Archive for May, 2011

Convergent evolution of obligate biotrophy in plant parasites

May 29th, 2011

Nearly all plants play host to a myriad of parasites. They can suffer from bacterial, viral and fungal attack, but fungal parasites are by far the most prevalent plant pathogenic organism. Over 20,000 species of fungi are parasites and cause disease in crops and plants. Parasitic fungi require a living host to survive. This forces them to achieve a delicate balance, extracting enough nutrients to ensure their own survival but not so much that they kill the plant. Common plant fungi such as powdery mildews, rusts and smuts require a living plant to sustain them.

The genomes of the hemibiotrophic (Magnaporthe oryzae, Leptopshaeria maculans, Fusarium spp., Mycosphaerella spp.) and necrotrophic (Stagnospora nodorum, Botrytis cinerea, Nectria haematoccoca), fungal plant pathogens have been released at a regular pace over the last five years and allowed a better understanding of the evolution of pathogenesis. These studies highlighted the value of comparative genomics in identifying important virulence genes with host-specific functions. Given that almost nothing is known about the molecular basis or evolution of obligate biotrophy in plant pathogens, the recent publication of papers describing the genomes of  three pathogens representing two independent evolutions of obligate biotrophy in the powdery mildews [Blumeria graminis, Spanu et al. (2010)] and the rusts [Puccinia graminis f. sp tritici & Melampsora larici-populina, Duplessis et al. (2011)] is a key step in our understanding of plant-pathogen interactions.

The genome analysis of the barley powdery mildew (Blumeria graminis) revealed a genome size expansion caused by transposon proliferation concomitant with a striking reduction in gene content, i.e., genes encoding sugar-cleaving enzymes, transporters and assimilatory enzymes for inorganic nitrate and sulfur (Spanu et al., 2010).  To identify the genetic idiosyncrasies underlying pathogenesis and biotrophic ability of rust pathogens, we have analyzed the genome sequences of the rust fungi M. larici-populina and P. graminis f. sp. tritici. in a joint collaboration between the JGI and the MIT Broad Institute (see my previous post). Our comparisons of M. larici-populina and P. graminis f. sp. tritici to other saprotrophic, pathogenic, and symbiotic basidiomycetes indicate that developmental innovations in the rust fungi lineages did not involve major changes in the ancestral repertoire of conserved proteins with known function. However, gene family expansions observed for oligopeptide transporters, auxin efflux carriers and signaling elements could reflect specific adaptations to this extreme parasitic lifestyle of these fungi. No massive gene loss was observed in M. larici-populina and P. graminis f. sp. tritici,

However, our comparisons of these three genomes with genomes from non-obligate fungal relatives have confirmed several startling commonalities amongst powdery mildews and  rusts:

  • Dramatic reduction of plant cell-wall degrading enzymes and other pathogenicity genes (i.e., evolution for “stealth”).
  • Loss of sulphite and nitrite assimilation genes (i.e., metabolic dependency on the host).
  • The massive proliferation of (retro)transposable elements which lead to significant enlargement of the overall genome size (the increase in genetic, heritable variability may confers an adaptive advantage to obligate life on a live host).
  • The deployment of large arrays of secreted effector proteins that act within and outside of host cells to counteract plant immunity and may facilitate other processes that are integral to survival within a hostile host.

Our study on the wheat and poplar rusts is the latest in a series of papers that investigates genomic attributes of biotrophy in obligate plant parasites. Another recent works focused on the oomycete Hyaloperonospora arabidopsidis, which causes downy mildew of Arabidopsis (Baxter et al., 2010). Interestingly, the genomic features discussed above were also identified  in this non-fungal lineage. Collectively, these draft genomes of these microbial parasites thus provided the first opportunity to gain insight into the genomic signatures and convergent evolution of obligate biotrophy.

A series of papers describing the transcriptome and secretome of the poplar rust will soon be published. Now, it remains to determine the role of hundreds of effector-like secreted proteins released in planta by M. larici-populina and P. graminis.


John McDowell’s comparison of our and Pietro’s papers appears as an invited Commentary in the PNAS early edition published online the week of May 16.

Duplessis et al. (2011) Obligate biotrophy features unraveled by the genomic analysis of rust fungi. Proc Ntl Acad Sci USA, Published online before print May 2, 2011, doi:10.1073/pnas.1019315108.

Spanu et al. (2010) Genome expansion and gene loss in powdery mildew fungi reveal tradeoffs in extreme parasitism. Science 330: 1543–1546.

