Skip to main content
NCBI home page
AoB Plants logoLink to AoB Plants
. 2025 Mar 31;17(3):plaf018. doi: 10.1093/aobpla/plaf018

An ecological, phenotypic, and genomic survey of duckweeds with their associated aquatic environments in the United Kingdom

Kellie E Smith 1,2,✉, Laura Cowan 3, Paulina Flis 4, Chris Moore 5, Matthew Heatley 6, Carlos A Robles-Zazueta 7, Adam Lee 8, Levi Yant 9
Editor: Anna Locke
PMCID: PMC12190800  PMID: 40574895

Abstract

The duckweeds feature global distributions and diverse applications in phytoremediation and nutrition, as well as use in fundamental studies of development. Existing collections have minimal environmental data linked to natural habitats. Thus, there is a lack of understanding of natural variation in the context of native habitats. Here, a novel collection of 124 duckweed accessions from 115 sites across the United Kingdom was characterized by genome sequencing and ionomics. In common nutrient-replete experimental conditions, all accessions hyperaccumulated P, K, Mg and Ca. Local but not large-scale associations were revealed between elemental composition of duckweed in common, replete conditions and native water profiles. Lemna minor was the most prevalent species in the UK, with a closely related hybrid L. japonica frequently found in waters with higher micronutrient concentrations. Invasive L. minuta was common in the southern and midland regions, but restricted in Scotland. Lemna accessions accumulated heavy metal contaminants typically together with macronutrients, suggesting phytoremediation potential, but some limitations as food. Furthermore, monitoring the ecological interactions between native, hybrid and invasive Lemna species should be ongoing in the interest of biodiversity.

Keywords: duckweed, ionomics, adaptation, invasive species, hybridization, phytoremediation

Duckweed species from a large UK collection were characterized using genotyping, phenotyping, and elemental analysis of both aquatic environments and duckweed plants. The native Lemna minor was the most widespread species. Invasive Lemna minuta and hybrid Lemna japonica were also common, though absent in Northern Scotland. Lemna japonica was found in environments with higher levels of several elements than Lemna minor , while Lemna minuta thrived in magnesium-rich environments, suggesting enhanced elemental adaptability in these invasive and hybrid species. Additionally, use of common nutrient environments revealed differences in Lemna species accumulation of macronutrients. Some individuals of each species showed exceptional ability to accumulate or exclude certain elements, highlighting potential applications for food production or water remediation.

Introduction

Duckweeds (Lemnaceae) represent some of the fastest growing flowering plants in the world (Sree et al., 2015; Ziegler et al., 2015) and are powerful models for studies in development (Ware et al., 2023), bioremediation and ecotoxicology (Laird and Barks, 2018). There are at least thirty-six species of duckweeds, consisting of simple stem-leaf structures called fronds in the rootless Wolffia and Wolffiella genera, to early diverged root-bearing genera Spirodela, Landoltia and Lemna (Lam and Michael, 2022; Ware et al., 2023). Duckweed has increasing roles in wastewater purification through uptake of excessive nutrients, metals and toxic elements leached from industrial and agricultural activities (Landesman et al., 2010; Ekperusi et al., 2019). Moreover, other varieties are emerging as food sources when grown hydroponically and axenically in vertical farms, providing comparable protein and nutritional contents to wheat (Appenroth et al., 2017; Xu et al., 2023). However, duckweed applications development requires improved knowledge of trait natural variation between accessions, species and environments (Barton, 2024). For context, traditional crops like wheat and maize were optimized for abiotic stress resilience and nutritional quality by assessing wild relatives in their natural habitats (Zhang et al., 2017; Reynolds and Braun, 2022). Similarly, harvesting wild duckweed accessions which are well adapted to specialized water environments or those showing enhanced nutrient accumulation potentials could be key to unlock their development in future food and phytoremediation applications. To this end, duckweed cultivation in controlled environments is a powerful phenotypic approach to select for candidates with desirable traits (Bergmann et al., 2000), which can be later translated into a genetic approach to engineer superior varieties (Stomp, 2005; Lam and Michael, 2022).

Presently duckweed varieties are surface sterilized and adapted to axenic artificial conditions (Sree and Appenroth, 2020). Collection dates, original environmental data and genome sequencing for existing clones are largely unavailable, limiting studies of accession optimization to bioremediation or food production. Ecological characterization at regional scales have been performed from European (Kirjakov and Velichkova, 2016), Middle Eastern (Friedjung Yosef et al., 2022; Taghipour et al., 2022) and Asian collections (Xu et al., 2015; Chen et al., 2022; Kadono and Iida, 2022; Tran et al., 2022). These works have uncovered various invasive and hybrid Lemna (L.) species. Within Europe, four invasive alien species have been identified, with L. minuta dominating (Lansdown, 2008; Fedoniuk et al., 2022; GBIF, 2021). Lemna minuta is a species that opportunistically outgrows native L. minor in high nutrient and light controlled conditions (Njambuya et al., 2011; Paolacci et al., 2016, 2018a) with potentially destructive ecological consequences, including reduction of L. minor observations in wild populations in Italy (Ceschin et al., 2016). However, wild species characterization is limited in the UK due to paucity of clones with only four classified as L. minuta species (Rutgers Duckweed Stock Cooperative, 2018).

Clear temporal and spatial patterns in L. minuta dispersal drive species invasion fronts (Ceschin et al., 2018a). For example, humans and avian species are an important vector of dispersal (Coughlan et al., 2015; Silva et al., 2018) whereas, increasing ambient temperature and nutrient availability promotes invasive L. minuta growth (Njambuya et al., 2011; Peeters et al., 2013). From initial invasion fronts upon European Atlantic coasts in the 1960s, L. minuta is now firmly established in the UK and registered in the Global Register of Introduced and Invasive species (GRIIS) with alien status causing negative impact (Ceschin et al., 2018a). Although, negative impacts of invasive L. minuta dense infestation include thick mat formations which decrease light penetration, pH and oxygenation into water bodies, thereby reducing native biodiversity and causing problems for aquatic flora and fauna (Janes et al., 1996; Ceschin et al., 2019). Although notably, native L. minor can also form dense mats and achieve species dominance compared to L. minuta in some aquatic environments (Paolacci et al., 2018b). Wetland habitats in the UK of high conservation status are now threatened by hyper-eutrophication, ecosystem imbalance and duckweed invasion (Feller et al., 2024). It is therefore timely to conduct regional surveys of both native and invasive duckweed species in wild wetlands with a view to assessing adaptations in these environments.

Additionally, particularly ‘extremophile’ duckweed have great promise for the development of phytoremediation and food applications. Consideration of the plant ‘ionome’ refers to its whole-tissue or organismal levels of macro-, micronutrients and trace minerals (Salt et al., 2008). The applications of ionomics ranges from assessments of nutrient uptake and soil/water relations to understanding the nutritional composition of food and biofortification of crops. In duckweed, clones of L. minor and Wolffia globosa may have phytoremediation potential as they bioconcentrate by a factor of 100-fold and hyperaccumulate over 1 g/kg dry weight of heavy metals such as Cd, Cu and As (Zayed et al., 1998; Zhang et al., 2009). From a worldwide collection, L. yungensis clones displayed local-scale tissue variation in macronutrients Mg, S, and Mn (Smith et al., 2024b). This suggests that differences in nutrient accumulation could be due to either short-term phenotypic plasticity or linked to adaptations to previous micro-habitats as shown for other traits (Roubeau Dumont et al., 2019; Huber et al., 2021; Strzałek and Kufel, 2021; Antro et al., 2022; Jewell and Bell, 2023; Vámos et al., 2023). However, there is still a lack of understanding of the scale of variation in either water environments or the attendant accumulation potential of native duckweed accessions.

This paper presents a genomic, ecological and environmental assessment of novel UK duckweed accessions, detailing 115 environments and 124 accessions. We discover elemental variation using ionomics at local scales and document the spread of invasive L. minuta, as well as new reports of hybrid species. A common garden experiment with replete nutrient media was used to measure differences in duckweed whole-plant tissue ionomes. Native environmental water chemistry was also measured using inductively coupled mass spectrometry (ICP-MS). Overall this work provides a local-scale and UK-wide assessment of duckweed variation and water habitats, providing accessions with promising elemental accumulation profiles with potential for food and phytoremediation applications.

Materials and methods

Selection of site locations

To assess distributions across fine- to moderate geographic scales, duckweeds were collected from across England, Wales and Scotland, totalling 115 wild environments from the UK. For temporal assessment, a subset of 19 sites along an inland-coastal transect were selected locally with initial duckweed and water collection performed during autumn 2020, and subsequent water collections performed in summer 2021, autumn 2021 and winter 2022. For spatial assessment of the UK, duckweed and water collections were conducted in spring 2021, starting at southern locations in early April and finishing mid-May 2021 in northern Scotland, to account for variation in springtime across UK. Regions were chosen to span the UK using duckweed observations reported recently using the Global Biodiversity Information Facility (GBIF, 2021). Locations from GBIF.org were mapped onto Google maps and several potential sites were searched within each region to give n=>6 local sites with duckweed presence. Names and descriptions of sites are given in Table S1 and a map for sampling regions presented in Fig. S1.

Collection of duckweeds and morphological assessment

A total of 124 duckweeds were collected as described in (Smith et al., 2024a) from temporal and spatial collections. The primary latitudinal axis was between 41 sites in southern England and Wales (regions HAS, COR, BRI, NEW) and 37 sites across Scotland (regions ABE, ELG, GLA), giving a total of 103 accessions. The central UK consisted of five sampling regions LAN, BFD, YOR, HUL, and MID, yielding a total of 44 duckweed accessions. All accession names, sampling coordinates, dates and characteristics are provided in Table S2A. In sites with more than one suspected duckweed species, these were collected and cultured separately based on size and denoted as A, B, or C. From across 19 sites along the seasonal transect, duckweeds were phenotyped at each time point, and a handful re-sequenced and denoted as 1, 2 or 3, when they showed differences from species previously characterized there. For other sites, duckweeds were collected at a single time point.

