Protection studies following bronchopulmonary and intramuscular immunisation with Yersinia pestis F1 and V subunit vaccines coencapsulated in biodegradable microspheres: a comparison of efficacy

We have compared the ability of intramuscularly and intratracheally administered recombinant F1 and V subunit antigens to safeguard mice from a lethal systemic challenge with plague. The combined subunits (1μg V plus 5 μg F1) were inoculated either in the ‘free’ state as a solution, or entrapped within microspheres composed of a biodegradable polyester (Poly- -lactide), on day 1 and 60 of the experiment. In comparison to the other regimens, introduction of microsphere suspensions into the respiratory tract resulted in statistically elevated levels of specific immunoglobulins in day 82 lung wash samples. A subcutaneous challenge with virulent Yersinia pestis bacteria on day 137, equivalent to more than 10⁵ mouse LD₅₀s, was comparatively well tolerated by all subunit treatment groups (with survival rates between 66 and 90%). In contrast, 80% of the mice injected intramuscularly with soluble F1 and V were defeated by a 10⁷ MLD₅₀ subcutaneous challenge, whereas the group immunised intramuscularly with microparticles were significantly better protected (p<0.1) with 50% survival. Similarly, mice immunised intratracheally with microparticles were significantly better safeguarded (56% survival) compared with the group immunised with soluble subunits intramuscularly (p<0.01). Soluble sub-units delivered intratracheally afforded 33% protection against 10⁷ MLD₅₀s. These data indicate that bronchopulmonary administration of microsphere co-encapsulated recombinant F1 and V antigens elicits a similar level of protective immunity against systemic plague infection as that evoked by injecting co-encapsulated subunits into the muscle. Such findings corroborate the thesis that introduction of appropriately formulated F1 and V subunits into the respiratory tract may be an alternative to parenteral immunisation schedules for protecting individuals from plague.     

Immunisation against plague by transcutaneous and intradermal application of subunit antigens

We have investigated immunological responses in BALB/c mice following transcutaneous (TC) delivery of fraction 1 (F1) and V subunits from Yersinia pestis in conjunction with an enterotoxin-derived adjuvant (cholera toxin, CT). It was found that two or more TC applications of F1 and V subunits (admixed with cholera toxin) served to elicit significant levels of anti-F1 and V antibodies in the serum of immunised mice. IL-6 secretion from cultured splenocytes derived from immunised mice indicated that a single TC application of F1 and V subunits (admixed with cholera toxin) conferred a cell-mediated response. As compared with intranasal or direct intradermal injection of F1 and V, the numbers of F1/V-specific antibody-forming cells in the spleens of animals immunised by TC application of F1 and V (admixed with CT) was relatively low. It was noted that TC application of F1 and V admixed with CT was very effective for priming responses that were boosted by intranasal or intradermal routes. Similarly, it was found that TC application of F1 and V admixed with CT could be used to efficiently boost pre-existing responses engendered by intradermal injection or intranasal instillation of F1 and V. In order to assess if TC application of F1 and V admixed with CT could protect experimental animals from plague, immunised mice were injected with a virulent strain of Y. pestis. It was found that two TC applications of F1 and V admixed with CT conferred only limited protection against 10(2) MLDs. However, three TC applications of F1 and V admixed with CT conferred solid protection against 10(2) MLDs. Hence we have shown, for the first time, that TC application of F1 and V admixed with CT can protect animals against challenge with a virulent strain of plague causing bacteria. These data suggest that transcutaneous immunisation may be a simple and non-invasive method for immunising individuals against plague.     

Analysis of local and systemic immunological responses after intra-tracheal, intra-nasal and intra-muscular administration of microsphere co-encapsulated Yersinia pestis sub-unit vaccines

