Ptbp1 Condensates With Phase Separation‐Like Features in Cardiomyocyte Nuclei Regulate Splicing of Proliferative Genes in Neonatal Cardiomyocytes

ABSTRACT

Cardiovascular diseases remain the leading cause of morbidity and mortality worldwide; however, effective clinical therapies that achieve myocardial regeneration are still lacking, underscoring the urgent need for novel therapeutic strategies. In this study, we identified an RNA‐binding protein Ptbp1, a key component of RNA metabolism, as a critical regulator of cardiomyocyte proliferation via the formation of biomolecular condensates. We found that Ptbp1 exhibits granule‐like structures in cardiomyocyte nuclei during fetal to early postnatal stages. These Ptbp1 foci are phase‐separated liquid droplets formed via its multivalent RNA binding domains. Functional analyses revealed that Ptbp1 is essential for cardiomyocyte proliferation in neonatal mouse primary culture assays. Furthermore, proximity labeling using a primary antibody‐based approach combined with RIP‐seq in highly purified cardiomyocyte populations demonstrated that these condensates are enriched with multiple proliferation‐associated factors. Our findings identify a novel phase‐separated condensate in cardiomyocytes and provide a new potential therapeutic target for cardiac regeneration.

Phase‐separated Ptbp1 droplets form molecular complexes containing multiple proliferation‐associated molecules. Ptbp1 granules containing four RNA recognition motifs (RRMs) domains are present in the nuclei of mouse cardiomyocytes specifically during the proliferative stage. Within these Ptbp1 droplets, proliferation‐associated RNA‐binding proteins (RBPs) and RNAs assemble through specific domain‐mediated interactions, forming a highly dynamic and complex molecular network with phase‐separation‐like properties. Images were created with BioRender.com.

1. Introduction

Cardiovascular diseases are the leading cause of death worldwide (Global Burden of Cardiovascular Diseases and Risks 2023 Collaborators ²⁰²⁵). One of the major reasons for this is the lack of myocardial regenerative capacity in adult mammals. Mammalian cardiomyocytes retain the ability to proliferate and mount regenerative responses during the neonatal period; however, this capacity is rapidly lost soon after birth (Porrello et al. 2011; Walsh et al. ²⁰¹⁰; Soonpaa et al. ¹⁹⁹⁶). In adulthood, cardiomyocytes rarely proliferate, and following injury such as myocardial infarction, the heart fails to regenerate functional myocardium and instead undergoes fibrotic scar formation, ultimately leading to heart failure (Bergmann et al. 2009; Porrello et al. ²⁰¹¹). Although recent advances in medical care have enabled early intervention for myocardial infarction, these approaches are largely limited to preventing further loss of viable and functional myocardium and there remains a critical need to develop simple and accessible therapeutic strategies to promote cardiac regeneration after injury (Thrane et al. 2023).

In recent years, biomolecular condensates formed through phase separation have emerged as a novel class of therapeutic targets beyond conventional drug discovery approaches (F. Wang and Zhang ²⁰²⁴). These dynamic assemblies, often driven by multivalent interactions among RNA molecules and RNA‐binding proteins (RBPs), regulate diverse cellular processes depending on their molecular composition (Roden and Gladfelter 2021; Lin et al. ²⁰¹⁵; Jain and Vale ²⁰¹⁷). Accumulating evidence suggests that targeting such condensates using peptides or small molecules can exert therapeutic effects in diseases such as cancer and neurological disorders, highlighting their potential as druggable structures (F. Xie et al. ²⁰²⁵; Datar et al. ²⁰²⁵; Deng et al. ²⁰²⁵). In the cardiovascular field, although several types of condensates have been reported in cardiomyocytes, including those associated with cell cycle regulation and cardiac function (Jiang et al. 2024; Gong et al. ²⁰²³; Gan et al. ^{2022, 2024}; Lei et al. ²⁰²⁵), their functions in cardiomyocytes have yet to be fully understood.

We therefore sought to identify novel RBP candidates that may regulate proliferation through phase separation, by reanalyzing publicly available omics datasets (Gan et al. 2022). Through reanalysis of publicly available datasets, we identified polypyrimidine tract binding protein 1 (Ptbp1), a splicing‐related RBP containing multiple RNA recognition motif (RRM) domains, as a factor strongly associated with proliferative cardiomyocytes. Interestingly, Ptbp1 exhibited a punctate nuclear localization pattern specifically during the neonatal period, consistent with phase‐separation‐like properties. As expected, Ptbp1 was required for neonatal cardiomyocyte proliferation, and comprehensive identification and analysis of proteins proximal to Ptbp1 condensates in purified cardiomyocytes revealed enrichment of proliferation‐associated factors. Collectively, our findings identify a previously unrecognized phase‐separation‐like Ptbp1 condensate in cardiomyocyte nuclei.

2. Results

2.1. Expression of RRM Genes Is Associated With Proliferation in the Mouse Embryonic Heart

To identify RBPs that regulate cardiomyocyte proliferative capacity, we reanalyzed previously published single‐cell RNA sequencing (scRNA‐seq) data from mouse hearts collected at different developmental stages (R. H. Wang and Thakar ²⁰²⁴). The cell population annotated as cardiomyocytes was extracted based on the cell type labels from embryonic (E12.5) to postnatal (P9) stages. To quantify gene set‐level expression in individual cells, we used the AddModuleScore method, which assigns a score to each cell based on the aggregated expression of a predefined gene set (R. H. Wang and Thakar ²⁰²⁴). Using this approach, proliferation scores and RBP scores were assigned to individual cardiomyocytes based on cell cycle markers built into Seurat and an RBP gene list from the EuRBPDB database, respectively (Figure 1A,B). Based on these scores, cells in the top 10% proliferation score or RBP score across all stages were defined. As expected, the fraction of cells with top 10% proliferation scores progressively decreased toward the postnatal stage P9 (Figure 1C). Similarly, the fraction of cells with top 10% RBP scores were enriched during embryonic stages, when cardiomyocytes have higher proliferative capacity (Figure 1D). Comparison of proliferation scores between cardiomyocytes in the top 10% and bottom 10% of RBP scores across all stages showed that proliferation scores were significantly higher in the RBP top 10% group (Figure 1E). These results indicate an association between higher RBP expression and increased proliferative capacity in cardiomyocytes.

FIGURE 1.

Expression analyses of RBPs in embryonic and postnatal cardiomyocytes. (A) Schematic overview of the analysis pipeline. Previously published scRNA‐seq data from mouse hearts across developmental stages were reanalyzed. Cardiomyocytes were scored based on the expression of cell cycle markers and RNA‐binding proteins (RBPs) annotated in the EuRBPDB to screen for candidate regulators of proliferative capacity. Images were created with BioRender.com. (B) UMAP visualization of cardiomyocytes extracted from E12.5, E14.5, E16.5, E18.5, P1, and P9 datasets. Left: all cardiomyocytes projected onto a shared UMAP space. Right: cardiomyocytes from each stage shown separately in the same UMAP space. (C, D) Distribution of proliferation scores (C) and RBP scores (D) across individual cardiomyocytes projected onto UMAP space (left). Bar plots (right) show the proportion of cells in the top 10% for each score, defined across all stages within each stage. (E) Comparison of proliferation scores between cardiomyocytes in the top 10% and bottom 10% of RBP scores. Statistical significance was assessed using the Wilcoxon rank‐sum test (F) Classification of RBPs expressed in neonatal cardiomyocytes based on bulk RNA‐seq data. Functional domain annotations were derived from the Mus musculus dataset in EuRBPDB. Left: bar plot showing the number of genes per domain. Right: pie chart showing the proportion of each domain among expressed RBPs. (G) Comparison of RBP domain‐associated expression between neonatal and adult cardiomyocytes using bulk RNA‐seq data. Each dot represents the mean log2(CPM + offset) of RBPs within each domain. Statistical significance was assessed using an unpaired t‐test with Benjamini–Hochberg (BH) correction. Adjusted p‐values (Padj_BH) are indicated as follows: ns, not significant; *p < 0.05; **p < 0.01. (H) Pearson correlation between proliferation scores (y‐axis) and RBP domain scores (x‐axis) across individual cardiomyocytes from scRNA‐seq data. Each dot represents a single cell and is colored by developmental stage. Correlation coefficient (r) and FDR are indicated in each RBP domain. (I) Gene Ontology (GO) analysis of cardiomyocytes in the top 10% of RRM scores. GO terms were summarized using simplified categories. RRM genes themselves were excluded from the analysis. Terms are ranked by adjusted p‐value. The x‐axis indicates gene counts assigned to each term.

To further investigate which types of RBPs are related to proliferative capacity, bulk RNA‐seq data comparing neonatal and adult cardiomyocytes were analyzed (Quaife‐Ryan et al. ²⁰¹⁷). Based on annotations from the EuRBPDB database, functional domains associated with RBPs expressed in cardiomyocytes were identified (Figure 1F). Given that cardiomyocytes lose proliferative capacity after birth, we examined the expression of genes containing the 10 RBP‐associated domains most frequently found among genes expressed in cardiomyocytes, ranked by the number of domain‐containing genes. Genes containing KH_1, ResIII, RRM, and RNase domains showed significantly decreased expression from neonatal to adult stages (Figure 1G). RBP genes containing these four domains were scored using the scRNA‐seq dataset to assess correlations with proliferation scores at the single‐cell level. While all of these four gene groups showed statistically significant correlations (false discovery rate [FDR] < 0.05), the genes containing RRM showed the strongest correlation with proliferation score (Pearson's r = 0.59; Figure 1H). In addition, Gene Ontology (GO) analysis of the top 10% of cardiomyocytes with high RRM scores revealed significant enrichment of proliferation‐related terms, such as chromosome segregation and DNA replication (Figure 1I). Taken together, these results indicate that RBP expression levels are associated with proliferative capacity in mouse cardiomyocytes, with RRM domain‐containing genes showing the highest correlation and decreased expression in adult stages. We therefore further investigated the role of RRM genes in cardiomyocyte proliferation.

2.2. The RRM‐Containing Gene Ptbp1 Is Enriched in Proliferative Cardiomyocytes

To identify RRM domain‐containing genes (hereinafter referred to as RRM genes) that may play key roles in regulating cardiomyocyte proliferation, we analyzed previously published transcriptome datasets together with the EuRBPDB database (Quaife‐Ryan et al. ²⁰¹⁷). Cardiac‐specific expression was not included as a selection criterion. Among the RRM genes listed in EuRBPDB, 156 were found to be expressed in neonatal cardiomyocytes (Figure 2A). From these, we selected RRM genes that showed sufficiently high expression levels in neonatal cardiomyocytes (Figure 2A, mean_P1 ≥ 50 across technical replicates) and exhibited at least a twofold decrease in expression from neonatal to adult stages (Figure 2A, log2FC_P56_vs_P4 < −1.0). This resulted in 29 candidate RRM genes, ranked by the degree of downregulation (Figure 2A).

FIGURE 2.

The RRM domain‐containing protein Ptbp1 forms nuclear foci in cardiomyocytes during the neonatal period. (A) Schematic overview of the screening pipeline for RRM domain‐containing RNA‐binding proteins (RRM genes). Previously published bulk RNA‐seq data of mouse hearts were reanalyzed. RRM genes were filtered based on the following criteria in cardiomyocytes: high expression in neonatal samples (mean_P1 ≥ 50), significant downregulation from neonatal to adult stages (log2FC_P56_vs_P1 < −1.0), and the presence of embryonic lethal phenotypes in MGI. (B) UMAP visualization of cardiomyocytes showing the distribution of cells with the top 10% Ptbp1 expression defined across all stages. Red dots indicate top 10% cells, while gray dots represent the remaining cells. (C) Correlation analysis between Ptbp1 expression and gene expression across individual cardiomyocytes. Each dot represents a gene. The x‐axis indicates Pearson's correlation coefficient (r), and the y‐axis indicates −log10(FDR). Genes meeting the thresholds (r ≥ 0.2, FDR < 0.05, and nonzero cells ≥ 10%) are highlighted in red. (D) Immunofluorescence analysis of primary‐cultured cardiomyocytes isolated from neonatal and adult hearts. Left: schematic overview of the experimental design. Right: representative images of immunofluorescence. Blue, DAPI (nuclear marker); red, Ptbp1; gray, cTnT (cardiac troponin T, cardiomyocyte marker). Scale bars indicate 20 μm. Schematic images were created with BioRender.com. (E) Immunofluorescence images of mouse heart tissue across developmental stages. Blue, DAPI (nuclear marker); green, Ptbp1; red, cTnT; white, WGA (wheat germ agglutinin, cell boundary marker). White arrowheads indicate Ptbp1 nuclear foci. Scale bars indicate 2 μm. (F) Quantification of the proportion of cardiomyocyte nuclei containing Ptbp1 foci across developmental stages (E18.5, P1, P7, and P14). Data are presented as mean ± SD (n = 3 biological replicates). Statistical significance was assessed using one‐way ANOVA followed by Dunnett's multiple comparison test. ns, not significant; ***p < 0.001; ****p < 0.0001.

Because these genes were suggested to be involved in cardiomyocyte proliferation during embryonic and neonatal stages, we reasoned that the loss of key regulators may lead to developmental defects, potentially including embryonic lethality. Therefore, we examined the knockout phenotypes of the top 10 candidate RRM genes. Among them, Tra2b, Ptbp1, Srsf3, and Srsf1 were associated with embryonic lethality in knockout mice (Yu et al. 2006; Shibayama et al. ²⁰⁰⁹; Mende et al. ²⁰¹⁰; Jumaa et al. ¹⁹⁹⁹). Among these candidates, Ptbp1 contains multiple RRM domains, suggesting a high capacity to interact with diverse RNA molecules. This structural feature raises the possibility that Ptbp1 may coordinate a broad range of cellular functions, including cell proliferation, through RNA‐mediated regulatory mechanisms. Proteins containing multiple RRM domains can potentially recognize distinct RNA sequences or structures and participate in the formation of different ribonucleoprotein complexes, thereby enabling the coordinated regulation of multiple RNA targets. Such broad post‐transcriptional regulatory capacity may be particularly relevant to cardiomyocyte proliferation, which requires coordinated changes not only in cell cycle regulators but also in cellular growth, metabolism, and structural organization. Therefore, among the identified candidates, Ptbp1 represented a plausible regulatory hub that could integrate multiple cellular processes required for cardiomyocyte proliferation. Indeed, Ptbp1 is known to regulate multiple biological processes through alternative splicing and has been implicated in the control of cell proliferation in multiple types of cells, including cancer cells and B cells (Monzón‐Casanova et al. ²⁰¹⁸; Zhu et al. ²⁰²⁰). However, whether Ptbp1 has a comparable function in cardiomyocytes and contributes to the regulation of their proliferative capacity remained unclear. Based on this rationale, we selected Ptbp1 for further analysis.

Cardiomyocytes with the top 10% Ptbp1 expression levels were enriched at proliferative stages on UMAP (Figure 2B). Furthermore, although only a small number of genes were positively correlated with Ptbp1 expression across all stages (r ≧ 0.2, FDR < 0.05), these genes were mostly associated with cell cycle regulation (Figure 2C). These findings suggest that the RRM gene Ptbp1 is strongly associated with the proliferative state and may regulate cardiomyocyte proliferation.

Ptbp1 is known to regulate alternative splicing and plays diverse roles in multiple cellular processes, including RNA metabolism, cellular maturation, metabolic switching, and apoptosis (Zhu et al. 2020; Z. Liu et al. ²⁰¹⁷; H.‐L. Liu et al. ²⁰²³; Caruso et al. ²⁰¹⁷). In immature cardiomyocytes, Ptbp1 is highly expressed and regulates metabolic states, development, and isoform switching to maintain an immature phenotype in coordination with other RBPs and cellular networks (Z. Li et al. ²⁰²⁵; Fochi et al. ²⁰²⁰). In contrast, its role in cardiomyocyte proliferation has remained unclear, with previous reports suggesting context‐dependent pro‐proliferative and proapoptotic effects (J. Chen et al. ²⁰²⁵; Zhang et al. ²⁰⁰⁹).

To investigate the role of Ptbp1 in cardiomyocyte cell cycle regulation, we first examined its expression pattern in cardiomyocytes at neonatal and adult stages. Consistent with the transcriptome analysis, Ptbp1 protein was predominantly localized within the nuclei of cardiomyocytes during embryonic and early postnatal stages, whereas no detectable signal was observed in the cytoplasm (Figure 2D,E). Furthermore, the nuclear Ptbp1 signal markedly diminished within the first postnatal week and became undetectable in adult cardiomyocytes (Figure 2D,E). Notably, Ptbp1 formed granule‐like structures within cardiomyocyte nuclei, and these nuclear Ptbp1 granules were significantly diminished after the first postnatal week (Figure 2F). To further characterize the properties of these nuclear Ptbp1 condensates, we next examined their distribution patterns and subnuclear localization in cardiomyocytes, showing a majority of cardiomyocyte nuclei had one or two relatively large granules (Figure S1A). In contrast, non‐cardiomyocytes exhibited strong, diffuse nuclear Ptbp1 signals regardless of developmental stage, revealing a clear difference in the subnuclear organization of Ptbp1 between cardiomyocytes and non‐cardiomyocytes (Figure S1B). Interestingly, although the expression of Ptbp1 is enriched in proliferative cardiomyocytes, Ptbp1 condensates were predominantly observed in the G0 phase of the cell cycle in neonatal cardiomyocytes (Figure S1C,D). Previous studies have shown that Ptbp1 localizes as punctate structures adjacent to nucleoli in cancer cells, known as perinucleolar compartments (PNCs) (Huang et al. 1997). However, nuclear Ptbp1 granules in cardiomyocytes were not positioned adjacent to nucleoli, marked by a nucleolar marker Fibrillarin, suggesting that these structures are distinct from canonical PNCs (Figure S1E). These findings indicate that the expression of Ptbp1, a protein containing four RRM domains, forms nuclear granules and is associated with proliferative cardiomyocytes.

2.3. Ptbp1 Foci Exhibit Properties of Liquid–Liquid Phase Separation

We next investigated the properties of Ptbp1 to form its nuclear granule. Interestingly, overexpression of human PTBP1 in primary‐cultured neonatal cardiomyocytes resulted in predominantly cytoplasmic localization and failed to induce nuclear condensate formation (Figure S2A). Previous studies have shown that Ptbp1 localization is regulated not only by its nuclear localization signal (NLS) but also by phosphorylation at serine 16 (S16), which promotes nuclear export (J. Xie et al. ²⁰⁰³). Therefore, we examined the phosphorylation status of Ptbp1 in cardiac tissue using antibodies against non‐phosphorylated Ptbp1 at serine 16 (anti‐S16) and phosphorylated Ptbp1 at serine 16 (anti‐pS16) (J. Xie et al. ²⁰⁰³). As expected, Ptbp1 condensates in cardiomyocytes were not detected by the anti‐pS16 antibody, whereas they were clearly co‐localized with the anti‐S16 signal (Figure 3A). We therefore overexpressed a phosphorylation‐deficient mutant (S16A) form of human PTBP1 in cultured cardiomyocytes. Consistent with the result from immunostaining with phosphorylation‐specific antibodies, S16A increased the proportion of cardiomyocytes with nuclear localization of PTBP1 (Figure S2A,B), although the fraction of cardiomyocytes showing nuclear condensates remained limited (Figure S2C). We then generated multiple constructs based on known mouse Ptbp1 variants and evaluated their localization and condensate‐forming ability in cardiomyocytes. Mouse Ptbp1 contains two distinct NLS regions surrounding S16 within the N‐terminal 60 amino acids, and the linker region between RRM2 and RRM3 is an intrinsically disordered region (IDR) (de Vries et al. ²⁰²⁴; Romanelli and Morandi ²⁰⁰²; H.‐L. Liu et al. ²⁰²³). Several Ptbp1 isoforms differ in these regions, including variants lacking part of the N‐terminal 60 amino acids or containing shortened linker regions between RRM2 and RRM3 (Figure S2D). Among these, mouse Ptbp1 isoform 1 showed the highest tendency toward nuclear localization and condensate formation (Figure S2E–G). Furthermore, to enhance nuclear restriction, we introduced an additional NLS sequence as shown for isoform 1 and 2 in Figure S2D. By combining this modification with the phosphorylation‐deficient S16A mutation, we generated a mouse Ptbp1 isoform 1 S16A‐NLS‐Flag construct (Figure S2H). This construct exhibited near‐complete nuclear localization and markedly improved the proportion of cells containing Ptbp1 condensates (Figure S2I,J).

FIGURE 3.

