Journal: The EMBO Journal

Article Title: METTL3/MYCN cooperation drives neural crest differentiation and provides therapeutic vulnerability in neuroblastoma

doi: 10.1038/s44318-024-00299-8

Figure Lengend Snippet: ( A ) Schematic diagram showing key steps involved in the differentiation process of human embryonic stem cells (hESC) into trunk neural crest cells (tNCC), followed by their further differentiation into sympathoadrenal progenitors (SAP) and, ultimately, into sympathetic neurons (SN). ( B ) Representative immunofluorescence (IF) images illustrating the expression of distinct lineage markers at different stages of differentiation: HOXC9 for tNCC, PHOX2B for SAP, and PRPH for SN. The scale bar is indicative of 50 µm. ( C ) Representative immunoblot shows the levels of METTL3, METTL14 RBM15, and WTAP across various stages of differentiation, including hESC, tNCC, SAP, and SN. Vinculin, A-tubulin, and GAPDH were loading controls. The values below the blots indicate the fold change (normalized to loading control) in the levels of METTL3, METTL14, RBM15, and WTAP. The experiments were repeated three times. ( D ) The total number of m 6 A peaks in hESC and tNCC, the p value was calculated using a permutation test, and the number of permutations was set to 1000. ( E ) Identified motifs from de novo motif analysis of m 6 A peaks enriched in hESC and tNCC and P values were obtained using the HOMER tool. ( F ) Metagene analysis showing relative m 6 A peak density at genes in hESC and tNCC. ( G ) Venn diagram showing overlap of the m 6 A positive (m 6 A + ) [containing at least one m 6 A peak] genes in hESC and tNCC. ( H ) Left: Venn diagram comparison of differentially expressed genes (DEGs) [hESC vs. tNCC] and m 6 A + . Right: Top enriched terms associated with m 6 A-containing DEGs (hESC vs. tNCC) were identified using enrichGO, with p values obtained through Fisher’s exact test. ( I ) Representative immunoblot shows the levels of METTL3, in hESC and tNCC with control (shCtrl) or stable METTL3 KD (shMETTL3-1, shMETTL3-2). Vinculin and GAPDH were loading control. The values below the blots indicate the fold change (normalized to loading control) in the levels of METTL3. The experiments were repeated three times. ( J ) Top enriched terms associated with DEGs (shCtrl vs. shMETTL3-1) that are m 6 A+ in tNCC were identified using enrichGO, with P values obtained through Fisher’s exact test. ( K ) Genome browser screenshots of HOXC8 and HOXC9 3´UTR, showing the presence of m 6 A in tNCC, neural crest stem cells (NCSC) at day 7 and at day 14. ( L ) RT-qPCR data showing the relative expression of HOXC8 and HOXC9 in SAP following METTL3 KD. GAPDH was used to normalize the qPCR data. Data are shown as mean ± SEM of three independent biological replicates. Two-way ANOVA with Šídák’s multiple comparisons test was used. ( M ) shCtrl, shMETTL3-1, and shMETTL3-2 hESC were differentiated to SN, followed by IF with PRPH antibody to assess neurite length and PRPH signal intensity. Data are represented by box-whisker plots where the median is indicated by a horizontal line, the boxes represent the 25th to 75th percentiles, the whiskers show the 10th to 90th percentiles, and any outliers beyond this range are displayed as individual dots. This analysis was conducted across three independent biological replicates and statistical significance was determined using one-way ANOVA with Dunnett’s multiple comparisons test. Scale bar represents 100 µm. .

Article Snippet: TetO shCtrl (pLKO-Tet-On-shRNA-Control) , Addgene , Cat# 98398.

Techniques: Immunofluorescence, Expressing, Western Blot, Control, Comparison, Quantitative RT-PCR, Whisker Assay

Journal: The EMBO Journal

Article Title: METTL3/MYCN cooperation drives neural crest differentiation and provides therapeutic vulnerability in neuroblastoma