Intra-tracheal, intra-nasal and intra-muscular immunisation with admixed Y. pestis sub-units (3 μg V, 0.47 μg F1) or equivalent doses of poly-L-lactide microsphere co-encapsulated antigens was done. Systemic and mucosal responses to F1 and V differed according to immunisation route, and encapsulated status of the sub-units. Irrespective of immunisation site, particulated sub-units stimulated statistically superior primary systemic reactions, with intra-tracheal and nasal microsphere immunisations eliciting superior serum anti-V IgG titres in comparison to intra-muscular injection of free vaccines (p<0.001 beyond day 8). Pulmonary and nasal delivery of microspheres induced primary serum anti-V IgG titres which were greater (p<0.039) or equal to (p > 0.056) those after intra-muscular injection of spheres. In terms of serum anti-F1 titres, mice responded best to intra-muscular, and comparatively poorly to intra-nasal immunisations. Intra-tracheal administration of microspheres induced strongest responses in the respiratory tract, dominated by the IgG rather than IgA isotype. An intra-nasal booster immunisation on day 63 potentiated strong local and circulating anti-V IgG titres in microsphere vaccinees. Priming and boosting with free vaccines induced significantly depressed secondary serum anti-F1 titres relative to microsphere immunisations (p<0.024 at days 78 and 120). In contrast to other priming sites, intra-tracheal instillation of encapsulated vaccines facilitated the induction of IgG antibody to both F1 and V in day 146 broncho-alveolal washings. With the exception of primary responses to F1 in mice immunised intra-tracheally with microspheres, IgG₁ was the dominant subclass of anti-F1/V IgG in serum. We conclude that introduction of biodegradable microspheres containing the F1 and V sub-units into to the upper or lower respiratory tract engenders immune responses of a magnitude comparable with that induced by parenteral immunisation, and may present a means of protecting individuals from plague.     

Intranasal Protollin/F1-V vaccine elicits respiratory and serum antibody responses and protects mice against lethal aerosolized plague infection

F1-V is a recombinant plague antigen comprising the capsular (F1) and virulence-associated (V) proteins. Given intramuscularly with Alhydrogel, it protects mice against challenge, but is less effective in non-human primates against high-dose aerosolized Yersinia pestis challenge, perhaps because it fails to induce respiratory immunity. Intranasal immunization of mice with F1-V formulated with a Proteosome-based adjuvant (Protollin), elicited high titers of specific IgA in lungs whereas intranasal F1-V alone or intramuscular Alhydrogel-adsorbed F1-V did not. The Protollin-adjuvanted F1-V vaccine also induced high serum titers of specific IgG, comparable to those induced by intramuscular Alhydrogel-adsorbed F1-V. Mice immunized intranasally with Protollin-F1-V were 100% protected against aerosol challenge with 170 LD50 of Y. pestis and 80% against 255 LD50.     

Kinetics of the immune response to the (F1+V) vaccine in models of bubonic and pneumonic plague

Protection against aerosol challenge with > 300 MLD of Yersinia pestis was observed 7 days after a single immunisation of mice with the F1+V vaccine. At day 60, mice were protected against injected challenge (10(7)MLD) in a vaccine dose-related manner. Recall responses to rV in splenocytes ex vivo at day 98 correlated significantly (p<0.001) with the immunising dose-level of V antigen; no memory response or anti-V serum IgG was detected in killed whole cell vaccine (KWCV) recipients. This may explain the susceptibility of KWCV recipients to aerosol challenge and the enhanced protection conferred by the F1+V sub-unit vaccine, particularly since the anti-F1 responses induced by either vaccine were similarly IgG1-polarised.     

Microsphere translocation and immunopotentiation in systemic tissues following intranasal administration

With a view to developing improved mucosal immunisation strategies, we have quantitatively investigated the uptake of fluorescent polystyrene carboxylate microspheres (1.1 μm diameter), using histology and fluorescence-activated cell sorting, following intranasal delivery to BALB/c mice. To qualify these biodistribution data, antigen specific memory and effector responses in the spleens of mice immunised nasally with Yersinia pestis V antigen loaded poly(lactide) (PLA) microspheres (1.5 μm diameter) were assessed at 4, 7 and 11 days. Irrespective of administration vehicle volume (10 or 50 μl), appreciable numbers of fluorescent microspheres were detected within nasal associated lymphoid tissues (NALT) and draining cervical lymph nodes. Nasal administration of the particles suspended in 50 μl volumes of phosphate-buffered saline (PBS) served to deposit the fluorescent microspheres throughout the respiratory tract (P<0.05). In these animals, appreciable particle uptake into the mediastinal lymph node was noted (P<0.05). Also, spleens removed from mice 10 days after fluorescent particle application contained significantly more microspheres if the suspension had been nasally instilled using a 50 μl volume (P<0.05). Appreciable memory (and effector from day 7) responses were detected in mediastinal lymph nodes removed from mice immunised nasally with 50 μl volumes of microparticulated or soluble V antigen. Immunological responses in splenic tissue removed 7 days after intranasal immunisation corroborated the thesis that the spleen can act as an inductive site following bronchopulmonary deposition of particulated antigen: upon exposure to V in vitro, splenic T-cells from mice nasally immunised with 50 μl volumes of microspheres incorporated statistically greater (P<0.05) quantities of [³H]thymidine into newly synthesised DNA than did T-cells from cohorts nasally immunised with 50 μl volumes of V in solution. Similarly, significant numbers of anti-V IgG secreting cells were only detected in spleens from mice immunised intramuscularly or nasally with microparticles. These immunological and biodistribution data support the tenet that, following an appropriate method of mucosal delivery, microparticles can translocate to tissues in the systemic compartment of the immune system and thence provoke immunological reactions therein.     