Ptbp1 foci exhibit properties of liquid–liquid phase separation in cardiomyocytes.** (A) Immunofluorescence images of heart tissue at E18.5. Blue, DAPI; green, Ptbp1; red, phospho‐S16 Ptbp1 (upper) or S16Ptbp1 (lower); white, cTnT. White arrowheads indicate Ptbp1 nuclear foci. Scale bars indicate 2 μm. (B) Schematic overview of Ptbp1 constructs. All contain the S16A mutation. In addition to full‐length Ptbp1, four truncated Ptbp1 constructs—ΔRRM1, ΔRRM2, ΔRRM3, and ΔRRM4—were generated. Monomeric GFP was fused to the N‐terminus, and FLAG and an additional NLS were added to the C‐terminus. Images were created with BioRender.com. (C) Immunofluorescence images of primary‐cultured neonatal cardiomyocytes overexpressing exogenous Ptbp1. Blue, DAPI; green, GFP (exogenous Ptbp1); red, FLAG (exogenous Ptbp1); gray, α‐actinin (cardiomyocyte marker). Red arrowheads indicate exogenous Ptbp1 showing double positivity for GFP and FLAG. Lower panels show higher‐magnification images of the boxed regions. Scale bars indicate 10 μm. (D) Quantification of exogenous Ptbp1 granule formation. The denominator represents the number of transfected (GFP‐positive) cardiomyocytes, and the numerator represents GFP/FLAG double‐positive granule‐containing cardiomyocytes. Data are presented as mean ± SD (n = 3 biological replicates). Statistical significance was assessed using one‐way ANOVA followed by Dunnett's multiple comparison test. ns, not significant; ***p < 0.001; ****p < 0.0001. (E) Time‐lapse imaging of Ptbp1–overexpressing primary cardiomyocytes. Left: bright‐field and GFP images. The scale bar indicates 20 μm. Right: time‐lapse images (0 min, 10 min, 20 min, 30 min, 50 min, and 70 min). Arrowheads indicate fusion events (green and red) and fused granules (yellow). White dashed circles indicate putative fission events. Scale bars indicate 5 μm. (F, G) Fluorescence recovery after photobleaching (FRAP) analysis in H9c2 cells. Left: representative FRAP images. Right: quantification of fluorescence intensity recovery over time. The white arrow indicates the bleached areas. The blue line indicates the average trace, and the shaded areas around the trace represent SD (n = 12 cells). Scale bar, 5 μm.

Using this optimized Ptbp1 construct, we next investigated whether RRM domains, which have been implicated in the droplet formation by RBPs (Banani et al. 2016), are also required for Ptbp1 condensate formation in cardiomyocyte nuclei. A full‐length S16A Ptbp1 construct was generated with monomeric GFP fused to the N‐terminus and FLAG tag and NLS added to the C‐terminus. Subsequently, deletion constructs lacking each individual RRM domain (ΔRRM1, ΔRRM2, ΔRRM3, and ΔRRM4) were overexpressed in primary‐cultured neonatal cardiomyocytes (Figure 3B,C). Only puncta showing co‐localization of GFP and FLAG signals were considered Ptbp1 granules to minimize the detection of artifact signals (Figure S3A). Although deletion of RRM1 did not significantly affect granule formation, deletion of RRM2, RRM3, or RRM4 resulted in a marked and nearly complete loss of granule formation (Figure 3D). These findings suggest that all RRM domains except RRM1 are required for the formation of Ptbp1 nuclear condensates in cardiomyocytes. It has been reported that Ptbp1 can form RNA‐dependent condensates in cell‐free systems (P. Li et al. ²⁰¹²; de Vries et al. ²⁰²⁴; Lin et al. ²⁰¹⁵); however, whether it forms liquid‐like droplets in living cells remains unclear. To further assess the dynamic properties of these granules, we performed time‐lapse imaging of full‐length Ptbp1 condensates in cultured cardiomyocytes. As expected, these granules exhibited high mobility, including fusion events between droplets and fission into multiple smaller droplets (Figure 3E, Video S1). In addition, fluorescence recovery after photobleaching (FRAP) analysis of whole granules revealed rapid recovery of fluorescence, suggesting dynamic exchange of Ptbp1 with the surrounding environment (Figure 3F,G, Video S2). Taken together, these results indicate that in cardiomyocytes, Ptbp1 undergoes changes in subcellular localization depending on its phosphorylation status at S16, and forms dynamic liquid–liquid phase‐separated‐like condensates that require RRM2–4 for their formation.

2.4. Ptbp1 Is Required for Neonatal Cardiomyocyte Proliferation

To determine whether Ptbp1 is required for cardiomyocyte proliferation, we next performed loss‐of‐function experiments using siRNA‐mediated knockdown of Ptbp1 in neonatal cardiomyocytes (Figure 4A). As expected, Ptbp1 foci were efficiently eliminated upon knockdown (Figure 4B,C). Concomitantly, the proportion of cardiomyocytes positive for cell cycle markers Ki67 and phospho‐histone H3 (pH 3) was significantly reduced (Figure 4D–G). These results indicate that Ptbp1 is required for maintaining proliferative capacity in neonatal cardiomyocytes.

FIGURE 4.

Ptbp1 is required for neonatal cardiomyocyte proliferation. (A) Schematic overview of the experimental design for Ptbp1 knockdown analysis. Primary neonatal cardiomyocytes were cultured and transfected with siRNA (final concentration, 20 nM) at 4 h after seeding and again 48 h later. Cells were fixed 5 days after seeding and subjected to immunostaining to assess cell cycle status. Images were created with BioRender.com. (B) Representative immunofluorescence images of primary‐cultured neonatal cardiomyocytes following Ptbp1 knockdown. Blue, DAPI; red, Ptbp1; gray, cTnT. White arrowheads indicate nuclei containing Ptbp1 foci. White arrow indicates non‐cardiomyocyte with diffused Ptbp1 signal. Scale bars indicate 20 μm. (C) Rate of Ptbp1 foci–positive cardiomyocyte nuclei following Ptbp1 knockdown. (D) Immunofluorescence images of primary‐cultured neonatal cardiomyocytes stained for Ki67. Blue, DAPI (nuclear marker); green, Ki67; gray, cTnT (cardiomyocyte marker). White arrowheads indicate Ki67‐positive nuclei. Scale bars indicate 20 μm. (E) Rate of Ki67‐positive cardiomyocytes following Ptbp1 knockdown. (F) Immunofluorescence images of primary‐cultured neonatal cardiomyocytes stained for phospho‐histone H3 (pH 3). Blue, DAPI (nuclear marker); green, pH 3; gray, cTnT (cardiomyocyte marker). White arrowheads indicate pH 3‐positive nuclei. Scale bars indicate 20 μm. (G) Rate of pH 3‐positive cardiomyocytes following Ptbp1 knockdown. Data in panels C, E, and G are presented as mean ± SD (n = 3 biological replicates). Statistical significance was assessed using an unpaired t‐test. *p < 0.05; **p < 0.01; ***p < 0.001.

To further investigate whether Ptbp1 is sufficient to induce cell cycle reentry in mature cardiomyocytes, we next overexpressed Ptbp1 in adult cardiomyocytes, which normally lack proliferative activity and show no detectable Ptbp1 nuclear foci (Figure 2D). Cardiomyocytes were isolated from adult mouse hearts and cultured, followed by overexpression of a granule‐inducing Ptbp1 construct, and proliferation was assessed by EdU incorporation (Figure S4A). Although the efficiency of mRNA overexpression in adult cardiomyocytes was moderate (~30%–40%, data not shown), only a subset of cells (~10%) exhibited detectable Ptbp1 foci, while the majority of cells showed a diffuse nuclear distribution of Ptbp1 (Figure S4B). Under these conditions, EdU incorporation was only observed in non‐cardiomyocytes, but not in cardiomyocytes (Figure S4B). These results indicate that, at least under short‐term culture conditions, overexpression of Ptbp1 alone is insufficient to induce cell cycle reentry in adult cardiomyocytes.

2.5. Proximity Labeling of Ptbp1 Condensates Reveals a Multifunctional Complex Enriched for RNA Splicing and Proliferation‐Related Factors

We then aimed to assess the factors required for the formation of Ptbp1 nuclear foci in cardiomyocytes, based on the hypothesis that the formation of Ptbp1 condensates requires Ptbp1‐interacting proteins. Indeed, Ptbp1 is known to interact not only with RNAs but also with multiple RBPs (Masuda et al. 2024). We therefore sought to identify proteins enriched within Ptbp1 condensates in cardiomyocyte nuclei. As shown in Figure 4B, Ptbp1 is more diffusely expressed in non‐cardiomyocytes within cardiac tissue. To capture proteins specifically enriched in the granular state within cardiomyocytes, we first established a method to remove non‐cardiomyocytes from the cardiomyocyte culture system.

Neonatal cardiomyocytes are capable of utilizing lactate as an energy source (Fisher et al. 1981), and this property has previously been exploited for the purification of cardiomyocytes derived from induced pluripotent stem cells (Tohyama et al. 2013, 2016). Based on this feature, we applied a similar strategy to primary neonatal cardiomyocyte cultures by using a lactate‐ and BSA‐based nutrient‐restricted medium (Figure S5A). Under standard culture conditions, non‐cardiomyocyte populations occupied a considerable proportion of the cultured cells, showing diffuse Ptbp1 patterns in nuclei (Figure S5B). In contrast, culture in lactate‐ and BSA‐only medium markedly reduced the presence of non‐cardiomyocytes, resulting in highly purified cardiomyocyte cultures (Figure S5B). Importantly, Ptbp1 maintained its punctate condensate pattern under these purified conditions (Figure S5B).

To identify proteins co‐localizing with Ptbp1 nuclear foci in cardiomyocytes, we then performed a proteomic analysis (Figure S5C,D; Figure 5A). To capture the dynamic and weak protein–protein interactions within phase‐separated condensates, we employed an antibody‐based proximity labeling method using biotinylation after fixation (hereafter referred to as BAR) in purified cardiomyocytes, rather than conventional pull‐down approaches (Figure S5C,D; Figure 5A) (Bar et al. 2018). Candidate proteins robustly enriched in Ptbp1 foci within cardiomyocyte nuclei were defined as those showing greater than fivefold enrichment compared with non‐cardiomyocytes, in which Ptbp1 showed predominantly diffuse nuclear localization, and these were subjected to GO analysis (Figure 5B). As expected, enriched proteins included numerous splicing‐related factors and RNA metabolism‐associated factors, as well as those involved in proliferation‐related pathways (Figure 5C). Furthermore, to determine whether the enrichment of proliferation‐associated proteins reflected the intrinsic binding properties of Ptbp1 itself or was specific to condensate‐associated Ptbp1, we excluded proteins commonly enriched in non‐cardiomyocytes and reanalyzed the dataset. This analysis revealed a stronger enrichment of GO terms related to cell proliferation. (Figure S6A). Together, these findings suggest that Ptbp1 condensates function as multifunctional regulatory complexes involved in both splicing and proliferation, potentially through interactions with both DNA and RNA in Ptbp1 foci (Figure 5C). In addition, based on phase‐separation‐associated proteins curated in the PhaSePro database, Ptbp1 condensates contained a higher proportion of phase‐separation‐associated proteins compared with non‐cardiomyocytes (Figure S6B) (Mészáros et al. 2020). These findings further support that Ptbp1 condensates exhibit characteristics consistent with phase‐separated structures.

FIGURE 5.

Proximity labeling of Ptbp1 condensates reveals enrichment of RNA splicing and proliferation‐associated proteins. (A) Overview of the BAR (biotinylated antibody recognition) proximity labeling method. Highly purified cardiomyocytes were fixed and incubated with a primary antibody against Ptbp1, followed by an HRP‐conjugated secondary antibody recognizing the anti‐Ptbp1 primary antibody. Biotin‐tyramide labeling was used to tag proteins proximal to Ptbp1, which were subsequently captured and analyzed by mass spectrometry. Inset: immunofluorescence validation of BAR efficiency. Green, endogenous Ptbp1; red, biotin signal. Biotinylation is absent without a primary antibody. White arrowheads indicate colocalization of endogenous Ptbp1 and biotin signal. Schematic images were created with BioRender.com. (B) Identification of proteins enriched in Ptbp1 condensates. After quality filtering, 6354 proteins were detected; proteins showing ≥ fivefold enrichment in Ptbp1 antibody (+) samples were defined as enriched. In total, 240 proteins were identified as Ptbp1 condensate‐enriched proteins. (C) Gene Ontology (GO) analysis of proteins enriched in Ptbp1 condensates. Top 10 terms ranked by adjusted p‐value. GO terms were summarized into simplified categories to reduce redundancy among closely related terms. (D) Immunofluorescence showing subcellular localization of Hnrnph1 and Matr3 relative to Ptbp1. Domain structures are indicated above each panel. White arrowheads indicate colocalization with Ptbp1 condensates. Schematic images were created with BioRender.com. (E) Expression of Ptbp1, Hnrnph1, and Matr3 in cardiomyocytes from neonatal and adult stages based on previously published transcriptomic data. The y‐axis indicates expression levels (CPM). Statistical significance was assessed using an unpaired t‐test with Benjamini–Hochberg (BH) correction. ns, not significant; *p < 0.05; **p < 0.01; ***p < 0.001. (F) Immunofluorescence images of primary‐cultured neonatal cardiomyocytes. Blue, DAPI; red, Ptbp1; gray, cTnT. White arrowheads indicate nuclei with Ptbp1 condensates. Scale bars indicate 20 μm (yellow) and 10 μm (white). (G) Quantification of the proportion of cardiomyocyte nuclei containing Ptbp1 condensates. Data are presented as mean ± SD (n = 3 biological replicates). Statistical significance was assessed using one‐way ANOVA followed by Dunnett's multiple comparison test. ns, not significant; ***p < 0.001; ****p < 0.0001. (H) Immunofluorescence images of primary‐cultured neonatal cardiomyocytes stained for cell cycle markers. Blue, DAPI; green, Ki67 (upper) or phospho‐histone H3 (pH 3, lower); gray, cTnT. Scale bars indicate 20 μm. (I, J) Quantification of Ki67‐ (I) and pH 3‐positive (J) cardiomyocytes. Data are presented as mean ± SD (n = 3 biological replicates). Statistical significance was assessed using one‐way ANOVA followed by Dunnett's multiple comparison test. *p < 0.05; **p < 0.01; ***p < 0.001.

Next, to further investigate the relationship between Ptbp1 condensate‐associated proteins and cardiomyocyte proliferation, we focused on highly enriched factors within Ptbp1 condensates whose expression decreases postnatally. Among these, we selected Hnrnph1, which contains multiple RRM domains and is associated with RNA splicing, and Matr3, which possesses both DNA‐ and RNA‐binding capabilities and has been implicated in cardiomyocyte maturation (Taliani et al. 2023; Quintero‐Rivera et al. ²⁰¹⁵) (Figure 5D,E). Immunostaining revealed that both proteins colocalized with Ptbp1 condensates within cardiomyocyte nuclei (Figure 5D). Interestingly, they exhibited distinct localization patterns: Hnrnph1 displayed a diffuse nuclear distribution with locally enhanced signals overlapping Ptbp1 condensates, whereas Matr3 showed a more discrete pattern that closely overlapped with Ptbp1 foci (Figure 5D). These results indicate that proteins identified by the BAR method are in close proximity to Ptbp1 condensates in cardiomyocyte nuclei.

To determine whether proteins enriched within Ptbp1 condensates are required for their formation and function, we next examined the roles of Hnrnph1 and Matr3. Using siRNA‐mediated knockdown, we first confirmed by qPCR that the expression levels of Ptbp1, Hnrnph1, and Matr3 were effectively reduced without significantly affecting the expression of the other genes (Figure S7A–C). In addition, immunofluorescence analysis confirmed a corresponding reduction in protein signal intensity following knockdown of each gene, consistent with the qPCR results (Figure S7D–F). Notably, Matr3 knockdown led to the loss of Ptbp1 condensates, resulting in its diffused nuclear distribution, whereas Hnrnph1 knockdown did not affect the formation or morphology of Ptbp1 condensates (Figure 5F,G). These results suggest that Ptbp1 condensates consist of functionally distinct subsets of components, including factors essential for condensate formation and others dispensable for structural assembly. To examine whether these two factors are functionally involved in proliferation, we assessed cell cycle activity. In contrast to their differential roles in condensate formation, knockdown of either Hnrnph1 or Matr3 resulted in a significant reduction in proliferative activity, as indicated by the decreased number of Ki67‐ and pH 3‐positive cardiomyocytes (Figure 5H–J). These results suggest that Ptbp1 condensates are enriched with proteins that contribute to the regulation of proliferation and that the structural maintenance of the condensates can be dissociated from their biological function.

2.6. Ptbp1 Condensates Exhibit Context‐Dependent Functions Through Association With RNAs

Based on the proteomic analysis, we found that Ptbp1 condensates contain factors involved in proliferation, condensate formation, and splicing, suggesting that RNAs in these condensates also contribute to their function. To further investigate RNA‐mediated functions of Ptbp1, we performed RNA immunoprecipitation (RIP) in purified cardiomyocyte (CM) and non‐cardiomyocyte (non‐CM) fractions (Figure 6A). Unexpectedly, despite the more diffuse nuclear distribution of Ptbp1 in non‐CMs, the number of enriched transcripts was comparable between CM and non‐CM fractions (Figure 6B). This result suggests that the condensed or diffuse state of Ptbp1 does not substantially affect the overall diversity of transcripts that are associated with Ptbp1 condensates. We next categorized the enriched transcripts into three groups: CM‐specific, non‐CM–specific, and shared between both cell types, and performed GO analysis (Figure 6C–E). Interestingly, shared transcripts were enriched for metabolic processes, while CM‐specific and non‐CM–specific transcripts showed distinct enrichment patterns: non‐CM–specific transcripts were enriched for cell adhesion and cytoskeletal organization, whereas CM‐specific transcripts were enriched for terms related to ER stress, response to hypoxia, and response to nutrient levels (Figure 6C–E). These results indicate that the landscape of Ptbp1‐associated transcripts differs substantially between CM and non‐CMs.

FIGURE 6.

Ptbp1 exhibits cell type–specific RNA interaction and functional landscapes. (A) Schematic overview of RNA immunoprecipitation sequencing (RIP‐seq). Highly purified cardiomyocyte (CM) and non‐cardiomyocyte (non‐CM) fractions were subjected to RIP‐seq using anti‐Ptbp1 antibody. (B) Identification of transcripts enriched by Ptbp1 in CM and non‐CM. Transcripts were defined as enriched when the enrichment score satisfied: (CM_log2[Ptbp1/IP over Input] − CM_log2[IgG/IP over Input]) ≥ 1. The number of enriched transcripts was 1801 in CM and 1735 in non‐CM. (C–E) Gene Ontology (GO) analysis of Ptbp1‐enriched transcripts categorized into shared (C), CM‐specific (D), and non‐CM–specific (E) groups. Top 10 terms are shown, ranked by adjusted p‐value. GO terms were summarized into simplified categories to reduce redundancy among closely related terms. (F) Landscape of alternative splicing changes following Ptbp1 knockdown. Using publicly available transcriptomic data (NTC vs. siPtbp1 in CM and non‐CM), splicing changes were defined using rMATS‐based criteria (Z. Li et al. ²⁰²⁵). The x‐axis indicates splicing event types (“All” represents the sum of all events), and the y‐axis indicates the number of unique transcripts exhibiting significant changes. (G) Schematic overview of the identification of direct Ptbp1 splicing targets. Transcripts that were both enriched in RIP‐seq and exhibited splicing changes upon Ptbp1 knockdown were defined as direct targets. (H, I) GO analysis of direct Ptbp1 target transcripts in CM (H) and non‐CM (I). Top 10 terms are shown, ranked by adjusted p‐value. GO terms were summarized into simplified categories to reduce redundancy among closely related terms. Schematic images in this figure were created with BioRender.com.

To further identify functionally relevant targets, we integrated the RIP data with previously published transcriptomic datasets (Z. Li et al. ²⁰²⁵). We first examined transcripts that showed significantly altered expression upon Ptbp1 knockdown and were enriched in CM or non‐CM by RIP analysis. Among these genes, in both CM and non‐CM, very few transcripts directly associated with Ptbp1 exhibited significant changes in gene expression (51 genes in CM and 9 genes in non‐CM, Figure S8A,B), suggesting that Ptbp1 may not primarily regulate gene expression at the transcript abundance level through direct binding. In contrast, analysis of alternative splicing by rMATS revealed that the number and pattern of splicing events affected by Ptbp1 knockdown were largely similar between CM and non‐CM (Figure 6F). However, transcripts both directly bound by Ptbp1 and those exhibiting splicing changes upon Ptbp1 knockdown showed striking differences in GO terms between cell types. In CM, these genes were enriched for metabolic and cell cycle–related processes (Figure 6H), whereas in non‐CMs, proliferation‐related terms were not enriched (Figure 6I). Despite the comparable overall diversity of transcripts directly bound to Ptbp1 in CM and non‐CM (Figure 6B), cell cycle–related processes were selectively enriched in CM, suggesting that by forming condensates, Ptbp1 may preferentially recruit proliferation‐associated splicing targets in CMs. Taken together, these findings indicate that although Ptbp1 regulates similar splicing patterns across cell types, its functional output may be determined by cell type–specific target selection and contribute to CM proliferation through splicing regulation within nuclear foci.