doi: 10.1038/s44318-024-00299-8

Figure Lengend Snippet: ( A ) Kaplan–Meier plot illustrates event-free survival in neuroblastoma (NB) patients ( n = 498, SEQC cohort) with either low or high expression of HOXC8 and HOXC9 . Statistical analysis of survival was performed with a log-rank test. ( B ) Box-whisker plots show HOXC8 and HOXC9 expression in NB patients from the SEQC cohort, classified based on MYCN amplification status (non-MNA: non- MYCN -amplified, n = 401; MNA: MYCN -amplified, n = 92). The centerlines of the boxes represent the medians, the boxes extend from the 25th to 75th percentiles, and the whiskers depict the minimum and maximum values. Statistical analysis was performed using a two-tailed unpaired t test. ( C ) Left: Box-whisker plots show HOXC9 protein levels in non-MNA ( n = 22) and MNA ( n = 12) NB patients (Hartlieb et al, ). The centerlines of the boxes represent the medians, the boxes extend from the 25th to 75th percentiles, and the whiskers depict the minimum and maximum values. Statistical significance was determined using a two-sided Mann-Whitney test. Right: Immunoblot shows the levels of HOXC9 in NB patient samples. MYCN status, risk stratification, stage, and HOXC9 expression levels determined by RNA sequencing (expression score) are also provided. HSP90 was used as a loading control. The experiments were repeated three times. ( D ) Browser screenshot of m 6 A RIP-seq tracks at 3´UTR of HOXC8 and HOXC9 genes in MNA NB tumors. ( E ) Top enriched terms associated with m 6 A+ genes in both MNA NB tumors were identified using enrichGO, with P values obtained through Fisher’s exact test. ( F ) Genome browser screenshot showing the presence of m 6 A enrichment at 3´UTR of HOXC8 and HOXC9 genes in MNA NB cell line, SK-N-BE(2). ( G ) Differentially expressed posterior HOXC genes between control and METTL3 KD SK-N-BE(2) cells, and the number of m 6 A peaks identified using MACS peak caller in these genes are indicated. ( H ) Stability of HOXC8 and HOXC9 transcripts detected by RT-qPCR following Actinomycin D (10 µg/ml) mediated transcription blocking for the time points indicated in control (TetO shCtrl) and METTL3 KD (TetO shM3-1) SK-N-BE(2). Assay was conducted following 3 and 6 days of doxycycline (Dox) addition. Line plots present the quantification of remaining levels of HOXC8 and HOXC9 transcript at the indicated time points. Half-life ( t 1/2 ) values are also denoted. Experiments were performed in three independent biological replicates. Data are presented as mean ± SEM. Two-way ANOVA with Šídák’s multiple comparisons test was employed. ( I ) Left: Line plots showing tumor volume in control (TetO shCtrl) and METTL3 KD (TetO shM3-1) SK-N-BE(2) mouse xenograft with representative tumors from each group ( n = 4 mice per group). Data are presented as mean ± SEM. Middle: Box-whisker plots show tumor weight in control and METTL3 KD xenograft tumors. The median is indicated by a horizontal line, the boxes represent the 25th to 75th percentiles, the whiskers show the 10th to 90th percentiles, and any outliers beyond this range are displayed as individual dots. Two-way ANOVA with Šídák’s multiple comparisons test was employed to compare tumor volumes and two-tailed unpaired t test for tumor weights. Right: Immunoblot showing expression of METTL3, METTL14, and HOXC9 in control and METTL3 KD xenografted tumors. GAPDH and vinculin were used as loading controls. The values below indicate the fold change (normalized to loading control) in the individual METTL3 KD xenografts compared to the mean expression of the control xenografts for METTL3, METTL14, and HOXC9. The experiments were repeated three times. ( J ) Representative IF showing PRPH (green), TUBB3 (red) staining in control (TetO shCtrl) and METTL3 KD (TetO shM3-1, TetO shM3-2) SK-N-BE(2) cells were pretreated with Dox for 1 day followed by 3 days of Dox and retinoic acid (RA) mediated differentiation. Box-whisker plots show the quantification of the neurite length, TUBB3, and PRPH intensity. The median is indicated by a horizontal line, the boxes represent the 25th to 75th percentiles, the whiskers show the 10th to 90th percentiles, and any outliers beyond this range are displayed as individual dots. Scale bar represents 50 µm. Experiments were performed in three independent biological replicates and statistical significance was determined using one-way ANOVA with Dunnett’s multiple comparisons test. .

Article Snippet: TetO shCtrl (pLKO-Tet-On-shRNA-Control) , Addgene , Cat# 98398.

Techniques: Expressing, Whisker Assay, Amplification, Two Tailed Test, MANN-WHITNEY, Western Blot, RNA Sequencing, Control, Quantitative RT-PCR, Blocking Assay, Staining

Journal: The EMBO Journal

Article Title: METTL3/MYCN cooperation drives neural crest differentiation and provides therapeutic vulnerability in neuroblastoma

doi: 10.1038/s44318-024-00299-8

Figure Lengend Snippet: ( A ) Left: Co-IP of METTL3 or MYCN from lysates of Flag-MYCN overexpressing SHEP cells (SHEP MYCN ) after Dox induction for 24 h, blotted with MYCN or METTL3 antibodies. IgG served as a negative control. Right: Proximity ligation assay (PLA) in tNCC and 24 h Dox-induced SHEP MYCN cells depicting METTL3 and MYCN PLA signal (green) in the nucleus (marked by DAPI). The negative control shows PLA with only the METTL3 antibody. Scale bar is 5 μm. The experiments were repeated three times. ( B ) Co-IP of METTL3 or MYCN from lysates of SHEP MYCN after 24 h of Dox induction and 1 h treatment with 300 nM flavopiridol (FP), followed by western blotting with MYCN or METTL3 antibodies. IgG was used as a negative control. The experiments were repeated three times. ( C ) Venn diagram comparison of METTL3 and MYCN binding sites determined from ChIP-seq experiments performed in tNCC. ( D ) Distribution and heatmaps of normalized ChIP-seq reads for METTL3, MYCN, and H3K27ac over the MYCN and METTL3 overlapping peak coordinates. The data is centered on MYCN peaks (− 4 kb to +4 kb). ( E ) Distribution of METTL3 ChIP signal in a metagene profile. The data is centered at the transcription start site (TSS) [−1 kb to +1 kb], at genes that are co-bound by METTL3 and MYCN or bound by METTL3 only in tNCC. ( F ) Genome browser screenshot showing METTL3, MYCN, and H3K27ac ChIP-seq signals in hESC and tNCC over the HOXC8 and HOXC9 gene locus. ( G ) Venn diagram comparison of METTL3 and MYCN binding sites determined from ChIP-seq experiments performed in SHEP MYCN cells before and after Dox induction for 24 h. ( H ) Distribution and heatmaps of normalized ChIP-seq reads for METTL3 and MYCN overlapping peaks centered on MYCN peaks (− 4 kb to +4 kb) in SHEP MYCN cells after Dox induction. ( I ) Left: Browser screenshot showing m 6 A RIP-seq tracks at 3´UTR of HOXC8 and HOXC9 genes in SHEP MYCN cells before and after Dox induction for 24 h. Right: m 6 A RIP-qPCR data showing enrichment of both HOXC8 and HOXC9 in SHEP MYCN cells before and after Dox induction for 24 h. Data are represented as a percentage of input. IgG was used as a negative control. Data are from three independent experiments and shown as mean ± SEM. Two-way ANOVA with Tukey’s multiple comparisons test was used. ( J ) RT-qPCR data showing the expression of HOXC8 and HOXC9 in SHEP MYCN cells Left: SHEP cells Right: with either control (TetO shCtrl) or METTL3 KD (TetO shM3-1) after Dox induction for 6 days. GAPDH was used to normalize the qPCR data. Data are shown as mean ± SD of three independent biological replicates. Two-way ANOVA with Šídák’s multiple comparisons test was used. ( K ) Left: Distribution of METTL3 ChIP signal in a metagene profile. The data are centered at the transcription start site (TSS) [−1 kb to +1 kb], at genes that are co-bound by METTL3 and MYCN or bound by METTL3 only in SHEP MYCN cells after 24 h Dox induction. ( L ) Venn diagram comparing m 6 A+ and METTL3/MYCN co-bound genes SHEP MYCN cells after Dox 24 h induction. METTL3 and MYCN co-bound regions were determined using the ChIP-seq experiments. ( M ) Metagene analysis showing relative m 6 A peak density at genes co-bound by METTL3 and MYCN or the rest of m 6 A-containing genes in SHEP MYCN cells after Dox induction. ( N ) Box-whisker plots showing the number of m 6 A peaks/genes that are co-bound by METTL3 and MYCN (median = 2) or the rest of m 6 A-containing genes (median = 1) in SHEP MYCN cells after Dox induction. The number of co-bound peaks and Rest peaks used for this analysis are 3127 and 3349, respectively. Whiskers indicate the 1st to 99th percentiles, and any outliers beyond this range are shown as individual dots. Statistical analysis was performed using the Wilcoxon matched-pairs signed rank test. ( O ) PLA in SHEP MYCN cells with or without Dox induction for 24 h depicting METTL14 and H3K36me3 PLA signal (green) in the nucleus (marked by DAPI). The negative control shows PLA with only the H3K36me3 antibody. Signal intensity measurements were taken from over 50 cells. Data are from three independent experiments and presented as box-whisker plots where the median is indicated by a horizontal line, the boxes represent the 25th to 75th percentiles, the whiskers show the 10th to 90th percentiles and any outliers beyond this range are displayed as individual dots. Statistical analysis was performed using a two-tailed unpaired t test. Scale bar represents 10 μm. .

