Courtesy of Lynx Technologies
By conscious design, plant genomics initiatives have devoted initial resources to new technology development. Part of that money went to developing functional genomics approaches, and part to new sequencing technologies. Lynx's Massively Parallel Signature Sequencing (MPSS) approach, shown here, can decipher millions of sequence fragments – each represented by one bead in this image – simultaneously.
Plants are quieter than people. They might flaunt pollen at a passing bee or exude fragrance, but except for flashy bloomers, most do little to attract attention. Perhaps it is fitting, then, that plant genome initiatives have kept a low profile relative to the human genome project. Yet the lack of fanfare does not stem – so to speak – from a lack of merit: These initiatives are transforming how biologists do plant science and expanding what they hope to accomplish.
Key to that transformation, as in...
Several common threads wove programs together. The first was agreement on model systems. While researchers continued to work on everything from the potato to the snapdragon, they chose rice and the flowering mustard plant, Arabidopsis thaliana, for more intense scrutiny. (See sidebar, "A New Model") These two plants were anointed the Rosetta stones of the plant world, the templates that researchers could apply to other members of the monocot and dicot plant families, repectively.
"The ultimate end point of this will be the clickable plant," Jane Silverthorne, director of NSF's Plant Genome Research Program, says of efforts to find the function of every Arabidopsis gene by 2010. "It will be a database you can go to find out about all the genes that give rise to a given organ, where they're expressed, how much they're expressed, how they interact with each other."
Technology is another common thread. While sequencing efforts zoomed ahead, a large portion of the early money was invested in developing not only high-throughput methods, but also services to help biologists do functional genomics with the completed sequences. GARNet, for example, will take a scientist's RNA sample and hybridize it on an Affymetrix GeneChip®containing about 23,000 matches for every known gene in the Arabidopsis genome.
The US-based Arabidopsis Gene Expression Database gives scientists access to a new gene expression map that shows when and where more than 22,000 genes turn on and off in cells from the root of Arabidopsis.1 Philip Benfey at Duke University in Durham, NC, and his colleagues used a combination of cutting-edge technologies, including Affymetrix microarrays and a fluorescence-activated cell sorter, to develop the map.
Biologists had to develop new methodologies, says Benfey; studying genes one at a time would never reveal how they worked as a system in complex organisms. "The mind can't comprehend and make use of 23,000 data points," he says. "Ultimately, the goal is to turn on the lights and see the whole thing."
For University of Arizona, Tucson, biologist David Galbraith, who developed tools used in the map, it is an example of how the technology and biology have become tightly integrated. "Really, it's the putting together of all these technologies that enables the work to be done," says Galbraith, who provides his own microarrays to plant scientists at relatively low cost.
MORE NEW TECHNOLOGIES
Courtesy of The Samuel Roberts Noble Foundation (top right). Courtesy of Kenneth Bimbaum (left)
Medicago truncatula (top right)
A) Arabidopsis seed-lings expressing green fluorescent protein (GFP) in specific types, such as the one shown marking the lateral root cap cells, are subjected to an enzymatic treatment that breaks the plants apart into their constituent cells. B) A subset of cells (proto-plasts) expresses GFP, marking the targeted cell types (compare B with inset, which is the same field of cells viewed under white light). These marked cells are then isolated using a fluorescence activated cell sorter and expression profiled using microarrays. (Reprinted with permission, Science, 303:291, Jan. 16, 2004.) (left)
Plant scientists can also obtain gene expression information from the Massively Parallel Signature Sequencing (MPSS) project. Lynx Therapeutics in Hayward, Calif., provides Blake C. Meyers of the University of Delaware in Newark with raw MPSS data for Arabidopsis, which he puts into usable form on a Web site; a rice project is underway as well.
Another new technology, Targeted Induced Local Lesions in Genomes (TILLING), supports reverse genetics in plants.2 This high-throughout technique creates point mutations through chemical mutagenesis and then screens for lesions. Steven Henikoff of the Fred Hutchinson Cancer Research Center in Seattle offers Arabidopsis TILLING from a Web site. It delivers an average of 10 mutations, says Henikoff, who is developing TILLING for maize and rice. Other scientists are applying it to zebrafish, he says. "TILLING is general. You can do it on any organism."
