Fig 2: IDO1 reduces the prognostic benefit of tumor infiltrating CD8+ T cells in human ovarian cancer and reduces overall survival in a murine model of ovarian cancer. (A) Scatterplot and (B) Kaplan-Meier curves of 4 distinct populations comprised of 265 high grade serous ovarian cancer patients from The Cancer Genome Atlas (TCGA) data set. RNA-seq data was analyzed in the context of 44 genes from the tryptophan metabolism and AHR(Aryl Hydrocarbon Receptor) signaling pathways, and CD3E, CD8A, IL2, and Granzyme B. Confidence intervals for the stratified population of patients include OS (Overall Survival): black line median 42.0[38.0, 46], red line median 48.7[31.2, NA] and green line median NA[57.4, NA] p=0.02; DFS (Disease Free Survival): black line median 17.3[15.1, 19.9], red line median 18.2[13.0, 48.5], and green line median 29.9[18.2, NA] p=0.03). (C) 6- to 8- week old WT C57BL/6 mice challenged i.p. with 1x107 IE9mp1-EV (n=10) or IE9mp1-mIDO1 (n=12) tumor cells. Tumor progression was quantified by measuring the abdominal circumference of tumor-bearing mice. (D) Kaplan-Meier curves of the survival analysis of IE9mp1-EV (n=14) and IE9mp1-mIDO1 (n=15) tumor-bearing WT C57BL/6 mice. *p < 0.05, ***p < 0.001, by the Log-rank (Mantel-Cox) test (B, D), or Student’s t test (C). NA, Not available.

Fig 3: Growth delay of indoleamine 2,3-dioxygenase 1 (IDO1)-positive CT26 tumors by prophylactic vaccination with major histocompatibility complex (MHC) class I-directed or II-directed IDO1 peptides requires intact adaptive immunity and host IDO1. (A) Western blot analysis of tumors formed by six different mouse tumor-derived cell lines for comparison of IDO1 protein levels (top panel) with β-tubulin as a loading control (bottom panel). Epididymis lysates from wild-type (WT) and Ido1-/- male BALB/c strain mice were included as positive and negative controls. M designates the molecular weight marker. (B) Growth curves of CT26 tumors in mice administered single doses of five different MHC class I-directed, IDO1 peptide vaccines (EP1–5) 7 days prior to tumor engraftment. Overall responses are plotted as means±SEM together with concurrent results from both untreated and vehicle-treated animals. Growth curves for all groups are represented with gray lines except for the treatment cohort exhibiting the strongest response (EP2) and the untreated cohort which are distinguished with black lines. (n=6 tumors/cohort). (C) Growth curves of CT26 tumors in mice treated with two different MHC class II-directed, IDO1 peptide vaccines (EP6 left, EP7 right) 7 days prior to tumor engraftment. Overall responses (black lines) are plotted as means±SEM. Concurrent results from both vehicle and EP2 peptide-treated animals (gray lines) are included on each graph for comparison (n=10 tumors/cohort). (D) Growth curves of CT26 tumors in Rag1–/– mice vaccinated with the MHC class I-directed and II-directed IDO1 peptides EP2 and EP6 7 days prior to tumor engraftment. Overall responses (black lines) are plotted as means±SEM. Concurrent results from vehicle-treated animals (gray lines) are included on each graph for comparison (n≥6 tumors/cohort). (E) Growth curves of CT26 tumors in Ido1-/- mice vaccinated with EP2 and EP6 7 days prior to tumor engraftment. Overall responses (black lines) are plotted as means±SEM. Concurrent results from vehicle-treated animals (gray lines) are included on each graph for comparison (n=10 tumors/cohort). P values from longitudinal analysis of tumor growth for each peptide vaccine-treated group compared with untreated or vehicle-treated animals are included on each graph.