Baxter et al. (2010) Signatures of adaptation to obligate biotrophy in the Hyaloperonospora arabidopsidis genome. Science 330: 1549-1551.

McDowell JM (2011) Genomes of obligate plant pathogens reveal adaptations for obligate parasitism. Proc Ntl Acad Sci USA, doi:10.1073/pnas.1105802108, published ahead of print May 16, 2011.

Photo: A poplar leaf infected by the leaf rust M. larici-populina (© F Martin).


The Poplar Microbiome project: Step 2. The Back-Story.

May 13th, 2011

Last week, we have witnessed something that many of us thought we’d never see – the completed publication of the genomes from the poplar leaf rust, Melampsora larici-populina, and the wheat rust, Puccinia graminis f. sp tritici. The story appeared on May 2 in PNAS Early View.

But what is the story behind a story that started 8 years ago? As someone that was there from the beginning, I think it is helpful to recap the highlights and lowlights that did not reach the journal article. It adds much more texture to the news release and gives a much better understanding of the process of getting from dream idea to final publication of a genome paper.

At the Tree Biotech meeting in Umeå (Sweden), in June 2003, I met Jerry Tuskan, leader of the Populus Genome Consortium, in a corner of the conference hall. I shared a few good beer moments with Jerry and we talked about sequencing microbes interacting with poplar – a dream idea for those working on non-model symbionts and pathogens associated to the only perennial species with its genome (freshly) sequenced.

As an unexpected follow-up, a month later, on July 24th, 2003, I heard from Jerry and Steve DiFazio about a possible U.S. Department of Energy (DOE) initiative to sequence micro-organisms associated with poplar, i.e. the Poplar microbiome. They have been floating this idea with various managers in DOE, and the initial response was positive and they received the go-ahead to submit a proposal. That was the good news. The bad news was that we’ve been encouraged to submit this proposal in time to be reviewed with the a batch of proposals submitted to the Office of Biological and Environmental Research (OBER) sequencing programme by July 25th!!! We thus wrote a quick proposal in less than two days.

In the proposal, we argued that the successful deployment of Populus-based systems for bioenergy tree plantations depends not only on the ability of Populus as a genome to address the demands placed upon it by environmental stresses and stimuli but also by the suite of micro-organisms that exist around and within the deployed genotypes. We stressed that this is this consortium of endophytes, symbionts and pathogens that in part determines the effectiveness of the deployed tree system. In order to gain a predictive understanding of the complex biological systems that evolve from plant-microbial interactions, we therefore proposed the draft sequencing of several known Populus microbial associates: the ectomycorrhizal symbiont Laccaria bicolor, the poplar rust pathogen Melampsora larici-populina (Mlp) and the arbuscular mycorrhizal symbiont Glomus intraradices. For each of these fungal associates there were large groups of researchers waiting and willing to contribute to the initial sequencing efforts as well as the follow-on functional genomics work. In addition to these individual species, we proposed random shotgun profiling of three anonymous communities: the rhizosphere community, stem endophytic symbionts and leaf endophytic symbionts. In addition to Jerry and Steve, the initial participants to this Populus Community Genome project included Peter Lammers, the late Gopi Podila, and George Newcombe.

On October 23rd, Jerry cheerfully announced that our proposal was met with partial, but welcomed, success!! We were successful in getting Glomus intraradices and Laccaria bicolor in the queue for complete genome sequencing in 2004 – a very naive prediction. The poplar leaf rust, Melampsora larici-populina, and the anonymous community portions of the proposal did not rank as high and therefore were excluded from the pipeline in 2004.

With the crunch time for the poplar and Laccaria genomes and all kinds of other things happening in 2004, we’re not able to get the Melampsora project together for the Community Sequencing Programme (CSP) 2004 deadline. We eventually builded a stronger case for Melampsora sequencing and submitted a letter of intent for the CSP January 27, 2005 deadline. The LoI was well received and we submitted the final proposal on February 25, 2005. On May 10, Jim Bristow from JGI let us know that our proposal (Project 05-SE-07) was approved for sequencing. To celebrate such occasion, a french scientist always goes for the best: Champagne flowed in the lab!!!

