CoChat

Pharmacological inhibition of LSD1 triggers myeloid differentiation by targeting GSE1 oncogenic functions in AML

Open access

The DOI matches its Crossref record; 72 of the 72 references it cites were found in a registry or an index.

Abstract

The histone demethylase LSD1 is over-expressed in hematological tumors and has emerged as a promising target for anticancer treatment, so that several LSD1 inhibitors are under development and testing, in preclinical and clinical settings. However, the complete understanding of their complex mechanism of action is still unreached. Here, we unraveled a novel mode of action of the LSD1 inhibitors MC2580 and DDP-38003, showing that they can induce differentiation of AML cells through the downregulation of the chromatin protein GSE1. Analysis of the phenotypic effects of GSE1 depletion in NB4 cells showed a strong decrease of cell viability in vitro and of tumor growth in vivo. Mechanistically, we found that a set of genes associated with immune response and cytokine-signaling pathways are upregulated by LSD1 inhibitors through GSE1-protein reduction and that LSD1 and GSE1 colocalize at promoters of a subset of these genes at the basal state, enforcing their transcriptional silencing. Moreover, we show that LSD1 inhibitors lead to the reduced binding of GSE1 to these promoters, activating transcriptional programs that trigger myeloid differentiation. Our study offers new insights into GSE1 as a novel therapeutic target for AML.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1038/s41388-021-02123-7

