Structural Features of Pathogenic Aggregates Correlate with Cell Pathology in Differentiated Cells

Structural Features of Pathogenic Aggregates Correlate with Cell Pathology in Differentiated Cells | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Structural Features of Pathogenic Aggregates Correlate with Cell Pathology in Differentiated Cells Vered Raz, Sander Mallon, Erik Bos, Vahid Sheikhhassani, Milad Shademan, and 3 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5676243/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract The accumulation of intracellular protein aggregates is a hallmark of aging. In hereditary adult-onset neuromuscular diseases (NMDs), these aggregates are not only characteristic but also pathogenic, marking age-related neuromuscular disorders. The transition from age-associated non-pathogenic aggregates to disease-driving pathogenic aggregates remains poorly understood. Poly(A) binding protein nuclear 1 (PABPN1) forms non-pathogenic nuclear aggregates in post-mitotic aged cells. However, a short trinucleotide expansion in PABPN1 leads to muscle dysfunction in Oculopharyngeal Muscular Dystrophy (OPMD), where insoluble nuclear aggregates in skeletal muscle become a defining disease feature. Combining an array of advanced imaging modalities, we examined the morphological differences between nuclear aggregates formed by non-pathogenic and pathogenic PABPN1 variants. Through micro- to nanoscale analyses, we identified key structural differences in the aggregation propensity of these variants in both differentiated and undifferentiated muscle cells and linked these differences to mRNA cellular dysfunctions. Our findings provide new insights into the structural distinctions between pathogenic and non-pathogenic aggregates and their implications for cellular dysfunction in neuromuscular diseases. Biological sciences/Neuroscience/Molecular neuroscience Biological sciences/Cell biology/Cellular imaging/Super-resolution microscopy Biological sciences/Structural biology/Electron microscopy/Cryoelectron microscopy protein aggregates aggregates structure imaging facilities muscle OPMD Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Introduction Misfolded proteins are normally degraded by the cell's quality control systems, such as proteasomes and autophagy pathways. However, in diseases such as neurodegeneration and some forms of muscular dystrophy, mutations lead to proteins that resist degradation 1 . Loss of proteostatic homeostasis is a natural process that leads to protein aggregation during aging 2 – 4 . Over time, the efficiency of proteostatic mechanisms declines, leading to an increase in protein misfolding and aggregation 5 . Protein aggregation is also a hallmark of several genetic diseases, particularly neurodegenerative and neuromuscular diseases, in which the mutated encoded proteins form aggregates 1 . These proteins accumulate and form pathogenic aggregates that contribute to cell dysfunction. The pathogenic protein aggregates can be cytoplasmic, such as in Parkinson's disease and Alzheimer's disease, or nuclear as found in polyglutamine repeat expansion disorders, including Huntington’s disease (HD) and Spinocerebellar ataxia diseases (SCAs), or polyalanine repeat disorders expansion like Oculopharyngeal muscular dystrophy (OPMD) 6 . OPMD is caused by a short expansion mutation in the gene encoding poly(A)-binding protein nuclear 1 (PABPN1) 7 . Wild type PABPN1 has a repeat of 10 alanine residues at the N-terminus, but the expansion results in the mutant protein containing an expansion of 11–18 alanine residues. PABPN1 is ubiquitously expressed and essential in all eukaryotic cells, but symptoms are limited to skeletal muscles. Both wild-type and the expanded PABPN1 are prone to aggregate but only aggregates of the mutant protein are pathogenic 8 . Discriminating between structural features of the pathogenic PABPN1 aggregates from the non-pathogenic form has been addressed in previous studies using mitotic, non-muscle cells 9 , 10 , but has not captured the nuances of aggregation in post-mitotic cells, and its effect on muscle cell biology 11 . Understanding the differences between pathogenic and non-pathogenic aggregated proteins is crucial for deciphering how protein aggregation in disease differs from natural aging. This knowledge can inform the development of targeted, disease-specific therapies. In this study, we investigated the structural distinctions between pathogenic and non-pathogenic PABPN1 protein variants using a human muscle cell model. We generated stable muscle cells expressing the wild-type PABPN1 (Ala10) or the pathogenic allele (Ala16) under the tetracycline-inducible promoter to bypass PABPN1 toxic effect in mitotic cells. We compared the structure of pathogenic and non-pathogenic PABPN1 aggregates using four imaging modalities, ranging from micrometer to nanometer resolution and linked to muscle cell biology, to provide new structural insights into the structure-pathogenesis relationship of PABPN1. Materials and Methods PABPN1 constructs and lentivirus production The full length PABPN1 wild type (Ala10) or the expanded PABPN1 (Ala16) differentiated to YFP, previously described in 9 were cloned into the pCW57-MCS1-2A-MCS2 doxycycline (Dox) inducible lentiviral vector (Addgene plasmid #71782). Cloning was confirmed by sanger sequencing. Lentivirus production was performed as detailed in 12 . Lentivirus particle titers were determined in HeLa cells. Cell culture Cells were cultured in growth medium (F10 (Gibco) medium supplemented with 15% FCS, 1 ng/ml bFGF, 10 ng/ml EGF and 0.4 µg/ml Dexamethasone). Cells were propagated in confluence 50–80%. Cell cultures did not reach 100% confluence to avoid spontaneous differentiation. Cell differentiation was induced at high confluency (85–95%) in DMEM + 2% horse serum for 3–5 days. The 2417 immortal human muscle cells were transduced with lentiviruses encoding Ala10-YFP or Ala16-YFP, and stable cell cultures were created using puromycin selection. Induction of the PABPN1-YFP transgene was carried out with 4 mg/ml doxycycline hyclate (D5207, Sigma Aldrich). For high content screening (HCS), cells were seeded in a Nunc 96 well plate; for live cell confocal microscopy, cells were seeded in either a µ-Slide 15 Well 3D ibiTreat (81506, IBIDI) or a µ-Slide 8 Well high ibiTreat (80806, IBIDI) slide. For electron microscopy, cells were seeded in µ-Dish 35 mm, high Grid-500 ibiTreat (81166, IBIDI) dishes. For refractive index imaging, cells were cultured on a µ-Dish 35 mm, low ibiTreat (80136, IBIDI) dish. Cell cultures were treated with 0.5 µM epoxomicin (Sigma-Aldrich #134381-21-8) for 6 hours; Leptomycin B (LMB) 20nM for 4 hours; or 20 µg/ml cycloheximide (Sigma-Aldrich # 01810) for one hour. Cells were incubated in the growth medium. Protein extraction and Western blotting Proteins were lysed from cells using RIPA buffer (20 mM Tris, pH 7.4, 150 mM NaCl, 5 mM EDTA, 1% NP40, 5% glycerol and 1 mM DTT and protease inhibitor cocktail). After sonication and centrifugation (1 min, 13000g, at 4°C), the supernatant containing the soluble proteins was transferred to a new tube and the pellet, containing the insoluble proteins, was washed once in PBS, dissolved in loading buffer, sonicated and spin down prior to heat inactivation. Protein aliquots were separated on 10% SDS-PAGE. Western blotting was carried out with a PVDF membrane. Bulk proteins were visualized with the No-Stain Protein Labeling Reagent (#A44717, ThermoFisher) and imaged using the iBright Imaging System (ThermoFisher). The membrane was blocked with 5% dried milk powder (T145.2, Carl Roth), first antibody incubation was carried out at 4 degrees overnight, and secondary antibody incubation at room temperature for one hour. Antibodies are listed in Table S1 . An Odyssey CLx Infrared imaging system (LiCOR, NE. USA) was used to detect the fluorescent signal. Quantification of protein abundance was done using ImageJ. Values were corrected for background and normalized to loading controls. Western blot quantification was carried out with ImageJ. Normalization was made for both the No-Stain and house-keeping signal. All Western blotting experiments were repeated 6 times. Immunohistochemistry Insoluble PABPN1 was detected in 1M KCl pre-treatment for 15 minutes. Immunohistochemistry with or without KCl pre-treatment was performed using standard procedures: fixation (2% formaldehyde in PBS) for 5 minutes, permeabilization (1% triton in PBS) for 10 minutes, PBS washing, incubation with primary antibodies for one hour at room temperature, PBS washing, incubation with secondary antibodies + DAPI (4′,6-Diamidino-2-phenylindole dihydrochloride) for 30 minutes, and PBS washing. Cells were kept in PBS during imaging. Antibodies are listed in Table S1 . Cellular Assays All cellular assays were repeated 3 times in biological triplicates. The results shown in figures are from a representative experiment in biological triplicates. The Mitochondrial activity assays were performed in differentiated or proliferating cell cultures grown in 96 well plate and treated with Tetramethylrhodamine methyl ester perchlorate (TMRM). Cell cultures grown in 96 well plates were washed with PBS and incubated with a staining solution (5nM TMRM (Sigma-Aldrich #115532-50-8) and Hoechst were diluted in growth media and incubated for 45 minutes. After twice PBS washing, cells were kept in differentiation media during imaging. The Protein Synthesis Assays were performed in differentiated or proliferating cell cultures grown in 96 well plate. The protein synthesis assay kit (Cayman Chemicals #601100) was employed according to the manufacturer protocol, with the following modifications: azido- O -propargyl-puromycin (OPP)-488 was replaced with OPP-555 (Vector laboratories, #CCT-1494). For the negative control, 30 minutes pre incubation with 20µM cycloheximide was used. Hoechst was added after fixation. For RNA hybridization with Oligo-dT, differentiated cell cultures were fixed using 3.7% FA for 15 minutes at RT. After two PBS washes the cells were incubated in Protease III diluted 1:30 in PBS (#322337 Advanced Cell Diagnostics) for 15 minutes at RT. After twice PBS washes cells were incubated in hybridization buffer (#10369 Cepham Life Sciences) for 15 minutes at RT. Incubation with 5’-Cy5-Oligo d(T)12–18 probe (#26-4400-02 Gene Link), diluted 1:1000 in hybridization buffer, was carried out overnight at 40 degrees in a humidified chamber. The following day, washes were carried out at 40 degrees for 5 minutes with 4x, 2x, 1x SSC buffer, and with PBS. Finally, the cells were incubated with Hoechst and kept in PBS during imaging. For differentiation index calculation, cells in 96 well plate were treated with Dox for 24 hours and then were incubated in differentiation medium for 72 hours. Differentiated cells were marked for MyHC expression using Immunohistochemistry procedure and the MF20 antibody. Imaging and Image quantification The CellInsight CX7 LZR high content screening (HCS) platform was used for imaging and image quantification using the accompanied HCS toolbox spot detector and co-localization (ThermoFisher Scientific). Between 2000–12000 nuclei objects were imaged and each experiment were made in 3–4 biological replicates, at least two times. Cells were imaged with 405 nm (DAPI), 488 nm (YFP) filters, and per cellular assay with the following filters: imaging: TMRM with 560 nm, OPP-555 with 560 nm, oligo-dT-Cy5 with 647 nm, and MF20 antibody with 647 nm. Imaging for calculation of differentiation index was made with a 10x objective covering over 12,000 nuclei per replicate, and imaging for nuclear YFP quantification with a 20x objective, covering at least 5000 nuclei per replicate. The co-localization toolbox was used for the quantification of differentiation index by the percentage of myonuclei without MyHC objects, and the spot detection toolbox for the YFP puncta, TMRM, OPP-555 and oligo-dT-Cy5. Bulk YFP intensity was considered with a low threshold (50–150) and the high threshold of puncta (650–1200). The exact threshold was adjusted per experiment, per experiment both low and high thresholds were quantified. Analysis of both TMRM and 555-OPP MFI was made from the perinuclear region. Oligo-dT signal and overlap with YFP was measured from both nuclear and perinuclear regions in YFP positive myonuclei. Confocal imaging imaging of single nuclei was made with Leica SP8 confocal microscope using a 63 x/1.3 oil objective and HyD detectors, or the Dragonfly spinning disc module using a 40x/1.3 or 63x/1.3 oil immersion objective. In fixed cells, nuclear counterstain was with DAPI (Sigma–Aldrich # D9542). In living cells nuclear counterstain was with Sir700-DNA kit (Spirochrome # SC015). Imaging of DAPI was at 405nm, YFP – 488nm, anti-Cy5-647 and Sir-700 at 700nm. Imaging settings including exposure time, laser power, the excitation-emission range, and Z-stacks step size were consistent between Ala10 and Ala16 within an experiment. Time-lapse imaging was made with Z-stack acquisition taking 4.20 minutes (1 frame per 2 seconds). Maximum-projection of Z-stacks was made with Imaris. Experiments were repeated 3 times. The results shown in figures are from a representative experiment. Quantifications of confocal images were carried out in ImageJ (v1.54f). 1. Overlay of time-lapse images was made from frames at 0, 130, and 260 seconds. Each frame was assigned a distinct RGB color, and an overlap was made with imageJ. 2. YFP puncta quantification was carried out with a macro in ImageJ. The YFP channel was processed with a Gaussian blur (1.5 sigma (radius) ‘Blur’ and ‘Despeckle’ functions. Masking of the YFP puncta was applied with a constant threshold across all images within an experiment. The threshold was manually determined to match the YFP puncta (examples are in Figure S1 A). Particles > 0.01 µm 2 in area were considered for analysis. From each punctum, the mean fluorescence intensity (MFI), the area, and the circularity were recorded. Analysis was conducted on gated myonuclei from differentiated or undifferentiated cells (single nucleus). 