Research ArticleCANCER

Lymphangiogenesis-inducing vaccines elicit potent and long-lasting T cell immunity against melanomas

See allHide authors and affiliations

Science Advances  24 Mar 2021:
Vol. 7, no. 13, eabe4362
DOI: 10.1126/sciadv.abe4362
  • Fig. 1 Irradiated VEGF-C–overexpressing tumor cells induce local lymphangiogenesis, naïve T cell infiltration, and enhanced lymphatic transport to the dLNs.

    (A to F) Lethally irradiated B16-Ctrl or B16-VEGFC cells, either OVA-expressing or not, were injected intradermally, and the skin from the injection sites was analyzed 8 days later. (A) Representative images of skin sections immunostained for Lyve-1 (lymphatic vessels, red), CD3 (T cells, green), and 4′,6-diamidino-2-phenylindole (DAPI) (nuclei, blue). Scale bars, 50 μm. (B) Flow cytometry–based quantification of lymphatic and blood endothelial cells (LECs and BECs). (C) VEGF-C concentration over time by enzyme-linked immunosorbent assay (ELISA). (D) CCL21 concentration at day 8 by ELISA. (E) Representative flow cytometry plots of T cell subsets, gated on total CD8+ or CD4+ T cells. (F) Frequencies of total CD8+ and CD4+ T cells (left) and relative fractions of subsets: naïve = CD62L+ CD44, CM (central memory) = CD62L+ CD44+, EM (effector and effector memory) = CD62L CD44+. (G and H) Seven days after irradiated tumor cell injection, 0.5-μm fluorescein isothiocyanate–labeled beads were injected intradermally in the same spot and, after 24 hours, injection sites and dLNs were analyzed by flow cytometry. (G) Percentages of bead-positive cells within each indicated antigen-presenting cell (APC) subset. (H) Frequencies of APC subsets in the dLNs and injection sites. Legend in (B) applies to the whole figure. All experiments were done in duplicate with n = 4 each. Values are reported as means ± SEM. *P < 0.05, **P < 0.01, and ***P < 0.001 using a two-tailed Student’s t test.

  • Fig. 2 Naïve T cells can be activated in situ in VEGFC vax injection sites.

    (A) Experimental design: Irradiated (irr) B16-OVA-Ctrl or B16-OVA-VEGFC cells were injected intradermally into the back skin, and IMQ was applied on the injection site 7 and 9 days later. On day 7, mice received an adoptive transfer of naïve CD8+ T cells isolated from OT-1 and pmel transgenic mice, labeled with either CFSE or CellTrace fluorescent dyes, and daily administration of FTY720 [intraperitoneally (i.p.)] was started on the same day. Mice were euthanized on day 11 for analysis. (B and C) Activated (proliferated) OT-1 and pmel CD8+ T cells in the (B) blood (as % CD45+ cells) and (C) dLNs (as total numbers). (D) Representative flow cytometry plots for OT-1 and pmel CD8+ T cells in the injection sites after gating on total CD8+ T cells. (E) Numbers of activated OT-1 and pmel CD8+ T cells in the vaccine injection site. (F) Ratios of activated OT-1 and pmel CD8+ T cells (expressed as % of total CD8+ T cells) in vaccine injection sites versus draining LNs. Pooled data from two independent experiments (n = 9 to 10). Data are reported as means ± SE. *P < 0.05, **P < 0.01, and ***P < 0.001 using Kruskal-Wallis with Dunn’s posttest. ns, not significant.

  • Fig. 3 VEGF-C–driven lymphangiogenesis boosts vaccine-induced T cell immunity.

    (A) Schematic of the immunization protocol. Irradiated B16-Ctrl or B16-VEGFC cells were injected intradermally on day 0 (d0). On days 4, 6, 8, and 10, IMQ was applied onto the skin, and on days 8 and 10, MB-αCD40 was injected intradermally, both at the cell injection site. Mice were euthanized at day 17, and splenocytes were restimulated ex vivo against the indicated antigens. (B) Frequencies of IFN-γ–producing splenocytes measured by ELISPOT. (C) IFN-γ and IL-2 secretion measured by ELISA. (D to F) Mice were immunized with VEGFC vax as shown in (A) but also received intraperitoneal injections of mF4-31C1 (αVEGFR-3) or control immunoglobulin G antibodies every 3 to 4 days. (D) Representative images of skin sections at day 17 immunostained for Lyve-1 (red) and DAPI (blue). Scale bars, 50 μm. (E) Frequencies of IFN-γ–producing splenocytes measured by ELISPOT. (F) IFN-γ and IL-2 secretion measured by ELISA. (G to I) Mice were immunized as in (A) but with irr BP-Ctrl or BP-VEGFC cells rather than B16. (G) Schematic of the immunization protocol. (H) Frequencies of IFN-γ–producing splenocytes measured by ELISPOT. (I) ELISA quantification of IFN-γ and IL-2 secretion. (J and K) Mice were immunized as in (A), and ex vivo T cell reactivity was tested against four different melanoma-associated peptides at day 17. (L) Frequencies of antigen-specific IFN-γ–producing T cells (after subtracting unstimulated control wells) by ELISPOT. (M) Breadth of reactivity against each antigen for each individual mouse expressed as number of antigen-specific IFN-γ spots. Data from one of two repeated experiments, shown as means ± SE. *P < 0.05, **P < 0.01, and ***P < 0.001 by Welch’s analysis of variance (ANOVA) with Dunnett’s T3 posttest (B, H, and I), one-way ANOVA with Tukey’s posttest (C, E, and F), Mann Whitney test (J), or paired Student’s t test.

