To the editor

In an influential paper in 1989, Meyerowitz1 described Arabidopsis thaliana as a “useful weed” and indeed he was right. This weed, endogenous to temperate climes, has been the corner stone of a transformation in plant biology over the past decade and has been widely employed as a model for development and differentiation in plants. In addition, data about its genes and the isolation of cloned cDNAs have facilitated the improvement of crop plants.

Though not widely known, for several years, Arabidopsis has been a model for wood and fiber production in trees2,3,4. Repeated clipping of inflorescence stems of the weed induces several common genotypes (and probably many others) to produce much more secondary wood than usual in all plant parts. The increased secondary xylem production of Arabidopsis is sufficient not only for developmental studies2,3 but also for the cloning of xylem-specific genes4.

Turning weeds into crops is an ancient practice5. Now that sequencing of the Arabisopsis genome is complete6, it should be possible to turn Arabidopsis from a useful weed into a useful crop. Why ignore its tremendous potential for genetic manipulation when many other crops are much less easy to transform or unresponsive to tissue culture protocols?

For many purposes, when large biomass is not required, this small plant is sufficient. We can make it sweet (with sugar or with sugar substitutions), sour or hot, enrich it with antioxidants, vitamins, pigments, amino acids, or fats, and use it to express many other medicines and natural products.

We already eat or process quite a number of the Brassicaceae, and Arabidopsis could join these crops. Turning a model plant into a crop is a test and challenge for modern biology. It is time to start Arabidopsis agriculture.