Research ArticleBIOMATERIALS

A biologic scaffold–associated type 2 immune microenvironment inhibits tumor formation and synergizes with checkpoint immunotherapy

See allHide authors and affiliations

Science Translational Medicine  30 Jan 2019:
Vol. 11, Issue 477, eaat7973
DOI: 10.1126/scitranslmed.aat7973
  • Fig. 1 Tissue-derived UBM particles inhibit tumor formation in a CD4+ T

    cell–dependent manner. (A) Scanning electron microscopy (SEM; top view and side view) of decellularized bladder (UBM) particles. UBM particles were hydrated and injected with cancer cell lines into mice to monitor the effect the UBM microenvironment on tumor formation. Tumor volume and survival in (B) B16-F10 melanoma (C57BL/6 mice; n = 5) and (C) CT26 colon carcinoma (BALB/c mice; n = 5). (D) Bioluminescence imaging of luciferase-expressing B16-F10 melanoma cells 1 to 5 days after implantation with saline or UBM in mice. (E) Histologic and macroscopic appearance of B16-F10 tumors at a similar external size (200 mm3) with saline or UBM implantation [hematoxylin and eosin (H&E) stain; 50× mosaic images left; 200× bottom right]. Tumors are denoted by arrowheads (50× mosaics) and the “Tu” label (200× images). UBM particles are enclosed by a dashed line and the “UBM” label. (F) Immunofluorescence staining for CD3+ T cells (red) and B220+ B cells (green) with 4′,6-diamidino-2-phenylindole (DAPI) counterstain (blue) after 7 days. (n = 3 animals; 200×). (G) CD4 (green) and CD3 (red) T cell costaining within the UBM microenvironment after 7 days (200×). (H) Flow cytometry quantifying the proportions of CD4+ and CD8+ T cells after 7 days (n = 5, means ± SE). (I) Flow cytometry of FoxP3 expression in tumor CD4+ T cells [and UBM fluorescence minus one (FMO)/isotype control], quantified (J) in tumors and tumor draining lymph nodes (DLNs) 7 days after injection (n = 5, means ± SE). (K) B16-F10 tumor growth and survival in wild-type (WT), Rag1−/−, and CD4+ T cell repopulated Rag1−/− mice with UBM or saline delivery (n = 5, means ± SE). Flow cytometry: *P < 0.05, ***P < 0.001, ****P < 0.0001, Student’s t test (saline versus UBM). Tumor volume: *P < 0.05 (WT saline versus WT UBM), †P < 0.05 (Rag1−/− UBM versus Rag1−/− + CD4 UBM), two-way repeated measures analysis of variance (ANOVA) with post hoc Tukey test at each time point before sacrifice. Survival: *P < 0.05, log-rank test of each group compared to WT saline with the Sidak correction (significance indicators in legend).

  • Fig. 2 T cells isolated from the UBM B16-F10 tumor microenvironment have an activated T

    H2 and myeloid-regulating phenotype. T cells were sorted from UBM or saline-delivered B16-F10 tumors for multiplex gene expression analysis using the NanoString platform. (A) Volcano plot of genes differentially regulated in UBM-associated T cells compared to saline 14 days after injection. Significantly different regulation was determined from false discovery rate–adjusted P values. (B) Normalized counts of TH2-associated (Il4 and Il13) and myeloid-regulating (Csf1 and Cd40lg) gene transcripts between 7 and 21 days after injection (n = 3 to 4 except 7-day saline pooled from three animals, means ± SD). (C) Differential expression of TH2-related, T cell activation, and cytotoxic gene sets in UBM relative to saline 14 days after injection. (D) Intracellular cytokine staining of IL-4 and IFN-γ in CD4+ T cells from saline and UBM B16-F10 tumors 14 days after injection compared to FMO controls (n = 5, means ± SE). (E) T cell (CD3+NK1.1), NKT cell (CD3+NK1.1+), and NK cell (CD3NK1.1+) density in tumors and implants (cells/mm3) 14 days after injection (n = 5, means ± SE). (F) Effect of exogenous IL-4c codelivery with saline or UBM on B16-F10 tumor formation and survival (10 μg of IL-4c per injection; n = 5, means ± SE). (G) Immunofluorescence histology of F4/80+ macrophages (green) and the nuclear proliferation marker Ki67 (red) in the UBM microenvironment after 7 days with DAPI counterstain (200×; dashed line indicates UBM implant border). Flow cytometry and gene transcript counts: *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, Student’s t test (saline versus UBM). Tumor volume: ‡P < 0.05 (saline versus UBM, saline + IL-4c, and UBM + IL-4c), two-way repeated measures ANOVA with post hoc Tukey test at each time point before sacrifice. Survival: *P < 0.05, **P < 0.01, log-rank test compared to WT saline with the Sidak correction (significance indicators in legend).

