Main

The lifestyle of oomycetes varies from saprophytic to pathogenic. Phytophthora infestans is one of the most destructive potato pathogens, causing losses of US$6.7 billion annually worldwide. P. infestans str. T30-4 is the third Phytophthora species to be sequenced, after Phytophthora ramorum and Phytophthora sojae . At 240 Mbp, it has the largest genome of the sequenced chromalveolates1,2. Comparison of the 3 Phytophthora spp. genomes with those of 10 other eukaryotes3 revealed numerous small duplications of 2–3 consecutive genes in the 3 Phytophthora spp., possibly indicating that a whole genome duplication event occurred in their common ancestor.

The P. infestans genome contains around 18,000 protein-coding genes, of which half are shared between the 3 Phytophthora spp. genomes. Strikingly, around two-thirds of the genome contains repetitive DNA that is non-randomly distributed, giving rise to repeat-rich and repeat-poor regions. These repeats may have mediated the species-specific rearrangements that shape the Phytophthora spp. genomes. The repeat-rich regions are mostly composed of expanded families of Gypsy DNA elements, which, together with a second element termed 'Albatross', account for at least 29% of the genome and are the main cause of the differences in genome sizes between Phytophthora spp.

Pathogenicity of Phytophthora spp. is mediated by secreted effector proteins that modulate host physiology4,5. Effector proteins include plant tissue-degrading hydrolytic enzymes (such as proteases and lipases), inhibitors of host defence-related proteins, and toxins. The RLXR and Crinkler (CRN) effectors are mainly encoded in the repeat-rich regions and contain a conserved amino-terminal signal for host targeting and a highly diverse carboxy-terminal effector domain. The 563 predicted RXLR effector proteins in P. infestans represent a highly diverse lineage-specific expansion containing 151 distinct subfamilies. Expansion of the effector gene families seems to have been facilitated by non-allelic homologous recombination and tandem gene duplication, and it is likely that their localization in repeat-rich regions of the genome facilitated their rapid evolution.

The genome of the related oomycete Pythium ultimum 6, another plant pathogen, has also been sequenced recently, allowing a further in-depth analysis of the genomic changes in the oomycete lineages. Most Pythium spp. are soil inhabitants that live as saprophytes or facultative plant pathogens. Despite there being many phenotypic features (such as lifestyle, host range and host specificity) that differentiate the phytopathogenic Pythium spp. from the sequenced Phytophthora spp., there is considerable sequence conservation between the genomes. The gene order is largely the same between the sequenced species, although it is frequently interrupted by local segmental DNA inversions.

The 42.8 Mb P. ultimum DAOM BR144 genome contains 15,297 protein-coding genes but lacks genes encoding RXLR effectors and has fewer genes encoding CRN effectors. The conservation of the export signal of CRN proteins and its variable expansion across sequenced oomycetes highlights its early evolution in the last common ancestor of Pythium and Phytophthora. In addition, using known properties of effector genes, such as localization to gene-poor regions, the presence of long non-coding regions and inclusion in the predicted secretome, a new family of candidate effectors was identified and found to be expanded in Phytophthora spp.

Key host-degrading enzymes such as cutinases and endoxylanases are absent in P. ultimum, and there are fewer carbohydrate hydrolases than in Phytophthora spp., reflecting the differences in host specificities and pathogenicity mechanisms between the genera. P. ultimum also contains an expanded repertoire of subtilisin-like proteases. Surprisingly, P. ultimum and Phytophthora spp. genomes encode cell adhesion proteins of the cadherin family, marking the first time that these proteins have been detected outside metazoa.

These studies show how multispecies comparisons can address questions such as the evolution of pathogenicity. They also highlight the fact that not all oomycete plant pathogens contain a “similar toolkit for survival and pathogenesis” (Ref. 6).