Type VI secretion system sheaths as nanoparticles for antigen display

Significance

Critical aspects that limit use of nanoparticles as vaccine delivery systems include the technical difficulty of preparing particles under conditions that allow controlled protein loading and folding. Biologic nanoparticles offer advantages over chemically derived particles that require harsh organic solvents for preparation. Here, we demonstrate that the multimeric sheath structure of the bacterial type 6 secretion system (T6SS) can be exploited to generate particles that display on their surface many copies of specific proteins of interest. T6SS sheaths represent a functional vaccine delivery system with remarkable flexibility due to the ability to biochemically modify these structures to expand or enhance potential immunogenicity.

Abstract

The bacterial type 6 secretion system (T6SS) is a dynamic apparatus that translocates proteins between cells by a mechanism analogous to phage tail contraction. T6SS sheaths are cytoplasmic tubular structures composed of stable VipA-VipB (named for ClpV-interacting protein A and B) heterodimers. Here, the structure of the VipA/B sheath was exploited to generate immunogenic multivalent particles for vaccine delivery. Sheaths composed of VipB and VipA fused to an antigen of interest were purified from Vibrio cholerae or Escherichia coli and used for immunization. Sheaths displaying heterologous antigens generated better immune responses against the antigen and different IgG subclasses compared with soluble antigen alone. Moreover, antigen-specific antibodies raised against sheaths presenting Neisseria meningitidis factor H binding protein (fHbp) antigen were functional in a serum bactericidal assay. Our results demonstrate that multivalent nanoparticles based on the T6SS sheath represent a versatile scaffold for vaccine applications.

The bacterial type 6 secretion system (T6SS) is a dynamic apparatus that translocates proteins between effector cells and target cells (1–4). It is conserved in 25% of Gram-negative bacteria, including Vibrio cholerae, Pseudomonas aeruginosa and Escherichia coli. The T6SS plays a crucial role in bacterial pathogenicity and symbiosis, targeting either eukaryotic cells or competitor bacterial cells (5). The assembled and functional T6SS apparatus has structural homology to bacteriophage T4 phage tail components and can be divided into two distinct assemblies: a contractile phage tail-like structure and a transmembrane complex (6). V. cholerae VipA and VipB (named for ClpV-interacting protein A and B) and orthologous proteins in other bacteria build within the cytosol of effector cells a tubular sheath structure that is anchored to the various layers of the cell envelope through its association with the T6SS transmembrane complex (7).

VipA/B sheaths are composed of six protofilaments arranged as a right-handed six-start helix similar to early T4 tail sheaths (8). Each protofilament is formed by a VipA/B heterodimer, and the atomic-resolution structure of a native contracted V. cholerae sheath has been recently determined by cryo-electron microscopy (9). Stable expression of VipB in V. cholerae requires the presence of VipA, and VipA/B heterodimers can be recruited into assembled tubular sheath structures spontaneously (10, 11). Because both ends of VipA are exposed on the external surface of the sheath tubules, a C-terminal fusion of VipA protein with superfolded green fluorescent protein (sfGFP) is functional in T6SS sheath assembly and activity, as previously demonstrated (3).

Because these tubular structures are assembled in cytoplasm and can be purified from bacteria (3), we explored the possibility that T6SS sheaths could be used as a new particle-based delivery system for vaccine antigens. It is thought that particulate structures used for vaccine formulations are efficiently targeted for uptake by antigen-presenting cells (APCs) and interact directly with antigen-specific B cells generating humoral responses (12). Although particulate protein antigens may be more resistant to degradation, they are eventually proteolytically processed, and the resulting peptides are presented by the major histocompatibility complex (MHC) class I and class II molecules in a process that leads to activation of CD4⁺ and CD8⁺ T-cell helper and effector responses. Examples of particulate vaccine delivery systems include lipid-based systems [emulsions, immune-stimulating complexes (ISCOMs), liposomes, virosomes], polymer-based structures (e.g., nano-/microparticles), and virus-like particles (VLPs), with each of these systems presenting their own spectrum of advantages and disadvantages for practical use as human immunogens (13).

In this work, VipA/B sheaths displaying heterologous protein antigens on the surface were generated and tested as a particulate vaccine antigen delivery system. Our results show that sheath-like structures displaying different antigens were immunogenic and that antibodies elicited against one of these, the Neisseria meningitidis factor H binding protein (fHbp), were functional in a serum bactericidal assay. The T6SS antigen delivery system demonstrates potential as a multivalent particle to deliver one or more antigens simultaneously into the same antigen-presenting cell. Moreover, the use of heterologous VipA and VipB sheaths displaying a common antigen in sequential vaccine booster regimens minimizes immune responses against the delivery system itself and focuses the immune responses against the common antigen of interest.

Results

Generation of Sheath-Like Structures Exposing sfGFP on the Surface.

A recombinant Duet vector system was used to enable assembly of sheath structures made by VipA, VipB, and multiple VipA fusions (see SI Materials and Methods for details). One copy of vipB and two copies of vipA DNA constructs were generated to express VipB, VipA with a C-terminal 6xHis tag (VipAhis), and a second VipA fusion protein displaying sfGFP as a C-terminal extension. Sheath-like structures were purified from either V. cholerae 2740-80 or E. coli BL21-DE3 strains, using a two-step protocol. The first step applies ultracentrifugation to collect larger particulate sheath structures, and the second step uses Ni-affinity chromatography to purify proteins associated with VipAhis (Fig. 1A). Purified sheaths from either V. cholerae or E. coli were analyzed by negative staining using electron microscopy (EM). EM analysis showed that purified sheaths are organized into 40- to 200-nm tubular structures (Fig. 1B) with an appearance that corresponded to contracted sheaths (3). Thus, VipA and VipB can autoassemble in tubular structures in the E. coli cytoplasm without the need for other T6SS components not expressed in this bacterial strain. Furthermore, sheaths were tested for the ability to autoassemble in cell-free in vitro conditions. To this end, recombinant VipAhis, VipBhis, and VipA-sfGFPhis were purified under denaturing conditions in the presence of urea, mixed, and allowed to refold. Tubular structures similar to typical sheaths were detected by EM analysis (Fig. S1). Surface sfGFP exposure on sheath particles was determined by immunogold staining with anti-GFP antibodies and Protein A-gold particles (Fig. 1C). These results indicate exposure of sfGFP on the surface and the inclusion of VipA-sfGFP fusion protein in each assembled sheath.

Fig. 1.

Purification and characterization of T6SS sheaths from V. cholerae and E. coli. (A) Purified T6SS sheaths fused to sfGFP (1 μg) were separated by SDS/PAGE and visualized by Coomassie blue staining. All three sheath components—VipAhis, VipB, and VipA-sfGFP—were observed when purified from either V. cholerae (Vc) or E. coli (Ec). (B) Electron microscopy analysis of sheath preparations shows assembled tubular structures with a length of 40–200 nm. (C) Immunogold staining by anti-sfGFP antibody at EM confirmed sfGFP exposure on the sheath surface.

