Final Report Summary - RUST-SAFE (Effector discovery and validation of Puccinia striiformis sp. f. tritici, a wheat pathogen)
Wheat resistance research has a profound social and economic impact because the crop forms such an important part of the diet of the world population: 40% used wheat as a staple and it provides about 20% of the calories eaten globally. In 2009 new and highly aggressive strains of yellow rust devastated crucial wheat-producing areas in China, northern and eastern Africa, western and central Asia and the Middle East. Epidemics of this rust have been spreading since 2000 in United States and Australia, and there every prospect that such outbreaks will occur in Europe. Despite the importance of wheat, little is known about the molecular mechanisms underlying wheat-rust interaction, which is the determinant for plant immunity and our ability to breed crop plants for long-lasting resistance. This project was based on existing knowledge of the biology of Puccinia striiformis f. sp. tritici (PST), yellow rust, its interactions with wheat, and it utilized existing sources resistance, to improve our understanding of the role of effectors in determining effective disease resistance.
To enhance the finding of new PST virulence effectors we compared five isolates from two different origins: isolates PST-87/7 and PST-08/21 were obtained from fields of the United Kingdom in 1987 and 2008 respectively, and isolates PST-21, PST-43 and PST-130 were obtained in the USA in 2007. PST isolates are usually identified through the inoculation of a differential set of wheat cultivars with known resistance genes. All of these isolates were scored for virulence in their country of origin. The two UK isolates had a common virulence profile of ten Yr genes, but differed in their virulence to wheat varieties Robigus (YrRob) and Solstice (YrSol). All the isolates were sequenced and assembled into contings. From RNA of wheat-infected tissue we obtained the transcriptome, and selecting those sequences with signal peptide we got the secretome. All the secretome was filtered based in different features to identify the putative effectors, for that reason a bioinformatics pipeline was executed. The putative effectors were grouped in tribes following sequence homology. The pipeline gave a score to each tribe based in probable effector protein features such as defined avirulence activity, nuclear localization signal, flanking intergenic regions, repeat containing proteins and small cysteine rich proteins. The filtering diminish the number of total tribes from 1037 to 969, where all of contained at least one PST secreted protein. From infected wheat leaves haustoria was isolated and RNAseq analysis of UK PST-08/21 was performed, identifying after that haustoria- enriched transcripts. Thus, haustorial expression of each individual candidate effector was analysed. After the whole process of filtering 5 effector candidates remained, all with high haustorial expression and highly polymorphic between the UK races 87/7 and 08/21. They were subsequently chosen to be validated in vivo.
The second objective of the proposal was to establish a functional validation method for rust effectors in wheat. The original protocol proposed was the wheat protoplast isolation and effector transfection. The protocol was developed but the system was not reproducible enough to validate the putative effectors in a high throughput fashion.
Alternatively, we set up a system that delivers by the type III secretion system the effector candidates directly into wheat cells by expressing them in Pseudomonas fluorescens (Pfo). We used two different genes to establish the system, AvrRpt2 and AvrSr22, which are able to induce HR in two different wheat varieties: Avocet ‘S’ or a variety carrying the R gene Sr22. They were cloned in two different vector backbones with either AvrRps4 or AvrRpm1 signal peptides. From infected Avocet S tissue with the race PST-08/21 the cDNA of putative effectors were cloned using the pCR8 TOPO® TA Gateway® Cloning System. All the putative effectors, once cloned, were sequenced looking for different allelic variants, which accounted for more than 90 constructs. One allelic variant of each candidate were transferred to the vector pAvrRpm1 by Gateway® recombination. These plasmids were after transferred to Pfo. The transformation was carried out by conjugation, using as a donor the E. Coli stripe carrying the putative effector, Pfo as a receptor and an E. Coli strain carrying the helper plasmid pRK2013. All the strains were tested for HR, inoculating them in a differential set of wheat varieties for pathotyping PST isolates. This set of differentials is composed of 20 varieties for which most of them PST 08/21 is avirulent, according to its virulence profile (Yr1, Yr2, Yr3, Yr4, Yr6, Yr7, Yr9, Yr17, Yr27, Yr32, YrRob, YrSo).
