Identification of 53BP1 as a GMCL1 interactor

A) Schematics for the immunoprecipitation-mass spectrometry (IP-MS) workflow using wild-type GMCL1 (GMCL1 WT) and mutants (GMCL1 EK and GMCL1 BBO). Color coding: Red, GMCL1; orange, putative substrates/interacting partners; blue, CUL3; green, RBX1; purple, E2 ubiquitin-conjugating enzyme. B) HEK293T cells were transfected with FLAG-GMCL1 WT, FLAG-GMCL1 EK, or FLAG-GMCL1 BBO. After 24 hours, FLAG-tagged proteins were immunoprecipitated and analyzed by MS/MS. Left panel: proteins enriched with GMCL1 WT vs. BBO; right panel: proteins enriched withGMCL1 EK vs. BBO. Significant interactors were identified using SAINT scores > 0.70 and FDR < 5%. C) HEK293T cells transfected with empty vector (EV), FLAG-GMCL1 WT, FLAG-GMCL1 BBO, FLAG-GMCL1 WKE_AAA (broadly disrupts the binding to CUL3) and FLAG-GMCL1 EK were treated with MLN4924 (3h). 53BP1 and CUL3 were immunoprecipitated with FLAG beads and analyzed by western blot. Asterisk indicates non-specific bands. This experiment was performed four times, and a representative blot is shown. D) HEK293T cells were transfected with EV, FLAG-GMCL1 WT, FLAG-GMCL1 EK, or FLAG-GMCL1 RA. FLAG immunoprecipitations were probed for 53BP1 and CUL3. This experiment was performed four times, and a representative blot is shown. E) HEK293T cells were transfected with EV, FLAG-53BP1 WT, FLAG-53BP1 ΔMFF and FLAG-53BP1 IEDI_AAAA. After MLN4924 treatment (3h), 53BP1 was immunoprecipitated and immunoblotted. This experiment was performed three times, and a representative blot is shown. F) M phase-synchronized GMCL1 FLAG knock-in HCT116 cells were collected. GMCL1 was immunoprecipitated using FLAG-beads and analyzed by immunoblotting.

GMCL1 targets 53BP1 for degradation during M phase

A) Asynchronous or M phase-synchronized parental or GMCL1 knockout (KO) U2OS cells were collected. Whole-cell extracts (WCE) were prepared using RIPA buffer, and other lysates were fractionated into soluble and chromatin-bound fractions for immunoblotting. Arrow indicates GMCL1-specific bands. Asterisk indicates non-specific bands. B) Stable U2OS cell lines expressing EV, FLAG-GMCL1 WT, FLAG-GMCL1 EK, or FLAG-GMCL1 RA in a GMCL1 KO background were synchronized into M phase and fractionated. Immunoblots show the chromatin fraction. C) M phase synchronized FLAG-GMCL1-expressing U2OS cells were collected by mitotic shake-off and cultured in fresh FBS-containing medium for 7 h. G1 phase daughter cells were treated with CHX for the indicated time points, fractionated into soluble and chromatin-bound fractions, and analyzed by immunoblotting. Immunoblots show the chromatin fraction. Independent experiments were performed in triplicate. Differences between KO and KO+WT were tested by two-way ANOVA followed by a Bonferroni test (** p < 0.005, **** p < 0.0001). D) HEK293T cells were transfected with EV, V5-GMCL1 WT, or V5-GMCL1 EK, together with FLAG-TR-TUBE where indicated. FLAG immunoprecipitates were probed for 53BP1 and GMCL1. To visualize ubiquitinated 53BP1, the 53BP1 blot was resolved on a 3-8% gel. The arrow indicates the band corresponding to TR-TUBE.

Cell Cycle Fate Determination of Daughter Cells Following Prolonged Mitosis

A) Dot plots and graphs show the proportions of RPE1 cells in S, G1, and G2/M phases at the indicated time points following mitotic shake-off. Cells were synchronized in mitosis by nocodazole treatment for 16 h and were subsequently released into fresh medium. Cell cycle distribution was determined by EdU pulse labeling and PI staining. EdU was added 1 h prior to each indicated time point. Cells had been transfected 48 h before the experiment with siNT, siGMCL1, or co-transfected with siGMCL1 and either siUSP28 or siTP53BP1. Error bars represent standard deviation. Differences among four groups were tested by one-way ANOVA followed by Tukey’s multiple comparisons test (* p < 0.05, ** p < 0.005, **** p < 0.0001). B) Representative immunoblot showing the silencing efficiencies for panel (A). C) Dot plots and graphs show the proportions of RPE1 cells stably expressing either an empty vector or V5-GMCL1, in S, G1, and G2/M phases at the indicated time points following mitotic shake-off. Cells were synchronized in mitosis by nocodazole treatment for 16 h and were subsequently released into fresh medium. Cell cycle distribution was determined by EdU pulse labeling and PI staining. EdU was added 1 h prior to each indicated time point. Error bars represent standard deviation. Differences between EV and V5-GMCL1 were tested by a two-tailed Welch’s t test (**** p < 0.0001). D) Representative immunoblot showing the over-expression of V5-GMCL1 for panel (C).