Fig. S1.

In vitro self-assembly of sheath-like structures. Denatured VipAhis, VipBhis, and VipA-sfGFPhis were allowed to refold in vitro, and then sheath assembly was checked by electron microscopy analysis (EM), showing similar structures as sheaths assembled in V. cholerae and E. coli. The three panels show examples of in vitro refolded sheaths as imaged by EM using different magnifications.

Interaction of Sheath Nanoparticles with Phagocytic Eukaryotic Cells.

Because the size of purified sheath structures is comparable with other particles that can be engulfed by antigen presenting cells (APCs) (14, 15), we explored their interaction with phagocytic eukaryotic cells. Sheath-like structures exposing sfGFP (sfGFP sheaths) purified from V. cholerae or E. coli were incubated with RAW264-7 macrophages as an in vitro model for antigen uptake. Compared with a negative control, after 6 h of incubation, GFP-positive RAW264-7 macrophages were detected, indicating that sheaths purified from V. cholerae or E. coli are taken up or associated to the surface of macrophages (Fig. 2A). The number of positive RAW264-7 cells also increased over time (Fig. 2A). Confocal microscopy was used to determine localization of sheaths inside of the eukaryotic cells (Fig. 2B). Z-stack analysis showed the presence of sfGFP sheaths, on cell surface and inside the cells. These experiments demonstrated that sheath-like structures interact with macrophages and therefore are likely to interact with other APCs in the context of immunization protocols.

Fig. 2.

Sheath uptake by macrophages. (A) sfGFP sheaths purified from V. cholerae (Vc) or E. coli (Ec) were incubated with RAW264.7 macrophages for 6 or 20 h. The percentage of GFP-positive cells (highlighted green) increased over time, as determined by FACS. (B) Confocal images of cells incubated with sfGFP sheaths from E. coli after 20 h. (Upper) The 3D reconstruction of the Z-stack image. Cell membranes are visualized in red, and nuclei are in blue. (Lower) Two cross-sectional views of the vertical top or bottom section of cells, to show the presence of green dots, corresponding to sfGFP sheaths, inside the cells, as observed in xz and yz cross-section in smaller side panel.

Immunogenicity of Sheaths and Induction of an Immune Response Against the Exposed Heterologous Antigen sfGFP.

Mice were immunized with sfGFP sheaths as a proof-of-concept experiment to determine whether a heterologous protein exposed on sheaths was more immunogenic compared with the soluble recombinant protein alone. Groups of five BALB/c female mice were immunized three times biweekly with sfGFP sheaths purified from V. cholerae or E. coli with or without alum as an adjuvant. In addition, two groups of mice were immunized with “empty sheaths” from V. cholerae or E. coli (i.e., formed only with VipA and VipB) mixed with sfGFP recombinant protein to determine whether sheaths have adjuvant properties for “bystander antigens” that were not tightly associated with the particles. These experiments showed that sfGFP sheaths were immunogenic and elicited a strong humoral immune response against sfGFP (Fig. 3A). No significant differences in total IgG ELISA titers were observed against sfGFP when exposed on a sheath’s surface or as a recombinant protein in combination with empty sheaths. In both cases, titers were similar or slightly higher compared with recombinant sfGFP alone.

Fig. 3.

sfGFP sheaths are immunogenic, and heterologous VipA-VipB pairs boost immune responses against the displayed antigen. (A) The presence of sfGFP-specific antibodies in serum from individual immunized mice was determined by ELISA. Serum was collected after immunization with recombinant sfGFP plus alum or the indicated sheath preparation with or without alum. (B) The repertoire of IgG subclasses in sera was measured by ELISA for mice immunized with sfGFP sheaths from E. coli. Titers are reported as absorbency units at OD₄₀₅ for serum dilutions (dil.): dil. 1:32,000 for total IgG and IgG1; dil. 1:4,000 for IgG2a, IgG2b, and IgG3. (C) Mice were immunized with either (i) three rounds of sfGFP sheaths containing V. cholerae VipA/VipB or (ii) sequential injections of sfGFP sheaths containing V. cholerae VipA/VipB, P. aeruginosa VipA/VipB, or A. baylyi VipA/VipB. Sera were collected after each immunization (P1, post-1; P2, post-2; P3 post-3) and tested by ELISA to determine specific antibodies against sfGFP, V. cholerae VipA, and V. cholerae VipB. *P value < 0.05; **P value < 0.01; ***P value < 0.001; ****P value < 0.0001; ns, not statistically significant. If not specified, there was no statistical difference between samples.

Because there was no significant difference between sheaths isolated from the two host systems in terms of sheath structure/composition and immune response, E. coli was chosen as the preferred host strain for sheath assembly and purification because it is (i) a common host for recombinant protein expression and (ii) amenable to scale-up processes that meet good manufacturing guidelines. To dissect the features of the immune response induced by sheaths purified from E. coli, IgG subclass repertoires were determined by ELISA (Fig. 3B). IgG1 antibodies were the most represented IgG, reflecting results obtained measuring total IgG and correlating with the use of alum (16). Interestingly, levels of IgG2a and IgG2b were significantly higher when immunization was done with VipA/B/A-sfGFP compared with a recombinant sfGFP alone or in combination with empty VipA/B sheaths.

Furthermore, we tested the feasibility to incorporate multiple proteins in a single nanoparticle structure by coexpressing two fusion proteins, VipA-sfGFP and VipA-mCherry, together with VipAhis and VipB. Although the system requires optimization, our results showed that tubular structures containing both fusion proteins were assembled and were able to induce an immune response against both sfGFP and mCherry (Fig. S2).

Fig. S2.

Coexposure of two antigens on a sheath surface. VipA-sfGFP and VipA-mCherry were coexpressed in E. coli to generate sheaths carrying two different antigens. (A) Immunogold staining at EM using specific anti-sfGFP and anti-mCherry antibodies showed coexposure of both proteins on assembled sheaths. (B) VipA-sfGFP and VipA-mCherry amounts in sheath preparation were compared with VipA/B/A-sfGFP sheaths and purified recombinant sfGFP and mCherry by loading 1 μg of total proteins for each sample in SDS/PAGE. Both VipA-sfGFP and VipA-mCherry fusion proteins were detected in the sheath sample by Coomassie blue staining, even if the preparation was less homogeneous compared with purified VipA-sfGFP sheaths. (C) Sera from mice immunized with VipA/B/A-sfGFP/A-mCherry sheaths or a mixture of sfGFP and mCherry recombinant proteins were tested to measure antigen-specific antibodies. ELISA titers showed the induction of specific anti-sfGFP and anti-mCherry antibodies by sheath preparation although it was lower compared with the mixture of recombinant proteins. Because VipA/B/A-sfGFP/A-mCherry sheath preparation was not homogeneous, it was not possible to confidently estimate the amount of the two VipA fusion proteins in the sheath sample by Coomassie blue staining and normalize molar amounts of recombinant antigens for immunization experiments. Thus, VipA/B/A-sfGFP/A-mCherry sheaths seemed inferior, but this result was likely an artifact due to disparate levels of antigens in the two samples. Total mouse IgG ELISA titers are reported as absorbency units at OD₄₀₅ for the same serum dilution, 1:32,000. *P value < 0.05; **P value < 0.01.

