Vaccines have transformed the fight against critical infectious diseases; however, developing vaccines against cancer has proved challengi1ngT.herapeutic cancer vaccines require type 1 cellular immunity4both helper type 1 (Th1 +C)D 4and cytotoxic (CD8+) T cells that recognize and kill cancer cel2ls;therefore, strategies to induce tumor-specific type 1 T icnells vivo are needed. Dendritic cells (DCs) are the immune system9s central player for initniattii-ng a tumor immunity.3 As professional antigen-presenting cells (APCs), DCs internalize pathogens and present processed epitopes on major histocompatibility complex (MHC) molecules forllT ce priming.4 Additionally, in response to pathogen engagement, DCs secrete cytokines that activate cellular (Th1 and CD8+), humoral (Th2), or tolerogenic (Treg) T cell immun5 itDy.ue to their key role in modulating immunity via T cell polarization, specific activation of indiviCdual D signaling pathways is critical for inducing tumor antigen-specific, type 1 cellulaurniimtym.

DC-based immunotherapies primarily rely oenx vivo activation and engineering of DCs to present tumor antigens for T cell primi6nEgx. vivo loaded DCs are compromised by limited migration to lymph nodes for T cell interactions and a diminished capacity to maintain proinflammatory status upon injection, which can inadvertently lead to the undesired induction of Treg and Th2 responses6. In contrast, targeting of antigens to DinCsvivo could enable the direct activation of DCs for generation of pro-inflammatory, tumor-specific immune responses. This approach would also circumvent the resource-intensievxe vivo DC generation process to enable scalable vaccine production.

DCs employ lectin recognition of cell surface glycans as the first extracellular dneatteiromn i of self and non-self. DC lectins can identify pathogen-associated glycans as foreign threats to ramp up immunity and self-associated glycans as signals to induce tolerance, preventing autoimmunity.7 However, aberrant glycosylation exhibited by cancer cells often suppresses immunity through inhibitory lectin signaling, notably exemplified by the sialic acid-Sigle.c8 axis The ability of lectins to shape DC signaling and modulate downstream immunity make them interesting targets for immunomodulation9,.10 The dendritic cell-specific surface lectin DC-SIGN (CD209) binds mannose or fucose-substituted oligosaccharides on numerous exogenous pathogens with distinct immune outcome11s-.13 While DC-SIGN binding facilitates efficient pathogen uptake, engagement of DC-SIGN alone may induce regulatory responses and immune evasion.14-16 Lectins are therefore pivotal regulators of DC immunity, but they do not act alone.

Toll-like receptors (TLRs) bind a broader range of viral and bacterial molecular structures determined by their location on the cell surface or within endosomes of DCs. The endosomal

TLR7 binds viral and microbial nucleic aciidnsc,reasing formation of MHC-peptide complexes, upregulation of costimulatory molecules (CD80, CD86, CD40), and secretion of IL.-117-219 TLR7 agonists and TLR combinations have been studied as vaccine adjuvants and cancer therapeutics,20-23 but challenges in achieving suitable efficacy while avoiding toxicity remain a significant barrier to their us2e4.While nonselective lectin-targeting has been used to facilitate uptake of TLR agonists by APCs, the reported results have varying degrees of success, potentially due to poorly understood immune outcomes of engaging both lectins and25T-2L9Rs. As co-delivery of ligands for lectins and TLRs has the potential to switch immune outcomes from regulatory to immunostimulatory30, we envisaged that selective dual lectin-TLR signaling could be leveraged to activate immunity. We recently reported an aryl mannoside ligan)d (Ma that selectively binds to DC-SIGN and found that synergistic engagement of DC-SIGN and TLR7 by Man-functionalized virus-like particles (VLPs) resulted in Th1 sign3a1linTgh.e chemical structure of the mannose ligand was critical for DC-SIGN signaling. Man bound DCSIGN with higher avidity compared to a trimannose ligand at lower pH, a feature designed to afford enhanced engagement of DC-SIGN and TLR7 in the endosome for robust DC activation. Given that type 1 cellular immunity is critical for therapeutic cancer vaccinaecye,ffwice envisioned that delivering tumor antigens within this framework could generate tumor-specific

