A stable protein-based condensate and method of its making and use

Abstract

A system for the preparation of a synthetic and protein-based condensate. Methods for the dynamic stabilization of the condensates using engineered proteins and methods of using the same are also disclosed.

Description

Atty Docket TUOE-P2010WQ-00676481A STABLE PROTEIN-BASED CONDENSATE AND METHOD OF ITS MAKING AND USECROSS REFERENCE TO RELATED APPLICATIONS

[0001] This application claims priority to US Provisional Application 63 / 742,254 filed on January 6, 2025, the content of which is incorporated herein by reference in its entirety for all purposes.FIELD OF THE INVENTION

[0002] The present disclosure relates to systems and methods for the preparation of synthetic and protein-based condensates. More particularly, it relates to systems and methods for dynamic stabilization of such condensates using engineered proteins.BACKGROUND

[0003] Membraneless organelles (MLOs), cellular structures without a surrounding lipid membrane, are formed through liquid-liquid phase separation (LLPS), where multivalent interactions between proteins and / or nucleic acids drive their formation.1,2MLOs play a crucial role in regulating cellular processes, including the spatial organization and compartmentalization of biomolecules, dynamic responses to environmental stimuli, and the concentration and sequestration of specific molecules.1,3,4

[0004] Complex condensates (or coacervates) have shown great potential as model systems to mimic and understand the properties of MLOs. Biomolecular condensates are one type of are cellular structures that play a crucial role in regulating cellular processes, including the spatial organization and compartmentalization of biomolecules, dynamic responses to environmental stimuli, and the concentration and sequestration of specific molecules. As a result, condensate droplets have emulsion-like characteristics and are enriched in proteins or other phase separating molecular matter. Their size ranges from nano-to-micrometers.

[0005] Condensates are found inside eukaryotic and prokaryotic cells but can also be reconstituted chemically (depending on the origin of the molecular matter, either referred to as condensates, coacervates, or liquid-liquid phase-separated systems). The shape and size of condensates give them an emulsion-like characteristic, similar to what is observed in oil-in-water emulsions. However, in the case of condensates, both phases are aqueous.

[0006] Chemically reconstituted condensates (or synthetic condensates) are characterized by long-term instability. They tend to aggregate, fuse, disassemble, and adhere to surfaces. ThisAtty Docket TUOE-P2010WQ-00676481general instability limits their usage in applications that require long-term stability, such as in food, pharma, paint, cosmetics, and biotech formulations.

[0007] It has been reported that condensates can be stabilized by providing them with a synthetic membrane. Reported examples include membranes via lipids, polymers, surfactants, protein-polymer conjugates, and inorganic nanoparticles. However, while functionalizing condensates with such synthetic membranes can enhance their stability, it also introduces significant challenges. The introduction of these membranes often restricts the natural dynamic behavior of the condensates, particularly their ability to rapidly and reversibly exchange molecules with their surrounding environment. This altered dynamicity limits their functional resemblance to biological systems. Additionally, the use of synthetic materials could raise concerns about biocompatibility and safety.SUMMARY

[0008] The present disclosure provides a composition and methodology to solve the problems outlined above. In one embodiment, a technology is disclosed which helps stabilize condensates via protein-based material only that preserves the dynamic nature (exchange between condensate and dilute phase) characteristic of natural condensates. Also, the technology enables long-term storage / stability of the condensates. Additionally, the disclosed procedure does not require any heating steps and can preserve the functional structure of the proteins which form the condensate or are embedded within the condensate.

[0009] In one embodiment, the disclosed condensate-stabilizing proteins are molecularly engineered to feature an anchoring group and a buoyancy group. In one aspect, these two functional groups can be molecularly adapted to align physically and chemically with the internal molecular characteristics of the condensates and provide modularity to the system. In another aspect, the condensate-stabilizing proteins work by forming a dynamic protein layer on the surface of the condensate.

[0010] In some embodiments, various phase-separating macromolecules with distinct chemical properties can be used to produce condensates with a wide range of colloidal properties.

[0011] In some embodiments, modulating the size and composition of the condensates allows for optimizing the condensate / emulsion properties.

[0012] In some embodiments, a dynamic protein-based layer is used to provide stability for the condensates against droplet fusion and dissolution, thereby enabling long-term condensate storage.Atty Docket TUOE-P2010WQ-00676481BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 shows the method for the production of protein-stabilized condensates. A) Two oppositely charged proteins / polypeptides are mixed at 1500rpm for 5 minutes in the presence of a condensate stabilizing protein. After mixing, stable condensates are formed. B) Confocal brightfield micrograph illustrating the emulsion-like characteristics of the protein-stabilized condensates. Scalebar is 30pm.

[0014] Figure 2 shows the mechanism of stability involving electrostatic docking and buoyancy repulsion. A) The stabilizer protein consists of two domains: The Tau anchor domain, with which it interacts with negative condensate interfaces, and the GFP buoyancy group, which provides buoyancy against internalization of the stabilizer protein into the condensate. B) Cryo-TEM microscopy demonstrating a protein monolayer at the interface of the condensate. Scale bar is 200nm. C) A zoom of the area marked by the yellow box in panel B. Scale bar is lOnm. D-E) Brightfield images of negatively or positively charged condensates in the presence of stabilizing protein. The positively charged tau domain is electrostatically attracted to negative condensates, resulting in stable droplets (panel D). As the positive charge of the condensates increases, electrostatic repulsion towards the tau domain also increases, leading to reduced condensate stability (panel E). F) Unstable condensates are observed in the presence of non-connected anchor and buoyancy domains. All scalebars are 30pm.

