Powdery mildew genomes reloaded


Third International Powdery Mildew Workshop in Copenhagen, Denmark, August 2013

Nearly two years ago the Second International Powdery Mildew Workshop in Zürich (Switzerland) brought together 34 delegates from 11 countries to discuss current aspects of powdery mildew biology (Spanu & Panstruga, 2012). Powdery mildews are ascomycete fungi that colonize many plant species and cause severe disease resulting in significant yield losses (Glawe, 2008). A particularly startling aspect of their biology is the obligate biotrophic lifestyle, which these fungi share with very distantly related phytopathogens such as the rust fungi (basidiomycetes) and downy mildews (oomycetes; Kemen & Jones, 2012). At the 2012 meeting, kindly supported by the New Phytologist Trust as the Third New Phytologist Workshop, a major focus was the discussion on in silico and experimental analyses of powdery mildew genomes. Back then, only the draft genome of the barley powdery mildew pathogen, Blumeria graminis f.sp. hordei (Bgh; isolate DH14; Fig. 1a), and the very fragmented genomes of the pea- and Arabidopsis-colonizing powdery mildew fungi, Erysiphe pisi and Golovinomyces orontii, were accessible (Spanu et al., 2010). Part of the joint activities planned and discussed at the 2012 gathering related to the genome-wide analysis of genes encoding effectors – small, secreted proteins that the fungus supposedly uses to undermine host defence and to accommodate its dedicated infection structures (haustoria; Fig. 1b; Panstruga & Dodds, 2009). Close inspection by a subgroup of the workshop attendees revealed the existence of c. 500 genes coding for such ‘candidate secreted effector proteins’ (CSEPs) in the Bgh genome, of which 72 have a distant relationship to secreted fungal ribonucleases (Pedersen et al., 2012). Experimental evidence indicates that at least some of these ribonuclease-like CSEPs are bona fide effectors that function within the plant cell by contributing to the infection process and interfering with pathogen-triggered host cell death (Pliego et al., 2013). Another aspect recognized by the delegates of the 2012 workshop was the need for further powdery mildew genomes to reach better conclusions about the evolution, pathogenicity mechanisms and biology of these widespread plant parasites. In the following year, the draft genome of one isolate of the wheat powdery mildew pathogen, Blumeria graminis f.sp. tritici (Bgt), a close relative of Bgh, and the resequenced genomes of three additional Bgt isolates, were published by a team led by some of the workshop participants (Wicker et al., 2013). Comparative analysis of these four Bgt genomes revealed that they are mosaic of ancient haplotypes (Wicker et al., 2013). The same conclusion was reached from the analysis of two additional Bgh isolate genomes (A6 and K1) by a different team, which suggests that at least the cereal-colonizing powdery mildews Bgh and Bgt largely propagate by asexual reproduction and that mating and sexual recombination is a rare event for these plant parasites (Hacquard et al., 2013). The latter study revealed also an unexpected conservation of Bgh transcriptional programmes during early stages of compatible and incompatible interactions on host (barley) and immune-compromised Arabidopsis thaliana nonhost plants (Hacquard et al., 2013). Notably, many of the earlier-mentioned publications build to some extent upon discussions at the 2012 workshop, demonstrating the strong impact this meeting had on the recent progress made within the community.

Figure 1.

Blumeria graminis f.sp. hordei structures. (a) Sporulating Bgh colonies. (b) Fluorescently-labelled Bgh haustorium. Bars: (a) 2 mm; (b) 20 μm.

‘… a vibrantly discussed topic was the evolutionary history of “formae speciales” of cereal powdery mildews …’

Inspired by the fruitful collaborations (including several joint publications and grant applications) initiated at the workshop in Zürich, the Third International Powdery Mildew Workshop took place in August 2013. More than 50 participants from seven different countries gathered at the University of Copenhagen (Denmark) to exchange progress and debate future directions on various aspects of the analysis of plant–powdery mildew interactions. This time a vibrantly discussed topic was the evolutionary history of ‘formae speciales’ of cereal powdery mildews such as Bgh and Bgt. Two contrasting hypotheses place their divergence from a common ancestor either back in the Miocene (c. 4.7–11 million yr ago), shortly after the divergence of the host plants barley (Hordeum vulgare) and wheat (Triticum aestivum) from their last common ancestor (Takamatsu & Matsuda, 2004; Inuma et al., 2007; Wicker et al., 2013), or in the Holocene, coinciding with early agriculture and wheat domestication, < 10 000 yr ago (Wyand & Brown, 2003). While the former estimate relies on molecular clock calculations, the latter rests on the seeming incongruence of host and pathogen phylogenies, which is inconsistent with a classical model of host–pathogen co-evolution. A suboptimal choice of molecular clock genes, unusually high mutation rates of sunlight (UV)-exposed epiphytic fungal structures in combination with a lack of genetic variation in introduced populations may account for an overestimation of the BghBgt divergence time by conventional molecular clock analysis (Troch et al., 2014). More precise calculations will require estimates on the basis of experimentally supported mutation rates in powdery mildews – a challenging task for the future. With this and several other questions (e.g. development of transformation protocols, cloning of novel avirulence genes) remaining unresolved for the time being, we look forward to the Fourth Powdery Mildew Meeting, which will be held at Kew Gardens (London, UK) in September 2014 (http://www.blugen.org/index.php?page=4thpowderymildewworkshop).