Morphological characteristics were used for species determination. Frond characteristics (length, width, length width ratio (L:W), number of fronds per colony and anthocyanin presence) were assessed initially from images taken with a Zeiss SV6 stereo microscope (Ziess, Oberkochen, Germany) (n = 10 colonies per accession). Then each of these characteristics in addition to root lengths were measured at two later timepoints after lab sterilization and cultivation, to capture any phenotypic adaptations to common controlled environments. Anthocyanin was estimated by RGB colour analyses of photographs. Root characteristics were measured using duckweeds floating on water in a modified square petri dish (Greiner Bio-One) and stood vertically with inclusion of a ruler for photographs of the air–water interface. Stomatal counts were performed for n = 3 whole fronds of a subset of 24 accessions using a Leica TCS SP5 confocal microscope (Leica, Wetzlar, Germany) using preparations as described in (Kurihara et al., 2015; Smith et al., 2024b). The biomass of three frond colonies (n = 3 per accession) and presence or absence of turions were assessed from cultures exhausted in nutrient media over three years. All images were analysed with Fiji (Schindelin et al., 2012).

Collection of water samples

Environmental assessment was performed concurrent with plant collections, focusing on water body analysis for elemental composition. For seasonal water collection, 100 ml samples were taken four times at each site, unless accessibility issues or water was not present. Solid Phase Microextraction Polytetrafluoroethylene (SPME PTFE) amber bottles were pre-washed with ultra-pure 10% nitric acid overnight followed by soaking in MilliQ water (Milipore, USA) and then thoroughly air dried. At each site, water bottles were washed at the top surface of the water, filled to 100 ml and 0.5% ultra-pure 1 ml nitric acid added, before storage at 4°C. Later, 18 ml water was filtered through a 1.45 µm syringe filter into 2 ml 10% Primar grade nitric acid to acidify samples to release suspended metals from the solid phase into solution. For UK-wide samples, water samples were collected in triplicate per site at a single time point, from the top water surface and filtered through a 1.45 µm syringe filter into High density polyethylene (HDPE) Universal 30 ml tubes (Sarstedt, Leicester, UK). HDPE tubes were pre-weighed, then 2 ml Primar grade 10% nitric acid added and then re-weighed. All samples were stored at 4°C before ICP-MS analyses.

Plant care and harvesting for DNA sequencing

Duckweeds were sterilized using 0.5% sodium hypochlorite and grown in GEN2000 SH controlled environment cabinets (Conviron, Winnipeg, Canada). Among all collections, four accessions from the south and 14 accessions from Scotland could not be successfully cultured in laboratory conditions. The majority were L. trisulca and provisional L. gibba clones, possibly due to hypersensitivity to sodium hypochlorite or specific adaptation to locality so were not included for sequencing or ionomics. After sterilization and weekly media changes of successful cultures, independent sealed flasks of UK accessions were grown for four weeks for DNA harvesting. Duckweeds from the Landolt collection and available at Rutger’s stock database (www.ruduckweed.org) were also grown and DNA harvested to provide known species controls. For each accession, 20-100 mg fresh duckweed tissue was harvested into liquid nitrogen and then stored at −80°C.

DNA isolation, short-read library preparation and sequencing

Accessions were ground using a Tissuelyser II (Qiagen, Hilden, Germany) and DNA extracted using DNAeasy Plant kit (Qiagen, Hilden, Germany). DNA quantification was performed using dsDNA HS assay (Thermo Fisher Scientific, Massachusetts, USA) and Qubit 2.0. DNA was diluted to < 20 ng/µl with sterile MilliQ water. Individual Illumina DNA Prep (Illumina, San Diego, USA) sequencing libraries were prepared on a Mosquito HV (SPT Labtech, Melbourn, UK) liquid handling robot using 1/10th volumes at all steps. A total of 9-48 ng of DNA was used as library input and 5 cycles of polymerase chain reaction were used for the library amplification step. Final libraries were normalized and pooled on a Fluoroskan Ascent fluorometer (Thermo Fisher, Massachusetts, USA) and the resulting pools were size selected using 0.65X Ampure XP (Beckman Coulter, California, USA) to remove library fragments < 300 bp. Short-read sequencing using paired-end reads with Illumina HiSeq 2500 platform sequencing was performed using a target of 20x coverage.

Variant calling

The processing pipeline involved three parts: (1) preparing the raw sequencing data, (2) mapping and re-aligning the sequencing data and (3) variant discovery (GATK v.4 following GATK best practices). In addition to newly sequenced samples, previous sequencing data for existing clones were downloaded from the National Centre for Biotechnology Information (NCBI) Sequence Read Archive (SRA) and summarized in Table S2B. To prepare raw sequencing data for mapping, the different sequencing lanes were concatenated, followed by quality trimming using Trimmomatic (Bolger et al., 2014). All genomes were then aligned to reference genome L. minor 7210 (SRR10958743) using BWA 0.7.17 (Li and Durbin 2009) and processed using Samtools v1.9 (Li et al., 2009) and duplicate reads flagged using ‘MarkDuplicates’ from picard-tools 1.13464 followed by GATK v.4 to re-align reads around indels (McKenna et al., 2010). The variant dataset was filtered for biallelic sites and mapping quality with GATK using QD < 2.0, FS > 60.0, MQ < 40.0, MQRankSum < -12.5, ReadPosRankSum < -8.0, HaplotypeScore < 13.0 and sites remaining after depth filtering DP < 141 carried forward for analysis. The scripts for batch processing are available at https://github.com/mattheatley/ngs_pipe.

Genomic analysis

Degenotate (https://github.com/harvardinformatics/degenotate) was used to identify genome variants encoding fourfold degenerate sites (4FDS) as proxies for neutrally evolving sites. These sites were further filtered > 20% missingness to reduce the cohort from 143 varieties to 135. The dataset was pruned by linkage disequilibrium to obtain independent segregating markers out of linkage using a custom script (Hämälä et al., 2024). The final genomic analysis included only biallelic single nucleotide polymorphisms at minor allele frequencies > 2.5%, with one SNP per 100 kb sliding windows with a step size of 50 kb and r2 of 0.1. Species allocation was confirmed using a mixture of PCA, tree, and structure-based approaches using R v3.6.3. The PCA was produced for variants using ggplot2 (Gómez-Rubio 2017). Unrooted neighbour joining trees were compiled using ape v5.4 package (Paradis and Schliep 2018) for L. minor and L. japonica. For structure plots, 4FDS variants with > 20% missingness were dropped, removing two varieties, and then converted into genotype call files using PLINK v1.9 (Chang et al., 2015) based on Hardy-Weinberg equilibrium and Fisher exact tests. Allele frequencies were used for group allocation and admixture proportions by FastStructure v1 (Raj et al., 2014) with K groups between 4-10. Selection of K = 4 provides the most interpretable grouping when visualized with a Structure plot v2 using Omicsspeaks omicsspeaks.com/strplot2/

Duckweed growth and harvesting for ionomics experiments

After nine months of subculturing, ionomic experiments were conducted for accessions grown in controlled environment cabinets. Two colonies of each accession were grown in 500 ml Erlenmeyer flasks containing 250 ml Nutrient medium, replenished weekly for six weeks. N-medium was used as described in (Appenroth et al., 1996; Appenroth & Sree, 2015) and contains KH2PO4 (0.15 mM), Ca(NO3)2 (1 mM), KNO3 (8 mM), MgSO4 (1 mM), H3BO3 (5 μM), MnCl2 (13 μM), Na2MoO4 (0.4 μM) and FeEDTA (25 μM). Duckweeds were grown at 25°C day and 18°C night, with 16 h day lengths providing a light intensity of 100 µmol m-2 s-1. To harvest, duckweed were rinsed for two minutes each with three MilliQ water washes. Three replicates were obtained for each accession from three independent flasks using 150 mg duckweed tissue per sample. Duckweed were dried in an oven at 88°C overnight and stored in a desiccator before analysis. Weights of the dried tubes were made using a 5 dp precision balance (Mettler Toledo, Ohio, USA).

Ionomics processing using inductively coupled plasma mass spectrometry (ICP-MS)

Elemental analysis for water and duckweed samples were analysed on a NexION 2000 ICP-MS (PerkinElmer, Massachusetts, USA) in Helium collision mode. For each set of water and duckweed analyses, calibration standards were run throughout using single element standards (Inorganic 226 Ventures; Essex Scientific Laboratory Supplies Ltd, Essex, UK), to subtract against background samples. Concentrations of elements in water samples were measured in µg/L for Na, Mg, Si, S, K, Ca, Al, P, Li, B, Ti, Cr, Mn, Fe, Co, Ni, Cu, Zn, As, Rb, Sr, Mo, Cd, Pb and Sn. Ti, Cr, Sn were removed from water analysis as they were below the limit of detection (LOD). Ba was also removed as it was not measured across all sites. Duckweed samples were digested with 2 ml 63% nitric acid at 115°C for 4 hrs (spiked with the element Indium as an internal standard) and then digested with 0.5 ml hydrogen peroxide for a further 1.5 hrs at 115°C, before dilution into 10 ml MilliQ water. Elements Li, B, Na, Mg, Al, Si, P, S, K, Ca, Ti, Cr, Mn, Fe, Co, Ni, Cu, Zn, As, Rb, Sr, Mo, Cd, Sn, Ba, Pb were measured in duckweed tissue by dry weight (mg/kg). Elements with low levels in duckweed Li, Cr, Sn < 1 mg/kg and Ni < 3 mg/kg were removed from further analysis as they were below the LOD of ICP-MS.