In addition, several researchers developed techniques to make complex plant genomes more manageable. Teams led by Lance Palmer at Cold Spring Harbor Laboratory, NY, and Catherine A. Whitelaw at the Institute for Genomic Research in Rockville, Md., used C0t selection and methylation filtration (which select for low-copy and undermethylated sequences, respectively) to weed out large segments of repetitive sequences in maize, thereby focusing their efforts on likely coding sequences.34
COOPERATION AND DISCORD
The third thread weaves cooperation and bioinformatics through the programs. The NSF's research program has appropriated $375 million and is currently supporting 120 strategically chosen projects. Much of the money has gone to what Silverthorne calls virtual centers: collaborations by laboratories rather than single investigators. As all groups must make their results immediately available to the scientific community, every project has a bioinformatics component, and some were created expressly to develop online resources such as the Plant Genome Database and Gramene, which is a comparative mapping resource for grains.
Many European programs also emphasize collaboration and shared databases. In Great Britain, scientists associated with GARNet suggest that such collaboration resulted partly because they don't have nearly as much money to spend. "People have to devise cunning community efforts to allow maximum value to be milked from every research pound and research dollar," says Malcolm M. Campbell at Oxford University in the UK, where he researches gene expression in trees.
Yet despite official policies endorsing collaborations, US researchers complain that European colleagues occasionally refuse to share sequence information. Some European initiatives are partnerships formed by public and private sponsors, responds Freitag, so researchers are more likely to face restrictions on releasing what could be proprietary information. These collaborations also have a greater responsibility to work on local agricultural interests, he adds, describing barley as a model plant for Germany.
Sharing information has not been a problem for the IRGSP, though, says Takuji Sasaki, director of the Genome Research Department at the National Institute of Agrobiological Sciences in Japan. The project completed a draft of the rice genome in 2002 and plans to publish the complete sequence by the end of 2004.5 Information is centralized in Japan's Rice Genome Program, which publishes the Integrated Rice Genome Explorer, a database known as INE, the Japanese word for rice. The Rice Genome Automated Annotation System is another open resource.
Sasaki expresses concern, however, that sharing could become difficult if scientists try to transfer gene knockout plants for research into functional genomics. "The transfer of these materials, even as seeds, is very difficult practically and politically," he suggests, citing quarantine systems in rice-producing countries and in the United States, which, he says, has become stricter since Sept. 11, 2001. Some countries are concerned about protecting intellectual property rights and genetic resources, he adds. Sasaki suggests that for functional genomics, small-scale collaborations between institutes or individual researchers may be easier to establish than open consortia.
ONLY THE BEGINNING
Courtesy of Gramene.org
New publicly accessible Web resources provide researchers with data integration unimaginable only a few years ago. This series of images from Gramene shows how it is possible to zoom in on the genome from the chromosomal level to the base pair level. These data views show, among other things, expressed sequence tags, transcription units, and homologous genes in related organisms.
Despite these differences, plant biologists are excited about the completion of gene sequencing leading to a heyday for functional genomics. Consider David E. Salt at Purdue University. He describes four pillars: the proteome, the metabalome, the transcriptome, and the ionome (which covers all the mineral ions going in and out of cells). Instead of looking at one or two ions, say calcium or potassium, the appropriately named Salt examines how ions relate to each other and to the genes that regulate them.6
Salt cites three reasons to learn how plants process minerals. First, more efficient use of nutrients such as nitrogen and phosphorus might allow plants to grow better on marginal land where people cannot afford fertilizer. Second, raising mineral levels in staple food crops might improve nutrition for at-risk populations. Finally, there is the possibility of reclaiming metal-contaminated soil using nonedible crops, which could remove high levels of metal either for reclamation or removal to a landfill.
Salt is hardly alone in his vision of functional genomics rising to the challenges of feeding the world's population in a changing environment. The Oryza Map Alignment Project, headed by Rod Wing of the University of Arizona, will map the genomes of 12 wild species of rice and align them to the sequenced genome for the cultivated varieties. If the project can identify ancestral genes and functions that have been lost though rice's evolution into an edible cereal, it might find ways to improve one of the world's leading food crops. Wing will be looking for genes that allow rice to survive in shade or salty water, for instance. "Probably close to 99% of the gene pool hasn't been tapped yet," he says, emphasizing that anything learned about the rice genome could have broad application to other cereal grains.
Courtesy of Lynx Therapeutics
George Golda, an engineer, with the Lynx Massively Parallel Signature Sequencing (MPSS) machine he helped create. The machine allows two flow cells containing each 1.6 million beads to be sequenced in parallel.
