Hedgehog/Gli promotes epithelial-mesenchymal transition in lung squamous cell carcinomas

Tissue specimens of the UCSF cohort were collected from 14 patients who underwent surgical resection for lung SCC at the Thoracic Oncology Program at UCSF. Tissue specimens of the Tianjin cohort were collected from 177 patients who underwent surgical resection for lung SCC at the Tianjin Medical University Cancer Institute and Hospital. Samples were fixed in formalin and embedded in paraffin to make tissue slides. The study with UCSF patient tissues was approved by the Committee on Human Research (CHR approval number: H8714-11647-10) at the University of California, San Francisco (UCSF), and that with Tianjin cohort was approved by the Tianjin Medical University Cancer Institute and Hospital. Written, informed consent was obtained from each patient before specimen collection.

Immunohistochemistry, immunofluorescence and western blot were performed following standard procedures. Antibodies applied to detect protein expressions in IHC and IF were Gli1 (sc-20687 Santa Cruz Biotechnology, Santa Cruz, CA) at 1:100, Shh (ab 50515 Abcam, Cambridge, MA) at 1:100, Smo (ab 72130 Abcam) at 1:200, Ptch1 (Santa Cruz that Biotechnology,) and E-Cadherin (EMD Millipore) Smo (Sigma, St. Louis, MO) at 1:100, E-cadherin (sc-7870, Santa Cruz Biotechnology) at 1:100, and β-catenin (BD Biosciences, San Jose, California) at 1:400. Antibodies used in Western blot were Gli (C68H3, Abcam) at 1:1000, E-Cad (HECD-1 MED Milliopore, Darmstadt, Germany) at 1:1000 and Actin (A5441, Sigma) at 1:5000. Total protein extraction was performed with M-PER Mammalian Protein Extraction Solution (Thermo Scientific, Waltham, MA), and 40ug of proteins were analyzed in Western blot.

Human lung SCC cell lines H2170, H1703, H1869 and SK-MES-1 were purchased from the Cell Culture Core Facility at Harvard University (Boston, MA, USA). The cell lines were cultured in RPMI 1640 (Life Technologies, Carlsbad, CA) supplemented with 10% fetal bovine serum (FBS) and antibiotics. Cells were seeded one day before treatment with Gli-I and vismodegib (Selleck) at different concentrations and Shh recombinant proteins (eBioscience) for 30 hours, with vehicle (DMSO) as controls. Cells were subjected to the following analyses of immunofluorescence and migration assay. In migration assays, four wounds were made in each condition, and cell migration was presented by the average of distance differences between 30 hr and 0 hr. All experiments have been conducted for more than three times, and representative results were included in the text.

Kappa test was used to evaluate the association between the expressions of Hh pathway components and EMT markers, and between Gli1 and recurrence/metastasis. IHC scores of 1–3 were grouped as positive “+” , and 0 was grouped as negative “-” for dichotomized analysis. Non-parametric Kendall’s tau-b statistics was used to determine the correlation between IHC staining of Hh components. Two-sided student’s t-test was performed for migration assays. A p value <0.05 was indicated as *, 0.01 as **, and 0.001 as *** in corresponding figures. Data analysis was performed using SPSS 17.0 software.

We first investigated the protein expression of key Shh pathway components in lung SCC tissue samples. Formalin-fixed, paraffin-embedded (FFPE) tissue specimens were collected from 177 lung SCC patients who underwent surgical resection at the Lung Cancer Center at Tianjin Medical University Cancer Institute and Hospital. The protein expressions of Shh, Smo, Ptch1 and Gli1 were characterized by immunohistochemistry (IHC), and scored on a scale of 0–3 (negative, mild positive, positive, and strong positive). Representative samples in each score category were summarized in Figure 1A. More than 90% of the lung SCC tissue samples examined were positive for the signal molecule Shh, while 53% and 61% were positive for downstream components and transcriptional targets Ptch1 and Gli1 respectively (Figure 1B). Previous studies have demonstrated limited expressions of Shh components in normal lung tissues at the mRNA and protein levels in NSCLC [27, 28], therefore the expression of key Shh signaling components indicates the activation of Shh pathway. No significant association was found between expressions of Shh, Smo, Ptch1 and Gli1 and patients’ characteristics (sex, age, tumor size, or degree of tumor differentiation) (Table 1) (P > 0.05, data note shown).