Given that specific lectin-TLR dual activation is poorly understood, we sought to engineer VLPs that could specifically engage DC-SIGN and TLR7 to elicit robust DC activationP.s VL are non-replicating nanostructures comprised of self-assembling virus-derived coat protein monomers.32 Encapsulation of single-stranded RNAs (ssRNAs) from the bacterial expression host enable TLR7 signaling for immune activati3o3nV.LPs have been successfully used to induce humoral immunity comprising long-lasting IgG antibody respon3s4-e3s9, but engineering

VLPs to induce effective cellular immunity remains challenging. We previously reportted tha VLPs engineered from bacteriophage Q³ to present a DC-SIGN-selective, pH-independent aryl mannoside ligand and bearing genetically encoded OVA(323-339) (OvaII) peptides could induce antigen-specific Th1 CD4+ T cellsin vivo.31 Motivated by these results, we hypothesized that tumor antigen delivery in this context could induce both tumor antigen-specific Th+1 aCnDd4 CD8 + T cells, providing holistic type 1 cellular immunity. However, analogous Q³ VLPs bearing a similar density of Man and genetically encoded OVA(257-264) (OvaI) peptides towards induction of antigen-specific CD8+ T cells were not accessible, as the internal lysine residue of OvaI would likely be acylated in the course of particle conjugation. Attempts to chyemicall derivatize Q³ VLPs with high densities of glycans and peptides (>540 Man and >40 peptides per particle) were not successful, since these particles were prone to irreversible agigorne.gat

To maximize the loading of peptide antigens on VLPs, we turned to the bacteriophage PP7PP7 platform, which has a higher density of accessible amines for functionalization (~1200 per particle), and a higher tolerance for peptide incorporat4i0onT.o enable particle loading with diverse peptide antigens, including peptides containing nucleophilic amino acei.dgs., (lysine, cysteine), we designed a two-stage modification strategy. Amine-reactive esterrse wuseed to install a low density of fluorophores and a high density of azides on VLPs. The alkynecontaining aryl mannoside ligand was synthesized by a facile, three-step synthesis (Supplementary Scheme 1). Following particle purification, the intermediate paersticwlere then coupled to alkyne-containing glycomimetic or control ligands (Man, PE) and peptide antigens (OvaI, OvaII) via a copper(I)-catalyzed Huisgen azide3alkyne cycloaddition (CuAAC), a powerful click reaction (Fig. 1c41),.42 The model tumor antigens OvaI and OvaII were modified with a cathepsin D-sensitive linker to facilitate release by endosomal protceoplryoticessing and loading on MHC molecules4.3 The extent of ligand and antigen functionalization on each particle was controlled by sequential addition of peptide-alkynes and Man-/PE-alkynes. The resulting conjugates were analyzed by LC-ESI-TOF-MS to determine the average loading density of each functional unit (Supplementary Fig. 1) and by DLS and FPLC to confirm particle integrity (Supplementary Fig. 2,3), demonstrating that all particles displayed equivalent ligand loadings and size (~50 nm diameter). Thus, we generated PP7 particles displayi n10g00 P(VELP-PE), PE900-OvaI100 (VLP-PE-OvaI), PE900-OvaII100 (VLP-PE-OvaII), Ma1n000 (VLP-Man), Ma9n00OvaI100 (VLP-Man-OvaI), and Ma9n00-OvaII100 (VLP-Man-OvaII) (Supplementary Table 1,2). We hypothesized that particles constructed in this manner could enhance DC activation via costimulatory signaling by dual mannose-DC-SIGN and ssRNA-TLR7 interactions to achieve antigen-selective immunity. Additionally, we postulated that combinations of paersticdlelivering MHC-I and MHC-II epitopes (e.g., VLP-Man-OvaI/II) could induce both CD+4and CD8+ T cells, where Th1 CD 4+T cells could help augment C D+8T cell proliferation, for complete and robust anti-tumor immunity.

