Genomics in azalea: defence against broad mite (Polyphagotarsonemus latus) infection as a case study

Plant material of different tissues (flowers, leaves and buds) from R. simsii genotypes was harvested after specific treatments. This resulted in 16 tissue samples for comprehensive transcriptome assembly. Six of these samples were related to plant defence: leaf samples of a mock treatment were compared to foliar application of methyl jasmonate (MeJa) or P. latus infestation on both a sensitive (‘Nordlicht’) and a mite tolerant (‘Elien’) genotype. The other samples were related to flower colour and branching. Poly-A selected RNA seq libraries were prepared and sequencing was performed using Illumina HiSeq PE-100. In total, 711M reads were used for de novo assembly according to the Orthology Guided Assembly routine described by Ruttink et al. (2013) using the A. thaliana proteome as a reference. The azalea transcriptome contains a total of 32367 transcript fragments representing the orthologs of 14860 unique A. thaliana genes; 8012 of these coding sequences presumably represent full-length proteins. For differential gene expression (DEG) analysis, reads were mapped to the R. simsii transcriptome using the RNA-seq analysis utility of CLCbio GW. In ‘Nordlicht’, more than 1400 genes were differentially expressed in response to either MeJa treatment or P. latus infestation; both treatments shared over 500 DEGs when compared to the mock treatment. For the mite tolerant genotype ‘Elien’ more than 2400 DEGs were reported upon mite infestation, but only 1342 DEGs were reported as a result of MeJa treatment. Compared to the mock, both treatments had 851 DEGs in common. Gene Ontology (GO) enrichment was analysed using the BINGO plugin for Cytoscape (version 3.2.1). Jasmonic acid (JA) involvement in the response to the treatment was clear in 3 out of 4 mock-treatment comparisons. Only when comparing the mock to P. latus infestation in ‘Elien’, no GO terms indicating the direct involvement of JA were enriched. These results provided evidence for the involvement of the JA-pathway in response to P. latus infestation. On the other hand, also genes involved in the salicylic acid (SA) pathway were differentially expressed. Hence, the DEG results were validated in gene expression analysis using genes for the biosynthesis pathways of JA (LOX, AOS, AOC, OPR3 and JMT) and SA (ICS and PAL) and the PPO gene involved in oxidative stress response. Multiple experiments using either MeJa treatment or P. latus infestation on 6 different genotypes were used for this purpose. The results of this validation in relation to the transcriptome analysis will be discussed.