A Modular Design for Synthetic Membraneless Organelles Enables Compositional and Functional Control

Abstract

Living cells organize a wide variety of processes through compartmentalization into membraneless organelles, known as biomolecular condensates. Given their ubiquitous presence across a wide spectrum of different organisms and cell types, biomolecular condensates are increasingly considered to offer great potential for biotechnological applications. However, native condensates remain difficult to harness for engineering applications, both due to their intertwined mechanisms of assembly and compositional control, and potential disruptions to native cellular processes. Here, we demonstrate a modular framework for the formation of synthetic condensates designed to decouple cluster formation and protein recruitment. Synthetic condensates are built through constitutive oligomerization of intrinsically-disordered regions (IDRs), which drive the formation of condensates whose composition can be independently defined through fused interaction domains. The composition of the proteins driven to partition into the condensate can be quantitatively described using a binding equilibrium model, demonstrating predictive control of how component expression levels and interaction affinity determine the degree of protein recruitment. Finally, the engineered system is utilized to regulate protein interactions and metabolic flux by harnessing the system’s compositional tunability.

Emergence of Multiphase Condensates from a Limited Set of Chemical Building Blocks

“…In conclusion, our theory and inverse design approach advance our understanding of multiphase condensate formation in biology as well as our ability to engineer chemically diverse artificial condensates. Our predictions could be tested using existing experimental technologies in both biological − and synthetic − systems. Our theory could also be applied to design “patchy” colloidal particles, in which case the feature matrix would indicate the surface area covered by a sticker type and the feature–feature interactions would represent the interactions between different types of stickers.…”

“…Finally, we perform molecular dynamics simulations of phase coexistence using these optimized heteropolymer mixtures to validate our theoretical predictions and the inverse design approach. Our results suggest an extensible and generalizable framework for exploring molecular “grammars” in multicomponent biomolecular systems − as well as a practical strategy for engineering complex artificial condensates using experimentally realizable molecules. − …”

Hybrid Physics-Informed Metabolic Cybergenetics: Process Rates Augmented with Machine-Learning Surrogates Informed by Flux Balance Analysis

“…At the posttranslational level, metabolic modulation targets enzymes after protein synthesis, often constitutively expressed. For instance, modifying the physical clustering of constitutively expressed enzymes can impact metabolic fluxes . However, constitutive expression of enzymes can impose a metabolic burden on cells …”