[0015] Figure 3 shows the size of the condensates can be altered. A-F) Condensates formed in the presence of stabilizing protein and varying concentrations of phase -separating molecules. The total charge concentrations used for images A-F are respectively: 0.5mM, ImM, 2mM, 4mM, 6mM, 8mM. G) Quantification of stabilized condensate areas at various condensate formulations.

[0016] Figure 4 shows a variety of condensates that have been stabilized using a subset of condensate stabilizing proteins. A) Various charged polyelectrolytes, such as polypeptides, disordered proteins, or folded proteins can form condensates and many of these combinations have been successfully stabilized using condensate stabilizing proteins. Green and red boxes indicate stabilized and non-stabilized conditions, respectively. Grey boxes indicate conditions that have not been tested. B-D) Examples of conditions in panel A that have been successfully stabilized by condensate stabilizing proteins. E) Condensate stabilizing proteins with various biochemical properties can be engineered, in this example by substituting the moderately dimerizing GFP buoyancy group with the strongly dimerizing GST protein to produce GST-Tau, or with the non-dimerizing muGFP protein to produce muGFP-Tau. F-H) Brightfield microscopy images of poly-(aspartic acid)3o / poly-(lysine)ioo condensates formed in theAtty Docket TUOE-P2010WQ-00676481presence of GST-Tau (F), GFP-Tau (G), or muGFP-Tau (H). Based on the condensate circularity and wetting behavior, condensates are classified as stable (F), semi-stable (G), or unstable (H). Semi-stable addresses an intermediate category that describes formulations in which a mixture of both stable and unstable condensates is observed. All scalebars are 50pm. I-K) Overview of the stabilizing effect of GST-Tau (I), GFP-Tau (J), or muGFP-Tau (K) upon addition to a library of 18 phase -separating polypeptide combinations. Green, yellow, and red, respectively, indicate stable, semi-stable, and unstable formulations.

[0017] Figure 5 shows phase separating behavior of various commercially available polyelectrolytes, as determined by brightfield microscopy. A) The screening process involved using 8mM of charged monomers and five different positive-to-negative charge ratios (3:7, 4:7, 5:5, 6:4, 7:3). Green color indicates phase-separated combinations under at least one experimental condition. In contrast, yellow indicates combinations that did not phase separate under any of the conditions. Red indicates incompatible polyelectrolyte combinations that were found to form solid aggregate structures. Illustrative brightfield images of the polyelectrolyte combinations marked in A) are shown in B-I). Images B-C are representative images that were classified as phase separation. Images D-H represent different types of aggregation behaviors. Scalebars are 50pm.

[0018] Figure 6 shows protein-based coacervate stabilization. A) Schematic overview of the formulation of protein-stabilized coacervates by the addition of a GFP-Tau fusion protein. B) Confocal micrograph displaying the localization of the stabilizing protein (green) outside of the coacervate environment. C) Normalized fluorescence intensity profiles of the dotted lines in B), demonstrating that there is no accumulation of stabilizing protein at the interface of the coacervates. D-F) Brightfield images of coacervates formed in the presence of varying concentrations of stabilizing protein, demonstrating increased coacervate stability at increasing GFP-Tau concentrations. G) Coacervate circularity at different concentrations of protein stabilizer. Each datapoint is presented as mean ± sd (n>50) and the exponential fit serves to aid the eye. All scalebars are 30pm.

[0019] Figure 7 shows the mechanism of stability involving electrostatic docking and buoyancy repulsion. A) Illustration of the stabilizing Tau-GFP protein and its orientation on the coacervate interface. A TEV protease recognition site is included between the Tau and GFP domains, such that the protein can be cleaved upon the addition of TEV protease. The inset shows the expected orientation of the stabilizing protein on the interface, with the positively charged tau domain pointing towards the negatively charged coacervate interior. In contrast, the GFP protein provides buoyancy against the internalization of the protein and is orientedAtty Docket TUOE-P2010WQ-00676481towards the di-lute phase. B-D) Brightfield images of negative-, neutral-, or positively charged coacervates made up of poly-(lysine)100 and poly-(aspartic acid)250 in the presence of IpM stabilizing protein. The positively charged tau domain is electrostatically attracted to negative coacervates, resulting in stable droplets. As the positive charge of the coacervates increases, electrostatic repulsion towards the tau domain also increases, leading to reduced coacervate stability. E-G) Brightfield images demonstrating the necessity of employing the fusion protein GFP-Tau on the stability of poly-(lysine)100 / poly-(aspartic acid)250 coacervates. Unstable coacervates are formed in the presence of Tau alone (E), GFP alone (F), or a combination of both individual domains (G). All scalebars are 30pm.

