Preclinical efficacy of a cell division protein candidate gonococcal vaccine identified by artificial intelligence

A safe and effective vaccine is urgently needed to combat the global threat of multidrug-resistant (MDR) Neisseria gonorrhoeae. We screened 26 gonococcal proteins discovered by an artificial intelligence-driven platform called Efficacy Discriminative Educated Network (EDEN) trained to identify novel, protective vaccine antigens against pathogenic bacteria for efficacy in the mouse vaginal colonization model of gonorrhea. Combinations of two to three antigens adjuvanted with GLA-SE (induces T(H)1 responses) yielded 11 groups that were used to vaccinate mice. An inverse correlation was noted between the complement-dependent bactericidal activity of antisera from each of the 11 groups and the burden of gonococcal colonization. The combination of NGO1549 (FtsN; cell divisome protein) and NGO0265 (predicted cell division protein) most substantially reduced the burden of colonization by MDR strain WHO X. The EDEN prediction score for each group of antigens correlated positively with reductions in overall bacterial burden, providing evidence for its predictive potential. FtsN and NGO0265 administered either individually, in combination, or as a chimeric protein significantly attenuated gonococcal vaginal colonization by all three test strains. IgG in antisera from mice immunized with the chimeric NGO0265-FtsN protein supported the complement-dependent killing of all 50 (100%) gonococcal isolates tested. The efficacy of the chimeric NGO0265-FtsN vaccine required the membrane attack complex (C5b-9) of complement, evidenced by loss of efficacy in complement C9(-/- )mice. In conclusion, a chimeric molecule comprising NGO0265 and FtsN adjuvanted with GLA-SE elicits IgG with broad anti-gonococcal bactericidal activity, attenuates gonococcal colonization in a complement-dependent manner, and represents a promising gonococcal vaccine candidate.