347 12.7*1.24mm Stainless steel coiled tubing ,Molecular mechanism of synchronous electrostatic condensation and coaggregation of α-synuclein and tau

Thank you for visiting Nature.com. You are using a browser version with limited CSS support. For the best experience, we recommend that you use an updated browser (or disable Compatibility Mode in Internet Explorer). In addition, to ensure ongoing support, we show the site without styles and JavaScript.
Sliders showing three articles per slide. Use the back and next buttons to move through the slides, or the slide controller buttons at the end to move through each slide.

347 Stainless Steel Pipe Specification

347 12.7*1.24mm Stainless steel coiled tubing 

Outside Diameter: 6.00 mm OD up to 914.4 mm OD, Sizes up to 24” NB available Ex-stock, OD Size Steel Tubes available Ex-stock

SS 347 Pipe Thickness Range: 0.3mm – 50 mm, SCH 5, SCH10, SCH 40, SCH 80, SCH 80S, SCH 160, SCH XXS, SCH XS
W.T: SCH5S, SCH10S, SCH40S, SCH80S, SCH160S, etc. (0.5-12mm) Or non-regular size to be tailored as required

Type: SS 347 Seamless Pipes | SS 347 ERW Pipes | SS 347 Welded Pipes | SS 347 Fabricated Pipes | SS 347 CDW Tubes, LSAW Pipes / Seam-welded / Redrawn

Form: SS 347 Round Pipes/ Tubes, SS 347 Square Pipes/ Tubes, SS 347 Rectangular Pipe/ Tubes, SS 347 Coiled Tubes, SS 347 “U” Shape, SS 347 Pan Cake Coils, SS 347 Hydraulic Tubes

Length: Single Random, Double Random & Required Length End: Plain End, Beveled End, Treaded

End Protection: Plastic Caps | Outside Finish: 2B, No.4, No.1, No.8 Mirror Finish for Stainless Steel Pipes, Finish as per customer Requirements

Delivery Condition: Annealed and Pickled, Polished, Bright Annealed, Cold Drawn

Inspection, Test Reports: Mill Test Certificates, EN 10204 3.1, Chemical Reports, Mechanical Reports, PMI Test Reports, Visual Inspection Reports, Third Party Inspection Reports, NABL Approved Lab Reports, Destructive Test Report, Non Destructive Test Reports

Packing: Packed in Wooden Boxes, Plastic Bags, Steel Strips Bundled, or as per Customers Requests

Specials: Sizes and Specifications other than above can be manufactured on request

SS 347 Pipe Size Range:1/2 inch NB, OD to 24 inch

ASTM A312 347: Seamless and straight-seam welded austenitic pipe intended for high temperature and general corrosive service. Filler metal not permitted during welding.

ASTM A358 347: Electric fusion welded austenitic pipe for corrosive and/or high temperature service. Typically only pipe up to 8 inch is produced to this specification. Addition of filler metal is permitted during welding.

ASTM A790 347: Seamless and straight-seam welded ferritic/austenitic (duplex) pipe intended for general corrosive service, with a particular emphasis on resistance to stress corrosion cracking.

ASTM A409 347: Straight-seam or spiral-seam electric fusion welded large diameter austenitic light-wall pipe in sizes 14” to 30” with walls Sch5S and Sch 10S for corrosive and/or high

ASTM A376 347: Seamless austenitic pipe for high temperature applications.

ASTM A813 347: Single-seam, single- or double- welded austenitic pipe for high temperature and general corrosive applications.

ASTM A814 347: Cold-worked welded austenitic pipe for high temperature and general corrosive service.