During Summer, Pascal Frey at INRA sprayed hundred of poplar leaves with Mlp spores to produce the millions Mlp germlings required for the large-scale prep of HMW DNA for genomic libraries. On early 2006, we shipped the first batch of DNA to JGI. This DNA was okay for small insert (3 and 8 kb) libraries but not suitable for constructing 40 kb libraries. Sebastien Duplessis worked hard on DNA preps and a new batch was forwarded to the JGI in June. Jan-Fang Cheng let us know that the Mlp project was doing pretty good in the Summer. They have successfully constructed an 8 Kb library with the new DNA that Sebastien shipped in June. It passed QC in July, and they have generated 125 Mb of Sanger sequences from this new library. They have also generated 135 Mb of sequences from the 3 Kb library. The fosmid library did also pass QC and they have generated about 18 Mb of sequences from that fosmid library. An initial sequence assembly was completed on early September 2007 and the first assembly was released for Christmas 2007. A great gift !!! … but the expected genome size (~100 Mbp) was much larger than expected and larger than the fungal genomes sequenced so far. The assembly was packed with repeated sequences. JGI therefore did further sequencing and the pre-release for the initial annotation of the 6.79x assembly of the Melampsora larici-populina v1.0 assembly was available on April 11, 2008. The genome browser came online with expert input from Andrea Aerts and Igor Grigoriev at JGI. Let me tell you that the first click on a newly released genome web portal is an unique excitement … and a major achievement thanks to the collective expertise of dozen of unknown JGI techies acting in backstage. The quality of the assembly and annotation was very good despite the high content in transposable elements (~50%) and the dikaryotic stage of the material sequenced.

The genome of the symbiont Laccaria bicolor (the Poplar Microbiome project: Step 1) was published in the March 2008 issue of Nature and during 2008 we were very busy publishing a series of companion papers.

We organized the 1st Annotation Workshop of the Melampsora Genome Consortium on August 20-21, 2008 at INRA-Nancy. We had had a very exciting first glimpse of the Mlp genome coding space and a lot of fun with our ‘cousins québecois’ from Richard Hamelin’s group. Annotation was greatly facilated by additional genomic resources, such as the EST produced by Erika Lindquist’s team at JGI, transcript profiling generated by Annegret Kohler and Emile Tisserant, and protein sequencing by Philippe Tanguay. During the following months, the various teams produced a large amount of data sets, including a very informative analysis of the expansion and contraction of multigene families by Yao-Cheng Lin and Pierre Rouzé, the analysis of CAZymes by Bernard Henrissat’s group, a database of repeated elements by Joelle Anselem, Claude Murat, and Hadi Quesneville, and an outstanding manual annotation of the secretome by Stéphane Hacquard, David Joly and Nicolas Feau. Of course, many other labs donated time and expertise. The value of these contributions cannot be understated, as literally hundreds of man months were committed to this project with uncertain guarantee of reward.

In October 2008, I contacted Christina Cuomo leading the wheat rust, Puccinia graminis f. sp tritici (Pgt) effort at the Broad Institute to explore the possibility to join forces in analyzing the two rust genomes. We agreed that working together to co-submit or on a joint paper would be helpful to both projects, and would provide an appealing package for a high-ranking journal. We aimed for a publication for the end of 2008. An unrealistic aggressive timeline. Time went on, annotation reports accumulated, but our INRA team efforts were mainly dedicated to the analysis of the Perigord Black Truffle genome released by the Genoscope on early March 2008. Dozen of drafts of the truffle paper with its humongous (125 pages !!!) supplemental materials were written between February and August 2009. During Fall, we wrote the three revisions requested by Nature (referees can sometime be very picky!) and the final manuscript version was submitted in January 2010. Writing a genome paper for Nature is like a ‘mission impossible’: squeezing the equivalent of a dozen of regular papers in a 1,800 words main text. We were thus delighted when the truffle paper made it in the March 2010 issue of Nature, but we were burned out.

During this period, Christina, Les Szabo and the Puccinia consortium have carried out numerous analyses on the multigene family evolution, the secretome, transcript profilings during in planta infection, and the repeated elements of Pgt. We thus started synchronizing the analyses across the two rust genomes and we drafted the final joint paper outline in the Marriott lounge during the 5th JGI Users meeting in Walnut Creek in March 2010. Time went on and the draft was still in limbo.

