In the last decade, advances in sequencing technology have significantly increased the reports of gene fusions, from a few dozen to hundreds every year. Gene fusions arise from combining two different genes via chromosomal rearrangements and can result in fusion proteins with detrimental cellular functions such as cancer development. Gene fusions of the transcriptional regulator YAP1 are present in various kinds of human cancers, including brain and epithelial tumors. Tumors with YAP1 gene fusions lack other cancer-inducing mutations, suggesting that YAP1 gene fusions are the initiators of tumor formation. Postdoctoral fellow Dr. Frank Szulzewsky from the Holland lab led a multi-lab study in the Human Biology Division that evaluated the oncogenic potential of four distinct YAP1 gene fusions and identified a unifying feature as a potential therapeutic target. They recently published this work in the journal Genes & Development.
The investigators first sought to determine if YAP1 gene fusions YAP1-MAMLD1, YAP1-FAM118B, YAP1-TFE3, and YAP1-SS18 were sufficient to cause tumor formation in vivo. Since several of these YAP1 gene fusions are present in human brain tumors, the investigators used a transgenic mouse model that allows for tissue-specific gene delivery to express the YAP1 proteins in the central nervous system (CNS). They observed that all of the YAP1 gene fusions, but none of the wild-type versions of the gene partners or wild-type YAP1 itself, led to tumor formation in mouse brains. Considering that YAP1 gene fusions are not exclusive to brain tumors, and some are present in epithelial cancers, the investigators further expressed the gene fusions in peripheral tissues and similarly observed tumor formation. Dr. Szulzewsky explained these findings: “All analyzed YAP1 gene fusions are oncogenic when expressed in mice and thereby the causal tumor-initiating events and oncogenic drivers in the human cancers [in which] they are found and are therefore promising therapeutic targets.”
The regulatory and oncogenic functions elicited by YAP1 are attributed to its interactions with the TEAD transcriptional networks. All four YAP1 gene fusions retained the N-terminal segment of wild-type YAP1, which contains the TEAD binding domain. Thus, the authors hypothesized that all of the YAP1 gene fusions might exert the canonical YAP1-TEAD activity. To test their hypothesis, they transfected a panel of plasmids encoding for the YAP1 gene fusions, wild-type YAP1, or the wild-type gene partners into a human cell line. They analyzed the expression of five genes that are typically responsive to YAP1 activity by RT-qPCR. Only transfection with YAP1 gene fusions or wild-type YAP1 led to a significant increase of all five YAP1-responsive genes. Transfection of the same YAP1 panel (wild type or gene fusions), along with a YAP1 reporter plasmid, showed that YAP1 fusions exert similar activity as wild-type YAP1. Using a combination of RNA-seq and Cut&Run, the investigators also showed that all four YAP1 fusions exert YAP activity and bind to YAP1 target regions, such as enhancers located in intronic or distal intergenic regions. Taken together, this data demonstrates that YAP1 fusions exert the canonical YAP1-TEAD activity.
Given the similarities between wild-type YAP1 and the YAP1 gene fusions on downstream activity, the investigators tested if YAP1 gene fusions were sensitive to negative regulation by the Hippo pathway, which results in wild-type YAP1 phosphorylation, nuclear exclusion, and proteasomal degradation. Using the YAP1 reporter plasmid system, the investigators measured YAP activity for each one of the gene fusions in the presence of Hippo proteins. Whereas reporter activity induced by wild-type YAP activity was reduced when co-expressed with the Hippo proteins, reporter activity exerted by most YAP1 gene fusions was not affected. Sequence analyses showed that all YAP1 fusion genes encode for nuclear localization signals in the C-terminal. Using immunofluorescence, the investigators showed that YAP1 fusions are retained in the nucleus even in growth conditions that induce nuclear exclusion and degradation of wild-type YAP1. Dr. Szulzewsky explained the implications of these findings: “YAP1 fusions are oncogenic by exerting YAP activity that is resistant to inhibitory Hippo signaling, by 1) constitutive nuclear localization and 2) resistance to proteasomal degradation of the fusion proteins.
To identify a potential therapeutic strategy common to all tumors harboring YAP1 fusions, the investigators evaluated if the activity of YAP1 fusions was dependent on the interactions with TEAD transcription factors, as previously reported for wild-type YAP1.To this end, they generated YAP1 fusion mutants that cannot bind to TEAD transcription factors and tested their activity using the YAP1 reporter plasmid. For most mutants, YAP activity was reduced, showing that interaction with TEAD transcription factors is necessary for the activity of YAP1 fusions. In mice, the mutant YAP1 fusion proteins lost the ability to form tumors, showing that interaction with TEAD transcription factors is necessary for oncogenesis. To pharmacologically target the YAP1-TEAD interaction necessary for YAP activity, the investigators –in collaboration with the lab of Dr. Taran Gujral– used two small molecule inhibitors that reduce YAP1 functions by blocking the interaction between YAP1 and TEAD. In in vitro experiments, they observed a significant reduction of YAP activity in the YAP reporter system and reduced tumor cell growth in vitro. “The YAP activity of the different YAP1 fusions relies on the interaction with TEAD transcription factors. Inhibition of the interaction between the YAP1 fusions and TEAD was sufficient to reduce their YAP activity and inhibit the growth of tumor cell lines,” Dr. Szulzewsky added.
In a continued collaborative effort with the Gujral lab, the current research in the lab is focused on testing if pharmacological disruption of the YAP1-TEAD interaction sufficient to inhibit the growth of established YAP1 fusion-driven tumors in vivo. The investigators are also interested in identifying other pathways that are activated by these fusions and to determine if inhibition of these pathways be a feasible approach to inhibit the growth of these tumors. Finally, Dr. Szulzewsky added: “Some fusions also exert activity of the C’-terminal fusion partners. Therefore, it is crucial to determine what roles does the activity of the C’-terminal fusion partner play and how does it contribute to the function of the fusions and the biology of the tumors [in which] they are found in.”
This work was supported by grants from the National Institutes of Health, the Fred Hutch Pilot Project funding, the American Cancer Society, and the Discovery Award from the American Lung Association
Fred Hutch/UW Cancer Consortium members Eric Holland, Taran Gujral, Patrick Paddison, and Valeri Vasioukhin contributed to this work.
Szulzewsky, F., Arora, S., Hoellerbauer, P., King, C., Nathan, E., Chan, M., Cimino, P. J., Ozawa, T., Kawauchi, D., Pajtler, K. W., Gilbertson, R. J., Paddison, P. J., Vasioukhin, V., Gujral, T. S., & Holland, E. C. 2020. Comparison of tumor-associated YAP1 fusions identifies a recurrent set of functions critical for oncogenesis. Genes & Development, 34(15-16), 1051–1064. https://doi.org/10.1101/gad.338681.120