347H Stainless Steel Pipes Chemical Composition

Stainless Steel 347H Pipe Mechanical Properties

Stainless Steel 347H Pipes Physical Properties

Equivalent Grades for 347H Stainless Steel Pipe

Amyloid alpha-synuclein (αS) aggregation is a hallmark of Parkinson’s disease and other synucleinopathies. Recently, the tau protein commonly associated with Alzheimer’s disease has been associated with αS pathology and found to co-localize in αS-rich inclusions, although the molecular mechanism of coaggregation of the two proteins remains unclear. We report here that the αS phase separates into liquid condensates via electrostatic complex condensation with positively charged polypeptides such as tau. Depending on the affinity of αS for polycations and the rate of valence depletion of the coagulation network, clots undergo rapid gelation or coalescence followed by slow amyloid aggregation. By combining a suite of advanced biophysical techniques, we were able to characterize the liquid-liquid αS/Tau phase separation and identify the key factors that lead to the formation of heterogeneous aggregates containing both proteins in a liquid protein condensate.
In addition to membrane compartments, spatial separation in cells can also be achieved by the formation of protein-rich, liquid-like dense bodies called biomolecular condensates or droplets, through a process known as liquid-liquid phase separation (LLPS). These droplets are formed by multivalent temporal interactions, usually between proteins or proteins and RNA, and serve a variety of functions in almost all living systems. A large number of LLP-capable proteins exhibit sequences of low complexity that are highly disordered in nature and in the formation of biomolecular condensates3,4,5. Numerous experimental studies have revealed the flexible, often disordered, and multivalent nature of the proteins that make up these liquid-like condensates, although little is known about the specific molecular determinants that control the growth and maturation of these condensates to a more solid-like state. .
The new data support the hypothesis that aberrant protein-driven LLPS and transformation of droplets into solid structures may be relevant cellular pathways leading to the formation of insoluble toxic aggregates that are often hallmarks of degenerative diseases. Many LLPS-associated intrinsically disordered proteins (IDPs), often highly charged and flexible, have long been associated with neurodegeneration through the process of amyloid aggregation. In particular, biomolecular IDP condensates such as FUS7 or TDP-438 or proteins with large low complexity domains such as hnRNPA19 have been shown to age into gel-like or even solid forms through a process called fluidization. compound. to solid phase transition (LSPT) as a function of time or in response to certain post-translational modifications or pathologically significant mutations1,7.
Another IDP associated with LLPS in vivo is Tau, a microtubule-associated disordered protein whose amyloid aggregation has been implicated in Alzheimer’s disease10 but has also recently been implicated in Parkinson’s disease (PD) and other synaptic nuclear proteinopathies 11, 12, 13 are related. Tau has been shown to spontaneously dissociate from solution/cytoplasm due to favorable electrostatic interactions14, resulting in the formation of tau-enriched droplets known as electrostatic coacervates. It has also been observed that this type of non-specific interaction is the driving force behind many biomolecular condensates in nature15. In the case of a tau protein, electrostatic aggregation can be formed by simple aggregation, in which oppositely charged regions of the protein trigger the cleavage process, or by complex aggregation through interaction with negatively charged polymers such as RNA.
Recently, α-synuclein (αS), an amyloid IDP implicated in PD and other neurodegenerative diseases collectively known as synucleinopathy17,18, has been demonstrated in cellular and animal models19,20 concentrated in protein condensates with fluid-like behavior. In vitro studies have shown that αS undergoes LLPS by simple aggregation through predominantly hydrophobic interactions, although this process requires exceptionally high protein concentrations and atypically long incubation times19,21. Whether the αS-containing condensates observed in vivo are formed by this or other LLPS processes remains a key unresolved issue. Similarly, although αS amyloid aggregation has been observed in neurons in PD and other synucleinopathies, the exact mechanism by which αS undergoes intracellular amyloid aggregation remains unclear, as overexpression of this protein does not appear to trigger this process by itself. Additional cellular damage is often required, suggesting that certain cellular locations or microenvironments are required for the renucleation of intracellular αS amyloid assemblies. One cellular environment that is particularly prone to aggregation may be the interior of protein condensates 23 .
Interestingly, αS and tau have been found to co-localize in characteristic disease inclusions in humans with Parkinson’s disease and other synucleinopathies 24,25 and experiments have reported a synergistic pathological relationship between the two proteins 26,27 suggesting a potential relationship between aggregation αS and tau in neurodegenerative diseases. illness. αS and tau have been found to interact and promote each other’s aggregation in vitro and in vivo 28,29 and heterogeneous aggregates composed of these two proteins have been observed in the brains of patients with synucleinopathies 30 . However, little is known about the molecular basis of the interaction between αS and tau and the mechanism of its co-aggregation. αS has been reported to interact with tau through an electrostatic attraction between the highly negatively charged C-terminal region of αS and the central proline-rich region of tau, which is also enriched in positively charged residues.
In this study, we show that αS can indeed dissociate into droplets via electrostatic complex condensation in the presence of tau protein, in contrast to its interaction with other positively charged polypeptides such as poly-L-lysine (pLK), and in this process . αS acts as a scaffold molecule for the droplet network. We have identified noticeable differences in the process of maturation of electrostatic αS coacervates, which are associated with differences in the valency and strength of the interaction of proteins involved in the coacervate network. Interestingly, we observed co-aggregation of αS and tau amyloid proteins in long-lived liquid coacervates and identified some key factors that lead to co-aggregation of these two proteins in such coacervates. Here we describe in detail this process, which is a possible molecular mechanism underlying the colocalization of two proteins in disease-specific inclusions.
αS has a highly anionic C-terminal tail at neutral pH (Fig. 1a), and we hypothesized that it could undergo LLPS through the condensation of electrostatic complexes with polycationic disordered polypeptide molecules. We used a 100-residue poly-L-lysine (pLK) as the starting model molecule due to its positively charged and disordered polymeric nature at neutral pH 32. First, we confirmed that pLK interacts with the Ct domain of αS via Solution NMR spectroscopy (Figure 1b) using 13C/15N-labeled αS in the presence of increasing αS:pLK molar ratios. The interaction of pLK with the Ct-domain of αS manifests itself in perturbations of the chemical shift and a decrease in the peak intensity in this region of the protein. Interestingly, when we mixed αS with pLK at an αS concentration of approx. 5–25 µM in the presence of polyethylene glycol (5–15% PEG-8) (typical LLPS buffer: 10 mM HEPES pH 7.4, 100 mM NaCl, 15% PEG-8) we immediately went through a wide field of protein formation. droplets were observed using fluorescence (WF) and bright-field (BF) microscopy (Fig. 1c). 1-5 µm droplets containing concentrated αS (added 1 µM AlexaFluor488-labeled αS, AF488-αS), their electrostatic properties can be derived from their resistance to 10% 1,6-hexanediol (1,6-HD) and its sensitivity to an increase in NaCl concentration (Fig. 1c). The fluid-like nature of the coacervates of the αS/pLK electrostatic complex is demonstrated by their ability to fuse within milliseconds (Fig. 1d). Using turbidimetry, we quantified the formation of droplets under these conditions, confirmed the electrostatic nature of the main interaction associated with its stability (Fig. 1e), and evaluated the effect of various polymer ratios on the LLPS process (Fig. 1f). Although droplet formation is observed over a wide range of polymer ratios, the process is very favorable when pLK is in excess of αS. LLPs have also been observed using the chemically different displacing agent dextran-70 (70 kDa) or using a variety of sample formats, including glass slide drops, microplate wells of various materials, Eppendorf or quartz capillaries.
a Schematic representation of different protein regions in the WT-αS and ΔCt-αS variants used in this study. The amphipathic N-terminal domain, the hydrophobic amyloid-forming (NAC) region, and the negatively charged C-terminal domain are shown in blue, orange, and red, respectively. The Net Charge Per Residual (NCPR) map of WT-αS is shown. b NMR analysis of the αS/pLK interaction in the absence of macromolecular clumps. As pLK concentration increases (αS:pLK molar ratios of 1:0.5, 1:1.5, and 1:10 are shown in light green, green, and dark green, respectively). c Coacervate αS/pLK (molar ratio 1:10) at 25 µM (1 µM AF488-labeled αS or Atto647N-labeled pLK for WF imaging) in LLPS buffer (top) or supplemented with 500 mM NaCl (bottom left) or after 10% 1,6-hexanediol (1,6-HD; bottom right). Scale bar = 20 µm. d Representative microscopic images of BF droplet fusion of αS/pLK (molar ratio 1:10) at a concentration of 25 μM; arrows indicate the merging of individual drops (red and yellow arrows) into a new drop (orange arrow) within 200 ms) . Scale bar = 20 µm. e Light scattering (at 350 nm) αS/pLK aggregation in LLPS buffer before and after addition of 500 mM NaCl or 10% 1,6-HD at 25 µM αS (N = 3 sample replicates, mean and standard deviation also indicated). f BF image (top) and light scattering analysis (at 350 nm, bottom) of αS/pLK aggregation at 25 μM αS with increasing αS:pLK molar ratio (N = 3 sample replicates, mean and standard deviation also indicated). Scale bar = 10 µm. The scale bar on one image indicates the scale of all images in one panel. Raw data is provided in the form of raw data files.
Based on our observations of αS/pLK electrostatic complex condensation and previous observations of αS as a client molecule of the tau/RNA condensate through direct interaction with tau31, we hypothesized that αS and tau could co-segregate with the solvent in the absence of RNA condensation. through electrostatic complexes, and αS is the scaffold protein in αS/Tau coacervates (see tau charge distribution in Figure 2e). We observed that when 10 μM αS and 10 μM Tau441 (containing 1 μM AF488-αS and 1 μM Atto647N-Tau, respectively) were mixed together in LLPS buffer, they easily formed protein aggregates containing both proteins, as seen by WF microscopy. (Fig. 2a). The colocalization of the two proteins in the droplets was confirmed by confocal (CF) microscopy (Supplementary Fig. 1a). Similar behavior was observed when dextran-70 was used as a aggregation agent (Supplementary Fig. 1c). Using either FITC-labeled PEG or dextran, we found that both crowding agents were evenly distributed across the samples, showing neither segregation nor association (Supplementary Fig. 1d). Rather, it suggests that in this system they promote phase separation through macromolecular crowding effects, since PEG is a preferentially stable crowding agent, as seen in other LLP systems33,34. These protein-rich droplets were sensitive to NaCl (1 M) but not to 1,6-HD (10% v/v), confirming their electrostatic properties (Supplementary Fig. 2a, b). Their fluid behavior was confirmed by observing millisecond merging droplet events using BF microscopy (Fig. 2b).
a Confocal (CF) microscopy images of αS/Tau441 coacervates in LLPS buffer (10 μM of each protein, 0.5 μM of AF488-labeled αS and Atto647N-labeled Tau441). b Representative differential interference contrast (DIC) images of αS/Tau441 droplet fusion events (10 μM for each protein). c Phase diagram based on light scattering (at 350 nm) of Tau441 LLPS (0–15 µM) in the absence (left) or presence (right) of 50 µM αS. Warmer colors indicate more scattering. d Light scattering of αS/Tau441 LLPS samples with increasing αS concentration (Tau441 at 5 µM, N = 2–3 sample repetitions as indicated). e Schematic representation of some tau protein variants and different regions of the protein used in this study: negatively charged N-terminal domain (red), proline-rich region (blue), microtubule-binding domain (MTBD, highlighted in orange), and amyloid-forming pair spiral. filament regions (PHF) located within the MTBD (gray). The Net Charge Per Residue (NCPR) map of Tau441 is shown. f Using 1 µM AF488-labeled αS and Atto647N-labeled ΔNt-, using 1 µM AF488-labeled αS or ΔCt-αS in the presence of ΔNt-Tau (top, 10 µM per protein) or K18 (bottom, 50 µM per protein ) ) ) micrographs of WF condensed in LLPS or K18 buffer. The scale bars in one image represent the scale of all images in one panel (20 µm for panels a, b and f). Raw data for panels c and d are provided as raw data files.
To test the role of αS in this LLPS process, we first investigated the effect of αS on droplet stability by nephelometry using increasing concentrations of NaCl (Fig. 2c). The higher the salt concentration in the samples containing αS, the higher the light scattering values ​​(at 350 nm), which indicates the stabilizing role of αS in this LLPS system. A similar effect can be observed by increasing the αS concentration (and hence the αS:Tau441 ratio) to approx. 10-fold increase relative to the tau concentration (5 µM) (Fig. 2d). To demonstrate that αS is a scaffold protein in coacervates, we decided to investigate the behavior of the LLPS-disrupted Tau mutant, which lacks a negatively charged N-terminal region (residues 1–150, see Fig. 2e) called ΔNt-Tau. WF microscopy and nephelometry confirmed that ΔNt-Tau itself did not undergo LLPS (Fig. 2f and Supplementary Fig. 2d), as previously reported 14. However, when αS was added to dispersion solutions of this truncated Tau variant, the LLPS process was completely restored with droplet density close to the droplet density of full-size solutions of Tau and αS under similar conditions and protein concentrations. This process can also be observed under conditions of low macromolecular crowding (Supplementary Fig. 2c). The role of the C-terminal αS region, but not its entire length, in the LLPS process was demonstrated by inhibition of droplet formation using a C-terminal truncated αS variant lacking residues 104–140 (Fig. 1a) of the (ΔCt-αS) protein (Fig. 2f and Supplementary Fig. 2d). The colocalization of αS and ΔNt-Tau was confirmed by confocal fluorescence microscopy (Supplementary Fig. 1b).
To further test the LLPS mechanism between Tau441 and αS, an additional Tau variant was used, namely the paired helical filament core (PHF) fragment in the microtubule-binding domain (MTBD), which if it contains four characteristic repeat domains, commonly also known as the K18 fragment ( see Fig. 2e). It has recently been reported that αS preferentially binds to a tau protein located in a proline-rich domain in a sequence that precedes the microtubule-binding domain. However, the PHF region is also rich in positively charged residues (see Figure 2e), especially lysine (15% residues), which prompted us to test whether this region also contributes to the condensation of the αS/Tau complex. We observed that K18 alone could not trigger LLPS at concentrations up to 100 μM under the conditions tested (LLPS buffer with 15% PEG or 20% dextran) (Figure 2f). However, when we added 50 µM αS to 50 µM K18, rapid formation of protein droplets containing K18 and αS was observed by nephelometry (Supplementary Fig. 2d) and WF microscopy (Fig. 2f). As expected, ΔCt-αS was unable to restore the LLPS behavior of K18 (Fig. 2f). We note that αS/K18 aggregation requires slightly higher protein concentrations to induce LLPS compared to αS/ΔNt-Tau or αS/Tau441, other things being equal. This is consistent with a stronger interaction of the αS C-terminal region with the proline-rich Tau domain compared to the microtubule-binding domain, as described previously 31 .
Given that ΔNt-Tau cannot perform LLPS in the absence of αS, we chose this Tau variant as a model for characterizing αS/Tau LLPS given its simplicity in LLPS systems with full-length Tau (isotype, Tau441/Tau441). with complex (heterotypic, αS/Tau441) aggregation processes. We compared the degree of αS aggregation (as part of the condensed phase protein, fαS,c) in the αS/Tau and αS/ΔNt-Tau systems by centrifugation and dispersed phase SDS-PAGE analysis (see 2e), found very similar values ​​for all proteins at the same concentration. In particular, we obtained fαS,c 84 ± 2% and 79 ± 7% for αS/Tau and αS/ΔNt-Tau, respectively, suggesting that the heterotypic interaction between αS and tau is superior to the interaction between tau molecules. between.
The interaction with various polycations and the effect of the condensation process on the αS kinetics were first studied by the fluorescence recovery after photobleaching (FRAP) method. We tested αS/Tau441, αS/ΔNt-Tau and αS/pLK coacervates (100 μM αS supplemented with 2 μM αS AF488-αS and 100 μM Tau441 or ΔNt-Tau or 1 mM pLK). Data were obtained within the first 30 minutes after mixing the sample components. From representative FRAP images (Fig. 3a, αS/Tau441 condensation) and their corresponding time course curves (Fig. 3b, Supplementary Fig. 3), it can be seen that αS kinetics are very similar to those of Tau441 coacervates. and ΔNt-Tau, which is much faster with pLK. The calculated diffusion coefficients for αS inside the coacervate according to FRAP (as described by Kang et al. 35) are D = 0.013 ± 0.009 µm2/s and D = 0.026 ± 0.008 µm2/s for αS/Tau441 and αS/ΔNt- for the αS/ system. pLK, Tau, and D = 0.18 ± 0.04 µm2/s, respectively (Fig. 3c). However, the diffusion coefficient αS in the dispersed phase is several orders of magnitude higher than all condensed phases, as determined by Fluorescence Correlation Spectroscopy (FCS, see Supplementary Fig. 3) under the same conditions (LLPS buffer) but in the absence of polycations ( D = 8 ± 4 µm2/s). Therefore, the kinetics of αS translation is significantly reduced in coacervates compared to proteins in the dispersed phase due to pronounced molecular crowding effects, although all coacervates retain liquid-like properties during the first half hour after their formation, in contrast to the tau phase. faster kinetics in pLK condensate.
a–c FRAP analysis of αS dynamics (2% AF488-labeled αS) in electrostatic coacervates. Representative images of αS/Tau441 FRAP assays in triplicate are shown in (a), where red circles indicate decolorized areas. The scale bar is 5 µm. b Average FRAP curves and (c) calculated diffusion coefficients (D) for 5–6 (N) different droplets from three experiments using 100 µM αS and equimolar concentrations of Tau441 (red) or ΔNt-Tau (blue) or pLK (green) at ten times the concentration of LLPS. The standard deviation of the FRAP curve is shown in shaded color. For comparison, the diffusion coefficient αS in the dispersed phase was determined in triplicate using fluorescence correlation spectroscopy (FCS) (see Supplementary Figure 3 and methods for more information). d Continuous X-band EPR spectra of 100 μM TEMPOL-122-αS in LLPS buffer without any polycation (black) or in the presence of 100 μM Tau441 (red) or ΔNt-Tau (blue) or 1 mM pLK (green). The inset shows a magnified view of the strong field lines where the most dramatic changes occur. e Binding curves of 50 μM TEMPOL-122-αS with various polycations in the absence of LLPS (no PEG). The reduced amplitude of band III compared to band II (IIII/III) of the normalized EPR spectrum is shown to increase the molar ratios of Tau441 (red), ΔNt-Tau (blue) and pLK (green). Colored lines show fit to data using a rough binding model with n identical and independent binding sites on each curve. Raw data are provided in the form of raw data files.
