Journal: bioRxiv

Article Title: Nuclear speckle proteins form intrinsic and MALAT1 -dependent microphases

doi: 10.1101/2025.02.26.640430

Figure Lengend Snippet: (A) Schematic of a homopolymer that undergoes macrophase separation. Homopolymers or effective homopolymers are defined by uniformity of interactions (blue lines) that are on the order of thermal energy. Above a threshold concentration of the homopolymer, known as the saturation concentration (c sat ), a single micron-scale dense phase forms that coexists with the dilute phase. The inset shows a schematic of how some of the millions of molecules might be organized within the dense phase. (B) Schematic of a di-block copolymer featuring A-(purple) and B-blocks (green). A normalized interaction coefficient Δ XY is used to define the interactions between blocks. In the interaction matrix, the homotypic inter-block interactions Δ AA and Δ BB are attractive, whereas the incompatibility of the blocks leads to repulsive heterotypic interactions denoted by positive values of Δ AB . (C) An alternative scenario for the di-block system is one where the homotypic interactions are repulsive, positive values of Δ AA and Δ BB , whereas heterotypic interactions are attractive (negative values of Δ AB ). (D) Sequence architectures of hnRNP-A1 and SRSF1 showing the domain boundaries. (E) Mapping of the computed surface electrostatic potentials (SEPs) onto the surfaces of the RRMs from hnRNP-A1 and SRSF1. Regions in blue are positive potentials, regions in red are negative potentials, and regions in white are neutral. The magnitudes and signs of the site-specific SEPs are shown in Figure S1 . (F) Non-random sequence features within the LCD of hnRNP-A1 and IDR1 and IDR2 of SRSF1. The color-coded sequences from which the features were extracted are shown below the NARDINI+ analysis. The colors correspond to polar residues (green), positively charged residues (blue), negatively charged residues (red), glycine residues (magenta), aromatic residues (orange), and others (black). The inventory of sequence features is provided in the Methods section. A positive z-score implies either an enrichment of a compositional bias when compared to the human proteome or a non-random linear segregation of specific pairs of amino acid types. A negative z-score implies either a depletion of a compositional bias when compared to the human proteome or a non-random linear mixing of specific pairs of amino acid types. (G) Interaction coefficients Δ XY computed from atomistic simulations for each pair of distinct domains within SRSF1 (left) and hnRNP-A1 (right).

Article Snippet: pGEMT- MALAT1 was cloned by amplifying the MALAT1 cDNA from pCMV-MALAT1 and ligating it into the pGEM-T easy vector (Promega, USA) according to the manufacturer’s protocol. pET28-SRSF1 was cloned by sub-cloning SRSF1 cDNA from YFP-SRSF1 into the pET28 vector by restriction digestion and ligation. pET28-SRSF3 (Uniprot: P84103), pET28-SRSF5 (Uniprot: Q13243), and pET28-SRSF7 (Uniprot: Q16629) were generated by gene synthesis (GenScript, USA).

Techniques: Concentration Assay, Blocking Assay, Sequencing

Journal: bioRxiv

Article Title: Nuclear speckle proteins form intrinsic and MALAT1 -dependent microphases

doi: 10.1101/2025.02.26.640430

Figure Lengend Snippet: (A) Data from RALS measurements for SRSF1. The scattering intensity is plotted as a function of SRSF1 concentration. Results are shown from a single representative measurement (solid circles joined by dashed lines) and the spread over three technical replicates across different protein preps (pale blue envelope). The discontinuity in the scattering intensities at 0.45 ± 0.05 µM is denoted as c µ . (B) Decay rate constants ( k decay ) of the fluorescence lifetime of pyrene. The solid circles are rate constants from a single representative measurement. The pale blue envelope shows the range of values obtained from three technical replicates. The rate constant shows a plateauing behavior (horizontal dashed line) near the c µ (∼0.45 µM) estimated from RALS. The increase in the rate constant beyond the plateau value is proportional to the increase in the concentration of microphases with increasing [SRSF1]. (C) QF-DEEM images at three different magnifications show biconcave morphologies of SRSF1 microphases and their reproducibility across a large field-of-view. (D) Three-dimensional rendering of the morphologies observed in panel (C) . The method of rendering, described in Methods , uses a diameter of 8 units, the biconcave (dent) thickness as 0.5 units, and the maximum thickness of the biconcave structure near the rim as 2 units. (E) Histogram of diameters (in nm) constructed from the analysis of 639 distinct images. (F) Results from TRFQ experiments were fit as described in the Methods . The intercept along the ordinate is 49.7 ± 1.8 molecules, and it quantifies the number of SRSF1 molecules within a microphase at the c µ .

Article Snippet: pGEMT- MALAT1 was cloned by amplifying the MALAT1 cDNA from pCMV-MALAT1 and ligating it into the pGEM-T easy vector (Promega, USA) according to the manufacturer’s protocol. pET28-SRSF1 was cloned by sub-cloning SRSF1 cDNA from YFP-SRSF1 into the pET28 vector by restriction digestion and ligation. pET28-SRSF3 (Uniprot: P84103), pET28-SRSF5 (Uniprot: Q13243), and pET28-SRSF7 (Uniprot: Q16629) were generated by gene synthesis (GenScript, USA).

Techniques: Concentration Assay, Fluorescence, Construct

Journal: bioRxiv

Article Title: Nuclear speckle proteins form intrinsic and MALAT1 -dependent microphases

doi: 10.1101/2025.02.26.640430

Figure Lengend Snippet: (A) Schematic showing how clustering of SRSF1 microphases might arise via SALR-like interactions. An accumulation of surface charge shown by blue + marks will draw a counterion cloud (dashed red envelope) around each microphase (solid blue circles). Increasing the concentration of microphases increases the likelihood that microphases cluster via effective short-range attractions that result from the sharing of counterion clouds among pairs and higher-order clusters of microphases. The attractions saturate beyond a certain size because the buildup of charge creates long-range repulsions. Hence, clusters of microphases do not grow beyond a specific size. (B) MRPS measurements of SRSF1 above the c µ are shown as average and standard deviation about the mean. The data show peaks at a constant position while the amplitude increases with increasing concentration of SRSF1. The implication is that the mean sizes stay the same, while the abundance of the clusters increases with increasing concentrations of SRSF1. Results are shown from a single representative measurement (solid lines) and the spread over three technical replicates (pale envelopes). (C) Confocal microscopy at concentrations above c µ . With increased SRSF1 concentration, the abundance of SRSF1 clusters increases. The schematic depicts how microphases form clusters by touching to form incipient networks of clusters. (D) Box and whisker plot from analysis of QF-DEEM images yielded estimates for the number of microphases within a ∼380 nm sized cluster. The box extends from the 25 th to 75 th percentiles. The line in the middle of the box is at the median value of 23 microphases per cluster. The whiskers extend from the minimum to the maximum number of microphases detected in the probe spheres. (E) Single-molecule microscopy with photoactivatable JaneliaFluor 549 (PA-JF549)-labeled SRSF1 shows that the clusters are composed of multiple microphases. The image is produced as a maximum projection of 200 frames collected where the color bar shows the fluorescence intensity. (F) Histograms of fluorophores with trajectories of less than 50 nm (slow) or more than 50 nm (fast) over each 20 ms frame for 20,000 frames each with an inset for a zoomed-in field of view of the outlined area. The histograms are represented as heatmaps mapped onto the image. (G) Excess variance (see Methods ) computed over bins that quantify the displacements of slow-versus fast-moving molecules. (H) RALS extracted c µ values in the absence and presence of 200 mM L-arginine in experimental buffer are within the error. (I) Confocal microscopy images of AF488-SRSF1 in the presence of 50 mM, 100 mM, 200 mM, and 500 mM L-arginine in the experimental buffer. (J) GFP-tagged SRSF1 is expressed in X. laevis GVs by microinjection of mRNA and SRSF1-positive bodies are observed. (K) Merge of images collected simultaneously via brightfield microscopy and the intensity in the GFP channel (see S3J ) shows that the SRSF1 bodies are juxta identifiable Cajal bodies in the GVs. (L) Cumulative distribution functions (CDF) of the diameters of SRSF1-positive bodies remain essentially invariant at different levels of the injected mRNA. Latrunculin A (Lat-A) treatments (orange dotted) to oocytes injected with 100 ng/µL (orange) mRNA did not result in an increase in the size of the intra-GV SRSF1-positive bodies.

Article Snippet: pGEMT- MALAT1 was cloned by amplifying the MALAT1 cDNA from pCMV-MALAT1 and ligating it into the pGEM-T easy vector (Promega, USA) according to the manufacturer’s protocol. pET28-SRSF1 was cloned by sub-cloning SRSF1 cDNA from YFP-SRSF1 into the pET28 vector by restriction digestion and ligation. pET28-SRSF3 (Uniprot: P84103), pET28-SRSF5 (Uniprot: Q13243), and pET28-SRSF7 (Uniprot: Q16629) were generated by gene synthesis (GenScript, USA).

