(1) To determine the functional impact of recurrent FBXW7 and SPOP mutations uncovered in endometrial cancers Ubiquitin-mediated proteasomal degradation plays a critical role in maintaining proper levels of cellular proteins, including proteins that promote tumorigenesis. The degradation process is tightly regulated, and is mediated by a large number of substrate-specific E3 ubiquitin ligases including the SKP1-CUL1-FBXW7 and SPOP-CUL3-RBX1 ubiquitin ligase complexes. In previous work, we discovered that FBXW7 and SPOP are highly mutated in serous and clear cell endometrial carcinomas, with most mutations residing in the substrate-binding regions of the encoded proteins (Le Gallo et al., Nature Genetics 2012). We hypothesize that the mutations act in a loss-of-function or dominant-negative manner to improperly regulate proteasomal degradation of one or more substrates. We are currently testing this hypothesis in ongoing functional studies using biochemical and cell-based approaches. We are focusing our efforts on recurrent mutations in these genes, as well as unique mutations that are predicted (in silico) to impact protein function. Progress is as follows: (1a) In previous reporting periods, using shRNA knockdown and Western blotting, Dr. Mary Ellen Urick identified several substrates of FBXW7 that exhibit increased protein levels upon knockdown of FBXW7 in serous endometrial cancer cell lines. We found that a number of these substrates also reproducibly show increased protein levels upon transient overexpression of mutant FBXW7 constructs compared to wildtype FBXW7 constructs in endometrial cancer cells. These findings suggest that the mutant forms of FBXW7 may be functionally altered and unable to properly regulate the turnover of key protein substrates. In the current reporting period, the Genome Engineering and iPSC Center at Washington University St. Louis successfully used CRISPR/Cas9 technology to stably introduce FBXW7 mutations into endometrial cancer cells, on a fee-for-service basis for the Bell laboratory. In ongoing studies that will extend into the next reporting period, we are utilizing the CRISPR/Cas9-edited cell lines to determine the biochemical consequences of the introduced FBXW7 mutations, and to determine whether the mutations confer sensitivity to drugs targeting downstream proteins. (1b) In previous reporting periods we assessed the ability of SPOP mutations to regulate an SPOP-CUL3-RBX1 protein substrate upon transient expression in the HEK293 cell line. Specifically, we transiently expressed thirteen mutants of interest, as well as a known loss-of-function mutant of SPOP that serves as an internal control, in HEK293 cells. We reproducibly found that three of the thirteen SPOP mutants correlated with increased levels of a specific substrate protein in transiently transfected cells. In parallel, we used co-immunoprecipitation and immunoblotting to determine whether the observed increase in substrate levels might result from impaired SPOP-substrate binding. In ongoing studies, Postdoctoral Fellow Dr. Fred Lozy is assessing the potential cellular effects of the three SPOP mutants in viability assays, clonogenic assays, and wound healing assays. (2) To identify somatically mutated driver genes for clear cell endometrial carcinomas In previous reporting periods, we performed a comprehensive mutational analysis of clear cell endometrial cancers, using a combination of whole exome sequencing of 16 tumor-nontumor pairs and targeted Sanger sequencing of 22 genes in another 47 tumors. In the current reporting period we finalized a manuscript describing our major findings and it was published in Cancer, accompanied by an Editorial. In the incoming reporting period we aim to assemble a cohort of clear cell endometrial tumors, and matched non-tumor tissues, for future whole genome sequencing and potentially RNA-sequencing. (3) To identify somatically mutated genes that drive advanced-stage endometrioid endometrial cancers In the previous reporting period we extracted and purified genomic DNAs from 19 advanced stage (stage III or IV) primary endometrioid endometrial tumor tissues as well as matched non-tumor tissues from the same patients. The purified DNAs were whole exome sequenced by the NIH Intramural Sequencing Center. The Roche SeqCap EZ Exome+UTR Library capture kit was used for exome capture and the captured libraries were sequenced on Illumina HiSeq2500 instruments. The resulting short sequence reads for each exome were aligned to the human reference sequence to assess a number of parameters including, but not limited to, the percentage of target regions covered, the average read depth for target regions, and the total number of within-target sequence variants. In the current reporting period, somatic variants were called in each tumor exome using four algorithms: Strelka, Shimmer, Mutect, and SomaticSniper. Within the Bell laboratory, Postdoctoral Fellow Xiaolu Zhang and Biologist Meghan Rudd rigorously filtered and annotated the somatic variant calls to delineate high-confidence somatic single nucleotide variants (SNVs) within the 19 advanced-stage tumor exomes. In parallel we performed microsatellite instability testing to determine whether tumors are microsatellite stable (mismatch repair proficient) or microsatellite unstable (mismatch repair deficient). We are in the process of further annotating our variant dataset to define high-confidence insertions and deletions, in a so-called validation screen. In the incoming reporting period, we aim to complete the validation screen, and subsequently rank (prioritize) genes based on the likelihood that the mutations they sustained are pathogenic driver mutations. Our prioritization strategy will consist of: 1. Prioritization of statistically significantly mutated genes, defined as genes that have acquired nonsynonymous and splice junction mutations at statistically significantly higher rates than the background mutation rate across all genes targeted by our exome sequencing. 2. Prioritization of somatically mutated genes that are consensus cancer genes. 3. Prioritization of potential cancer driver genes and druggable genes using in silico data-mining tools. 4. Prioritization of somatically mutated genes that form one or more significantly enriched functional groupings. 5. Prioritization of somatically mutated genes that form mutually exclusive networks. Subsequently, we will initiate a mutation prevalence screen in which we will sequence a subset of the prioritized genes in a larger cohort of advanced-stage endometrioid endometrial carcinomas with the goal of identifying novel driver genes based on mutation frequency, rate, and spectrum. Depending on the scale of the prevalence screen (i.e. the number of individual exons), it will be performed via high-throughput Sanger sequencing of tumor DNAs and subsequent sequencing of variant positions in matched normal DNA, or by custom capture and targeted next-generation sequencing of both tumor and matched normal DNAs. (4) A longer-term aim of this study is to determine the genomic etiology of serous and clear cell endometrial carcinomas that arise in women who have a prior history of tamoxifen use. We hypothesize that the mutational landscape differs between non-endometrioid tumors arising in tamoxifen users versus non-users. In the incoming reporting period, we will initiate efforts to obtain clinical specimens for future investigations designed to test this hypothesis.
Showing the most recent 10 out of 21 publications
