Poster V7 - Understanding mechanisms and consequences of condensation in oncogeneic receptor tyrosine kinase fusions.

Introduction: Receptor tyrosine kinase (RTK) fusions are a large class of oncogenes found in ~5% of all cancers. RTK fusions are expressed in the cytoplasm, and their oncogenic signaling is driven by activity of the kinase domain. While potent kinase inhibitors exist and elicit initial responses in patient tumors, drug resistance inevitably occurs. There is thus a continued need to better understand mechanisms of signal transmission of this large class of oncogenes. It was recently reported that, in certain patient tumors, RTK fusions like EML4-ALK form large, micron-scale protein condensates. Several subsequent studies ultimately concluded that fusion condensation was essential for its signaling. However, it is not mechanistically clear how condensation promotes signaling through the many downstream pathways. Moreover different variants of EML4-ALK and other RTK fusions do not obviously form large condensates when expressed in cells, further calling into question the roles of condensation. The aim of our study was to systematically determine the extent to which protein condensation causally controls downstream signaling from EML4-ALK and related RTK fusions.

Materials and

Methods: RTK fusions were heterologously expressed in isogenic cell lines. Fusion expression, condensation, RTK activity and downstream Ras/Erk signaling were quantified on a single-cell level using high content fixed-cell immunofluorescence. To determine minimal requirements for RTK fusion signaling, synthetic fusions were generated by fusing the intracellular domain of an RTK to stoichiometrically defined multimers. These synthetic proteins were then expressed and quantified as described for the oncogenic RTK fusion. To modulate the degree of condensation, fusions were co-expressed with exogenous adapter proteins or CluMPS reporters, which amplify protein clusters through multivalency (Mumford 2024, Cell Systems).

Results, Conclusions, and Discussions: Contrary to a current model, EML4-ALK condensation negatively correlated with both downstream Ras/Erk signaling, as well as with EML4-ALK concentration. By contrast, Ras/Erk signaling showed strong correlation with fusion expression (Figure 1A). Condensate formation could be increased either by supplementation with exogenous Grb2 or by co-expression with CluMPS. However, enhanced condensation did not correspond to higher Ras/Erk activity (Figure 1B). Thus, we find no evidence that fusion condensation promotes its signaling. In agreement, when we measured the above parameters in a small library of oncogenic RTK fusions, we found that, uniformly, signaling correlated with expression, but the relationship of signaling and condensation varied for each individual fusion. Finally, our synthetic RTK fusion screen found that, surprisingly, cytoplasmic dimerization of RTK tails was sufficient to elicit strong signaling, despite the absence of fusion condensation. Our data indicate that micron-scale condensation is dispensable for signaling from RTK fusions, and that small-scale multimerization is sufficient. These results clarify our mechanistic understanding of RTK fusion biology and motivate further mechanistic studies to delineate how these small oncogene assemblies drive signaling from the cytoplasm.

Acknowledgements (Optional):