Phytohormone signaling in arbuscular mycorhiza development

https://doi.org/10.1016/j.pbi.2014.04.003Get rights and content

Highlights

  • Distinct structural needs for strigolactone perception by plants and AM fungi.

  • DELLA proteins are required for intraradical colonization and arbuscule formation.

  • Expression of DELLA in the vasculature is sufficient to promote arbuscule formation in the cortex.

  • DELLAs act genetically downstream of CYCLOPS in promoting arbuscule development.

  • An increasing number of GRAS proteins is required for AM development.

To establish arbuscular mycorhiza (AM) symbiosis glomeromycotan fungi colonize the interior of roots. This process is associated with developmental changes of root cells as well as fungal hyphae. The formation of fungal colonization-structures and the extent of root colonization are largely under plant control, depending on environmental conditions and the resulting physiological state of the host. Phytohormone signaling pathways are currently emerging as important regulators of AM development. Root exuded strigolactones activate AM fungi before colonization and a host strigolactone receptor component is required for AM development. Auxin quantitatively influences AM colonization and might perform an additional cell-autonomous function in the promotion of arbuscule development. Gibberellin signaling inhibits AM and conversely DELLA proteins are required for AM formation. Given the importance of phytohormone signaling in plant developmental responses to the environment it can be predicted that elucidating how phytohormones regulate AM development will provide a lead into understanding how plants orchestrate AM symbiosis with their physiological needs under changing environmental conditions.

Introduction

Arbuscular mycorhiza (AM) is a widespread mutualistic symbiosis between most land plants from mosses to angiosperms and fungi of the Glomeromycota [1, 2]. During symbiosis the obligate biotrophic AM fungi receive up to 20% of the plant's photosynthetically fixed carbon. Thus AM significantly contributes to carbon sequestration in the soil and to global carbon cycles with an estimated carbon flow of 5 mega tons per year [3, 4]. In turn, AM fungi improve plant mineral nutrition, especially with phosphate. They take up mineral nutrients from the soil via their extraradical hyphal network and deliver them into root cortex cells [reviewed in 5]. The geographically and phylogentically widespread occurrence of AM is likely an adaptation of plants to scarcity and inaccessibility of mineral nutrients in natural habitats, considered particularly important on tropical soils, where rapid nutrient cycling leads to permanently low nutrient availability. The capacity of AM fungi to increase plant nutrition can also greatly contribute to sustainable agriculture with reduced chemical fertilizer input [6, 7].

To establish symbiosis, AM fungi colonize the root interior. This involves developmental changes of plant cells as well as fungal hyphae. Root colonization is preceded by mutual recognition via diffusible molecules released by both symbionts, culminating in differentiation of the fungal tip growing hypha into a lense-shaped hyphopodium for docking at the root surface (Figure 1) [8, 9]. The fungus then penetrates the outer root cell layers, spreads longitudinally within the root cortex and forms branched tree-shaped hyphal structures, the arbuscules, in cortical cells [10, 11•]. The formation of intracellular fungal structures and the extent of root colonization are dynamically regulated by the plant likely to optimize symbiotic benefit according to the plants physiological and developmental status that results from environmental conditions [12, 13, 14]. To achieve this control, the plants cellular remodeling program, which involves transient restructuring of already differentiated root cells to accommodate the fungus, must be intertwined with molecular mechanisms that control plant development and physiology. Such interconnecting mechanisms are largely unknown.

Phytohormones act as determinants of single cell fate decisions as well as systemic integrators of physiological and developmental states. They are involved in essentially all plant developmental processes and play a particularly important role in developmental responses to abiotic and biotic external signals. Evidence is accumulating that they are also pivotal for AM formation. During this process they do not only influence plant cell behavior but some also have an impact on the fungus. Hence, phytohormones might emerge as important interconnectors between AM and whole plant development. Here available evidence for phytohormone function  except ethylene, jasmonate (JA) and salicylic acid (SA) [15, 16]  in AM development will be discussed.