Cites

72 of 72 references verified0 citation errors · 0 not checkable

Statistics

Figures

Fig. 1 LSD1 inhibitors alter LSD1–GSE1 interaction by reducing GSE1 protein levels in AML. A Experimental design of the SILAC/LSD1 co-IP strategy setup to study the changes in LSD1 interactome upon 24 h treatment with 2 μΜ of MC2580, as described in ref. [13]. B Volcano plot displaying the alteration in the LSD1 interactome upon MC2580 treatment. Significantly evicted and recruited proteins are marked as red dots, while the proteins previously defined as specific LSD1 binders [13] are indicated by red triangles. The light-blue dashed line indicates the p-value threshold used to define the interactors modulated by the LSD1 inhibitors (p value < 0.05 calculated with limma [72]). Analysis of the evicted and recruited LSD1 interactors after pharmacological treatment is performed using the Differential Enrichment analysis of Proteomics data (DEP) R software package [40]. C Upper panel: MS1 spectrum of the SILAC peak doublet for the peptide 897-ALSAAVADSLTNSPR- 911 of GSE1 protein in the SILAC-LSD1 co-immuno-precipitated (co-IP) sample from NB4 cells treated for 24 h with 2 μΜ of MC2580 (heavy peak) and DMSO-treated control cells (light peak). Lower panel: Table displaying the H/L (forward replicate) and L/H (reverse replicate) ratios for GSE1 in SILAC-LSD1 co-IPs of NB4 cells treated with the compound MC2580 or DMSO. The table also shows the ratio count (RC) of each replicate, which corresponds to the number of peptides used for SILAC-based protein quantitation. The inhibitor was added to the heavy channel in the forward replicate and the light channel in the reverse one. D Western blot analysis of GSE1, LSD1, and H3 in LSD1 co-IPs (left panel) and GSE1 co-IPs (right panel), both in control and MC2580-treated cells (2 μΜ) for 24 h. Rabbit IgG was used as negative control co-IP. E Western blot analysis of GSE1, LSD1, GFI1, and H3 in in vitro LSD1 co-IPs using as input NB4 nuclear cell extract co-incubated for 6 h with increasing doses of MC2580 (2, 10 and 20 μM) and DMSO as control. F Left panel: Western blot analysis of GSE1 in NB4 cells treated with MC2580 (2 µM) DDP-38003 (2 µM) and control DMSO for 12 and 24 h. Vinculin was used as loading control. Right panel: Bar graph displaying the quantitation results of GSE1 protein, normalized over the Vinculin, in three independent replicates of NB4 cells treated with MC2580 (2 µM), DDP-38003 (2 µM), and DMSO. The results are plotted as fold change (FC) of GSE1 protein level in the treated samples over control (DMSO). The chart represents mean ± standard deviation (SD) (n = 3 biological replicates; one-sample T-test, **p value < 0.01, *p-value < 0.05). G RT-qPCR analysis on GSE1 transcript in NB4 cells treated for 12 and 24 h with MC2580 (2 µM), DDP-38003 (2 µM), and DMSO as control. GSE1 ct values are normalized against GAPDH. The results are plotted as FC of GSE1 mRNA levels in LSD1 inhibitors-treated conditions over DMSO. The chart represents mean ± SD of three (n = 3) biological replicates. Primer list in Table S7. H Upper panel: Representative Western blot analysis of GSE1 and c-MYC in NB4 cells treated with cycloheximide (0.1 mg/ml) in combination with MC2580 (2 µM) and DMSO as control for 3, 6, 9, and 12 h. Vinculin was used as loading control. Lower panel: Line-plot displaying the percentage of GSE1 protein level normalized over Vinculin in cycloheximide-treated NB4 cells, together with 2 µM of MC2580 or DMSO as control. The chart represents mean ± SD (n = 2 biological replicates). I Left panel: Western blot analysis of GSE1 and LSD1 in NB4 WT, NB4 LSD1-KO, and LSD1-KO cells transduced with either a retroviral empty PINCO vector (EV) or a PINCO vector containing an exogenous, N-terminal truncated (172–833) form of LSD1 [13]. Vinculin was used as loading control. Arrows indicate the endogenous and the exogenous LSD1. Right panel: Bar graph displaying quantitation of GSE1 protein level, normalized over Vinculin level in three replicates of NB4 WT, NB4 LSD1-KO, and LSD1-KO cells, transduced with either an empty PINCO vector (EV) or a PINCO vector containing an exogenous, N-terminal truncated (172–833) form of LSD1 [13]. In all samples, the data are plotted as FC of GSE1 protein level compared with the NB4 WT, with mean ± SD from n = 3 biological replicates (one-sample T-test, **p value < 0.01, ***p-value < 0.001).
Fig. 1 LSD1 inhibitors alter LSD1–GSE1 interaction by reducing GSE1 protein levels in AML. A Experimental design of the SILAC/LSD1 co-IP strategy setup to study the changes in LSD1 interactome upon 24 h treatment with 2 μΜ of MC2580, as described in ref. [13]. B Volcano plot displaying the alteration in the LSD1 interactome upon MC2580 treatment. Significantly evicted and recruited proteins are marked as red dots, while the proteins previously defined as specific LSD1 binders [13] are indicated by red triangles. The light-blue dashed line indicates the p-value threshold used to define the interactors modulated by the LSD1 inhibitors (p value < 0.05 calculated with limma [72]). Analysis of the evicted and recruited LSD1 interactors after pharmacological treatment is performed using the Differential Enrichment analysis of Proteomics data (DEP) R software package [40]. C Upper panel: MS1 spectrum of the SILAC peak doublet for the peptide 897-ALSAAVADSLTNSPR- 911 of GSE1 protein in the SILAC-LSD1 co-immuno-precipitated (co-IP) sample from NB4 cells treated for 24 h with 2 μΜ of MC2580 (heavy peak) and DMSO-treated control cells (light peak). Lower panel: Table displaying the H/L (forward replicate) and L/H (reverse replicate) ratios for GSE1 in SILAC-LSD1 co-IPs of NB4 cells treated with the compound MC2580 or DMSO. The table also shows the ratio count (RC) of each replicate, which corresponds to the number of peptides used for SILAC-based protein quantitation. The inhibitor was added to the heavy channel in the forward replicate and the light channel in the reverse one. D Western blot analysis of GSE1, LSD1, and H3 in LSD1 co-IPs (left panel) and GSE1 co-IPs (right panel), both in control and MC2580-treated cells (2 μΜ) for 24 h. Rabbit IgG was used as negative control co-IP. E Western blot analysis of GSE1, LSD1, GFI1, and H3 in in vitro LSD1 co-IPs using as input NB4 nuclear cell extract co-incubated for 6 h with increasing doses of MC2580 (2, 10 and 20 μM) and DMSO as control. F Left panel: Western blot analysis of GSE1 in NB4 cells treated with MC2580 (2 µM) DDP-38003 (2 µM) and control DMSO for 12 and 24 