Characterization of transcriptional complexity during berry development in Vitis vinifera using RNA-Seq

Abstract

The development of massively parallel sequencing technologies enables the sequencing of total cDNA (RNA-Seq) to derive accurate measure of individual gene expression, differential splicing activity, and to discover novel regions of transcription, dramatically changing the way that the functional complexity of transcriptomes can be studied. Here we report on the first use of RNA-Seq to gain insight into the wide range of transcriptional responses that are associated with berry development in Vitis vinifera 'Corvina'. More than 59 million sequence reads, 36 to 44 bp in length, were generated from three developmental stages: post setting, véraison, and ripening. The sequence reads were aligned onto the 8.4-fold draft sequence of the Pinot Noir 40024 genome and then analyzed to measure gene expression levels, to detect alternative splicing events, and expressed single nucleotide polymorphisms. We detected 17,324 genes expressed during berry development, 6,695 of which were expressed in a stage-specific manner, suggesting differences in expression for genes in numerous functional categories and a significant transcriptional complexity. This exhaustive overview of gene expression dynamics demonstrates the utility of RNA-Seq for identifying single nucleotide polymorphisms and splice variants and for describing how plant transcriptomes change during development.

Figure 1.

Candidate gene undergoing alternative splicing. A, Summary of reads falling onto constitutive and alternative junctions of the GSVIVT00023307001 gene in berry developmental stages. B, GSVIVT00023307001 gene model is represented by four exons (red tracks) and three introns (black lines). The V. vinifera ESTs (green tracks below) confirm the existence of alternative spliced mRNAs lacking the third exon. C, RT-PCR confirmation of GSVIVT00023307001 constitutive and alternative splice events in berry post fruit-set (PFS), véraison (V), and ripening (R) developmental stages. NC is the negative control. The predicted fragment sizes are 295 bp for the constitutive splice and 149 bp for the alternative splice.

Figure 2.

Comparative displays from Maqview showing MAQ-aligned reads. The top track shows the reference sequence and the next shows the MAQ-called consensus derived from the aligned individual reads depicted below. Differences from the reference base are shown in red.

Figure 3.

Venn diagram showing the genes expressed in each of the three stages of berry development.

Figure 4.

Categories distribution in the eight expression clusters. Clusters were obtained by the k-means method on the gene expression profiles of the 6,695 modulated genes. Histogram representation of the functional categories distribution is expressed as percentage of the amount of genes belonging to the cluster. Gene coding for unknown products were not considered in the analysis. PFS, Post fruit set; R, ripening; V, véraison.