[0020] Fig. 8 shows the effect of protein dimerization on condensate stability. A) The role of the buoyancy group’s dimerization is studied by substituting the moderately dimerizing GFP protein domain (PDB: 5NHN) of GFP-Tau with the strongly dimerizing GST protein (PDB: 1PKW) to produce GST-Tau, or with the non-dimerizing muGFP protein (PDB: 5JZL) to produce muGFP-Tau. B-D) Brightfield microscopy images of poly-(aspartic acid)3o / poly-(lysine)ioo coacervates formed in the presence of GST-Tau (B), GFP-Tau (C), or muGFP-Tau (D). Based on the coacervate circularity and wetting behavior, coacervates are classified as stable (B), semi-stable (C), or unstable (D). Semi-stable addresses an intermediate category that describes formulations in which a mixture of both stable and unstable coacervates is observed. All scalebars are 50pm. E-G) Overview of the stabilizing effect of GST-Tau (E), GFP-Tau (F), or muGFP-Tau (G) upon addition to a library of 18 phase-separating polypeptide combinations. Green, yellow, and red, respectively, indicate stable, semi-stable, and unstable formulations.

[0021] Figure 9 shows the effect of protein charge on coacervate stability. A) GFP-Tau effectively stabilizes negative coacervates, hypothesized to result from balanced attractive and repulsive forces between the stabilizer and the coacervate; Tau is electrostatically attracted to the coacervate, while the GFP domain provides adequate buoyancy due to its slightly negative charge (-7) and folded nature. B) Substitution of GFP (-7) into a positive variant (+15) reduces the buoyancy character of the GFP moiety, thereby disrupting its specific interaction with the coacervate surface. C) Substitution of GFP (-7) into a more negative variant (-25) potentially disrupts the protein dimerization due to repulsion as well as increases the electrostatic repulsion towards the negative coacervates. D-E) The stabilizing effect of +15GFP-Tau (D) and -25GFP-Tau (E) upon addition to a library of 18 phase-separating polypeptide combinations. Using the same metrics as in Figure 8, Green, yellow, and red indicate stable, semi-stable, and unstable formulations. The presented structures for +15GFP (D) and -25GFP (E) are AlphaFoldAtty Docket TUOE-P2010WQ-00676481prediction structures, where positive amino acids are represented as blue spheres, negative amino acids as red spheres, and the protein structure as grey cartoon.

[0022] Fig. 10 shows Cryo-TEM imaging of the coacervate membrane topology. The images depict poly-(lysine)100 / poly-(aspartic acid)250 coacervates and their interface in the absence (B-C) or presence of stabilizer proteins GST-Tau (D-E), GFP-Tau (F-G), mEOS3.2-GST-Tau (H-I). The lower panel of images provides a closer view of the areas marked by the yellow boxes in the upper panel. The scale bar in the upper panel indicates 200nm, while the lower panel indicates lOnm. I) From the Cryo-TEM images in A-H, the coacervate membrane width is determined. Non-stabilized coacervates had no observable membrane width, whereas stabilized samples showed a membrane thickness corresponding to a protein monolayer.

[0023] Figure 11 shows dynamicity of the stabilizing proteins at the coacervate interface. A) The stabilizing proteins can interact with the coacervate interface in three ways: through diffusion, docking, and undocking. To study this behavior, photoconvertible fluorophore-labeled proteins (mEOS3.2-GST-Tau) are illuminated with UV light, causing them to change color from green to red. This allows for the imaging of bulk proteins using 488nm light and single molecules overtime using 405nm and 561nm lasers. B-C) Micrographs displaying the stabilizing proteins at the coacervate interface. B) shows the imaging of bulk mEOS3.2-GST-Tau, while C) shows a sptPALM reconstruction of a continuous acquisition of individual proteins. Scalebars are 10pm. D-F) Diffusion of stabilizing proteins at the coacervate interface that undergo a green-to-red shift, enabling protein tracking over time, as illustrated by the schematic D. E) shows four representative tracks of individual proteins that exhibited X / Y movement over the interface. An image reconstruction is shown in F. Scalebar is 2pm. G-L) Confirmation of the docking of stabilizing proteins at the interface, as indicated by the schematic G. H) shows the experimental set-up: GST-Tau stabilized coacervates were prepared in the presence of 500nM GFP-Tau. After imaging at the green (I) and red (J) channel, lOOnM mEOS3.2-GST-Tau was added to the coacervates, whereafter they were again imaged at the green (K) and red (L) channel, showing the exchange of mEOS3.2-GST-Tau. Panel J and L are sptPALM reconstructions of a continuous acquisition of individual proteins. Scalebars are 1pm.DETAILED DESCRIPTION

[0024] In cells, condensates formed through a liquid-liquid phase separation process where specific proteins and / or RNA molecules, potentially with special domains, interact with each other via weak, multivalent bonds, causing them to cluster together and separate from theAtty Docket TUOE-P2010WQ-00676481surrounding liquid, forming a distinct liquid-like droplet, known as a condensate. This disclosure provides a condensate comprising a first molecule and a second molecule, wherein the first molecule and a second molecule are oppositely charged. The term “synthetic” means the condensate is created artificially by molecular engineering, but the condensate can be composed of naturally occurring biomolecules or compounds that does not exist in nature.

[0025] The disclosure will now be illustrated with working examples, which are intended to illustrate the working of disclosure and not intended to restrict the scope of the present disclosure. Unless otherwise defined in this disclosure, all technical and scientific terms used herein have the same meaning as commonly understood to one of ordinary skill in the art to which this disclosure belongs. Although methods and materials similar or equivalent to those described herein can be used in the practice of the disclosed methods and compositions, the exemplary methods, devices and materials are described herein.EXAMPLES

[0026] Example 1

[0027] A new method has been developed to reconstitute stable condensates in vitro (Figure 1A). The method involves the mixing of two oppositely charged phase separating molecules in the presence of a condensate-stabilizing protein. After five minutes of mixing stable condensates were formed with adaptable sizes in the nano-to-micrometer range (Figure IB).

