Extended Data Fig. 8: Models and readouts of GPR183 function in the cancer-lymphocyte interplay.

a, Expression of GPR183 ligand synthesis genes in a collection of 70 breast cancer cell lines. Left: The top bars stratify breast cancer cell lines by the GPR183 ligand synthesis genes they express (bottom rows). The number of breast cancer cell lines expressing the pertinent gene is shown in the bottom-left bars. Right: Expression (log1p TPM) of the GPR183 ligand synthesis genes CYP27A1, CYP7B1, and CH25H, shown separately for the DCIS/non-cancerous (n = 6) and invasive breast carcinoma cell lines (n = 64). P-values were quantified via a one-tailed unpaired Student’s t-test, comparing the expression of the gene (log1p-TPM) in the two groups. Boxplots middle line: median; box edges: 25^th and 75^th percentiles; whiskers: most extreme points that do not exceed ± IQR x 1.5; further outliers are marked individually with circles (minima/maxima). b, Flow cytometry plots of primary human NK and CD8 T cells isolated from PBMCs. Left: NK cell purity assessed based on the T cell marker CD3 and NK cell marker CD56. Right: CD8 T cell purity assessed based on CD8a and human TCRab. c, Retroviral vectors used to express either control-GFP or human GPR183-GFP constructs (top). Representative flow cytometry analysis of primary NK cells after transduction (bottom). d, Flow cytometry plots show the gating strategy to detect and quantify the relative abundance of injected NK-92 cells in the tumor. Only live (7-AAD negative) cells were considered. mCD45⁻hCD45⁺ were identified as injected NK-92 cells in the NSG mice. mCherry⁺/GFP⁺ ratio was calculated. mCD45, mouse CD45; hCD45, human CD45; GFP, green fluorescent protein. e, Fold increases in IFNγ (left) and CD107a (middle) expression at baseline and upon PMA/Ionomycin stimulation, and cytotoxicity against breast cancer cells, in GPR183^OE CAR T cells (right), primary NK cells (pNK), and NK-92 cells relative to their respective controls (n = 3 CAR T cell donors per group, n = 2 pNK cell donors per group, and n = 6 technical replicates per group for NK-92 cells). Statistical significance was determined via paired two-sided t-tests for CAR T and pNK cells (pairing donors), two-sided unpaired t-tests for NK-92 cytotoxicity, and two-sided mixed effect tests for NK-92 IFNγ and CD107a production, testing whether GPR183^OE cells significantly diverge from the matching control ones.

Source data