Receptor recognition and signaling facilitates DC activation and antigen uptake for efficient T cell priming4. To investigate the efficiency of DC-SIGN-mediated particle uptake and DC activation, we compared the uptake of mannosylated VLPs against control particles. Fluorophore-labeled VLPs were added to monocyte-derived dendritic cells (moDCs), and VLP uptake was monitored by flow cytometry. DC uptake of VLP-Man-OvaI/II was significantly higher (5-fold) than that of VLP-PE-OvaI/II over 30 min (Fig. 2a,b), demonstrating that conjugation of tumor antigens did not interfere with the glycan-lectin interactions. To intveestiga the endocytic role of DC-SIGN over other mannose-binding lectins, moDCs were pre-treated with blocking antibodies against DC-SIGN, mannose receptor (MR), Dectin-2, or Mincle and then exposed to VLPs. Pre-treatment with DC-SIGN4but not MR, Dectin-2, or Mincle4 resulted in a significant decrease in VLP-Man-OvaI/II uptake, confirming that DC-SIGNtheis major endocytic receptor for VLP-Man-OvaI/II (Fig. 2c). These data indicate that VaLnP--M OvaI/II uptake occurs primarily by DC-SIGN-mediated endocytosis.

We envisioned that the endocytic ability of DC-SIGN would facilitate delivery of VLPencapsulated ssRNAs to endosomal TLR474. To assess co-engagement of DC-SIGN and TLR7, we performed confocal microscopy on moDCs stimulated with VLPs and labeled with anti-DCSIGN and anti-TLR7 antibodies. Confocal images revealed that VLP-Man-OvaI/II was efficiently endocytosed by moDCs within 30 min whereas VLP-PE-OvaI/II was present only at low levels in the cells, corroborating the enhanced uptake of mannosylated VLPs ausremdebays flow cytometry. Additionally, colocalization of only VLP-Man-OvaI/II with both DC-SIGN and TLR7 was observed (Fig. 2d), further confirming DC-SIGN and TLR7 engagement by VLPMan-OvaI/II.

To assess DC activation by VLPs, moDCs were incubated with VLP-Man-OvaI/II or- VLP PE-OvaI/II for 24 h. VLP-Man-OvaI/II induced more robust DC activation than VLP-PEOvaI/II, as demonstrated by the significant increase in surface expression of activaatiroknersm and costimulatory molecules CD40, CD80, CD83, and CD86 (Fig. 2e). Taken together, the mannosylated VLPs target DCs via binding to the cell surface lectin DC-SIGN, fatcinilgita highly efficient endocytosis and engagement of TLR7 for robust DC activation. incubated with 8 nM AF647-labeled VLPs for 30 min, and uptake of VLP by moDCs was measured by + flow cytometry. b, Mean fluorescence intensity of AF647-labeled VLPs in CD11c moDCs indicated at 30 min. c, moDCs were pre-treated with blocking antibodies against lectins for 20 min and then incubated with 8 nM VLP-Man-OvaI/II for 15 min. The uptake of VLPs was assessed by flow cytometry. Relative uptake was calculated by normalization against a control sample without pretreatment of blocking antibodies. d, Confocal microscopy images showing VLP localization in moDCs after 1 h incubation with 8 nM VLPs. Green: AF405 anti-DC-SIGN; Pink: AF488 anti-TLR7; Cyan: AF647-VLP. Scale bars, 5 µm. e, Activation of moDCs by VLPs. moDCs were incubated with 8 nM VLPs for 24 h, and the expression levels of activation surface markers CD40, CD83, CD86, and CD80 were measured by flow cytometry. Data represent mean +/- standard error of the mean (s.e.m.) from a representative experiment (n=3). Statistical analysis was performed using one-way ANOVA with

Bonferroni9s multiple comparisons test. *P<0.1, **P<0.01, ***P<0.001, ****P<0.0001. Each experiment was repeated at least twice with similar results.

To better understand the basis for the enhanced upregulation of activation markers observed upon stimulation with mannosylated VLPs, we performed bulk RNA-seq experiments on moDCs treated with VLP-Man-OvaI/II and VLP-PE-OvaI/II or unstimulated control moDCs. A total of 486 genes were differentially expressed after treatment with VLP-Man-OvaI/II vs. VLP-PEOvaI/II (log2FC >1.5 and p<0.05) (Fig. 3a). The significantly upregulated genes included CXCL1, CXCL5, ISG15, IL6, CCL4, ISG20, andSOCS3, all of which are implicated in the TLR7 downstream signaling pathway4,5 suggesting a higher extent of TLR7 activation with the mannosylated VLPs (Fig. 3b). Gene set enrichment analysis (GSEA) between VLP-ManOvaI/II- and VLP-PE-OvaI/II-treated moDCs revealed that hallmark DC activation pathways were also upregulated in VLP-Man-OvaI/II-treated DCs, including pathways related tNo-aI F, IFN-g, and TNFa- signaling via nuclear factor-»B (NF-kB) (Fig. 3c).

