Journal: Nucleic Acids Research

Article Title: DeepCLIP: predicting the effect of mutations on protein–RNA binding with deep learning

doi: 10.1093/nar/gkaa530

Figure Lengend Snippet: DeepCLIP predicts increased TDP-43 binding as mechanism behind ACADM exon 6 skipping. ( A ) DeepCLIP TDP-43 profile across the 5′ss of ACADM exon 6 with wt indicated in black and patient mutation indicated in red. Along the first axis the sequence is shown and along the second axis the DeepCLIP BLSTM values are shown. SPRi oligo location and SSO locations are indicated in blue and red bars above and below the sequence, respectively. ( B ) Splicing of wt and mutant minigenes with either TDP-43 targeting siRNA or non-targeting siRNA determined by RT-PCR. ( C ) Western blot of TDP-43 and HPRT from siRNA and minigene transfected samples. ( D ) Splicing of wt and mutant minigenes treated with either a control SSO (Ctrl-SSO), SSO1, or SSO2 determined by RT-PCR. ( E ) DeepCLIP profile of short RNA oligos used in SPRi measurement, reference in black and +7A>G variant in red. ( F ) The difference in DeepCLIP binding profiles in (E) between reference and variant. Positive score indicates higher score in variant. ( G ) SPRi measurements of TDP-43 binding to the wt oligo in (E). ( H ) SPRi measurements of TDP-43 binding to the variant oligo in (E). In both (G) and (H), the black line indicates the fitted binding model.

Article Snippet: Surface plasmon resonance imaging (SPRi) by IBIS MX-96 was used to measure the kinetics of recombinant hnRNP A1 (ab224866, Abcam), SRSF1 (GenScript, Piscataway, NJ, USA) and TDP-43 (R&Dsystems, AP-190) binding to the immobilized RNA oligonucleotides.

Techniques: Binding Assay, Mutagenesis, Sequencing, Reverse Transcription Polymerase Chain Reaction, Western Blot, Transfection, Variant Assay

Journal: Nucleic Acids Research

Article Title: DeepCLIP: predicting the effect of mutations on protein–RNA binding with deep learning

doi: 10.1093/nar/gkaa530

Figure Lengend Snippet: DeepCLIP analysis of TDP-43 repressed pseudoexons indicate position-dependent tissue-specificity. ( A ) The average DeepCLIP TDP-43 profile scores of 58 neuron-specific and 79 muscle-specific pseudoexons activated in TDP-43-null mice in the areas covering the 25 first and last nucleotides of the pseudoexon, and the 50 nt spanning intronic regions. 95%-confidence intervals are indicated by shaded areas.

Article Snippet: Surface plasmon resonance imaging (SPRi) by IBIS MX-96 was used to measure the kinetics of recombinant hnRNP A1 (ab224866, Abcam), SRSF1 (GenScript, Piscataway, NJ, USA) and TDP-43 (R&Dsystems, AP-190) binding to the immobilized RNA oligonucleotides.

Techniques:

Journal: bioRxiv

Article Title: Pathological TDP-43 filaments accumulate at synapses and cause synaptic dysfunction

