{"paper_id":"48e8cb35-d641-4022-afe9-eb0b073b3fea","body_text":"Synthesis and Characterization of Phase-Separated Extracellular Condensates \n in Interactions with Cells \nAida Naghilou1,2, Tom M.J. Evers1,2, Oskar Armbruster3, Vahid Satarifard1,4, Alireza Mashaghi1,2* \n \n1 Medical Systems Biophysics and Bioengineering, Leiden Academic Centre for Drug Research, Faculty \nof Science, Leiden University, 2333CC, Leiden, The Netherlands \n2 Laboratory for Interdisciplinary Medical Innovations, Centre for Interdisciplinary Genome Research, \n2333CC, Leiden University, Leiden, The Netherlands \n3 Institute of Synthetic Bioarchitectures, Department of Bio technology and Food Science , BOKU \nUniversity, 1190 Vienna, Austria \n4 Yale Institute for Network Science, Yale University, New Haven, CT 06520, USA \n*Correspondence: a.mashaghi.tabari@lacdr.leidenuniv.nl  \n \nAbstract \n \nBiomolecular condensates formed through liquid-liquid phase separation play key roles in intracellular \norganization and signaling, yet their function in extracellular environments remains largely unexplored.  \nHere, we establish a model using heparan sulfate, a key component of the extracellular matrix, to study \nextracellular condensate -cell interactions . We demonstrate that heparan sulfate can form \ncondensates with a positively charged counterpart  in serum-containing solutions, mimicking the \ncomplexity of extracellular fluid, and supporting cell viability . We observe that these condensates \nadhere to cell membranes and remain stable, enabling a versatile platform for examining extracellular \ncondensate dynamics and quantifying their rheological properties as well as their adhesion forces with \ncellular surfaces. Our findings and methodology open new avenues for understanding the organizational \nroles of condensates beyond cellular boundaries. \n \n \nIntroduction \nPhase-separated biomolecular condensates have emerged as key players in  cellular \ncompartmentalization, enabling a range of critical functions  such as  gene regulation, membrane \nremodeling, and intracellular sensing  1-6. Liquid-liquid phase separation has been identified as the \nunderlying mechanism of condensate formation and has been linked to physicochemical properties of \nbiomolecules7-9. Although much of the research on biomolecular condensates has focused on their \nroles inside cells, recent findings suggest that extracellular molecules can also form condensates 10-12. \nThese observations raise questions about the role of condensates in extracellular organization, where \nthey may contribute to interactions with cells and the extracellular matrix.  Despite these possibilities, \ndata on extracellular condensates and their interactions with cells are scarce. Developing assays for \nstudying extracellular condensate s and their cellular interactions could open new avenues for \nunderstanding and engineering biochemical processes, as phase separation may modulate enzyme \nkinetics, transport processes, and mechanochemistry 13,14. \n \nModeling condensate-cell interactions in an extracellular setting requires considerations distinct from \nthose for intracellular condensates. While many studies use simplified buffer systems to mimic the \nintracellular environment by introducing ions and crowding age nts 15-18, extracellular condensates \nrequire assays in culture medium conditions that support cell viability. In addition, the extracellular fluid \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 26, 2025. ; https://doi.org/10.1101/2025.03.24.644961doi: bioRxiv preprint \n\nin humans resembles plasma 19, implying that  extracellular condensate s would  encounter plasma \nproteins and other soluble factors, necessitating the development of a condensate model that sustains \nstability within such an environment to study its properties. \n \nOne characteristic feature of the extracellular environment is its abundance of charged \npolysaccharides, which can contribute to condensate formation alongside proteins . \nGlycosaminoglycans (GAGs) are long, unbranched, and highly negatively charged polysaccharides that \nare key components of the extracellular environment  20. They are involved in a variety of biological and \npathological processes, including development, angiogenesis, and inflammatory responses  21,22. One \nsuch GAG, heparan sulfate (HS), interacts with a broad spectrum of molecules at the cellular interface, \nplaying a vital role in extracellular signaling 23-25. Heparin (H) is another GAG known for its anticoagulant \nproperties 26 has been shown to form  coacervates with charged proteins 23,24. Furthermore, in silico  \napproaches have predicted the condensate formation propensity of numerous extracellular proteins 27. \nThese lines of evidence suggest that phase separation of GAGs and extracellular proteins is ubiquitous \nand may play a role in various (patho)physiological settings. \n \nIn this article, we demonstrate that HS phase separates to form condensate droplets in the presence of \na positively charged counterpart. Importantly, these condensates are also formed in serum-like media \ncontaining proteins and nutrients necessary to sustain cell viability . We used a combination of optical \nand scanning probe microscopy (SPM) as well as fluorescence recovery after photobleaching (FRAP) to \ninvestigate the liquid-like behavior of these condensates. In addition, we developed assays enabling the \ndetermination of condensate mechanical proper ties such as viscoelasticity and adhesion forces on \ncells, marking the first application of SPM in analyzing cell -condensate interactions . Our assay is \nindependent of condensate nature and cell types, providing a versatile and adaptable platform for \ninvestigating interactions between extracellular condensates, cells, and the extracellular matrix (ECM).  \n \nResults \nHeparan sulfate phase separates with poly-L-lysine with distinct differences to heparin  \nWhile H phase separates with poly-L-lysine (pK) at 1  M KCl with crowding agents, no droplets were \nformed with pK -HS at the same buffer concentration ( Figure 1A). To gain insights into the nature of \ncondensation of pK -HS, we performed a concentration gradient analysis of KCl to determine the \nconditions at which condensate formation occurs ( Figure 1B). At lower salt contents, pK-HS phase \nseparation leads to condensate formation, whereas at higher KCl  concentrations, the system \ntransitions back to a single-phase regime, highlighting the differences between HS and H condensates. \nNotably, pK-H interactions at low salt concentrations, comparable to those that leads to pK-HS droplet \nformation, causes an aggregation of pK-H, as depicted in Figure 1C. As 0.15 M KCl is routinely used for \nstudying biological systems, for further investigation of pK -HS condensates we employed this salt \nconcentration. \n \nIn order to demonstrate the liquid-like nature of pK-HS condensates and to decipher their viscoelastic \ncharacteristics, we employed SPM and FRAP. Figure 1D displays the rheological properties of droplets \nat room temperature (RT) . The real (elastic modulus, G’) and imaginary (viscous modulus, G”) shear \nmoduli are presented as blue and red circles, respectively. The larger values of G” in comparison to G’ \ndemonstrated a dominant viscous behavior of droplets at all experimentally accessible frequencies. A \nlinear fit of G” was performed to determine the viscosity,  ƞ which was found to  be (26.1  ± 0.5) Pa s \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 26, 2025. ; https://doi.org/10.1101/2025.03.24.644961doi: bioRxiv preprint \n\n(Error! Reference source not found. A). The surface energy density of the droplets , 𝛾 was calculated \nfrom the quasi-static indentation curve (see Methods) and is shown in in Error! Reference source not \nfound.B. Average recovery traces, the half-time of recovery, 𝜏1/2 and diffusion coefficients, D as well as \nrepresentative confocal images of condensates during FRAP experiments is exemplified in Figure 1E, \nalso indicating the liquid-like behavior of pK-HS condensates based on the fast recovery.  \n \nAs we aimed to study cell-condensate interactions, a crucial prelude is biological temperatures. Hence, \nwe repeated the SPM and FRAP measurements after heating the droplets to 37 °C, shown in Figure 1F \nand G, respectively. In accordance with Arrhenius’ law 28, ƞ decreased at 37°C to (23.5 ± 0.7) Pa s. This \nis also indicated by the smaller values of 𝜏1/2 and larger D at higher temperature. A comparison of the \nmaterial properties of condensates at RT and 37°C is depicted in Error! Reference source not found.. \n \n \n \nFigure 1: pK-HS condensates differ from pK-H. A) Representative confocal micrographs of pK-H droplets and pK-HS at 1 M \nKCl, scale bar is 50 µm. B) Exemplary confocal micrographs of pK -HS droplets at various KCl concentrations, scale bar is \n50 µm. C) Representative confocal micrographs of pK-H droplets at 0.15 M KCl D) Elastic (G’, blue) and viscous (G”, red) shear \nmoduli of pK -HS condensates formed at 0.15  M KCl  and RT, measured at RT  (mean ± SD, n = 3). E) Time-dependent \nnormalized fluorescence intensity depicting the recovery of the bleached area (circular data points) and the fit with Eq. 2, \nresulting in half-time of recovery (𝜏1/2) and diffusion constant (D) with Eq. 3 pK-HS condensates formed at 0.15 M KCl and RT, \nmeasured at RT (mean ± SD, n = 1) as well as representative confocal micrographs of droplet recovery after photobleaching, \nscale bar is 10 µm. F) Elastic (G’, blue) and viscous (G”, red) shear moduli of pK-HS condensates formed at 0.15 M KCl and RT, \nmeasured at 37°C (mean ± SD, n = 3). G) Time-dependent normalized fluorescence intensity depicting the recovery of the \nbleached area (circular data points) and the fit with Eq. 2, resulting in half-time of recovery (𝜏1/2) and diffusion constant (D) with \nEq. 3 for pK-HS condensates formed at 0.15 M KCl and RT, measured at 37°C (mean ± SD, n = 1) as well as representative \nconfocal micrographs of droplet recovery after photobleaching, scale bar is 10 µm. \n \nHeparan sulfate phase separates with poly -L-lysine in a serum-containing medium without the \nneed for crowding agents or surface passivation  \nDeciphering condensate-cell interactions in the extracellular environment requires culture conditions \nthat preserve cell viability. This prompted us to study whether pK-HS condensates also form in a culture \nmedium. Furthermore, extracellular fluid in human tissues is rich in plasma proteins, which may impact \ncondensate formation and properties. As such, we prepared serum containing media typically used for \nin vitro cell culture studies  (see Methods). pK and HS were added to the medium with the same end \nconcentrations used in buffer. Figure 2 A shows phase contrast micrographs of the pK-HS condensates \nin the culture medium, with morphologies similar to those formed in buffer. To confirm that the \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 26, 2025. ; https://doi.org/10.1101/2025.03.24.644961doi: bioRxiv preprint \n\ncondensates are indeed pK -HS and not the result of interactions between any medium components \nwith either pK or HS, we repeated the experiments using medium with added pK without HS (Figure 2 B \ntop), and vice versa  (Figure 2 B, bottom). No condensates were formed unless both pK and HS were \npresent, thereby demonstrating that the droplets formed are indeed pK-HS condensates. Interestingly, \ndue to the high protein content of culture medium, neither a crowding agent nor a surface coating was \nnecessary. Condensates retained their natural spherical morphology, despite a lack of any surface \npassivation, which is in contrast to the pK-HS condensates prepared in buffer (Error! Reference source \nnot found.). This hints at the fact that the molecules of the culture medium may have bounded to the \nsurface of the dish, acting as a passivation.  To gain deeper insights on the differences between H and \nHS, we also investigated pK -H interactions in the same culture media as used for pK -HS. Consistent \nwith the behavior seen in buffer ( Figure 1A and C), while pK -HS leads to condensate formation, pK -H \nforms aggregates in culture medium (Error! Reference source not found.)  \n \n \nFigure 2: pK-HS condensates formed in culture medium. A) Representative phase contrast micrograph of pK-HS droplets \nformed in culture medium; scale bar is 50 µm. B) Exemplary phase contrast micrographs of culture medium with added pK \nwithout HS (top) and with added HS without pK (bottom), scale bar is 50 µm. C) Elastic (G’, blue) and viscous (G”, red) shear \nmoduli of pK-HS condensates formed in culture medium and RT, measured at RT  (mean ± SD, n = 3). D) Time-dependent \nnormalized fluorescence intensity depicting the recovery of the bleached area (circular data points) and the fit with Eq. 2, \nresulting in half-time of recovery (𝜏1/2) and diffusion constant (D) with Eq. 3 pK-HS condensates formed in culture medium and \nRT, measured at RT  (mean ± SD, n = 1) as well as representative confocal micrographs of droplet recovery after \nphotobleaching, scale bar is 10 µm. E) Elastic (G’, blue) and viscous (G”, red) shear moduli of pK-HS condensates formed in \nculture medium and RT, measured at 37°C  (mean ± SD, n = 3). F) Time-dependent normalized fluorescence intensity \ndepicting the recovery of the bleached area (circular data points) and the fit with Eq. 2, resulting in half-time of recovery (𝜏1/2) \nand diffusion constant (D) with Eq. 3 pK-HS condensates formed in culture medium and RT, measured at 37°C (mean ± SD, \nn = 1) as well as representative confocal micrographs of droplet recovery after photobleaching, scale bar is 10 µm. G) Elastic \n(G’, blue) and viscous (G”, red) shear moduli of pK-HS condensates formed in culture medium and 37°C, measured at 37°C \n(mean ± SD, n = 3). H) Time-dependent normalized fluorescence intensity depicting the recovery of the bleached area \n(circular data points) and the fit with Eq. 2, resulting in half-time of recovery (𝜏1/2) and diffusion constant (D) with Eq. 3 pK-HS \ncondensates formed in culture medium and 37°C, measured at 37°C (mean ± SD, n = 1) as well as representative confocal \nmicrographs of droplet recovery after photobleaching, scale bar is 10 µm. \nThe material properties of pK-HS condensates in medium vary in comparison to those formed in buffer. \nSPM measurements (Figure 2C) showed that, the condensates are closer to their cross -over point at \nhigher frequencies and their ƞ at RT is slightly lower (21.7 ± 0.7) Pa s than the pK-HS droplets in buffer. \nInterestingly, we did not see the same trend in FRAP ( Figure 2D), where 𝜏1/2 was in fact slightly higher \nthan the values for measurements in buffer at RT. In contrast to the slight changes of ƞ between medium \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 26, 2025. ; https://doi.org/10.1101/2025.03.24.644961doi: bioRxiv preprint \n\nand buffer, 𝛾 values were noticeably lower for pK-HS droplets formed in medium in comparison to those \nin buffer ( Error! Reference source not found. B). This behavior could be related to the absent of the \ncrowding agent in the medium, greatly impacting the interfacial 𝛾. Afterwards, we increased the \ntemperature from RT to 37°C, and observed lower viscosities in SPM (Figure 2E), as well as smaller 𝜏1/2 \nand larger D in FRAP (Figure 2F).  \n \nFor studying condensate interactions with cells, it is crucial to avoid temperature fluctuations for cells. \nHence, we also investigated the phase separation with medium pre-heated to 37°C before adding pK \nand HS to form condensates. The resulting droplets showed material properties comparable to those \nformed by heating the medium from RT to 37°C, with lower ƞ and faster fluorescent recovery in \ncomparison to pk -HS condensate prepared at RT, as shown in Figure 2G and H. A c omparison of \nmaterial properties of droplets in all conditions measured by SPM and FRAP is depicted in Error! \nReference source not found. . These results demonstrate that pK -HS droplets serve as a suitable \nmodel system for studying condensate-cell interactions, as they form under biologically relevant fluids \nand temperature while retaining their liquid-like properties. \n \nScanning probe microscopy reveals cell-condensate interactions  \nAfter demonstrating that pK-HS condensates remain stable in culture medium and maintain their liquid-\nlike characteristics, we proceeded to investigate their interactions with fibroblast cells (Figure 3A). For \nthis purpose, pK and HS were mixed to a fresh culture medium at 37°C and immediately added on the \ncells. An exemplary phase contrast micrograph after 2  h of incubation, depicting the condensates \nsettling both on top and next to the cells is shown in Figure 3B. The condensate size on the cells is larger \nin comparison to the ones on the dish, which may be due to the movement of cells, increasing the \nchances of condensate fusion. This is an indication that cells are viable and migrating within the pK-HS \ncontaining medium, also visible by similar morphologies of cells in culture medium  without and with \ncondensates (Figure 3A and B, respectively). \n \nTo determine the mechanical properties of droplets resting on cells, it is crucial to consider that the \ncells’ elastic moduli are much lower than those of polystyrene dishes, which may lead to a deformation \nof cell while indenting the condensate. To address this, we derived a stacked Hertzian contact model, \nallowing us to extract the G’ and G” of condensates by taking into account the shear moduli of the cell \nas a known parameter (see Methods). Hence, we first measured the rheological properties of the cells \nand in the same manner the condensates resting on cells (Figure 3C and D, respectively). Both G’ and \nG” of fibroblast cells show a rise at higher frequencies with a dominant G’, hinting at a mainly elastic \nbehavior (Figure 3E). Employing the stacked Hertzian contact model (Eq. 4-10), the viscoelastic \ncharacteristics of condensates on cells were determined (Figure 3F). The condensates on cells showed \nsimilar ƞ and slightly higher 𝛾 in comparison to those formed in medium without the presence of cells. \nA comparison of ƞ and 𝛾 for condensates in medium, with and without the cells is depicted in  \nSupplementary Figure S 4. \n \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 26, 2025. ; https://doi.org/10.1101/2025.03.24.644961doi: bioRxiv preprint \n\n \nFigure 3: pK-HS condensates’ rheological properties in contact with fibroblast cells. A) Representative phase contrast \nmicrograph of fibroblast cells in culture medium, scale bar is 50 µm. B) Exemplary phase contrast micrographs of pK -HS \ncondensates in culture medium residing on fibroblast cells, scale bar is 50 µm. C) Schematic representation of the SPM probe, \nindenting the fibroblast cells. D) Schematic representation of the SPM probe, indenting the condensates residing on cells D) \nFibroblast cells in culture medium at 37°C E) Elastic (G’, blue) and viscous (G”, red) shear moduli of fibroblast cells measured \nat 37°C (mean ± SD, n = 4). F) Elastic (G’, blue) and viscous (G”, red) shear moduli of pK-HS condensates on the fibroblast \ncells formed in culture medium and 37°C, measured at 37°C (mean ± SD, n = 3). \n \nNext, we measured the adhesion forces between the condensates and the cells with SPM (Figure 4A). \nTo achieve this, a tipless cantilever was coated with laminin, chosen due to its strong interactions with \nHS 29. A representative force ( F) curve as a function of time ( t) is shown in Figure 4B, illustrating the \nmeasurement process. During the experiment, the cantilever head was lowered towards the \ncondensate until a predefined deflection was achieved (indicated by the green line in Figure 4B). This \nwas followed by a pause period (red line) to allow sufficient time for the laminin -coated cantilever to \nbind firmly to the condensate. Subsequently, the cantilever was retracted, initiating the detachment \nphase of the curve, represented by the dark grey line.  \n \nImportantly, the alignment of the baseline during the approach and detachment phases demonstrates \nthat while the cantilever lost contact with the condensate, the condensate itself remained securely \nattached to the cell surface. This observation underscores the strong adhesive interaction between the \ncondensates and the cellular surface , hinting that the pK -HS droplets could function as a biological \nadhesive on the cell. The minimum force value observed during the detachment phase corresponds to \nthe adhesion  forces Fad between the cantilever and the cell (highlighted by the violet arrow)  and \naverages to 0.98 ± 0.09 nN. The distribution of the adhesion forces for all measurements is depicted in \nSupplementary Figure S 5.  \n \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 26, 2025. ; https://doi.org/10.1101/2025.03.24.644961doi: bioRxiv preprint \n\nEmploying the JKR model , and droplet sizes  (see Methods, Eq. 11) the interfacial surface energy \ndistribution 𝛾 between the droplet and the cantilever were calculated . This was compared with 𝛾 \ndetermined from the quasi-static indentation curves while considering the contribution of the cell \ndeformation to the overall indentation (See Methods, Eq. 12 -15). The results of the comparison  are \nshown in  Figure 4C. The higher 𝛾 between the condensate and the cantilever in comparison to the \ncondensate and dilute phase indicates the adhesive interactions of the pK-HS droplets with the laminin \ncoated cantilever. \n \n \nFigure 4: pK-HS condensates’ adhesion forces with fibroblast cells. A) Schematic representation of the SPM cantilever, \ndetaching the condensates residing on cells. B) A representative curve of cantilever deflection force over time, F(t) showing \nthe approach in green, pause in red, and retraction in dark grey. The violet arrow indicates the lowest force value during \ndetachment, corresponding to the adhesion force Fad. C) Comparison of the interfacial surface energy density (𝛾) between the \npK-HS condensates condensate and dilute phase denoted as condensate and between the condensate and laminin coated \ncantilever denoted as condensate-cantilever on the fibroblast cells formed in culture medium and 37°C, measured at 37°C \n(mean ± SD, n = 3). \n \nDiscussion and Conclusions \nLiquid-liquid phase separation underpins key cellular functions  30. It drives the formation of functional \ncondensates in the cytoplasm and nucleoplasm  31,32 and can occur on the plasma membrane or \nendoplasmic reticulum, supporting condensate formation that aids in cell signaling, tight junctions, and \nsynaptic transmission 33,34. The intracellular environment with its high protein density facilitat es liquid-\nliquid phase separation 23, and thus has been the primary focus on condensate research over the last \ndecade. The extracellular environment of tissues is also rich in biopolymers with a predicted propensity \nfor condensate formation  27, and recent studies have experimentally identified the presence of \nmolecules in the extracellular environment  capable of forming condensates, including elastin and \ngalectin-3 10,11. The discovery of extracellular condensates is paramount as the condensates may reside \non cells, a configuration that could enable various mechanical, biochemical, and transport processes \nat the cell ular interface 23. These findings underscore the need for technical development that enable \ncharacterization of extracellular condensates and their cellular interactions to understand their roles \nand mechanisms. In this study, we developed a methodology using SPM for quantifying interactions \nbetween extracellular condensates and cells. While SPM has recently been utilized to study \nbiomolecular condensates, particularly for probing liquid-gel phase transitions 35, our work extends its \npotential for analyzing extracellular condensates and establishes the groundwork for precise \nmeasurements of their interactions with cellular surfaces.  \n \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 26, 2025. ; https://doi.org/10.1101/2025.03.24.644961doi: bioRxiv preprint \n\nIn this regard, we created biomolecular condensates in an extracellular fluid mimic. Widely present on \ncell surfaces, within the ECM, basement membranes, and even intracellularly, HS plays a pivotal role in \nnumerous biological processes  36. Through dynamic multivalent ionic interactions, it binds various \npositively charged extracellular proteins, aiding in the structural organization of the ECM 25. \nFurthermore, its interactions with positively charged molecules, such as chemokines and growth \nfactors, have been shown to results in condensate formation 23,24. Given its crucial role in the ECM and \nits affinity for positive molecules, we utilized pK -HS condensates in a culture medium containing \ncomponents that support cell viability.  