The case-control study comprised a case group of 571 adult patients aged younger than 60 years with newly diagnosed primary lung cancer, recruited between December 2006 and February 2014, and a control group of 609 gender- and age-matched healthy individuals (±2 years) enrolled from July 2006 and February 2014. All the participants were residents of Fujian Province, living in Fujian for more than 10 years. The case and control groups were further stratified by age as ≤50, 51-54, and 55-59 years. In the present study, young and middle-aged was defined as an adult younger than 60 years.
Diagnoses of primary lung cancer were based on pathology after surgery or endoscopy. The patients were recruited from 3 hospitals: First Affiliated Hospital of Fujian Medical University; Fujian Medical University Union Hospital; and the Nine Hundredth Hospital of the People's Liberation Army Joint Logistics Support Force. Individuals in the control group were randomly selected from the three hospitals and the community, but excluded the following: direct relatives of the patients; or with pulmonary inflammation, secondary lung cancer, a previous history of cancer, or other critical diseases.
Data and sample collection
All epidemiological data were obtained through person-to-person interviews using a standardized questionnaire. The questionnaire covered general demographics; body mass index; food, tea, and alcohol consumption; living environment; smoking; passive smoking; history of lung diseases; occupational history; family history of cancer; and physical activity. All research variables were clearly defined.
Using the inquiry method for surveying dietary habits, respondents recalled their average frequency of consumption of foods (grams per day) in the last year (the year prior to study enrolment for all subjects) for a variety food items including seafood (fish, shellfish, snails, salted fish), fried food, fruits, vegetables, and salted vegetables. Questions regarding living environment included ventilation, use of exhaust fan, cooking oil fumes, irritating odors, and pollution in the area of residence. Smoking habits were defined as non-smoking, ≤30 pack-years (including smoking now and quit smoking now), and >30 pack-years (including smoking now and quit smoking now). Drinking tea was defined as drinking at least 1 cup per week for more than half a year. Drinking alcohol was defined as drinking at least once a week for more than half a year. Physical activity included physical exercise (none, 1-3×/wk, >3×/wk) and walking (none, occasionally, often).
A 5-mL non-fasting blood sample was collected from each participant for genotyping.
This study was approved by the Institutional Review Board of Fujian Medical University (Fuzhou, China) and all participants proved signed informed consent.
Selection of SNPs and prediction of binding sites with microRNAs
The KEGG (Kyoto Encyclopedia of Genes and Genomes) database (http://www.genome.jp/kegg/) was used to retrieve the inflammatory signaling pathways downstream of RIPK1 related to lung cancer, and the node genes in the pathway were searched. Sequences and functional information of these genes were found using the NCBI Gene Functional Annotation Database (http://www.ncbi.nlm.nih.gov/gene/).
SNPs at the target microRNAs on the 3′-UTR of the RIPK1 gene were predicted in reverse and screened using the following online prediction tools: microRNAanda (http:///www.microrna.org//); PicTar (http://pictar.mdc-berlin.de/); Diana-microT v 3.0 (http://diana.cslab.ece.ntua.gr/microT/); TargetScan Human 6.0 (http:///www.targetscan.org//); Target (PicTar (http://pictartar.mhttar.mdc.mdc-berlin.de/); Diana-MicroT v3.0 (http://diana.cslab.ece.ece.ntua.ntua.ntua.gr/://www.bioguo.org/microRNA SNP/).
Using the dbSNP (http://www.ncbi.nlm.nih.gov/SNP/) database, basic information regarding SNPs of the RIPK1 gene was obtained, including chromosome location, genome location, allele variation, risk allele, and minimum allele frequency. SNPs with minor allele frequency > 5% in the Han Chinese population were selected. RNAhybrid (http://bibiserv.techfak.uni-bielefeld.de/rnahybrid) was used to calculate the Gibbs free energy (G) difference (ΔΔG) between different alleles of SNPs and microRNAs, to evaluate the effects of these SNPs on the interaction between microRNAs and target genes. The greater the absolute difference (|ΔΔG|), the greater the influence of the SNP. Finally, the larger SNPs were selected.
Multiplex SNaPshot typing technology was used to achieve simultaneous automatic typing of the multiple SNP sites based on multiple primer extension. An SNP positively associated with lung cancer among the young and middle-aged Han population was selected as an experimental object of the present study.
The sequence information of the RIPK1 3′-UTR was downloaded from the University of California Santa Cruz website (http://genome.ucsc.edu/). The microRNAs and their target gene sequence variations were combined to predict the possible binding of microRNAs upstream and downstream of rs17548629 on the RIPK1 3′-UTR. The microRNA with the highest absolute context score value was selected as the experimental object.