To assess downstream cytokine secretion, moDCs were incubated with VLPs for 32 h and the supernatant was collected for analysis. The moDCs exposed to VLP-Man-OvaI/IrIetesedc significantly higher levels of type 1, pro-inflammatory cytokines, including aT,NFIL--6, CXCL10, IL-8, and IFN l1, compared to those treated with VLP-PE-OvaI/II (Fig. 3d). Notably, moDCs pre-treated with blocking antibodies against DC-SIGN showed a significant reduction in TNF-a production following VLP-Man-OvaI/II addition (Fig. 3e). This finding shows that engagement of DC-SIGN and TLR7 by VLP-Man-OvaI/II triggers a signaling cascade leading to type 1, pro-inflammatory cytokine secretion.

Mannosylated VLPs are taken up preferentially by DCs in vivo

Motivated by the efficient uptake of mannosylated VLPs in moDCs, we examined the uptake of VLPs by DCsin vivo. Trafficking of vaccines to the lymph nodes (LNs) is a key step for efficient uptake by DCs and antigen presentation to T cells. Particles on the order of 20-200 nm have been shown to drain directly to LNs within hours of subcutaneous (s.c.) administration.44 The size of PP7 VLPs (~50 nm diameter) is therefore ideal for entering the lymphatic system, rendering them capable of immune priming even when administered at sites distal from both the lymph vessels and the tu4m4oTr.o assess the trafficking of the VLPs,

C57BL/6J mice were immunized s.c. with AF647-labeled VLPs, and fluorescence intensity in the isolated organs was quantified. We observed selective accumulation of VLPs in the immune cell-rich LNs within 4 hours and clearance through the liver over the course of 72 hours (Supplementary Fig. 4a,b). We also inoculated mice with B16F10-OVA cells and allowed tumors to grow to ~200 m m3 followed by s.c. administration of VLPs and measurement of VLP fluorescence intensity within the tumor over time. No localization of the VLPs withuimnotrhe t was observed, suggesting that there is no direct VLP activity within the tumor. (Supplementary Fig. 4c). Active targeting with mannoside ligands did not significantly alter VLP distribution to measured organs at 24 h (Supplementary Fig. 4d). Instead, mannosylation of VLPs enhanced their uptake by LN-resident DCs. VLP-Man-OvaI/II uptake was significantly higher than VLPPE-OvaI/II uptake by DCs in the LNs after 24 h (Supplementary Fig. 4e,f). We hypothesized that this increased VLP-Man-OvaI/II uptake in combination with greater DC activation would facilitate more robust anti-tumor immunity. between VLP-Man-OvaI/II- and VLP-PE-OvaI/II-treated moDCs. moDCs were treated with 8 nM VLPs for 6 h. b, TLR7 downstream gene signature module score for unstimulated, VLP-PE-OvaI/II-, and VLP-Man-OvaI/II-treated moDCs. c, GSEA analysis of upregulated and downregulated pathways in DCs treated with VLP-Man-OvaI/II and VLP-PE-OvaI/II. d, The moDCs were treated with VLPs for 32 h and supernatants were collected for cytokine analysis. e, The moDCs were pre-treated with blocking antibodies against DC-SIGN for 20 min followed by stimulation with VLP-PE-OvaI/II or VLP-Man-OvaI/II for 32 h, and TNF-a secretion was measured. Data represent mean ± s.e.m. from a representative experiment (n=3). Statistical analysis was performed by one-way ANOVA with Bonferroni9s multiple comparisons test. *P<0.1, **P<0.01, ***P<0.001, ****P<0.0001. For d and e, experiments were repeated at least twice with similar results. + + Mannosylated VLPs induce antigen-specific Th1 CD4 and CD8 T-cell responses in vivo

