Chemically Programmed Vaccines: Iron Catalysis in Nanoparticles Enhances Combination Immunotherapy and Immunotherapy-Promoted Tumor Ferroptosis

Abstract

Immunotherapy has yielded impressive results, but only for a minority of patients with cancer. Therefore, new approaches that potentiate immunotherapy are a pressing medical need. Ferroptosis is a newly described type of programmed cell death driven by iron-dependent phospholipid peroxidation via Fenton chemistry. Here, we developed iron oxide-loaded nanovaccines (IONVs), which, chemically programmed to integrate iron catalysis, drug delivery, and tracking exploiting the characteristics of the tumor microenvironment (TME), improves immunotherapy and activation of ferroptosis. The IONVs trigger danger signals and use molecular disassembly and reversible covalent bonds for targeted antigen delivery and improved immunostimulatory capacity and catalytic iron for targeting tumor cell ferroptosis. IONV- and antibody-mediated TME modulation interfaced with imaging was important toward achieving complete eradication of aggressive and established tumors, eliciting long-lived protective antitumor immunity with no toxicities. This work establishes the feasibility of using nanoparticle iron catalytic activity as a versatile and effective feature for enhancing immunotherapy.

Keywords: Biomaterials; Cancer; Immunology; Nanomaterials; Therapy.

Graphical abstract

Scheme 1

Design, Mechanism and Features of IONVs Design and mechanism of action of inherently therapeutic IONVs with chemically programmed multi-functional elements to activate and reprogram immune cells and cancer cells, harnessing the TME to sensitize tumors for immunotherapy and ferroptosis.

Figure 1

In Vivo Evaluation of the Effect of Vaccination with Different TLR Agonists and Cisplatin Chemotherapy as Adjuvants (A–C) In the prophylactic approach, C57BL/6 mice (n = 5) were immunized subcutaneously on days 0 and 14 with 5 μg of OVA free or in combination with TLR4 agonists Xcc-LOS, MPLA, LPS (1 μg/mouse), TLR3 agonist pIC (4 μg/mouse), TLR7 agonist R837 (2 μg/mouse), pIC-R837 or TLR6 agonist CpG ODNs (5 μg/mouse). Mice were challenged with 3 × 10⁵ B16-F10(OVA) cells in the right back on day 21 after first immunization. (A) Average tumor growth curves and (B) Kaplan-Meier survival curves. (C) Comparison of tumor volume of different experimental groups at day 17 after tumor challenge. The data show mean ± SEM from a representative experiment. (D–F) To exploit for cancer therapy C57BL/6 mice (n = 5) were treated on days 7, 10, and 13 after B16-F10(OVA) cells injection in the right back. Animals were subcutaneously administered with 10 μg of OVA free or in combination with pIC-R837 (4 and 2 μg, respectively), CpG ODNs (10 μg), and cisplatin (intraperitoneal, 100 μg). In this last group, cisplatin was administered with 60 μg of OVA antigen subcutaneously, in order to mimic the literature. (D) Average tumor growth curves, (E) Kaplan-Meier survival curves, and (F) animal body weight change over the experiment. Arrows indicate the days of therapy administration. Data are shown as mean ± SEM. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001 by two-tailed unpaired Student's t test (A, D, and F), by log rank (Mantel-Cox) test against OVA (SC) group (B and E), and one-way ANOVA followed by Tukey's test (C). See also Figure S1.

Figure 2

Radiolabeling and SPECT/CT Tracking of IONVs Allow Non-invasive Monitoring of Therapy Outcome (A and B) The radiocation [⁶⁷Ga][Ga(H₂O)₆]Cl₃(aq.) is loaded by heating with the anionic IONVs and following purification by ultrafiltration using a 100k membrane results in ⁶⁷Ga-doped IONVs inradiolabeling yields > 80%. (C) The chelator DOTA (present in 10⁶ molar excess relative to IONP) is able to remove <10% of the IONP-bound ⁶⁷Ga over a 24-h period. (D–F) Biodistribution study of ⁶⁷Ga labeled mIONP-CpG ODNs (23.6 μg of magnetite, 43.4 MBq) injected in the hock of melanoma tumor-bearing mice. (D) SPECT/CT images 3, 24, and 48 h post injection (coronal and sagittal views). Ex vivo analysis by (E) SPECT/CT and (F) gamma counter 48 h after injection. T, tumor; bLN, brachial LN; aLN, axillary LN; Lu, lungs; L, liver; S, spleen; K, kidneys; iLN, iliac LN; sLN, sciatic LN; inLN, inguinal LN; pLN, popliteal LN; I, injection site. See also Figures S6 and S7.

