Fig 2: HEL impairs NPM1 mutant AML cells and induces cell differentiation through targeting RPS2. (A) Anti-FLAG, anti-RPS2 and loading control anti-ß-actin protein expression in OCI-AML3 cells stably expressing FLAG-GFP, FLAG-RPS2 WT or FLAG-RPS2 C222S mutant assessed using western blotting. (B)–(E) Cell viability (B, n = 6); cycles (C, n = 5), apoptotic rates (D, n = 3) and differentiation status (E, n = 3) of OCI-AML3 cells transiently expressing pcDNA3.1 vector, FLAG-RPS2 WT or a FLAG-RPS2 C222S mutant treated with DMSO or HEL (5 µmol/L) for 48 h, cell viability was tested by MTT method and cell cycle, apoptosis, and differentiation were all tested by flow cytometry method. Representative flow cytometry charts of experiments C&D are shown in Supporting Information Figs. S10 and S11. All the above experiments were analyzed by at least three biological replicates, presented as Median ± IQR; ns, not significant, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Fig 3: HEL targets C222 of RPS2 in NPM1 mutant AML cells. (A) General protocol for QTRP. The proteome from HEL or DMSO (control) treatment with OCI-AML3 cells are labeled with IPM probe in vitro and then digested by trypsin, the peptides were conjugated with Light (L) and Heavy (H) Az-UV-biotin reagents, respectively, mixed and captured with streptavidin beads. The probe modified peptides are photo-released for further LC–MS/MS analysis, followed by pFind3 software analysis and post-processing. (B) Distribution of competitive QTRP ratios (RH/L values) quantified from reactions with the human OCI-AML3 proteome treated with 10 µmol/L HEL (three technical replicates from three biological replicates). A cutoff of threefold or greater blockade of IPM probe labeling (RH/L > 3) is shown by black line to mark cysteines that exhibit high sensitivity to HEL, and proteins with cysteines showing the strongest competitive reactivity with HEL are labeled by names. (C) Left: representative MS1 profiles for multiple cysteine-containing peptides from the RPS2 protein, only the peptide containing C222 (MS2 profile underneath) shows sensitivity to HEL competition. Right: representative MS1 profiles for HEL-sensitive C222 peptide in RPS2 showing concentration-dependent blockade of IPM labeling by HEL. (D) Box plots showing the RH/L values C222 peptide of RPS2 quantified from competitive QTRP experiments with the OCI-AML3 proteome treated with 1, 10 and 50 µmol/L HEL, respectively, data were shown in three biological replicates. Data are presented as Median ± IQR, ns, not significant, *P < 0.05. (E) Left: Successful plasmid mutagenesis was verified by Sanger sequencing. Mutation site is presented in red line. Right: Showing bands in the expected size (37 kDa) for the purified RPS2 WT and C222S mutant recombinant proteins. 12% SDS-PAGE gel stained with Coomassie blue or Western blot with RPS2 antibody. (F) Gel-based ABPP analysis of HEL against recombinant human RPS2 WT or C222S mutant proteins. Proteins were preincubated with DMSO vehicle or different concentrations of HEL before IA-rhodamine labeling. Proteins were resolved by SDS-PAGE and visualized by in-gel fluorescence, and protein loading was assessed by Coomassie staining. (G) MST traces and dose–response curves of labeled RPS2 WT or C222S mutant proteins bound with HEL. Error bars indicate the SD between the performed three technical replicates. Dissociation constants (KD) are shown.