3. Oligo-dT signal analysis was carried out on gated myonuclei in differentiated cells in ImageJ. YFP puncta and oligo-dT analysis was carried out with a Gaussian blur of 1.0, per fluorophore, a constant threshold was used for all images (Figure S1 B). The MFI and area were recorded. The overlap and correlation between YFP and oligo-dT were assessed with the JACoP plugin 13 in ImageJ, using the M1 & M2 coefficients and the Pearson correlation. Transmission electron microscopy – Differentiated cell cultures were fixed in 1.5% glutaraldehyde in 0.1 M Sodium Cacodylate buffer for 2 hours and were successively incubated in 1% Osmium Tetroxide in 0.1 M cacodylate buffer for 1 hour and in 1% Uranyl Acetate in water for 1 hour. The cells were then dehydrated through a series of incubations in Ethanol (70–100%) for 90 minutes and embedded in Epon. The flat embedded cells were sectioned with an ultramicrotome (UC6, Leica, Vienna) using a 35 degrees diamond knife (Diatome, Biel, Switzerland) at a nominal section thickness of 90 nm. The sections were transferred to a formvar, and carbon coated 1×2 mm copper slot grid and stained for 20 minutes with 7% uranyl acetate in water and for 10 minutes with lead citrate. EM images were recorded using a Tecnai 12 electron microscope (Thermo Fisher Scientific) equipped with an EAGLE 4k×4k digital camera. For navigation on EM images, montages of images at 11,000× were generated using stitching software 14 . Morphology was assessed by sampling 100 nuclei on stitched EM images. For 3D reconstruction, 8 consecutive serial sections with a nominal thickness of 200 nm were manually aligned, segmented using the software program, Ais 15 and rendered as 3D isosurfaces in ChimeraX 16 . For correlative light and electron microscopy, cells were grown on a gridded µ-Dish 35 mm plate and stained with DAPI. The living cells were imaged in an EVOS FL Digital Inverted Fluorescence Microscope (Invitrogen) equipped with a 20× objective. YFP was visualised with a GFP filter and DAPI with a UV filter. After imaging in the light microscope, the cells were processed for electron microscopy as described above. Superimposition and correlation of light and electron microscopy images was performed using Photoshop. In Photoshop the LM image was copied as a layer into the EM image and made 50% transparent. The LM image required transformation to align with the broader scale of the EM image. This involved isotropic scaling and rotation. Alignment was facilitated by utilizing the nuclear DAPI staining alongside cell morphology. Holo-tomographic microscopy (HTM) , in combination with epifluorescence, was performed on the 3D Cell-Explorer Fluo (Nanolive, Ecublens, Switzerland) using a 60× air objective (NA = 0.8) at a wavelength of λ = 520 nm (Class 1 low power laser, sample exposure 0.2 mW/mm2) and CMOS Sony sensor, with quantum efficiency (typical) 70% (at 545 nm), dark noise (typical) 6.6 e-, dynamic range (typical) 73.7 dB, field of view 90 × 90 × 30 µm, axial resolution 400 nm, and maximum temporal resolution 0.5 3D RI volume per second. Acquired RI images were processed with built-in software (Nanolive). ImageJ/Fiji ( https://imagej.nih.gov/ ) was used for the final processing and quantifications. Experiments were performed in two independent experiments, and the reported results are from one representative experiment. Nuclei texture analysis - RI images of each cell type were initially converted to the 8-bit format using ImageJ. Subsequently, the areas, including the cell nuclei, were outlined, and extracted using a free-hand selection tool. Following this, the texture of these selected areas was analyzed utilizing the GLCM (Gray Level Co-occurrence Matrix) texture analysis plug-in (version 0.4) developed by Julio E. Cabrera. This plug-in facilitated the computation of various statistical parameters associated with the GLCM of the image, including the Inverse Difference Moment (IDM) and Entropy. Statistics and online analysis Statistical tests were performed in GraphPad Prism. AlphaFold predictions were carried out in AlphaFold 3.0 https://alphafoldserver.com Results Ala16-YFP aggregation is higher than Ala10-YFP in an inducible muscle cell model. To investigate differences in aggregation between a pathogenic and non-pathogenic form in a disease-relevant cell model, our study was performed in human muscle cells. We generated cells stably expressing the Ala10 wild-type PABPN1 or the Ala16 extended PABPN1 tagged with YFP (hereafter referred to as Ala10 and Ala16, respectively). Both transgenes were expressed under the tetracycline-inducible promoter to eliminate PABPN1 toxicity due to overexpression. Western blot analysis confirmed the inducible expression of both transgenes after doxycycline (Dox) treatment (Fig. 1 A). Both Ala10-YFP and Ala16-YFP proteins accumulated in the insoluble fraction, but A16-YFP accumulation was 1.5 times higher than A10-YFP (Fig. 1 B). In addition, the average ratio of insoluble to soluble protein levels was 3.5 times higher for A16-YFP than for A10-YFP (Fig. 1 B). We next examined whether transgene overexpression was associated with endogenous PABPN1 levels and found a higher ratio in Ala16 compared to Ala10 or in vehicle cultures (Fig. 1 C). This suggests that an increase in insoluble PABPN1 depletes the levels of the soluble form. Image quantification confirmed nuclear accumulation of both Ala10 and Ala16 (Fig. 1 D). Consistent with Western blot analysis, YFP fluorescence intensity was 2-fold higher in Ala16 puncta than in Ala10 puncta (Fig. 1 E). To verify that the YFP puncta corresponded to PABPN1 aggregates, we treated live cells with KCl prior to imaging. KCl treatment improves solubility and stabilizes aggregates of hydrophobic polypeptides 17 , and has been used to detect PABPN1 aggregates in OPMD models 9 , 18 . KCl treatment in the muscle cell model resulted in higher YFP intensity compared to untreated cell cultures (Fig. 1 F and Figure S2). Notably, the difference between KCl-treated and untreated cells was greater in Ala10 than in Ala16, indicating that the YFP signal in Ala16 predominantly represents aggregated proteins. To identify PABPN1 aggregates without KCl treatment, we applied two thresholds during image quantification, which allowed discrimination between bulk signal at low threshold and signal in puncta at high threshold (Fig. 1 G). Consistent with the KCl treatment, a greater difference between low and high thresholds was found for Ala10 (Fig. 1 H). This analysis demonstrates that PABPN1 aggregates can be distinguished from the soluble protein puncta by image quantification, which may implicate structural differences between pathogenic and non-pathogenic protein aggregates. Structural features discriminate between non-pathogenic and pathogenic PABPN1 aggregates The PABPN1 protein comprises a poly-alanine stretch within the N-terminal intrinsically disordered region (IDR), a coiled coil domain (CCD), and a C-terminal RNA recognition motif (RRM) (Figure S3A). Logicoil predicts the CCD to form a tetrameric coiled coil 19 , which has been shown to be critical for aggregation 20 , 21 . The N-terminal alanine stretch alone is not critical for aggregation 20 , suggesting that folding of the contiguous IDR + CCD region may form a stable structure leading to aggregation. AlphaFold3 22 prediction of the monomeric IDR + CCD predicts, with low confidence, that the alanine tract in Ala16 folds on the CCD (Figure S3B), which may be more stable than the Ala10 structure. Taking the Logicoil prediction of a stable CCD tetramer, AlphaFold3 predicts that the wild type (Ala10) IDR N-terminus is parallel to the tetrameric coiled-coil domain. However, the Ala16 expansion is consistently predicted to intercalate with the tetrameric CCD (Figure S3C). These predictions hint at a structural role for the Ala16 expansion that might lead to pathogenesis, although the mechanism is still unclear. Considering the limitations of AlphaFold for IDR prediction 23 , the difference between Ala10 and Ala16 justified the investigation of structural features using puncta structure analysis. Structural features of PABPN1 puncta were assessed in confocal images from single nuclei (Fig. 2 A), and YFP puncta were segmented with a constant threshold. YFP puncta segmentation was made at two thresholds: a low threshold that visually matched the Ala10 and a high threshold that visually matched the Ala16 puncta but was too high for the Ala10 puncta (Fig. 2 A). A quantitative assessment of puncta segmentation confirmed the visual evaluation and showed that the average number of puncta per nucleus was significantly higher in Ala16 compared to Ala10 (Fig. 2 B). For comparative studies of puncta features, we considered the low threshold. The average puncta area per nucleus was 2.5-fold larger in Ala16 than in Ala10, whereas the average circularity was smaller in Ala16 (Fig. 2 D). The intranuclear variability of puncta area and circularity were larger in Ala16, indicating high heterogeneity in puncta structure (Fig. 2 C and 2 D). The aggregation process was assessed by the relation between puncta area and circularity. In Ala10 puncta, larger puncta remained with a similar circularity (Fig. 2 E). In contrast, Ala16 puncta showed a negative correlation between area and circularity (Fig. 2 E). Together, this suggests that Ala16 aggregation is disorganized and heterogeneous, while the Ala10 aggregates keep their shape during growth. As the quantification of puncta structure could be affected by imaging and image processing, we verified the differences in puncta structural features using refractive index (RI) imaging combined with a fluorescence imaging platform using NanoLive holo-tomographic microscopy. This allowed us to simultaneously acquire morphological and molecular density information in selected cells (Fig. 2 F). Our imaging analysis showed that, per nucleus, the average fluorescence intensity in puncta was higher, size was larger, and circularity was lower in Ala16 compared to Ala10 (Fig. 2 G). Overall, the results from combined RI/fluorescence imaging are consistent with those obtained from confocal imaging. Taken together, nuclei in Ala16 cells were more densely packed with protein aggregates than in Ala10. This consistency across different imaging modalities and analytical techniques increases the robustness of these observations. To verify the structure of the aggregates, we used transmission electron microscopy (TEM) to characterize PABPN1 aggregates in differentiated cells at nanometer resolution. A stitched image of a differentiated cell created with in-house software 14 , allowed for detailed analysis of the entire differentiated cell (Fig. 3 A). In the Dox-treated cell cultures, we observed electron-dense structures of variable morphology. These structures, which were absent in the vehicle control myonuclei, had a different electron density than the nucleoli (Fig. 3 B). We classified the myonuclei based on the morphology of these electron density structures; ‘punctate’ morphology was smaller and circular (cyan circles in Fig. 3 B), and ‘bouquet’ morphology was larger and not circular (encircled in blue, Fig. 3 B). To confirm that these novel structures corresponded to PABPN1-YFP aggregates, we used correlative light and electron microscopy (CLEM). We correlated and overlayed fluorescence images with TEM images showing that each electron-dense aggregate corresponded to a YFP signal (Fig. 3 C, circled in pink). Some YFP signals did not correspond to an electron-dense structure because the aggregate was above or below the 90 nm thick section imaged by electron microscopy (Fig. 3 C, encircled in orange). Some punctate aggregates showed a toroidal architecture (Fig. 3 B, cyan circles). Sectioning through 3D spherical, cylindrical, or toroidal structures may result in a torus in the resulting 2D section. To determine the morphology of these aggregates, we imaged 8 consecutive 200 nm thick sections using TEM for 3D reconstruction (Figs. 3 D and S4). This revealed that the toroidal structures represent hollow spheres of aggregated material, and that these spheres can further aggregate into larger structures. To assess the structural differences between Ala10 and Ala16 we manually scored the aggregate structure in 90 nm thick sections of 100 nuclei. Bouquet morphology and found only in Ala16, and in 2/3rd of the nuclei with electron-dense structures (Fig. 3 E). In around 50% of the nuclei in both Ala10 and Ala16 myonuclei, electron-dense structures were not found (Fig. 3 E), which is consistent with the 3D reconstruction analysis of PABPN1 nuclear aggregates. Taken together, confocal, holographic tomography and TEM imaging consistently confirm the structural differences, both in area and circularity, between Ala10 and Ala16 aggregates. Aggregation differs between non-differentiated and differentiated muscle cells Since PABPN1 aggregation is pathogenic in multinucleated muscle fibers, we investigated whether Ala10 and Ala16 aggregation differs between differentiated and non-differentiated muscle cells. In culture, high cell density and starvation stress drive muscle cell fusion into multinucleated cells, characterized by the expression of the myosin heavy chain (MyHC). Induction of Ala10 or Ala16 reduces the differentiation index compared to vehicle-treated cells, and the differentiated cells in Ala16 cultures detached faster compared to Ala10 cultures (Figure S5). Therefore, to assess aggregation in differentiated cells, the following protocol was used: cells were cultured in differentiation condition for 48 hours, followed by Dox treatment for 12 to 72 hours. Cell cultures were fixed, immunolabelled with MyHC and imaged by confocal microscopy (Fig. 4 A). The Ala16-YFP signal appeared after 12 hours of Dox treatment, earlier than in Ala10-YFP cells (Fig. 4 A). At 72 hours after Dox treatment, Ala10-YFP fluorescence was stronger in MyHC-negative cells compared to the nuclei within the MyHC-positive cells (Fig. 4 A). To assess the statistical significance of aggregate accumulation in multinucleated versus mononucleated cells, we discriminated the multinucleated and MyHC cells from the mononucleated cells (Figure S5C) and confirmed a faster accumulation of Ala16 in differentiated cell cultures compared to Ala10 (Fig. 4 B). For Ala10, the fluorescence intensity is higher in mononucleated cells compared to multinucleated cells, whereas for Ala16, the fluorescence intensity is higher in multinucleated cells (Fig. 4 B). The difference in YFP intensity between Ala16 and Ala10 was much greater in multinucleated cells compared to mononucleated cells (Fig. 4 B), and the average number of puncta in multinucleated cells was also higher in Ala16 compared to Ala10 (Fig. 4 C). We then compared the puncta structure of the pathogenic PABPN1 form between mononucleated and multinucleated cells. To eliminate potential image quantification artefacts due to fixation, live cell imaging was performed and the multinucleated and mononucleated objects were segmented based on nuclear density (Fig. 4 D). With this nuclear-based segmentation, some mononucleated objects are potentially false positives. The puncta area was significantly larger (Fig. 4 E), while the circularity was smaller (Fig. 4 F) in multinucleated objects compared to mononucleated cells. Next, we investigated whether puncta dynamics might contribute to the differences in puncta accumulation and structure. We assessed puncta dynamics in live cells using spinning disk imaging. Cells were imaged 16 hours after Dox treatment (Video S1A and S1B in Ala10 and Ala16, respectively). Comparative kinetic analysis between Ala10 and Ala16 was performed in nuclei with similar YFP intensity levels. To visualize puncta dynamics, three consecutive time frames (2 seconds apart) were superimposed and the overlap between time frames was quantitatively assessed. An overlap between frames was colored white and indicated low puncta dynamics, whereas monochromatic puncta indicated highly dynamic puncta (Fig. 4 G). The proportional area of white was significantly larger in Ala10 than in Ala16 (Fig. 4 H), indicating that puncta dynamics were lower in Ala10 than in Ala16. The higher puncta dynamics could indicate unstructured aggregation, which is consistent with the reduced circularity in Ala16. Taken together, these results suggest that Ala16 aggregation is more affected in differentiated cells, consistent with its pathogenicity. PABPN1 pathogenic aggregates affect cell nucleus morphology cell function To assess if aggregate structure correlates with the cell nucleus morphology, we employed label-free RI imaging, which allows textural features referring to myonuclei structure to be extracted. We imaged myonuclei in differentiated cells (Fig. 5 A), and calculated entropy that measures the complexity and randomness of intensity patterns within the nuclear texture, and Inverse Difference Moment (IDM) that evaluates image homogeneity 24 . In Ala16 myonuclei entropy was significantly lower, indicating lower texture complexity, than in Ala10 (Fig. 5 A). In contrast, higher IDM values were measured in Ala16, indicating a more uniform texture complexity (Fig. 5 B). The entropy and IDM values in Ala10 nuclei were close to those in the vehicle nuclei (Fig. 5 B and 5 C). While our approach did not uncover the chemical nature or the mechanism behind the structural changes, the significant difference in nuclear texture measures between Ala10 and Al16 suggests that the pathogenic aggregates impact nuclear morphology. We then investigated whether puncta intensity correlates with cellular mechanisms that are associated with PABPN1 levels, including mitochondrial activity 25 , proteasomal activity 26 , 27 and translational efficiency 28 . TMRM is sequestered in active mitochondria, and higher TMRM intensity indicates higher mitochondrial activity. TMRM intensity was reduced in both Ala10 and Ala16 multinucleated cells compared to vehicle-treated cells, but did not differ between Ala10 and Ala16 (Fig. 5 D and 5 E), but TMRM intensity was not affected by YFP puncta intensity (Fig. 5 F). TMRM intensity was higher in differentiating cell cultures compared to proliferating conditions, but even in proliferating conditions, TMRM intensity did not differ between Ala10 and Ala16 (Figure S6A). This suggests that PABPN1 aggregation negatively affects mitochondrial activity regardless of aggregate size or differentiation condition. Proteasome inhibition by epoxomicin treatment in differentiated