Fig 4: Indoleamine 2,3-dioxygenase 1 (IDO1) expression is localized to infiltrating immune cells within CT26 tumors. (A) Confocal images of CT26 tumor sections. (top row, left to right) Immunofluoresence staining of sections from wild-type (WT) and Ido1-/- mice for IDO1 (Cy3, red), and from WT mice for programmed death-ligand 1 (PDL1) (Cy3, red) and the combination of IDO1 (fluorescein isothiocyanate (FITC), green) and PDL1 (Cy3, red). Nuclei were stained on all sections (DAPI, blue). (bottom row, left to right) Staining of CT26 tumor sections from WT mice for combinations of IDO1 (Cy3, red) with CD45, CD11b, Gr1 and CD11c (FITC, green). Nuclei were stained on all sections (DAPI, blue). (B) Confocal images of a field of FACS-isolated CD45+, CD11b+, CD11c+ cells from a CT26 tumor in a WT host. (left to right) Immunofluoresence staining for nuclei (DAPI, blue), CD11c (FITC, green), IDO1 (Cy3, red), CD19 (Cy5, purple) and the composite image.

Fig 5: Major histocompatibility complex (MHC) class I-directed and II-directed responses are predominantly mediated by CD8+ and CD4+ T cells, respectively. (A) Flow cytometry characterization of splenic T-cell populations from a complete responder (CR) mouse (bottom panels) compared with a naïve mouse (top panels). CD4+ CD8- (left panels) and CD8+ CD4- (right panels) gated populations assessed for levels of CD44 (X-axis) and CD62L (Y-axis). (B) Quantitative comparison of the proportional representation of CD44hi CD62Lhi (activated) populations of CD4+ and CD8+ T cells in the spleens of CR mice relative to naïve mice identified by flow cytometry analysis as shown in (E). Graphed as means±SEM with significance determined by two-tailed Student’s t-test (n≥3 mice/group). (C) Quantitative comparison of the proportional representation of CD44hi CD62Llo (memory) populations as described for B. (D) Schematic illustration of the basic strategy for adoptive transfer of T cells from IDO1 peptide(s)+programmed cell death protein 1-binding antibody (anti-PD1 antibody)-treated, CR mice to treatment naïve, tumor-bearing mice. (E) Growth curves of CT26 tumors in WT mice that received T cells isolated 10 days following tumor rechallenge from a CR mouse previously treated with anti-PD1 antibody and the MHC class I-directed IDO1 peptide EP2. Responses to PBS (gray line) and T cells (black line) are plotted as means±SEM (n=8 tumors/cohort). (F) Growth curves of CT26 tumors in WT mice that received flow cytometry sorted T-cell subsets isolated 10 days following tumor rechallenge from a CR mouse treated with anti-PD1 and EP2. Responses to PBS, CD4+, CD8+ (black lines), and combined CD4+ and CD8+ T cells (gray line) are plotted as means±SEM (n=4 tumors/cohort). (G) Growth curves of CT26 tumors in WT mice that received flow cytometry sorted T-cell subsets isolated 10 days following tumor rechallenge from CR mice treated with either anti-PD1 and the MHC class II-directed IDO1 peptide EP6 or anti-PD1 and EP2. Responses to PBS, CD4+ and CD8+ T cells from EP6+anti-PD1 (black lines) or EP2+anti-PD1 (gray lines) treated donors are plotted as means±SEM (n=6 tumors/cohort). (H) Growth curves of CT26 tumors in WT mice that received flow cytometry sorted T-cell subsets isolated 10 days following tumor rechallenge from a CR mouse treated with anti-PD1 and EP2+EP6. Responses to PBS, CD4+, CD8+ (black lines) and combined CD4+ and CD8+ T cells (gray line) are plotted as means±SEM (n=6 tumors/cohort). (I) As described above in G except the recipient mice were Ido1-/- (n=6 tumors/cohort). P values for longitudinal tumor growth comparisons between the PBS (no cells) and different T-cell adoptive transfer cohorts are included on graphs E–I. P values for comparisons between the combined and individual CD4+ and CD8+ adoptive transfer cohorts are also included for graphs H and I. *P<0.05;**p<0.01; ***p<0.001.