Use of Heterologous VipA-VipB Sheaths to Boost a Focused Immune Response Against a Common Displayed Antigen.

We explored the immunogenicity of heterologous VipA-VipB sheaths displaying the same antigen decoration (sfGFP) and tested whether boosting drove the immune response selectively against the common antigen and not the scaffold proteins. VipA and vipB genes from P. aeruginosa or Acinetobacter baylyi were used to express VipAhis, VipB, and VipA-sfGFP in E. coli. Resulting sheaths were analyzed by Coomassie blue staining and EM (Fig. S3 A and B). P. aeruginosa proteins produced homogeneous and relatively long sheaths whereas sheaths produced from A. baylyi proteins were shorter and less uniform. Flow cytometry confirmed that both sheath preparations interacted with RAW264-7 cells and that a higher percentage of GFP-positive cells was observed after incubation with A. baylyi sheaths compared with P. aeruginosa (Fig. S3C). Sheath preparations were used to immunize two groups of mice. The first group was immunized with sfGFP sheaths made by V. cholerae VipA/B sequences as a reference condition. The second group was immunized with sfGFP sheaths prepared with V. cholerae VipA/B for the first injection, P. aeruginosa sfGFP-VipA/B for the second injection, and A. baylyi sfGFP-VipA/B for the third injection. Sera were collected 1 wk postimmunization, and ELISA was performed to detect antibodies against V. cholerae VipA and VipB (used as reference) and sfGFP. As shown in Fig. 3C, an immune response against sfGFP was developed in both immunization groups. In contrast, antibody titers against V. cholerae VipA and VipB were significantly lower for the group immunized with the three different sheath preparations, suggesting that the immune response was focused on the common displayed sfGFP antigen rather than conserved epitopes that exist between the VipA and VipB proteins present in these heterologous sheaths. The initial boost with V. cholerae sfGFP-VipA/B sheaths resulted in higher antibody response than boosting with P. aeruginosa sfGFP-VipA/B sheaths, suggesting that stimulation of VipA/VipB-specific CD4 T cells may provide B-cell help for the induction of sfGFP antibodies.

Fig. S3.

Generation of sheaths using heterologous VipA-VipB pairs. (A) SDS/PAGE analysis of sfGFP sheaths with P. aeruginosa VipA/VipB (lane 2) or A. baylyi VipA/VipB (lane 3) in comparison with V. cholerae VipA/VipB (lane 1). Band 1, P. aeruginosa VipB; band 2, A. baylyi VipB; band 3, A. baylyi VipA-sfGFP; band 4, P. aeruginosa VipA-sfGFP; band 5, A. baylyi VipA; band 6, P. aeruginosa VipA. The results showed a clean and homogeneous sample for sheaths obtained using P. aeruginosa VipA/VipB whereas the sheath preparation was less clean for A. baylyi. (B) Assembled tubular sheath structures were detected for both preparations by negative staining electron microscopy analysis. (C) FACS analysis of RAW264-7 macrophages incubated for 20 h with 20 μg of sfGFP sheaths with P. aeruginosa VipA/VipB or A. baylyi VipA/VipB demonstrated efficient sheath uptake by cells. GFP-positive cells are marked in red.

Use of Sheaths to Deliver Neisseria meningitidis fHbp Antigen.

To test the feasibility of using sheaths as a delivery system for more complex proteins, Neisseria meningitidis group B antigen factor H binding protein (fHbp) was used. The fHbp lipoprotein is a surface-exposed molecule characterized by a beta-barrel at its C terminus (17), and it is currently a part of two recently licensed vaccines against meningococcal B disease (18, 19). The N terminus of fHbp was fused to the C terminus of V. cholerae VipA and expressed in E. coli together with VipAhis and VipB. Results show fully assembled sheaths in the presence of fHbp fusion protein (Fig. S4 A and B). Sheaths and recombinant fHbp were used for immunization experiments, with alum as an adjuvant. As a further control, one group of mice was also immunized with empty VipA/B sheaths mixed with recombinant fHbp protein. Total IgG ELISA titers showed that fHbp sheaths were immunogenic, with antibody levels comparable with the recombinant antigen alone (Fig. 4A). Interestingly, sheaths with incorporated fHbp induced significantly higher immune response compared with recombinant fHbp mixed with VipA/B sheaths. IgG1 titers reflected the results obtained measuring total IgGs, as observed for sfGFP sheaths. In the case of IgG2a and IgG2b, although the differences were not statistically significant, fHbp sheaths induced generally higher levels of antigen-specific antibodies, compared with other soluble immunogens (Fig. S4C).

Fig. S4.

Characterization of fHbp sheaths for assembly and elicitation of immune response. SDS/PAGE (A) and EM analysis (B) of an fHbp sheath sample showed clean purification of tubular structures composed of VipA, VipB, and VipA-fHbp. (C) Sera from mice immunized with fHbp sheaths or recombinant fHbp were analyzed by ELISA to measure IgG subclasses. Titers are reported as absorbency units at OD₄₀₅ for the same dilution (dil.) for each subclass: dil. 1:32,000 for IgG1 and dil. 1:4,000 for IgG2a, IgG2b, and IgG3. For each immunization group, the geometric mean value was calculated and compared using one-way ANOVA statistical analysis. ***P value < 0.001; ns, not statistically significant. If not specified, there was no statistical difference between samples.

Fig. 4.

fHbp sheaths are immunogenic, and specific anti-fHbp antibodies are functional in SBA. (A) Serum from individual mice was collected after immunization with fHbp sheaths, with or without alum, or recombinant fHbp. The presence of fHbp-specific antibodies was determined by ELISA. Total mouse IgG ELISA titers are reported as absorbency units at OD₄₀₅ for the same serum dilution 1:32,000. **P value < 0.01; ***P value < 0.001; ns, not statistically significant. (B) Mice sera were used in a serum bactericidal assay (SBA) using rabbit complement serum and N. meningitidis strain NZ98/254. SBA data are reported as the percentage of bacterial survival (cfu/mL after 60 min of incubation/cfu/mL at time 0) for each serum dilution tested. Pooled sera for each group were used. The horizontal line represents the SBA titer as the reciprocal of serum dilution resulting in a 50% decrease in bacterial cfu/mL at the end of the assay compared with control cfu/mL at time 0. sfGFP sheaths (VipA/B/A-sfGFP) were used as negative controls. Titer = 16 is the limit of detection of the assay.

