PROJECT SUMMARY
Human cancers can be classified according to their cellular lineage, which is often defined by the presence of
master regulator transcription factors. In several clinical contexts, opportunities exist to target master regulators
as a therapeutic strategy (e.g. antagonists of the Androgen Receptor or degraders of Ikaros/Aiolos). However,
scenarios also exist in which the aberrant expression of a master regulator can promote lineage plasticity, which
can allow tumor cells to evade targeted therapies. These clinical observations highlight the importance of
investigating the biology and mechanisms of lineage master regulators in oncology. Our lab recently reported
the existence of a tuft cell-like lineage of human carcinoma, which is a tumor defined by the expression of
POU2F3. We described tuft cell-like carcinomas as a non-neuroendocrine form of small cell lung cancer (SCLC),
however subsequent studies have shown that POU2F3+ carcinomas exist in several other organs as well. In this
research proposal, we seek to define the unique biology and molecular mechanisms that distinguish POU2F3+
carcinomas from other forms of human cancer, with a long-term objective of advancing therapeutic solutions for
this newly identified tumor lineage. The first Aim of this project will focus on revealing molecular mechanisms,
building upon our recent discovery of OCA-T1/OCA-T2 as obligate coactivators of POU2F3 in the tuft cell lineage.
For this work, we will employ our latest innovations in genome editing technology and epigenomics to define the
effectors of the POU2F3/OCA-T complex in POU2F3+ SCLC. These efforts will have potential to reveal
biochemical opportunities that allow for pharmacological modulation of POU2F3 in cancer. The second Aim will
evaluate the unique biological properties of neoplastic tuft cells in vivo, which builds upon our preliminary data
showing that engineering of cancer gene mutations into pulmonary tuft cells leads to the generation of SCLC-
like tumors that express POU2F3. These unique genetic models will allow us to explore biological questions that
would otherwise be impossible using human cancer cell lines. Specifically, we will investigate a) which specific
tumor genotypes are permissive for forming tuft cell carcinomas b) which tumor genotypes drive lineage plasticity
in this context, particularly transitions between tuft and neuroendocrine fates and c) how cytotoxic chemotherapy
selects for specific cellular states in the setting of a tuft cell carcinoma. The third Aim of this proposal seeks to
understand tuft cell identity as an acquired state and as a drug resistance mechanism in prostate
adenocarcinoma. We have developed methods for reprogramming carcinomas into a tuft cell state using
POU2F3/OCA-T co-expression, which we will incorporate into the RapidCaP model of prostate cancer to study
its impact on disease progression and castration resistance. Collectively, the research outlined in this proposal
will advance our fundamental understanding of the biology and therapeutic vulnerabilities of POU2F3+ human
carcinomas.