[0028] The mechanism of stability involves interactions between the stabilizing protein and the interface of the condensates (Figure 2A). These interactions result in the formation of a protein monolayer at the condensate interface (Figure 2B-C). These interactions prevent the droplets from dissolution and coalescence, likely by reducing the droplet surface tension. Cryo-TEM microscopy results indicate that the stabilizing proteins are tightly packed at the interface of the condensates in a monolayer fashion. These binding events are expected to reduce the surface tension of the condensate, thereby effectively stabilizing them against dissolution and coalescence. In this example, the stabilizer consisted of two protein domains: A positively charged protein domain derived from the protein tau (the anchor domain), and a highly folded, slightly negatively charged green fluorescent protein domain (the buoyancy group). The positive charge of the anchor domain facilitates an electrostatic attraction of the stabilizer towards negatively charged condensates (Figure 2D), but not towards positively charged condensates (Figure 2E). The buoyancy group provides electrostatic and steric repulsion against internalization of the entire stabilizer into the condensate. When the stabilizer was split into its two separate domains, the individual anchor and buoyancy group were no longer ableAtty Docket TUOE-P2010WQ-00676481to stabilize the condensates (Figure 2F), demonstrating that the combination of both elements in the stabilizing agent is necessary. The physical connection between the anchor and the buoyancy group is thus crucial to balance the electrostatic attraction between the anchor and the condensate and the repulsion of the folded buoyancy group.

[0029] Example 2

[0030] The size of the condensates can be altered to optimize the emulsion-like properties of the condensates (Figure 3A-G). This can be achieved by altering the concentrations of the phase separating molecules.

[0031] Many molecules are capable of phase-separating and can form condensates. Due to varying chemical properties of these molecules, a wide range of condensate types, varying in dynamicity, stiffness, size, charge density, hydrophobicity, and internal pH, can be produced by mixing different types of molecules. A toolbox of different condensate stabilizers has been developed to stabilize different types of condensates. With this toolbox, a wide variety of condensates can be stabilized, varying from charged polypeptides to disordered proteins to folded protein structures. An overview of the current protein combinations that have been successfully stabilized by one or more of these in-house stabilizing proteins is presented in Figure 4. These results indicate the broad range of condensate types and condensate stabilizers that can be mixed and matched to form stable condensates.

[0032] Example 3 Engineering of a coacervate stabilizing protein

[0033] The experiments in this Example were aimed to replicate the stability and endurance of metastable membraneless organelles (MLOs). To mimic their protein environment, several polypeptides were screened for their phase -separating behavior and stability. Mainly, polypeptides that have been commonly used to study the mechanisms of phase separation were tested.37 40Similar to the findings of others, many polypeptides could phase separate at a wide range of charge ratios and polypeptide lengths (Figure 5). However, not a single combination yielded stable coacervates that could replicate the endurance seen in metastable MLOs. These initial findings suggested the need for a third bio-mimetic component that stabilizes the condensates. To replicate the protein environment in the cell, protein engineering was used to develop novel protein-based coacervate stabilizers.

[0034] The IDP Tau was used as a starting point for engineering coacervate stabilizers. This protein has previously been shown to be involved in stabilizing cellular structures.41Specifically, the microtubule-binding region of Tau contains an excess of positive charges that can interact with acidic tubulin to stabilize the microtubules in neurons.42The highly positive charge of this Tau domain could potentially be electrostatically attracted towards negativelyAtty Docket TUOE-P2010WQ-00676481charged coacervates, functioning as an anchor to engineer a surface -active coacervate stabilizing protein. To prevent the protein from being internalized into the coacervate core, a highly soluble, slightly negative, green fluorescent protein (GFP) was fused to the N-terminus of the Tau domain (Figure 6A). This provides steric resistance and electrostatic repulsion against internalization. The fluorescent nature of the protein was used as a bonus for visualization purposes.

[0035] The resulting, overall positively charged, engineered fusion protein (GFP-Tau) was tested for its ability to stabilize conventional negatively charged poly-(lysine)100 / poly-(aspartic acid)250 complex coacervates. The GFP-Tau protein was added before the onset of phase separation, gratifyingly resulting in stable coacervate droplets (Figure 6A). Under physiological conditions (pH 7.4, lOOmM NaCl), 2.5pM of stabilizer protein made that the coacervates showed no coalescence and Ostwald ripening at room temperature for more than one week. Using confocal microscopy, we observed that GFP-Tau was present in the dilute phase (Figure 6B-C). The stabilizing effect of GFP-Tau did not occur at lower stabilizer concentrations (Figure 6D-F). This suggests that the stabilizing protein requires a minimum effective concentration.