cells resulted in higher Ala10 YFP fluorescence, but Ala16 YFP intensity was unchanged (Fig. 5 G). Furthermore, YFP puncta intensity was significantly correlated in Ala10 mock-treated differentiated and epoxomicin-treated cells, but no correlation was found in Ala16 differentiated cells (Fig. 5 H). This suggests that proteasomal activity is impaired in Ala16 cells, in agreement with previous studies 26 , 27 . In OPMD, mRNA is sequestered in nuclear aggregates 29 , therefore, we examined mRNA co-localization with PABPN1-YFP signal in differentiated and proliferating cells. Nuclear sequestering of mRNA was observed in both Ala10 and Ala16 cells, but not in vehicle cell cultures (Fig. 6 A). Quantification of oligo-dT revealed a 3-fold higher signal in Ala16 compared to Ala10 in differentiating cell cultures (Fig. 6 B). In proliferating cell cultures, the oligo-dT signal was only 1.7-fold higher in Ala16 compared to Ala10 (Figure S6C). The signal overlap between oligo-dT and YFP was also significantly higher in Ala16 compared to Ala10 in both differentiated cells and proliferating cell cultures, although the correlation was higher in differentiated cells (Fig. 6 C and Figure S6D). This suggests that higher mRNA nuclear inclusion is caused by Ala16 aggregates and is exacerbated by cell differentiation. To verify the results obtained from HCS imaging and to investigate whether oligo-dT co-localizes with PABPN1 puncta, we used confocal imaging in differentiated cells, Z-stack imaging showed co-localization between oligo-dT and YFP puncta (Fig. 6 E). The average number of oligo-dT puncta and the overlap with YFP puncta in multinucleated cells were significantly higher for Ala16 (Fig. 6 F and 6 G). The consistency of results between the two imaging platforms suggests that mRNA is entrapped in PABPN1 aggregates, which could imply limited levels of mRNA in the cytosol and thus reduced translational efficiency. We assessed nuclear export of mRNA by the ratio of nuclear to perinuclear oligo-dT intensity in the HCS platform and found a higher ratio in Ala16 (Fig. 6 D), supporting nuclear export of mRNA. Treatment with leptomycin B (LMB), a nuclear export inhibitor, showed a higher nuclear to perinuclear ratio in Ala10, indicating nuclear export (Figure S6H). However, LMB treatment had no effect in Ala16 cells (Figure S6E). This suggests impaired nuclear export in Ala16 cells. LMB treatment in differentiating cell cultures did not affect the subcellular accumulation of oligo-dT (Figure S6E), indicating that nuclear export of the protein differs between cell culture conditions. Last, we examined whether translation efficiency, as measured by OPP-cy5, correlated with PABPN1 aggregation and mRNA nuclear inclusion. Translation efficiency was significantly lower in Ala16 differentiating cells compared to Ala10 cells (Fig. 6 H-I and Figure S6G). Translation efficiency was more affected in Ala16 differentiated cells compared to proliferation conditions (Figure S6F-G). YFP puncta intensity negatively correlated with translation efficiency (Fig. 6 J). Taken together, our results demonstrate a strong correlation between Ala16-PABPN1 aggregates and mRNA nuclear entrapment, which negatively affects reduced translation efficiency. Discussion Using advanced microscopy across four distinct imaging modalities, we identified aggregation patterns ranging from the micro- to nano-scale and uncovered differences between pathogenic and non-pathogenic aggregates. At the nanoscale, pathogenic aggregates exhibit a unique "bouquet" structure, consisting of connected punctate units, whereas non-pathogenic aggregates formed unconnected punctae (Fig. 3 ). At the microscale, pathogenic aggregates were larger in size, and less circular compared to the non-pathogenic aggregates (Fig. 2 ). We demonstrated that structural differences between pathogenic and non-pathogenic aggregates were more pronounced in differentiated multinucleated cells than in non-differentiated cells (Fig. 4 ). These findings suggest that Ala16 aggregation is more pronounced in differentiated cells, consistent with its pathogenic nature. Previous studies in non-muscle proliferating cells have reported limited differences between aggregates of wild-type and expanded PABPN1 proteins 10 , 26 . These studies also suggested that slow protein dynamics of expanded PABPN1 correlate with higher aggregation 9 , 10 . However, our time-lapse imaging in differentiating cells showed slower puncta dynamics in Ala10 compared to Ala16. The higher dynamics of Ala16 puncta may indicate unstructured aggregation, aligning with the reduced circularity observed in Ala16 aggregates. Understanding the structure and dynamics of protein aggregates should ideally be complemented by predictive models. In the case of PABPN1, the region between the alanine stretch (or the pathogenic expansion stretch) and the coiled-coil domain is primarily intrinsically disordered. Consequently, AlphaFold3's prediction of the monomer structure is inconclusive 23 . Nevertheless, predictions of the tetramer structure revealed that the extended alanine stretch becomes entangled with the coiled-coil domain (CCD) in a stable configuration. Future studies should clarify how the expanded alanine stretch structure affect cell pathogenesis, and investigate whether disaggregation preserves PABPN1's functional integrity. Most disease-associated protein aggregates are found in post-mitotic cells, such as neurons or differentiated muscle cells 1 . RNA-protein interaction patterns, as well as protein-protein interactions, are significantly altered during cell differentiation 30 . For example, the chaperone network system undergoes notable changes in differentiated neuronal cells, reflecting differences in proteostasis maintenance between proliferating and differentiated cells 31 . Our experiments in muscle cells demonstrate higher levels of insoluble Ala16 compared to Ala10, a finding that contrasts with studies in non-muscle proliferating cells, which report little to no difference between the insoluble levels of wild-type and pathogenic PABPN1 10 . These results emphasize the critical importance of studying nuclear aggregates in cell models that closely mimic disease conditions to gain a deeper understanding of disease mechanisms. In our cell-based analysis, we investigated the relationship between cellular function and PABPN1 aggregates. Decreased mitochondrial activity, a hallmark of NMDs 1 , 32 , aging muscles and OPMD 25 , 33 . We did not find significant difference in mitochondrial activity between cells overexpressing the wild-type or pathogenic PABPN1, or a correlation between PABPN1 puncta and mitochondrial activity (Fig. 5 ). However, proteasome inhibition discriminated between wild-type PABPN1 and pathogenic aggregates. When the proteasome was inhibited using epoxomicin, we observed increased accumulation of Ala10-YFP aggregates, indicating that these are typically degraded by the proteasome. In contrast, the same treatment had no effect on Ala16-YFP aggregates, suggesting that either Ala16 aggregates evade proteasome recognition or that proteasome function is impaired in Ala16-expressing cells (Fig. 5 ). This impaired proteasome activity in Ala16 cells aligns with previous studies in mouse models, which indicate that PABPN1-mediated reduced expression of proteasomal components in OPMD models compromises proteasome function 34 , 35 . Furthermore, our imaging studied show that mRNA is entrapped in pathogenic aggregates is significantly more than in the non-pathogenic aggregates (Fig. 6 ). In OPMD, poly(A) RNA sequestration within myonuclei has been reported 29 , which strengthens the relevance of our cell model. Additionally, our analysis revealed that mRNA sequestration in nuclear pathogenic aggregates corroborated with mRNA nuclear export and reduced translational efficiency (Fig. 6 ). These findings are consistent with previous studies showing that impaired mRNA export reduces translational efficiency 36 . In summary, our findings revealed structural differences between pathogenic and non-pathogenic protein aggregates and an impact on cellular changes affecting muscle function. Pathogenic aggregates were larger, less circular, and more dynamic compared to non-pathogenic aggregates. Furthermore, we demonstrated a correlation between aggregate size, mRNA nuclear entrapment and nuclear export, and translational efficiency in differentiated muscle cells. Key differences in aggregate behavior between proliferating and differentiating cells highlight the importance of studying protein aggregation under disease-relevant cellular conditions. Declarations Acknowledgements We thank Danish Khan for assisting in the generation of the muscle cell model and Dino Rocca for assisting in cell culture. Funding This study was financed by PPP Health~Holland, The Netherlands and by argenx B.P. Competing interests The authors report no competing interests. Author contributions: Conceptualization: VR, TS, AM Methodology: VR, TS, AM, LMV, MS Investigation: SDM, EB, VS, MS, DR Visualization: SDM, EB, VS Supervision: VR, TS, AM Writing—original draft: SDM, EB, VS, VR Writing—review & editing: VR, TS, AM Supplementary material Supplementary material is available at Brain online References Shastry BS. Neurodegenerative disorders of protein aggregation. Neurochemistry International . 2003/07/01/ 2003;43(1):1-7. doi:https://doi.org/10.1016/S0197-0186(02)00196-1 Taylor RC, Dillin A. Aging as an event of proteostasis collapse. Cold Spring Harb Perspect Biol . May 1 2011;3(5)doi:10.1101/cshperspect.a004440 Cuanalo-Contreras K, Schulz J, Mukherjee A, Park KW, Armijo E, Soto C. Extensive accumulation of misfolded protein aggregates during natural aging and senescence. Front Aging Neurosci . 2022;14:1090109. doi:10.3389/fnagi.2022.1090109 Frankowska N, Lisowska K, Witkowski JM. Proteolysis dysfunction in the process of aging and age-related diseases. Review. Frontiers in Aging . 2022-July-22 2022;3doi:10.3389/fragi.2022.927630 Morimoto RI, Cuervo AM. Proteostasis and the aging proteome in health and disease. J Gerontol A Biol Sci Med Sci . Jun 2014;69 Suppl 1(Suppl 1):S33-8. doi:10.1093/gerona/glu049 Abu-Baker A, Rouleau G. Polyalanine And Polyglutamine Diseases: Possible Common Mechanisms? Genetic Instabilities and Neurological Diseases, Second Edition . 12/31 2006:487-513. doi:10.1016/B978-012369462-1/50034-X Brais B, Bouchard JP, Xie YG , et al . Short GCG expansions in the PABP2 gene cause oculopharyngeal muscular dystrophy. Nat Genet . Feb 1998;18(2):164-7. doi:10.1038/ng0298-164 Yamashita S. Recent Progress in Oculopharyngeal Muscular Dystrophy. J Clin Med . Mar 29 2021;10(7)doi:10.3390/jcm10071375 Raz V, Abraham T, van Zwet EW, Dirks RW, Tanke HJ, van der Maarel SM. Reversible aggregation of PABPN1 pre-inclusion structures. Nucleus . May-Jun 2011;2(3):208-18. doi:10.4161/nucl.2.3.15736 Guan W-L, Jiang L-L, Yin X-F, Hu H-Y. PABPN1 aggregation is driven by Ala expansion and poly(A)-RNA binding, leading to CFIm25 sequestration that impairs alternative polyadenylation. Journal of Biological Chemistry . 2023;299(8)doi:10.1016/j.jbc.2023.105019 Bar-Lavan Y, Shemesh N, Dror S, Ofir R, Yeger-Lotem E, Ben-Zvi A. A Differentiation Transcription Factor Establishes Muscle-Specific Proteostasis in Caenorhabditis elegans. PLOS Genetics . 2016;12(12):e1006531. doi:10.1371/journal.pgen.1006531 Carlotti F, Bazuine M, Kekarainen T , et al . Lentiviral vectors efficiently transduce quiescent mature 3T3-L1 adipocytes. Mol Ther . Feb 2004;9(2):209-17. doi:10.1016/j.ymthe.2003.11.021 BOLTE S, CORDELIÈRES FP. A guided tour into subcellular colocalization analysis in light microscopy. Journal of Microscopy . 2006;224(3):213-232. doi:https://doi.org/10.1111/j.1365-2818.2006.01706.x Faas FG, Avramut MC, van den Berg BM, Mommaas AM, Koster AJ, Ravelli RB. Virtual nanoscopy: generation of ultra-large high resolution electron microscopy maps. J Cell Biol . Aug 6 2012;198(3):457-69. doi:10.1083/jcb.201201140 Last MGF, Abendstein L, Voortman LM, Sharp TH. Ais: streamlining segmentation of cryo-electron tomography datasets. bioRxiv . 2024:2024.04.04.586917. doi:10.1101/2024.04.04.586917 Pettersen EF, Goddard TD, Huang CC , et al . UCSF ChimeraX: Structure visualization for researchers, educators, and developers. Protein Sci . Jan 2021;30(1):70-82. doi:10.1002/pro.3943 Yu C, Chen L, Ouyang K , et al . Effect of partial substitution of NaCl by KCl on aggregation behavior and gel properties of beef myosin. Food Chemistry . 2024/11/15/ 2024;458:140178. doi:https://doi.org/10.1016/j.foodchem.2024.140178 Trollet C, Anvar SY, Venema A , et al . Molecular and phenotypic characterization of a mouse model of oculopharyngeal muscular dystrophy reveals severe muscular atrophy restricted to fast glycolytic fibres. Hum Mol Genet . Jun 1 2010;19(11):2191-207. doi:10.1093/hmg/ddq098 Vincent TL, Green PJ, Woolfson DN. LOGICOIL--multi-state prediction of coiled-coil oligomeric state. Bioinformatics . Jan 1 2013;29(1):69-76. doi:10.1093/bioinformatics/bts648 Tavanez JP, Calado P, Braga J, Lafarga M, Carmo-Fonseca M. In vivo aggregation properties of the nuclear poly(A)-binding protein PABPN1. Rna . May 2005;11(5):752-62. doi:10.1261/rna.7217105 Verheesen P, de Kluijver A, van Koningsbruggen S , et al . Prevention of oculopharyngeal muscular dystrophy-associated aggregation of nuclear poly(A)-binding protein with a single-domain intracellular antibody. Human Molecular Genetics . 2005;15(1):105-111. doi:10.1093/hmg/ddi432 Abramson J, Adler J, Dunger J , et al . Accurate structure prediction of biomolecular interactions with AlphaFold 3. Nature . 2024/06/01 2024;630(8016):493-500. doi:10.1038/s41586-024-07487-w Ruff KM, Pappu RV. AlphaFold and Implications for Intrinsically Disordered Proteins. Journal of Molecular Biology . 2021/10/01/ 2021;433(20):167208. doi:https://doi.org/10.1016/j.jmb.2021.167208 Pantic I, Dimitrijevic D, Nesic D, Petrovic D. Gray level co-occurrence matrix algorithm as pattern recognition biosensor for oxidopamine-induced changes in lymphocyte chromatin architecture. J Theor Biol . Oct 7 2016;406:124-8. doi:10.1016/j.jtbi.2016.07.018 Anvar SY, Raz Y, Verway N , et al . A decline in PABPN1 induces progressive muscle weakness in oculopharyngeal muscle dystrophy and in muscle aging. Aging (Albany NY) . Jun 2013;5(6):412-26. doi:10.18632/aging.100567 Raz V, Routledge S, Venema A , et al . Modeling oculopharyngeal muscular dystrophy in myotube cultures reveals reduced accumulation of soluble mutant PABPN1 protein. Am J Pathol . Oct 2011;179(4):1988-2000. doi:10.1016/j.ajpath.2011.06.044 Raz V, Buijze H, Raz Y , et al . A Novel Feed-Forward Loop between ARIH2 E3-Ligase and PABPN1 Regulates Aging-Associated Muscle Degeneration. The American Journal of Pathology . 2014/04/01/ 2014;184(4):1119-1131. doi:https://doi.org/10.1016/j.ajpath.2013.12.011 Mei H, Boom J, El Abdellaoui S , et al . Alternative Polyadenylation Utilization Results in Ribosome Assembly and mRNA Translation Deficiencies in a Model for Muscle Aging. J Gerontol A Biol Sci Med Sci . Jun 1 2022;77(6):1130-1140. doi:10.1093/gerona/glac058 Calado A, Tomé FMS, Brais B , et al . Nuclear inclusions in oculopharyngeal muscular dystrophy consist of poly(A) binding protein 2 aggregates which sequester poly(A) RNA. Human Molecular Genetics . 2000;9(15):2321-2328. doi:10.1093/oxfordjournals.hmg.a018924 Trendel J, Schwarzl T, Horos R , et al . The Human RNA-Binding Proteome and Its Dynamics during Translational Arrest. Cell . Jan 10 2019;176(1-2):391-403.e19. doi:10.1016/j.cell.2018.11.004 Vonk WIM, Rainbolt TK, Dolan PT, Webb AE, Brunet A, Frydman J. Differentiation Drives Widespread Rewiring of the Neural Stem Cell Chaperone Network. Molecular Cell . 2020/04/16/ 2020;78(2):329-345.e9. doi:https://doi.org/10.1016/j.molcel.2020.03.009 Connolly NMC, Theurey P, Adam-Vizi V , et al . Guidelines on experimental methods to assess mitochondrial dysfunction in cellular models of neurodegenerative diseases. Cell Death & Differentiation . 2018/03/01 2018;25(3):542-572. doi:10.1038/s41418-017-0020-4 Chartier A, Klein P, Pierson S , et al . Mitochondrial Dysfunction Reveals the Role of mRNA Poly(A) Tail Regulation in Oculopharyngeal Muscular Dystrophy Pathogenesis. PLOS Genetics . 