Protective efficacy of a fully recombinant plague vaccine in the guinea pig

A fully recombinant sub-unit vaccine comprising the protein antigens rF1 + rV has been demonstrated to protect immunised guinea pigs against exposure to 10(5) colony-forming units (CFU) of virulent Yersinia pestis. Additionally, IgG purified from rF1 + rV-immunised guinea pig serum, protected the mouse by passive immunisation against challenge with Y. pestis whereas IgG purified from the serum of guinea pigs immunised with a licensed killed whole cell (KWC) vaccine for plague, protected less well. Guinea pigs immunised with the licensed killed whole cell vaccine developed an IgG titre for fraction 1 (F1) but not for V antigen. The differential in protection conferred on the mouse by passive immunisation with guinea pig IgG, was abrogated by the use of IgG purified from guinea pigs immunised with killed whole cell vaccine supplemented with V antigen. These findings indicate that the reduced efficacy of the licensed killed whole cell vaccine formulation previously observed in the mouse can be attributed to lack of the V antigen. Cross-protection of the mouse with guinea pig IgG suggests that the recognition of neutralising epitopes in the F1 and V proteins is conserved between these two species.     

Efficacy of the live attenuated Francisella tularensis vaccine (LVS) in a murine model of disease

A live attenuated vaccine Francisella tularensis live vaccine strain (LVS), that confers protection against tularemia infection in a number of animal models including man was developed during the 1960s in the US. In this study, we have established the median lethal dose (MLD) after intraperitoneal (i.p.) or intravenous (i.v.) delivery of NDBR Lot 4 F. tularensis LVS to be 4 cfu and 2.24 x 10(4) cfu, respectively, in BALB/c mice and less than 1 cfu and 1.29 x 10(4) cfu, respectively, in C57BL/6 mice. When delivered subcutaneously, the MLD for F. tularensis LVS was greater then 1 x 10(8) cfu in both strains of mouse. Using mouse models of systemic tularemia infection it was demonstrated that F. tularensis LVS immunised BALB/c mice were fully protected after challenge with approximately 1000 MLD of a strain of F. tularensis subsp. tularensis or a strain of F. tularensis subsp. holarctica. Under similar challenge conditions, protection in C57BL/6 mice was only evident against a subsp. holarctica strain. In BALB/c mice, protection against a subsp. holarctica strain was achieved 4 days after F. tularensis LVS immunisation whereas protection against a subsp. tularensis strain was only evident 14 days after F. tularensis LVS immunisation.     

Quantitative anti-F1 and anti-V IgG ELISAs as serological correlates of protection against plague in female Swiss Webster mice

A recombinant fusion protein composed of Yersinia pestis fraction 1 capsule (F1) and virulence-associated V antigen (V) (F1–V) has been developed as the next-generation vaccine against plague. In this study, female Swiss Webster mice received a single intramuscular vaccination with one of eight doses of the F1–V vaccine and exposed 4 weeks later to either Y. pestis CO92 or C12 organisms by the subcutaneous or aerosol routes of infection. Quantitative anti-F1 and anti-V immunoglobulin G (IgG) ELISAs were used to examine the relationship between survival outcome and antibody titers to F1 and V. Results suggested that each 1 log₁₀ increase in week 4 quantitative anti-F1 and anti-V IgG ELISA titers were associated with a 1.7-fold (p = 0.0051) and 2.5-fold (p = 0.0054) increase in odds of survival, respectively, against either bubonic or pneumonic plague and may serve as serological correlates of protection.     

