Date of Award
Doctor of Philosophy (PhD)
Benjamin A. Garcia
Recent studies have identified recurrent mutations affecting the activity of polycomb repressive complex 2 (PRC2) in a variety of human tumors. These genetic lesions include point mutations in histone genes (as in pediatric high-grade glioma; pHGG), homozygous loss of core PRC2 subunits (as in malignant peripheral nerve sheath tumor; MPNST), or chromosomal translocations affecting PRC2 subunits (as in endometrial stromal sarcoma; ESS). The overarching hypothesis of this work was that the integration of histone modification and proteome analysis with available molecular diagnostic information would lead to an improved understanding of oncogenic mechanisms and reveal new immunohistochemistry (IHC) targets. We examined both artificial model systems and patient-derived cell lines to explore the consequences of these mutations, and we also developed a novel method to extract protein from formalin-fixed paraffin-embedded (FFPE) tumor tissue to enable characterization of archival specimens for which model systems are not widely available. We report that in pHGG, the oncogenic mechanism of histone H3 K27M mutation is related to a decrease in H3 K27me3 and an increase in H3 K27ac, but, critically, the H3 K27M mutation does not render the PRC2 complex inactive. In MPNST, SUZ12 or EED loss leads to a complete loss of H3 K27me3 as well as H3 K27me2, as well as a gain of H3 K27ac. The report of H3 K27me2 loss in MPNST is novel and appears to be disease-specific, and consequently represents the best IHC target to date to reliably distinguish MPNST from histologic mimics. The new FFPE workflow we developed enables faithful characterization of archival tissue samples, potentially beginning a new era of translational proteomics research. Finally, we report that in ESS, the JAZF1-SUZ12 does not lead to global PRC2 dysfunction, but may instead cause a mislocalization of PRC2 on chromatin, leading to derepression of PRC2 targets such as EMILIN3. Indeed, EMILIN3 appears to be a promising IHC marker for the differentiation of ESS from its closest histologic mimic highly cellular leiomyoma (HCL). Collectively, this body of work demonstrates the value of combining proteomic and immunohistochemical analysis of model systems and pathology specimens for translational cancer epigenetics research.
Marchione, Dylan M., "The Intersection Of Proteomics And Pathology For Translational Cancer Epigenetics Research" (2019). Publicly Accessible Penn Dissertations. 3626.