  • Fig. 4 VEGFC vax provides complete protection against melanoma challenge and long-term immunological memory.

    (A) Experimental design. Mice were vaccinated intradermally with Ctrl Vax, VEGFC vax, or GVAX over the right shoulder, and on day 17, B16-F10 cells were injected intradermally on the contralateral side. Ctrl vax and VEGFC vax were composed of irradiated tumor cells (B16-Ctrl or B16-VEGFC) plus IMQ and MB-αCD40, while GVAX consisted of irradiated GM-CSF–overexpressing B16-F10 without additional immune adjuvants. Tumor growth was recorded, and mice that rejected tumors were rechallenged 320 days later with B16-F10 cells intradermally. (B) Frequencies of circulating Trp2-specific CD8+ T cells at day 16 assessed by pentamer staining. (C) Individual tumor growth curves showing ratios of mice with complete tumor rejection following the first challenge. (D and E) Survival curves following the (D) first and (E) second tumor challenges. In (E) naïve mice were used as positive controls for tumor growth. (F to H) The above experiment was repeated (without second challenge) to further compare GVAX supplemented with IMQ and MB-αCD40 (GVAX + adjuvants) against VEGFC vax. (F) Individual tumor growth curves and ratios of mice with complete tumor rejection. (G) Survival curves. (H) Frequencies of circulating Trp2-specific CD8+ T cells at day 19. Data shown are from (B to E) one of two repeated experiments or (F to H) one experiment, n = 7 to 11 mice per group. *P < 0.05, **P < 0.01, and ***P < 0.001 using Kruskal-Wallis with Dunn’s posttest (B and H) or log-rank test (D, E, and G).

  • Fig. 5 VEGFC vax combined with PD-1 blockade delays the growth of preexistent B16 melanomas.

    (A) Treatment schedule: Mice were inoculated intradermally with B16-F10 tumor cells and then therapeutically vaccinated on the contralateral side with VEGFC vax or Ctrl vax according to the protocol described, starting 2 days following tumor injection. αPD-1 blocking antibodies were administered intraperitoneally every 3 to 4 days starting from day 6, for a total of four injections. (B) Average tumor growth curves (means ± SEM). (C) Individual tumor growth curves. (D) Survival curves. Shown are representative data from one of two repeated experiments with n = 7 to 9 mice per group. *P < 0.05 and **P < 0.01 using (B) two-way ANOVA with Sidak’s multiple comparisons test or (D) log-rank test.

  • Fig. 6 Proposed model for VEGFC vax mechanism of action.

    Lethally irradiated tumor cells, injected intradermally, undergo radiation-induced cell death and provide a source of tumor-associated antigens. When transduced to overexpress VEGF-C, these irradiated cells also activate local lymphatics to undergo proliferation and increase antigen transport to the dLNs. In addition, VEGF-C stimulates lymphatics to secrete increased levels of chemokines that modulate the immune infiltrate in the vaccine site, particularly CCL21, which recruits naïve T cells and APCs. Later, when immune adjuvants are locally administered, APCs are further recruited to the vaccine site and activated, allowing presentation of tumor cell–derived antigens to T cells and in situ priming. Together with enhanced LN priming due to increased antigen transport, this supports a more robust and long-lasting antitumor immune response.

Supplementary Materials

  • Supplementary Materials

    Lymphangiogenesis-inducing vaccines elicit potent and long-lasting T cell immunity against melanomas

    Maria Stella Sasso, Nikolaos Mitrousis, Yue Wang, Priscilla S. Briquez, Sylvie Hauert, Jun Ishihara, Jeffrey A. Hubbell, Melody A. Swartz

    Download Supplement

    This PDF file includes:

    • Figs. S1 to S5

    Files in this Data Supplement:

Stay Connected to Science Advances

Navigate This Article