Robust analysis of 5'-transcript ends (5'-RATE): a novel technique for transcriptome analysis and genome annotation

Abstract

Complicated cloning procedures and the high cost of sequencing have inhibited the wide application of serial analysis of gene expression and massively parallel signature sequencing for genome-wide transcriptome profiling of complex genomes. Here we describe a new method called robust analysis of 5'-transcript ends (5'-RATE) for rapid and cost-effective isolation of long 5' transcript ends (approximately 80 bp). It consists of three major steps including 5'-oligocapping of mRNA, NlaIII tag and ditag generation, and pyrosequencing of NlaIII tags. Complicated steps, such as purification and cloning of concatemers, colony picking and plasmid DNA purification, are eliminated and the conventional Sanger sequencing method is replaced with the newly developed pyrosequencing method. Sequence analysis of a maize 5'-RATE library revealed complex alternative transcription start sites and a 5' poly(A) tail in maize transcripts. Our results demonstrate that 5'-RATE is a simple, fast and cost-effective method for transcriptome analysis and genome annotation of complex genomes.

Figure 1

Experimental procedure for the 5′-RATE. The mRNA from maize is treated with bacterial alkaline phosphatase and acid pyrophosphatase to modify the cap structure at the 5′ regions. The 5′ decapped mRNA is divided into pools 1 and 2 and ligated with RNA oligos (A and B). The cDNA is synthesized and tags are released from the 5′ regions of cDNA using the NlaIII enzyme. Tags from the two pools are self-ligated to generate ditag cassettes. Ditags are amplified using PCR and linkers are removed by XhoI digestion. Ditag fragments are sequenced using the 454 pyrosequencer at DOE Joint Genome Institute (JGI), CA.

Figure 2

Size distribution of the 5′-RATE tags.

Figure 3

Sequence alignment of the jasmonate-induced gene (ID: Q564C9) with its alternative TSSs and 5′ poly(A) tail. The nucleotides in underlined are non-template sequences.

Figure 4

FL-cDNA sequences with 5′ poly(A) tail from plants and animals. The 5′ poly(A) sequences are shown in boldface letters and translation initiation codon (ATG) is shown in capital letters.