The functionality of antibodies raised against fHbp alone or exposed on the surface of sheaths was tested by serum bactericidal assay (SBA), which is currently used as an immunological correlate of protection for meningococcal vaccines (20). Increasing concentrations of sera were incubated with N. meningitidis clinical isolate NZ98/254 in the presence of an active baby rabbit complement. Bacterial colony-forming units (cfu) corresponding to viable bacteria were counted after 1 h of incubation. The results showed similar SBA titers for sera obtained from mice immunized with fHbp sheaths, with or without alum as adjuvant, and higher titers than sera obtained using recombinant protein alone or mixed with empty VipA/B sheaths (Fig. 4B). These results suggest that the different IgG subclasses elicited by sheath particles were more functional in binding to the antigen on the N. meningitidis surface, fixing complement, and causing bacterial killing.

Modification of Sheath Surface by SrtA-Mediated Ligation.

To broaden the T6SS sheath platform for antigen delivery, we developed a system that enables antigen fusion on preassembled, purified sheaths. Sortases are membrane-associated transpeptidases that anchor proteins to the cell wall in Gram-positive bacteria (21) and have recently been exploited to label a variety of proteins at the C terminus, N terminus, or internal regions (22). To apply sortase technology to the T6SS sheath system, appropriately tagged proteins were engineered to be substrates for sortase. First, VipA was modified at its C terminus to contain the LPETGG sortase-recognition motif of Staphylococcus aureus sortase A (SrtA), and VipA-sort/VipB sheaths were purified from E. coli. Secondly, 6X-his-SUMO-(Gly)₅-tagged versions of sfGFP, mCherry, and fHbp were generated and purified as recombinant proteins (Fig. 5). SUMO protease was used to cleave the SUMO tag, allowing release of (Gly)₅-sfGFP (Fig. 5, lane 5). When mixed with the VipA-sort/VipB sheaths, (Gly)₅-sfGFP was recognized by a recombinant SrtA and ligated to the LPETGG motif exposed on the external surface of sheaths (Fig. 5, lane 6). After 10 min of incubation with SrtA, an extra band (∼45 kDa) corresponding to the predicted molecular mass of the VipA-sfGFP fusion protein was observed. This band was confirmed as VipA-sfGFP by Western blot analysis (Fig. S5A). The SrtA reaction was also performed by ligating either (Gly)₅-sfGFP and (Gly)₅-mCherry to the same reaction mixture, generating sheaths decorated with both sfGFP and mCherry in equimolar amounts (Fig. 5, lane 11). Similar results were obtained using (Gly)₅-fHbp, demonstrating the versatility of the reaction (Fig. S5 B and C).

Fig. 5.

Modification of sheath surface by SrtA-mediated ligation. Different reaction mixtures, where single components were added sequentially, were separated by SDS/PAGE and visualized by Coomassie blue staining. (A) sfGFP was conjugated to VipA/B sheaths by SrtA. Lane 6 displays an extra band corresponding to VipA-sfGFP of ∼45 kDa. (B) SrtA conjugation was expanded to fuse mCherry (lane 10) or a combination of sfGFP and mCherry to VipA/B sheaths (lane 11). *, This band corresponds to VipA-sortase intermediate of ∼40 kDa.

Fig. S5.

Modification of sheath surface by SrtA-mediated ligation. (A) Western blot analysis of different sortase reaction mixtures, using specific anti-sfGFP or anti-mCherry antibodies, confirmed the generation of VipA-sfGFP or VipA-mCherry fusion proteins. (B) Sortase activity to fuse fHbp to VipA/B sheaths is shown in lane 6, where an extra band corresponding to VipA-fHbp of ∼45 kDa is present. *, This band corresponds to VipA-sortase intermediate of ∼40 kDa. (C) Western blot analysis of two different sortase reaction mixtures, using a specific anti-fHbp antibody, confirmed the generation of the VipA-fHbp fusion protein.

SI Materials and Methods

Bacterial Strains, Cells, and Culture Media.

To express and purify sheaths from Vibrio cholerae, the previously described 2740-80 ΔclpVΔflgGΔvipA strain was used, where ClpV, FlgG, and VipA genes had been deleted, to avoid sheath disassembly by ClpV and flagellar filament contamination during purification (3, 10). Escherichia coli strains DH5a and BL21-DE3 Gold (Life Technologies) were used for cloning and protein expression, respectively. Antibiotic concentrations used were as follows: streptomycin (100 μg/mL), kanamycin (30 μg/mL), chloramphenicol (5 μg/mL for V. cholerae and 20 μg/mL for E. coli), and carbenicillin (75 μg/mL). Luria–Bertani (LB) broth was used for all growth conditions. Liquid cultures were grown aerobically at 37 °C at 180 rpm shaking. Neisseria meningitidis clinical isolate NZ98/254 (38) was used in the serum bactericidal assay. Bacterial strains details are reported in Table S1.

Table S1.

Strains and plasmids used in this study

Name	Features	Source
Strains
Vibrio cholerae 2740–80 ΔClpVΔFlgGΔVipA	Streptomycin resistant, lacZ2 derivative of V. cholerae 2740-80 deleted for ClpV, FlgG and VipA genes	(3)
Escherichia coli DH5-α	supE44 hsdR17 recA1 endA1 gyrA96 thi-1 relA1	Life Technologies
Escherichia coli BL21-DE3	hsdS gal (λcIts857 ind1 Sam7 nin5 lacUV5-T7 gene 1)	Life Technologies
Neisseria meningitidis NZ98/254	ST-41/44 complex/lineage 3; B:4:P1.4	(38)
Plasmids
pET-Duet	The vector encodes two multiple cloning sites with T7 promoter, pBR322-derived ColE1 replicon, lacI gene and AmpR	Novagen
pCOLA-Duet	The vector encodes two multiple cloning sites with T7 promoter, COLA replicon from ColA (1) and KanR	Novagen
pET-Duet-sfGFP-his	Construct to express recombinant sfGFP protein in E.coli	This study
pET-Duet-mCherry-his	Construct to express recombinant mCherry protein in E.coli	This study
pET-Duet-VipA-his	Construct to express recombinant V. cholerae VipA protein in E.coli	This study
pET-Duet-VipB-his	Construct to express recombinant V.cholerae VipB protein in E.coli	This study
pETDuet-VipAhis/VipA-sfGFP	Construct to coexpress V. cholerae VipAhis and VipA-sfGFP fusion protein in V. cholerae and E. coli	This study
pET-Duet-VipAhis/VipA-mCherry	Construct to coexpress V. cholerae VipAhis and VipA-mCherry fusion protein in E. coli	This study
pET-Duet-VipAhis/VipA-fHbp	Construct to coexpress V. cholerae VipAhis and VipA-fHbp fusion protein in E. coli	This study
pET-Duet-VipAhis/VipA-NHBA	Construct to coexpress V. cholerae VipAhis and VipA-NHBA fusion protein in E. coli	This study
pET-Duet-aVipAhis/aVipA-sfGFP	Construct to coexpress A. baylyi VipAhis and VipA-sfGFP fusion protein in E. coli	This study
pET-Duet-pVipAhis/pVipA-sfGFP	Construct to coexpress P. aeruginosa VipAhis and VipA-sfGFP fusion protein in E. coli	This study
pCOLA-Duet-VipAhis/VipB	Construct to coexpress V. cholerae VipAhis and VipB protein in V. cholerae and E. coli	This study
pCOLA-Duet-aVipAhis/aVipB	Construct to coexpress A. baylyi VipAhis and VipB protein in E. coli	This study
pCOLA-Duet-pVipAhis/pVipB	Construct to coexpress P. aeruginosa VipAhis and VipB protein in E. coli	This study
pET21b-fHbp-his	Construct to express recombinant N.meningitidis fHbp protein in E.coli	This study
pET21b-NHBA-his	Construct to express recombinant N.meningitidis NHBA protein in E.coli	This study
pTARA	pBAD33 plasmid carrying T7 RNA polymerase under the control of AraPBAD promoter	Addgene

