Gene Expression Profiling of Sweetpotato Storage Roots Through de novo Root Transcriptome and Custom Microarray Analysis

We developed a custom 4x44K sweetpotato oligonucleotide microarray (SPOArrav1) based on the Agilent platform. The array included probes for 14K transcripts selected from a de novo  sweetpotato transcriptome assembly that was built integrating next generation sequencing (NGS) data from root, leaf and stem tissues and expressed sequence tags (ESTs) from GenBank.  SPOArrav1 was used to analyze gene expression profile of both storage and fibrous roots, each with six biological replicates. In total, 1,111 unique transcripts involved in regulatory role, transport, molecular signaling and cell wall organogenesis were differentially expressed in early developing storage roots vs. fibrous root, in which 783 and 328 transcripts were upregulated and downregulated, respectively, in storage roots. Expression profiling of selected genes through quantitative RT-PCR validated the microarray results. Comparison of the expression of 54 selected genes in storage root showed a significantly high positive correlation of microarray and digital gene expression analysis for 32 genes in the sweetpotato root transcriptome. The novel finding of this study was the association of seven BEL1-like genes and six calcium binding proteins with the sweetpotato storage root development. In addition, 68 transcripts with no functional annotation also showed enhanced expression in young storage roots from both microarray and digital expression analyses; these may represent putative genes specific to the onset of storage root formation in sweetpotato. This research demonstrated, for the first time, the utility of global transcriptome analysis through combination of NGS and custom oligonucleotide microarray in the genome-wide gene expression profiling of sweetpotato root development.