Strigolactones (SLs) are carotenoid-derived plant hormones involved in seed germination, suppression of shoot branching and in shaping root architecture [17, 18, 19, 20, 21]. Under phosphate or nitrogen limiting conditions plants exude elevated amounts of SLs into the rhizosphere [22, 23], where they trigger developmental responses of AM fungi such as spore germination, mitosis, hyphal growth and hyphal branching (Figure 1) [24, 25, 26]. Thus, SLs probably enhance the chance of fungal encounter with the host root by activating fungal growth and by providing a directional cue. Plant mutants impaired in SL production or export support this assumption as they display reduced levels of AM colonization, but morphologically intact intraradical fungal structures [17, 27, 28, 29]. In addition, SLs seem to stimulate fungal production of some chitin oligomers: chitin tetramers and pentamers increased by 38-fold and 69-fold, respectively in the supernatant of germinating Rhizophagus irregularis spores treated with the synthetic SL analogue GR24, while the amount of chitin trimers and hexamers increased by only 4.6-fold and 15-fold, respectively [30]. This raises the possibility that the production of chitin tetramers and pentamers represents a specific response to SLs and is not a simple side effect of enhanced fungal growth. Chitin tetramer and pentamers but not the other chitin oligomers induced nuclear calcium spiking in the rhizodermis a well-described symbiotic root response [30]. Thus, their perception by the plant might increase fungal success in colonizing the root.

The structural requirements for fungal responses to SLs differ from those of plants (Figure 2); and synthetic SL analogs have been developed that suppress shoot branching of plants but do not induce hyphal branching of AM fungi [31]. The known naturally occurring SLs consist of a tricyclic lactone (A-C ring) that is connected via an enol-ether bond with a methylbutenolide (D-ring; [Figure 2). Genetic studies in plants implement the α/β-fold hydrolase protein D14/DAD2 and the F-box subunit of the ubiquitin E3 ligase complex MAX2/D3/RMS4 in SL perception [reviewed in 32]. Evidence has been collected that the SL molecule is cleaved via the Ser-His-Asp catalytic triad of the α/β-fold hydrolase D14 resulting in a free hydroxylated D-ring [33, 34, 35]. A D14 crystal structure binding the D-OH hydrolysis product has been solved [34]. However, it is unclear whether D-OH alone can act as the endogenous signaling molecule since a modest suppression of shoot branching by exogenous application of D-OH required concentrations that were orders of magnitude higher than those needed of the synthetic SL analogue GR24 [33, 34, 36]. Structure-function studies revealed that at low applied concentrations the minimum structural requirements for shoot branching repression are the D-ring, the enol-ether or a thio-ether bond and a Michael-acceptor [36], raising the possibility that the cleavage reaction itself is important for signal transduction [34]. For the induction of hyphal branching in AM fungi the tricyclic lactone as well as the methylbutenolide are needed [37]. This indicates that the fungal SL receptor is different from D14 and possibly not an α/β-fold hydrolase protein. Thus, SL perception in plants and AM fungi is most likely a result of convergent evolution.

SL production in roots is controlled by the GRAS proteins NODULATION SIGNALING PATHWAY1 (NSP1) and NSP2. They are both required for full expression of DWARF27 (D27), that encodes a β-carotene isomerase which catalyzes the first committed step of SL biosynthesis [38, 39]. Colonization levels of Lotus japonicus nsp1 and Medicago truncatula nsp1 and nsp2 mutants are lower than of wild-type, while arbuscule morphology is normal (Figure 3a [40, 41, 42]). These phenotypes seem consistent with a role of SL in activating the fungus before physical contact with the root. However, colonization of L. japonicus nsp1 mutants could not be restored by exogenous application of GR24 [41]. Thus, NSP1 likely performs additional functions in AM development. NSP1 and NSP2 have been shown to interact in yeast and in tobacco leaves (Figure 3b) [43]. Whether this interaction is necessary to promote D27 expression remains to be determined.

Given the well-established role of exuded SLs in activating fungal growth before colonization it was a surprising discovery that part of the putative SL receptor complex, namely MAX2/D3/RMS4 is required for AM development in planta [44•, 45]. Rice dwarf3 (d3) mutant roots, block colonization at the rhizodermis. The fungus forms aberrant hyphopodia at the root surface and only very rarely penetrates into the inner cell layers. The few arbuscules that develop as a consequence of these rare penetration events have wild type-like appearance suggesting that MAX2/D3/RMS4-mediated signaling is needed in the rhizodermis rather than the cortex [44]. The AM phenotype of a pea ramosus4 (rms4) mutant was similar to rice d3. However, it differed quantitatively among independent greenhouse experiments, indicating, that the requirement for RMS4 might depend on environmental conditions or fungal inoculum strength [45]. Interestingly, AM colonization was independent of D14 and d14 rice mutants were even more strongly colonized than wild type [44]. This calls for an alternative receptor protein, which interacts with MAX2/D3/RMS4 during AM development and either binds SL or another small molecule.

