The emergence of seed-producing plants more than 300 million years ago was an evolutionary watershed, opening new environments to plants and ultimately leading to the flowering plants that brighten our world and supply much of our food. But it was less of a leap than it seems, newly published DNA sequences suggest.
The genomes, from three fern species and a cycad, one of the oldest kinds of seed-bearing plants, show genes key to making seeds are the same as those in the spore-producing machinery of ferns, which emerged tens of millions of years earlier . They evidently existed in a common ancestor but were recruited into different reproductive functions as plants diverged.
The fern and cycad genomes, published in a series of papers over the past several months “fill the gap of the gene flow during plant evolution,” says Shu-Nong Bai, a plant developmental biologist emeritus at Peking University who helped sequence a member of the maidenhair fern genus. “Evolutionary innovation [can] come from the alternative use of existing genes or networks, not new genes.” The genomes also teach a second striking lesson: that plants acquired some of their genes not through mutation and selection, but straight from fungi or other microbes through a controversial process dubbed horizontal gene transfer.
Because of the daunting size of most fern genomes and the focus on crops such as rice, wheat, and maize, the majority of the more than 800 plant genomes sequenced so far have come from seed plants. Until now, just two were from ferns—ones with unusually small genomes. As a result, “We have only had a small snapshot of plant evolution,” says Blaine Marchant, a plant evolutionary geneticist at Stanford University.
Thanks to advances in sequencing long stretches of DNA and reductions in costs, his team and three other groups have now tackled ferns with more typical, large genomes as well as a species of cycad, a nonflowering plant with bare seeds, like those of pines and other conifers. “It is wonderful to finally see more diverse plant genomes being sequenced,” says Jennifer Wisecaver, an evolutionary biologist at Purdue University.
The fern genomes, with some 30,000 genes each, reveal a panoply of genes previously tied to flowering plants, which evolved more than 200 million years later. For example, Marchant and his colleagues reported on September 1 in Nature Plants that a water fern, Ceratopteris richardiihas 10 members of a gene family known to control flowering time, seed germination, and flower shape in a small flowering plant, Arabidopsis. Their roles in the fern are unclear, but seven of these genes are active in leaves where spores are produced, suggesting they play a role in reproduction in ferns as well as in seed plants.
Jianbin Yan, a plant physiologist at the Chinese Academy of Agricultural Sciences’s Agricultural Genomics Institute, and colleagues found similar parallels in a maidenhair fern, Adiantum capillus-veneris. Its DNA contains genes for transcription factors called EMS1 and TPD1, proteins that in maize and other seed plants regulate genes involved in pollen development, Yan’s team reported in the same issue of Nature Plants. These pollen gene controllers are active in maidenhair’s sporangia, the tissue where spores develop.
That fern’s genome also contains a trio of genes that regulate seed development in flowering plants, adds Hongzhi Kong, a plant evolutionary developmental biologist at the Chinese Academy of Sciences’ Institute of Botany. Ferns, Yan says, are “evolutionarily pivotal for a comprehensive understanding of the origin and diversification of the seed.” The cycad genome contains similar networks, showing they were active in the earliest seed plants, notes Shouzhou Zhang, the botanist at the Fairy Lake Botanical Garden in Shenzan, who led its sequencing.
The new genomes shed light on one reason such insights were slow in coming: Ferns “are notoriously known to have gigantic genomes,” says Fay-Wei Li, a plant evolutionary biologist at Cornell University. Researchers had assumed a process called whole genome duplication, in which an organism’s complement of DNA is doubled during reproduction, explains their genome size. But, “We’re not seeing the genome doubling footprint that we thought we would,” says Paul Wolf, a plant geneticist at the University of Alabama, Huntsville. Instead, the ferns and the cycad gained the bulk of their DNA from the accumulation of mobile DNA—transposons and other genetic elements that infect genomes and multiply, or repetitive DNA, short sequences that got copied over and over again.
The four new genomes are also changing views about whether plants experience horizontal gene transfer. Microbes are known to swap genes all the time, helping them adapt to new conditions, but multicellular organisms seem to borrow genes only rarely. However, the genomes of the ferns and cycad contain a surprising number of genes from bacteria and fungi. “It is remarkable that we see genes of bacterial and fungal origin in vascular plants,” Kong says.
For example, the cycad sequenced has four copies of a fungal gene for a cytotoxin, a protein that can bore holes in foreign cells, and the Ceratopteris genome has 36 copies of another cytotoxin gene from a bacterium. These acquired genes could have bolstered their new hosts’ defenses against pathogens or herbivores.
Verónica Di Stilio, a botanist at the University of Washington, Seattle, expects more surprises from the newly unveiled genomes. “Having reference genomes representative of each of the major plant lineages opens up so many possibilities,” she says. “Genomes are tools, the tip of the iceberg.”