h. Vinculin was used as loading control. Right panel: Bar graph displaying the quantitation results of GSE1 protein, normalized over the Vinculin, in three independent replicates of NB4 cells treated with MC2580 (2 µM), DDP-38003 (2 µM), and DMSO. The results are plotted as fold change (FC) of GSE1 protein level in the treated samples over control (DMSO). The chart represents mean ± standard deviation (SD) (n = 3 biological replicates; one-sample T-test, **p value < 0.01, *p-value < 0.05). G RT-qPCR analysis on GSE1 transcript in NB4 cells treated for 12 and 24 h with MC2580 (2 µM), DDP-38003 (2 µM), and DMSO as control. GSE1 ct values are normalized against GAPDH. The results are plotted as FC of GSE1 mRNA levels in LSD1 inhibitors-treated conditions over DMSO. The chart represents mean ± SD of three (n = 3) biological replicates. Primer list in Table S7. H Upper panel: Representative Western blot analysis of GSE1 and c-MYC in NB4 cells treated with cycloheximide (0.1 mg/ml) in combination with MC2580 (2 µM) and DMSO as control for 3, 6, 9, and 12 h. Vinculin was used as loading control. Lower panel: Line-plot displaying the percentage of GSE1 protein level normalized over Vinculin in cycloheximide-treated NB4 cells, together with 2 µM of MC2580 or DMSO as control. The chart represents mean ± SD (n = 2 biological replicates). I Left panel: Western blot analysis of GSE1 and LSD1 in NB4 WT, NB4 LSD1-KO, and LSD1-KO cells transduced with either a retroviral empty PINCO vector (EV) or a PINCO vector containing an exogenous, N-terminal truncated (172–833) form of LSD1 [13]. Vinculin was used as loading control. Arrows indicate the endogenous and the exogenous LSD1. Right panel: Bar graph displaying quantitation of GSE1 protein level, normalized over Vinculin level in three replicates of NB4 WT, NB4 LSD1-KO, and LSD1-KO cells, transduced with either an empty PINCO vector (EV) or a PINCO vector containing an exogenous, N-terminal truncated (172–833) form of LSD1 [13]. In all samples, the data are plotted as FC of GSE1 protein level compared with the NB4 WT, with mean ± SD from n = 3 biological replicates (one-sample T-test, **p value < 0.01, ***p-value < 0.001).CC BY · PubMed Central
Fig. 2 GSE1 knockdown (KD) affects NB4 cell viability and in vivo tumor growth in NOD/SCID gamma (NSG) mice. A Left panel: Western blot analysis of GSE1 protein in NB4 cells transduced with either an empty lentiviral pLKO.1 puro vector (EV) or pLKO.1 puro containing the shRNA constructs targeting GSE1 (shA1 and shB2) for 72 hours. Vinculin was used as loading control. Right panel: Bar graph displaying the quantitation of GSE1 protein level normalized over the Vinculin in three independent replicates (n = 3), whereby the results are plotted as fold change (FC) of GSE1 level in the knockdown (KD) samples over the control EV. The chart represents mean ± SD (n = 3 biological replicates; one-sample T-test, *p value < 0.05). B Growth curve of NB4 cells transduced with either an empty pLKO.1 puro vector (EV) or pLKO.1 puro containing the shA1 and shB2 targeting GSE1. Graph represents mean ± SD from two biological replicates (n = 2). C Bar graph displaying the percentage of dead cells in control EV and GSE1 KD cells. The analysis is performed after 48, 72, 96, 120, and 144 hours of infection. Cell viability was measured by trypan blue staining. The chart represents mean ± SD from two (n = 2) biological replicates. D Western blot analysis of GSE1, cleaved caspase-3, and total caspase-3 in cells transduced with either the control pLKO.1 puro (EV) or pLKO.1 containing the two shRNAs targeting GSE1 (shA1 and shB2) for 72 and 96 h. Vinculin was used as loading control. The arrow indicates the cleaved form of the caspase-3. E Left panel: Contour-plot representation of multiparametric analysis, with propidium iodide (PI) at x axis and AlexaFluor®488 for BrdU at y axis, of BrdU incorporation in NB4 GSE1 KD and control EV cells, at 48- and 72 h post infection. Right panel: Bar graph displaying the percentage of cells at each cell cycle phase, extrapolated from the BrdU incorporation assay. The chart represents mean ± SD from two (n = 2) biological replicates. F Tumor-growth curve in NSG mice transplanted subcutaneously with NB4 cells transduced 24 h earlier with either an empty pLKO.1 puro or pLKO.1 puro with the insert of the shRNAs (shA1 and shB2) targeting GSE1 until day22 (when all mice transplanted with EV-transduced cells were dead). Graph represents mean ± SD of the tumor volume in each condition (n = 8, 8 distinct mice for each group; unpaired T-test, **p-value < 0.01 calculated at day 22). G Kaplan–Meier survival curve of mice transplanted with NB4 cells previously transduced with either an empty pLKO.1 vector or pLKO.1 containing shA1 or shB2 targeting GSE1 (n = 8 for each group, log-rank Mantel–Cox test, ****p-value < 0.0001).
Fig. 2 GSE1 knockdown (KD) affects NB4 cell viability and in vivo tumor growth in NOD/SCID gamma (NSG) mice. A Left panel: Western blot analysis of GSE1 protein in NB4 cells transduced with either an empty lentiviral pLKO.1 puro vector (EV) or pLKO.1 puro containing the shRNA constructs targeting GSE1 (shA1 and shB2) for 72 hours. Vinculin was used as loading control. Right panel: Bar graph displaying the quantitation of GSE1 protein level normalized over the Vinculin in three independent replicates (n = 3), whereby the results are plotted as fold change (FC) of GSE1 level in the knockdown (KD) samples over the control EV. The chart represents mean ± SD (n = 3 biological replicates; one-sample T-test, *p value < 0.05). B Growth curve of NB4 cells transduced with either an empty pLKO.1 puro vector (EV) or pLKO.1 puro containing the shA1 and shB2 targeting GSE1. Graph represents mean ± SD from two biological replicates (n = 2). C Bar graph displaying the percentage of dead cells in control EV and GSE1 KD cells. The analysis is performed after 48, 72, 96, 120, and 144 hours of infection. Cell viability was measured by trypan blue staining. The chart represents mean ± SD from two (n = 2) biological replicates. D Western blot analysis of GSE1, cleaved caspase-3, and total caspase-3 in cells transduced with either the control pLKO.1 puro (EV) or pLKO.1 containing the two shRNAs targeting GSE1 (shA1 and shB2) for 72 and 96 h. Vinculin was used as loading control. The arrow indicates the cleaved form of the caspase-3. E Left panel: Contour-plot representation of multiparametric analysis, with propidium iodide (PI) at x axis and AlexaFluor®488 for BrdU at y axis, of BrdU incorporation