Changes in root auxin content upon AM colonization or increases in AM colonization upon treatment of roots with auxin or auxin transport inhibitors have been observed several times during the last thirty years [reviewed in 46]. Genetic evidence for a promotion of AM development by auxin signaling is only recently emerging. Pea and tomato mutants with phenotypic aberrancies in indole-acetic acid (IAA) biosynthesis, transport or signaling display reduced levels of AM colonization but no defects in the development of fungal structures [47, 48]. Low AM colonization of the IAA deficient bushy mutant of pea was attributed to insufficient expression of the strigolactone biosynthesis gene CAROTENOID CLEAVAGE DIOXIGENASE 8 (CCD8) and to the resulting reduction in SL biosynthesis and exudation [47, 49]. In fact, addition of GR24 could partially restore colonization of bushy [47]. However, a causal relationship between IAA levels, strigolactone exudation and AM colonization was not established and it remains unknown whether IAA application would restore strigolactone exudation and/or AM colonization of bushy. The nature and the genomic location of the bushy mutation are unknown and the exact cause of reduced CCD8 expression and incomplete AM colonization remains to be determined.

AM assays with the tomato mutants diageotropica (dgt) and polycotyledon (pct) revealed striking differences in AM development between whole plants and root organ culture (ROC, [48]). The dgt mutant is defective in IAA signaling due to a mutation in a CYCLOPHILIN gene. ROCs of dgt did not support AM development and on agar plates the fungal hypha grew away from the root. By contrast, ROCs of pct that carries a mutation in an unknown locus and suffers from increased IAA transport were colonized more rapidly than wild type ROCs [48]. Surprisingly, when intact plants of both mutants were subjected to AM fungi they displayed an equivalent phenotype: both were colonized but root length colonization was reduced by 50% as compared to wild type plants. This striking discrepancy with their phenotype in ROCs might be caused by additive effects of the mutations with alterations in hormone balance that typical occur in Agrobacterium rhizogenes transformed hairy roots. Alternatively, it might indicate that a systemic component, for example, shoot derived auxin or another shoot factor mitigates the effects that the dgt and pct mutations have in ROCs.

Novel data support an additional and possibly cell-autonomous function of auxin signaling during the development of arbuscules. The IAA reporter DR5-GUS was specifically activated in arbuscule-containing cells of tomato, M. truncatula and rice indicating an elevated auxin-response associated with arbuscule development. Furthermore, overexpression of miR393, which targets transcripts of the the IAA receptor TIR1/AFB, caused an arrest in arbuscule formation in all three species (Etemadi, Gutjahr, Combier et al., unpublished), suggesting that TIR1/AFB-mediated IAA perception is required for arbuscule development (Figure 1). During arbuscule development root cortex cells become strongly polarized, as the cytoskeleton reorganizes and a distinct membrane domain, the peri-arbuscular membrane, forms and surrounds the arbuscule [50, 51]. Since IAA application can stimulate cytoskeletal rearrangement and local IAA maxima can lead to a TIR1-depedent re-polarization of cells [52, 53, 54] it is tempting to speculate that TIR1/AFB-dependent IAA signaling mediates cytoskeleton rearrangement and polarization of cortex cells during arbuscule formation.

