Portal de Periódicos da CAPES

Skimming the surface: duckweed as a model system in ecology and evolution

2018; Wiley; Volume: 105; Issue: 12 Linguagem: Inglês

10.1002/ajb2.1194

1537-2197

Robert A. Laird, Patrick Barks,

Botany and Plant Ecology Studies

With almost 400,000 species to choose from, plant biologists seeking a study species have no shortage of raw material. In the face of such diversity, plant model systems have come to the fore in many disciplines, with Arabidopsis thaliana being the most conspicuous example. The main benefits of model systems are that they (1) are amenable to the collection of data relevant to a variety of research questions (i.e., due to intrinsic species traits such as small stature and short generation time) and (2) come with value-added properties by virtue of their history of use in research, such as well-developed experimental protocols and molecular tools, a deep and broad literature, and an international community of experts (Fox, 2012; Chang et al., 2016). While the wisdom of relying on a small number of model systems is a matter of debate, there is little doubt that working on easily studied and well understood plant species has paid off in many disciplines. Here we highlight the potential of one particular group—duckweeds—as a model system for research in the disciplines of ecology and evolution. Duckweeds are floating or submergent aquatic monocots that comprise the "simplest and smallest of flowering plants" (Hillman, 1961). Their small stature and morphological simplicity are just two of the many traits that make duckweeds well suited for addressing a variety of questions in ecology and evolution (Tables 1, 2). Furthermore, because of a long history of use in fields such as ecotoxicology (e.g., Wang, 1990) and plant development (e.g., Hillman, 1976), and ongoing interest in using duckweed for industrial applications such as feed and biofuel production (e.g., Cheng and Stomp, 2009) and bioremediation (e.g., Ziegler et al., 2016), researchers studying duckweed today benefit from a mature literature, established research procedures, repositories with hundreds of live strains, and ample molecular resources including genome sequences for Spirodela polyrhiza (Wang et al., 2014) and Lemna minor (Van Hoeck et al., 2015). Before exploring the potential of duckweeds in ecology and evolution, we briefly review their natural history (based on Landolt, 1986). Individual duckweed ramets consist of a single "frond" (also called a thallus) with between zero and several roots emanating from the lower surface, depending on the species. Fronds are tiny, with surface areas on the order of 1 mm2 to 1 cm2. Though duckweeds can flower, the vast majority of their reproduction is asexual, via meristematic pockets from which clonal daughters successively bud and detach. An individual frond may produce up to a couple dozen daughters over its life, which is typically on the order of weeks. Taxonomically, the five genera and 37 species of duckweed form a monophyletic subfamily, Lemnoideae, within the family Araceae (though some authors prefer to put duckweeds into their own separate family, Lemnaceae; Sree et al., 2016). They are found in lentic and slow-moving freshwater systems around the world and are particularly successful in systems subject to cultural eutrophication. The duckweed traits that are most broadly useful in ecology and evolution are an extremely short lifespan and rapid rate of asexual growth. Duckweeds are ranked among the fastest growing and most productive higher plants (e.g., Ziegler et al., 2015), often completely blanketing the surface of their water body (Fig. 1). This rapid growth is part of what makes duckweed useful for industrial applications, but is also advantageous for fundamental research in ecology and evolution because it allows for a quick buildup of large sample sizes. Furthermore, the short lifespan of individual ramets makes it possible to track cohorts longitudinally and even multigenerationally, allowing in a matter of months the collection of data that would take years to obtain even in annual plants. For example, the relative ease of tracking cohorts over multiple generations has made duckweeds ideal for studying parental age effects in the contexts of both senescence (Ashby and Wangermann, 1949; Barks and Laird, 2015) and bet hedging (Mejbel and Simons, 2018). Despite the dominance of asexual reproduction, populations of duckweed maintain relatively high levels of genetic diversity. For instance, Vasseur et al. (1993) sampled 157 unique genotypes of L. minor among eight small ponds in Ontario, Canada, separated by a maximum distance of just 12 km. In many parts of the world, at least a few duckweed species are locally abundant, with multiple duckweed species often co-occurring in distinct communities (Landolt, 1986). This diversity at multiple biological levels and spatial scales makes duckweed a good candidate for research on competition, coexistence, and the maintenance of biodiversity (e.g., Vasseur et al., 1995), as well as distributional ecology and biogeography (e.g., Keddy, 1976). Another trait that makes duckweeds amenable to research in ecology and evolution is the ability to tolerate a wide range of conditions, as might be supposed based