We first set out to determine whether MYC and MYCN bind to RNA in cells. This analysis has been completed and showed that both oncoproteins bind to most nascent RNAs in cells during or shortly after transcription. For both oncoproteins, we have shown that they recruit the RNA degrading machinery to their target sites on RNA, firmly establishing MYC proteins as part of the cellular RNA degrading machinery. The subsequent analysis also showed that MYC and MYCN exist in a dynamic equilibrium between DNA and RNA-bound forms; for example, delaying transcription elongation, which causes an accumulation of non-spliced nascent RNA, leads to an almost complete shift of MYC from its canonical position at promoters to a localization on nascent RNA.
Under this aim, we also analysed how RNAs are recognized by MYC proteins. To do so, we performed an siRNA screen that tested whether multiple proteins of the MYC interactome are required for the localization of MYC or MYCN on RNA. This screen identified multiple proteins involved in RNA degradation as inducers of MYC´s phase transition, but no RNA binding proteins that are required for MYC´s phase transition in response to stress. This result suggested that MYC and MYCN do not rely on other proteins to recognise RNA, leading to the key observation that they are direct RNA-binding proteins with multiple RNA-binding domains. This finding also enabled the identification of mutant MYC and MYCN alleles that exhibit reduced RNA binding but unaltered DNA binding. In particular, analysing an MYC allele (MYCRBRIIIMUT), which carries several point mutations in the major MYC RNA-binding domain, has enabled us to dissect the contributions of DNA- and RNA-binding-dependent mechanisms to MYC's oncogenic functions.
We also established both a quantitative immune fluorescence assay for MYC multimerization in response to stress and Nanobret assays that measure the formation of MYC.MYC complexes and MYC.MAX complexes in cells in a quantitative manner. Establishing these assays enabled us to screen a library of approximately 2,000 FDA-approved compounds for molecules that inhibit MYC multimerization. This led to the demonstration that mTORC1 activity is required for MYC multimerization. The functional consequences and the underlying of this regulation will now be analysed.
We have identified an allele of MYC that is deficient in RNA binding but indistinguishable in its ability to activate transcription from wtMYC (MYCRBRIIIMUT). We have shown that this allele fails to suppress innate immune signalling in PDAC cells since it fails to suppress the accumulation of double-stranded RNAs and DNA/RNA-hybrids that are derived from nuclear R-loops onto TLR3, a pattern recognition receptor that directly activates the TBK1 innate immune kinase. Intriguingly, MYCWT but to a much lesser degree MYCRBRIIIMUT binds to these RNAs, arguing that this regulation is directly mediated by RNA-bound MYC. This allele maintains cell proliferation in culture and is able to activae transcription, but is unable to sustain tumorigenesis in vivo. Additional alleles that separate DNA- from RNA binding are being analysed.
We have also begun to study mutant alleles of MYC that retain the ability to bind RNA but are specifically defective in the ability to form multimers. These alleles shows defects in multimerization that are independent of RNA binding and show strong defects in MYC function, which we are analysing at the moment.
We have adapted the MYC.MYC and MYC.MAX proximity ligation assays that we established in tissue culture for the analysis of MYC multimer formation in human tumours. Initial data from human colon tumours show that MYC.MAX heterodimers are present both in the tumour tissue and in the adjacent normal proliferating tissue at similar levels. In contrast, MYC.MYC multimers are virtually absent in normal tissue, but highly enriched in the tumour tissue.
The finding that MYC and MYCN are direct RNA binding protein has shifted the focus of the latter part of the project to the analysis of mutant alleles of MYC that are defect in RNA binding but retain the ability to bind to DNA and activate transcription. We have also completed a screen of MYC cofactors for their ability to promote immune evasion, as originally proposed. This points to factors involved in transcription termination as being critical for immune evasion, and we are completing the analysis of the underlying mechanism.
