Oral Squamous Cell Carcinomas (OSCCs) represent the vast majority cases of squamous cell carcinomas of the head and neck. Currently the 5-year survival rate for OSCCs remains at ~50-60%, largely due to their therapy resistance, recurrence, distant metastasis, and late diagnosis. It has been proposed that in many cancer types, including OSCCs, cancer cells are organized into aberrant cell hierarchies, including cancer stem cells [CSCs, also known as cancer initiating cells (CICs)] and their more differentiated daughter cancer cells. Several recent studies in murine cancer models have used lineage tracing to identify cancer cells that bear markers of tissue stem cells, and which give ris to long-lived clones that drive tumor expansion. This in vivo lineage tracing approach typically involves tamoxifen-induced transient activation of a Cre recombinase-estrogen receptor fusion protein (CreERT), which leads to activation of a conditional Cre reporter (e.g., the Rosa26-Stop-YFP conditional reporter). CreERT expression can be controlled by a cell type-specific promoter and CreERT-expressing cells and their daughter cells can be genetically labeled by the Cre reporter (e.g., YFP) upon Cre-mediated recombination (thus forming a cell lineage). Although these recent lineage tracing studies in murine models have provided compelling evidence to support the existence of CSCs in intact tumors, they give few clues about the molecular mechanisms required to ensure CSC self- renewal, nor about whether different signaling pathways might control distinct CSC fates. It is also unclear whether CSCs form a homogeneous population or whether different subpopulations of CSCs might exhibit different behaviors (e.g., differential survival) upon chemotherapy. In OSCCs, although it is thought that CSCs are responsible for their metastasis, recurrence and therapy resistance, the identity of OSCC CSCs remains elusive. In this R21 project, an innovative approach is proposed to study CSCs during OSCC development and upon therapy in a well-established carcinogen-induced murine OSCC models by a novel pathway-based in vivo lineage tracing strategy. This approach involves genetically marking subsets of cancer cells based on activation of a panel of pathways known to be involved in regulating stem cells (e.g., Axin2-CreERT or Gli1-CreERT for labeling Wnt or Hedgehog signaling-responsive cells, respectively) (Aim 1). By quantitatively comparing clonal outcomes between different pathway-labeled cell populations, one can reveal which pathways are active in CSCs and how cell fate choices are affected by pathway activity. In Aim 2, how oral cancer-initiating cells/CSCs in this murine model (i.e., defined based on their activit for stem cell-related pathway) respond to therapeutic interventions at the clonal level will be further determined by lineage tracing. The long-term goal of this project is to use this pathway-based lineage tracing approach to identify which pathways correlate with generation and regeneration (upon therapy) of CSCs or with therapeutic resistance, and through this to develop and test novel therapy to target OSCC CSCs.