As a complement, we investigated the dynamics of αS in various coacervates using directed spin labeling (SDSL) and continuous electron paramagnetic resonance (CW-EPR). This method has proven to be very useful in reporting the flexibility and dynamics of the IDP with a realistic residual resolution36,37,38. To this end, we constructed cysteine ​​residues in single Cys mutants and used the 4-hydroxy-2,2,6,6-tetramethylpiperidine-N-oxyl (TEMPOL) spin probe. Maleimide derivatives label them. More specifically, we inserted TEMPOL probes at position 122 or 24 αS (TEMPOL-122-αS and TEMPOL-24-αS). In the first case, we target the C-terminal region of the protein, which is involved in interaction with polycations. Instead, position 24 can give us information about the overall dynamics of the proteins in the condensate. In both cases, the EPR signals obtained for proteins of the dispersed phase corresponded to nitroxide radicals in the fast moving state. After phase separation in the presence of tau or pLK (100 μM TEMPOL-αS, Tau441 or ΔNt-Tau at a ratio of 1:1 or pLK at a ratio of 1:10), an increase in the relative peak intensity was observed in the EPR spectrum of αS. The loss line broadened, indicating reduced αS reorientation kinetics in droplets compared to protein in dilute phase (Fig. 3d, Supplementary Fig. 4a). These changes are more pronounced at position 122. While at position 24 the presence of pLK did not affect the kinetics of the probe, at position 122 the spectral line shape changed significantly (Supplementary Fig. 4a). When we attempted to model the spectra at position 122 of two αS/polycation systems using the isotropic model (Supplementary Figure 5a) commonly used to describe the dynamics of spin-labeled IDP38,39, we were unable to reconstruct the experimental spectra. . Spectral simulation of the position of 24 spins contrasts (Supplementary Fig. 5a). This suggests that there are preferential positions in the space of spin configurations of the C-terminal region of αS in the presence of polycations. When considering the fraction of αS in the condensed phase under experimental EPR conditions (84 ± 2%, 79 ± 7%, and 47 ± 4% for αS/Tau441, αS/ΔNt-Tau, and αS/pLK, respectively—see Supplementary Fig. 2e of data analysis c), it can be seen that the broadening detected by the EPR method mainly reflects the interaction of the C-terminal region of αS with various polycations in the condensed phase (the main change when using TEMPOL-122-αS), and not protein condensation. An increase in microviscosity is observed in the probe. As expected, the EPR spectrum of the protein under conditions other than LLPS was completely restored when 1 M NaCl was added to the mixture (Supplementary Fig. 4b). Overall, our data suggest that the changes detected by CW-EPR mainly reflect the interaction of the C-terminal region of αS with various polycations in the condensed phase, and this interaction appears to be stronger with pLK than with Tau.
In order to obtain more structural information about the proteins in the coacervate, we decided to study the LLPS system using NMR in solution. However, we could only detect the αS fraction remaining in the dispersed phase, which may be due to reduced protein dynamics inside the coacervate and a dense phase at the bottom of the solution in NMR analysis. When we analyzed the structure and dynamics of the protein remaining in the dispersed phase of the LLPS sample using NMR (Supplementary Fig. 5c, d), we noticed that the protein behaved almost identically in the presence of pLK and ΔNt-Tau, both of which were in secondary structure and dynamics of the protein backbone, revealed by experiments on secondary chemical shift and R1ρ relaxation. NMR data show that the C-terminus of αS suffers a significant loss of conformational flexibility while retaining its disordered nature, like the rest of the protein sequence, due to its interactions with polycations.
Since the CW-EPR signal broadening observed in the TEMPOL-122-αS condensed phase reflects the interaction of the protein with polycations, we performed an EPR titration to evaluate the binding affinity of αS to various polycations in the absence of LLPS (no accumulation of Buffer LLPS), suggesting that the interactions are the same in dilute and concentrated phases (which is confirmed by our data, Supplementary Fig. 4a and Supplementary Fig. 6). The goal was to see if all coacervates, despite their common fluid-like properties, exhibit any underlying differential behavior at the molecular level. As expected, the EPR spectrum broadened with increasing polycation concentration, reflecting a decrease in molecular flexibility due to molecular interactions of all interaction partners almost to saturation (Fig. 3e, Supplementary Fig. 6). pLK achieved this saturation at a lower molar ratio (polycation:αS) compared to ΔNt-Tau and Tau441. In fact, comparison of the data with an approximate binding model assuming n identical and independent binding sites showed that the apparent dissociation constant of pLK (~5 μM) is an order of magnitude lower than that of Tau441 or ΔNt-Tau (~50 μM). µM). Although this is a rough estimate, this suggests that αS has a higher affinity for simpler polycations with continuous positive charge regions. Given this difference in affinity between αS and various polycations, we hypothesized that their liquid properties may change differently over time and thus suffer from different LSPT processes.
Given the highly crowded environment within the protein coacervate and the amyloid nature of the protein, we observed the behavior of the coacervate over time to detect possible LSPT processes. Using BF and CF microscopy (Figure 4), we observed that αS/Tau441 coacervates to a large extent in solution, forming large droplets that contact and wet the surface at the bottom of the well/slide as full droplets, as expected (Supplementary Fig. 7d); we call these bottom-formed structures “protein rafts”. These structures remained fluid as they retained the ability to fuse (Supplementary Fig. 7b) and could be seen for several hours after LLPS was triggered (Fig. 4 and Supplementary Fig. 7c). We observed that the wetting process is favored on the surface of hydrophilic rather than hydrophobic materials (Supplementary Fig. 7a), as expected for electrostatic coacervates with unbalanced charges and thus high electrostatic surface potentials. Notably, αS/ΔNt-Tau coalescence and rafting were significantly reduced, while αS/pLK condensates were significantly reduced (Fig. 4). During the short incubation time, the αS/pLK droplets were able to coalesce and wet the hydrophilic surface, but this process quickly stopped and after 5 hours of incubation, only limited coalescence events and no wetting were observed. – gel-drip transition.
Representative BF (greyscale panels) and CF (right panels, AF488-labeled αS in green) of coacervate samples containing 100 µM αS (1% fluorescent label) in LLPS buffer in the presence of 100 µM Tau441 (top) fluorescence) microscopic images ΔNt-Tau (center) or 1 mM pLK (bottom) at different incubation times and focal heights (z, distance from the bottom of the plate well). The experiments were repeated 4-6 times independently of each other with the same results. αS/Tau441 coacervates are wetted after 24 hours, forming rafts larger than the image. The scale bar for all images is 20 µm.
We then asked whether the large fluid-like protein pools formed in αS/Tau441 LLPS would lead to amyloid aggregation of any of the proteins studied. We followed the maturation of αS/Tau441 droplets over time with WF microscopy under the same conditions as above, but using 1 μM AF488-labeled αS and Atto647N-labeled Tau441 (Fig. 5a). As expected, we observed complete protein localization throughout the maturation process. Interestingly, from ca. After 5 hours, more intense non-circular structures were observed inside the rafts, which we called “points”, some of which were colocalized with αS, and some were enriched in Tau441 (Fig. 5a, white arrows). These spots have always been observed within rafts to a greater extent for αS/ΔNt-Tau than for αS/ΔNt-Tau. There were no distinct spots in the droplets of pLK and Tau systems incompetent for fusion/wetting. To test whether these stains containing αS and Tau441 are amyloid-like aggregates, we performed a similar experiment using CF microscopy in which Tau441 was labeled with Atto647N and 12.5 μM amyloid-specific thioflavin-T (ThT) was added from the start. dye. Although ThT-staining of αS/Tau441 droplets or rafts was not observed even after 24 h of incubation (Fig. 5b, top row—remaining droplets over protein rafts), ThT-positive structures containing Atto647N-Tau441 inside the rafts were very weak. this replicates the size, shape, and location of the previously described spots (Fig. 5b, middle and bottom rows), suggesting that the spots may correspond to amyloid-like aggregates formed in aging fluid coacervates.
WF 25 μM αS at various incubation times and focal heights (z, distance from unbound bottom) in the presence of 25 μM Tau441 (1 μM AF488-labeled αS and Atto647N-labeled Tau441) in a well of a microscope plate with LLPS buffer). Six experiments were independently repeated with similar results. b CF microscopic image of 25 μM αS in the presence of 25 μM Tau441 (1 μM Atto647N-labeled Tau441) and 12.5 μM thioflavin-T (ThT). Weighted protein droplets and deposited protein rafts and spots are shown in the top and middle rows, respectively. The bottom row shows images of rafts and drops from 3 independent replicates. White arrows indicate ThT-positive dots in both panels. The scale bar for all images is 20 µm.
