A scaffold vaccine to promote tumor antigen cross-presentation via sustained toll-like receptor-2 (TLR2) activation

Abstract

Cancer vaccination holds great promise for cancer treatment, but its effectiveness is hindered by suboptimal activation of CD8⁺ cytotoxic T lymphocytes, which are potent effectors to mediate anti-tumor immune responses. A possible solution is to switch antigen-presenting cells to present tumor antigens via the major histocompatibility complex class I (MHC-I) to CD8⁺ T cells - a process known as cross-presentation. To achieve this goal, we develop a three-dimensional (3D) scaffold vaccine to promote antigen cross-presentation by persisted toll-like receptor-2 (TLR2) activation after one injection. This vaccine comprises polysaccharide frameworks that "hook" TLR2 agonist (acGM) via tunable hydrophobic interactions and forms a 3D macroporous scaffold via click chemistry upon subcutaneous injection. Its retention-and-release of acGM enables sustained TLR2 activation in abundantly recruited dendritic cells in situ, inducing intracellular production of reactive oxygen species (ROS) in optimal kinetics that crucially promotes efficient antigen cross-presentation. The scaffold loaded with model antigen ovalbumin (OVA) or tumor specific antigen can generate potent immune responses against lung metastasis in B16-OVA-innoculated wild-type mice or spontaneous colorectal cancer in transgenic Apc^Min/+ mice, respectively. Notably, it requires neither additional adjuvants nor external stimulation to function and can be adjusted to accommodate different antigens. The developed scaffold vaccine may represent a new, competent tool for next-generation personalized cancer vaccination.

Keywords: Cross-presentation; Polysaccharide; Reactive oxygen species; Scaffold vaccine; TLR2 agonist.

Graphical abstract

Scheme 1

Schematic illustration of the scaffold vaccine that promotes antigen cross-presentation to trigger anti-tumor immune responses. To create optimal intracellular ROS kinetics for lysosomal disruption and antigen escape (as discovered in this study), a 3D injectable scaffold is designed for the “retention-and-release” of a TLR2 agonist glycan upon subcutaneous implantation. Installation of tunable hydrophobic interactions between the scaffold and the “hooked” agonist enables sustained TLR activation in the abundantly recruited DCs in situ, which triggers consequently efficient cross-presentation of tumor antigens. The efficacy of this novel scaffold vaccine is subsequently undergoing comprehensive evaluations in both wild-type and transgenic mice against localized, metastatic, and spontaneous tumors.

Fig. 1

The effect of intracellular ROS kinetics on the mode of antigen presentation. (A) ROS kinetics in BMDCs induced by acGM at different concentrations and frequencies. Left panel: 10 μg/mL acGM; Middle panel: 100 μg/mL acGM; Right panel: 1000 μg/mL acGM. (B) Cell viability of BMDCs following acGM treatment at different concentrations and frequencies. (C) The lysosome disruption caused by ROS. Lysosomes were stained with Lyso-Tracker (red). Nuclei were stained with Hoechst 33342 (blue). ROS signal was green. Scale bar: 20 μm. Freq.: frequency; Con.: concentration. (D) Antigen escapes from lysosomes. Lysosomes were stained with Lyso-Tracker (red). Nuclei were stained with Hoechst 33342 (blue). OVA were labeled with FITC (green). Scale bar: 20 μm. Freq.: frequency; Con.: concentration. (E) ROS induction through TLR2 activation. (F) MHC-I antigen presentation of OVA/acGM (4, 100). BMDCs were pulsed with OVA (50 μg/mL) for 15 min and then treated with acGM, followed by co-culturing with CFSE-labeled, OVA-specific CD8⁺ or CD4⁺ T cells for 48 h as described in Materials and Methods. MHC-I and MHC-II antigen presentation were determined by measuring DC-primed CD8⁺ and CD4⁺ T cell proliferation (defined as CFSE^low), respectively. (G) MHC-I antigen presentation pre-treated by bortezomib (Bzb), a proteasome inhibitor. (H) Confocal image of BMDC treated with OVA/acGM (X, Y) or Bzb + OVA/acGM (X, Y). Green signal showed the expression of 25D1.16. Red signal showed the expression of CD11c. Nuclei were stained with Hoechst 33342 (blue). Scale bar: 20 μm. (I) The effect of TLR2 activation-induced ROS kinetics on lysosome disruption and antigen processing. n = 3, values represent means ± standard deviation. Statistical significance was calculated by one-way ANOVA using the Tukey posttest. ns: no significance; *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Fig. 2

Design of the injectable, 3D scaffold vaccine for acGM “retention-and-release” in situ upon subcutaneous implantation. (A) Schematic illustration of hydrophobic modification of hyaluronic acid to enhance its interaction with acGM. (B) Interactions between HA derivatives and acGM. (C) Representative spectra excited at 550 nm, showing the effect of altering Cy3-Cy5 donor-acceptor spacing in acGM-Cy3 and Cy5 labeled HA derivatives. (D) Schematic of preparation of scaffold vaccine based on click chemistry. (E) Morphology of DOCA-HA/OVA@acGM. (F) Retention of acGM from scaffold vaccine. 1: HA/OVA@acGM; 2: C6-HA/OVA@acGM; 3: C14-HA/OVA@acGM; 4: DOCA-HA/OVA@acGM. n = 3, values represent means ± standard deviation. Statistical significance was calculated by one-way ANOVA using the Tukey posttest. ns: no significance; *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Fig. 3