Analysis of ionomics data

For water and duckweeds analyses, elements were grouped as those present in duckweed growth N-medium or trace/heavy metals (negligible presence in N-medium) for separate analysis. Water site replicates were combined to form site averages (n = 3) and replicates per accession combined for accession averages (n = 3). Standardization for each element by z-scores were obtained by subtracting raw data for each element from the panel mean and dividing by standard deviation (SD) to produce heat maps, radar plots and PCAs. Linear regression with Spearman correlation was performed between each accession’s ionome and their originating water elemental concentration averages, grouped initially by species, by region and where possible by species and regions to find positive and negative relationships. Permutation analysis was performed using 5000 iterations for each element from each region and P values < 0.05 considered significant as they represent < 5% probability of extreme values. For comparison of species ionomes, those with fewer accessions were dropped including potential Lmu/Lmo hybrids, L. gibba, L. turionifera and S. polyrhiza. Differences between the remaining Lemna species elemental concentrations and originating water elemental profiles were compared separately with a Kruskal–Wallis test and a post-hoc Dunn’s test with Bonferroni adjustment using P = < 0.05.

Results

Phenotype-based species identification

Morphological factors were used to determine species membership of accessions, including frond and root characteristics and turion production. Phenotypes were quantified first upon collection and then confirmed during laboratory growth for three years cultivated in controlled growth environments. Principal component analysis (PCA) was used with a subset of UK accessions suspected to be different species to discriminate species clusters based on morphological characteristics (Fig. 1B). Morphological assessment confirmed that UK duckweed consisted of species in the Lemna and Spirodela genera.

Figure 1.

ALT TEXT: Photographs and graphs comparing duckweed species based on morphological traits, including frond size and shape, number of fronds per colony, three-frond biomass, stomatal density, and root length. Statistical analyses are shown to indicate significant differences between species.

Open in a new tab

Frond and root characteristics used to classify six UK duckweed species. (A) Photographs of four duckweed species growing in native sites. Genus Lemna (L.) L. minor (red), L. minuta (green), and L. trisulca (blue) and Spirodela (S.) S. polyrhiza (purple). (B) Species confirmation using multiple morphological traits represented on a PCA as assessed in a laboratory using a subset of 30 accessions. Principal components PC1 and PC2 explain ~50% data variation. Ellipses display 90% confidence intervals for species groups and overlaps indicate reduced morphological criteria to differentiate between four Lemna species: L. minor, L. minuta, L. turionifera and L. japonica. Within species groups the number of accessions were: L. japonica = 11, L. minor = 5, L. minuta = 5, L. trisulca = 4, L. turionifera = 2 and S. polyrhiza = 3. Arrows are coloured by squared cosine (Cos2) with > 0.5 indicating phenotypic traits contributing most to dataset variation on PC1 and PC2. (C–F) Differences in morphological traits between four Lemna species (C) Frond number per colony, (D) stomatal counts, (E) root length, and (F) three frond biomass per colony (3-fr biomass). Boxes display median and 25% and 75% percentiles for each species. Phenotypes which are significantly different between species are indicated on the top of each plot using Kruskal–Wallis P = <0.05. Different letters indicate significant differences between species using a Dunn’s post-hoc test with Bonferroni adjustment using P = <0.05. For L. minor and L. minuta differences in frond number (P = 0.001), and for L. japonica and L. minuta differences in root length (P = 0.004), stomatal counts (P = 0.003) and biomass of a three-frond colony (P = 0.0002).

Spirodela and Lemna species can be differentiated by frond and root characteristics (Fig. 1). Criteria for membership in the Spirodela genus included larger fronds and multiple roots per colony (Landolt, 1986), anthocyanin accumulation, shorter roots and lower length-to-width frond ratios (L:W) than Lemna (Fig. 1A, B). Within Lemna, L. trisulca had thin, pointed fronds, giving the highest L:W ratios and higher fronds per colony connected by long stipes (Fig. 1A, B). Lemna turionifera were deduced from other Lemna species by observations of turions (overwintering bodies) produced in nutrient-depleted conditions. Lemna minuta produced fewer fronds (three per colony), compared to other species including L. minor, producing a maximum of eight (Fig. 1C). Roots were shorter in L. minuta but frond adaxial stomatal counts were almost two-fold higher than other Lemna species (Fig. 1D:E). In contrast, Lemna japonica and L. minor could not be differentiated by morphological criteria (Fig. 1B, 1C:E).

Genomics-based species identification

To extend the criteria for distinction of these two Lemna species and further confirm other species definitions, a genetic structure analysis was carried out. Whole genome sequencing of 122 new UK accessions was performed and mapped to a common L. minor 7210 reference genome. These were processed with ten additional varieties sequenced from the Rutgers duckweed collection and four other available sequences were downloaded from public repositories (see Methods, Table S2A:B) for a total of 136 accessions. All accessions were classified into species clusters by a variety of genomic clustering methods, including PCA and FastStructure analysis.

This PCA shows genetic groupings by species, when using the primary and secondary principal components (Fig. 2). PC1 vastly explained 72% of the variance and clearly discriminated species groups L. minuta and L. minor, with Spirodela emerging on PC2. Overall UK species were clustered with known species and genomic analysis aided species discrimination, compared to just using morphology alone. Native Lemna trisulca species formed a cluster with (L. trisulca 7192), invasive L. turionifera (L. turionifera 6002) and Spirodela (S. intermedia 9394) (Fig. 2A). Genome analysis was key for distinction of native Lemna minor and hybrid L. japonica which showed similar phenotypic traits. Two clusters of L. minor (Lmo) clones were initially observed and differentiated as C1 and C2 (Fig. 2A, C, D). These two groups clustered close together in the PCA (Fig 2A) but were much better discriminated by FastStructure and especially tree-based approaches (Fig. 2B:D). The C1 cluster includes English Lmo7016 and Irish Lmo5500. Therefore this cluster is inferred as L. minor. The C2 cluster is located between L. minor (C1) and L. turionifera clusters on the PCA (Fig. 2A). In cluster C2, very strongly admixed European L. japonica 9250, Canadian L. japonica 7123, South African Lmo8389 and North African Lmo7295 were found along with hybrid L. japonica (Ljp) species.

Figure 2.

ALT TEXT: Graphs illustrating genetic relationships among UK duckweed accessions, including the interspecific hybrid Lemna japonica. A scatter plot shows species clusters in low-dimensional space based on genetic similarity. A stacked bar plot represents the proportional genetic contributions of each species to individual accessions. Phylogenetic trees depict genetic distances among accessions and species.

Open in a new tab

Genetic structure of 136 duckweed accessions, primarily from the novel UK cohort. 122 novel UK accessions, four previously published (L. minor 7016, L. minor 5500, L. gibba 131, L. turionifera 6002), and ten newly sequenced duckweeds from the Rutgers duckweed collection. (A) PCA of 11,088 quality-filtered four-fold degenerate (neutrally evolving) SNPs. Species are coloured by clusters determined by previously identified clones. The L. minor (red) clade shows two clusters labelled as cluster one (C1) and two (C2) conforming to membership with established, previously phenotyped clones. (B) FastStructure analysis differentiates accessions by group membership. Lemna minor accessions from the ABE region are labelled. Accessions with Bayesian probability assigning them to two or more species groups show admixture and are determined as hybrid species. K = 9 was best supported under Bayesian model selection showing the highest marginal likelihood, however no further meaningful groups were found beyond K = 4 in the structure plots. The scale represents Bayesian probability of likelihood of species membership. (C) Neighbour-joining tree showing genetic differentiation between species. Lemna minor (C1) clustered with L. minor 7016, 5500 and L. japonica (C2) grouped with L. minor 7295 and 8389 and L. japonica 7123. D. Close-up of a neighbour-joining tree distinguishing L. minor from L. japonica. Lemna minor accessions from COR with a common ancestor are labelled. Clone sequences from the Sequence Read Archive (SRA) or newly sequenced in this study from the duckweed stock collection are labelled in italics with their corresponding identifying number.

Structure analysis was used to estimate ancestry and to assign membership of each accession to species (Fig. 2B). This confirmed that C1 group were entirely L. minor species, and the C2 cluster contained accessions with substantial admixture with invasive L. turionifera gene pools. The C2 cluster is therefore likely composed of L. japonica, an interspecific hybrid of L. minor and L. turionifera (Lmo/Ltu) (Fig. 2B). Six accessions showed potential admixture between L. minuta and L. minor (Lmu/Lmo hybrid; Fig. 2B) however, morphological criteria did not differentiate these from L. minuta.

Interestingly, some Lemna species are difficult to resolve by genetic structure alone. Lemna yungensis clone 9208 from Bolivia clustered with UK L. minuta accessions, showing a high degree of similarity between these species, both are part of the ‘Uninerves’ section of Lemna, consisting of one frond nerve (Fig. 2A,C, Bog et al., 2020). Lemna turionifera (Ltu) and L. trisulca (Ltr) are robustly separated using morphology (Fig. 1) but were undifferentiated by FastStructure (Fig. 2B) possibly due to having few representative accessions in each group (2 and 5) and low mapping efficiency to the L. minor reference (Table S2C). Furthermore, putative Lmu/Lmo hybrids could not be reliably confirmed by genetic structure due to the same limitations as other minority species (Fig 2A, B) and could not be distinguished from L. minuta by phenotyping.

Highly variable species distributions by region

Lemna minor (n = 81) were the most common species in number and diversity across the UK survey (Fig. 3A). This species was found both in monocultures and co-existing with other species. Lemna minuta were also frequent (n = 30), and exhibited a marked latitudinal contrasting distribution (Fig. 3D). Lemna minuta prevalence in south England and Wales was greatest (20/41 sites; 49% prevalence), with presence at 11/32 sites in central England (34% prevalence) compared to negligible presence in Scotland (3/36 sites; 8% prevalence; Fig. 3D). Lemna minor was the only species found across all sampling sites within the ABE region in the north of Scotland, (Figs. 2B, 3). In contrast, the southwestern BRI and NEW regions had the greatest species diversity both between sites (Fig. 3) and within sites, with up to three species co-existing in several sites (Table S2A, Fig. S2I, J and S3I, J), including the less frequent S. polyrhiza and L. gibba.

Figure 3.