2015;11(3):e1005092. doi:10.1371/journal.pgen.1005092 Anvar SY, t Hoen PA, Venema A , et al . Deregulation of the ubiquitin-proteasome system is the predominant molecular pathology in OPMD animal models and patients. Skelet Muscle . Apr 4 2011;1(1):15. doi:10.1186/2044-5040-1-15 Ribot C, Soler C, Chartier A , et al . Activation of the ubiquitin-proteasome system contributes to oculopharyngeal muscular dystrophy through muscle atrophy. PLoS Genet . Jan 2022;18(1):e1010015. doi:10.1371/journal.pgen.1010015 Katahira J. Nuclear export of messenger RNA. Genes (Basel) . Mar 31 2015;6(2):163-84. doi:10.3390/genes6020163 Additional Declarations (Not answered) Supplementary Files Suppldata.pdf supplemental material originaldata.jpg original data Graphicalsummary.docx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-5676243","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":400054586,"identity":"9e7fb0b9-1423-4909-952b-d1bf72c794de","order_by":0,"name":"Vered Raz","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAlElEQVRIiWNgGAWjYNCCCgsI/YB4LWckIHQC0ToY20jRYnCA9+HHn/MkEte2H2B7QKQWdmMJyW0SidvOJLAbEKVFsoGNQcIQpOUGA5sEsVqYfyTOIUULPwMbm8TBBpK0MLOxWTYckzDediaxjTgtbOxtzDd/1NjIbjt++JjEB2K0MDDDWYwNRGkYBaNgFIyCUUAEAACIeiq/9PGVtwAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0000-0003-3152-1952","institution":"Leiden University Medical Centre","correspondingAuthor":true,"prefix":"","firstName":"Vered","middleName":"","lastName":"Raz","suffix":""},{"id":400054587,"identity":"30de3f29-0c38-4966-ae0f-efa28e9f84bb","order_by":1,"name":"Sander Mallon","email":"","orcid":"","institution":"Leiden University Medical Centre","correspondingAuthor":false,"prefix":"","firstName":"Sander","middleName":"","lastName":"Mallon","suffix":""},{"id":400054588,"identity":"dcaba584-8099-469a-b80e-fa9ae674ab2e","order_by":2,"name":"Erik Bos","email":"","orcid":"","institution":"Leiden University of Medical Center","correspondingAuthor":false,"prefix":"","firstName":"Erik","middleName":"","lastName":"Bos","suffix":""},{"id":400054589,"identity":"ddee90d2-9a78-4ed1-bb09-5bc3065225e9","order_by":3,"name":"Vahid Sheikhhassani","email":"","orcid":"","institution":"Leiden University","correspondingAuthor":false,"prefix":"","firstName":"Vahid","middleName":"","lastName":"Sheikhhassani","suffix":""},{"id":400054590,"identity":"d2883c28-c8d6-4bda-9cb1-e6bf69f04e33","order_by":4,"name":"Milad Shademan","email":"","orcid":"","institution":"Leiden University of Medical Center","correspondingAuthor":false,"prefix":"","firstName":"Milad","middleName":"","lastName":"Shademan","suffix":""},{"id":400054591,"identity":"a38c1520-0ec8-448a-aa6e-c389e4018297","order_by":5,"name":"Lennard Voortman","email":"","orcid":"","institution":"Leiden University medical center","correspondingAuthor":false,"prefix":"","firstName":"Lennard","middleName":"","lastName":"Voortman","suffix":""},{"id":400054592,"identity":"f9a4a749-2a5c-4d65-af46-c6942cd797e3","order_by":6,"name":"Alireza Mashaghi","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Alireza","middleName":"","lastName":"Mashaghi","suffix":""},{"id":400054593,"identity":"d7ccfb2a-d021-425f-9329-2f98322d7f43","order_by":7,"name":"Thom Sharp","email":"","orcid":"","institution":"School of Biochemistry/Bristol University","correspondingAuthor":false,"prefix":"","firstName":"Thom","middleName":"","lastName":"Sharp","suffix":""}],"badges":[],"createdAt":"2024-12-19 11:11:02","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5676243/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5676243/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":73784879,"identity":"15c19ad8-d329-4006-bc5d-c9b0522b9245","added_by":"auto","created_at":"2025-01-14 16:05:01","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":851688,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eAggregation characterization of Ala10 and Ala16 in muscle cells. \u003c/strong\u003eExperiments were performed in cell cultures that were treated with Dox for 72 hr. \u003cstrong\u003eA.\u003c/strong\u003e A representative Western blot of soluble (sol) and insoluble (InS) fractions from vehicle (V), Ala10-YFP and Ala16-YFP cell cultures. PABPN1-YFP is 75 kDa, endogenous PABPN1 is 50 kDa, GAPDH and H2B are controls for soluble and insoluble fractions, respectively. No-Stain is a loading control. \u003cstrong\u003eB. \u003c/strong\u003eBox plots of insoluble PABPN1-YFP levels in Ala10 and Ala16 (left) or insoluble/soluble ratio (right). Expression levels were normalized to the No-stain and the Ala10 soluble fraction. Mean and standard deviation are from N=6. \u003cstrong\u003eC.\u003c/strong\u003eBoxplot of endogenous (Endo.) PABPN1 insoluble/soluble ratio in vehicle (V), Ala10 and Ala16. Expression levels were normalized to No staining. Mean and standard deviation are from N=4. \u003cstrong\u003eD-H\u003c/strong\u003e HCS imaging and image quantification. \u003cstrong\u003eD.\u003c/strong\u003e Representative images of YFP (green) and the segmented puncta in yellow (spots) in vehicle, Ala10, and Ala16 myonuclei. Scale bar is 10 µm. \u003cstrong\u003eE.\u003c/strong\u003e Accumulation of YFP puncta fluorescence intensity over time. Mean and standard deviation are from N=3 biological replicates, each replicate represent ~3000 cells. \u003cstrong\u003eF. \u003c/strong\u003eDot plot showing average YFP bulk intensity in Ala10 and Ala16 myonuclei with and without KCl treatment. Mean and standard deviation are from N=3 biological replicates, each replicate represent ~5000 cells. \u003cstrong\u003eG.\u003c/strong\u003e A schematic representation of the intensity of Ala10-YFP (orange) or Ala16-YFP (blue) after KCl treatment (dashed curved line) or at low or high YFP threshold. The low threshold measures the bulk YFP intensity and the high threshold measures the aggregated signal. The gray area represents the aggregated puncta. H. Dot plot shows YFP puncta intensity at low or high threshold. Mean and standard deviation are from N=3 biological replicates, each replicate represent ~5000 cells. Statistical significance was evaluated using a parametric t-test. P\u0026lt;0.05 and \u0026lt;0.0001 are indicated with *, ****.\u003c/p\u003e","description":"","filename":"image2.png","url":"https://assets-eu.researchsquare.com/files/rs-5676243/v1/991da6f1a905240ede9f0038.png"},{"id":73784190,"identity":"7afd5b2d-ccc7-4109-a2be-dab6fab2ffcd","added_by":"auto","created_at":"2025-01-14 15:57:01","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":958533,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eStructural characteristics of Ala10 and Ala16 aggregates. \u003c/strong\u003eA-E. Experiments in proliferation conditions 72 hr Dox-treatment; \u003cstrong\u003eF-G.\u003c/strong\u003e Experiments in differntion conditions 72 hr Dox-treatment. \u003cstrong\u003eA.\u003c/strong\u003e Representative confocal images of Ala10-YFP and Ala16-YFP (in green) in a single myonucleus (DAPI counterstain is in blue) and masks of YFP puncta at low and high threshold (TH25 and TH400, respectively). The scale bar is 10 µm. \u003cstrong\u003eB.\u003c/strong\u003eBoxplot showing the average number of YFP puncta per nucleus at low and high thresholds. N=25 nuclei. \u003cstrong\u003eC-E.\u003c/strong\u003e Analysis was performed in TH25. \u003cstrong\u003eC-D.\u003c/strong\u003eBox plots show the mean and intranuclear variation of puncta area or circularity. Circles represent single nuclei. \u003cstrong\u003eE.\u003c/strong\u003e A linear regression analysis of puncta [Log2 (area and circularity)], each point represents a single punctum. The significance of the F-test, the slope (S) and the coefficient (R) are indicated. \u003cstrong\u003eF.\u003c/strong\u003e Representative refractive index images of differentiated cells overlaid with fluorescence signal. The scale bar is 20 µm. \u003cstrong\u003eG. \u003c/strong\u003eBox plots of mean YFP intensity (left), puncta area (middle), and circularity (right). Statistical significance was determined by Anova. P\u0026lt;0.05, 0.005, 0.001 are indicated by *, ***, ****, respectively.\u003c/p\u003e","description":"","filename":"image3.png","url":"https://assets-eu.researchsquare.com/files/rs-5676243/v1/3c9b1d6e4998639a2fc78550.png"},{"id":73785859,"identity":"c71174bf-64a3-4b30-bbe0-912a991ff04b","added_by":"auto","created_at":"2025-01-14 16:13:01","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":3211578,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eDetection of PABPN1 aggregate structures using TEMand correlative light and electron microscopy (CLEM). \u003c/strong\u003eDifferentiated cell cultures were treated with Doc for 72 hr. \u003cstrong\u003eA.\u003c/strong\u003e An EM image of multinucleated cell. The scale bar is 10 µm. \u003cstrong\u003eA.\u003c/strong\u003e A TEM image of a multinucleated Ala16-YFP cell. \u003cstrong\u003eB.\u003c/strong\u003eTEM images of a nucleus in vehicle cell culture followed by aggregated structures: a punctate (circled in cyan) and two bouquets (circled in blue) that differ in size and shape. Nucleoli (N) are annotated. The scale bar is 2 µm. \u003cstrong\u003eC.\u003c/strong\u003e Light microscopy (FM) and electron microscopy (EM) images and overlay of the same cell area. YFP signal correlating with protein aggregate structures is encircled in pink and YFP signal without correlation is encircled in orange. \u003cstrong\u003eD.\u003c/strong\u003e 3D reconstruction of PABPN1 aggregates in the Ala16 cell. Left, a central section; right, a 3D reconstruction based on 8 serial sections. Aggregates are green; nuclear envelope is cyan. \u003cstrong\u003eE. \u003c/strong\u003eBar graph showing the percentage of nuclear structure (punctate or bouquet) or nuclei without aggregates in Ala10 and Ala16 myonuclei. 100 nuclei were sampled per genotype.\u003c/p\u003e","description":"","filename":"image4.png","url":"https://assets-eu.researchsquare.com/files/rs-5676243/v1/5375da60b0525cf7932572ee.png"},{"id":73784881,"identity":"35850581-8520-407b-8f0a-be46b9a3a5ab","added_by":"auto","created_at":"2025-01-14 16:05:01","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":1150267,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eKinetics of Ala10 and Ala16 puncta accumulation.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eA. \u003c/strong\u003eRepresentative confocal images of Ala10 and Ala16 cell cultures treated with Dox for 0, 12 and 72 hours. Cell cultures were immunostained with anti-MyHC, nuclei counterstained with DAPI, YFP is in green. The scale bar is 20 µm. \u003cstrong\u003eB.\u003c/strong\u003eYFP puncta MFI accumulation over time in multinucleated cells expressing MyHC (right angle and solid line), and in mononucleated cells (circle and dashed line). Mean and standard deviation are based on 100 nuclei. \u003cstrong\u003eC.\u003c/strong\u003e Box plot of the average number of puncta per nucleus in multinucleated cells after 36 hours of Dox treatment. Each dot represents the average number of puncta in one multinucleated cell. Mean and standard deviation are from 15 multinucleated cells. D-F. Imaging in living cells. Nuclear counterstain was made with Sir-700. \u003cstrong\u003eD.\u003c/strong\u003e Representative confocal images of Ala16 differentiated cell cultures after 12 and 72 hours of Dox treatment. Multinucleated and mononucleated objects are surrounded by solid or dashed line, respectively. The scale bar is 50 µm. \u003cstrong\u003eE.-F. \u003c/strong\u003eQuantification of Ala16-YFP puncta area or circularity in multinucleated or mononucleated objects, denoted by continuous line or dashed line, respectively. Mean and standard deviation are from single nuclei: N=41-73 mononucleated objects and N=63-137 multinucleated objects. \u003cstrong\u003eG-H. \u003c/strong\u003eImaging was carried out in proliferating cell cultures starting att 16 hours after Doc-treatment. \u003cstrong\u003eG.\u003c/strong\u003e An overlay image of three frames (2-second interval) from time-lapse imaging. Each time frame is represented by a color (red, green, or blue), white indicates an overlap of puncta. The magnification of the boxed area and the segmented area of the white area in cyan are shown in the right panels. The scale bar is 10 µm. \u003cstrong\u003eH.\u003c/strong\u003eA dot plot shows the ratio of white area to colored area in Ala10 and Ala16 nuclei, and the standard deviation is from N=10 nuclei from two experiments. Statistical significance was assessed by one-way ANOVA. P\u0026lt;0.05; \u0026lt;0.01; is indicated with *.\u003c/p\u003e","description":"","filename":"image5.png","url":"https://assets-eu.researchsquare.com/files/rs-5676243/v1/d8822ecb4ec2e069a2bf6dae.png"},{"id":73784211,"identity":"50bf7ae4-bd97-437a-ab64-c8ab9342edee","added_by":"auto","created_at":"2025-01-14 15:57:02","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":824675,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eEvaluation of textural and cellular differences between Ala10 and Ala16 differentiating cell cultures. \u003c/strong\u003eExperiments were done in Dox- or vehicle- treated cell cultures for 48 hr.\u003cstrong\u003e A-C. \u003c/strong\u003eRefractive index. \u003cstrong\u003eA.\u003c/strong\u003e Representative images of a nucleus. \u003cstrong\u003eB-C.\u003c/strong\u003e Box plots of entropy (B) or inverse difference moment (C). \u003cstrong\u003eC-F. \u003c/strong\u003eTMRM assay in differentiated cells. \u003cstrong\u003eC. \u003c/strong\u003eRepresentative images of TMRM staining. Segmented multinucleated cells that were used for quantification are outlined. \u003cstrong\u003eE.\u003c/strong\u003e TMRM fluorescence intensity. The scale bar is 50 µm.\u003cstrong\u003e F.\u003c/strong\u003e Dot plot and correlation analysis of mean YFP intensity in puncta and mean TMRM MFI, each dot represents a biological replicate. Regression line and 95% confidence level are depicted. \u003cstrong\u003eG-H.\u003c/strong\u003eProteasome inhibition assay in differentiation conditions followed by 6 hours of epoxomicin or mock treatment. \u003cstrong\u003eE.\u003c/strong\u003e Bulk YFP intensity in mock or epoxomicin (Epox) treated cell cultures. \u003cstrong\u003eF.\u003c/strong\u003e Dot plot and correlation analysis between mean YFP intensity and YFP intensity in puncta. The regression line and 95% confidence level are depicted, and a significant slope (S) is denoted. Statistical significance was assessed by ANOVA test. P\u0026lt;0.01; \u0026lt;0.005 or \u0026lt;0.0001 are indicated by **, *** and ****, respectively. Data in C-H are from four biological replicates. Imaging and analysis in panels D-H were with the HCS platform, the data is from four biological replicates; each replicate represents the mean from ~4000 nuclei.\u003c/p\u003e","description":"","filename":"image6.png","url":"https://assets-eu.researchsquare.com/files/rs-5676243/v1/0555cd2ca893aefe9f35fb99.png"},{"id":73784882,"identity":"1980fe1a-9705-4d51-a1d4-b89f086f80ba","added_by":"auto","created_at":"2025-01-14 16:05:01","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":1093982,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003emRNA binding properties of Ala10 and Ala16 aggregates. \u003c/strong\u003eExperiments were made in 48 hr Dox-treated cell cultures.\u003cstrong\u003e \u0026nbsp;A. \u003c/strong\u003eRepresentative images of vehicle Ala10 and Ala16 cells labeled with Oligo-dt-Cy5. The scale bar is 50 µm.\u003cstrong\u003e B-D.\u003c/strong\u003e Bar graphs of image quantification in Ala10 and Ala16 differentiated cells: (B) nuclear oligo-dT intensity, (C) signal overlap between Oligo-dT and YFP, (D) oligo-dT (log2[nuclear/cytoplasmic]) ratio. Averages are taken from four biological replicates. Per replicate \u0026gt;1000 nuclear objects. \u003cstrong\u003eE.\u003c/strong\u003e Representative confocal images of YFP and oligo-dT. The scale bar is 15 µm. An XY and XZ plane from Z-stacks in Ala10 and Ala16 myonuclei. Scale bar is 5 µm. YFP-oligo-dT overlap is yellow. Multinucleated objects are surrounded by a white line. \u003cstrong\u003eF-G.\u003c/strong\u003e Analysis of confocal YFP-Oligo-dT images per multinucleated object. Box plots of average Oligo-dT puncta number (\u003cstrong\u003eF\u003c/strong\u003e) and Oligo-dT - YFP overlap (\u003cstrong\u003eG\u003c/strong\u003e). Means and standard deviations are from 17 and 14 multinucleated objects in Ala10 and Ala16, respectively. \u003cstrong\u003eH-J.