To examine in more detail the changes in the coacervate protein network during the transition from liquid to solid, we used fluorescence lifetime imaging (FLIM) and Förster resonance energy transfer microscopy (FRET) (Figure 6 and Supplementary Figures 8 and 9). We hypothesized that the coacervate maturation of the layer into a more condensed or even solid-like aggregated protein structure leads to closer contact between the protein and the fluorescent probe attached to it, potentially producing a quenching effect manifested in a shortened probe lifetime (τ), as described previously40. ,41 ,42. In addition, for double labeled samples (AF488 and Atto647N as FRET donor and acceptor dyes, respectively), this decrease in τ can also be accompanied by coacervate condensation and an increase in FRET(E) efficiency during LSPT. We monitored raft and spot formation over time in LLPS αS/Tau441 and αS/ΔNt-Tau samples (25 µM of each protein in LLPS buffer containing 1 µM AF488 labeled αS and/or Atto647N labeled Tau441 or ΔNt-Tau). We observed a general trend in that the fluorescence lifetime of the AF488 (τ488) and Atto647N (τ647N) probes decreased slightly as the coacervates matured (Fig. 6 and Supplementary Fig. 8c). Interestingly, this change was significantly enhanced for dots within rafts (Fig. 6c), indicating that further protein condensation occurred at dots. In support of this, no significant change in fluorescence lifetime was observed for αS/ΔNt-Tau droplets aged for 24 h (Supplementary Fig. 8d), suggesting that droplet gelation is a process distinct from spotting and that is not accompanied by significant molecular reorganization within coacervates. It should be noted that the dots have different sizes and variable content in αS, especially for the αS/Tau441 system (Supplementary Fig. 8e). The decrease in spot fluorescence lifetime was accompanied by an increase in intensity, especially for Atto647N labeled Tau441 (Supplementary Fig. 8a), and higher FRET efficiencies for both αS/Tau441 and αS/ΔNt-Tau systems, indicating further condensation in LLPS Five hours after triggering, the proteins inside the static electricity condensed. Compared to αS/ΔNt-Tau, we observed lower τ647N and somewhat higher τ488 values ​​in αS/Tau441 spots, accompanied by lower and more inhomogeneous FRET values. Possibly, this may be related to the fact that in the αS/Tau441 system, the observed and expected αS abundance in aggregates is more heterogeneous, often substoichiometric compared to Tau, since Tau441 itself may also undergo LLPS and aggregation (Supplementary Fig. 8e). However, the degree of droplet coalescence, raft formation, and, importantly, protein aggregation within liquid-like coacervates is maximal when both Tau441 and αS are present.
a Lifetime fluorescence microscopy (FLIM) images of αS/Tau441 and αS/ΔNt-Tau at 25 μM of each protein (1 μM AF488-labeled αS and 1 μM Atto647N-labeled Tau441 or ΔNt-Tau) in LLPS buffer. The columns show representative images of LLPS samples at different maturation times (30 min, 5 h and 24 h). The red frame shows the region containing αS/Tau441 spots. Life spans are shown as color bars. Scale bar = 20 µm for all images. b Zoomed-in FLIM image of the selected area, shown in the red box in panel a. Life ranges are shown using the same color scale as in panel a. Scale bar = 5 µm. c Histograms showing AF488 (attached to αS) or Atto647N (attached to Tau) for different protein species (droplets-D-, raft-R- and speckle-P) identified in FLIM images recorded for αS-) timing distributions lifetime of Tau441 and αS/ΔNt-Tau coacervate samples (N = 17-32 ROI for D, 29-44 ROI for R, and 21-51 ROI for points). Mean and median values ​​are shown as yellow squares and black lines inside boxes, respectively. The lower and upper bounds of the box represent the first and third quartiles, respectively, and the minimum and maximum values ​​within the 1.5-fold interquartile range (IQR) are shown as whiskers. Outliers are shown as black diamonds. Statistical significance between pairs of distributions was determined using a two-sample t-test, assuming unequal variances. Two-tailed t-test p-values ​​are shown with asterisks for each pair of compared data (* p-value > 0.01, ** p-value > 0.001, *** p-value > 0.0001, **** p-value > 0.00001), ns Indicates negligibility (p-value > 0.05). The exact p values ​​are given in Supplementary Table 1, and the original data are presented as raw data files.
To further demonstrate the amyloid-like nature of speckles/aggregates, we treated unstained coacervate samples for 24 hours with high concentrations of (1 M) NaCl, which resulted in the separation of aggregates from protein coacervates. When isolated aggregates (i.e., a dispersed solution of aggregates) were observed using atomic force microscopy (AFM), we observed a predominantly spherical morphology with a regular height of about 15 nm, which tends to associate under conditions of high salt concentration, similar to the behavior of typical amyloid fibrils due to the strong hydrophobic effect on the surface (note that fibrils typically have a height of ~10 nm) (Supplementary Fig. 10a). Interestingly, when isolated aggregates were incubated with ThT in a standard ThT fluorescence assay, we observed a dramatic increase in ThT fluorescence quantum yield, comparable to that observed when the dye was incubated with typical αS amyloid fibrils (Supplementary Fig. 10b), suggesting that the coacervate aggregates contain amyloid-like structures. . In fact, the aggregates were tolerant to high salt concentrations but sensitive to 4 M guanidine chloride (GdnHCl), like typical amyloid fibrils (Supplementary Fig. 10c).
Next, we analyzed the composition of the aggregates using single molecule fluorescence, specific fluorescence correlation/cross-correlation spectroscopy (FCS/FCCS), and burst analysis of two-color coincidence detection (TCCD). To this end, we isolated aggregates formed after 24 hours of incubation in 100 μl LLPS samples containing αS and Tau441 (both 25 μM) together with 1 μM AF488-labeled αS and 1 μM Atto647N-labeled Tau441. Dilute the resulting dispersed aggregate solution to a monomolecular state using the same PEG-free buffer and 1 M NaCl (the same buffer used to separate aggregates from coacervate) to prevent any possible electrostatic interactions between LLPS and protein. An example of the time trajectory of a single molecule can be seen in Fig. 7a. FCCS/FCS analysis (cross-correlation, CC and autocorrelation, AC) showed that aggregates containing αS and tau were abundant in the samples (see CC curve in Fig. 7b, left panel), and an excess of residual monomeric protein arose as a result of the dilution process (see AC curves in Figure 7b, left panel). Control experiments performed under the same solution conditions using samples containing only monomeric proteins showed no CC curves, and the AC curves fit well with the one-component diffusion model (Eq. 4), where monomeric proteins have the expected diffusion coefficients (Fig. 7b), right panel). The diffusion coefficient of aggregated particles is less than 1 µm2/s, and that of monomeric proteins is about 1 µm2/s. 50–100 µm/s; values ​​are similar to previously published values ​​for sonicated αS amyloid fibrils and monomeric αS separately under similar solution conditions44. When we analyzed the aggregates with TCCD explosion analysis (Fig. 7c, top panel), we found that in each isolated aggregate (αS/Tau heteroaggregate), about 60% of the detected aggregates contained both αS and tau, about 30% contained only tau, about 10% αS only. Stoichiometric analysis of αS/Tau heteroaggregates showed that most of the heteroaggregates were enriched in tau (stoichiometry below 0.5, the average number of tau molecules per aggregate is 4 times more than αS molecules), which is consistent with our work observed in FLIM in situ experiments. . FRET analysis showed that these aggregates contained both proteins, although the actual FRET values ​​in this case are not of great importance, since the distribution of fluorophores in each aggregate was random due to an excess of unlabeled protein used in the experiment. Interestingly, when we performed the same analysis using the 45,46 mature amyloid aggregation-deficient Tau variant (see Supplementary Fig. 11a,b), we noticed that although the αS electrostatic aggregation was the same (Supplementary Fig. 11c, d), the ability to form aggregates within the coacervate was drastically reduced and FLIM detected several spots in in situ experiments, and weak cross-correlation curves were observed for isolated aggregate samples. However, for a small number of detected aggregates (only one tenth of Tau441), we observed that each aggregate was enriched in αS than this Tau variant, with approximately 50% of the detected aggregates containing only αS molecules, and αS was heterogeneous in excess. aggregates (see Supplementary Fig. 11e), in contrast to the heterogeneous aggregates generated by Tau441 (Fig. 6f). The results of these experiments showed that although αS itself is capable of accumulating with tau within the coacervate, tau nucleation is more favorable under these conditions, and the resulting amyloid-like aggregates are able to act as a form of αS and tau. However, once a tau-rich core is formed, heterotypic interactions between αS and tau are favored in aggregates over homotypic interactions between tau molecules; we also observe protein networks in liquid αS/tau coacervates.
a Representative fluorescence temporal traces of single molecules of isolated aggregates formed in αS/Tau441 electrostatic coacervates. Bursts corresponding to αS/Tau441 coaggregates (bursts above the indicated threshold) were observed in three detection channels (AF488 and Atto647N emission after direct excitation, blue and red lines, Atto647N emission after indirect excitation), FRET, violet line). b FCS/FCCS analysis of a sample of isolated αS/Tau441 aggregates obtained from LLPS (left panel). Autocorrelation (AC) curves for AF488 and Atto647N are shown in blue and red, respectively, and cross-correlation (CC) curves associated with aggregates containing both dyes are shown in purple. The AC curves reflect the presence of labeled monomeric and aggregated protein species, while the CC curves show only the diffusion of double-labeled aggregates. The same analysis, but under the same solution conditions as in isolated spots, samples containing only monomeric αS and Tau441 are shown as controls in the right panel. c Fluorescence flash analysis of single molecules of isolated aggregates formed in αS/Tau441 electrostatic coacervates. Information for each aggregate found in four different repeats (N = 152) is plotted against their stoichiometry, S values, and FRET efficiency (top panel, color bar reflects occurrence). Three types of aggregates can be distinguished: -αS-only aggregates with S~1 and FRET~0, Tau-only aggregates with S~0 and FRET~1, and heterogeneous Tau/αS aggregates with intermediate S and FRET Estimates of the amount of both marker proteins detected in each heterogeneous aggregate (N = 100) are shown in the lower panel (the color scale reflects the occurrence). Raw data are provided in the form of raw data files.