[0036] As a measure of coacervate stability, we here introduce circularity as a quantifiable metric. By assessing the circularity at different concentrations of protein stabilizers, the minimum amount required for robust stabilization of coacervates can be determined. (Figure 6G). To accurately quantify the circularity, the interior of coacervates was visualized with Cy5-labelled poly-(lysine) 100, and confocal micrographs were made. A stability threshold (red line, Figure 6G) was determined based on the circularity of the stable coacervates at 2.5 pM, and more specifically, based on the average minus one standard deviation of the stable coacervates at 2.5pM . Circularity below the stability threshold was observed below IpM of stabilizing protein, which is consistent with the micrographs shown in Figure 6D-F. Together, these results indicate that a minimum concentration of stabilizer is required and that the GFP-Tau protein operates via a mechanism different from “classical” surfactants, which selectively occupy the interface between two phases.

[0037] Example 4 The mechanism of inducing coacervate stability

[0038] The mechanism of binding of the Tau domain at the coacervate interface was investigated by producing negative, neutral, and positively charged coacervates. These coacervates were formed by mixing different charge ratios of poly-(lysine)100 and poly-(aspartic acid)250. It was anticipated that the positively charged tau domain would only effect negatively charged coacervates because of matching electrostatic interactions (Figure 7A). InAtty Docket TUOE-P2010WQ-00676481line with these hypotheses, stable, spherical droplets for the coacervates produced with an excess negative charge were observed (Figure 7B). Due to non-matching electrostatic interactions, increasing instability and wetting were observed for neutral and positive coacervate formulations (Figure 7C-D). Complementary, a fusion protein was produced which contained a negatively charged disordered region derived from the alpha-synuclein protein and a positively engineered GFP protein as a buoyancy domain. This protein was capable of specifically stabilizing positively charged coacervates. Together, these results suggest that molecular recognition and coacervate stabilization occur via complementary electrostatic interactions between the intrinsically disordered anchoring region and the charged coacervate interface.

[0039] A Tobacco Etch Virus (TEV) cleavage site43was introduced between the IDP and GFP domains. After TEV protease cleavage, Tau and GFP were separately purified from the full-length construct. Neither Tau nor GFP nor their mixture was capable of stabilizing the coacervates (Figure 7E-G). The physical connection between GFP and Tau in GFP-Tau is thus crucial to balance the electrostatic attraction between the Tau sequence and the coacervates and the repulsion of the folded GFP buoyancy group.

[0040] The effective buoyancy behavior of the relatively small GFP protein was surprisingly efficient. One possible explanation could be the fact that GFP can dimerize, and we therefore wanted to investigate if this feature was critical for effective stabilization. A set of proteins was designed to evaluate the contribution of the buoyancy group dimerization (Figure 8A). The strong dimerizing Glutathione S-Transferase (GST)44protein was used to yield GST-Tau, and the non-dimerizing monomeric ultra-stable GFP (muGFP)45was the basis for muGFP-Tau. All three buoyancy proteins are comparable in charge, such that primarily their dimerization character is varied among the stabilizing proteins.

[0041] Coacervates formed by mixing an excess of poly-(aspartic acid)30 with poly-(lysine)100 were studied in the presence of the three fusion proteins (Figures 8B-D). The coacervates' circularity and wetting behavior criteria revealed that GST-Tau stabilizes this specific poly-electrolyte composition most effectively. GFP-Tau stabilizes to a lesser extent and, strikingly, muGFP-Tau cannot stabilize this poly-electrolyte combination. The latter is potentially due to muGFP-Tau's inability to pack the stabilizing proteins tightly on the coacervate interface.

[0042] The three fusion proteins were screened against an 18-membered in-house library of phase-separating polypeptide combinations to obtain a more concise overview of the relationship between buoyancy group dimerization and stabilization of complex coacervates.Atty Docket TUOE-P2010WQ-00676481Based on their wetting behavior and circularity, the resulting coacervates were classified as stable, semi-stable, and unstable. To clarify, “semi-stable” describes an intermediate category of formulations in which a mixture of stable and unstable coacervates is observed. A summary of the stability classifications for all three fusion proteins against the polypeptide library is presented in Figures 8E-G. Interestingly, a clear trend between buoyancy dimerization and coacervate stability was observed for the whole complex coacervate library. Almost no complex coacervate compositions could be stabilized by the non-dimerizing muGFP-Tau protein. In contrast, the dimerizing proteins GFP-Tau and GST-Tau stabilized numerous complex coacervates. These findings indicate a favorable effect of buoyancy dimerization on coacervate stability. It should be noted that the dimerization affinity of GFP is in the high micromolar range46, whereas GST forms dimers in the low nanomolar range.44Since a stabilizer concentration of 2.5pM was used for the screening, we expect that GST-Tau predominantly exists as a dimeric protein in both the bulk phase and at the coacervate interface. In contrast, GFP-Tau is unlikely to form dimers in the dilute phase under our experimental conditions but potentially dimerizes at the surface of the coacervate. As such, the strongly dimerizing GST protein does not necessarily outperform the more weakly dimerizing GFP protein. Together, these results suggest that a dimerizing buoyancy group is highly beneficial in engineering a protein-based coacervate stabilizer.