DELLA proteins belong to the GRAS protein family and act as repressors of gibberellin (GA) signaling thereby restricting plant growth [55]. Three independent studies have now shown that DELLA proteins are required for AM development [56••, 57, 58•]. A remarkable phenotype was observed in roots of an M. truncatula della1 della2 double mutant: arbuscule numbers were strongly reduced while root length colonization was normal and intraradical hyphae seemed to proliferate even more than in the wild type [56••]. The few arbuscules, which were able to form in della1 della2 developed to full maturity, indicating that DELLA proteins are required for the initiation of arbuscule formation but not for later stages of arbuscule development (see [11] for a review on stages of arbuscule development). However, the M. truncatula genome contains three DELLA genes and it is unknown how perturbation of all three DELLA genes would affect AM. Indeed, the rice mutant slender rice1 (slr1), which is entirely DELLA-deficient showed a severe and general reduction in intraradical AM colonization (Figure 3a) while hyphopodium formation was unaffected [58]. It is possible that a della triple mutant of Medicago would display a similarly severe phenotype. According to Southern and Western blots the pea genome contains two DELLA encoding genes LA and CRY [57, 59], however, since the pea genome is not sequenced the presence of additional DELLA genes cannot be formally excluded. AM development in la cry-s double mutant roots was generally reduced but reduction of arbuscules was most severe resulting in a ratio of arbuscules to intraradical hyphae that resembled the phenotype of the Medicago della1/della2 double mutant [57]. The range of phenotypes observed in Medicago, pea and rice points to a dual role of DELLA proteins in promoting colonization of the outer cell layers as well as arbuscule formation in the cortex (Figure 1). This assumption is consistent with a previous study showing a dose dependent inhibition of AM development by GA-treatment of pea where lower GA-concentrations inhibited the formation of arbuscules while higher concentrations fully suppressed colonization [60].

Treatment of M. truncatula roots with exogenous GA phenocopied the della1 della2 mutant. Arbuscule formation upon GA treatment could be restored by transgenic expression of a non-degradable DELLA version (della1-Δ18) which lacks the DELLA-domain [56••]. This supports a role of DELLA in suppressing GA signaling to enable arbuscule formation. Congruously, AM colonization of the rice GA-receptor mutant gibberellin insensitive dwarf1 (gid1) was also resistant to GA-treatment [58]. The M. truncatula DELLA1 promoter-GUS fusion was predominantly activated in the vasculature and endodermis and only sporadically in the cortex. Interestingly, expression of della1-Δ18 driven by the vasculature-specific PHOSPHATE TRANSPORTER 9 (PT9) promoter sufficed to restore arbuscule formation in the della1 della2 mutant [56••]. Assuming that the PT9 promoter is really vasculature-specific, this surprising finding hints at a DELLA-dependent factor that moves from the vasculature to the cortex to promote arbuscule formation. As cell-to-cell movement is a well-known feature of GRAS proteins [61], the translocated factor could be DELLA itself (Figure 4a).

AM development depends on a conceptual signaling network called common SYM signaling that is also required for root nodule symbiosis [62]. This signal transduction cascade translates the perception of microbial signaling molecules at the plasmamembrane into nuclear calcium-spiking [62]. A nuclear localized calcium calmodulin dependent kinase CCAMK/DMI3 likely interprets the calcium signal and then phosphorylates the transcription factor CYCLOPS/IPD3 that activates the transcription of genes required for symbiosis [63••]. AM development in L. japonicus ccamk mutants is blocked at the rhizodermis while cyclops mutants permit colonization of the inner cortex by intraradical hyphae but not by arbuscules (reviewed in [11]). The phenotypic resemblance of cyclops with the della1 della2 mutant points to the possibility that DELLA and CYCLOPS act in the same pathway. Indeed, arbuscule formation in an L. japonicus cyclops mutant was restored by transgenic expression of della1-Δ18 [56••], suggesting that CYCLOPS might activate processes which lead to the stabilization of DELLA (Figure 4b). However, colonization of ccamk roots was not restored. Therefore CCAMK must activate at least one additional factor that is required for arbuscule development (Figure 4b). Since ABA treatment has been shown to stabilize DELLA proteins in the presence of GA [64] one mechanism by which CYCLOPS could promote DELLA stability could be through encouragement of ABA signaling (Figure 1). Consistent with this idea, the tomato mutant sitiens, which is deficient in ABA biosynthesis, displays a reduced number of arbuscules and a reduction in arbuscule viability [65]. However, DELLA stability in the sitiens background has not been examined.