on their widespread occurrence, and use as bioremediation agents in contaminated sites. One particular growth environment that can be useful for research in ecology and evolution is axenic culture (i.e., free of microorganisms), which allows for precise control over environmental conditions. Obtaining axenic cultures from wild-collected duckweed is straightforward (Bowker et al., 1980), and we have ourselves exploited this property to ensure constant environmental conditions in common garden experiments (Barks et al., 2018). Axenic culture is also useful in the study of plant microbiomes, a topic currently of intense interest (e.g., Gilbert et al., 2018). Finally, under axenic conditions, duckweed cultures can be maintained at low temperature for up to a year without requiring human intervention, which allows for simple long-term maintenance of stock cultures. Of course, no model system is perfect for all questions, and some of duckweed's limitations must also be acknowledged. For instance, some duckweed species pairs are difficult to distinguish based on morphology alone; they can reliably be distinguished using laboratory manipulations or molecular identification techniques, but these approaches require time and money. In addition, because individual plants are small and typically free-floating, it can be challenging to mark and track individuals in a laboratory setting, let alone in the wild. That said, we think it will be possible to conduct individual-based studies in the wild using mark–recapture techniques or floating structures to limit plant movement—approaches we hope to pursue in the future. Furthermore, computer vision-based approaches are likely to become increasingly feasible for tracking and quantifying duckweed. Another potential limitation is that it may be difficult to generate de novo genetic diversity in duckweeds due to low rates of mutation and recombination (Xu et al., 2018 [preprint]); however, there have been successful efforts at flower induction and artificial cross pollination in some duckweeds (e.g., Fu et al., 2017). Finally, generalizing results based on duckweeds to other plant taxa must be done with caution, as some of the traits that make duckweeds easy to study (e.g., their size and anatomical simplicity; Table 1) also make them somewhat unusual. Relatedly, some of the functional traits on which plant ecologists commonly rely for comparative analyses are absent in duckweeds or at least difficult to compare with other plant taxa (e.g., height, rooting depth, phenology). These potential criticisms should be taken seriously when attempting to extrapolate results; however, we do not see them as a serious bar to using duckweeds in the first place. Notwithstanding the long history of research on duckweeds in ecology and evolution (e.g., Table 2), their use in these disciplines has only "skimmed the surface" of possible applications. Indeed, despite what we argue are clear advantages to studying duckweeds, the ecology and evolution research communities have never coalesced around them to the extent of other fields such as ecotoxicology and bioremediation, where they continue to play a major role (Ziegler et al., 2016). Why not? Perhaps practitioners of ecology and evolution tend to favor theory and study species in which the mode of reproduction is primarily sexual; because duckweeds are almost exclusively clonal, they may not be a natural fit within the culture of these research communities. We also speculate that there may be more general resistance to model systems, particularly in ecology, compared to, say, molecular genetics or developmental biology (Fox, 2012 made a similar argument). Biodiversity is justifiably of perennial interest in ecology. Moreover, it is almost axiomatic that variation per se is often of direct focus in both ecology and evolution. It may be that these features predispose scientists in these disciplines away from model species, especially those like duckweeds whose charisma is subtle. Regardless, we think there are many fundamental questions in ecology and evolution for which duckweeds would be an ideal study system, particularly those whose study requires large sample sizes, short generation times, and easy experimental manipulability. To close with three representative examples, we foresee duckweeds as being highly profitable in studies on (1) experimental plant evolution, (2) processes that straddle the line between evolutionary and ecological timescales (an avenue beginning to be explored by S. P. Hart, M. M. Turcotte, and J. M. Levine [unpublished manuscript]), and (3) experimental tests of the effects of complex competitive networks (e.g., intransitive "rock–paper–scissors" competition) on strain or species coexistence. We thank Lonnie Aarssen, Andrew Simons, Martin Turcotte, Suzanne Chmilar, Jake Hayden, and Austin Paiha for comments on earlier drafts. We thank Dave Armitage, an anonymous reviewer, and Editor-in-Chief Pamela Diggle for helpful reviews. We thank Martin Turcotte for providing the photograph in Fig. 1D (used with his permission) and S. P. Hart, M. M. Turcotte, and J. M. Levine for providing their unpublished manuscript. Funding was provided by the Natural Sciences and Engineering Research Council of Canada (grant no. RGPIN-2015-05486).