Transcriptome analysis of woodland strawberry (Fragaria vesca) response to the infection by Strawberry vein banding virus (SVBV)

We chose strawberry plants infected with SVBV as a model to investigate host transcriptome responses. The inoculated strawberry leaves developed typical vein banding symptoms at 40 days post inoculation (Fig. 1a), while the mock-inoculated strawberry leaves did not show any symptoms (Fig. 1b). Southern blot analysis validated that symptomatic leaves were infected by SVBV (Fig. 1c). These leaves were collected for RNA extraction followed by Illumina sequencing. Two cDNA libraries, T1 (mock-inoculated control) and T2 (SVBV inoculated leaf material) were sequenced resulting in 76,157,614 and 72,701,140 sequence reads that encompassed 7.69 Gb and 7.34 Gb of sequence data, respectively. The reads’ length from both libraries ranged from 50 to 101 base pairs (bp) (Table 1). The tag density was sufficient for the quantitative analysis of gene expression. Sequence data from T1 and T2 libraries were combined to obtain 36,850 unigenes. The GC content of sequences from these was ~50 % with the phred quality score (Q30) above 90 %, indicating that the accuracy and quality of the sequencing data were sufficient for further analysis. The length distributions of total unigenes had similar patterns between the libraries, suggesting that there was no bias in the library construction (Additional file 2: Figure S1).

Further analysis revealed that 517 unigenes showed differential expression as compared to mock-inoculated control. Among these 517 differentially expressed genes (DEGs), 280 were up regulated, and 237 were down regulated. BLAST results were obtained for these DEGs against five protein databases, namely the non-redundant (nr, NCBI) protein database, Swiss-Prot database, Clusters of Orthologous Groups of proteins (COG, NCBI) database, the Gene Ontology (GO) database and Kyoto Encyclopedia of Genes and Genomes (KEGG) database. Among them, 498 DEGs had nr annotation, 388 DEGs had Swiss-Prot annotation, 195 DEGs had COG annotations, 425 DEGs had GO annotations and 79 DEGs had KEGG functional annotations.

Sequences with BLAST hits were further classified with COG, GO and KEGG pathway analysis. COG analysis indicated that 195 DEGs could be grouped into 21 clusters, and the DEGs mainly distributed in the cluster of ‘posttranslational modification, protein turnover and chaperones, replication, recombination, repair, transcription, and signal transduction mechanisms’. The cluster of ‘general function prediction’ accounted for 12 % of the total annotated DEGs, and their functions are still unknown (Fig. 2). The 425 DEGs annotated to the GO database were distributed in 47 functional types, including “growth and development”, “cell death”, “metabolism”, and “transcriptional regulation” (Fig. 3). Particularly, among these GO functional classes, a class of ‘cell killing’ has the largest differences between the SVBV-infected and virus-free samples, suggesting that the corresponding genes might be related to the pathogenic mechanism of SVBV. In conclusion, no matter which database we used, most of DEGs were putatively involved in cellular metabolism, signaling transduction, and plant-pathogen interactions. These genes appear to be relevant to the interactions between strawberry and SVBV.

DEGs of strawberry infected with SVBV were annotated to 20 KEGG pathways (Fig. 4). The top three KEGG pathways containing the largest numbers of DEGs were DNA replication, plant-pathogen interactions and homologous recombination (Table 2). In addition, pigment metabolism and photosynthesis pathways also have a certain degree of enrichment. In these KEGG pathways, we found that the expressions of 14 genes were significantly altered (Fig. 5).

Glutamic acid decarboxylase (GAD) gene was significantly up-regulated by SVBV infection. GAD has been reported to be involved in “taurine and hypotaurine metabolism” of plant pigment metabolism and is the key enzyme in γ-aminobutyric acid (GABA) synthesis. In response to the biotic stress, plants rapidly increase GAD expression and accumulate quantity of GABA [12, 13], leading to reduce other free amino acid content and inhibiting the growth of stems [14]. The carotenoid biosynthesis pathway that mainly related to the photosynthesis also has a certain degree of enrichment, which might be highly associated with strawberry regulatory network in response to SVBV infection. The pathway might be highly associated with strawberry regulatory network in response to SVBV infection. Abscisic acid-8'-hydroxylase was related to “carotenoid biosynthesis” upon the pathogens and environmental stresses, the content of ABA caused by abscisic acid-8'-hydroxylase rise so rapidly, ABA depresses entire plant growth and vitro organs by functioning as a strong plant growth inhibitor [15].

Plant WRKY transcription factors and respiratory burst oxidase homolog protein were involved in plant-pathogen interaction, the up-regulation of factors is the embodiment of the plant immune responses [16]. Anthocyanin reductase and dihydroflavonol 4-reductase were both related to “flavonoid biosynthesis”. Anthocyanins, the plant secondary metabolites, play a key role in the process of fruit mature of strawberry [17].

In order to validate our DGE data, ten DEGs with annotations were selected for qRT-PCR analysis. Among them, eight out of ten DGEs exhibited a consistent expression pattern between RNA-Seq and qRT-PCR (Fig. 6). For example, both qRT-PCR and DGE analyses showed that genes encoding dihydroflavonol-4 reductase, glutamate decarboxylase, anthocyanidin reductase and abscisic acid hydroxylase were strongly up regulated in SVBV-infected strawberry compared to uninfected control. Analogously, the reduced expression of genes encoding the gibberellin 20-oxidase, heat shock protein 83 (Hsp83), eukaryotic translation initiation factor-5 (ETIF5) and auxin-induced protein 15 A revealed by RNA-Seq analysis were also validated by qRT-PCR. In addition, we analyzed expression of two genes encoding calmodulin and abscisic-acid receptor PYL4 by qRT-PCR. According to our data, neither of these two genes exhibited detectable differential expression during SVBV infection in the DEG dataset. Once again, qRT-PCR analysis validated these results. Thus, we have validated the conclusions of RNA-Seq analysis by independent qRT-PCR approach for all three major categories of genes: those up regulated, down regulated and unchanged in response to SVBV infection of strawberry.