Reconstructing Trait Evolution in Plant Evo–Devo Studies

Current Biology Magazine

Essay Marchantia is genetically tractable [10]. The diversifi cation of model species Reconstructing trait evolution in plant offers unprecedented opportunities to explore fundamental biological evo–devo studies processes. Furthermore, comparative studies with other plant clades have Pierre-Marc Delaux1,*, Alexander J. Hetherington2, Yoan Coudert3, the potential to unravel evolutionary Charles Delwiche4, Christophe Dunand1, Sven Gould5, Paul Kenrick6, events that shaped the diversity of Fay-Wei Li7,8, Hervé Philippe9,10, Stefan A. Rensing11,12,13, Mélanie Rich1, extant plant species. The access to Christine Strullu-Derrien6,14, and Jan de Vries15,16,17 many plant genomes, transcriptomes and new model species therefore Our planet is teeming with an astounding diversity of plants. In a mere single makes it an exciting time for studying group of closely related species, tremendous diversity can be observed in their plant evolution. However, as the form and function — the colour of petals in fl owering plants, the shape of the number of model species expands fronds in ferns, and the branching pattern of the gametophyte in mosses. Diversity and diversifi es, it is important to pay can also be found in subtler traits, such as the resistance to pathogens or the close attention to the method used ability to recruit symbiotic microbes from the environment. Plant traits can also to draw evolutionary comparisons be highly conserved — at the cellular and metabolic levels, entire biosynthetic between species. This becomes pathways are present in all plant groups, and morphological characteristics particularly important when drawing such as vascular tissues have been conserved for hundreds of millions of years. conclusions with the benchmark of The research community that seeks to understand these traits — both the Arabidopsis, when the species last diverse and the conserved — by taking an evolutionary point-of-view on plant shared a common ancestor with this biology is growing. Here, we summarize a subset of the different aspects of Brassicaceae hundreds of millions of plant evolutionary biology, provide a guide for structuring comparative biology years ago. approaches and discuss the pitfalls that (plant) researchers should avoid when The aim of this article is to provide a embarking on such studies. guide for structuring the rationale that is used when employing comparative biology approaches. We start fi rst by Plants are extremely diverse, whether biology is growing. There are a number highlighting the importance of drawing this be in the range of petal colours of reasons for this. Foremost is the conclusions based on precisely in angiosperms [1], the shape of the availability of whole plant genomes and reconstructed species phylogenies, fronds in ferns [2], the branching transcriptomes covering the breadth of and go on to outline a 5-step guide pattern of the gametophyte in mosses the plant phylogeny. This goes hand- for structuring evolutionary studies, [3] or their interactions with microbes in-hand with the development of model illustrated with examples from the and the environment. To understand systems beyond the fl owering plant literature. this diversity, it is essential to explore Arabidopsis thaliana (L.) Heynh. Such the genetic framework underlying any prospering models include a huge Species phylogenies and tree of these traits in light of evolution. diversity of angiosperm species, from thinking Researchers studying the evolution trees (such as Populus trichocarpa), Evolutionary relationships between of traits aim at determining “which to a range of species to study fl ower organisms are best expressed genes and what kinds of changes development (e.g., Antirrhinum majus through phylogenetic trees. The use in their sequences are responsible and Aquilegia caerulea), together with of DNA sequences to reconstruct for the evolution of [morphological] enormous progress with gymnosperms, phylogenies, particularly multi-loci diversity” [4]. This way of approaching including the fi rst genome assemblies phylogenomics, has improved the diversity was initiated by developmental and studies of gene expression and resolution of the plant tree of life biologists, leading to the emergence conservation. Major advances have (Figure 1). It is now widely accepted of evolutionary developmental biology also been made with lycophytes or that the closest extant relatives of land (evo–devo) that was later expanded monilophytes, such as the model fern plants are the streptophyte algae in to all aspects of plant biology, such Ceratopteris richardii, and fi nally the the Zygnematophyceae class [11–13]. as the interactions between plants development of liverwort, moss as well Within the land plants, the relationship and microbes [5,6] (evo–MPMI, for as algal model systems. among the major lineages is relatively evolutionary molecular plant–microbe The liverwort Marchantia polymorpha well supported (Figure 1). Living land interactions) or the study of cellular is being established as a major model plants constitute two main groups, biology [7] (evo–cell biology). The in plant science, mainly because it vascular and non-vascular plants. terms are different, but these fi elds of has a short life cycle and a relatively Vascular plants (tracheophytes) form research rely on the same comparative small genome with fewer paralogs a monophyletic group, encompassing approaches to characterise trait (see Glossary box) than most other the lycophytes, ferns and seed plants. evolution — they, hence, face similar land plant species [8]. Similar to the Uncertainty still remains regarding the challenges. model moss Physcomitrella patens — branching order of the non-vascular The research community taking an which has been extensively used plants, the bryophytes (hornworts, evolutionary point-of-view on plant throughout the last two decades [9] — liverworts and mosses), with three main

A Angiosperms DEMosses Gymnosperms Vascular Gymnosperms Liverworts Angiosperms plants Ferns Hornworts Ferns or Lycophytes Angiosperms Lycophytes Tracheophytes Mosses Gymnosperms Hornworts Non-vascular Liverworts Ferns Liverworts plants or Hornworts Lycophytes Mosses Bryophytes Zygnematophyceae Rotate Zygnematophyceae Rotate Zygnematophyceae Coleochaetophyceae Coleochaetophyceae Coleochaetophyceae Streptophyte Charophyceae Charophyceae Charophyceae algae Klebsormidiophyceae Klebsormidiophyceae Klebsormidiophyceae or Chlorokybophyceae Chlorokybophyceae Chlorokybophyceae Charophytes Mesostigmatophyceae Mesostigmatophyceae Mesostigmatophyceae Chlorophytes Chlorophytes Chlorophytes

B F ae Klebsormidiophyceae Angiosperms s e alg te hyt hy top Gymnosperms p ep to tr ae S e ep Ferns tr Charophyceae S Coleochaetoph

Lycophytes tophyc Hornworts Mosses Zygnematophyceae Chlorokybophyceae yceae Liverworts Mesostigma Zygnematophyceae Coleochaetophyceae Charophyceae Chlorophytes Klebsormidiophyceae Land plants Chlorokybophyceae Mesostigmatophyceae G Streptophyte algae Chlorophytes Chlorophytes Land plants

C H Angiosperms Lycophytes Gymnosperms es Br yt yo Ferns h ph op y e te h s Lycophytes c F ra erns T Mosses osses Liverworts Gymnos M s Hornworts pe rms erwort Zygnematophyceae Liv Coleochaetophyceae Charophyceae Klebsormidiophyceae Angiosperms Hornworts Chlorokybophyceae Mesostigmatophyceae I Tracheophytes Bryophytes Chlorophytes Current Biology

Figure 1. Navigating the plant phylogeny — one tree, different views. (A–C) The dendrograms depict the three most highly supported branching orders for algal and land plant groups forming the green lineage (after Puttick et al. [12]). (A) Bryophyte monophyletic; (B) Mosses and liverworts monophyletic, hornworts sister to tracheophytes; (C) Mosses and liverworts monophyletic with hornworts sister to all land plants. (D,E) Alternative views of (A) showing that nodes can be rotated in a tree without changing the topology or relationships between sister groups. (F) Alternative view of cladograms (A–C) depicting the paraphyly of streptophyte algae, the position of the Zygnematophyceae as sister lineage to land plants, and the position of the chlorophyte sister lineage to the streptophytes. (G) Condensed view of (H). (H) Cladogram depicting the relationships between the major groups of land plants as in (A,D,E). (I) Condensed view of (H) highlighting bryophyte and tracheophyte monophyly.

