Microarray-based comparative genomic profiling of reference strains and selected Canadian field isolates of Actinobacillus pleuropneumoniae

Abstract

Background: Actinobacillus pleuropneumoniae, the causative agent of porcine pleuropneumonia, is a highly contagious respiratory pathogen that causes severe losses to the swine industry worldwide. Current commercially-available vaccines are of limited value because they do not induce cross-serovar immunity and do not prevent development of the carrier state. Microarray-based comparative genomic hybridizations (M-CGH) were used to estimate whole genomic diversity of representative Actinobacillus pleuropneumoniae strains. Our goal was to identify conserved genes, especially those predicted to encode outer membrane proteins and lipoproteins because of their potential for the development of more effective vaccines.

Results: Using hierarchical clustering, our M-CGH results showed that the majority of the genes in the genome of the serovar 5 A. pleuropneumoniae L20 strain were conserved in the reference strains of all 15 serovars and in representative field isolates. Fifty-eight conserved genes predicted to encode for outer membrane proteins or lipoproteins were identified. As well, there were several clusters of diverged or absent genes including those associated with capsule biosynthesis, toxin production as well as genes typically associated with mobile elements.

Conclusion: Although A. pleuropneumoniae strains are essentially clonal, M-CGH analysis of the reference strains of the fifteen serovars and representative field isolates revealed several classes of genes that were divergent or absent. Not surprisingly, these included genes associated with capsule biosynthesis as the capsule is associated with sero-specificity. Several of the conserved genes were identified as candidates for vaccine development, and we conclude that M-CGH is a valuable tool for reverse vaccinology.

Figure 1

Number of reference strains representing the 15 serovars of A. pleuropneumoniae where gene is divergent or missing for each gene, ordered as in A. pleuropneumoniae L20 genome sequence and based on M-CGH results. (Blue: -3<Log2Ratio<-1, Red: Log2Ratio<-3). RM, DNA restriction/modification enzymes; TNP, transposon; LPS, lipopolysaccharide biosynthesis genes; RTX, toxin genes; CHO, carbohydrate biosynthesis genes.

Figure 2

Hierarchical clustering of A. pleuropneumoniae reference strains based on M-CGH data excluding phage and transposase genes. The dendrogram was produced using the MEV software from the J. Craig Venter Institute with Euclidean distance and average linkage clustering (n = 1000 bootstrap iterations).

Figure 3

Variability of genes involved in toxin and capsule biosynthesis across the A. pleuropneumoniae reference strains representing the 15 serovars. Green indicates that the gene is divergent in sequence or absent in the tester strain.

Figure 4

Number of strains from 15 fresh field isolates of A. pleuropneumoniae serovars 1, 7 and 15 where the sequence is divergent or absent for each gene, ordered as in A. pleuropneumoniae L20 genome sequence and based on M-CGH results.

Figure 5

Hierarchical clustering of field isolates of A. pleuropneumoniae based on M-CGH data excluding phage and transposase genes.

Figure 6

Comparison between M-CGH Log2Ratio data from A. pleuropneumoniae serovar 1 versus serovar 5b hybridizations independently collected, scanned and processed using either ArrayPro software or SpotFinder/MIDAS software. At least 2 independent hybridizations were performed for each dataset, and the processed normalized data was averaged between replicate experiments and between pairs of duplicate spots on the microarray.