ALT TEXT: Maps of the UK showing the distribution of five duckweed species, represented by coloured dots. The maps illustrate the widespread occurrence of Lemna minor and the absence of hybrid and invasive Lemna species in northern Scotland.

Open in a new tab

Different prevalence of Lemna species within UK sampling regions. Five Lemna species, coloured by species. (A) L. minor, (B) L. turionifera, (C) L. japonica, (D) L. minuta, (E) Putative Lmu/Lmo. The five species include common native duckweed (L. minor), invasive species (L. minuta, L. turionifera), hybrid species (L. japonica (Lmo/Ltu)) and potential hybrid Lmu/Lmo. The total regions n = 12, and sample sites within regions n = < 10.

Lemna turionifera was sparse throughout the UK; searches yielded only two accessions isolated in the northeast of England (Fig. 3B). The L. japonica (Lmo/Ltu) hybrids were abundant and overlapped with the locations of two L. turionifera accessions, from which interspecies hybridization may have occurred (Fig. 3B, C). Lemna japonica were more prevalent than the L. turionifera parental species but not as cosmopolitan as the L. minor parental species, as they were not found in Scotland (Fig. 3A, C). Conversely, the putative Lmu/Lmo hybrids could not be distinguished from L. minuta by distribution as they were found in southern regions, mirroring the pattern of L. minuta (Fig. 3D, E). Low numbers of accessions of Lemna japonica and Lemna minuta were found in GLA compared to regions in England. There were no hybrid or invasive species found in regions in the north of Scotland ABE and ELG during this survey (Fig. 3C:E).

Duckweed species broadly classified as native, invasive and hybrid types following morphological and genomic assessments. Native UK species included L. minor, L. trisulca and S. polyrhiza. We aimed to characterize presence of L. minuta invasive species, but we found that and an additional invasive species, L. turionifera. Furthermore, L. japonica hybrids formed between native and invasive species were detected. Species types exhibited different regional distributions (Fig. 3) and further showed contrasting whole-plant ionomes in common, replete conditions, along with native water elemental differences between derived habitats (Fig. 4).

Figure 4.

ALT TEXT: Violin plots showing elemental concentrations in duckweed plants and their water habitats, colour-coded by species. Potassium and magnesium levels vary between duckweed species, while magnesium and arsenic differ across water habitats. Statistical values highlight significant differences.

Open in a new tab

Macronutrients (K, Mg) composition varies between species, with additional variation of Mg and As between water environments. (A–C). Elemental composition of K, Mg and As, for whole duckweed tissue. (D–F). Environmental water concentrations of K, Mg and As. Whole tissue ionome element content (mg/kg) averaged per accession and grouped into species. Site water average elemental composition in µg/L from n = 3 water replicates per site. Significance was assessed by a Kruskal–Wallis test and Dunn’s post hoc test with Bonferroni adjustment using P = <0.05 to indicate species significant differences using letters above plots.

Variable ionomic profiles are species-specific

In total, twenty-six elements were measured in 116 accessions from 100 water sampling sites. After classification into species using phenotyping and genomic clustering, species differences between plant ionomes detected in a common garden and their home water chemistries were compared (Fig. 4). Overall, tissue levels of Mg and K contents varied significantly between the predominant Lemna species when grown in common conditions (Fig. 4A, C, D). The highest Mg content overall was found in invasive L. minuta (P = < 0.0001, Fig. 4B, Table S3). The hybrid L. japonica had higher K levels than other Lemna species and higher S, Si and Mo than its parent L. minor (P = 0.0015, Fig. 4A, Table S3). Additionally, L. japonica was found on waters higher in other macro- and trace minerals Mg, P, S, Ca, B, Mo and Sr than L. minor (P < 0.05, Table S4).

Species showed contrasting accumulation profiles in replete nutrient conditions and also showed differing originating water elemental profiles. Lemna minuta accumulated more Mg overall and was found in higher Mg environments than L. minor (Fig. 4B, E, Tables S3, S4), demonstrating one example of higher elemental compositions in both native water and in ionomes of species. Lemna japonica was found in water environments with higher As contamination but did not differ from other species in As accumulation in replete nutrient conditions (Fig. 4C, F, Tables S3, S4), Lemna japonica accumulated higher K levels than other Lemna (Fig. 4A), however, originating water levels of K did not vary between species (Fig. 4D, Table S4). Therefore, with the exception of Mg, there was generally no clear trend between elemental chemistry of water sites and elemental accumulation by species in replete nutrient conditions.

Widespread within-species ionomic variation in common conditions

In common conditions all accessions accumulated macronutrients P, K, Mg, Ca above 1 g/kg. All accessions hyperaccumulated B, Mg, Fe, Zn, Cd and Ba using a bioconcentration factor threshold of at least 100-fold. Overall, the largest variation of tissue concentrations between duckweed accessions were found for Mn and Pb, followed by S (Table S5). For Mn, 82/116 accessions hyperaccumulated it, including all of the Scottish accessions. Relative to the duckweed cohort as a whole, B was hyperaccumulated by accession HAW, Si by accession BOG and Fe by accession LAN (Table 1).

Table 1.

Accessions showing variable accumulation of elements as measured by ICP-MS

Accession Region Species Duckweed ionome elemental variation
BOG GLA Lmu Inline graphic B, Na, Si, S, Ba, Pb, Al, Ti, Zn, Cd Inline graphic P
LAN HAS Lmu Inline graphic Fe, S, Si, Mg, Pb, Al, Ti, Zn
HAW NEW Lmo Inline graphic B, Ca, Fe, Cd, Ba, Pb Inline graphic K, Rb
ALL2 BFD Lmo/Ltu hybrid (Ljp) Inline graphic B, Na, Sr, Ba, P Inline graphic Fe
MAV ELG Lmo Inline graphic B, Fe, Cu, Ba, Cd
APP BFD Lmo/Ltu hybrid (Ljp) Inline graphic Mo, Na, P
CRO HAS Lmo Inline graphic B, Ca, Cu
Open in a new tab

Green triangles (Inline graphic) indicate higher accumulation and red triangles (Inline graphic) indicate reduced accumulation compared to cohort average. Accumulation differences are considered significant when z-scores exceed ±2 SD for n = 116 accessions. Lmo—L. minor, Lmu—L. minuta, Ljp—L. japonica.

Often hyperaccumulation of one element is accompanied by changes in suites of others (Table 1). The hyperaccumulating accessions (HAW, BOG, LAN) all had higher concentrations of multiple other elements, including heavy metals, compared to the cohort average (Table 1, Fig. S2C,J,L S3C,J,L), indicating some interdependence. BOG showed the most differential ionome, accumulating ten different elements. Overall, B was the most accumulated element in five out of seven higher accumulator accessions. Higher accumulation of elements co-occurred with reduced levels of macronutrients P, K and Fe in some instances (Table 1). High accumulators consist of several Lemna species and originated from a range of collection regions (Table 1, Fig. S2 and S3).

Regional and local-scale site water elemental variation

Simultaneous with duckweed collection, water samples were collected for elemental composition in order to relate environmental chemistry with those ionomes of specific accessions. High nutrient water bodies included the BRI and BFD regions (summarized in Table 2, Fig. S6). Different regions showed ranges of Ca, Mg and especially Mn concentrations (summarized in Table S6). Additionally, the levels of Mg, Ca, Mn and Fe were highly variable seasonally, with the site ALL showing the largest variation overall (Fig. S7, Table S7).

Table 2.

Native duckweed environmental water sites with significant levels of five or more elements as measured by ICP-MS

Duckweed site Region Species Significant elemental variation in water habitat
ALL BFD Lmo/Ltu hybrid (Ljp) Inline graphic K, P, Mo, B, Pb, Sn, Rb, Ni
MLA BRI Lmo Inline graphic Mg, Si, S, Ca, Li, Sr
NEW BRI Lmu Inline graphic Mg, S, Ca, Li, Sr
KEY HUL Lmo Inline graphic P, K, Cd, Ni, Co
Open in a new tab

Summary of water sampling sites with increased levels of macronutrients and heavy metals. Green triangles (Inline graphic) indicate higher accumulation compared to water site average and is considered significant when z-scores exceed ±2 SD. Species refers to the duckweed species found at water sites, Lmo—L. minor, Lmu—L. minuta, Ljp—L. japonica.

Duckweed ionome responses in relation to native aquatic environments

To infer presence of local adaptation, native water chemistry was compared with ionomes of plants grown in common nutrient-replete conditions, using linear models with spearman’s rank correlation. At the grossest scale, no significant relationships were observed for twenty-two elements between water and duckweed ionomes across the sampling range as a whole. Furthermore, relationships between native environments and accumulation of elements within species overall showed no clear trends, with the exception of negative relationships for B (R = -0.66, P = 0.0069) and Cu (R = -0.75, P = 0.0012) for L. japonica. At a finer scale, associations by region were evident (Fig. 5) showing that region- or concentration- specific levels of elements in water habitats may therefore drive specific duckweed adaptation.

Figure 5.

ALT TEXT: Violin plots showing correlations between elemental concentrations in duckweed tissue and their respective water environments. Regional variation is observed, with elements such as lead, magnesium, manganese, and sulfur exhibiting both positive and negative trends across different sites.

Open in a new tab

Specific regions show significant directional linear associations between site water elemental concentrations and Lemna duckweed tissue concentrations. (A and B) Elements with significant correlations in northern UK regions. (C and E) Elements with significant correlations in southern UK regions. Normalized water elemental concentration (µg/L) are recorded on the x axis with normalized duckweed whole tissue element concentration on the y axis (mg/kg) to scale values between 0 and 1. Duckweed accessions and sites within each region is n=>5. y and x represents line slope and intersect, Spearman correlation coefficients are significantly directional if R values are <0.50 or <−0.50. Permutation analysis was used to report significant relationships when P = <0.05. Non-Lemna (Spirodela) accessions were removed from analysis.

