DESCRIPTION (provided by applicant): Non-melanoma skin cancers are the most common types of cancers in the United States, with over 1 million new cases per year, while increasing exposure to sunlight (UV) and environmental carcinogens appear to be the major causative factors. The long term goal of this application is to develop a safe and effective strategy using relatively non-toxic dietary cancer chemopreventive compounds to prevent skin cancer. The rationale for these studies is based on investigation from our laboratory and others, showing that: (1) Sulforaphane (SFN) presents abundantly in these berries and crucifers, respectively, strongly inhibits skin carcinogenesis in various mouse models in vivo; (2) Skin carcinogenesis is greatly enhanced in knockouts of Nrf2, a transcription factor critical in mediating anti-oxidative stress/inflammatory responses, and, importantly, SFN has no cancer protective effect in these Nrf2(-/-) mice; (3) Nrf2 and Nrf2-targeted HO-1 expression is decreased in skin tumors in Nrf2(+/+) mice; (4) Enhanced expression of c-Myc, Bcl-2, cyclin D1, beta-catenin, VEGF, Cox-2, and Bax is seen in all skin tumors; (5) UVB enhanced expression of inflammatory biomarkers in the skin of Nrf2(-/-) mice is substantially decreased compared with Nrf2(+/+) mice; (6) SFN induces cellular antioxidant proteins in MEFs (mouse embryonic fibroblast) and MEKs (mouse epidermal keratinocytes) of Nrf2 (+/+) but not of Nrf2 (- /-); (7) SFN blocks TPA-induced ear inflammation in C57BL/6J mice and not in Nrf2(-/-) mice; (8) Ursolic acid (UA), a natural triterpenoid with potent anti-inflammatory and cancer preventive activities inhibits TPA-induced ear edema and tumor promotion; (9) Increasing evidence suggests that epigenetics plays an important role in cancer development and progression including non-melanoma skin cancer; both UA and SFN modify CpG methylation of Nrf2 in JB6 mouse epidermal cell line; (10) Significant difference are observed in global DNA methylation profiles between DMBA/TPA papilloma vs normal skin and UVB-induced skin tumor vs non- irradiated skin analyzed by MeDIP-Seq analysis; and (11) UA and SFN inhibit TPA-induced JB6 transformation in soft agar. Based on the results summarized above we hypothesize that UA and SFN prevent skin carcinogenesis by altering epigenetic modifications with 3 specific aims: (1) To investigate the epigenetic events during in vivo skin cancer development and progression in B[a]P/TPA-induced skin carcinogenesis in Nrf2(+/+) vs Nrf2(-/-) C57BL6 mice, UVB-induced skin tumorigenesis in SKH-1 mice, and the cancer preventive efficacy coupled with epigenetic changes of UA and SFN; (2) To examine the epigenetic changes during transformation in vitro of mouse epidermal JB6 and human keratinocyte HaCaT cells and the effect of UA and SFN on these epigenetic changes; (3) To elucidate the molecular mechanisms involved in the epigenetic regulation of inflammatory/oxidative stress genes identified in Aims 1 and 2 by UA and SFN in mouse epidermal JB6 and human keratinocyte HaCaT cells. This study will advance our knowledge in developing better chemopreventive compounds for cancer chemoprevention clinical trials in skin cancer.