Another way by which CYCLOPS could contribute to DELLA persistence would be by activating a stabilizing interaction partner. A yeast two-hybrid screen identified a novel GRAS protein called DELLA INTERACTING PROTEIN1 (DIP1) that interacts with the rice DELLA SLR1 [58]. DIP1 expression is induced by AM colonization. However, it is unknown whether this induction is CYCLOPS-dependent. Nevertheless, DIP1 is required for AM development as DIP1 RNAi reduced AM colonization to a similar degree as the slr1 mutant (Figure 3a) [58]. Interestingly, DIP1 also interacts with REQUIRED FOR ARBUSCULAR MYCORRHIZA (RAM1, [Figure 3b), another GRAS protein that had been described earlier to be required for hyphopodium formation at the root surface and for arbuscule branching [58•, 66, 67]. It remains to be seen whether SLR1, DIP1 and RAM1 unite in a ternary complex or whether DIP1 forms alternative complexes with SLR1 and RAM1. Other open questions relate to the tissue-specific localization of putative GRAS protein complexes in mycorrhizal roots and to whether CYCLOPS influences DELLA stability in a cell-autonomous or non-cell-autonomous manner.

The plant nutrient status plays an important role in regulating AM development. Especially the effect of high phosphate conditions on AM development has been well described [13, 14]. However, the mechanisms of how plants control AM according to their nutrient status are unknown. A role of phytohormones in root system architecture responses to phosphate starvation has been well established [68]. It thus represents an attractive hypothesis that phytohormones act as interconnectors between the host physiological status and AM development. IAA signaling has been implicated in phosphate starvation responses [69, 70, 71]. Arabidopsis plants grown under Pi starvation display an increased IAA sensitivity due to elevated expression of the auxin receptor gene TIR1 [71]. Arbuscule development might be inhibited at high phosphate because TIR1 expression is not sufficiently high (Etemadi, Gutjahr, Combier et al., unpublished). How TIR1 expression is regulated according to phosphate level is unknown and it would be interesting to interrogate the role of miR393 in this process. Similarly, DELLA proteins have been implicated in phosphate starvation responses and at high phosphate GA levels increase and DELLAs are destabilized [72]. Therefore DELLA/GA could be an ideal signaling module to integrate arbuscule numbers with the phosphate status of the plant [56••]. Although SL exudation is highly sensitive to plant nutrient status [23] addition of GR24 to phosphate sufficient plants did not restore AM colonization [13, 14]. Thus, deficient AM colonization in plants with high Pi status cannot be explained by insufficient SL exudation and resulting insufficient activation of the fungus before colonization. However, it is possible that high Pi also reduces AM-crucial SL signaling inside the root that might not be restored by exogenous addition of GR24 but might require a natural, structurally different SL.

Section snippets

Conclusions and outlook

Phytohormone signaling in AM development is an emerging area of research. We know that most phytohormones regulate AM development but we currently lack insight into most molecular players involved, into the targets of and into possible interactions between hormone pathways. The use of increasing genetic resources for AM model plants coupled with smart reporter assays and physiological experiments will close this gap and can be predicted to soon provide an understanding of how plants orchestrate

References and recommended reading

Papers of particular interest, published within the period of review, have been highlighted as:

  • • of special interest

  • •• of outstanding interest

Acknowledgements

I thank Andreas Binder for improving the design of the figures and Jean-Philippe Combier for permitting the description of unpublished results. Research is supported by the collaborative research center SFB924 ‘Molecular mechanisms regulating yield and yield stability in plants’ of the German Research Foundation (DFG) and scientific exchange by the COST action FA1206 ‘Strigolactones biological roles and applications’ (STREAM).

References (76)

  • N. Pumplin et al.

    Live-cell imaging reveals periarbuscular membrane domains and organelle location in Medicago truncatula roots during arbuscular mycorrhizal symbiosis

    Plant Physiol

    (2009)
  • E. Gobbato et al.

    A GRAS-type transcription factor with a specific function in mycorrhizal signaling

    Curr Biol

    (2012)
  • E. Gobbato et al.

    RAM1 and RAM2 function and expression during arbuscular mycorrhizal symbiosis and Aphanomyces euteiches colonization

    Plant Signal Behav

    (2013)
  • C. Shen et al.

    OsARF16, a transcription factor, is required for auxin and phosphate starvation response in rice (Oryza sativa L.)

    Plant Cell Environ

    (2013)
  • A. Robert-Seilaniantz et al.

    Hormone crosstalk in plant disease and defense: more than just jasmonate-salicylate antagonism

    Annu Rev Phytopathol

    (2011)
  • B. Wang et al.