The preclinical testing of a formaldehyde inactivated Ross River virus vaccine designed for use in humans

Ross River virus was grown in industrial facilities in vaccine-certified Vero cells in the absence of serum, inactivated using standard formalin-inactivation protocols, treated with Benzonase to digest host cell DNA and purified on a sucrose gradient. Mice given two subcutaneous injections of 0.625 microg of this vaccine or two doses of 0.156 microg vaccine with aluminium hydroxide adjuvant failed to develop a detectable viraemia after intravenous challenge with 10(6)TCID50 of the prototype strain of Ross River virus (T48). Guinea pigs immunised with one or two10 microg doses of vaccine with adjuvant also failed to develop a detectable viraemia following a similar challenge. The levels of neutralising antibody (neutralisation index 1.9-3.1) in the mice protected against challenge with 10(6)TCID50 Ross River virus were similar to those in 16 former epidemic polyarthritis patients (1.1-3.5) who had not experienced a second clinical infection with Ross River virus in the 20 years following their initial infection.     

Evaluation of cross-protection between S. Pneumoniae serotypes 35B and 29 in a mouse model

Pneumococcal conjugate vaccines (PCVs) have reduced vaccine-type pneumococcal disease but in turn have also resulted in replacement with non-vaccine serotypes. One such serotype, 35B, a multidrug resistant type, has been associated with an increase in disease.Mice were immunized intramuscularly with monovalent pneumococcal polysaccharide 35B conjugated to CRM197 containing aluminum phosphate adjuvant on days 0, 14, and 28. Pneumococcal enzyme-linked immunosorbent assay, opsonophagocytic killing assays, and competition OPA were performed for STs 35B and 29 to measure serotype-specific binding and functional antibodies. On day 52, mice were intratracheally challenged with S. pneumoniae ST29 to evaluate cross-protection.35B-CRM197 immunized mice had binding and functional antibodies to both PnPs 35B and 29. 35B-CRM197 immunized mice were 100% protected from IT challenge with S. pneumoniae ST29 as compared to 30% survival in the naïve group.Future vaccines containing polysaccharide 35B, such as the investigational 21-valent PCV, V116, may provide cross protection against the non-vaccine serotype 29 due to structural similarity.     

Differential efficacy of vaccinia virus envelope proteins administered by DNA immunisation in protection of BALB/c mice from a lethal intranasal poxvirus challenge

DNA vaccines might offer an alternative to the live smallpox vaccine in providing protective efficacy in an orthopoxvirus (OPV) lethal respiratory challenge model. BALB/c mice were immunised with DNA vaccines coding for 10 different single vaccinia virus (VACV) membrane proteins. After an intranasal challenge with the VACV IHD strain, three gene candidates B5R, A33R and A27L produced > or =66% survival. The B5R DNA vaccine consistently produced 100% protection and exhibited greatest efficacy after three 50 microg intramuscular doses in this model. Sero-conversion to these vaccines was often inconsistent, implying that antibody itself was not a correlate of protection. The B5R DNA vaccine induced a strong and consistent gamma interferon (IFNgamma) response in BALB/c mice given a single DNA vaccine dose. Strong IFNgamma responses were also measured in pTB5R immunised C57BL6 mice deficient for MHC class I molecules, suggesting that the memory response was mediated by a CD4+ T cell population.     

A single dose sub-unit vaccine protects against pneumonic plague

In this study, the protection afforded against aerosolised Yersinia pestis by injection of a single dose of an alhydrogel-adsorbed sub-unit vaccine has been compared with that given by an existing killed whole cell vaccine licensed for human use. The sub-unit vaccine, prepared by admixing F1 antigen derived from a Y. pestis cell culture supernatant with recombinant V antigen derived from an E. coli cell lysate, fully protected an outbred strain of mouse against exposure to 10(6) CFU of virulent plague organisms (10(4) mouse lethal doses, MLD). In contrast, the whole cell vaccine provided only 16% protection against the same level of challenge. Furthermore, sub-unit vaccinees were able to clear the bacteria from their lungs post-challenge whereas bacteria were cultured from the lungs of a surviving KWC vaccinee post-challenge. In killed whole cell vaccinees, physiologically significant levels of IgG to F1 only were detectable and the levels of F1-specific IgG in serum and in broncho-alveolar washings were significantly lower (p<0.05) compared with sub-unit vaccinees. In sub-unit vaccinees, an IgG titre to the F1 and V antigens was detected in serum where it was significantly higher (p<0.05) compared with broncho-alveolar washings suggesting that, at the time of challenge, protection is attributable mainly to the combined circulating IgG titre to the F1 and V sub-units. The enhanced protective efficacy of this sub-unit vaccine administered as a single dose compared with an existing vaccine has been demonstrated in an outbred animal model of pneumonic plague.     

