Synthesis of Chirally Chimeric Protein Nanoparticle Vaccines via Mirror-Image Spy Chemistry

Abstract

Mirror-image proteins and nucleic acids exhibit remarkable biostability and bioorthogonality, offering a promising strategy to enhance the longevity of biological therapeutics. Here, we present a modular approach for constructing complex protein architectures that integrate both natural-chirality (L-) and mirror-image (D-) motifs. Key to this strategy is the development of D-SpyStapler-a chemically synthesized, chirally inverted ligase that enables the efficient conjugation of D-SpyTag and D-BDTag. By exploiting the achirality of glycine (Gly), we used L-sortase to covalently link D-peptides (e.g., SpyTag or BDTag, bearing an N-terminal poly-Gly motif) to L-proteins (e.g., GFP, VLP-forming Mi3, or antigens) containing a C-terminal LPXTG sorting signal. The resulting D-SpyTagged and D-BDTagged proteins were further assembled via D-SpyStapler. This method enabled the construction of chirally chimeric VLP vaccines displaying antigens derived from malaria parasites and coronaviruses in various forms-recombinant proteins or synthetic peptides-providing significant flexibility and modularity for vaccine design. The resulting chirally hybrid vaccines exhibited enhanced proteolytic resistance in vitro and elicited potent immune responses in vivo. This study provides a versatile platform for developing long-acting therapeutics and vaccines.

Keywords: Chirally chimeric protein; Mirror‐image protein; Protein nanoparticle vaccine; Proteolytic resistance; Spy chemistry.

Figure 1

Schematic illustration of the synthesis of chimeric D/L VLP vaccines. a) SpyStapler‐mediated BDTag/SpyTag ligation by catalyzing the isopeptide bond formation between the aspartic acid residue of SpyTag and the lysine residue of BDTag. b) Sortase A‐mediated transpeptidation for conjugating chirally inverted GGG‐D‐peptides with recombinant L‐Mi3‐LPETGG and L‐antigen‐LPETGG proteins. c) Assembly of antigen‐decorated Mi3 nanoparticle vaccines through chirally inverted Spy chemistry, where D‐SpyStapler facilitates the ligation of D‐SpyTag and D‐BDTag.

Figure 2

Sortase‐mediated YFP a), CFP b) and Mi3 c) reactions with varied concentration of D‐BDTag and D‐SpyTag at 37 °C for 2 h. The conjugation products, including YFP‐D‐SpyTag, YFP‐D‐BDTag, CFP‐D‐SpyTag, CFP‐D‐BDTag, Mi3‐D‐SpyTag, and Mi3‐D‐BDTag, are highlighted. The suspected hydrolysis products of YFP and CFP resulting from sortase‐mediated cleavage are also indicated.^{[ 32 ]} Uncropped gel images are provided in the Supporting Information.

Figure 3

Effects of spacer linkers on D‐SpyStapler‐mediated decoration of VLPs. a) Gene constructs, sequences, and predicted structures of Mi3‐Linker(x)‐LPETGG. The structures were generated by AlphaFold2. b) and c) SDS‐PAGE analyses showing the covalent conjugation of b) Mi3‐Linker(x)‐D‐SpyTag with CyRPA‐D‐BDTag and c) Mi3‐Linker(x)‐D‐BDTag with CyRPA‐D‐SpyTag. d) Conjugation efficiencies for the tested spacer linker variants for Mi3‐D‐SpyTag with CyRPA‐D‐BDTag and Mi3‐D‐BDTag with CyRPA‐D‐SpyTag. The conjugation efficiency was estimated by calculating the ratio of conjugated Mi3 to undecorated Mi3. Note: Uncropped gel images are provided in the Supporting Information.

Figure 4

Stability of protein nanoparticles toward proteolytic degradation. a) SDS‐PAGE analysis of trypsin digestion of Mi3‐(L‐Spy)‐YFP and Mi3‐(D‐Spy)‐YFP at various time points. b) SDS‐PAGE analysis of proteinase K digestion of Mi3‐(L‐Spy)‐YFP and Mi3‐(D‐Spy)‐YFP at various time points. c) and d) Regions corresponding to degraded fragments and SpyStapler in a) and b) were enlarged and contrast‐optimized to resolve low‐abundance proteolytic intermediates. e) and f) Quantification analysis of degradation coefficient at different time points for trypsin and proteinase K digestion, respectively. The degradation coefficient was calculated as the normalized intensity ratio of degraded fragments to undigested conjugates. Uncropped gel images are provided in the Supporting Information.

Figure 5

TEM characterization of chirally chimeric protein nanoparticles decorated with various antigens. a) SDS‐PAGE analysis of conjugating various proteins onto Mi3. Each lane is numbered, and the corresponding sample names are listed below the gel image. b) SDS‐PAGE analysis of conjugating antigenic peptides onto Mi3. The synthetic chimeric D/L peptides, which possess D‐SpyTag at the N‐terminus, were conjugated to Mi3‐D‐BDTag in the presence of D‐SpyStapler. c) TEM analysis confirming the integrity of Mi3 nanoparticles analyzed in a) and b). Their size distribution is shown in the embedded bar graphs, with the x‐ and y‐axis representing the diameter and the number of nanoparticles, respectively. Scale bars: 50 nm. Uncropped gel images are provided in the Supporting Information.

Figure 6

Mirror‐image ligation of the protein antigen CyRPA onto VLPs elicits potent antibody responses. a) Schematic representation of the Mi3‐CyRPA VLPs used in this study. VLP‐L1‐CyRPA, VLP‐L‐SpyTag + L‐SpyStapler + CyRPA‐L‐BDTag. VLP‐L2‐CyRPA, VLP‐L‐BDTag + L‐SpyStapler + CyRPA‐L‐SpyTag. VLP‐D1‐CyRPA, VLP‐D‐SpyTag + D‐SpyStapler + CyRPA‐D‐BDTag. VLP‐D2‐CyRPA, VLP‐D‐BDTag + D‐SpyStapler + CyRPA‐D‐SpyTag. b) Immunization schedule. c) Area‐under‐curve (AUC) analysis of anti‐CyRPA antibody production in mice immunized and boosted with CyRPA in various forms. One‐way ANOVA analysis was performed by comparing the VLP‐treated groups with the free CyPRA‐treated one. **p < 0.01; ***p < 0.001; ****p < 0.0001; n = 5 or 6 mice.

Figure 7

Mirror‐image ligation of peptide antigens onto VLPs enhances antibody responses. a) Schematic representation of the VLPs used in this study. VLP‐(D‐Spy)‐Mal, VLP‐D‐BDTag + D‐SpyStapler + D‐SpyTag‐Mal; VLP‐(D‐Spy)‐SARS, VLP‐D‐BDTag + D‐SpyStapler + D‐SpyTag‐SARS. b) Immunization schedule. c) AUC analyses of antibody production in mice immunized and boosted with peptide antigens in various forms on Day 14, 28, 42, and 56. d) Comparison of antibody levels among VLP‐treated groups, undecorated VLP, and free antigen groups on Day 56. One‐way ANOVA was utilized for all analyses. *p < 0.05; **p < 0.01; n = 5 or 6 mice.