The cold tumor population drives an overall immunosuppressive phenotype in mixed-population tumors. (A–E) Flow cytometric immune profiling of single-population (homogeneous) CIT6-YFP and CIT9-RFP tumors and 1:1 mixed-population tumors, including (A) T cells, CD4 T cells, and CD8 T cells (n = 6–7), (B) frequency of tumor cells (CD45⁻ YFP⁻ or CD45⁻ RFP⁻) expressing MHC-I molecules (H2-K^q and H2-L^qD^q) (left) and median fluorescence intensity (MFI) of the MHC-I molecules on H2-K^q and H2-L^qD^q double-positive tumor cells (right) (n = 5), (C) Treg (Foxp3⁺ CD25⁺) frequency, CD8 T cell-to-Treg ratio, and Teff (Foxp3⁻ CD25⁻)-to-Treg ratio (n = 5–6), (D) B cells (B220⁺) and NK or innate lymphoid cell (ILC) (NKp46⁺) lymphocytes (n = 6–7), and (E) macrophages (F4/80⁺ CD11b⁺ MHC-II⁺) (n = 6–7). Data in A–E are a representation of at least two independent experiments. (F–H) In vitro CD8 T cell suppression by macrophages sorted from the tumors. CFSE-stained and CD3/28 beads–stimulated CD8 T cells isolated from naïve Confetti mice were co-cultured with TAMs at varying ratios (1:4, 1:2, and 1:1 TAM:T cell ratios; 12,500, 25,000, or 50,000 TAMs with 50,000 T cells) and analyzed for proliferation via CFSE signal and activation markers. Data are a representation of two independent experiments, with three biological replicates in each experiment. (F) Representative histogram showing the CFSE signal in CD8 T cells, with gates marking divided cells defined as cells having undergone ≥2 divisions, in no TAM condition or 1:1 TAM:T cell ratio (50K T cell + 50K TAM conditions) (left), and graph showing the count of divided cells (right). “No TAM,” T cell stimulation control without macrophages; “No stim,” control with no T cell stimulation. (G) Count of activated (CD69⁺ CD25⁺) CD8 T cells. (H) Count of effector CD8 T cell subsets marked by combinations of CD62L and CD44 expression. (I) Flow cytometric analysis of T cell infiltration in mixed-population tumors derived from varying ratios of CIT6:CIT9 cell lines. n = 5–7. Statistical significance was determined by one-way ANOVA (A–E and I) or two-way ANOVA (F–H) with Tukey’s correction posttest. *P < 0.05, **P < 0.01, and ***P < 0.001.