Next we analyzed the association between expressions of key components in the Shh pathway. Kendall’s tau-b correlation tests yielded significant correlations between every two factors (p = 0.000), while Kappa’s test suggested strong positive association between SHh and Gli1(p = 0.000) (Figure 1C), suggesting the canonical Shh pathway is activated in lung SCC. These data are consistent with previous reports that the upstream Shh signaling has correlations with downstream targets in NSCLC [29, 30]. Taken together, our results suggest that aberrant activation of the Shh pathway plays an important role in lung SCC.

E-Cadherin is a well-established EMT biomarker, and its expression has been suggested to be associated with cancer recurrence and metastasis [5]. The expression of β-Catenin also serves as a biomarker for EMT [31]. To investigate whether the Shh/Gli signaling plays a role in EMT regulation in lung SCC, we first examined 14 lung SCC patients who underwent surgical resection for lung SCC at the Thoracic Oncology Program at UCSF. Eight of fourteen samples showed reverse correlation between E-Cadherin and Gli1 expressions (three representative samples were shown in Figure 2A). To confirm the reverse correlation between EMT markers and Gli1 expressions in lung SCC, we further analyzed E-Cadherin and β-Catenin expressions and correlated with Gli1 expression in the Tianjin cohort. Our results revealed strong reverse correlations between Gli1 and E-Cadherin (p = 0.003), as well as Gli1 and β-Catenin (p = 0.004) (Figures 2B and C). We also observed reverse correlation between Gli1 and E-Cadherin expression at different areas within one sample in multiple cases due to the heterogeneity of tumor cells (Figure 2), further supporting the reverse correlation between Gli1 and EMT marker expressions. Moreover, our analysis revealed that Gli1 significantly correlated with recurrence and metastasis of lung SCC in the Tianjin cohort (p = 0.033; Figure 2C). Consistent with the tissue expression analysis, we observed that Gli1 expression reversely correlated with E-Cadherin expression in four human lung SCC cell lines, H1703, H1869, H2170 and SK-MES-1 (Figure 2D). Taken together, our results indicate the essential role of Gli1, a downstream effector of Shh pathway, in enhancing EMT, which in turn promotes recurrence and metastasis in lung SCC.

To further understand the role of Shh/Gli in EMT regulation in lung SCC, we manipulated the Shh/Gli signaling pathway in lung SCC cell lines to examine its impact on cell migration and E-Cadherin expression. To inhibit the Shh/Gli activity, we applied two small molecule compounds: Vismodegib and a novel Gli inhibitor. Vismodegib (also known as GDC-0449) is a Smo inhibitor recently approved by the U.S. Food and Drug Administration to treat adult patients with basal cell carcinoma [32‐35]. Multiple clinical trials are evaluating the use of vismodegib in other types of cancer, in addition to other candidate drugs that targets Hh signaling [32, 36]. The novel Gli inhibitor (Gli-I) developed by our lab specifically inhibits Gli1 and Gli2 transcriptional activity [28]. To stimulate the pathway, we applied recombinant Shh proteins.

We first performed cell migration assay in lung SCC cell lines H1703 and H2170 after the treatments with either Shh/Gli inhibitors or recombinant Shh proteins. Cells treated with Vismodegib and Gli-I exhibited significantly slower migration in 30 hours; on the other hand, cells stimulated by Shh proteins migrated significantly faster (Figure 3). This data strongly suggests that Shh/Gli signaling plays an essential role in regulating the migration of lung SCC cells. Next we examined E-Cadherin expression in these cells by immunofluorescence staining. We observed that E-Cadherin expression was up-regulated in those lung SCC cells treated with Shh/Gli inhibitors and down-regulated in the cells stimulated by Shh proteins (Figure 4). This is consistent with the mobility of lung SCC cells after the different treatments (Figure 3). Therefore, our results indicate that Shh/Gli signaling may promote cell migration by down-regulating E-Cadherin expression in lung SCC.