Relationships between native environments and duckweed accumulation were element and region specific. Notably, regions BFD, YOR, HUL and BRI gave rise to outlier accumulator accessions but showed no strong or significant correlations between environment and internal ionomes. In contrast, higher accumulation of Pb by GLA accessions was positively correlated with higher water levels of Pb in native environments (Fig. 5B), forming the only positive relationship documented for elements within regions. However, the concentrations of macro- and micronutrients including Mg, S, Mn and Si showed the opposite trend in specific regions, with higher concentrations in duckweed tissue correlating with lower concentrations of these elements in water sites (Fig. 5A,C:E). Therefore, in these specific regional cases, heavy metals such as Pb tended to increase in duckweeds originating from contaminated water environments, but for some macro- and micronutrients, low levels in the water environment potentially stimulated higher relative accumulation in duckweeds from these native sites when grown in replete nutrient conditions. Accounting for species variation, L. minor which was the most dominant species throughout regions, showed negative relationships for Pb in ABE (R = -0.63, P = 0.044) and additionally S in COR (R = -0.73, P = 0.021) providing support for (Fig. 5D). However, lesser species L. minuta and L. japonica showed no strong relationships between elemental concentrations in water sites and accumulation, possibly due to lack of replication.

Water body elemental levels did not appear to drive broadly deviant plant ionomes. Whereby, none of the accessions with particularly many elements outside the normal range had higher or lower elemental concentrations in their source water profiles compared to the UK average (Tables 1, 2, Fig. S2, S5). From the seven accumulating accessions presented in Table 1, (including hyperaccumulators of B, Si and Fe), six of these came from water bodies with broadly normal levels of elements, including B, Si and Fe (Fig. S5C). An excellent example is the accession BOG which exhibited the most extreme ionome overall, hyperaccumulating ten macronutrients and heavy metals with reduced P (Figs. S2C, S3C). However, the source water chemistry from which BOG was collected harboured no elements significantly differing from the cohort average (Figs. S4C, S5C). Conversely, accession ALL1 from the most nutrient-dense and contaminated ALL site (Figs. S4D, S5D) only highly accumulated Fe in replete conditions (Tables 1, 3, Figs. S2E, S3E). It is noteworthy that Fe, K and B were highly variable at this site, with Fe showing maximum levels in autumn 2020 but in decline until 2022 (Fig. S7).

From this survey of duckweed in the UK, native L. minor was diverse and commonly found. Presence of invasive species L. minuta and L. turionifera were more limited as were new reports of hybrid species L. japonica and potential hybrid Lmu/Lmo. Species showed different potentials to uptake macronutrients (Mg, K) and heavy metals (As) and also inhabited waters with diverse elemental profiles. Hyperaccumulators and high accumulator accessions were detected, usually coincident with higher uptake of multiple elements. Relationships between duckweed ionomes and water profiles were complex. Some elements show specific trends only in a few localized regions, possibly due to varied water chemistries and species composition.

Discussion

Region-wide genomic assessments of duckweed diversity are scarce. Furthermore, existing duckweed collections lack data on source environmental parameters, and no study has assessed local-scale whole-plant ionomes in common conditions using regional sampling. Nor is there any genomic assessment of invasive duckweed impact on native accessions. In this study we fill these gaps, and further interpret this information with a view toward identification of useful accessions tailored to phytoremediation and food development applications.

UK duckweeds are composed of native, invasive and hybrid Lemna species

This survey confirms that native L. minor is cosmopolitan and thus still more prevalent than invasive L. minuta, which had a more limited distribution, especially in the north. Lemna minor is evidently well-adapted to the UK environments here, although hybridization with invasives is a liability. Lemna minuta was only prevalent in the south and midlands of England with some ingress into Scotland. Lemna minuta has been added to the GRIIS of Great Britain; however observations have been in decline since 2019, according to the Global Biodiversity Information Facility (GBIF, 2021). Thus, presence of L. minuta is possibly not as damaging to native species as previously reported (Paolacci et al., 2018b) and this study found evidence of co-occurrence of both species in water bodies. Although invasive species can be opportunistic in in-vitro high light and Mg conditions (Paolacci et al., 2016, 2018a), dominance of invasive or non-invasive species depends on competition in particular environments (Paolacci et al., 2016; Van Echelpoel et al., 2016; Gérard and Triest 2016; Ceschin et al., 2018b; Paolacci et al., 2018a). It appears that environments in Scotland are less suited to promote opportunism in L. minuta species.

Within the worldwide duckweed collection (such as the Landolt collection), many L. minor species have been reclassified as hybrids including L. japonica and L. mediterranea (Braglia et al., 2021, 2024), indicating that the previous assignments of L. minor, L. japonica and other Lemna species may not be fully correct. Genetic contribution from both L. minor and L. turionifera in L. japonica accessions indicated admixture and therefore hybrid presence in the UK. The presence of both L. japonica and potential Lmu/Lmo hybrid varieties have not been previously reported in the UK. In part, because the morphology of hybrid and native species can be very similar making it difficult to differentiate between them without genomic or ionomic confirmation, as performed here.

Indeed, L. japonica was not easily differentiated from L. minor by morphology here (Fig. 1) and in Eastern Europe (Volkova et al., 2023). Among these hybrids, there is heterogeneity of parental introgression, putatively resulting in fitness performance attributable to parental gene variants in the context of different ecological backgrounds (Gompert and Buerkle, 2012). Similarly, L. japonica accessions exhibit variable propensities to form turions under inductive conditions, probably in line with receiving varied allelic contributions from L. minor (non-turionating) or L. turionifera (turionating) parents (Ernst et al., 2025).

Hybrid L. japonica appears more successful than its invasive parent L. turionifera, which showed limited geographic presence in this UK survey (Fig. 3B) and has only been reported twice previously, also localized in the east England region (Lansdown, 2008). In the case of L. japonica, hybridization has afforded wider spread than one of the parental genotypes (Volkova et al., 2023). Hybridization thus may occur to access wider adaptative potential of genetic variants from native species. Supporting this, we find evidence for hybrid and parental differences in both water elemental niches and nutrient uptake.

Species show differences in ionomes and native water chemistry

Hybridization is a proposed mechanism to generate ionome variation in plants and aid their adaptation to the water environment (Chen et al., 2017; Wang et al., 2021). Hybridisation also increases vigour in plants by increasing allelic diversity, especially in stressful environments (Washburn and Birchler, 2014). Lemna japonica reportedly form both diploid and triploids, and contain more transposable elements compared to parental species, hinting at possible increased capacity for environmental adaptation (Tran et al., 2022; Ernst et al., 2025). There is also experimental support for increased adaptation to high light irradiance in L. japonica, relative to both parental species (Smith et al., 2024a). Hybrids and parental species showed some differences in elemental composition between their water environments too, in line with possible adaptive speciation to specialized environmental niches.

Here, L. japonica water sites were contaminated with more As than L. minor sites and included higher concentrations of Mg, P, S, Ca, Mo, B and Sr. Many of these elements (Mg, S, Ca, Sr) were typically high in BRI-region water bodies (Fig. S6), linking species distribution with adaptation to a differential water environmental niche in the southwest. These elemental differences may point to transgressive phenotypes, which can provide an evolutionary advantage to hybrid species in challenging elemental environments. Furthermore, transgressive phenotypes can arise commonly in interspecific hybrid plants to give them higher abiotic tolerance during niche establishment, to enable divergence away from competition from parental species (Rieseberg et al., 1999). Moreover, there are examples of hybrids which show improved NaCl and Cd tolerance from that of their parents in other plant species (Xue et al., 2021; Ortega-Albero et al., 2023).

In this study, invasive L. minuta was found in higher Mg-containing sites than native L. minor, accumulating more Mg than other Lemna species in common conditions. This finding provides affirmation that this species is a high Mg-tolerator, supporting both invasive behaviour in foreign environments, possible tolerance to hardwater areas and enhanced potential for phytoremediation of high Mg-containing wastewater (Paolacci et al., 2016; Ceschin et al., 2020; Walsh et al., 2020). Higher Mg in L. minuta is consistent with enhanced Mg accumulation occurring within species within the ‘Uninerves’ section of Lemna (Smith et al., 2024b) but for the first time this work links accumulation to higher Mg tolerance in native habitats.

From the UK panel, L. japonica shows higher K tissue concentration compared to parent L. minor, which is consistent with findings from a worldwide duckweed ionome comparison (Smith et al., 2024b). As K is widely available, provided at the highest concentrations in water sites found here and all accessions accumulated it, it can be inferred that increased accumulation of K has a functional purpose in L. japonica, and may enhance tolerance to other elements, such as As to allow niche environment establishment. Some support comes from Arabidopsis thaliana and Vicia faba (Broad bean), whereby higher K uptake mitigated As and NaCl toxicity (Chao et al., 2013; Che et al., 2022; Shah et al., 2022).

Duckweed accumulator accessions and their potential applications

Using native water data, duckweed tolerance to macronutrients in the environment could be defined at a regional-scale, further highlighting specific UK accessions with tolerance traits for phytoremediation. Here, UK native water sites showed 81%, 74% and 44% higher maximum values for Mg, Ca and K than previously reported (Linton and Goulder, 1998). Additionally the maximum concentrations of K, Mg, Ca, Mn and Fe in native water exceeded concentrations of these elements tolerated by L. minor grown on dairy wastewater (O’Mahoney et al., 2022). Thus, inhabiting accessions may have developed useful tolerance and accumulating traits for phytoremediation and enhanced supply of macronutrients for nutrition.

