A precarious balance

a haploid organism directly leads to an altered phenotype, making forward and reverse genetic approaches more straightforward in haploid moss than in diploid seed plants. A major additional asset of mosses came from the discovery that, in the moss Physcomitrella patens, recombination occurs between DNA introduced into cells by transfection and homologous sequences in its nuclear DNA. This occurs as efficiently as in yeast — five orders of magnitude (!) more efficiently than in any other plant species that has been tested. Since then, this technique has been used to study gene–function relationships in single gene knockout mosses. Additionally, homologous recombination has been used to generate tagged, saturated Physcomitrella mutant collections as the basis for genome-wide studies of plant gene functions. A special offer from moss? The last common ancestor of mosses and seed plants lived about 450 million years ago. Mosses have not changed much since then, and, consequently, they offer the chance to learn more about plant evolution and diversity. Are there differences between gametophytic and sporophytic gene regulation? How do single cells decide to differentiate into new tissues? Are basic mechanisms of regulatory networks and cross-talk conserved between mosses and seed plants? Can novel genes or metabolites be identified from moss? Mosses offer a variety of metabolites that are not known from seed plants. Some of them, like very long-chain polyunsaturated fatty acids, are of significant commercial value in improving the human diet and consequently the relevant moss genes are being transferred into seed plants to alter their fatty acid composition into a moss-like one. Moss can be grown efficiently in large-scale bioreactors to produce foreign proteins, including human proteins. Inactivating, by homologous recombination, the genes for the enzymes that mediate plant-specific protein glycosylation alters the modification patterns of moss proteins to a human-like pattern; a milestone in the production of biopharmaceuticals in plants. Sequence information from the Physcomitrella transcriptome is rapidly increasing and presently covers more than 95% of the estimated 25,000 protein-encoding moss genes. Mosses have conserved, ancient biochemical pathways; unlike seed plants they show no real codon-bias; on average, they have fewer representative members per protein family; and they have more than 5,000 genes with no clear homolog in seed plants. This impressive set of novel genes is attracting more and more scientists. However, to fully understand — and exploit — land plant diversity, the full genome sequence of Physcomitrella is needed. The genome …