GMCL1 expression shows positive correlation with taxane resistance in cancel cell lines

A) Schematic overview of DepMap and PRISM data integration used in the analysis, including GMCL1 mRNA (protein not available) and drug response for taxanes across DepMap cancer cell lines (left panel). Method for classification of cell lines into GMCL1-high and GMCL1-low groups based on its median mRNA expression levels within tissue types (middle panel). Schematic depicting comparison of taxane sensitivity between GMCL1-high and GMCL1-low groups (right panel). (B) Boxplots visualizing comparison of taxane sensitivity (i.e., cabazitaxel, docetaxel and paclitaxel; log-fold change in cell viability) between GMCL1-high and GMCL1-low groups. Statistical significance was assessed using a two-sided, unpaired Wilcoxon-rank sum test (* p < 0.05). (C) Boxplots visualizing comparison of taxane sensitivity (i.e., cabazitaxel, docetaxel and paclitaxel; log-fold change in cell viability) between GMCL1 and TP53BP1 High_Low and Low_High groups, respectively, further stratified by TP53 mutation status. Statistical significance was assessed using a two-sided, unpaired Wilcoxon-rank sum test (* p < 0.05).

GMCL1 deficiency sensitizes cancers with wild-type p53 to paclitaxel-induced apoptosis

(A-G) MCF7 (A), U2OS (C), HeLa (E), HEC-1-A (G) cells were transfected with GMCL1-targeting siRNAs or non-targeting (NT) control for 72 h. Cells were treated with DMSO or 100 nM paclitaxel for 48 h, and cell viability was assessed using the CellTiter-Glo Cell Viability Assay from four or six independent measurements. (B-H) Apoptosis was measured in the same conditions, i.e., MCF7 (B), U2OS (D), HeLa (F), HEC-1-A (H), using the RealTime-Glo Annexin V Apoptosis and Necrosis Assay from four or six independent measurements. For comparisons between two independent groups (A-H), a two-tailed Welch’s test was applied (* p < 0.05, ** p < 0.005, *** p < 0.001). (I) hTERT-RPE1cells were transfected with a non-targeting (NT) control or GMCL1-targeting siRNAs alone or in combination with either TP53BP1 or USP28 targeting siRNAs for 72 h, followed by treatment with 100 nM paclitaxel for 48h. Cell viability was assessed using the CellTiter-Glo Cell Viability Assay from five independent measurements. (J) Apoptosis was measured in the same conditions, i.e., RPE1 using the RealTime-Glo Annexin V Apoptosis and Necrosis Assay from five independent measurements. Error bars represent standard deviation. For analysis involving four groups (I and J), one-way ANOVA followed by Tukey’s multiple-comparisons test was applied (* p < 0.05, ** p < 0.005, *** p < 0.001). (K) Schematic model of this study. During prolonged mitosis, GMCL1 promotes degradation of 53BP1, thereby releasing p53 from the 53BP1-p53-USP28 ternary complex and leading to p53 degradation. As a result, daughter cells proceed through the cell cycle. In the absence of GMCL1, excessive accumulation of 53BP1 results in inheritance of the 53BP1-p53-USP28 ternary complex into daughter cells, where p21 expression is induced and cell cycle progression is arrested.

