Materials and Methods

Ethics Statement
The study was approved by the Medical Ethics Committee of the Shanghai Chest Hospital, Shanghai Zhongshan Hospital, Shanghai Changhai Hospital, and Shanghai Pulmonary Hospital. Written informed consent was obtained from all patients who participated in this study. In addition, the experiments in this study were conducted in accordance with approved guidelines and regulations.

Study subjects
The test cohort included 663 patients who had been diagnosed with either clinical stage IIIA or IV NSCLC and had received first-line platinum-based chemotherapy (no prior surgery, radiotherapy, or concurrent chemoradiotherapy) between March 2005 and January 2010. The association of the WEE1 tag SNP with the efficacy of chemotheraputic regimens was reviewed and evaluated. The validation cohort included 264 NSCLC patients who accepted the platinum-gemicitabine regimen in Shanghai Pulmonary Hospital between June 2010 and May 2013. All the patients had histologically confirmed NSCLC with the presence of at least one measurable and evaluable lesion. Eastern Cooperative Oncology Group performance status (ECOG PS) for the patients is 0–2 and the cardiovascular, hepatic, hematologic, and renal functions of the patients have been reviewed to assess chemotherapy tolerance.

Treatment
All patients in the test cohort underwent first-line platinum-based chemotherapy for 2 to 6 cycles with one of the following chemotherapy doublet regimens: (1) DNA-damaging agents regimen: either cisplatin 75 mg/m2 or carboplatin AUC 5 administered on day 1 every 3 weeks, in combination with gemcitabine 1250 mg/m2 on days 1 and 8 every 3 weeks, (2) tubulin-targeting drugs regimens: Either cisplatin 75 mg/m2 or carboplatin AUC 5 administered on day 1 every 3 weeks, in combination with navelbine 25 mg/m2 on days 1 and 8 every 3 weeks, or paclitaxel 175 mg/m2 on day 1 every 3 weeks, or docetaxel 75 mg/m2 on day 1 every 3 weeks, (3) other combination: Either cisplatin 75 mg/m2 or carboplatin AUC 5 administered on day 1 every 3 weeks, in combination with etoposide or bevacizumab. For the patients in validation cohort, either cisplatin 75 mg/m2 or carboplatin AUC 5 was administered on day 1 every 3 weeks, in combination with gemcitabine 1250 mg/m2 on days 1 and 8.

Tumor response
According to the Response Evaluation Criteria in Solid Tumors (RECIST), the short-term responses to chemotherapy treatment were evaluated after the first 2 cycles. The disease control rate (DCR) included complete response (CR), partial response (RR) and stable disease (SD). The objective response rate (ORR) consisted of complete response (CR) and partial response (PR). The progression-free survival (PFS) was assessed as the date from the chemotherapy treatment to objective disease progression or death (whichever occurred first) or the last progression-free follow-up. The overall survival (OS) was defined as the first day to receive chemotherapy treatment to the day of death or to the final follow-up.

WEE1 tag SNPs genotyping and Linkage Disequilibrium Analysis
The four tag SNPs (rs3829254, rs3910384, rs4370932 and rs1049403) representing 50 SNPs of WEE1 were selected from the CHB data of the phase 2 HapMap SNP database (http://www.hapmap.org/) using a minor allele frequency (MAF) cutoff of 0.05 and a correlation coefficient (r2) threshold of 0.8. Genomic DNA was extracted from patients’ peripheral blood samples using the QIAamp DNA Maxi Kit (Qiagen GmbH, Hilden, Germany) according to the manufacturer’s instructions. The SNPs were genotyped using the iSelect HD Bead-Chip (Illumina, San Diego, CA, USA) with the following quality-control criteria: genotyping call rate >0.95; MAF >0.05; and GenCall score >0.2. The concordance among replicates was >99.9% and the GeneMap software was used to analyze the data and prepare reports.
Linkage disequilibrium was analyzed in Haploview Software (Broad Institute of Harvard and Massachusetts Institute of Technology, Cambridge, MA, USA). Three putative functional SNPs, i.e. rs6486433, rs3763869 and rs3763868 in the WEE1 promoter region which are closely linked with rs3910384 were selected using MAF >0.2 and D’ >0.8 cutoff. The function of SNPs was predicted with FuncPred tools (National Institute of Environmental Health Sciences, USA). Individual haplotype frequencies for the selected SNPs were evaluated using the Hyploview or PHASE 2.0 Program (version 2.0.2) based on the Bayesian algorithm.