[0043] Next the effect of the overall charge of the buoyancy group on coacervate stabilization was investigated (Figure 9A-C). For this experiment, the buoyancy group of GFP-Tau with an overall charge of -7 (-7GFP-Tau) was changed both into a positively charged variant (+15GFP-Tau)47 and a more negatively charged variant (-25GFP-Tau)48. The stabilizing effects of these fusion proteins were tested against our 18-membered in-house library of polyelectrolytes. All formulations were classified using the same metrics as described in Figure 3 (stable, semistable, and unstable). Contrary to previous results, in which GFP-Tau could stabilize many formulations, +15GFP-Tau and -25GFP-Tau could not stabilize most coacervate formulations (Figure 9D-E). Most likely, a strong positive charge of the buoyancy group causes this protein to be attracted to the coacervate core (Figure 9B) . On the contrary, a buoyancy group with a highly negative charge may lead to electrostatic repulsion; this potentially lowers the dimeric character, as well as providing an energy barrier for the stabilization of the negatively charged coacervate core (Figure 9C). For both proteins, we did find a few stabilized formulations, specifically for coacervate combinations that consisted of longer polypeptides. Due to multivalent charge interactions between the polypeptides, these combinations might be inherentlyAtty Docket TUOE-P2010WQ-00676481more easily stabilized by suboptimal protein stabilizers, similar to what was observed for the non-dimerizing muGFP-Tau.

[0044] Example 5 Topography of the membrane

[0045] Cryo-Transmission Electron Microscopy (Cryo-TEM) studies were performed to explore the topography of the membrane on the coacervates (Figure 10). The cryogenic conditions preserve the vitrified frozen-hydrated state of the coacervates. The surface of nonstabilized coacervates was compared with the interface of coacervates that were produced in the presence of stabilizing proteins with different hydrodynamic volumes

[0046] The coacervates formed without stabilizing protein revealed no well-defined membrane (Figure 10A-B). In stark contrast, addition of either GST-Tau or GFP-Tau both resulted in a clear dark membrane, observable at the periphery of the coacervates (Figure 10C-F). The membrane widths for GST-Tau and GFP-Tau were 3.4±0.7nm and 4.6±0.8nm, respectively, as determined using image analysis (Figure 101). These findings suggest a protein monolayer since GFP and GST are about 4nm in diameter. As an additional control, a triple-domain fusion protein was examined. This fusion protein contains be-sides Tau both a GST domain and an N-terminal GFP -like barrel-shaped mEOS3.2 (mEOS3.2-GST-Tau). This triple-domain protein formed a thicker monolayer membrane of 5.7±0.8nm as a consequence of its increase in size.

[0047] Example 6 Dynamics of stabilizing proteins at the coacervate interface

[0048] Due to the surface-active design of the stabilizing proteins, an accumulation of stabilizer proteins was expected at the interface between the condensed and polymer dilute phase. Although the Cryo-TEM analysis corroborated this concept, confocal microscopy did not provide direct evidence of an apparent accumulation of stabilizing proteins on the coacervate interface (Figure 6B). Single -particle tracking photoactivated localization microscopy (sptPALM) experiments using total internal reflection fluorescence (TIRF) were therefore carried out to study the accumulation and dynamics of the protein-based stabilizers at the coacervate interface. The sptPALM technique has emerged as a powerful tool for high-resolution imaging. The photo-switchable mEOS3.2 protein has a bright green (516nm) fluorescence and can stochastically switch to red fluorescence (58 Inm) upon UV irradiation of ~390nm.50 This UV photo-switching enables the imaging of individual proteins in addition to the proteins in bulk. Specifically, here, sptPALM enables the selective tracking of motion-restricted individual proteins at the condensate interface, reducing background signal from freely diffusing proteins and providing insights into system dynamics.Atty Docket TUOE-P2010WQ-00676481

[0049] The triple-domain fusion protein mEOS3.2-GST-Tau was optimally suited for sptPALM imaging as super-resolution construct (Figure 11A). mEOS3.2-GST-Tau was mixed with an excess of unlabeled GST-Tau and excited in the green channel to investigate the whole population of stabilizing proteins and in the red channel to allow single-molecule tracking. The excitation of fluorescent proteins in the green channel confirmed that the stabilizing proteins accumulate on the interface (Figure 11B). After re-construction of the excitation of single fluorescent proteins in the red channel, we were able to reduce the background signal significantly and could resolve a high-resolution membrane at the interface of the coacervates (Figure 11C).

[0050] The protein stabilizer system was observed to be highly dynamic, and docked proteins went out of focus due to z-directed motion within milliseconds. For dynamicity studies, we therefore only studied the proteins at the top of the coacervate, as these proteins have limited z-movement and are more likely to move in the X / Y directions. Individual proteins were traced, and several representative traces are shown in Figure 1 IE; a reconstruction is shown in Figure 11F.

[0051] These results confirm that the membrane is not static but that the stabilizer proteins dynamically move across the monolayer. Furthermore, for this particular experiment, the stabilizer proteins seem to form clusters at the coacervate membrane. This characteristic is also found in other forms of condensate stabilizers, such as Pickering agents.35

[0052] Finally, an exchange experiment demonstrated that the proteins not only show lateral dynamics and diffusivity, but can also effectively exchange between the interface and the bulk solution (Figure 11G). First, coacervates were prepared in the presence of both GST-Tau and GFP-Tau to allow for visualization. Upon inspection of these coacervates by exciting all stabilizing proteins in the green channel, a stabilizing layer of GFP-Tau could be observed (Figure 1 II). As a control, no signal was seen by exciting the proteins in the red channel (Figure 11 J). Subsequently, lOOnM of photoactivatable mEOS3.2- GST-Tau was added to the preformed coacervates. As before, a stabilizing layer of proteins was observed in the green channel. But within a few minutes, an additional fluorescent signal was observed in the red channel (Figure 11K-L). These findings demonstrate the dynamic exchange of proteins between the bulk and the interface.