Article Snippet: TetO shCtrl (pLKO-Tet-On-shRNA-Control) , Addgene , Cat# 98398.

Techniques: Co-Immunoprecipitation Assay, Negative Control, Proximity Ligation Assay, Western Blot, Comparison, Binding Assay, ChIP-sequencing, Quantitative RT-PCR, Expressing, Control, Whisker Assay, Two Tailed Test

Journal: The EMBO Journal

Article Title: METTL3/MYCN cooperation drives neural crest differentiation and provides therapeutic vulnerability in neuroblastoma

doi: 10.1038/s44318-024-00299-8

Figure Lengend Snippet: ( A ) Top 10 transcription factor binding motifs enriched in the promoter region of the DEGs (Flag- MYCN overexpressed Dox- vs. Dox + ) in SAP. P values were obtained using the HOMER tool. ( B ) IF showing expression of PRPH (green) and TUBB3 (red) in Flag-MYCN overexpressed (Dox + , from day 5 onwards) SN-stage cells with either control (Ctrl) or HOXC9 overexpression (OE). HOXC9 OE was performed from day 9 of differentiation. Box-whisker plots show the quantification of the neurite length, TUBB3, and PRPH intensity. The median is indicated by a horizontal line, the boxes represent the 25th to 75th percentiles, the whiskers show the 10th to 90th percentiles, and any outliers beyond this range are displayed as individual dots. Experiments were performed in three independent biological replicates. Two-tailed unpaired t test was used. Scale bar represents 100 µm. ( C ) Illustration describing the recruitment of dCasRx-FTO at target RNA. Representative IF showing expression of PRPH (green) and TUBB3 (red) in SK-N-BE(2) cells expressing dCasRx-FTO Mutanat (catalytically dead-H231A and D233A mutant)/dCasRx-FTO WT (wild-type) with either non-template control (NTC gRNA) or HOXC9 guide RNAs (HOXC9 gRNA-1, HOXC9 gRNA-2). Dox induction was performed for 72 h followed by 3 days of RA-mediated differentiation in the presence of Dox. Box-whisker plots show neurite length. The median is indicated by a horizontal line, the boxes represent the 25th to 75th percentiles, the whiskers show the 10th to 90th percentiles, and any outliers beyond this range are displayed as individual dots. Data are from three independent experiments. Two-way ANOVA with Šídák’s multiple comparisons test was used. Scale bar represents 50 μm. ( D ) Representative IF showing expression of HOXC9 (green) in SK-N-BE(2) cells in the same condition as detailed above in ( C ). Box-whisker plots show mean HOXC9 intensity normalized with DAPI intensity. The median is indicated by a horizontal line, the boxes represent the 25th to 75th percentiles, the whiskers show the 10th to 90th percentiles, and any outliers beyond this range are displayed as individual dots. Signal intensity measurements were taken from over 2000 cells. Data are from three independent experiments. Two-way ANOVA with Šídák’s multiple comparisons test was used. Scale bar represents 50 μm. ( E ) Representative immunoblot showing expression of HOXC9 in control (TetO shCtrl) and METTL3 KD (TetO shM3-1, TetO shM3-2) SK-N-BE(2) cells, along with shRNA-mediated KD of HOXC9 (shHOXC9). Dox induction was performed for 72 h. Vinculin was loading control. The values below indicate the fold change in levels of HOXC9. The experiments were repeated three times. ( F ) IF showing expression of PRPH (green) and TUBB3 (red) in TetO shCtrl, TetO shM3-1, and TetO shM3-2 SK-N-BE(2) cells in similar conditions as described in ( E ), except Dox was added for 24 h after HOXC9 shRNA transduction followed by 3 days RA-mediated differentiation in the presence of Dox. Box-whisker plots show the quantification of the neurite length. The median is indicated by a horizontal line, the boxes represent the 25th to 75th percentiles, the whiskers show the 10th to 90th percentiles, and any outliers beyond this range are displayed as individual dots. Experiments were performed in three independent biological replicates. Two-way ANOVA with Šídák’s multiple comparisons test was used. Scale bar represents 50 µm. .

