Strigolactones affect development in primitive plants. The missing link between plants and arbuscular mycorrhizal fungi?

Abstract

The question of how aquatic green algae evolved the adaptation to terrestrial conditions, which enabled them to colonize land, has intrigued scientists for a long time. It is generally believed that a symbiotic relationship with arbuscular mycorrhizal (AM) fungi stimulated this process. Indeed, fossil evidence for the existence of this symbiotic relationship dates back to the earliest land plants, c. 480–430 million yr (MY) ago. In 2005 it was discovered that modern land plants facilitate AM fungal symbiosis by exudation of chemical compounds – called strigolactones – into the rhizosphere (Akiyama et al., 2005). In this issue of New PhytologistDelaux et al. (pp. 849–863) shed new light on this by showing that freshwater green algae belonging to the Charales produce and exude strigolactones. ‘This would imply that strigolactone biosynthesis predates AM fungal colonization.’ Strigolactones were initially discovered to be germination stimulants of root parasitic plants of the Orobanchaceae, causing severe damage to their host plants (Bouwmeester et al., 2003). The discovery that they play a role in symbiosis was believed to provide an answer to the evolutionary question as to why plants produce strigolactones. However, more recently, strigolactones were also found to play an endogenous hormonal role, where they inhibit shoot branching in higher plants (Gomez-Roldan et al., 2008; Umehara et al., 2008). Since then, many other biological functions have been assigned to this ‘new’ class of plant hormones, including roles in lateral root initiation and development, primary root elongation, root hair elongation, seed germination, hypocotyl elongation and secondary growth. Also, in the primitive plant species Physcomitrella patens strigolactones have a physiological role, stimulating the growth of protonema (thread-like chains of cells that mosses form from their spores; Proust et al., 2011). These recent discoveries raised new questions about the evolutionary origin of strigolactone biosynthesis. Delaux et al. now show that species belonging to the Charales, some of the closest freshwater green algal relatives of land plants, produce and exude strigolactones. This would imply that strigolactone biosynthesis predates AM fungal colonization. Furthermore, they nicely demonstrate that application of a synthetic strigolactone to Chara coralina, a member of the Charales, stimulates rhizoid elongation. These intriguing results show that the stimulatory role that strigolactones have in AM fungal symbiosis is reminiscent of their endogenous plant hormonal origin and this sheds new light on the evolution of plant adaptation to a terrestrial existence. The green lineage (Viridiplantae) comprises the green algae (Chlorophytes and Charophytes) and their descendants, the land plants (Embryophytes; Delaux et al.; Fig. 1). The Chlorophytes, represented by freshwater and marine unicellular green algae, and the freshwater Charophytes, which are the ancestors of the land plants, diverged c. 725–1200 MY ago. The oldest order of the Charophytes, the Mesostigmatales, are unicellular organisms that possess flagellata which are lost in the Chlorobykales. The Klebsormidiales and Zygnematales are unbranched filamentous multicellular organisms, while a branched filamentous structure is first acquired by the Coleochaetales and Charales (Delaux et al.; Fig. 1). The most primitive land plants, represented by the Bryophytes, include the mosses, hornworts and liverworts, and are the closest living relatives of the first plants that colonized land. The exact relationship between land plants and their primitive relatives, however, remains uncertain. It has been suggested that the physiological adaptations in Charophyte green algae to a terrestrial lifestyle occurred gradually in a freshwater habitat in which they were periodically exposed to moist terrestrial conditions (Becker & Marin, 2009). Colonization of land c. 480–430 MY ago required various adaptations such as protection against ultra-violet (UV) radiation and temperature fluctuations as well as physical support and water/mineral uptake. With respect to the latter, rhizoid development (roots appeared at a later stage only, during evolution of the Monilophytes c. 385–400 MY ago; Fig. 1) and the potential to accommodate symbiotic associations with AM fungi, as indeed observed in Bryophytes, were of crucial importance. Indeed, the earliest fossil evidence of an AM fungal–plant association dates back to c. 400 MY ago (Remy et al., 1994), while the first fossil of a Glomeromycota-like fungus, although not associated with plant material, is 460 MY old (Redecker et al., 2000). These dates coincide with the colonization of land by plants. Phylogenetic representation of the plant kingdom showing major green plant lineages, the