The spike glycoprotein of infectious bronchitis virus (IBV), a coronavirus, is translated as a precursor protein (So), then cleaved into two subunits (S1 and S2) by host cell serine proteases. In this study, we compared the cleavage recognition site of 55 IBV isolates to determine if the cleavage recognition site sequence, which consists of five basic amino acid residues, correlates with host cell range, serotype, geographic origin, and pathogenicity as it does in orthomyxoviruses and paramyxoviruses. The most common cleavage recognition site observed (33 of 55 viruses) was Arg-Arg-Ser-Arg-Arg, representing at least 11 different serotypes. Thus, cleavage recognition site does not appear to correlate with serotype. We also determined that cleavage recognition site sequence does not correlate with pathogenicity because attenuated and pathogenic isolates (different passages of the same virus) contain identical cleavage recognition site sequences. In addition, nephropathogenic strains had the same cleavage recognition site sequence as many nonnephropathogenic isolates. Cleavage recognition site sequence does correlate with viruses in different geographic regions, which may be an important characteristic to examine in epidemiologic studies. An IBV monoclonal antibody neutralization-resistant mutant (NR 18) had an unusual substitution of Ile for Arg at the fourth position, giving the sequence Arg-Arg-Ser-Ile-Arg, which likely prevents cleavage and, thus, destroys the conformationally dependent monoclonal antibody binding epitope. Six residues on the amino-terminal side of the cleavage recognition site are conserved in 31% of the isolates and consist of only one or two basic amino acids. Thus, the number of basic residues around the cleavage recognition site does not appear to correlate with increased cleavability, host cell range, and increased virulence as it does with envelope glycoproteins in orthomyxoviruses and paramyxoviruses.