Fig 2: Tumor-intrinsic PD-1 does not require PD-L1 ligation to promote BTIC growth.(A and B) Representative immunofluorescence images and quantification of GBM specimens stained for PD-1, PD-L1, and SOX2. Dashed outlines correspond to areas where there is minimum PD-L1 around PD-1–positive BTICs. Quantification was performed in three to four FOVs of each patient. Each circle is of a different patient with GBM. Green, 026-1; purple, 101220; orange, 110907; blue, 100819. (C) Proliferation of BT048-OE PD-1 and control in the presence or absence of PD-L1. (D and E) Representative immunofluorescence and quantification of p65 translocation in BTICs after treatment with PD-L1 (5 µg/ml) for 1 hour. For each well of cells, four FOVs were quantified. TNF-a (100 ng/ml) was used as a control to activate p65 translocation. (F) Schematic of human PD-1 K78A version with defective PD-L1 binding. (G) Flow cytometry plots of BTICs OE wild-type or K78A PD-1. (H and I) Representative flow cytometry and bar plots of EdU assay. (J) Schematic of knocking out of mouse PD-L1 (Cd274) gene by CRISPR-Cas9 and overexpressing Pdcd1 in mBT0309. (K) Proliferation assay of Cd274 knockout (KO) clones 2-5 and 3-5 of mBT0309-OE PD-1 compared to control. (L and M) Representative images of 72- to 96-hour outcomes of tumor spheres and quantification in clones 2-5 and 3-5. Proliferation of human BTICs overexpressing PD-1 and control after exposure to (N) PD-1 (pembrolizumab) or (O) PD-L1 (avelumab) antibodies for 72 hours. Results are representative of n = 2 to 3 independent experiments. Unpaired (two-tailed) t test was used to assess significance between two groups. For more than two groups, one-way ANOVA with Tukey’s post hoc was used: *P < 0.05, ***P < 0.001, and ****P < 0.0001. ns, not significant. Data are represented as means ± SEM. See also fig. S6.

Fig 3: PD-1 on BTICs promotes intracranial tumor growth in mice.(A) Schematic diagram depicts implantation of Pdcd1-knockdown or OE cells into the striatum of mice. Kaplan-Meier analysis to assess the survival of wild-type mice i4mplanted with (B) Pdcd1 knockdown or (C) overexpressing murine BT0309. Cells transduced with lentiviral vectors coding nontarget shRNA or transfected with empty PiggyBac vector were considered as controls for knockdown (shRNA control) or OE (OE control) BTICs, respectively. (D) Representative images of immunofluorescence staining of PD-1, SOX2, and p65 in sections collected from C57BL/6 wild-type mice implanted with Pdcd1-knockdown or shRNA controls. Nuclear translocation (yellow arrowheads) and cytoplasmic localization (blue arrowheads) of p65 in BTICs are shown on the images. (E) Kaplan-Meier analysis of Rag1 KO mice implanted with Pdcd1-knockdown or shRNA controls of mBT0309. Representative T2-weighted MRI images (F) at day 50 of Rag1 KO mice implanted with Pdcd1-knockdown or (G) at day 54 of Rag1 KO mice implanted with Pdcd1 OE mBT0309 (tumor demarcated with red outlines). Quantitation of tumor volume from MRI images of mice implanted with (H) Pdcd1 knockdown or (I) OE BTICs versus respective vector controls. Results are representative of n = 2 to 3 independent experiments. Unpaired (two-tailed) t test was used to assess significance between two groups in (I). For more than two groups in (H), one-way ANOVA with Tukey’s post hoc was used. Kaplan-Meier survival curves were analyzed for statistical differences between groups using the log-rank Mantel-Cox test: ***P < 0.001 and ****P < 0.0001 compared to respective vector control. Data are represented as means ± SEM.

Fig 4: (A) Flow cytometric characterization of IFPSCs at day 14 (P2), with unstained cells serving as the negative control. The cells were characterized according to the surface antigen expression of CD90, CD105, CD166 and CD45, as well as the transcription factor Sox2 and early neuronal marker nestin. (B) Immunocytochemical analysis of IFPSCs for stem cell markers at day 14 (P2; magnification, ×20). FITC-stained images show the positive expression of the following stem cell-specific markers: (a) Nucleostemin, (b) Nanog, (c) CD166, (d) CD105 and (e) CD13, with (f) unstained cells serving as the negative control. Scale bar, 50 µm. IFPSCs, infrapatellar fat pad-derived stem cells; FITC, fluorescein isothiocyanate; PE, phycoerythrin; APC, allophycocyanin; Sox2, SRY-box 2.

Fig 5: The expression of PD-1 on BTICs in resected clinical GBM specimens in situ.Representative images of immunofluorescence staining and corresponding 3D reconstruction of images of (A and B) CD45 and PD-1, (C and D) PD-1 and BTIC marker SOX2, and (E and F) PD-1 and BTIC marker nestin in sections of human GBM specimens resected from three patients (1085, 110907, and 101029). Representative immunofluorescence images of human normal brains stained for (G) PD-1 and SOX2 or (H) PD-1 and nestin. (I) Quantification of immunofluorescence staining of PD-1+ SOX2+ or PD-1+ nestin+ cells within the tumor microenvironment. Quantification was performed in three to four fields of view (FOVs) of each patient. Each circle is of a different patient with GBM, and the color of the dots in both panels match to the same subject. Green, 026-1; purple, 101220; orange, 110907; blue, 100819; red, 110512; black, 101029; yellow, 1085. Nuclei were counterstained with 4',6-diamidino-2-phenylindole (DAPI). Data are represented as means ± SEM. See also fig. S1 and table S1.