The aim of this study was to compare the safety and efficacy of a vaccine formulated with MOMP, CpG-1826 and Montanide ISA 720 VG to induce in mice protective humoral and cell mediated immune responses against a C. muridarum respiratory challenge. Montanide ISA 720 VG was used at four different concentrations 70%, 50%, 30% and 10% (v/v) while the quantities of MOMP and CpG-1826 were kept constant in the four vaccines. When compared with the 30% and 10% concentrations, i.m. vaccination with the 70% and 50% formulations of Montanide ISA 720 VG produced higher reactogenicity at the site of immunization that lasted for the length of the experiment. Humoral and cellular immune memory responses were similar in mice vaccinated with the 70%, 50% and 30% Montanide ISA 720 VG but were weaker in the group immunized with 10%. Following the i.n. challenge, based on changes in body weight, weights of the lungs, and the number of C. muridarum IFU recovered from the lungs, vaccines containing 70%, 50% and 30% Montanide ISA 720 VG elicited similar robust protection while the 10% did not. Our results indicate that vaccines using 30% Montanide ISA 720 VG should be compared with the 70% formulation in humans and animal models for their safety and efficacy at inducing protection against pathogens.
The intensity of the reactogenicity of a vaccine is dependent on several factors including the site of immunization, volume of the vaccine, and type of antigen and adjuvant used 55. The immunological status of the vaccinee will also impact the level of reactogenicity. When formulated as "a water-in-oil" emulsion (70% v/v), a shortcoming of Montanide ISA 720 VG is the production of a granuloma at the site of immunization that can last for weeks or months 44. The Montanide ISA 720 VG induced granuloma, by creating a depot effect, helps to slowly release the antigen and maintain local immune responses over a long period of time. In addition to the depot effect, other mechanisms appear to be involved in the induction of immune responses by Montanide ISA 720 VG. For example, injection of the antigen in one site, and of Montanide ISA 720 VG at a different location, still has an adjuvant effect although it is weaker when compared with delivery at the same site 44. Our results confirm that the depot effect of Montanide ISA 720 VG is only one of the components that affect its adjuvanticity. A similar phenomenon occurs with Alum. The depot effect is not necessary for induction of innate immune responses 56. While the 70% formulation generated bullae, that measured ~ 2–5 mm in diameter, and were still present at the end of the experiment, the 30% Montanide ISA 720 VG formulation resulted in a ~ 1–2 mm indurations that disappeared over a period of 3–4 weeks. In spite of the differences in local reactogenicity between the Montanide ISA 720 VG at a 70% versus a 30% concentration both elicited the same immune responses indicative that they are independent of the depot effect. Attraction of APC to the immunization site could be one the mechanism by which alum and Montanide ISA 720 VG may enhance immunity 44.
Most studies in animal models, and data collected from patients, indicate that CD4 + T-cells producing IFN-g, are required to protect against C. trachomatis infections, while CD8 + T-cells play a secondary role 33,57,58. Although the role of antibodies in protection is still controversial, they are probably important particularly during the early stages of the infection 32,59,60. For example, Brunham et al. 61 demonstrated that levels of C. trachomatis secretory IgA specific antibodies in the cervix inversely correlated with the number of recoverable EB. Therefore, a combination of a Th1 and a Th2 adjuvant may be required for a chlamydia vaccine to optimize protection.
CpG-1826, delivered alone with MOMP, has a limited adjuvant effect likely because it readily diffuses systemically 62,63. Enhancement of the immune responses has been observed when CpGs, or other TLR agonists, are delivered with Montanide ISA 720 VG 35,40,42,64. These adjuvants combinations elicit more robust APC activation, resulting in enhanced expression of CCR7, MHC class II and co-stimulatory molecules that lead to robust T-cell activation in the lymph nodes 43,65.