Current Biology 29, R1105–R1121, November 4, 2019 R1111 Current Biology Magazine

hypotheses equally well supported A guideline for plant evolutionary the origin of the fl ower, a synapomorphy (Figure 1A–C) [11,12,14]. biology studies of angiosperms (the ingroup), one One requirement for the study of can choose to compare with the evolution is the ability to navigate Step 1: Inferring trait evolution gymnosperms (the outgroup) that lack phylogenies. Rooted phylogenetic trees Any study on the evolution of form fl owers. By mapping a character of contain deep and shallow branches. and function (traits) should start by interest onto a species tree of both the In addition, for practical reasons trees precisely defi ning the trait of interest. in- and outgroup, it is possible to defi ne often include more species that are Given that the same term can be used the evolution of a trait based on extant closely related to the focal organisms — by different authors or by different fi elds species. which is referred to as selective, to mean different things, it is essential A trait mapped onto a phylogeny only or biased, taxon sampling — often to be explicit about the defi nition of the tells us about the presence of the trait resulting in depictions that can resemble trait of interest. With all the diversity in the living species. However, it does ladders with certain clades of organisms of the species investigated and their not reveal how the trait evolved in the of interest (e.g., humans, angiosperms) bouquet of traits brought to the table, past. To do this, we must infer ancestral at the ‘top’ of the trees. However, it might be necessary to clearly defi ne character states. An ancestral character phylogenetic trees can be rotated the trait of interest — for example, state is the defi nition of a character at any node without changing their what is meant by broad terms such at a node within the tree rather than evolutionary meaning (Figure 1A,D,E). as ‘multicellular’ or ‘complex body in living species (which sit on the Consequently, to put trait evolution into plan’? It further is essential to bear tips, i.e. ‘leaves’, of the tree). Inferring an evolutionary context, proper tree in mind that over broad evolutionary ancestral character states hinges on thinking — which is not intuitive — is timescales, ancestral traits derive the usage of the appropriate statistical required [15]. independently in different lineages methods (Figure 2) [19,20]. For instance, While different extant organisms and, while being homologous, may combining three different methods might share a more recent or distant result in completely different forms, (maximum parsimony, maximum common ancestor, they have all been thus obscuring homology at a fi rst likelihood, and Bayesian approaches) on subject to evolutionary changes. glance. For instance, the 3D leaves a database listing fl ower morphologies In the case of land plant evolution, of succulent plants are homologous of 792 extant fl owering plant species, for example, this means that the to the leaves of all euphyllophytes, Sauquet et al. recently inferred a angiosperms are not ‘higher plants’ but display a completely different, morphology for the common ancestor of and, in turn, bryophytes are not tube-like structure at the macroscopic all extant fl owers: bisexual and radially ‘lower’, ‘basal’, or ‘primitive’ plants. level [17]. Conversely, convergent symmetric [21]. In a different context, The best practice, hence, is to refer evolution, or homoplasy, may lead to Werner et al. [22 ] inferred losses and to any given organism by the name similar yet not homologous structures, gain of the arbuscular mycorrhizal (AM) of the group to which it belongs. such as the megaphyllous leaves in symbiosis formed by most land plants Marchantia, Physcomitrella, and euphyllophytes and microphyllous with members of the Glomeromycotina any other extant bryophyte are as leaves in lycophytes [18]. fungi, a mutualistic symbiosis that ‘evolved’ as any other plant that is Once defi ned, one can map the trait improves plant nutrition [23]. Using a living today. Extant bryophytes and of interest onto a species tree that database of 3,736 seed plant species angiosperms are equally divergent captures a range of organisms salient to they predicted that AM symbiosis was from the most recent common the question. Mapping means scoring present in the MRCA of seed plants, ancestor (MRCA) of all land plants all (or as many as possible) species in and was lost multiple times during [12]. Similarly, any trait present in the tree for the state of the trait (typically seed plant evolution [22]. Importantly, the MRCA of Zygnematophyceae presence or absence). The range of they compared their inferences of AM and land plants has experienced an organisms is defi ned by starting with a symbiosis losses with the evolution of equal opportunity for divergence focal species harboring the trait. This is alternative nutrient-uptake strategies in each lineage since the time they key to applying the comparative method: and discovered a strong correlation [22]. derived from their MRCA. The fact given a rooted phylogeny, in an iterative Such inferences are the basis for that every organism is composed process related species have to be comparative studies, and in most of ancestral and derived traits — investigated — from closely to distantly cases are directly followed by step two refl ected by independent gains and related. Eventually, a deep-enough node (see below). However, an ancestral losses of genomic parts, expansions, in the phylogeny will be reached that state reconstruction is a prediction diversifi cations and genome includes species lacking the trait. For that can only be tested by looking rearrangements — has long been instance, to study the evolution of fl ower for direct or indirect evidence of the discussed as ‘mosaic evolution’ or development, one may start with any character of interest in the past. When heterobathmy [16]. model angiosperm, such as Arabidopsis available, the fossil record is therefore Plant evolutionary biology studies or Antirrhinum, and expand the sampling essential for testing predictions of therefore fi rst involve the investigation to the entire angiosperm clade. To be ancestral character states and fossils of the species of interest and where informative, comparative analyses must can be integrated with well-supported they fall on a phylogenetic tree, keeping include species that fall beyond the phylogenies to improve such inferences in mind that all extant lineages have clade of interest (the ingroup), forming [24]. DNA data are never available evolved from an ancient MRCA. the outgroup. For example, to defi ne from ancient fossils, but various forms