On early February 2010, Pietro Spanu and John McDowell leading the Blumeria graminis and Hyaloperonospora arabidopsidis genome consortia, respectively, contacted us suggesting that there were a compelling scientific case to submit the genome papers together if at all possible. Having a “raft” of papers on the genomes of obligate biotrophs (Hyaloperonospora, Rusts and Powdery Mildews) would likely make a greater impact. We were learning such a lot about the nature of obligate parasitism and biotrophy based on the collective analyses. The metabolic differences identified were really exciting, coupled with the novel genome organisation, and potentially large differences in secondary metabolic potential compared with non-obligate pathogens and saprotrophs. However, we were wondering about the timing of this. We approached Nature as joint grouping proposing three manuscripts on the comparative analysis of the genomes of the oomycete H. arabidopsidis (model pathogen of Arabidopsis, representative of downy mildews), the ascomycete fungus B. graminis (barley powdery mildew), and basidiomycetes P. graminis and M. larici-populina (wheat and poplar rusts, respectively). Our solicitation was declined about two weeks later. This was a bit of a setback. What next? We contacted Science to get their opinion on whether they might be interested in a series of manuscripts that describe remarkable genome-level convergences in the evolution of obligate biotrophy in three different plant pathogen lineages. Science encouraged us to send in the work. This was great news!!!

The Blumeria and Hyaloperonospora papers were submitted on early Summer, whereas we were still struggling in merging the Pgt and Mlp datasets. Sebastien and Christina did an outstanding job in pulling together all the available results during Summer. In September, the main text and the huge supplemental material section (133 pages) were almost done. However, there was much more to do than first thought, mostly around the significant reduction in word number requested by Science while trying to keep the story together. After another round of edits, we submitted the manuscript on October 6 when the two other papers were already in revision … a fatal delay. It was sent for in-depth evaluation on October 21! First step passed! We hoped that our paper would sail smoothly through review/revision and that we can all come out together! … never be too confident! … On November 18, we received the final decision from the Science editorial office – the kind that nobody likes to read: ‘We have now received the detailed reviews of your paper. Unfortunately they are not positive enough to support publication of the paper in Science. Although we recognize that you could likely address many of these specific criticisms in a revised manuscript, the overall nature of the reviews is such that the paper would not be able to compete for our limited space.’ This was a blow and I’m still disappointed (and baffled) that it was rejected. Reviewers were positive and none cite any scientific weakness in our paper. In my opinion, we were unfortunate being not able to meet the aggressive timeline. We were so sorry that it turned out this way. The two other papers were published in the December 10 issue of Science.

Despite our frustration, Sebastien quickly reformatted the paper for a submission to Nature on early December, but it was not considered for publication either. Yaah! Nit-picked to death!

On January 5, the paper was submitted to PNAS and less than three weeks after we received the editorial decision: ‘your work in principle suitable for publication in PNAS’. Thanks God! The suggested changes were mainly editorial, although additional analyses were carried out to definitely exclude genes that were probable transposable elements. Sebastien, Christina and Yao-Cheng were able to turn this around quickly. High quality, appropriately critical reviews led us to reshape to a stronger version of the work and then resubmit in March 25 and the final acceptance came a week later. Que du bonheur! The galley proofs were downloaded on April 25, but with a request to cut down this printed version from 8 to 6 pages. The paper had had little flesh remaining, but after this extra streamlining only the barebones remained! … Sebastien had had at least a few sleepless nights near the end.

The final version was published online on May 2, 2011 … 8 years flashed by between the initial dream idea and the final product. A long hard road for 6 thin pages, but it’s great to see it out and I anticipate that it will receive well-deserved emphasis. It was challenging to merge the efforts from the Melampsora and Puccinia consortia, but thanks to the 50 co-authors and the ‘unsung heroes’ of the technical staffs at JGI and Broad, the rust genome paper is a superb piece of work. It was a joy to work on a this joint project with dozen of friendly colleagues, but also to endure the ups and downs together.

In the next post, I will highlight the major features of the obligate biotroph genomes and one day, I may tell you the back-story of my ‘nightmare’ genome, Glomus intraradices, i.e. The Poplar Microbiome project: Step 3.

Duplessis et al. (2011) Obligate biotrophy features unraveled by the genomic analysis of rust fungi. Proc Natl Acad Sci USA, Published online before print May 2, 2011, doi: 10.1073/pnas.1019315108

Photos: Poplar leaf infected by M. larici-populina (F Martin); Urediniospores of M. larici-populina (B Pêtre/INRA); Wheat leaf infected by Puccinia graminis f. sp tritici (Marc Fouchard/INRA)

Sous l’effet de la rouille, l’immunité des plantes se grippe

May 5th, 2011

Lire le communiqué de presse INRA …

Rust Genome Released

May 5th, 2011

Read more …

Photo: Confocal microscopy image of Melampsora-infected leaves from poplar cv. Image by Stéphane Hacquard (INRA Nancy).