Research Projects

A fundamental challenge in cell biology is to understand how biochemical reactions are organized and precisely regulated in space and time in the densely packed cellular microenvironment. One way cells achieve this specificity is by compartmentalization of the cellular space. Besides the classical membrane bound organelles, there exists a repertoire of membraneless organelles (MLOs) such as nucleolus, germ granules which are now considered to be biomolecular condensates. Condensates may form through the physical process of liquid-liquid phase separation (LLPS) and are characterized by the ability to concentrate proteins and nucleic acids. LLPS occurs when an initially homogeneous solution spontaneously demixes into two distinct phases above a critical concentration of the scaffold molecule. LLPS is primarily driven by weak, multivalent interactions between proteins and/or nucleic acids. While there has been substantial studies focusing on 3D LLPS in bulk solution, only a handful of instances exist for membrane associated 2D phase separation. Though thermodynamically governed by the same principles as 3D LLPS, membrane association can lower the critical concentration of phase separation by an order of magnitude. By localizing a scaffold protein to a membrane surface, cells take the advantage of site-specifically driving condensate assembly at much lower critical concentration of the scaffold protein. In the fruit fly germline, oskar mRNA is locally translated at the posterior of the developing oocyte and fertilized embryo into two isoforms, Long and Short Oskar, which differ by 138 amino acids present exclusively at the N-terminus of the long isoform. The short isoform is essential for germ granule assembly, while the long isoform, by yet unknown mechanisms, nucleates the posterior F-actin network, which in turn anchors the germ granules. Therefore, the two classes of Oskar protein condensates provide an excellent and rather unique paradigm to explore how functional specificity can be achieved in biological assemblies by differential subcellular targeting. Our preliminary observations indicate that Long Oskar forms condensates on endomembrane surfaces, such as endosomes and yolk vesicles (YV) distinct form the Short Oskar assemblies. In this project we aim to characterize the membrane associated Long Oskar assemblies and investigate the molecular mechanism behind membrane targeting and phase separation of the long isoform. Employing a combination of in vitro studies with purified protein and reconstitution on model membrane (liposomes, GUVs), we aim to identify the biophysical principles that direct 2D LLPS on membrane surfaces. The identified mechanism would be tested in vivo by generating transgenic fly lines to study the importance of membrane binding-coupled phase separation in Drosophila germline development.