Review for "Wheat blast disease caused by Pyricularia graminis-tritici sp. nov."

Completed on 1 May 2016 by Sophien Kamoun . Sourced from

Login to endorse this review.

Comments to author

Author response is in blue.

It's great that the authors have posted this article. It's obviously very timely given the current outbreak in Bangladesh and the Open Wheat Blast initiative.

These are my thoughts after a first read of the paper.

I’m not sure about naming a plant pathogen species after a specific host plant when the pathogen has additional hosts or is known to have experienced host jumps in its recent evolutionary history. There is an unfortunate tendency to do this in plant pathology.

A more neutral name, say Pyricularia infestans (or whatever), is less likely to convey the impression that these strains can only infect wheat. What if in the future this taxon causes an epidemic on oats? The same confusion that reigns today about rice vs wheat blast would apply with such a host-specific name. What about proper quarantine of other host plants that may carry this pathogen outside Brazil and now Bangladesh? Would the authorities be more likely to ignore other potential hosts when the pathogen has such a defined name?

I agree that a distinct name is needed for the wheat blast strains to clearly convey the message to the community and the authorities that this pathogen is distinct from the rice blast strain. There is no question that this has important implications for quarantine and disease management. But here two names are proposed. It is important to ensure that it is justified to divide the wheat isolates into two taxa. In the recently posted report by Croll and McDonald (Github 2016, attached), all wheat isolates are clustered in one well supported clade unlike Figure 1 of Castroagudin et al. Will the two taxa proposed here for the wheat blast isolates remain valid when genome-wide analyses are performed? Could the less-defined position of the “PoT” strains in Castroagudin et al. reflect genetic exchange between the main clusters of blast fungi?

Unfortunately, Linnaean binoms are outdated. They’re becoming more and more useless and inappropriate. It’s great that the fungal community has moved to the one species-one name concept but there is still work to do. For one thing, the current scheme fails to provide stability - a key, some would say the main reason we name things. See the twitter exchange linked below for a discussion on the topic. Perhaps, virologists have figured out a solution?

It would be unfortunate if the blast community gets bogged down in this naming issue despite what seems like a consensus that the wheat blast fungus is a distinct OTU. There are clearly more important topics to discuss and debate in relation to this invasive disease.

Croll and McDonald. Github 2016.


Open Wheat Blast

I recognize that naming Pyricularia has been a controversial topic in the Pyricularia community (mainly those working on rice blast), which has shown some historical
resistance to splitting the P. oryzae infecting rice from the P. oryzae infecting Triticum, even though these are two clearly distinct pathogens. I have even heard other investigators state that "it was abundantly clear" they were a single species. Perhaps if P. oryzae was not the official name associated with the wheat blast pathogen, appropriate quarantine laws specifically targeting the wheat blast pathogen would have blocked its introduction into Bangladesh. But the past is the past. We propose that now is the time to move forward.

Dr. Kamoun presented some suggestions for the naming system for the wheat blast disease complex. I would like to point out that there are two other species recently
described by Sylvia Kaublauf et al. 2014 (from Pedro Crous' group, available at, named Pyricularia pennisetigena and P. zingibericola, associated with local grasses and wheat, that can also cause wheat blast in Brazil. As confusing as it sounds, we are, in fact, dealing with a species complex that causes wheat blast. Before the publications of Couch and Kohn and Klaubauf and Crous, these species were all called Pyricularia grisea. And with 0.01% of the genome data

Based on our sample of around 500 Pyricularia strains sampled in sympatry with wheat fields in Brazil, we have information on the frequency distribution of all four Brazilian species described in our paper (P. graminis tritici sp. nov., P. oryzae pathotype Triticum, P. pennisetigena and P. zingibericola), as well as local P. oryzae pathotype Oryza and P. grisea. Pyricularia graminis tritici sp. nov. and P. oryzae pathotype Triticum predominated in our collections, but P. graminis tritici sp. nov. is distributed mainly in the warmer region of the Brazilian Cerrado while P. oryzae pathotype Triticum is associated mainly with the cooler regions of Southern Brazil.

Because of our comprehensive sampling strategy, that encompassed seven states from Central to Southern Brazil, we do not consider this distribution to be an artifact. It appears to reflect local adaptation of the corresponding wheat blast populations. For more details, please check the slides we presented at the recent Wheat Blast meeting in Florianapolis Brazil at

It is important to recognize that wheat blast has been tightly connected with the name given to the rice blast pathogen for 14 years already. As more and better genome sequences become available, we think it is possible that P. oryzae pathotype Triticum could eventually be merged into the P. graminis-tritici clade. But because knowledge of wheat blast has emerged in a stepwise process, we have published our findings in the same stepwise manner. The bioRxiv paper represents
the most recent step in our wheat blast research efforts. The naming system proposed in this paper highlights the existence of P. graminis-tritici, but does not conflict with a possible future merger of P. oryzae pathotype Triticum into P. graminis-tritici. We hope that more genome datasets will help us to fill the current gaps in the phylogenetic relationships of the wheat blast species complex.

When we chose the name "graminis-tritici", it was to acknowledge that this pathogenic fungus can infect other Poaceae in addition to Triticum (wheat). There was not a strict association with Triticum in our collections, though of course we sampled mainly from wheat fields across Brazil. Our important innovation was to sample also from the grasses growing nearby those wheat fields and to characterize these strains.

I would like to comment specifically on Dr. Kamoun's last suggestion: "and then work with the community to develop a new naming system using the best possible data set." At this point, without diminishing the importance of other data sets available in other labs, I would like to keep our data set restricted to the hundreds of sympatric samples we collected across Brazil, which we believe is the center of diversity of the wheat blast pathogen and its likely center of origin. We also would like to give the proper credit to the group of mostly Brazilian scientists who have already invested 5-11 years of effort into this research. We think that the most ethical decision is to let this group of dedicated scientists decide how to name this important disease complex.

Paulo Ceresini
Corresponding author