in NB4 GSE1 KD and control EV cells, at 48- and 72 h post infection. Right panel: Bar graph displaying the percentage of cells at each cell cycle phase, extrapolated from the BrdU incorporation assay. The chart represents mean ± SD from two (n = 2) biological replicates. F Tumor-growth curve in NSG mice transplanted subcutaneously with NB4 cells transduced 24 h earlier with either an empty pLKO.1 puro or pLKO.1 puro with the insert of the shRNAs (shA1 and shB2) targeting GSE1 until day22 (when all mice transplanted with EV-transduced cells were dead). Graph represents mean ± SD of the tumor volume in each condition (n = 8, 8 distinct mice for each group; unpaired T-test, **p-value < 0.01 calculated at day 22). G Kaplan–Meier survival curve of mice transplanted with NB4 cells previously transduced with either an empty pLKO.1 vector or pLKO.1 containing shA1 or shB2 targeting GSE1 (n = 8 for each group, log-rank Mantel–Cox test, ****p-value < 0.0001).CC BY · PubMed Central
Fig. 3 GSE1 downregulation elicited by LSD1 inhibitors induces transcription of genes associated with cytokine-mediated signaling and immune response pathways. A Volcano plot displaying up- and downregulated genes upon 48 h of transduction with pLKO.1 vector containing shA1 and shB2 inserts. The x axis shows the log2 fold-change (FC) values of each gene in the shRNAs-transduced cells compared with control empty vector (EV)-transduced, while the y axis displays the −log10-adjusted p-values (p-adj). FC value is calculated with DEseq2 program [59], using two biological replicates for each condition (n = 2). B Venn diagrams with the number of individual and overlapping upregulated and downregulated genes identified 48 h after transduction with shA1 and shB2 constructs. C Voronoi tree map displaying the statistically significant GO biological process (BP) terms associated with the common 422 upregulated genes upon GSE1 KD with the two shRNA constructs. GO analysis and calculation of the statistically significant BP was performed through EnrichR [60] (adjusted p-value < 0.05), and then BP was grouped using Revigo [46]. Voronoi plot includes the BP terms mapped to a high hierarchical level. The tassel size corresponds to the −log10 p-value of the enrichment while the color intensity to the number of genes belonging to each category. D Unsupervised hierarchical clustering of all significantly upregulated genes upon MC2580 treatment (p-adj < 0.01, log2 FC > 1). Heatmap displays the log2 FC values of these genes in response to both GSE1 KD and MC2580 treatment in NB4 cells. Each row in the heatmap represents an individual gene. Log2 FC values of each gene are plotted as the average of two independent biological replicates, for each condition (n = 2). Genes significantly upregulated by both shA1 and shB2 (p-adj < 0.01, log2 FC > 1) are flagged in red. Below is shown the pie-chart indicating the percentage of the 109 genes upregulated by MC2580 that are also significantly induced by GSE1 KD with both shRNAs. E Bubble-plot displaying the statistically significant IPA pathways (−log10 p-value > 2) obtained from the analysis of the differentially expressed genes (DEG) after MC2580 treatment and comparison with the IPA results achieved from the DEG upon GSE1 KD. The bubble size reflects the number of genes belonging to that pathway, while colors indicate the −log10 p-value associated with each IPA term. Nonsignificant IPA pathway terms are colored in white. IPA pathways are clustered manually, according to specific biological processes. Some IPA pathway terms are not displayed in the current figure panel, but the complete list is in Table S4. F RT-qPCR analysis on a panel of genes in MLL- translocated BB104 primary patient cells treated for 24 h with MC2580 (2 µM), DDP-38003 (2 µM) and DMSO as control. Ct values of different genes are normalized against β-Actin. The results are plotted as FC of the mRNA levels in the treated conditions compared with DMSO. The chart represents mean ± SD (n = 3 biological replicates, one sample T-test, **p-value < 0.01, *p-value < 0.05). Primers list in Table S7. G Left panel: Western blot analysis of GSE1 protein in MLL- translocated BB104 primary patient cells transduced with either a nontargeting shRNA construct (NTC) cloned in pLKO.1 puro vector or pLKO.1 puro containing the shRNA constructs targeting GSE1 (shA1 and shB2) for 72 hours. Vinculin was used as loading control. Right panel: RT-qPCR analysis on a panel of genes at 48- and 72-h post transduction with shA1, shB2 and NTC as negative control in MLL- translocated BB104 primary patient cells. Ct values of the different genes are normalized over β-Actin. The results are plotted as FC of the mRNA levels in the GSE1 KD samples, over the corresponding levels in the NTC-transduced cells. Bar graph represents mean ± SD (n = 2 biological replicates). Primer list in Table S7. H RT-qPCR analysis on a panel of genes upregulated by both LSD1 inhibitors and GSE1 KD, in control and GSE1-over-expressing (OE) NB4 cells treated with 2 µM of MC2580 treatment. GSE1 OE cells are generated by lentiviral transduction with the pLEX_307 vector containing the GSE1 coding sequence (NM_001134473.3), while control cells are transduced with the empty pLEX_307 (pLEX_307 EV). Ct values for each gene are normalized against GAPDH. The results are plotted as FC of the mRNA levels in the treated conditions relative to the control DMSO. Bar graph represents mean ± SD from three (n = 3) biological replicates; paired T-test, *p-value < 0.05. Primer list in Table S7.
Fig. 3 GSE1 downregulation elicited by LSD1 inhibitors induces transcription of genes associated with cytokine-mediated signaling and immune response pathways. A Volcano plot displaying up- and downregulated genes upon 48 h of transduction with pLKO.1 vector containing shA1 and shB2 inserts. The x axis shows the log2 fold-change (FC) values of each gene in the shRNAs-transduced cells compared with control empty vector (EV)-transduced, while the y axis displays the −log10-adjusted p-values (p-adj). FC value is calculated with DEseq2 program [59], using two biological replicates for each condition (n = 2). B Venn diagrams with the number of individual and overlapping upregulated and downregulated genes identified 48 h after transduction with shA1 and shB2 constructs. C Voronoi tree map displaying the statistically significant GO biological process (BP) terms associated with the common 422 upregulated genes upon GSE1 KD with the two shRNA constructs. GO analysis and calculation of the