Article Snippet: TetO shCtrl (pLKO-Tet-On-shRNA-Control) , Addgene , Cat# 98398.

Techniques: Binding Assay, Expressing, Control, Over Expression, Whisker Assay, Two Tailed Test, Mutagenesis, Western Blot, shRNA, Transduction

Journal: The EMBO Journal

Article Title: METTL3/MYCN cooperation drives neural crest differentiation and provides therapeutic vulnerability in neuroblastoma

doi: 10.1038/s44318-024-00299-8

Figure Lengend Snippet: ( A ) HOXC8 (red), and Flag (green) IF were performed in Flag- MYCN overexpressed (Dox + , from day 5 onwards) SAP after DMSO or STM2457 (10 μM) treatment. STM2457 or DMSO was added on day 9 of differentiation. Box-whisker plots show HOXC8 signal intensity normalized to DAPI intensity. The median is indicated by a horizontal line, the boxes represent the 25th to 75th percentiles, the whiskers show the 10th to 90th percentiles, and any outliers beyond this range are displayed as individual dots. Signal intensity measurements were taken from over 2200 cells and data are from three independent experiments. Statistical analysis was performed using a two-tailed unpaired t test. Scale bar represents 10 μm. ( B ) HOXC9 (green), and MYCN (red) IF were performed in Flag- MYCN overexpressed (Dox + , from day 5 onwards) SAP after DMSO or STM2457 (10 μM) treatment. STM2457 or DMSO was added on day 9 of differentiation. Box-whisker plots show HOXC9 intensity normalized to DAPI intensity. The median is indicated by a horizontal line, the boxes represent the 25th to 75th percentiles, the whiskers show the 10th to 90th percentiles, and any outliers beyond this range are displayed as individual dots. Signal intensity measurements were taken from over 1500 cells and data from three independent experiments. Statistical analysis was performed using a two-tailed unpaired t test. Scale bar represents 10 μm. ( C ) PRPH (green), and MYCN (red) IF were performed in Flag- MYCN overexpressed (Dox + , from day 5 onwards) SN-stage cells after DMSO or STM2457 (10 μM) treatment. STM2457 or DMSO was added from day 9 of differentiation. Box-whisker plots show quantification of PRPH signal intensity and neurite length. The median is indicated by a horizontal line, the boxes represent the 25th to 75th percentiles, the whiskers show the 10th to 90th percentiles, and any outliers beyond this range are displayed as individual dots. Data are from three independent experiments. Statistical analysis was performed using a two-tailed unpaired t test. Scale bar is 50 μm. ( D ) Representative IF images of TUBB3 (green) in SK-N-BE(2) cells that were pretreated with either DMSO or STM2457 (10 μM) for 24 h, followed by RA treatment for another 3 days. Box-whisker plots show the quantification of neurite length. The median is indicated by a horizontal line, the boxes represent the 25th to 75th percentiles, the whiskers show the 10th to 90th percentiles, and any outliers beyond this range are displayed as individual dots. Data are from three independent experiments. Statistical analysis was performed using a two-tailed unpaired t test. Scale bar is 50 μm. ( E ) RPA32 (red) [top] and gamma-H2AX (green) [bottom] IF were performed in Flag-MYCN overexpressed (Dox + , from day 5 onwards) SN-stage cells after DMSO or STM2457 (10 μM) treatment. STM2457 or DMSO was added from day 13 of differentiation. Box-whisker plots show either RPA32 or gamma-H2AX signal intensity normalized to DAPI intensity. The median is indicated by a horizontal line, the boxes represent the 25th to 75th percentiles, the whiskers show the 10th to 90th percentiles, and any outliers beyond this range are displayed as individual dots. Signal intensity measurements were taken from over 90 cells and data are from three independent experiments. Statistical analysis was performed using a two-tailed unpaired t test. Scale bar is 10 μm. ( F ) RPA32 (red) IF was performed in SK-N-BE(2) cells with TetO shCtrl or TetO shM3-1 after 48 h Dox induction. Box-whisker plots show RPA32 signal intensity normalized to DAPI intensity. The median is indicated by a horizontal line, the boxes represent the 25th to 75th percentiles, the whiskers show the 10th to 90th percentiles, and any outliers beyond this range are displayed as individual dots. Signal intensity measurements were taken from over 800 cells and data are from three independent experiments. Statistical analysis was performed using a two-tailed unpaired t test. Scale bar is 5 μm. ( G ) Left: Representative IF showing expression of RPA32 (red) in SK-N-BE(2) cells treated either with DMSO, STM2457 (10 μM), doxorubicin (1 μM), or a combination of STM2457 with doxorubicin for 24 h. Scale bar is 100 μm. Middle: Box-whisker plots show RPA32 signal intensity normalized to DAPI intensity. The median is indicated by a horizontal line, the boxes represent the 25th to 75th percentiles, the whiskers show the 10th to 90th percentiles, and any outliers beyond this range are displayed as individual dots. Signal intensity measurements were taken from over 1900 cells. Data are from three independent experiments. Right: Bar plots show relative cell viability in SK-N-BE(2) cells treated for 72 h with DMSO, STM2457, Doxorubicin, and a combination of STM2457 with doxorubicin. Data are presented as mean ± SEM from three independent experiments. Statistical analysis was conducted using a one-way ANOVA with Tukey’s post hoc test. ( H ) Left: Cartoon demonstrating the experimental strategy used for the mouse in vivo experiment performed with patient-derived xenograft (PDX) cells. MNA COG-N-415x, PDX cells were injected into NSG mice. Once tumors reached 170 mm 3 mice were randomly allocated into four treatment groups ( n = 4–6 mice per group) and treated for 14 days with either vehicle (20% hydroxypropyl-beta cyclodextrin) daily, STM2457 (50 mg/kg in vehicle) daily, doxorubicin (0.2 mg/kg in vehicle) every three days or a combination of STM2457 and doxorubicin at the same doses. Line plots show tumor volume (middle) and body weight (right) in the treatment groups. Data are presented as mean ± SEM. Statistical analysis was conducted using a two-way ANOVA with Tukey’s post hoc test. .

