Wen Zhiyou
Editor|Wen Zhiyou
«——[·Preface·] ——»
Aquatic plants face many new challenges compared to terrestrial plants. The habitat they occupy is often highly dispersed, with isolated bodies of water surrounded by large swathes of "arid deserts". This can lead to reduced gene flow, inbreeding, and potential local extinction.
The level of gene flow and the degree of genetic structure of these species may also be affected by the mating systems they employ. To test this hypothesis, we compared the phylogeographic structure of two freshwater plants in the British Isles, mainly cloned angiosperms and isosporidous pinus plants.
We sampled the two plants from the lakes where they co-emerged and used restriction site-associated DNA sequencing to infer relationships between them. The genetic structure between lakes is higher in angiosperms, which we associate with reduced sexual reproduction, so the level of gene flow between lakes is lower.
In addition, we found evidence of lineage-specific associations between specific lake trophic types in Singleflower Lake, which may be the result of environmental filtering of specific ecotypes.
Overall, we conclude that the reproductive system of lycopods is less specific to terrestrial conditions, and that it provides an advantage after secondary colonization of aquatic habitats, which allows for frequent gene exchange between populations and may promote faster adaptation.
«——[Ferns.]——»
The transition from aquatic to terrestrial environments has occurred many times in flora and fauna. This is often accompanied by a series of challenges related to survival and reproduction.
In plants, ancestral reproductive patterns are intrinsically linked to the presence of water, and once plants become terrestrial, it is necessary to increase the specialization of the reproductive system to adapt to dry conditions and several lineages then transition back to the aquatic environment, which may disproportionately affect their ability to spread depending on their reproductive strategy.
In the basal groups of terrestrial plants such as bryophytes, ferns, and lycophytes, male gametes are flagellar and intolerant to dryness, and sexual reproduction often requires moist habitats, even in terrestrial environments.
As a result, secondary aquatic species of these groups are able to sexually reproduce underwater.
In contrast, submerged flowering plants, angiosperms, share the mating system of their terrestrial ancestors and typically use only flowers for sexual reproduction on the surface of the water, although underwater sexual reproduction has evolved in some taxa.
The type of propagule will further affect the spread in the aquatic environment. Aquatic propagules can be efficiently disseminated in aquatic environments, but the production of resistant to dry propagules, such as fruits and seeds, may help them spread in "dry deserts" between isolated aquatic habitats.
Gene flow between populations depends on their ability to transmit and mating systems. This, in turn, affects the genetic structure of populations, affecting their adaptive potential and resilience to environmental changes.
While population size and its spatial distribution also affect intraspecific genetic structure, in plants, the reproductive system is arguably the most important factor. This has important evolutionary consequences, especially in highly dispersed habitats.
Plants from freshwater environments, such as rivers and lakes, are particularly likely to experience habitat fragmentation. These environments are ephemeral in evolutionary time and are not necessarily directly related to other suitable habitats, leading to a high risk of local extinction, small effective population sizes, and decline in inbreeding.
Despite these limitations, plants are widespread in freshwater environments and do have a very large range of species compared to terrestrial plants, a paradox that has long fascinated biologists. Solving this paradox requires the use of population genetics methods to estimate effective diffusion rates and gene flow.
Many studies have deduced the genetic structure of angiosperms and more basal flora. However, the genetic structure of angiosperms and basal vascular plants settled in the same freshwater environment has never been directly compared.
Lycopods and angiosperms co-occur in the lakes of northern Europe. Despite 400 million years of independent evolution, the two species exhibit converging ecological and phenotypic characteristics.
Both species independently adapted to carbon-depleted aquatic environments through relatively slow growth rates, evergreen leaves, isogrowth forms, internal voids that allow access to sediment CO2, and sedum acid metabolism.
Although their distribution, ecology, and plant types are converging, the two species retain different reproductive systems corresponding to their taxa. Angiosperms reproduce asexually by producing short creeping stems when submerged, although the buoyancy and longevity of the floating whole plant may also allow for short asexual transmission within the lake.
In contrast, the reproduction of lycopodians occurs through the fusion of small and large spores. Because the spores are scattered in water, genetic exchange between submerged plants is possible, although the ratio of heterohybridization to self-inbreeding is unknown.
