Introduction

Lung cancer is the most commonly diagnosed cancer type and a leading cause of cancer death globally. Non-small cell lung cancer (NSCLC) accounts for approximately 85% of all lung cancer cases. Despite the availability of surgical therapy, radiotherapy, and chemotherapy, prognosis of NSCLC is still poor with overall five-year survival rate being as low as 15%, mainly due to development of resistance to chemo- and radiotherapy, postoperative recurrence and early metastasis [1,2,3,4,5,6]. Even though molecular targeted therapeutic drugs, e.g. EGFR tyrosine kinase inhibitors (TKIs), have shown encouraging efficacies on NSCLC patients in recent years, the vast majority of NSCLC patients who are initially sensitive to TKIs acquire TKI resistance and undergo relapse, metastasis, or other progressions ultimately [7, 8].

Cancer stem cells (CSCs) are subpopulations of malignant cells that possess the abilities to self-renew and differentiate within a tumor [9]. The biological properties of CSCs have been linked to tumor resistance to chemotherapy and radiation, post-treatment recurrence, and metastasis, and presumably, specific, effective CSC targeting strategies might suppress cancer relapse [10, 11]. Notably, while the molecular mechanism via which cancer cells acquire stemness and the acquired stemness is maintained remains to be understood, Wnt/β-catenin signaling has been evidently associated with the development of cellular stemness in both cancer and benign tissues [12, 13]. Canonically, activation of the Wnt/β-catenin pathway is initiated by binding of Wnt ligands to their transmembrane receptors, followed by sequestration of β-catenin in the cytoplasm away from the destined destruction complex so that β-catenin can enter the nucleus and activate transcription of its target genes, many of which have been found to contribute to the development of cellular stemess [14]. Of note, activation of β-catenin signaling has been well demonstrated in various cancer types, most of which is attributable to gene alterations of the key components of β-catenin signaling. Typically, in colorectal tumors, the vast majority (80–90%) of clinical cases contain frameshift or truncating mutations in APC, resulting in the loss of ability to binding β-catenin [15]. Mutations of AXIN, which also lead to disruption of the destruction complex, have been identified likewise. In addition, mutations of β-catenin phosphorylation sites and consequent abrogation of β-catenin phosphorylation have been found in melanoma, which leads to β-catenin accumulation in the nucleus and transcription activation of its target genes [16, 17]. In such a context, of great interest is the fact that while enhanced nuclear localization of β-catenin has been observed in NSCLC [18] and hyperactive Wnt/β-catenin signaling is associated with increased drug resistance and distant metastasis of NSCLC [19], the aforementioned mutations are rare in NSCLC [20]. Hence, the molecular mechanisms underlying the activation of the pro-stemness β-catenin signaling in NSCLC remain to be investigated.

Of note, activating mutations of EGFR are common in NSCLC. Previous reports have shown a positive correlation between the presence of activating EGFR mutations and activation of β-catenin signaling in NSCLC [21], and the convergences between these two pathways have been indicated at multiple subcellular levels [21,22,23,24,25]. Notably, EGFR signaling reportedly increases cytoplasmic accumulation of β-catenin and nuclear translocation by either promoting release of β-catenin from the cytoplasmic membrane or disrupting the β-catenin destruction complex [24,25,26,27,

Materials and methods

Cell cultures

All tumor cell lines were obtained from the Cell Bank of Shanghai Institutes of Biological Sciences (Shanghai, China), Fu Erbo Biotechnology Co., Ltd (Guangzhou, China), or ATCC, and cultured as previously described [55]. Primary NLE cells were obtained and cultured in keratinocyte-SFM medium (KSFM) [55]. Authenticity of the cell lines was verified by short tandem repeat (STR) fingerprinting at the Medicine Laboratory of Forensic Medicine Department of Sun Yat-sen University (Guangzhou, China).

Tumor specimens from patients

Clinical tissue specimens were histopathologically diagnosed at the Sun Yat-Sen University Cancer Center from 2000 to 2004. The histological characterization and clinicopathologic staging of the cases were determined by following the standard provided in the current Union for International Cancer Control (UICC) Tumor-Node-Metastasis (TNM) classification. Each tumor and adjacent non-cancerous lung tissue pair was obtained according to our previous reports [55, 56]. Prior patients’ consents and approval from the Institutional Research Ethics Committee were obtained.

Plasmids and transfection

The SHCBP1 expression plasmid was generated by PCR subcloning the human SHCBP1 coding sequence into the lentiviral transfer plasmid pSin-puro (Clontech, Palo Alto, CA) to generate plasmid pSin-SHCBP1. To deplete SHCBP1 expression, two human shRNA sequences (sh1: GCGATTCAGAGCCTATCAA; sh2: CCATAGTGATCCATTGTCT) were cloned into the pSuper-retro-puro plasmid to generate pSuper-retro-SHCBP1-sh1 and pSuper-retro-SHCBP1-sh2. Human EGF-coding sequence was subcloned into the lentiviral transfer plasmid to generate plasmid pSin-EGF. Retroviral and lentiviral production and infection were performed according to the manufacturer’s instructions.