doi: 10.64898/2026.01.27.701787

Figure Lengend Snippet: A. Schematic summarising the extraction of TDP-43 filaments and their incubation with neuronal cultures. B. Immunogold negative-stain electron microscopy of extracted TDP-43 filaments from FTLD-TDP Type A patient brain, using a primary antibody against the N-terminus of TDP-43 and a secondary antibody conjugated to 10 nm gold particles. Scale bar, 100 nm. C. Immunoblot of extraction fractions and the final TDP-43 filament sample from FTLD-TDP Type A patient brain, probed with an antibody against pS409/410 TDP-43 (pTDP-43). The arrow indicates full-length (FL) TDP-43 and the bar indicates C-terminal fragments (CTFs) of TDP-43. D. Immunofluorescence confocal microscopy images of mouse primary cortical neurons incubated with (+ filaments) or without (- filaments) FTLD-TDP Type A patient-derived TDP-43 filaments for 7 d. Neurons were labelled using an antibody against class III β-tubulin (βIII-Tub; cyan) and TDP-43 filaments were labelled using the antibody against pTDP-43 (magenta). E. pTDP-43 signal (white) within images masked using the βIII-tubulin signal (yellow). Signal of 0.05-3 μm 2 is shown in cyan. This size filter was used to exclude noise and large extracellular TDP-43 filament accumulations from quantification in (F) . Scale bar, 20 µm. F. Quantification of βIII-tubulin-masked pTDP-43 signal from (E) at different timepoints. The total area of pTDP-43 signal as a percentage of the βIII-tubulin signal mask is plotted. Each data point represents one technical repeat. Data points are colour coded by biological replicate. n = 3 biological replicates. Means +/- SD are shown. A one-way ANOVA with Tukey’s multiple comparison test was performed, ***p<0.001, ****p<0.0001. G. Immunofluorescence confocal microscopy images of mouse primary cortical neurons incubated with FTLD-TDP Type A patient brain-derived TDP-43 filaments for 1 d. TDP-43 filaments were labelled with the antibody against pTDP-43 pre- (yellow) and post- (magenta) detergent permeabilisation. Neurons were labelled with an antibody against class III β-tubulin (βIII-Tub; cyan). Arrows indicate examples of pTDP-43 signal unique to permeabilised neurons. H. pTDP-43 signal (white) within images masked using the βIII-tubulin signal (yellow). Signal of 0.05-3 μm 2 shown in cyan. This size filter was used to exclude noise and large extracellular TDP-43 filament accumulations from quantification in (I) . Scale bars, 5 and 10 μm, as indicated. I. Quantification of βIII-tubulin-masked pTDP-43 signal pre- and post-permeabilisation from (H) . The total area of pTDP-43 signal as a percentage of the βIII-tubulin signal mask is plotted. Each pair of data points represent one technical repeat. n = 3 biological replicates. A paired t-test was performed, **p<0.01.

Article Snippet: The grids were then blotted and incubated with anti-TDP-43 primary antibody (Proteintech, 10782-2-AP; 1:25) diluted in blocking buffer for 3 h at 21°C.

Techniques: Extraction, Incubation, Staining, Electron Microscopy, Western Blot, Immunofluorescence, Confocal Microscopy, Derivative Assay, Comparison

Journal: bioRxiv

Article Title: Pathological TDP-43 filaments accumulate at synapses and cause synaptic dysfunction

doi: 10.64898/2026.01.27.701787

Figure Lengend Snippet: A. Schematic summarising the antibody-targeted proximity labelling strategy. TDP-43 filaments (red) were targeted using a primary antibody against pS409/410 TDP-43 (magenta). Horseradish peroxidase (grey) conjugated to a secondary antibody (green) was then used to catalyse the conversion of biotin-phenol to biotin-phenoxyl radicals (yellow) in the presence of hydrogen peroxide, thereby biotinylating proximal biomolecules. B. Immunofluorescence confocal microscopy images of mouse primary cortical neurons incubated with (+ filaments) or without (- filaments) FTLD-TDP Type A patient brain-derived TDP-43 filaments for 1 d, followed by antibody-targeted proximity labelling. TDP-43 filaments were labelled using an antibody against pS409/410 TDP-43 (pTDP-43; magenta), biotin was labelled using fluorescence-conjugated streptavidin (yellow) and neurons were labelled with an antibody against class III β-tubulin (βIII-Tub; cyan). Scale bars, 10 μm. C. Immunoblots of affinity purified biotinylated proteins from mouse primary cortical neurons (beads, b) and the flowthrough (f), probed with antibodies against biotin (left) and TDP-43 (right). Neurons were incubated with (+ filaments) or without (- filaments) FTLD-TDP Type A patient brain-derived TDP-43 filaments, in the presence or absence (- antibody) of the antibody against pTDP-43, followed by antibody-targeted proximity labelling.

Article Snippet: The grids were then blotted and incubated with anti-TDP-43 primary antibody (Proteintech, 10782-2-AP; 1:25) diluted in blocking buffer for 3 h at 21°C.

