The major adhesin of Actinobacillus pleuropneumoniae, the causative agent of porcine pleuropneumonia, has been previously identified as lipopolysaccharide (LPS). The purpose of the present study was to isolate and characterize A. pleuropneumoniae LPS mutants. Screening of LPS mutants was performed with colony dot and sensitivity to novobiocin. One mutant obtained by colony dot (F19) and one mutant selected for its increased sensitivity to novobiocin (33.1) did not react with a monoclonal antibody against A. pleuropneumoniae serotype 1 O-antigen compared with the parent strain. Mutants F19 and 33.1 did not express high-molecular-mass LPS bands as determined in silver-stained SDS-PAGE gels. The core-lipid A region of mutant 33.1 and of the parent strain had similar relative mobilities and reacted with serum from a pig experimentally infected with the serotype 1 reference strain of A. pleuropneumoniae, while the same region in mutant F19 showed faster migration and did not react with this serum. Use of piglet tracheal frozen sections indicated that mutant F19 was able to adhere to piglet trachea as well as the parent strain, while mutant 33.1 adhered [half as much as] the parent strain. Finally, both LPS mutants were markedly less virulent in mice than the parent strain. Taken together, our observations support the idea that LPS is an important virulence factor of A. pleuropneumoniae.

A previous study indicated that lipopolysaccharides (LPS) extracted from Actinobacillus pleuropneumoniae bind two low-molecular-mass proteins, of approximately 10 and 11 kDa, present in porcine respiratory tract secretions (M. Bélanger, D. Dubreuil, and M. Jacques, Infect. Immun. 62:868-873, 1994). In the present study, we determined the N-terminal amino acid sequences of these two proteins, which revealed high homology with the ␣ and ␤ chains of pig hemoglobin. Some isolates of A. pleuropneumoniae were able to use hemoglobin from various animal species as well as other heme compounds as sole sources of iron for growth, while other isolates were unable to use them. Immunoelectron microscopy showed binding of pig hemoglobin at the surface of all A. pleuropneumoniae isolates as well as labeling of outer membrane blebs. We observed, using Western blotting (immunoblotting), that the lipid A-core region of LPS of all isolates was binding pig hemoglobin. Furthermore, lipid A obtained after acid hydrolysis of LPS extracted from A. pleuropneumoniae was able to bind pig hemoglobin and this binding was completely abolished by preincubation of lipid A with polymyxin B but was not inhibited by preincubation with glucosamines. Fatty acids constituting the lipid A of A. pleuropneumoniae, namely, dodecanoic acid, tetradecanoic acid, 3-hydroxytetradecanoic acid, hexadecanoic acid, and octadecanoic acid, were also binding pig hemoglobin. Our results indicate that LPS of all A. pleuropneumoniae isolates tested bind pig hemoglobin and that lipid A is involved in this binding. Our results also indicate that some A. pleuropneumoniae isolates are, in addition, able to use hemoglobin for growth. Binding of hemoglobin to LPS might represent an important means by which A. pleuropneumoniae acquires iron in vivo from hemoglobin released from erythrocytes lysed by the action of its hemolysins.

No abstract

Lipopolysaccharides (LPS) of Actinobacillus pleuropneumoniae were separated by Tricine-SDS-polyacrylamide gel electrophoresis, which has been shown to improve resolution of low-molecular-mass fast migrating bands. Strains representing the 12 serotypes of A. pleuropneumoniae can be divided in two groups according to the gel mobility of the core - lipid A region of their LPS. The first electromorphic core type (core type I), found in serotypes 1, 6, 9, and 11, had a migration slower than Salmonella typhimurium Ra LPS. The second electromorphic core type (core type II), found in the remaining serotypes (i.e., 2, 3, 4, 5, 7, 8, 10, and 12) had a migration similar to S. typhimurium Ra LPS. Furthermore, we observed that these two core types were antigenically different. Western blot analyses indicated that core - lipid A region of LPS from electromorphic core type I strains reacted when probed with serum from a pig experimentally infected with a core type I strain but not when probed with serum from a pig experimentally infected with a core type II strain. Conversely, core - lipid A region of LPS from electromorphic core type II strains reacted only when probed with serum from a pig experimentally infected with a core type II strain. Our results, based on both electrophoretic mobility and antigenicity, suggest the presence of two LPS core types in A. pleuropneumoniae.