That said, higher accumulators did not tend to come from sites with higher elemental concentrations (accession BOG), nor did higher accumulators necessarily come from contaminated UK environments (accession ALL1). Therefore, there is some degree of unlinking of elemental tolerance from accumulation potential. The accession GLA-Lmu-BOG showed enhanced concentrations of Ba, Pb, Al, Ti, Zn and Cd (Fig. 3D, Table 1) and can be considered for remediation of contaminated water courses, provided the tendency to accumulate can be tested on real-world water conditions. For example, Zn and Pb pollution in GLA waterbodies require remediation and the accession is already inhabiting the region (Fordyce et al., 2019; Eschenfelder et al., 2023). Unfortunately, accumulation of several macronutrients including B and Fe co-occurred with heavy metal uptake, even in low level controlled conditions. Whilst willingness to uptake heavy metals is an optimal trait for phytoremediation, this is problematic for direct applications in nutrition. Thus, mechanisms to retain high macronutrients but mitigate heavy metal levels such as inoculation with a synthetic microbiome or post-harvest washing or cooking steps may be future directed targets for duckweed consumption (Stout et al., 2010; Sattar et al., 2015; Amir et al., 2019).

Conclusion

This collection is the first of its nature, presenting a unique mixture of large-scale duckweed genomics, ionomics, and species assessment with environmental water data. Hybrid and invasive species were associated with novel water chemistry niches compared to parental species but were more restricted in their ranges. Therefore, high water hardness may be a predictor for future L. minuta colonization in new regions as well as highlighting their potential in bioremediation. It is possible that invasive species are not currently establishing well in north Scotland but their introduction should be mitigated. Heavy metal accumulating accessions identified from this study (BOG, LAN and others) should be further explored for phytoremediation potential using outside transplantation experiments and in-vitro elemental spiking experiments to maximize hyperaccumulation. This UK collection serves as a useful resource to explore desired traits for human consumption, bioremediation or ecological population studies and promotes the further genetic understanding of hybrid and parental duckweed species.

Supplementary Material

plaf018_suppl_Supplementary_Material
plaf018_suppl_supplementary_material.pdf (7.2MB, pdf)

Acknowledgements

Thanks to Tuomas Hämälä for PCA and pruning by linkage disequilibrium scripts. Thanks to Ana Da Silva for NJ tree script. Thanks to Claire Smith and Bethany Severn for harvesting DNA from some of the accessions. Thanks to Chris Dalton for aiding with collections and to Laura Briers for reading and revision of the manuscript.

Contributor Information

Kellie E Smith, Division of Plant and Crop Sciences, School of Biosciences, University of Nottingham, Sutton Bonington, Loughborough LE12 5RD, UK; School of Life Sciences, University of Nottingham, Nottingham NG7 2RD, UK.

Laura Cowan, School of Life Sciences, University of Nottingham, Nottingham NG7 2RD, UK.

Paulina Flis, Future Food Beacon, School of Biosciences, University of Nottingham, Sutton Bonington, Loughborough LE12 5RD, UK.

Chris Moore, Deepseq DNA Sequencing Laboratory, School of Life Sciences, Queens Medical Centre, Nottingham NG7 2UH, UK.

Matthew Heatley, School of Life Sciences, University of Nottingham, Nottingham NG7 2RD, UK.

Carlos A Robles-Zazueta, Division of Plant and Crop Sciences, School of Biosciences, University of Nottingham, Sutton Bonington, Loughborough LE12 5RD, UK.

Adam Lee, Division of Microbiology, Brewing and Biotechnology, School of Biosciences, University of Nottingham, Sutton Bonington, Loughborough LE12 5RD, UK.

Levi Yant, School of Life Sciences, University of Nottingham, Nottingham NG7 2RD, UK.

Funding

This work was supported by a Biotechnology and Biological Sciences Research Council (BBSRC) PhD scholarship BB/M008770/1 and the University of Nottingham Future Food Beacon of Excellence for K.E.S. and L.Y. The authors wish to thank the European Research Council (ERC) grant no. 679056 attributed to L.Y.

Author contributions

All authors approved the final manuscript. Kellie E. Smith (Conceptualization, Data curation, Methodology, Validation, Formal analysis, Investigation, Visualization, Writing—original draft, Writing—review & editing, Funding acquisition, Administration, Resources), Laura Cowan (Investigation), Paulina Flis (Investigation), Chris Moore (Investigation), Matthew Heatley (Software, Resources, Data curation), Carlos A. Robles-Zazueta (Resources, Writing—review and editing), Adam Lee (Resources, Writing—review and editing), and Levi Yant (Conceptualization, Data curation, Methodology, Visualization, Writing—original draft, Writing—review and editing, Funding acquisition, Supervision, Administration).

Conflict of interest

The authors declare no competing interests.

Data availability

Sequence data that support the findings of this study have been deposited in the Sequence Read Archive (SRA; https://www.ncbi.nlm.nih.gov/sra) with the primary accession code PRJNA1030266 (available at http://www.ncbi.nlm.nih.gov/bioproject/PRJNA1030266). Scripts for statistical and graphical outputs are available at https://github.com/Duckweed-KS/UK_genomics_ionomics