2.7. Ptbp1 Condensate Responds to DNA Damages by Oxidative Stress

We next investigated upstream regulators of Ptbp1 condensate formation and dissolution. Based on our findings that Ptbp1 granules exhibit phase‐separation‐like properties (Figure 3), we hypothesized that changes in the surrounding environment may influence condensate integrity. First, we induced pH changes ranging from physiological to nonphysiological conditions. However, Ptbp1 condensates in CM nuclei remained largely unaffected (Figure S9A). Next, because RIP‐seq analysis revealed enrichment of transcripts associated with metabolic processes (Figure 6C), we examined the effects of fatty acid (FA) loading to mimic postnatal metabolic stimuli. However, FA treatment did not significantly alter condensate morphology or integrity (Figure S9B). Thirdly, because RIP‐seq analysis revealed that hypoxia‐related terms were enriched among CM‐specific transcripts (Figure 6D), and as oxidative stress is known to influence postnatal CM proliferative capacity (Puente et al. 2014), we induced oxidative stress in CM. Interestingly, oxidative stress induced by H₂O₂ resulted in partial dissolution of Ptbp1 condensates, and this effect was rescued by the antioxidant N‐acetylcysteine (NAC) (Figure S10A,B). We next examined whether NAC treatment could prolong the period during which Ptbp1 granules are detectable in vivo. Although the proportion of CMs containing Ptbp1 granules tended to increase in the NAC‐treated group, the difference did not reach statistical significance (Figure S10C–E). Taken together, these findings suggest that oxidative stress may be one of the factors regulating the condensate state of Ptbp1 in CMs, although additional factors are likely required.

3. Discussion

3.1. Mechanism and Function of Ptbp1 Granules

Mammalian CMs retain proliferative capacity only within a limited neonatal window, and this capacity is rapidly lost within approximately 1 week after birth; however, the mechanisms underlying this transition remain poorly understood (Porrello et al. 2011; Walsh et al. ²⁰¹⁰). In this study, we identify a previously unrecognized aspect of Ptbp1 in CMs, demonstrating that Ptbp1 forms granules that are highly enriched in proliferation‐associated factors. Although Ptbp1‐containing granules have been reported in other systems—such as PNCs (Figure S1D) in cancer cells that suppress apoptosis and MATR3‐associated structures in skeletal muscle that regulate chromatin organization (Yap et al. 2018; Desideri et al. ²⁰²⁰)—our findings provide the first evidence for Ptbp1 granules functioning in proliferation regulation.

Previous studies on Ptbp1 function in the heart partly relied on bulk cardiac tissue, where signals from non‐CMs with high Ptbp1 expression could confound interpretation (Zhang et al. 2009; Z. Chen et al. ²⁰²³). By applying a stringent purification strategy, we were able to define a CM‐specific landscape of Ptbp1‐associated proteins and RNAs. Both BAR‐based proteomics and RIP‐seq consistently revealed enrichment of proliferation‐related factors within Ptbp1 condensates in CMs. Knockdown of these factors, such as Hnrnph1 and Matr3, resulted in reduced proliferative capacity, indicating that Ptbp1 condensates contain factors that positively regulate proliferation. At the transcript level, our RIP‐seq analysis in this highly purified system showed that, consistent with previous reports, Ptbp1 knockdown primarily affected alternative splicing rather than global gene expression (Z. Li et al. ²⁰²⁵), and this tendency was also observed in our RIP‐enriched fractions, suggesting that Ptbp1 promotes proliferation mainly through the regulation of splicing without substantially affecting transcript stability. Taken together, these findings support a model in which Ptbp1 condensates regulate proliferation through distinct mechanisms at the protein and RNA levels. For protein targets, Ptbp1 condensates may promote proliferation by modulating the function or activity of proximal proteins within condensates. In contrast, for transcript targets, Ptbp1 may selectively facilitate the generation of proliferation‐associated splicing isoforms within condensates (Iannone et al. 2023).

The effects of the knockdown of Hnrnph1 and Matr3, both of which are Ptbp1 condensate components, on condensate integrity and CM proliferation suggest that certain proteins may regulate CM proliferation by mediating Ptbp1 condensate formation, and other proteins may regulate it by different mechanisms, including regulation of alternative splicing. Further studies are required to clarify how these three factors interact with one another, either directly or indirectly. Because these proteins are RBPs, their association could be direct protein–protein interactions, RNA‐mediated interactions, or indirect multivalent interactions. Distinguishing these possibilities and determining how these interactions change upon condensate dissolution will require systematic biochemical analyses, which represents an important direction for future investigation.

Beyond interactions among RBPs, previous studies have reported interactions between Ptbp1 and long noncoding RNAs (lncRNAs), and some reports have suggested that lncRNAs such as CHARME and PNCTR contribute to condensate formation (Pandya‐Jones et al. ²⁰²⁰; Desideri et al. ²⁰²⁰; J. Chen et al. ²⁰²⁵; Yap et al. ²⁰¹⁸; Du et al. ²⁰²⁴). These observations raise the possibility that lncRNAs are also involved in Ptbp1 condensate assembly in CM nuclei. Consistent with this idea, multiple lncRNAs were enriched in our RIP‐seq analysis (data not shown). In addition, transcription and/or degradation of RNA may contribute directly to the assembly and dissolution of these nuclear foci. Because Ptbp1 is an RBP, changes in RNA transcription, processing, localization, or degradation could alter the availability of RNA scaffolds required for condensate formation, resulting in formation or dissolution of Ptbp1 foci to exert some biological effects on CMs such as cell cycle regulation. Interestingly, despite its RNA binding property, regulation of chromosome organization was among the most enriched categories in the proximity proteome (Figure 5C). Given that chromatin architecture undergoes dynamic reorganization during cell cycle progression and that Ptbp1 condensates are predominantly observed in quiescent (G0) CMs, chromatin state may influence both the formation and function of these condensates (Naumova et al. 2013; Goel et al. ²⁰²⁵) (Figures S1C and 5C).

Furthermore, given their multicomponent nature, Ptbp1 condensates may localize to specific genomic regions, potentially linking RNA processing, chromatin organization, and cell cycle regulation. Recent studies have shown that RBPs can associate with specific genomic loci and regulate transcriptional activity, and Ptbp1 itself has been reported to localize to chromatin regions via interactions with Matr3 and the lncRNA CHARME to control muscle cell maturation (Desideri et al. 2020; Xiao et al. ²⁰¹⁹; Rafiee et al. ²⁰²¹). Our findings extend this concept by suggesting that Ptbp1 condensates may integrate RNA‐ and chromatin‐based regulatory mechanisms to control CM proliferation. Taken together, our study suggests that Ptbp1 condensates may function as regulatory hubs that modulate the activity of RNA and protein factors, thereby influencing proliferative capacity.

3.2. Phase Separation Properties and Regulatory Roles

We demonstrate that Ptbp1 forms dynamic, liquid‐like condensates in CMs, suggesting CMs harbor molecular environments that support biomolecular phase separation, consistent with several previous findings (Jiang et al. 2024; Gan et al. ²⁰²⁴; Gong et al. ²⁰²³; Lei et al. ²⁰²⁵). In vitro studies have shown that Ptbp1 requires RNA for liquid droplet formation through weak multivalent protein‐RNA interactions (de Vries et al. ²⁰²⁴; Lin et al. ²⁰¹⁵; Protter et al. ²⁰¹⁸; P. Li et al. ²⁰¹²; Banani et al. ²⁰¹⁶). Consistent with this, we found that the RNA‐binding motifs (RRM2–4) are essential for granule formation in non‐CMs (Figure 3). Previous studies have demonstrated that Ptbp1 preferentially recognizes pyrimidine‐rich RNA sequences (e.g., UCUCU motifs), primarily through RRM3 and RRM4, which function as a relatively rigid unit due to their short inter‐domain linker (de Vries et al. ²⁰²⁴; Carico and Placzek ²⁰²³). These findings are consistent with our observation that deletion of RRM3–4 completely abolishes Ptbp1 foci formation, suggesting that multivalent interactions between Ptbp1 and pyrimidine‐rich RNA sequences are critical for condensate formation. However, given that in the present study, the liquid‐like properties of Ptbp1 condensates in CMs were assessed using exogenously expressed proteins, we cannot exclude the possibility that overexpression influences their formation or properties. Thus, whether endogenous Ptbp1 undergoes similar liquid‐like properties under physiological conditions remains to be determined.

Although several upstream regulatory factors of Ptbp1 have been described in cancer and neural systems (Zhu et al. 2020), the mechanisms governing its regulation, especially its phase‐separated droplet formation, remain largely unclear. Previous studies have shown that Ptbp1 localization can be influenced by metabolic conditions and mechanical stress (Tseng et al. 2025; Knoch et al. ²⁰⁰⁴). Indeed, our RIP‐seq analysis revealed enrichment of metabolism‐associated transcripts among Ptbp1‐interacting factors in both CMs and non‐CMs (Figure 6C). Importantly, oxidative stress by ROS, one of the inducers of postnatal CMs cell cycle arrest (Puente et al. 2014; Sada and Kimura ²⁰²⁴; X. Li et al. ²⁰²³), partly dissolved Ptbp1 condensates (Figure S10). Although these conditions may not fully reflect physiological postnatal environments, these findings raise the possibility that Ptbp1 condensates function as sensors of postnatal environmental stress and changes upon birth. We speculate that RBP condensates may assemble at specific genomic loci and that DNA damage or associated chromatin remodeling may disrupt their recruitment or retention at these sites. Previous studies have shown that DNA damage can alter chromatin organization and that biomolecular condensates can accumulate at defined genomic regions and participate in transcriptional regulation. Using the property of Ptbp1 as highly fluid droplets, Ptbp1 might work especially during the quite short window just after birth in response to dynamically changing environmental conditions. However, given that in the present study the liquid‐like properties of Ptbp1 condensates in CMs were assessed using exogenously expressed proteins, we cannot exclude the possibility that overexpression influences their formation or properties. Thus, whether endogenous Ptbp1 undergoes similar liquid‐like properties under physiological conditions remains to be determined.

In summary, our study identifies a previously unrecognized condensate‐like structure in CM nuclei that is specifically present during the neonatal stage and enriched with proliferation‐associated factors. Future systematic analyses will identify essential factors involved in this condensate formation and proliferative regulation. Recent studies have demonstrated the therapeutic potential of targeting phase‐separated condensates in various diseases, whereas such approaches remain largely unexplored in the heart. Elucidating the molecular mechanisms and functional roles of Ptbp1 condensates may therefore provide new strategies for the treatment of cardiovascular diseases.

4. Limitations

One of the major limitations of this study is that the direct functional significance of Ptbp1 condensate formation in adult CMs remains insufficiently addressed. Although our data demonstrate that individual condensate‐associated factors regulate proliferation and are required for condensate formation, we have not directly established whether condensate formation itself is causally linked to functional outcomes. This is partly due to technical challenges in gene delivery systems for adult CMs: transfection efficiency is low, viral vectors are constrained by cargo size, and mRNA‐based approaches are insufficient for delivering nuclear RNAs such as long noncoding RNAs or nascent transcripts. As a result, it remains difficult to simultaneously manipulate multiple components required for condensate formation. An alternative approach would be genome‐wide perturbation strategies, such as CRISPR‐based activation using dCas9 or RNA‐targeting systems such as Cas13d, may enable systematic identification and coordinated regulation of factors required for condensate formation, providing a potential avenue to reconstruct Ptbp1 condensates and assess their functional roles in adult CMs.

Another major limitation of this study is that we were unable to quantitatively assess nuclear Ptbp1 protein levels in cardiomyocytes across developmental stages, mainly because isolating cardiomyocytes from later postnatal and adult hearts is technically challenging. Although our imaging analysis showed a marked decline in detectable Ptbp1 nuclear foci in cardiomyocytes from E18.5 to P14, we cannot determine whether this decline reflects reduced nuclear Ptbp1 abundance, a developmental change in the capacity of cardiomyocytes to support granule formation, or a combination of both. Future quantitative analyses of nuclear Ptbp1 protein levels will be required to distinguish these possibilities.

A final limitation is that the highly purified cardiomyocyte culture system may influence cellular metabolism. Although cardiomyocytes were returned to standard medium for subsequent assays after purification, the nutrient‐restricted conditions used for the process of purification may have irreversibly altered their metabolic state, as reported for induced pluripotent stem cell derived cardiomyocytes (Tohyama et al. 2013). Nevertheless, Ptbp1 nuclear condensates were still observed after purification, suggesting that their formation is not solely an artifact of the purification culture system.

5. Experimental Procedures

5.1. Animals

All animal experiments were performed on CD1 (ICR) mice obtained from the Laboratory for Animal Resources and Genetic Engineering in RIKEN. Unless otherwise specified, primary CM cultures were prepared from neonatal mice at postnatal day 0–2 (P0–P2) and from adult mice aged 1–2 months.

5.2. Heart Tissue Preparation and Immunohistochemistry

Tissues were fixed overnight at 4°C in 4% paraformaldehyde (PFA) in PBS, followed by multiple washes with Phosphaste Buffered Saline (PBS). The samples were then incubated in 30% sucrose in PBS at 4°C until the tissues sank to the bottom of the tube. Subsequently, tissues were processed for cryosectioning, embedded in tissue‐freezing medium, and sectioned at a thickness of 8 μm. For immunostaining, sections were subjected to antigen retrieval using an epitope retrieval solution with a steamer (IHC‐Tek Epitope Retrieval Steamer Set) for 30 min, followed by washing with PBS. Sections were then blocked and permeabilized with 10% normal donkey serum (NDS) in 0.3% Triton X‐100/PBS for 20 min and incubated with primary antibodies diluted in 3% NDS and 0.3% Tween‐20/PBS overnight at 4°C in a humidified chamber. The following day, sections were washed with PBS and incubated with the appropriate secondary antibodies (1:400) diluted in PBS containing 1.5% NDS for 1 h at room temperature in a humidified chamber protected from light. After washing, sections were stained with DAPI (1:5000 in PBS) and mounted with Fluoro‐KEEPER, followed by coverslipping. Details of reagents and primary antibodies are provided in a separate table (Tables 1 and 2).

TABLE 1.

Reagents_list.

Cell_culture
	Product_name	Company	Catalog_number
Primary_culture_for_neonate	DMEM/F12	gibco	11,320–033
	Insulin solution from bovine pancrease	Sigma‐Aldrich	I0516
	Sodium Pyruvate	gibco	11,360–070
	MEM NEAA	gibco	11,140–050
	Pen Strep	gibco	15,070–063
	L‐glutamine	gibco	25,030–081
	Fetal Bovine Serum	gibco	A5256701
	Horse Serum	gibco	16,050,122
	L‐Lactic acid	Sigma	L6402‐1G
	DMEM (no glutamine、no glucose、no pyruvate)	nacalai	19,362–45
	BSA,10 g	Sigma	A3059‐10G
	Fibronectin coat solution	Cosmo‐bio	SFN01
Cardiomyocyte_isolation_and_purification	Red Blood Cell Lysis Solution	Miltenyi Biotec	130–094‐183
	MACS BSA Stock Solution	Miltenyi Biotec	130–091‐376
	autoMACS Rinsing Solution	Miltenyi Biotec	130–091‐222
	Neonatal Heart Dissociation Kit, mouse and rat	Miltenyi Biotec	130–098‐373
	Neonatal Cardiomyocyte Isolation Kit, mouse	Miltenyi Biotec	130–100‐825
	gentle MACS Octro DIssociator with Hearters	Miltenyi Biotec	130–093‐235
	MiniMACS starting Kit	Miltenyi Biotec	130–090‐312
Primary_culture_for_Adult	Heparin, Sodium (injection, USP)	Wako	087–00133
	CD (Chemically defined) lipid concentrate	Thermo	11,905–031
	Protease from Streptomyces griseus typeXIV	Sigma	P5147‐100MG
	collagenase 2	Worthington	LS004176
	collagenase 4	Worthington	LS004188
	ITS(100×)	Sigma	I3146‐5ML
	1 mol/L‐HEPES	Nacalai	17,557–94
	(−)‐Blebbistatin 10 mM (1 mL in DMSO)	Selleck	S7099
	Laminin (Natural, Mouse), 1 mg	Invitrogen	23,017,015
Transfection	Opti‐MEM	Invitrogen	31,985–062
	Lipofectamine 3000 Transfection Kit	Thermo	L3000008
	Lipofectamine RNAiMAX Transfection Reagent, 0.3 mL	Invitrogen	13,778,030
	Lipofectamine MessengerMAX Transfection Reagent 1.5 ml	Invitrogen	LMRNA015

Immunostaining
Product_name	Company	Catalog_number
4% PFA	Nacalai	09154–85
Donkey serum	Merck‐Millipore	S30‐100ML
Tween‐20	Sigma	P1379‐500 mL
TRITON X‐100	MP Biomedicals	807,426
WGA (Wheat Germ Agglutinin) Alexa Fluor 647 Conjugated	Thermo Fisher	W32466
DAPI	nacalai	11,034–56
IHC‐Tek Epitope Retrieval Solution	IHCWORLD	IW‐1100‐1 L
TFM	GeneralData	TFM‐5
Tissue‐Tek O.C.T. Compound	Sakura finetek	45,833
Fluoro‐KEEPER	Nacalai	1,259,364

BAR
Product_name	Company	Catalog_number
Donkey Anti‐Goat IgG H&L (HRP polymer)	Abcam	ab214881
Avidin/Biotin Blocking Kit	Abcam	ab64212
Tyramide Conjugates	Invitrogen	B40951
Pierce Streptavidin Magnetic Beads(1 ml)	Thermo Fisher	88,817
DTT, Molecular Grade (DL‐Dithiothreitol)	Promega	P1171
Trypsin Platinum, Mass Spectrometry Grade	Promega	VA9000
Streptavidin, Horseradish Peroxidase	PGI	SA00001‐0
Pierce Streptavidin Magnetic Beads	Thermo Fisher	88,817
Sodium Ascorbate	Sigma	A4034‐100G
Biotinyl tyramide	Invitrogen	B40951
Sodium Deoxycholate, 10 g	Wako	190–08313

RIP
Product_name	Company	Catalog_number
Magna RIP	Merck	17–700
NEBNext Ultra II Directional RNA Library Prep Kit for Illumina	NEB	E7760S
NEBNext Multiplex Oligos for Illumina (96 UDI Primer Pairs)	NEB	E6440S
KAPA SYBR FAST qPCR Master Mix	Roche	KK4603

Cloning_IVT
Product_name	Company	Catalog_number
TRIzol Reagent, 100 mL	Thermo Fisher	15,596,026
Direct‐zol RNA Micro Kit without TRI‐reagent, 50 preps	ZYR	R2050
mMESSAGE mMACHINE T7 ULTRA Transcription Kit	Invitrogen	AM1345
PrimeSTAR Max DNA Polymerase(400回分)	Takara	R045B
In‐Fusion cloning kit	Takara	639,649
Gene art site‐directed mutagenesis PLUS kit	invitrogen	A14604

qPCR
Product_name	Company	Catalog_number
TB Green Premix Ex Taq II (Tli RNaseH Plus)	Takara	RR820A
PrimeScript RT Master Mix	Takara	RR036A

Cell_treatment
Product_name	Company	Catalog_number
Click‐iT EdU Imaging Kit	invitrogen	C10338
Nigericin(10 mg)	Abcam	ab120494
BSA Control for BSA‐Fatty Acid Complexes (5 mM)	Cayman Support Science	29,556
Palmitate‐BSA	Cayman Support Science	29,558
Hydrogen Peroxide(35%)	Nacalai	20,779–65
NAC	Sigma	A7250‐5G

Note: All reagents used in this study are listed.

TABLE 2.

Ptbp1_project_Antibody_list.

target_name_in_paper	Antibody_name	company	Number	Dilution for IHC	Dilution for ICC
Actinin	sarcomeric alpha actinin EP2529Y	Abcam	ab68167‐100	N/A	1/1000
cTnT	Cardiac Troponin T Monoclonal Antibody (13–11)	Invitrogen	MA5‐12960	1/800	1/1500
Ptbp1	IHC‐plus Polyclonal Goat anti‐Human PTBP1 Antibody (aa2‐14, IHC, WB) LS‐B3235	LS‐Bio	LS‐B3235‐50	1/800	1/1500
Ptbp1 pS16	rabbit serum	kindly provided by Dr. Douglas Black lab.	N/A	1/2000	N/A
Ptbp1 S16	rabbit serum	kindly provided by Dr. Douglas Black lab.	N/A	1/2000	N/A
Matr3	Rabbit monoclonal anti‐MATR3	Abcam	Cat# ab151714	N/A	1/1000
Hnrnph1	hnRNP‐H Polyclonal Antibody 10ul	Bethyl Laboratories	IHC‐00087‐T	N/A	1/1000
GFP	chick polyclonal	ThermoFisher	A10262	N/A	1/500
Flag	mouse monoclonal	Sigma	F3165	N/A	1/500
Ki67	Ki‐67 Monoclonal Antibody (SolA15)	Invitrogen	14–5698‐82	1/1000	1/1500
pH 3	Anti‐phospho‐Histone H3 (Ser10) Antibody, Mitosis Marker	Merck	06–570	N/A	1/1500
8‐OHG	Anti DNA/RNA Damage	abcam	ab62623	N/A	1/500
FITC	Streptavidin, (FITC)	Invitrogen	SA1001	N/A	1/400
Fibrillarin	rabbit monoclonal	CST	#2639	N/A	1/500
Secondary antibody	Antibody_name
	alexa fluor488 donkey anti‐goat IgG (H + L)	invitrogen	A11055	1/400	1/1500
	alexa fluor488 donkey anti‐Rat IgG (H + L)	invitrogen	A21208	1/400	1/1500
	alexa fluor488 donkey anti‐mouse IgG (H + L)	invitrogen	A21202	N/A	1/1500
	alexa fluor488 donkey anti‐rabbit IgG (H + L)	invitrogen	A21206	N/A	1/1500
	alexa fluor488 donkey anti‐Chick IgY	invitrogen	A78948	N/A	1/1500
	alexa fluor555 donkey anti‐goat IgG (H + L)	invitrogen	A21432	1/400	N/A
	alexa fluor555 donkey anti‐mouse IgG (H + L)	invitrogen	A31570	1/400	1/1500
	alexa fluor555 donkey anti‐rabbit IgG (H + L)	invitrogen	A31572	1/400	1/1500
	alexa fluor647 donkey anti‐goat IgG (H + L)	invitrogen	A21447	N/A	1/1500
	alexa fluor647 donkey anti‐mouse IgG (H + L)	invitrogen	A31571	1/400	1/1500
	alexa fluor647 donkey anti‐Rabbit IgG (H + L)	invitrogen	A32795	N/A	1/1500
	WGA (Wheat Germ Agglutinin) Alexa Fluor 647 Conjugated	Thermo Fisher	W32466	0.05 mg/mL	N/A
RIP
Normal goat IgG	Normal Goat IgG	MBL	PM094	5 μg/μl for RIP

Note: All antibodies used in this study are listed.

