Back

Mapping interactions between disordered regions reveals promiscuity in biomolecular condensate formation

Gilat, A.; Dubrueil, B.; Levy, E. D.

2023-07-04 synthetic biology
10.1101/2023.07.04.547715 bioRxiv
Show abstract

Intrinsically-disordered regions (IDRs) promote intracellular phase separation and the formation of biomolecular condensates through interactions encoded in their primary sequence. While these condensates form spatially distinct assemblies in cells, it is unclear whether such specificity can be conferred by IDRs alone. Indeed, IDRs exhibit high conformational flexibility whereas specificity in protein recognition is generally associated with well-defined 3D structures. To characterize IDR-IDR interactions and assess their ability to mediate self-specific partitioning, we developed a synthetic system of Multivalent IDRs forming Constitutive DROPlets (micDROP). We investigated ten natural IDRs that underwent phase separation in micDROP. These IDRs exhibited a wide range of saturation concentrations in vivo, which correlated well with their total sequence stickiness. We then probed IDR-IDR specificity by co-expressing pairs of IDRs fused to homologous scaffolds that did not co-assemble. We observed a high degree of promiscuity, particularly among IDRs from the proteins Ddx4, DYRK3, ER, FUS, hnRNPA1, HspB8, RBM14 and TAF15, whereas TDP43 and UBQ2 formed spatially distinct condensates regardless of their partner. Further investigation revealed the short and conserved -helical segment of TDP43s IDR was governing its specific self-recognition. Our findings imply that IDRs can tune their phase separation propensity through sequence composition, while their formation of discrete condensates likely requires additional cellular or structural determinants. O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=200 SRC="FIGDIR/small/547715v1_ufig1.gif" ALT="Figure 1"> View larger version (60K): org.highwire.dtl.DTLVardef@1bc6a1forg.highwire.dtl.DTLVardef@db4325org.highwire.dtl.DTLVardef@1154717org.highwire.dtl.DTLVardef@1ee619e_HPS_FORMAT_FIGEXP M_FIG C_FIG

Published in Nature Communications (predicted rank #1) · training set

Matching journals

The top 2 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.