RAW264.7 murine macrophages (ATCC TIB-71) were cultured in Dulbecco's modified Eagle medium (Gibco) supplemented with 10% (vol/vol) FBS, 1% penicillin/streptomycin solution, and 2 mM l-glutamine.

Molecular Cloning.

The Duet Expression System (Novagen) was used to coexpress VipA, VipB, and VipA-antigen fusion proteins in V. cholerae or E. coli. This system comprises four different bicistronic plasmids. Each plasmid carries two multilocus cloning sites to allow isopropyl β-d-1-thiogalactopyranoside (IPTG)-dependent expression of two different proteins for each plasmid, and all of the plasmids are compatible to be cotransformed in the same host.

V. cholerae VipA and VipB gene sequences (VCA0107 and VCA0108, respectively) were amplified from pBAD24 plasmids carrying respective genes (3) and cloned in a pCOLA-Duet vector. A 6xHis-tag was added at the C terminus of the VipA gene for the final step of sheath purification by nickel affinity chromatography. Another copy of VipA-his and different VipA fusion proteins were cloned in a pET-Duet vector. For coexpression of VipA-sfGFP and VipA-mCherry fusion proteins, genes were amplified from pBAD24 plasmids carrying respective genes (3) and cloned in a pET-Duet vector.

To generate VipA/B sheaths for sortase A (SrtA) conjugation, a SrtA-recognition motif [(Gly-Ser)₃-Leu-Pro-Glu-Thr-Gly-Gly-His₆] was added at the C terminus of VipA.

To generate fusion proteins of VipA with N. meningitidis antigen, the fHbp gene without the signal peptide and lipidation motif was amplified from plasmids pET21b-fHbp-his (received from Novartis Vaccines) and cloned in pET-Duet using a linker encoding 3xAla 3xGly to provide separation from VipA. To test different VipA-VipB couples in sheath assembly and immunization, Pseudomonas aeruginosa genes PA0083 and PA0084 and Acinetobacter baylyi ACIAD2691 and ACIAD2690 were amplified from genomic DNA to clone VipA and VipB, respectively. To allow recombinant protein expression in the V. cholerae strain, bacteria were cotransformed with the pTARA plasmid carrying AraC-araBAD promoter-regulated T7 RNA polymerase (Addgene). sfGFP, mCherry, VipA, and VipB genes were PCR-amplified from plasmids carrying the genes and cloned in a pET-Duet vector with the addition of a 6His-tag at the C terminus for affinity chromatography purification. The plasmids were then transformed in E. coli BL21-DE3. To obtain recombinant proteins for sortase reaction, aminoglycine pentapeptides (Gly₅) were inserted by PCR at the N terminus of sfGFP, mCherry, and fHbp, and then the fragments were cloned into a pET-SUMO vector using the Champion pET SUMO Expression System (Life Technologies) to add a His-SUMO-(Gly)₅ tag at the N terminus of the protein of interest. All plasmids used in this work are listed in Table S1, and primers are listed in Table S2.

Table S2.

Primers used in this study

Primer	Sequence 5′– 3′
Cloning in pDUET vectors for sheath assembly
VipB_Fn	GGAATTCCATATGATGTCTACGACTGAAAAGGT
VipB_Rx	CCGCTCGAGGGCTTGATCAAGACGTCCAAC
VipA_Fn	GGAATTCCATATGTCTAAAGAAGGAAGTGTAGCTC
VipA_Re	GAGAGATATCCGCTTGTGGCTCTTCTTGAC
VipA_F_his	ATATCCATGGATGTCTAAAGAAGGAAGTGTAGCTCCCAAAGAG
VipA-R_his	ATATGCGGCCGCTCAGTGGTGATGATGGTGATGACCCGCTTGTGGCTCTTCTTGACCACTGAG
fHbp_Fe	GAGAGATATCGGCGGAGGAGGATCTGTCGCCGCCGACATCGGCGCGGGGC
fHbp_Rx	CCGCTCGAGTTATTGCTTGGCGGCAAGAC
VipA_SeqF	GTTGGCAGAACTGAATCTGC
fHbpSeq_F	CACAAGGTGCGGAAAAAACT
fHbpSeq_R	GGAAGCTTGTCAAAAGATGTATG
Cloning for expression of recombinant proteins
rVipA_Fn	GGAATTCCATATGTCTAAAGAAGGAAGTGTAGCTCCCAAAGAG
rVipA_HisR	CCGCTCGAGTCAGTGGTGATGATGGTGATGACCCGCTTGTGGCTCTTCTTGACCACTGAG
rsfGFP_Fn	GGAATTCCATATGAAAGGTGAAGAACTGTTCACCGGTGTTGTT
rsfGFP_HisR	CCGCTCGAGTCAGTGGTGATGATGGTGATGACCTTTGTAGAGCTCATCCATGCCGTGCGT
rVipB_Fn	GGAATTCCATATGATGTCTACGACTGAAAAGGTATTGGAAAGG
rVipB_HisR	CCGCTCGAGTCAGTGGTGATGATGGTGATGACCGGCTTGATCAAGACGTCCAACTAATGA
rmCh_Fn	GGAATTCCATATGGTGAGCAAGGGCGAGGAGGATAACATGGCC
rmCh_hisR	CCGCTCGAGTCAGTGGTGATGATGGTGATGACCCTTGTACAGCTCGTCCATGCCGCCGGT
Cloning for sortase reaction substrates
VipA-SoF	GGAATTCCATATGTCTAAAGAAGGAAGTGTAGCTCCCAAAG
VipA-SoR5	CCGCTCGAGTCAGTGGTGATGATGGTGATGTCCTCCGGTTTCTGGAAGAGAGCCTCCGCCCGCTTGTGGCTCTTCTTGACCACTGAGCAGATTCAG
mCher-SoF	GGCGGAGGTGGAGGTGTGAGCAAGGGCGAGGAGGATAACATG
mCher-SoR	TTACTTGTACAGCTCGTCCATGCCGCC
GFP-SoF	GGCGGAGGTGGAGGTAAAGGTGAAGAACTGTTCACCGGTGTT
GFP-SoR	TCATTTGTAGAGCTCATCCATGCCGTGCGT
fHbp-SoF	GGCGGAGGTGGAGGTGTCGCCGCCGACATCGGC
fHbp-SoR	TTATTGCTTGGCGGCAAGGCCGATATG

Expression and Purification of Recombinant Proteins.