5.3. Isolation and Culture of Primary Neonatal CMs

Primary CMs were isolated from neonatal mice using a method adapted from our previously published protocol (Sakaguchi et al. 2025) for use with the Neonatal Heart Dissociation Kit according to the manufacturer's protocol with minor modifications. Detailed procedures are described in the manufacturer's instructions.

Prior to heart isolation (approximately 2 h), culture vessels were coated with fibronectin and incubated at 37°C in a humidified incubator with 5% CO₂. Neonatal ICR mice were euthanized, and hearts were rapidly excised and immediately transferred into PBS on ice. Atria and major vessels were carefully removed, and the remaining ventricular tissue was subjected to enzymatic dissociation using a tissue dissociator with a heater for approximately 1 h under the recommended settings. Following dissociation, the cell suspension was passed through a 70 μm mesh strainer to remove undigested tissue fragments. The filtrate was centrifuged at 600 × g for 5 min, and the cell pellet was collected. Red blood cells were lysed using RBC lysis solution, after which the procedure was continued according to the manufacturer's protocol.

For CM enrichment, a pre‐plating step was performed on a sterile dish for 1 h 10 min to reduce non‐CM populations. The attached non‐CMs were collected and cultured as the non‐CM population. Subsequently, CMs were further purified using magnetic‐activated cell sorting according to the isolation kit protocol. Purified CMs were seeded onto fibronectin‐coated 96‐well plates at a density of 3–5 × 10⁴ cells per well in 100 μL of culture medium containing 25% serum. The culture medium was first replaced 4 h after plating. Thereafter, starting from the following morning, the medium was replaced once every 24 h. At each medium change, cells were gently washed with pre‐warmed PBS prior to the addition of fresh medium.

Details of reagents and kits are provided in a separate table (Tables 1 and 3).

TABLE 3.

Neonate_primary_culture_reagent_list.

Culture_medium_component_50ml_scale	Company	Catalog_number	Final_conc
FBS	gibco	A5256701	20%
Horse serum	gibco	16,050,122	5%
Sodium Pyruvate	gibco	11,360–070	3 mM
NEAA	gibco	11,140–050	0.1 mM
Insulin solution from bovine pancrease	Sigma‐Aldrich	I0516	1 μg/mL
L‐glutamine	gibco	25,030–081	2 mM
Pen Strep	gibco	15,070–063	100 mg/mL
DMEM/F12	gibco	11,320–033	to 50 ml
Nutrient‐free medium
Lactate	Sigma	L6402‐1G	4 mM
BSA	Sigma	A3059‐10G	0.10%
Pen Strep	gibco	15,070–063	100 mg/ml
DMEM (No glucose, No pyruvate, No glutamine)			to 50 ml

Note: All reagents used for neonatal primary culture in this study are listed.

5.4. Stringent Purification System for BAR and RIP

A stringent protocol for CM purification was established based on a previously reported method (Tohyama et al. 2013). After one round of pre‐plating followed by MACS purification, cells were seeded onto 6‐well plates at a high density (5 × 10⁶ cells per well) and cultured. Four hours after seeding, the medium was replaced with nutrient‐free medium (composition provided in a separate table) to suppress the expansion of non‐CMs, and cells were cultured for 24 h. The medium was then refreshed with fresh nutrient‐free medium, and cells were further cultured for approximately 20 h. Subsequently, the medium was switched to standard culture medium containing 25% serum. On the following day, the medium was replaced again with fresh culture medium containing 25% serum, and cells were cultured for an additional 10 h before collection. Details of reagents and kits are provided in a separate table (Table 1).

5.5. Langendorff‐Free Isolation of Adult CMs

Adult CMs were isolated using a modified Langendorff‐free protocol based on a previous method (Ackers‐Johnson et al. ²⁰¹⁶). Details of reagents and buffer components, including any further modification, are provided in a separate table (Table 1).

Culture plates were pre‐coated with laminin (5 μg/mL in PBS) for at least 1 h at 37°C prior to use.

5.5.1. Heparinization

Mice were intraperitoneally injected with heparin (500 μL of 10,000 U/10 mL solution) using a 27G needle 20 min prior to heart excision.

5.5.2. Heart Excision

All buffers, including collagenase buffer, stop solution, and calcium reintroduction buffer, were pre‐warmed to 37°C. While the buffers were being warmed, mice were euthanized by cervical dislocation, and the thoracic cavity was rapidly opened to expose the heart.

Approximately 7 mL of EDTA buffer was immediately injected into the right ventricular (RV) region. Due to the thinner wall of the RV, this allowed efficient perfusion, resulting in rapid cessation of cardiac contraction and clearance of blood from the heart. Following this initial EDTA perfusion, the aorta was grasped, and surrounding tissues were trimmed while maintaining the hold on the aorta.

5.5.3. Heart Dissociation

To remove blood and induce relaxation, EDTA buffer (~10 mL total) was gently perfused through the left ventricular apex at a rate of approximately 1 mL per 2 min, avoiding excessive pressure to prevent atrial rupture. The heart was then transferred to perfusion buffer, and ~3 mL was gently perfused to remove residual EDTA. Subsequently, the heart was transferred to collagenase buffer and enzymatic digestion was initiated by repeated gentle perfusion and recirculation of the enzyme solution for approximately 10 min. Digestion was monitored visually based on tissue softening and color change.

After digestion, the ventricles were isolated, minced into small fragments, and gently triturated using a transfer pipette for 2 min. The suspension was transferred to a dish containing 5 mL of warm stop solution and further gently triturated for 2 min. The resulting cell suspension was filtered through a 100‐μm cell strainer into a 50 mL tube to remove undigested tissue. Cells were allowed to settle by gravity for 20 min at room temperature, after which the supernatant was carefully removed.

5.5.4. Calcium Reintroduction

Calcium was gradually reintroduced by stepwise addition of calcium‐containing buffer in three increments, with 10 min incubation at each step. After each step, cells were allowed to settle by gravity and the supernatant was removed. Following calcium adaptation, cells were resuspended in culture medium at the desired density.

5.5.5. Culture Conditions

Cells were plated onto laminin‐coated plates. After 1–3 h, the medium was replaced to remove unattached cells. Medium was replaced again the following morning, and subsequently every 24 h. Gentle washing was performed after the first day of culture.

5.6. Immunocytochemistry

Cultured cells were fixed for 15 min in 4% PFA in PBS at room temperature, followed by five washes with PBS. Samples were then blocked and permeabilized with 10% NDS in 0.3% Triton X‐100/PBS for 20 min and incubated with primary antibodies diluted in 3% NDS and 0.3% Tween‐20/PBS overnight at 4°C in a humidified chamber. The following day, samples were washed with PBS and incubated with the appropriate secondary antibodies (1:1500) diluted in PBS containing 1.5% NDS for 1 h at room temperature, protected from light. After washing, samples were stained with DAPI (1:5000 in PBS) and mounted with Fluoro‐KEEPER or PBS‐based mounting medium. Details of primary antibodies and reagents are provided in a separate table (Tables 1 and 2).

5.7. EdU Incorporation Assay Combined With Immunocytochemistry

Adult CMs were treated with 10 μM EdU in culture medium at 16 h and 40 h after transfection. Cells were fixed with 4% PFA 24 h after each EdU treatment. Following fixation, cells were processed for immunocytochemistry as described above, including washing, blocking, and permeabilization steps. Primary and secondary antibody incubations were performed sequentially for 1.5 h each at room temperature, followed by nuclear staining with DAPI (1:5000 in PBS). EdU detection was then performed using the Click‐iT EdU Imaging Kit according to the manufacturer's instructions, and signals were observed using an APX100 microscope. Non‐CMs present within the same wells were used as internal positive controls. Details of reagents and kits are provided in a separate table (Table 1).

5.8. Plasmid Construction

Plasmid DNA constructs were obtained from Origene. Fragments of interest were amplified by PCR using gene‐specific primers and PrimeSTAR Max DNA Polymerase. PCR products were separated by agarose gel electrophoresis, excised, and purified using a DNA purification kit. Multiple PCR fragments were then assembled using the In‐Fusion cloning system to generate the desired constructs. For certain constructs, sub‐cloning into a vector containing a T7 promoter was performed to enable in vitro transcription–mediated mRNA synthesis. Details of reagents and plasmid sequences are provided in a separate table (Table 4).

TABLE 4.

Construct_list.

Origene
construct_name	Origene_number
ORIGENE MR208803 PTBP1(NM_001077363)MOUSE TAGGED ORF CLONE	64,168,350

