SuperSAGE array: the direct use of 26-base-pair transcript tags in oligonucleotide arrays.

We developed a new platform for genome-wide gene expression analysis in any eukaryotic organism, which we called SuperSAGE array. The SuperSAGE array is a microarray onto which 26-bp oligonucleotides corresponding to SuperSAGE tag sequences are directly synthesized. A SuperSAGE array combines the advantages of the highly quantitative SuperSAGE expression analysis with the high-throughput microarray technology. We demonstrated highly reproducible gene expression profiling by the SuperSAGE array for 1,000 genes (tags) in rice. We also applied this technology to the detailed study of expressed genes identified by SuperSAGE in Nicotiana benthamiana, an organism for which sufficient genome sequence information is not available. We propose that the SuperSAGE array system represents a new paradigm for microarray construction, as no genomic or cDNA sequence data are required for its preparation. The analysis of complete transcriptomes by expression microarrays 1 and serial analysis of gene expression (SAGE) 2 are now indispensable tools for various biological and biomedical studies. Microarrays allow high-throughput large-scale gene expression analysis in many samples at a time. For some model organisms, cDNA or oligonucleotide arrays covering entire gene sets or even the whole transcriptome are now commercially available. But for the majority of organisms researchers have to design their own arrays spotted with DNA obtained from cDNA libraries, which is a time- and cost-intensive procedure. Furthermore, as a ‘closedarchitecture’ technology, microarrays allow detection of expression changes of spotted genes only and therefore exclude the discovery of new genes. In contrast, SAGE is an ‘open architecture’ system that enables both the discovery of new expressed genes and the accurate quantification of the resulting transcripts 2 . But lengthy experimental steps and the fact that it is not suitable for the analysis of multiple samples, are major drawbacks of SAGE. Additionally, tags of only 14 bp in the original SAGE 2 , and 21 bp in the LongSAGE procedure 3 , are too short for completely unambiguous tag-to-gene annotations, hampering the application of SAGE to ‘non-model’ organisms for which genome sequence information is not available. Recently, we developed SuperSAGE, an improved version of SAGE, allowing the isolation of 26-bp tags using the type III restriction endonuclease EcoP15I 4 . Our SuperSAGE protocol greatly improves the efficiency of tag-to-gene annotation. This high gene annotation power of SuperSAGE already had allowed the simultaneous and quantitative gene expression profiling of both host cells and their eukaryotic pathogens (parasites) as well as transcript profiling in ‘‘non-model’’ organisms without existing databases of genome sequence information 4,5 . We hypothesized that the high information content of 26-bp SuperSAGE tags should make them suitable hybridization probes in microarray analyses. We named this experimental system SuperSAGE array (Fig. 1). In the present study, we demonstrate highly reproducible expression analysis by SuperSAGE array in both model and non-model organisms, and show that this SuperSAGE-array platform indeed combines the advantages of both SAGE and microarray techniques. RESULTS