Fig 2: MRG15 undergoes liquid–liquid phase separation in vitro.A Fluorescence microscopy images of purified recombinant MRG15 at variable NaCl concentrations. Scale bar: 5 μm. B Fluorescence microscopy images of purified recombinant MRG15 at variable protein concentrations. Scale bar: 5 μm. C, D FRAP of EGFP-MRG15 condensation. Representative images of early droplets at indicated time points. C FRAP recovery curves of MRG15 droplets. Data are presented as the mean ± SD; n = 3 biological replicates (D). E Time series showing fusion and fission of MRG15 droplets in vitro. Scale bar: 5 μm.

Fig 3: Constructive model describing the role of MRG15 in alleviating hMSC senescence.In youthful human mesenchymal stem cells (hMSCs), MRG15 facilitates the assembly of protein-DNA complexes through a phase separation mechanism, enabled by interactions between its own and partner proteins. This process promotes the targeted colocalization of MRG15 and its cofactors to the regulatory regions of specific genes, including CDKN1A, LMNB1, CCNB1, HMGB2, NPM1, and MYC, thereby modulating their expression. Conversely, the disruption of protein-DNA complex assembly due to the obstruction of MRG15’s liquid-liquid phase separation (LLPS) results in the mislocalization of MRG15 and its cofactors at critical gene loci, leading to the dysregulation of gene expression (this figure was partially created by Figdraw).

Fig 4: Acceleration of cellular senescence due to MRG15 inactivation in hMSCs.A, B Relative protein levels of MRG15 and p16 in early (P6) and late (P20) passages of hMSCs. Data are presented as mean ± SD, n = 3 biological replicates. C, D SA-β-gal staining of WT and MRG15-knockdown hMSCs, MRG15-KD hMSCs were generated by 48 h lentiviral infection, followed by two passages before analysis. n ≧ 9 biological replicates. Scale bar:100 μm. E–G Clonal expansion analysis (E), IF staining of Ki67 (F, G) MRG15-KD hMSCs were generated by 48 h lentiviral infection, followed by two passages before analysis. White arrows indicate Ki67-positive cells. Scale bars: 200 μm in (E) and 20 μm in (F). H Representative image of p53 and p53 phosphorylation in vector- and MRG15-KD cells. MRG15-KD hMSCs were generated by 48 h lentiviral infection, followed by two passages before analysis. All cells (vector control and knockdown) were treated with 125 nM doxorubicin for 24 h prior to analysis. I Semi-quantified band intensities normalized against GADPH expression. Data are presented as the mean ± SD; n = 3 biological replicates. J IF staining of MRG15 in early (P6) and late (P20) passage hMSCs. The coefficient of variation (CV) for each MRG15 dot in young or aged MSCs was computed to assess the potential differences in nuclear heterogeneity. n = 9 biological replicates. Scale bar: 5 μm.

Fig 5: IDR-dependent MRG15 phase separation relevant to MSC senescence.A Representative image of SA-β-gal staining of MRG15-KD hMSCs and MRG15-KD hMSCs supplemented with MRG15 or MRG15 ΔIDR. MRG15-KD hMSCs were re-expressed with MRG15 or MRG15 ΔIDR for 72 h, followed by two passages before analysis. Scale bars: 100 μm. B Representative image of Ki67 staining of MRG15-KD hMSCs and MRG15-KD hMSCs supplemented with MRG15 or MRG15 ΔIDR. MRG15-KD hMSCs were re-expressed with MRG15 or MRG15 ΔIDR for 72 h, followed by two passages before analysis. Scale bars: 20 μm. C The percentage of SA-β-gal-positive cells in MRG15-KD hMSCs and MRG15-KD hMSCs supplemented with MRG15 or MRG15 ΔIDR, n = 10 biological replicates. D The percentage of Ki67-positive cells in MRG15-KD hMSCs and MRG15-KD hMSCs supplemented with MRG15 or MRG15 ΔIDR, n = 10 biological replicates. E Representative image of p53 and p53 phosphorylation in MRG15-KD cells and MRG15-KD cells supplemented with MRG15 or MRG15 ΔIDR. MRG15-KD hMSCs were re-expressed with MRG15 or MRG15 ΔIDR for 72 h, followed by two passages all constructed cell lines were treated with 125 nM doxorubicin for 24 h before detection. F Schematic of IDR replacement. The IDR of MRG15 was replaced with the IDR1 of AKAP95. G Representative images of 293 T cells expressing MRG15 ΔIDR or MRG15 IDR1. Scale bar: 2 μm. HSA-β-gal staining and Ki67 staining of MRG15-KD hMSCs supplemented with MRG15 ΔIDR or MRG15 IDR1. MRG15-KD hMSCs were re-expressed with MRG15 ΔIDR or MRG15 IDR1 for 72 h, followed by two passages before detection. Top panel: representative image of SA-β-gal staining; bottom panel: representative image of Ki67 staining. I The percentage of SA-β-gal-positive and Ki67-positive cells in MRG15-KD hMSCs supplemented with MRG15 ΔIDR or MRG15 IDR1. MRG15-KD hMSCs were re-expressed with MRG15 ΔIDR or MRG15 IDR1 for 72 h, followed by two passages before detection.