sub_cloning
construct	sequence(CDS)
hPtbp1‐TurboGFP	ATGGACGGCATTGTCCCAGATATAGCCGTTGGTACAAAGCGGGGATCTGACGAGCTTTTCTCTACTTGTGTCACTAACGGACCGTTTATCATGAGCAGCAACTCGGCTTCTGCAGCAAACGGAAATGACAGCAAGAAGTTCAAAGGTGACAGCCGAAGTGCAGGCGTCCCCTCTAGAGTGATCCACATCCGGAAGCTCCCCATCGACGTCACGGAGGGGGAAGTCATCTCCCTGGGGCTGCCCTTTGGGAAGGTCACCAACCTCCTGATGCTGAAGGGGAAAAACCAGGCCTTCATCGAGATGAACACGGAGGAGGCTGCCAACACCATGGTGAACTACTACACCTCGGTGACCCCTGTGCTGCGCGGCCAGCCCATCTACATCCAGTTCTCCAACCACAAGGAGCTGAAGACCGACAGCTCTCCCAACCAGGCGCGGGCCCAGGCGGCCCTGCAGGCGGTGAACTCGGTCCAGTCGGGGAACCTGGCCTTGGCTGCCTCGGCGGCGGCCGTGGACGCAGGGATGGCGATGGCCGGGCAGAGCCCCGTGCTCAGGATCATCGTGGAGAACCTCTTCTACCCTGTGACCCTGGATGTGCTGCACCAGATTTTCTCCAAGTTCGGCACAGTGTTGAAGATCATCACCTTCACCAAGAACAACCAGTTCCAGGCCCTGCTGCAGTATGCGGACCCCGTGAGCGCCCAGCACGCCAAGCTGTCGCTGGACGGGCAGAACATCTACAACGCCTGCTGCACGCTGCGCATCGACTTTTCCAAGCTCACCAGCCTCAACGTCAAGTACAACAATGACAAGAGCCGTGACTACACACGCCCAGACCTGCCTTCCGGGGACAGCCAGCCCTCGCTGGACCAGACCATGGCCGCGGCCTTCGGTGCACCTGGTATAATCTCAGCCTCTCCGTATGCAGGAGCTGGTTTCCCTCCCACCTTTGCCATTCCTCAAGCTGCAGGCCTTTCCGTTCCGAACGTCCACGGCGCCCTGGCCCCCCTGGCCATCCCCTCGGCGGCGGCGGCAGCTGCGGCGGCAGGTCGGATCGCCATCCCGGGCCTGGCGGGGGCAGGAAATTCTGTATTGCTGGTCAGCAACCTCAACCCAGAGAGAGTCACACCCCAAAGCCTCTTTATTCTTTTCGGCGTCTACGGTGACGTGCAGCGCGTGAAGATCCTGTTCAATAAGAAGGAGAACGCCCTAGTGCAGATGGCGGACGGCAACCAGGCCCAGCTGGCCATGAGCCACCTGAACGGGCACAAGCTGCACGGGAAGCCCATCCGCATCACGCTCTCGAAGCACCAGAACGTGCAGCTGCCCCGCGAGGGCCAGGAGGACCAGGGCCTGACCAAGGACTACGGCAACTCACCCCTGCACCGCTTCAAGAAGCCGGGCTCCAAGAACTTCCAGAACATATTCCCGCCCTCGGCCACGCTGCACCTCTCCAACATCCCGCCCTCAGTCTCCGAGGAGGATCTCAAGGTCCTGTTTTCCAGCAATGGGGGCGTCGTCAAAGGATTCAAGTTCTTCCAGAAGGACCGCAAGATGGCACTGATCCAGATGGGCTCCGTGGAGGAGGCGGTCCAGGCCCTCATTGACCTGCACAACCACGACCTCGGGGAGAACCACCACCTGCGGGTCTCCTTCTCCAAGTCCACCATCACGCGTACGCGGCCGCTCGAGATGGAGAGCGACGAGAGCGGCCTGCCCGCCATGGAGATCGAGTGCCGCATCACCGGCACCCTGAACGGCGTGGAGTTCGAGCTGGTGGGCGGCGGAGAGGGCACCCCCGAGCAGGGCCGCATGACCAACAAGATGAAGAGCACCAAAGGCGCCCTGACCTTCAGCCCCTACCTGCTGAGCCACGTGATGGGCTACGGCTTCTACCACTTCGGCACCTACCCCAGCGGCTACGAGAACCCCTTCCTGCACGCCATCAACAACGGCGGCTACACCAACACCCGCATCGAGAAGTACGAGGACGGCGGCGTGCTGCACGTGAGCTTCAGCTACCGCTACGAGGCCGGCCGCGTGATCGGCGACTTCAAGGTGATGGGCACCGGCTTCCCCGAGGACAGCGTGATCTTCACCGACAAGATCATCCGCAGCAACGCCACCGTGGAGCACCTGCACCCCATGGGCGATAACGATCTGGATGGCAGCTTCACCCGCACCTTCAGCCTGCGCGACGGCGGCTACTACAGCTCCGTGGTGGACAGCCACATGCACTTCAAGAGCGCCATCCACCCCAGCATCCTGCAGAACGGGGGCCCCATGTTCGCCTTCCGCCGCGTGGAGGAGGATCACAGCAACACCGAGCTGGGCATCGTGGAGTACCAGCACGCCTTCAAGACCCCGGATGCAGATGCCGGTGAAGAAAGAGTTTAAACGGCCGGCCGCGGTCATAGCTGTTTCCTGA
S16AhPtbp1‐TurboGFP	ATGGACGGCATTGTCCCAGATATAGCCGTTGGTACAAAGCGGGGATCTGACGAGCTTTTCTCTACTTGTGTCACTAACGGACCGTTTATCATGAGCAGCAACTCGGCTTCTGCAGCAAACGGAAATGACAGCAAGAAGTTCAAAGGTGACAGCCGAAGTGCAGGCGTCCCCTCTAGAGTGATCCACATCCGGAAGCTCCCCATCGACGTCACGGAGGGGGAAGTCATCTCCCTGGGGCTGCCCTTTGGGAAGGTCACCAACCTCCTGATGCTGAAGGGGAAAAACCAGGCCTTCATCGAGATGAACACGGAGGAGGCTGCCAACACCATGGTGAACTACTACACCTCGGTGACCCCTGTGCTGCGCGGCCAGCCCATCTACATCCAGTTCTCCAACCACAAGGAGCTGAAGACCGACAGCTCTCCCAACCAGGCGCGGGCCCAGGCGGCCCTGCAGGCGGTGAACTCGGTCCAGTCGGGGAACCTGGCCTTGGCTGCCTCGGCGGCGGCCGTGGACGCAGGGATGGCGATGGCCGGGCAGAGCCCCGTGCTCAGGATCATCGTGGAGAACCTCTTCTACCCTGTGACCCTGGATGTGCTGCACCAGATTTTCTCCAAGTTCGGCACAGTGTTGAAGATCATCACCTTCACCAAGAACAACCAGTTCCAGGCCCTGCTGCAGTATGCGGACCCCGTGAGCGCCCAGCACGCCAAGCTGTCGCTGGACGGGCAGAACATCTACAACGCCTGCTGCACGCTGCGCATCGACTTTTCCAAGCTCACCAGCCTCAACGTCAAGTACAACAATGACAAGAGCCGTGACTACACACGCCCAGACCTGCCTTCCGGGGACAGCCAGCCCTCGCTGGACCAGACCATGGCCGCGGCCTTCGGTGCACCTGGTATAATCTCAGCCTCTCCGTATGCAGGAGCTGGTTTCCCTCCCACCTTTGCCATTCCTCAAGCTGCAGGCCTTTCCGTTCCGAACGTCCACGGCGCCCTGGCCCCCCTGGCCATCCCCTCGGCGGCGGCGGCAGCTGCGGCGGCAGGTCGGATCGCCATCCCGGGCCTGGCGGGGGCAGGAAATTCTGTATTGCTGGTCAGCAACCTCAACCCAGAGAGAGTCACACCCCAAAGCCTCTTTATTCTTTTCGGCGTCTACGGTGACGTGCAGCGCGTGAAGATCCTGTTCAATAAGAAGGAGAACGCCCTAGTGCAGATGGCGGACGGCAACCAGGCCCAGCTGGCCATGAGCCACCTGAACGGGCACAAGCTGCACGGGAAGCCCATCCGCATCACGCTCTCGAAGCACCAGAACGTGCAGCTGCCCCGCGAGGGCCAGGAGGACCAGGGCCTGACCAAGGACTACGGCAACTCACCCCTGCACCGCTTCAAGAAGCCGGGCTCCAAGAACTTCCAGAACATATTCCCGCCCTCGGCCACGCTGCACCTCTCCAACATCCCGCCCTCAGTCTCCGAGGAGGATCTCAAGGTCCTGTTTTCCAGCAATGGGGGCGTCGTCAAAGGATTCAAGTTCTTCCAGAAGGACCGCAAGATGGCACTGATCCAGATGGGCTCCGTGGAGGAGGCGGTCCAGGCCCTCATTGACCTGCACAACCACGACCTCGGGGAGAACCACCACCTGCGGGTCTCCTTCTCCAAGTCCACCATCACGCGTACGCGGCCGCTCGAGATGGAGAGCGACGAGAGCGGCCTGCCCGCCATGGAGATCGAGTGCCGCATCACCGGCACCCTGAACGGCGTGGAGTTCGAGCTGGTGGGCGGCGGAGAGGGCACCCCCGAGCAGGGCCGCATGACCAACAAGATGAAGAGCACCAAAGGCGCCCTGACCTTCAGCCCCTACCTGCTGAGCCACGTGATGGGCTACGGCTTCTACCACTTCGGCACCTACCCCAGCGGCTACGAGAACCCCTTCCTGCACGCCATCAACAACGGCGGCTACACCAACACCCGCATCGAGAAGTACGAGGACGGCGGCGTGCTGCACGTGAGCTTCAGCTACCGCTACGAGGCCGGCCGCGTGATCGGCGACTTCAAGGTGATGGGCACCGGCTTCCCCGAGGACAGCGTGATCTTCACCGACAAGATCATCCGCAGCAACGCCACCGTGGAGCACCTGCACCCCATGGGCGATAACGATCTGGATGGCAGCTTCACCCGCACCTTCAGCCTGCGCGACGGCGGCTACTACAGCTCCGTGGTGGACAGCCACATGCACTTCAAGAGCGCCATCCACCCCAGCATCCTGCAGAACGGGGGCCCCATGTTCGCCTTCCGCCGCGTGGAGGAGGATCACAGCAACACCGAGCTGGGCATCGTGGAGTACCAGCACGCCTTCAAGACCCCGGATGCAGATGCCGGTGAAGAAAGAGTTTAAACGGCCGGCCGCGGTCATAGCTGTTTCCTGA
mPtbp1_iso1_Flag(Origene_MR208803)	ATGGACGGCATCGTCCCAGACATAGCAGTCGGTACAAAGCGGGGATCCGACGAGCTCTTCTCCACGTGTGTCAGCAACGGCCCCTTCATCATGAGCAGCTCTGCCTCAGCAGCCAATGGAAACGATAGCAAGAAGTTCAAAGGTGACAACAGGAGCGCAGGAGTCCCTTCCAGAGTCATCCATGTCAGAAAGCTGCCCAGCGATGTCACTGAGGGCGAGGTCATCTCCCTAGGGCTGCCCTTTGGAAAGGTTACCAACCTTCTCATGCTGAAGGGGAAGAACCAGGCCTTCATTGAGATGAACACAGAGGAGGCTGCCAACACTATGGTTAACTACTATACATCGGTGGCGCCAGTGCTTCGTGGACAGCCCATCTACATCCAGTTCTCCAACCACAAAGAGCTCAAGACCGACAGCTCGCCCAACCAGGCACGTGCCCAGGCAGCCCTGCAGGCTGTAAACTCCGTCCAGTCTGGAAACCTGGCCTTGGCAGCGTCCGCTGCTGCCGTGGATGCAGGAATGGCAATGGCAGGGCAGAGCCCAGTGCTCAGGATCATTGTGGAAAACCTTTTCTACCCAGTGACCCTGGACGTGCTGCACCAGATCTTCTCTAAGTTTGGCACCGTCCTGAAGATCATCACGTTCACCAAGAACAACCAGTTCCAGGCGCTGCTGCAGTATGCTGACCCTGTGAGCGCCCAGCATGCCAAGCTGTCCCTGGATGGCCAGAACATCTACAACGCCTGCTGCACGCTGCGCATCGACTTCTCCAAGCTCACCAGTCTCAATGTCAAGTACAACAATGATAAGAGCAGAGACTACACTCGACCTGACCTGCCCTCTGGAGACAGCCAGCCTTCACTAGACCAGACCATGGCAGCAGCCTTTGGTGCGCCCGGCATAATGTCAGCCTCTCCGTATGCAGGAGCCGGGTTCCCTCCCACCTTTGCCATCCCTCAGGCCGCAGGCCTCTCTGTCCCTAATGTCCATGGAGCCTTGGCCCCCCTGGCCATCCCGTCTGCTGCTGCTGCTGCTGCGGCCAGCCGCATTGCCATCCCAGGGTTGGCAGGTGCTGGGAATTCTGTCCTTTTGGTCAGCAATCTGAACCCTGAGAGAGTCACACCCCAAAGCCTCTTTATTCTCTTCGGCGTCTACGGTGATGTGCAGCGGGTGAAGATCCTGTTCAATAAGAAGGAGAACGCACTTGTGCAGATGGCAGACGGCAGCCAGGCCCAGCTGGCCATGAGCCACCTGAACGGGCACAAGCTGCACGGGAAGTCAGTGCGCATTACACTGTCCAAGCATCAGAGTGTGCAGCTGCCTCGGGAGGGTCAGGAGGACCAGGGCCTGACCAAGGACTATGGCAGCTCCCCGCTGCACCGCTTCAAGAAACCAGGCTCCAAGAACTTCCAGAACATCTTTCCACCCTCAGCTACCCTGCACCTCTCCAACATCCCGCCCTCTGTGTCAGAGGACGACCTCAAGAGCCTCTTCTCCAGCAACGGTGGTGTGGTCAAAGGCTTCAAGTTCTTCCAGAAGGACCGCAAGATGGCACTGATCCAGATGGGCTCTGTGGAGGAGGCTGTGCAGGCGCTGATTGAACTGCACAACCATGACCTGGGCGAGAACCACCACCTGCGAGTGTCCTTTTCCAAGTCCACCATCACGCGTACGCGGCCGCTCGAGCAGAAACTCATCTCAGAAGAGGATCTGGCAGCAAATGATATCCTGGATTACAAGGATGACGACGATAAGGTTTAA
mPtbp1_iso2_Flag	atggacggcatcgtcccagacatagcagtcggtacaaagcggggatccgacgagctcttctccacgtgtgtcagcaacggccccttcatcatgagcagctctgcctcagcagccaatggaaacgatagcaagaagttcaaaggtgacaacaggagcgcaggagtcccttccagagtcatccatgtcagaaagctgcccagcgatgtcactgagggcgaggtcatctccctagggctgccctttggaaaggttaccaaccttctcatgctgaaggggaagaaccaggccttcattgagatgaacacagaggaggctgccaacactatggttaactactatacatcggtggcgccagtgcttcgtggacagcccatctacatccagttctccaaccacaaagagctcaagaccgacagctcgcccaaccaggcacgtgcccaggcagccctgcaggctgtaaactccgtccagtctggaaacctggccttggcagcgtccgctgctgccgtggatgcaggaatggcaatggcagggcagagcccagtgctcaggatcattgtggaaaaccttttctacccagtgaccctggacgtgctgcaccagatcttctctaagtttggcaccgtcctgaagatcatcacgttcaccaagaacaaccagttccaggcgctgctgcagtatgctgaccctgtgagcgcccagcatgccaagctgtccctggatggccagaacatctacaacgcctgctgcacgctgcgcatcgacttctccaagctcaccagtctcaatgtcaagtacaacaatgataagagcagagactacactcgacctgacctgccctctggagacagccagccttcactagaccagaccatggcagcagcctttggcctctctgtccctaatgtccatggagccttggcccccctggccatcccgtctgctgctgctgctgctgcggccagccgcattgccatcccagggttggcaggtgctgggaattctgtccttttggtcagcaatctgaaccctgagagagtcacaccccaaagcctctttattctcttcggcgtctacggtgatgtgcagcgggtgaagatcctgttcaataagaaggagaacgcacttgtgcagatggcagacggcagccaggcccagctggccatgagccacctgaacgggcacaagctgcacgggaagtcagtgcgcattacactgtccaagcatcagagtgtgcagctgcctcgggagggtcaggaggaccagggcctgaccaaggactatggcagctccccgctgcaccgcttcaagaaaccaggctccaagaacttccagaacatctttccaccctcagctaccctgcacctctccaacatcccgccctctgtgtcagaggacgacctcaagagcctcttctccagcaacggtggtgtggtcaaaggcttcaagttcttccagaaggaccgcaagatggcactgatccagatgggctctgtggaggaggctgtgcaggcgctgattgaactgcacaaccatgacctgggcgagaaccaccacctgcgagtgtccttttccaagtccaccatcacgcgtacgcggccgctcgagcagaaactcatctcagaagaggatctggcagcaaatgatatcctggattacaaggatgacgacgataaggtttaa
mPtbp1_iso3_Flag	ATGGACGGCATCGTCCCAGACATAGCAGTCGGTACAAAGCGGGGATCCGACGAGCTCTTCTCCACGTGTGTCAGCAACGGCCCCTTCATCATGAGCAGCTCTGCCTCAGCAGGGCTGCCCTTTGGAAAGGTTACCAACCTTCTCATGCTGAAGGGGAAGAACCAGGCCTTCATTGAGATGAACACAGAGGAGGCTGCCAACACTATGGTTAACTACTATACATCGGTGGCGCCAGTGCTTCGTGGACAGCCCATCTACATCCAGTTCTCCAACCACAAAGAGCTCAAGACCGACAGCTCGCCCAACCAGGCACGTGCCCAGGCAGCCCTGCAGGCTGTAAACTCCGTCCAGTCTGGAAACCTGGCCTTGGCAGCGTCCGCTGCTGCCGTGGATGCAGGAATGGCAATGGCAGGGCAGAGCCCAGTGCTCAGGATCATTGTGGAAAACCTTTTCTACCCAGTGACCCTGGACGTGCTGCACCAGATCTTCTCTAAGTTTGGCACCGTCCTGAAGATCATCACGTTCACCAAGAACAACCAGTTCCAGGCGCTGCTGCAGTATGCTGACCCTGTGAGCGCCCAGCATGCCAAGCTGTCCCTGGATGGCCAGAACATCTACAACGCCTGCTGCACGCTGCGCATCGACTTCTCCAAGCTCACCAGTCTCAATGTCAAGTACAACAATGATAAGAGCAGAGACTACACTCGACCTGACCTGCCCTCTGGAGACAGCCAGCCTTCACTAGACCAGACCATGGCAGCAGCCTTTGGTCTCTCTGTCCCTAATGTCCATGGAGCCTTGGCCCCCCTGGCCATCCCGTCTGCTGCTGCTGCTGCTGCGGCCAGCCGCATTGCCATCCCAGGGTTGGCAGGTGCTGGGAATTCTGTCCTTTTGGTCAGCAATCTGAACCCTGAGAGAGTCACACCCCAAAGCCTCTTTATTCTCTTCGGCGTCTACGGTGATGTGCAGCGGGTGAAGATCCTGTTCAATAAGAAGGAGAACGCACTTGTGCAGATGGCAGACGGCAGCCAGGCCCAGCTGGCCATGAGCCACCTGAACGGGCACAAGCTGCACGGGAAGTCAGTGCGCATTACACTGTCCAAGCATCAGAGTGTGCAGCTGCCTCGGGAGGGTCAGGAGGACCAGGGCCTGACCAAGGACTATGGCAGCTCCCCGCTGCACCGCTTCAAGAAACCAGGCTCCAAGAACTTCCAGAACATCTTTCCACCCTCAGCTACCCTGCACCTCTCCAACATCCCGCCCTCTGTGTCAGAGGACGACCTCAAGAGCCTCTTCTCCAGCAACGGTGGTGTGGTCAAAGGCTTCAAGTTCTTCCAGAAGGACCGCAAGATGGCACTGATCCAGATGGGCTCTGTGGAGGAGGCTGTGCAGGCGCTGATTGAACTGCACAACCATGACCTGGGCGAGAACCACCACCTGCGAGTGTCCTTTTCCAAGTCCACCATCACGCGTACGCGGCCGCTCGAGCAGAAACTCATCTCAGAAGAGGATCTGGCAGCAAATGATATCCTGGATTACAAGGATGACGACGATAAGGTTTAA
mPtbp1_iso4_Flag	atgagcagctctgcctcagcagccaatggaaacgatagcaagaagttcaaaggtgacaacaggagcgcaggagtcccttccagagtcatccatgtcagaaagctgcccagcgatgtcactgagggcgaggtcatctccctagggctgccctttggaaaggttaccaaccttctcatgctgaaggggaagaaccaggccttcattgagatgaacacagaggaggctgccaacactatggttaactactatacatcggtggcgccagtgcttcgtggacagcccatctacatccagttctccaaccacaaagagctcaagaccgacagctcgcccaaccaggcacgtgcccaggcagccctgcaggctgtaaactccgtccagtctggaaacctggccttggcagcgtccgctgctgccgtggatgcaggaatggcaatggcagggcagagcccagtgctcaggatcattgtggaaaaccttttctacccagtgaccctggacgtgctgcaccagatcttctctaagtttggcaccgtcctgaagatcatcacgttcaccaagaacaaccagttccaggcgctgctgcagtatgctgaccctgtgagcgcccagcatgccaagctgtccctggatggccagaacatctacaacgcctgctgcacgctgcgcatcgacttctccaagctcaccagtctcaatgtcaagtacaacaatgataagagcagagactacactcgacctgacctgccctctggagacagccagccttcactagaccagaccatggcagcagcctttggtgcgcccggcataatgtcagcctctccgtatgcaggagccgggttccctcccacctttgccatccctcaggccgcaggcctctctgtccctaatgtccatggagccttggcccccctggccatcccgtctgctgctgctgctgctgcggccagccgcattgccatcccagggttggcaggtgctgggaattctgtccttttggtcagcaatctgaaccctgagagagtcacaccccaaagcctctttattctcttcggcgtctacggtgatgtgcagcgggtgaagatcctgttcaataagaaggagaacgcacttgtgcagatggcagacggcagccaggcccagctggccatgagccacctgaacgggcacaagctgcacgggaagtcagtgcgcattacactgtccaagcatcagagtgtgcagctgcctcgggagggtcaggaggaccagggcctgaccaaggactatggcagctccccgctgcaccgcttcaagaaaccaggctccaagaacttccagaacatctttccaccctcagctaccctgcacctctccaacatcccgccctctgtgtcagaggacgacctcaagagcctcttctccagcaacggtggtgtggtcaaaggcttcaagttcttccagaaggaccgcaagatggcactgatccagatgggctctgtggaggaggctgtgcaggcgctgattgaactgcacaaccatgacctgggcgagaaccaccacctgcgagtgtccttttccaagtccaccatcacgcgtacgcggccgctcgagcagaaactcatctcagaagaggatctggcagcaaatgatatcctggattacaaggatgacgacgataaggtttaa
mPtbp1_isor6_Flag	atgagcagctctgcctcagcagccaatggaaacgatagcaagaagttcaaaggtgacaacaggagcgcaggagtcccttccagagtcatccatgtcagaaagctgcccagcgatgtcactgagggcgaggtcatctccctagggctgccctttggaaaggttaccaaccttctcatgctgaaggggaagaaccaggccttcattgagatgaacacagaggaggctgccaacactatggttaactactatacatcggtggcgccagtgcttcgtggacagcccatctacatccagttctccaaccacaaagagctcaagaccgacagctcgcccaaccaggcacgtgcccaggcagccctgcaggctgtaaactccgtccagtctggaaacctggccttggcagcgtccgctgctgccgtggatgcaggaatggcaatggcagggcagagcccagtgctcaggatcattgtggaaaaccttttctacccagtgaccctggacgtgctgcaccagatcttctctaagtttggcaccgtcctgaagatcatcacgttcaccaagaacaaccagttccaggcgctgctgcagtatgctgaccctgtgagcgcccagcatgccaagctgtccctggatggccagaacatctacaacgcctgctgcacgctgcgcatcgacttctccaagctcaccagtctcaatgtcaagtacaacaatgataagagcagagactacactcgacctgacctgccctctggagacagccagccttcactagaccagaccatggcagcagcctttggtctctctgtccctaatgtccatggagccttggcccccctggccatcccgtctgctgctgctgctgctgcggccagccgcattgccatcccagggttggcaggtgctgggaattctgtccttttggtcagcaatctgaaccctgagagagtcacaccccaaagcctctttattctcttcggcgtctacggtgatgtgcagcgggtgaagatcctgttcaataagaaggagaacgcacttgtgcagatggcagacggcagccaggcccagctggccatgagccacctgaacgggcacaagctgcacgggaagtcagtgcgcattacactgtccaagcatcagagtgtgcagctgcctcgggagggtcaggaggaccagggcctgaccaaggactatggcagctccccgctgcaccgcttcaagaaaccaggctccaagaacttccagaacatctttccaccctcagctaccctgcacctctccaacatcccgccctctgtgtcagaggacgacctcaagagcctcttctccagcaacggtggtgtggtcaaaggcttcaagttcttccagaaggaccgcaagatggcactgatccagatgggctctgtggaggaggctgtgcaggcgctgattgaactgcacaaccatgacctgggcgagaaccaccacctgcgagtgtccttttccaagtccaccatcacgcgtacgcggccgctcgagcagaaactcatctcagaagaggatctggcagcaaatgatatcctggattacaaggatgacgacgataaggtttaa
mPtbp1_iso1_Flag‐NLS‐GFP‐Apex2	atggacggcatcgtcccagacatagcagtcggtacaaagcggggatccgacgagctcttctccacgtgtgtcagcaacggccccttcatcatgagcagctctgcctcagcagccaatggaaacgatagcaagaagttcaaaggtgacaacaggagcgcaggagtcccttccagagtcatccatgtcagaaagctgcccagcgatgtcactgagggcgaggtcatctccctagggctgccctttggaaaggttaccaaccttctcatgctgaaggggaagaaccaggccttcattgagatgaacacagaggaggctgccaacactatggttaactactatacatcggtggcgccagtgcttcgtggacagcccatctacatccagttctccaaccacaaagagctcaagaccgacagctcgcccaaccaggcacgtgcccaggcagccctgcaggctgtaaactccgtccagtctggaaacctggccttggcagcgtccgctgctgccgtggatgcaggaatggcaatggcagggcagagcccagtgctcaggatcattgtggaaaaccttttctacccagtgaccctggacgtgctgcaccagatcttctctaagtttggcaccgtcctgaagatcatcacgttcaccaagaacaaccagttccaggcgctgctgcagtatgctgaccctgtgagcgcccagcatgccaagctgtccctggatggccagaacatctacaacgcctgctgcacgctgcgcatcgacttctccaagctcaccagtctcaatgtcaagtacaacaatgataagagcagagactacactcgacctgacctgccctctggagacagccagccttcactagaccagaccatggcagcagcctttggtgcgcccggcataatgtcagcctctccgtatgcaggagccgggttccctcccacctttgccatccctcaggccgcaggcctctctgtccctaatgtccatggagccttggcccccctggccatcccgtctgctgctgctgctgctgcggccagccgcattgccatcccagggttggcaggtgctgggaattctgtccttttggtcagcaatctgaaccctgagagagtcacaccccaaagcctctttattctcttcggcgtctacggtgatgtgcagcgggtgaagatcctgttcaataagaaggagaacgcacttgtgcagatggcagacggcagccaggcccagctggccatgagccacctgaacgggcacaagctgcacgggaagtcagtgcgcattacactgtccaagcatcagagtgtgcagctgcctcgggagggtcaggaggaccagggcctgaccaaggactatggcagctccccgctgcaccgcttcaagaaaccaggctccaagaacttccagaacatctttccaccctcagctaccctgcacctctccaacatcccgccctctgtgtcagaggacgacctcaagagcctcttctccagcaacggtggtgtggtcaaaggcttcaagttcttccagaaggaccgcaagatggcactgatccagatgggctctgtggaggaggctgtgcaggcgctgattgaactgcacaaccatgacctgggcgagaaccaccacctgcgagtgtccttttccaagtccaccatcacgcgtacgcggccgctcgaggtgagcaagggcgaggagctgttcaccggggtggtgcccatcctggtcgagctggacggcgacgtaaacggccacaagttcagcgtgtccggcgagggcgagggcgatgccacctacggcaagctgaccctgaagttcatctgcaccaccggcaagctgcccgtgccctggcccaccctcgtgaccaccctgacctacggcgtgcagtgcttcagccgctaccccgaccacatgaagcagcacgacttcttcaagtccgccatgcccgaaggctacgtccaggagcgcaccatcttcttcaaggacgacggcaactacaagacccgcgccgaggtgaagttcgagggcgacaccctggtgaaccgcatcgagctgaagggcatcgacttcaaggaggacggcaacatcctggggcacaagctggagtacaactacaacagccacaacgtctatatcatggccgacaagcagaagaacggcatcaaggtgaacttcaagatccgccacaacatcgaggacggcagcgtgcagctcgccgaccactaccagcagaacacccccatcggcgacggccccgtgctgctgcccgacaaccactacctgagcacccagtccgccctgagcaaagaccccaacgagaagcgcgatcacatggtcctgctggagttcgtgaccgccgccgggatcactctcggcatggacgagctgtacaagggaggagagcagaaactcatctccgaagaggacctcggaaagtcttacccaactgtgagtgctgattaccaggacgccgttgagaaggcgaagaagaagctcagaggcttcatcgctgagaagagatgcgctcctctaatgctccgtttggcattccactctgctggaacctttgacaagggcacgaagaccggtggacccttcggaaccatcaagcaccctgccgaactggctcacagcgctaacaacggtcttgacatcgctgttaggcttttggagccactcaaggcggagttccctattttgagctacgccgatttctaccagttggctggcgttgttgccgttgaggtcacgggtggacctaaggttccattccaccctggaagagaggacaagcctgagccaccaccagagggtcgcttgcctgatcccactaagggttctgaccatttgagagatgtgtttggcaaagctatggggcttactgaccaagatatcgttgctctatctgggggtcacactattggagctgcacacaaggagcgttctggatttgagggtccctggacctctaatcctcttattttcgacaactcatacttcacggagttgttgagtggtgagaaggaaggtctccttcagctaccttctgacaaggctcttttgtctgaccctgtattccgccctctcgttgataaatatgcagcggacgaagatgccttctttgctgattacgctgaggctcaccaaaagctttccgagcttgggtttgctgatgcctaa
mPtbp1_iso1_Flag‐NLS	atggacggcatcgtcccagacatagcagtcggtacaaagcggggatccgacgagctcttctccacgtgtgtcagcaacggccccttcatcatgagcagctctgcctcagcagccaatggaaacgatagcaagaagttcaaaggtgacaacaggagcgcaggagtcccttccagagtcatccatgtcagaaagctgcccagcgatgtcactgagggcgaggtcatctccctagggctgccctttggaaaggttaccaaccttctcatgctgaaggggaagaaccaggccttcattgagatgaacacagaggaggctgccaacactatggttaactactatacatcggtggcgccagtgcttcgtggacagcccatctacatccagttctccaaccacaaagagctcaagaccgacagctcgcccaaccaggcacgtgcccaggcagccctgcaggctgtaaactccgtccagtctggaaacctggccttggcagcgtccgctgctgccgtggatgcaggaatggcaatggcagggcagagcccagtgctcaggatcattgtggaaaaccttttctacccagtgaccctggacgtgctgcaccagatcttctctaagtttggcaccgtcctgaagatcatcacgttcaccaagaacaaccagttccaggcgctgctgcagtatgctgaccctgtgagcgcccagcatgccaagctgtccctggatggccagaacatctacaacgcctgctgcacgctgcgcatcgacttctccaagctcaccagtctcaatgtcaagtacaacaatgataagagcagagactacactcgacctgacctgccctctggagacagccagccttcactagaccagaccatggcagcagcctttggtgcgcccggcataatgtcagcctctccgtatgcaggagccgggttccctcccacctttgccatccctcaggccgcaggcctctctgtccctaatgtccatggagccttggcccccctggccatcccgtctgctgctgctgctgctgcggccagccgcattgccatcccagggttggcaggtgctgggaattctgtccttttggtcagcaatctgaaccctgagagagtcacaccccaaagcctctttattctcttcggcgtctacggtgatgtgcagcgggtgaagatcctgttcaataagaaggagaacgcacttgtgcagatggcagacggcagccaggcccagctggccatgagccacctgaacgggcacaagctgcacgggaagtcagtgcgcattacactgtccaagcatcagagtgtgcagctgcctcgggagggtcaggaggaccagggcctgaccaaggactatggcagctccccgctgcaccgcttcaagaaaccaggctccaagaacttccagaacatctttccaccctcagctaccctgcacctctccaacatcccgccctctgtgtcagaggacgacctcaagagcctcttctccagcaacggtggtgtggtcaaaggcttcaagttcttccagaaggaccgcaagatggcactgatccagatgggctctgtggaggaggctgtgcaggcgctgattgaactgcacaaccatgacctgggcgagaaccaccacctgcgagtgtccttttccaagtccaccatcacgcgtacgcggccgctcgagcagaaactcatctcagaagaggatctggcagcaaatgatatcctggattacaaggatgacgacgataaggttccaaaaaagaagagaaaggtataa
S16AmPtbp1_iso1_Flag‐NLS	atggacggcatcgtcccagacatagcagtcggtacaaagcggggagccgacgagctcttctccacgtgtgtcagcaacggccccttcatcatgagcagctctgcctcagcagccaatggaaacgatagcaagaagttcaaaggtgacaacaggagcgcaggagtcccttccagagtcatccatgtcagaaagctgcccagcgatgtcactgagggcgaggtcatctccctagggctgccctttggaaaggttaccaaccttctcatgctgaaggggaagaaccaggccttcattgagatgaacacagaggaggctgccaacactatggttaactactatacatcggtggcgccagtgcttcgtggacagcccatctacatccagttctccaaccacaaagagctcaagaccgacagctcgcccaaccaggcacgtgcccaggcagccctgcaggctgtaaactccgtccagtctggaaacctggccttggcagcgtccgctgctgccgtggatgcaggaatggcaatggcagggcagagcccagtgctcaggatcattgtggaaaaccttttctacccagtgaccctggacgtgctgcaccagatcttctctaagtttggcaccgtcctgaagatcatcacgttcaccaagaacaaccagttccaggcgctgctgcagtatgctgaccctgtgagcgcccagcatgccaagctgtccctggatggccagaacatctacaacgcctgctgcacgctgcgcatcgacttctccaagctcaccagtctcaatgtcaagtacaacaatgataagagcagagactacactcgacctgacctgccctctggagacagccagccttcactagaccagaccatggcagcagcctttggtgcgcccggcataatgtcagcctctccgtatgcaggagccgggttccctcccacctttgccatccctcaggccgcaggcctctctgtccctaatgtccatggagccttggcccccctggccatcccgtctgctgctgctgctgctgcggccagccgcattgccatcccagggttggcaggtgctgggaattctgtccttttggtcagcaatctgaaccctgagagagtcacaccccaaagcctctttattctcttcggcgtctacggtgatgtgcagcgggtgaagatcctgttcaataagaaggagaacgcacttgtgcagatggcagacggcagccaggcccagctggccatgagccacctgaacgggcacaagctgcacgggaagtcagtgcgcattacactgtccaagcatcagagtgtgcagctgcctcgggagggtcaggaggaccagggcctgaccaaggactatggcagctccccgctgcaccgcttcaagaaaccaggctccaagaacttccagaacatctttccaccctcagctaccctgcacctctccaacatcccgccctctgtgtcagaggacgacctcaagagcctcttctccagcaacggtggtgtggtcaaaggcttcaagttcttccagaaggaccgcaagatggcactgatccagatgggctctgtggaggaggctgtgcaggcgctgattgaactgcacaaccatgacctgggcgagaaccaccacctgcgagtgtccttttccaagtccaccatcacgcgtacgcggccgctcgagcagaaactcatctcagaagaggatctggcagcaaatgatatcctggattacaaggatgacgacgataaggttccaaaaaagaagagaaaggtataa

