A: Mechanisms of mobility of pathogenicity islands

Minimal recombination modules of four different asn tDNA-associated genomic islands (HPI from Y. pseudotuberculosis, Ecoc54N from uropathogenic E. coli CFT073, HAI7 and HAI13 from the plant pathogen Erwinia carotovora) were reconstructed. All modules are highly homologous at the DNA sequence level.  
The ability of island integrases to use foreign attP sites as substrates for the site-specific recombination were investigated by in vivo trans-recombination assay. For this each attP site was recovered and cloned in a pir-dependent suicide plasmid. Each plasmid was evaluated for recombinogenic activity in E. coli RecA- cells with each of the four island-encoded integrase proteins.

Only integrases from HPI and Ecoc54N islands were able to support recombination not only of their own cognate attP sites. HPI integrase was able to assist recombination of attPEcoc and Ecoc54N integrase - of attPHPI. Surprisingly, integrases from HAI7 and HAI13 islands were not able to substitute each other in integrative site-specific recombination assay although they have even higher homology. Functional organization of the integrase protein might explain this phenomenon. Most differences between IntHAI7 and IntHAI13 accumulate in the N-terminal part of the proteins. For lambda integrase it has been shown that the N-terminal domain is responsible for attP site recognition. Higher discrepancy of the attPHAI7 and attPHAI13 nucleotide sequences serves as an additional proof of this proposal.

Aim: To address the cross-talk between four different asn-tRNA-associated genomic islands, especially the genes and proteins involved in recombination.

This project is executed by Uladzimir Antonenka (PhD) and supported by the DFG "Yersinia pestis High-pathogenicity island: Mechanisms and Structures responsible for its Dissemination".

B: Comparative phylogenomics of Yersinia

The complete sequence of Yersinia enterocolitica palearctica Y11 virulence plasmid pYVO3 has been extracted from the sequencing data as a set of 15 single contigs that were mapped to the pYVe227 plasmid (Y. enterocolitica O9) chosen for reference. Contigs were accommodated in right order by PCR approach and gaps were filled in by directed sequencing. Finally, a single sequence consisting of ~ 72 kb is now available and under investigation for single nucleotide changes in comparison with the serogroup O:9 plasmid pYVe227. Both plasmids show an amazingly highly related plasmid sequence but are different in organization. Interestingly, an additional IS element (ISyen2) was found in pYVO3 plasmid.
Using the Staden package software, different data sets from two genome sequencing performances were combined: ABI derived Sanger data and “454” pyrosequencing results. Due to the software limitations, the single read traces are not complete for exact analysis yet. With Staden package program gap4 and Roche GS software the contigs number was reduced to 726 (including 341 large ones) with 4473296 bases and a median sequencing deepness of 17. 99.69 % of the large size contigs show high quality values of or above 40, expressed statistically as a 99.99 % probability for every nucleotide.

Aim: To complete Y. enterocolitica palearctica O3 strain Y11 sequence and compare it with the published Yersinia sequences. Manual curation of the sequence differences will reveal genetic changes that will be used for construction of patho-arrays. Additional focus will be placed on the rtxA-like gene in Yersinia enterocolitica O3. Creation of a mutant strain should give insight into the mode of action of this novel toxin gene product.

This project is executed by Julia Batzilla. It is supported by the BMBF Interdisciplinary Research Network, FBI-Zoo  "Molecular characterization of Yersinia enterocolitica serotype O3: A genome-based approach".