Techniques: Immunofluorescence, Confocal Microscopy, Incubation, Derivative Assay, Fluorescence, Western Blot, Affinity Purification

Journal: bioRxiv

Article Title: Pathological TDP-43 filaments accumulate at synapses and cause synaptic dysfunction

doi: 10.64898/2026.01.27.701787

Figure Lengend Snippet: A. Venn diagrams of the number of significantly-enriched proximal proteins from mouse primary cortical neurons and human ESC-derived cortical neurons incubated with FTLD-TDP Type A patient-brain derived TDP-43 filaments for 1 d and 3 d. B. Bar plot of numbers of significantly-enriched Gene Ontology (GO) Biological Process terms for proximal proteins from mouse primary cortical neurons and human ESC-derived cortical neurons incubated with FTLD-TDP Type A patient-brain derived TDP-43 filaments for 1 d and 3 d. C. Schematic summarising the significantly-enriched GO Cellular Compartment terms (red) for mouse primary cortical neurons and human ESC-derived cortical neurons incubated with FTLD-TDP Type A patient-brain derived TDP-43 filaments for 1 d and 3 d. D. Representative significantly-enriched proximal proteins from mouse primary cortical neurons and human ESC-derived cortical neurons incubated with FTLD-TDP Type A patient-brain derived TDP-43 filaments for 1 d and 3 d.

Article Snippet: The grids were then blotted and incubated with anti-TDP-43 primary antibody (Proteintech, 10782-2-AP; 1:25) diluted in blocking buffer for 3 h at 21°C.

Techniques: Derivative Assay, Incubation

Journal: bioRxiv

Article Title: Pathological TDP-43 filaments accumulate at synapses and cause synaptic dysfunction

doi: 10.64898/2026.01.27.701787

Figure Lengend Snippet: A. Representative significantly-enriched proximal pre- and postsynaptic proteins from mouse primary cortical neurons and human ESC-derived cortical neurons incubated with FTLD-TDP Type A patient-brain derived TDP-43 filaments for 1 d and 3 d. Presynaptic active zone complex proteins are highlighted in red. B. Immunofluorescence Airyscan confocal microscopy images of mouse primary cortical neurons incubated with FTLD-TDP Type A patient-derived TDP-43 filaments for 3 d. Neurons were labelled using antibodies against microtubule-associated protein 2 (MAP2; grey); the presynaptic active zone complex protein bassoon (yellow); and PSD95 (cyan). TDP-43 filaments were labelled using the antibody against pTDP-43 (magenta). The arrow indicates an example of pTDP-43 signal colocalising with bassoon. Scale bars, 5 µm and 1 µm, as indicated. C. Percentage of MAP2-masked pTDP-43 signal colocalised with bassoon (left) and PSD95 (right), and vice versa, from (C) in three-dimensional space. Each data point represents one technical replicate. n= 2 biological replicates. Means +/- SD are shown.

Article Snippet: The grids were then blotted and incubated with anti-TDP-43 primary antibody (Proteintech, 10782-2-AP; 1:25) diluted in blocking buffer for 3 h at 21°C.

Techniques: Derivative Assay, Incubation, Immunofluorescence, Confocal Microscopy

Journal: bioRxiv

Article Title: Pathological TDP-43 filaments accumulate at synapses and cause synaptic dysfunction

doi: 10.64898/2026.01.27.701787

Figure Lengend Snippet: A. Immunoblot of synaptosome isolation fractions (s, supernatant; cp, cell pellet; cyt, cytosolic; syn, synaptosomes) from mouse primary cortical neurons incubated with (+ filaments) and without (- filaments) FTLD-TDP Type A patient brain-derived TDP-43 filaments for 3 d, probed with antibodies against postsynaptic density protein 95 (PSD95, top), pS409/410 TDP-43 (pTDP-43, middle) and TDP-43 (bottom). The arrows indicate full-length (FL) TDP-43 and C-terminal fragments (CTFs) of TDP-43. B. Denoised tomographic slices of synaptosomes from mouse primary cortical neurons incubated with FTLD-TDP Type A patient brain-derived TDP-43 filaments for 3 d. TDP-43 filaments (magenta arrows), example synaptic vesicles (cyan arrows) and the postsynaptic density (yellow arrow) are indicated. C and D. Denoised tomographic slices showing TDP-43 filaments (magenta arrows). making contacts (green arrows) with synaptic vesicles (cyan arrows) (C) and the presynaptic plasma membrane (yellow arrows) (D) . B-D. Scale bars, 50 nm.