R1112 Current Biology 29, R1105–R1121, November 4, 2019 Current Biology Magazine

of information about the fossils can ABAngiosperms Angiosperms CAngiosperms be integrated to understand trait Gymnosperms Gymnosperms Gymnosperms evolution. Although fossils are not Ferns Ferns Ferns always straightforward to interpret and may represent derived traits of extinct Lycophytes Lycophytes Lycophytes lineages, integration of fossil data Mosses Mosses Mosses may lead to strong reinterpretation of Liverworts Liverworts Liverworts predicted ancestral states, such as in Hornworts Hornworts Hornworts the case of fl owers [25]. Here, we will Zygnematophyceae Zygnematophyceae Zygnematophyceae provide two examples — fi rst from the DE F evolution of plant–fungal symbiosis and Angiosperms Angiosperms Angiosperms second from the anatomical evolution of Gymnosperms Gymnosperms Gymnosperms rooting structures. Ferns Ferns Ferns Most extant land plants (85%) form Lycophytes Lycophytes Lycophytes AM symbiosis, which likely evolved in Hornworts Hornworts Hornworts the MRCA of land plants, as proposed Liverworts Liverworts Liverworts by phylogenetic inferences [22,26]. However, members of two bryophyte Mosses Mosses Mosses lineages (i.e., liverworts and hornworts) Zygnematophyceae Zygnematophyceae Zygnematophyceae as well as some vascular plants (i.e., Current Biology lycopods and ferns) [27] can develop endosymbiotic associations with not Figure 2. Reconstructing trait evolution and ancestral states from phylogenetic trees. just the Glomeromycotina, but also Top (A–C) and bottom (D–F) panels show two plausible topologies of the land plant phylogeny, where bryophyte (green) and tracheophyte (yellow) lineages are monophyletic (A–C), or mosses the Mucoromycotina, sometimes and liverworts are monophyletic, and hornworts form the sister group to tracheophytes (D–F). simultaneously [28]. Based on this White- and red-fi lled circles represent the absence and presence of a given hypothetical character observation, it can be proposed that in extant groups, respectively. In the fi rst scenario (A,D), the tree topology has no impact on an- dual colonization might have been a cestral character state reconstruction. Reconstructions suggest that the character was present in trait present in the earliest land plants. the land plant common ancestor and that it was lost after the divergence between lycophytes and This prediction can be tested by other tracheophytes, because under a parsimonious model of evolution a single loss (in the MRCA of ferns and seed plants) is more probable than two independent gains (in the MRCA of bryo- examining fossil plants. Fossil sites, phytes and in lycophytes). In the second scenario (B,E), the character is present only in bryophytes such as the 407 million year old Rhynie and the tree topology impacts on ancestral character state reconstruction. In (B), reconstruction chert, approaching the predicted age suggests that the character was present in the bryophyte common ancestor and absent in the of the fi rst land plants that originated tracheophyte common ancestor, and the reconstruction for the land plant common ancestor is c. 515–475 million years ago, provides uncertain. In (E), reconstruction suggests that the character was present in the land plant common a window into the early evolution of ancestor and subsequently lost in the tracheophyte common ancestor. The number of required changes (losses/gains) of the characters remains the same in each scenario. In the third scenario land plants [29]. Two exceptionally (C,F), the character is present only in mosses and liverworts. In (C), reconstruction suggests that preserved fossils of early land plant the character was absent in the land plant common ancestor and was gained in the moss/liverwort described from the Rhynie chert, common ancestor. In (F), reconstruction suggests that the character was gained in the moss/liv- Aglaophyton majus (a non-vascular erwort common ancestor, but the situation in the land plant common ancestor is uncertain. Again, plant [30]) and Horneophyton lignieri (a the required number of changes to explain the evolution of the trait remains the same. vascular plant [31]) show hallmarks of symbioses with both Glomeromycotina allow for confi dent placement of these chert fossil, Asteroxylon mackiei. and Mucoromycotina, thus validating species on a phylogeny of land plants The meristem of the rooting axes of the ancestral state inference [32–34]. [30,38]. It was found that species in this lycophyte lacks root caps and These fossils, when examined within the Rhynie chert spanned the origin of therefore displays a transitional suite a phylogenetic framework, allow us the vascular lineage and the majority of characters with some but not all to confi rm that early non-vascular of species developed rhizoid-based characters of extant plant roots [44]. and vascular species developed (fi lamentous outgrowth) rooting This extinct transitional stage sheds endosymbioses with one or both groups systems [39–41 ]—a fi nding at odds light on gradual character evolution of fungi [35,36]. with the prediction that the common leading to the roots of extant plants. Almost all extant tracheophytes ancestor of vascular plants developed The aim of the fi rst step is to infer the develop specialized rooting organs a true root. This suggests that roots of trait evolution, mostly relying on extant termed ‘true’ roots that develop from extant vascular plants do not originate species. Fossils are not essential but, a root meristem with a root cap [37]. from a shared common ancestor, when available, strongly improve these Hence, the prediction based on living that is, they are not homologous inferences. species is that the common ancestor of but in fact evolved independently at vascular plants possessed a true root. least twice [40,42,43] — making their Step 2: Reconstructing the evolution To test this hypothesis, we can again similarities in anatomy the product of of genes associated with the trait of call upon the Rhynie chert plants, whose convergent evolution. This theory was interest exceptional preservation provides cemented with the examination of a More than two decades of genetics numerous anatomical characters that rooting structure from another Rhynie in model angiosperms such as

Current Biology 29, R1105–R1121, November 4, 2019 R1113 Current Biology Magazine

Box 1. Glossary of terms.

Term Deﬁ nition Deep homology Deep homology describes the sharing of the genetic regulatory apparatus that is used to build morphologically and phylogenetically disparate features. Extant A group of organisms (species/clade) for which living representatives exist — that is, a group that is not extinct. Heterobathmy The occurrence of both ancestral and derived character states in a given organism. Applied to groups of organisms this means that ancestral and derived characters occur in different numbers and combinations across the sampled diversity of life. In turn, heterobathmy highlights that no single organism is solely made up of ancestral or derived characters. Homology Two sequences are homologous if they share a common ancestor. Homoplasy A character state/trait that is found in several species but was not present in their common ancestor. Often the result of convergent evolution. Monophyletic Monophyletic genes/organisms cluster together in a phylogeny and form a supported clade (a monophylum). All genes/organisms of that monophylum share the same last common ancestor. In turn, the monophylum includes all descendants from that last common ancestor. Neofunctionalization One ortholog has acquired a new function that was absent in the orthologs of the MRCA. Ortholog Orthologs are homologous genes that derive from speciation (not duplication) events. Outgroup Distantly related species or gene family in the context of the analysed group of organisms that are often used to root the tree. Paralogs Homologs of a gene that originate from a duplication event. Paraphyletic A paraphylum is a group of organisms that share a last common ancestor but that does not include all descendants of that last common ancestor. Also called a grade (as opposed to a clade). Plesiomorphy Ancestral character present in the MRCA of the group. Pro-orthologs Pro-orthologs are gene orthologs to paralogs in another species. Sister group The group that in a cladogram or phylogenetic analysis is most closely related to the analysed/ discussed group and therefore clusters with said group. The discussed group and its sister group together form a monophyletic group. Subfunctionalization Two paralogs have each retained a subset of the original functions of the ancestral gene. Synapomorphy A synapomorphy is a feature that was gained in the most recent common ancestor (MRCA) and is shared by the descendants of the MRCA.

