CREB (cAMP Response Element Binding protein) is a transcription factor that is overexpressed in primary Acute Myeloid Leukemia (AML) cells and is associated with a decreased event-free survival and increased risk of relapse. We previously demonstrated that CREB overexpression increases leukemia cell growth and survival. Transgenic mice overexpressing CREB in myeloid cells develop a myeloproliferative neoplasm and myelodysplasia. CREB knockdown inhibits AML cell proliferation but not normal hematopoietic stem cell activity in vivo. To demonstrate the feasibility of targeting CREB for treatment for AML, we recently described a small molecule inhibitor of CREB, N-(4-cyanophenyl)-3-hydroxy-2-naphthamide (XX-650-23), which is a compound originally based on naphthol AS-E phosphate first identified as an inhibitor of CREB interaction with its coactivator, CBP (CREB Binding Protein). To identify a lead candidate with improved potency and physicochemical properties, we performed structure-activity relationships (SAR) studies for a series of salicylamides derived from naphthol AS-E phosphate. Development of this series led to the identification of the anthelmintic niclosamide as a potent agent that suppresses cell viability of five AML cell lines (IC50= 280 nM (HL60), 340 nM (KG1), 420 nM (MOLM13), 560 nM (MV411), 360 nM (U937), without a significant decrease in colony forming activity of normal bone marrow cells up to 10 μM (18- to 36-fold therapeutic window). Niclosamide binds CBP with a KD of 22.3 nM by Surface Plasmon Resonance (Biacore) analysis. To determine whether niclosamide specifically inhibits CREB-mediated gene expression in cells, luciferase reporter gene activity under the control of a promoter containing two CRE elements was measured after treatment of niclosamide for 6 hours. Niclosamide inhibited CREB-driven luciferase activity in HL60 cells with an IC50 of 1.09 μM. We also examined the efficacy of niclosamide in an AML patient-derived xenograft (PDX) mouse model. Niclosamide significantly inhibited the progression of AML in mice injected with primary AML cells. The percentage of circulating AML cells in the peripheral blood (%), vehicle vs. niclosamide treatment 5 weeks after engraftment were 28.75 ± 3.507 vs. 0.5363 ± 0.2744 (n=8, p< 0.001, mean ± SEM). In Kaplan Meier analysis, the median survival of PDX mice was 41 days vs. 51.5 days (p = 0.0015, log-rank test). To characterize the cellular effects of niclosamide, we analyzed the DNA profile, apoptosis, DNA-damage, cell cycle regulators, and other signaling molecules using flow cytometry. Niclosamide treatment increased DNA-damaged and apoptosis populations during the G1/S cell cycle phase, which also showed reduced phosphorylated CREB levels. To examine the functional requirement of CREB, we determined the effects of CREB knockdown in HL60 cells treated with niclosamide. CREB knockdown protected HL60 cells from niclosamide treatment-mediated cytotoxic effects (IC50=670 nM for CREB knockdown vs. 200 nM for vector control cells). Furthermore, combination treatment of niclosamide with XX-650-23 in HL60 cells showed an additive antiproliferative effect, suggesting that niclosamide and XX-650-23 regulate the same targets or pathways to inhibit viability of AML cells. To further identify genes that confer resistance or sensitivity to niclosamide, we performed a functional shRNA screen using subsets of whole genomic shRNA libraries (apoptosis, motility, other cancer; 35154 elements). We identified 53 genes, including tumor necrosis factor receptor superfamily members, which when knocked downed conferred resistance to niclosamide at a 10% false discovery rate. Taken together, our results demonstrate that niclosamide is a potential drug to treat AML by inducing DNA-damage, apoptosis and cell cycle arrest through the inhibition of CREB-dependent pathways in AML cells.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.