Techniques: Concentration Assay, Standard Deviation, Confocal Microscopy, Whisker Assay, Microscopy, Labeling, Produced, Fluorescence, Microinjection, Injection

Journal: bioRxiv

Article Title: Nuclear speckle proteins form intrinsic and MALAT1 -dependent microphases

doi: 10.1101/2025.02.26.640430

Figure Lengend Snippet: (A) Results from NARDINI+ analysis comparing the sequence features of IDRs in SRSF1, SRSF3, SRSF5, and SRSF7. For clarity, only the sequence features in which at least one sequence had a z-score greater than or equal to 3 are shown. The Ser / Arg rich IDRs (C-terminal IDR in SRSF3 and SRSF7 and IDR2 in SRSF1 and SRSF5) share similarities in terms of compositional biases toward Ser and Arg and the presence of Ser patches (red box). The two distinct IDRs are IDR1 of SRSF1 and the linker between the N-terminal RRM and C-terminal IDR of SRSF7. (B) Inter-domain interaction matrix of Δ XY coefficients for SRSF3. (C) Inter-domain interaction matrix of Δ XY coefficients for SRSF5. (D) Inter-domain interaction matrix of Δ XY coefficients for SRSF7. (E) Bar plot of threshold concentrations c µ for SRSF3, SRSF5, and SRSF7 extracted from RALS measurements. The data for SRSF1 are from the RALS measurements shown in and these are incorporated into the bar plot for comparison. The values of c µ increase in the order of SRSF1 (0.45 ± 0.05 µM) < SRSF7 (1.0 ± 0.14 µM) < SRSF3 (2.0 ± 0.14 µM) < SRSF5 (9.5 ± 0.71 µM). (F) The number of molecules per microphases determined from TRFQ experiments for the four SRSFs (see Figure S4F for values). (G) QF-DEEM images of 5 µM, SRSF3 which is above the measured c µ , shows a spheroidal morphology. (H) Three-dimensional rendering of the morphologies observed in panel (G) . The method of rendering, described in the Methods , uses a diameter of 8 units, the biconcave (dent) thickness as 5 units, and the maximum thickness of the biconcave structure near the rim as 5 units. (I) Histogram of diameters (in nm) constructed from the analysis of images of 269 distinct particles. (J) Summary of results from MRPS measurements of SRSFs over a range of protein concentrations. The abscissa is normalized by the intrinsic c µ of each protein for comparison across the SRSFs. The ordinate shows the diameter of clusters in nanometers. The average diameter is 355 nm for SRSF1 (dark blue), 295 nm for SRSF7 (light blue), 298 nm for SRSF5 (yellow), and 276 nm for SRSF3 (green). G. Sub-micron-scale clusters of microphases are observed by confocal microscopy for SRSF7, SRSF3, and SRSF5.

Article Snippet: pGEMT- MALAT1 was cloned by amplifying the MALAT1 cDNA from pCMV-MALAT1 and ligating it into the pGEM-T easy vector (Promega, USA) according to the manufacturer’s protocol. pET28-SRSF1 was cloned by sub-cloning SRSF1 cDNA from YFP-SRSF1 into the pET28 vector by restriction digestion and ligation. pET28-SRSF3 (Uniprot: P84103), pET28-SRSF5 (Uniprot: Q13243), and pET28-SRSF7 (Uniprot: Q16629) were generated by gene synthesis (GenScript, USA).

Techniques: Sequencing, Comparison, Construct, Confocal Microscopy

Journal: bioRxiv

Article Title: Nuclear speckle proteins form intrinsic and MALAT1 -dependent microphases

doi: 10.1101/2025.02.26.640430

Figure Lengend Snippet: (A) Computed interaction matrices for pairs of domains drawn from SRSF1 and SRSF3. (B) QF-DEEM images at different magnifications of nanoscale structures of microphases formed in mixtures of SRSF1 with SRSF3. (C) Three-dimensional rendering of the morphologies observed in panel (B) . The rendered images show the types of biconcave (same as 2D ) and spheroidal structures (same as 4H ) that were observed. (D) Histogram of diameters (in nm) constructed from the analysis of 1359 distinct particle images. (E) Confocal images of sub-micron scale structures formed in mixtures of SRSF3 and SRSF1. (F) Confocal images of sub-micron scale structures formed in mixtures of SRSF5 and SRSF1. (G) Confocal images of sub-micron scale structures formed in mixtures of SRSF7 and SRSF1. In panels (E), (F), and (G), the concentrations of SRSF3, SRSF5, and SRSF7 were below their c µ whereas the concentration of SRSF1 was above its c µ . (H) Impact of each of SRSF3, SRSF5, and SRSF7 on the c µ of SRSF1.

Article Snippet: pGEMT- MALAT1 was cloned by amplifying the MALAT1 cDNA from pCMV-MALAT1 and ligating it into the pGEM-T easy vector (Promega, USA) according to the manufacturer’s protocol. pET28-SRSF1 was cloned by sub-cloning SRSF1 cDNA from YFP-SRSF1 into the pET28 vector by restriction digestion and ligation. pET28-SRSF3 (Uniprot: P84103), pET28-SRSF5 (Uniprot: Q13243), and pET28-SRSF7 (Uniprot: Q16629) were generated by gene synthesis (GenScript, USA).

Techniques: Construct, Concentration Assay

Journal: bioRxiv

Article Title: Nuclear speckle proteins form intrinsic and MALAT1 -dependent microphases

doi: 10.1101/2025.02.26.640430

Figure Lengend Snippet: (A) Sequence architecture of TDP-43. Also shown are the sequence details, including the sequence of the C-terminal LCD. (B) Computed SEPs of each of the RRMs mapped onto the surface sites. RRM1 features patches of basic regions, and RRM2 features highly acidic faces. (C) Computed inter-domain interaction maps for TDP-43. (D) Representative RALS trace of TDP-43 showing a discontinuity at ∼0.12 µM. Results are shown from a single representative measurement (solid circles) and the spread over three technical replicates across different protein preps (pale envelope). On average, the value of c µ was found to be 0.18 ± 0.08 µM. (E) QF-DEEM images 0.7 µM TDP-43 at different magnifications. (F) A rendering of the biconcave structures seen in QF-DEEM images (see Methods ) uses a diameter of 8 units, the biconcave (dent) thickness as 1 units, and the maximum thickness of the biconcave structure near the rim as 2 units. The biconcavity thickness is larger than 0.5 used for SRSF1 (see ) to portray dents in TDP-43 structures that are less evident than SRSF1. (G) Histogram of diameters of TDP-43 microphases from analysis of 855 particle images. (H) Results from TRFQ measurements performed over a range of TDP-43 concentrations. The concentrations of microphases were fit to a line at the plateau value and the intercept yields an estimate of the number of TDP-43 molecules per microphase to be ∼13. (I) Mixture of 1 µM TDP-43 and 2 µM SRSF1 show co-localization of the two proteins in size-limited clusters. TDP-43 is observed to accumulate on the surface SRSF1 clusters. (J) Effect of TDP-43 on the c µ of SRSF1, which was measured at different molar ratios of TDP-43 and SRSF1.

Article Snippet: pGEMT- MALAT1 was cloned by amplifying the MALAT1 cDNA from pCMV-MALAT1 and ligating it into the pGEM-T easy vector (Promega, USA) according to the manufacturer’s protocol. pET28-SRSF1 was cloned by sub-cloning SRSF1 cDNA from YFP-SRSF1 into the pET28 vector by restriction digestion and ligation. pET28-SRSF3 (Uniprot: P84103), pET28-SRSF5 (Uniprot: Q13243), and pET28-SRSF7 (Uniprot: Q16629) were generated by gene synthesis (GenScript, USA).

Techniques: Sequencing

Journal: bioRxiv

Article Title: Nuclear speckle proteins form intrinsic and MALAT1 -dependent microphases

doi: 10.1101/2025.02.26.640430

Figure Lengend Snippet: A. In the presence of MALAT1 , the c µ , as extracted from RALS, of SRSF1 decreased from ∼0.45 µM to ∼0.22 µM. Conversely, the c µ of TDP-43 increased from ∼0.18 µM to 0.21 µM in the presence of MALAT1 . (B) TRFQ experiments, performed over a range of SRSF1 concentrations at a constant molar ratio of SRSF1-to- MALAT1 at 1-to-5×10 -4 , yielded an estimate of ∼13 SRSF1 molecules per microphase. (C) QF-DEEM images of SRSF1- MALAT1 microphases show spherical morphologies, which are different from the biconcave structures observed in the absence of MALAT1 . (D) Histograms collected across 118 particle images of the diameters (in nm) of the spheroids of SRSF1 microphases that form in the presence of MALAT1 . (E) QF-DEEM images of TDP-43+ MALAT1 microphases. The biconcave structures formed in the absence of MALAT1 are maintained in the presence of MALAT1 . (F) Histograms collected across 230 particle images of the diameters (in nm) of the spheroids of TDP-43 microphases that form in the presence of MALAT1 .

Article Snippet: pGEMT- MALAT1 was cloned by amplifying the MALAT1 cDNA from pCMV-MALAT1 and ligating it into the pGEM-T easy vector (Promega, USA) according to the manufacturer’s protocol. pET28-SRSF1 was cloned by sub-cloning SRSF1 cDNA from YFP-SRSF1 into the pET28 vector by restriction digestion and ligation. pET28-SRSF3 (Uniprot: P84103), pET28-SRSF5 (Uniprot: Q13243), and pET28-SRSF7 (Uniprot: Q16629) were generated by gene synthesis (GenScript, USA).