Article Snippet: TetO shCtrl (pLKO-Tet-On-shRNA-Control) , Addgene , Cat# 98398.

Techniques: Whisker Assay, Two Tailed Test, Expressing, In Vivo, Derivative Assay, Injection

Journal: The EMBO Journal

Article Title: METTL3/MYCN cooperation drives neural crest differentiation and provides therapeutic vulnerability in neuroblastoma

doi: 10.1038/s44318-024-00299-8

Figure Lengend Snippet: Reagents and tools table

Article Snippet: TetO shCtrl (pLKO-Tet-On-shRNA-Control) , Addgene , Cat# 98398.

Techniques: Sequencing, Cloning, Imaging, Staining, Software, Control, Mutagenesis, Plasmid Preparation

Journal: bioRxiv

Article Title: A targetable dependency on nonsense-mediated decay for proteostasis and immune control in small cell lung cancer

doi: 10.64898/2026.03.31.715503

Figure Lengend Snippet: A) Schematic representation of the NMD pathway initiation and approaches for NMD inhibition by siRNA-knockdown (KD) of SMG1 (siSMG1) or UPF1 (siUPF1), or chemical inhibition of the SMG1 kinase (SMG1i). B) Quantification of SMG1 and UPF1 KD protein levels (relative to GAPDH) in n=6 TMB high cell lines 72h post-transfection with siSMG1 and siUPF1, respectively. Band densitometry analysis was done on western blots from , with comparisons by unpaired t-tests. C) Quantification of P-UPF1 protein levels (relative to UPF1/HSP90) in n=7 TMB high cell lines 24h post-treatment with SMG1i in western blots from , with comparisons by unpaired t-tests. * B and C include not only SCLC cell lines but also H841 and H1341, recently reclassified as SMARCA4-UT and small cell carcinoma of the cervix, respectively . D-E) Kinome assay for SMG1i treatment on 436 kinases showing the number of kinases (%) affected and the degree of inhibition ( D ), highlighting affected kinases at 100 nM ( E ). Colors in D-E indicate high (>70%), medium (50-70%) and low (<50%) relative kinase activity (see Table S2A and ). F) Pearson correlation of transcriptomic changes (bulk RNA-Seq, DESeq2 analysis for differential transcript expression (DTE)) in H841 cells following chemical NMD inhibition (0.5µM SMG1i for 24h) vs genetic KD (siRNA-mediated for 72h). Each dot represents the Log2FC expression of a single transcript in SMG1i vs DMSO (n=3, x -axis) against the Log2FC for siUPF1 or siSMG1 vs siCTRL (n=3, y -axis). G) UpSet plots for differential transcript expression (DTE) overlaps in H841 following chemical NMD inhibition with SMG1i and siRNA-mediated KD. For each treatment (siSMG1, siUPF1, SMG1i), numbers of differentially up/downregulated transcripts (Log2FC>|1|, Padj<0.05, n=3) are indicated by colored numbers in brackets (see Table S2D-E ). H) Number of PTC-containing transcripts (annotated as nonsense mediated decay or retained intron by GENCODE) significantly up/downregulated upon NMD inhibition (see Table S2B-C ). I) Proliferation assays for CTRL vs SMG1i-treated cell lines showing the number of cell divisions after 6 days based on the loss of the proliferation dye eFluor670 fluorescence intensity over time (see Methods and ). Comparisons by paired t-tests (n≥3). J-L) Cell death assays in human cell lines (hSCLC) monitored by live-cell imaging (DRAQ7 staining) for 4 days. Representative death curves shown for H146 cells treated with DMSO vs SMG1i and additional cell death inhibitors (QvD=apoptosis inhibitor; Fer1=ferroptosis inhibitor; Nec1s=necroptosis inhibitor) (J) ; the area under the curve (AUC) was quantified and compared by unpaired t-tests (n≥3), as shown exemplified for H146 (K) , with average normalized cell death plotted as a heatmap for n=7 human TMB High cell lines (L) . M-N) Cell death assays in DMSO vs SMG1i-treated murine TMB Low cell lines (mSCLC) monitored by live-cell imaging (DRAQ7 staining) for 4 days. Representative death curves shown for RP1380 cells (M) with AUC quantification compared by unpaired t-tests (n≥3) for n=3 mSCLC RP cell lines (see ) (N). O) SMG1i-induced cell death in n=3 TMB Low mSCLC RP cell lines vs n=7 TMB High hSCLC cell lines (from N and L , respectively) compared by Mann-Whitney test. P) Pearson correlation of SMG1i-induced cell death (from N and L ) with total number of exome mutations (see Table S5A ). Q) Sensitivity to SMG1i treatment of n=6 non-neoplastic (NORM) cell lines vs n=7 hSCLC cell lines compared by Mann-Whitney test (see ). R) Schematic representation for mSCLC cell lines derived from lung tumors arising in RP ( Rb1 fl/fl ;Trp53 fl/fl , TMB low ) and RP-Msh2 mice ( Rb1 fl/fl ;Trp53 fl/fl ;Msh2 fl/fl , TMB high ) following Adeno-CMV-Cre inhalation. S) Western blot for n=3 RP and n=3 RP-Msh2 cell lines representative cell lines compared to MEFs as control (C) probing for Rb1, p53 and MSH2, and confirming expression of the SCLC-specific NCAM1/CD56 140KDa-isoform . T-V) RP vs RP-Msh2 cell lines comparing total number of exome mutations ( T ), sensitivity to SMG1i treatment ( U ) (see ), and CTG-viability assays following 6 days of SMG1i treatment ( V ), by Mann-Whitney tests. For all graphs, error bars represent the SEM, and P-values are specified as ***P<0.001; **P <0.01; *P<0.05; ns=non-significant.

