NEW YORK (GenomeWeb News) – The small, aquatic plant duckweed — a potential biofuel source — has a tiny genome and hangs on to its juvenile state, according to a paper appearing today in Nature Communications from researchers led by Rutgers University's Joachim Messing.
Duckweed is a fast-growing flowering plant — under certain favorable circumstances its population can double in a manner of days as it relies on vegetative propagation — found in small lakes and ponds where it can form thick mats. It is also a simple plant, consisting of a kidney-shaped frond that lacks a stem and, in some derived species, roots. Its leaves resemble cotyledons, the embryonic leaves found in plant seeds.
Messing and his colleagues in Germany and the US sequenced Spirodela polyrhiza, or Greater Duckweed, and found that it contains some 158 million basepairs and less than 20,000 protein-coding genes, or about 27 percent fewer protein-coding genes than Arabidopsis thaliana and 50 percent less than rice. Duckweed, they found, has increased numbers of repressors of the transition from juvenile to adults and fewer promoters of the process, as compared to A. thaliana.
"The most surprising find was insight into the molecular basis for genes involved in maturation — a forever-young lifestyle," senior author Messing, director of the Waksman Institute of Microbiology at Rutgers, said in a statement.
Because, among other characteristics, duckweed has a high biomass yield density, low levels of lignin, and can be grown on marginal land, it is viewed as a promising biofuel source. Having a better understanding of what its genome contains and mechanisms therein will enable its use as a biofuel, the researchers said.
"Classical breeding or genetics does not apply here because of its clonal propagation and rare flowering, but these organisms can be transformed with DNA," Messing added. "[N]ew variants can be created with modified pathways for industrial applications. These variants would be an enhancement over what can be done now."
As a representative of duckweed, Messing and his colleagues sequenced a cluster of three to five fronds of a clonally grown S. polyrhiza strain by whole-genome shotgun sequencing using both single-end and pair-end Roche/454 next-generation sequencing in conjunction with pair-end Sanger sequencing.
They assembled the 158 megabase genome into some 250 scaffolds that they aligned to a Spirodela BAC library and joined into 32 pseudomolecules. A masking method indicated the assembly was at least 90 percent complete for genic sequences and more than 80 percent complete for the rest.
Based on a homology search, the researchers reported that about 13 percent of the Spirodela genome is LTR-retrotransposon-derived, which is comparable to other plants with its genome size. As compared to A. thaliana, though, the LTR-retrotransposon-derived regions of the S. polyrhiza genome are much older, and its genome lacks younger insertions. While they noted that the apparent absence of younger insertions could be a sequencing artifact, the researchers said that the "atypical age distribution suggests an 'ancient' genome state without much recent transposon activity in combination with small removal rates."
Because of both the small size of the S. polyrhiza genome and its apparent control of transposon activity, the researchers hypothesized that the features could be linked to the plant's propensity toward clonal propagation.
By comparing the S. polyrhiza and Oryza sativa genomes, the researchers peeked into how the S. polyrhiza genome evolved. They found 2:4 syntenic relations between the genomes that, in addition to chromosomal copy number duplication patterns, indicated that the plant underwent two rounds of whole-genome duplication. Further, drawing on synonymous mutations rates, they said that those two duplication rounds likely took place within a short timeframe, which they traced to about 95 million years ago.
S. polyrhiza also contains fewer genes than other plants — 19,623 protein-coding genes, or 28 percent less than Arabidopsis' 27,416, and 50 percent less than O. sativa's 39,049 protein-coding genes — as determined through gene models based on transcriptome sequencing from diurnal time courses and stress conditions.
Even though the S. polyrhiza genome is small, it shares more than 8,000 common gene families with other plants, such as Arabidopsis, tomato, banana, and rice. It tends, however, to have gene clusters that were on the low side of average for gene expansions and copy number.
It also lacks 750 orthoMCL gene clusters that other plants have. Those clusters include genes involved in water transport, lignin biosynthesis, and cell wall organization. This, the researchers noted, "is consistent with the specialized morphology and lifestyle of Spirodela."
Further, they also reported that Spirodela is missing a number of the cell wall loosening α-expansin genes and has only three β-expansins, even though monocotyledonous plants underwent an expansion of β-expansins. It also has a limited copy number of, or is missing, genes involved in cell wall biosynthesis and lignification. Spirodela does, though, contain nearly the full lignin biosynthesis pathway, but the copy number is reduced as compared to other plants.
Meanwhile, Spirodela has an increased number of repressors targeting the transition from juvenile to adult, as compared to the other plants, as well as fewer components that enhance the transition to the adult phase. In addition, Spirodela has up to 32 copies of miRNA 156, which has been shown in Arabidopsis to promote the juvenile stage and inhibit the adult; other plants have about a dozen copies of that microRNA gene.
"The low gene count of Spirodela could in part be due to the structural reduction and juvenile nature, reducing the need for and consequently the retention or duplication of genes acting in the adult phase," Messing and his colleagues said.
With the duckweed genome in hand, Messing and his colleagues said that it could now be scoured to optimize the plant as a biofuel source. "The genome sequence of Spirodela provides the first step to identify, understand, and improve relevant traits for specific target applications," they said.