  • Fig. 3 Myeloid cell recruitment and macrophage polarization in the UBM B16-F10 tumor microenvironment is lymphocyte dependent.

    Myeloid cell infiltration was characterized 7 days after B16-F10 and UBM implantation in WT and Rag1−/− C57BL/6 mice. (A) SEM of cell infiltration in and around acellular UBM after implantation in WT mice. (B) Flow cytometry of CD11b+ myeloid cells in the UBM microenvironment and (C) the number of myeloid cells recruited to UBM in WT and Rag1−/− mice 7 days after injection (n = 5, means ± SE). (D) Flow cytometry plots of saline and UBM B16-F10 microenvironments in WT and Rag1−/− mice. (E) Myeloid cell quantification of eosinophil (Siglec-F+MHCII), granulocyte (Ly6G+), and monocyte (Ly6C+) infiltration in WT and Rag1−/− mice [saline tumors pooled from five animals; n = 5 for UBM-delivered cells (means ± SE)]. (F) Mean fluorescence intensity (MFI) of the macrophage polarization markers CD86 (M1) and CD206 (M2) in F4/80+ and CD11c+ macrophage subpopulations [saline-delivered tumors in Rag1−/− mice were pooled from five animals; n = 4 for saline delivered in WT and n = 5 for UBM-delivered cells (means ± SE)]. (G) B16-F10 tumor growth after macrophage ablation with clodronate liposomes (ClodLipo) or control phosphate-buffered saline liposomes (PBSLipo). Liposomes were delivered before B16-F10 injection and maintained during tumor growth, and tumor volume and survival were monitored [n = 5 for PBSLipo groups and n = 3 for ClodLipo groups (means ± SE)]. Flow cytometry: *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, one-way ANOVA with post hoc Tukey test. The variance for saline tumors in Rag1−/− mice was estimated as equivalent to the measurement value. Tumor volume: Two-way repeated measures ANOVA with post hoc Tukey test at each time point before sacrifice. #P < 0.05 for saline + PBSLipo versus UBM + PBSLipo, saline + PBSLipo versus saline + ClodLipo, and UBM + PBSLipo versus UBM + ClodLipo.

  • Fig. 4 Macrophages isolated from the UBM B16-F10 microenvironment have a complex polarization phenotype.

    Macrophages were sorted from UBM or saline-delivered B16-F10 tumors for gene expression analysis using the NanoString platform. (A) Volcano plot of genes differentially regulated in UBM-associated macrophages compared to saline B16-F10 tumors 14 days after injection. Significantly different regulation was determined from false discovery rate–adjusted P values. (B) Normalized counts of selected M2-associated (Arg1 and Mrc1), M1-associated (Cd86 and Cd80), and chemokine (Ccl8 and Ccl24) gene transcripts between 7 and 21 days after injection [n = 3 to 4 except 7-day saline pooled from three animals, mean ± SD). (C) Differential expression of M1- and M2-associated genes in UBM-associated macrophages 14 days after injection compared to saline delivery. (D) Immunofluorescence histology of the macrophage surface marker F4/80 (green) and the intracellular M2 marker Fizz1 (red) with DAPI counterstain. Examples of cells coexpressing F4/80 and Fizz1 are denoted with arrowheads (n = 2; 200×). (E) Differential expression of complement, angiogenesis, and cell regulation genes in UBM relative to saline B16-F10 tumor macrophages 14 days after injection (n = 4, means ± SE). Gene transcript counts: *P < 0.05, **P < 0.01, ****P < 0.0001, Student’s t test.

  • Fig. 5 Synergistic tumor inhibition with UBM and immune checkpoint blockade immunotherapy.

    B16-F10 delivery with saline or UBM was followed by treatment with monoclonal antibodies blocking PD-1, PD-L1, PD-L2, or isotype controls. (A) Individual tumor growth curves comparing the effect of anti–PD-1 treatment in the UBM microenvironment compared to saline (shaded region indicates the time range of terminal tumor growth in the UBM and isotype group). (B) Average tumor volume and (C) survival for treatments noted in (A) (n = 8 to 10, means ± SE). Arrowheads indicate treatment frequency. (D) Tumor volume when UBM or saline was delivered 1 day after B16-F10 cells, followed by anti–PD-1 treatment or isotype controls 4 days later. (E) Survival for delayed UBM implantation with anti–PD-1 treatment as noted in (D) (n = 5). (F) B16-F10 cell titration in the UBM microenvironment with anti–PD-1 treatment compared to saline (n = 5, means ± SE). Initial B16-F10 cell doses range between 1 × 103 and 1 × 106 cells per injection. Tumor volume: §P < 0.05 (UBM + isotype versus UBM + PD-1 or PD-L1), ****P < 0.0001 (all UBM treatments versus saline + isotype). For cell titration, ****P < 0.01 (UBM versus saline). Two-way repeated measures ANOVA with post hoc Tukey test at each time point before sacrifice. Survival: *P < 0.05, **P < 0.01, log-rank test with the Sidak correction.