Arabidopsis have established causal While the evolutionary history of gene Phylogeny of the candidate gene, links between traits and genes, families is tied to vertical evolution of the Tma12, indicated its likely horizontal thus allowing insights to be gained species, additional genetic processes gene transfer from bacteria, providing on genetic and biochemical levels. (gene duplication, loss, and — a putative scenario for the origin of this Studying the evolution of these well- occasionally — horizontal gain) add an trait in ferns [47]. described genes by phylogenetic additional layer of complexity to this. For To be fully informative, phylogenetic inference opens the door towards a instance, the scattered distribution of a tree reconstruction must be conducted better understanding of the evolution of trait may be the result of convergent gain on an appropriately curated dataset the traits and the formulation of working (homoplasy) or convergent losses, such covering the entire species-space of hypotheses. as complex leaves in the Brassicaceae interest deﬁ ned in step one. Indeed, In step one, we have outlined how to [45]. Resolving the phylogeny of restricting the analysis to a too narrow work with a given species phylogeny. REDUCED COMPLEXITY (RCO), a set of species may yield misleading Through a mapping approach, traits gene known to regulate complex leaf results. For instance, using the advent were projected onto the species morphology in Cardamine hirsute of more genomes and transcriptome phylogeny and ancestral character [46 ], identiﬁ ed multiple independent sequence of Zygnematophyceae, it has states of the MRCAs of the species in losses in species with simpler leaves, recently been demonstrated that many that phylogeny were reconstructed. indicative of convergent losses of of the genes once considered to be Here, we will hone in on gene the trait [45]. Another example is typical of or even unique to land plants phylogenies to understand the evolution the well characterized resistance are in fact present in the algal sister of the genetics that underpin these traits. to phytophagous insects in ferns. group of land plants [26,48–51].

R1114 Current Biology 29, R1105–R1121, November 4, 2019 Current Biology Magazine

A number of DNA sequence the hypothesis that such genes have mutants with orthologs from multiple databases are now available for plant a conserved molecular function — plant clades. biologists. These include Phytozome sometimes termed functional orthologs. (https://phytozome.jgi.doe.gov/pz/ Molecular function encompasses all Step 4: Determining the evolution of portal.html), which hosts a select set the features that are important for the a gene’s biological role of well-curated genomes, and the protein action, such as protein–protein Conserved biochemical properties of a 1KP database (https://db.cngb.org/ interactions, enzymatic activities, protein does not mean that its biological onekp/), which includes RNA-seq data DNA-binding abilities or regulation role itself is conserved. Testing the from a range of land plant and algal sites (such as phosphorylation). conservation of a gene’s biological role sequences, many of which still lack In the hypothetical case where all is typically performed by generating genomes. A database covering both these functions are known, in vitro mutants that are devoid of the function species diversity and sequencing depth assays can be conducted. However, of the respective gene (often known as (i.e., genomes and not transcriptomes) a more comprehensive approach reverse genetics) in multiple species. If should result from the recently launched to test all these features at once is knock-out mutants of an orthologous 10KP initiative that aims at sequencing an inter-species complementation gene in two species display similar all plant genera [52]. assay [55]. If the gene of one defects, it can be hypothesized that After identifying homologs of the species can complement the loss- the biological role of that gene has target genes by similarity searches of-function mutation of an ortholog been conserved since the divergence (such as BLAST), orthology is in another species, it indicates of these two species, since their MRCA determined using phylogenetic analyses conservation of its molecular function. lived. As an example, the function [53]. Two genes are orthologs if they There are many examples of deep of ABI3 as a key transcription factor have been vertically inherited from conservation, between the fl owering controlling the response to dehydration a common ancestor only through plant Arabidopsis, the bryophytes has been found in both Arabidopsis speciation — in other words, if Physcomitrella patens or Marchantia and Physcomitrella, suggesting the gene phylogeny matches the [51,56] and streptophyte and its conservation at least since the species phylogeny. Orthology is often chlorophyte algae [26,51]. MRCA of land plants [59]. Another complicated by paralogy (duplicated For instance, complementation example, where such analyses have genes), but phylogenetic inference helps assays of the Medicago truncatula been extended to a broader range of to defi ne ancestral states of gene family dmi3 mutant (which is unable to form species, is the conserved function of and sub-family evolution [54]. symbiosis with Glomeromycotina a clade of bHLH transcription factors The aim of step two is, once a gene fungi) have been conducted with in the formation of cells with rooting of interest has been identifi ed, to DMI3 orthologs from liverworts, function in dicots, monocots, mosses reconstruct the evolutionary history hornworts [57], Zygnematophyceae, and liverworts, a clear example of deep of the gene based on phylogenetic Chlorokybophyceae and homology [60–64]. analysis, making use of the extensive chlorophytes [26]. While liverwort, When inferring a gene’s biological role sequence databases. Ultimately, hornwort, Zygnematophyceae and using comparisons between multiple comparisons between species Chlorokybophyceae orthologs were species, it is important to keep in mind phylogenies and gene phylogenies able to complement the M. truncatula that the power of the comparative allows for the correlation of genotype dmi3 mutant, the chlorophyte analysis is limited by the number of and phenotype and the formulation of orthologs failed to do so [26,57]. This species sampled. For instance, if a hypotheses. It is prudent to note that indicates that the molecular functions biological role is not conserved between testing such hypotheses that hinge on of DMI3 are not completely conserved two species, it may either be that the correlative predictions require functional between the chlorophyte and the role was not present in their MRCA, validation, which are being explored in streptophyte orthologs, suggesting or alternatively, that the biological role the next two steps. the gain of a new function in the was present in the MRCA but has MRCA of the streptophytes [26]. subsequently been lost in one or the Step 3: Determining the evolution of Another example is the fl oral regulator other species. To distinguish between biochemical properties LEAFY, which evolved novel functions these two scenarios, it is of crucial Orthology is a statement about apparent in seed plants that are not importance to add more lineages, linear descent that is inferred from shared with, for example, mosses or represented by emerging model species, phylogenies. That means that it does the streptophyte alga Klebsormidium to the equation [65]. For instance, not have to coincide with (completely) nitens [58]. Note that functional data the transcription factor LEAFY has conserved biochemical properties. This do not impact orthology of genes/ been studied in a dozen eudicots, two assumption needs to be experimentally proteins. Orthologs are orthologous monocots, Physcomitrella and more tested. to one another, no matter whether recently the fern Ceratopteris [66]. A Many genes have been functionally shown to be functionally conserved or conserved defect in leafy loss of function characterized in the fl owering plant divergent. mutants in both Physcomitrella and Arabidopsis. If phylogenetic inference In step three, we suggest to Ceratopteris is the absence of division (step two) shows that there are study the evolution of biochemical in the sporophyte zygote — a defect orthologous sequences of a gene in properties by conducting inter-species not found in angiosperm mutants [66]. other plants, one can derive and test complementation assays of knock-out Because this role is found in mosses

