Open Access Open Badges Research article

Transcriptomic analysis of the interaction between Helianthus annuus and its obligate parasite Plasmopara halstedii shows single nucleotide polymorphisms in CRN sequences

Falah As-sadi1, Sébastien Carrere1, Quentin Gascuel1, Thibaut Hourlier1, David Rengel1, Marie-Christine Le Paslier2, Amandine Bordat1, Marie-Claude Boniface1, Dominique Brunel2, Jérôme Gouzy1, Laurence Godiard1* and Patrick Vincourt1*

Author Affiliations

1 INRA, Laboratoire des Interactions Plantes-Microorganismes (LIPM), UMR441, F-31326 Castanet-Tolosan, France. CNRS, Laboratoire des Interactions Plantes-Microorganismes (LIPM), UMR2594, F-31326 Castanet-Tolosan, France

2 Etude de Polymorphisme des Génomes Végétaux (INRA), CEA, Institut de Génomique, Centre National de Génotypage, 2 rue Gaston Crémieux CP 5721, F-91057 Évry Cedex, France

For all author emails, please log on.

BMC Genomics 2011, 12:498  doi:10.1186/1471-2164-12-498

Published: 11 October 2011



Downy mildew in sunflowers (Helianthus annuus L.) is caused by the oomycete Plasmopara halstedii (Farl.) Berlese et de Toni. Despite efforts by the international community to breed mildew-resistant varieties, downy mildew remains a major threat to the sunflower crop. Very few genomic, genetic and molecular resources are currently available to study this pathogen. Using a 454 sequencing method, expressed sequence tags (EST) during the interaction between H. annuus and P. halstedii have been generated and a search was performed for sites in putative effectors to show polymorphisms between the different races of P. halstedii.


A 454 pyrosequencing run of two infected sunflower samples (inbred lines XRQ and PSC8 infected with race 710 of P. halstedii, which exhibit incompatible and compatible interactions, respectively) generated 113,720 and 172,107 useable reads. From these reads, 44,948 contigs and singletons have been produced. A bioinformatic portal, HP, was specifically created for in-depth analysis of these clusters. Using in silico filtering, 405 clusters were defined as being specific to oomycetes, and 172 were defined as non-specific oomycete clusters. A subset of these two categories was checked using PCR amplification, and 86% of the tested clusters were validated. Twenty putative RXLR and CRN effectors were detected using PSI-BLAST. Using corresponding sequences from four races (100, 304, 703 and 710), 22 SNPs were detected, providing new information on pathogen polymorphisms.


This study identified a large number of genes that are expressed during H. annuus/P. halstedii compatible or incompatible interactions. It also reveals, for the first time, that an infection mechanism exists in P. halstedii similar to that in other oomycetes associated with the presence of putative RXLR and CRN effectors. SNPs discovered in CRN effector sequences were used to determine the genetic distances between the four races of P. halstedii. This work therefore provides valuable tools for further discoveries regarding the H. annuus/P. halstedii pathosystem.