101

Mechanisms & Processes

Erwinia carotovora subsp. atroseptica (Eca) is an eco-nomically important pathogen of potato, causing

blackleg of plants in the field and soft rot of tuberspost-harvest. Its pathogenicity is primarily dependanton the tightly regulated production of large amountsof extracellular enzymes that degrade plant cell walls,with other factors such as iron acquisition and mecha-nisms to defend against plant attack also playing arole. In recent years, however, it has become clearthat soft rot pathogenesis is more complex than previ-ously thought and the relationship between Eca andpotato / non-host plants is still far from understood.As a new approach togene discovery in Eca,the complete genomesequence and annota-tion of Eca was deter-mined incollaboration withthe Sanger Institute,Cambridge, UK andSCRI through SEER-AD funding. Thegenome is ca 5 Mbwith 4, 491 codingsequences.

Analysis of thegenome, and compar-ison with 60 otherbacterial genomesusing bioinformaticshas revealed a wealthof new information,including putativepathogenicity factorspreviously unknownin this organism. Forexample, we have dis-covered i) a numberof putative toxingenes, including those possibly involved in the forma-tion of the polyketide-based coronafacic acid (part ofthe plant toxin coronatine produced by Pseudomonassyringae during infection); ii) a cluster of genes similarto a type IV secretion system that, in the plantpathogen Agrobacterium tumefaciens, plays a majorrole in the disease process.

We have also found that the number of pathogenicitygenes acquired from more distantly-related bacteria,possibly via horizontal gene transfer, was greater thanexpected. Many of these distantly-related bacteria areplant pathogenic or plant associated, suggesting thatEca may have developed its plant pathogenic lifestylethrough gain of important genes, following exchangeof DNA with bacteria relevant to a plant associatedlifestyle.

In collaboration with the Phytophthora infestans groupat SCRI, we have developed a ‘transposon mutation

grid’, allowing pooledlibraries of transposonmutants to be searchedrapidly for mutations inany given gene in thegenome. We also havepotato plants, obtainedas miniplants from acommercial source,available for disease test-ing throughout theyear. Using this dualapproach over the last 6months, we have isolat-ed over 20 Eca mutantsand determined the roleof some importantnovel genes inpathogenicity, includingthose associated withboth the coronafacicacid and type IV secre-tion system.

Finally, a number ofother functionalgenomics programmesare being developed i) atSCRI, including micro-arrays containing the

complete set of Eca coding sequences, to study thegenome at the gene expression level both in vitro andin planta; ii) in collaboration with other institutions,such as Cambridge University and Moredun ResearchInstitute, including proteomics to study the genomeat the protein level.

New discoveries with Erwinia genomicsI.K. Toth, L. Pritchard, M.C. Holeva, L.J. Hyman, K.S. Bell, S.C. Whisson, A.O. Avrova & P.R.J. Birch

Figure 1 Comparison of the Eca genome sequence with other bacterial genomes: Inner to outer tracks: the locations of reciprocal best hits found by reciprocal FASTA of Eca CDSs against those from 32 bacterial genomes: Gram+ (grey); Shewanella oneidensis (ochre); non-enteric animal pathogens (green); plant-associated bacteria (brown); non-enteric plant pathogens (red); enterobacteria (blue). The locations of CDSs on the Eca genome, coloured by functional class. Two tracks indicating islands listed in Table 1: islands with evidence of recent acquisition (red bars), possible islands based on reciprocal FASTA analysis (green bars). A plot of G+C skew (red) and %GC content (blue).

1

1000001

2000001

3000001

4000001

5000001

PAI2

PAI3

PAI4

PAI6

PAI8

PAI1

0

PAI12

PAI13

PAI16

PAI17

PAI1

PAI5

PAI7

PA

I9

PAI1

1

PAI14

PAI15