orders within Bryophytes and Charophytes discussed by Delaux et al. (in this issue of New Phytologist, pp. 849–863) and pictures of some of the species for which Delaux et al. provide evidence for the absence (Chlorophytes) or presence (Charales) of strigolactones. MYBP, million years before present. Leftside of figure reproduced with permission from Hay & Tsiantis (2010). Images of Nitella hyalina and Chara corallina courtesy of John Clayton (reproduced with permission); Chlamydomonas reinhardtii (http://remf.dartmouth.edu/Chlamydomonas/); Volvox carteri adapted from Nematollahi et al. (2006). In addition to molecular, biochemical and ultra-structural characterization, physiological features may support phylogenetic relationships. In order to place strigolactone biosynthesis in an evolutionary context, Delaux et al. performed a detailed study, analysing species representing various orders and divisions for the presence of strigolactones in exudates and extracts. Furthermore, the ability of these extracts and exudates to stimulate AM fungal branching was investigated and the conservation of strigolactone biosynthetic and downstream signalling genes analysed. Finally, the physiological effect of exogenous application of the synthetic strigolactone GR24 was studied. Similar to abscisic acid (ABA), strigolactones are derived from the carotenoid pathway (Matusova et al., 2005). The putative precursor of the strigolactones is all-trans-β-carotene, which is isomerized to 9-cis-β-carotene by D27 (Alder et al., 2012; Fig. 2).Two subsequent enzymatic cleavage steps catalysed by the carotenoid cleavage dioxygenases CCD7 and CCD8 (Umehara et al., 2008; Ariyaratne et al., 2009) yield an intermediate called carlactone (Alder et al., 2012; Fig. 2). Further conversion is probably mediated by a cytochrome P450 enzyme, MAX1, which indeed in Arabidopsis was demonstrated to be required for strigolactone biosynthesis (Kohlen et al., 2011). So far, no orthologues of MAX1 have been proven to exist in other plant species, possibly due to gene redundancy. Interestingly, in the moss genome no MAX1 homolog was detected at all (Proust et al., 2011). Schematic representation of the strigolactone biosynthetic pathway including biosynthetic enzymes and proteins likely involved in downstream signalling, as discussed by Delaux et al. (in this issue of New Phytologist, pp. 849–863). The perception of strigolactones and/or downstream signalling is mediated by an F-box leucine rich repeat protein (MAX2; Stirnberg et al., 2002) and/or two α/β-hydrolases (D14 and D14-like; Arite et al., 2009; Waters et al., 2012; Fig. 2). Using all available protein sequences, expressed sequence tag (EST) collections and (draft) genome and transcript assemblies from the early Prasinophyceae belonging to the Chlorophytes until the modern Angiosperms, Delaux et al. performed a comprehensive phylogenetic analysis of these strigolactone-related genes. Interestingly, D27 was highly conserved in all annotated genomes throughout the green lineage, while CCD7 and CCD8 were less consistently conserved in the Chlorophytes. Only in the genome sequences of the Chlorophyceae Chlamydomonas reinhardtii and Volvox carteri putative homologs for both CCDs were identified. However, a close examination of short conserved motifs present in CCD/NCED proteins revealed that these motifs were not conserved in the latter two species. It will be of interest to investigate the functional relevance of these conserved motifs by expressing the CCD7 and CCD8 genes of these Chlorophytes in heterologous organisms such as carried out by, for example Alder et al. (2012) to prove their catalytic functionality. In addition, significant homologs for strigolactone downstream signalling genes seem also to be lacking in the Chlorophytes. Unfortunately, due to the lack of complete genome sequences, conclusions from homology searches within the EST databases of all Charophytes, and also the Bryophytic liverworts and hornworts, are difficult to make. The detection of clear D27 homologs in most Charophyte orders and absence of CCD and downstream signalling genes may hint at an alternative role for D27. Possibly its function in strigolactone biosynthesis was only acquired later during evolution. Strigolactones and ABA partly share a common pathway and ABA has been reported in Chlorophytes (Hartung, 2010). It would be of interest to compare the conservation of strigolactone biosynthetic genes with the conservation of genes specifically involved in ABA production throughout the green lineage. Although (predicted) protein sequences may give clues about the functionality of biosynthetic pathways, the final evidence should come from analytical characterization of the anticipated pathway products and this is what Delaux et al. did. So far, the moss P. patens (belonging to the phylum of the non-vascular Bryophytes) was the most primitive plant in which strigolactones