It is known that delivering antigens and adjuvants to the same APC helps to enhance immune responses 52,66. The negative charges of MOMP likely interact with the positively charged polar head of cationic lipids on Montanide ISA 720 VG. CpG-1826 also carries negative electric charges and has strong electrostatic attraction to the surface of Montanide ISA 720 VG. The complex of MOMP + Montanide ISA 720 + CpG-1826 can then bind to the positive charged surface of APC. As a result, colocalized delivery of antigen and adjuvants can enhance immune responses at the vaccination site 43,64,66–69. Furthermore, by trapping multiple molecules, the adjuvant increases the density of the antigen leading to enhanced B-cell responses 67,70–72.
Following immunization, the C. muridarum EB-specific lgG2a/IgG ratio in serum showed that the Montanide ISA 720 VG + CpG-1826 vaccines all elicited Th1-biased responses while MOMP alone induced Th2-skewed responses. Neutralizing antibodies in serum were present in the three groups of mice immunized with high concentrations of Montanide ISA 720 VG but were negative in the 10% group. These findings were supported by the high levels of IFN-g present in the supernatants of T-cells stimulated with EB from mice vaccinated with 70%, 50% and 30% Montanide ISA 720 VG, but not with the 10% formulation.
Levels of protection against the i.n. challenge with C. muridarum correlated with the immune responses. Mice vaccinated with 70%, 50% or 30% Montanide ISA 720 VG loss less body weight than those immunized with the 10% Montanide ISA 720 VG. A similar trend was observed when the lungs' weights of mice were determined at 10 days following i.n. challenge. Similarly, the number of C. muridarum IFU in the lungs were significantly higher in mice receiving the 10% Montanide ISA 720 VG than in the other three groups of mice. The high levels of C. muridarum specific of IFN-g in T-cell supernatants, the presence of neutralizing antibodies in serum and of C. muridarum specific IgA in lung’s supernatants induced by formulations containing the 70%, 50% or 30% Montanide ISA 720 VG, likely are responsible for the protection observed against the respiratory challenge.
A limitation of this study is the use of the respiratory tract model rather than the genital tract model for infection with C. muridarum. Testing for local and systemic reactogenicity is not a limitation since the i.m. route will likely be implemented when a vaccine for C. trachomatis genital infections becomes available. Furthermore, these results can likely be directly applied to vaccines developed for respiratory bacterial and viral pathogens such as Chlamydia pneumoniae, Mycobacterium tuberculosis, Streptococcus pneumoniae, Haemophilus influenzae, Bordetella pertussis, influenza viruses, coronaviruses, and respiratory syncytial virus.
We have used the respiratory model extensively to test Chlamydia antigens and adjuvants. The respiratory and the genital tract have a mucosal and as systemic component and therefore, we can evaluate immune responses in both compartments. Immunization and effects on protection can be tested in the respiratory model in less than three months while experiments with the genital tract model take seven months to complete. This is a major difference that significantly affects supplies and personnel costs. We have performed several experiments, and our conclusion is that, if we cannot induce protection in the respiratory model, that vaccine formulation and delivery system, will not protect against a genital tract challenge.
Another shortcoming of this experiment is that the immune responses and protective activity of these vaccines formulations were tested only four weeks following the boost. It will be therefore important to test these immunization protocols for their ability to induce long-term protection. It is possible that the 70% Montanide ISA 720 VG formulation, by having a longer depot effect, may induce extended memory immune responses when compared to the 30% Montanide ISA 720 VG vaccine. However, it is also possible that a long exposure to the antigen will result in tolerance with changes in the immune responses that will lead to increase susceptibility to infection 73.
To summarize, for the first time, we have shown that a C. muridarum MOMP vaccine, formulated with 30% Montanide ISA 720 VG, combined with CpG-1826, elicits minimal local reactogenicity at the site of immunization, while inducing robust protective immune responses, similar to the 70% formulation, against a respiratory C. muridarum challenge. Our next step is to determine, in the genital tract mouse model, if the protective immune responses, elicited by the vaccine formulation containing 30% Montanide ISA 720 VG, induce an equivalent protection to that obtained with the 70% concentration. Studies in humans should then validate the results obtained with animal models. Positive data could help to move forwards the licensing of Montanide ISA 720 VG for clinical use.