Article Snippet: For in vivo experiments in immunocompetent mice, C57BL/6J mice (Janvier) were injected subcutaneously with 1.5-3x10 6 RP1380 TetO-shCTRL or RP1380 TetO-shSMG1 cells per flank.

Techniques: Inhibition, Knockdown, Transfection, Western Blot, Activity Assay, RNA Sequencing, Expressing, Fluorescence, Live Cell Imaging, Staining, MANN-WHITNEY, Derivative Assay, Control

Journal: bioRxiv

Article Title: A targetable dependency on nonsense-mediated decay for proteostasis and immune control in small cell lung cancer

doi: 10.64898/2026.03.31.715503

Figure Lengend Snippet: A) Enzymatic in vitro phosphorylation assay for full-length recombinant SMG1 protein using p53-Ser15 as a phosphorylation substrate in the presence of increasing SMG1i concentrations. IC50 was calculated by non-linear adjustment of the kinase activity inhibitory curve. B-C) Kinome assay on 436 kinases for selected SMG1i concentrations showing the relative kinase activity (%) for each individual kinase upon NMD inhibition with 100 nM (B) and 1000 nM SMG1i (C) , with kinases showing medium (yellow) to low (red) activity highlighted (see Table S2A ). D) Western blot evaluation of the off-target inhibition of selected kinases in COR-L88 cells (NE High SCLC) treated with increasing concentrations of SMG1i, assessing for the respective kinase activity by probing for the phosphorylated downstream substrate as indicated in the scheme (upper panel). Quantifications based on band densitometry are provided (lower panels). Bars represent mean + SEM (n = 3); ***P<0.001; *P<0.05 (Unpaired t-tests vs untreated sample). E) On-target inhibition of NMD by SMG1i as shown by cell-based luciferase reporter assays in human SCLC (H1694), human SMARCA4-UT (H841) and murine SCLC (RP1380) cell lines transfected with the NMD-sensitive PTC-reporter and treated with increasing SMG1i concentrations. IC50 values were calculated by non-linear adjustment of NMD inhibitory curves (n = 3). F) Off-target inhibition of PIP5Kg by SMG1i as shown by NanoBRET assays, with IC50 value calculated by non-linear adjustment of PIP5Kg inhibitory curve, accompanied by a schematic representation of the NanoBRET assay principle.

Article Snippet: For in vivo experiments in immunocompetent mice, C57BL/6J mice (Janvier) were injected subcutaneously with 1.5-3x10 6 RP1380 TetO-shCTRL or RP1380 TetO-shSMG1 cells per flank.

Techniques: In Vitro, Phospho-proteomics, Recombinant, Activity Assay, Inhibition, Western Blot, Luciferase, Transfection

Journal: bioRxiv

Article Title: A targetable dependency on nonsense-mediated decay for proteostasis and immune control in small cell lung cancer

doi: 10.64898/2026.03.31.715503

Figure Lengend Snippet: A-D) Representative curves for cell growth (confluence ratio, A ) and cell death (DRAQ7 staining, C ) determined by live-cell imaging accompanied by respective AUC quantifications ( B and D , respectively) for individual experiments (n = 3) in several murine TMB low SCLC cell lines. All graphs represent mean + SEM. ***P < 0.001; **P < 0.01; *P < 0.05; ns = non-significant (Unpaired t-tests on AUC quantification). The three upper panels refer to chemical NMD inhibition treatment with DMSO vs SMG1i, and the two lower panels refer to DMSO vs doxycycline (DOXY) treatment of RP1380 TetO-shSMG1 cells for genetic NMD inhibition via SMG1-KD, and in RP1380 TetO-shCTRL cells as a control to account for doxycycline-associated effects. Doxycycline-inducible SMG1-KD cellular models are described in .

Article Snippet: For in vivo experiments in immunocompetent mice, C57BL/6J mice (Janvier) were injected subcutaneously with 1.5-3x10 6 RP1380 TetO-shCTRL or RP1380 TetO-shSMG1 cells per flank.

Techniques: Staining, Live Cell Imaging, Inhibition, Control

Journal: bioRxiv

Article Title: A targetable dependency on nonsense-mediated decay for proteostasis and immune control in small cell lung cancer

