OBJECTIVE To evaluate efficacy of a recombinant Moraxella bovis pilin-cytotoxin-Moraxella bovoculi cytotoxin subunit vaccine to prevent naturally occurring infectious bovine keratoconjunctivitis (IBK). METHODS 107 beef steers. METHODS 2 groups of calves were inoculated SC with an immunostimulating complex (ISCOM) matrix adjuvant (control group; n = 54) or a recombinant M bovis pilin-cytotoxin-M bovoculi cytotoxin subunit antigen with the ISCOM matrix adjuvant (vaccine group; 53); calves received booster injections 21 days later. Calves were examined once weekly for 16 weeks. Investigators and herd managers were not aware of the inoculum administered to each calf throughout the trial. Primary outcome of interest was the cumulative proportion of calves that developed IBK. Serum samples were obtained before inoculation (day 0) and on days 42 and 112. Serum hemolysin-neutralizing titers against native M bovis and M bovoculi cytotoxin were determined. RESULTS No difference was detected between groups for the cumulative proportion of calves that developed IBK at weeks 8 and 16 after inoculation. Non-IBK-affected calves in the vaccine group had a significantly higher fold change in serum hemolysin-neutralizing titer against native M bovoculi cytotoxin from day 0 to 42 compared to control calves. CONCLUSIONS The M bovis pilin-cytotoxin-M bovoculi cytotoxin subunit vaccine with the ISCOM matrix adjuvant was not effective at preventing naturally occurring IBK. It is likely that the incorporation of additional protective antigens in a recombinant Moraxella spp subunit vaccine will be required to yield a product that can be used for effective immunization of cattle against IBK.