The human lung cancer cell line A549 was purchased from the American Type Culture Collection (ATCC) and cultured in RPMI-1640 with Dulbecco’s modified Eagle’s medium (Sigma-Aldrich, St. Louis, MO, USA) supplemented with 20% fetal calf serum, 0.05% L-glutamine, 150 UI/ml penicillin and 50 μg/ml streptomycin in a humidified atmosphere with 5% CO₂ at 37°C. The cultures were split every second day by dilution at a concentration of 2×10⁵ cells/ml. The cell counts were performed with a hemocytometer; the cell membrane integrity was determined using the trypan blue exclusion technique. Cell lines in the maximal range of up to 20 passages were used for the present study. All cell lines tested negative for mycoplasma contamination by polymer chain reaction (PCR) methods (16). Cell lines were authenticated using the short tandem repeats (STR) testing.

A549 cells (1×10⁶) were plated in 75-cm² culture flasks and irradiated with 4 Gy of γ-rays using a Theratron Cobalt-60 treatment unit (Siemens, Concord, CA, USA) at a dose rate of 1 Gy/min when the cells were at ~60% confluency in the culture flask. Immediately following irradiation, the culture medium was renewed, and the cells were returned to the incubator. When the A549 cells reached ~90% confluency, they were trypsinized, counted and passaged into new culture flasks. Again, the cells were treated with 4-Gy γ-rays when they reached ~60% confluency. The irradiation was performed 13 times for a total dose of 60 Gy (irradiated with 2 Gy of γ-rays at the final irradiation) over 5 months. The parental cells were trypsinized, counted and passaged under the same conditions without irradiation.

The final cell density was adjusted to 1×10⁶ cells/ml, and the samples were placed at 37°C in a 5% CO₂ incubator. X-irradiation was performed using a Theratron Cobalt-60 treatment unit (Siemens) at a dose rate of 2 Gy/min. Non-irradiated cells were treated in similar way, but at a zero radiation dose.

The cells were trypsinized and resuspended in T25 flasks, followed by γ-ray exposure at room temperature at single doses of 0, 4, 8, and 12 Gy. After irradiation, the cells were seeded onto 6-cm dishes and incubated for 14 days without disturbance. The seeded cell number was increased as the radiation dose increased. Formed colonies were visualized by staining with 0.02% crystal violet solution (w/v in 75% ethanol). The plating efficiency was determined as the ratio of the number of colonies divided by the number of cells seeded. The surviving fraction was determined from the ratio of plating efficiencies of irradiated cells compared to the no irradiated control. Each data value represents the mean of three independent experiments ± SEM.

The samples were ground by using liquid nitrogen. Total RNA was extracted from lung tissues using an animal tissue RNA purification kit (Norgen Biotek, Thorold, ON, USA) as recommended by the manufacturer. RNA samples were measured using Bioanalyzer to determine RNA integrity number.

miRNA expression profiling was performed using the TaqMan Rodent miRNA Array Card A and Card B (Applied Biosystems, Foster City, CA, USA). Each chip contained multiple quality control probes and employed dual-color chip to examine miRNA expression profiling in mouse lung tissues. Probes were synthesized in situ with photosensitive PGR. The sequence consisted of 2 fragments, namely, a chemically modified oligonucleotide encoding fragment complementary to target miRNA or other target RNA, and an extension arm at a certain distance to the connected encoding sequence, which lessened the hybridization spatial impairment. The temper of probe hybridization was balanced by the chemical modification. Cy3- and Cy5-specific fluorescent labels and Axon GenePix 4000B microarray scanner were used to capture hybridization images. ArrayPro (Media Cybernetics, Bethesda, MD, USA) was utilized to complete the digital transformation.

