Use of Single-Enzyme Amplified Fragment Length Polymorphism for Typing Pasteurella multocida subsp. multocida Isolates from Pigs

Bacteriology.

Ninety-seven strains of P. multocida subsp. multocida were identified through biochemical characterization, including production of catalase, oxidase, and indol, urease activity, production of ornithine decarboxylase, and carbohydrate fermentation (1, 7). The strains were obtained between 1999 and 2002 from 67 herds from nine states in Brazil: São Paulo, Santa Catarina, Paraná, Espírito Santo, Minas Gerais, Mato Grosso do Sul, Goiás, Rio de Janeiro, and Rio Grande do Sul. The animals presented pneumonia, atrophic rhinitis, and septicemia. For PCR and SE-AFLP, cultures were grown in brain heart infusion broth (Difco) at 37°C for 18 to 24 h, and DNA was extracted using the guanidine thiocyanate method (2).

Capsular typing.

Capsular serotyping was conducted employing the indirect hemagglutination test as previously described (17). Results were confirmed by the multiplex capsular PCR assay recently described and which generates fragments of 1,044, 760, 657, 511, and 851 bp by P. multocida types A, B, D, E and F, respectively (18).

Toxin gene detection.

All strains were tested for the presence of the gene that codifies toxin production using the PCR previously described (13).

SE-AFLP subtyping.

Ten microliters of DNA was digested overnight at 37°C with 24 U of HindIII in the buffer of enzyme and water in a volume of 20 μl. Five microliters of digested DNA was used in a ligation reaction containing 0.2 μg of each adapter oligonucleotide (15), 1 U of T4 DNA ligase, ligase buffer, and water, in a final volume of 20 μl incubated at room temperature for 3 h. Ligated DNA was heated to 80°C for 10 min and diluted 1/5 in sterile distilled water, and 5 μl was used for each PCR.

PCRs consisted of 5 μl of ligated DNA, 2.5 mM MgCl₂, 300 ng of primer (15), and 1.25 U of Taq DNA polymerase, PCR buffer in a final volume of 50 μl. Cycles consisted of 94°C for 4 min followed by 35 cycles of 94°C for 1 min, 60°C for 1 min, and 72°C for 2.5 min.

The amplicons were analyzed on a 2.0% agarose gel and registered by an image capturing system (ImageMaster VDS, Amersham Pharmacia Biotech). The 100-bp DNA ladder (Invitrogen) was included twice on each gel. Banding patterns were assessed visually and under a code, with no knowledge about serotyping results or epidemiological data.

The discriminatory power of typing methods was calculated following the description provided by Hunter and Gaston (11).

Infection with P. multocida subsp. multocida is widespread in Brazilian swine herds. Capsular typing of the 97 strains showed 72 strains to be type A, 22 strains to be type D, and three strains to be type F. Eleven strains were toxigenic, where three were type A and eight were type D strains. The distribution of type A and type D strains in the lungs and nasal cavity was similar to the data provided previously (12, 14). There had been no reports of type F infection in pigs before.

Using SE-AFLP, 7 to 12 DNA fragments ranging between approximately 400 and 1,400 bp were studied. Eighteen SE-AFLP profiles, designated A to R, were observed (Fig. 1). To test the reproducibility, three separate preparations of 22 P. multocida isolates were subjected to SE-AFLP, and no band variation was seen. However, some variations in the intensities of the bands were observed in different PCR runs.

• Open in new tab
• Download powerpoint

FIG. 1.

SE-AFLP profiles of P. multocida subsp. multocida produced by primer HIG. Lanes 1 and 20, 100-bp molecular weight standard. The remaining lanes show AFLP profiles (A to R).

The discriminatory index of the SE-AFLP was 0.87. A comparison between plasmid profile, phage typing, and HindIII ribotyping showed discriminatory indexes of 0.16, 0.77, and 0.78, respectively (9). Discriminatory indexes of 0.74 and 0.84 were registered using RFLP of the ompH gene with HinfI and RsaI restriction enzymes (3).

The relationship that exists among SE-AFLP profiles and the phenotypic data are illustrated in Fig. 2. Strains from the same herd showed one to three different SE-AFLP profiles; however, they were clustered together. The majority of type D strains isolated from the nasal cavity and toxin-producing strains were allocated in the same cluster, i.e., AIIa. Strains isolated from the nasal cavity were present in other clusters, too. Some authors suggest that this occurs because P. multocida from pneumonic lung lesions can colonize the nasal cavity and vice versa (9). Type F strains showed two different profiles, which were not observed in type A and D strains but which were allocated in the same cluster of three type A strains (AIIId) with 80% similarity.

• Open in new tab
• Download powerpoint

FIG. 2.

Dendrogram showing the relationship between P. multocida subsp. multocida isolates on the basis of AFLP patterns. Percentages of similarity between patterns were calculated by use of Jaccard's coefficient. The dendrogram was constructed by use of UPGMA (unweighted pair group method using arithmetic average).

Among type A strains, a high level of genetic heterogeneity was observed. Previous reports had also described the heterogeneous nature of serogroup capsular A isolated from different animal species (8).

With respect to the states of origin of pigs, strains with common profiles were observed in different regions of Brazil. This can be attributed to the distribution of breeders from different genetic sources across the country.

SE-AFLP was simple and easy to use or standardize and proved to have great potential for subtyping strains of P. multocida subsp. multocida isolated from pigs.