Hsinchu, Taiwan) and β-tubulin (cat. no. sc-73242; Santa Cruz Biotechnology, Dallas, TX, USA) were used at a dilution of 1:1,000 and incubated at 4°C for overnight. Secondary antibodies to anti-mouse-HRP and anti-rabbit-HRP (cat. no. 31430 and 31460, Invitrogen; Thermo Fisher Scientific, Inc.) were used at a dilution of 1:5,000 and incubated at room temperature for 1 h. Signals were detected with chemiluminescence (Amersham; GE Healthcare, Chicago, IL, USA) and exposed to X-ray film (Kodak, Rochester, NY, USA). Signal quantification was done by using ImageJ.

Culture medium (400 µl) was harvested after cells were treated with CpG ODN-M362 for 24 h according to the manufacturer's instructions (Human Cytokine Array Panel A; R&D Systems, Inc., Minneapolis, MN, USA). CpG-ODN-M362 sequence, 5′-TCGTCGTCGTTC:GAACGACGTTGAT-3′; and negative control, pCpG was a synthetic sequence called pCpG Giant that failed to trigger downstream TLR9 signaling and served as a negative control. pCpG Giant was purchased from InvivoGen (San Diego, CA, USA) (cat. no. tlrl-cpgg).

All experiments were repeated at least 3 times. Mean values and standard deviation (SD) were calculated and analyzed with a one-way analysis of variance (ANOVA) or Student's t-test with PRISM 5.0 (GraphPad Software, Inc., San Diego, CA, USA). Scheffe's post hoc test was used following one-way ANOVA. Data with P<0.05 was considered to indicate a statistically significant difference.

In order to evaluate the CpG level of different stages of tumorigenesis in lung cancer patients, a 172-bp mtDNA segment between nucleotide positions 3130 and 3301 was amplified from genomic DNA of a healthy volunteer with the primers. Mt3130-3301 was successfully amplified (Fig. 1A). The mtDNA amplicons were then cloned into the pCRII TOPO TA plasmid (Invitrogen; Thermo Fisher Scientific, Inc.) and were confirmed with restriction enzyme digestion and sequencing (Fig. 1B and C). The cloned mtDNA plasmid was to be used as a calibrator for the mtDNA copy number.

To calibrate the correlation between threshold cycle (C_T) in a TaqMan assay and the serum mtDNA copy number, a plot of the C_T against the input target quantity, the plotted on a common log scale, was generated (Fig. 1D). The linearity of the quantitative assay was assessed with a calibrator, which was 172 bp mtDNA flanked with a plasmid. The calibrator was serially diluted to concentrations from 10 to 10⁷ copies/µl before use. The mtDNA copy number was determined by division of the total DNA concentration by the weight of each plasmid molecule (25). The assay was sensitive and able to detect 10 copies of mitochondrial DNA/µl. As the plot indicated, the threshold cycle (C_T) was set to 15 and was proportional to the target copy number from 10 to 10⁷ mtDNA copies.

To investigate the mtDNA copy number at different malignancy stages of lung cancer, we performed quantitative PCR analysis to evaluate the levels of mtDNA. Total DNA extraction from 13 healthy volunteers, 5 patients with non-metastatic, and 11 patients with metastatic lung adenocarcinoma (clinical characteristics of healthy subjects and lung adenocarcinoma patients are summarised in Table I) was conducted, and we determined the concentration of serum mtDNA. Among these patients, 11 had lung adenocarcinoma of stage IV with metastasis, and 5 had stage II lung adenocarcinoma without metastasis. Serum mtDNA levels of the M0 (non-metastatic) group were significantly higher than those of the Meta (metastatic) and HS (healthy volunteer) groups (Fig. 2A). In addition, the Meta group had the lowest serum mtDNA level that was about a half of that of the M0 group. The qPCR results were then used to calculate mtDNA copy numbers with the calibrator (Fig. 2B). The mtDNA copy number was significantly higher in the M0 group, and up to 4×10⁴, ~4 times higher than that in the HS group, indicating an associtaion between serum mtDNA levels and metastasis status during lung carcinogenesis.

To confirm that the high copy number of mtDNA functionally affects its receptor, TLR9, two human lung adenocarcinoma cell lines were treated with different concentrations of a synthetic unmethylated CpG ODN, ODN-M362, which has been known to serve as an agonist of TLR9, thereby simulating to the role of mtDNA by triggering TLR9-mediated signalling. A549 cells were treated with 1 or 3 µM ODN-M362 for 24 h before harvesting. The expression level of TLR9 in A549 was induced by ODN-M362, and the downstream adaptor protein, MyD88, was also upregulated (Fig. 3A). In addition, the HCC827 cell line was also treated with 1, 3, or 5 µM ODN-M362 for 24 h. Compared with the A549 cells, the HCC827 cell line is also a lung adenocarcinoma cell line however it carries an acquired mutation in the EGFR tyrosine kinase domain. TLR9 and MyD88 in HCC827 cells were significantly upregulated by ODN-M362 treatment in a dose-dependent manner (Fig. 4A). Quantitative analysis of the western blotting was carried out relative to untreated cells (Figs. 3B and 4B). Thus, ODN-M362 was able to stimulate the expression of TLR9 in both lung cancer cell lines, and TLR9 may be regulated by mtDNA copy numbers in A549 and HCC827 cells.

We hypothesised that induction of TLR9 in lung adenocarcinoma cell lines activated downstream signalling through ODN-M362. Therefore, we performed a cytokine array analysis to assess the expression levels of 36 human cytokines with/without ODN-M362 treatment of A549 and HCC827 cell lines (Fig. 5A and C). Among the changes of cytokine expression, ODN-M362-treated A549 cells secreted IL-8 >15-fold compared to the untreated cells, while C5/C5a was induced 1.6-fold in ODN-M362-treated A549 cells (Fig. 5B). It is known that C5/C5a is a direct inducer of interleukin-8 (IL-8) secretion (26). Therefore, marked increase of IL-8 supports the previous finding and may be due to serial activation from the C5/C5a-mediated complement cascade. IL-8, an inflammation-associated chemokine, is also likely involved in cancer-associated processes, such as angiogenesis, proliferation, migration, and invasion (27). Moreover, IL-8 has been identified as an immunosuppressive cytokine that promotes carcinogenesis (28). Therefore, the ODN-M362-mediated induction of TLR9 signalling may cause inflammation-induced carcinogenesis through the IL-8 abnormality. Conversely, CCL5 was downregulated in ODN-M362-treated A549 cells (Fig. 5B). Although CCL5 has been reported as a protein that promotes cancer cell proliferation and migration (29), CCL5 also induces proper immune responses to tumour formation, and the mechanism is still unclear (28).

Both A549 and HCC827 are NSCLC cell lines; however, HCC827 cells are EGFR-mutated and highly sensitive to EGFR inhibitors with deletions from E746 to A750, whereas A549 cells encode wild-type EGFR, and these cell lines may induce different immune responses (2,30). We found that after HCC827 cells were treated with ODN-M362, monocyte chemoattractant protein-1 (MCP-1), CXCL1, and macrophage migration inhibitory factor (MIF) were significantly downregulated (Fig. 5B and D). These results indicated that impaired EGFR signalling not only affects TLR9-mediated cytokine expression levels but also fails to induce chemotaxis for monocytes, neutrophils, and macrophages when TLR9 signalling is activated.