Tumor necrosis factor-alpha (TNF-α) blockade is an effective treatment for rheumatoid arthritis (RA) and other inflammatory diseases, but in patients, it is associated with reduced resistance to the infectious agents Mycobacterium tuberculosis and Listeria monocytogenes, among others. Our goal was to model infection and arthritis in mice and to compare etanercept, a currently used anti-TNF-α inhibitor, to an anti-TNF-α vaccine. We developed a murine surrogate of the TNF-α kinoid and produced an anti-murine TNF-α vaccine (TNFKi) composed of keyhole limpet hemocyanin conjugated to TNF-α, which resulted in anti-TNF-α antibody production in mice. We also used etanercept (a soluble receptor of TNF commonly used to treat RA) as a control of TNF neutralization. In a mouse model of collagen-induced arthritis, TNFKi protected against inflammation similar to etanercept. In a mouse model of acute L. monocytogenes infection, all TNFKi-treated mice showed cleared bacterial infection and survived, whereas etanercept-treated mice showed large liver granulomas and quickly died. Moreover, TNFKi mice infected with the virulent H37Rv M. tuberculosis showed resistance to infection, in contrast with etanercept-treated mice or controls. Depending on the TNF-α blockade strategy, treating arthritis with a TNF-α inhibitor could result in a different profile of infection suceptibility. Our TNFKi vaccine allowed for a better remaining host defense than did etanercept.
Keywords: host-defense; infection; rheumatoid arthritis; tumor necrosis factor; vaccine.