Sheath purification protocol.

Sheath-like structures were purified from both V. cholerae and E. coli host strains by using a two-step purification method. Protein expression was induced at OD₆₀₀ = 0.5–0.7 by 1 mM IPTG and 0.2% arabinose (only for V. cholerae strains carrying the pTARA plasmid, to induce expression of T7 RNA polymerase) for 3 h. Cells were lysed by sonication in lysis buffer [400 mM NaCl, 20 mM Tris⋅HCl, pH 8.0, lysozyme 600 μg/mL, 4 μL of Benzonase Dnase (Sigma), 0.5 mL of CelLytic B (Sigma), 1 tablet of EDTA-free protease inhibitor mixture (cOmplete; Roche), 1% (vol/vol) Triton X-100]. Previously cleared lysates were subject to ultraspeed centrifugation at 150,000 × g for 1 h at 4 °C to concentrate the sample and remove soluble proteins not assembled as sheaths, as previously reported (3). In addition, the use of Triton X-100 in lysis buffer reduced the possibility to copurify membrane structures and outer membrane vesicles during the ultracentrifugation. After the ultracentrifugation, the pellet containing sheaths was resuspended in 400 mM NaCl, 20 mM Tris⋅HCl, pH 8.0, and 0.5% Triton X-100 and used for Ni-affinity chromatography purification by gravity flow using 2 mL of Ni²⁺-NTA agarose (Qiagen), following the manufacturer’s instructions. The column was washed with 10 column volumes of wash buffer (20 mM Tris⋅HCl, pH 8.0, 150 mM NaCl, and 10 mM imidazole), and sheaths were eluted with wash buffer supplemented with 250 mM imidazole. The addition of 0.5% Triton X-114 to wash buffer at 4 °C before elution was instrumental to reduce LPS contamination of samples (39). Eluted fractions were subject to buffer exchange using Econo-Pac 10DG Columns (Bio-Rad) and stored in 20 mM Tris⋅HCl, 400 mM NaCl, and 10% (vol/vol) glycerol at −20 °C.

Purification of recombinant proteins.

Protein expression was induced as described for sheath purification. His-tagged proteins in the cleared lysate were bound to Ni²⁺-NTA resin (Qiagen), washed with wash buffer (20 mM Tris⋅HCl, pH 8.0, 150 mM NaCl, and 20 mM imidazole), and eluted with wash buffer supplemented with 250 mM imidazole. The resulting purified proteins were exchanged into imidazole-free buffer (20 mM Tris⋅HCl, pH 8.0 and 150 mM NaCl) using automated Akta purifier (GE Healthcare Biosciences). Removal of LPS contaminant was achieved by using Pierce High Capacity Endotoxin Removal Spin Columns (Life Technologies), and LPS content was checked with a Pierce LAL Chromogenic Endotoxin Quantitation Kit (Life Technologies). Recombinant fHbp was provided by Novartis Vaccines.

In vitro refolding experiment.

VipAhis, VipBhis, and VipA-sfGFPhis were purified under denaturing conditions in 6 M urea using Ni-NTA agarose (Qiagen) following the manufacturer’s instructions. Denatured proteins were then mixed as 50 μg/mL VipAhis, 100 μg/mL VipBhis, and 50 μg/mL VipA-sfGFPhis and dialyzed overnight against refolding buffer (50 mM phosphate buffer, 5 mM reduced glutathione, 0.5 mM oxidized glutathione, 10% vol/vol glycerol, pH 8.8) with 4 M urea, and then refolding buffer with 2 M urea, and finally with a buffer without urea.

SDS/PAGE and Immunoblot Analysis.

Purified proteins (100 ng) or sheath preparations (1 μg of total proteins) were loaded in 4–12% (wt/vol) precast polyacrylamide gels (Life Technologies) and then stained with Coomassie Brilliant Blue R-250 solution, according to standard protocol. For Western blot analysis, after the gel run, proteins were transferred to a nitrocellulose membrane (Life Technologies). Membrane was blocked by 5% (wt/vol) milk in PBS containing Tween 20 0.05% (PBS-T), incubated with primary antibody in 3% (wt/vol) milk in PBS-T for 1 h. Mouse monoclonal antibodies anti-sfGFP or anti-mCherry (Abcam) or mouse polyclonal serum anti-fHbp (received from Novartis Vaccines) were used. Membrane was then washed with PBS-T, incubated for 1 h with horseradish peroxidase-labeled anti-mouse antibody (The Jackson Laboratory), and washed with PBS-T. The peroxidase reaction was detected by SuperSignalWest Pico Chemiluminescent Substrate (Pierce) or an Opti-4CN Substrate Kit (Bio-Rad).

Flow Cytometry Analysis.

RAW 264.7 macrophages were seeded (1 × 10⁵ cells per well) in 12-well plates 2 d before the experiment. Cells were exposed to 10 μg or 20 μg of sfGFP sheaths (purified from V. cholerae or E. coli) diluted in cell culture medium, in triplicate. After 6 or 20 h of incubation, cells were washed three times with PBS and collected using Accutase (Life Technologies). Cells from three wells were pooled and fixed with 3.7% (wt/vol) formaldehyde for 15 min at room temperature. Formaldehyde was removed, cells were resuspended in PBS, and 10,000 events were analyzed by flow cytometry to quantify GFP-positive cells using the FITC channel.

Confocal Microscopy Analysis.

RAW 264.7 macrophages (1 × 10⁴) were seeded in four-well glass slides (Corning BioCoat) 2 d before the experiment. After incubation with 10 μg of sfGFP sheaths for 20 h, cells were washed with PBS and stained with CellMask Deep Red Plasma Membrane Stain (Life Technologies) for 20 min. Cells were fixed with 3.7% formaldehyde for 15 min at room temperature and incubated with Hoechst (1 μg/mL) (Thermo Scientific) diluted in PBS for 30 min. Coverslip was mounted using ProLong Gold Antifade mounting medium (Life Technologies), and cells were imaged by confocal microscopy using a Nikon A1R laser scanning confocal microscope.

Negative Stain Electron Microscopy and Immunogold Staining.

