Infection with Mycobacterium tuberculosis remains a leading cause of death worldwide, despite the widespread use of the Mycobacterium bovis bacille Calmette–Guérin (BCG) vaccine. Although BCG provides protection against disseminated childhood tuberculosis, protection against pulmonary tuberculosis in adults is limited, in part because of its inefficient activation of T-cell immunity.

Now, reporting in the Journal of Clinical Investigation, Stefan Kaufmann and colleagues describe the generation of a novel BCG strain equipped with a virulence protein from Listeria monocytogenes. The authors provide evidence for the increased protective effect of this vaccine and propose a mechanism to explain its enhanced efficacy.

Protection against M. tuberculosis requires optimal CD4+ and CD8+ T-cell immunity, but it is the generation of the latter that poses a problem. Both M. tuberculosis and BCG reside in the phagosomes of antigen-presenting cells (mainly macrophages and dendritic cells), which channel mycobacterial antigens into the MHC class II pathway. As CD4+ T cells, but not CD8+ T cells, interact with MHC class II molecules, the isolation of infecting mycobacteria in phagosomes favours the activation of the helper-T-cell response.

To redress the balance of the T-cell response to mycobacterial infection, Kaufmann and colleagues generated a recombinant BCG (rBCG) strain that secreted listeriolysin, a protein used by L. monocytogenes to perforate the phagosome membrane. As the 'hole punching' function of listeriolysin is optimal at an acid pH, the BCG urease C gene, which encodes an enzyme that neutralizes the pH of the phagosome, was also deleted. The authors reasoned that disruption of the phagosome by functional listeriolysin would dislodge BCG into the cytoplasm, activating CD8+ T cells through the MHC class I pathway.

Indeed, macrophages infected with the ΔureC hly+ rBCG strain accumulated mycobacterial antigens in the cytoplasm. And the new vaccine strain was significantly more effective than unmodified BCG in protecting mice against infection with both a laboratory strain and a new clinical isolate of M. tuberculosis.

Although the enhanced protection offered by ΔureC hly+ rBCG might be due to increased cytoplasmic antigen loading of MHC class I molecules, experiments conducted by Kaufmann and his team suggested an additional possibility. Previous work has shown that disruption of the phagosome induces programmed cell death in infected cells, which release apoptotic blebs with mycobacterial antigen cargo. Neighbouring dendritic cells take up these apoptotic blebs and channel antigens into the MHC class I pathway, with efficient activation of CD8+ T cells. It is this phenomenon, known as cross priming, that seems to be crucial for superior vaccine efficacy.