Rationally designed mariner vectors for functional genomic analysis of Actinobacillus pleuropneumoniae and other Pasteurellaceae species by transposon-directed insertion-site sequencing (TraDIS)

Abstract

Comprehensive identification of conditionally essential genes requires efficient tools for generating high-density transposon libraries that, ideally, can be analysed using next-generation sequencing methods such as Transposon Directed Insertion-site Sequencing (TraDIS). The Himar1 (mariner) transposon is ideal for generating near-saturating mutant libraries, especially in AT-rich chromosomes, as the requirement for integration is a TA dinucleotide, and this transposon has been used for mutagenesis of a wide variety of bacteria. However, plasmids for mariner delivery do not necessarily work well in all bacteria. In particular, there are limited tools for functional genomic analysis of Pasteurellaceae species of major veterinary importance, such as swine and cattle pathogens, Actinobacillus pleuropneumoniae and Pasteurella multocida, respectively. Here, we developed plasmids, pTsodCPC9 and pTlacPC9 (differing only in the promoter driving expression of the transposase gene), that allow delivery of mariner into both these pathogens, but which should also be applicable to a wider range of bacteria. Using the pTlacPC9 vector, we have generated, for the first time, saturating mariner mutant libraries in both A. pleuropneumoniae and P. multocida that showed a near random distribution of insertions around the respective chromosomes as detected by TraDIS. A preliminary screen of 5000 mutants each identified 8 and 14 genes, respectively, that are required for growth under anaerobic conditions. Future high-throughput screening of the generated libraries will facilitate identification of mutants required for growth under different conditions, including in vivo, highlighting key virulence factors and pathways that can be exploited for development of novel therapeutics and vaccines.

Keywords: Actinobacillus pleuropneumoniae; Mariner; Pasteurella multocida; Pasteurellaceae; TraDIS; Transposon.

Fig. 1

General map of mariner vectors, pTsodCPC9 and pTlacPC9, differing only in the promoter for the tnp gene. Genes tnp (mariner transposase C9 mutant), cat (chloramphenicol resistance), traJ (plasmid transfer gene), and bla (ampicillin resistance) are indicated by the appropriately labelled solid red arrows. The origin of plasmid transfer (oriT) is shown as a filled lollipop (), and the origin of plasmid replication (colE1) as a hollow lollipop (). Coloured blocks flanking the cat gene indicate the locations of Himar1 inverted repeats (thick orange), and paired copies of DNA uptake sequences for Neisseria spp. (thin green), H. influenzae (thin pink) and A. pleuropneumoniae (thin blue). Arrowhead upstream of the C9 tnp gene indicates promoter sequences for either A. pleuropneumoniae sodC or E. coli lac genes, depending on the vector (pTsodCPC9 and pTlacPC9, respectively)

Fig. 2

Linker PCR products for 12 randomly selected A. pleuropneumoniae mutants. Sequences flanking mariner insertions were amplified from AluI-digested linker-ligated DNA fragments using primers L-PCR-C and IR-Left_out (for amplification of the left flank). Lanes 1–12 show single amplicons for each of selected mutants, indicating the presence of a single insertion of transposon in each case

Fig. 3

Comparison of linker PCR-products generated from pooled A. pleuropneumoniae library genomic DNA +/− ISceI digestion. Lanes: 1) 100 bp ladder; 2) ISceI-treated sample; 3) untreated sample. Amplification products were generated for the left flanking sequences, as in Fig. 2. The dominant plasmid band in the untreated sample is indicated

Fig. 4

Distribution of mariner insertions identified in the A. pleuropneumoniae and P. multocida genomes. TraDIS reads for the respective pooled libraries were mapped to A the complete genome of A. pleuropneumoniae MIDG2331 (accession number LN908249); and D the draft genome (pseudochromosome) of P. multocida MIDG3277 (accession number ERZ681052). Each spike plotted around the chromosome represents a single insertion site, with the length of each spike proportional to the number of mapped sequence reads from that insertion site. A total of 78,638 unique insertion sites were identified in the A. pleuropneumoniae MIDG2331 library, and 147,613 in the P. multocida MIDG3277 library. In the MIDG3277 dataset, there were several insertion sites with large numbers of mapped reads. To enable insertions with fewer reads to be seen clearly, read coverage has been capped to a maximum of 50,000 in D (the true maximum coverage at an insertion site was 136,032 reads). The pseudochromosome of P. multocida MIDG3277 was assembled based on ordering of the draft sequence contigs following alignment, using NUCmer 4.0 (43), with the complete genome of P. multocida Pm70, as shown in C. Arrows indicate the position and orientation of the contigs. Red blocks indicate matches in the same orientation, blue blocks indicate matches in the reverse orientation. Plots of the cumulative insertion counts across the MIDG2331 chromosome B and MIDG3277 pseudochromosome E are shown in red, with a dotted line indicating the expected relationship for uniformly distributed insertions. Both libraries deviate from this, with a bias towards insertions close to the origin of replication, but insertions are found across the genome in both libraries