Current Biology 29, R1105–R1121, November 4, 2019 R1115 Current Biology Magazine

and ferns, it can be proposed as the (step one). Phylogenetic analyses Concluding remarks ancestral state (since one loss in seed identifi ed a clear COI1 ortholog in all Plant science is experiencing a plants is more parsimonious and likely investigated land plants (step two). diversifi cation of species accessible for than two independent gains in mosses However, the protein from bryophytes genetic work. Comparisons of functional and ferns). Although previously proposed does not allow JA–Ile perception data gathered in model angiosperms, [67], experimentally determining whether (step three), indicating non-conserved mosses, liverworts as well as multiple that role is ancestral (lost or reduced in biochemical properties. A single representatives of the major plant angiosperms) or derived (gain in mosses) mutation discriminating COI1 from clades will allow the plant community with only Physcomitrella and angiosperm tracheophytes from those in bryophytes to paint a more comprehensive picture mutants was almost impossible. was identifi ed [71]. of the evolutionary mechanisms that Obviously, such inferences will become Importantly, the authors included have led to the diversity observed in stronger as model systems become lycophytes in the sequence extant species. Reporting such fi ndings available from other major clades, such comparison, pinpointing the amino- requires appropriate methodology and as the Zygnematophyceae [65,68]. acid switch at the origin of the common terminology that we describe Besides the limited number of tracheophytes. While the metabolic in this commentary. model species, a common limitation precursor of JA–Ile, OPDA, inhibits of reverse genetic approaches is the growth of Marchantia and other ACKNOWLEDGEMENTS redundancy, often caused by a close bryophytes, the Marchantia coi1 mutant The authors would like to thank Liam Dolan homolog or a paralog. Because gene is insensitive to that molecule. Using (University of Oxford) for his critical reading duplications are very common in a biochemical screening, the authors and suggestions on earlier drafts of the plant genomes, it remains diffi cult [71] showed that the ligand of the manuscript. P.-M.D., M.R. and C.D. are part to reject the presence of recent, Marchantia COI1 receptor is dinor- of the LRSV laboratory, which belongs to species-specifi c, paralogs. Thus, OPDA, another molecule derived from the TULIP Laboratoire d’Excellence (LABEX) absence of phenotypes must be taken OPDA. Phenotypically, the coi1 mutant (ANR-10-LABX-41). P.-M.D. and M.R. are with caution. In addition, following was more sensitive to a generalist funded by the ANR grant EVOLSYM (ANR-17- duplication, subfunctionalization may herbivore, suggesting a biological role CE20-0006-01). J.d.V. gratefully acknowledges occur differently between the paralogs reminiscent of the one in angiosperms funding through a Research Fellowship in different species. For instance, in despite differences in the biochemical (VR132/1-1) and Return Grant (VR132/3-1) Arabidopsis, AGAMOUS regulates the function (Step 4). Altogether, these data awarded by the German Research Foundation specifi cation of reproductive organs suggest that COI1 functioned in the (DFG). Y.C. acknowledges funding from the while this function is taken over by its perception of OPDA-derived molecules French National Center for Scientifi c Research paralog, PLENA, in Antirrhinum [69]. in the MRCA of land plants. In (ATIP-AVENIR research fellowship) and the In step four, the biological roles of tracheophytes, possibly a single point University of Lyon (IMPULSION grant). A.J.H. orthologous genes are explored in mutation led to the capacity of COI1 to was funded by the George Grosvenor Freeman phylogenetically diverse model plants perceive JA–Ile. The signaling pathways Fellowship by Examination in Sciences, using reverse genetics. downstream of COI1 have likely been Magdalen College (Oxford). C.F.D. is supported conserved since the land plant MRCA. in part by the Maryland Agricultural Experiment Step 5: Synthesizing the molecular To test this hypothesis, the authors Station (MAES). F.-W.L. is supported by the National Science Foundation grant DEB- evolution of the trait of interest [71] mimicked evolution by replacing 1831428. Projecting gene phylogeny, in the Marchantia COI1 sequence the conservation of the biochemical codon discriminating bryophytes and properties, and conservation of the tracheophytes with the tracheophyte REFERENCES biological role onto the species’ version. This was suffi cient to make 1. Moyroud, E., and Glover, B.J. (2017). The phylogenetic tree provides a means Marchantia sensitive to JA–Ile, thus evolution of diverse fl oral morphologies. Curr. Biol. 27, R941–R951. of unravelling the evolutionary validating the proposed scenario [71]. 2. Vasco, A., Moran, R.C., and Ambrose, B.A. mechanisms that shaped the trait All the previous steps are combined (2013). The evolution, morphology, and of interest. An elegant example is in step fi ve, hereby producing a development of fern leaves. Front. Plant Sci. 4, 345. the evolution of the jasmonate (JA) holistic understanding of the molecular 3. Coudert, Y., Bell, N.E., Edelin, C., and receptor COI1. evolution of the trait of interest. Harrison, C.J. (2017). Multiple innovations underpinned branching form diversifi cation in Jasmonates are a class of plant Such inference is valid depending mosses. New Phytol. 215, 840–850. hormones that play multiple essential on available data. Experimental 4. Carroll, S.B. (2008). Evo-devo and an expanding roles in plants [70]. In tracheophytes, validations in additional species and evolutionary synthesis: a genetic theory of morphological evolution. Cell 134, 25–36. COI1 perceives the conjugated form the sequencing of more genomes 5. Rich, M., and Delaux, P.-M. (2018). Taking the of JA (JA–Ile), while the bryophyte may lead to refi ned likely scenarios. step: from Evo-Devo to plant-microbe interaction evolution with the liverwort Marchantia. New version does not [71]. So, what is Importantly, ‘likely scenario’ is the Phytol. 218, 882–884. COI1’s story? The trait here is the appropriate wording here — it conveys 6. Upson, J.L., Zess, E.K., Białas, A., Wu, C. perception of JA–Ile, which is found the due humbleness. Since we can hang, and Kamoun, S. (2018). The coming of age of EvoMPMI: evolutionary molecular plant– in tracheophytes but not in any of the only investigate extant species at the microbe interactions across multiple timescales. bryophytes that the authors tested, genetic level, we will always capture a Curr. Opin. Plant Biol. 44,108–116 7. Lynch, M., Field, M.C., Goodson, H.V., including Physcomitrella, Marchantia, mere snapshot of the phenotypes that Malik, H.S., Pereira-Leal, J.B., Roos, D.S., and the hornwort Anthoceros agrestis evolution has given rise to. Turkewitz, A.P., and Sazer, S. (2014). Evolutionary