References

  1. Amir  RM, Randhawa  MA, Sajid  MW  et al.  Evaluation of various soaking agents as a novel tool for heavy metal residues mitigation from spinach. Food Sci Technol  2019;39:176–80. 10.1590/fst.00118 [DOI] [Google Scholar]
  2. Antro  MV, Prelovsek  S, Ivanovic  S  et al.  DNA methylation in clonal Duckweed lineages (Lemna minor L.) reflects current and historical environmental exposures. Mol Ecol  2022;32:428–43. 10.1111/mec.16757 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Appenroth  KJ, Teller  S, Horn  M. Photophysiology of turion formation and germination in Spirodela polyrhiza. Biol Plant  1996;38:95–106. 10.1007/BF02879642 [DOI] [Google Scholar]
  4. Appenroth  KJ, Sree  KS, Böhm  V  et al.  Nutritional value of duckweeds (Lemnaceae) as human food. Food Chem  2017;217:266–73. 10.1016/j.foodchem.2016.08.116 [DOI] [PubMed] [Google Scholar]
  5. Barton  KE.  The ontogenetic dimension of plant functional ecology. Funct Ecol  2024;38:98–113. 10.1111/1365-2435.14464 [DOI] [Google Scholar]
  6. Bergmann  BA, Cheng  J, Classen  J  et al.  In vitro selection of duckweed geographical isolates for potential use in swine lagoon effluent renovation. Bioresour Technol  2000;73:13–20. 10.1016/s0960-8524(99)00137-6 [DOI] [Google Scholar]
  7. Bog  M, Sree  KS, Fuchs  J  et al.  A taxonomic revision of Lemna sect. Uninerves (Lemnaceae). Taxon  2020;69:56–66. 10.1002/tax.12188 [DOI] [Google Scholar]
  8. Bolger  AM, Lohse  M, Usadel  B.  Trimmomatic: A flexible trimmer for Illumina sequence data. Bioinformatics  2014;30:2114–20. 10.1093/bioinformatics/btu170 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Braglia  L, Breviario  D, Gianì  S  et al.  New insights into interspecific hybridization in Lemna l. Sect. lemna (lemnaceae martinov). Plants  2021;10:2767. 10.3390/plants10122767 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Braglia  L, Ceschin  S, Iannelli  MA  et al.  Characterization of the cryptic interspecific hybrid Lemna×mediterranea by an integrated approach provides new insights into duckweed diversity. J Exp Bot  2024;75:3092–110. 10.1093/jxb/erae059 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Ceschin  S, Abati  S, Leacche  I  et al.  Does the alien Lemna minuta show an invasive behavior outside its original range? Evidence of antagonism with the native L. minor in central Italy. Int Rev Hydrobiol  2016;101:173–81. 10.1002/iroh.201601841 [DOI] [Google Scholar]
  12. Ceschin  S, Abati  S, Ellwood  NTW, Zuccarello  V. Riding invasion waves: Spatial and temporal patterns of the invasive Lemna minuta from its arrival to its spread across Europe. Aquat Bot  2018a;150:1–8. 10.1016/j.aquabot.2018.06.002 [DOI] [Google Scholar]
  13. Ceschin  S, Abati  S, Leacche  I, Zuccarello  V. Ecological comparison between duckweeds in central Italy: The invasive Lemna minuta vs the native L. minor. Plant Biosyst  2018b;152:674–83. 10.1080/11263504.2017.1317671 [DOI] [Google Scholar]
  14. Ceschin  S, Abati  S, Traversetti  L  et al.  Effects of the invasive duckweed Lemna minuta on aquatic animals: evidence from an indoor experiment. Plant Biosyst  2019;153:749–55. 10.1080/11263504.2018.1549605 [DOI] [Google Scholar]
  15. Ceschin  S, Crescenzi  M, Iannelli  MA.  Phytoremediation potential of the duckweeds Lemna minuta and Lemna minor to remove nutrients from treated waters. Environ Sci Pollut Res  2020;27:15806–14. 10.1007/s11356-020-08045-3 [DOI] [PubMed] [Google Scholar]
  16. Chao  D, Dilkes  B, Luo  H  et al.  Polyploids exhibit higher Potassium uptake and salinity tolerance in Arabidopsis. Science (New York, NY)  2013;341:658–9. 10.1126/science.1240561 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Chang  CC, Chow  CC, Tellier  LCAM, Vattikuti  S, Purcell  SM, Lee  JJ. Second-generation PLINK: Rising to the challenge of larger and richer datasets. Gigascience  2015;4:1–16. 10.1186/s13742-015-0047-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Che  Y, Fan  D, Wang  Z  et al.  Potassium mitigates salt-stress impacts on photosynthesis by alleviation of the proton diffusion potential in thylakoids. Environ Exp Bot  2022;194:104708. 10.1016/j.envexpbot.2021.104708 [DOI] [Google Scholar]
  19. Chen  Z, Taylor  AA, Astor  SR  et al.  Removal of boron from wastewater: Evaluation of seven poplar clones for B accumulation and tolerance. Chemosphere  2017;167:146–54. 10.1016/j.chemosphere.2016.09.137 [DOI] [PubMed] [Google Scholar]
  20. Chen  G, Stepanenko  A, Lakhneko  O  et al.  Biodiversity of duckweed (Lemnaceae) in water reservoirs of Ukraine and China assessed by chloroplast DNA barcoding. Plants  2022;11:1468–14. 10.3390/plants11111468 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Coughlan  NE, Kelly  TC, Jansen  MAK. Mallard duck (Anas platyrhynchos)-mediated dispersal of Lemnaceae: A contributing factor in the spread of invasive Lemna minuta? Plant Biol  2015;17:108–114. 10.1111/plb.12182 [DOI] [PubMed] [Google Scholar]
  22. Ekperusi  AO, Sikoki  FD, Nwachukwu  EO.  Application of common duckweed (Lemna minor) in phytoremediation of chemicals in the environment: State and future perspective. Chemosphere  2019;223:285–309. 10.1016/j.chemosphere.2019.02.025 [DOI] [PubMed] [Google Scholar]
  23. Eschenfelder  J, Lipp  AG, Roberts  GG.  Quantifying excess heavy metal concentrations in drainage basins using conservative mixing models. J Geochem Explor  2023;248:107178. 10.1016/j.gexplo.2023.107178 [DOI] [Google Scholar]
  24. Ernst  E, Abramson  B, Acosta  K, Hoang  PTN, Mateo-elizalde  C, Schubert  V, Lam  E, Michael  TP, Martienssen  RA. Duckweed genomes and epigenomes underlie triploid hybridization and clonal reproduction. Curr Biol  2025;35:1–20. 10.1016/j.cub.2025.03.013 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Feller  J, Taylor  M, Henry  P. Predicting Lemna growth based on climate change and eutrophication in temperate freshwater drainage ditches. Hydrobiologia. 2024;851:2529–41. 10.1007/s10750-024-05477-7 [DOI] [Google Scholar]
  26. Fedoniuk  T, Bog  M, Orlov  O  et al.  Lemna aequinoctialis migrates further into temperate continental Europe—A new alien aquatic plant for Ukraine. Feddes Repert  2022;133:305–12. 10.1002/fedr.202200001 [DOI] [Google Scholar]
  27. Fordyce  FM, Everett  PA, Bearcock  JM  et al.  The geosciences in Europe’s urban sustainability: lessons from Glasgow and beyond (CUSP) Soil metal/metalloid concentrations in the Clyde Basin, Scotland, UK: implications for land quality. Earth Environ Sci Trans R Soc Edinburgh  2019;108:191–216. [Google Scholar]
  28. Friedjung Yosef  A, Ghazaryan  L, Klamann  L  et al.  Diversity and differentiation of duckweed species from Israel. Plants (Basel)  2022;11:3326. 10.3390/plants11233326 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Gérard  J, Triest  L. Competition between invasive Lemna minuta and native L. minor in indoor and field experiments. Hydrobiologia  2018;812:57–65. 10.1007/s10750-016-2754-2 [DOI] [Google Scholar]
  30. Gómez-Rubio  V. ggplot2 - Elegant Graphics for Data Analysis (2nd Edition). J Stat Softw  2017;77:3–5. 10.18637/jss.v077.b02 [DOI] [Google Scholar]
  31. Gompert  Z, Buerkle  CA. bgc: Software for Bayesian estimation of genomic clines. Mol Ecol Resour. 2012;12:1168–76. 10.1111/1755-0998.12009.x [DOI] [PubMed] [Google Scholar]
  32. Global Biodiversity Information Facility. (GBIF)  Lemna L. in GBIF Secretariat (2021). GBIF Backbone Taxonomy. 2021. United Kingdom of Great Britain and Northern Ireland. https://www.gbif.org/ [Google Scholar]
  33. Hämälä  T, Moore  C, Cowan  L  et al.  Impact of whole-genome duplications on structural variant evolution in the plant genus Cochlearia. Nat C  2024;15:1–13. 10.1038/s41467-024-49679-y [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Huber  M, Gablenz  S, Höfer  M.  Transgenerational non-genetic inheritance has fitness costs and benefits under recurring stress in the clonal duckweed Spirodela polyrhiza. Proc R Soc B Biol Sci  2021;288:20211269. 10.1098/rspb.2021.1269 [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Janes  RA, Eaton  JW, Hardwick  K.  The effects of floating mats of Azolla filiculoides Lam. and Lemna minuta Kunth on the growth of submerged macrophytes. Hydrobiologia  1996;340:23–6. 10.1007/bf00012729 [DOI] [Google Scholar]
  36. Jewell  MD, Bell  G.  Environmental and genetic variation in an asexual plant. Aquat Bot  2023;188:103675. 10.1016/j.aquabot.2023.103675 [DOI] [Google Scholar]
  37. Kadono  Y, Iida  S.  Identification of a small, spring water-associated duckweed with special reference to the taxonomy of sect. uninerves of the genus Lemna (Lemnaceae) in Japan. Acta Phytotaxon Geobot  2022;73:57–65. 10.18942/apg.202112 [DOI] [Google Scholar]
  38. Kirjakov  IK, Velichkova  KN.  Invasive species Lemna L. (Lemnaceae) in the flora of Bulgaria. Period Biol  2016;118:131–8. 10.18054/pb.2016.118.2.4165 [DOI] [Google Scholar]
  39. Kurihara  D, Mizuta  Y, Sato  Y  et al.  ClearSee: A rapid optical clearing reagent for whole-plant fluorescence imaging. Development (Cambridge, England)  2015;142:4168–79. 10.1242/dev.127613 [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Laird  RA, Barks  PM.  Skimming the surface: duckweed as a model system in ecology and evolution. Am J Bot  2018;105:1962–6. 10.1002/ajb2.1194 [DOI] [PubMed] [Google Scholar]
  41. Lam, E.  Rutgers Duckweed Stock Cooperative. 2018. http://www.ruduckweed.org (13 May 2020, date last accessed). [Google Scholar]
  42. Lam  E, Michael  TP.  Wolffia, a minimalist plant and synthetic biology chassis. Trends Plant Sci  2022;27:430–9. 10.1016/j.tplants.2021.11.014 [DOI] [PubMed] [Google Scholar]
  43. Landesman  L, Fedler  C, Duan  R.  Plant nutrient phytoremediation using duckweed. In: Ansari  A, Singh Gill  S, Lanza  G, Rast  W (eds.), Eutrophication: Causes, Consequences and Control, 2010, 341–354. Dordrecht: Springer. 10.1007/978-90-481-9625-8_17 [DOI] [Google Scholar]
  44. Landolt  E.  Biosystematic investigations in the family of duckweeds (Lemnaceae). In: Landolt  E (ed.). Vol. 2. Zürich: Geobotanisches Institut der Eidgenossiesche Technische Hochschule. 1986, 466–470. [Google Scholar]
  45. Lansdown  RV.  Red duckweed (Lemna turionifera Landolt) new to Britain. Watsonia  2008;27:127–30. [Google Scholar]
  46. Li  H, Durbin  R. Fast and accurate short read alignment with Burrows-Wheeler transform. Bioinformatics. 2009;25:1754–60. 10.1093/bioinformatics/btp324 [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Li  H, Handsaker  B, Wysoker  A, Fennell  T, Ruan  J, Homer  N, Marth  G, Abecasis  G, Durbin  R. The Sequence Alignment/Map format and SAMtools. Bioinformatics. 2009;25:2078–79. 10.1093/bioinformatics/btp352 [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Linton  S, Goulder  R.  The duckweed Lemna minor compared with the alga Selenastrum capricornutum for bioassay of pond-water richness. Aquat Bot  1998;60:27–36. 10.1016/s0304-3770(97)00067-3 [DOI] [Google Scholar]