To evaluate tumor antigen-specific cellular immune responses, we examined the immunomodulatory effects of the VLPs in an immune-competent and syngeneic mouse transplant using the aggressive B16F10-OVA murine melanoma model. This solid tumor model expresses ovalbumin antigens and is non-responsive to anti-PD-1 treatment. We first invedstigate the ability of VLP-Man-OvaI/II to induce tumor antigen-specific +CDa4nd CD8+ T cells that mediate tumor growth inhibition. A dinucleotide agonist for STING, 2232 cyclic GMP-AMP (cGAMP), was used as an adjuvant to boost T cell activ4a6,ti4o7na,nd the immune checkpoint inhibitor anti-PD-1 was used so that cytotoxic capacity of T cells could be directly measured. Six-week-old C57BL/6J mice were inoculated with B16F10-OVA cells, followed by three immunizations (at 6-day intervals) with VLPs, 2939-cGAMP, and anti-PD-1 (Fig. 4a,b). On day 21 (6 days following the last dose), spleens and tumors were isolated for immune cell analysis.

We first measured the ability of mannosylated VLPs conjugated with tumor antigens to induce tumor-antigen specific Th1 CD+4and CD8+ T cells. To this end, we compared T cell induction by VLP-Man, VLP-Man-OvaII, VLP-Man-OvaI, and a combination of both VLPs (VLP-Man-OvaI/II) (Fig. 4a). Immunization with VLP-Man-OvaII and VLP-Man-OvaI resulted in increased percentages of CD+4and CD8+ T cells, respectively, compared to immunization with VLP-Man only (Fig. 4c). Interestingly, immunization with the combination particPl-es VL Man-OvaI/II led to a significantly higher percentage of +CDT8 cells in the spleen compared to VLP-Man-OvaI alone (Fig. 4c), in line with ability of Th+1 TCDc4ell help to augment C D+8T cell proliferation.48 To assess T cell specificity against tumor antigens, we restimulated splenocytes with the antigens OVA(323-339) or OVA(257-264) and used intracellular cytokine staining to quantify the percentage of C D+ 4and CD8+ T cells expressing IFN-³ and TNF-³, cytokines characteristic of type 1 polyfunctional T cells. Upon restimulation with(3O23V- A 339), a significant increase in the percentage of IF+CND-4³+ T cells and TNF+-C³D4+ T cells among spleen cells was detected in the VLP-Man-OvaII immunization group compared to those treated with VLP-Man alone (Fig. 4d). Notably, a significant increase in+CDIF8N+-³T cells (6.6-fold) and TNF-³+CD8 + T cells (4.7-fold) among splenocytes was detected in the Man-OvaI group compared to the Man group upon restimulation with OVA(257-264) (Fig. 4e). Both tumor-antigen specific CD4+ and CD8+ T cells were induced by immunization with VLP-ManOvaI/II (Fig. 4d,e). Thus, simultaneous presentation of exogenous glycoligands with tumor antigens by VLPs enabled the differentiation of naïve T cells into tumor antigen-spTehc1ific

The engagement of DC-SIGN by VLP-Man-OvaI/II induced differential DC activation and signaling profilesin vitro compared to the nonspecific PE particles. We next assessed if this superior DC activation translated to a more robust T cell respionnsveivo relative to an immunogenic presentation of the target tumor antigens alone. Thus, while administration of

