MEK inhibition enhances the response to tyrosine kinase inhibitors in acute myeloid leukemia

FMS-like tyrosine kinase 3 (FLT3) is a key driver of acute myeloid leukemia (AML). Several tyrosine kinase inhibitors (TKIs) targeting FLT3 have been evaluated clinically, but their effects are limited when used in monotherapy due to the emergence of drug-resistance. Thus, a better understanding of drug-resistance pathways could be a good strategy to explore and evaluate new combinational therapies for AML. Here, we used phosphoproteomics to identify differentially-phosphorylated proteins in patients with AML and TKI resistance. We then studied resistance mechanisms in vitro and evaluated the efficacy and safety of rational combinational therapy in vitro, ex vivo and in vivo in mice. Proteomic and immunohistochemical studies showed the sustained activation of ERK1/2 in bone marrow samples of patients with AML after developing resistance to FLT3 inhibitors, which was identified as a common resistance pathway. We examined the concomitant inhibition of MEK-ERK1/2 and FLT3 as a strategy to overcome drug-resistance, finding that the MEK inhibitor trametinib remained potent in TKI-resistant cells and exerted strong synergy when combined with the TKI midostaurin in cells with mutated and wild-type FLT3. Importantly, this combination was not toxic to CD34+ cells from healthy donors, but produced survival improvements in vivo when compared with single therapy groups. Thus, our data point to trametinib plus midostaurin as a potentially beneficial therapy in patients with AML.


Extended cell cultures, AML patients, healthy donors and drugs
Human MOLM-13 (FLT3 ITD/WT ) and OCI-AML3 (FLT3 WT/WT ) AML cell lines were obtained from the DSMZ culture collection (Braunschweig, Germany). Other mutations are available at the Catalogue of Somatic Mutations (COSMIC) website (COSMIC sample ID: COSS1330947 and COSS1290455 respectively). MOLM-13 cells were cultured in RPMI-1640 medium (Biowest, Nuaillé, France) supplemented with 10% fetal bovine serum (FBS) (Thermo Fisher Scientific, Waltham, MA, USA) and antibiotics (100 U/ml penicillin and 100 μg/ml streptomycin). To induce resistance to sorafenib, the parental MOLM-13 cell line was grown in culture in the presence of 5 nM sorafenib, which was gradually increased to 20 nM over 17 days to obtain MOLM-13 resistant (MOLM-13R) cells. Culture media for MOLM-13R cells were supplemented with a final concentration of 20 nM sorafenib. The OCI-AML3 cell line was cultured in IMDM medium (Biowest, Nuaillé, France) supplemented with 5% FBS. To induce resistance to midostaurin, the parental OCI-AML3 cell line was grown in culture in the presence of 0.1 M midostaurin, which was gradually increased to 5 M over 10 days to obtain the resistant culture (OCI-AML3R). All cell lines were grown at 37ºC in a humidified atmosphere containing 5% CO2, and maintained by periodic subculture every 2-3 days.

Extended whole exome sequencing
A total of 1 µg of high-quality RNA-free genomic DNA was isolated from the bone marrow of patient #1 at diagnosis. Exonic sequences were captured through probe-hybridization, amplified, and purified following the Ion TargetSeq™ Exome Enrichment manual (MAN0006730, Life Technologies S.A., Madrid, Spain). The purified DNA was then sequenced on the Ion Proton™ System (ThermoFisher Scientific, Waltham, MA, USA). Obtained sequence reads were aligned to the human genome reference sequence (hg19) in Torrent Suite 4.4.3.
Point mutations and 1 base-pair indels were restricted to variants with mean allele frequency (MAF) < 0.1%, P < 0.05, and a minimum of 10 reads, supporting a call of at least 20 total reads. Indels were restricted to variants with MAF < 0.1%, P < 0.05, and a minimum of five reads supporting a call. The Ion PGM system (ThermoFisher Scientific) was used to validate the results.

Extended LC-MS/MS analysis
Phospohoproteomic experiments were conducted at the Proteomics Core Unit of the Centro Nacional de Investigaciones Oncológicas (CNIO).

Data analysis
Raw files were processed with MaxQuant (v 1.5.0.12) using the standard settings against a human protein database (UniProtKB/Swiss-Prot, December 2013, 20,584 sequences) supplemented with contaminants. Label-free quantification was done with match between runs (match window of 0.7 min and alignment window of 20 min). Carbamidomethylation of cysteines was set as a fixed modification whereas oxidation of methionines, protein N-term acetylation and phosphorylation of serines, threonines and tyrosines were set as variable modifications. Minimal peptide length was set to 7 amino acids and a maximum of two tryptic missed-cleavages were allowed. Results were filtered at 0.01 false discovery rate (peptide and protein level). Subsequently, the "phospho(STY)sites.txt" file was loaded in Perseus (v1.5.1.6) for further analysis.

Extended histopathology and immunohistochemistry
Paraffin embedding and sectioning of the different tissues (bone marrow clots, sternum, spleen, liver and urinary bladder) from patients with AML or mice were performed according to standard protocols. For phospho-ERK1/2 and human CD45 staining, formalin-fixed paraffin-

Extended immunoblotting assays
Cultured cells after treatment (200 nM of each treatment for 3 hours; or DMSO in control sample) or mononuclear cells were washed with ice-cold phosphate buffered saline (PBS) (Lonza, Basel, Switzerland) and pellets were collected. Cell pellets were treated with a cold Page 7 of 13 lysis buffer mixture (50 mM Tris, 150 mM NaCl, 1% Triton, 0.5% sodium deoxycholate, 0.1% SDS, protease and phosphatase inhibitors, 100 U/ml DNase I, distilled H2O) for 30 minutes on ice and, subsequently centrifuged at 14000 × g for 5 minutes at 4ºC. Supernatants were collected and stored at -80ºC until analysis. Protein concentrations were determined using the Quick Start™ Bradford 1× Dye Reagent (Bio-Rad Laboratories), following the manufacturer's instructions, and read with the Epoch microplate reader running Gen5 software (both from BioTek, Winooski, VT, USA).
Whole cell lysates were separated by SDS-PAGE and then transferred to polyvinylidene difluoride Immun-Blot® membranes (Bio-Rad Laboratories). Proteins were detected after overnight incubation at 4ºC with primary antibodies (see Supplementary Table 1). Antibodies against α-tubulin, GAPDH or ERK1/2, AKT, STAT5, MAPK14 and their phosphorylated forms were used. Then, secondary antibodies (see Supplementary Table 1)

Extended colony-forming unit assays
To test for treatment-related toxicity, colony-forming assays in CD34+ control cells were

Extended in vivo studies
Female JAX TM NSG mice (NOD.Cg-Prkdc scid Il2rg tm1Wjl /SzJ) from The Jackson Laboratory (Bar Harbor, ME, USA) were used at the age of 5-6 weeks. Standard rodent diet and water were available ad libitum throughout the study. Mice were housed seven per cage, each cage being a treatment group. Mice were injected with 5×10 6 OCI-AML3 cells resuspended in PBS by tail injection. Three days after injection, mice were treated with vehicle (10% DMSO, n=7), trametinib (0.5 mg/kg, n=7), midostaurin (50 mg/kg, n=7) or the combination of both