Because of β-lactamase-mediated resistance, β-lactam antibiotics were long considered ineffective drugs for tuberculosis (TB) treatment. However, some β-lactams, including meropenem and faropenem, are being re-evaluated in patients infected with TB. Penicillin-binding protein (PBP) 3, or ftsI, is an essential transpeptidase in Mycobacterium tuberculosis (Mtb) required for cell division, and thus it is an important drug target. Structures of apo MtbPBP3 and of complexes with five β-lactams, including meropenem and faropenem, reveal how they cause inactivation via formation of hydrolytically stable acyl-enzyme complexes. The structures reveal unique features of the antibiotic interactions, both in terms of differences in their binding to MtbPBP3 and in comparison with structures of other PBPs and serine β-lactamases, including the tautomerization status of the carbapenem-derived acyl-enzyme complexes. The results suggest that rather than hoping PBP inhibitors developed for other infections will work against TB, work should focus on developing PBP inhibitors specialized for treating TB. SIGNIFICANCE STATEMENT: The structures of Mycobacterium tuberculosis penicillin-binding protein 3, an essential protein in M. tuberculosis, in complex with a number of widely used β-lactam antibiotics (e.g., meropenem, aztreonam, and amoxicillin) were solved. These data provide new insights for next-generation rational approaches to design tuberculosis (TB)-specific β-lactam or nonlactam antibiotics. This manuscript is a seminal article in the field of anti-TB drug discovery and suitable for the broad readership.