Recombinant raccoon pox vaccine protects mice against lethal plague

Using a raccoon poxvirus (RCN) expression system, we have developed new recombinant vaccines that can protect mice against lethal plague infection. We tested the effects of a translation enhancer (EMCV-IRES) in combination with a secretory (tPA) signal or secretory (tPA) and membrane anchoring (CHV-gG) signals on in vitro antigen expression of F1 antigen in tissue culture and the induction of antibody responses and protection against Yersinia pestis challenge in mice. The RCN vector successfully expressed the F1 protein of Y. pestis in vitro. In addition, the level of expression was increased by the insertion of the EMCV-IRES and combinations of this and the secretory signal or secretory and anchoring signals. These recombinant viruses generated protective immune responses that resulted in survival of 80% of vaccinated mice upon challenge with Y. pestis. Of the RCN-based vaccines we tested, the RCN-IRES-tPA-YpF1 recombinant construct was the most efficacious. Mice vaccinated with this construct withstood challenge with as many as 1.5 million colony forming units of Y. pestis (7.7 x 10(4)LD(50)). Interestingly, vaccination with F1 fused to the anchoring signal (RCN-IRES-tPA-YpF1-gG) elicited significant anti-F1 antibody titers, but failed to protect mice from plague challenge. Our studies demonstrate, in vitro and in vivo, the potential importance of the EMCV-IRES and secretory signals in vaccine design. These molecular tools provide a new approach for improving the efficacy of vaccines. In addition, these novel recombinant vaccines could have human, veterinary, and wildlife applications in the prevention of plague.     

Further development of raccoon poxvirus-vectored vaccines against plague (Yersinia pestis)

In previous studies, we demonstrated protection against plague in mice and prairie dogs using a raccoon pox (RCN) virus-vectored vaccine that expressed the F1 capsular antigen of Yersinia pestis. In order to improve vaccine efficacy, we have now constructed additional RCN-plague vaccines containing two different forms of the lcrV (V) gene, including full-length (Vfull) and a truncated form (V307). Mouse challenge studies with Y. pestis strain CO92 showed that vaccination with a combination of RCN-F1 and the truncated V construct (RCN-V307) provided the greatest improvement (P = 0.01) in protection against plague over vaccination with RCN-F1 alone. This effect was mediated primarily by anti-F1 and anti-V antibodies and both contributed independently to increased survival of vaccinated mice.     

Role of antibody to lipopolysaccharide in protection against low- and high-virulence strains of Francisella tularensis

Mice immunised with lipopolysaccharide (LPS) from Francisella tularensis were protected against challenge with the live vaccine strain (LVS). However, when similarly immunised mice were challenged using the fully virulent F. tularensis strain Schu4, only an increase in the time to death was observed. Passive transfer of serum from LPS-immunised mice to naive mice afforded protection against F. tularensis LVS. LPS-immunised mice depleted of either CD4+ or CD8+ T-cells survived a F. tularensis LVS challenge although the rate of clearance of bacteria from the spleen was significantly reduced in the CD8+ depleted group. LPS-immunised mice boosted with F. tularensis LVS were re-challenged with F. tularensis Schu4. This cohort was significantly protected (LD(50) increased from <1 to >1000 CFU). However, passive transfer of serum did not confer protection and mice depleted of CD4+ or CD8+ T-cells did not survive.     