Little is known about the mechanism of lake-to-lake dispersal of heterosporids Pinus species, with waterfowl and wind-mediated transmission being the most prominent recommendations. Long-distance transmission of this species may remain challenging because dried spores of two closely related species, Lycopodus and Angiosperms, lead to failure to germinate.
In this study, we compared the intraspecific structure of British and lycophytes. Until about 12,000 years ago, the ice sheet covered much of Northern Europe, including the United Kingdom, and these geographical areas were subsequently resettled from the refuges.
As such, the UK's Lycopods and populations are highly similar in ecological and demographic history and therefore represent a good system in which to understand the impact of their contrasting reproductive systems on the genetic structure of populations, and their adaptive evolution to these species.
«——[Plant material and sequencing·] ——»
Lycopods and angiosperms were collected, dried and stored in silica gel from mainland Scotland, as well as the Outer and Inner Hebrides in August-September 2016. In addition, a single sample of stone pines and stones was collected from Snowdonia, Wales in 2016.
Follow the manufacturer's protocol to extract DNA from silica-dried leaf material using the Disney Botanical Mini Extraction Kit, except for the elution step, which is performed once using 50 μl AE buffer.
Briefly, DNA (approximately 200–700 ng) is double-digested with Eco RI and Mse I. The barcode connector is connected to the Eco RI side and the normal connector is connected to the Mse I side. After ligation, PCR amplification of the library using standard Illumina sequencing primers.
Based on relative estimates of library concentration, a total of 96 samples from the same and different items were pooled. Select the size of the library cell by gel extraction with a target size of 300–600 bp and purify using the gel extraction kit.
Two-ended sequencing (2x125bp) was performed on a single HiSeq2500 channel at the Edinburgh Genome Centre following standard protocols.
«——[Assembly and analysis of chloroplast genomes.] ——»
Readings of cleaning and pruning were mapped to previously assembled lycophytes and plastic bodies collected from Wales, using the default settings of bowtie2v.2.2.3 with paired end readings.
Base calls and maximum likelihood phylogeny for each plastid genome location were extracted using an internally developed shell script under the RAxMLv.8.2.11 under the GTR+G surrogate model. The node supports evaluation with 100 bootstrap replications.
«——[Firm analysis of nuclear polymorphisms.] ——»
Use the program ipyradv.0.7.2, with default parameters for clustering and assembly. To avoid incorporation of plastid and mitochondrial sites in the final assembly, only clusters with less than 100x coverage are processed.
The maximum number of alleles per single nucleotide polymorphism is set to two, and only loci present in at least 40% of the sample are incorporated into the final assembly. All samples of each genus are used for two separate clusters.
Deletion data less than 60% of random single nucleotide polymorphisms were extracted from each assembled RAD locus using vcftools v.0.1.15. The generated unlinked SNP dataset was used for phylogenetic and genetic structure analysis.
«——[Genetic structure of angiosperms.] ——»
On average, 83% of the chloroplast genome is covered by filtered readings. The inferred plastid phylogeny is generally poorly conclusive and has low support values.
Interestingly, some geographically distant populations were grouped together, while geographically close populations were placed in different regional trees.
Overall, a high degree of diversity was observed, including a single lake from Snowdonia, Wales, with clean readings ranging from 800,000 to 2.4 million per sample, likely reflecting variations in the quality and quantity of DNA and libraries entered.
A total of 128,359 RAD sites were assembled for assembly. After filtration, 14,669 of them with 1.7% polymorphic sites are retained for analysis. The level of homozygosity is moderate.
The first two principal components (PC) in explain 16.7% and 12.4% of the variation in the data, respectively. The first PC separates the two samples (30 and 19) from all the others, reflecting chloroplast phylogeny, where the two samples form a single lineage.
The remaining samples formed three groups on the second PC, one of which corresponded to the Welsh samples (A, B, and C), while the other two groups represented different types of lakes independent of geography. This model is broadly generalized in maximal likelihood nuclear phylogeny.
In the remaining samples, monophyletic support for the mesotrophic and oligotrophic groups was strongly supported. However, some important inconsistencies have been observed between chloroplast and nuclear phylogeny, such as the lack of aggregation by lake type in chloroplast phylogeny.