Article Snippet: The grids were then blotted and incubated with anti-TDP-43 primary antibody (Proteintech, 10782-2-AP; 1:25) diluted in blocking buffer for 3 h at 21°C.

Techniques: Western Blot, Isolation, Incubation, Derivative Assay, Clinical Proteomics, Membrane

Journal: bioRxiv

Article Title: Pathological TDP-43 filaments accumulate at synapses and cause synaptic dysfunction

doi: 10.64898/2026.01.27.701787

Figure Lengend Snippet: A. Representative false-coloured images of mouse primary cortical neurons expressing SypHy prior to stimulation (Rest), during stimulation with a train of 300 action potentials delivered at 10 Hz (10 Hz), following stimulation (Recovery) and during challenge with ammonium chloride (NH 4 ) buffer to reveal the total synaptic vesicle pool. Arrows indicate examples of terminals that respond to stimulation. Scale bar, 2 μm. B. Time course of relative SypHy fluorescence intensity (ΔF/F0) in mouse primary cortical neurons incubated with FTLD-TDP Type A patient brain-derived TDP-43 filaments (+ filaments) or with aged-matched control brain extracts (+ control) for 3 d, normalised to challenge with NH 4 buffer (total synaptic vesicle pool). The bar indicates the period of stimulation. n = 4 biological replicates. Means +/- SEM are shown. C. Rate of activity-dependent relative SypHy fluorescence increase (ΔF/F0) over the first 6 seconds of stimulation (stimulation triggered at 30 s). n = 4 biological replicates. Means +/- SEM are shown. Linear regression analysis was performed (lines), **p=0.0055. D. Representative traces of excitatory postsynaptic currents (EPSCs) evoked by paired presynaptic stimuli (50 ms interval) from mouse primary hippocampal neurons incubated with (+ filaments) or without (- filaments) FTLD-TDP Type A patient brain-derived TDP-43 filaments or with aged-matched control brain extracts (+ control) for 7 d. E. Plot of the ratios of the second to the first EPSC (paired-pulse ratio; PPR) from mouse primary hippocampal neurons incubated with (+ filaments) or without (- filaments) FTLD-TDP Type A patient brain-derived TDP-43 filaments or aged-matched control brain extracts (+ control) for 3-4 d (d3/4) and 6-7 d (d6/7). Each data point represents one paired recording. Data points are colour coded by biological replicate. n = 3 biological replicates. Means +/- SD are shown. A two-way ANOVA with Tukey’s multiple comparison test was performed, *p<0.05, ****p<0.0001. F. Widefield fluorescence microscopy images of mouse primary cortical neurons expressing the fluorescent calcium indicator jRGECO1a (red) over 1200 ms. Images were taken at 100 ms intervals. Scale bar, 100 μm. G. The spontaneous firing frequency calculated from (F) for mouse primary cortical neurons incubated with (+ filaments) or without (- filaments) FTLD-TDP Type A patient brain-derived TDP-43 filaments or aged-matched control brain extracts (+ control) for 9 d (d9) and 14 d (d14). Each data point represents one time-lapse recording. Data points are colour coded by biological replicate. n = 5 biological replicates. Means +/- SD are shown. A two-way ANOVA with Tukey’s multiple comparison test was performed, **p<0.01, ****p<0.0001.

Article Snippet: The grids were then blotted and incubated with anti-TDP-43 primary antibody (Proteintech, 10782-2-AP; 1:25) diluted in blocking buffer for 3 h at 21°C.

Techniques: Expressing, Fluorescence, Incubation, Derivative Assay, Control, Activity Assay, Comparison, Microscopy