Mapping 53BP1 binding sites on GMCL1

(A) Predicted structure of GMCL1, domain architecture overview, and comparative analysis of the substrate-binding domain across Drosophila, fish, chicken, mouse, and human. Conserved amino acids are indicated by asterisks, with the human R433 residue highlighted in red. (B) Schematic representation of 53BP1 domains. (C) HEK293T cells were co-transfected with FLAG-GMCL1 and either EV, HA-53BP1 WT, or deletion mutants: HA-53BP1 ΔMFF (minimal focus forming), HA-53BP1 ΔN (lacking the N-terminus of MFF), HA-MFF, HA-MFF ΔOD (oligomerization domain), HA-MFF ΔGAR (glycine-arginine-rich motif), HA-MFF Δ1270-1484, or HA-MFF Δ1370-1484. Immunoprecipitation of 53BP1 was performed using HA beads, followed by immunoblotting of co-purified proteins. (D) HEK293T cells were transfected with EV or FLAG-GMCL1, together with HA-53BP1 WT or mutants: HA-MFF (minimal focus forming), HA-53BP1 ΔMFF, HA-53BP1 ΔN (lacking the N-terminus of MFF), HA-53BP1 ΔTudor, and HA-53BP1 ΔC (lacking the C-terminus of MFF). 53BP1 was immunoprecipitated with HA-beads, followed by immunoblotting of co-purified proteins. Asterisk indicates non-specific bands. (E) To narrow down the GMCL1-binding region on 53BP1, sequential 20-amino-acid deletions within the MFF (minimal focus forming) domain were generated. HEK293T cells were co-transfected with EV or FLAG-GMCL1, along with HA-MFF, HA-MFF Δ1410-1430, and site-specific mutants (every three amino acids mutated within 1410–1430 region of 53BP1). Immunoprecipitation of 53BP1 was conducted with HA-beads, followed by immunoblotting. (F) HEK293T cells were transfected with FLAG-GMCL1 or FLAG-GMCL2. GMCL1 and GMCL2 were immunoprecipitated with FLAG-beads and analyzed by immunoblotting. This experiment was performed two times, and a representative blot is shown. (G) RNA was extracted from asynchronous parental, GMCL1 silenced, and GMCL1 knockout U2OS cells (Clone 1). GMCL1 Mrna levels were quantified by Qpcr from three independent experiments. Error bars represent standard deviation.

Mitotic stress imprints apoptotic memory in daughter cells

(A) Stable U2OS cell lines expressing EV, FLAG-GMCL1 WT, FLAG-GMCL1 EK, or FLAG-GMCL1 RA in a GMCL1 KO background were synchronized into M phase and fractionated and analyzed by immunoblotting; this panel presents the soluble fraction corresponding to Figure 2B. (B) HEK293T cells were transfected with EV, V5-GMCL1 WT, or V5-GMCL1 EK, together with FLAG-TR-TUBE where indicated. FLAG immunoprecipitates were probed for ubiquitin. (C) Stable U2OS cell lines expressing EV, FLAG-GMCL1 WT, FLAG-GMCL1 EK, or FLAG-GMCL1 RA in a GMCL1 KO background were synchronized into M phase by mitotic shake-off following and subsequently release into fresh FBS-containing medium for 20 hours. Daughter cells were fractionated into chromatin-bound fractions and analyzed by immunoblotting. (D) RNA was extracted from the same cells as in (C), and NOXA and PUMA Mrna levels were quantified by Qpcr from three independent experiments. Error bars represent standard deviation from three independent experiments. Differences among four groups were tested by one-way ANOVA followed by Tukey’s multiple comparisons test (* p < 0.05, *** p < 0.001, **** p < 0.0001).

Cell Cycle Fate Determination of Daughter Cells Following Prolonged Mitosis in GMCL1 Knockdown Cells with 53BP1 or USP28.

(A) Dot plots and graphs show RPE1 cell cycle distribution 2 hours and 12 hours following nocodazole release, corresponding to the data in Figure 3A. The y-axis indicates EdU incorporation, and the x-axis represents DNA content measured by PI staining. (B) Dot plots and graphs show RPE1 cell cycle distribution 12 hours following nocodazole release, corresponding to the data in Figure 3C. The y-axis indicates EdU incorporation, and the x-axis represents DNA content measured by PI staining. (C) Cell cycle distribution of asynchronous parental or the indicated FLAG-GMCL1 over-expressing U2OS cells were determined using EdU pulse and PI. Error bars represent standard deviation from three independent experiments.

(A) Boxplots visualizing comparison of taxane sensitivity (i.e., cabazitaxel, docetaxel and paclitaxel; log-fold change in cell viability) between GMCL1-high and GMCL1-low groups. Statistical significance was assessed using a two-sided, unpaired Wilcoxon-rank sum test (* p < 0.05). (B) Boxplots visualizing comparison of taxane sensitivity (i.e., cabazitaxel, docetaxel and paclitaxel; log-fold change in cell viability) between GMCL1 and TP53BP1 High_Low and Low_High groups, respectively, further stratified by TP53 mutation status. Statistical significance was assessed using a two-sided, unpaired Wilcoxon-rank sum test (* p < 0.05).