R1116 Current Biology 29, R1105–R1121, November 4, 2019 Current Biology Magazine

cell biology: Two origins, one objective. Proc. angiosperms: the “war of the whorls” is heating 46. Vlad, D., Kierzkowski, D., Rast, M.I., Vuolo, F., Natl. Acad. Sci. USA 111, 16990–16994. up. J. Exp. Bot. 70, 2615–2622. Dello Ioio, R., Galinha, C., Gan, X., Hajheidari, M., 8. Bowman, J.L., Kohchi, T., Yamato, K.T., 26. Delaux, P.-M., Radhakrishnan, G.V., Hay, A., Smith, R.S., et al. (2014). Leaf shape Jenkins, J., Shu, S., Ishizaki, K., Yamaoka, S., Jayaraman, D., Cheema, J., Malbreil, M., evolution through duplication, regulatory Nishihama, R., Nakamura, Y., Berger, F., et al. Volkening, J.D., Sekimoto, H., Nishiyama, T., diversifi cation and loss of a homeobox gene. (2017). Insights into land plant evolution garnered Melkonian, M., Pokorny, L., et al. (2015). Algal Science 343, 780–783. from the Marchantia polymorpha genome. Cell ancestor of land plants was preadapted for 47. Li, F.W., Brouwer, P., Carretero-Paulet, L., Cheng, 171, 287–304.e15. symbiosis. Proc. Natl. Acad. Sci. USA 112, S., De Vries, J., Delaux, P.M., Eily, A., Koppers, 9. Lang, D., Ullrich, K.K., Murat, F., Fuchs, J., 13390–13395. N., Kuo, L.Y., Li, Z., et al. (2018). Fern genomes Jenkins, J., Haas, F.B., Piednoel, M., 27. Rimington, W.R., Pressel, S., Duckett, J.G., and elucidate land plant evolution and cyanobacterial Gundlach, H., Van Bel, M., Meyberg, R., et al. Bidartondo, M.I. (2015). Fungal associations of symbioses. Nat. Plants. 4, 460–472. (2018). The Physcomitrella patens chromosome- basal vascular plants: Reopening a closed book? 48. Nishiyama, T., Sakayama, H., de Vries, J., scale assembly reveals moss genome structure New Phytol. 205, 1394–1398. Buschmann, H., Saint-Marcoux, D., Ullrich, K.K., and evolution. Plant J. 93, 515–533. 28. Strullu-Derrien, C., Selosse, M.-A., Kenrick, P., Haas, F.B., Vanderstraeten, L., Becker, D., 10. Ishizaki, K., Nishihama, R., Yamato, K.T., and and Martin, F.M. (2018). The origin and evolution Lang, D., et al. (2018). The Chara genome: Kohchi, T. (2016). Molecular genetic tools and of mycorrhizal symbioses: from palaeomycology secondary complexity and implications for plant techniques for marchantia polymorpha research. to phylogenomics. New Phytol. 220, 1012–1030. terrestrialization. Cell 174, 448–464.e24. Plant Cell Physiol. 57, 262–270. 29. Edwards, D., Kenrick, P., and Dolan, L. (2018). 49. Wilhelmsson, P.K.I., Mühlich, C., Ullrich, K.K., 11. Wickett, N.J., Mirarab, S., Nguyen, N., History and contemporary signifi cance of the and Rensing, S.A. (2017). Comprehensive Warnow, T., Carpenter, E., Matasci, N., Rhynie cherts-our earliest preserved terrestrial genome-wide classifi cation reveals that many Ayyampalayam, S., Barker, M.S., ecosystem. Philos. Trans. R. Soc. Lond. B. Biol. plant-specifi c transcription factors evolved Burleigh, J.G., Gitzendanner, M.A., et al. (2014). Sci. 373, 20160489. in Streptophyte algae. Genome Biol. Evol. 9, Phylotranscriptomic analysis of the origin and 30. Kenrick, P., and Crane, P.R. (1997). The origin 3384–3397. early diversifi cation of land plants. Proc. Natl. and early evolution of plants on land. Nature 389, 50. de Vries, J., Curtis, B.A., Gould, S.B., and Acad. Sci. USA 111, E4859–E4868. 33–39. Archibald, J.M. (2018). Embryophyte stress 12. Puttick, M.N., Morris, J.L., Williams, T.A., 31. Cascales-Miñana, B., Steemans, P., Servais, T., signaling evolved in the algal progenitors of land Cox, C.J., Edwards, D., Kenrick, P., Pressel, S., Lepot, K., and Gerrienne, P. (2019). An alternative plants. Proc. Natl. Acad. Sci. USA 115, E3471– Wellman, C.H., Schneider, H., Pisani, D., et al. model for the earliest evolution of vascular E3480. (2018). The interrelationships of land plants and plants. Lethaia. 52, 445–453. 51. Ju, C., Van De Poel, B., Cooper, E.D., the nature of the ancestral embryophyte. Curr. 32. Remy, W., Taylor, T.N., Hass, H., and Kerp, H. Thierer, J.H., Gibbons, T.R., Delwiche, C.F., and Biol. 28, 733–745.e2. (1994). Four hundred-million-year-old vesicular Chang, C. (2015). Conservation of ethylene 13. Timme, R.E., Bachvaroff, T.R., and Delwiche, C.F. arbuscular mycorrhizae. Proc. Natl. Acad. Sci. as a plant hormone over 450 million years of (2012). Broad phylogenomic sampling and USA 91, 11841–11843. evolution. Nat. Plants 1, 14004. the sister lineage of land plants. PLoS One 7, 52. Cheng, S., Melkonian, M., Smith, S.A., e29696. 33. Taylor, T.N., Remy, W., Hass, H., and Kerp, H. (1995). Fossil arbuscular mycorrhizae from the Brockington, S., Archibald, J.M., Delaux, P.-M., 14. de Sousa, F., Foster, P.G., Donoghue, P.C.J., Li, F., Melkonian, B., Mavrodiev, E. V, Sun, W., Schneider, H., and Cox, C.J. (2018). Nuclear Early Devonian. Mycologia 87, 560–573. 34. Strullu-Derrien, C., Kenrick, P., Pressel, S., et al. (2018). 10KP: A phylodiverse genome protein phylogenies support the monophyly of sequencing plan. Gigascience. 7, 1–9. the three bryophyte groups (Bryophyta Schimp.). Duckett, J.G., Rioult, J.P., and Strullu, D.G. (2014). Fungal associations in Horneophyton 53. Mindell, D.P., and Meyer, A. (2001). Homology New Phytol. 222, 565–575. evolving. Trends Ecol. Evol. 16, 434–440. 15. Baum, D.A., Smith, S.D., and Donovan, S.S.S. ligneri from the Rhynie Chert (c. 407 million year old) closely resemble those in extant lower land 54. Altenhoff, A.M., and Dessimoz, C. (2012). (2005). Evolution. The tree-thinking challenge. Inferring orthology and paralogy. In Methods in Science. 