Resistance to glucocorticoids (GC) is a hallmark of relapsed acute lymphoblastic leukemia (ALL) and is a predictor of outcome at diagnosis. In spite of the importance of GC in the treatment of ALL and other hematological malignancies the molecular mechanisms that lead to effective eradication of leukemic cells is incompletely understood. To address this problem we have performed a functional screen for genes involved in prednisolone resistance in ALL cell line (Reh) and correlated these results with our previously published results using an integrated genomic analysis to discover genes (pathways) altered at relapse (Hogan et al 2011).

Cells were infected with a pooled whole genome shRNA library that contained approximately 80,000 shRNAs targeting 18,000 genes. Deep sequencing was used to identify shRNAs enriched or depleted upon treatment with prednisolone. Three computational methods including; bioinformatics for next generation sequencing analysis (BiNGS), redundancy & fold change analysis (RFC) and strict standardized mean difference (SSMD) were applied to the sequencing data in efforts to obtain the most robust set of candidate genes for validation (Porter et. al, Leukemia 2012, Zhang XD, J Biomol Screen, 2007).

Through our primary screen a total of 263 genes were identified to modulate prednisolone sensitivity in ALL. Upon knockdown, 142 genes increased the sensitivity of the cells to prednisolone and 121 genes increased resistance to prednisolone. Five genes overlapped with genes previously identified to be altered at relapse compared to matched diagnosis samples including SLC6A18, AARSD1, MIER3, CDC42BPB, and YAP1. We hypothesize that genes that are altered at relapse in ALL and identified through functional genomics screening to modulate chemosensitivity in vitro, are likely drivers of chemoresistance and eventual relapse.

We also performed gene ontology (GO) analysis using DAVID Bioinformatics to identify pathways that may be responsible for altered resistance to prednisolone. This analysis strongly implicated the mitogen-activated kinase (MAPK) pathway. The MAPK was also identified as a pathway with increased activity at relapse through our integrative genomics analysis(Hogan et al 2011). One gene of particular interest was MAP2K4 which encodes for MEK4, an upstream kinase involved in JNK phosphorylation and c-Jun activation. Knockdown of MAP2K4 by shRNA in B-precursor ALL cell lines (Reh and RS4;11) results in statistically significant (p-value < 0.5) increased sensitivity to prednisolone induced apoptosis at a range of prednisolone concentrations but not to other chemotherapy tested (etoposide, doxorubicin, and 6-thioguanine).604. Molecular Pharmacology, Drug Resistance: Poster I Upon treatment with prednisolone MAP2K4 knockdown cells have increased levels of prednisolone responsive genes GILZ (1.4-2.4 fold in Reh, 2.1-3.8 fold in RS4;11) and TXNIP (1.8-5.7 fold in Reh, 2.5-2.6 in RS4;11). Increased sensitivity to prednisolone and increased levels of prednisolone responsive genes was associated with decreased levels of p-JNK that has been previously implicated in regulating glucocorticoid signaling through phosphorylation of the glucocorticoid receptor (GR) at S226 (Roatsky et. al, PNAS 1998, Itoh et. al, Mol. Endo. 2002).s Overall this data suggests that decreased levels of MAP2K4 results in increased sensitivity to GC by increasing GC signaling and implicates MEK4 as novel drug target in ALL.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.