Biologists have vastly expanded understanding of the biological machinery controlling the intricate process by which plant roots burgeon from single cells into complex tissues. A Duke University-led team's discovery of new components of the root-development pathway in the mustard plant Arabidopsis thaliana represents both a scientific and technical achievement, the scientists said.

Scientifically, Arabidopsis is a basic biological model for all flowering plants, so the finding offers insights into a critical function of all such plants, including crop plants, the researchers said. Further, since the root is a useful model for tissue development in general, basic findings regarding the root-development pathway could offer insights into how complex tissues are generated from immature cells, called stem cells.

Technically, the genomic analytical method they used also will offer biologists a highly useful approach to discovering the components of complex biological pathways governing development, the researchers said. Their statistical "meta-analysis" technique involved using computational methods to integrate data from multiple genetic analyses using several DNA microarrays -- popularly known as "gene chips." Each of these chips contained some 24,000 genes representing nearly the entire genome of the Arabidopsis plant.

The team's findings appeared May 2, 2006, in the online edition of the journal Public Library of Science Biology and will be published in the journal's May 2006 edition.

Philip Benfey, professor and chair of Duke's Department of Biology and a member of the Duke Institute for Genome Sciences & Policy, is senior first author of the report. Joint first authors are Mitchell Levesque and Teva Vernoux, who performed the work in Benfey's laboratory.

The research was sponsored by the National Institutes of Health and the National Science Foundation.

Before the latest work, Benfey and his colleagues had discovered a gene called SHORT-ROOT (SHR), which produces a protein that appears to be a central regulatory molecule in the root-development pathway. SHR was so named because gene mutations that cause malfunction result in stunted roots with incompletely differentiated tissues. The researchers had also discovered a second gene, SCARECROW, that appeared to be controlled by SHR.

The researchers believed the SHR protein to be a "transcription factor," a master switch that controls activation, or "transcription," of a multitude of target genes. In turn, those target genes might control still other biological regulatory pathways which form a biological network that governs plant root development.

"Before this paper, we had two big questions," Benfey said. "One was whether SHORT-ROOT was actually a transcription factor that activated or repressed genes. The other was the identity of the target genes beyond SCARECROW, which we believed was a target but had no direct evidence for. We basically had no clue what was involved in the root-development process beyond those genes."

To search for genes that were SHR targets, the researchers used the technique of microarray analysis. Basically, such analysis involves isolating the messenger RNA produced by all of the genes in the Arabidopsis genome. Using the microarrays, the researchers could determine the level of messenger RNA produced by each gene, which reflects its activity. Thus, they could pinpoint among thousands of genes only those activated or repressed under different conditions, such as mutation of SHR.

However, in the search for SHR targets, the scientists went a step further in their microarray analysis. They modulated the SHR pathway in several distinct ways by using chemicals or mutation to switch it on or block it under different conditions. By using statistical methods to compare the subtle differences in activity of the multitude of genes under those different conditions, they could identify genes that were likely targets of SHR.

That "meta-analysis" of the results from the microarrays revealed eight genes that appeared likely to be direct targets, as well as numerous genes that appeared to be indirectly affected by SHR activity, the scientists said.

"The next thing we did was to determine whether these target genes were, indeed, expressed in domains that were consistent with SHORT-ROOT expression," Benfey said. "After all, we did not preselect for genes that were expressed in the root."

That analysis, he said, showed that many of the genes were expressed in the same region of the plant as SHR. Also, further studies showed evidence that the SHR protein directly bound some of the target gene promoters. Even though not all of the genes showed such binding, many factors compromise such analyses, he said, such that the lack of evidence does not rule out any of the eight as SHR targets.

Also, the researchers' analysis of the expression of the SHR targets revealed evidence for a role for SHR where it had not been demonstrated before -- in the central vascular tissue, or stele, of the plant root.

"So, this meta-analysis enabled us to identify a major new function for a gene that we had been studying for quite a while," Benfey said. "We had seen effects on the stele of mutating SHORT-ROOT in previous studies, but they were so subtle that we couldn't say for sure whether they were significant."

The researchers also analyzed SHR's effects on indirect target genes revealing that many of the secondary targets are turned off.

"What we found striking was that over three-quarters of the indirect targets we identified were repressed and only one-quarter upregulated," Benfey said. "That's not a common finding with transcription factors. So, it leads us to believe that of the direct targets we identified, most are probably working more to repress their target genes than to activate them."

Further studies of the treasure trove of new SHR gene targets will aim at exploring the details of their functions using biochemical and genetic techniques, Benfey said. Also, the new findings will enable the researchers to begin to understand the complex network regulated by SHR.

"We think that SHORT-ROOT is part of a cascade of transcription factors, and that there are many intermediate steps before the end-stage differentiation of cells in tissues," he said. "We're now actively exploring those intermediate steps."

More broadly, the technique of meta-analysis -- used currently in epidemiological studies -- will find wider use in studying the development of organisms, Benfey said.

"To our knowledge, this is the first time this statistical technique has been used in a developmental context," he said. "It is much more powerful than simply comparing the results from different microarrays. And there is nothing in this technique that is specific to plants, meaning that it can be applied to analyze any organism."

Other co-authors on the paper are Hongchang Cui, Jean Wang, Keji Nakajima and Noritaka Matsumoto, who worked in the Benfey laboratory; Wolfgang Busch and Jan Lohmann of the Max Planck Institute for Developmental Biology; and Hala Hassan and Ben Scheres of Utrecht University.