The antigenic lipopolysaccharide 0 polysaccharides of capsular serotypes 9 and 11 were examined by chemical, immunological, and nuclear magnetic resonance methods. Immunodiffusion tests carried out on these 0 antigens indicated that both contained common epitopes which were also shared by Actinobacilus peuropneumonae serotype 1. Chemical analysis and high-field nuclear magnetic resonance spectroscopy showed that the 0 antigens of serotypes 9 and 11 were high-molecular-weight polymers consisting of a backbone of repeating trisaccharide units composed of a-L-rhamnopyranosyl and a-D-glucopyranosyl residues (2:1). One of the c-L-rhamnose units forms a branch point and is stoichiometrically substituted with terminal 2-acetamido-2-deoxy-1-D-glucose residues in the serotype 11 0 polysaccharide, but only to the extent of 25% in the serotype 9 0 polysaccharide. Thus, the serotype 9 0 polysaccharide contains two different repeating units: a tetrasaccharide unit with the same structure as that of the serotype 11 0 polysaccharide and a trisaccharide unit:where R = j3-D-GlcpNAc for serotype 1 and 11 0 polysaccharides, and R = H (75%) and R = f-D-GlcPNAc (25%o) for serotype 9. The structure of the previously determined serotype 1 0 polysaccharide (E. Altman, J.-R Brisson, and M. B. Perry, Biochem. Cell. Biol. 64:17-25, 1986) is identical to that of the serotype 11 0 polysaccharide. We propose a more complete serotyping scheme for A. pleuropneumoniae which includes designation of both the capsular (K) and 0 antigens. Actinobacillus pleuropneumoniae is a gram-negative, pleomorphic rod which causes a highly contagious disease in pigs. The occurrence of A. pleuropneumoniae-associated pleuropneumonia has increased since 1976 (29) because of intensified pig production, and A. pleuropneumoniae lipopolysaccharide (LPS) is recognized as an important virulence factor (30). In attempts to combat the disease, serological methods of identification of infected animals based on the identification and characterization of serotype-specific antigens would greatly assist in solving the problem. Information on the structures of the LPS 0 antigens of A. pleuropneumoniae involved as virulence factors, in protection, and in serodiagnosis could provide data leading to the rational development of suitable, protective vaccines and the production of specific serotyping and serodiagnostic reagents.Currently, 12 serotypes ofA. pleuropneumoniae based on capsular polysaccharides (CPS) are recognized (26). Capsular serotypes 1, 5, and 7 are the predominant disease isolates from A. pleuropneumoniae outbreaks in the United States and Canada (28), whereas serotype 9 is predominant in The Netherlands (19). Serological cross-reactivity between A. pleuropneumoniae serotypes 1 and 9 has been detected by * Corresponding author. t NRCC no. 34260. complement fixation and gel diffusion methods (25). This cross-reactivity was attributed to common polysaccharide antigens by Mittal (24), who also concluded that these serotypes do not share CPS epitopes but that the cross...