Luciferase reporter assay
The base sequence of the predicted microRNAs (hsa-mir-1197) combined with the SNP locus (rs17548629), was downloaded from the authoritative database for microRNAs (http://www.mirbase.org), i.e., 5'-UAGGACAUGGUCUACUUCU-3'. Using this, hsa-mir-1197 MIC (mimics) fragments were synthesized for subsequent cell transfection experiments.
To obtain the DNA fragment of the target gene, primers were designed by Oligo 7 software to amplify 51 base pair regions of rs17548629, wildtype (Wt) and mutant (Mut), on RIPK1 3′-UTR and hsa-mir-1197 inhibitor sponge sequences (Table 1). The resulting product solution was diluted 50-fold to a final concentration of 200 nM for the ligation reaction.
Carrier Psi-CHECK2 (Figure 1A) was digested by SacI and XhoI at 37 ℃ for 2 hours. The digestion system consisted of: 5 µL 10× buffer, 5 µL Psi-CHECK2, 1 µL SacI, 1 µL XhoI, and 38 µL ddH2O. After agarose gel electrophoresis, the Psi-CHECK2 vector strip was recovered using a DNA gel Recovery Kit (Takara, Japan). Psi-CHECK2 bands obtained by double enzyme digestion, and the double-stranded template obtained by amplification, were linked using T4 DNA ligase to construct the recombinant plasmid for inserting into different target sequence fragments.
The plasmids were transferred into competent cells and evenly coated on Luria-Bertani broth plates containing 50 µg/mL ampicillin. The plasmids were cultured in a 37 ℃ incubator. After 16 hours, 200 µL with positive clones were collected and sequenced. The sequencing results were compared with the target gene sequence, and the plasmids were extracted after confirmation.
Cells of the human renal epithelial cell line 293T and lung adenocarcinoma cell line A549 were cultured respectively in Dulbecco's modified Eagle's medium containing 10% fetal bovine serum (with 1.5 mML-glutamine, 100 U/mL penicillin, 100 μg/mL streptomycin), in an incubator with 37 ºC ambient temperature and 5% CO2 saturated humidity. Cells in logarithmic growth phase were selected for the experiments.
The 293T cells and A549 cells were transiently transfected separately. Cells of each cell line were apportioned into the following 6 groups: positive control (psiCHECK2 vector inserted into a 1197 inhibitor sequence) and positive + miR-1197 mimics; wildtype psiCHECK2-RIPK1 control and wildtype psiCHECK2-RIPK1 + miR-1197 mimics; and mutant psiCHECK2-RIPK1 control and mutant psiCHECK2-RIPK1 + miR-1197 mimics. Cells were harvested 24 hours after transfection and luciferase activity was measured using the Dual-Luciferase Reporter Assay System (Promega, Madison, Wisconsin, USA) in accordance with the manufacturer’s protocol. Experiments were repeated three times in triplicate.
The first pair (positive control without and with miR-1197 mimics) was used to exclude false negatives caused by experimental system problems. To observe whether the mimics bind to the allelic type plasmid, the controls (wildtype or mutant psiCHECK2-RIPK1 without miR-1197 mimics) were used as blanks to the corresponding mimics group (wildtype or mutant psiCHECK2-RIPK1 with miR-1197 mimics). The relative activity values of the groups wildtype psiCHECK2-RIPK1 + miR-1197 mimics and mutant psiCHECK2-RIPK1 + miR-1197 mimics were cross-referenced to determine if the Wt and Mut forms of SNPs differ with changes in luciferase activity.
The chi-squared test (χ2) was used to compare the distribution differences between the case and control groups with regard to age, gender, education level, marital status, and other general demographic characteristics. The distribution of SNPs at a single site in the control group was analyzed to determine whether the distribution of SNP accorded with Hardy-Weinberg's law of genetic balance. The SNP allele frequencies of the case and control groups were calculated, and the difference in allele frequencies was compared with the χ2 test. A binary unconditional logistic regression model was applied to analyze the association of each site with lung cancer among the study population under co-dominant, dominant, over-dominant, recessive, and additive genetic models. A multi-classification logistic regression model was used to evaluate the association of each site with histopathological types of lung cancer in the population under these genetic models. The odds ratios (ORs), adjusted ORs, and 95% confidence intervals (CI) were calculated.
For target gene validation, the t-test was used to compare the relative activity of the reporter gene between the recombinant plasmid + mimics co-transfection groups and the corresponding control group. Analysis of variance of factorial design was applied to compare the activity of the reporter gene after the psiCHECK2 recombinant plasmid interacted with mimics. All the P values were bilateral test results, and the test level α was 0.05.