Gene Expression Profiling

Changes in gene expression underlie many biological phenomena. The use of DNA microarrays provides insight into tissue- and developmental-specific expression of genes and the response of gene expression to environmental stimuli. Oligonucleotide- and cDNA-based microarrays are being generated on the BioRobotics Ltd, MicroGrid system by using unique cDNA isolates identified in the M. truncatula EST programs. These glass slide arrays are hybridized with fluorescently labeled cDNA probes and analyzed with the GSI Lumonics ScanArray 4000 two-color microarray analysis system.9-11 Arrays will be used to profile transcript levels in 1) elicited cell cultures, 2) developmental comparisons within a particular tissue type (e.g., stems for the study of lignin biosynthesis), and 3) host-symbiont/host-pathogen interactions. Many of the gene features (or probes) on the arrays represent unknown, previously undescribed genes. The correlation of transcript profiles, probe clustering, and expression patterns in a particular tissue or developmental stage may help to assign biological relevance to many of the unknown genes.

Qiagen Operon, in collaboration with the Noble Foundation, Chris Town (The Institute for Genomic Research) and Kate VandenBosch (The University of Minnesota) have developed a commercially available Array Ready Genome Oligonucleotide Set for M. truncatula. This set of 16,000 bioinformatically optimized oligonucleotides are being used as probes in our microarray analysis.

To supplement the expression analyses data generated by microarrays, we have added an "open system" serial analysis of gene expression (SAGE) to our set of transcript profiling tools. Such open system approaches allow for the identification and analysis of genes not previously characterized. With "closed systems" such as microarrays, analysis is limited to only those species previously identified and assigned to an array. SAGE analyses have already been used to study gene expression in plant systems.12-13 Among the high-throughput, comprehensive technological methods used to analyze transcript expression levels, array-based hybridization and SAGE are currently the most common approaches. In a recent comparison of SAGE and array-based technologies, the two methods correlated quite well in both absolute expression analyses and comparative analyses during differentiation.14 The correlation was better for genes with higher expression levels and greater changes in expression.

Plants are continuously exposed to biotic and abiotic (notably harmful UV radiation) stress elements and have evolved mechanisms to reduce their deleterious effects. The UV-inducibility of a number of DNA repair15 and metabolic pathways has previously been determined.16-20 To date, no global assessment of the effects of UV irradiation on transcript abundance in plants has been performed. As an example of the application of SAGE to monitor changes in gene expression, we have profiled modulation in transcript accumulation through analysis of libraries generated from mRNAs isolated from UV-, g-treated, and non-treated Arabidopsis plantlets. Ten days post germination, whole plantlets were exposed to either UV-B light at a dosage rate of 20 J M-2S-1 and harvested after a 8000 J M-2 exposure, or were irradiated with g-rays at a dose rate of 181cGy min-1 and harvested after 217Gy of g-rays irradiation. More than 73,000 tags were identified between the control and treatment libraries and were analyzed using SAGE 2000 software (http://www.sagenet.org/sage_protocol.htm).

More than 300 transcripts, identified in treatment comparisons, had a five-fold or greater difference in abundance. Greater than 36,000 of the more than 73,000 SAGE tags were unique. The largest differences in tag abundance were between a 16-fold increase and a 13-fold decrease for transcripts in a comparison between untreated and UV-treatment SAGE tag libraries. A comparison of the g-ray and untreated SAGE tag libraries reveals that the greatest fluctuation in transcript levels are an approximately ten-fold increase or decrease in relative tag abundance. A number of the SAGE tags had identity and mapped to "putative proteins" corresponding to predicted open reading frames (ORFs) in the Arabidopsis genome. The identification of SAGE tags that correspond to putative proteins confirms that the predicted ORFs are indeed expressed.

The sequence of the yeast genome has been completed for some time, and analyses of yeast SAGE databases have led to the discovery of new, previously un-annotated ORFs.21 In these experiments, SAGE tags were identified that did not correspond to known or predicted genes in the yeast genome. Likewise, our comparison of the UV, g-ray and untreated SAGE tag libraries has led to the discovery of new ORFs in the Arabidopsis genome. SAGE tag counts in the non-treated control library reflected transcript quantitation results similar to those in previous studies reported in the literature that utilized other profiling technologies. SAGE analysis of the UV- and gamma-treated library demonstrated an increase in stress-related transcripts. Genes for many of these transcripts are in the phytochrome signaling and flavonoid biosynthetic pathways, both of which were previously described as UV-inducible. 22 These studies illustrate the power of SAGE technology as a tool for both transcript profiling and gene discovery and its use in examining global changes in plant gene expression patterns. As additional plant genomes such as M. truncatula are sequenced and plant-specific SAGE databases become publicly available, the use of SAGE in understanding fundamental changes in gene expression should gain broad appeal in the plant research community.

References