Abstract 1665 Amino Acid Composition Defines a Class of DNA Binding and Phase Separating Intrinsically Disordered Regions

Protein intrinsically disordered regions (IDRs) are abundant in eukaryotic proteomes where they function as multivalent interaction scaffolds. IDR-dependent interactions often require short linear sequence motifs embedded within the longer region of disorder. Interestingly, some IDRs do not depend on a specific primary sequence but instead rely on the overall amino acid composition of the disordered region. Indeed, our lab has recently demonstrated that three proteins essential for DNA replication initiation (Orc1, Cdc6, and Cdt1) contain long IDRs that are necessary and sufficient for chromatin recruitment and DNA-dependent phase separation. Despite their common functionality, these initiator's IDR (inIDR) sequences share no sequence similarity, but instead maintain a similar sequence composition, a feature we refer to as 'compositional homology'. Furthermore, randomly scrambled inIDRs retain the ability to bind DNA, phase separate, and form co-condensates with wild type initiators, demonstrating that composition alone is sufficient to direct the function of inIDRs. We therefore hypothesized that IDRs that are functionally equivalent to inIDRs could be found through compositional homology and that these 'initiator-type' IDRs (itIDRs) would lead to the discovery of additional factors important in similar chromatin-contextualized processes. To test this, we developed a computational algorithm to identify compositional homology in disordered regions and have used this to search the Drosophila melanogaster proteome for itIDRs. Gene Ontology analysis of proteins containing itIDRs showed an enrichment for "Chromatin Organization", "Chromatin", and "Chromatin Binding", and amongst our hit proteins were both known phase separating proteins (e.g., Set2 [human homolog]) and proteins with no record of phase separation (e.g., XPC). To experimentally test the functional equivalency of identified IDRs, a representative set of itIDRs have been purified and fluorescently labeled to determine DNA-binding affinity via Electrophoretic Mobility Shift Assay (EMSA) and phase separation activity via confocal fluorescence microscopy. Notably, DNA binding activity was universally conserved across itIDRs. However, only a subset was found to mediate DNA-dependent phase separation. These results not only reveal fully uncharacterized sequences with high-affinity DNA binding capabilities that open the door for future functional studies, but they also add to our mechanistic understanding of known chromatin-contextualized processes. These preliminary results, along with our bioinformatic analyses, support our hypothesis that a distinct class of chromatin acting proteins can be identified based on IDR sequence composition. This work was supported by the Welch Foundation.