Wing attended the First International Symposium on Rice Functional Genomics in Shanghai in November; he will host the second meeting this fall. "A lot of really good functional genomics is going to be coming out of China and Japan," he predicts. "I kind of envy the grad students or undergrads thinking of going in science now," says Wing. "By the time they get their PhDs, the genomes of the majority of crop plants and the genes will already be there at their fingertips."
Jane Salodof MacNeil firstname.lastname@example.org is a freelance writer in Groveland, Mass.
A NEW MODEL
As much as scientists hope to learn from Arabidopsisand rice, it has become apparent that neither can explain processes such as nitrogen fixation and symbiosis, which are specific to legumes. Consequently, a third model plant is emerging: Medicago truncatula, a close relative to alfalfa. Nevin D. Young of the University of Minnesota, St. Paul, heads a multinational group that is sequencing Medicago.
"To impact world food you have to understand legume biology as well," Young says. "Legumes can take nitrogen out of the atmosphere and convert it to a biologically useful form," he adds, envisioning a day when societies use legumes for pollution control and bioremediation as well as protein.
Young predicts the project will have a draft sequence available in three years on interconnected Web sites. When a biologist wants to study legumes in 2007, he says, "the scientist ought to be able to sit at a computer and do research without ever walking into a lab or greenhouse."
SELECTED WEB RESOURCES
Arabidopsis TILLING Project http://tilling.fhcrc.org:9366
AREX: The Arabidopsis Gene Expression Database http://www.arexdb.org
GABI: Genome Analysis of the Plant Biological System http://www.gabi.de
GARNet:Genomic Arabidopsis Resource Network http://www.york.ac.uk/res/garnet/garnet.htm
INE: Integrated Rice Genome Explorer http://rgp.dna.affrc.go.jp/giot/INE.html
International Rice Genome Sequencing Project http://rgp.dna.affrc.go.jp/IRGSP
Maize Genetics and Genomics Database http://www.maizegdb.org
Massively Parallel Signature Sequencing Project http://mpss.udel.edu
Plant Genome Database http://www.plantgdb.org
Rice Genome Automated Annotation System http://RiceGAAS.dna.affrc.go.jp
GMO DEBATE CLOUDS FUTURE
At a time when plant biologists foresee their work relieving world hunger and healing the environment, many fear that political opposition to genetically modified organisms (GMOs) will prevent society from reaping the benefits of plant genomics.
The controversy is already having an impact in Europe, where volatile debate is constricting programs. British science has taken a strong stand in defense of plant genomics, but Oxford University's Malcolm M. Campbell says European Union money is being channeled from pure research to "food security issues," and private funding is no longer available. "Plant people are the bad boys," laments pro-GMO Jens Freitag of the Genome Analysis of the Plant Biological System (GABI) project.
Opposition is not expected to be as strong in Asia and other parts of the world under pressure to improve their food supply, but in more affluent Japan, Takuji Sasaki of the National Institute of Agrobiological Sciences reports GMO is a hot issue. His institute has started a new section to respond to GMO opinions, but he is not sure it can satisfy opponents. "I feel that the anti-GMO [argument] is an emotional one, and it is not easy to find out a solution only by a scientific explanation," he says.
Many plant scientists express frustration with what they see as inconsistencies, omissions, and misleading arguments. They ask why critics do not object to wearing genetically modified cotton, producing seedless grapes, or eating foods that are the result of thousands of years of selective breeding. Sasaki worries that the Japanese dependence on food imports is not being discussed. Ottoline Leyser, GARNet coordinator, complains that environmental groups are scaring people about technology; she says that their real concern is the possibility that an herbicide-tolerant crop could harm birds. "There's a fundamental confusion between the technology and the application for which it's used," she says.
Genomics, many point out, does not mean genetically modified food. Genomics is "agnostic," according to Nevin D. Young at the University of Minnesota, St. Paul. "Changing things is not built into genomics," Young says. "Genomics is discovering how things work. It's up to scientists and policy makers and corporations and entrepreneurs to decide what to do with that information." One option is application of the knowledge gained to conventional plant breeding. Inevitably, however, many foresee genetic modification as offering substantial benefits to a hungry third world, only to be rejected by GMO opponents in developed nations.
- Jane Salodof MacNeil