The tumor suppressor Fbw7 is a ubiquitin ligase that targets many proteins for proteasomal degradation. This targeted degradation is a way to keep protein levels in check and prevent them from becoming too abundant and causing problems for the cell. Fbw7’s targeted proteins read like an all-star cast of tumor biology. Heading the A-list are oncogenic transcription factors such as Myc and Notch, which bind DNA to directly regulate their target genes and if dysregulated, can lead to abnormal cell growth and cancer. Unsurprisingly, tumors often contain mutations in the gene encoding Fbw7 which lead to oncogenic signaling through aberrant regulation of proteins like Myc. However, due to the complexity of Fbw7’s substrate network, it has been challenging to uncover transcriptional mechanisms underlying Fbw7-associated tumorigenesis. A recent paper published in eLife from the lab of Dr. Bruce Clurman, a member of the Human Biology and Clinical Research Divisions, sought to identify the transcriptional consequences of oncogenic Fbw7 mutations. The work represents a close collaboration with two Fred Hutch labs: the Henikoff lab, who developed many of the methods and computational approaches employed, and the Paddison lab, who pioneered high efficiency gene editing in neural stem cells. This study, led by graduate student Dr. Nayanga Thirimanne, took genome-wide approaches to uncover how Fbw7 mutations result in gene transcription changes that are in part due to altered binding of Myc and Jun transcription factors. Importantly, understanding the transcriptional mechanism underlying Fbw7-associated cancers may help in the development for Fbw7-based therapies in the future.
To understand the global transcriptional regulation by Fbw7, the authors first asked how gene expression changes with Fbw7 mutations in colorectal cancer cells. First, they identified genes similarly regulated by Fbw7 based on altered expression in both Fbw7 null cells (loss of two copies) and in cells harboring an Fbw7 heterozygous missense mutation (one mutated copy), as compared to wild-type cells. Missense mutations are the most common Fbw7 mutations in cancers, but the biologic consequences of these mutations are still poorly understood. Differentially regulated genes included ones related to epithelial to mesenchymal transition (EMT) and cell migration, cell proliferation, and the tumor suppressive p53 pathway. These analyses also suggested that regulation of the KLF family of transcription factors by Fbw7 may play an important role in shaping the Fbw7-dependent transcriptome. The researchers then asked how Fbw7 mutations globally influenced how these genes were being regulated by examining silencing and active chromatin modifications. Chromatin consists of DNA compacted around histone protein octamers, called nucleosomes. This helps pack over 2m of DNA into the tiny nucleus. Modifying the histone proteins can influence how accessible the DNA is to transcription factors, where different histone modifications can help activate or repress gene expression. With the help of the Henikoff lab, Thirimanne et al. looked at two histone modifications, histone 3 (H3) lysine-27 trimethylation (H3K27me3) and H3 lysine-27 acetylation (H3K27ac) to provide simple readouts of transcriptionally repressed or transcriptionally active chromatin, respectively, that were altered upon loss of Fbw7. As expected, increased H3K27ac levels near promoters of genes positively correlated with gene expression levels, while increased H3K27me3 correlated with decreased expression. To determine whether differential H3K27ac resulted from the altered binding of known Fbw7 substrates, the researchers looked for DNA motifs enriched at these regions that were indicative of specific transcription factor binding sites. Intriguingly, in regions with increased H3K27ac in Fbw7 mutants, they identified enrichment of the Jun DNA binding motif. To validate this finding, the researchers identified regions physically bound by Jun to confirm that altered H3K27ac enrichment at Jun binding motifs was indeed due to increased binding of Jun. In a similar fashion, the researchers also looked at binding of Myc, another critical oncogene. As Fbw7 promotes degradation of Jun and Myc, they predicted Jun and Myc binding would be globally increased across all binding regions. Contrary to this expectation, binding was altered at specific sites with a preference towards distal regulatory elements versus binding near promoter regions. This surprising finding revealed that rather than Fbw7 mutations causing widespread changes in gene expression, its loss results in dysregulation of only certain genes which could be advantageous for developing Fbw7-based therapies.
Wanting to investigate how Fbw7 might cause more site-specific rather than global changes, the researchers hypothesized that perhaps some sites were co-regulated by Myc and Jun, which are known to regulate common pathways. Strikingly, the authors found that about 20% of Myc and Jun binding sites were shared with each other and observed a coordinated gain in binding of both factors at many of these regions. One of the genes co-regulated by Myc and Jun in an Fbw7-dependent manner was CIITA, a gene that regulates expression of Major Histocompatibility Complex (MHC) Class II genes that are important for immune regulation. The authors found that CIITA and MHC class II genes were upregulated in Fbw7 mutant cells in addition to finding CIITA upregulated in Fbw7 mutant patient cancers. They speculate that upregulation of CIITA in colorectal cancers with Fbw7 mutations may provoke immune responses due to inappropriate expression of class II HLA proteins in tumor cells. Furthermore, Thirimanne et al. went on to investigate if the Fbw7-dependent transcriptional regulation observed in colorectal cancer cells was generalizable to other cell types. Here, they investigated gene expression and Jun occupancy in neural stem cells and found that surprisingly, both Fbw7-dependent gene expression changes and Jun occupancy closely mirrored their findings in colorectal cancer cells, especially with respect to activated pathways in EMT and related activities. Dr. Thirimanne explains that future work will investigate “Fbw7's role in KLF regulation and biological processes such as epithelial mesenchymal transition that we identified in this study.”
Collectively this work establishes a framework for understanding the transcriptional consequences of Fbw7 mutations across cell types and cancers in addition to highlighting how Fbw7 substrates may act synergistically to control gene expression. It also suggests that phenotypic changes associated with EMT may prove to be underappreciated and very important normal functions of Fbw7, and equally important consequences of oncogenic Fbw7 mutations in cancers. Dr. Clurman notes that “most important pathways in cancer are very complex, and we need to use a number of different approaches to understand their functions.” Another recently highlighted paper from the Clurman Lab “attacked this problem one amino acid at a time, while Nayanga studies it through thousands of transcription factor binding sites. It is only through a combination of these disparate approaches that we will make real progress.”
This work was supported by the National Institutes of Health, the National Cancer Institute, and the Genomics & Bioinformatics Shared Resource of the Fred Hutch/University of Washington Cancer Consortium.
UW/Fred Hutch Cancer Consortium members Bruce Clurman, Steven Henikoff, and Patrick Paddison contributed to this work.
Thirmanne HN, Wu F, Janssens DH, Swanger J, Diab A, Feldman H, Amezquita RA, Gottardo R, Paddison PJ, Henikoff S, Clurman BE. Global and context-specific transcriptional consequences of oncogenic Fbw7 mutations. Elife. 2022 Feb 28;11:e74338. doi: 10.7554/eLife.74338. Epub ahead of print. PMID: 35225231.