mRNA
mRNA_name	sequence(T7promotor‐end of CDS)
T7_GFP‐S16APtbp1var1‐Flag‐NLS	tcagaattttgtaatacgactcactatagggcggccgggaattcgtcgactggatccggtaccgaggagatctgccgccgcgatcggccaccatggtgagcaagggcgaggagctgttcaccggggtggtgcccatcctggtcgagctggacggcgacgtaaacggccacaagttcagcgtgtccggcgagggcgagggcgatgccacctacggcaagctgaccctgaagttcatctgcaccaccggcaagctgcccgtgccctggcccaccctcgtgaccaccctgacctacggcgtgcagtgcttcagccgctaccccgaccacatgaagcagcacgacttcttcaagtccgccatgcccgaaggctacgtccaggagcgcaccatcttcttcaaggacgacggcaactacaagacccgcgccgaggtgaagttcgagggcgacaccctggtgaaccgcatcgagctgaagggcatcgacttcaaggaggacggcaacatcctggggcacaagctggagtacaactacaacagccacaacgtctatatcatggccgacaagcagaagaacggcatcaaggtgaacttcaagatccgccacaacatcgaggacggcagcgtgcagctcgccgaccactaccagcagaacacccccatcggcgacggccccgtgctgctgcccgacaaccactacctgagcacccagtccgccctgagcaaagaccccaacgagaagcgcgatcacatggtcctgctggagttcgtgaccgccgccgggatcactctcggcatggacgagctgtacaagatggacggcatcgtcccagacatagcagtcggtacaaagcggggagccgacgagctcttctccacgtgtgtcagcaacggccccttcatcatgagcagctctgcctcagcagccaatggaaacgatagcaagaagttcaaaggtgacaacaggagcgcaggagtcccttccagagtcatccatgtcagaaagctgcccagcgatgtcactgagggcgaggtcatctccctagggctgccctttggaaaggttaccaaccttctcatgctgaaggggaagaaccaggccttcattgagatgaacacagaggaggctgccaacactatggttaactactatacatcggtggcgccagtgcttcgtggacagcccatctacatccagttctccaaccacaaagagctcaagaccgacagctcgcccaaccaggcacgtgcccaggcagccctgcaggctgtaaactccgtccagtctggaaacctggccttggcagcgtccgctgctgccgtggatgcaggaatggcaatggcagggcagagcccagtgctcaggatcattgtggaaaaccttttctacccagtgaccctggacgtgctgcaccagatcttctctaagtttggcaccgtcctgaagatcatcacgttcaccaagaacaaccagttccaggcgctgctgcagtatgctgaccctgtgagcgcccagcatgccaagctgtccctggatggccagaacatctacaacgcctgctgcacgctgcgcatcgacttctccaagctcaccagtctcaatgtcaagtacaacaatgataagagcagagactacactcgacctgacctgccctctggagacagccagccttcactagaccagaccatggcagcagcctttggtgcgcccggcataatgtcagcctctccgtatgcaggagccgggttccctcccacctttgccatccctcaggccgcaggcctctctgtccctaatgtccatggagccttggcccccctggccatcccgtctgctgctgctgctgctgcggccagccgcattgccatcccagggttggcaggtgctgggaattctgtccttttggtcagcaatctgaaccctgagagagtcacaccccaaagcctctttattctcttcggcgtctacggtgatgtgcagcgggtgaagatcctgttcaataagaaggagaacgcacttgtgcagatggcagacggcagccaggcccagctggccatgagccacctgaacgggcacaagctgcacgggaagtcagtgcgcattacactgtccaagcatcagagtgtgcagctgcctcgggagggtcaggaggaccagggcctgaccaaggactatggcagctccccgctgcaccgcttcaagaaaccaggctccaagaacttccagaacatctttccaccctcagctaccctgcacctctccaacatcccgccctctgtgtcagaggacgacctcaagagcctcttctccagcaacggtggtgtggtcaaaggcttcaagttcttccagaaggaccgcaagatggcactgatccagatgggctctgtggaggaggctgtgcaggcgctgattgaactgcacaaccatgacctgggcgagaaccaccacctgcgagtgtccttttccaagtccaccatcacgcgtacgcggccgctcgagcagaaactcatctcagaagaggatctggcagcaaatgatatcctggattacaaggatgacgacgataaggttccaaaaaagaagagaaaggtataa
T7_GFP‐S16APtbp1var1‐Flag‐NLSΔRRM1	tcagaattttgtaatacgactcactatagggcggccgggaattcgtcgactggatccggtaccgaggagatctgccgccgcgatcggccaccatggtgagcaagggcgaggagctgttcaccggggtggtgcccatcctggtcgagctggacggcgacgtaaacggccacaagttcagcgtgtccggcgagggcgagggcgatgccacctacggcaagctgaccctgaagttcatctgcaccaccggcaagctgcccgtgccctggcccaccctcgtgaccaccctgacctacggcgtgcagtgcttcagccgctaccccgaccacatgaagcagcacgacttcttcaagtccgccatgcccgaaggctacgtccaggagcgcaccatcttcttcaaggacgacggcaactacaagacccgcgccgaggtgaagttcgagggcgacaccctggtgaaccgcatcgagctgaagggcatcgacttcaaggaggacggcaacatcctggggcacaagctggagtacaactacaacagccacaacgtctatatcatggccgacaagcagaagaacggcatcaaggtgaacttcaagatccgccacaacatcgaggacggcagcgtgcagctcgccgaccactaccagcagaacacccccatcggcgacggccccgtgctgctgcccgacaaccactacctgagcacccagtccgccctgagcaaagaccccaacgagaagcgcgatcacatggtcctgctggagttcgtgaccgccgccgggatcactctcggcatggacgagctgtacaagatggacggcatcgtcccagacatagcagtcggtacaaagcggggagccgacgagctcttctccacgtgtgtcagcaacggccccttcatcatgagcagctctgcctcagcagccaatggaaacgatagcaagaagttcaaaggtgacaacaggagcgcaggagtcccttccagagtcatccatgtcagaaagctgcccagcgatgtcactgagggcgaggtcatctccctagggctgccctttggaaaggttaccaaccttctcatgctgaaggggaagaaccaggccttcattgagatgaacacagaggaggctgccaacactatggttaactactatacatcggtggcgccagtgcttcgtggacagcccatctacatccagttctccaaccacaaagagctcaagaccgacagctcgcccaaccaggcacgtgcccaggcagccctgcaggctgtaaactccgtccagtctggaaacctggccttggcagcgtccgctgctgccgtggatgcaggaatggcaatggcagggcagagcccagtgctcaggatcattgtggaaaaccttttctacccagtgaccctggacgtgctgcaccagatcttctctaagtttggcaccgtcctgaagatcatcacgttcaccaagaacaaccagttccaggcgctgctgcagtatgctgaccctgtgagcgcccagcatgccaagctgtccctggatggccagaacatctacaacgcctgctgcacgctgcgcatcgacttctccaagctcaccagtctcaatgtcaagtacaacaatgataagagcagagactacactcgacctgacctgccctctggagacagccagccttcactagaccagaccatggcagcagcctttggtgcgcccggcataatgtcagcctctccgtatgcaggagccgggttccctcccacctttgccatccctcaggccgcaggcctctctgtccctaatgtccatggagccttggcccccctggccatcccgtctgctgctgctgctgctgcggccagccgcattgccatcccagggttggcaggtgctgggaattctgtccttttggtcagcaatctgaaccctgagagagtcacaccccaaagcctctttattctcttcggcgtctacggtgatgtgcagcgggtgaagatcctgttcaataagaaggagaacgcacttgtgcagatggcagacggcagccaggcccagctggccatgagccacctgaacgggcacaagctgcacgggaagtcagtgcgcattacactgtccaagcatcagagtgtgcagctgcctcgggagggtcaggaggaccagggcctgaccaaggactatggcagctccccgctgcaccgcttcaagaaaccaggctccaagaacttccagaacatctttccaccctcagctaccctgcacctctccaacatcccgccctctgtgtcagaggacgacctcaagagcctcttctccagcaacggtggtgtggtcaaaggcttcaagttcttccagaaggaccgcaagatggcactgatccagatgggctctgtggaggaggctgtgcaggcgctgattgaactgcacaaccatgacctgggcgagaaccaccacctgcgagtgtccttttccaagtccaccatcacgcgtacgcggccgctcgagcagaaactcatctcagaagaggatctggcagcaaatgatatcctggattacaaggatgacgacgataaggttccaaaaaagaagagaaaggtataa
T7_GFP‐S16APtbp1var1‐Flag‐NLSΔRRM2	tcagaattttgtaatacgactcactatagggcggccgggaattcgtcgactggatccggtaccgaggagatctgccgccgcgatcggccaccatggtgagcaagggcgaggagctgttcaccggggtggtgcccatcctggtcgagctggacggcgacgtaaacggccacaagttcagcgtgtccggcgagggcgagggcgatgccacctacggcaagctgaccctgaagttcatctgcaccaccggcaagctgcccgtgccctggcccaccctcgtgaccaccctgacctacggcgtgcagtgcttcagccgctaccccgaccacatgaagcagcacgacttcttcaagtccgccatgcccgaaggctacgtccaggagcgcaccatcttcttcaaggacgacggcaactacaagacccgcgccgaggtgaagttcgagggcgacaccctggtgaaccgcatcgagctgaagggcatcgacttcaaggaggacggcaacatcctggggcacaagctggagtacaactacaacagccacaacgtctatatcatggccgacaagcagaagaacggcatcaaggtgaacttcaagatccgccacaacatcgaggacggcagcgtgcagctcgccgaccactaccagcagaacacccccatcggcgacggccccgtgctgctgcccgacaaccactacctgagcacccagtccgccctgagcaaagaccccaacgagaagcgcgatcacatggtcctgctggagttcgtgaccgccgccgggatcactctcggcatggacgagctgtacaagatggacggcatcgtcccagacatagcagtcggtacaaagcggggagccgacgagctcttctccacgtgtgtcagcaacggccccttcatcatgagcagctctgcctcagcagccaatggaaacgatagcaagaagttcaaaggtgacaacaggagcgcaggagtcccttccagagtcatccatgtcagaaagctgcccagcgatgtcactgagggcgaggtcatctccctagggctgccctttggaaaggttaccaaccttctcatgctgaaggggaagaaccaggccttcattgagatgaacacagaggaggctgccaacactatggttaactactatacatcggtggcgccagtgcttcgtggacagcccatctacatccagttctccaaccacaaagagctcaagaccgacagctcgcccaaccaggcacgtgcccaggcagccctgcaggctgtaaactccgtccagtctggaaacctggccttggcagcgtccgctgctgccgtggatgcaggaatggcaatggcagggcagagcccagtgctcaggatcattgtggaaaaccttttctacccagtgaccctggacgtgctgcaccagatcttctctaagtttggcaccgtcctgaagatcatcacgttcaccaagaacaaccagttccaggcgctgctgcagtatgctgaccctgtgagcgcccagcatgccaagctgtccctggatggccagaacatctacaacgcctgctgcacgctgcgcatcgacttctccaagctcaccagtctcaatgtcaagtacaacaatgataagagcagagactacactcgacctgacctgccctctggagacagccagccttcactagaccagaccatggcagcagcctttggtgcgcccggcataatgtcagcctctccgtatgcaggagccgggttccctcccacctttgccatccctcaggccgcaggcctctctgtccctaatgtccatggagccttggcccccctggccatcccgtctgctgctgctgctgctgcggccagccgcattgccatcccagggttggcaggtgctgggaattctgtccttttggtcagcaatctgaaccctgagagagtcacaccccaaagcctctttattctcttcggcgtctacggtgatgtgcagcgggtgaagatcctgttcaataagaaggagaacgcacttgtgcagatggcagacggcagccaggcccagctggccatgagccacctgaacgggcacaagctgcacgggaagtcagtgcgcattacactgtccaagcatcagagtgtgcagctgcctcgggagggtcaggaggaccagggcctgaccaaggactatggcagctccccgctgcaccgcttcaagaaaccaggctccaagaacttccagaacatctttccaccctcagctaccctgcacctctccaacatcccgccctctgtgtcagaggacgacctcaagagcctcttctccagcaacggtggtgtggtcaaaggcttcaagttcttccagaaggaccgcaagatggcactgatccagatgggctctgtggaggaggctgtgcaggcgctgattgaactgcacaaccatgacctgggcgagaaccaccacctgcgagtgtccttttccaagtccaccatcacgcgtacgcggccgctcgagcagaaactcatctcagaagaggatctggcagcaaatgatatcctggattacaaggatgacgacgataaggttccaaaaaagaagagaaaggtataa
T7_GFP‐S16APtbp1var1‐Flag‐NLSΔRRM3	tcagaattttgtaatacgactcactatagggcggccgggaattcgtcgactggatccggtaccgaggagatctgccgccgcgatcggccaccatggtgagcaagggcgaggagctgttcaccggggtggtgcccatcctggtcgagctggacggcgacgtaaacggccacaagttcagcgtgtccggcgagggcgagggcgatgccacctacggcaagctgaccctgaagttcatctgcaccaccggcaagctgcccgtgccctggcccaccctcgtgaccaccctgacctacggcgtgcagtgcttcagccgctaccccgaccacatgaagcagcacgacttcttcaagtccgccatgcccgaaggctacgtccaggagcgcaccatcttcttcaaggacgacggcaactacaagacccgcgccgaggtgaagttcgagggcgacaccctggtgaaccgcatcgagctgaagggcatcgacttcaaggaggacggcaacatcctggggcacaagctggagtacaactacaacagccacaacgtctatatcatggccgacaagcagaagaacggcatcaaggtgaacttcaagatccgccacaacatcgaggacggcagcgtgcagctcgccgaccactaccagcagaacacccccatcggcgacggccccgtgctgctgcccgacaaccactacctgagcacccagtccgccctgagcaaagaccccaacgagaagcgcgatcacatggtcctgctggagttcgtgaccgccgccgggatcactctcggcatggacgagctgtacaagatggacggcatcgtcccagacatagcagtcggtacaaagcggggagccgacgagctcttctccacgtgtgtcagcaacggccccttcatcatgagcagctctgcctcagcagccaatggaaacgatagcaagaagttcaaaggtgacaacaggagcgcaggagtcccttccagagtcatccatgtcagaaagctgcccagcgatgtcactgagggcgaggtcatctccctagggctgccctttggaaaggttaccaaccttctcatgctgaaggggaagaaccaggccttcattgagatgaacacagaggaggctgccaacactatggttaactactatacatcggtggcgccagtgcttcgtggacagcccatctacatccagttctccaaccacaaagagctcaagaccgacagctcgcccaaccaggcacgtgcccaggcagccctgcaggctgtaaactccgtccagtctggaaacctggccttggcagcgtccgctgctgccgtggatgcaggaatggcaatggcagggcagagcccagtgctcaggatcattgtggaaaaccttttctacccagtgaccctggacgtgctgcaccagatcttctctaagtttggcaccgtcctgaagatcatcacgttcaccaagaacaaccagttccaggcgctgctgcagtatgctgaccctgtgagcgcccagcatgccaagctgtccctggatggccagaacatctacaacgcctgctgcacgctgcgcatcgacttctccaagctcaccagtctcaatgtcaagtacaacaatgataagagcagagactacactcgacctgacctgccctctggagacagccagccttcactagaccagaccatggcagcagcctttggtgcgcccggcataatgtcagcctctccgtatgcaggagccgggttccctcccacctttgccatccctcaggccgcaggcctctctgtccctaatgtccatggagccttggcccccctggccatcccgtctgctgctgctgctgctgcggccagccgcattgccatcccagggttggcaggtgctgggaattctgtccttttggtcagcaatctgaaccctgagagagtcacaccccaaagcctctttattctcttcggcgtctacggtgatgtgcagcgggtgaagatcctgttcaataagaaggagaacgcacttgtgcagatggcagacggcagccaggcccagctggccatgagccacctgaacgggcacaagctgcacgggaagtcagtgcgcattacactgtccaagcatcagagtgtgcagctgcctcgggagggtcaggaggaccagggcctgaccaaggactatggcagctccccgctgcaccgcttcaagaaaccaggctccaagaacttccagaacatctttccaccctcagctaccctgcacctctccaacatcccgccctctgtgtcagaggacgacctcaagagcctcttctccagcaacggtggtgtggtcaaaggcttcaagttcttccagaaggaccgcaagatggcactgatccagatgggctctgtggaggaggctgtgcaggcgctgattgaactgcacaaccatgacctgggcgagaaccaccacctgcgagtgtccttttccaagtccaccatcacgcgtacgcggccgctcgagcagaaactcatctcagaagaggatctggcagcaaatgatatcctggattacaaggatgacgacgataaggttccaaaaaagaagagaaaggtataa
T7_GFP‐S16APtbp1var1‐Flag‐NLSΔRRM4	tcagaattttgtaatacgactcactatagggcggccgggaattcgtcgactggatccggtaccgaggagatctgccgccgcgatcggccaccatggtgagcaagggcgaggagctgttcaccggggtggtgcccatcctggtcgagctggacggcgacgtaaacggccacaagttcagcgtgtccggcgagggcgagggcgatgccacctacggcaagctgaccctgaagttcatctgcaccaccggcaagctgcccgtgccctggcccaccctcgtgaccaccctgacctacggcgtgcagtgcttcagccgctaccccgaccacatgaagcagcacgacttcttcaagtccgccatgcccgaaggctacgtccaggagcgcaccatcttcttcaaggacgacggcaactacaagacccgcgccgaggtgaagttcgagggcgacaccctggtgaaccgcatcgagctgaagggcatcgacttcaaggaggacggcaacatcctggggcacaagctggagtacaactacaacagccacaacgtctatatcatggccgacaagcagaagaacggcatcaaggtgaacttcaagatccgccacaacatcgaggacggcagcgtgcagctcgccgaccactaccagcagaacacccccatcggcgacggccccgtgctgctgcccgacaaccactacctgagcacccagtccgccctgagcaaagaccccaacgagaagcgcgatcacatggtcctgctggagttcgtgaccgccgccgggatcactctcggcatggacgagctgtacaagatggacggcatcgtcccagacatagcagtcggtacaaagcggggagccgacgagctcttctccacgtgtgtcagcaacggccccttcatcatgagcagctctgcctcagcagccaatggaaacgatagcaagaagttcaaaggtgacaacaggagcgcaggagtcccttccagagtcatccatgtcagaaagctgcccagcgatgtcactgagggcgaggtcatctccctagggctgccctttggaaaggttaccaaccttctcatgctgaaggggaagaaccaggccttcattgagatgaacacagaggaggctgccaacactatggttaactactatacatcggtggcgccagtgcttcgtggacagcccatctacatccagttctccaaccacaaagagctcaagaccgacagctcgcccaaccaggcacgtgcccaggcagccctgcaggctgtaaactccgtccagtctggaaacctggccttggcagcgtccgctgctgccgtggatgcaggaatggcaatggcagggcagagcccagtgctcaggatcattgtggaaaaccttttctacccagtgaccctggacgtgctgcaccagatcttctctaagtttggcaccgtcctgaagatcatcacgttcaccaagaacaaccagttccaggcgctgctgcagtatgctgaccctgtgagcgcccagcatgccaagctgtccctggatggccagaacatctacaacgcctgctgcacgctgcgcatcgacttctccaagctcaccagtctcaatgtcaagtacaacaatgataagagcagagactacactcgacctgacctgccctctggagacagccagccttcactagaccagaccatggcagcagcctttggtgcgcccggcataatgtcagcctctccgtatgcaggagccgggttccctcccacctttgccatccctcaggccgcaggcctctctgtccctaatgtccatggagccttggcccccctggccatcccgtctgctgctgctgctgctgcggccagccgcattgccatcccagggttggcaggtgctgggaattctgtccttttggtcagcaatctgaaccctgagagagtcacaccccaaagcctctttattctcttcggcgtctacggtgatgtgcagcgggtgaagatcctgttcaataagaaggagaacgcacttgtgcagatggcagacggcagccaggcccagctggccatgagccacctgaacgggcacaagctgcacgggaagtcagtgcgcattacactgtccaagcatcagagtgtgcagctgcctcgggagggtcaggaggaccagggcctgaccaaggactatggcagctccccgctgcaccgcttcaagaaaccaggctccaagaacttccagaacatctttccaccctcagctaccctgcacctctccaacatcccgccctctgtgtcagaggacgacctcaagagcctcttctccagcaacggtggtgtggtcaaaggcttcaagttcttccagaaggaccgcaagatggcactgatccagatgggctctgtggaggaggctgtgcaggcgctgattgaactgcacaaccatgacctgggcgagaaccaccacctgcgagtgtccttttccaagtccaccatcacgcgtacgcggccgctcgagcagaaactcatctcagaagaggatctggcagcaaatgatatcctggattacaaggatgacgacgataaggttccaaaaaagaagagaaaggtataa

Note: All construct sequences used for overexpression in this study are listed.

