A framework for understanding the functions of biomolecular condensates across scales.

Biomolecular condensates are found throughout eukaryotic cells, including in the nucleus, in the cytoplasm and on membranes. They are also implicated in a wide range of cellular functions, organizing molecules that act in processes ranging from RNA metabolism to signalling to gene regulation. Early work in the field focused on identifying condensates and understanding how their physical properties and regulation arise from molecular constituents. Recent years have brought a focus on understanding condensate functions. Studies have revealed functions that span different length scales: from molecular (modulating the rates of chemical reactions) to mesoscale (organizing large structures within cells) to cellular (facilitating localization of cellular materials and homeostatic responses). In this Roadmap, we discuss representative examples of biochemical and cellular functions of biomolecular condensates from the recent literature and organize these functions into a series of non-exclusive classes across the different length scales. We conclude with a discussion of areas of current interest and challenges in the field, and thoughts about how progress may be made to further our understanding of the widespread roles of condensates in cell biology.

Nuclear Import Receptor Inhibits Phase Separation of FUS through Binding to Multiple Sites.

Liquid-liquid phase separation (LLPS) is believed to underlie formation of biomolecular condensates, cellular compartments that concentrate macromolecules without surrounding membranes. Physical mechanisms that control condensate formation/dissolution are poorly understood. The RNA-binding protein fused in sarcoma (FUS) undergoes LLPS in vitro and associates with condensates in cells. We show that the importin karyopherin-ß2/transportin-1 inhibits LLPS of FUS. This activity depends on tight binding of karyopherin-ß2 to the C-terminal proline-tyrosine nuclear localization signal (PY-NLS) of FUS. Nuclear magnetic resonance (NMR) analyses reveal weak interactions of karyopherin-ß2 with sequence elements and structural domains distributed throughout the entirety of FUS. Biochemical analyses demonstrate that most of these same regions also contribute to LLPS of FUS. The data lead to a model where high-affinity binding of karyopherin-ß2 to the FUS PY-NLS tethers the proteins together, allowing multiple, distributed weak intermolecular contacts to disrupt FUS self-association, blocking LLPS. Karyopherin-ß2 may act analogously to control condensates in diverse cellular contexts.

Compositional Control of Phase-Separated Cellular Bodies.

Cellular bodies such as P bodies and PML nuclear bodies (PML NBs) appear to be phase-separated liquids organized by multivalent interactions among proteins and RNA molecules. Although many components of various cellular bodies are known, general principles that define body composition are lacking. We modeled cellular bodies using several engineered multivalent proteins and RNA. In vitro and in cells, these scaffold molecules form phase-separated liquids that concentrate low valency client proteins. Clients partition differently depending on the ratio of scaffolds, with a sharp switch across the phase diagram diagonal. Composition can switch rapidly through changes in scaffold concentration or valency. Natural PML NBs and P bodies show analogous partitioning behavior, suggesting how their compositions could be controlled by levels of PML SUMOylation or cellular mRNA concentration, respectively. The data suggest a conceptual framework for considering the composition and control thereof of cellular bodies assembled through heterotypic multivalent interactions.

Conserved interdomain linker promotes phase separation of the multivalent adaptor protein Nck.

The organization of membranes, the cytosol, and the nucleus of eukaryotic cells can be controlled through phase separation of lipids, proteins, and nucleic acids. Collective interactions of multivalent molecules mediated by modular binding domains can induce gelation and phase separation in several cytosolic and membrane-associated systems. The adaptor protein Nck has three SRC-homology 3 (SH3) domains that bind multiple proline-rich segments in the actin regulatory protein neuronal Wiskott-Aldrich syndrome protein (N-WASP) and an SH2 domain that binds to multiple phosphotyrosine sites in the adhesion protein nephrin, leading to phase separation. Here, we show that the 50-residue linker between the first two SH3 domains of Nck enhances phase separation of Nck/N-WASP/nephrin assemblies. Two linear motifs within this element, as well as its overall positively charged character, are important for this effect. The linker increases the driving force for self-assembly of Nck, likely through weak interactions with the second SH3 domain, and this effect appears to promote phase separation. The linker sequence is highly conserved, suggesting that the sequence determinants of the driving forces for phase separation may be generally important to Nck functions. Our studies demonstrate that linker regions between modular domains can contribute to the driving forces for self-assembly and phase separation of multivalent proteins.

Evidence of Kinetic Cooperativity in Dimeric Ketopantoate Reductase from Staphylococcus aureus.

Ketopantoate reductase (KPR) catalyzes the NADPH-dependent production of pantoate, an essential precursor in the biosynthesis of coenzyme A. Previous structural studies have been limited to Escherichia coli KPR, a monomeric enzyme that follows a sequential ordered mechanism. Here we report the crystal structure of the Staphylococcus aureus enzyme at 1.8 Å resolution, the first description of a dimeric KPR. Using sedimentation velocity analysis, we show that the S. aureus KPR dimer is stable in solution. In fact, our structural analysis shows that the dimeric assembly we identify is present in the majority of KPR crystal structures. Steady state analysis of S. aureus KPR reveals strong positive cooperativity with respect to NADPH (Hill coefficient of 2.5). In contrast, high concentrations of the substrate ketopantoate (KP) inhibit the activity of the enzyme. These observations are consistent with a random addition mechanism in which the initial binding of NADPH is the kinetically preferred path. In fact, Förster resonance energy transfer studies of the equilibrium binding of NADPH show only a small degree of cooperativity between subunits (Hill coefficient of 1.3). Thus, the apparently strong cooperativity observed in substrate saturation curves is due to a kinetic process that favors NADPH binding first. This interpretation is consistent with our analysis of the A181L substitution, which increases the Km of ketopantoate 844-fold, without affecting kcat. The crystal structure of KPRA181L shows that the substitution displaces Ser239, which is known to be important for the binding affinity of KP. The decrease in KP affinity would enhance the already kinetically preferred NADPH binding path, making the random mechanism appear to be sequentially ordered and reducing the kinetic cooperativity. Consistent with this interpretation, the NADPH saturation curve for KPRA181L is hyperbolic.

Human UDP-α-D-xylose synthase and Escherichia coli ArnA conserve a conformational shunt that controls whether xylose or 4-keto-xylose is produced.

Human UDP-α-D-xylose synthase (hUXS) is a member of the short-chain dehydrogenase/reductase family of nucleotide-sugar modifying enzymes. hUXS contains a bound NAD(+) cofactor that it recycles by first oxidizing UDP-α-D-glucuronic acid (UGA), and then reducing the UDP-α-D-4-keto-xylose (UX4O) to produce UDP-α-D-xylose (UDX). Despite the observation that purified hUXS contains a bound cofactor, it has been reported that exogenous NAD(+) will stimulate enzyme activity. Here we show that a small fraction of hUXS releases the NADH and UX4O intermediates as products during turnover. The resulting apoenzyme can be rescued by exogenous NAD(+), explaining the apparent stimulatory effect of added cofactor. The slow release of NADH and UX4O as side products by hUXS is reminiscent of the Escherichia coli UGA decarboxylase (ArnA), a related enzyme that produces NADH and UX4O as products. We report that ArnA can rebind NADH and UX4O to slowly make UDX. This means that both enzymes share the same catalytic machinery, but differ in the preferred final product. We present a bifurcated rate equation that explains how the substrate is shunted to the distinct final products. Using a new crystal structure of hUXS, we identify the structural elements of the shunt and propose that the local unfolding of the active site directs reactants toward the preferred products. Finally, we present evidence that the release of NADH and UX4O involves a cooperative conformational change that is conserved in both enzymes.