Abstract 3380: Synthetic lethal interaction between ARID1A mutation and BET bromodomain inhibition in ovarian clear cell carcinoma

Introduction:

Current treatment for advanced stage ovarian clear cell cancer is severely hampered by a lack of effective systemic therapy options, leading to a poor outlook for these patients. Given that ARID1A is inactivated by mutation in over 50% of ovarian clear cell carcinomas, we pursued an ARID1A synthetic lethal screening strategy to identify druggable targets in OCCC.

Experimental procedures:

We performed synthetic lethal kinome short hairpin (shRNA) screens in a large panel (n=14) of OCCC cell lines having different ARID1A mutation status. Hit validation was performed with isogenic ARID1A ko cell line pairs and in (patient-derived) xenograft mouse models.

Summary of the data:

We show here that BRD2 inhibition is synthetic lethal with ARID1A mutation in ovarian clear cell cancer cells. Importantly, inhibiting the BET family of proteins, to which BRD2 belongs, with small molecules specifically inhibits proliferation of ARID1A mutated cell lines both in vitro and in ovarian clear cell cancer xenografts and patient-derived xenograft models. We demonstrate that ARID1A loss leads to upregulation of the WNT ligand WNT10B, possibly causing a WNT dependency in the ARID1A mutant lines. BET inhibitors cause a reduction in WNT10B expression and WNT target genes such as MYC, JUN and WISP1, providing a potential explanation for the observed synthetic lethal interaction with ARID1A loss.

Conclusions:

Our study uncovered a new synthetic lethal interaction between ARID1A mutation and BET bromodomain inhibition, suggesting a new treatment strategy for ARID1A mutant ovarian clear cell carcinomas.

Citation Format: Katrien Berns, Joseph J. Caumanns, E Marielle Hijmans, Annemiek Gennissen, Bastiaan Evers, Bea A. Wisman, Gert Jan Meersema, Cor Lieftink, Roderick L. Beijersbergen, Hiroaki Itamochi, Ate G. van der Zee, Steven de Jong, René Bernards. Synthetic lethal interaction between ARID1A mutation and BET bromodomain inhibition in ovarian clear cell carcinoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 3380. doi:10.1158/1538-7445.AM2017-3380

BS69, a novel adenovirus E1A-associated protein that inhibits E1A transactivation

The adenovirus E1A gene products are nuclear phosphoproteins that can transactivate the other adenovirus early genes as well as several cellular genes, and can transform primary rodent cells in culture. Transformation and transactivation by E1A proteins is most likely to be mediated through binding to several cellular proteins, including the retinoblastoma gene product pRb, the pRb-related p107 and p130, and the TATA box binding protein TBP. We report here the cloning of BS69, a novel protein that specifically interacts with adenovirus 5 E1A. BS69 has no significant homology to known proteins and requires the region that is unique to the large (289R) E1A protein for high affinity binding. BS69 and E1A proteins coimmunoprecipitate in adenovirus-transformed 293 cells, indicating that these proteins also interact in vivo. BS69 specifically inhibits transactivation by the 289R E1A protein, but not by the 243R E1A protein. BS69 also suppressed the E1A-stimulated transcription of the retinoic acid receptor in COS cells, but did not affect the cellular E1A-like activity that is present in embryonic carcinoma cells. Our data indicate that BS69 is a novel and specific suppressor of E1A-activated transcription.

Isolation and expression of two novel Wnt/wingless gene homologues in Drosophila

Wingless (wg), the Drosophila homologue of the mouse Wnt-1 proto-oncogene, is a segment polarity gene essential in each segment for normal Drosophila development. We here report the isolation of two novel Drosophila Wnt homologues, DWnt-2 and DWnt-3, and thus the existence of a Wnt/wingless gene family in Drosophila. DWnt-2 and DWnt-3 map to chromosome 2 position 45E and chromosome X position 17A/B, respectively. DWnt-2 and DWnt-3, like the other known Wnt genes, encode amino-terminal signal peptides suggesting that the gene products are secreted proteins. The putative translation product of DWnt-2 and the carboxy-terminal half of the deduced DWnt-3 product are both rich in conserved cysteine residues. In comparison with other Wnt gene products, mostly about 40 × 10(3) relative molecular mass, the DWnt-3 protein has an extended amino terminus and a long internal insert, and its predicted relative molecular mass is 113 × 10(3). The expression patterns of these two Wnt/wg homologues are dynamic during Drosophila embryogenesis. The distribution of DWnt-2 transcripts is predominantly segmented, with the additional presence of transcripts in the presumptive gonads. Transcripts of both DWnt-2 and DWnt-3 appear to be associated with limb primordia in the embryo and may therefore specify limb development. DWnt-3 is also expressed in mesodermal and neurogenic regions. The distribution of DWnt-3 transcripts in cells of the central nervous system (CNS) during Drosophila embryogenesis suggests that DWnt-3 could be involved in CNS development.

The Wnt-1 (int-1) oncogene promoter and its mechanism of activation by insertion of proviral DNA of the mouse mammary tumor virus

Wnt-1 (int-1) is a cellular oncogene often activated by insertion of proviral DNA of the mouse mammary tumor virus. We have mapped the 5' end and the promoter area of the Wnt-1 gene by nuclease protection and primer extension assays. In differentiating P19 embryonal carcinoma cells, in which Wnt-1 is naturally expressed, two start sites of transcription were found, one preceded by two TATA boxes and one preceded by several GC boxes. In P19 cells, a 1-kilobase upstream sequence of Wnt-1 was able to confer differentiation-specific expression on a heterologous gene. We have investigated how Wnt-1 transcription was affected by mouse mammary tumor virus proviral integrations in various configurations near the promoters of the gene. One provirus has been inserted in the 5' nontranslated part of Wnt-1, in the same transcriptional orientation, and has functionally replaced the Wnt-1 promoters. Wnt-1 transcription in this tumor starts in the right long terminal repeat of the provirus, with considerable readthrough transcription from the left long terminal repeat. Another provirus has been inserted in the orientation opposite that of Wnt-1 into a GC box, disrupting the first Wnt-1 transcription start site but not the downstream start site. Most insertions have not structurally altered the Wnt-1 transcripts and have enhanced the activity of the normal two promoters.

Two domains in alpha interferons influence the efficacy of the antiviral response

Murine interferon-alpha 1 and murine interferon-alpha 4 share 80% of their amino acids, yet the proteins differ considerably in their ability to protect mouse or hamster cells against viral infection. With the aim of localizing areas within these proteins which influence the biological response we have constructed hybrid alpha 1 alpha 4 genes by means of homologous recombination of the parent genes. When the antiviral activities of these proteins were compared, it appeared that there are at least two domains that affect the biological response to these proteins: area A (amino acids 10-20) and area B (amino acids 55-67). These areas are presumably involved in the interaction between ligand and receptor. Most interestingly, hybrids in which area A from IFN-alpha 1 is combined with area B from alpha 4, have antiviral activities on homologous cells that are one to two orders of magnitude higher than those of the parent proteins.