White-tailed deer (Odocoileus virginianus) have emerged as reservoirs of bovine tuberculosis in northern America. For tuberculosis surveillance of deer, antibody-based assays are particularly attractive because deer are handled only once and immediate processing of the sample is not required. Sera collected sequentially from 25 Mycobacterium bovis-infected and 7 noninfected deer were evaluated by enzyme-linked immunosorbent assay (ELISA), immunoblotting, and multiantigen print immunoassay (MAPIA) for immunoglobulin specific to M. bovis antigens. Various routes of experimental M. bovis infection, such as intratonsillar inoculation (n = 11), aerosol (n = 6), and exposure to infected deer (in contact, n = 8), were studied. Upon infection, specific bands of reactivity at approximately 24 to 26 kDa, approximately 33 kDa, approximately 42 kDa, and approximately 75 kDa to M. bovis whole-cell sonicate were detected by immunoblot. Lipoarabinomannan-specific immunoglobulin was detected as early as 36 days postchallenge, and responses were detected for 94% of intratonsillarly and "in-contact"-infected deer. In MAPIA, sera were tested with 12 native and recombinant antigens coated on nitrocellulose. All in-contact-infected (8 of 8) and 10 of 11 intratonsillarly infected deer produced antibody reactive with one or more of the recombinant/native antigens. Responses were boosted by injection of tuberculin for intradermal tuberculin skin testing. Additionally, three of six deer receiving a very low dose of M. bovis via aerosol exposure produced antibody specific to one or more recombinant proteins. M. bovis was isolated from one of three nonresponding aerosol-challenged deer. Of the 12 antigens tested, the most immunodominant protein was MPB83; however, a highly sensitive serodiagnostic test will likely require use of multiple antigens.