Figure 3

In Vivo Evaluation of the Therapeutic Effect Derived from the “Stealth” Hydrophobic IONP Coating and IONP Size C57BL/6 mice (n = 5) were treated on days 7, 10, and 13 after B16-F10(OVA) cells injection in the right back. (A) intrinsic adjuvant activity of mIONP was analyzed by treating animals with 10 μg of OVA antigen alone (IT) or in combination (as protein corona adsorbed by the nanoparticle) with mIONP (SC injection into the hock [ankle]; 61 μg magnetite); (B) nanoparticle size and exposure of coating was studied by injecting animals with different IONP-filled micelle designs (101 μg magnetite) peritumorally (SC, 2 cm from tumor); and (C) the effect of IONP coating nature was examined by comparing the co-administration (SC, right back) of B16-F10(OVA) cancer cells with IONPs with polar (ferumoxytol) or apolar (mIONP) coating (12 mg Fe/kg, 332 μg magnetite). (A–C) Results are shown as average tumor growth curves (upper graphs) and Kaplan-Meier survival curves for each case group (lower graphs). Arrows indicate the days of therapy administration. Data are shown as mean ± SEM. ∗p < 0.05, ∗∗p < 0.01 by two-tailed unpaired Student's t test (A–C upper graphs) and by log rank (Mantel-Cox) test against PBS/NaCl (0.9%) control group (A–C lower graphs).See also Figures S8–S10 for further in vivo studies and Figure S2 for characterization of the hydrophobic IONPs.

Figure 4

Comparison of DC Activation Elicited by mIONPs and Potent Immunostimulatory TLRa Primary culture of BMDCs was incubated overnight with increasing concentrations of mIONPs (magnetite μg/mL) and standard amounts of TLR agonists (μg/mL) diluted in media. (A) Activation and immunosuppressive marker expression measured by flow cytometry and expressed as relative mean fluorescent intensity (MFI). (B and C) (B) Cytokine production on cell culture supernatant measured by ELISA and (C) cytotoxic effect of the different stimulus after overnight incubation (MTT assay). Results are shown as mean ± SEM of three independent experiments. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001 by one-way ANOVA (A and C) and by two-way ANOVA (B) followed by Tukey's test. See also Figure S3 for characterization of mIONPs

Figure 5

Comparison of DC Activation with mIONP-OVA (Magnetite-Antigen, μg/mL) and OVA (A–C) (A) Expression of surface markers measured by flow cytometry, (B) cytotoxic effect (MTT assay), and (C) cytokine production measured by ELISA. Results are shown as mean ± SEM of three independent experiments. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗∗p < 0.0001 by one-way ANOVA (A and C) and by two-way ANOVA (B) followed by Tukey's test. See Figure S3 for mIONP-OVA characterization.

Figure 6

In Vivo Evaluation of the Therapeutic Efficacy Arising from an Optimized Antigen Nanoparticle Attachment and Adjuvant Core Material C57BL/6 mice (n = 5) were treated on days 7, 10, and 13 after 3 × 10⁵ B16-F10(OVA) cells injection in the right back with vehicle, 10 μg of OVA adsorbed by IONP (mIONP-OVA(ads)), conjugated to IONP by hydrazone linkages (mIONP-OVA(hyd)) alone or co-administered with adjuvants (mIONP-pIC-R837 or AuNP@PEI-pIC-R837 (4 μg of pIC and 2 μg of R837, 30 μg magnetite, and 5 μg of gold)). (A–C) (A) Average tumor growth, (B) Kaplan-Meier survival plot, and (C) individual tumor growth curves. CR, complete rejection. Arrows indicate the days of therapy administration. Data are shown as mean ± SEM. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001 by unpaired Student's t test between indicated groups (A) and by log rank (Mantel-Cox) test against AuNP@PEI-pIC-R837 group (B). See also Figures S3–S5 for IONV characterization, Figure S11 for AuNP@PEI characterization, and Figure S12 for further statistical analysis.

