Development of a Versatile Cancer Vaccine Format Targeting Antigen-Presenting Cells Using Proximity-Based Sortase A-Mediated Ligation of T-Cell Epitopes

Abstract

Cancer vaccines are a promising strategy to increase tumor-specific immune responses in patients who do not adequately respond to checkpoint inhibitors. Cancer vaccines that contain patient-specific tumor antigens are most effective but also necessitate the production of patient-specific vaccines. This study aims to develop a versatile cancer vaccine format in which patient-specific tumor antigens can be site-specifically conjugated by a proximity-based Sortase A (SrtA)-mediated ligation (PBSL) approach to antibodies that specifically bind to antigen-presenting cells to stimulate immune responses. DEC205 and CD169 are both receptors expressed on antigen-presenting cells that can be targeted to deliver antigens and stimulate T-cell responses. We used the CRISPR/HDR platform to produce mouse heavy chain IgG2a antibodies with DEC205 or CD169 specificity containing an SrtA recognition motif followed by a SpyTag at the C-terminus. Using a recombinant protein of SrtA linked to SpyCatcher, we applied proximity-based SrtA-mediated ligation to ligate fluorescein isothiocyanate (FITC)-labeled or antigenic peptides to the antibodies. Ligated antibodies bound to DEC205-expressing dendritic cells or CD169-expressing macrophages both in vitro and in vivo. More importantly, immunization with DEC205- or CD169-specific Abs linked to T-cell epitopes efficiently stimulated T-cell responses in vivo. To conclude, we have developed a cancer vaccine format using PBSL that enables the rapid incorporation of tumor antigens and could potentially be implemented for the synthesis of personalized cancer vaccines.

Figure 1

CRISPR/HDR engineering of CD169-, DEC205-specific, and isotype control rat IgG2a hybridomas to produce mouse IgG2a with SrtA recognition motif and SpyTag. (A) Hybridomas producing rat IgG2a Abs specific for mouse CD169 or DEC205 were engineered with CRISPR/HDR to produce murine mIgG2a_silent isotype Abs containing the SrtA recognition site LPETG and SpyTag. (B) Western blot of the original rat IgG2a and the switched (sw)-mouse IgG2a Abs specific for mouse CD169, DEC205, and isotype control. The blot was incubated with anti-rat (green) and anti-mouse secondary Ab (red). (C) Flow cytometry analysis of isotype-sw-Abs bound to CHO cells that express CD169 or DEC205. Anti-rat IgG Abs and anti-mouse IgG Abs were used to detect the original Abs or the sw-Abs, respectively.

Figure 2

Optimization of PBSL reaction of fluorescent substrate to sw-mouse IgG2a Abs. (A) Schematic diagram of ligation of fluorescent GGGGGK-(FITC) probe to Ab-HC C-terminus by PBSL. (B) Investigation of different incubation temperatures and times for conjugation of Abs to SrtA-SpyCatcher protein using InstantBlue-stained SDS-PAGE gel (left panel). The relative intensity values of the conjugated HC were quantified and normalized to the bead band intensity (right panel). (C) Analysis of different incubation temperatures and times for the SrtA ligation using the GGGGGK-(FITC) substrate using fluorescent imaging or InstantBlue-stained SDS-PAGE gel (left panel). Reactions were assessed at RT and 37 °C and after 2, 4, 6, and 24 h and quantified (right panel). (D) SDS-PAGE gel of SrtA-SpyCatcher reaction with Ab-LPETG-SpyTag in the absence of a FITC peptide substrate. The relative intensity values of the HC were quantified and normalized to the LC intensity value (right panel). (E) Representative SDS-PAGE visualizing PBSL labeling of anti-CD169-LPETG-SpyTag, anti-DEC205-LPETG-SpyTag, and isotype-LPETG-SpyTag with GGGGGK-(FITC).

Figure 3

PBSL of FITC to sw-Abs result in functional Abs that bind to their ligands in vitro and in vivo. (A, B) Binding of PBSL-mediated FITC-ligated or NHS-labeled anti-CD169, anti-DEC205 Abs, and isotype controls to CD169 (A) or DEC205-expressing (B) CHO cells was determined by flow cytometry. (C, D) In vitro binding of PBSL-mediated FITC-ligated Abs and NHS-labeled original Abs to splenic CD169 macrophages (C) and cDC1 (D) were determined by flow cytometry. (E, F) In vivo binding of PBSL-mediated FITC-ligated Abs and NHS-labeled original Abs to splenic CD169 macrophages (E) and cDC1 (F) were determined 30 min after intravenous (i.v.) injection by flow cytometry. Mice were injected (i.v.) with 20 μg sw-anti-CD169-FITC, sw-anti-DEC205-FITC, sw-isotype-FITC or NHS-labeled original anti-CD169-FITC, original anti-DEC205-FITC, original isotype-FITC. Indicated are representative histogram overlays from triplicates (A–D) or 6 mice from two independent experiments (E, F) and geometric mean fluorescence intensity (gMFI) quantified and indicated as mean ± SD. Statistical analysis one-way ANOVA with Šidák’s multiple comparison test with respective isotype controls. *p < 0.05, ***p < 0.001, ****p < 0.0001.

Figure 4

PBSL of OVA_247–279 peptides to sw-anti-CD169, anti-DEC205, and isotype control mouse IgG2a Abs. (A) Schematic diagram of GGGGG-OVA_247–279 peptide ligation to sw-mouse IgG2a containing SrtA and SpyTag motifs by PBSL. (B) Representative SDS-PAGE visualizing the different steps of the PBSL procedure of GGGGGOVA_247–279 peptide to the sw-anti-CD169, anti-DEC205, and isotype mouse IgG2a Abs. The cross-linked HC-HC, conjugated HC (red arrow), original HC, peptide-ligated HC (green arrow), SrtA-SpyCatcher, and LC were labeled adjacent to the corresponding protein bands on the gel image. (C, D) CHO cells expressing CD169 or DEC205 were stained with PBSL-ligated anti-CD169-OVA_247–279, anti-DEC205-OVA_247–279, and isotype-OVA_247–279 Abs. The gMFI signals of the anti-mouse secondary Ab were determined by flow cytometry. Representative histogram and graphs with mean ± SD from triplicates were shown. Statistical analysis one-way ANOVA with Šidák’s multiple comparison test. ****p < 0.0001.

Figure 5

Immunization with anti-CD169-OVA_247–279 and anti-CD169-Adpgk_295-313 Abs induces robust CD8⁺ and CD4⁺ T-cell responses. Mice were immunized with 3 μg of ligated antibodies, 25 μg of anti-CD40 antibody, and 25 μg of Poly(I:C), with immune responses assessed on day 7. (A, B) Splenic OVA-specific CD8⁺ and CD4⁺ T-cell responses were evaluated using H-2K^b-OVA_257–264 tetramer and I-A^b-OVA_262–276 tetramer staining, respectively, with representative dot plots shown alongside quantification of frequencies (mean ± SD; n = 6, pooled from two experiments). (C) Adpgk_295–313-specific CD8⁺ T-cell activation was measured after 5 h of restimulation with Adpgk_299–307 peptide, and percentage of IFNγ-producing cells is displayed (mean ± SD; n = 5). Statistical analysis one-way ANOVA with Šidák’s multiple comparison test. ***p < 0.001.