Additional sex combs-like 1 (ASXL1) is recurrently mutated in various myeloid neoplasms and its mutations confer poor prognosis. Most of the ASXL1 mutations in myeloid malignancies are frameshift or nonsense mutations, leading to expression of C-terminally truncated form of ASXL1 (ASXL1-MT). Recently, we have demonstrated that monoubiquitination of ASXL1-MT at lysine 351 plays an important role in enhancing the catalytic activity of BAP1, resulting in reduction of H2AK119ub and upregulation of posterior HOXA genes to promote myeloid leukemogenesis (Asada et al. Nat commun. 2018). However, roles of other post-translational modifications of ASXL1 including phosphorylation remain to be elucidated.

In this study, we performed nano liquid chromatography tandem mass spectrometry (nano LC-MS/MS) and found that ASXL1 was phosphorylated at multiple serine (S) and threonine (T) residues including S503 residue, the major phosphorylated site of ASXL1. We generated an antibody that specifically recognizes phosphorylated S503 of ASXL1, and showed that CDK1/2 phosphorylated both wildtype ASXL1 and ASXL1-MT. However, phosphorylation status of ASXL1-MT at serine 503 did not affect its oncogenic activity. Thus, we considered the possibility that phosphorylations of other residues would compensate the effects of phosphorylation of ASXL1-MT at serine 503. Using public database and our data from mass spectrometry, we found additional five S/T residues containing a CDK phosphorylation motif (S/T-P). We then established a non-phosphomimetic mutant form of ASXL1-MT (ASXL1-MT-6A) whose six S/T residues were substituted to alanine.

To compare the oncogenic functions between ASXL1-MT-6A and ASXL1-MT, we first assessed the impact of these mutants on differentiation of hematopoietic cells. Expression of ASXL1-MT-6A more strongly impaired differentiation of human CD34+ cord blood (CB) cells towards granulocytes and erythrocytes than that of ASXL1-MT. These effects of ASXL1-MT-6A on differentiation block were also observed in HL-60 cells. Next, we assessed the impact of phosphorylation of ASXL1-MT on its leukemogenic functions using RUNX1-ETO and N-Ras G12V-induced leukemia models, because RUNX1-ETO fusion or N-Ras mutations frequently coexist with ASXL1 mutations in patients with myeloid neoplasms. In agreement with our previous report, expression of ASXL1-MT enhanced proliferation of RUNX1-ETO expressing CB cells. ASXL1-MT-6A showed the stronger growth promoting effect than ASXL1-MT in RUNX1-ETO expressing CB cells in vitro by inhibiting apoptosis and progressing cell cycle. Moreover, ASXL1-MT-6A induced more rapid development of myeloid leukemia driven by N-Ras G12V than ASXL1-MT in vivo. ASXL1-MT-6A induced upregulation of Hoxa5, Hoxa7 and Hoxa9 more strongly than ASXL1-MT in myeloid progenitor cells. These data indicated that ASXL1-MT-6A is an active form of ASXL1-MT.

Mechanistically, cycloheximide chasing assay revealed that ASXL1-MT-6A had increased protein stability than ASXL1-MT. Furthermore, cell cycle synchronization and immunoprecipitation assay demonstrated that degradation of ASXL1 protein began at early M phase by interacting with Cdc20, one of the coactivators of anaphase-promoting complex (APC/C), an E3 ubiquitin ligase complex involved in regulating cell cycle progression. Overexpression of Cdc20 promoted degradation of ASXL1, whereas depletion of Cdc20 using CRISPR-Cas9 system increased expression of ASXL1. These biochemical features were also observed for ASXL1-MT. Intriguingly, ASXL1-MT-6A showed resistance to degradation induced by APC/CCdc20. Therefore, CDK-mediated phosphorylation of ASXL1-MT triggers its degradation induced by APC/CCdc20.

Collectively, we demonstrated that stability and leukemogenic functions of ASXL1-MT are regulated by CDK-mediated phosphorylation. Dynamics of protein degradation of ASXL1-MT triggered by phosphorylation and ubiquitination could be targets for therapeutic interventions of myeloid neoplasms harboring an ASXL1 mutation.

Disclosures

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution