Background: We have identified the FOXO3-AMPK-AKT pathway as important to HbF regulation through unbiased analysis of whole exome sequencing from patients with sickle cell disease (SCD). Both AMPK and AKT are activated by phosphorylation; AMPK is a FOXO3 activator; AKT inhibits AMPK and FOXO3. We have also treated hematopoietic stem and progenitor cells (HSPCs) from patients with SCD with metformin, a known AMPK/FOXO3 activator, and produced a significant rise in HbF from 17.6% at baseline, to 44.9%. This data is the foundation for an IRB approved clinical trial of metformin in patients with hemoglobinopathies, enrolling patients now.

Objectives: We seek to investigate other drugs which act on the FOXO3-AMPK-AKT pathway as possible HbF inducing agents. Resveratrol and piceatannol are both known activators of the pathway; however, resveratrol is not bioavailable. We will investigate piceatannol as a HbF inducing agent, assess toxicity, and verify its action is via the FOXO3-AMPK-AKT pathway.

Methods: HSPCs from 5 patients with SCD were treated with serial doses of piceatannol at 5, 12.5, 25, 50 and 100 µM on day 7 of two phase erythroid culture. Cell lysate was collected on day 14, and hemoglobin profile analyzed by HPLC. 2 µM Compound C was added with piceatannol to block AMPK activity, and the effect on HbF measured by HPLC. The effect of piceatannol on the amount of phosphorylated AMPK and AKT in HSPCs was measured by quantitative western blot in triplicate, and the effect of piceatannol on FOXO3 cellular localization was determined by fluorescent microscopy and fractionation of the cells into nuclear and cytoplasmic compartments, followed by quantitative western blot analysis.

Results: Erythroid progenitor cells from 5 unique patients with SCD treated with 25 µM piceatannol all exhibited a 3-4 fold rise in HbF as measured by HPLC. Toxicity studies showed that 12.5-50 µM of piceatannol induced HbF without significantly altering CFU generation of human CD34+ HSPCs; treatment with 100 µM piceatannol showed a reduction in CFU. In the presence of Compound C, piceatannol induction of HbF was attenuated. 5-25 µM piceatannol markedly increased the amount of phosphorylated AMPK present in the cell, caused an increase in FOXO3 localization in the nucleus, and effectively inhibited AKT phosphorylation.

Conclusions: Piceatannol induces HbF in SCD patient derived HSPCs. While piceatannol has been reported to have HDAC inhibitor activity, the fact that pharmacologic blockage of AMPK activity with Compound C results in reduction of HbF induction by piceatannol supports the hypothesis that piceatannol acts through the FOXO3-AMPK-AKT pathway. Additionally, piceatannol reduced phosphorylation of a known FOXO3 inhibitor, AKT, and increased phosphorylation of AMPK, a FOXO3 activator, thereby increasing FOXO3 activity and localization to the nucleus. With further safety studies, piceatannol could be a novel HbF inducing agent for patients with hemoglobinopathies.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.