Wnt Signaling in Oral Cancer Initiating Cells

Abstract

Oral cavity and pharynx cancer is the eighth most common malignancy amongst men in the United States, and more than five times this number are diagnosed on a global scale annually. Owing to the extremely heterogeneous nature of these tumors, the prognosis and survival rates are dismal and remain similar to what was reported over 20 years ago. Patients continue to be managed surgically with/without chemoradiation. More recently, targeted therapy was introduced with promising outcomes. To explain the heterogeneity of cancer, the concept of cancer initiating cells (CICs) has been extensively studied in epithelial cancers. These subpopulations of cancer cells can self-propagate and form heterogeneous clones that are capable of metastasis, recurrence and escaping conventional therapy. Targeting these specialized subpopulations will provide a relatively unexplored avenue to overcoming therapy resistance. As previously established, CICs and normal stem cells share common signaling pathways related to developmental and self-renewal programs. One such crucial pathway is the Wnt/β-catenin signaling pathway, the constitutive activation of which is associated with many epithelial cancers, including oral squamous cell carcinoma. We thus hypothesize that the presence of Wnt signaling responsive CICs in oral premalignant lesions and squamous cell carcinoma contributes to the initiation, progression in these tumors and possibly resistance to standard therapy. To address this hypothesis, we generated two aims; at the onset, we wanted to induce and bank premalignant lesions and oral squamous cell carcinoma (OSCC) and characterize these lesions at a histopathologic and molecular level. To generate OSCC tumors, we used a 4 NQO induced model in mice that were amenable to Cre/lox-based reporter lines for further characterization. After we successfully generated and banked primary tumors in this model, we characterized these tumors from a histological standpoint and immunolabeled these tumor cells for Wnt signaling-related markers. The progression of these tumors shared the phenotype of their human counterparts thus providing an effective platform to study the various tumor subpopulations. Other results revealed variable expression of β-catenin in premalignancy and OSCC. Interestingly, β-catenin nuclear positivity was noted in scattered cells at the invasive front of the malignant tumor islands. Further, rare populations of basal cells were positive for downstream markers of Wnt signaling such as β-catenin and LEF-1. Of note, this pattern of labeling was not noted in normal epithelium. Next, we planned to identify and study CICs in OSCC based on their activity for Wnt/β-catenin signaling pathway. We resorted to utilizing lineage-tracing experiments to identify and study CICs in OSCC. To visualize the fluorescent reporter, tamoxifen was administered, and lesions were traced for three days. Axin2 positive cells were noted in OSCC and not in papillary epithelial hyperplasia. To overcome the difficulties associated with in vivo tracing, a 3-D organoid platform was established to perform further lineage testing. Organoids from primary OSCC tumors were established and passaged. Further, upon Axin2 induction, evidence of Wnt signaling was noted in these organoids. These results suggest that Wnt activity is a key pathway upregulated in OSCC and further, it is plausible that these clones of cells will be susceptible to signaling pathway specific targeted therapeutics in the future for better outcomes.