Sheath samples were adsorbed for 5 min on a carbon-coated grid made hydrophilic by a 30-s exposure to a glow discharge. For negative staining electron microscopy, grids were washed with water, stained with 0.75% (wt/vol) uranyl formate, and imaged using a JEOL 1200EX transmission electron microscope with an AMT 2k CCD camera. For immunogold staining, grids were incubated in blocking solution [PBS plus 1% bovine serum antigen (BSA)] for 10 min and exposed to mouse primary antibody diluted 1:30 in PBS plus 1% BSA for 30 min. The same antibodies, specific for each protein exposed on the sheath surface, were used for Western blotting and immunogold staining. Grids were washed in PBS and incubated with rabbit anti-mouse bridging antibody in 1% BSA for 20 min. Grids were washed three times and incubated for 20 min with protein A-gold particles (10 nm) diluted 1:50 in PBS plus 1% (wt/vol) BSA. Grids were finally washed in PBS and water, stained with 0.75% (wt/vol) uranyl formate for 30 s, and examined as stated above.

Animal Protocol.

All animal studies were performed in accordance with National Institutes of Health guidelines and were approved by the Harvard and Boston University Medical Center Institutional Care and Use Committees.

Immunization Experiments.

Five- to 7-wk-old female BALB/c mice were purchased from Charles River Laboratories, Wilmington, MA. All mice were housed using sterile setup and were allowed a 1-wk acclimatization period before the initiation of experiments.

Groups of five mice were immunized by intraperitoneal (i.p.) injection three times biweekly with 20 μg of sheath preparations, with or without 2 mg/mL aluminum hydroxide as adjuvant (alum, purchased as Alhydrogel; Invivogen). The amount of antigen in sheath preparations was calculated as band intensity in Coomassie-stained SDS/PAGE gel, using ImageJ software, and compared with a standard curve of the known amount of recombinant protein. Relative antigen amount was calculated as 5 μg for 20 μg of total proteins in a sheath sample; therefore, 5 μg of protein was used for groups immunized with recombinant sfGFP, mCherry, or fHbp. Other groups were immunized with 15 μg of empty VipA/B sheaths plus 5 μg of recombinant proteins plus alum. Two weeks after the last immunization, all animals were killed, and blood was collected from each mouse by cardiac puncture. For mice immunized with sfGFP sheaths generated using V. cholerae, P. aeruginosa, and A. baylyi VipA/B sequences, sera were collected also 1 wk after first and second immunization by cheek blood collection. Finally, serum was obtained by centrifugation at 11,000 × g for 5 min and used for the determination of protein-specific responses by ELISA and serum bactericidal assays.

Antibody Titers.

An ELISA was designed to measure antibodies specific for each single antigen exposed on a sheath surface, and V. cholerae VipA and VipB, used as scaffold for sheath assembly. Purified proteins were coated onto Nunc Maxisorp flat bottom plates (Thermo Scientific) at 1 mg/mL and incubated over night at 4 °C. The next day, plates were washed three times with PBS-T, and the remaining adsorption sites were blocked with 3% (wt/vol) BSA in PBS-T for 1 h at room temperature. Serum from individual mice was serially diluted twofold (starting at 1:1,000) and incubated for 2 h at room temperature. Plates were washed three times with PBS-T and incubated with alkaline phosphatase-conjugated goat secondary antibody against total mouse IgG, or specific mouse IgG1, IgG2a, IgG2b, and IgG3, (Southern Biotech) diluted 1:10,000 into PBS-T plus 1.5% (wt/vol) BSA for 2 h at room temperature. Plates were visualized by the addition of p-nitrophenol substrate (Thermo Scientific). After 30 min, reactions were stopped with 2 M NaOH, and absorbance at OD₄₀₅ was determined on a spectrophotometer. Data were reported as ELISA absorbency units at OD₄₀₅ for the same serum dilution (reported for each graph).

Statistical Analysis.

ELISA data are reported as geometrical means for each group tested. Statistical analyses were performed using one-way ANOVA and unpaired t test with Welsh correction. In all cases, a P value of <0.05 was considered significant.

Serum Bactericidal Assay.

Serum bactericidal activity against N. meningitidis strains was evaluated as previously described (37), with pooled baby rabbit serum (Cedarlane) used as a complement source. Briefly, N. meningitidis strains were grown overnight on chocolate agar plates at 37 °C in 5% CO₂. A few colonies were inoculated in Mueller–Hinton (MH) broth containing 0.25% glucose to reach an OD₆₀₀ of 0.05–0.08 and incubated at 37 °C with shaking until OD₆₀₀ 0.23–0.24. The bacteria were diluted in Dulbecco’s modified PBS (Sigma), 0.1% glucose, and 1% BSA (assay buffer) at the working dilution of 10⁴ cfu/mL. The total volume in each well was 50 μL, with 25 μL of serial twofold dilutions of test mice serum, 12.5 μL of bacteria at the working dilution, and 12.5 μL of baby rabbit complement. Controls included bacteria incubated with complement serum alone or immune sera incubated with bacteria and heat inactivated complement. Immediately after the addition of the baby rabbit complement, 10 μL of the controls was plated on MH agar (t0). After 1 h of incubation at 37 °C, 10 μL of each sample and control was spotted on Mueller–Hinton agar plates (t60). The agar plates were incubated for 18 h at 37 °C, and the colonies corresponding to t0 and t60 were counted to determine colony-forming units (cfu). Serum bactericidal titers were defined as the serum dilution resulting in 50% decrease in bacterial cfu/mL at t60 compared with control cfu/mL at t0. The bactericidal titers reported in this study are related to pooled mouse sera for each group.

SrtA Sortase Reaction.

Antigens harboring N-terminal Gly₅ peptides were generated by cleavage of the His-SUMO tag by SUMO protease (31). VipA/B sheaths (34 μM final concentration), containing a SrtA recognition motif (LPETGG), were incubated with a 10-fold excess of Gly₅-sfGFP, Gly₅-mCherry, or Gly₅-fHbp in sortase reaction buffer (50 mmol/L Tris⋅HCl, 10 mmol/L CaCl₂, 150 mmol/L NaCl, pH 7.5) supplemented with 5 μM SrtA enzyme at room temperature for 15 min. An evolved form of Staphylococcus aureus SrtA enzyme (harboring mutations P94S/D160N/K196T and lacking the residues 1–58 of membrane-spanning domain) (31) was used. Samples were analyzed by SDS/PAGE and Western blot.

Discussion

Despite decades of effort, there remains significant interest in development of new vaccine delivery systems that efficiently elicit a protective immune response against pathogens and cancer, particularly in multivalent formats (13). Recently, attention has been focused on nanoparticles where the vaccine antigen is either encapsulated within the particles (i.e., liposome, ISCOMs, and polymeric nanoparticles) or decorated on the surface of particles (i.e., virus-like particles and nondegradable nanoparticles). In general, these particle-based approaches are thought to improve the immunogenicity of antigens by increasing the influx of professional APCs into the injection site, promoting antigen uptake by APCs and protecting antigens from early degradation. Critical aspects that limit use of these vaccine delivery systems include the technical feasibility of consistent particle preparation at reasonable yields, protein antigen stability during manufacture, and the toxicity associated with components present in some particle formulations.