Protection against anthrax and plague by a combined vaccine in mice and rabbits

The protective antigen (PA) of Bacillus anthracis and the Fraction 1 Capsular Antigen (F1 antigen), V antigen of Yersinia pestis have been demonstrated to be potential immunogens and candidate vaccine sub-units against anthrax and plague respectively. In this study, the authors have investigated the antibody responses and the protective efficacy when the antigens were administered separately or in combination intramuscularly formulation adsorbed to an aluminum hydroxide adjuvant. Results show that immunized rF1 + rV and rPA antigen together was as effective as separately for induction of serological antibody response, and these titers were maintained for over 1 year in mice. An isotype analysis of the serum indicates that the co-administration of these antigens did not influence the antigen-specific IgG1/IgG2a ratio which was consistent with a Th2 bias. Furthermore, the combined vaccine comprising the protein antigens rF1 + rV + rPA has been demonstrated to protect mice from subcutaneous challenge with 10⁷ colony-forming units (CFU) virulent Y. pestis strain, and to fully protect rabbit against subcutaneous challenge with 1.2 × 10⁵ colony-forming units (CFU) virulent B. anthracis spores. These data show that the protective efficacy was unaffected when the antigens were administered in combination.     

Expression of a recombinant form of the V antigen of Yersinia pestis, using three different expression systems.

Yersinia pestis, the causative organism of plague, produces V antigen (LcrV), a bifunctional protein with regulatory and virulence roles that has been shown to be highly protective against a plague challenge. A combined sub-unit vaccine, comprising recombinant V and Fraction 1 antigens is currently being developed. We report here the expression and purification of recombinant V antigen (rV) using three different expression systems: the N-terminal GST fusion pGEX-5X-2 and pGEX-6P-2 systems from Pharmacia Biotech, and the C-terminal CBD fusion (IMPACT I) system from New England Biolabs. After cleavage from the carrier protein, the yields of rV were 25 mg l(-1) (pGEX-5X-2), 31 mg l(-1) (pGEX-6P-2) and 0.75 mg l(-1) (IMPACT I). All of the recombinant proteins were immunogenic in mice, although there were some differences in their protective efficacy against subcutaneous challenge with Y. pestis. Whilst rV antigen derived from the IMPACT I and pGEX-6P-2 systems and given in two immunising doses protected fully against challenge with 1 x 10(7) colony forming units (cfu) of Y. pestis, there was breakthrough in protection against 1 x 10(5) cfu of Y. pestis in animals immunised twice with rV from the pGEX-5X-2 system. From this study, the pGEX-6P-2 has been selected for the production of rV as a vaccine component. The pGEX-6P-2 system utilises a GST tagged PreScission Protease (a recombinant human rhinovirus 3C protease) to cleave the fusion protein, thereby allowing efficient removal of the enzyme from the final product. In addition, the enzyme is not of animal origin, therefore making it suitable for vaccine production.     

Single-dose, virus-vectored vaccine protection against Yersinia pestis challenge: CD4+ cells are required at the time of challenge for optimal protection

We have developed an experimental recombinant vesicular stomatitis virus (VSV) vectored plague vaccine expressing a secreted form of Yersinia pestis low calcium response protein V (LcrV) from the first position of the VSV genome. This vector, given intramuscularly in a single dose, induced high-level antibody titers to LcrV and gave 90-100% protection against pneumonic plague challenge in mice. This single-dose protection was significantly better than that generated by VSV expressing the non-secreted LcrV protein. Increased protection correlated with increased anti-LcrV antibody and a bias toward IgG2a and away from IgG1 isotypes. We also found that the depletion of CD4+ cells, but not CD8+ cells, at the time of challenge resulted in reduced vaccine protection, indicating a role for cellular immunity in protection.     

Protection against plague following immunisation with microencapsulated V antigen is reduced by co-encapsulation with IFN-gamma or IL-4, but not IL-6

We have investigated intranasal delivery of novel vaccines for plague, based on poly-L-lactide (PLLA) microencapsulated recombinant V antigen (rV) of Yersinia pestis. Microspheres containing rV alone or co-encapsulated with the cytokines IFN-gamma, IL-4 or IL-6 were administered in a two-dose regimen and antibody responses and protective efficacy were monitored. All treatment groups stimulated high rV-specific antibody titres in serum, predominantly of the IgG1 isotype, which were maintained over several months. There was evidence of both IgG and IgA responses in lung samples from all groups. Formulations based on rV antigen alone or rV co-encapsulated with IL-6 provided complete protection against systemic challenge with Y. pestis strain GB; however protective efficacy was impaired by co-encapsulating either IFN-gamma or IL-4 with rV.