doi: 10.64898/2026.03.31.715503

Figure Lengend Snippet: A) Enzymatic in vitro phosphorylation assay for full-length recombinant SMG1 protein using p53-Ser15 as a phosphorylation substrate in the presence of increasing SMG1i and KVS0001 concentrations (see ). Absolute IC50 was calculated by non-linear adjustment of the inhibitory curve. B-F) In vivo pharmacological NMD inhibition treatment of NSG mice with KVS0001 monitoring (B) systemic toxicity through weight changes at 50mpk KVS0001 applied in a 5-days-on-2-day-off regimen, with each time point compared by unpaired t-tests (n=16); (C-D) pharmacokinetics (PK) after a single shot intraperitoneal (ip) injection of KVS0001 at the indicated concentrations (n=3 mice per timepoint), quantifying KVS0001 levels by mass spectrometry in plasma (C) and inside subcutaneous tumors (D) (see Methods and ) ; (E) intratumor NMD inhibition shown by qPCR for the NMD target TBL2 in control vs 50mpk KVS0001-treated tumors 2h post-injection, compared by Mann-Whitney test (n=6); (F) tumor growth for NSG mice treated with control vs KVS0001 in n=3 hSCLC and n=2 mSCLC xenograft models. Treatment was initiated at tumor sizes of 100mm 3 (indicated by “X”) and conducted on a 5-days-on-2-day-off regimen, Statistics represent unpaired t-tests on AUC (n≥4). G) Pearson correlation between total number of exome mutations and tumor growth control (fold change in end tumor size of control vs KVS0001 treatment from F ). H-I) In vitro CTG-viability assays for ( H ) TetO-SMG1 n=1 mSCLC RP and n=3 hSCLC cell lines following doxycycline-induced SMG1-KD for 6 days ( I ) and for chemical NMD inhibition with SMG1i for 6 days in n=2 mSCLC RP, n=2 mSCLC RP-Msh2, n=6 hSCLC cell lines and n=5 patient-derived tumor cells, with comparisons by Two-way ANOVA (n≥3). J) Schematic representation of tumor-targeted genetic NMD inhibition in vivo employing doxycycline-inducible SMG1-KD models (TetO-shSMG1 cell lines) grown as subcutaneous tumors in NSG mice fed with either normal or doxycycline-diet (see ). K) Tumor growth for NSG mice xenograft models for TetO-shSMG1 cell lines ( left ) and parental control cell lines ( right ) fed with normal or doxycycline-diet initiated at tumor sizes of 100mm 3 (indicated by “X”; note that for RP1380-TetO-shSMG1 model, doxycycline treatment started directly after tumor cell injection, to act as a proper control for experiments shown in and ). Statistics represent unpaired t-tests on AUC (n≥4). For all graphs, error bars represent the SEM, and P-values are specified as ***P<0.005; **P <0.01; *P<0.05; ns=non-significant.

Article Snippet: For in vivo experiments in immunocompetent mice, C57BL/6J mice (Janvier) were injected subcutaneously with 1.5-3x10 6 RP1380 TetO-shCTRL or RP1380 TetO-shSMG1 cells per flank.

Techniques: In Vitro, Phospho-proteomics, Recombinant, In Vivo, Inhibition, Drug discovery, Injection, Mass Spectrometry, Clinical Proteomics, Control, MANN-WHITNEY, Derivative Assay

Journal: bioRxiv

Article Title: A targetable dependency on nonsense-mediated decay for proteostasis and immune control in small cell lung cancer

doi: 10.64898/2026.03.31.715503

Figure Lengend Snippet: A) Western blot on the human COR-L88 SCLC cell line (NE High , ASCL1+) treated in vitro with increasing concentrations of the SMG1i-derivate compound KVS0001 for 24 h. B) Pharmacokinetic parameters for KVS0001 compound detected by mass spectrometry in plasma and tumors of NSG mice after a single intraperitoneal injection with 25 or 50 milligram per kg (mpk) at timepoints post-injection indicated in . C) Schematic representation for the doxycycline-inducible SMG1-KD operating within TetO-shSMG1 models, which include the murine SCLC cell line RP1380, the human SMARCA-UT cell line H841, the human NE High SCLC cell line GLC-1 and the human NE Low SCLC cell line H1048. D) Western blots for TetO-shSMG1 cell lines probing for downregulation of SMG1 and consequently decreased P-UPF1 levels upon adding 2 ug/ml doxycycline to the cell culture for 72h. For human TetO-shSMG1 cell lines, the corresponding parental cell lines served as control to account for the effects of doxycycline treatment. For the murine SCLC cell line RP1380, a TetO-shCTRL cell line was generated to exclude immunogenic effects of the TetO system in subsequent experiments shown in . E) Band densitometry quantification for SMG1 (relative to TUBG) and P-UPF1 (relative to UPF1/TUBG) protein levels in TetO-shSMG1 cell lines referring to the data in panel D . F) qPCR quantification of the bona fide NMD target TBL2 in human TetO-shSMG1 cell lines after 72h of in vitro doxycycline treatment. G) Scheme for in vivo induction of SMG1-KD through the administration of doxycycline-containing diet to either NSG or C57BL/6J mice subcutaneously implanted with TetO-shSMG1 cells. H) Representative blot for SMG1 protein levels from allograft RP1380 TetO-shSMG1 subcutaneous tumors derived from C57BL/6J mice fed with normal or doxycycline diet. I) Densitometry quantification of protein levels determined in western blots for data in panel H (n ≥ 14 tumors grown in vivo ). I) Mouse weight changes over time in mixed NSG and C57BL/6J mice receiving normal or doxycycline diet (n = 16).

Article Snippet: For in vivo experiments in immunocompetent mice, C57BL/6J mice (Janvier) were injected subcutaneously with 1.5-3x10 6 RP1380 TetO-shCTRL or RP1380 TetO-shSMG1 cells per flank.

Techniques: Western Blot, In Vitro, Mass Spectrometry, Clinical Proteomics, Injection, Cell Culture, Control, Generated, In Vivo, Derivative Assay

Journal: bioRxiv

Article Title: A targetable dependency on nonsense-mediated decay for proteostasis and immune control in small cell lung cancer