These condensates exhibited morphologies similar to  those \nformed in the buffer system with crowding agents. However, in the culture medium, neither additional \ncrowding agents nor surface passivation were needed, suggesting that the protein -rich environment \nprovided by serum acts as a natural crowding agent, while protein -surface interactions mimic surface \npassivation. We characterized the material properties of pK -HS condensates using SPM and FRAP at \nroom and physiological temperatures. Our results demonstrated that condensate viscosity decreases \nat higher temperatures, which is in line with previous studies investigating the impact of temperature on \ncondensates 28. While H is often used as a mimetic for HS in biochemical interaction studies, our study \nreveals clear differences in condensation behavior s of HS and H . Specifically, under the same \nconditions in which HS forms liquid condensates, H instead forms aggregates. As such, we recommend \nthat for extracellular condensate studies, H should not be used as a replacement for HS. These distinct \nbehaviors and physicochemical properties may be explained the differing roles of HS and H in the \nextracellular space 37. \n \nWe observed that pK -HS condensates form in culture media containing cells and stably adhere to \ncellular surfaces. Previous studies have highlighted the potential role of condensates in mediating cell-\ncell adhesion, suggesting that they may act as molecular bridges that facilitate intercellular \nconnections 10,38. Our findings not only align with these observations but also provide a new dimension \nby enabling the quantitative assessment of condensate properties during cell interactions. Using SPM \nand introducing a novel stacked Hertzian model, we were able to measure the rheological properties of \nthese condensates while they were directly on the cells. In addition to rheology , by determination of \nsurface tension, as well as adhesion forces with SPM, we offer a novel approach to understanding the \nmechanical dynamics of condensates on cells, paving the way for deeper insights into their biological \nimplications, including their roles in cellular communication and tissue organization. \n \nCondensate interaction with neighboring cells is fundamentally important with many implications for \nbiology and therapeutic development. These direct physical interactions  could enable various \nmechanical, biochemical, and transport processes at the condensate-cell interface. By organizing ECM \ncomponents, receptor ligands, or signaling molecules near the cell surface, these condensates may \ninfluence cell behavior through biochemical and mechanical processes21. They could enhance or inhibit \nenzymatic reactions, regulate substrate transport, and provide spatial control over signaling \npathways 22. Acting as dynamic microenvironments, extracellular condensates may regulate the \navailability and diffusion of key molecular players, thereby potentially exerting local control over nutrient \nuptake, drug delivery, or growth factor responses  39. Additionally, their adhesion to cell membranes \nsuggests a role in modulating cellular mechanical properties, such as stiffness or elasticity, with \nimplications for processes like migration, mechanosensing, and ECM remodeling 12. Condensates’ role \nin regulating mechanics could be particularly relevant in pathological contexts, such as cancer, where \nECM organization is disrupted 40. Future investigations will shed light on how extracellular condensates \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 26, 2025. ; https://doi.org/10.1101/2025.03.24.644961doi: bioRxiv preprint \n\nmediate interactions with the cells, the surrounding matrix, and potentially contribute to ECM \nremodeling. \n \nMethods \nCondensate preparation \nExperiments were performed on virgin polystyrene culture dishes  (93040, TPP, Switzerland) , with \nreagents purchased from Sigma-Aldrich unless noted otherwise. Dishes were treated with 1%w/v BSA \n(A2153) for 30 minutes followed by rinsing 5–6 times with deionized water. For pK-H condensates, the \ncoacervation buffer contained 1 M KCl (p9333), 10 mM imidazole (56750), 1 g/l NaN3 (8.22335), and 50 \ng/l Ficoll® PM 70 (F2878). H (H3393) and pK (P2658) were added sequentially with 30 µM concentration \nfor each. For the pK -HS condensates, the same buffer was used except for the co ncentration of KCl, \nwhich was varied between 0.05 and 1 M to determine the best condition, which was chosen as 0.15 M \nKCl, for all other pK -HS experiments. Same pK and HS  (H7640) concentrations of 30 µM were \nemployed. \n \nFor pK-HS droplet experiments in culture medium, DMEM (Capricorn, Germany) containing glucose \n(4.5 g/l) and Sodium Pyruvate was supplemented with 10% fetal calf serum, 2.5% HEPES, and 1% \npenicillin-streptomycin. Afterwards without any further additives, pK and HS were added to an end \nconcentration of 30 µM each. \n \nImaging of droplets and cells \nFor imaging the morphology of condensates as shown in Figure 1, they were prepared in an ultra -low \nbinding 384-Wells (Corning™, USA) and confocal microscopy was carried out with a Nikon Ti2 Eclipse \nmicroscope and a 20x/0.75 objective. Phase contrast microscopy as see n in Figure 2 and 3 was \nperformed with a Nikon Ti2 microscope and a 20x/0.4 objective and the droplets and cells were \nprepared in uncoated polystyrene TPP culture dishes.  \n \n \nFluorescent Recovery After Photobleaching (FRAP) \nFor FRAP, a 5% FITC -labeled pK (P3543) solution was mixed with the unlabeled counterpart. \nExperiments were conducted using a Nikon Ti2 Eclipse microscope with a 20x/0.75 objective. With a \n488 nm laser and an average power of 250  µW, a region with radius of  rn = 1.5 µm on the droplets was \nbleached for 1 s, followed by a 2 min recovery scan. This procedure was repeated for 7-9 condensates \nin one sample preparation. Data were normalized to an unbleached reference droplet to correct for \nphotodecay. The diffusion coefficient was calculated by fitting the recovery curve using the half-time of \nrecovery 41,42. Timepoint zero, t = 0 s is defined as the first measured point after the bleaching pulse. The \nnormalized fluorescence intensity Fnorm was calculated from the fluorescence intensity of the region of \ninterest Froi, the fluorescence intensity of the reference area Fref, and the background fluorescence \nintensity Fbkgd as  \n 𝐹!\"#$(𝑡) =\n%!\"#('))%$%&'(')\n%!()('))%$%&'(')\n%*+,\n# )%$%&'\n#\n%!()\n# )%$%&'\n#  Eq.1 \nWhere Fi denotes the time average of the fluorescence intensities for t < 0 s. Fnorm is subsequently fitted \nwith \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 26, 2025. ; https://doi.org/10.1101/2025.03.24.644961doi: bioRxiv preprint \n\n 𝐹(𝑡) =\n%-*%.+' , / /1- .