Dimethyl fumarate and vitamin D derivatives cooperatively enhance VDR and Nrf2 signaling in differentiating AML cells in vitro and inhibit leukemia progression in a xenograft mouse model.

Acute myeloid leukemia (AML) is one of the deadliest hematological malignancies without effective treatment for most patients. Vitamin D derivatives (VDDs) - active metabolites 1α,25-dihydroxyvitamin D (1,25D2) and 1α,25-dihydroxyvitamin D (1,25D3) and their analogs - are differentiation-inducing agents which have potential for the therapy of AML. However, calcemic toxicity of VDDs limits their clinical use at doses effective against cancer cells in vivo. Here, we demonstrate that in AML cell cultures, moderate pro-differentiation effects of low concentrations of VDDs can be synergistically enhanced by structurally distinct compounds known to activate the transcription factor Nuclear Factor (Erythroid-derived 2)-Like 2 (NFE2L2 or Nrf2). Particularly, dimethyl fumarate (DMF), which is clinically approved for the treatment of multiple sclerosis and psoriasis, strongly cooperated with 1,25D3, PRI-5100 (19-nor-1,25D2; paricalcitol) and PRI-5202 (a double-point modified 19-nor analog of 1,25D2). The pro-differentiation synergy between VDDs (1,25D3 or PRI-5202) and Nrf2 activators (DMF, tert-butylhydroquinone or carnosic acid) was associated with a cooperative upregulation of the protein levels of the vitamin D receptor (VDR) and Nrf2 as well as increased mRNA expression of their respective target genes. These data support the notion that VDDs and Nrf2 activators synergize in inducing myeloid cell differentiation through the cooperative activation of the VDR and Nrf2/antioxidant response element signaling pathways. We have previously reported that PRI-5202 is more potent by approximately two orders of magnitude than 1,25D3 as a differentiation inducer in AML cell lines. In this study, we found that PRI-5202 was also at least 5-fold less calcemic in healthy mice compared to both its direct precursor PRI-1907 and 1,25D3. In addition, PRI-5202 was remarkably more resistant against degradation by the human 25-hydroxyvitamin D-24-hydroxylase than both 1,25D2 and 1,25D3. Importantly, using a xenograft mouse model we demonstrated that co-administration of PRI-5202 and DMF resulted in a marked cooperative inhibition of human AML tumor growth without inducing treatment toxicity. Collectively, our findings provide a rationale for clinical testing of low-toxic VDD/DMF combinations as a novel approach for differentiation therapy of AML.