Figure 5.

PERK–ATF4–C/EBPβ signaling reprograms splenic LSK cells to support tumor-associated myelopoiesis in the spleen. (A–F) Naive BM LSK cells were pretreated with vehicle (veh; 0.1% DMSO), PERK inhibitors (GSK, 250 nM; AMG, 5 μM), or ATF4 siRNAs before stimulation with SSN of different origins. After 4 d of culture, the proportions of HSPC subsets (A), percentages of GM-CSF⁺ cells in MPP3 (B), CFU-C activity (C), ability to generate myeloid descendants capable of suppressing CD8⁺ T cell proliferation (D and F), and antigen-specific cytotoxicity (E) were determined. (G) Cartoon depicting the gene-edited HSPC transplantation assay. (H and I) IB analysis of PERK, ATF4, and C/EBPβ isoform expression in splenic LSK cells from mice that received control, Eif2ak3-KO, or Atf4-KO HSPC transplantation. t, total. (J–M) Changes in the HSPC number (J), LSK cell constitution (K), percentage of GM-CSF⁺ cells in MPP3 (L), and frequencies of splenic myeloid cells (M). (N) Suppressive activity of Gr-1⁺ descendants generated from splenic LSK cells in vitro. (O) Suppressive activity of tumor-infiltrating PMN-like cells. (P) Percentages of tumor-infiltrating IFN-γ⁺ CTLs in CD45⁺ cells. In J–M, O, and P, n = 6–9 mice per group. (Q) Tumor growth in mice that received control, Eif2ak3-KO, or Atf4-KO, and Cebpb-KO HSPC transplantation (n = 9–10 mice per group). Error bars indicate the means ± SEM. Statistics: One-way ANOVA corrected by Dunnett’s test (J, L, M, and P) or Tukey’s test (B and E); two-way ANOVA corrected by Dunnett’s test (A, C, D, F, K, N, O, and Q). *, P < 0.05; **, P < 0.01; ***, P < 0.001. Data are from two independent experiments (J–M and O–Q), or three to four independent experiments (A–F and N; n = 3–4 samples per group, with cells pooled from three mice) or are representative of two independent experiments (H and I). SP, spleen; T, tumor. Source data are available for this figure: SourceDataF5.