An approximate length of 21–23 nt of miRNA resulted in difficulties in conventional PCR test. We used TaqMan^® microRNA assays (Applied Biosystems) to examine the miRNA differential expression profiling in PTC as recommended by the manufacturer. Sample RNA (10 ng) was reversely transcribed into cDNA by using specific stem-loop primers and TaqMan^® microRNA reverse transcription kit. With cDNA as the template, TaqMan microRNA assay and TaqMan^® Universal PCR Master Mix were used for the serial real-time PCR. RNU48 was used as an internal control to minimize the variation among reverse transcription, PCR and samples. Data were collected, analyzed, and normalized using the Applied Biosystems analysis software to determine the differential expression profiles of the miRNAs. All of the experiments were conducted in triplicate. Expression levels were calculated using the relative quantification method (ΔΔCT) in the ABI PRISM 7500 sequence detection system (Applied Biosystems), according to the manufacturer’s instructions.

miR-1323 was overexpressed by reverse transfection of 100 nM pre-miR precursor molecule using siPORT NeoFX transfection reagent as per the manufacturer’s instructions and compared with the pre-miR negative control (miR-control) (all from Life Technologies/Ambion). Inhibition of miR-1323 was achieved by reverse transfection of 30 nM anti-miR-1323 inhibitor (Life Technologies/Ambion) using siPORT NeoFX transfection reagent and compared with the anti-miR negative control.

Full-length human PRKDC 3′UTR sequences were amplified by PCR and cloned into the Spe1/HindIII restriction sites of the pMIR Report Luciferase plasmid (Life Technologies/Invitrogen, Carlsbad, CA, USA). The putative miR-1323 target sites in the 3′UTR of PRKDC were mutated using the QuikChange XL Site-Directed Mutagenesis kit (Stratagene, La Jolla, CA, USA). Resistant cells were transfected with 750 ng wild-type or mutant PRKDC 3′UTR luciferase reporter constructs and stable cell lines generated by puromycin selection. Stable cell lines were reverse transfected with 10 nM miR-1323 or miR-control. Cells were incubated for a further 24 h, and luciferase activity was assessed using the Dual-Luciferase assay system (Promega, Fitchburg, WI, USA).

Proteins were resolved on 12% polyacrylamide gels, transferred to a nitrocellulose membrane (Bio-Rad, Hercules, CA, USA) and blocked with 5% non-fat dairy milk in Tris-buffered saline (20 mM Tris and 150 mM NaCl, pH 7.4) with 0.1% Tween-20.

In the process and analysis of data, the background was initially excluded. Any ‘bad spot’ that showed a signaling value deviation above 50% of the average of repeated spots and/or a spot CV above 0.5 was deleted preceding the calculation of the mean and standard deviation for repeated spots. The normalization was completed by locally weighted regression. For the double-labeling experiments, the ratio of 2 sets of detected signals (log2 transformed) and P-value for the Student-t test were calculated. A P-value <0.05 was considered to be statistically significant. Significantly differential miRNAs were defined as those showing a logarithm (log2) for the ratio of the treatment group signal to control signal ≥1 or ≤−1. Multiple target gene prediction software including miRanda (http://www.mocrorna.org), TargetScan (http://www.targetscan.org), mirTarget2 (http://mirdb.org/miRDB), RNA22 (http://cbcsrv.watson.ibm.com/rna22.html), and microTv3.0 (http://diana.cslab.ece.ntua.gr/microT) were used to forecast some potential target genes showing a sequential differential expression profiling. The genes identified by at least 3 software, were taken as target genes to minimize the false positivity. In order to further characterize the molecular functions of forecasted target genes, MAS database (http://bioinfo.capitalbio.com/mas3/) and GDM database (http://gdm.fmrp.usp.br/) were used to make a preliminary analysis of forecasted target genes.

The radioresistant cells designated as A549R, were obtained by subjecting A549 cells to 5 months of fractionated irradiation with a total dose of 60 Gy and 10 additional passages without irradiation. No obvious change in the cell morphology was observed following irradiation (Fig. 1A). The radiosensitivity of the A549 and A549R cells was compared using a colony formation assay (Fig. 1B). Each point on the survival curve represents the mean surviving fraction from triplicate experiments. As expected, the A549R cells had a higher survival rate than the A549 cells, indicating that the A549R cells were more radioresistant than the A549 cells.

