Back

Incomplete developmental silencing of cancer-testis antigen BORIS in humanized mouse model promotes cancer susceptibility

Price, E.; Pugacheva, E.; Bhatt, D. N.; Ji, Y. J.; Yeboah, S.; Scott, A.; Loukinov, D.; Lobanenkov, V.

2025-12-26 cancer biology
10.64898/2025.12.24.696438 bioRxiv
Show abstract

Cancer-testis antigens are genes normally restricted to the germline but aberrantly activated in many cancers, where their role in tumorigenesis remains unclear. BORIS (CTCFL), a testis-specific paralog of the chromatin organizer CTCF, is one such antigen that shows limited conservation between humans and mice, complicating in vivo functional studies. Here, we generated humanized mice in which both alleles of the endogenous Boris locus are replaced with the full-length human BORIS gene, including its highly diverged cis-regulatory elements. These fully humanized mice maintain normal fertility, indicating accurate germline expression and preserved BORIS function despite evolutionary divergence. However, unlike the strictly testis-specific expression of mouse Boris, human BORIS escapes complete somatic silencing, producing mosaic expression in a minority of mouse somatic cells. This ectopic expression is associated with reduced survival, increased tumor incidence and a shift of tumor spectrum toward aggressive lymphomas. Transcriptomic and chromatin profiling revealed that ectopic human BORIS reactivates testes-specific genes in the soma, including regulators of meiosis and DNA repair, through direct chromatin binding. This transcriptional reprogramming was consistent across tissues and clonal cell lines, revealing a dominant tissue-independent gene activation program. These findings demonstrate that when human BORIS escapes epigenetic silencing in somatic cells, it connects aberrant germline gene activation with increased cancer susceptibility in vivo.

Matching journals

The top 6 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.