The two most common primary liver tumors—hepatocellular carcinoma (HCC) in adults and hepatoblastoma in children—share a common molecular feature: aberrant activation of the Wnt/β-catenin pathway. Somatic mutations in CTNNB1, the gene encoding β-catenin, are found in approximately 30% of HCCs, whereas deletions of exon 3 of the same gene occur in more than 80% of hepatoblastomas. These alterations prevent the degradation of β-catenin, which then accumulates in the nucleus where, by associating with the TCF-4 factor, it drives the expression of a repertoire of genes involved in metabolism and proliferation. In hepatoblastoma, induction of the DLK1/DIO3 locus—the largest cluster of non-coding RNAs in the genome, comprising in particular 54 microRNAs and several long non-coding RNAs such as MEG3—correlates with CTNNB1 mutations and is associated with a poor prognosis. However, the mechanism linking β-catenin activation to this induction remained poorly understood.
To unravel it, the authors used two mouse models that recapitulate the development of these cancers through oncogenic activation of β-catenin: one based on hepatocyte-specific loss of function of Apc (ApcΔhep), the other on deletion of exon 3 of Ctnnb1 (β-cateninΔExon3). Human liver cell lines carrying CTNNB1 mutations complemented this experimental setup. By combining in vivo CRISPR-Cas9 editing, chromosome conformation capture, ATAC-seq, and chromatin immunoprecipitation, the team identified a regulatory site located upstream of Meg3, termed DLK1-WRE (Wnt responsive element). This site, bound by β-catenin/TCF-4 complexes when β-catenin activation is sustained, then adopts an open conformation and stimulates transcription of the DLK1/DIO3 locus in both human hepatoblastomas and mouse models.
CRISPR-Cas9 editing of the DLK1-WRE site reduced expression of the locus and slowed tumor growth in three contexts: subcutaneous grafts of CTNNB1-mutated tumor cells, ApcΔhep hepatoblastomas, and β-cateninΔExon3 HCCs. The mechanisms underlying this inhibition differed across models: in the first case, silencing of the locus was accompanied by increased expression of FADD and cleavage of caspase-3, reflecting reactivation of apoptosis; in the other two, it led to decreased expression of cell-cycle effectors regulated by the FoxM1 factor.
This work thus establishes that the DLK1/DIO3 locus is an essential determinant of FoxM1-dependent cell proliferation during β-catenin-driven hepatic tumorigenesis. According to the authors, targeting the DLK1-WRE enhancer to silence this locus could represent a promising therapeutic strategy to limit the growth of primary liver cancers harboring CTNNB1 mutations.