Back

Loss of the MLL3 tumor suppressor accelerates breast tumor onset via HIF1a-induced CCL2-mediated recruitment of CCR2+ regulatory T cells

Boutet, M.; Erler, P.; Nishitani, K.; Couturier, N.; Zhang, Z.; Barbieux, E.; Guillen, E.; Suzuki, M.; Lauvau, G.; Guo, W.

2022-10-05 immunology
10.1101/2022.10.02.510540 bioRxiv
Show abstract

While essential gatekeepers of immune homeostasis, Foxp3+ regulatory T (Treg) cells infiltrating tumors acquire distinct phenotypes and become highly immunosuppressive, promoting tumor immune escape and growth. How this occurs and relates to tumor-driver mutations is largely uncharacterized. Herein, we created a mouse mammary stem cell-based tumor model using CRISPR gene editing in which we introduced known human cancer-driver mutations. These included functional loss of the MLL3 histone methyltransferase and p53, and constitutive PI3-kinase activation, recapitulating the genetic makeup of aggressive breast cancers. We show that MLL3 loss fosters tumorigenesis by promoting the rapid establishment of an immunosuppressive microenvironment through induction of HIF1, which increases the secretion of the chemokine CCL2 by tumor cells and the recruitment of higher numbers of Foxp3+ Treg cells via CCR2. Greater infiltration of Treg cells also correlates with MLL3 downregulation and mutations in human breast cancer biopsies. Interestingly, HIF1 enforces the differentiation of tumor-infiltrating Treg cells into highly immunosuppressive ICOShiGITRhi Blimp-1hi effector Treg cells that enable rapid tumor escape. Monoclonal antibody targeting of ICOS or GITR inhibits tumorigenesis in most mice even two months after the cessation of treatment as well as the growth of established tumors, suggesting possible therapeutic opportunities for MLL3-mutant breast cancers.

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.