Techniques:

Journal: bioRxiv

Article Title: Nuclear speckle proteins form intrinsic and MALAT1 -dependent microphases

doi: 10.1101/2025.02.26.640430

Figure Lengend Snippet: (A) Schematic of a homopolymer that undergoes macrophase separation. Homopolymers or effective homopolymers are defined by uniformity of interactions (blue lines) that are on the order of thermal energy. Above a threshold concentration of the homopolymer, known as the saturation concentration (c sat ), a single micron-scale dense phase forms that coexists with the dilute phase. The inset shows a schematic of how some of the millions of molecules might be organized within the dense phase. (B) Schematic of a di-block copolymer featuring A-(purple) and B-blocks (green). A normalized interaction coefficient Δ XY is used to define the interactions between blocks. In the interaction matrix, the homotypic inter-block interactions Δ AA and Δ BB are attractive, whereas the incompatibility of the blocks leads to repulsive heterotypic interactions denoted by positive values of Δ AB . (C) An alternative scenario for the di-block system is one where the homotypic interactions are repulsive, positive values of Δ AA and Δ BB , whereas heterotypic interactions are attractive (negative values of Δ AB ). (D) Sequence architectures of hnRNP-A1 and SRSF1 showing the domain boundaries. (E) Mapping of the computed surface electrostatic potentials (SEPs) onto the surfaces of the RRMs from hnRNP-A1 and SRSF1. Regions in blue are positive potentials, regions in red are negative potentials, and regions in white are neutral. The magnitudes and signs of the site-specific SEPs are shown in Figure S1 . (F) Non-random sequence features within the LCD of hnRNP-A1 and IDR1 and IDR2 of SRSF1. The color-coded sequences from which the features were extracted are shown below the NARDINI+ analysis. The colors correspond to polar residues (green), positively charged residues (blue), negatively charged residues (red), glycine residues (magenta), aromatic residues (orange), and others (black). The inventory of sequence features is provided in the Methods section. A positive z-score implies either an enrichment of a compositional bias when compared to the human proteome or a non-random linear segregation of specific pairs of amino acid types. A negative z-score implies either a depletion of a compositional bias when compared to the human proteome or a non-random linear mixing of specific pairs of amino acid types. (G) Interaction coefficients Δ XY computed from atomistic simulations for each pair of distinct domains within SRSF1 (left) and hnRNP-A1 (right).

Article Snippet: pGEMT- MALAT1 was cloned by amplifying the MALAT1 cDNA from pCMV-MALAT1 and ligating it into the pGEM-T easy vector (Promega, USA) according to the manufacturer’s protocol. pET28-SRSF1 was cloned by sub-cloning SRSF1 cDNA from YFP-SRSF1 into the pET28 vector by restriction digestion and ligation. pET28-SRSF3 (Uniprot: P84103), pET28-SRSF5 (Uniprot: Q13243), and pET28-SRSF7 (Uniprot: Q16629) were generated by gene synthesis (GenScript, USA).

Techniques: Concentration Assay, Blocking Assay, Sequencing

Journal: bioRxiv

Article Title: Nuclear speckle proteins form intrinsic and MALAT1 -dependent microphases

doi: 10.1101/2025.02.26.640430

Figure Lengend Snippet: (A) Data from RALS measurements for SRSF1. The scattering intensity is plotted as a function of SRSF1 concentration. Results are shown from a single representative measurement (solid circles joined by dashed lines) and the spread over three technical replicates across different protein preps (pale blue envelope). The discontinuity in the scattering intensities at 0.45 ± 0.05 µM is denoted as c µ . (B) Decay rate constants ( k decay ) of the fluorescence lifetime of pyrene. The solid circles are rate constants from a single representative measurement. The pale blue envelope shows the range of values obtained from three technical replicates. The rate constant shows a plateauing behavior (horizontal dashed line) near the c µ (∼0.45 µM) estimated from RALS. The increase in the rate constant beyond the plateau value is proportional to the increase in the concentration of microphases with increasing [SRSF1]. (C) QF-DEEM images at three different magnifications show biconcave morphologies of SRSF1 microphases and their reproducibility across a large field-of-view. (D) Three-dimensional rendering of the morphologies observed in panel (C) . The method of rendering, described in Methods , uses a diameter of 8 units, the biconcave (dent) thickness as 0.5 units, and the maximum thickness of the biconcave structure near the rim as 2 units. (E) Histogram of diameters (in nm) constructed from the analysis of 639 distinct images. (F) Results from TRFQ experiments were fit as described in the Methods . The intercept along the ordinate is 49.7 ± 1.8 molecules, and it quantifies the number of SRSF1 molecules within a microphase at the c µ .

Article Snippet: pGEMT- MALAT1 was cloned by amplifying the MALAT1 cDNA from pCMV-MALAT1 and ligating it into the pGEM-T easy vector (Promega, USA) according to the manufacturer’s protocol. pET28-SRSF1 was cloned by sub-cloning SRSF1 cDNA from YFP-SRSF1 into the pET28 vector by restriction digestion and ligation. pET28-SRSF3 (Uniprot: P84103), pET28-SRSF5 (Uniprot: Q13243), and pET28-SRSF7 (Uniprot: Q16629) were generated by gene synthesis (GenScript, USA).

Techniques: Concentration Assay, Fluorescence, Construct

Journal: bioRxiv

Article Title: Nuclear speckle proteins form intrinsic and MALAT1 -dependent microphases

doi: 10.1101/2025.02.26.640430

Figure Lengend Snippet: (A) Schematic showing how clustering of SRSF1 microphases might arise via SALR-like interactions. An accumulation of surface charge shown by blue + marks will draw a counterion cloud (dashed red envelope) around each microphase (solid blue circles). Increasing the concentration of microphases increases the likelihood that microphases cluster via effective short-range attractions that result from the sharing of counterion clouds among pairs and higher-order clusters of microphases. The attractions saturate beyond a certain size because the buildup of charge creates long-range repulsions. Hence, clusters of microphases do not grow beyond a specific size. (B) MRPS measurements of SRSF1 above the c µ are shown as average and standard deviation about the mean. The data show peaks at a constant position while the amplitude increases with increasing concentration of SRSF1. The implication is that the mean sizes stay the same, while the abundance of the clusters increases with increasing concentrations of SRSF1. Results are shown from a single representative measurement (solid lines) and the spread over three technical replicates (pale envelopes). (C) Confocal microscopy at concentrations above c µ . With increased SRSF1 concentration, the abundance of SRSF1 clusters increases. The schematic depicts how microphases form clusters by touching to form incipient networks of clusters. (D) Box and whisker plot from analysis of QF-DEEM images yielded estimates for the number of microphases within a ∼380 nm sized cluster. The box extends from the 25 th to 75 th percentiles. The line in the middle of the box is at the median value of 23 microphases per cluster. The whiskers extend from the minimum to the maximum number of microphases detected in the probe spheres. (E) Single-molecule microscopy with photoactivatable JaneliaFluor 549 (PA-JF549)-labeled SRSF1 shows that the clusters are composed of multiple microphases. The image is produced as a maximum projection of 200 frames collected where the color bar shows the fluorescence intensity. (F) Histograms of fluorophores with trajectories of less than 50 nm (slow) or more than 50 nm (fast) over each 20 ms frame for 20,000 frames each with an inset for a zoomed-in field of view of the outlined area. The histograms are represented as heatmaps mapped onto the image. (G) Excess variance (see Methods ) computed over bins that quantify the displacements of slow-versus fast-moving molecules. (H) RALS extracted c µ values in the absence and presence of 200 mM L-arginine in experimental buffer are within the error. (I) Confocal microscopy images of AF488-SRSF1 in the presence of 50 mM, 100 mM, 200 mM, and 500 mM L-arginine in the experimental buffer. (J) GFP-tagged SRSF1 is expressed in X. laevis GVs by microinjection of mRNA and SRSF1-positive bodies are observed. (K) Merge of images collected simultaneously via brightfield microscopy and the intensity in the GFP channel (see S3J ) shows that the SRSF1 bodies are juxta identifiable Cajal bodies in the GVs. (L) Cumulative distribution functions (CDF) of the diameters of SRSF1-positive bodies remain essentially invariant at different levels of the injected mRNA. Latrunculin A (Lat-A) treatments (orange dotted) to oocytes injected with 100 ng/µL (orange) mRNA did not result in an increase in the size of the intra-GV SRSF1-positive bodies.

Article Snippet: pGEMT- MALAT1 was cloned by amplifying the MALAT1 cDNA from pCMV-MALAT1 and ligating it into the pGEM-T easy vector (Promega, USA) according to the manufacturer’s protocol. pET28-SRSF1 was cloned by sub-cloning SRSF1 cDNA from YFP-SRSF1 into the pET28 vector by restriction digestion and ligation. pET28-SRSF3 (Uniprot: P84103), pET28-SRSF5 (Uniprot: Q13243), and pET28-SRSF7 (Uniprot: Q16629) were generated by gene synthesis (GenScript, USA).