RNA extraction and real-time PCR

Total RNA extraction, reverse transcription, and real-time PCR were performed as described previously [55]. Primers were purchased from Invitrogen. The cDNA was acquired by using the GoScript™ Reverse Transcription Mix (Promega). All results were normalized for the expression of GAPDH, and relative quantification was calculated using the 2−△△CT formula.

Immunoprecipitation and protein purification

Lysates were prepared from 3 × 107 293T cells transduced with Flag-tagged SHCBP1 or vector using lysis buffer. Lysates were then incubated with FLAG affinity agarose (Sigma-Aldrich, St Louis, MO) overnight at 4 °C. Beads containing affinity-bound proteins were washed six times with immunoprecipitation wash buffer. Proteins were separated on SDS polyacrylamide gels stained with Coomassie blue, and all bands were subjected to mass spectrometry analysis. SHCBP1 and β-catenin protein purification was acquired by using immunoprecipitation and FLAG/HA competing peptides (MedChemExpress, Monmouth, NJ, USA) (see Supplementary Materials and Methods for details).

Western blotting analysis

Western blotting analysis was performed as described previously [55, 57]. The antibodies used are listed in Supplementary Information.

Luciferase reporter assay

Cells were seeded in triplicates in 24- or 48-well plates and allowed to settle for 24 h. Indicated plasmids plus 10 ng pRL-TK renilla plasmid was transfected into the cells using the Lipofectamine 3000 reagent (Invitrogen). Forty-eight hours after transfection, Dual-Luciferase reporter assays were performed according to the manufacturer’s protocol of Dual Luciferase Reporter Assay Kit (Promega, Madison, WI). The reporter plasmids containing wild-type (CCTTTGATC; TOPflash) or mutated (CCTTTGGCC; FOPflash) TCF/LEF DNA-binding sites were purchased from Upstate Biotechnology (Lake Placid, NY).

Nuclear and cytoplasmic extraction

NE-PER Nuclear and Cytoplasmic Extraction Reagents Kit (Thermo Scientific ™ Pierce, USA) was used and the assays were performed according to the manufacturer’s instructions.

Primary sphere formation

Tumor spheres were cultured according to a previous report [58]. Cells were allowed to further grow for 10 days, and the numbers of spheres were microscopically counted. The sphere yields were calculated by the number of spheres derived from cells based on the initially plated 2500 cells.

IHC assays

IHC assays in 207 NSCLC tissues were performed by using primary SHCBP1 antibody (Abgent, San Diego, CA) and quantified according to our previous report [59, 60]. The degree of immunostaining of indicated proteins was evaluated and scored by two independent observers as previously described, scoring both the proportions of positive staining tumor cells and the staining intensities. Scores representing the proportion of positively stained tumor cells was graded as: 0 (no positive tumor cells), 1 (<10%), 2 (10–50%), and 3 (>50%). The intensity of staining was determined as: 0 (no staining); 1 (weak staining = light yellow), 2 (moderate staining = yellow brown), and 3 (strong staining = brown). The staining index (SI) was calculated as the product of staining intensity × percentage of positive tumor cells, resulting in scores as 0, 1, 2, 3, 4, 6, and 9. Cutoff values for high- and low-expression of protein of interest were chosen based on a measurement of heterogeneity using the log-rank test with respect to overall survival. The optimal cutoff was identified as: the SI score of ≥4 was considered as high expression, and ≤3 as low expression.

Animal studies

Female BALB/c-nu mice (5–6 weeks of age, 18–20 g) were purchased and housed in specific pathogen-free facilities on a 12-h light/dark cycle. All experimental procedures were approved by the Institutional Animal Care and Use Committee of Sun Yat-Sen University. At least five nude mice per group were used to ensure the adequate power and each mouse with different weight was randomly allocated. ICG-001 or vehicle has begun to be used 7 days after tumor cell injections in animals. ICG-001 (50 mg/kg) or vehicle (PBS) were intraperitoneally injected every 3 days for 3 weeks as described previously [50, 61]. For bioluminescent imaging assay, 15 min prior to imaging, mice were injected intraperitoneally (i.p.) with 150 mg/kg luciferin. Following general anesthesia, images were taken and analyzed with Spectrum Living Image 4.0 software (Caliper Life Sciences). Tumor growth was monitored weekly by in vivo imaging and photon radiance measurement, and the final monitoring was performed after 4 or 6 weeks after the injection. Bioluminescent imaging of primary tumors and metastases was performed in a blinded manner.

Statistical analysis

Sample size was determined by power analysis to achieve a minimum effect size of 0.5 with P < 0.05 and all sample sizes were appropriate for assumption of normal distribution. Variance within each group of data was estimated and was similar between compared groups. Data analysis was performed by two independent investigators who were blinded to the sample groups. All statistical analyses were performed using the SPSS 13.0 (IBM) statistical software package. The Kaplan–Meier method was used to establish survival curves. The statistical significance of various variables for survival was analyzed using the Cox proportional hazards model in the multivariate analysis. Correlation between SHCBP1 expression and T-, N-, M-classification was analyzed using chi-square test. Comparisons between groups were performed with a two-tailed paired Student’s t-test. In all cases, P < 0.05 was defined as statistically significant.