Abstract
The inhibition of the lysine-specific demethylase 1 (LSD1) has been shown to exert a potent antileukemic effect on the aggressive MLL-AF9+ acute myeloid leukemia (AML) in preclinical settings by inducing myeloid differentiation and apoptosis. In clinical studies, this drug is however not yet approved for treatment of AML patients and several reports have suggested that this treatment alone is not sufficient to provide the patient with significant benefits. Therefore, combination with other agents might represent a useful strategy for treatment of certain subtypes of AML.
LSD1 is known to demethylate histone- and non-histone proteins. Notably, regulation of p53 at both mono- and di-methylated Lysine K370 by LSD1 have been reported (Huang et al., 2007, Nature). A crosstalk between LSD1 activity and phosphorylation changes has been already reported and phosphorylation of LSD1 itself plays a role in metastatic events in solid tumors and inflammatory response (Metzger et al., 2010, Nature; Kim et al., 2018, Molecular Cell; Tu et al., 2020, Front Immunol). Here, we assessed the role of phosphorylation in the LSD1 inhibition (LSD1i) therapy of murine and human MLL-AF9+ AML cells in vitro.
Transcriptome analysis revealed that LSD1i induced upregulation of MAP kinases signaling in both murine and human AML cells (NES=1.87, p<0.05, FDR= 0.18 and NES=1.85, p=0.002, FDR=0.008, respectively) (RNA-seq analysis). Surprisingly, phosphoproteomic data analysis in MOLM14 cells also revealed a strong perturbation of phosphorylation landscape with 372 significantly differentially phosphorylated residues (increased phosphorylated: 212; decreased phosphorylated: 159) from 262 proteins upon 48h of LSD1i treatment. As reported in previous works where AML cells rapidly develop resistance to treatment via activation of phospho-pathways, we hypothesised that the MAPK-induction was an escape route for survival of leukemic cells. To this end, we tested the effect of combining LSD1- (GSK-LSD1, 0.5 μM ) with MEK-inhibitors (MEKi) (Trametinib, 0.005 μM) in murine and human AML cells. After 5 days-treatment with MEKi+LSD1i, cells displayed significantly decreased proliferation in comparison to the single drug treatments (Combination vs MEKi: p= 0.0079 (murine), p=0.0010 (human); Combination vs LSD1i: p=0.00099 (murine), p= 0.0058 (human)). Serial dilutions treatment revealed a synergistic effect (Highest Single Agent (HSA) Synergy score: 45.4 and 34.2 in murine AML and MOLM14 cells, respectively). As expected, LSD1i alone induced myeloid differentiation, while the double treatment did not lead to significant additional morphological and immunophenotipycal changes. However, at transcriptome level we observed an enhanced expression of several pro-inflammatory genes, as possible sign of further differentiation or triggering of pyroptosis. Interestingly, the combination treatment significantly increased the apoptosis (Combination vs MEKi: p<0.0001 in murine / p=0.0017 in human; Combination vs LSD1i: p=0.0002 in murine / p=0.0031 in human). The transcriptome analysis also revealed 1969 genes exclusively modulated upon MEK- and LSD1- inhibition (Figure 1) and Gene Ontology analysis of the differentially expressed genes highlighted the "Regulation of TP53 activity through phosphorylation” reactome pathway as signature exclusively modulated under treatment with MEKi+LSD1i. Phosphoproteomic analysis showed that MEKi+LSD1i treatment induced a specific dephosphorylation of p53 at Serine 315 (p<0.05; Log2Fc=-0.69) in comparison to the single treatments and the untreated cells (Figure 2). This residue has been reported to be more phosphorylated in MLL- rearranged AML compared to normal bone marrow samples and seems to be involved in p53 stability and activity (Kampen et al., 2014, Leukemia; Katayama et al., 2014, Nat. Genet; Nantajit, et al., 2010, PLoS One); in the present work we report for the first time, that this Post-translational modification (PTM) correlates alone with an effective antileukemic therapy.
In conclusion, MEKi+LSD1i significantly increases the antileukemic activity of the single treatments in this leukemia in vitro and this effect correlates with increased apoptosis and decreased phosphorylation of Ser315 of p53. Ongoing molecular investigations are aiming to confirm the involvement of p53 phosphorylation at Ser315 under MEK- and LSD1- inhibition.
Disclosures
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.
This feature is available to Subscribers Only
Sign In or Create an Account Close Modal