Techniques: Concentration Assay, Standard Deviation, Confocal Microscopy, Whisker Assay, Microscopy, Labeling, Produced, Fluorescence, Microinjection, Injection

Journal: bioRxiv

Article Title: Nuclear speckle proteins form intrinsic and MALAT1 -dependent microphases

doi: 10.1101/2025.02.26.640430

Figure Lengend Snippet: (A) Results from NARDINI+ analysis comparing the sequence features of IDRs in SRSF1, SRSF3, SRSF5, and SRSF7. For clarity, only the sequence features in which at least one sequence had a z-score greater than or equal to 3 are shown. The Ser / Arg rich IDRs (C-terminal IDR in SRSF3 and SRSF7 and IDR2 in SRSF1 and SRSF5) share similarities in terms of compositional biases toward Ser and Arg and the presence of Ser patches (red box). The two distinct IDRs are IDR1 of SRSF1 and the linker between the N-terminal RRM and C-terminal IDR of SRSF7. (B) Inter-domain interaction matrix of Δ XY coefficients for SRSF3. (C) Inter-domain interaction matrix of Δ XY coefficients for SRSF5. (D) Inter-domain interaction matrix of Δ XY coefficients for SRSF7. (E) Bar plot of threshold concentrations c µ for SRSF3, SRSF5, and SRSF7 extracted from RALS measurements. The data for SRSF1 are from the RALS measurements shown in and these are incorporated into the bar plot for comparison. The values of c µ increase in the order of SRSF1 (0.45 ± 0.05 µM) < SRSF7 (1.0 ± 0.14 µM) < SRSF3 (2.0 ± 0.14 µM) < SRSF5 (9.5 ± 0.71 µM). (F) The number of molecules per microphases determined from TRFQ experiments for the four SRSFs (see Figure S4F for values). (G) QF-DEEM images of 5 µM, SRSF3 which is above the measured c µ , shows a spheroidal morphology. (H) Three-dimensional rendering of the morphologies observed in panel (G) . The method of rendering, described in the Methods , uses a diameter of 8 units, the biconcave (dent) thickness as 5 units, and the maximum thickness of the biconcave structure near the rim as 5 units. (I) Histogram of diameters (in nm) constructed from the analysis of images of 269 distinct particles. (J) Summary of results from MRPS measurements of SRSFs over a range of protein concentrations. The abscissa is normalized by the intrinsic c µ of each protein for comparison across the SRSFs. The ordinate shows the diameter of clusters in nanometers. The average diameter is 355 nm for SRSF1 (dark blue), 295 nm for SRSF7 (light blue), 298 nm for SRSF5 (yellow), and 276 nm for SRSF3 (green). G. Sub-micron-scale clusters of microphases are observed by confocal microscopy for SRSF7, SRSF3, and SRSF5.

Article Snippet: pGEMT- MALAT1 was cloned by amplifying the MALAT1 cDNA from pCMV-MALAT1 and ligating it into the pGEM-T easy vector (Promega, USA) according to the manufacturer’s protocol. pET28-SRSF1 was cloned by sub-cloning SRSF1 cDNA from YFP-SRSF1 into the pET28 vector by restriction digestion and ligation. pET28-SRSF3 (Uniprot: P84103), pET28-SRSF5 (Uniprot: Q13243), and pET28-SRSF7 (Uniprot: Q16629) were generated by gene synthesis (GenScript, USA).

Techniques: Sequencing, Comparison, Construct, Confocal Microscopy

Journal: bioRxiv

Article Title: Nuclear speckle proteins form intrinsic and MALAT1 -dependent microphases

doi: 10.1101/2025.02.26.640430

Figure Lengend Snippet: (A) Computed interaction matrices for pairs of domains drawn from SRSF1 and SRSF3. (B) QF-DEEM images at different magnifications of nanoscale structures of microphases formed in mixtures of SRSF1 with SRSF3. (C) Three-dimensional rendering of the morphologies observed in panel (B) . The rendered images show the types of biconcave (same as 2D ) and spheroidal structures (same as 4H ) that were observed. (D) Histogram of diameters (in nm) constructed from the analysis of 1359 distinct particle images. (E) Confocal images of sub-micron scale structures formed in mixtures of SRSF3 and SRSF1. (F) Confocal images of sub-micron scale structures formed in mixtures of SRSF5 and SRSF1. (G) Confocal images of sub-micron scale structures formed in mixtures of SRSF7 and SRSF1. In panels (E), (F), and (G), the concentrations of SRSF3, SRSF5, and SRSF7 were below their c µ whereas the concentration of SRSF1 was above its c µ . (H) Impact of each of SRSF3, SRSF5, and SRSF7 on the c µ of SRSF1.

Article Snippet: pGEMT- MALAT1 was cloned by amplifying the MALAT1 cDNA from pCMV-MALAT1 and ligating it into the pGEM-T easy vector (Promega, USA) according to the manufacturer’s protocol. pET28-SRSF1 was cloned by sub-cloning SRSF1 cDNA from YFP-SRSF1 into the pET28 vector by restriction digestion and ligation. pET28-SRSF3 (Uniprot: P84103), pET28-SRSF5 (Uniprot: Q13243), and pET28-SRSF7 (Uniprot: Q16629) were generated by gene synthesis (GenScript, USA).

Techniques: Construct, Concentration Assay

Journal: bioRxiv

Article Title: Nuclear speckle proteins form intrinsic and MALAT1 -dependent microphases

doi: 10.1101/2025.02.26.640430

Figure Lengend Snippet: (A) Sequence architecture of TDP-43. Also shown are the sequence details, including the sequence of the C-terminal LCD. (B) Computed SEPs of each of the RRMs mapped onto the surface sites. RRM1 features patches of basic regions, and RRM2 features highly acidic faces. (C) Computed inter-domain interaction maps for TDP-43. (D) Representative RALS trace of TDP-43 showing a discontinuity at ∼0.12 µM. Results are shown from a single representative measurement (solid circles) and the spread over three technical replicates across different protein preps (pale envelope). On average, the value of c µ was found to be 0.18 ± 0.08 µM. (E) QF-DEEM images 0.7 µM TDP-43 at different magnifications. (F) A rendering of the biconcave structures seen in QF-DEEM images (see Methods ) uses a diameter of 8 units, the biconcave (dent) thickness as 1 units, and the maximum thickness of the biconcave structure near the rim as 2 units. The biconcavity thickness is larger than 0.5 used for SRSF1 (see ) to portray dents in TDP-43 structures that are less evident than SRSF1. (G) Histogram of diameters of TDP-43 microphases from analysis of 855 particle images. (H) Results from TRFQ measurements performed over a range of TDP-43 concentrations. The concentrations of microphases were fit to a line at the plateau value and the intercept yields an estimate of the number of TDP-43 molecules per microphase to be ∼13. (I) Mixture of 1 µM TDP-43 and 2 µM SRSF1 show co-localization of the two proteins in size-limited clusters. TDP-43 is observed to accumulate on the surface SRSF1 clusters. (J) Effect of TDP-43 on the c µ of SRSF1, which was measured at different molar ratios of TDP-43 and SRSF1.

Article Snippet: pGEMT- MALAT1 was cloned by amplifying the MALAT1 cDNA from pCMV-MALAT1 and ligating it into the pGEM-T easy vector (Promega, USA) according to the manufacturer’s protocol. pET28-SRSF1 was cloned by sub-cloning SRSF1 cDNA from YFP-SRSF1 into the pET28 vector by restriction digestion and ligation. pET28-SRSF3 (Uniprot: P84103), pET28-SRSF5 (Uniprot: Q13243), and pET28-SRSF7 (Uniprot: Q16629) were generated by gene synthesis (GenScript, USA).

Techniques: Sequencing

Journal: bioRxiv

Article Title: Nuclear speckle proteins form intrinsic and MALAT1 -dependent microphases

doi: 10.1101/2025.02.26.640430

Figure Lengend Snippet: A. In the presence of MALAT1 , the c µ , as extracted from RALS, of SRSF1 decreased from ∼0.45 µM to ∼0.22 µM. Conversely, the c µ of TDP-43 increased from ∼0.18 µM to 0.21 µM in the presence of MALAT1 . (B) TRFQ experiments, performed over a range of SRSF1 concentrations at a constant molar ratio of SRSF1-to- MALAT1 at 1-to-5×10 -4 , yielded an estimate of ∼13 SRSF1 molecules per microphase. (C) QF-DEEM images of SRSF1- MALAT1 microphases show spherical morphologies, which are different from the biconcave structures observed in the absence of MALAT1 . (D) Histograms collected across 118 particle images of the diameters (in nm) of the spheroids of SRSF1 microphases that form in the presence of MALAT1 . (E) QF-DEEM images of TDP-43+ MALAT1 microphases. The biconcave structures formed in the absence of MALAT1 are maintained in the presence of MALAT1 . (F) Histograms collected across 230 particle images of the diameters (in nm) of the spheroids of TDP-43 microphases that form in the presence of MALAT1 .

Article Snippet: pGEMT- MALAT1 was cloned by amplifying the MALAT1 cDNA from pCMV-MALAT1 and ligating it into the pGEM-T easy vector (Promega, USA) according to the manufacturer’s protocol. pET28-SRSF1 was cloned by sub-cloning SRSF1 cDNA from YFP-SRSF1 into the pET28 vector by restriction digestion and ligation. pET28-SRSF3 (Uniprot: P84103), pET28-SRSF5 (Uniprot: Q13243), and pET28-SRSF7 (Uniprot: Q16629) were generated by gene synthesis (GenScript, USA).