    Phylogenetic distribution and evolution of mycorrhizas in land plants

    Mycorrhiza

    (2006)
  • B. Bago et al.

    Carbon metabolism and transport in arbuscular mycorrhizas

    Plant Physiol

    (2000)
  • S.E. Smith et al.

    Roles of arbuscular mycorrhizas in plant nutrition and growth: new paradigms from cellular to ecosystem scales

    Annu Rev Plant Biol

    (2011)
  • S. Gianinazzi et al.

    Agroecology: the key role of arbuscular mycorrhizas in ecosystem services

    Mycorrhiza

    (2010)
  • C. Gutjahr et al.

    Cell and developmental biology of the arbuscular mycorrhiza symbiosis

    Annu Rev Cell Dev Biol

    (2013)
  • R.T. Koide et al.

    Regulation of the vesicular-arbuscular mycorrhizal symbiosis

    Annu Rev Plant Physiol Plant Mol Biol

    (1992)
  • C. Balzergue et al.

    The regulation of arbuscular mycorrhizal symbiosis by phosphate in pea involves early and systemic signalling events

    J Exp Bot

    (2010)
  • F. Breuillin et al.

    Phosphate systemically inhibits development of arbuscular mycorrhiza in Petunia hybrida and represses genes involved in mycorrhizal functioning

    Plant J

    (2010)
  • C. Gutjahr et al.

    Weights in the balance: jasmonic acid and salicylic acid signaling in root–biotroph interactions

    Mol Plant-Microbe Interact

    (2009)
  • B. Hause et al.

    The role of jasmonates in mutualistic symbioses between plants and soil-born microorganisms

    Phytochemistry

    (2009)
  • M. Umehara et al.

    Inhibition of shoot branching by new terpenoid plant hormones

    Nature

    (2008)
  • C. Ruyter-Spira et al.

    Physiological effects of the synthetic strigolactone analog GR24 on root system architecture in Arabidopsis: another belowground role for strigolactones?

    Plant Physiol

    (2011)
  • Y. Kapulnik et al.

    Strigolactones affect lateral root formation and root-hair elongation in Arabidopsis

    Planta

    (2011)
  • D.C. Nelson et al.

    F-box protein MAX2 has dual roles in karrikin and strigolactone signaling in Arabidopsis thaliana

    Proc Natl Acad Sci USA

    (2011)
  • K. Yoneyama et al.

    Phosphorus deficiency in red clover promotes exudation of orobanchol, the signal for mycorrhizal symbionts and germination stimulant for root parasites

    Planta

    (2007)
  • K. Yoneyama et al.

    Nitrogen and phosphorus fertilization negatively affects strigolactone production and exudation in sorghum

    Planta

    (2013)
  • K. Akiyama et al.

    Plant sesquiterpenes induce hyphal branching in arbuscular mycorrhizal fungi

    Nature

    (2005)
  • A. Besserer et al.

    Strigolactones stimulate arbuscular mycorrhizal fungi by activating mitochondria

    PloS Biol

    (2006)
  • A. Besserer et al.

    GR24, a synthetic analog of strigolactones, stimulates the mitosis and growth of the arbuscular mycorrhizal fungus Gigaspora rosea by boosting its energy metabolism

    Plant Physiol

    (2008)
  • W. Kohlen et al.

    The tomato CAROTENOID CLEAVAGE DIOXYGENASE8 (SlCCD8) regulates rhizosphere signaling, plant architecture and affects reproductive development through strigolactone biosynthesis

    New Phytol

    (2012)
  • C. Gutjahr et al.

    The half-size ABC transporters STR1 and STR2 are indispensable for mycorrhizal arbuscule formation in rice

    Plant J

    (2012)
  • T. Kretzschmar et al.

    A petunia ABC protein controls strigolactone-dependent symbiotic signalling and branching

    Nature

    (2012)
  • A. Genre et al.

    Short-chain chitin oligomers from arbuscular mycorrhizal fungi trigger nuclear Ca2+ spiking in Medicago truncatula roots and their production is enhanced by strigolactone

    New Phytol

    (2013)
  • Cited by (155)

    • Signaling in mycorrhizal symbioses

      2023, The Chemical Dialogue between Plants and Beneficial Microorganisms
    View all citing articles on Scopus
    View full text