\u003c/strong\u003e Translation efficiency in fused cell cultures assessed by OPP-Cy5 incorporation using the HCS platform. \u003cstrong\u003eH. R\u003c/strong\u003eepresentative images of Ala16 vehicle and Dox-treated (48 hr) differentiated cells labeled with OPP-555. The scale bar is 50 µm.\u003cstrong\u003e I.\u003c/strong\u003e Boxplot of OPP MFI cell cultures. MFI values indicate the proportion of vehicle cell cultures. \u003cstrong\u003eJ.\u003c/strong\u003eDot plot and correlation analysis between mean YFP intensity in puncta and OPP MFI. The regression line and 95% confidence level are depicted. Means, and standard deviations are from N=5. Statistical significance was assessed by ANOVA test, p\u0026lt;0.01; \u0026lt;0.005 or \u0026lt;0.0001 are indicated by **, ***, ****, respectively. Images and analysis in panels A-D and H-J were made with the HCS platform.\u003c/p\u003e","description":"","filename":"image7.png","url":"https://assets-eu.researchsquare.com/files/rs-5676243/v1/575b5c4c51fed5ffd58480e3.png"},{"id":79693023,"identity":"21f0fbea-72f0-467b-a615-da896c4b26ec","added_by":"auto","created_at":"2025-04-01 14:59:52","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":10954928,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5676243/v1/0662daa1-499d-4f1d-a18f-13f850639206.pdf"},{"id":73784188,"identity":"567200fc-5378-40c2-a9b0-928f61d3ed14","added_by":"auto","created_at":"2025-01-14 15:57:01","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":1754698,"visible":true,"origin":"","legend":"supplemental material","description":"","filename":"Suppldata.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5676243/v1/e5e8544d6594df1c7a007ec5.pdf"},{"id":73784194,"identity":"6cadfa3e-0208-4f21-9b9f-255dbd641ac6","added_by":"auto","created_at":"2025-01-14 15:57:01","extension":"jpg","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":13854281,"visible":true,"origin":"","legend":"original data","description":"","filename":"originaldata.jpg","url":"https://assets-eu.researchsquare.com/files/rs-5676243/v1/9dc92e17b7d610d95fc6e6b5.jpg"},{"id":73784196,"identity":"0662e779-6c6c-4303-b3b1-b59b7ff3da13","added_by":"auto","created_at":"2025-01-14 15:57:01","extension":"docx","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":222893,"visible":true,"origin":"","legend":"","description":"","filename":"Graphicalsummary.docx","url":"https://assets-eu.researchsquare.com/files/rs-5676243/v1/eb2bb6432308481e3168ae7a.docx"}],"financialInterests":"(Not answered)","formattedTitle":"Structural Features of Pathogenic Aggregates Correlate with Cell Pathology in Differentiated Cells","fulltext":[{"header":"Introduction","content":"\u003cp\u003eMisfolded proteins are normally degraded by the cell's quality control systems, such as proteasomes and autophagy pathways. However, in diseases such as neurodegeneration and some forms of muscular dystrophy, mutations lead to proteins that resist degradation \u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e. Loss of proteostatic homeostasis is a natural process that leads to protein aggregation during aging \u003csup\u003e\u003cspan additionalcitationids=\"CR3\" citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u003c/sup\u003e. Over time, the efficiency of proteostatic mechanisms declines, leading to an increase in protein misfolding and aggregation \u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e. Protein aggregation is also a hallmark of several genetic diseases, particularly neurodegenerative and neuromuscular diseases, in which the mutated encoded proteins form aggregates \u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e. These proteins accumulate and form pathogenic aggregates that contribute to cell dysfunction. The pathogenic protein aggregates can be cytoplasmic, such as in Parkinson's disease and Alzheimer's disease, or nuclear as found in polyglutamine repeat expansion disorders, including Huntington\u0026rsquo;s disease (HD) and Spinocerebellar ataxia diseases (SCAs), or polyalanine repeat disorders expansion like Oculopharyngeal muscular dystrophy (OPMD) \u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eOPMD is caused by a short expansion mutation in the gene encoding poly(A)-binding protein nuclear 1 (PABPN1) \u003csup\u003e\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u003c/sup\u003e. Wild type PABPN1 has a repeat of 10 alanine residues at the N-terminus, but the expansion results in the mutant protein containing an expansion of 11\u0026ndash;18 alanine residues. PABPN1 is ubiquitously expressed and essential in all eukaryotic cells, but symptoms are limited to skeletal muscles. Both wild-type and the expanded PABPN1 are prone to aggregate but only aggregates of the mutant protein are pathogenic \u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e. Discriminating between structural features of the pathogenic PABPN1 aggregates from the non-pathogenic form has been addressed in previous studies using mitotic, non-muscle cells \u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e,\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u003c/sup\u003e, but has not captured the nuances of aggregation in post-mitotic cells, and its effect on muscle cell biology \u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u003c/sup\u003e. Understanding the differences between pathogenic and non-pathogenic aggregated proteins is crucial for deciphering how protein aggregation in disease differs from natural aging. This knowledge can inform the development of targeted, disease-specific therapies.\u003c/p\u003e \u003cp\u003eIn this study, we investigated the structural distinctions between pathogenic and non-pathogenic PABPN1 protein variants using a human muscle cell model. We generated stable muscle cells expressing the wild-type PABPN1 (Ala10) or the pathogenic allele (Ala16) under the tetracycline-inducible promoter to bypass PABPN1 toxic effect in mitotic cells. We compared the structure of pathogenic and non-pathogenic PABPN1 aggregates using four imaging modalities, ranging from micrometer to nanometer resolution and linked to muscle cell biology, to provide new structural insights into the structure-pathogenesis relationship of PABPN1.\u003c/p\u003e"},{"header":"Materials and Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003ePABPN1 constructs and lentivirus production\u003c/h2\u003e \u003cp\u003eThe full length PABPN1 wild type (Ala10) or the expanded PABPN1 (Ala16) differentiated to YFP, previously described in \u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e were cloned into the pCW57-MCS1-2A-MCS2 doxycycline (Dox) inducible lentiviral vector (Addgene plasmid #71782). Cloning was confirmed by sanger sequencing. Lentivirus production was performed as detailed in \u003csup\u003e\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e. Lentivirus particle titers were determined in HeLa cells.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eCell culture\u003c/h3\u003e\n\u003cp\u003eCells were cultured in growth medium (F10 (Gibco) medium supplemented with 15% FCS, 1 ng/ml bFGF, 10 ng/ml EGF and 0.4 \u0026micro;g/ml Dexamethasone). Cells were propagated in confluence 50\u0026ndash;80%. Cell cultures did not reach 100% confluence to avoid spontaneous differentiation. Cell differentiation was induced at high confluency (85\u0026ndash;95%) in DMEM\u0026thinsp;+\u0026thinsp;2% horse serum for 3\u0026ndash;5 days. The 2417 immortal human muscle cells were transduced with lentiviruses encoding Ala10-YFP or Ala16-YFP, and stable cell cultures were created using puromycin selection. Induction of the PABPN1-YFP transgene was carried out with 4 mg/ml doxycycline hyclate (D5207, Sigma Aldrich). For high content screening (HCS), cells were seeded in a Nunc 96 well plate; for live cell confocal microscopy, cells were seeded in either a \u0026micro;-Slide 15 Well 3D ibiTreat (81506, IBIDI) or a \u0026micro;-Slide 8 Well high ibiTreat (80806, IBIDI) slide. For electron microscopy, cells were seeded in \u0026micro;-Dish 35 mm, high Grid-500 ibiTreat (81166, IBIDI) dishes. For refractive index imaging, cells were cultured on a \u0026micro;-Dish 35 mm, low ibiTreat (80136, IBIDI) dish. Cell cultures were treated with 0.5 \u0026micro;M epoxomicin (Sigma-Aldrich #134381-21-8) for 6 hours; Leptomycin B (LMB) 20nM for 4 hours; or 20 \u0026micro;g/ml cycloheximide (Sigma-Aldrich # 01810) for one hour. Cells were incubated in the growth medium.\u003c/p\u003e\n\u003ch3\u003eProtein extraction and Western blotting\u003c/h3\u003e\n\u003cp\u003eProteins were lysed from cells using RIPA buffer (20 mM Tris, pH 7.4, 150 mM NaCl, 5 mM EDTA, 1% NP40, 5% glycerol and 1 mM DTT and protease inhibitor cocktail). After sonication and centrifugation (1 min, 13000g, at 4\u0026deg;C), the supernatant containing the soluble proteins was transferred to a new tube and the pellet, containing the insoluble proteins, was washed once in PBS, dissolved in loading buffer, sonicated and spin down prior to heat inactivation. Protein aliquots were separated on 10% SDS-PAGE. Western blotting was carried out with a PVDF membrane. Bulk proteins were visualized with the No-Stain Protein Labeling Reagent (#A44717, ThermoFisher) and imaged using the iBright Imaging System (ThermoFisher). The membrane was blocked with 5% dried milk powder (T145.2, Carl Roth), first antibody incubation was carried out at 4 degrees overnight, and secondary antibody incubation at room temperature for one hour. Antibodies are listed in Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e. An Odyssey CLx Infrared imaging system (LiCOR, NE. USA) was used to detect the fluorescent signal. Quantification of protein abundance was done using ImageJ. Values were corrected for background and normalized to loading controls. Western blot quantification was carried out with ImageJ. Normalization was made for both the No-Stain and house-keeping signal. All Western blotting experiments were repeated 6 times.\u003c/p\u003e\n\u003ch3\u003eImmunohistochemistry\u003c/h3\u003e\n\u003cp\u003eInsoluble PABPN1 was detected in 1M KCl pre-treatment for 15 minutes. Immunohistochemistry with or without KCl pre-treatment was performed using standard procedures: fixation (2% formaldehyde in PBS) for 5 minutes, permeabilization (1% triton in PBS) for 10 minutes, PBS washing, incubation with primary antibodies for one hour at room temperature, PBS washing, incubation with secondary antibodies\u0026thinsp;+\u0026thinsp;DAPI (4\u0026prime;,6-Diamidino-2-phenylindole dihydrochloride) for 30 minutes, and PBS washing. Cells were kept in PBS during imaging. Antibodies are listed in Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e.\u003c/p\u003e\n\u003ch3\u003eCellular Assays\u003c/h3\u003e\n\u003cp\u003eAll cellular assays were repeated 3 times in biological triplicates. The results shown in figures are from a representative experiment in biological triplicates.\u003c/p\u003e \u003cp\u003eThe Mitochondrial activity assays were performed in differentiated or proliferating cell cultures grown in 96 well plate and treated with Tetramethylrhodamine methyl ester perchlorate (TMRM). Cell cultures grown in 96 well plates were washed with PBS and incubated with a staining solution (5nM TMRM (Sigma-Aldrich #115532-50-8) and Hoechst were diluted in growth media and incubated for 45 minutes. After twice PBS washing, cells were kept in differentiation media during imaging.\u003c/p\u003e \u003cp\u003eThe Protein Synthesis Assays were performed in differentiated or proliferating cell cultures grown in 96 well plate. The protein synthesis assay kit (Cayman Chemicals #601100) was employed according to the manufacturer protocol, with the following modifications: azido-\u003cem\u003eO\u003c/em\u003e-propargyl-puromycin (OPP)-488 was replaced with OPP-555 (Vector laboratories, #CCT-1494). For the negative control, 30 minutes pre incubation with 20\u0026micro;M cycloheximide was used. Hoechst was added after fixation.\u003c/p\u003e \u003cp\u003eFor RNA hybridization with Oligo-dT, differentiated cell cultures were fixed using 3.7% FA for 15 minutes at RT. After two PBS washes the cells were incubated in Protease III diluted 1:30 in PBS (#322337 Advanced Cell Diagnostics) for 15 minutes at RT. After twice PBS washes cells were incubated in hybridization buffer (#10369 Cepham Life Sciences) for 15 minutes at RT. Incubation with 5\u0026rsquo;-Cy5-Oligo d(T)12\u0026ndash;18 probe (#26-4400-02 Gene Link), diluted 1:1000 in hybridization buffer, was carried out overnight at 40 degrees in a humidified chamber. The following day, washes were carried out at 40 degrees for 5 minutes with 4x, 2x, 1x SSC buffer, and with PBS. Finally, the cells were incubated with Hoechst and kept in PBS during imaging. For differentiation index calculation, cells in 96 well plate were treated with Dox for 24 hours and then were incubated in differentiation medium for 72 hours. Differentiated cells were marked for MyHC expression using Immunohistochemistry procedure and the MF20 antibody.\u003c/p\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eImaging and Image quantification\u003c/h2\u003e \u003cp\u003e \u003cb\u003eThe CellInsight CX7 LZR high content screening (HCS) platform\u003c/b\u003e was used for imaging and image quantification using the accompanied HCS toolbox spot detector and co-localization (ThermoFisher Scientific). Between 2000\u0026ndash;12000 nuclei objects were imaged and each experiment were made in 3\u0026ndash;4 biological replicates, at least two times. Cells were imaged with 405 nm (DAPI), 488 nm (YFP) filters, and per cellular assay with the following filters: imaging: TMRM with 560 nm, OPP-555 with 560 nm, oligo-dT-Cy5 with 647 nm, and MF20 antibody with 647 nm.\u003c/p\u003e \u003cp\u003eImaging for calculation of differentiation index was made with a 10x objective covering over 12,000 nuclei per replicate, and imaging for nuclear YFP quantification with a 20x objective, covering at least 5000 nuclei per replicate. The co-localization toolbox was used for the quantification of differentiation index by the percentage of myonuclei without MyHC objects, and the spot detection toolbox for the YFP puncta, TMRM, OPP-555 and oligo-dT-Cy5. Bulk YFP intensity was considered with a low threshold (50\u0026ndash;150) and the high threshold of puncta (650\u0026ndash;1200). The exact threshold was adjusted per experiment, per experiment both low and high thresholds were quantified. Analysis of both TMRM and 555-OPP MFI was made from the perinuclear region. Oligo-dT signal and overlap with YFP was measured from both nuclear and perinuclear regions in YFP positive myonuclei.\u003c/p\u003e \u003cp\u003e \u003cstrong\u003eConfocal imaging\u003c/strong\u003e \u003cp\u003eimaging of single nuclei was made with Leica SP8 confocal microscope using a 63 x/1.3 oil objective and HyD detectors, or the Dragonfly spinning disc module using a 40x/1.3 or 63x/1.3 oil immersion objective. In fixed cells, nuclear counterstain was with DAPI (Sigma\u0026ndash;Aldrich # D9542). In living cells nuclear counterstain was with Sir700-DNA kit (Spirochrome # SC015). Imaging of DAPI was at 405nm, YFP \u0026ndash; 488nm, anti-Cy5-647 and Sir-700 at 700nm. Imaging settings including exposure time, laser power, the excitation-emission range, and Z-stacks step size were consistent between Ala10 and Ala16 within an experiment. Time-lapse imaging was made with Z-stack acquisition taking 4.20 minutes (1 frame per 2 seconds). Maximum-projection of Z-stacks was made with Imaris. Experiments were repeated 3 times. The results shown in figures are from a representative experiment.\u003c/p\u003e \u003c/p\u003e \u003cp\u003eQuantifications of confocal images were carried out in ImageJ (v1.54f).\u003c/p\u003e \u003cp\u003e1. Overlay of time-lapse images was made from frames at 0, 130, and 260 seconds. Each frame was assigned a distinct RGB color, and an overlap was made with imageJ.\u003c/p\u003e \u003cp\u003e \u003cem\u003e2.