Techniques:

Journal: bioRxiv

Article Title: Nuclear speckle proteins form intrinsic and MALAT1 -dependent microphases

doi: 10.1101/2025.02.26.640430

Figure Lengend Snippet: (A) Schematic of a homopolymer that undergoes macrophase separation. Homopolymers or effective homopolymers are defined by uniformity of interactions (blue lines) that are on the order of thermal energy. Above a threshold concentration of the homopolymer, known as the saturation concentration (c sat ), a single micron-scale dense phase forms that coexists with the dilute phase. The inset shows a schematic of how some of the millions of molecules might be organized within the dense phase. (B) Schematic of a di-block copolymer featuring A-(purple) and B-blocks (green). A normalized interaction coefficient Δ XY is used to define the interactions between blocks. In the interaction matrix, the homotypic inter-block interactions Δ AA and Δ BB are attractive, whereas the incompatibility of the blocks leads to repulsive heterotypic interactions denoted by positive values of Δ AB . (C) An alternative scenario for the di-block system is one where the homotypic interactions are repulsive, positive values of Δ AA and Δ BB , whereas heterotypic interactions are attractive (negative values of Δ AB ). (D) Sequence architectures of hnRNP-A1 and SRSF1 showing the domain boundaries. (E) Mapping of the computed surface electrostatic potentials (SEPs) onto the surfaces of the RRMs from hnRNP-A1 and SRSF1. Regions in blue are positive potentials, regions in red are negative potentials, and regions in white are neutral. The magnitudes and signs of the site-specific SEPs are shown in Figure S1 . (F) Non-random sequence features within the LCD of hnRNP-A1 and IDR1 and IDR2 of SRSF1. The color-coded sequences from which the features were extracted are shown below the NARDINI+ analysis. The colors correspond to polar residues (green), positively charged residues (blue), negatively charged residues (red), glycine residues (magenta), aromatic residues (orange), and others (black). The inventory of sequence features is provided in the Methods section. A positive z-score implies either an enrichment of a compositional bias when compared to the human proteome or a non-random linear segregation of specific pairs of amino acid types. A negative z-score implies either a depletion of a compositional bias when compared to the human proteome or a non-random linear mixing of specific pairs of amino acid types. (G) Interaction coefficients Δ XY computed from atomistic simulations for each pair of distinct domains within SRSF1 (left) and hnRNP-A1 (right).

Article Snippet: For covalent conjugation of SRSF1 or other SRSFs with NHS (h-hydroxysuccinimide ester) AlexaFluor 488 or AlexaFluor 205 (Thermofisher Scientific, USA), 1 mg of SRSF1 was dialyzed into labeling Buffer (10% glycerol, 20 mM HEPES (pH 8.3), 1.25 M NaCl) using Pierce Microdialysis Plates (ThermoFisher Scientific, USA).

Techniques: Concentration Assay, Blocking Assay, Sequencing

Journal: bioRxiv

Article Title: Nuclear speckle proteins form intrinsic and MALAT1 -dependent microphases

doi: 10.1101/2025.02.26.640430

Figure Lengend Snippet: (A) Data from RALS measurements for SRSF1. The scattering intensity is plotted as a function of SRSF1 concentration. Results are shown from a single representative measurement (solid circles joined by dashed lines) and the spread over three technical replicates across different protein preps (pale blue envelope). The discontinuity in the scattering intensities at 0.45 ± 0.05 µM is denoted as c µ . (B) Decay rate constants ( k decay ) of the fluorescence lifetime of pyrene. The solid circles are rate constants from a single representative measurement. The pale blue envelope shows the range of values obtained from three technical replicates. The rate constant shows a plateauing behavior (horizontal dashed line) near the c µ (∼0.45 µM) estimated from RALS. The increase in the rate constant beyond the plateau value is proportional to the increase in the concentration of microphases with increasing [SRSF1]. (C) QF-DEEM images at three different magnifications show biconcave morphologies of SRSF1 microphases and their reproducibility across a large field-of-view. (D) Three-dimensional rendering of the morphologies observed in panel (C) . The method of rendering, described in Methods , uses a diameter of 8 units, the biconcave (dent) thickness as 0.5 units, and the maximum thickness of the biconcave structure near the rim as 2 units. (E) Histogram of diameters (in nm) constructed from the analysis of 639 distinct images. (F) Results from TRFQ experiments were fit as described in the Methods . The intercept along the ordinate is 49.7 ± 1.8 molecules, and it quantifies the number of SRSF1 molecules within a microphase at the c µ .

Article Snippet: For covalent conjugation of SRSF1 or other SRSFs with NHS (h-hydroxysuccinimide ester) AlexaFluor 488 or AlexaFluor 205 (Thermofisher Scientific, USA), 1 mg of SRSF1 was dialyzed into labeling Buffer (10% glycerol, 20 mM HEPES (pH 8.3), 1.25 M NaCl) using Pierce Microdialysis Plates (ThermoFisher Scientific, USA).

Techniques: Concentration Assay, Fluorescence, Construct

Journal: bioRxiv

Article Title: Nuclear speckle proteins form intrinsic and MALAT1 -dependent microphases

doi: 10.1101/2025.02.26.640430

Figure Lengend Snippet: (A) Schematic showing how clustering of SRSF1 microphases might arise via SALR-like interactions. An accumulation of surface charge shown by blue + marks will draw a counterion cloud (dashed red envelope) around each microphase (solid blue circles). Increasing the concentration of microphases increases the likelihood that microphases cluster via effective short-range attractions that result from the sharing of counterion clouds among pairs and higher-order clusters of microphases. The attractions saturate beyond a certain size because the buildup of charge creates long-range repulsions. Hence, clusters of microphases do not grow beyond a specific size. (B) MRPS measurements of SRSF1 above the c µ are shown as average and standard deviation about the mean. The data show peaks at a constant position while the amplitude increases with increasing concentration of SRSF1. The implication is that the mean sizes stay the same, while the abundance of the clusters increases with increasing concentrations of SRSF1. Results are shown from a single representative measurement (solid lines) and the spread over three technical replicates (pale envelopes). (C) Confocal microscopy at concentrations above c µ . With increased SRSF1 concentration, the abundance of SRSF1 clusters increases. The schematic depicts how microphases form clusters by touching to form incipient networks of clusters. (D) Box and whisker plot from analysis of QF-DEEM images yielded estimates for the number of microphases within a ∼380 nm sized cluster. The box extends from the 25 th to 75 th percentiles. The line in the middle of the box is at the median value of 23 microphases per cluster. The whiskers extend from the minimum to the maximum number of microphases detected in the probe spheres. (E) Single-molecule microscopy with photoactivatable JaneliaFluor 549 (PA-JF549)-labeled SRSF1 shows that the clusters are composed of multiple microphases. The image is produced as a maximum projection of 200 frames collected where the color bar shows the fluorescence intensity. (F) Histograms of fluorophores with trajectories of less than 50 nm (slow) or more than 50 nm (fast) over each 20 ms frame for 20,000 frames each with an inset for a zoomed-in field of view of the outlined area. The histograms are represented as heatmaps mapped onto the image. (G) Excess variance (see Methods ) computed over bins that quantify the displacements of slow-versus fast-moving molecules. (H) RALS extracted c µ values in the absence and presence of 200 mM L-arginine in experimental buffer are within the error. (I) Confocal microscopy images of AF488-SRSF1 in the presence of 50 mM, 100 mM, 200 mM, and 500 mM L-arginine in the experimental buffer. (J) GFP-tagged SRSF1 is expressed in X. laevis GVs by microinjection of mRNA and SRSF1-positive bodies are observed. (K) Merge of images collected simultaneously via brightfield microscopy and the intensity in the GFP channel (see S3J ) shows that the SRSF1 bodies are juxta identifiable Cajal bodies in the GVs. (L) Cumulative distribution functions (CDF) of the diameters of SRSF1-positive bodies remain essentially invariant at different levels of the injected mRNA. Latrunculin A (Lat-A) treatments (orange dotted) to oocytes injected with 100 ng/µL (orange) mRNA did not result in an increase in the size of the intra-GV SRSF1-positive bodies.

Article Snippet: For covalent conjugation of SRSF1 or other SRSFs with NHS (h-hydroxysuccinimide ester) AlexaFluor 488 or AlexaFluor 205 (Thermofisher Scientific, USA), 1 mg of SRSF1 was dialyzed into labeling Buffer (10% glycerol, 20 mM HEPES (pH 8.3), 1.25 M NaCl) using Pierce Microdialysis Plates (ThermoFisher Scientific, USA).