5.9. In Vitro Transcription

In vitro transcription was performed using the mMESSAGE mMACHINE T7 ULTRA Transcription Kit according to the manufacturer's instructions. The coding sequence (CDS) inserted downstream of the T7 promoter was amplified as a linearized template by PCR using primers spanning from the T7 promoter to the end of the CDS. The PCR product was gel‐purified and used as the template for transcription. Following transcription, RNA was purified using the spin column from the Direct‐zol RNA MiniPrep Kit with TRIzol. The purified mRNA was stored at −80°C until use. Details of reagents and kits are provided in a separate table (Table 1).

5.10. Transfection of Plasmid DNA

Plasmid DNA transfection was performed in 8 well chamber slides. Transfections were carried out 16 h after cell seeding. Medium changes were performed as described for standard culture conditions. Cell collection or fixation was performed at least 48 h after transfection.

For transfection, a DNA plasmid mixture (25 μL per well) was prepared and added to 275 μL of culture medium. For neonatal CMs, the mixture consisted of two components: Mixture 1 (12.5 μL Opti‐MEM and 1.5 μL Lipofectamine 3000) and Mixture 2 (12.5 μL Opti‐MEM, 200–500 ng plasmid DNA, and 0.5 μL Lipofectamine P3000 reagent). Mixture 1 and Mixture 2 were combined prior to addition to the cells.

Details of reagents are provided in a separate table (Table 1).

5.11. mRNA Transfection

mRNA transfection was performed in 96‐well plates at a dose of 30–60 ng per well. Transfections were carried out 4 h after cell seeding. Medium changes were performed as described for standard culture conditions. Live‐cell imaging, cell collection, or fixation was performed at least 48 h after transfection.

For transfection, an mRNA transfection mixture (10 μL per well) was prepared and added to 90 μL of culture medium. For neonatal CMs, the mixture consisted of two components: Mixture 1 (5 μL Opti‐MEM and 0.1 μL Lipofectamine MessengerMAX) and Mixture 2 (5 μL Opti‐MEM and 30–60 ng mRNA). For adult CMs, Mixture 1 contained 5 μL Opti‐MEM and 0.3 μL Lipofectamine MessengerMAX, and Mixture 2 contained 5 μL Opti‐MEM and 100–200 ng mRNA. Mixture 1 and Mixture 2 were combined prior to addition to the cells.

Details of reagents and kits are provided in a separate table (Tables 1 and 4).

5.12. siRNA‐Mediated Gene Silencing

siRNA transfection was performed at a final concentration of 20 nM using either non‐targeting control (NTC) or gene‐specific siRNAs (see Table 5 for details). Transfections were carried out twice for robust silencing, at 4 and 48 h after cell seeding. Medium changes were performed as described for standard culture conditions. Cells were harvested at least 48 h after the final transfection.

TABLE 5.

siRNA_list.

	Company	Product_Number	Assay_ID
siPtbp1	ThermoFisher	4,390,771	s72336
siHnrnph1	ThermoFisher	4,390,771	s81783
siMatr3	ThermoFisher	4,390,771	s69630
siTardbp	ThermoFisher	4,390,771	s106686

Note: All siRNAs used in this study are listed. Related to Figures 5 and S7.

For transfection, a silencing mixture (10 μL per well) was prepared and added to 90 μL of culture medium. The silencing mixture consisted of two components: Mixture 1 (5 μL Opti‐MEM and 0.1 μL Lipofectamine iMAX) and Mixture 2 (5 μL Opti‐MEM and 0.2 μL of 10 μM siRNA). Mixture 1 and Mixture 2 were combined prior to addition to the cells.

Details of siRNA and reagents are provided in a separate table (Tables 1, 5).

5.13. RNA Immunoprecipitation (RIP)

Confluent CMs and non‐CMs purified using the stringent CM purification system described above were used for RIP experiments. Cells cultured in 6‐well plates were washed with 1 mL of ice‐cold PBS and harvested by scraping. Cell suspensions were centrifuged at 15,000 rpm for 5 min at 4°C, and the supernatant was discarded to collect the cell pellet. Subsequent procedures were performed according to the manufacturer's protocol of the Magna RIP kit, which includes Protein A/G magnetic beads. For each sample, 10% of the lysate was reserved as input prior to antibody addition. CM and Non‐CM lysates were subjected to immunoprecipitation with anti‐Ptbp1 antibody or control IgG. Antibodies used for immunoprecipitation included goat anti‐Ptbp1 antibody and normal goat IgG as a control, with 5 μg antibody added per sample. RNA yields ranged from 20 to 110 ng for the input samples, and 1.1 to 4.2 ng for the RIP samples. RNA quality was assessed using a Bioanalyzer, confirming stable baselines and the presence of expected RNA peaks. The resulting RNA was subsequently used for RIP‐seq analysis. Details of reagents and kits are provided in a separate table (Tables 1 and 2).

5.14. RIP‐Seq Library Preparation and Sequencing

For library preparation, 20 ng of input RNA and the entire amount of the recovered RIP RNA were used. RNA was ethanol‐precipitated by mixing the RNA sample with 3 M sodium acetate, Gene‐packman co‐precipitant, and ethanol at the volume ratio of 1:0.1:0.05:2.875, followed by centrifugation at 12,000 × g for 15 min, washing with 70% ethanol, and resuspension in 5.5 μL of nuclease‐free water. Libraries were prepared from 5 μL of ethanol‐precipitated RNA using the NEBNext Ultra II Directional RNA Library Prep Kit for Illumina and NEBNext Multiplex Oligos for Illumina (96 UDI Primer Pairs) according to the manufacturer's instructions. RNA fragmentation was performed at 94°C for 8 min. The optimal number of PCR cycles was determined using KAPA SYBR FAST qPCR Master Mix with one‐sixth of adaptor‐ligated cDNA prior to library amplification. Library amplification was then performed using the remaining five‐sixths of the adaptor‐ligated cDNA, with 9 cycles of PCR for the input samples and 13–15 cycles of PCR for the RIP samples. The peak size and overall fragment size distribution were confirmed to be comparable across all libraries. Libraries were sequenced on Illumina NovaSeq 6000 platform at Azenta Inc. to generate paired‐end 150‐nt reads, yielding 81–121 million paired‐end reads for the input samples, and 16–22 million paired‐end reads for the RIP samples. Details of reagents and kits are provided in a separate table (Table 1).

5.15. RIP‐Seq Data Processing and Analysis

Raw sequencing reads were subjected to quality control and adapter trimming using FastQC (v0.12.1) and fastp (v1.0.1), respectively (Andrews 2010, S. Chen et al. ²⁰¹⁸). Processed reads were aligned to the mouse reference genome (GRCm39) using STAR (v2.7.11b, Dobin et al. ²⁰¹³). Mitochondrial reads were removed using samtools (v1.19.2), and reads overlapping rRNA‐annotated regions were excluded using bedtools (v2.31.1) based on rRNA coordinates extracted from the Ensembl GTF annotation (Quinlan and Hall 2010; H. Li et al. ²⁰⁰⁹). Gene‐level read counts were quantified using featureCounts (v2.0.6, Liao et al. ²⁰¹⁴).

For RIP‐seq data analysis, enrichment was calculated for each gene based on the ratio of immunoprecipitated (IP) to input samples. Specifically, log₂ enrichment values were computed as:

• log₂(Ptbp1 RIP/Input)

• log₂(IgG RIP/Input)

A corrected enrichment score was defined as the difference between these values:

Enrichment = log₂(Ptbp1 RIP/Input) − log₂(IgG RIP/Input).

To define robust Ptbp1‐associated transcripts, the following filtering criteria were applied independently for CMs and non‐CM:

• Sufficient read coverage in both RIP and input samples.

• Positive enrichment over input (log₂[Ptbp1/IP over Input] > 0)

• Positive enrichment relative to IgG control.

• Removal of transcripts with low or negligible signal in IgG controls.

Finally, transcripts were defined as Ptbp1‐enriched when the corrected enrichment score satisfied:

log₂(Ptbp1 RIP/Input) − log₂(IgG RIP/Input) ≥ 1.

GO analysis was performed using transcripts with enrichment scores ≥ 1.

Note that Ptbp1 protein levels were not normalized between the two cell types and that the RIP results should be interpreted as relative enrichment patterns within each cell type because of limited samples.

5.16. Biotinylation by Antibody Recognition (BAR) and Sample Preparation for Mass Spectrometry

A modified biotin‐based proximity labeling (BAR) protocol was used based on a previously reported method (Bar et al. 2018). Confluent CMs and non‐CMs, purified using the stringent purification system described above, were used. Cells cultured in 6‐well plates were washed with ice‐cold PBS and fixed with 4% PFA at room temperature.

5.16.1. Day 1: Fixation, Permeabilization, and Primary Antibody Incubation

Samples were washed three times with PBST (PBS containing 0.1% Tween‐20) for 5 min each at room temperature with gentle shaking. Cells were permeabilized with 0.5% Triton X‐100 in PBS for 30 min, followed by three washes with PBST. Endogenous peroxidase activity was quenched by incubation with 1.5% hydrogen peroxide in PBST for 1 h at room temperature. After washing, samples were blocked with 5% NDS in PBST for 1 h and incubated overnight at 4°C with primary antibodies diluted in 3% NDS and 0.3% Tween‐20/PBS.

5.16.2. Day 2: HRP Labeling and TSA Reaction

Samples were washed three times with PBST for 10 min each and incubated with HRP‐conjugated secondary antibodies for 4.5 h at room temperature. After washing, biotin‐tyramide signal amplification (TSA) was performed by incubating samples with TSA reaction mixture (1× TSA and 0.015% H₂O₂ in PBST) for 10 min with gentle shaking. The reaction was quenched by incubation with 500 mM sodium ascorbate for 5 min, followed by three washes with PBST. At this stage, samples were either stored in PBS at 4°C or subjected to validation by incubation with FITC‐conjugated streptavidin (1:400) and DAPI staining.

5.16.3. Day 3: Protein Extraction Under Denaturing Conditions

Cells were harvested by scraping and lysed in denaturing buffer containing SDS and sodium deoxycholate. Lysates were heated at 99°C for 1 h with intermittent mixing to reverse cross‐linking and solubilize proteins. After centrifugation (15,000 rpm, 5 min), the supernatant was collected and diluted to 1 mL with PBST.

Biotinylated proteins were captured by incubation with magnetic streptavidin beads (250 μL per sample) for 2 h at room temperature followed by overnight incubation at 4°C.

5.16.4. Day 4: Bead Washing and Sample Preparation

Beads were collected using a magnetic stand, and the supernatant (unbound fraction) was removed. Beads were washed sequentially with PBST, high‐salt buffer (1 M NaCl in PBST), and PBS to remove nonspecific binding. Finally, beads were resuspended in PBS and stored at −80°C prior to downstream LC–MS/MS analysis.

Details of reagents are provided in a separate table (Tables 1 and 2).

5.17. LC–MS/MS Analysis

Peptides were analyzed using an Orbitrap Eclipse Tribrid mass spectrometer coupled to a Vanquish Neo nanoLC system (Thermo Fisher Scientific). Peptides were loaded onto a PepMap Neo Trap Cartridge (0.3 × 5 mm, 5 μm; Thermo Fisher Scientific) and separated on an Aurora C18 analytical column (75 μm × 250 mm, 1.7 μm; IonOpticks). Peptides were separated on a C18 analytical column using a 110‐min gradient from 0% to 35% acetonitrile in 0.1% formic acid. MS1 spectra were acquired at a resolution of 60,000 over an m/z range of 295–745. DIA‐MS/MS spectra were acquired at a resolution of 30,000 using HCD with a normalized collision energy of 30%. DIA was performed using 60 variable isolation windows covering the precursor m/z range of 500–740. Raw data were analyzed using DIA‐NN (version 1.9.2) in library‐free mode against the UniProt mouse proteome database (UP000000589, downloaded on August 26, 2024). Trypsin specificity with up to two missed cleavages was allowed. Carbamidomethylation of cysteine was set as a fixed modification, while methionine oxidation and protein N‐terminal acetylation were set as variable modifications. FDR was controlled at 1% at both precursor and protein levels. RT‐dependent cross‐run normalization, match‐between‐runs, and gene‐level proteotypicity were enabled.

In our study, only the bead‐bound fractions were analyzed by LC–MS/MS; unbound fractions were not used for downstream proteomic analysis. Four sample groups were analyzed: CM Ab+, CM Ab−, non‐CM Ab+, and non‐CM Ab−. Samples without primary antibody treatment (Ab−) were used as negative controls for each cell type. Proteins were defined as CM‐enriched when the abundance in CM Ab+ samples was at least fivefold higher than in CM Ab− controls (CM Ab+/CM Ab− ≥ 5). To identify CM‐specific enriched proteins, proteins that were also enriched in non‐CM samples (non‐CM Ab+/Non‐CM Ab− ≥ 5) were excluded and classified as shared proteins. The remaining proteins were defined as CM‐specific enriched proteins.

5.18. FRAP Assays

H9c2(2–1) cells were cultured in Advanced DMEM/F‐12 medium supplemented with 10% FBS and 2 mM L‐glutamine. Cells were transiently transfected with a plasmid encoding GFP‐tagged PTBP1 under the control of the CMV promoter using PEI‐MAX. FRAP analysis was performed 36 h after transfection for H9c2(2–1) cells using a Nikon A1 confocal microscope equipped with a 60× objective lens. A nuclear region containing a PTBP1 granule was photobleached using a 488‐nm laser at 100% power, and fluorescence recovery was recorded over time using low laser power to minimize additional photobleaching. Fluorescence intensity in the bleached region was normalized to that in an unbleached nuclear region.

5.19. pH Assay

For pH manipulation, neonatal CMs were treated at day 2 after plating (day 0 defined as the day of seeding), when cells reached a semi‐confluent state. Cells were incubated for 30 min with culture medium adjusted to the indicated pH and supplemented with 10 μM nigericin. Media were equilibrated to the desired pH immediately prior to treatment with a pH meter.

Control cells were treated with 10 μM nigericin in standard culture medium (pH 7.2). Following treatment, cells were immediately fixed with 4% PFA and subjected to immunostaining to assess the status of Ptbp1 foci.

Details of reagents are provided in a separate table (Table 1).

5.20. Time‐Lapse Imaging

Neonatal mouse CMs were transfected with monomeric GFP‐S16A Ptbp1‐NLS‐Flag mRNA on day 0. Time‐lapse imaging was performed 3 days after transfection using an APX100 microscope equipped with an incubation system. Imaging was performed under controlled temperature and CO₂ conditions. Images were acquired every 10 min using a 40× high‐resolution objective.

CMs were identified based on their spontaneous contractile activity. For each time point, Z‐stack images were acquired at 1 μm intervals (5 slices per time point). Time points with significant stage drift were excluded from analysis.

5.21. Palmitate Treatment

At day 2 after plating (day 0 defined as the day of seeding), when cells reached a semi‐confluent state, cells were treated with either BSA (control) or palmitate‐BSA (PA‐BSA) at a final concentration of 100 μM. After 24 h (day 3), cells were washed and subjected to a second round of the same treatment. Following an additional 24 h (day 4), cells were fixed with 4% PFA and subjected to immunostaining to assess the status of Ptbp1 foci. Details of reagents are provided in a separate table (Table 1).

5.22. H₂O₂ and NAC Treatment

At day 3 after plating, when cells were nearly confluent, four treatment conditions were prepared: control (PBS), NAC alone, H₂O₂ alone, and NAC pretreatment followed by H₂O₂. For NAC treatment, cells were incubated with 4 mM NAC for 1.5 h. For H₂O₂ treatment, cells were exposed to 200 μM hydrogen peroxide (H₂O₂) for 1 h. For the combined treatment, cells were pretreated with 4 mM NAC for 1.5 h followed by 200 μM H₂O₂ for 1 h. Following treatment, cells were fixed with 4% PFA and subjected to immunostaining to assess Ptbp1 foci and DNA damage using 8‐oxo‐dG (8‐OHdG) staining. Details of reagents are provided in a separate table (Table 1).

5.23. NAC Injection

NAC was dissolved in PBS to a concentration of 10 mg/mL. ICR mice received daily subcutaneous or intraperitoneal injections of NAC at 75 mg/kg from P7 to P14.

5.24. RT‐qPCR

Total RNA was isolated from NTC‐ and siRNA‐treated cultured CMs in 24‐well plates using TRIzol reagent and further purified using the spin column from the Direct‐zol RNA MiniPrep Kit with DNase treatment. cDNA was synthesized from total RNA using PrimeScript RT Master Mix according to the manufacturer's instructions. Quantitative real‐time PCR (RT‐qPCR) was performed using TB Green Premix Ex Taq II on a QuantStudio 1 system (Thermo Fisher Scientific) with gene‐specific primers.

The cycling conditions consisted of an initial denaturation at 95°C for 30 s, followed by 35 cycles of 95°C for 5 s and 60°C for 30 s. Melting curve analysis was subsequently performed with a dissociation step from 60°C to 95°C to confirm amplification specificity. GAPDH (glyceraldehyde‐3‐phosphate dehydrogenase) was used as a housekeeping gene for normalization, and relative gene expression was calculated using the ΔΔCt method. Each sample was analyzed in triplicate.

Details of reagents and primers are provided in a separate table (Tables 1, 6).

TABLE 6.

qPCR_Primer_list.

Target_genes	Direction	Sequence(5′ ⇒ 3′)	Primer_name
Hnrnph1	Fw	GGTGCTTATGGTGGAGGCTA	260227_Hnrnph1_qPCR_F‐p
	Rev	CTGTGGTCAGACATTCCTGAAA	260227_Hnrnph1_qPCR_R‐p
Ptbp1	Fw	CGTTTCCATTGGCTGCTGAG	251009_Ptbp1_Fp_qPCR
	Rev	TTTGAGAGGATGAGGCGTGG	251009_Ptbp1_Rp_qPCR
Gapdh	Fw	GTGAACCACGAGAAATATGACAAC	GAPDH qPCR Fp
	Rev	AGTGATGGCATGGACTGTG	GAPDH qPCR Rp
Matr3	Fw	GGTTATCCCCATCTGTGCTCT	260310_Matr3_qPCR_F‐p
	Rev	CATCGACGACTGTGACTTGC	260310_Matr3_qPCR_R‐p

Note: All qPCR primer sequences used in this study are listed. Related to Figure S7.

