De Novo Characterization Of Blueberry Fruit Transcriptome Using Solexa Sequencing To Discover Putative Genes Related To Anthocyanins

Anthocyanin antioxidant can provide effective protection by neutralizing free radicals which are unstable molecules that are linked to the development of a number of degenerative diseases and conditions, such as diabetes, cancer and ulcers. Blueberry(Vaccinium spp.) is an important small fruit crop rich in antioxidants. The content of the anthocyanin in blueberry is considered to be of primary importance and the biosynthesis of anthocyanins is considered to be one of the most important secondary metabolic pathways during the fruit development. However, tissue-specific transcriptome and genomic data in public databases are not sufficient for an understanding of the molecular mechanisms associated with antioxidants, especially the biosynthesis of anthocyanins.Aimed at broadening the scope of pigment diversity in blueberry breeding, we have carried out the first global analysis of the blueberry transcriptome during berry development using the Illumina/Solexa RNA-Seq method. To experimentally confirm that the unigenes obtained from sequencing and computational analysis were indeed expressed, qRT-PCR was performed on five key enzymes genes associated with anthocyanin biosynthetic genes. Two full-length cDNA sequences of anthocyanin pathway key enzymes genes were obtained from skins using RT-PCR and RACE. This study generated a large set of blueberry transcript sequences that can be used to discover mechanisms of tissue specific functions and secondary metabolism, which will lay the molecular biology foundation for further research on the molecular mechanism of anthocyanin biosynthesis and new varieties breeding of blueberry.Main results are as follows:1. Two pools of mRNA samples, one from blueberry skin and one from pulp, were used to build libraries for high-throughput parallel sequencing using Illumina sequencing technology on the Genome Analyzer IIX platform. The64,312,746raw reads are available in the Sequence Read Archive at the NCBI under Accession number SRA046311.2. De novo assemblies yielded34,464All-unigenes from both samples with an average length of735bp. The length greater than1,000nt was7,812(22.66%of all) and no gaps was32,032(92.3%).3. To identify the putative functions of the All-unigene sequences, we performed Blast X alignments (E-value<1.0e-5) against the NCBI non-redundant databases NR, Swiss-Prot, GO, COG and KEGG databases. This approach returned25,376All-unigenes (73.6%of all) with matches to one or more of the database that were below the cut-off value. Based on their sequence homology, a total of8,498All-unigenes (24.6%) were categorized into three main categories and found that a total of9,175(26.6%) sequences had a COG classification.4. To identify the biological pathways that are active in blueberry, we mapped the25,376annotated sequences to the reference canonical pathways in KEGG. In total,12,547sequences were assigned to119KEGG pathways. The pathways most represented in the unique sequences were metabolic pathways (2,982), biosynthesis of secondary metabolites (1,568).19subclasses including1,236transcripts were mapped to antioxidant related pathways.5. All-unigenes are firstly aligned by Blast X (E-value<1.0e-5) to protein databases to generate CDS25,436(5’-3’). The length greater than1,000nt was5,229(20.6%of all).6. Differences in tag frequencies in the skin and pulp libraries were used to estimate differences in gene expression levels between the two libraries.9,703transcripts (FDR≤0.001and|log2Ratio|≥1) between the two libraries are significantly differences;76DEGs were present at higher levels in the skin library with differences greater than15-fold. The highest DEG was the GDSL-motif lipase/hydrolase family protein gene which was expressed in the skin library at～20-fold higher than in pulp; A total of2,171DEGs were categorized into three main categories and a total of116pathways were affected by3,176DEGs, among them,28pathways with Q-value≤0.05were identified as being significantly enriched. The pathways most represented in the unique sequences was620sequence mapped to the biosynthesis of secondary metabolites.7.862proteins were identified as putative transcription factors (TFs) or regulators belonging to22different families. TFs play an essential role in regulating the overall activity of flavonoid biosynthesis, including MYB (117), WD40(91), WRKY (63), AP2/ERF(53) and bHLH (5).8.92DEGs were selected may encode14key enzyme in flavonoids biosynthetic pathway. To experimentally confirm that the All-unigene obtained from sequencing and computational analysis were indeed expressed, five transcripts related to anthocyanin biosynthetic genes PAL (phenylalanine ammonia-lyase), CHS (chalcone synthase), F3H (flavonoid3-hydroxylase), DFR (dihydroflavonol4-reductase) and ANS (anthocyanidin synthase) were used to examine the expression pattern during blueberry fruit development. Most of the selected genes were those that were highly expressed in the skin library. The anthocyanin pathway genes were expressed in flowers at the very beginning of the blueberry fruits development; the lowest expression levels were in the green fruits, and then expression levels increased from pink fruits to blue fruits and finally reached the highest levels in the skin of the blue fruits9. Anthocyanin content was investigated in different tissues and organs during blueberry devlepment. The results showed that the expression levels of PAL, CHS, F3H, DFR and ANS increased in concurrence with the accumulation of anthocyanins excluding in flowers, and much more highly expressed in the skins of blue fruits compared with other organs or tissues of the fruit10. A full-length cDNA sequence of anthocyanin synthase (ANS) gene was obtained from skins of the blueberry using RT-PCR and RACE, named VcANS (GenBank accession No. JN654701). Sequence analyses indicated that VcANS was1,424bp in full length, containing a5’-UTR of93bp, a3’-UTR of248bp, and an ORF of1,083bp encoding a protein of360amino acids, which possessed the conserved structural domain of the2-oxoglutarate and Fe2+-dependent oxygenase superfamily. Sequence alignment and phylogenetic tree analyses showed that VcANS had the closest phylogenetic relationship with the Ericaceae.