We validated all the 5 effectors and found 2 possible matches between putative effector-Yr gene. For the two of them, positive HR was shown in two out of three replicates, at least.
In the most advanced one, effector 7, we are working now in coexpression experiments in Nicotiana benthamiana, where we have found different protein sublocalization when different domains of the effector are expressed separately. Wheat plants were transformed recently to overexpress this effector. A second method of validation for the match between the putative effector 7 and the Yr gene is being carried out, using biolistics and GUS expression over Yr and non Yr wheat lines.
To enhance the finding of new PST virulence effectors we compared five isolates from two different origins: isolates PST-87/7 and PST-08/21 were obtained from fields of the United Kingdom in 1987 and 2008 respectively, and isolates PST-21, PST-43 and PST-130 were obtained in the USA in 2007. PST isolates are usually identified through the inoculation of a differential set of wheat cultivars with known resistance genes. All of these isolates were scored for virulence in their country of origin. The two UK isolates had a common virulence profile of ten Yr genes, but differed in their virulence to wheat varieties Robigus (YrRob) and Solstice (YrSol). All the isolates were sequenced and assembled into contings. From RNA of wheat-infected tissue we obtained the transcriptome, and selecting those sequences with signal peptide we got the secretome. All the secretome was filtered based in different features to identify the putative effectors, for that reason a bioinformatics pipeline was executed. The putative effectors were grouped in tribes following sequence homology. The pipeline gave a score to each tribe based in probable effector protein features such as defined avirulence activity, nuclear localization signal, flanking intergenic regions, repeat containing proteins and small cysteine rich proteins. The filtering diminish the number of total tribes from 1037 to 969, where all of contained at least one PST secreted protein. From infected wheat leaves haustoria was isolated and RNAseq analysis of UK PST-08/21 was performed, identifying after that haustoria- enriched transcripts. Thus, haustorial expression of each individual candidate effector was analysed. After the whole process of filtering 5 effector candidates remained, all with high haustorial expression and highly polymorphic between the UK races 87/7 and 08/21. They were subsequently chosen to be validated in vivo.
The second objective of the proposal was to establish a functional validation method for rust effectors in wheat. The original protocol proposed was the wheat protoplast isolation and effector transfection. The protocol was developed but the system was not reproducible enough to validate the putative effectors in a high throughput fashion.
Alternatively, we set up a system that delivers by the type III secretion system the effector candidates directly into wheat cells by expressing them in Pseudomonas fluorescens (Pfo). We used two different genes to establish the system, AvrRpt2 and AvrSr22, which are able to induce HR in two different wheat varieties: Avocet ‘S’ or a variety carrying the R gene Sr22. They were cloned in two different vector backbones with either AvrRps4 or AvrRpm1 signal peptides. From infected Avocet S tissue with the race PST-08/21 the cDNA of putative effectors were cloned using the pCR8 TOPO® TA Gateway® Cloning System. All the putative effectors, once cloned, were sequenced looking for different allelic variants, which accounted for more than 90 constructs. One allelic variant of each candidate were transferred to the vector pAvrRpm1 by Gateway® recombination. These plasmids were after transferred to Pfo. The transformation was carried out by conjugation, using as a donor the E. Coli stripe carrying the putative effector, Pfo as a receptor and an E. Coli strain carrying the helper plasmid pRK2013. All the strains were tested for HR, inoculating them in a differential set of wheat varieties for pathotyping PST isolates. This set of differentials is composed of 20 varieties for which most of them PST 08/21 is avirulent, according to its virulence profile (Yr1, Yr2, Yr3, Yr4, Yr6, Yr7, Yr9, Yr17, Yr27, Yr32, YrRob, YrSo).
We validated all the 5 effectors and found 2 possible matches between putative effector-Yr gene. For the two of them, positive HR was shown in two out of three replicates, at least.
In the most advanced one, effector 7, we are working now in coexpression experiments in Nicotiana benthamiana, where we have found different protein sublocalization when different domains of the effector are expressed separately. Wheat plants were transformed recently to overexpress this effector. A second method of validation for the match between the putative effector 7 and the Yr gene is being carried out, using biolistics and GUS expression over Yr and non Yr wheat lines.