310, 979–980. plants: Novel insights into ancestral plant-fungus molecular biology (Clifton, N.J.), pp. 259–279. 16. Hennig, W. (1950). Grundzüge einer Theorie symbioses. New Phytol. 203, 964–979. der phylogenetischen Systematik (Deutscher 35. Bidartondo, M.I., Read, D.J., Trappe, J.M., 55. Kramer, E.M. (2015). A stranger in a strange land: Zentralverlag). Merckx, V., Ligrone, R., and Duckett, J.G. (2011). The utility and interpretation of heterologous 17. Ogburn, R.M., and Edwards, E.J. (2013). The dawn of symbiosis between plants and expression. Front. Plant Sci. 6, 734. Repeated origin of three-dimensional leaf fungi. Biol. Lett. 7, 574–577. 56. Rensing, S.A. (2018). Great moments in evolution: the conquest of land by plants. Curr. venation releases constraints on the evolution of 36. Feijen, F.A.A., Vos, R.A., Nuytinck, J., and succulence in plants. Curr. Biol. 23, 722–726. Merckx, V.S.F.T. (2018). Evolutionary dynamics Opin. Plant Biol. 42, 49–54. 18. Harrison, C.J., Coriey, S.B., Moylan, E.C., of mycorrhizal symbiosis in land plant 57. Wang, B., Yeun, L.H., Xue, J.Y., Liu, Y., Alexander, D.L., Scotland, R.W., and diversifi cation. Sci. Rep. 8, 10698. Ané, J.M., and Qiu, Y.L. (2010). Presence of three mycorrhizal genes in the common ancestor of Langdale, J.A. (2005). Independent recruitment 37. Foster, A.S. (Adriance S., and Gifford, E.M. land plants suggests a key role of mycorrhizas in of a conserved developmental mechanism (1974). Comparative Morphology of Vascular the colonization of land by plants. New Phytol. during leaf evolution. Nature. 434, 509–514. Plants (W.H. Freeman). 186, 514–525. 19. Ronquist, F. (2004). Bayesian inference of 38. Kenrick, P.R., and Crane, P.R. (1997). The origin character evolution. Trends Ecol. Evol. 19, 58. Sayou, C., Monniaux, M., Nanao, M.H., Moyroud, and early diversifi cation of land plants: a cladistic E., Brockington, S.F., Thévenon, E., Chahtane, 475–481. study. Nature 389, 33–39. 20. Nunn, C.L. (2011). The Comparative Approach H., Warthmann, N., Melkonian, M., Zhang, Y., et 39. Edwards, D. (2003). Embryophytic sporophytes in Evolutionary Anthropology and Biology al. (2014). A promiscuous intermediate underlies in the Rhynie and Windyfi eld cherts. Trans. R. (University of Chicago Press). the evolution of LEAFY DNA binding specifi city. Soc. Edinb. Earth Sci. 94, 397–410. 21. Sauquet, H., von Balthazar, M., Magallón, S., Science. 343, 645–648. Doyle, J.A., Endress, P.K., Bailes, E.J., 40. Kenrick, P., and Strullu-Derrien, C. (2014). The 59. Khandelwal, A., Cho, S.H., Marella, H., Barroso de Morais, E., Bull-Hereñu, K., Carrive, origin and early evolution of roots. Plant Physiol. Sakata, Y., Perroud, P.-F., Pan, A., and Quatrano, L., Chartier, M., et al. (2017). The ancestral fl ower 166, 570–580. R.S. (2010). Role of ABA and ABI3 in desiccation of angiosperms and its early diversifi cation. Nat. 41. Hetherington, A.J., and Dolan, L. (2018). tolerance. Science 327, 546–546. Commun. 8, 16047. Bilaterally symmetric axes with rhizoids 60. Menand, B., Yi, K., Jouannic, S., Hoffmann, L., 22. Werner, G.D.A., Cornelissen, J.H.C., composed the rooting structure of the common Ryan, E., Linstead, P., Schaefer, D.G., and Dolan, Cornwell, W.K., Soudzilovskaia, N.A., Kattge, J., ancestor of vascular plants. Philos. Trans. R. L. (2007). An ancient mechanism controls the West, S.A., and Kiers, E.T. (2018). Symbiont Soc. B Biol. Sci. 373, 20170042. development of cells with a rooting function in switching and alternative resource acquisition 42. Raven, J.A., and Edwards, D. (2001). Roots: land plants. Science 316, 1477–1480. strategies drive mutualism breakdown. Proc. evolutionary origins and biogeochemical 61. Proust, H., Honkanen, S., Jones, V.A.S., Natl. Acad. Sci. USA 115, 5229–5234. signifi cance. J. Exp. Bot. 52, 381–401. Morieri, G., Prescott, H., Kelly, S., Ishizaki, K., 23. Vigneron, N., Radhakrishnan, G.V, and 43. Boyce, C.K. (2010). The evolution of plant Kohchi, T., and Dolan, L. (2016). RSL class i Delaux, P.-M. (2018). What have we learnt development in a paleontological context. Curr. genes controlled the development of epidermal from studying the evolution of the arbuscular Opin. Plant Biol. 13, 102–107. structures in the common ancestor of land mycorrhizal symbiosis? Curr. Opin. Plant Biol. 44, 44. Hetherington, A.J., and Dolan, L. (2018). plants. Curr. Biol. 26, 93–99. 49–56. Stepwise and independent origins of roots 62. Kim, C.M., and Dolan, L. (2016). ROOT HAIR 24. Betts, H.C., Puttick, M.N., Clark, J.W., among land plants. Nature 561, 235–238. DEFECTIVE SIX-LIKE class I genes promote root Williams, T.A., Donoghue, P.C.J., and Pisani, D. 45. Kiefer, C., Willing, E.-M., Jiao, W.-B., Sun, H., hair development in the grass Brachypodium (2018). Integrated genomic and fossil evidence Piednoël, M., Hümann, U., Hartwig, B., distachyon. PLoS Genet. 12, e1006211. illuminates life’s early evolution and eukaryote Koch, M.A., and Schneeberger, K. (2019). 63. Kim, C.M., Han, C.D., and Dolan, L. (2017). origin. Nat. Ecol. Evol. 2, 1556–1562. Interspecies association mapping links reduced RSL class I genes positively regulate root hair 25. Rümpler, F., and Theißen, G. (2019). CG to TG substitution rates to the loss of gene- development in Oryza sativa. New Phytol. 213, Reconstructing the ancestral fl ower of extant body methylation. Nat. Plants 5, 846–855. 314–323.

