Arabidopsis genome. A milestone in plant biology.

The completion of the Arabidopsis genome sequence is the culmination of a remarkable decade of world-wide growth and collaboration in developing Arabidopsis as a model genetic and genomic system. Since the 1930s, plant geneticists have been drawn to the study of Arabidopsis because of its small size, its predilection to self-pollinate, its quick generation time, and its copious production of tiny seeds. Parallels with Drosophila melanogaster were commonly discussed, but with the exception of a few pioneers, Arabidopsis was not widely embraced by the broader plant biology community until the last 10 years when it became apparent that genomic tools had the potential to revolutionize plant biology. I made the decision to switch the focus of my laboratory from work on symbiotic nitrogen fixation to Arabidopsis-pathogen interactions in the summer of 1984, but the appeal of Arabidopsis had already been percolating in my mind for a number of years. Influenced by the work of George Rédei and John Langridge on the isolation of Arabidopsis biosynthetic mutants (Langridge, 1965; Rédei and Li, 1969), I devoted about a year of my postdoctoral training in 1973/1974 to the development of an Arabidopsis somatic cell system. As an assistant professor in 1975, the major goal of my first National Science Foundation (NSF) grant was the “transfer of functioning bacterial nitrogen fixation genes to plants.” Although the plant in this proposal was Arabidopsis, the seeds that I had worked on as a postdoc remained in an unopened vial in my desk drawer for many years. In fact, I did not fully appreciate the power of Arabidopsis as a genetic model until the early 1980s when I heard Chris Somerville describe the genetic analysis of photorespiratory mutants at a Plant Molecular Biology Gordon Conference, became aware that Elliot Meyerowitz’s laboratory had shown not only that Arabidopsis has a small genome but also had begun construction of a correlated physical genetic map, and took a close look at Maarten Koornneef’s and David Meinke’s work on the isolation of Arabidopsis developmental mutants (Meinke and Sussex, 1979; Koornneef et al., 1980; Somerville and Ogren, 1982; Leutwiler et al., 1984). I then concluded (with some feeling of regret that my Arabidopsis seeds had lain dormant for so long) that it was just a matter of time before Arabidopsis would be adopted as a universal model system for plant biologists and that it would be a good idea to get in on the ground floor. The amount of research on Arabidopsis has grown from a trickle in the 1970s and 1980s to a steadily increasing flood in the 1990s (Fig. 1). The completion of the Arabidopsis genome sequence no doubt will stimulate even more Arabidopsis-related work. It is noteworthy that the highly productive Arabidopsis genomics enterprise that we have today is not simply a consequence of plant biologists systematically recognizing the advantages of a model plant system. On the contrary, in the 1970s and 1980s many influential plant scientists resisted attempts to adopt a model system, arguing that the diversity of crop plants and the realities of funding plant-related research demanded that the majority of plant research should be carried out on agronomically important species. Interestingly, I think that the development of recombinant DNA technologies initially reinforced the view that a model plant system was not necessary because rDNA technology allowed genes to be cloned from any plant. This led to many unrealistic predictions about the coming of a second green revolution, but in reality, because of a lack of funding and a lack of focus, plant research languished in comparison to the stunning advances that were being made by our colleagues working with animal systems. Thus, I believe that it was important, at least in the United States, that the NSF began to orchestrate a worldwide Arabidopsis genome project. In so doing, NSF had the willing assistance of an activist group of Arabidopsis converts who organized themselves into the Multinational Coordinated Arabidopsis Genome Research Project and the North American Arabidopsis Steering Committee. Gradually, an Arabidopsis infrastructure was established that included stock centers in the United States and Europe, a database, an annual meeting, recombinant inbred mapping lines, identification of PCR-based mapping markers, an expressed sequence tag sequencing project, and a coordinated genome sequencing project. More recent additions to the infrastructure include knock-out libraries with more than 200,000 lines in the public domain, expression tagged lines, transcriptional profiling tools, and a large database of single nucleotide polymorphisms and small insertions and deletions between the Columbia and Landsberg erecta accessions. The latter database, which should greatly facilitate map-based cloning (Drenkard et al., 2000), is notable because it was posted as a public service by Cereon Genomics, setting an important precedent for what will hopefully be the beginning of a new era of cooperation between industry and academia. So now, only 10 years after the establishment of the Multinational Coordinated Arabidopsis Genome Project, there is a complete Arabidopsis sequence. Remarkably, the sequence was completed only 1 year after the Caenorhabditis elegans and a few months after the D. melanogaster genomes were finished, a major achievement for plant biology. It is important to explicitly and forcefully make the point that the com-