statistically significant BP was performed through EnrichR [60] (adjusted p-value < 0.05), and then BP was grouped using Revigo [46]. Voronoi plot includes the BP terms mapped to a high hierarchical level. The tassel size corresponds to the −log10 p-value of the enrichment while the color intensity to the number of genes belonging to each category. D Unsupervised hierarchical clustering of all significantly upregulated genes upon MC2580 treatment (p-adj < 0.01, log2 FC > 1). Heatmap displays the log2 FC values of these genes in response to both GSE1 KD and MC2580 treatment in NB4 cells. Each row in the heatmap represents an individual gene. Log2 FC values of each gene are plotted as the average of two independent biological replicates, for each condition (n = 2). Genes significantly upregulated by both shA1 and shB2 (p-adj < 0.01, log2 FC > 1) are flagged in red. Below is shown the pie-chart indicating the percentage of the 109 genes upregulated by MC2580 that are also significantly induced by GSE1 KD with both shRNAs. E Bubble-plot displaying the statistically significant IPA pathways (−log10 p-value > 2) obtained from the analysis of the differentially expressed genes (DEG) after MC2580 treatment and comparison with the IPA results achieved from the DEG upon GSE1 KD. The bubble size reflects the number of genes belonging to that pathway, while colors indicate the −log10 p-value associated with each IPA term. Nonsignificant IPA pathway terms are colored in white. IPA pathways are clustered manually, according to specific biological processes. Some IPA pathway terms are not displayed in the current figure panel, but the complete list is in Table S4. F RT-qPCR analysis on a panel of genes in MLL- translocated BB104 primary patient cells treated for 24 h with MC2580 (2 µM), DDP-38003 (2 µM) and DMSO as control. Ct values of different genes are normalized against β-Actin. The results are plotted as FC of the mRNA levels in the treated conditions compared with DMSO. The chart represents mean ± SD (n = 3 biological replicates, one sample T-test, **p-value < 0.01, *p-value < 0.05). Primers list in Table S7. G Left panel: Western blot analysis of GSE1 protein in MLL- translocated BB104 primary patient cells transduced with either a nontargeting shRNA construct (NTC) cloned in pLKO.1 puro vector or pLKO.1 puro containing the shRNA constructs targeting GSE1 (shA1 and shB2) for 72 hours. Vinculin was used as loading control. Right panel: RT-qPCR analysis on a panel of genes at 48- and 72-h post transduction with shA1, shB2 and NTC as negative control in MLL- translocated BB104 primary patient cells. Ct values of the different genes are normalized over β-Actin. The results are plotted as FC of the mRNA levels in the GSE1 KD samples, over the corresponding levels in the NTC-transduced cells. Bar graph represents mean ± SD (n = 2 biological replicates). Primer list in Table S7. H RT-qPCR analysis on a panel of genes upregulated by both LSD1 inhibitors and GSE1 KD, in control and GSE1-over-expressing (OE) NB4 cells treated with 2 µM of MC2580 treatment. GSE1 OE cells are generated by lentiviral transduction with the pLEX_307 vector containing the GSE1 coding sequence (NM_001134473.3), while control cells are transduced with the empty pLEX_307 (pLEX_307 EV). Ct values for each gene are normalized against GAPDH. The results are plotted as FC of the mRNA levels in the treated conditions relative to the control DMSO. Bar graph represents mean ± SD from three (n = 3) biological replicates; paired T-test, *p-value < 0.05. Primer list in Table S7.CC BY · PubMed Central
Fig. 4 MC2580 decreases GSE1 localization on the LSD1-bound promoters of genes involved in cytokine-mediated signaling and immune-response pathways. A Heatmap representing the normalized V5-GSE1 ChIP-seq intensities ±5 kb around the center of the GSE1 target loci, at basal state and after 24 h of treatment with MC2580 (2 μM). V5-GSE1 ChIP-seq was performed in NB4 GSE1 OE cells produced by lentiviral transduction of the pLEX_307 construct containing the GSE1 coding sequence (NM_001134473.3) (pLEX_307 GSE1), while the negative control ChIP was carried out in cells transduced with the empty pLEX_307 (pLEX_307 EV). B Bar graph displaying the number of peaks extrapolated from the V5-GSE1 ChIP-seq in DMSO and MC2580-treated cells. C Genome annotation of the V5-GSE1 ChIP-seq peaks in control DMSO and LSD1-inhibited NB4 GSE1 OE cells. D Heatmap displaying the normalized V5-GSE1, LSD1, and H3K4me3 ChIP-seq intensities ±5 kb around the center of the GSE1 target loci after 24 h of treatment with MC2580 or DMSO as control. The heatmap is grouped in different subgroups, according to the presence or absence or differential intensity of the GSE1 binding regions in the two conditions. The subset of “Common GSE1 targets with reduced intensity in DMSO” contains only 32 regions and is not displayed in the figure. The LSD1 and H3K4me3 ChIP-seq data used to generate the heatmap were published in ref. [13]. E Heatmap displaying the V5-GSE1, LSD1, and H3K4me3 ChIP-seq intensities at the promoters of genes upregulated after LSD1 inhibition, which show reduced binding of V5-GSE1 in MC2580-treated cells. Each row in the heatmap represents an individual gene. For the V5-GSE1 ChIP-seq, read intensities in regions identified as peaks and annotated to each gene are plotted as log2 RPKM. Instead, for the LSD1 and H3K4me3 ChIP-seq, read coverage (log2 RPKM) at promoter (±2.5 kb distance from the TSS) was represented. For each gene, its chromatin state is shown as referred in Fig. 4D, F) V5-GSE1 ChIP-qPCR profiling of a panel of GSE1-bound promoters of genes significantly upregulated by LSD1 inhibition in DMSO and MC2580 (2 μM)-treated NB4 (upper panel) and THP-1 (lower panel) cells overexpressing GSE1 by pLEX_307 GSE1 construct transduction. V5 ChIP in cells transduced with the pLEX_307 EV was used as negative control ChIP. NEG indicates a negative control region. Data are normalized over the respective input and displayed as fold change (FC) over the control EV. Bar graph represents mean ±standard deviation (SD) from two (n = 2) biological replicates. Primers list in Table S7.