VLP-PE-OvaI/II slowed tumor growth, vaccination with VLP-Man-OvaI/II, displaying the DCSIGN ligand, significantly inhibited tumor growth and extended survival in the B16F10-OVA model (Fig. 5a-d). Examination of the immune landscape on day 21 revealed that DC activation paralleled T cell generation and therapeutic efficacy. Immunization with VLP-Man-OvaI/II significantly increased percentages of both CD+4and CD8+ T cells in the spleen compared to immunization with VLP-PE-OvaI/II (Fig. 5e). Not only did the mannosylated particles recruit more T cells to the spleen, but they also induced a significantly higher percentage of tumor antigen-specific T cells compared to the control particles. Specifically, IFN-³ and-³ TNF secretion upon restimulation with OVA(323-339) or OVA(257-264) were significantly higher for mice immunized with VLP-Man-OvaI/II (Fig. 5f,g). Compared to the VLP-PE-OvaI/II control, we detected a significant increase in the percentage of +CDan4d CD8+ T cells among all tumor cells and among CD3+ T cells in the tumors of VLP-Man-OvaI/II immunized mice (Fig. 5h,i). Consistent with our in vitro results, these data suggest that the mannosylated VLPs drive greater DC activation, antigen presentation, and more potently activating immune signaling to enhance T cell responses. + Fig. 4 | Mannosylated VLP conjugation with tumor antigens induces specific Th1 CD4 and CD8 T cells. a, Schematic representation of VLPs with Ova antigens. b, C57/BL6 mice were inoculated 5 with 5´10 B16F10-OVA cells (s.c.) on day 0 and treated with VLPs and 2939-cGAMP (s.c.) and anti+ + PD-1 intraperitoneally (i.p.) on days 3, 9, and 15. c, Percentages of CD4 and CD8 T cells in the + spleens were analyzed on day 21. d, IFN-³ and TNF-³-secreting CD4 T cells in the spleen were + measured following restimulation with OVA(323-339). e, IFN-³ and TNF-³-secreting CD4 T cells in + the spleen were measured following restimulation with OVA(257-264). f, Percentages of CD4 and + + + + CD8 T cells in the tumors were analyzed on day 21. g, Percentages of CD4 and CD8 among CD3 T cells within the tumor. h, Mean tumor growth curves. i, Tumor volumes on day 21. j, Survival curves. The difference between VLP-Man and VLP-Man-OvaI/II was significant (p < 0.01, Log-rank (Mantel3

Cox) test). Data represent mean ± s.e.m. from a representative experiment (n=5 (c-g) and n=8 (h-j). Statistical analysis was performed by one-way ANOVA with Bonferroni9s multiple comparisons test or by log-rank (Mantel-Cox) test for the survival analysis. *P<0.1, **P<0.01, ***P<0.001, ****P<0.0001. Each experiment was repeated twice with similar results.

As melanoma is a highly metastatic and recurrent disease, robust protection against melanoma recurrence is neede5d1. To test our VLP strategy on this front, surviving mice from the VLP-Man-OvaI/II cohort were re-challenged with B16F10-Ova on Day 28 (Fig. 5j). While naïve mice rapidly developed tumors, the VLP-Man-OvaI/II cohort was protected against tumor rechallenge (Fig. 5k). The immune protection observed following therapeutic vaccine administration suggests the generation of long-term memory T cells that provide prophylactic immunity and combat disease recurrence.

Having demonstrated that VLP-Man-OvaI/II vaccination induces tumor-antigen specific Th1 CD4+ and CD8+ T cells, we sought to assess the potential benefits of simultaneous STING activation with the adjuvant cGAMP. Mice inoculated with B16F10-OVA tumor celrles we immunized with VLP-Man-OvaI/II and anti-PD-1 with and without the adjuvant 2939-cGAMP (Fig. 6a). No significant difference in tumor growth or animal survival was observed (Fig. 6b-d). Immune cell analysis showed a decrease in the percentage of antigen-specific+ CTD 4cells and no significant difference in the percentage of antigen-specific +CDT8 cells (Fig. 6e-f). The same trends were observed in the tumor, with a decrease in the percentage of tumor-infiltrati+ng CD4 T cells and no significant decrease in the percentage of tumor-infiltrating+ CTD8cells (Fig. 6g). These data indicate that VLP-Man-OvaI/II can serve as a self-adjuvanting cancceirnevadcue to the immunostimulatory signaling achieved via combinatorial engagement of TLR7 and DCSIGN.

Fig. 6 | VLP-Man-OvaI/II has self-adjuvanting capacity. a, C57/BL6 mice were inoculated with 5 5´10 B16F10-OVA cells (s.c.) on day 0 and treated with VLPs and 2939-cGAMP (s.c.) and anti-PD-1 intraperitoneally (i.p.) on days 3, 9, and 15. b, Mean tumor growth curves. c, Tumor volumes on day + significant. e, IFN-³ and TNF-³-secreting CD4 T cells in the spleen were measured following tumors were analyzed on day 21. Data represent mean +/- s.e.m. from a representative experiment (n=8 (b-d) and n=5 (e-g)). Statistical analysis was performed by one-way ANOVA with Bonferroni9s multiple comparisons test or by log-rank (Mantel-Cox) test for the survival analysis. *P<0.1, **P<0.01, ***P<0.001, ****P<0.0001.

