An evolutionarily conserved PTEN-C/EBPalpha-CTNNA1 axis controls myeloid development and transformation.

Loss of function of tumor suppressor genes, such as PTEN, CEBPAlpha, and CTNNA1 (encoding the alpha-catenin protein), has been found to play an essential role in leukemogenesis. However, whether these genes genetically interact remains largely unknown. Here, we show that PTEN-mammalian target of rapamycin signaling acts upstream to dictate the ratio of wild-type p42 C/EBPalpha to its dominant-negative p30 isoform, which critically determines whether p30 C/EBPalpha (lower p42/p30 ratio) or p42 C/EBPalpha (higher p42/p30 ratio) binds to the proximal promoter of the retained CTNNA1 allele. Binding of p30 C/EBPalpha recruits the polycomb repressive complex 2 to suppress CTNNA1 transcription through repressive H3K27me3 modification, whereas binding of p42 C/EBPalpha relieves this repression and promotes CTNNA1 expression through activating H3K4me3 modification. Loss of Pten function in mice and zebrafish induces myelodysplasia with abnormal invasiveness of myeloid progenitors accompanied by significant reductions in both wild-type C/EBPalpha and alpha-catenin protein. Importantly, frame-shift mutations in either PTEN or CEBPA were detected exclusively in the primary LICs with low CTNNA1 expression. This study uncovers a novel molecular pathway, PTEN-C/EBPalpha-CTNNA1, which is evolutionarily conserved and might be therapeutically targeted to eradicate LICs with low CTNNA1 expression.