Construction of PGL4 WEE1 luciferase reporter plasmids and luciferase reporter Assay
Genomic DNA was extracted from the peripheral blood lymphocytes of patients. The WEE1 haplotype promoter region (650 bp;−24 bp ∼+626 bp) containing the 3 putative functional SNPs was PCR-cloned into PGL4 luciferase reporter plasmid. Forward primer: 5’-CACGGTACCAGCCAGTGTCCAGCCTAAGCA-3’ (Kpn I); Reverse primer: 5’-CTAACGCGTAGGTCTCCTCAGGTCCAGTCTCA-3’ (Mlu I). The PCR reaction was carried out at 95 °C, 3 minutes followed by 35 cycles (95 °C 30 s, 68 °C 30 s, 72 °C 30 s) and an extension period of 5 minutes at 72°C. PCR products were ran on an agarose gel and purified from the bands (MiniBEST Agarose Gel DNA Extraction Kit, TaKaRa, China), ligated, and transformed into DH5α. PGL4 WEE1 luciferase reporter plasmids containing two major promoter haplotypes, i.e. rs3910384 “A” allele-linked G/A/C (rs6486433/rs3763869/rs3763868) (termed “haplotype 1”) or rs3910384 “G” allele-linked T/G/G (rs6486433/rs3763869/rs3763868) (termed “haplotype 2”) were obtained by sequencing of the selected single clones.
The PGL4 WEE1 luciferase reporter and TK plasmid (DNA dosage PGL-4: TK = 9:1) were co-transfected into A549 or H1299 (human lung adenocarcinoma) cell lines using Lipo FiterTM liposomal Transfection regent (Hanbio, China). After the indicated time, the cells were lysed, and relative luciferase units (RLU) were measured using the dual-luciferase reporter assay system (Promega, WI, USA) according to the manufacturer’s instructions. To determine the WEE1 RLU after DNA damaging agents treatment, the transfected cells were subjected to 8 μg/mL cisplatin (DDP) or 40 μg/mL gemcitabine (GEM) for two hours and RLU was measured at 12, 24, and 36 hours after drug removal. Each assay was repeated at least three times.

Electrophoretic Mobility Shift Assay
Nuclear extracts of lung cancer cell lines H1299 and A549 were prepared using the method described by Edmead et al15. The sequences of the biotin-labeled double-stranded oligonucleotide probes of rs6486433, rs3763869 and rs3763868 are listed in Table S1. EMSA was carried out according to the manufacturer’s protocol (EMSA Kit ES009, Beyotime, China). Briefly, 10 μg nuclear extract were incubated in 20 μl binding reaction buffer (10 mM Tris-HCl, pH7.5, 50 mM KCl, 1 mM DTT, 10% glycerol) with the 2.5 nM biotin-labeled probe and 0.5 μg poly dI:dC (Sigma) for 30 min at room temperature. DNA–protein complexes were subjected to electrophoresis on 5% polyacrylamide gel, transferred to nylon membrane, and visualized by chemiluminescence imaging.

Statistical analysis
All statistical analyses were accomplished using SPSS® version 20.0 (SPSS Inc., Chicago, IL, USA). The association between OS or PFS of the NSCLC patients and the tag SNPs was evaluated by the Kaplan-Meier curves and verified by log-rank tests. Univariate and multivariate analyses were executed utilizing the Cox proportional hazard model to validate the significant factors related to the OS or PFS. Variables yielding P-values < 0.05 in the univariate analyses were used for the multivariate analyses, including gender (female vs. male), smoking status (ever vs. never), clinical stage (TNM IV vs. IIIb vs. IIIa), histology (adenocarcinoma, squamous, adenosquamous or others), the rs3910384 dominant model, the rs3829254 recessive model, and the rs4370932 additive model. For the dominant, recessive and additive models of each SNP, the model with the smallest P-value in univariate analyses was used for multivariate analyses. The relationship between the significant SNPs and the clinical characteristics was determined using χ2 test. All statistical analyses were two-sided and the differences were considered as statistically significant at P < 0.05.