Figure 7

Comparison of Catalytic Properties of mIONPs AuNPs and Ferumoxytol (A and B) (A) Peroxidase-like activity of mIONPs at different pH values, AuNP@PEI, ferumoxytol, and released free iron cations in catalyzing the oxidation of 3,3′,5,5′-tetramethylbenzidine (TMB). (B) Catalytic oxidation of o-phenylenediamine (OPD). Concentrations of [Fe] = 150 μM for mIONPs and 550 μM for ferumoxytol were used. (C and D) Michaelis-Menten kinetics plots for the oxidation of TMB in the presence of H₂O₂ at varying concentrations of (C) H₂O₂ and (D) TMB, based on data (mean ± SEM) from three independent experiments.

Figure 8

Cell Death Triggered by mIONPs and IONVs in Cell-to-Cell Contact Co-cultures of Melanoma Cells and Macrophages and Induction of Ferroptosis (A and B) Cell viability was measured by flow cytometry following 24-h treatment (A) B16-F10(OVA) melanoma cells and (B) RAW 264.7 macrophages in cell-to-cell contact co-culture approach (77 μg/mL magnetite = 1 mM of Fe). Data are shown as mean ± SEM (n = 3). (C) Effects of Erastin on the cell viability of RAW 264.7 (green bars) and B16-F10(OVA) (blue bars) in their respective mono-cultures, after 24 h of exposure determined by MTT assay. Data were normalized to the absence of treatment. (D) Effects of mIONPs (0.5 mM of Fe) on the cell viability of B16-F10(OVA) after 24 h of exposure upon 24-h preconditioning with increasing concentrations of Erastin. Data were normalized to the absence of mIONP treatment for each Erastin concentration (solid green bars). Data are shown as mean ± SEM (n = 4). (E) Reaction of Liperfluo with hydroperoxide lipids to generate fluorescence Liperfluo-Ox. (F) B16-F10(OVA) incubated with IONPs (1 mM of Fe) for 24 h. Representative flow cytometric profiles are shown to demonstrate lipid peroxides with Liperfluo signals, where cells treated with cumene hydroperoxide are used as positive control. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001 by one-way ANOVA followed by Tukey's test (A, B) and by two-way ANOVA followed by Sidak's test (C, D). See also Figure S13.

Figure 9

In Vivo Detection of the Expression of PD-L1 and OX40 Immune Receptors by Immuno-PET Imaging and Subsequent Analysis of Infiltrating Immune Cells Allowed Characterization of TME Status (A) PET images were acquired 24, 48, and 72 h after administration of the radiolabeled anti-PD-L1 and anti-OX40 antibodies (80–100 μg, 1–7 MBq). (B–E) (B) Antibody accumulation in tumor and spleen expressed as injected dose percentage per cm³ (% ID/cm³). Analysis of immune cells phenotype found in the spleen, TME and TDLN: (C) CD4⁺ and CD8⁺ T cell percentages, (D) OX40⁺ T cell0020percentages, and (E) activation profile of CD4⁺OX40⁺ T cells. Data are shown as mean ± SEM (n = 7).

Figure 10

In Vivo Evaluation of the Enhanced Effect of Combination Immunotherapy with IONVs and Immunostimulatory mAbs C57BL/6 mice (n = 5) were inoculated with 3 × 10⁵ B16-F10(OVA) cells and treated with nanovaccines (10 μg of OVA, 4 μg of pIC, and 2 μg of R837; 42 μg of magnetite) on days 4, 11, and 18. ICIs (100 μg of anti-PD-L1 (IP) and 10 μg of anti-OX40 (IT)) were administered on days 1 and 4 after each immunization. (A–E) (A) Average tumor growth, (B) Kaplan-Meier survival plot, and (C) individual tumor growth curves. (D) Average tumor growth and (E) Kaplan-Meier survival plot following re-challenge of melanoma cells in the contralateral back of cured mice. CR, complete rejection. Arrows indicate the days of nanovaccines (black) and checkpoint inhibitors (blue) administration. Data are shown as mean ± SEM. ∗p < 0.05, ∗∗p < 0.01 by unpaired Student's t test between indicated groups (A) and by log rank (Mantel-Cox) test against mIONP-OVA(hyd) + mIONP-pIC-R837 group (B).