Techniques: Concentration Assay, Standard Deviation, Confocal Microscopy, Whisker Assay, Microscopy, Labeling, Produced, Fluorescence, Microinjection, Injection

Journal: bioRxiv

Article Title: Nuclear speckle proteins form intrinsic and MALAT1 -dependent microphases

doi: 10.1101/2025.02.26.640430

Figure Lengend Snippet: (A) Results from NARDINI+ analysis comparing the sequence features of IDRs in SRSF1, SRSF3, SRSF5, and SRSF7. For clarity, only the sequence features in which at least one sequence had a z-score greater than or equal to 3 are shown. The Ser / Arg rich IDRs (C-terminal IDR in SRSF3 and SRSF7 and IDR2 in SRSF1 and SRSF5) share similarities in terms of compositional biases toward Ser and Arg and the presence of Ser patches (red box). The two distinct IDRs are IDR1 of SRSF1 and the linker between the N-terminal RRM and C-terminal IDR of SRSF7. (B) Inter-domain interaction matrix of Δ XY coefficients for SRSF3. (C) Inter-domain interaction matrix of Δ XY coefficients for SRSF5. (D) Inter-domain interaction matrix of Δ XY coefficients for SRSF7. (E) Bar plot of threshold concentrations c µ for SRSF3, SRSF5, and SRSF7 extracted from RALS measurements. The data for SRSF1 are from the RALS measurements shown in and these are incorporated into the bar plot for comparison. The values of c µ increase in the order of SRSF1 (0.45 ± 0.05 µM) < SRSF7 (1.0 ± 0.14 µM) < SRSF3 (2.0 ± 0.14 µM) < SRSF5 (9.5 ± 0.71 µM). (F) The number of molecules per microphases determined from TRFQ experiments for the four SRSFs (see Figure S4F for values). (G) QF-DEEM images of 5 µM, SRSF3 which is above the measured c µ , shows a spheroidal morphology. (H) Three-dimensional rendering of the morphologies observed in panel (G) . The method of rendering, described in the Methods , uses a diameter of 8 units, the biconcave (dent) thickness as 5 units, and the maximum thickness of the biconcave structure near the rim as 5 units. (I) Histogram of diameters (in nm) constructed from the analysis of images of 269 distinct particles. (J) Summary of results from MRPS measurements of SRSFs over a range of protein concentrations. The abscissa is normalized by the intrinsic c µ of each protein for comparison across the SRSFs. The ordinate shows the diameter of clusters in nanometers. The average diameter is 355 nm for SRSF1 (dark blue), 295 nm for SRSF7 (light blue), 298 nm for SRSF5 (yellow), and 276 nm for SRSF3 (green). G. Sub-micron-scale clusters of microphases are observed by confocal microscopy for SRSF7, SRSF3, and SRSF5.

Article Snippet: For covalent conjugation of SRSF1 or other SRSFs with NHS (h-hydroxysuccinimide ester) AlexaFluor 488 or AlexaFluor 205 (Thermofisher Scientific, USA), 1 mg of SRSF1 was dialyzed into labeling Buffer (10% glycerol, 20 mM HEPES (pH 8.3), 1.25 M NaCl) using Pierce Microdialysis Plates (ThermoFisher Scientific, USA).

Techniques: Sequencing, Comparison, Construct, Confocal Microscopy

Journal: bioRxiv

Article Title: Nuclear speckle proteins form intrinsic and MALAT1 -dependent microphases

doi: 10.1101/2025.02.26.640430

Figure Lengend Snippet: (A) Computed interaction matrices for pairs of domains drawn from SRSF1 and SRSF3. (B) QF-DEEM images at different magnifications of nanoscale structures of microphases formed in mixtures of SRSF1 with SRSF3. (C) Three-dimensional rendering of the morphologies observed in panel (B) . The rendered images show the types of biconcave (same as 2D ) and spheroidal structures (same as 4H ) that were observed. (D) Histogram of diameters (in nm) constructed from the analysis of 1359 distinct particle images. (E) Confocal images of sub-micron scale structures formed in mixtures of SRSF3 and SRSF1. (F) Confocal images of sub-micron scale structures formed in mixtures of SRSF5 and SRSF1. (G) Confocal images of sub-micron scale structures formed in mixtures of SRSF7 and SRSF1. In panels (E), (F), and (G), the concentrations of SRSF3, SRSF5, and SRSF7 were below their c µ whereas the concentration of SRSF1 was above its c µ . (H) Impact of each of SRSF3, SRSF5, and SRSF7 on the c µ of SRSF1.

Article Snippet: For covalent conjugation of SRSF1 or other SRSFs with NHS (h-hydroxysuccinimide ester) AlexaFluor 488 or AlexaFluor 205 (Thermofisher Scientific, USA), 1 mg of SRSF1 was dialyzed into labeling Buffer (10% glycerol, 20 mM HEPES (pH 8.3), 1.25 M NaCl) using Pierce Microdialysis Plates (ThermoFisher Scientific, USA).

Techniques: Construct, Concentration Assay

Journal: bioRxiv

Article Title: Nuclear speckle proteins form intrinsic and MALAT1 -dependent microphases

doi: 10.1101/2025.02.26.640430

Figure Lengend Snippet: (A) Sequence architecture of TDP-43. Also shown are the sequence details, including the sequence of the C-terminal LCD. (B) Computed SEPs of each of the RRMs mapped onto the surface sites. RRM1 features patches of basic regions, and RRM2 features highly acidic faces. (C) Computed inter-domain interaction maps for TDP-43. (D) Representative RALS trace of TDP-43 showing a discontinuity at ∼0.12 µM. Results are shown from a single representative measurement (solid circles) and the spread over three technical replicates across different protein preps (pale envelope). On average, the value of c µ was found to be 0.18 ± 0.08 µM. (E) QF-DEEM images 0.7 µM TDP-43 at different magnifications. (F) A rendering of the biconcave structures seen in QF-DEEM images (see Methods ) uses a diameter of 8 units, the biconcave (dent) thickness as 1 units, and the maximum thickness of the biconcave structure near the rim as 2 units. The biconcavity thickness is larger than 0.5 used for SRSF1 (see ) to portray dents in TDP-43 structures that are less evident than SRSF1. (G) Histogram of diameters of TDP-43 microphases from analysis of 855 particle images. (H) Results from TRFQ measurements performed over a range of TDP-43 concentrations. The concentrations of microphases were fit to a line at the plateau value and the intercept yields an estimate of the number of TDP-43 molecules per microphase to be ∼13. (I) Mixture of 1 µM TDP-43 and 2 µM SRSF1 show co-localization of the two proteins in size-limited clusters. TDP-43 is observed to accumulate on the surface SRSF1 clusters. (J) Effect of TDP-43 on the c µ of SRSF1, which was measured at different molar ratios of TDP-43 and SRSF1.

Article Snippet: For covalent conjugation of SRSF1 or other SRSFs with NHS (h-hydroxysuccinimide ester) AlexaFluor 488 or AlexaFluor 205 (Thermofisher Scientific, USA), 1 mg of SRSF1 was dialyzed into labeling Buffer (10% glycerol, 20 mM HEPES (pH 8.3), 1.25 M NaCl) using Pierce Microdialysis Plates (ThermoFisher Scientific, USA).

Techniques: Sequencing

Journal: bioRxiv

Article Title: Nuclear speckle proteins form intrinsic and MALAT1 -dependent microphases

doi: 10.1101/2025.02.26.640430

Figure Lengend Snippet: A. In the presence of MALAT1 , the c µ , as extracted from RALS, of SRSF1 decreased from ∼0.45 µM to ∼0.22 µM. Conversely, the c µ of TDP-43 increased from ∼0.18 µM to 0.21 µM in the presence of MALAT1 . (B) TRFQ experiments, performed over a range of SRSF1 concentrations at a constant molar ratio of SRSF1-to- MALAT1 at 1-to-5×10 -4 , yielded an estimate of ∼13 SRSF1 molecules per microphase. (C) QF-DEEM images of SRSF1- MALAT1 microphases show spherical morphologies, which are different from the biconcave structures observed in the absence of MALAT1 . (D) Histograms collected across 118 particle images of the diameters (in nm) of the spheroids of SRSF1 microphases that form in the presence of MALAT1 . (E) QF-DEEM images of TDP-43+ MALAT1 microphases. The biconcave structures formed in the absence of MALAT1 are maintained in the presence of MALAT1 . (F) Histograms collected across 230 particle images of the diameters (in nm) of the spheroids of TDP-43 microphases that form in the presence of MALAT1 .

Article Snippet: For covalent conjugation of SRSF1 or other SRSFs with NHS (h-hydroxysuccinimide ester) AlexaFluor 488 or AlexaFluor 205 (Thermofisher Scientific, USA), 1 mg of SRSF1 was dialyzed into labeling Buffer (10% glycerol, 20 mM HEPES (pH 8.3), 1.25 M NaCl) using Pierce Microdialysis Plates (ThermoFisher Scientific, USA).