The maturation or aging of liquid protein condensates into gel-like or solid structures over time has been reported to be involved in several physiological functions of the condensate47 as well as in disease, as an abnormal process preceding amyloid aggregation 7, 48, 49. Here we study phase separation and behavior in detail. LSPT αS in the presence of random polycations in a controlled environment at low micromolar concentrations and physiologically relevant conditions (note that the calculated physiological concentration of αS is >1 µM50), following typical thermodynamically driven behavior of LPS. We found that αS, which contains a highly negatively charged C-terminal region at physiological pH, is able to form protein-rich droplets in aqueous solution via LLPS in the presence of highly cationic disordered peptides such as pLK or Tau through the process of electrostatic complex condensation in the presence of aggregation macromolecules. This process may have relevant effects in the cellular environment where αS encounters various polycationic molecules associated with its disease-associated aggregation both in vitro and in vivo51,52,53,54.
In many studies, protein dynamics within droplets have been considered as one of the key factors determining the maturation process55,56. In electrostatic αS coacervates with polycations, the maturation process apparently depends on the strength of interactions with polycations, the valence, and the multiplicity of these interactions. Equilibrium theory suggests that an equilibrium landscape of two liquid states would be the presence of a large droplet rich in biopolymers that drive LLPS57,58. Droplet growth can be achieved by Ostwald maturation59, coalescence60 or consumption of free monomer in the dispersed phase61. For αS and Tau441, ΔNt-Tau or pLK, most of the protein was concentrated in the condensate under the conditions used in this study. However, while full-size tau droplets coalesced rapidly upon surface wetting, droplet coalescence and wetting were difficult for ΔNt-Tau and pLK, suggesting a rapid loss of liquid properties in these two systems. According to our FLIM-FRET analysis, the aged pLK and ΔNt-Tau droplets showed a similar degree of protein aggregation (similar fluorescence lifetime) as the original droplets, suggesting that the original protein network was retained, albeit more rigid.
We rationalize our experimental results in the following model (Figure 8). The initially temporarily formed droplets are often protein networks without electrostatic compensation, and thus there are areas of charge imbalance, especially at the droplet interface, resulting in droplets with a high electrostatic surface potential. To compensate for charge (a phenomenon commonly referred to as valence depletion) and minimize the surface potential of the droplets, the droplets can include new polypeptides from the dilute phase, reorganize protein networks to optimize charge-charge interactions, and interact with other droplets. with surfaces (wetting). The αS/pLK droplets, due to their simpler protein network (only heterotypic interactions between αS and pLK) and greater affinity for protein-protein interactions, seem to be able to balance the charge of the condensate more quickly; indeed, we observed faster protein kinetics in initially formed αS/pLK coacervates than in αS/Tau. After valence depletion, the interactions become less ephemeral and the droplets lose their liquid properties and turn into gel-like, non-flammable droplets with a low electrostatic surface potential (and therefore unable to wet the surface). In contrast, αS/Tau droplets are less efficient at optimizing droplet charge balance due to more complex protein networks (with both homotypic and heterotypic interactions) and the weaker nature of protein interactions. This results in droplets that retain liquid behavior for extended periods of time and exhibit a high electrostatic surface potential that tends to be minimized by coalescing and growing (thus minimizing the surface area/volume ratio of the droplets) and by wetting the hydrophilic surface chem. This creates large concentrated protein libraries that retain fluid properties as the interactions remain very transient due to the constant search for charge optimization in the protein network. Interestingly, N-terminally truncated forms of Tau, including some naturally occurring isoforms62, exhibit intermediate behavior, with some coacervates aging with αS into long-lived gel-like droplets, while others transform into large liquid condensates. This duality in the maturation of αS electrostatic coacervates is consistent with recent LLPS theoretical and experimental studies that have identified a correlation between valence depletion and electrostatic sieving in condensates as a key to controlling condensate size and fluid properties. Mechanism 58.61.
This scheme shows the putative amyloid aggregation pathway for αS and Tau441 via LLPS and LSPT. With additional anion-rich (red) and cation-rich (blue) regions, αS and tau electrostatic coacervates with satisfactory valence have lower surface energy and therefore less coalescence, resulting in rapid droplet aging. A stable non-agglomerated gel state is achieved. . This situation is very favorable in the case of the αS/pLK system due to its higher affinity and simpler protein-pair interaction network, which allows for a fast gel-like transition. On the contrary, droplets with unsatisfactory valence and, therefore, protein-charged regions available for interaction, make it easier for the coacervate to fuse and wet the hydrophilic surface in order to reduce its high surface energy. This situation is preferable for αS/Tau441 coacervates, which have a multivalent complex network consisting of weak Tau-Tau and αS-Tau interactions. In turn, larger coacervates will more readily retain their fluid-like properties, allowing other protein-to-protein interactions to occur. Eventually, amyloid heterogeneous aggregates containing both αS and tau form within the coacervate fluid, which may be related to those found in inclusion bodies, which are hallmarks of neurodegenerative diseases.
The large fluid-like structures formed during maturation of αS/Tau441 with a highly congested but dynamic protein environment and, to a lesser extent, αS/ΔNt-Tau coacervates are ideal reservoirs for the nucleation of protein aggregation. We have indeed observed the formation of solid protein aggregates in this type of protein coacervates, often containing both αS and tau. We have shown that these heteroaggregates are stabilized by non-electrostatic interactions, are able to bind amyloid-specific ThT dyes in the same way as typical amyloid fibrils, and indeed have similar resistance to various influences. The αS/tau aggregates formed by LLPS were shown to have amyloid-like properties. Indeed, the mature variant of Tau deficient in amyloid aggregation is significantly impaired in the formation of these heterogeneous αS aggregates within the liquid electrostatic coacervate. The formation of αS/Tau441 aggregates was observed only inside the coacervates, which retained liquid-like properties, and never, if the coacervates/droplets did not reach the gel state. In the latter case, the increased strength of electrostatic interactions and, as a result, the rigidity of the protein network prevent the necessary conformational rearrangements of proteins to establish new protein interactions necessary for amyloid nucleation. However, this can be achieved in more flexible, liquid-like coacervates, which in turn are more likely to remain liquid as they increase in size.
The fact that the formation of aggregates within the condensed phase is preferable in large αS/Tau condensates than in small droplets that rapidly gel, highlights the relevance of identifying the factors that control droplet coalescence. Thus, not only is there a tendency for phase separation, but the size of the condensate must be controlled for proper functioning as well as disease prevention58,61. Our results also highlight the importance of the balance between LLPS and LSPT for the αS/Tau system. While droplet formation may protect against amyloid aggregation by reducing the amount of protein monomers available under saturation conditions, as has been proposed in other systems63,64, droplet fusion at high droplet levels may lead to internal protein aggregation through slow conformational rearrangements. protein networks. .
Overall, our data strongly emphasize the relevance of cohesive valence and satisfied/unsatisfied interactions in drop networks in the context of LSPT. In particular, we show that full-length αS/Tau441 condensates are able to efficiently fuse and nucleate to form amyloid-like heteroaggregates that include both proteins and propose a molecular mechanism based on our experimental results. The co-aggregation of two proteins in the αS/Tau fluid coacervate that we report here may indeed be related to the co-localization of two proteins in inclusions, which are hallmarks of the disease, and may contribute to understanding the relationship between LLPS and amyloid aggregation, paving the way for highly charged IDP in neurodegeneration.