Fig. 4 MC2580 decreases GSE1 localization on the LSD1-bound promoters of genes involved in cytokine-mediated signaling and immune-response pathways. A Heatmap representing the normalized V5-GSE1 ChIP-seq intensities ±5 kb around the center of the GSE1 target loci, at basal state and after 24 h of treatment with MC2580 (2 μM). V5-GSE1 ChIP-seq was performed in NB4 GSE1 OE cells produced by lentiviral transduction of the pLEX_307 construct containing the GSE1 coding sequence (NM_001134473.3) (pLEX_307 GSE1), while the negative control ChIP was carried out in cells transduced with the empty pLEX_307 (pLEX_307 EV). B Bar graph displaying the number of peaks extrapolated from the V5-GSE1 ChIP-seq in DMSO and MC2580-treated cells. C Genome annotation of the V5-GSE1 ChIP-seq peaks in control DMSO and LSD1-inhibited NB4 GSE1 OE cells. D Heatmap displaying the normalized V5-GSE1, LSD1, and H3K4me3 ChIP-seq intensities ±5 kb around the center of the GSE1 target loci after 24 h of treatment with MC2580 or DMSO as control. The heatmap is grouped in different subgroups, according to the presence or absence or differential intensity of the GSE1 binding regions in the two conditions. The subset of “Common GSE1 targets with reduced intensity in DMSO” contains only 32 regions and is not displayed in the figure. The LSD1 and H3K4me3 ChIP-seq data used to generate the heatmap were published in ref. [13]. E Heatmap displaying the V5-GSE1, LSD1, and H3K4me3 ChIP-seq intensities at the promoters of genes upregulated after LSD1 inhibition, which show reduced binding of V5-GSE1 in MC2580-treated cells. Each row in the heatmap represents an individual gene. For the V5-GSE1 ChIP-seq, read intensities in regions identified as peaks and annotated to each gene are plotted as log2 RPKM. Instead, for the LSD1 and H3K4me3 ChIP-seq, read coverage (log2 RPKM) at promoter (±2.5 kb distance from the TSS) was represented. For each gene, its chromatin state is shown as referred in Fig. 4D, F) V5-GSE1 ChIP-qPCR profiling of a panel of GSE1-bound promoters of genes significantly upregulated by LSD1 inhibition in DMSO and MC2580 (2 μM)-treated NB4 (upper panel) and THP-1 (lower panel) cells overexpressing GSE1 by pLEX_307 GSE1 construct transduction. V5 ChIP in cells transduced with the pLEX_307 EV was used as negative control ChIP. NEG indicates a negative control region. Data are normalized over the respective input and displayed as fold change (FC) over the control EV. Bar graph represents mean ±standard deviation (SD) from two (n = 2) biological replicates. Primers list in Table S7.CC BY · PubMed Central
Fig. 5 GSE1 downregulation contributes to the myeloid differentiation reactivation enforced by LSD1 inhibition. A Percentage of NB4 CD11b-positive cells measured by flow cytometry 48 h post transduction with the shA1 and shB2 cloned in pLKO.1 puro, or with the empty pLKO.1 puro vector (EV) used as negative control. Dot plots show the results obtained in one of the two biological replicates of the experiment. The bar chart represents mean ± SD (n = 2 biological replicates). B Left panel: Percentage of CD11b- and CD14- positive THP-1 cells, assessed by flow cytometry analysis 48 h upon transduction with pLKO.1 puro vector containing shA1, shB2 or EV as negative control. Dot plots display the data of one of the two biological replicates of the experiment. The bar chart represents the mean + SD from two (n = 2) biological replicates. Right panel: Bar graph of the fold change (FC) of CD14 mean fluorescence intensity in THP-1 cells transduced for 48 h with shA1 and shB2, over the corresponding level in the EV-transduced cells. The chart represents mean + SD (n = 2 biological replicates). C Left panel: Representative dot plots showing the levels of CD11b- and CD86- positive cells as well as the intensity of these markers 48 h and 72 h post transduction with the shA1 and shB2 constructs targeting GSE1, or a nonspecific shRNA construct (NTC) cloned in pLKO.1 puro vector as negative control in the BB104 MLL- translocated primary patient sample. Right panel: Bar graph displaying the mean fluorescence intensity of CD11b, CD86 and CD14 in BB104 primary patient cells transduced for 48 h and 72 h with NTC, shA1 and shB2. The results are plotted as FC of the mean fluorescence intensity of these markers in the GSE1 KD samples over the corresponding levels in the NTC-transduced cells. The chart represents mean ± SD (n = 2 biological replicates). D Percentage of BB104 CD14-positive cells measured by flow cytometry at 48- and 72-h post transduction with the shA1 and shB2 cloned in pLKO.1 puro, or with the NTC used as negative control. Data are plotted as FC of the CD14 mean fluorescence intensity in the GSE1 KD samples over the corresponding levels in the NTC-infected cells. The chart represents mean ± SD (n = 2 biological replicates). E Percentage of CD11b-positive cells evaluated by flow cytometry in control (EV) and NB4 GSE1 OE cells, treated for 24 h with MC2580 (2 μM). NB4 GSE1 OE is produced by transducing the pLEX_307 vector with the GSE1 coding sequence (NM_001134473.3), while control cells are transduced with the pLEX_307 EV. The chart represents mean ± SD (n = 4 biological replicates; paired T-test, **p-value < 0.01). F Left panel: Flow cytometry analysis of the percentage of CD11b- and CD14- positive cells in EV and GSE1 OE THP-1 cells treated for 24 hours with MC2580. THP-1 GSE1 OE is generated by lentiviral transduction of the pLEX_307 vector including GSE1 coding sequence (NM_001134473.3), while control cells are transduced with the pLEX_307 EV. The chart represents mean ± SD (n = 3 biological replicates; paired T-test, *p-value < 0.05). Right panel: Bar graph displaying the CD14 mean fluorescence intensity in DMSO, MC2580 and MC2580 + GSE1 OE THP-1 cells. The chart represents mean ± SD from three (n = 3) biological replicates (paired T-test, *p-value < 0.05).
Fig. 5 GSE1 downregulation contributes to the myeloid differentiation reactivation enforced by LSD1 inhibition. A Percentage of NB4 CD11b-positive cells measured by flow cytometry 48 h post transduction with the shA1 and shB2 cloned in pLKO.1 puro, or with the empty pLKO.1 puro vector (EV) used as negative control. Dot plots show the results obtained in one of the two biological replicates of the experiment. The bar chart represents mean ± SD (n = 2 biological replicates). B Left panel: Percentage of CD11b- and CD14- positive THP-1 cells, assessed by flow cytometry analysis 48 h upon transduction with pLKO.1 puro vector containing shA1, shB2 or EV as negative control. Dot plots display the data of one of the two biological replicates of the experiment. The bar chart represents the mean + SD from two (n = 2) biological replicates. Right panel: Bar graph of the fold change (FC) of CD14 mean fluorescence intensity in THP-1 cells transduced for 48 h with shA1 and shB2, over the corresponding level in the EV-transduced