[0053] To conclude, it is disclosed here that stabilizing proteins can interact with the interface of condensates, thereby providing stability against droplet disintegration and coalescence. By modulating the molecular characteristics of the protein stabilizers, highly effective and robust condensate stabilizers can be designed for various condensate formulations. In oneAtty Docket TUOE-P2010WQ-00676481embodiment, the stabilizer requires: 1) a charged disordered region to enable electrostatic attraction to the oppositely charged condensates and 2) a buoyancy group with balanced electrostatic and steric repulsion towards the condensates. The ability to stabilize condensates using an entirely protein-based strategy presents opportunities for their implementation as healthy and biocompatible colloids with potential applications as fat substitutes in food products, as colloid substitutes in paint and cosmetic formulations, and as model screening systems in pharmaceutical and biotechnology industry.Atty Docket TUOE-P2010WQ-00676481ReferencesAll refs listed below or cited throughout the disclosure are hereby incorporated by reference into this disclosure.1. Hirose, T., Ninomiya, K., Nakagawa, S. & Yamazaki, T. A guide to membraneless organelles and their various roles in gene regulation. Nat. Rev. Mol. Cell Biol. 24, 288-304 (2023).2. Pappu, R. V., Cohen, S. R., Dar, F., Farag, M. & Kar, M. Phase Transitions of Associative Biomacromolecules. Chem. Rev.123, 8945-8987 (2023).3. Gomes, E. & Shorter, J. The molecular language of membraneless organelles. / . Biol. Chem.294, 7115-7127 (2019).4. Wollny, D. et al. Characterization of RNA content in individual phase-separated coacervate microdroplets. Nat. Commun. 13, 2626 (2022).5. Banani, S. F., Lee, H. 0., Hyman, A. A. & Rosen, M. K. Biomolecular condensates: organizers of cellular biochemistry. Nat Rev. Mol. Cell Biol. 18, 285-298 (2017).6. Spruijt, E. Open questions on liquid-liquid phase separation. Commun. Chem. 6, 1-5 (2023).7. Donau, C. etal. Active coacervate droplets as a model for membraneless organelles and protocells. Nat. Commun. 11, 5167 (2020).8. Yewdall, N. A., Andre, A. A. M., Lu, T. & Spruijt, E. Coacervates as models of membraneless organelles. Curr. Opin. Colloid Interface Sci. 52, 101416 (2021).9. Lin, Z., Beneyton, T., Baret, J.-C. & Martin, N. Coacervate Droplets for Synthetic Cells. Small Methods 7, 2300496 (2023).10. Morin, J. A. et al. Sequence-dependent surface condensation of a pioneer transcription factor on DNA. Nat. Phys. 18, 271-276 (2022).11. Frankel, E. A., Bevilacqua, P. C. & Keating, C. D. Polyamine / Nucleotide Coacervates Provide Strong Compartmentalization of Mg2+, Nucleotides, and RNA. Langmuir 32, 2041-2049 (2016).12. Lindhoud, S. & Claessens, M. M. A. E. Accumulation of small protein molecules in a macroscopic complex coacervate. Soft Matter 12, 408-413 (2015).13. van Veldhuisen, T. W. et al. Enzymatic Regulation of Protein-Protein Interactions in Artificial Cells. Adv. Mater. Deerfield Beach Fla 35, e2300947 (2023).14. Altenburg, W. J. et al. Programmed spatial organization of biomacromolecules into discrete, coacervate-based protocells. Nat Commun. 11, 6282 (2020).15. Poudyal, R. R. et al. Template-directed RNA polymerization and enhanced ribozyme catalysis inside membraneless compartments formed by coacervates. Nat. Commun. 10, 490 (2019).16. Erkamp, N. A. etal. Spatially non-uniform condensates emerge from dynamically arrested phase separation. Nat. Commun. 14, 684 (2023).17. Lu, T. & Spruijt, E. Multiphase Complex Coacervate Droplets. / . Am. Chem. Soc. 142, 2905-2914 (2020).18. Mountain, G. A. & Keating, C. D. Formation of Multiphase Complex Coacervates and Partitioning of Biomolecules within them. Biomacromolecules 21, 630-640 (2020).19. Feric, M. etal. Coexisting Liquid Phases Underlie Nucleolar Subcompartments. Cell 165, 1686-1697 (2016). 20. Nakashima, K. K., Vibhute, M. A. & Spruijt, E. Biomolecular Chemistry in Liquid Phase Separated Compartments. Front Mol. Biosci.6, 21 (2019).21. Cingil, H. E., Meertens, N. C. H. & Voets, I. K. Temporally Programmed Disassembly and Reassembly of C3Ms. Small 14, 1802089 (2018).22. Banerjee, P. R., Milin, A. N., Moosa, M. M., Onuchic, P. L. & Deniz, A. A. Reentrant Phase Transition Drives Dynamic Substructure Formation in Ribonucleoprotein Droplets. Angew. Chem. 129, 11512-11517 (2017).23. Nakashima, K. K., Baaij, J. F. & Spruijt, E. Reversible generation of coacervate droplets in an enzymatic network. Soft Matter 14, 361-367 (2018).24. Aumiller, W. M. & Keating, C. D. Phosphorylation-mediated RNA / peptide complex coacervation as a model for intracellular liquid organelles. Nat Chem.8, 129-137 (2016).25. Gao, N. & Mann, S. Membranized Coacervate Microdroplets: from Versatile Protocell Models to Cytomimetic Materials. Acc. Chem. Res. 56, 297-307 (2023).26. Zhang, Y. et al. Giant Coacervate Vesicles As an Integrated Approach to Cytomimetic Modeling. / . Am. Chem. Soc. 143, 2866-2874 (2021).27. Mason, A. F., Buddingh’, B. C., Williams, D. S. & van Hest, J. C. M. Hierarchical Self-Assembly of a Copolymer-Stabilized Coacervate Protocell. / . Am. Chem. Soc. 139, 17309-17312 (2017).28. Yin, C., Lin, Z., Jiang, X., Martin, N. & Tian, L. Engineering the Coacervate Microdroplet Interface via Polyelectrolyte and Surfactant Complexation. ACS Appl. Mater. Interfaces 15, 27447-27456 (2023).29. Li, J., Liu, X., Abdelmohsen, L. K. E. A., Williams, D. S. & Huang, X. Spatial Organization in Proteinaceous Membrane-Stabilized Coacervate Protocells. Small 15, 1902893 (2019).30. Gao, N., Xu, C., Yin, Z., Li, M. & Mann, S. Triggerable Protocell Capture in Nanoparticle-Caged Coacervate Microdroplets. / . Am. Chem. Soc. 144, 3855-3862 (2022).31. Uversky, V. N., Kuznetsova, I. M., Turoverov, K. K. & Zaslavsky, B. Intrinsically disordered proteins as crucial constituents of cellular aqueous two phase systems and coacervates. FEBS Lett. 589, 15-22 (2015).32. Boeynaems, S. etal. Protein Phase Separation: A New Phase in Cell Biology. Trends Cell Biol. 28, 420-435 (2018).33. Martin, E. W. & Holehouse, A. S. Intrinsically disordered protein regions and phase separation: sequence determinants of assembly or lack thereof. Emerg. Top. Life Sci.4, 307-329 (2020).Atty Docket TUOE-P2010WQ-00676481&&&&&&&&