Effective cancer vaccines require mobilization of type 1 cellular immunity, a ptrhoactess necessitates DC recognition of endogenous tumor antigens as pathoge5n2icC.urrently FDAapproved cancer vaccines involve adoptive transfer eoxfvivo loaded DCs;53 however, these individualized procedures are complex, lengthy, and expensive. Moreover, significant limitations of this approach, including poor immune cell persistence and low tumor infiltration, nresult i limited tumor control5.4 To overcome these challenges, there is a critical need for strategies that facilitate in vivo priming of DCs and T cells against tumor antig2eAnsc.hieving the desired T cell polarizationin vivo requires localization to the LNs, activation of distinct DC signaling pathways, and secretion of pro-inflammatory cytokin5e5sW.hile antibody conjugates efficiently target DC-specific lectins for tumor antigen delivery, they do not achieve sufficiennt lecti clustering to induce cellular immunit5y6. On the other hand, TLR agonists have demonstrated promise in promoting DC activation and cellular immunity, but their applicationbeheans limited by poor efficacy and systemic toxici2t0y,2.4 Enhancing delivery to APCs via particulate formulations of TLR agonists can increase efficacy while minimizing off-target e5f7f,e5c8tsA. strategy that combines lectin and TLR signaling holds promise to facilnitvaitveo priming of DCs and T cells against tumor antigens.

Mannose functionalization has previously been utilized to increase uptake of TLR sagonist via nonselective targeting of lectins on APCs with mixed res2u5l-2t9s. In addition to utilizing lectins to facilitate DC uptake and minimize off-target activity, we sought thoodmiceatlly leverage defined pathogen-mimetic glycans to selectively engage immune activatintginsle,c initiate lectin signaling, and amplify TLR activation. Indeed, our previous studies demteodnstra that the chemical structure of mannose ligands has profound effects on downstream immunity. Specifically, we observed an augmentation of Th1 responses only with an aryl mannoside ligand that binds DC-SIGN in a pH-independent manner, while pH-dependent mannose ligands only increased antigen uptake without enhancing immune respon3s1esW.e therefore hypothesized that leveraging lectin-driven immune signaling could amplify TLR activation to induce robust antitumor cellular immunity.

Our vaccine platform is uniquely designed to emulate exogenous pathogen signaling to reshape the immune response against endogenous tumor antigens. To achieve this pathogen mimicry, we engineered DC-SIGN-targeted VLPs that package ssRNA as maskedR TliLgands. The self-assembled PP7 VLP was functionalized with exogenous mannose ligannddsmaodel tumor antigens to yield the vaccine candidate VLP-Man-OvaI/II. The use of a synltyhetical accessible aryl mannoside derivative that binds the lectin DC-SIGN in a pH-independennnt emra facilitated efficient DC-SIGN engagement and signaling, without requiring isolation of native and otherwise prohibitively complex carbohydrat.e31s Attachment via powerful azide-alkyne bioconjugation chemistry allows the required high density of ligand presentation to be achieved in a manner compatible with the co-display of antigenic peptTidhees.VLPs traffic to the immune cell-rich LNs following subcutaneous administration, enabling induction of tumorspecific immunity without requiring intratumoral administration. This is advantageous to overcome limited tumor accessibility, the tolerogenic tumor environment, acceptability of delay before tumor excision, and tumor metastas5i9s.

DC-SIGN-mediated endocytosis facilitated highly efficient uptake of VLP-Man-OvaI/II and TLR7 recognition of encapsulated ssRNAs. RNA-Seq analysis revealed that VLP-ManOvaI/II upregulatedTLR7-associated gene expression compared to identical particles that do not engage DC-SIGN. Additionally, VLP-Man-OvaI/II induced higher levels of DC activation and pro-inflammatory cytokine secretion compared to the control VLP-PE-OvaI/II, demonstrating the benefit of simultaneously delivering ligands for DC-SIGN and TLR7. Efficient presentation of tumor antigens on activated DCs induced tumor antigen-specific Th1 +CDa4nd CD8+ T cells capable of tumor infiltration. We demonstrated that Th1 +CDT4cell help supported CD+8T cell proliferation, survival, and effector functions. Immunization with VLP-Man-OvaI/II significantly inhibited tumor growth and prolonged survivcaolmpared to VLP-PE-OvaI/II in a challenging murine melanoma mode.l Generation of tumor antigen-specific T ceinllsvivo established immune memory, providing protection against tumor re-challenge. This capability to prevent tumor recurrence following treatment adds significant value to our therapeutic cancer vaccine. Finally, we demonstrated the capacity of VLP-Man-OvaI/II to serve as a self-adjuvaannticnegr c vaccine, preventing the need for additional adjuvants that could overstimulate the immune system.