Monomeric WT-αS, cysteine ​​mutants (Q24C-αS, N122C-αS) and ΔCt-αS variants (Δ101-140) were expressed in E. coli and purified as previously described. 5 mM DTT was included in all steps in the purification of αS cysteine ​​mutants to prevent disulfide bond formation. Tau441 isoform (plasmid obtained from Addgene #16316), ΔNt-Tau variant (Δ1–150, obtained by cloning IVA with primers CTTTAAGAAGGAGATACATATGATCGCCACACCGCGG, CATATGTATATCCTCTCTTCTTAAAGTTAAAC) and AggDef-Tau variant (Δ275–311, purified with GGCTC5 primer) E. coli cultures were grown to OD600 = 0.6–0.7 at 37°C and 180 rpm, and expression was induced with IPTG for 3 hours at 37°C. Harvest cells at 11,500 x g for 15 min at 4 °C and wash with saline buffer containing 150 mM NaCl. Resuspend the pellet in lysis buffer (20 ml per 1 L LB: MES 20 mM, pH 6.8, NaCl 500 mM, EDTA 1 mM, MgCl2 0.2 mM, DTT 5 mM, PMSF 1 mM, benzamidine 50 μM, copeptin 100 μM). The sonication step was performed on ice with an amplitude of 80% for 10 pulses (1 min on, 1 min off). Do not exceed 60 ml in one ultrasound. E. coli lysates were heated at 95° C. for 20 minutes, then cooled on ice and centrifuged at 127,000×g for 40 minutes. The clarified supernatant was applied to a 3.5 kDa membrane (Spectrum™ Thermo Fisher Scientific, UK) and dialyzed against 4 L of dialysis buffer (20 mM MES, pH 6.8, NaCl 50 mM, EDTA 1 mM, MgCl2 2 mM, DTT 2 mM, PMSF 0.1 mM) for 10 hours. A 5 ml cation exchange column (HiTrap SPFF, Cytiva, MA, USA) was equilibrated with equilibration buffer (20 mM MES, pH 6.8, 50 mM NaCl, 1 mM EDTA, 2 mM MgCl2, 2 mM DTT, 0.1 mM PMSF) . The tau lysate was filtered through a 0.22 μm PVDF filter and injected into the column at a flow rate of 1 ml/min. Elution was carried out gradually, tau was eluted with 15–30% elution buffer (20 mM MES, pH 6.8, 1 M NaCl, 1 mM EDTA, 2 mM MgCl2, 2 mM DTT, 0.1 mM PMSF). Fractions were analyzed by SDS-PAGE, and any fractions containing one band with the expected molecular weight of tau were concentrated using a 10 kDa centrifuge filter and replaced with a buffer containing 10 mM HEPES, pH 7.4, NaCl 500 mM and DTT 2 mM for the final protein concentration was 100 μM. The protein solution was then passed through a 0.22 μm PVDF filter, quickly frozen and stored at -80°C. Protein K18 was kindly provided by Prof. Alberto Boffi. The purity of the preparation was >95% as confirmed by SDS-PAGE and MALDI-TOF/TOF. Various cysteines were chemically labeled with AlexaFluor488-maleimide (AF488, ThermoFisher Scientific, Waltham, MA, USA) or TEMPOL-maleimide (Toronto Research Chemicals, Toronto, Canada). were confirmed by absorbance and MALDI-TOF/TOF. Tau441, ΔNt-Tau, AggDef-Tau and K18 were labeled with native cysteine ​​residues at positions 191 and 322 using Atto647N-maleimide (ATTO-TEC GmbH, Siegen, Germany) following the same procedure. Net charge per residue maps for αS and Tau441 were generated using CIDER66.
Solid poly-L-lysine (pLK DP 90-110 according to NMR from supplier, Alamanda Polymers Inc, Huntsville, Alabama, USA) was dissolved in 10 mM HEPES, 100 mM NaCl, pH 7.4 to 10 mM concentration, process sonicated for 5 minutes in an ultrasonic water bath and store at -20°C. PEG-8, dextran-70, FITC-PEG-10 (Biochempeg, Watertown, MA, USA) and FITC-dextran-500 (Sigma -Aldrich, Sant Louis, MI, USA) are water soluble and widely distributed in LLPS buffer. Dialysis removes contaminating salts. They were then filtered through a syringe filter with a pore size of 0.22 μm, and their concentrations were calculated using a refractometer (Mettler Toledo, Columbus, Ohio, USA). LLPS samples were prepared at room temperature in the following order: buffer and extrusion were mixed and 1 mM tris(2-carboxyethyl)phosphine (TCEP, Carbosynth, Compton, UK), 1 mM 2,2,2,2-(Ethane-1, 2-diyldinitrile) tetraacetic acid (EDTA, carboxynth) and a mixture of 1% protease inhibitor (PMSF 100 mM, benzimide 1 mM, leupeptin 5 μM). Then αS and fused polycations (options pLK or Tau) are added. For thioflavin-T time series experiments (ThT, Carbosynth, Compton, UK), use the total ThT concentration to be half the αS concentration. Gently but thoroughly mix the samples to ensure they are homogeneous. The concentration of each component varied from experiment to experiment, as described in the Results section. Azide was used at a concentration of 0.02% (w/v) whenever the duration of the experiment exceeded 4 hours. For all analyzes using LLPS samples, allow the mixture to equilibrate for 5 minutes prior to analysis. For light scattering analysis, 150 µl of samples were loaded onto non-binding 96-well microplates (µClear®, black, F-Bottom/Chimney Well, Greiner bio-one, Kremsmünster, Austria) and covered with adhesive film. LLPs were monitored by measuring absorbance at 350 nm at the center of the solution in a CLARIOstar plate reader (BMG Labtech, Ortenberg, Germany). The experiments were carried out in triplicate at 25°C, and the errors were calculated as the standard deviation from the mean. The dilute phase was quantified by sample centrifugation and SDS-PAGE gel analysis, and the αS fraction in the dilute and concentrated phases was quantified in various LLPS solutions. A 100 μl LLPS sample containing 1 μM AF488-labeled αS was prepared by thorough mixing followed by centrifugation at 9600×g for 30 minutes, after which the precipitate was usually visible. The top 50 μl of the supernatant was used for protein quantification using SDS-PAGE gel. Gels were scanned with AF488 filters using a ChemiDoc gel imaging system (Bio-Rad Laboratories, Hercules, CA, USA) or stained with Coomassie stain and visualized with appropriate filters. The resulting bands were analyzed using ImageJ version 1.53i (National Institutes of Health, USA). The experiments were carried out in duplicate in two different experiments with similar results.
Typically, 150 μl of samples were applied to non-binding 96-well microplates and visualized at room temperature on a Leica DMI6000B inverted microscope (Leica Microsystems, Wetzlar, Germany). For spot experiments, µ-Slide Angiogenesis plates (Ibidi GmbH, Gräfelfing, Germany) or 96-well polystyrene microplates (Corning Costar Corp., Acton, Massachusetts) were also used. EL6000 halogen or mercury metal halide lamps were used as illumination sources (for BF/DIC and WF imaging, respectively). For WF microscopy, a 40x magnification air objective (Leica Microsystems, Germany) was used to focus the light on the sample and collect it. For AF488 and ThT labeled samples, filter excitation and emission with standard GFP filter sets, excitation and emission bandpass filters, respectively, 460–500 nm and 512–542 nm bandpass filters, and a 495 nm dichroic mirror. For samples labeled with Atto647N, a standard set of Cy5 filters with excitation and emission bandpass filters 628–40 nm and 692–40 nm, respectively, and a 660 nm dichroic mirror were used. For BF and DIC microscopy, use the same reflected light collection objective. The collected light was recorded on a Leica DFC7000 CCD camera (Leica Microsystems, Germany). The exposure time was 50 ms for BF and DIC microscopy imaging and 20-100 ms for WF microscopy imaging. For comparison, the exposure time for all experiments with ThT was 100 ms. Time-lapse experiments were performed to visualize droplet coalescence, with images being collected every 100 ms for several minutes. ImageJ (NIH, USA) was used for image analysis. The experiments were carried out in triplicate with similar results.
For colocalization experiments, FRAP and 3D reconstruction, images were acquired on a Zeiss LSM 880 inverted confocal microscope using a ZEN 2 blue edition (Carl Zeiss AG, Oberkochen, Germany). Samples of 50 µl were applied to µ-Slide Angiogenesis Petri dishes (Ibidi GmbH, Gröfelfing, Germany), treated with a hydrophilic polymer (ibiTreat) and mounted in a 63× oil immersion objective (Plan-Apochromat 63×/NA 1.4 Oil) on DIC). Images were acquired using 458 nm, 488 nm, and 633 nm argon laser lines with a resolution of 0.26 µm/pixel and an exposure time of 8 µs/pixel for excitation and emission detection windows of 470–600 nm, 493–628 nm, and 638–755 nm was used to visualize ThT, AF488 and Atto647N, respectively. For the FRAP experiments, time-lapse photography of each sample was recorded at 1 frame per second. The experiments were carried out in triplicate at room temperature with similar results. All images were analyzed using Zen 2 blue edition software (Carl Zeiss AG, Oberkochen, Germany). The FRAP curves were normalized, plotted and fitted to intensity/time data extracted from images using Zen 2 using OriginPro 9.1. The recovery curves were fitted to a mono-exponential model to account for molecular diffusion with an additional exponential term to account for the acquisition bleaching effect. We then calculated D using the nominal bleaching radius and the previously determined recovery half-life as in the equation of Kang et al. 5 35 shown.
Single cysteine ​​variants of αS were synthesized with 4-hydroxy-2,2,6,6-tetramethylpiperidine-N-oxyl (TEMPOL) at positions 24 (TEMPOL-24-αS) and 122 (TEMPOL-122-αS), respectively. Spin Labeling For EPR experiments, the αS concentration was set at 100 μM and the PEG concentration was 15% (w/v). For various aggregation conditions, the αS:pLK ratio was 1:10, while the αS:ΔNt-Tau and αS:Tau441 ratios were maintained at 1:1. For binding titration experiments in the absence of crowding, TEMPOL-122-αS was maintained at 50 μM and polycations were titrated at increasing concentrations, preparing each condition separately. CW-EPR measurements were carried out using a Bruker ELEXSYS E580 X-band spectrometer equipped with a Bruker ER4118 SPT-N1 resonator operating at a microwave (SHF) frequency of ~9.7 GHz. The temperature was set at 25°C and controlled by a liquid nitrogen cryostat. The spectra were obtained under unsaturated conditions at a MW power of 4 mW, a modulation amplitude of 0.1 mT, and a modulation frequency of 100 kHz. Spectral intensities were normalized to avoid differences in spin concentrations between samples and possible spin reduction due to residual concentrations of reducing agents in samples containing Tau441 or ΔNt-Tau (present in the original protein solutions). The given values ​​of g were obtained as a result of EPR spectral modeling performed using the Easyspin software (v. 6.0.0-dev.34) implemented in Matlab®67. One/two component isotropic models were used to model the data. After normalizing all signals, the residuals were calculated by subtracting each simulation from the corresponding experimental spectrum. For binding titration analysis, the relative intensity of the third band to the second band of the normalized EPR spectrum (IIII/III) was used to monitor the binding of polycations to αS. To estimate the dissociation constant (Kd), the resulting curve was fitted to an approximate model assuming n identical and independent binding sites.
