FLT3 internal-tandem-duplication (FLT3-ITD) mutations are highly recurrent in acute myeloid leukemias (AML). Although secondary messenger impacts of these receptor tyrosine kinase mutations are well characterized, downstream mechanisms by which myeloid precursors are transformed remain unclear. One clue is that FLT3-ITD and CEBPA bi-allelic mutations are mutually exclusive, suggesting that FLT3-ITD might impact CEBPA, a master transcription factor mandatory for granulo-monocytic lineage-differentiation. In fact, FLT3-signaling has been shown to phosphorylate CEBPA at serine 21 (S21-P), and inhibition of this post-translational modification resumed granulocytic-differentiation of FLT3-ITD AML cells (Radomska et al, J Exp Med 2006). How S21-P post-translational modification of CEBPA blocks granulocyte lineage-differentiation, however, was unknown.

To examine FLT3-ITD impacts on CEBPA function in primary AML cells, we examined expression of CEBPA target genes (24 genes identified by CEBPA ChIP-seq, Jakobsen et al, Genome Res 2013): CEBPA target genes positively correlated with CEBPA mRNA expression in normal granulopoiesis and in primary FLT3/CEBPA-wildtype (WT) AML cells (n=112) (Pearson correlation coefficient [ρ] 0.36, p<0.0001), but not in primary FLT3-ITD AML cells (n=35)(ρ =-0.15, p=0.4). Moreover, CEBPA target genes were almost 2-fold less expressed in the FLT3-ITD AML cells (p<0.0009), despite CEBPA mRNA expression comparable to the WT AML cells (RNA-sequencing, TCGA).

To investigate the disconnect between CEBPA mRNA expression and CEBPA target gene activation, we analyzed CEBPA protein, and found very low amounts in FLT3-ITD AML cells MOLM13 and MV411 compared to large CEBPA protein amounts detected in WT THP1 AML cells, both by Western blot and immuno-fluorescence, again despite similar CEBPA mRNA expression. This observation complemented published data from others that FLT3-ITD introduction into murine 32D cells suppressed total Cebpa protein (Mizuki et al, Blood 2003; Zheng et al, Blood 2004). The FLT3-inhibitors sunitinib and gilteritinib decreased CEBPA S21-P, increased CEBPA total protein measured by Western blot and immunofluorescence, and induced terminal granulocytic differentiation in the FLT3-ITD MOLM13 and MV4-11 AML cells, but not FLT3-WT THP1 or OCI-AML3 cells. Moreover, the proteosome inhibitors bortezomib or MG132 increased S21-P-, ubiquitinated-, and total-CEBPA protein in the FLT3-ITD but not the FLT3-WT AML cells.

We therefore further investigated interactions of endogenous CEBPA with the ubiquitin-proteosome system (UPS): unbiased analyses by immunoprecipitation (IP)-LCMS/MS demonstrated 3-4 fold enrichment of the deubiquitinase USP7 and E3 ubiquitin ligase UHRF1 in the CEBPA interactome of FLT3-ITD versus FLT3-WT AML cells. Reverse analyses of the endogenous USP7 protein interactome by IP-LCMS/MS and IP-Western blot confirmed USP7 interactions with CEBPA and UHRF1 and demonstrated USP7 phosphorylation at serine-18 (S18-P), a candidate for regulation by FLT3. The small molecule P22077 alters and inhibits USP7 catalytic site conformation: P22077 shifted USP7 deubiquitinase interactions from UHRF1 to CEBPA, decreased UHRF1 and increased total CEBPA (Figure). FLT3-ITD AML cells were >2-fold more sensitive to P22077 than FLT3-wt AML cells (IC50 5.8 and 7.1 µM for MV411 and MOLM13 versus 13.65 and 21.7 µM for THP1 and K562 cells) (Figure).

Another mystery in FLT3-ITD AMLs is relative chemo-resistance even though p53 is wild-type - USP7 regulates p53, and thus FLT3-ITD regulation of USP7 may contribute to chemo-resistance, a link we are currently consolidating.

In sum, FLT3-ITD signals for CEBPA protein degradation via USP7 and UHRF1, to thereby decouple proliferation from onward lineage-maturation in committed granulo-monocytic progenitors. Overall, this pathway explains FLT3-ITD: (i) origin in committed granulo-monocytic progenitors; (ii) link to suppressed CEBPA-target genes; (iii) mutual exclusivity with bi-allelic CEBPA mutations; (iv) differentiation-syndrome upon treatment with FLT3-inhibitors; (v) sensitivity to UPS inhibitors; (vi) mutual exclusivity with TP53-mutations yet chemo-resistance, and can inform understanding of how other receptor tyrosine kinase mutations, e.g., KIT mutations, transform committed lineage-progenitors.

Saunthararajah:EpiDestiny: Consultancy, Current equity holder in private company, Membership on an entity's Board of Directors or advisory committees, Patents & Royalties: intellectual property with royalty rights ; Novo Nordisk: Consultancy.

Author notes


Asterisk with author names denotes non-ASH members.

Sign in via your Institution