Fig 4: RPS2 is an essential target for NPM1 mutant AML treatment, and this target is p53 dependent. OCI-AML3 shNC (blue) or shp53 (red) cells transfected with RPS2 siRNA or siNC (control) and treated with DMSO or HEL (5 µmol/L), respectively for 48 h, the cells were then tested for the (A) proliferation, (B) apoptosis, (C) cycle arrest, and (D) differentiation status [a, DMSO; b, siNC; c, HEL (5 µmol/L)+siNC; d, siRPS2; e, HEL (5 µmol/L)+siRPS2] and (E) expressions of p53, RPS2, caspase 3, C-caspase 3, p21 and CEBP/a proteins. Experiments (A) was analyzed by five biological replicates; experiments (B–D) were analyzed by three biological replicates. Representative flow cytometry charts of experiments B&C are shown in Supporting Information Figs. S13 and S14. All data are presented as Median ± IQR; ns, not significant, *P < 0.05, **P < 0.01, unpaired t test; gels shown in experiment (E) are representative blots for three biological independent samples per group; Protein level quantification results are shown in Supporting Information Fig. S15. (F) Expression of RPS2 protein and loading control actin levels in OCI-AML3 cells transfected with RPS2 siRNA and siNC (control), assessed using Western blot. (G, H) Northern blot analysis of pre-rRNA processing phenotypes after transfected with RPS2 siRNA or siNC (control) treated with DMSO or HEL (10 µmol/L) for 48 h, respectively. Mature rRNAs were shown on the EtBr-stained gel. Bands on the Northern blots or EtBr gels (n = 3) were quantitated by ImageJ, presented as Median ± IQR; ns, not significant, *P < 0.05, **P < 0.01, ****P < 0.0001. (I) FBL (red) and NPM1 (green) localization in the RPS2 knock-down OCI-AML3 cells by IF. Hoechst 33342 (blue) was used to stain for nuclei. Images by Leica SP8 confocal microscope. Scale bar, 12.3 µm. Quantification of percentage of cells with nucleolar stress is shown in Supporting Information Fig. S16.

Fig 5: HEL inhibits RPS2 interaction with the ribosome proteins (RPs) therefore affecting the RPs–MDM2–p53 pathway. (A) Schematic workflow of MS-based Co-IP. OCI-AML3 cells were treated with 10 µmol/L HEL or DMSO, respectively, for 24 h. Subsequently, DSSO linker was added to covalently link protein complexes within cells. RPS2 (or MDM2) and linked proteins were enriched using an anti-RPS2 (or anti-MDM2) antibody immobilized on protein A/G beads. Tryptic peptides were measured via LC–MS/MS, and analyzed using MaxQuant and Perseus. (B) Volcano plots represent two-sample t-test results of anti-RPS2 Co-IP compared to isotype control Co-IP (n = 3). Cut-off criteria were defined as log2 = 2 (4-fold enrichment) enrichment factor and –log10 (t-test P-value) = 1.5 (dotted lines). RPS2 (or MDM2) is colored in red, RPs in blue and p53 in indigo. (C) Venn diagram showing the overlap between the anti-MDM2 Co-IP (DMSO treated) and anti-RPS2 Co-IP (HEL treated) enriched proteins. The common proteins are listed, where RPL11 (red) possess the highest enrichment score. (D) The OCI-AML3 cells stably expressing FLAG-GFP, FLAG-RPS2 WT or FLAG-RPS2 C222S were treated with DMSO or HEL (10 µmol/L) for 24 h, FLAG-GFP-, FLAG-RPS2-WT- and FLAG-RPS2-C222S-interacting proteins were subsequently enriched and subjected to Western blot analysis, protein levels of FLAG-tagged proteins, RPL11 and loading control GAPDH showing specific interaction of RPL11 with FLAG-RPS2 WT, however, when cells were treated with HEL, this interaction will be significantly reduced. Although the C222S mutant RPS2 protein can also specifically interact with RPL11, this interaction was not disrupted by HEL. (E) Western blot analysis of MDM2, p53 and RPL11 expression levels of OCI-AML3 either treated with DMSO or HEL (5 and 10 µmol/L) for 24 h and Co-IP of MDM2 in OCI-AML3 cells and detection of MDM2 and interactors p53/RPL11 by Western blot analysis. Gels shown in D and E are representative blots from n = 3 biologically dependent samples per group, bands were quantitated by ImageJ, presented as Median ± IQR; ns, not significant, *P < 0.05, **P < 0.01, ***P < 0.001.