\u003c/em\u003e YFP puncta quantification was carried out with a macro in ImageJ. The YFP channel was processed with a Gaussian blur (1.5 sigma (radius) \u0026lsquo;Blur\u0026rsquo; and \u0026lsquo;Despeckle\u0026rsquo; functions. Masking of the YFP puncta was applied with a constant threshold across all images within an experiment. The threshold was manually determined to match the YFP puncta (examples are in Figure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003eA). Particles\u0026thinsp;\u0026gt;\u0026thinsp;0.01 \u0026micro;m\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e in area were considered for analysis. From each punctum, the mean fluorescence intensity (MFI), the area, and the circularity were recorded. Analysis was conducted on gated myonuclei from differentiated or undifferentiated cells (single nucleus).\u003c/p\u003e \u003cp\u003e3. Oligo-dT signal analysis was carried out on gated myonuclei in differentiated cells in ImageJ. YFP puncta and oligo-dT analysis was carried out with a Gaussian blur of 1.0, per fluorophore, a constant threshold was used for all images (Figure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003eB). The MFI and area were recorded. The overlap and correlation between YFP and oligo-dT were assessed with the JACoP plugin \u003csup\u003e\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u003c/sup\u003e in ImageJ, using the M1 \u0026amp; M2 coefficients and the Pearson correlation.\u003c/p\u003e \u003cp\u003e \u003cb\u003eTransmission electron microscopy\u003c/b\u003e \u003cem\u003e\u0026ndash;\u003c/em\u003e Differentiated cell cultures were fixed in 1.5% glutaraldehyde in 0.1 M Sodium Cacodylate buffer for 2 hours and were successively incubated in 1% Osmium Tetroxide in 0.1 M cacodylate buffer for 1 hour and in 1% Uranyl Acetate in water for 1 hour. The cells were then dehydrated through a series of incubations in Ethanol (70\u0026ndash;100%) for 90 minutes and embedded in Epon. The flat embedded cells were sectioned with an ultramicrotome (UC6, Leica, Vienna) using a 35 degrees diamond knife (Diatome, Biel, Switzerland) at a nominal section thickness of 90 nm. The sections were transferred to a formvar, and carbon coated 1\u0026times;2 mm copper slot grid and stained for 20 minutes with 7% uranyl acetate in water and for 10 minutes with lead citrate. EM images were recorded using a Tecnai 12 electron microscope (Thermo Fisher Scientific) equipped with an EAGLE 4k\u0026times;4k digital camera. For navigation on EM images, montages of images at 11,000\u0026times; were generated using stitching software \u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e. Morphology was assessed by sampling 100 nuclei on stitched EM images. For 3D reconstruction, 8 consecutive serial sections with a nominal thickness of 200 nm were manually aligned, segmented using the software program, Ais \u003csup\u003e\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e and rendered as 3D isosurfaces in ChimeraX \u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e. For correlative light and electron microscopy, cells were grown on a gridded \u0026micro;-Dish 35 mm plate and stained with DAPI. The living cells were imaged in an EVOS FL Digital Inverted Fluorescence Microscope (Invitrogen) equipped with a 20\u0026times; objective. YFP was visualised with a GFP filter and DAPI with a UV filter. After imaging in the light microscope, the cells were processed for electron microscopy as described above. Superimposition and correlation of light and electron microscopy images was performed using Photoshop. In Photoshop the LM image was copied as a layer into the EM image and made 50% transparent. The LM image required transformation to align with the broader scale of the EM image. This involved isotropic scaling and rotation. Alignment was facilitated by utilizing the nuclear DAPI staining alongside cell morphology.\u003c/p\u003e \u003cp\u003e \u003cb\u003eHolo-tomographic microscopy (HTM)\u003c/b\u003e, in combination with epifluorescence, was performed on the 3D Cell-Explorer Fluo (Nanolive, Ecublens, Switzerland) using a 60\u0026times; air objective (NA\u0026thinsp;=\u0026thinsp;0.8) at a wavelength of λ\u0026thinsp;=\u0026thinsp;520 nm (Class 1 low power laser, sample exposure 0.2 mW/mm2) and CMOS Sony sensor, with quantum efficiency (typical) 70% (at 545 nm), dark noise (typical) 6.6 e-, dynamic range (typical) 73.7 dB, field of view 90 \u0026times; 90 \u0026times; 30 \u0026micro;m, axial resolution 400 nm, and maximum temporal resolution 0.5 3D RI volume per second. Acquired RI images were processed with built-in software (Nanolive). ImageJ/Fiji (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://imagej.nih.gov/\u003c/span\u003e\u003cspan address=\"https://imagej.nih.gov/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) was used for the final processing and quantifications. Experiments were performed in two independent experiments, and the reported results are from one representative experiment.\u003c/p\u003e \u003cp\u003e \u003cem\u003eNuclei texture analysis\u003c/em\u003e - RI images of each cell type were initially converted to the 8-bit format using ImageJ. Subsequently, the areas, including the cell nuclei, were outlined, and extracted using a free-hand selection tool. Following this, the texture of these selected areas was analyzed utilizing the GLCM (Gray Level Co-occurrence Matrix) texture analysis plug-in (version 0.4) developed by Julio E. Cabrera. This plug-in facilitated the computation of various statistical parameters associated with the GLCM of the image, including the Inverse Difference Moment (IDM) and Entropy.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eStatistics and online analysis\u003c/h3\u003e\n\u003cp\u003eStatistical tests were performed in GraphPad Prism.\u003c/p\u003e \u003cp\u003eAlphaFold predictions were carried out in AlphaFold 3.0 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://alphafoldserver.com\u003c/span\u003e\u003cspan address=\"https://alphafoldserver.com\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e \u003cb\u003eAla16-YFP aggregation is higher than Ala10-YFP in an inducible muscle cell model.\u003c/b\u003e \u003c/p\u003e \u003cp\u003eTo investigate differences in aggregation between a pathogenic and non-pathogenic form in a disease-relevant cell model, our study was performed in human muscle cells. We generated cells stably expressing the Ala10 wild-type PABPN1 or the Ala16 extended PABPN1 tagged with YFP (hereafter referred to as Ala10 and Ala16, respectively). Both transgenes were expressed under the tetracycline-inducible promoter to eliminate PABPN1 toxicity due to overexpression. Western blot analysis confirmed the inducible expression of both transgenes after doxycycline (Dox) treatment (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eA). Both Ala10-YFP and Ala16-YFP proteins accumulated in the insoluble fraction, but A16-YFP accumulation was 1.5 times higher than A10-YFP (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB). In addition, the average ratio of insoluble to soluble protein levels was 3.5 times higher for A16-YFP than for A10-YFP (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB). We next examined whether transgene overexpression was associated with endogenous PABPN1 levels and found a higher ratio in Ala16 compared to Ala10 or in vehicle cultures (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eC). This suggests that an increase in insoluble PABPN1 depletes the levels of the soluble form.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eImage quantification confirmed nuclear accumulation of both Ala10 and Ala16 (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eD). Consistent with Western blot analysis, YFP fluorescence intensity was 2-fold higher in Ala16 puncta than in Ala10 puncta (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eE). To verify that the YFP puncta corresponded to PABPN1 aggregates, we treated live cells with KCl prior to imaging. KCl treatment improves solubility and stabilizes aggregates of hydrophobic polypeptides \u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e, and has been used to detect PABPN1 aggregates in OPMD models \u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e,\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e. KCl treatment in the muscle cell model resulted in higher YFP intensity compared to untreated cell cultures (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eF and Figure S2). Notably, the difference between KCl-treated and untreated cells was greater in Ala10 than in Ala16, indicating that the YFP signal in Ala16 predominantly represents aggregated proteins. To identify PABPN1 aggregates without KCl treatment, we applied two thresholds during image quantification, which allowed discrimination between bulk signal at low threshold and signal in puncta at high threshold (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eG). Consistent with the KCl treatment, a greater difference between low and high thresholds was found for Ala10 (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eH). This analysis demonstrates that PABPN1 aggregates can be distinguished from the soluble protein puncta by image quantification, which may implicate structural differences between pathogenic and non-pathogenic protein aggregates.\u003c/p\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eStructural features discriminate between non-pathogenic and pathogenic PABPN1 aggregates\u003c/h2\u003e \u003cp\u003eThe PABPN1 protein comprises a poly-alanine stretch within the N-terminal intrinsically disordered region (IDR), a coiled coil domain (CCD), and a C-terminal RNA recognition motif (RRM) (Figure S3A). \u003cem\u003eLogicoil\u003c/em\u003e predicts the CCD to form a tetrameric coiled coil \u003csup\u003e\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u003c/sup\u003e, which has been shown to be critical for aggregation \u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e,\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e. The N-terminal alanine stretch alone is not critical for aggregation \u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e, suggesting that folding of the contiguous IDR\u0026thinsp;+\u0026thinsp;CCD region may form a stable structure leading to aggregation. AlphaFold3 \u003csup\u003e22\u003c/sup\u003e prediction of the monomeric IDR\u0026thinsp;+\u0026thinsp;CCD predicts, with low confidence, that the alanine tract in Ala16 folds on the CCD (Figure S3B), which may be more stable than the Ala10 structure. Taking the \u003cem\u003eLogicoil\u003c/em\u003e prediction of a stable CCD tetramer, AlphaFold3 predicts that the wild type (Ala10) IDR N-terminus is parallel to the tetrameric coiled-coil domain. However, the Ala16 expansion is consistently predicted to intercalate with the tetrameric CCD (Figure S3C). These predictions hint at a structural role for the Ala16 expansion that might lead to pathogenesis, although the mechanism is still unclear. Considering the limitations of AlphaFold for IDR prediction \u003csup\u003e\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e, the difference between Ala10 and Ala16 justified the investigation of structural features using puncta structure analysis.\u003c/p\u003e \u003cp\u003eStructural features of PABPN1 puncta were assessed in confocal images from single nuclei (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA), and YFP puncta were segmented with a constant threshold. YFP puncta segmentation was made at two thresholds: a low threshold that visually matched the Ala10 and a high threshold that visually matched the Ala16 puncta but was too high for the Ala10 puncta (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA). A quantitative assessment of puncta segmentation confirmed the visual evaluation and showed that the average number of puncta per nucleus was significantly higher in Ala16 compared to Ala10 (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB). For comparative studies of puncta features, we considered the low threshold. The average puncta area per nucleus was 2.5-fold larger in Ala16 than in Ala10, whereas the average circularity was smaller in Ala16 (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eD). The intranuclear variability of puncta area and circularity were larger in Ala16, indicating high heterogeneity in puncta structure (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eC and \u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eD). The aggregation process was assessed by the relation between puncta area and circularity. In Ala10 puncta, larger puncta remained with a similar circularity (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eE). In contrast, Ala16 puncta showed a negative correlation between area and circularity (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eE). Together, this suggests that Ala16 aggregation is disorganized and heterogeneous, while the Ala10 aggregates keep their shape during growth.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eAs the quantification of puncta structure could be affected by imaging and image processing, we verified the differences in puncta structural features using refractive index (RI) imaging combined with a fluorescence imaging platform using NanoLive holo-tomographic microscopy. This allowed us to simultaneously acquire morphological and molecular density information in selected cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eF). Our imaging analysis showed that, per nucleus, the average fluorescence intensity in puncta was higher, size was larger, and circularity was lower in Ala16 compared to Ala10 (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eG). Overall, the results from combined RI/fluorescence imaging are consistent with those obtained from confocal imaging. Taken together, nuclei in Ala16 cells were more densely packed with protein aggregates than in Ala10. This consistency across different imaging modalities and analytical techniques increases the robustness of these observations.\u003c/p\u003e \u003cp\u003eTo verify the structure of the aggregates, we used transmission electron microscopy (TEM) to characterize PABPN1 aggregates in differentiated cells at nanometer resolution. A stitched image of a differentiated cell created with in-house software \u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e, allowed for detailed analysis of the entire differentiated cell (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA). In the Dox-treated cell cultures, we observed electron-dense structures of variable morphology. These structures, which were absent in the vehicle control myonuclei, had a different electron density than the nucleoli (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB). We classified the myonuclei based on the morphology of these electron density structures; \u0026lsquo;punctate\u0026rsquo; morphology was smaller and circular (cyan circles in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB), and \u0026lsquo;bouquet\u0026rsquo; morphology was larger and not circular (encircled in blue, Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB). To confirm that these novel structures corresponded to PABPN1-YFP aggregates, we used correlative light and electron microscopy (CLEM). We correlated and overlayed fluorescence images with TEM images showing that each electron-dense aggregate corresponded to a YFP signal (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eC, circled in pink). Some YFP signals did not correspond to an electron-dense structure because the aggregate was above or below the 90 nm thick section imaged by electron microscopy (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eC, encircled in orange). Some punctate aggregates showed a toroidal architecture (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB, cyan circles). Sectioning through 3D spherical, cylindrical, or toroidal structures may result in a torus in the resulting 2D section. To determine the morphology of these aggregates, we imaged 8 consecutive 200 nm thick sections using TEM for 3D reconstruction (Figs.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eD and S4). This revealed that the toroidal structures represent hollow spheres of aggregated material, and that these spheres can further aggregate into larger structures.