The scaffold vaccine promoted DC maturation and ROS production in vivo. (A) The number of CD11c⁺ DCs recruited by HA/OVA@acGM and HA-DOCA/OVA@acGM at indicated time points after injection. (B) Maturation of CD11c⁺ DCs recruited by HA/OVA@acGM and HA-DOCA/OVA@acGM at indicated time points after injection. (C) Proportion of ROS⁺ in CD11c⁺ cells in implant. (D) Proportion of ROS⁺ in CD11c⁻ cells in implant. (E) Proportion of ROS⁺ in CD11c⁺ cells in lymph nodes. (F) Proportion of ROS⁺ in CD11c⁻ cells in lymph nodes. (G) Representative flow cytometry plots of CD8⁺ T cells and CD4⁺ T cells in CD3⁺ T cells in lymph nodes. (H) Proportion of CD8⁺ T cells and CD4⁺ T cells in CD3⁺ T cells in lymph nodes. n = 3, values represent means ± standard deviation. Statistical significance was calculated by one-way ANOVA using the Tukey posttest or student's t-test. ns: no significance; *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Fig. 4

The OVA-laden scaffold vaccine inhibited the growth of B16-OVA tumors subcutaneously localized in wide-type mice. (A) Therapeutic schedule for scaffold vaccine. Schematic of the experiment design. (B–D) Levels of anti-OVA IgG (B), IgG1 (C), and IgG2c (D) antibodies in serum were determined. (E) The ratio of IgG2c to IgG1. (F) The tumor growth curve of B16-OVA subcutaneous model. (G) Photographs of the tumors at the end of treatment. (H) Survival percentage of the B16-OVA tumor-bearing C57BL/6J mice receiving different treatments. (I) TUNEL staining of the tumor sections examined at the end of antitumor treatment. Scale bar: 100 μm. n ≥ 5, values represent means ± standard deviation. Statistical significance was calculated by one-way ANOVA using the Tukey posttest. ns: no significance; *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Fig. 5

The OVA-laden scaffold vaccine generated robust CD8⁺T cell-mediated immune responses against B16-OVA tumors subcutaneously localized in wild-type mice. (A) Representative flow cytometry analysis of the tumor-infiltrated T cells. (B) Quantitation of the percentage of CD4⁺ T cells in the tumor. (C) Quantitative analysis of Treg cells in tumor tissues. (D) Quantitation of the percentage of CD8⁺ T cells in the tumor. (E) CD8⁺ T cell staining of the tumor sections examined at the end of anti-tumor treatment. Scale bar: 50 μm. (F) Proportion of Ki67⁺ within CD8⁺ T cells. (G) Proportion of IFN-γ within CD8⁺ T cells. (H) Proportion of OVA-Tet⁺ within CD8⁺ T cells. G1: Control; G2: OVA@acGM; G3: HA/OVA@acGM; G4: DOCA-HA/OVA@acGM. n ≥ 5, values represent means ± standard deviation. Statistical significance was calculated by one-way ANOVA using the Tukey posttest. ns: no significance; *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Fig. 6

The scaffold vaccine generated lasting immune memory against metastatic lung cancer in wild-type mice. Scaffold vaccine prevent metastasis via long term immune memory effects. (A) Therapeutic schedule for scaffold vaccine to inhibit tumor metastasis. (B) Representative photographs of lung tissues. (C) Quantification of pulmonary metastasis nodes on mice pre-vaccinated with OVA@acGM, HA/OVA@acGM or DOCA-HA/OVA@acGM. (D) H&E staining of the lung tissue collected at day 14. Scale bars: 250 μm. I Survival percentage of C57BL/6J mice i.v. infused with B16-OVA cells after vaccination. (F) Flow cytometry plots of effector memory T cells in the spleen examined at the same day for i.v. infusion of the B16-OVA cells. (G) Proportions of effector memory T cells in the spleen examined at the same day for i.v. infusion of the B16-OVA cells. (H) The IFN-γ production in splenocyte supernatant after 3-day restimulation with OVA_257-264 measured by ELISA. G1: Control; G2: OVA@acGM; G3: HA/OVA@acGM; G4: DOCA-HA/OVA@acGM. (I) The tumor cells lysis by splenocyte restimulated with OVA_257-264. G1: Control; G2: OVA@acGM; G3: HA/OVA@acGM; G4: DOCA-HA/OVA@acGM. n = 5, values represent means ± standard deviation. Statistical significance was calculated by one-way ANOVA using the Tukey posttest. Ns: no significance; *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Fig. 7

The TSA-laden scaffold vaccine generated robust anti-tumor responses against spontaneous colorectal tumor in transgenic Apc^min/+mice. (A) Schematic of scaffold vaccine to prevent the progression of spontaneous colorectal tumor. (B) The IFN-γ production in supernatant of CD8⁺ T cells harvested from immunized mice and restimulated with TSA for 3 days. (C) Photographs of the tumors at the end of treatment (a) and total number of tumors in colon (b). (D) Number of tumors in different size range. (E) TUNEL staining of tumor in the colon after DOCA-HA/TSA@acGM treatment. Scale bar: 50 μm. (F) Proportion of CD8⁺ within CD3⁺ T cells. (G) Proportion of Ki67 within CD8⁺ T cells. (H) Immunostaining in the tumor at day 14 showing CD8⁺ T cells (Red), and IFN-γ (green). Scale bar: 50 μm. (I) Immunostaining in the tumor at day 14 showing CD8⁺ T cells (Red), and Granzyme B (Cyan). Scale bar: 50 μm. n ≥ 5, values represent means ± standard deviation. Statistical significance was calculated by student's t-test. ns: no significance; *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.