RNA extracted from the parental and resistant cells showed an RIN number >8, suggesting the high purity of total RNA without any protein or DNA residual (Fig. 2). The formaldehyde denaturing agarose gel electrophoresis image showed that the 28S:18S band neared 2:1 in a clear and trailing-free manner, without any non-specific band, suggesting the integrity of degradation-free RNA. The quality control experiment confirmed the suitability of the RNA sample for further miRNA microarray and reverse transcription PCR.

A microarray platform optimized for the analysis of a panel of human miRNAs was used to analyze and compare the pattern of miRNA expression between the parental A549 cell line and its counterpart resistant cell line. The expression profile of 43 miRNAs significantly changed (2.0- to 14.0-fold) including 25 upregulated miRNAs and 18 downregulated miRNAs in the resistant cells as compared to the parental cells (Table I). The four most significantly changed miRNA were chosen to be verified by RT-PCR which showed the same results as the array analysis (Fig. 3). Among these, miR-1323 had the highest fold-change in the exposure tissues. Compared to those in the parental cells, miR-1323 was significantly upregulated and miR-142-5p was significantly downregulated (P<0.05), respectively, in the resistance exposure group.

To identify miRNAs that regulate human PRKDC expression, we used the web-based algorithms TargetScan, miRanda and PITA. All these tools predicted a putative binding site for miR-1323 in the PRKDC 3′UTR (Fig. 4A). To validate the interaction between miR-1323 and the PRKDC 3′UTR, we cloned full-length wild-type PRKDC 3′UTR into the reporter vector to serve as the 3′UTR of the luciferase coding region. Acquired resistant cells were used to generate stable wild-type PRKDC-luciferase cell lines. These cells were then transfected with miR-1323 or miR-control, as a negative control, and luciferase activity was measured. miR-1323 overexpression increased luciferase activity driven by PRKDC 3′UTRs. To determine whether this effect is direct, the predicted miR-1323-binding sites in the 3′UTR of luciferase-PRKDC plasmids were mutated generating luciferase PRKDC-mutant constructs (Fig. 4B). Co-transfection of lung cells with the parental luciferase construct without the PRKDC 3′UTR did not significantly change the expression of the reporter. However, when the miR-1323 target site from the PRKDC 3′UTR was inserted into the luciferase construct, expression of luciferase was strongly increased when co-transfected with miR-1323 (Fig. 4C). This enhanced expression was suppressed by a single-base mutation in the binding site. Taken together, these data indicate that miR-1323 interacts with specific elements in the 3′UTR of PRKDC.

To investigate whether miR-1323 modulates endogenous PRKDC expression, we examined the effects of miR-1323 overexpression on PRKDC protein levels in human primary lung cell lines. Upon transfection, miR-1323 was overexpressed 5-fold in the resistant cells as determined by real-time quantitative PCR (RT-qPCR) (Fig. 5A). Nevertheless, miR-1323 overexpression caused an increase in the DNA-PKcs protein levels compared with the miR-control in the cell lines (Fig. 5A).

Owing to the role of DNA-dependent protein kinases (DNA-PKcs) in response to genotoxic insults, we explored the regulation of miR-1323 and DNA-PKcs following genotoxic stress stimuli. Firstly, resistant cells were treated with 8-Gy radiation. Radiation-induced accumulation of DNA-PKcs protein was found to correlate with a concomitant increase in miR-1323 levels (Fig. 5B). These data suggest that an increase in miR-1323 and accumulation of DNA-PKcs are a general response to DNA double-break DNA adducts caused by radiation. To understand the role of miR-1323 in response to DNA damage, we knocked down miR-1323 in resistant cells that were then treated with 8 Gy. Unlike the control cells, miR-1323-knockdown cells failed to recruit the DNA-PKcs protein (Fig. 5C).