doi: 10.64898/2026.03.31.715503

Figure Lengend Snippet: A-D) In vitro co-cultures of the TMB High human cell line H841 with PBMCs (>80% T cells, ). (A) Representative images 4 days post-stimulation with indicated stimuli promoting naive, primed or activated T cell status; (B) flow cytometry histograms showing B2M and MHC-ABC surface expression in WT vs B2M -KO H841 cells ; (C) Western blot for WT and B2M -KO H841 cells showing SMG1-KD and UPF1-KD 72 hpt. (D) Tumor cell viability represented by the fraction of living single H841 cells (CD45-) measured by flow cytometry in 4 day-co-cultures with naive, primed or activated T cells. WT ( left ) or B2M -KO ( right ) H841 cells were transfected with siCTRL/siSMG1/siUPF1 and tumor killing under primed/activated conditions was normalized internally to the naive condition to subtract possible effects of NMD inhibition on cell-intrinsic viability. A subsequent re-normalization to the siCTRL condition is shown for primed co-cultures (n≥3). E) Tumor growth of RP1380 TetO-shSMG1 cell allografts growing in wild-type immunocompetent C57BL/6J mice fed with normal or doxycycline-diet (n=13). F) Flow cytometry immunophenotyping of endpoint tumors from E (n≥7) (see ). G) Tumor growth of RP1380 TetO-shSMG1 cells allografts growing in partially immunocompromised, T/B cell-deficient RAG1-KO C57BL/6J mice fed with normal or doxycycline-diet (n≥6). H) T cell quantification in WT C57BL/6J vs RAG1-KO C57BL/6J mice (n=4) (see ). I) Flow cytometry immunophenotyping of endpoint tumors from G (n≥6) . J) Tumor growth of RP1380 TetO-shSMG1 cell allografts grown in C57BL/6J mice fed with doxycycline-diet (SMG1-KD models) and treated with PBS or anti-PD-1 (see Mehods ). K-L) Flow cytometry showing infiltrating T cell numbers ( K ) and activation markers ( L ) in endpoint tumors from J (n≥6). M-N) Single cell TCR sequencing of sorted CD45+ cells derived from digested enpoint tumors from E , displaying number of identified T cells within all CD45+ cells (M) and number and distribution of T cell clonotypes in control vs NMD-inhibited tumors (N) (n=2, see Table S5H ). All statistics by unpaired t-tests (for E , G and K unpaired t-tests for AUC). For all graphs, error bars represent the SEM, and P-values are specified as ***P<0.001; **P <0.01; *P<0.05; ns=non-significant.

Article Snippet: For in vivo experiments in immunocompetent mice, C57BL/6J mice (Janvier) were injected subcutaneously with 1.5-3x10 6 RP1380 TetO-shCTRL or RP1380 TetO-shSMG1 cells per flank.

Techniques: In Vitro, Flow Cytometry, Expressing, Western Blot, Transfection, Inhibition, Activation Assay, Single Cell, Sequencing, Derivative Assay, Control

Journal: bioRxiv

Article Title: A targetable dependency on nonsense-mediated decay for proteostasis and immune control in small cell lung cancer

doi: 10.64898/2026.03.31.715503

Figure Lengend Snippet: A) DNA Sanger sequencing of the amplified B2M locus in representative H841 CRISPR/Cas9 WT and B2M-KO clones aligned to the reference sequence (NCBI, NG_012920.2) with sgRNA and the identified insertion mutation in H841 B2M-KO cells highlighted. B-D) Pharmacological NMD inhibition with either KVS0001 or SMG1i-11j compounds in healthy PBMC donors upon T cell artificial activation. ( B ) Proportion of viable CD45+ cells (left) and T cells (right) within PBMC populations after 4 days treatment with a T cell artificial activation cocktail (Act = CD3+CD28+IL-2) vs unstimulated conditions (Naive). ( C ) Proliferation of CD4+ (left) and CD8+ (right) T cells derived from T cell counts expressed as fold change (FC) for activated (Act) conditions relative to the unstimulated (Naive) control. ( D ) Proportion of T cells (CD3+), B cells (CD19+), NK cells (CD56+) and Myeloid cells (CD11b+) within PBMC populations (CD45+) from healthy donors directly after thawing (d-1), at the beginning of stimulation (CD3+CD28+IL-2) (d0) 4 days post-stimulation (d4). Graphs represent mean + SEM (n = 7). ***P<0,001; **P<0.01; *P<0.05; ns=non-significant (One-way ANOVA). E) Representative flow cytometry dot plots for the staining of CD45 vs CD56 (for NK cells) and CD8 vs CD4 (both for T cells) following 4 days co-culture of healthy donor PBMCs with WT H841 tumor cells. F) In vivo tumor growth in immunocompetent C57BL/6J mice transplanted with murine RP1380 TetO-shCTRL fed with normal or doxycycline-containing diet (n ≥ 4). G) Flow cytometry immunophenotyping of RP1380 TetO-shCTRL tumors harvested at the end of experiments shown in panel F (n ≥ 4). H-I) MHC-I surface expression quantified by flow cytometry in the indicated human and murine cell lines following genetic NMD inhibition via siRNA-mediated SMG1-KD and UPF1-KD ( H ) or doxycycline-inducible SMG1-KD ( I ). J) In vivo assessment of MHC-I (H2-Kb) surface expression in control vs NMD-inhibited murine RP1380 TetO-shSMG1 tumors (following doxycycline diet, DOXY) grown subcutaneously in C57BL/6 mice. K) Flow cytometry data for murine RP1380 TetO-shSMG1 allograft models grown in C57BL/6 mice significantly correlating tumor-specific H2-Kb surface expression with levels of immune cell infiltration (CD45+ infiltration). L) In vivo levels of tumor-specific MHC-I (H2-Kb) surface expression in control vs NMD-inhibited RP1380 TetO-shSMG1 tumors (DOXY) grown subcutaneously in RAG1-KO C57BL/6 mice. K) Flow cytometry data for murine RP1380 TetO-shSMG1 allograft models grown in RAG1-KO C57BL/6 mice correlating tumor-specific H2-Kb surface expression with levels of immune cell (CD45+) infiltration. L) Representative flow cytometry contour plot showing T cells (CD45+/CD3+) and B cells (CD45+/CD19+) in the blood from WT C57BL/6 and RAG1-KO C57BL/6 mice.

Article Snippet: For in vivo experiments in immunocompetent mice, C57BL/6J mice (Janvier) were injected subcutaneously with 1.5-3x10 6 RP1380 TetO-shCTRL or RP1380 TetO-shSMG1 cells per flank.

Techniques: Sequencing, Amplification, CRISPR, Clone Assay, Mutagenesis, Inhibition, Activation Assay, Derivative Assay, Control, Flow Cytometry, Staining, Co-Culture Assay, In Vivo, Expressing