Limits...
A novel workflow correlating RNA-seq data to Phythophthora infestans resistance levels in wild Solanum species and potato clones.

Frades I, Abreha KB, Proux-Wéra E, Lankinen Å, Andreasson E, Alexandersson E - Front Plant Sci (2015)

Bottom Line: More transcript families were expanded in the resistant clones and species and the enriched functions of these were associated to expected gene ontology (GO) terms for resistance mechanisms such as hypersensitive response, host programmed cell death and endopeptidase activity.However, no differences in numbers of susceptibility (S-)gene homologs were seen between species and clones.In addition, we identified P. infestans transcripts including effectors in the early stages of P. infestans-Solanum interactions.

View Article: PubMed Central - PubMed

Affiliation: Department of Plant Protection Biology, Swedish University of Agricultural Sciences Alnarp, Sweden.

ABSTRACT
Comparative transcriptomics between species can provide valuable understanding of plant-pathogen interactions. Here, we focus on wild Solanum species and potato clones with varying degree of resistance against Phytophthora infestans, which causes the devastating late blight disease in potato. The transcriptomes of three wild Solanum species native to Southern Sweden, Solanum dulcamara, Solanum nigrum, and Solanum physalifolium were compared to three potato clones, Desiree (cv.), SW93-1015 and Sarpo Mira. Desiree and S. physalifolium are susceptible to P. infestans whereas the other four have different degrees of resistance. By building transcript families based on de novo assembled RNA-seq across species and clones and correlating these to resistance phenotypes, we created a novel workflow to identify families with expanded or depleted number of transcripts in relation to the P. infestans resistance level. Analysis was facilitated by inferring functional annotations based on the family structure and semantic clustering. More transcript families were expanded in the resistant clones and species and the enriched functions of these were associated to expected gene ontology (GO) terms for resistance mechanisms such as hypersensitive response, host programmed cell death and endopeptidase activity. However, a number of unexpected functions and transcripts were also identified, for example transmembrane transport and protein acylation expanded in the susceptible group and a cluster of Zinc knuckle family proteins expanded in the resistant group. Over 400 expressed putative resistance (R-)genes were identified and resistant clones Sarpo Mira and SW93-1015 had ca 25% more expressed putative R-genes than susceptible cultivar Desiree. However, no differences in numbers of susceptibility (S-)gene homologs were seen between species and clones. In addition, we identified P. infestans transcripts including effectors in the early stages of P. infestans-Solanum interactions.

No MeSH data available.


Related in: MedlinePlus

Clades of GO enriched terms for Biological Process (Siddappa et al., 2014) of the expanded and depleted OrthoMCL clusters determined quantitatively according to the resistance gradient (GO enriched terms based on the qualitative analysis is available in Figure S2). R and S denotes GO terms enriched in the resistant or susceptible groups, respectively. Broad, unspecific GO terms have been removed.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4585127&req=5

Figure 4: Clades of GO enriched terms for Biological Process (Siddappa et al., 2014) of the expanded and depleted OrthoMCL clusters determined quantitatively according to the resistance gradient (GO enriched terms based on the qualitative analysis is available in Figure S2). R and S denotes GO terms enriched in the resistant or susceptible groups, respectively. Broad, unspecific GO terms have been removed.

Mentions: Using the predicted proteomes of the 11 species and clones in OrthoMCL, 38,890 clusters were generated, in which leaf transcripts of Desiree, Sarpo Mira, SW93-1015, S. nigrum, S. dulcamara, and S. physalifolium were represented in 27,979 (Table S2). In the analyses of the transcriptomes only the longest representative transcript was used to avoid allelic variants caused by different polyploidy, mis-assemblies and sequencing errors (D'Agostino et al., 2013). The OrthoMCL cluster structure was analyzed by PCA (Figure 3) and GO term enrichment analysis (Figure 4; Figure S2). Displaying the OrthoMCL clusters constructed by transcripts of Desiree, Sarpo Mira, SW93-1015, S. nigrum, S. dulcamara, and S. physalifolium and genes S. tuberosum (PGSC or ITAG gene models), S. lycopersicum and A. thaliana, there are clear differences between transcript and gene data (principle component 1, PC1) as can be expected since not all genes are present in the leaf transcriptomes studied here. A division of S. tuberosum clones and wild Solanum follows PC2. Furthermore, OrthoMCL clusters expanded or depleted as for numbers of transcripts were identified and either linked to a gradient depending on the degree of resistance to P. infestans (quantitative analysis) or grouped as resistant vs. susceptible (qualitative analysis). These analyses found 143 and 1291 OrthoMCL clusters significantly (p < 0.05) expanded quantitatively and qualitatively, respectively, and 249 and 525 significantly depleted quantitatively and qualitatively, respectively (Table S1). In the qualitative analysis as we contrast two susceptible with four resistant species and clones and therefore we have unequal sequence amounts for the resistant and susceptible entities, 158.8 and 77.7 Mb, respectively, which could influence results. However, this unequal relation is not seen in the quantitative analysis and we find a substantial overlap of 86% in the clusters identified by both analyses showing coherence between both approaches.


