Background: Survival rates for acute lymphoblastic leukemia (ALL) have risen dramatically but up to 20% of patients relapse and their prognoses are dismal. Resistance to glucocorticoid (GC) agonists is a hallmark of relapsed ALL and a strong predictor of outcome at diagnosis (Dördelmann M et. al, Blood 1999, Schmiegelow K et. al, Leukemia 2001, Tissing WJ et. al, Leukemia 2003). However, the underlying biological pathways that mediate sensitivity to GCs remain to be determined. In this study, we performed a genome-scale shRNA screen to identify mediators of prednisolone sensitivity in ALL cells.

Methods: Genome-wide screening was performed using pooled shRNA libraries coupled with next generation sequencing to identify genes that when depleted promote GC sensitivity. Statistical analysis using Bioinformatics for Next Generation Sequencing (BiNGS) and Redundancy and Fold Change (RFC) were employed to identify candidate genes that mediate prednisolone sensitivity (Porter et. al, Leukemia 2012). Validation of hits from the primary screen were performed in Reh and RS4;11 cells. Knockdown of candidate genes MEK2 and MEK4 was determined by western blot. Changes in chemosensitivity upon MEK2 and MEK4 knockdown were determined by Cell Titer-Glo assay (Promega). The levels of apoptotic cells upon chemotherapy treatment in control and knockdown cell lines was determined by Annexin V-PE and 7-Amino-actinomycin D (7AAD) staining (Annexin V-PE Apoptosis Detection Kit, BD Pharmingen, San Diego, CA, USA), followed by flow cytometry using the FACSCalibur (Becton Dickinson, Franklin Lakes, NJ, USA). The levels of downstream GC target genes including NR3C1, GILZ, and BIM were determined by RT-PCR. The levels of target proteins including GR, pERK, ERK, MEK1, and p53 were determined by western blot. pERK levels in primary matched pairs were determined by multiparameter phosphoflow cytometry.

Results: In this study, we performed a genome-scale shRNA screen to identify mediators of prednisolone sensitivity in ALL cell lines. The incorporation of this data with integrated analysis of relapse-specific genetic and epigenetic changes (Hogan et. al, Blood 2012) allowed us to identify the mitogen-activated protein kinase (MAPK) pathway as a mediator of prednisolone resistance in pediatric ALL. Interestingly, depletion of MAPK pathway members, MEK2 and MEK4, increased sensitivity to prednisolone through distinct mechanisms. MEK4 knockdown increased sensitivity specifically to GCs by increasing the mRNA and protein levels of the glucocorticoid receptor (GR). This resulted in greater induction of the GR target genes GILZ and BIM upon prednisolone exposure over time. Importantly, depletion of MEK4 did not affect sensitivity of ALL cells to other chemotherapy agents (doxorubicin, etoposide, and 6-thioguanine). By contrast, MEK2 knockdown increased the sensitivity of cells to each of the chemotherapy agents tested including prednisolone, doxorubicin, etoposide, and 6-thioguanine. Depleting MEK2 decreased activated pERK and increased levels of p53. Over expression of a dominant negative p53 in MEK2 deficient cells reversed sensitivity to doxorubicin and prednisolone, indicating that MEK2 expression mediates chemosenstivity in a p53 dependent manner. Furthermore, inhibition of MEK1/2 pharmacologically with trametinib increased sensitivity of ALL cells to chemotherapy. Trametinib treatment also resulted in increased levels of p53. To determine if activation of the MAPK pathway in patients is associated with recurrent disease we examined seven matched diagnosis and relapse primary samples for MAPK activation as determined by pERK staining, and observed increased pERK levels at relapse in all samples tested.

Conclusion: Our data indicate that activation of the MAPK pathway promotes chemoresistance and may drive the development of recurrent disease in pediatric ALL.Asdisrupting MEK2 and MEK4 sensitizes cells to chemotherapy, this makes the MAPK pathway an attractive target for therapeutic intervention in relapsed ALL.

Disclosures

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

This icon denotes a clinically relevant abstract

Sign in via your Institution