were detected (Proust et al., 2011). Using MRM-LC-MS/MS (multiple reaction monitoring-liquid chromatography mass spectrometry/mass spectrometry), Delaux et al. provided analytical evidence for the presence of strigolactones in two liverwort, Marchantia, species. The liverworts are most likely the earliest diverging lineage of the Embryophytes. Excitingly, and considering the phylogenetic study on strigolactone biosynthesis genes even unexpected, LC-MS analysis revealed the presence of strigolactones in tissue extracts of two Nitella species (Fig. 1). However, in species belonging to the Zygnematales and Coleochaetales, two additional orders of the Charophytes, no strigolactones were detected. One further step back in evolution, no strigolactones were detected in C. reinhardtii (Chlorophytes; Fig. 1). However, because MRM-LC-MS analysis is based on the detection of known strigolactones, unknown strigolactones may have been overlooked by the authors. In addition, the sensitivity of the analysis may not have been sufficiently high. Even in the land plant Arabidopsis, strigolactone detection is still challenging (Kohlen et al., 2011). Also ABA levels in the Chlorophytes are very low and only become detectable under drought stress when species colonize terrestrial habitats, as is also observed in liverworts and mosses (Hartung, 2010). It could be that the induction of strigolactone biosynthesis only appeared as a response to some selective pressure during a later stage in the evolution of the green lineage. Nevertheless, Delaux et al. elegantly use an extremely sensitive bioassay – based on the branching stimulatory activity of strigolactones in the AM fungus Gigaspora rosea– to show that there is no evidence for the presence of strigolactones in species belonging to Chlorophyceae, Zygnematales and Coleochaetales, whereas in the Charales species, C. corallina (Fig. 1), clear activity was detected. In conclusion, Delaux et al. present compelling evidence that in the green lineage strigolactones are first detected in the Charales. They propose that this could either mean that strigolactones appeared in a putative common ancestor of advanced Charophytes and the Embryophytes and was later lost in the Coleochaetales and Zygnematales. Or, that the Charales form a sister clade of the Embryophytes. Either way, because a symbiotic relationship with AM fungi is only described to occur in the early land plants and not in the Charales, the original function of the strigolactones is very likely hormonal. With regard to the acquisition of a rhizosphere signalling role of the strigolactones the authors assume this first occurred in the Bryophytes even though they show that the Charales also exude strigolactones. Nevertheless, although there are a few records of possible symbiotic interactions in the Charales, Delaux et al. assume strigolactones are not involved in this interaction and only have an hormonal role in the Charales. Analysis of the appearance of homologs of the higher plant strigolactone transporter PDR1 (Kretzschmar et al., 2012; Fig. 2) during land plant evolution may shed light on this aspect. With regard to the supposed hormonal role of strigolactones in primitive plants, Delaux et al. indeed show that the synthetic strigolactone GR24 stimulates rhizoid growth not only in the moss P. patens and the liverwort Marchantia sp., but also in the Charales species C. corallina (Fig. 1). Since hormones need to be perceived, it is perhaps not surprising that specifically in the Charales the first sign of the conservation of the putative strigolactone downstream signalling gene D14-like is observed. Strikingly, D14 is only found to be highly conserved later during evolution in Gymnosperms and Angiosperms. What could have been the evolutionary driving force for this morphological effect in these fresh water algae? It has been hypothesized that the Charophytes gradually shifted towards moist terrestrial habitats (Becker & Marin, 2009). This environmental change may have formed the selective pressure that led to rhizoid development, which provided anchorage and the possibility to take up water and nutrients. Finally, from a physiological viewpoint, stimulation of rhizoid elongation perfectly fits within the current known strigolactone effects such as stimulation of protonema expansion in moss (Proust et al., 2011), root hair elongation in angiosperms (Kapulnik et al., 2011) and induction of hyphal growth and branching in AM fungi (Besserer et al., 2006), and thus seems to bring plants and AM fungi closer together. However, if strigolactone signalling evolved first when plants started to colonize gradually drier habitats, strigolactone signalling in AM fungi must have evolved independently. It will be of major interest to see if this is indeed the case and what then are the differences and/or similarities between plants and AM fungi in strigolactone downstream signalling.