Claims

Atty Docket TUOE-P2010WQ-00676481Claims1. A synthetic condensate comprising one or multiple types of molecules.

2. The synthetic condensate of claim 1, comprising a first molecule and a second molecule, wherein the first molecule and a second molecule are oppositely charged.

3. The synthetic condensate of any preceding claims, wherein both the first molecule and the second molecule belong to same type of molecule selected from the group consisting of a polypeptide, a DNA, and an RNA.

4. The synthetic condensate of any preceding claims, wherein both the first molecule and the second molecule are polypeptides.

5. The synthetic condensate of any of preceding claims, further comprising a stabilizing protein, wherein the stabilizing protein comprises two domains, a first domain and a second domain, the first domain and the second domain being connected via a linker, wherein the first domain is a charged disordered region contributing to electrostatic attraction to the condensate that is oppositely charged, and wherein the second domain is a buoyancy group having electrostatic and / or steric repulsion towards the condensate.

6. The synthetic condensate of claim 5, wherein the stabilizing proteins forms a protein layer on the outer surface of the synthetic condensate.

7. The synthetic condensate of any of claims 5-6, wherein the first domain is the Tau protein or fragment thereof, and the second domain is the green fluorescent protein (GFP) or fragment thereof.

8. The synthetic condensate of any of claims 5-6, wherein the first domain is the Tau protein or fragment thereof, and the second domain is the Glutathione S-Transferase (GST) or fragment thereof.

9. The synthetic condensate of any of claims 5-6, wherein the first domain is the a-synuclein (aSyn) protein or fragment thereof, and the second domain is the green fluorescent protein (GFP) or fragment thereof.

10. The synthetic condensate of any of claims 4-9, wherein the ratio of polypeptide length between the first polypeptide and the second polypeptide ranges between 1 / 15 and 15 / 1.Atty Docket TUOE-P2010WQ-0067648111. A complex comprising the synthetic condensate of any of claims 1-4 and a stabilizing protein, wherein the stabilizing protein comprises two domains, a first domain and a second domain, the first domain and the second domain being connected via a linker, wherein the first domain is a charged disordered region contributing to electrostatic attraction to the condensate that is oppositely charged, and wherein the second domain is a buoyancy group having electrostatic and / or steric repulsion towards the condensate.

12. The complex of claim 11, further comprising a payload, wherein the payload is selected from the group consisting of a chemical, a drug, a cosmetic, a food supplement13. A method of preparing the synthetic condensate of any of claims 1-10, comprising a) mixing a first molecule and a second molecule, wherein the first molecule and a second molecule are oppositely charged,b) adding a stabilizing protein to the mixture of (a), wherein the stabilizing protein comprises two domains, a first domain and a second domain, the first domain and the second domain being connected via a linker, wherein the first domain is a charged disordered region contributing to electrostatic attraction to the condensate that is oppositely charged, and wherein the second domain is a buoyancy group having electrostatic and / or steric repulsion towards the condensate; andc) obtaining the synthetic condensate.

Images