NMR spectroscopy experiments were carried out using a Bruker Neo 800 MHz (1H) NMR spectrometer equipped with a cryoprobe and Z-gradient. All experiments were performed using 130–207 µM αS and corresponding αS/ΔNt-Tau and pLK equivalents in 10 mM HEPES, 100 mM NaCl, 10% DO, pH 7.4 and were performed at 15°C. To monitor LPS by NMR, 10% PEG was added to the pre-mixed samples. The chemical shift perturbation plot (Fig. 1b) shows the average 1H and 15N chemical shifts. The αS 2D1H-15N HSQC spectra were assigned based on a previous assignment (BMRB entry #25227) and confirmed by recording and analyzing the 3D spectra of HNCA, HNCO and CBCAcoNH. 13Cα and 13Cβ chemical shifts were calculated in the presence of ΔNt-Tau or pLK to measure possible changes in secondary structure trends compared to αS chemical shifts in pure random coil conformation 68 (Supplementary Figure 5c). The R1ρ rates were measured by recording hsqctretf3gpsi experiments (obtained from the Bruker library) with delays of 8, 36, 76, 100, 156, 250, 400, and 800 ms, and the exponential functions were adjusted to the peak intensity delays at different times to determine the R1ρ and its experimental uncertainty.
Two-color time-resolved fluorescence microscopy experiments were performed on a commercial time-resolved MT200 fluorescence confocal microscope (PicoQuant, Berlin, Germany) with a time-correlated single photon counting (TCSPC) device. The laser diode head is used for pulsed interleaved excitation (PIE), the beam passes through a single mode waveguide and is tuned to a laser power of 10 to 100 nW for 481 nm and 637 nm laser lines measured after a dichroic mirror. This ensures an optimal photon counting rate, avoiding the effects of photon aliasing, photobleaching and saturation. μ-Slide angiogenesis coverslips or plates (Ibidi GmbH, Gräfelfing, Germany) were placed directly in immersion water over a Super Apochromat 60x NA 1.2 lens with a corrective collar (Olympus Life Sciences, Waltham, USA). A 488/640 nm dichroic mirror (Semrock, Lake Forest, IL, USA) was used as the main beam splitter. Unfocused radiation is blocked by a hole with a diameter of 50 microns, then the focused radiation is divided into 2 detection paths by a 50/50 beam splitter. Bandpass emission filters (Semrock, Lake Forest, IL, USA) 520/35 for green dye (AF488) and 690/70 for red dye (Atto647N) were used in front of the detector. Single-photon avalanche diodes (SPAD) (Micro Photon Devices, Bolzano, Italy) were used as detectors. Both data collection and analysis were performed using the commercially available SymphoTime64 software (PicoQuant GmbH, Berlin, Germany).
Fifty microliters of LLPS samples were applied to μ-Slide angiogenesis wells (Ibidi GmbH, Gräfelfing, Germany). The resulting images are focused to 20 µm above the well bottom for optimal objective working distance for suspended droplets and to ~1 µm for rafts and dots with an axial resolution of at least 0.25 µm/pixel and a delay time of 400 µs/pixel. Select data by applying an intensity threshold based on the average background signal intensity (PBG, mean + 2σ) for each channel so that only liquid protein droplets, rafts, or spots are selected, filtering out any possible origin from the dispersed phase. To analyze the lifespan of each species (τ) of each channel (green, “g” for AF488 and red, “r” for Atto647N), we selected regions of interest (ROIs) containing droplets, rafts, or spots (Supplementary Figure 1). 8b) and derived them by fitting their lifetime decay (τD, τR and τP for droplets, rafts or spots, respectively, see Supplementary Fig. 8c) in each channel using a tail-fit analysis and a two-component decay model. Average τ from τ . ROIs that produced too few photons for a multi-exponential fit were excluded from the analysis. The cutoff used was <104 photons for rafts and dots and 103 for drops. The droplets have a lower threshold because it is difficult to obtain decay curves with higher intensity values, since the droplets in the image field are usually smaller and less numerous. ROIs with photon counts above the photon accumulation limit (set to >500 counts/pixel) were also discarded for analysis. Match the intensity decay curve obtained from the region of interest with an intensity at 90% of the maximum (slightly after the maximum intensity of the decay) from the beginning of the service life to ensure minimal IRF interference while maintaining the same for all intensity decay settings Relative time window Were analyzed 25 to 50 ROI for rafts and spots and 15-25 ROI for drops, images selected from more than 4 replicates recorded from at least 3 independent experiments. Two-tailed t-tests have been used to evaluate statistical differences between species or between coacervate systems. For a pixel-by-pixel analysis of the lifetime (τ), the total attenuation of the lifetime over the field for each channel was calculated and an approximation of a 2/3-component exponential attenuation model was carried out. The lifetime attenuation for each pixel was then fitted using previously calculated τ values, resulting in a pseudocolor FLIM fit image. The tail-fit lifetime range was the same across all images of the same channel, and each decay produced enough photons to provide a reliable fit. For FRET analysis, pixels were selected by applying a lower intensity threshold of 100 photons, which averaged a background signal (FBG) of 11 photons. The fluorescence intensity of each channel was corrected by experimentally determined correction factors: 69 spectral crosstalk α was 0.004, direct excitation β was 0.0305, detection efficiency γ was 0.517. The FRET efficiency at the pixel level is then calculated using the following equation:
where FDD is the fluorescence intensity observed in the donor (green) channel, FDA is the fluorescence intensity observed in the acceptor (red) channel under indirect excitation, and FAA is the fluorescence intensity observed in the acceptor (red) channel under direct excitation (PIE). Fluorescence intensity pulses are observed in the channel).
Place 100 µl of LLPS reaction solutions containing 25 µM unlabeled monomeric Tau441 (with or without 25 µM αS) in LLPS buffer (supplemented as above) on non-binding 96-well microplates with adhesive foil coating and droplet formation was checked by WF microscopy after equilibration. within 10 min. After 48 hours of incubation at room temperature, the presence of protein rafts and spots was confirmed. Then carefully remove the liquid over the rafts from the wells, then add 50 L of dissociation buffer (10 mM HEPES, pH 7.4, 1 M NaCl, 1 mM DTT) and incubate for 10 min. The high salt concentration ensures that LLPS will not repeat due to residual PEG, and possible protein assemblies formed only by electrostatic interactions will be disassembled. The bottom of the well was then carefully scraped off with a micropipette tip and the resulting solution was transferred to an empty observation well. After incubation of the samples with 50 μM ThT for 1 h, the presence of isolated spots was checked by WF microscopy. Prepare sonicated αS fibrils by incubating 300 µl of a 70-µM αS solution in PBS with pH 7.4, sodium azide 0.01% at 37 °C and 200 rpm on an orbital shaker for 7 days. The solution was then centrifuged at 9600×g for 30 min, the pellet was resuspended in PBS pH 7.4 and sonicated (1 min, 50% cycle, 80% amplitude in a Vibra-Cell VC130 sonicator, Sonics, Newton, USA) fibril samples with a relatively uniform size distribution of small fibrils.
FCS/FCCS analysis and two-color coincidence detection (TCCD) were performed on the same MT200 time-resolved fluorescent confocal microscope (Pico-Quant, Berlin, Germany) used for the FLIM-FRET microscopy experiments using the PIE mode. The laser power for these experiments was added to 6.0 µW (481 nm) and 6.2 µW (637 nm). The combination of these laser powers was chosen to produce similar brightness for the pairs of fluorophores used while achieving optimal count rates and avoiding photobleaching and saturation. Both data collection and analysis were performed using the commercially available SymphoTime64 version 2.3 software (PicoQuant, Berlin, Germany).
Samples of isolated αS/Tau aggregates obtained using LLPS are diluted in isolation buffer to the appropriate monomolecular concentration (typically a 1:500 dilution, since aggregates are already at low concentrations when isolated from coacervate samples). Samples were applied directly to coverslips (Corning, USA) precoated with a BSA solution at a concentration of 1 mg/mL.
For the PIE-smFRET analysis in the green and red channels, a lower intensity threshold of 25 photons was applied to filter out low intensity signals caused by monomeric events (note that monomers outnumber aggregated samples compared to isolated aggregates). This threshold was calculated as five times the average intensity of monomeric αS obtained from the analysis of pure monomer samples in order to specifically select aggregates for analysis. The PIE drive circuit, together with TSCPC data acquisition, has enabled the application of a lifetime weighting filter that helps eliminate background and spectral crosstalk. The flare intensity selected using the above thresholds was corrected using the average background signal determined from the histograms of occurrence versus intensity/bin of the buffer-only samples. Bursts associated with large aggregates typically occupy several consecutive bins in the time trace (set to 1 ms). In these cases, a bin of maximum strength was chosen. For FRET and stoichiometric analysis, the theoretically determined gamma factor γ (0.517) was used. Spectral crosstalk and direct excitation contributions are negligible (determined experimentally) at the excitation laser power used. The efficiency and stoichiometry of FRET in an explosion is calculated as follows.