A novel workflow correlating RNA-seq data to Phythophthora infestans resistance levels in wild Solanum species and potato clones.

Frades I, Abreha KB, Proux-Wéra E, Lankinen Å, Andreasson E, Alexandersson E - Front Plant Sci (2015)

Clades of GO enriched terms for Biological Process (Siddappa et al., 2014) of the expanded and depleted OrthoMCL clusters determined quantitatively according to the resistance gradient (GO enriched terms based on the qualitative analysis is available in Figure S2). R and S denotes GO terms enriched in the resistant or susceptible groups, respectively. Broad, unspecific GO terms have been removed.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4585127&req=5

Figure 4: Clades of GO enriched terms for Biological Process (Siddappa et al., 2014) of the expanded and depleted OrthoMCL clusters determined quantitatively according to the resistance gradient (GO enriched terms based on the qualitative analysis is available in Figure S2). R and S denotes GO terms enriched in the resistant or susceptible groups, respectively. Broad, unspecific GO terms have been removed.
Mentions: Using the predicted proteomes of the 11 species and clones in OrthoMCL, 38,890 clusters were generated, in which leaf transcripts of Desiree, Sarpo Mira, SW93-1015, S. nigrum, S. dulcamara, and S. physalifolium were represented in 27,979 (Table S2). In the analyses of the transcriptomes only the longest representative transcript was used to avoid allelic variants caused by different polyploidy, mis-assemblies and sequencing errors (D'Agostino et al., 2013). The OrthoMCL cluster structure was analyzed by PCA (Figure 3) and GO term enrichment analysis (Figure 4; Figure S2). Displaying the OrthoMCL clusters constructed by transcripts of Desiree, Sarpo Mira, SW93-1015, S. nigrum, S. dulcamara, and S. physalifolium and genes S. tuberosum (PGSC or ITAG gene models), S. lycopersicum and A. thaliana, there are clear differences between transcript and gene data (principle component 1, PC1) as can be expected since not all genes are present in the leaf transcriptomes studied here. A division of S. tuberosum clones and wild Solanum follows PC2. Furthermore, OrthoMCL clusters expanded or depleted as for numbers of transcripts were identified and either linked to a gradient depending on the degree of resistance to P. infestans (quantitative analysis) or grouped as resistant vs. susceptible (qualitative analysis). These analyses found 143 and 1291 OrthoMCL clusters significantly (p < 0.05) expanded quantitatively and qualitatively, respectively, and 249 and 525 significantly depleted quantitatively and qualitatively, respectively (Table S1). In the qualitative analysis as we contrast two susceptible with four resistant species and clones and therefore we have unequal sequence amounts for the resistant and susceptible entities, 158.8 and 77.7 Mb, respectively, which could influence results. However, this unequal relation is not seen in the quantitative analysis and we find a substantial overlap of 86% in the clusters identified by both analyses showing coherence between both approaches.

Bottom Line: More transcript families were expanded in the resistant clones and species and the enriched functions of these were associated to expected gene ontology (GO) terms for resistance mechanisms such as hypersensitive response, host programmed cell death and endopeptidase activity.However, no differences in numbers of susceptibility (S-)gene homologs were seen between species and clones.In addition, we identified P. infestans transcripts including effectors in the early stages of P. infestans-Solanum interactions.

View Article: PubMed Central - PubMed

Affiliation: Department of Plant Protection Biology, Swedish University of Agricultural Sciences Alnarp, Sweden.

ABSTRACT
Comparative transcriptomics between species can provide valuable understanding of plant-pathogen interactions. Here, we focus on wild Solanum species and potato clones with varying degree of resistance against Phytophthora infestans, which causes the devastating late blight disease in potato. The transcriptomes of three wild Solanum species native to Southern Sweden, Solanum dulcamara, Solanum nigrum, and Solanum physalifolium were compared to three potato clones, Desiree (cv.), SW93-1015 and Sarpo Mira. Desiree and S. physalifolium are susceptible to P. infestans whereas the other four have different degrees of resistance. By building transcript families based on de novo assembled RNA-seq across species and clones and correlating these to resistance phenotypes, we created a novel workflow to identify families with expanded or depleted number of transcripts in relation to the P. infestans resistance level. Analysis was facilitated by inferring functional annotations based on the family structure and semantic clustering. More transcript families were expanded in the resistant clones and species and the enriched functions of these were associated to expected gene ontology (GO) terms for resistance mechanisms such as hypersensitive response, host programmed cell death and endopeptidase activity. However, a number of unexpected functions and transcripts were also identified, for example transmembrane transport and protein acylation expanded in the susceptible group and a cluster of Zinc knuckle family proteins expanded in the resistant group. Over 400 expressed putative resistance (R-)genes were identified and resistant clones Sarpo Mira and SW93-1015 had ca 25% more expressed putative R-genes than susceptible cultivar Desiree. However, no differences in numbers of susceptibility (S-)gene homologs were seen between species and clones. In addition, we identified P. infestans transcripts including effectors in the early stages of P. infestans-Solanum interactions.

No MeSH data available.


Related in: MedlinePlus