Our novel platform delivers tumor antigens in a pathogen costume, mimicking key features of extracellular pathogens to drive immunity against cancer cells. We harnessed pathogenic glycosylation to boost immune responses against otherwise tolerated tumor antigens, contrary to the glycan-lectin interactions commonly exploited by cancer to evade immunity. Our vaccine system mirrors the immunostimulatory signaling of foreign pathogens via simultaneous delivery of DC-SIGN and TLR7 agonists to trigger immune activation against tumor antigens. These findings constitute a significant conceptual advance in cancer vaccine development by leveraging specific lectin signaling to drive TLR activation and anti-tumor immunity. The strategy we described shows promise as a viable alternative to autologous cell immruapnioetsh,e possessing the ability to induce robust tumor antigen-specific T cells capable of solid tumor infiltration. The straightforward conjugation strategy used for VLP functionalization enables the installation of multiple human tumor neoantigens, facilitating potential translation sof thi platform to a clinical setting. This wohraks the potential to broaden the scope of cancer immunotherapies by presenting a modular and synthetically accessible platform capoafble directing type 1 cellular immunity. Finally, we envisage that this concept could bed atpoplie vaccine development for diverse intracellular infectious diseases that have eludecdinveac realization.

This research was supported by the National Institutes of Health (R01A1055258 to L.L.K, R01CA247484 to M.G.F., 2R01CA220468-06A1 to J.A.J., and F99CA264404 to S.H.B.), the Centers for Disease Control and Prevention (75D30121P12509 to M.G.F.), the Marie SkCodowska-Curie Individual Fellowship (MSCA-IF-GF 886709 to A.G.), and the National Science Foundation (NSF GRFP fellowship for V.L.). We thank the Koch Institute's Robert A. Swanson (1969) Biotechnology Center, specifically the Flow Cytometry Core and the Integrated Genomics and Bioinformatics Core, supported in part by the Koch Institute Support Grant P30CA014051 from the National Cancer Institute. We also thank the MIT Department of Chemistry Instrumentation Facility, the MIT Division of Comparative Medicine, and the Imaging Core of the Ragon Institute of MGH, MIT, and HarvaWrde. thank T. Diefenbach for assistance with microscopy. Figure 1a. was created with BioRender.com. A.K.S. reports compensation for consulting and/or Scientific Advisory Board (SAB) membership from Merck, Honeycomb Biotechnologies, Cellarity, Repertoire Immune Medicines, Hovione, Third Rock Ventures, Ochre Bio, FL82, Empress Therapeutics, Relation Therapeutics, Senda Biosciences, IntrECate biotherapeutics, Santa Ana Bio, and Dahlia Biosciences unrelated to this work. D.J.I. reports compensation for consulting and/or SAB membership from Elicio Therapeutics, Ankyra Therapeutics, Strand Therapeutics, Window Therapeutics, Venn Therapeutics, Alloy Therapeutics, Livzon Pharmaceuticals, SQZ Biotechnologies, Jupiter Therapeutics, Parallel Bio, Surge Therapeutics, Senda Biosciences, Gensaic Therapeutics, and Third Rock Ventures unrelated to this research. J.A.J. is a cofounder and shareholder of Window Therapeutics unrelated to this research. L.L.K. reports compensation for consulting and/or SAB membership from Exo Therapeutics, the ONO Pharmaceutical Foundation, and Coca Cola unrelated to this research. V.L., R.H., M.M.A., L.L.K., and M.G.F. are inventors on relevant patent applications held by the Massachusetts Institute of Technology and Georgia Institute of Technology. The remaining authors declare that they have no competing