  49. McKenna  A, Hanna  M, Banks  E  et al.  The genome analysis toolkit: a MapReduce framework for analyzing next-generation DNA sequencing data. Genome Res  2010;20:1297–303. 10.1101/gr.107524.110 [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Njambuya  J, Stiers  I, Triest  L.  Competition between Lemna minuta and Lemna minor at different nutrient concentrations. Aquat Bot  2011;94:158–64. 10.1016/j.aquabot.2011.02.001 [DOI] [Google Scholar]
  51. O’Mahoney  R, Coughlan  NE, Walsh  E  et al.  Cultivation of Lemna minor on industry-derived, anaerobically digested, dairy processing wastewater. Plants  2022;11:3027–13. 10.3390/plants11223027 [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Ortega-Albero  N, González-Orenga  S, Vicente  O  et al.  Responses to salt stress of the interspecific hybrid Solanum insanum × Solanum melongena and its parental species. Plants (Basel)  2023;12:295. 10.3390/plants12020295 [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Paolacci  S, Harrison  S, Jansen  MAK.  A comparative study of the nutrient responses of the invasive duckweed Lemna minuta, and the native, co-generic species Lemna minor. Aquat Bot  2016;134:47–53. 10.1016/j.aquabot.2016.07.004 [DOI] [Google Scholar]
  54. Paolacci  S, Harrison  S, Jansen  MAK.  The invasive duckweed Lemna minuta Kunth displays a different light utilisation strategy than native Lemna minor Linnaeus. Aquat Bot  2018a;146:8–14. 10.1016/j.aquabot.2018.01.002 [DOI] [Google Scholar]
  55. Paolacci  S, Jansen  MAK, Harrison  S.  Competition between Lemna minuta, Lemna minor, and Azolla filiculoides. Growing fast or being steadfast? Front Chem  2018b;6:1–15. 10.3389/fchem.2018.00207 [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Paradis  E, Schliep  K. Ape 5.0: An environment for modern phylogenetics and evolutionary analyses in R. Bioinformatics. 2019;35:526–8. 10.1093/bioinformatics/bty633 [DOI] [PubMed] [Google Scholar]
  57. Peeters  ETHM, van Zuidam  JP, van Zuidam  BG, Van Nes  EH, Kosten  S, Heuts  PGM, Roijackers  RMM, Netten  JJC, Scheffer  M. Changing weather conditions and floating plants in temperate drainage ditches. J Appl Ecol  2013;50:585–93. 10.1111/1365-2664.12066 [DOI] [Google Scholar]
  58. Raj  A, Stephens  M, Pritchard  JK. FastSTRUCTURE: Variational inference of population structure in large SNP data sets. Genetics  2014;197:573–89. 10.1534/genetics.114.164350 [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Reynolds  MP, Braun  HJ.  Chapter 1: Wheat Improvement. In: Reynolds  MP, Braun  HJ (eds.). Wheat Improvement: Food Security in a Changing Climate. Cham: Springer, 2022. 3-15. 10.1007/978-3-030-90673-3_1 [DOI] [Google Scholar]
  60. Rieseberg  LH, Archer  MA, Wayne  RK. Transgressive segregation, adaptation and speciation. Heredity  1999;83:363–72. 10.1038/sj.hdy.6886170 [DOI] [PubMed] [Google Scholar]
  61. Roubeau Dumont  E, Larue  C, Pujol  B  et al.  Environmental variations mediate duckweed (Lemna minor L.) sensitivity to copper exposure through phenotypic plasticity. Environ Sci Pollut Res  2019;26:14106–15. 10.1007/s11356-019-04630-3 [DOI] [PubMed] [Google Scholar]
  62. Salt  DE, Baxter  I, Lahner  B.  Ionomics and the study of the plant ionome. Annu Rev Plant Biol  2008;59:709–33. 10.1146/annurev.arplant.59.032607.092942 [DOI] [PubMed] [Google Scholar]
  63. Sattar  MU, Anjum  FM, Sameen  A.  Mitigation of heavy metals in different vegetables through biological washing techniques. Int J Food Allied Sci  2015;1:40. 10.21620/ijfaas.2015240-44 [DOI] [Google Scholar]
  64. Schindelin  J, Arganda-Carreras  I, Frise  E  et al.  Fiji: an open-source platform for biological-image analysis. Nat Methods  2012;9:676–82. 10.1038/nmeth.2019 [DOI] [PMC free article] [PubMed] [Google Scholar]
  65. Shah  AA, Ahmed  S, Malik  A  et al.  Potassium silicate and zinc oxide nanoparticles modulate antioxidant system, membranous H+-ATPase and nitric oxide content in faba bean (Vicia faba) seedlings exposed to arsenic toxicity. Funct Plant Biol  2022;50:146–59. 10.1071/FP21301 [DOI] [PubMed] [Google Scholar]
  66. Smith  KE, Cowan  L, Taylor  B  et al.  Physiological adaptation to high irradiance in duckweeds depends on light habitat niche and is ecotype and species-specific. J Exp Bot  2024a;75:2046–63. 10.1093/jxb/erad499 [DOI] [PMC free article] [PubMed] [Google Scholar]
  67. Smith  KE, Zhou  M, Flis  P  et al.  The evolution of the duckweed ionome mirrors losses in structural complexity. Ann. Bot.  2024b;133:997–1006. 10.1093/aob/mcae012 [DOI] [PMC free article] [PubMed] [Google Scholar]
  68. Silva  GG, Green  AJ, Weber  V, Hoffmann  P, Lovas-Kiss Stenert  C, Maltchik  L. Whole angiosperms Wolffia columbiana disperse by gut passage through wildfowl in South America. Biol Lett  2018;14:1–4. 10.1098/rsbl.2018.0703 [DOI] [PMC free article] [PubMed] [Google Scholar]
  69. Sree  KS, Appenroth  KJ.  Worldwide genetic resources of duckweed: stock collections. In: Cao  XH, Fourounjian  P, Wang  W (eds.), The Duckweed Genomes. Cham, Switzerland: Springer International Publishing, 2020, 39–46. 10.1007/978-3-030-11045-1_3 [DOI] [Google Scholar]
  70. Sree  KS, Sudakaran  S, Appenroth  KJ.  How fast can angiosperms grow? Species and clonal diversity of growth rates in the genus Wolffia (Lemnaceae). Acta Physiol Plant  2015;37:204. 1–7. 10.1007/s11738-015-1951-3 [DOI] [Google Scholar]
  71. Stomp  A-M.  The duckweeds: A valuable plant for biomanufacturing. Biotechnol Annu Rev  2005;11:69–99. 10.1016/S1387-2656(05)11002-3 [DOI] [PubMed] [Google Scholar]
  72. Stout  LM, Dodova  EN, Tyson  JF  et al.  Phytoprotective influence of bacteria on growth and cadmium accumulation in the aquatic plant Lemna minor. Water Res  2010;44:4970–9. 10.1016/j.watres.2010.07.073 [DOI] [PubMed] [Google Scholar]
  73. Strzałek  M, Kufel  L.  Light intensity drives different growth strategies in two duckweed species: Lemna minor L. and Spirodela polyrhiza (L.) Schleiden. PeerJ  2021;9:1–14. 10.7717/peerj.12698 [DOI] [PMC free article] [PubMed] [Google Scholar]
  74. Taghipour  E, Bog  M, Frootan  F  et al.  DNA barcoding and biomass accumulation rates of native Iranian duckweed species for biotechnological applications. Front Plant Sci  2022;13:1–15. 10.3389/fpls.2022.1034238 [DOI] [PMC free article] [PubMed] [Google Scholar]
  75. Tran  NBT, Tran  TN, Hoang  TNP.  Morphological variation, chromosome number, and DNA barcoding of giant duckweed (Spirodela polyrhiza) in Vietnam. Can Tho Univ J Sci  2022;14:61–7. 10.22144/ctu.jen.2022.029 [DOI] [Google Scholar]
  76. Vámos  S, Li  C, Moradi  A  et al.  High tolerance to zinc but no evidence for local adaptation in the aquatic plant Lemna minor. Nord J Bot  2023;2024:1–12. 10.1111/njb.04078 [DOI] [Google Scholar]
  77. Van Echelpoel  W, Boets  P, Goethals  PLM.  Functional response (FR) and relative growth rate (RGR) do not show the known invasiveness of Lemna minuta (Kunth). PLoS One  2016;11:1–18. 10.1371/journal.pone.0166132 [DOI] [PMC free article] [PubMed] [Google Scholar]
  78. Volkova  PA, Nachatoi  VA, Bobrov  AA.  Hybrid between Lemna minor and L. turionifera (L. × japonica, Lemnaceae) in East Europe is more frequent than parental species and poorly distinguishable from them. Aquat Bot  2023;184:103593. 10.1016/j.aquabot.2022.103593 [DOI] [Google Scholar]
  79. Walsh  E, Paolacci  S, Burnell  G  et al.  The importance of the calcium-to-magnesium ratio for phytoremediation of dairy industry wastewater using the aquatic plant Lemna minor L. Int J Phytoremediation  2020;22:694–702. 10.1080/15226514.2019.1707478 [DOI] [PubMed] [Google Scholar]
  80. Wang  Z, Jiang  Y, Bi  H  et al.  Hybrid speciation via inheritance of alternate alleles of parental isolating genes. Mol Plant  2021;14:208–22. 10.1016/j.molp.2020.11.008 [DOI] [PubMed] [Google Scholar]
  81. Ware  A, Jones  DH, Flis  P  et al.  Loss of ancestral function in duckweed roots is accompanied by progressive anatomical simplification and a re-distribution of nutrient transporters. Curr Biol  2023;33:1795–802.e4. 10.1016/j.cub.2023.03.025 [DOI] [PubMed] [Google Scholar]
  82. Washburn  JD, Birchler  JA.  Polyploids as a ‘model system’ for the study of heterosis. Plant Reprod  2014;27:1–5. 10.1007/s00497-013-0237-4 [DOI] [PubMed] [Google Scholar]
  83. Xu  Y, Ma  S, Huang  M  et al.  Species distribution, genetic diversity and barcoding in the duckweed family (Lemnaceae). Hydrobiologia  2015;743:75–87. 10.1007/s10750-014-2014-2 [DOI] [Google Scholar]
  84. Xu  J, Shen  Y, Zheng  Y  et al.  Duckweed (Lemnaceae) for potentially nutritious human food: A review. Food Rev Int  2023;39:3620–34. 10.1080/87559129.2021.2012800 [DOI] [Google Scholar]
  85. Xue  C, Gao  Y, Qu  B  et al.  Hybridization with an invasive plant of Xanthium strumarium improves the tolerance of its native congener X. sibiricum to Cadmium. Front Plant Sci  2021;12:1–14. 10.3389/fpls.2021.696687 [DOI] [PMC free article] [PubMed] [Google Scholar]
  86. Zayed  A, Gowthaman  S, Terry  N.  Phytoaccumulation of trace elements by wetland plants: I. Duckweed. J Environ Qual  1998;27:715–21. 10.2134/jeq1998.00472425002700030032x [DOI] [Google Scholar]
  87. Zhang  X, Zhao  FJ, Huang  Q  et al.  Arsenic uptake and speciation in the rootless duckweed Wolffia globosa. New Phytol  2009;182:421–8. 10.1111/j.1469-8137.2008.02758.x [DOI] [PubMed] [Google Scholar]
  88. Zhang  H, Mittal  N, Leamy  LJ  et al.  Back into the wild—apply untapped genetic diversity of wild relatives for crop improvement. Evol Appl  2017;10:5–24. 10.1111/eva.12434 [DOI] [PMC free article] [PubMed] [Google Scholar]
  89. Ziegler  P, Adelmann  K, Zimmer  S  et al.  Relative in vitro growth rates of duckweeds (Lemnaceae) - the most rapidly growing higher plants. Plant Biol (Stuttg)  2015;17:33–41. 10.1111/plb.12184 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

plaf018_suppl_Supplementary_Material
plaf018_suppl_supplementary_material.pdf (7.2MB, pdf)

Data Availability Statement

Sequence data that support the findings of this study have been deposited in the Sequence Read Archive (SRA; https://www.ncbi.nlm.nih.gov/sra) with the primary accession code PRJNA1030266 (available at http://www.ncbi.nlm.nih.gov/bioproject/PRJNA1030266). Scripts for statistical and graphical outputs are available at https://github.com/Duckweed-KS/UK_genomics_ionomics

Articles from AoB Plants are provided here courtesy of Oxford University Press

RESOURCES