Boosting with Poxviruses Enhances Mycobacterium bovis BCG Efficacy against Tuberculosis in Guinea Pigs

Mycobacterium tuberculosis, the bacillus that causes tuberculosis, kills approximately 2 million people each year (15). The induction of type 1 cytokine-producing T cells is central to protection from disease. The bacillus Calmette-Guérin (Mycobacterium bovis BCG) is the only licensed vaccine against tuberculosis (2). Each year 80% of children born globally are vaccinated with BCG (http://www.who.int/inf-fs/en/fact104.html). BCG affords significant protection from childhood and miliary tuberculosis but fails to protect adults from pulmonary disease, which is the major cause of mortality in the developing world (3, 5). BCG-mediated protection lasted 10 to 15 years after vaccination in a United Kingdom study, indicating that neonatal vaccination has negligible impact on the adult epidemic (8).

We have previously shown that BCG-induced T-cell responses can be significantly boosted in BALB/c mice (7) and rhesus macaques (6) by attenuated poxviruses expressing the major M. tuberculosis secreted antigen and mycolyl transferase, antigen 85A. In BALB/c mice, BCG followed by a single vaccination of modified virus Ankara (MVA) expressing antigen 85A (MVA85A), both given intranasally, produced significantly greater protection following an aerosol M. tuberculosis challenge than did vaccination with BCG alone. The level of protection observed was the highest induced by a vaccination regimen in this strain of mice. In rhesus macaques we observed that parenteral immunization with BCG and then MVA85A induced T-cell responses that could be further boosted by intradermal immunization with a recombinant fowlpox virus, FP9, also expressing Ag85A (FP9.Ag85A) (6). The promising levels of immunogenicity induced by parenteral BCG-MVA85A-FP9.Ag85A vaccination led us to test whether this regimen protected guinea pigs against M. tuberculosis aerosol challenge. In this species, survival is a stringent measure of protection. Previous studies of plasmid DNA and poxvirus vaccines in guinea pigs have shown protection equivalent to but not better than that of BCG used alone (A. Williams et al., unpublished).

Six Dunkin-Hartley guinea pigs (David Hall, Burton-on-Trent, United Kingdom) were vaccinated subcutaneously with 5 × 10⁴ CFU BCG and then 4 weeks later vaccinated with 10⁷ PFU MVA85A intradermally and then 4 weeks later with a further 10⁷ PFU FP9.Ag85A. Six other animals were vaccinated with BCG only on the day of the MVA immunization, and six animals were sham immunized with saline. Six weeks after the FP9 immunization animals were challenged by aerosol using a Collison nebulizer and Henderson apparatus delivering approximately 500 CFU of M. tuberculosis (H37Rv) into the airways. Animals were killed at 26 weeks postchallenge or at the humane end point (20% loss of maximal body weight). All saline controls died before the experimental end point, whereas two/six BCG-vaccinated animals survived. However, all six BCG-MVA85A-FP9.Ag85A-vaccinated animals survived until the end of the experiment at 26 weeks (versus BCG, P = 0.018) (Fig. 1). Over the 26-week period following challenge, the average weight gain for the BCG-MVA85A-FP9.Ag85A-vaccinated animals was 156 g (range, +2 to +221 g) (Table 1). In contrast, in the saline group the average weight loss was 79 g (range, −9 to −118 g). All animals in this group were killed at the humane end point. In the BCG-vaccinated group there was an average weight gain of 48 g (range of −21 to +149 g).

Only one other vaccination approach has been reported to show significantly greater efficacy than BCG in guinea pigs, a recombinant BCG (10) of the Tice strain overexpressing, from an episomal plasmid, antigen 85B, a paralogue of antigen 85A. Seventy-five percent of guinea pigs administered this rBCG30 strain survived to 26 weeks after administration of a challenge dose of M. tuberculosis 25-fold lower than that used in our study (9). The poxvirus boosting strategy has several advantages over use of any new replicating BCG strain expressing an episomal plasmid. The efficacy of new strains of BCG is likely to be impaired by immunity to environmental mycobacteria, as appears to be the case with BCG, leading to little efficacy in older individuals in the developing world (1, 14). In contrast poxvirus vectors appear to use preexisting immunity to mycobacteria to enhance their own immunogenicity (12). Nonreplicating poxviruses have been used safely in human immunodeficiency virus-positive individuals (e.g., reference 4) and, unlike replicating bacterial vectors, will likely be usable in this important target population for a new tuberculosis vaccine. Finally, poxvirus boosting of BCG leads to markedly enhanced T-cell responses in animals (6, 7) and adult humans (12), providing an immune marker to facilitate vaccine development.

We have now shown that poxviruses expressing antigen 85A can significantly boost T-cell-induced responses in mice (7), rhesus macaques (6), and humans (12) and produce significantly greater protection than BCG alone when delivered intranasally in mice (7) and intradermally in guinea pigs. In humans MVA boosting of BCG was equally immunogenic in recently BCG-vaccinated individuals and those BCG vaccinated more than 10 years earlier (12), supporting the use of poxvirus boosting in teenagers as a means of tuberculosis prevention in adults.

Recombinant nonreplicating poxviruses are increasingly widely used in clinical trials of candidate prophylactic and therapeutic vaccines, including several studies in the developing world (4, 11, 13), and demonstrate an excellent safety record. Their strong immunogenicity and protective efficacy in animal models of tuberculosis disease identify BCG prime-poxvirus boost regimens as a leading approach to an improved tuberculosis vaccine, a view supported by ongoing clinical evaluation of such regimens (12).

• Open in new tab
• Download powerpoint

FIG. 1.

Kaplan-Meier curves of the survival of guinea pigs post-aerosol challenge with M. tuberculosis. Six animals per group were vaccinated with BCG, BCG-MVA85A-FP9.Ag85A, or saline.

• American Society for Microbiology