By Mark D. Uehling
May 7, 2002 | It is not eaten or smoked, neither milled nor harvested. It is not sufficiently bothersome to have risen to the stature of a weed. Yet the humble Arabidopsis thaliana is the botanical equivalent of a fruit fly --- a model organism growing in laboratories around the world.
Now its genome has been deposited on no fewer than three new powerful microarrays.
“This is the first time that we could put the entire genome of any species on an array,” says Elizabeth Kerr, director of marketing for gene expression at Affymetrix Inc., referring to the company’s latest custom-designed GeneChip for agronomic giant Monsanto Co.
Meanwhile, Syngenta, a Swiss competitor, has talked up its chip at scientific meetings, and researchers in academia have just received the first test copies of its Arabidopsis chip. “It is still not exactly clear how big the market will be,” Kerr says. “The genome is still ramping up in agriculture.”
Molecular biology, however, has already ramped up for Arabidopsis. The mustard plant is beloved by scientists because it has both a small size (standing 6 inches tall) and compact genome (117 million base pairs). Hardy, Arabidopsis is able to thrive under a variety of conditions; and it conveniently fertilizes itself, progressing from sprout to seed in six to eight weeks. It’s also easy to work with in the greenhouse.
So traces of any part of an Arabidopsis plant, at any stage of its life cycle, grown under, say, bright sunlight, wet soil, and heavy herbicide doses (or untold other combinations of growing conditions), can be applied to an Affymetrix chip for analysis. Botanists will be able to chart which of the plant’s genes are expressed under any set of conditions. “It's a good check on what you’re doing,” says Chris Town, visiting investigator at The Institute for Genomic Research (TIGR).
Town says the chips will show if groups of genes are working together. “You can say, 'Aha, those genes may be involved in the same process,’ ” Town says. “You get an implied functional association with other genes.”
At Monsanto, the commercial appeal of the Arabidopsis technology is clear.
“It is one tool to understand a given genome in more detail and under a wider array of circumstances," says Monsanto spokesman Mark Buckingham. “We hope the chip contributes to faster discovery of new traits.” More precisely, genetically targeted insights into how Arabidopsis works at the cellular level, he says, will give the company greater confidence that additional research on a new seed or chemical will pay off.
Says Buckingham: “The accuracy and robust nature of the information provided by this chip are well worth having.” Indeed, it won’t come cheap, as estimated costs are significant: $300,000 for a custom chip, $150,000 for the equipment to read the chips, and custom chips themselves are purchased in lots of 90 with a list price of $500.
Academic researchers will have a less expensive option in the Affymetrix catalog: packages of just six chips. This is yet a third custom Arabidopsis chip, commissioned by Town, and intended for the entire academic community. Chris Somerville, director of plant biology for the Carnegie Institution at Stanford University, first canvassed his international colleagues to determine demand and persuaded Affymetrix to offer the chip in smaller lots.
Somerville and his wife conceived the central role of Arabidopsis as a model organism for molecular biology back in 1978, before the rest of the botanical community had recognized how useful the plant might become. In 1989, Somerville, Elliot Meyerowitz of the California Institute of Technology and their colleagues had already dreamed up a bid to sequence every gene in Arabidopsis.
“The only thing that surprised me about the project was that we actually got it done,” Somerville says with a laugh, referring to the international sequencing work that provided the sequences underlying what Affymetrix is putting on its chips.
For Somerville, the custom chips for Monsanto and Syngenta are only part of what will drive Arabidopsis research forward. “The industry-sponsored custom chips will not have much impact on academic research,” says Somerville, noting the academic chip. But all of the Arabidopsis chips will help with the next big goal for the field: a 10-year, $500 million international effort to determine the functions of every Arabidopsis gene, the first of its kind in the plant kingdom.
That effort, coordinated by the National Science Foundation and known as The 2010 Project, could eventually affect much of what Americans eat. This is because the genes and proteins in Arabidopsis will have easily identified counterparts in major crops, such as rice.
As Somerville explained in a scientific paper published in 1999: “We may expect to have extensive databases of quantitative information about the degree to which each gene responds to pathogens, pests, drought, cold, salt, photoperiod, and other environmental variation. We will soon know which genes respond to the phytohormones, growth regulators, safeners, herbicides, and related agrichemicals.”
Large groups of genes common to Arabidopsis and major U.S. crops could then be crossbred from one species to another in the lab. Environmentalists, already wary of agricultural genetic manipulation, are likely to view such experiments with suspicion. However, all of the challenges of figuring out the interplay of genes in Arabidopsis will be applicable to human medicine and help scientists work out the genetic machinery of more complex species.
For all of the potential, Somerville says some of the computational tools are still iffy. One problem: the torrents of data about to pour off the Arabidopsis chips at Monsanto and elsewhere. With tiny wells corresponding to 24,000 of the plant’s genes, each chip kicks out more than 500,000 data points per run. “We are going to be seriously awash in large datasets,” Somerville says. “These are much larger datasets than most biologists are used to.”
He says researchers at Stanford have collected more than 365 million rows of data from microarrays into a large database, perhaps 4 percent of which relate to Arabidopsis. Even that university’s homegrown database for managing all of the information, while formidable, is not for computer neophytes. Says Somerville of the impending mountain of data: “It will take an awful lot of software to make sense of it.”
For TIGR investigator Town, the tools to work with Arabidopsis data are not so deficient. He does acknowledge a shortage of software in one area: monitoring the evanescent patterns of groups of genes flickering on and off in various parts of the plant, under various ecological conditions. “The tools that don't exist are how to make the associations,” Town says. “Genes are going up and down all over the place --- what does it mean?”
One answer may be to store some of the chip data in a public location and let colleagues, graduate students and others inspect it. Town reports that the Arabidopsis community is in the early stages of developing public repositories of raw information from the Affymetrix chips any scientist could mine to find overlooked nuggets.
The logic to such sifting and sorting, Town says, is simple: “If I approach this in a different way, I may find patterns the other guy missed.”