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eTo assess the structural differences between Ala10 and Ala16 we manually scored the aggregate structure in 90 nm thick sections of 100 nuclei. Bouquet morphology and found only in Ala16, and in 2/3rd of the nuclei with electron-dense structures (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eE). In around 50% of the nuclei in both Ala10 and Ala16 myonuclei, electron-dense structures were not found (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eE), which is consistent with the 3D reconstruction analysis of PABPN1 nuclear aggregates. Taken together, confocal, holographic tomography and TEM imaging consistently confirm the structural differences, both in area and circularity, between Ala10 and Ala16 aggregates.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eAggregation differs between non-differentiated and differentiated muscle cells\u003c/h2\u003e \u003cp\u003eSince PABPN1 aggregation is pathogenic in multinucleated muscle fibers, we investigated whether Ala10 and Ala16 aggregation differs between differentiated and non-differentiated muscle cells. In culture, high cell density and starvation stress drive muscle cell fusion into multinucleated cells, characterized by the expression of the myosin heavy chain (MyHC). Induction of Ala10 or Ala16 reduces the differentiation index compared to vehicle-treated cells, and the differentiated cells in Ala16 cultures detached faster compared to Ala10 cultures (Figure S5). Therefore, to assess aggregation in differentiated cells, the following protocol was used: cells were cultured in differentiation condition for 48 hours, followed by Dox treatment for 12 to 72 hours. Cell cultures were fixed, immunolabelled with MyHC and imaged by confocal microscopy (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA). The Ala16-YFP signal appeared after 12 hours of Dox treatment, earlier than in Ala10-YFP cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA). At 72 hours after Dox treatment, Ala10-YFP fluorescence was stronger in MyHC-negative cells compared to the nuclei within the MyHC-positive cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA). To assess the statistical significance of aggregate accumulation in multinucleated versus mononucleated cells, we discriminated the multinucleated and MyHC cells from the mononucleated cells (Figure S5C) and confirmed a faster accumulation of Ala16 in differentiated cell cultures compared to Ala10 (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eB). For Ala10, the fluorescence intensity is higher in mononucleated cells compared to multinucleated cells, whereas for Ala16, the fluorescence intensity is higher in multinucleated cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eB). The difference in YFP intensity between Ala16 and Ala10 was much greater in multinucleated cells compared to mononucleated cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eB), and the average number of puncta in multinucleated cells was also higher in Ala16 compared to Ala10 (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eC).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eWe then compared the puncta structure of the pathogenic PABPN1 form between mononucleated and multinucleated cells. To eliminate potential image quantification artefacts due to fixation, live cell imaging was performed and the multinucleated and mononucleated objects were segmented based on nuclear density (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eD). With this nuclear-based segmentation, some mononucleated objects are potentially false positives. The puncta area was significantly larger (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eE), while the circularity was smaller (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eF) in multinucleated objects compared to mononucleated cells.\u003c/p\u003e \u003cp\u003eNext, we investigated whether puncta dynamics might contribute to the differences in puncta accumulation and structure. We assessed puncta dynamics in live cells using spinning disk imaging. Cells were imaged 16 hours after Dox treatment (Video S1A and S1B in Ala10 and Ala16, respectively). Comparative kinetic analysis between Ala10 and Ala16 was performed in nuclei with similar YFP intensity levels. To visualize puncta dynamics, three consecutive time frames (2 seconds apart) were superimposed and the overlap between time frames was quantitatively assessed. An overlap between frames was colored white and indicated low puncta dynamics, whereas monochromatic puncta indicated highly dynamic puncta (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eG). The proportional area of white was significantly larger in Ala10 than in Ala16 (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eH), indicating that puncta dynamics were lower in Ala10 than in Ala16. The higher puncta dynamics could indicate unstructured aggregation, which is consistent with the reduced circularity in Ala16. Taken together, these results suggest that Ala16 aggregation is more affected in differentiated cells, consistent with its pathogenicity.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003ePABPN1 pathogenic aggregates affect cell nucleus morphology cell function\u003c/h2\u003e \u003cp\u003eTo assess if aggregate structure correlates with the cell nucleus morphology, we employed label-free RI imaging, which allows textural features referring to myonuclei structure to be extracted. We imaged myonuclei in differentiated cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA), and calculated entropy that measures the complexity and randomness of intensity patterns within the nuclear texture, and Inverse Difference Moment (IDM) that evaluates image homogeneity \u003csup\u003e\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e. In Ala16 myonuclei entropy was significantly lower, indicating lower texture complexity, than in Ala10 (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA). In contrast, higher IDM values were measured in Ala16, indicating a more uniform texture complexity (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eB). The entropy and IDM values in Ala10 nuclei were close to those in the vehicle nuclei (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eB and \u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eC). While our approach did not uncover the chemical nature or the mechanism behind the structural changes, the significant difference in nuclear texture measures between Ala10 and Al16 suggests that the pathogenic aggregates impact nuclear morphology.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eWe then investigated whether puncta intensity correlates with cellular mechanisms that are associated with PABPN1 levels, including mitochondrial activity \u003csup\u003e\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e, proteasomal activity \u003csup\u003e\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e,\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u003c/sup\u003e and translational efficiency \u003csup\u003e\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u003c/sup\u003e. TMRM is sequestered in active mitochondria, and higher TMRM intensity indicates higher mitochondrial activity. TMRM intensity was reduced in both Ala10 and Ala16 multinucleated cells compared to vehicle-treated cells, but did not differ between Ala10 and Ala16 (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eD and \u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eE), but TMRM intensity was not affected by YFP puncta intensity (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eF). TMRM intensity was higher in differentiating cell cultures compared to proliferating conditions, but even in proliferating conditions, TMRM intensity did not differ between Ala10 and Ala16 (Figure S6A). This suggests that PABPN1 aggregation negatively affects mitochondrial activity regardless of aggregate size or differentiation condition.\u003c/p\u003e \u003cp\u003eProteasome inhibition by epoxomicin treatment in differentiated cells resulted in higher Ala10 YFP fluorescence, but Ala16 YFP intensity was unchanged (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eG). Furthermore, YFP puncta intensity was significantly correlated in Ala10 mock-treated differentiated and epoxomicin-treated cells, but no correlation was found in Ala16 differentiated cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eH). This suggests that proteasomal activity is impaired in Ala16 cells, in agreement with previous studies \u003csup\u003e\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e,\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eIn OPMD, mRNA is sequestered in nuclear aggregates \u003csup\u003e\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e, therefore, we examined mRNA co-localization with PABPN1-YFP signal in differentiated and proliferating cells. Nuclear sequestering of mRNA was observed in both Ala10 and Ala16 cells, but not in vehicle cell cultures (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eA). Quantification of oligo-dT revealed a 3-fold higher signal in Ala16 compared to Ala10 in differentiating cell cultures (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eB). In proliferating cell cultures, the oligo-dT signal was only 1.7-fold higher in Ala16 compared to Ala10 (Figure S6C). The signal overlap between oligo-dT and YFP was also significantly higher in Ala16 compared to Ala10 in both differentiated cells and proliferating cell cultures, although the correlation was higher in differentiated cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eC and Figure S6D). This suggests that higher mRNA nuclear inclusion is caused by Ala16 aggregates and is exacerbated by cell differentiation.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eTo verify the results obtained from HCS imaging and to investigate whether oligo-dT co-localizes with PABPN1 puncta, we used confocal imaging in differentiated cells, Z-stack imaging showed co-localization between oligo-dT and YFP puncta (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eE). The average number of oligo-dT puncta and the overlap with YFP puncta in multinucleated cells were significantly higher for Ala16 (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eF and \u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eG). The consistency of results between the two imaging platforms suggests that mRNA is entrapped in PABPN1 aggregates, which could imply limited levels of mRNA in the cytosol and thus reduced translational efficiency.\u003c/p\u003e \u003cp\u003eWe assessed nuclear export of mRNA by the ratio of nuclear to perinuclear oligo-dT intensity in the HCS platform and found a higher ratio in Ala16 (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eD), supporting nuclear export of mRNA. Treatment with leptomycin B (LMB), a nuclear export inhibitor, showed a higher nuclear to perinuclear ratio in Ala10, indicating nuclear export (Figure S6H). However, LMB treatment had no effect in Ala16 cells (Figure S6E). This suggests impaired nuclear export in Ala16 cells. LMB treatment in differentiating cell cultures did not affect the subcellular accumulation of oligo-dT (Figure S6E), indicating that nuclear export of the protein differs between cell culture conditions.\u003c/p\u003e \u003cp\u003eLast, we examined whether translation efficiency, as measured by OPP-cy5, correlated with PABPN1 aggregation and mRNA nuclear inclusion. Translation efficiency was significantly lower in Ala16 differentiating cells compared to Ala10 cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eH-I and Figure S6G). Translation efficiency was more affected in Ala16 differentiated cells compared to proliferation conditions (Figure S6F-G). YFP puncta intensity negatively correlated with translation efficiency (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eJ). Taken together, our results demonstrate a strong correlation between Ala16-PABPN1 aggregates and mRNA nuclear entrapment, which negatively affects reduced translation efficiency.\u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eUsing advanced microscopy across four distinct imaging modalities, we identified aggregation patterns ranging from the micro- to nano-scale and uncovered differences between pathogenic and non-pathogenic aggregates. At the nanoscale, pathogenic aggregates exhibit a unique \u003cem\u003e\"bouquet\"\u003c/em\u003e structure, consisting of connected punctate units, whereas non-pathogenic aggregates formed unconnected punctae (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). At the microscale, pathogenic aggregates were larger in size, and less circular compared to the non-pathogenic aggregates (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). We demonstrated that structural differences between pathogenic and non-pathogenic aggregates were more pronounced in differentiated multinucleated cells than in non-differentiated cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). These findings suggest that Ala16 aggregation is more pronounced in differentiated cells, consistent with its pathogenic nature. Previous studies in non-muscle proliferating cells have reported limited differences between aggregates of wild-type and expanded PABPN1 proteins \u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e,\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e. These studies also suggested that slow protein dynamics of expanded PABPN1 correlate with higher aggregation \u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e,\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u003c/sup\u003e. However, our time-lapse imaging in differentiating cells showed slower puncta dynamics in Ala10 compared to Ala16. The higher dynamics of Ala16 puncta may indicate unstructured aggregation, aligning with the reduced circularity observed in Ala16 aggregates. Understanding the structure and dynamics of protein aggregates should ideally be complemented by predictive models. In the case of PABPN1, the region between the alanine stretch (or the pathogenic expansion stretch) and the coiled-coil domain is primarily intrinsically disordered. Consequently, AlphaFold3's prediction of the monomer structure is inconclusive \u003csup\u003e\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e. Nevertheless, predictions of the tetramer structure revealed that the extended alanine stretch becomes entangled with the coiled-coil domain (CCD) in a stable configuration. Future studies should clarify how the expanded alanine stretch structure affect cell pathogenesis, and investigate whether disaggregation preserves PABPN1's functional integrity.\u003c/p\u003e \u003cp\u003eMost disease-associated protein aggregates are found in post-mitotic cells, such as neurons or differentiated muscle cells \u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e. RNA-protein interaction patterns, as well as protein-protein interactions, are significantly altered during cell differentiation \u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e. For example, the chaperone network system undergoes notable changes in differentiated neuronal cells, reflecting differences in proteostasis maintenance between proliferating and differentiated cells \u003csup\u003e\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e. Our experiments in muscle cells demonstrate higher levels of insoluble Ala16 compared to Ala10, a finding that contrasts with studies in non-muscle proliferating cells, which report little to no difference between the insoluble levels of wild-type and pathogenic PABPN1 \u003csup\u003e10\u003c/sup\u003e. These results emphasize the critical importance of studying nuclear aggregates in cell models that closely mimic disease conditions to gain a deeper understanding of disease mechanisms.