\n/*' , / /1-  Eq.2 \nwhere F0 is the normalized fluorescence intensity at t = 0 s, F∞ is the fluorescence intensity at t → ∞, and \n𝜏 1/2 is the half-time of recovery. From 𝜏 1/2, the diffusion coefficient D can be calculated as \n 𝐷 = 0.224\n021\n, / /1\n Eq. 3 \n \nScanning probe microscopy measurements \nThe procedure for measuring and evaluation of rheological properties of condensates with SPM is stated \nelsewhere in detail  35. In short, a 40 µl bulk drop of condensates was prepared on TPP dishes and \nallowed to settle on the dish for 2 hours. A 5  µl drop from the sample was carefully transferred to the \ncantilever, to ensure no air bubbles were formed during SPM head placement. SAA -SPH-5UM Si3N4 \ncantilevers (Bruker, Germany) with a 23 µm height and a 5.16 µm radius hemispheric tip were used. \nCalibration of the cantilever's spring constant and amplitude was done using the thermal noise, and the \ncontact-mode method on polystyrene dishes, respective ly 43. To prevent droplet adhesion, the \ncantilever was passivated with 1% Pluronic (P2443) for 30 minutes followed by washing with deionized \nwater 44,45. After thermal stabilization (10 -15 minutes), condensates and cells were indented with a \nforce of 0.3  nN, at a velocity of 1  µm s-1 and 40  nm oscillation amplitude. G' and G\" values were \nextracted using custom evaluation software (Mathematica 13.2, Wolfram). Condensate radii were \nmeasured from phase contrast images in ImageJ. For the hemispheric condensates in buffer the single \nHertzian contact model, and for the spheric droplets in medium the double Hertzian contact model were \nemployed 35. Condensate experiments were conducted using three independent sample preparations, \nwith 8–16 condensates analyzed per sample. Rheology of cell was performed with four independent \nsample preparations, with 6–12 cells measured per sample. \n \nTo study the interactions between condensates and cells, SV-80 cells, a human fibroblast cell line, were \ncultured in DMEM supplemented with 10% fetal calf serum, 2.5% HEPES, and 1% penicillin -\nstreptomycin. Once the cells reached 60-70% confluency, the medium was discarded. A 5×5 mm² area \nwas then confined using UV -sterilized adhesive tape ( 0.12 mm thickness, Grace Bio -Labs, USA) to \nrestrict the condensate application area and minimize the required volume. Thereafter, pK and HS were \nmixed to a fresh culture medium at 37°C and immediately added on the cells. The cultures were kept at \n37°C and 5% CO2 for 2 h, allowing most coalescence events of condensates to cease. Afterwards, the \nrheological measurements were conducted as stated above  on three independent sample \npreparations, with 12 condensates analyzed per sample. \n \nFor adhesion experiments, Arrow ™ TL2 cantilevers (NanoWorld , Switzerland) were coated with \n10 µg/ml laminin (L2020). Condensates on cells were prepared as stated above and were indented with \na force of 1  nN and a velocity of 1  µm s-1, and the contact was kept for 60  s before detaching the \ncantilever at a velocity of 1  µm s-1. Experiments were conducted using three independent sample \npreparations, with 12 condensates analyzed per sample. \n \nStacked Hertzian model \nFigure 5 schematically depicts the experimental condition for the indenter-condensate-cell stack. The \nhemispherical Si3N4 probe (subscript 1) is indenting the condensate (subscript 2) which is resting on the \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 26, 2025. ; https://doi.org/10.1101/2025.03.24.644961doi: bioRxiv preprint \n\ncell (subscript 3). While the indenter can be assumed as infinitely rigid, d ue to the comparable shear \nmoduli of the condensate and the cell, both may deform under the external load by the indenter. \n \n \nFigure 5: Schematic representation of the rheological stacked Hertzian model and parameters. Subscript 1 : Hemispheric \nSi3N4 indenter, G1 = ∞, R1 = 5.13 µm, n1 = 1/2. Subscript 2:  Condensate, G2 is to be determined , R2 is measured from \nmicrographs, n2 = 1/2. Subscript 3: Cell, G3 ~ 1 kPa (exact value is measured separately), R3 = ∞, n3 = 1/2. \n \nTo calculate the shear moduli of the condensate on the cell, firstly e ach of the two contacts (indenter-\ncondensate and condensate-cell) can be described by t he Hertzian contact model. The contact force \nbetween the probe and condensate (denoted as 12) F12 as a function of deformation at the interface δ/1  \nis given by \n 𝐹/1 = 𝑈/1 \tδ/1\n2/1 Eq.4 \n \nwith 𝑈/1 =\n4\n2 𝐸/1 \t𝑅/1\n/ 1-\n, 𝐸/1 = 2\n/)5/1\n6/\n+\n/)511\n61\n4\n)/\n, 𝑅/1 = 2\n/\n7/\n+\n/\n71\n4\n)/\n, and 𝐸/,1 = 2\t𝐺/,161 + 𝜈/,19 where \n𝑅/  and 𝑅1 are the radii of curvature, 𝐺/  and 𝐺1 are the shear moduli, and 𝜈/  and 𝜈1 are the Poisson's \nratios of the two bodies, 1 and 2. \n \nUsing Hooke’s law, 𝑑𝐹 = 𝑘\t𝑑𝑥\t → \t𝑘 =\n9%\n9:, the corresponding spring is given by \n 𝑋/1 =\n9%/1\n9;/1\n=\n2\n1 𝑈/1 \tδ/1\n/ 1-\n Eq.5 \n \nThe contact between the condensate and cell (denoted 23) can be described equivalently. Afterwards, \nthe two contacts result in a series connection of two springs  (Fig….). In the steady state, the forces at \nthe indenter-condensate interface 𝐹/1  and at the condensate -cell interface 𝐹12 are identical, hence \n𝐹/1 = 𝐹12. With the overall indentation of the cell and condensate (δ = δ/1 +δ12), which is known from \nthe SPM head height, the indentations δ/1 \tand δ12 can be calculated as: \n δ/1 = δ ?1 + 2\n</1\n<13\n4\n1/2\n@\n)/\n Eq.6 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 26, 2025. ; https://doi.org/10.1101/2025.03.24.644961doi: bioRxiv preprint \n\nand \n δ12 = δ ?1 + 2\n<13\n</1\n4\n1/2\n@\n)/\n Eq.7 \nand thus, the respective springs (cf. Eq.5) \n 𝑋/1 =\n9%/1\n9;/1\n=\n2\n1 𝑈/1 δ/1\n//1 =\n2\n1 𝑈/1 δ//1 ?1 + 2\n</1\n<13\n4\n1/2\n@\n)//1\n Eq.8 \nand \n 𝑋12 =\n9%13\n9;13\n=\n2\n1 𝑈12δ12\n//1 =\n2\n1 𝑈12δ//1 ?1 + 2\n<13\n</1\n4\n1/2\n@\n)//1\n Eq.9 \n \nfollow. The series connection of 𝑋/1  and 𝑋12, 𝑋 is given by \n \n 𝑋 = (𝑋/1\n)/ + 𝑋12\n)/ ))/  Eq.10 \nHence, by separately measuring the shear modulus of the cell (G3) the shear modulus of the condensate \n(𝐺1) can be calculated by means of numerical inversion. It must be noted that 𝐺1 still contains \ncontributions from surface energy at low frequencies which is compensated by subtracting the low -\nfrequency value of 𝐺1\n=  from 𝐺1(𝑓). This is described in detail elsewhere (ref CRPS). \n \nThe adhesion forces Fad were evaluated with the JPK data evaluation software (version 8.0.168) as the \nminimum force exerted on the cantilever in the detachment section of the curve. From this value, the \nsurface energy density 𝛾 can be calculated as based on the Johnson-Kendall-Roberts (JKR) model 46: \n \n 𝛾 =\n%45\n>\t71\n Eq.11 \n \nWhere R2 is the radius of the condensate. \n \nQuasi-static approach curve analysis \nAs demonstrated previously, the surface energy density of the liquid -condensate-interface 𝛾1 can be \ndetermined from the quasi-static indentation curve of the condensate 35. In the present configuration of \ndroplet resting on cell, the indentation curve contains the mechanical response of both the condensate \nand of the cell. This requires the separation of the deformation of the condensate and the deformation \nof the cell. To do so, the  quasi-static shear modulus of the cell ( 𝐺2) needs to be known. This quantity \ncan be determined from the quasi-static indentation curve of the cell  by fitting the Hertzian contact \nmodel \n 𝐹(δ) =\n/@\n2 𝐺2𝑅/\n//1δ2/1 Eq.12 \nto the experimental data. Knowing 𝐺2, the shear modulus of the condensate 𝐺1 as a function of the \nindentation δ can be calculated from the experimental force -indentation curve of the condensate \nresting on the cell by numeric inversion of \n 𝐹(δ) =\n/@\n2 𝐺1𝑅/1\n//1δ/1\n2/1 Eq.13 \nwhere \n δ/1 =\nA\n/*B+/*61\n63\n.7/1\n/ /171\n7/ /1C\n1/3 Eq.14 \nis the indentation at the indenter-condensate interface. As the mechanical response of the condensate \nis determined by surface energy changes due to spheroidal distortion, G2 depends on δ/1  given by \n 𝐺1(δ/1 ) = 𝛾1\t𝜋 2\nD8(2)E1)\nE1(E1)/)171\n3 +\nD8(2*E1)\n1\tE3(E1)/)71\n3 ln\t\n/*E\n/\t)\tE4\n2\n/@ (𝑅/\n)/ + 𝑅1\n)/ )//1δ/1\n)2/1 Eq.15 \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 26, 2025. ; https://doi.org/10.1101/2025.03.24.644961doi: bioRxiv preprint \n\nwith 𝑐 = 𝑅1 − 𝛿/1 , 𝑎 = 𝑅1\n2/1𝑐)//1, and 𝑒1 = 1 +\nD1\nF1. \n \n \nAcknowledgments  \nThis project is partly supported by the Dutch Research Council (NWO), Open Competition grants \nOCENW.XS23.3.105 and OCENW.XS22.4.185. The authors are grateful to the technical support of \nLaurens Heling and Vahid Sheikhhassani form Medical Systems Biophysics  and Bioengineering at \nLeiden University. The authors also thank Kostas Tassis for assistance with confocal microscopy and \nFRAP measurements. \n  \nAuthor Contributions \nConceptualization: A.M., V.S.  Methodology: A.M., A.N., O.A. Investigation: A.N. Software: O.A. \nVisualization: A.N. Formal Analysis: A.N., O.A. Writing & Draft Preparation: A.N., A.M., T.E., Writing-\nReview & Editing: all authors, Project Administration, Funding acquisition & Supervision: A.M. \n \nDeclaration of Interests \nThe authors have no conflicting interests to disclose. \n \nReferences \n1 Wagh, K., Garcia, D. A. & Upadhyaya, A. Phase separation in transcription factor \ndynamics and chromatin organization. Curr Opin Struct Biol  71, 148 -155, \ndoi:10.1016/j.sbi.2021.06.009 (2021). \n2 Agudo-Canalejo, J., Schultz, S. W., Chino, H., Migliano, S. M., Saito, C., Koyama -\nHonda, I., Stenmark, H., Brech, A., May, A. I., Mizushima, N. & Knorr, R. L. Wetting \nregulates autophagy of phase -separated compartments and the cytosol. Nature \n591, 142-146, doi:10.1038/s41586-020-2992-3 (2021). \n3 Alberti, S. & Hyman, A. A. Biomolecular condensates at the nexus of cellular \nstress, protein aggregation disease and ageing. Nature Reviews Molecular Cell \nBiology 22, 196-213, doi:10.1038/s41580-020-00326-6 (2021). \n4 Evers, T. M. J., Holt, L. J., Alberti, S. & Mashaghi, A. Reciprocal regulation of cellular \nmechanics and metabolism. Nature Metabolism  3, 456 -468, \ndoi:10.1038/s42255-021-00384-w (2021). \n5 Ravindran, R. & Michnick, S. W. Biomolecular condensates as drivers of \nmembrane trafficking and remodelling. Current Opinion in Cell Biology  89, \n102393, doi:https://doi.org/10.1016/j.ceb.2024.102393 (2024). \n6 Satarifard, V. & Lipowsky, R. Mutual remodeling of interacting nanodroplets and \nvesicles. Communications Physics  6, 6, doi:10.1038/s42005 -022-01104-w \n(2023). \n7 Lyon, A. S., Peeples, W. B. & Rosen, M. K. A framework for understanding the \nfunctions of biomolecular condensates across scales. Nature Reviews Molecular \nCell Biology 22, 215-235, doi:10.1038/s41580-020-00303-z (2021). \n8 Mittag, T. & Pappu, R. V. A conceptual framework for understanding phase \nseparation and addressing open questions and challenges. Molecular Cell  82, \n2201-2214, doi:https://doi.org/10.1016/j.molcel.2022.05.018 (2022). \n9 Heidari, M., Moes, D., Schullian, O., Scalvini, B. & Mashaghi, A. A topology \nframework for macromolecular complexes and condensates. Nano Research 15, \n9809-9817, doi:10.1007/s12274-022-4355-x (2022). \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 26, 2025. ; https://doi.org/10.1101/2025.03.24.644961doi: bioRxiv preprint \n\n10 Chiu, Y.-P., Sun, Y.-C., Qiu, D.-C., Lin, Y.-H., Chen, Y.-Q., Kuo, J.-C. & Huang, J.-r. \nLiquid-liquid phase separation and extracellular multivalent interactions in the \ntale of galectin -3. Nature Communications  11, 1229, doi:10.1038/s41467 -020-\n15007-3 (2020). \n11 Vidal Ceballos, A., Díaz, A. J., Preston, J. M., Vairamon, C., Shen, C., Koder, R. L. \n& Elbaum -Garfinkle, S. Liquid to solid transition of elastin condensates. \nProceedings of the National Academy of Sciences  119, e2202240119, \ndoi:10.1073/pnas.2202240119 (2022). \n12 Muiznieks, L. D., Sharpe, S., Pomès, R. & Keeley, F. W. Role of Liquid–Liquid Phase \nSeparation in Assembly of Elastin and Other Extracellular Matrix Proteins. Journal \nof Molecular Biology  430, 4741 -4753, \ndoi:https://doi.org/10.1016/j.jmb.2018.06.010 (2018). \n13 O'Flynn, B. G. & Mittag, T. The role of liquid -liquid phase separation in regulating \nenzyme activity. Curr Opin Cell Biol  69, 70 -79, doi:10.1016/j.ceb.2020.12.012 \n(2021). \n14 Laha, S., Bauermann, J., Jülicher, F., Michaels, T. C. T. & Weber, C. A. Chemical \nreactions regulated by phase-separated condensates. Physical Review Research \n6, 043092, doi:10.1103/PhysRevResearch.6.043092 (2024). \n15 Ghosh, A., Kota, D. & Zhou, H. -X. Shear relaxation governs fusion dynamics of \nbiomolecular condensates. Nature Communications  12, 5995, \ndoi:10.1038/s41467-021-26274-z (2021). \n16 André, A. A. M. & Spruijt, E. Liquid -Liquid Phase Separation in Crowded \nEnvironments. Int J Mol Sci 21, doi:10.3390/ijms21165908 (2020). \n17 Ghosh, A., Zhang, X. & Zhou, H. -X. Tug of War between Condensate Phases in a \nMinimal Macromolecular System. Journal of the American Chemical Society 142, \n8848-8861, doi:10.1021/jacs.0c01881 (2020). \n18 Ghosh, A., Mazarakos, K. & Zhou, H. -X. Three archetypical classes of \nmacromolecular regulators of protein liquid –liquid phase separation. \nProceedings of the National Academy of Sciences  116, 19474 -19483, \ndoi:doi:10.1073/pnas.1907849116 (2019). \n19 Guyton, K. S. & Hall, J. E. Textbook of Medical Physiology.  (Surg Neurol Int., 2017). \n20 Silva, J. C., Carvalho, M. S., Han, X., Xia, K., Mikael, P. E., Cabral, J. M. S., Ferreira, \nF. C. & Linhardt, R. J. Compositional and structural analysis of \nglycosaminoglycans in cell -derived extracellular matrices. Glycoconj J 36, 141-\n154, doi:10.1007/s10719-019-09858-2 (2019). \n21 Jaqaman, K. & Ditlev, J. A. Biomolecular condensates in membrane receptor \nsignaling. Current Opinion in Cell Biology  69, 48 -54, \ndoi:https://doi.org/10.1016/j.ceb.2020.12.006 (2021). \n22 Niu, X., Zhang, L., Wu, Y., Zong, Z., Wang, B., Liu, J., Zhang, L. & Zhou, F. \nBiomolecular condensates: Formation mechanisms, biological functions, and \ntherapeutic targets. MedComm (2020) 4, e223, doi:10.1002/mco2.223 (2023). \n23 Xue, S., Zhou, F., Zhao, T., Zhao, H., Wang, X., Chen, L., Li, J.-p. & Luo, S.-Z. Phase \nseparation on cell surface facilitates bFGF signal transduction with heparan \nsulphate. Nature Communications  13, 1112, doi:10.1038/s41467 -022-28765-z \n(2022). \n24 Yu, X., Duan, G., Pei, P., Chen, L., Gu, R., Hu, W., Zhang, H., Wang, Y. -D., Gong, \nL., Liu, L., Chu, T. -T., Li, J. -P. & Luo, S. -Z. Heparan sulfate -dependent phase \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 26, 2025. ; https://doi.org/10.1101/2025.03.24.644961doi: bioRxiv preprint \n\nseparation of CCL5 and its chemotactic activity. eLife 13, RP93871, \ndoi:10.7554/eLife.93871 (2024). \n25 Alotaibi, F. S., Alsadun, M. M. R., Alsaiari, S. A., Ramakrishnan, K., Yates, E. A. & \nFernig, D. G. Interactions of proteins with heparan sulfate. Essays Biochem 68, \n479-489, doi:10.1042/ebc20230093 (2024). \n26 Zang, L., Zhu, H., Wang, K., Liu, Y., Yu, F. & Zhao, W. Not Just Anticoagulation-New \nand Old Applications of Heparin. Molecules 27, doi:10.3390/molecules27206968 \n(2022). \n27 Hadarovich, A., Singh, H. R., Ghosh, S., Scheremetjew, M., Rostam, N., Hyman, \nA. A. & Toth-Petroczy, A. PICNIC accurately predicts condensate-forming proteins \nregardless of their structural disorder across organisms. Nature Communications \n15, 10668, doi:10.1038/s41467-024-55089-x (2024). \n28 Alshareedah, I., Singh, A., Yang, S., Ramachandran, V., Quinn, A., Potoyan, D. A. \n& Banerjee, P. R. Determinants of viscoelasticity and flow activation energy in \nbiomolecular condensates. Science Advances  10, eadi6539, \ndoi:10.1126/sciadv.adi6539. \n29 Woodley, D. T., Rao, C. N., Hassell, J. R., Liotta, L. A., Martin, G. R. & Kleinman, H. \nK. Interactions of basement membrane components. Biochim Biophys Acta 761, \n278-283, doi:10.1016/0304-4165(83)90077-6 (1983). \n30 Boeynaems, S., Alberti, S., Fawzi, N. L., Mittag, T., Polymenidou, M., Rousseau, F., \nSchymkowitz, J., Shorter, J., Wolozin, B., Van Den Bosch, L., Tompa, P. & Fuxreiter, \nM. Protein Phase Separation: A New Phase in Cell Biology. Trends in Cell Biology \n28, 420-435, doi:10.1016/j.tcb.2018.02.004 (2018). \n31 Gibson, B. A., Doolittle, L. K., Schneider, M. W. G., Jensen, L. E., Gamarra, N., \nHenry, L., Gerlich, D. W., Redding, S. & Rosen, M. K. Organization of Chromatin by \nIntrinsic and Regulated Phase Separation. Cell 179, 470 -484.e421, \ndoi:10.1016/j.cell.2019.08.037 (2019). \n32 Li, P., Banjade, S., Cheng, H. -C., Kim, S., Chen, B., Guo, L., Llaguno, M., \nHollingsworth, J. V., King, D. S., Banani, S. F., Russo, P. S., Jiang, Q. -X., Nixon, B. \nT. & Rosen, M. K. Phase transitions in the assembly of multivalent signalling \nproteins. Nature 483, 336-340, doi:10.1038/nature10879 (2012). \n33 Banjade, S. & Rosen, M. K. Phase transitions of multivalent proteins can promote \nclustering of membrane receptors. Elife 3, doi:10.7554/eLife.04123 (2014). \n34 Pombo-García, K., Adame -Arana, O., Martin -Lemaitre, C., Jülicher, F. & \nHonigmann, A. Membrane prewetting by condensates promotes tight -junction \nbelt formation. Nature 632, 647-655, doi:10.1038/s41586-024-07726-0 (2024). \n35 Naghilou, A., Armbruster, O., Mashaghi, A. & Scanning probe microscopy \nelucidates gelation and rejuvenation of biomolecular condensates. Cell Reports \nPhysical Science 6, doi:10.1016/j.xcrp.2025.102430 (2025). \n36 Sasisekharan, R., Shriver, Z., Venkataraman, G. & Narayanasami, U. Roles of \nheparan-sulphate glycosaminoglycans in cancer. Nature Reviews Cancer 2, 521-\n528, doi:10.1038/nrc842 (2002). \n37 Baik, J. Y., Wang, C. L., Yang, B., Linhardt, R. J. & Sharfstein, S. T. Toward a \nbioengineered heparin. Bioengineered 3, 227 -231, doi:10.4161/bioe.20902 \n(2012). \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 26, 2025. ; https://doi.org/10.1101/2025.03.24.644961doi: bioRxiv preprint \n\n38 Gouveia, B., Kim, Y., Shaevitz, J. W., Petry, S., Stone, H. A. & Brangwynne, C. P. \nCapillary forces generated by biomolecular condensates. Nature 609, 255-264, \ndoi:10.1038/s41586-022-05138-6 (2022). \n39 Conti, B. A. & Oppikofer, M. Biomolecular condensates: new opportunities for \ndrug discovery and RNA therapeutics. Trends Pharmacol Sci  43, 820 -837, \ndoi:10.1016/j.tips.2022.07.001 (2022). \n40 Boija, A., Klein, I. A. & Young, R. A. Biomolecular Condensates and Cancer. \nCancer Cell 39, 174-192, doi:https://doi.org/10.1016/j.ccell.2020.12.003 (2021). \n41 Day, C. A. & Kang, M. The Utility of Fluorescence Recovery after Photobleaching \n(FRAP) to Study the Plasma Membrane. Membranes 13, 492 (2023). \n42 Carnell, M., Macmillan, A. & Whan, R. in Methods in Membrane Lipids   (ed Dylan \nM. Owen)  255-271 (Springer New York, 2015). \n43 Bocklitz, T., Walter, A., Hartmann, K., Rösch, P. & Popp, J. How to pre -process \nRaman spectra for reliable and stable models? Analytica Chimica Acta  704, 47-\n56, doi:https://doi.org/10.1016/j.aca.2011.06.043 (2011). \n44 Yao, R. -W. & Rosen, M. K. Advanced surface passivation for high -sensitivity \nstudies of biomolecular condensates. Proceedings of the National Academy of \nSciences 121, e2403013121, doi:doi:10.1073/pnas.2403013121 (2024). \n45 Rigato, A., Miyagi, A., Scheuring, S. & Rico, F. High -frequency microrheology \nreveals cytoskeleton dynamics in living cells. Nature Physics  13, 771 -775, \ndoi:10.1038/nphys4104 (2017). \n46 Johnson, K. L., Kendall, K., Roberts, A. D. & Tabor, D. Surface energy and the \ncontact of elastic solids. Proceedings of the Royal Society of London. A. \nMathematical and Physical Sciences  324, 301 -313, \ndoi:doi:10.1098/rspa.1971.0141 (1971). \n \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is \nThe copyright holder for this preprintthis version posted March 26, 2025. ; https://doi.org/10.1101/2025.03.24.644961doi: bioRxiv preprint","source_license":"CC-BY-4.0","license_restricted":false}