Synergistic Induction of Differentiation in Leukemia Cells by All-Trans Retinoic Acid (ATRA) and Plant polyphenols: Involvment of Retinoic Acid Receptor Beta (RARβ)

Abstract

Current cancer therapies are highly toxic and often nonspecific. A potentially less toxic approach to treating this prevalent disease employs agents that modify cancer cell differentiation, termed "differentiation therapy". Acute promyelocytic leukemia (APL), bearing a characteristic t(15;17) chromosomal translocation with breakpoints within the retinoic acid receptor alpha (RARα) and promyelocytic leukemia (PML) gene, is the best example of a malignant disease successfully treated by differentiation-inducing agents, e.g. all-trans retinoic acid (ATRA). The specific sensitivity of APL cells to ATRA has modified the therapeutic approach of APL resulting in 90% complete remission. However, prolonged treatment with high (therapeutic) ATRA doses (~ 1 μM) results in retinoid resistance syndrome and relapses usually occur within months in nearly all patients who are initially sensitive to ATRA. Another powerful differentiation agent, 1,25- dihydroxyvitamin D3 (1,25D3), can induce differentiation in various myeloid leukemia cells in vitro. However, it is profoundly toxic (hypercalcemic) at pharmacologically active doses. Recently, we have shown that several dietary antioxidant micronutrients, such as the tomato carotenoid lycopene and the polyphenol carnosic acid (CA) found in rosemary, substantially enhance the differentiating and anti-proliferative effects of low, non-toxic concentrations of ATRA and 1,25-dihydroxyvitamin D3 in HL-60 and U937 human myeloid leukemia cells (in vitro) and in the mouse-leukemia model (in vivo). This enhancement of antileukemic effect in vitro was associated with an increase in the levels of nuclear receptors for retinoids and vitamin D (VDR). The current study characterized strong anti-proliferative and differentiation effects of combined treatment of different human myeloid leukemia cell lines and cells obtained from leukemic patients with low concentrations of ATRA (0.3-1 nM) and CA (2.5-5 µM), and explored the molecular mechanism underlying these effects. A synergistic cell growth inhibitory effect of CA/ATRA combinations in different leukemic cell lines correlated with cell accumulation in G0/G1 phase of cell cycle and was not associated with cell death. Furthermore, a synergistic induction of differentiation was observed, as was evident by the changes in cell morphology, reorganization of nuclear bodies, induction of CD11b, CD11c and CD18 surface markers and oxygen burst activity. Importantly, combinations of ATRA and CA showed anti-proliferative and differentiation effects on ATRA-resistant cell lines (NB4-R1 and HL-60R). Importantly, the above antileukemic effects of inducer/polyphenol combinations in all leukemia cell lines (NB4, HL-60, U937, PLB-985) and ex vivo cells obtained from leukemia patients correlated with strong induction of RAR and, in NB4 cells (the APL cell line), with PML-RAR degradation and RAR stabilization. RAR gene expression is often lost or reduced in human carcinoma cells, suggesting a tumor suppressor role of RAR. Interestingly, the resistance of NB4 cells to 1,25D3, and of HT-29 colon cancer cells, and some types of ex vivo leukemic cells (M2 and M4) to polyphenols, ATRA, 1,25D3 and their combinations were associated with undetectable levels of RAR in these cells that were not changed by the indicated agents. Overexpression of RAR, after transfection of the RAR-expression vector into HT-29 cells, enhanced cell growth inhibition induced by CA, ATRA and their combination by 20-40%. Overexpression of the same vector in U937 cells enhanced the ATRA-induced growth inhibition and differentiation by 20-30%, whereas the effects of CA/ATRA combinations were not increased. These data indicate that substantial upregulation of RAR by the CA/ATRA combination results in a maximal cell differentiation response which is not further enhanced by the ectopic expression of RAR. The opposite effects on cell growth and differentiation were detected after RAR downregulation. Expression of the dominant negative RAR, mutated in its DNA-binding domain, in HL-60 cells decreased the cell growth inhibition, induced by the CA/ATRA combination (by 20%), and differentiation induced by both ATRA at high concentration (by 30%) and the combination (by 25%). Even stronger reduction of ATRA- and CA/ATRA-induced differentiation (by about 40%) was obtained by silencing of RAR induced by RAR-siRNA oligos transfected into HL-60 cells. The mechanism, by which plant polyphenols synergistically enhance the anticancer effects of ATRA and 1,25D3, is largely unknown. Our data indicate the involvement of retinoic acid receptor regulation at both transcriptional (mRNA) and protein levels in the CA-ATRA synergism. This is evidenced by the correlation between the enhanced antileukemic effect of the CA/ATRA combination and upregulation and stabilization of RAR, induction of RAR, and reduction in PML- RAR levels. In addition, strong upregulation of vitamin D receptor (VDR) and its heterodimeric partner, retinoid X receptor (RXR), correlated with the polyphenol potentiation of the antileukemic effect of 1,25D3 in patient-derived leukemic cells. Besides CA-induced upregulation of RAR at the transcriptional level, as seen by mRNA induction and transactivation of RAR response element (DR5) derived from RAR gene promoter, the possible mechanism of nuclear receptor modulation may involve the inhibition by polyphenols of proteasome-dependent or -independent degradation of these receptors. Indeed, treatment of NB4 cells with the proteasome inhibitor, MG132, dramatically elevated RAR levels. However, in contrast to CA, MG123 only slightly increased the ATRA-induced differentiation, indicating that the upregulation of RAR alone is not sufficient for the differentiation enhancement. An additional putative mechanism, which emerged from this study and may contribute to the polyphenol-inducer synergism, is CA-induced inhibition of the ATRA degradation pathway. Our preliminary results obtained by RT-PCR indicate that in ATRA treated NB4 cells, CA downregulates CYP26, the member of the cytochrome P450 enzyme family which is responsible for specific oxidation and degradation of ATRA. CYP26 downregulation is likely to increase cellular levels of ATRA which, in turn, would enhance the differentiation response. These data are supported by the evidence that a synergistic potentiation by CA of ATRA-induced CD11b induction in NB4 cells was mimicked by ketoconazole, a broad range inhibitor of P450 related enzymes. Taken together, the results of the current mechanistic study suggest that the enhanced cooperative anticancer effects of the combinations of polyphenols and differentiation inducers in both leukemic and non-leukemic cancer cells are mediated by multiple complementing mechanisms (e.g., nuclear receptor upregulation and stabilization of inducer concentrations). This study may provide the basis for the development of strategies for the treatment of myeloid leukemias and other malignant diseases, both sensitive and resistant to ATRA treatment, using combinations of ATRA and dietary polyphenols. This work supports the important role of RAR as the therapeutic target in cancer prevention and treatment.