\u003c/p\u003e \u003cp\u003eIn our cell-based analysis, we investigated the relationship between cellular function and PABPN1 aggregates. Decreased mitochondrial activity, a hallmark of NMDs \u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e,\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e\u003c/sup\u003e, aging muscles and OPMD \u003csup\u003e\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e,\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u003c/sup\u003e. We did not find significant difference in mitochondrial activity between cells overexpressing the wild-type or pathogenic PABPN1, or a correlation between PABPN1 puncta and mitochondrial activity (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e). However, proteasome inhibition discriminated between wild-type PABPN1 and pathogenic aggregates. When the proteasome was inhibited using epoxomicin, we observed increased accumulation of Ala10-YFP aggregates, indicating that these are typically degraded by the proteasome. In contrast, the same treatment had no effect on Ala16-YFP aggregates, suggesting that either Ala16 aggregates evade proteasome recognition or that proteasome function is impaired in Ala16-expressing cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e). This impaired proteasome activity in Ala16 cells aligns with previous studies in mouse models, which indicate that PABPN1-mediated reduced expression of proteasomal components in OPMD models compromises proteasome function \u003csup\u003e\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e,\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eFurthermore, our imaging studied show that mRNA is entrapped in pathogenic aggregates is significantly more than in the non-pathogenic aggregates (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e). In OPMD, poly(A) RNA sequestration within myonuclei has been reported \u003csup\u003e\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e, which strengthens the relevance of our cell model. Additionally, our analysis revealed that mRNA sequestration in nuclear pathogenic aggregates corroborated with mRNA nuclear export and reduced translational efficiency (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e). These findings are consistent with previous studies showing that impaired mRNA export reduces translational efficiency \u003csup\u003e\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eIn summary, our findings revealed structural differences between pathogenic and non-pathogenic protein aggregates and an impact on cellular changes affecting muscle function. Pathogenic aggregates were larger, less circular, and more dynamic compared to non-pathogenic aggregates. Furthermore, we demonstrated a correlation between aggregate size, mRNA nuclear entrapment and nuclear export, and translational efficiency in differentiated muscle cells. Key differences in aggregate behavior between proliferating and differentiating cells highlight the importance of studying protein aggregation under disease-relevant cellular conditions.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe thank Danish Khan for assisting in the generation of the muscle cell model and Dino Rocca for assisting in cell culture.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003eThis study was financed by PPP Health~Holland, The Netherlands and by argenx B.P.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u0026nbsp;\u003c/strong\u003eThe authors report no competing interests.\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor contributions:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eConceptualization: VR, TS, AM\u003c/p\u003e\n\u003cp\u003eMethodology: VR, TS, AM, LMV, MS\u003c/p\u003e\n\u003cp\u003eInvestigation: SDM, EB, VS, MS, DR\u003c/p\u003e\n\u003cp\u003eVisualization: SDM, EB, VS\u003c/p\u003e\n\u003cp\u003eSupervision: VR, TS, AM\u003c/p\u003e\n\u003cp\u003eWriting\u0026mdash;original draft: SDM, EB, VS, VR\u003c/p\u003e\n\u003cp\u003eWriting\u0026mdash;review \u0026amp; editing: VR, TS, AM\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSupplementary material\u003c/strong\u003e Supplementary material is available at \u003cem\u003eBrain\u003c/em\u003e online\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eShastry BS. Neurodegenerative disorders of protein aggregation. \u003cem\u003eNeurochemistry International\u003c/em\u003e. 2003/07/01/ 2003;43(1):1-7. doi:https://doi.org/10.1016/S0197-0186(02)00196-1\u003c/li\u003e\n\u003cli\u003eTaylor RC, Dillin A. Aging as an event of proteostasis collapse. \u003cem\u003eCold Spring Harb Perspect Biol\u003c/em\u003e. May 1 2011;3(5)doi:10.1101/cshperspect.a004440\u003c/li\u003e\n\u003cli\u003eCuanalo-Contreras K, Schulz J, Mukherjee A, Park KW, Armijo E, Soto C. Extensive accumulation of misfolded protein aggregates during natural aging and senescence. \u003cem\u003eFront Aging Neurosci\u003c/em\u003e. 2022;14:1090109. doi:10.3389/fnagi.2022.1090109\u003c/li\u003e\n\u003cli\u003eFrankowska N, Lisowska K, Witkowski JM. Proteolysis dysfunction in the process of aging and age-related diseases. Review. \u003cem\u003eFrontiers in Aging\u003c/em\u003e. 2022-July-22 2022;3doi:10.3389/fragi.2022.927630\u003c/li\u003e\n\u003cli\u003eMorimoto RI, Cuervo AM. Proteostasis and the aging proteome in health and disease. \u003cem\u003eJ Gerontol A Biol Sci Med Sci\u003c/em\u003e. Jun 2014;69 Suppl 1(Suppl 1):S33-8. doi:10.1093/gerona/glu049\u003c/li\u003e\n\u003cli\u003eAbu-Baker A, Rouleau G. Polyalanine And Polyglutamine Diseases: Possible Common Mechanisms? \u003cem\u003eGenetic Instabilities and Neurological Diseases, Second Edition\u003c/em\u003e. 12/31 2006:487-513. doi:10.1016/B978-012369462-1/50034-X\u003c/li\u003e\n\u003cli\u003eBrais B, Bouchard JP, Xie YG\u003cem\u003e, et al\u003c/em\u003e. Short GCG expansions in the PABP2 gene cause oculopharyngeal muscular dystrophy. \u003cem\u003eNat Genet\u003c/em\u003e. Feb 1998;18(2):164-7. doi:10.1038/ng0298-164\u003c/li\u003e\n\u003cli\u003eYamashita S. Recent Progress in Oculopharyngeal Muscular Dystrophy. \u003cem\u003eJ Clin Med\u003c/em\u003e. Mar 29 2021;10(7)doi:10.3390/jcm10071375\u003c/li\u003e\n\u003cli\u003eRaz V, Abraham T, van Zwet EW, Dirks RW, Tanke HJ, van der Maarel SM. Reversible aggregation of PABPN1 pre-inclusion structures. \u003cem\u003eNucleus\u003c/em\u003e. May-Jun 2011;2(3):208-18. doi:10.4161/nucl.2.3.15736\u003c/li\u003e\n\u003cli\u003eGuan W-L, Jiang L-L, Yin X-F, Hu H-Y. PABPN1 aggregation is driven by Ala expansion and poly(A)-RNA binding, leading to CFIm25 sequestration that impairs alternative polyadenylation. \u003cem\u003eJournal of Biological Chemistry\u003c/em\u003e. 2023;299(8)doi:10.1016/j.jbc.2023.105019\u003c/li\u003e\n\u003cli\u003eBar-Lavan Y, Shemesh N, Dror S, Ofir R, Yeger-Lotem E, Ben-Zvi A. A Differentiation Transcription Factor Establishes Muscle-Specific Proteostasis in Caenorhabditis elegans. \u003cem\u003ePLOS Genetics\u003c/em\u003e. 2016;12(12):e1006531. doi:10.1371/journal.pgen.1006531\u003c/li\u003e\n\u003cli\u003eCarlotti F, Bazuine M, Kekarainen T\u003cem\u003e, et al\u003c/em\u003e. Lentiviral vectors efficiently transduce quiescent mature 3T3-L1 adipocytes. \u003cem\u003eMol Ther\u003c/em\u003e. Feb 2004;9(2):209-17. doi:10.1016/j.ymthe.2003.11.021\u003c/li\u003e\n\u003cli\u003eBOLTE S, CORDELI\u0026Egrave;RES FP. A guided tour into subcellular colocalization analysis in light microscopy. \u003cem\u003eJournal of Microscopy\u003c/em\u003e. 2006;224(3):213-232. doi:https://doi.org/10.1111/j.1365-2818.2006.01706.x\u003c/li\u003e\n\u003cli\u003eFaas FG, Avramut MC, van den Berg BM, Mommaas AM, Koster AJ, Ravelli RB. Virtual nanoscopy: generation of ultra-large high resolution electron microscopy maps. \u003cem\u003eJ Cell Biol\u003c/em\u003e. Aug 6 2012;198(3):457-69. doi:10.1083/jcb.201201140\u003c/li\u003e\n\u003cli\u003eLast MGF, Abendstein L, Voortman LM, Sharp TH. Ais: streamlining segmentation of cryo-electron tomography datasets. \u003cem\u003ebioRxiv\u003c/em\u003e. 2024:2024.04.04.586917. doi:10.1101/2024.04.04.586917\u003c/li\u003e\n\u003cli\u003ePettersen EF, Goddard TD, Huang CC\u003cem\u003e, et al\u003c/em\u003e. UCSF ChimeraX: Structure visualization for researchers, educators, and developers. \u003cem\u003eProtein Sci\u003c/em\u003e. Jan 2021;30(1):70-82. doi:10.1002/pro.3943\u003c/li\u003e\n\u003cli\u003eYu C, Chen L, Ouyang K\u003cem\u003e, et al\u003c/em\u003e. Effect of partial substitution of NaCl by KCl on aggregation behavior and gel properties of beef myosin. \u003cem\u003eFood Chemistry\u003c/em\u003e. 2024/11/15/ 2024;458:140178. doi:https://doi.org/10.1016/j.foodchem.2024.140178\u003c/li\u003e\n\u003cli\u003eTrollet C, Anvar SY, Venema A\u003cem\u003e, et al\u003c/em\u003e. Molecular and phenotypic characterization of a mouse model of oculopharyngeal muscular dystrophy reveals severe muscular atrophy restricted to fast glycolytic fibres. \u003cem\u003eHum Mol Genet\u003c/em\u003e. Jun 1 2010;19(11):2191-207. doi:10.1093/hmg/ddq098\u003c/li\u003e\n\u003cli\u003eVincent TL, Green PJ, Woolfson DN. LOGICOIL--multi-state prediction of coiled-coil oligomeric state. \u003cem\u003eBioinformatics\u003c/em\u003e. Jan 1 2013;29(1):69-76. doi:10.1093/bioinformatics/bts648\u003c/li\u003e\n\u003cli\u003eTavanez JP, Calado P, Braga J, Lafarga M, Carmo-Fonseca M. In vivo aggregation properties of the nuclear poly(A)-binding protein PABPN1. \u003cem\u003eRna\u003c/em\u003e. May 2005;11(5):752-62. doi:10.1261/rna.7217105\u003c/li\u003e\n\u003cli\u003eVerheesen P, de Kluijver A, van Koningsbruggen S\u003cem\u003e, et al\u003c/em\u003e. Prevention of oculopharyngeal muscular dystrophy-associated aggregation of nuclear poly(A)-binding protein with a single-domain intracellular antibody. \u003cem\u003eHuman Molecular Genetics\u003c/em\u003e. 2005;15(1):105-111. doi:10.1093/hmg/ddi432\u003c/li\u003e\n\u003cli\u003eAbramson J, Adler J, Dunger J\u003cem\u003e, et al\u003c/em\u003e. Accurate structure prediction of biomolecular interactions with AlphaFold 3. \u003cem\u003eNature\u003c/em\u003e. 2024/06/01 2024;630(8016):493-500. doi:10.1038/s41586-024-07487-w\u003c/li\u003e\n\u003cli\u003eRuff KM, Pappu RV. AlphaFold and Implications for Intrinsically Disordered Proteins. \u003cem\u003eJournal of Molecular Biology\u003c/em\u003e. 2021/10/01/ 2021;433(20):167208. doi:https://doi.org/10.1016/j.jmb.2021.167208\u003c/li\u003e\n\u003cli\u003ePantic I, Dimitrijevic D, Nesic D, Petrovic D. Gray level co-occurrence matrix algorithm as pattern recognition biosensor for oxidopamine-induced changes in lymphocyte chromatin architecture. \u003cem\u003eJ Theor Biol\u003c/em\u003e. Oct 7 2016;406:124-8. doi:10.1016/j.jtbi.2016.07.018\u003c/li\u003e\n\u003cli\u003eAnvar SY, Raz Y, Verway N\u003cem\u003e, et al\u003c/em\u003e. A decline in PABPN1 induces progressive muscle weakness in oculopharyngeal muscle dystrophy and in muscle aging. \u003cem\u003eAging (Albany NY)\u003c/em\u003e. Jun 2013;5(6):412-26. doi:10.18632/aging.100567\u003c/li\u003e\n\u003cli\u003eRaz V, Routledge S, Venema A\u003cem\u003e, et al\u003c/em\u003e. Modeling oculopharyngeal muscular dystrophy in myotube cultures reveals reduced accumulation of soluble mutant PABPN1 protein. \u003cem\u003eAm J Pathol\u003c/em\u003e. Oct 2011;179(4):1988-2000. doi:10.1016/j.ajpath.2011.06.044\u003c/li\u003e\n\u003cli\u003eRaz V, Buijze H, Raz Y\u003cem\u003e, et al\u003c/em\u003e. A Novel Feed-Forward Loop between ARIH2 E3-Ligase and PABPN1 Regulates Aging-Associated Muscle Degeneration. \u003cem\u003eThe American Journal of Pathology\u003c/em\u003e. 2014/04/01/ 2014;184(4):1119-1131. doi:https://doi.org/10.1016/j.ajpath.2013.12.011\u003c/li\u003e\n\u003cli\u003eMei H, Boom J, El Abdellaoui S\u003cem\u003e, et al\u003c/em\u003e. Alternative Polyadenylation Utilization Results in Ribosome Assembly and mRNA Translation Deficiencies in a Model for Muscle Aging. \u003cem\u003eJ Gerontol A Biol Sci Med Sci\u003c/em\u003e. Jun 1 2022;77(6):1130-1140. doi:10.1093/gerona/glac058\u003c/li\u003e\n\u003cli\u003eCalado A, Tom\u0026eacute; FMS, Brais B\u003cem\u003e, et al\u003c/em\u003e. Nuclear inclusions in oculopharyngeal muscular dystrophy consist of poly(A) binding protein 2 aggregates which sequester poly(A) RNA. \u003cem\u003eHuman Molecular Genetics\u003c/em\u003e. 2000;9(15):2321-2328. doi:10.1093/oxfordjournals.hmg.a018924\u003c/li\u003e\n\u003cli\u003eTrendel J, Schwarzl T, Horos R\u003cem\u003e, et al\u003c/em\u003e. The Human RNA-Binding Proteome and Its Dynamics during Translational Arrest. \u003cem\u003eCell\u003c/em\u003e. Jan 10 2019;176(1-2):391-403.e19. doi:10.1016/j.cell.2018.11.004\u003c/li\u003e\n\u003cli\u003eVonk WIM, Rainbolt TK, Dolan PT, Webb AE, Brunet A, Frydman J. Differentiation Drives Widespread Rewiring of the Neural Stem Cell Chaperone Network. \u003cem\u003eMolecular Cell\u003c/em\u003e. 2020/04/16/ 2020;78(2):329-345.e9. doi:https://doi.org/10.1016/j.molcel.2020.03.009\u003c/li\u003e\n\u003cli\u003eConnolly NMC, Theurey P, Adam-Vizi V\u003cem\u003e, et al\u003c/em\u003e. Guidelines on experimental methods to assess mitochondrial dysfunction in cellular models of neurodegenerative diseases. \u003cem\u003eCell Death \u0026amp; Differentiation\u003c/em\u003e. 2018/03/01 2018;25(3):542-572. doi:10.1038/s41418-017-0020-4\u003c/li\u003e\n\u003cli\u003eChartier A, Klein P, Pierson S\u003cem\u003e, et al\u003c/em\u003e. Mitochondrial Dysfunction Reveals the Role of mRNA Poly(A) Tail Regulation in Oculopharyngeal Muscular Dystrophy Pathogenesis. \u003cem\u003ePLOS Genetics\u003c/em\u003e. 2015;11(3):e1005092. doi:10.1371/journal.pgen.1005092\u003c/li\u003e\n\u003cli\u003eAnvar SY, t Hoen PA, Venema A\u003cem\u003e, et al\u003c/em\u003e. Deregulation of the ubiquitin-proteasome system is the predominant molecular pathology in OPMD animal models and patients. \u003cem\u003eSkelet Muscle\u003c/em\u003e. Apr 4 2011;1(1):15. doi:10.1186/2044-5040-1-15\u003c/li\u003e\n\u003cli\u003eRibot C, Soler C, Chartier A\u003cem\u003e, et al\u003c/em\u003e. Activation of the ubiquitin-proteasome system contributes to oculopharyngeal muscular dystrophy through muscle atrophy. \u003cem\u003ePLoS Genet\u003c/em\u003e. Jan 2022;18(1):e1010015. doi:10.1371/journal.pgen.1010015\u003c/li\u003e\n\u003cli\u003eKatahira J. Nuclear export of messenger RNA. \u003cem\u003eGenes (Basel)\u003c/em\u003e. Mar 31 2015;6(2):163-84. doi:10.3390/genes6020163\u003cstrong\u003e\u003c/strong\u003e\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"protein aggregates, aggregates structure, imaging facilities, muscle, OPMD","lastPublishedDoi":"10.21203/rs.3.rs-5676243/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5676243/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThe accumulation of intracellular protein aggregates is a hallmark of aging. In hereditary adult-onset neuromuscular diseases (NMDs), these aggregates are not only characteristic but also pathogenic, marking age-related neuromuscular disorders. The transition from age-associated non-pathogenic aggregates to disease-driving pathogenic aggregates remains poorly understood. Poly(A) binding protein nuclear 1 (PABPN1) forms non-pathogenic nuclear aggregates in post-mitotic aged cells. However, a short trinucleotide expansion in PABPN1 leads to muscle dysfunction in Oculopharyngeal Muscular Dystrophy (OPMD), where insoluble nuclear aggregates in skeletal muscle become a defining disease feature. Combining an array of advanced imaging modalities, we examined the morphological differences between nuclear aggregates formed by non-pathogenic and pathogenic PABPN1 variants. Through micro- to nanoscale analyses, we identified key structural differences in the aggregation propensity of these variants in both differentiated and undifferentiated muscle cells and linked these differences to mRNA cellular dysfunctions. Our findings provide new insights into the structural distinctions between pathogenic and non-pathogenic aggregates and their implications for cellular dysfunction in neuromuscular diseases.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e","manuscriptTitle":"Structural Features of Pathogenic Aggregates Correlate with Cell Pathology in Differentiated Cells","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-01-14 15:56:56","doi":"10.21203/rs.3.rs-5676243/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"2e18a573-13bb-4286-9b33-ec3ebf2aeac8","owner":[],"postedDate":"January 14th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":42635549,"name":"Biological sciences/Neuroscience/Molecular neuroscience"},{"id":42635550,"name":"Biological sciences/Cell biology/Cellular imaging/Super-resolution microscopy"},{"id":42635551,"name":"Biological sciences/Structural biology/Electron microscopy/Cryoelectron microscopy"}],"tags":[],"updatedAt":"2025-04-01T14:51:41+00:00","versionOfRecord":[],"versionCreatedAt":"2025-01-14 15:56:56","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-5676243","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5676243","identity":"rs-5676243","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}