Techniques:

Journal: bioRxiv

Article Title: Nuclear speckle proteins form intrinsic and MALAT1 -dependent microphases

doi: 10.1101/2025.02.26.640430

Figure Lengend Snippet: (A) Schematic of a homopolymer that undergoes macrophase separation. Homopolymers or effective homopolymers are defined by uniformity of interactions (blue lines) that are on the order of thermal energy. Above a threshold concentration of the homopolymer, known as the saturation concentration (c sat ), a single micron-scale dense phase forms that coexists with the dilute phase. The inset shows a schematic of how some of the millions of molecules might be organized within the dense phase. (B) Schematic of a di-block copolymer featuring A-(purple) and B-blocks (green). A normalized interaction coefficient Δ XY is used to define the interactions between blocks. In the interaction matrix, the homotypic inter-block interactions Δ AA and Δ BB are attractive, whereas the incompatibility of the blocks leads to repulsive heterotypic interactions denoted by positive values of Δ AB . (C) An alternative scenario for the di-block system is one where the homotypic interactions are repulsive, positive values of Δ AA and Δ BB , whereas heterotypic interactions are attractive (negative values of Δ AB ). (D) Sequence architectures of hnRNP-A1 and SRSF1 showing the domain boundaries. (E) Mapping of the computed surface electrostatic potentials (SEPs) onto the surfaces of the RRMs from hnRNP-A1 and SRSF1. Regions in blue are positive potentials, regions in red are negative potentials, and regions in white are neutral. The magnitudes and signs of the site-specific SEPs are shown in Figure S1 . (F) Non-random sequence features within the LCD of hnRNP-A1 and IDR1 and IDR2 of SRSF1. The color-coded sequences from which the features were extracted are shown below the NARDINI+ analysis. The colors correspond to polar residues (green), positively charged residues (blue), negatively charged residues (red), glycine residues (magenta), aromatic residues (orange), and others (black). The inventory of sequence features is provided in the Methods section. A positive z-score implies either an enrichment of a compositional bias when compared to the human proteome or a non-random linear segregation of specific pairs of amino acid types. A negative z-score implies either a depletion of a compositional bias when compared to the human proteome or a non-random linear mixing of specific pairs of amino acid types. (G) Interaction coefficients Δ XY computed from atomistic simulations for each pair of distinct domains within SRSF1 (left) and hnRNP-A1 (right).

Article Snippet: pGEMT- MALAT1 was cloned by amplifying the MALAT1 cDNA from pCMV-MALAT1 and ligating it into the pGEM-T easy vector (Promega, USA) according to the manufacturer’s protocol. pET28-SRSF1 was cloned by sub-cloning SRSF1 cDNA from YFP-SRSF1 into the pET28 vector by restriction digestion and ligation. pET28-SRSF3 (Uniprot: P84103), pET28-SRSF5 (Uniprot: Q13243), and pET28-SRSF7 (Uniprot: Q16629) were generated by gene synthesis (GenScript, USA).

Techniques: Concentration Assay, Blocking Assay, Sequencing

Journal: bioRxiv

Article Title: Nuclear speckle proteins form intrinsic and MALAT1 -dependent microphases

doi: 10.1101/2025.02.26.640430

Figure Lengend Snippet: (A) Data from RALS measurements for SRSF1. The scattering intensity is plotted as a function of SRSF1 concentration. Results are shown from a single representative measurement (solid circles joined by dashed lines) and the spread over three technical replicates across different protein preps (pale blue envelope). The discontinuity in the scattering intensities at 0.45 ± 0.05 µM is denoted as c µ . (B) Decay rate constants ( k decay ) of the fluorescence lifetime of pyrene. The solid circles are rate constants from a single representative measurement. The pale blue envelope shows the range of values obtained from three technical replicates. The rate constant shows a plateauing behavior (horizontal dashed line) near the c µ (∼0.45 µM) estimated from RALS. The increase in the rate constant beyond the plateau value is proportional to the increase in the concentration of microphases with increasing [SRSF1]. (C) QF-DEEM images at three different magnifications show biconcave morphologies of SRSF1 microphases and their reproducibility across a large field-of-view. (D) Three-dimensional rendering of the morphologies observed in panel (C) . The method of rendering, described in Methods , uses a diameter of 8 units, the biconcave (dent) thickness as 0.5 units, and the maximum thickness of the biconcave structure near the rim as 2 units. (E) Histogram of diameters (in nm) constructed from the analysis of 639 distinct images. (F) Results from TRFQ experiments were fit as described in the Methods . The intercept along the ordinate is 49.7 ± 1.8 molecules, and it quantifies the number of SRSF1 molecules within a microphase at the c µ .

Article Snippet: pGEMT- MALAT1 was cloned by amplifying the MALAT1 cDNA from pCMV-MALAT1 and ligating it into the pGEM-T easy vector (Promega, USA) according to the manufacturer’s protocol. pET28-SRSF1 was cloned by sub-cloning SRSF1 cDNA from YFP-SRSF1 into the pET28 vector by restriction digestion and ligation. pET28-SRSF3 (Uniprot: P84103), pET28-SRSF5 (Uniprot: Q13243), and pET28-SRSF7 (Uniprot: Q16629) were generated by gene synthesis (GenScript, USA).

Techniques: Concentration Assay, Fluorescence, Construct

Journal: bioRxiv

Article Title: Nuclear speckle proteins form intrinsic and MALAT1 -dependent microphases

doi: 10.1101/2025.02.26.640430

Figure Lengend Snippet: (A) Schematic showing how clustering of SRSF1 microphases might arise via SALR-like interactions. An accumulation of surface charge shown by blue + marks will draw a counterion cloud (dashed red envelope) around each microphase (solid blue circles). Increasing the concentration of microphases increases the likelihood that microphases cluster via effective short-range attractions that result from the sharing of counterion clouds among pairs and higher-order clusters of microphases. The attractions saturate beyond a certain size because the buildup of charge creates long-range repulsions. Hence, clusters of microphases do not grow beyond a specific size. (B) MRPS measurements of SRSF1 above the c µ are shown as average and standard deviation about the mean. The data show peaks at a constant position while the amplitude increases with increasing concentration of SRSF1. The implication is that the mean sizes stay the same, while the abundance of the clusters increases with increasing concentrations of SRSF1. Results are shown from a single representative measurement (solid lines) and the spread over three technical replicates (pale envelopes). (C) Confocal microscopy at concentrations above c µ . With increased SRSF1 concentration, the abundance of SRSF1 clusters increases. The schematic depicts how microphases form clusters by touching to form incipient networks of clusters. (D) Box and whisker plot from analysis of QF-DEEM images yielded estimates for the number of microphases within a ∼380 nm sized cluster. The box extends from the 25 th to 75 th percentiles. The line in the middle of the box is at the median value of 23 microphases per cluster. The whiskers extend from the minimum to the maximum number of microphases detected in the probe spheres. (E) Single-molecule microscopy with photoactivatable JaneliaFluor 549 (PA-JF549)-labeled SRSF1 shows that the clusters are composed of multiple microphases. The image is produced as a maximum projection of 200 frames collected where the color bar shows the fluorescence intensity. (F) Histograms of fluorophores with trajectories of less than 50 nm (slow) or more than 50 nm (fast) over each 20 ms frame for 20,000 frames each with an inset for a zoomed-in field of view of the outlined area. The histograms are represented as heatmaps mapped onto the image. (G) Excess variance (see Methods ) computed over bins that quantify the displacements of slow-versus fast-moving molecules. (H) RALS extracted c µ values in the absence and presence of 200 mM L-arginine in experimental buffer are within the error. (I) Confocal microscopy images of AF488-SRSF1 in the presence of 50 mM, 100 mM, 200 mM, and 500 mM L-arginine in the experimental buffer. (J) GFP-tagged SRSF1 is expressed in X. laevis GVs by microinjection of mRNA and SRSF1-positive bodies are observed. (K) Merge of images collected simultaneously via brightfield microscopy and the intensity in the GFP channel (see S3J ) shows that the SRSF1 bodies are juxta identifiable Cajal bodies in the GVs. (L) Cumulative distribution functions (CDF) of the diameters of SRSF1-positive bodies remain essentially invariant at different levels of the injected mRNA. Latrunculin A (Lat-A) treatments (orange dotted) to oocytes injected with 100 ng/µL (orange) mRNA did not result in an increase in the size of the intra-GV SRSF1-positive bodies.

Article Snippet: pGEMT- MALAT1 was cloned by amplifying the MALAT1 cDNA from pCMV-MALAT1 and ligating it into the pGEM-T easy vector (Promega, USA) according to the manufacturer’s protocol. pET28-SRSF1 was cloned by sub-cloning SRSF1 cDNA from YFP-SRSF1 into the pET28 vector by restriction digestion and ligation. pET28-SRSF3 (Uniprot: P84103), pET28-SRSF5 (Uniprot: Q13243), and pET28-SRSF7 (Uniprot: Q16629) were generated by gene synthesis (GenScript, USA).

