Dysregulation of Follistatin Promotes Leukemogenesis in Acute Myeloid Leukemia

Acute myeloid leukemia (AML) is characterized by an abnormal increase of myeloblasts in bone marrow (BM) and peripheral blood (PB). Chemotherapy and allogeneic hematopoietic stem cell transplantation (HSCT) are the mainstays of treatment. However, these approaches have reached an impasse and despite an initial remission, leukemia relapses are frequent, with an overall cure rate of only 30-40% (Rowe and Tallman 2010). Internal tandem duplication (ITD) of the fms-like tyrosine kinase 3 (FLT3) gene is the commonest mutation in AML, occurring in nearly 30% cases with inferior disease-free and overall survivals (Boissel et al. 2002; Meshinchi et al. 2001). FLT3-ITD as a therapeutic target has been tested in phase I/II clinical trials (Man et al. 2012; Fischer et al. 2010; Levis et al. 2011; Cortes et al. 2011). In most circumstances, there was an initial response to FLT3 inhibitors with reduction or clearance of myeloblasts in BM and PB. However, resistance invariably developed despite continuous treatment (Kindler, Lipka, and Fischer 2010). Therefore, the identification of novel therapeutic targets for FLT3-ITD+ AML becomes urgent.

In our previous study (He et al. 2014), we showed that over-expression of human leukemic gene FLT3-ITD resulted in significant myeloid cells expansion which could be effectively rescued by FLT3-ITD inhibitor AC220 treatment. Interestingly, a proportion of zebrafish embryos expressing FLT3-ITD showed double-headed and axis-duplication phenotype as shown by the whole mount in situ hybridization (WISH) of a notochord-specific marker col9a2 . Mechanistically, FLT3-ITD-induced axis-duplication was correlated with the expansion of Spemann's Organizer as shown by the ectopic expression of Spemann's Organizer-specific marker goosecoid (gsc) by WISH as well as by the -2,067gsc-GFP reporter assay. It has been reported that some key signaling pathways are conserved between embryogenesis and tumorigenesis. We hypothesized that certain Spemann's Organizers were dysregulated in FLT3-ITD+ AML.

To test this hypothesis, we performed an in silico analysis to quantify the gene expression of Spemann's Organizers in normal HSC and AML by the public database BloodSpot. Intriguingly, the Activin antagonist Follistatin (FST) was significantly increased in hematological malignancies, comparing to the normal HSC. In human malignancies, FST has been shown to express in prostate (Sepporta et al. 2013; Tumminello et al. 2010), ovarian (Ren et al. 2012; Karve et al. 2012; Di Simone et al. 1996), and liver cancers (Patella et al. 2006; Rossmanith et al. 2002) promote cancer cell growth (Gao et al. 2010; Rossmanith et al. 2002). However, the role of FST in human AML and other hematological malignancies is currently unknown. Firstly, we confirmed the upregulation of FST in AML by real-time quantitative PCR, Western Blotting, and immunofluorescence imaging. Mechanistically, FLT3-ITD up-regulated FST via the phosphorylation of CREB as demonstrated by ChIP-PCR. Moreover, the upregulation of FST by FLT3-ITD was effectively blocked by the pharmacological inhibition of CREB in a dosage-dependent manner. FST gene knock-down (by shRNA) or knock-out (by CRISPR/Cas9) in FLT3-ITD+ MOLM-13 AML cell line significantly reduced leukemic growth in vitro, reduced leukemic burden, and prolonged survival of mice in the xenotransplantation experiment. Over-expression of FST in MOLM-13 and ML-2 AML cell lines promoted leukemic growth in vitro and in vivo xenograft mouse model via AKT/p70S6K/mTOR signaling. Furthermore, proteins (RPS19, RPS6, RPS13, RPL6, RPS21, RPSA, RPS18, RPLP2, RPL10A, WIBG, and PPP2R1A) associated with nonsense-mediated decay of mRNA were decreased after FST over-expression in ML-2 AML cells, while the chemoresistant and anti-apoptotic protein CD44 was consistently increased by a label-free proteomics analysis. Notably, the serum FST levels were significantly increased in a patient-derived xenograft AML mouse model. Consistently, plasma FST levels from FLT3-ITD+ AML patients were significantly reduced after an FLT3-ITD inhibitor quizartinib treatment in a phase I/II clinical trial.

In conclusion, the CREB targeted gene FST promotes AML cell growth via AKT/p70S6K/mTOR signaling and CD44-mediated anti-apoptosis. Therefore, FST may be a promising therapeutic target for AML treatment.