Back

The Y chromosome gene KDM5D restrains CD8+ T cell antitumor immunity through TCR and cholesterol-exhaustion programs

Li, J.; Ching, C. Y.; Ben-Shmuel, A.; Tallon de Lara, P.; Liu, J.; Shan, J.; Li, C.; Zhang, Z.; Wu, W. H.; Slotnik, M.; Wang, X.; Montes, R. C.; Jain, A. K.; Hornstein, N.; Zeineddine, F.; Zeineddine, M.; Woodman, S. E.; Fuentes, N. R.; Spring, D. J.; Shen, J. P.; Kopetz, S.; DePinho, R. A.

2026-08-17 cancer biology
10.64898/2026.07.23.740424 bioRxiv
Show abstract

Sex differences in immunity shape cancer risk, autoimmunity, and responses to immunotherapy, yet the sex-chromosome genes that regulate antitumor T cell function remain incompletely defined. Here, we identify the Y chromosome-encoded KDM5D histone demethylase as a male-specific suppressor of CD8+ T cell antitumor immunity. In murine colorectal cancer (CRC) models, male CD8+ T cells displayed reduced cytokine production, proliferation, cytotoxicity, TCR{beta} abundance, and proximal TCR signaling relative to female CD8+ T cells. CRISPR-RNP-mediated KDM5D depletion in male CD8+ T cells enhanced effector function, increased TCR{beta} expression, augmented TCR signaling, and improved tumor control after adoptive transfer. Transcriptomic and functional analyses further linked KDM5D to cholesterol biosynthesis and exhaustion-associated programs, with KDM5D depletion reducing SREBP2/XBP1-associated cholesterol and exhaustion signatures. Correspondingly, human CRC single-cell analyses supported the clinical relevance of this axis, showing enrichment of exhausted and cholesterol-associated CD8+ T cell states in male tumors. Pharmacologic inhibition of cholesterol biosynthesis with lovastatin partially attenuated select exhaustion-associated markers in male CD8+ T cells and delayed tumor growth in vivo. Together, these findings define KDM5D as a sex chromosome-encoded regulator of male CD8+ T cell dysfunction and point to cholesterol-exhaustion programs as a potential therapeutic vulnerability in male CRC.

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.