Techniques: Concentration Assay, Standard Deviation, Confocal Microscopy, Whisker Assay, Microscopy, Labeling, Produced, Fluorescence, Microinjection, Injection

Journal: bioRxiv

Article Title: Nuclear speckle proteins form intrinsic and MALAT1 -dependent microphases

doi: 10.1101/2025.02.26.640430

Figure Lengend Snippet: (A) Results from NARDINI+ analysis comparing the sequence features of IDRs in SRSF1, SRSF3, SRSF5, and SRSF7. For clarity, only the sequence features in which at least one sequence had a z-score greater than or equal to 3 are shown. The Ser / Arg rich IDRs (C-terminal IDR in SRSF3 and SRSF7 and IDR2 in SRSF1 and SRSF5) share similarities in terms of compositional biases toward Ser and Arg and the presence of Ser patches (red box). The two distinct IDRs are IDR1 of SRSF1 and the linker between the N-terminal RRM and C-terminal IDR of SRSF7. (B) Inter-domain interaction matrix of Δ XY coefficients for SRSF3. (C) Inter-domain interaction matrix of Δ XY coefficients for SRSF5. (D) Inter-domain interaction matrix of Δ XY coefficients for SRSF7. (E) Bar plot of threshold concentrations c µ for SRSF3, SRSF5, and SRSF7 extracted from RALS measurements. The data for SRSF1 are from the RALS measurements shown in and these are incorporated into the bar plot for comparison. The values of c µ increase in the order of SRSF1 (0.45 ± 0.05 µM) < SRSF7 (1.0 ± 0.14 µM) < SRSF3 (2.0 ± 0.14 µM) < SRSF5 (9.5 ± 0.71 µM). (F) The number of molecules per microphases determined from TRFQ experiments for the four SRSFs (see Figure S4F for values). (G) QF-DEEM images of 5 µM, SRSF3 which is above the measured c µ , shows a spheroidal morphology. (H) Three-dimensional rendering of the morphologies observed in panel (G) . The method of rendering, described in the Methods , uses a diameter of 8 units, the biconcave (dent) thickness as 5 units, and the maximum thickness of the biconcave structure near the rim as 5 units. (I) Histogram of diameters (in nm) constructed from the analysis of images of 269 distinct particles. (J) Summary of results from MRPS measurements of SRSFs over a range of protein concentrations. The abscissa is normalized by the intrinsic c µ of each protein for comparison across the SRSFs. The ordinate shows the diameter of clusters in nanometers. The average diameter is 355 nm for SRSF1 (dark blue), 295 nm for SRSF7 (light blue), 298 nm for SRSF5 (yellow), and 276 nm for SRSF3 (green). G. Sub-micron-scale clusters of microphases are observed by confocal microscopy for SRSF7, SRSF3, and SRSF5.

Article Snippet: pGEMT- MALAT1 was cloned by amplifying the MALAT1 cDNA from pCMV-MALAT1 and ligating it into the pGEM-T easy vector (Promega, USA) according to the manufacturer’s protocol. pET28-SRSF1 was cloned by sub-cloning SRSF1 cDNA from YFP-SRSF1 into the pET28 vector by restriction digestion and ligation. pET28-SRSF3 (Uniprot: P84103), pET28-SRSF5 (Uniprot: Q13243), and pET28-SRSF7 (Uniprot: Q16629) were generated by gene synthesis (GenScript, USA).

Techniques: Sequencing, Comparison, Construct, Confocal Microscopy

Journal: bioRxiv

Article Title: Nuclear speckle proteins form intrinsic and MALAT1 -dependent microphases

doi: 10.1101/2025.02.26.640430

Figure Lengend Snippet: (A) Computed interaction matrices for pairs of domains drawn from SRSF1 and SRSF3. (B) QF-DEEM images at different magnifications of nanoscale structures of microphases formed in mixtures of SRSF1 with SRSF3. (C) Three-dimensional rendering of the morphologies observed in panel (B) . The rendered images show the types of biconcave (same as 2D ) and spheroidal structures (same as 4H ) that were observed. (D) Histogram of diameters (in nm) constructed from the analysis of 1359 distinct particle images. (E) Confocal images of sub-micron scale structures formed in mixtures of SRSF3 and SRSF1. (F) Confocal images of sub-micron scale structures formed in mixtures of SRSF5 and SRSF1. (G) Confocal images of sub-micron scale structures formed in mixtures of SRSF7 and SRSF1. In panels (E), (F), and (G), the concentrations of SRSF3, SRSF5, and SRSF7 were below their c µ whereas the concentration of SRSF1 was above its c µ . (H) Impact of each of SRSF3, SRSF5, and SRSF7 on the c µ of SRSF1.

Article Snippet: pGEMT- MALAT1 was cloned by amplifying the MALAT1 cDNA from pCMV-MALAT1 and ligating it into the pGEM-T easy vector (Promega, USA) according to the manufacturer’s protocol. pET28-SRSF1 was cloned by sub-cloning SRSF1 cDNA from YFP-SRSF1 into the pET28 vector by restriction digestion and ligation. pET28-SRSF3 (Uniprot: P84103), pET28-SRSF5 (Uniprot: Q13243), and pET28-SRSF7 (Uniprot: Q16629) were generated by gene synthesis (GenScript, USA).

Techniques: Construct, Concentration Assay

Journal: bioRxiv

Article Title: Nuclear speckle proteins form intrinsic and MALAT1 -dependent microphases

doi: 10.1101/2025.02.26.640430

Figure Lengend Snippet: (A) Sequence architecture of TDP-43. Also shown are the sequence details, including the sequence of the C-terminal LCD. (B) Computed SEPs of each of the RRMs mapped onto the surface sites. RRM1 features patches of basic regions, and RRM2 features highly acidic faces. (C) Computed inter-domain interaction maps for TDP-43. (D) Representative RALS trace of TDP-43 showing a discontinuity at ∼0.12 µM. Results are shown from a single representative measurement (solid circles) and the spread over three technical replicates across different protein preps (pale envelope). On average, the value of c µ was found to be 0.18 ± 0.08 µM. (E) QF-DEEM images 0.7 µM TDP-43 at different magnifications. (F) A rendering of the biconcave structures seen in QF-DEEM images (see Methods ) uses a diameter of 8 units, the biconcave (dent) thickness as 1 units, and the maximum thickness of the biconcave structure near the rim as 2 units. The biconcavity thickness is larger than 0.5 used for SRSF1 (see ) to portray dents in TDP-43 structures that are less evident than SRSF1. (G) Histogram of diameters of TDP-43 microphases from analysis of 855 particle images. (H) Results from TRFQ measurements performed over a range of TDP-43 concentrations. The concentrations of microphases were fit to a line at the plateau value and the intercept yields an estimate of the number of TDP-43 molecules per microphase to be ∼13. (I) Mixture of 1 µM TDP-43 and 2 µM SRSF1 show co-localization of the two proteins in size-limited clusters. TDP-43 is observed to accumulate on the surface SRSF1 clusters. (J) Effect of TDP-43 on the c µ of SRSF1, which was measured at different molar ratios of TDP-43 and SRSF1.

Article Snippet: pGEMT- MALAT1 was cloned by amplifying the MALAT1 cDNA from pCMV-MALAT1 and ligating it into the pGEM-T easy vector (Promega, USA) according to the manufacturer’s protocol. pET28-SRSF1 was cloned by sub-cloning SRSF1 cDNA from YFP-SRSF1 into the pET28 vector by restriction digestion and ligation. pET28-SRSF3 (Uniprot: P84103), pET28-SRSF5 (Uniprot: Q13243), and pET28-SRSF7 (Uniprot: Q16629) were generated by gene synthesis (GenScript, USA).

Techniques: Sequencing

Journal: bioRxiv

Article Title: Nuclear speckle proteins form intrinsic and MALAT1 -dependent microphases

doi: 10.1101/2025.02.26.640430

Figure Lengend Snippet: A. In the presence of MALAT1 , the c µ , as extracted from RALS, of SRSF1 decreased from ∼0.45 µM to ∼0.22 µM. Conversely, the c µ of TDP-43 increased from ∼0.18 µM to 0.21 µM in the presence of MALAT1 . (B) TRFQ experiments, performed over a range of SRSF1 concentrations at a constant molar ratio of SRSF1-to- MALAT1 at 1-to-5×10 -4 , yielded an estimate of ∼13 SRSF1 molecules per microphase. (C) QF-DEEM images of SRSF1- MALAT1 microphases show spherical morphologies, which are different from the biconcave structures observed in the absence of MALAT1 . (D) Histograms collected across 118 particle images of the diameters (in nm) of the spheroids of SRSF1 microphases that form in the presence of MALAT1 . (E) QF-DEEM images of TDP-43+ MALAT1 microphases. The biconcave structures formed in the absence of MALAT1 are maintained in the presence of MALAT1 . (F) Histograms collected across 230 particle images of the diameters (in nm) of the spheroids of TDP-43 microphases that form in the presence of MALAT1 .

Article Snippet: pGEMT- MALAT1 was cloned by amplifying the MALAT1 cDNA from pCMV-MALAT1 and ligating it into the pGEM-T easy vector (Promega, USA) according to the manufacturer’s protocol. pET28-SRSF1 was cloned by sub-cloning SRSF1 cDNA from YFP-SRSF1 into the pET28 vector by restriction digestion and ligation. pET28-SRSF3 (Uniprot: P84103), pET28-SRSF5 (Uniprot: Q13243), and pET28-SRSF7 (Uniprot: Q16629) were generated by gene synthesis (GenScript, USA).

Techniques:

Journal: IBRO Neuroscience Reports

Article Title: Exploring the roles and clinical potential of exosome-derived non-coding RNAs in glioma

doi: 10.1016/j.ibneur.2025.01.015

Figure Lengend Snippet: Summary of exosomal circRNA in glioma.

Article Snippet: After inhibiting the expression of these three circRNAs in the U87 cell line, it was found that the proliferative activity of glioma was also impaired, and it is speculated that the above circRNAs could be used as feasible biomarkers for the detection of GBM ( ).Stella and colleagues screened from the ExoRBase database and compared the differences in circ-RNA within exosomes between healthy individuals and glioma patients, discovering that the tumor suppressor gene circSMARCA5 and the oncogene circHIPK3 are statistically significant and can serve as diagnostic biomarkers for GBM ( ).CircSMARCA5 can modulate the SRSF1/SRSF3/PTB signaling pathway to exert an inhibitory effect on the migration of glioma cells ( ).Exosome-derived circ-HIPK3 has been identified as a key factor in glioma growth.

Techniques: Migration