  • Fig. 6 The UBM scaffold immune signature is associated with increased overall survival in patients with melanoma.

    (A) A UBM immune gene signature was defined by genes that were up-regulated in T cells and macrophages from UBM-delivered B16-F10 tumors compared to saline in mice. Patients with melanoma in TCGA were assigned an enrichment score on the basis of their homology to the UBM immune signature and binned into discrete categories on the basis of their relative scoring ranks: top, middle, and bottom thirds. (B) Heat map of UBM scaffold immune signature gene expression in patients with melanoma. Genes with the greatest contribution to enrichment score are listed. (C) Melanoma patient survival when grouped by patient enrichment score. **P < 0.01, log-rank test with the Sidak correction (significance indicators in legend).

Supplementary Materials

  • www.sciencetranslationalmedicine.org/cgi/content/full/11/477/eaat7973/DC1

    Materials and Methods

    Fig. S1. Biologic scaffolds from different tissue sources inhibit tumor formation but do not affect cancer cell viability.

    Fig. S2. UBM implantation does not promote tumor growth in an orthotopic breast cancer resection model.

    Fig. S3. CD4+ T cell purity in adoptive transfer experiments.

    Fig. S4. Differentially expressed genes in T cells sorted from saline and UBM tumors.

    Fig. S5. T lymphocyte characterization in the UBM-tumor microenvironment and DLNs.

    Fig. S6. Gating strategy for myeloid cell analysis with flow cytometry.

    Fig. S7. Detailed analysis of macrophage polarization.

    Fig. S8. Macrophage depletion using clodronate liposomes.

    Fig. S9. Differentially expressed genes in macrophages sorted from UBM and saline tumors.

    Fig. S10. The synthetic adjuvant material microenvironment is distinct from UBM.

    Fig. S11. Acellular UBM induces type 2 immune responses.

    Fig. S12. B16-F10 cell titration with the combination of UBM and anti–PD-1.

    Fig. S13. Tumor rejection occurs in the UBM microenvironment with anti–PD-1 treatment and leads to protection on rechallenge.

    Fig. S14. UBM dose response for tumor growth inhibition.

    Fig. S15. Gating strategy for T cell and macrophage cell sorting.

    Table S1. Sorted T cell gene expression data.

    Table S2. Sorted macrophage gene expression data.

    Table S3. Antibodies used in immunofluorescence histology.

    Table S4. Antibodies used in flow cytometry experiments.

    Table S5. Primary data.

  • The PDF file includes:

    • Materials and Methods
    • Fig. S1. Biologic scaffolds from different tissue sources inhibit tumor formation but do not affect cancer cell viability.
    • Fig. S2. UBM implantation does not promote tumor growth in an orthotopic breast cancer resection model.
    • Fig. S3. CD4+ T cell purity in adoptive transfer experiments.
    • Fig. S4. Differentially expressed genes in T cells sorted from saline and UBM tumors.
    • Fig. S5. T lymphocyte characterization in the UBM-tumor microenvironment and DLNs.
    • Fig. S6. Gating strategy for myeloid cell analysis with flow cytometry.
    • Fig. S7. Detailed analysis of macrophage polarization.
    • Fig. S8. Macrophage depletion using clodronate liposomes.
    • Fig. S9. Differentially expressed genes in macrophages sorted from UBM and saline tumors.
    • Fig. S10. The synthetic adjuvant material microenvironment is distinct from UBM.
    • Fig. S11. Acellular UBM induces type 2 immune responses.
    • Fig. S12. B16-F10 cell titration with the combination of UBM and anti–PD-1.
    • Fig. S13. Tumor rejection occurs in the UBM microenvironment with anti–PD-1 treatment and leads to protection on rechallenge.
    • Fig. S14. UBM dose response for tumor growth inhibition.
    • Fig. S15. Gating strategy for T cell and macrophage cell sorting.
    • Table S3. Antibodies used in immunofluorescence histology.
    • Table S4. Antibodies used in flow cytometry experiments.

    [Download PDF]

    Other Supplementary Material for this manuscript includes the following:

    • Table S1 (Microsoft Excel format). Sorted T cell gene expression data.
    • Table S2 (Microsoft Excel format). Sorted macrophage gene expression data.
    • Table S5 (Microsoft Excel format). Primary data.

Stay Connected to Science Translational Medicine

Navigate This Article