cells. The chart represents mean + SD (n = 2 biological replicates). C Left panel: Representative dot plots showing the levels of CD11b- and CD86- positive cells as well as the intensity of these markers 48 h and 72 h post transduction with the shA1 and shB2 constructs targeting GSE1, or a nonspecific shRNA construct (NTC) cloned in pLKO.1 puro vector as negative control in the BB104 MLL- translocated primary patient sample. Right panel: Bar graph displaying the mean fluorescence intensity of CD11b, CD86 and CD14 in BB104 primary patient cells transduced for 48 h and 72 h with NTC, shA1 and shB2. The results are plotted as FC of the mean fluorescence intensity of these markers in the GSE1 KD samples over the corresponding levels in the NTC-transduced cells. The chart represents mean ± SD (n = 2 biological replicates). D Percentage of BB104 CD14-positive cells measured by flow cytometry at 48- and 72-h post transduction with the shA1 and shB2 cloned in pLKO.1 puro, or with the NTC used as negative control. Data are plotted as FC of the CD14 mean fluorescence intensity in the GSE1 KD samples over the corresponding levels in the NTC-infected cells. The chart represents mean ± SD (n = 2 biological replicates). E Percentage of CD11b-positive cells evaluated by flow cytometry in control (EV) and NB4 GSE1 OE cells, treated for 24 h with MC2580 (2 μM). NB4 GSE1 OE is produced by transducing the pLEX_307 vector with the GSE1 coding sequence (NM_001134473.3), while control cells are transduced with the pLEX_307 EV. The chart represents mean ± SD (n = 4 biological replicates; paired T-test, **p-value < 0.01). F Left panel: Flow cytometry analysis of the percentage of CD11b- and CD14- positive cells in EV and GSE1 OE THP-1 cells treated for 24 hours with MC2580. THP-1 GSE1 OE is generated by lentiviral transduction of the pLEX_307 vector including GSE1 coding sequence (NM_001134473.3), while control cells are transduced with the pLEX_307 EV. The chart represents mean ± SD (n = 3 biological replicates; paired T-test, *p-value < 0.05). Right panel: Bar graph displaying the CD14 mean fluorescence intensity in DMSO, MC2580 and MC2580 + GSE1 OE THP-1 cells. The chart represents mean ± SD from three (n = 3) biological replicates (paired T-test, *p-value < 0.05).CC BY · PubMed Central
Fig. 6 GSE1 expression shows significant correlation with oncogenic features in AML patient samples. A Histogram showing the expression levels of GSE1 mRNA across 31 different tumor histotypes from TCGA database. The x axis shows the tumor histotypes while the y axis displays the GSE1 transcripts per million (TPM). B Scatter plots in logarithmic scale of relative KDM1A (on x axis) and GSE1 (on y axis) mRNA expression levels, assessed in TCGA [50] (left panel) and Tyner et al. [51]. datasets (right panel). Red lines represent the best fit. Pearson and Spearman’s rank correlation coefficients, the equation of the line of best fit and the coefficient of determination (R2) are reported in the right corner of each panel. C GO analysis of the enriched biological processes (BP) on the list of genes whose expression levels are significantly negatively correlated with those of GSE1 in the TCGA (left panel) and Tyner et al. datasets (right panel). Bar graphs represent the top10 enriched BP ranked according to the enriched ratio. The y axes represent the BP terms while the x axes the enrichment ratio. D Scatterplots in logarithmic scale of relative ITGAM (1st panel from the left; on y axis), CD14 (2nd panel from the left; on y axis), CD86 (3rd panel from the left; on y axis), c-KIT (4th panel from the left; on y axis) and GSE1 (on x axes) mRNA expression levels, assessed in TCGA (upper raw) and Tyner et al. datasets (lower row). Red lines represent the best fit. Pearson and Spearman’s rank correlation coefficients, the equation of the line of best fit and the coefficient of determination (R2) are reported in each panel. E Box and whisker plots showing on the y axes the percentage of peripheral blood (1st panel and 2nd panel from the left) and bone marrow (3rd panel from the left) blast count of AML patients grouped on x axes according to low (i.e., below median value; GSE1-low) and high (i.e., above median value; GSE1-high) GSE1 mRNA tumor expression levels. Data from TCGA and Tyner et al. datasets are represented in the 1st panel and in the 2nd and 3rd panels, respectively.
Fig. 6 GSE1 expression shows significant correlation with oncogenic features in AML patient samples. A Histogram showing the expression levels of GSE1 mRNA across 31 different tumor histotypes from TCGA database. The x axis shows the tumor histotypes while the y axis displays the GSE1 transcripts per million (TPM). B Scatter plots in logarithmic scale of relative KDM1A (on x axis) and GSE1 (on y axis) mRNA expression levels, assessed in TCGA [50] (left panel) and Tyner et al. [51]. datasets (right panel). Red lines represent the best fit. Pearson and Spearman’s rank correlation coefficients, the equation of the line of best fit and the coefficient of determination (R2) are reported in the right corner of each panel. C GO analysis of the enriched biological processes (BP) on the list of genes whose expression levels are significantly negatively correlated with those of GSE1 in the TCGA (left panel) and Tyner et al. datasets (right panel). Bar graphs represent the top10 enriched BP ranked according to the enriched ratio. The y axes represent the BP terms while the x axes the enrichment ratio. D Scatterplots in logarithmic scale of relative ITGAM (1st panel from the left; on y axis), CD14 (2nd panel from the left; on y axis), CD86 (3rd panel from the left; on y axis), c-KIT (4th panel from the left; on y axis) and GSE1 (on x axes) mRNA expression levels, assessed in TCGA (upper raw) and Tyner et al. datasets (lower row). Red lines represent the best fit. Pearson and Spearman’s rank correlation coefficients, the equation of the line of best fit and the coefficient of determination (R2) are reported in each panel. E Box and whisker plots showing on the y axes the percentage of peripheral blood (1st panel and 2nd panel from the left) and bone marrow (3rd panel from the left) blast count of AML patients grouped on x axes according to low (i.e., below median value; GSE1-low) and high (i.e., above median value; GSE1-high) GSE1 mRNA tumor expression levels. Data from TCGA and Tyner et al. datasets are represented in the 1st panel and in the 2nd and 3rd panels, respectively.CC BY · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 72 references