5.25. Single‐Cell RNA‐Seq Reanalysis

Single‐cell RNA‐seq (scRNA‐seq) data analysis was performed in R (v4.4.2) using Seurat (v5.4.0) based on a previous report (Feng et al. 2022; Stuart et al. ²⁰¹⁹). Details of the analysis are described below.

5.25.1. AddModuleScoring

The Seurat object GSE193346_C57BL6_seurat_object.rds was loaded into RStudio, and CMs from E12.5, E14.5, E16.5, E18.5, P1, and P9 were extracted based on metadata annotations.

Proliferation scores were calculated using Seurat's built‐in cell cycle gene sets by AddModuleScore. RBP scores were calculated using mouse RBP gene lists obtained from EuRBPDB, after retaining only genes detected in the Seurat object. Module scoring was performed using Seurat AddModuleScore with 100 control genes. Cells were ranked according to proliferation or RBP scores. CMs in the top 10% and bottom 10% across all stages were defined as high‐score and low‐score populations, respectively. RBP module scores and individual RBP gene expression levels were compared between these populations using two‐sided Wilcoxon rank‐sum tests, with Benjamini–Hochberg correction applied for multiple testing where applicable. For individual RBP analysis, only RBP genes present in the expression matrix were included.

5.25.2. Correlation Analysis in scRNA‐Seq

Correlation analyses were performed to evaluate the relationship between proliferation scores and RBP‐related features. Pearson's and Spearman's correlation coefficients were calculated between proliferation scores and RBP domain–based module scores (e.g., KH, RRM, RNase, and ResIII domains) across individual CMs.

For gene‐level correlation analysis, the expression of each gene was correlated with Ptbp1 expression across all CMs. Genes expressed in fewer than 10% of cells were excluded to reduce spurious correlations. Pearson correlation coefficients were calculated, and corresponding p‐values were derived and adjusted using the Benjamini–Hochberg method.

Genes were defined as significantly correlated when they satisfied both an absolute correlation threshold (|r| ≥ 0.2) and a FDR < 0.05. Positively and negatively correlated gene sets were defined based on the sign of the correlation coefficient.

5.25.3. RNA‐Binding Protein (RBP) Annotation

A list of mouse RBPs was obtained from EuRBPDB. Non‐canonical RBPs were excluded, resulting in a curated list of 1912 RBPs. Among these, 1657 RBPs were detected in the scRNA‐seq dataset and used for downstream analyses. Functional annotations provided by EuRBPDB were used without modification.

5.26. Bulk RNA‐Seq Reanalysis of CMs

Bulk RNA‐seq data analysis was performed based on a previous report (Quaife‐Ryan et al. ²⁰¹⁷). Publicly available bulk RNA‐seq data of mouse CMs (GSE95755) were obtained and analyzed using the provided CPM‐normalized expression matrix (GSE95755_MultiCellularRNAseq_EdgeR_CPM). Genes expressed in cardiac tissue were defined based on the gene list present in the dataset. RBPs expressed in CMs were identified by intersecting this gene set with the curated mouse RBP list from EuRBPDB after exclusion of non‐canonical RBPs. To identify RBPs enriched in the neonatal stage, genes were further filtered based on the following criteria: (i) sufficient expression in neonatal samples (mean expression > 50 CPM), and (ii) downregulation during maturation, defined as log₂ fold change (adult vs. neonatal) ≤ −1. These filtered genes were defined as neonatal‐enriched RBPs.

5.27. Bulk RNA‐Seq Analysis of siPtbp1 Datasets

Bulk RNA‐seq data analysis was performed based on a previous report (Z. Li et al. ²⁰²⁵). Details of the analysis are described below.

5.27.1. RNA‐Seq Data Processing

Raw sequencing reads were subjected to quality control using FastQC (v0.11.9) and adapter trimming using fastp (v0.20.1). Cleaned reads were aligned to the mouse reference genome (GRCm39) using HISAT2 (v2.2.1, Kim et al. ²⁰¹⁹). Aligned reads were converted to BAM format, sorted, and indexed using samtools (v1.13).

Gene‐level read counts were generated using featureCounts (v2.0.3) with the GRCm39 gene annotation.

5.27.2. Differential Expression Analysis

Gene‐level read counts generated by featureCounts were analyzed using DESeq2 (v1.46.0) for both CM and fibroblasts (FB) (Love et al. 2014). Lowly expressed genes were filtered prior to analysis. Differential expression between conditions was assessed using a negative binomial model implemented in DESeq2. Log₂ fold changes were estimated using shrinkage methods to obtain robust effect size estimates, particularly under low replicate conditions. Differentially expressed genes (DEGs) were defined as transcripts with an adjusted p‐value (Benjamini–Hochberg) < 0.05 and an absolute log₂ fold change (|log₂FC|) ≥ 1.

5.27.3. Alternative Splicing Analysis

Alternative splicing analysis was performed using rMATS (v4.3.0, Shen et al. ²⁰¹⁴). Splicing events were quantified using both junction counts (JC) and junction counts combined with exon counts (JCEC) methods. Events were first filtered based on statistical significance (FDR < 0.05) and a minimum splicing change (|ΔPSI| ≥ 0.1). To ensure robustness, only events detected by both JC and JCEC methods were retained. Robust splicing events were defined as those satisfying all of the following criteria: (i) consistent direction of ΔPSI between JC and JCEC methods; (ii) FDR < 0.05 in both methods; (iii) |ΔPSI| ≥ 0.1 in both methods; and (iv) overlap between JC and JCEC results (robust intersection). Additional filtering was applied to exclude low‐confidence events, including those with insufficient read support or inconsistent PSI estimates across replicates.

5.28. Quantification of Ptbp1 Foci

For tissue sections, nine fields per sample were imaged by confocal microscopy, and the number of Ptbp1 foci within CMs nuclei was manually counted. CMs nuclei were identified using confocal z‐stack images. A DAPI‐positive nucleus was defined as a CM nucleus when it was located within a cTnT‐positive CM. For cultured cells, CM nuclear regions were defined based on DAPI staining in combination with cTnT or α‐actinin signals. Within these regions, puncta were defined as signals exceeding predefined intensity and size thresholds, which were empirically determined based on representative images. Quantification was performed using ImageJ. Puncta other than the defined size threshold (size = 0.5–4.2 μm²) were excluded from the analysis.

5.29. Colocalization Analysis of Ptbp1 Foci

Co‐localization between Ptbp1 foci and other signals was assessed using line intensity profile analysis. Signals were considered co‐localized when the fluorescence intensity peaks of the two channels overlapped along a defined line. For exogenous Ptbp1, foci were defined as exogenously induced only when both FLAG and GFP signals overlapped, thereby excluding potential artifacts. Only puncta showing coincident FLAG and GFP signals were counted as exogenous Ptbp1 foci.

5.30. Microscopy

Immunostained heart tissue sections and plasmid DNA transfected samples were imaged using an Evident BX53 microscope and a Carl Zeiss LSM800 confocal microscope. Cell culture and time‐lapse imaging were performed using an Evident APX100 microscope.

5.31. Statistical Analysis

Unless otherwise indicated in the figure legends, statistical analyses were performed using GraphPad Prism 9. Comparisons between two groups were conducted using unpaired two‐tailed Student's t‐tests, and comparisons among multiple groups were performed using one‐way analysis of variance (ANOVA) followed by Tukey's or Dunnett's multiple comparison tests. Data are presented as mean ± SD.

Due to limited sample availability, some analyses were performed using technical replicates (n = 3–5). In this study, each technical replicate corresponds to an individual well containing CMs derived from pooled neonatal mice (3–4 different litters per preparation). For imaging‐based quantification, each data point represents the average of measurements obtained from 4–9 fields per well.

Author Contributions

Tai Sada: conceptualization; data curation; formal analysis; funding acquisition; investigation; methodology; software; validation; visualization; writing – original draft preparation; Writing – review and editing. Akane Sakaguchi: methodology; supervision. Yoichi Shinkai: methodology; investigation. Reiko Nakagawa: methodology; investigation. Chiharu Tanegashima, Mitsutaka Kadota and Takefumi Kondo: methodology; investigation. Wataru Kimura: conceptualization; funding acquisition; methodology; project administration; resources; supervision; validation; writing – review and editing. All authors have reviewed the manuscript and agree to its submission.

Funding

This work was supported by the RIKEN JRA program and JST SPRING (Grant Number JPMJSP2138) to T.S., grants from Japan Society for the Promotion of Science (KAKENHI 17H05083, 19K22629, 20H03680, 21K18273 and 23H02913), the FOREST program of the Japan Science and Technology Agency (JPMJFR210O), the PRIME program of the Japan Agency for Medical Research and Development, Takeda Science Foundation, Uehara Memorial Foundation, Japanese Circulation Society, Senri Life Science Foundation, Princess Takamatsu Cancer Research Fund, Mitsubishi Foundation, Bristol‐Myers Squibb, Mochida Memorial Foundation for Medical and Pharmaceutical Research, Naito Foundation, Toray Science Foundation, Miyata Cardiac Research Promotion Foundation, Yakult Bio‐Science Foundation, Hoansha Foundation, Japanese Society of Gout and Uric & Nucleic Acids, and a RIKEN CDB/BDR intramural grant to W.K.

Ethics Statement

All animal experiments are approved by the Institutional Animal Care and Use Committee (IACUC) of the RIKEN Kobe Branch.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Figure S1: Cell cycle‐dependent condensate formation of Ptbp1.

(A) Statistical analysis of Ptbp1 condensate distribution patterns. Data are presented as mean ± SD (n = 3 biological replicates). Statistical significance was assessed using repeated‐measures one‐way analysis of variance (ANOVA) followed by Tukey's multiple comparison test. ns, not significant; ***p < 0.001. (B) Representative immunofluorescence images of non‐cardiomyocytes in neonatal mouse primary heart cultures and mouse heart tissues across developmental stages. Blue, DAPI (nuclear marker); green, Ptbp1; red, cTnT; white, WGA (wheat germ agglutinin, cell boundary marker). White arrowheads indicate cardiomyocyte nuclei. Yellow arrowheads indicate non‐cardiomyocyte nuclei. Scale bars indicate 5 μm. (C) Immunofluorescence images of primary‐cultured neonatal cardiomyocytes across cell cycle phases. Blue, DAPI (nuclear marker); green, Ki67 (cell cycle marker); red, Ptbp1. White arrowheads indicate cardiomyocyte nuclei. Cell cycle phases were defined based on Ki67 staining patterns in combination with nuclear morphology. Scale bars indicate 10 μm. (D) Statistical analysis of Ptbp1 granules proportion among cell cycle according to Ki67 signals. Mean proportions are presented as 100% stacked bars. Error bars represent the SD of the Ptbp1 granule‐positive proportion only (n = 5 technical replicates). Statistical significance was assessed using a two‐tailed paired t‐test. ***p < 0.001. E. Immunofluorescence images of primary‐cultured neonatal cardiomyocytes comparing the localization of Ptbp1 condensates and nucleoli. Blue, DAPI; green, Fibrillarin (nucleoli marker); red, Ptbp1. Scale bars indicate 10 μm.

Figure S2: Ptbp1 construction for recapitulating endogenous condensate formation. (A) Immunofluorescence images of primary‐cultured neonatal cardiomyocytes expressing human Ptbp1 constructs. Left, hPtbp1‐GFP; right, hPtbp1‐GFP containing phosphorylation‐deficient mutation (S16A). Green, GFP. White arrowheads indicate cardiomyocyte nuclei. Scale bars indicate 5 μm. (B, C). Quantification of nuclear localization (B) and foci formation (C) of hPtbp1‐GFP and S16A hPtbp1‐GFP. Due to low transfection efficiency, statistical analysis was not performed. (D) Schematic overview of mouse Ptbp1 isoform structure. The nuclear localization signal (NLS) are split into two regions within the N‐terminal 60 amino acids. Isoform 1 represents the longest isoform. Green circles indicate regions absent from each isoform relative to isoform 1. The wavy linker line indicates the intrinsically disordered region (IDR). Images were created with BioRender.com. (E) Immunofluorescence images of primary‐cultured neonatal cardiomyocytes expressing different Ptbp1 variants. Blue, DAPI; red, FLAG (exogenous Ptbp1). White arrowheads indicate Ptbp1 condensate–like structures. White scale bars indicate 5 μm, and yellow scale bars indicate 10 μm. (F, G) Quantification of nuclear localization (E) and foci formation (F) among Ptbp1 variants. Data are derived from technical replicates (n = 3). Due to low numbers of transfected cells, statistical analysis was not performed. (H) Immunofluorescence images of primary‐cultured neonatal cardiomyocytes expressing mouse Ptbp1 variant 1 with or without additional modifications. Red, FLAG (exogenous Ptbp1). White arrowheads indicate condensate‐like structures. Scale bars indicate 5 μm. Schematic images were created with BioRender.com. (I, J). Quantification of nuclear localization (H) and foci formation (I) of overexpressed mouse Ptbp1 variant 1 constructs. Due to low transfection efficiency, statistical analysis was not performed.

Figure S3: Requirement of Ptbp1 RRM domains for condensate formation. (A) Immunofluorescence images of primary‐cultured neonatal cardiomyocytes expressing full‐length or truncated Ptbp1. Green, GFP (exogenous Ptbp1); red, FLAG (exogenous Ptbp1). The same image shown in Figure 3C is used to illustrate the details of the image analysis procedure. Co‐localization of GFP and FLAG signals along the indicated line was used to define Ptbp1 granules. Red arrowheads indicate exogenous Ptbp1 condensates in cardiomyocyte nuclei. Scale bars indicate 10 μm.

Figure S4: S16APtbp1 overexpression alone does not induce cell cycle reentry in adult cardiomyocytes. (A) Schematic overview of the experimental design for Ptbp1 overexpression analysis. Primary adult mouse ventricular cardiomyocytes were isolated and transfected with S16A‐Ptbp1‐NLS‐Flag mRNA at 4 h after seeding. EdU was added 24 h after transfection to assess proliferative activity. Images were created with BioRender.com. (B) Immunofluorescence images of adult cardiomyocytes following Ptbp1 overexpression. Blue, DAPI; green, EdU; gray, cTnT; red, Flag (exogenous Ptbp1). White arrowheads indicate EdU‐positive nuclei in non‐cardiomyocytes, while yellow arrowheads indicate cardiomyocyte nuclei with exogenous Ptbp1 foci. Scale bars indicate 50 μm.

Figure S5: Combination of stringent purification and biotinylated antibody recognition (BAR) enables specific labeling of proteins proximal to Ptbp1 in cardiomyocytes. (A) Schematic overview of the stringent purification strategy. Primary neonatal cardiomyocytes were cultured in a low‐nutrient medium supplemented with lactate and BSA for 2 days to enrich cardiomyocytes, followed by recovery in standard culture medium for 1 day. Images were created with BioRender.com. (B) Immunofluorescence images of highly purified primary‐cultured neonatal cardiomyocytes. Upper images show representative confocal z‐stack images of Ptbp1 foci in cardiomyocytes and non‐cardiomyocytes. Cardiomyocytes were identified based on the localization of the nucleus within a cTnT‐positive cell. Blue, DAPI; red, Ptbp1; green, cTnT. White arrowheads indicate nuclei with Ptbp1 condensates. Scale bars indicate 20 μm. (C) Schematic overview of the biotinylated antibody recognition (BAR) method. After fixation and permeabilization, samples were incubated with a primary antibody against Ptbp1, followed by an HRP‐conjugated secondary antibody. Tyramide signal amplification (TSA) was used to biotinylate proteins proximal to Ptbp1. After reverse cross‐linking, biotinylated proteins were captured with streptavidin beads and subjected to mass spectrometry analysis in cardiomyocyte and non‐cardiomyocyte samples. Images were created with BioRender.com. (D) Immunofluorescence validation of BAR labeling. Blue, DAPI; red, Ptbp1; green, FITC (biotin); purple, cTnT. Yellow arrowheads indicate Ptbp1‐positive regions. Yellow circles highlight co‐localization between Ptbp1 and biotin signals. Scale bars indicate 20 μm.

Figure S6: Gene Ontology (GO) analysis of proteins enriched specifically in Ptbp1 condensates in cardiomyocytes. (A) Top 10 GO terms are shown, ranked by adjusted p‐value. GO terms were summarized into simplified categories to reduce redundancy among closely related terms. (B) Comparison with the PhaSePro database. The proportion of proteins annotated as phase‐separation‐associated in the PhaSePro database (121 curated LLPS‐associated proteins) was calculated among proteins enriched in cardiomyocytes (CM; n = 240) and non‐cardiomyocytes (non‐CM; n = 1203). Proportions are shown as the fraction of PhaSePro‐annotated proteins within each group.

Figure S7: Effect of condensate‐enriched protein knockdown on Ptbp1 condensate formation. (A–C) qRT‐PCR validation of knockdown efficiency for Ptbp1 (A), Hnrnph1 (B), and Matr3 (C). Expression levels were normalized to Gapdh and are shown relative to non‐targeting control (NTC). Data are presented as mean ± SD (n = 3 biological replicates). Statistical significance was assessed using one‐way ANOVA followed by Dunnett's multiple comparison test. ns, not significant; *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. (D–F) Immunofluorescence images of primary‐cultured neonatal cardiomyocytes following siRNA‐mediated knockdown of Ptbp1 (D), Hnrnph1 (E), or Matr3 (F). Blue, DAPI; red, Ptbp1; green, Hnrnph1 (E) or Matr3 (F); gray, cTnT. Scale bars indicate 10 μm. White arrowheads indicate co‐localization with Ptbp1 condensates.

Figure S8: Limited overlap between Ptbp1‐bound transcripts and differentially expressed genes in Ptbp1 knockdown. (A, B). Gene Ontology (GO) analysis of transcripts that are both enriched in Ptbp1 RIP‐seq and exhibit significant expression changes upon Ptbp1 knockdown in cardiomyocytes (CM; A) and non‐cardiomyocytes (non‐CM; B). Top 10 terms are shown, ranked by adjusted p‐value. GO terms were summarized into simplified categories to reduce redundancy among closely related terms. Differentially expressed genes (DEGs) were defined as transcripts with an adjusted p‐value (padj) < 0.05 and an absolute log2 fold change (|log2FC|) ≥ 1. Log2 fold changes were estimated using shrinkage methods to obtain robust effect size estimates, particularly under low replicate conditions (e.g., fibroblasts, n = 2).

Figure S9: Ptbp1 condensates are resistant to changes in pH and palmitate treatment. (A) Immunofluorescence images of primary‐cultured neonatal cardiomyocytes following intracellular pH modulation. Cells were incubated for 30 min with nigericin (10 μM) in culture media adjusted to the indicated pH values (pH 7.2 as the control condition). Blue, DAPI; green, cTnT; red, Ptbp1. Lower panels show higher‐magnification images of Ptbp1 signals. White arrowheads indicate cardiomyocyte nuclei containing Ptbp1 condensates. (B) Immunofluorescence images of primary‐cultured neonatal cardiomyocytes following palmitate (PA) treatment. Cells were incubated with PA (10 μM or 100 μM) for 6 h. Blue, DAPI; gray, cTnT; red, Ptbp1. Scale bar, 10 μm. (C) Quantification of the proportion of cardiomyocyte nuclei containing Ptbp1 condensates after palmitate treatment. Data are presented as mean ± SD (n = 3 technical replicates). Statistical significance was assessed using one‐way ANOVA followed by Dunnett's multiple comparison test. ns, not significant. Similar trends were confirmed in independent biological replicates (n = 3, data not shown).

Figure S10: Oxidative stress affects Ptbp1 condensate formation. (A) Immunofluorescence images of primary‐cultured neonatal cardiomyocytes under ROS‐induced stress conditions. Blue, DAPI (nuclear marker); red, Ptbp1; gray, cTnT (cardiomyocyte marker). Scale bars, 15 μm (overview) and 20 μm (higher magnification). (B) Quantification of the proportion of cardiomyocyte nuclei containing Ptbp1 condensates under ROS induction. Data are presented as mean ± SD (n = 3 technical replicates). Statistical significance was assessed using one‐way ANOVA followed by Tukey's multiple comparison test. ns, not significant; ***p < 0.001; ****p < 0.0001. (C) Schematic overview of the experimental design for NAC injection. Images were created with BioRender.com. (D) Representative immunofluorescence images of NAC injected mouse heart tissue at P14. Blue, DAPI (nuclear marker); green, cTnT (cardiomyocyte marker); red, Ptbp1; white, WGA (cell boundary marker). White arrowheads indicate cardiomyocyte nuclei. Scale bars indicate 2 μm. (E) Quantification of the proportion of cardiomyocyte nuclei containing Ptbp1 condensates. Data are presented as mean ± SD (n = 3 biological replicates). Statistical significance was assessed using a two‐tailed Mann–Whitney U test. ns, not significant.

Video S1: Timelapse_imaging. Time‐lapse imaging of primary neonatal cardiomyocytes overexpressing EGFP‐Ptbp1.

Frames in which the image was completely out of focus due to drift were removed.

Video S2: FRAP_imaging. FRAP imaging of H9c2 cells overexpressing EGFP‐Ptbp1.

Data Availability Statement

BAR data: The accession numbers are PXD078957 for ProteomeXchange and JPST004653 for jPOST. RIP‐seq data: The data are available in the Gene Expression Omnibus (GEO) under accession number GSE334676. Other data will be made available on request.