With complete bacterial genome sequences now available, "reverse vaccinology" can be conducted to identify proteins that can function as subunit vaccines. The "gene first" approach of reverse vaccinology relies upon computer analysis of the genome sequence to identify encoded proteins with features common to known surface-exposed and secreted bacterial proteins. The selected genes can then be cloned and expressed as recombinant proteins. The proteins, which may number in the hundreds, are then purified for vaccine testing in the animal model appropriate for the bacterial pathogen. Reverse vaccinology has been employed successfully to find protective protein antigens against Neisseria meningitidis serogroup B, Streptococcus pneumoniae, group B Streptococcus, Bacillus anthracis, Porphyromonas gingivalis, and other bacterial pathogens (reviewed in this paper).
This approach sounds straightforward but in practice may not always lead to identification of effective subunit vaccines. An important study from Ben Adler's group down in Monash University illustrates the challenges of finding a subunit vaccine that prevents chronic Leptospira infections. They focused on serovar Hardjo, which causes chronic infections in cattle. They selected 263 Hardjo genes that were predicted to encode surface-exposed, secreted, or lipid-modified proteins. Among these they successfully cloned and expressed 223 genes as 238 protein antigens in E. coli. Some genes were expressed as two or more fragments because of their large size. 210 of the 238 (88%) aggregated into inclusion bodies during expression and had to be kept dissolved in urea during their purification. (Strangely, the urea was not removed by dialysis prior to immunization.) The 238 purified proteins were mixed with an aluminum hydroxide adjuvant and injected into hamsters. 169 of the 238 proteins (71%) generated an antibody response, yet none succeeded in preventing colonization of the kidneys following challenge with a Hardjo strain.
It's been hard enough to find leptospiral proteins that protect hamsters from lethal disease when tested as vaccines (see this article for a review), yet Murray and colleagues sought proteins that protected against Leptospira colonization, a more difficult endeavor that has never been achieved with subunit vaccines. The Hardjo strain they used easily colonizes the kidneys yet fails to produce any signs of disease in hamsters. Although several studies have demonstrated that certain versions of the LigA and LigB proteins, when administered as vaccines, protect hamsters and mice from being killed by lethal strains of Leptospira, survivors are left with infected kidneys. A vaccine that protects against disease or death but not infection may be adequate for humans, who eventually clear the spirochetes from their kidneys even following a natural infection (assuming the disease doesn't kill them). However, vaccinated cattle infected with Hardjo may not be able to clear the spirochetes and will continue to shed infectious Leptospira into the environment, placing the entire herd and the workers handling them at risk of infection. Hardjo infections generally don't cause signs of disease in cattle, but they can cause fetal death and drop in milk production in cows.
The choice of adjuvant and destruction of protective conformational epitopes by urea are possible reasons for failure to find a protective antigen. On the other hand, perhaps a different method for delivery of protein antigens into animals should been considered. Stay tuned.
Featured paper
Murray GL, Lo M, Bulach DM, Srikram A, Seemann T, Quinsey NS, Sermswan RW, Allen A, & Adler B (2013). Evaluation of 238 antigens of Leptospira borgpetersenii serovar Hardjo for protection against kidney colonisation. Vaccine, 31 (3), 495-499 PMID: 23176980
Helpful reviews
Dellagostin, O.A., Grassmann, A.A., Hartwig, D.D., Felix, S.R., da Silva, E.F., & McBride, A.J.A. (November 2011). Recombinant vaccines against leptospirosis. Human Vaccines 7(11):1215-1224. DOI: 10.4161/hv.7.11.17944
Serruto, D., Serino, L., Masignani, V., & Pizza, M. (May 26, 2009). Genome-based approaches to develop vaccines against bacterial pathogens. Vaccine 27(25-26):3245-3250. DOI: 10.1016/j.vaccine.2009.01.072