Current Biology 29, R1105–R1121, November 4, 2019 R1117 Current Biology Magazine

64. Shubin, N., Tabin, C., and Carroll, S. (2009). Deep homology and the origins of evolutionary novelty. Quick guide Nature 457, 818–823. 65. Rensing, S.A. (2017). Why we need more non- seed plant models. New Phytol. 216, 355–360. Bedbugs 66. Plackett, A.R., Conway, S.J., Hewett Hazelton, K.D., Rabbinowitsch, E.H., Langdale, J.A., and Di Klaus Reinhardt Stilio, V.S. (2018). LEAFY maintains apical stem cell activity during shoot development in the fern Ceratopteris richardii. eLife 7, 39625. 67. Moyroud, E., Kusters, E., Monniaux, M., What are bedbugs? Bedbugs are Koes, R., and Parcy, F. (2010). LEAFY blossoms. the 100 or so species of heteropteran Trends Plant Sci. 15, 346–352. bugs of the family Cimicidae. However, 68. Chang, C., Bowman, J.L., and Meyerowitz, E.M. (2016). Field guide to plant model systems. Cell only three species are known to live 167, 325–339. in human beds — possibly four, if oral 69. Causier, B., Castillo, R., Zhou, J., Ingram, R., Xue, Y., Schwarz-Sommer, Z., and Davies, B. tales of the Hopi native Americans are (2005). Evolution in action: Following function in included as cultural evidence. Of those, duplicated fl oral homeotic genes. Curr. Biol. 15, the common bedbug Cimex lectularius 1508–1512. 70. Wasternack, C., and Hause, B. (2013). and the tropical bedbug C. hemipterus Jasmonates: Biosynthesis, perception, signal are offi cially classifi ed as urban pests transduction and action in plant stress response, growth and development. Ann. Bot. 111, by the World Health Organization. 1021–1058. Bedbugs are cultural synonyms for fi lth, 71. Monte, I., Ishida, S., Zamarreño, A.M., unhygienic conditions and poverty, yet Hamberg, M., Franco-Zorrilla, J.M., García-Casado, G., Gouhier-Darimont, C., none of this is supported by evidence. Reymond, P., Takahashi, K., García-Mina, J.M., The Cimicidae are obligate blood Figure 1. A scanning electron image of a et al. (2018). Ligand-receptor co-evolution shaped the jasmonate pathway in land plants. suckers, drawing their one and only mating pair fi xed in liquid nitrogen. Nat. Chem. Biol. 14, 480–488. favourite liquid from humans, bats or To adhere to the female (brown colouration), birds. the male (pale colouration) does not use the claws at the end of the feet but a special at- 1 tachment device more proximal to the claws. Laboratoire de Recherche en Sciences What are bedbugs not? Disease Végétales, Université de Toulouse, CNRS, © D. Voigt and K. Reinhardt. UPS, Castanet-Tolosan, France. 2Department vectors. Certainly, many pathogens of Plant Sciences, University of Oxford, can persist in the guts of bedbugs for one would expect that the geographic South Parks Road, Oxford, OX1 3RB, UK. weeks and, when one breaches the origin of C. lectularius would harbour 3Laboratoire Reproduction et Développement skin of experimental host animals, the largest genetic diversity of this des Plantes, Ecole Normale Supérieure de these pathogens may enter the new species, but this may be challenging Lyon, CNRS, INRA, Université Claude Bernard host. Martin bugs, also from the to prove because the bedbugs found Lyon 1, INRIA, 46 Allée d’Italie, Lyon, 69007, France. 4CBMG, University of Maryland, genus Cimex, can vector arboviruses in large cities such as New York and College Park, MD 20742, USA. 5Institute for from swallows to sparrows. These London have been transported there Molecular Evolution, Heinrich Heine University, fi ndings have led some to conclude from all over the world, also resulting in 40225 Düsseldorf, Germany. 6Department of that bedbugs could potentially act highly genetically diverse populations. Earth Sciences, The Natural History Museum, as vectors. But according to dozens, Exploiting genomic information from the Cromwell Road, London, SW7 5BD, UK. perhaps hundreds of studies, bedbugs tropical bedbug C. hemipterus could 7Boyce Thompson Institute, Ithaca, NY, USA. 8Plant Biology Section, Cornell University, have not yet been shown to spread help to address similar questions of Ithaca, NY, USA. 9Centre de Théorisation et disease in humans. how and when this species conquered de Modélisation de la Biodiversité, Station the tropics. d’Écologie Théorique et Expérimentale, UMR How did they come to like us? Fossil CNRS 5321, Moulis, France. 10Département and molecular dating places the origin It seems that bedbugs are becoming de Biochimie, Université de Montréal, of Cimicidae several million years terribly common— should we be Montréal, Québec, Canada. 11Plant Cell Biology, Faculty of Biology, University of before the origin of bats and, of course, worried? Bedbugs have always thrived Marburg, 35043 Marburg, Germany. 12BIOSS before humans, and so their ancestral in many parts of the world, although it Centre for Biological Signalling Studies, host remains unknown. C. lectularius has sometimes been suggested that University Freiburg, Germany. 13SYNMIKRO is an ancient companion of humans, they were extinct in the west. But yes, Research Center, University of Marburg, 35043 shown by a semi-fossil from 4,500 it appears that they are becoming more 14 Marburg, Germany. Institut de Systématique, years ago. Originally, this species lived common in some areas. However, Évolution, Biodiversité, UMR 7205, Muséum National d’Histoire Naturelle, Paris, France. exclusively on vespertilionid bats, but this notion rests mainly on non-peer 15Department of Biochemistry and Molecular now a morphologically and genetically reviewed newspaper articles and internal Biology, Dalhousie University, Halifax, distinct host race dwells on humans. administration reports, some of which Nova Scotia B3H 4R2, Canada. 16Institute The bat clade continues to exist in may be infl ated. For regions where we of Microbiology, Technische Universitaet Europe and Asia Minor, and molecular have good evidence of an increase, Braunschweig, 38106 Braunschweig, dating places the split between the Australia for example, increased travel, Germany. 17Institute for Microbiology and Genetics, Bioinformatics, University clades to several hundred thousand temperature rise, change in lifestyle and of Göttingen, Goldschmidtstr. 1, 37077 years ago, before the early movement insecticide resistance, or a combination Göttingen, Germany. of Homo sapiens out of Africa. Where of these factors, are suggested as *E-mail: [email protected] this split occurred is still unknown; possible reasons.