We hypothesized the structure of T6SS VipA/B sheaths could be exploited to generate biologically derived, multivalent particles as a nanoparticle delivery system for vaccine antigens. This study confirmed the predicted utility. T6SS sheaths can be easily purified from E. coli bacterial cell lysates in a two-step protocol to generate particulate structure with size ranging from 40 to 200 nm, optimal for uptake by APCs (23). Sheath particles were immunogenic with increased representation of IgG isotypes compared with soluble protein antigens. Even in the absence of adjuvant, induction of antigen-specific IgGs demonstrated functionality in a serum bactericidal assay; this result suggests that different isotype classes have different functionality despite recognizing the same antigen with different affinities. The particulate nature of T6SS sheaths is likely responsible for induction of a Th1-derived response, as proposed elsewhere (13). It will be interesting to dissect the immune response induced by sheath particles and determine the role of proinflammatory pathways (i.e., the stimulation of NLRP3 inflammasome) in the process. We expected sheaths to have adjuvant properties, but our results showed that mixing recombinant protein with empty sheaths did not result in a significant increase in immune response against the antigen. Thus, antigen fusion to VipA is critical to expose the protein with correct folding on the sheath surface to present immunogen to the immune system in a multivalent fashion similar to what happens when antigens are displayed on pathogen surfaces. In addition, sheath structures may enhance immune responses by retaining antigen at the site of injection and reducing the rate of degradation. Although such “depotting” can also be achieved by conjugating the antigen to gold or silica particles, our system has the advantage of preventing toxicity associated with the use of nondegradable material.

Another key advantage of T6SS sheaths as a delivery system is the potential for these structures to incorporate multiple proteins in a single nanoparticle and deliver them simultaneously into the same APC, as we demonstrated using two VipA fusion proteins, VipA-sfGFP and VipA-mCherry. Alternatively, sheaths may allow assembly of recombinant proteins separately purified under denaturing conditions. The advantage of this approach is that a variety of different VipA fusion proteins can in theory be expressed and purified separately in an unfolded state, mixed in optimal ratios based on their inherent immunogenicity, and assembled into multivalent sheath particles after refolding in vitro.

In contrast with chemically defined nanoparticle structures, one drawback of biologic-based vaccine delivery platforms is the likelihood that an immune response will be generated against the delivery system itself, potentially limiting use of the delivery platform to a single dose. Thus, a platform that minimizes this predicted antisystem response would be highly desirable. We took advantage of the presence of T6SS in many Gram-negative bacteria and, by using heterologous VipA-VipB pairs during boosting, we showed a focused immune response to the common displayed antigen target (i.e., sfGFP). This feature is potentially advantageous when selection of functional or high affinity antibodies is required in an application, as shown for HIV and RSV epitopes using other display systems (24–26). It is also worth noting that, given their structural similarity to T6SS sheaths (8, 9), the sheath proteins of bacteriophage contractile tails could also be used to produce heterologous sheaths for such sequential antigen boost applications. It remains to be determined whether different VipA/B pairs affect the folding of different protein antigens or prevent degradation of unstable or weak immunogens.

T6SS can also be modified to carry other functional moieties. We used sortase to covalently attach protein antigens (i.e., sfGFP, mCherry, and fHbp) to the surface of preassembled, purified VipA/B sheaths in vitro under native conditions. Other groups have exploited sortase in vitro to label proteins with chemical moieties, such as fluorescent probes (27, 28), lipids (29), sugars (30), and proteins (31). Sortase may be applied to decorate sheaths with appropriate chemical ligands or adhesins that might give sheaths binding activity for particular target cells, such as APC or tumor cells. In this case, the desired functional domains can be expressed in different cell lines than the one expressing sheaths (e.g., in eukaryotic cells) and subsequently ligated to the surface of sheaths. It is predicted that fusion of sheaths with cell-penetrating peptides (32, 33) will increase the endocytic uptake of sheath particles and facilitate their endosome escape, as proposed for TAT-peptides conjugated to albumin-based lyophilisomes (34). If such an approach can deliver sheaths to the cytosol of APCs, it might be possible to decorate sheaths with epitopes for processing and presentation by the class I MHC pathway, or with enzymes that can synthesize adjuvant molecules (e.g., DncV-related enzymes that could synthesize cyclic di-nucleotides within target cells) (35, 36). This approach would increase the potential of the T6SS sheath delivery system to induce a cytotoxic T-cell–mediated immune response.

Future experiments will be required to optimize and control the process of antigen exposure on sheaths to consistently generate uniform particles. Nevertheless, the versatility of the T6SS sheath-based delivery system may provide for a broad array of applications and advantages that are difficult to achieve using other described nanoparticle systems available up to now (13).

Materials and Methods

Expression and Purification of Recombinant Proteins.

V. cholerae 2740-80 ΔclpVΔflgGΔvipA and E. coli BL21-DE3 Gold strains were used to coexpress VipA, VipB, and VipA-antigen fusion proteins with the Duet Expression System. Plasmids and primers used for cloning strategies are listed in Tables S1 and S2. Sheath-like structures were purified from V. cholerae and E. coli host strains using a two-step purification method: (i) ultraspeed centrifugation of bacterial cell lysates to concentrate the sample and remove not assembled soluble proteins; and (ii) Ni-affinity chromatography purification to purify sheaths carrying VipAhis.

Sheath Characterization.

For negative staining electron microscopy, samples were loaded on grids, stained with 0.75% (wt/vol) uranyl formate, and imaged with a transmission electron microscope. For immunogold staining, grids were exposed to mouse primary antibody against the exposed antigen and rabbit anti-mouse bridging antibody, followed by protein A-gold particles. To test sheath uptake by phagocytic cells, RAW 264.7 macrophages were incubated with sfGFP sheaths, and, after 6 or 20 h of incubation, cells were collected and analyzed by flow cytometry to quantify GFP-positive cells. Alternatively, after incubation, cells were stained and imaged by confocal microscopy.

Immunization Experiments and Serology Analysis.

The methods for animal experimentation were approved by the Harvard and Boston University Medical Center Institutional Care and Use Committees. Groups of five BALB/c mice were immunized three times intraperitoneally with 20 μg of sheath preparations, with or without alum. Other groups were immunized with 5 μg of recombinant sfGFP, mCherry, or fHbp. An ELISA was designed to measure mouse IgG1, IgG2a, IgG2b, and IgG3 antibodies specific for each exposed antigen, and V. cholerae VipA and VipB, used as scaffold for sheath assembly. Serum bactericidal activity against an N. meningitidis strain was evaluated as previously described (37).

SrtA Sortase Reaction.

VipA/B sheaths, containing an SrtA recognition motif (LPETGG), were incubated with a 10-fold excess of Gly₅-sfGFP, Gly₅-mCherry, or Gly₅-fHbp in a sortase reaction buffer, supplemented with SrtA enzyme. Samples were analyzed by SDS/PAGE and Western blot.

Further details for all of the experimental protocols are provided in SI Materials and Methods.