Paired FST values show modest differences based on geographical origin, with values ranging from 0.14 to 0.22 between populations from different regions. However, pairwise F ST in the phylogenetic population mainly confirms the genetic structure we observe.
«——[Genetic structure in lycophytes ·] ——»
On average, 51% of lycopods are covered by sequencing readings. Phylogeny inferred from the phylogenetic system revealed two distinct groups in the genus Leeks, with bootstrap support of 100.
A previously published comparison of lycopod and angiosperm sequences identified diagnostic SNPs in this trnL gene, suggesting that members of the smaller clade are angiosperms and members of the larger clade are lycopods. Bootstrap support is generally low within the Pinus group.
A total of 134,378 RAD sites were assembled for leeks, of which 16,451 were retained after filtration. These loci contain 4.4% of polymorphic loci.
A second assembly using only Pinus samples yielded a total of 99,672 RAD sites, of which 19,855 were retained after filtration and 3.5% showed polymorphism. On average, samples of lycophytes had lower levels of homozygosity than those of Pinus species.
Principal component analysis of all leek samples clearly separates the two leek species identified in chloroplast phylogeny.
Similarly, the nuclear phylogeny of leeks clearly divides the two species into two highly supportive monophyletic clades. In the Lycophytes, the evidence for clustering is less pronounced than in Lycophytes, with samples widely distributed over the first PC and little clustering between geographic areas or lake types.
The second PC explained the 14.0% variation and roughly separated one sample (48) from the eutrophication lake in North Uist from the rest. Clade support values in the phylogeny of lycophyte in Pinus carinaceae are low, and there are no obvious geographic or lake type clusters.
The F ST value between geographical areas in lycophytes is usually lower than that of L. Uniflora (0.09-0.11 versus 0.14-0.22), similar levels of difference between samples from Wales, mainland Scotland, and the Outer Hebrides (0.09-0.11), and poor and mesotrophic samples showed limited genetic differentiation (F ST = 0.10).
«——[The wave of well-known British Isles.] ——»
As the ice sheet retreated in the post-glacial period in Britain, and the lycophytes were exposed to aquatic habitats by early colonists. However, this model does not appear to involve a single wave of colonization from a limited number of sources.
For both species, we found different genetic lineages in geographically adjacent lakes. Cohabitation of different genomes is consistent with multiple, independent colonization.
Different groups of angiosperms found in some lakes of the Hebrides or postcolonization by glaciers from unique glacial refuges.
The two conditions could not be distinguished without additional sampling outside the British Isles, but the coexistence of different genomes suggests that freshwater plant populations are not homogenized.
In addition, this view is supported by overall high chloroplast diversity, coupled with the lack of clear isolation of distances estimated with paired FST.
«——[·Author's View·] ——»
In this study, we compared the genetic structure of two freshwater plants belonging to a highly differentiated taxa within the British Isles, Lycophytes and Angiosperms. Our survey revealed a high level of population structure. We believe this stems from the increased chance of underwater sexual reproduction of lycopods of the Lycophytes, which depend on aquatic structures.
We also show that certain lineages of the genus Cyprus monopolis appear to be limited to lakes with specific nutritional status. We believe this pattern is due to the early adaptation of some populations to new habitats, followed by intense ecological filtering.
This pattern is not observed in lycopods, which can be explained by frequent genetic exchanges in this species, allowing adaptive alleles to spread more rapidly between lineages.
«——[References] ——»
1. Ashworth, "Reproductive Sensitivity of Plants to Habitat Fragmentation: Review and Synthesis through Meta-Analysis," Eco Letters, 2006.
2. Banksy, "Cypress and 400 Million Years of Separation," Annals of Plant Biology, 2009.
3. Barrett, "Evolutionary Process of Aquatic Plant Populations," Aquatic Botany, 1993.
4. Barrett, "The Ecology of Mating and Its Evolutionary Consequences in Seed Plants," Annual Review of Ecology, Evolution, and Systematics, 2017.
5. Burks, "Vegetation Development of Large Aquatic Plants in Lake Kråkenes, Western Norway during the Late Glacial Age and Early Holocene", Journal of Paleolacology, 2000.