Cited by

See all 33

Metadata

Authors
Nicosia, L., Boffo, F. L., Ceccacci, E., Conforti, F., Pallavicini, I., Bedin, F., Ravasio, R., Massignani, E., Somervaille, T. C. P., Minucci, S., & Bonaldi, T.
Affiliations
  • University of Manchester (GB)
  • European Institute of Oncology (IT)
  • University of Milan (IT)
Corresponding author
Bonaldi, Tiziana
Venue
Oncogene
Published
2021-12-03
Volume
41
Issue
6
Pages
878-894
ISSN
  • 0950-9232 (ISSN-L)
  • 1476-5594
OpenAlex
W3215349615
Cited by
31
References
72
Citations by year
2026: 3 · 2025: 8 · 2024: 6 · 2023: 7 · 2022: 7
FWCI
3.26
Type
journal article
Language
en
Open access
Licence
Locations
  • Oncogene (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • Research Explorer (The University of Manchester) (publishedVersion, other-oa)
  • Archivio Istituzionale della Ricerca (Universita Degli Studi Di Milano) (submittedVersion, cc-by)
  • Archivio Istituzionale della Ricerca (Universita Degli Studi Di Milano) (submittedVersion, other-oa)
Funders
  • Cancer Research UK ((C5759/A20971, A20971, 29389, C5759)
  • European Commission (823839, HORIZON2020)
  • Sapienza Università di Roma
  • Associazione Italiana per la Ricerca sul Cancro (15741, 823839)
  • Fondazione Italiana per la Ricerca sul Cancro
Topics
  • Health Sciences › Medicine › Hematology › Acute Myeloid Leukemia Research
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Epigenetics and DNA Methylation
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Retinoids in leukemia and cellular processes
Keywords
LSD1, LSD1 inhibitors, acute myeloid leukemia, myeloid differentiation, histone demethylase, chromatin proteins, transcriptional silencing, promoter binding, immune response genes, cytokine signaling, therapeutic target
MeSH terms
Animals, Cell Differentiation, Humans, Leukemia, Myeloid, Acute, Myeloid Cells, Cell Line, Tumor, Mice, Histone Demethylases

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Pharmacological inhibition of LSD1 triggers myeloid differentiation by targeting GSE1 oncogenic functions in AML. https://app.cochat.ai/papers/10.1038/s41388-021-02123-7
The paper
Nicosia, L., Boffo, F. L., Ceccacci, E., Conforti, F., Pallavicini, I., Bedin, F., Ravasio, R., Massignani, E., Somervaille, T. C. P., Minucci, S., & Bonaldi, T. (2022). Pharmacological inhibition of LSD1 triggers myeloid differentiation by targeting GSE1 oncogenic functions in AML. Oncogene. https://doi.org/10.1038/s41388-021-02123-7

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.