Logo of plosbiolPLoS BiologySubmit to PLoSGet E-mail AlertsContact UsPublic Library of Science (PLoS)View this Article
PLoS Biol. Mar 2008; 6(3): e68.
Published online Mar 4, 2008. doi:  10.1371/journal.pbio.0060068
PMCID: PMC2270324

Functional Adaptation of a Plant Receptor- Kinase Paved the Way for the Evolution of Intracellular Root Symbioses with Bacteria

Detlef Weigel, Academic Editor

Abstract

Nitrogen-fixing root nodule symbioses (RNS) occur in two major forms—Actinorhiza and legume-rhizobium symbiosis—which differ in bacterial partner, intracellular infection pattern, and morphogenesis. The phylogenetic restriction of nodulation to eurosid angiosperms indicates a common and recent evolutionary invention, but the molecular steps involved are still obscure. In legumes, at least seven genes—including the symbiosis receptor-kinase gene SYMRK—are essential for the interaction with rhizobia bacteria and for the Arbuscular Mycorrhiza (AM) symbiosis with phosphate-acquiring fungi, which is widespread in occurrence and believed to date back to the earliest land plants. We show that SYMRK is also required for Actinorhiza symbiosis of the cucurbit Datisca glomerata with actinobacteria of the genus Frankia, revealing a common genetic basis for both forms of RNS. We found that SYMRK exists in at least three different structural versions, of which the shorter forms from rice and tomato are sufficient for AM, but not for functional endosymbiosis with bacteria in the legume Lotus japonicus. Our data support the idea that SYMRK sequence evolution was involved in the recruitment of a pre-existing signalling network from AM, paving the way for the evolution of intracellular root symbioses with nitrogen-fixing bacteria.

Author Summary

As an adaptation to nutrient limitations in terrestrial ecosystems, most plants form Arbuscular Mycorrhiza (AM), which is a symbiotic relationship between phosphate-delivering fungi and plant roots that dates back to the earliest land plants. More recently, a small group including the legumes and close relatives has evolved the ability to accommodate nitrogen-fixing bacteria intracellularly. The resulting symbiosis is manifested by the formation of specialized root organs, the nodules, and comes in two forms: the interaction of legumes with rhizobia, and the more widespread Actinorhiza symbiosis of mostly woody plants with Frankia bacteria. The symbiosis receptor kinase SYMRK acts in a signalling pathway that legume plants require to trigger the development of nodules and the uptake of fungi or bacteria into their root cells. Here we show that the induction of Actinorhiza nodulation also relies on SYMRK, consistent with the idea that both types of nodulation evolved by recruiting common signalling genes from the pre-existing AM program. We observed that SYMRK from different land plant lineages differs significantly in exon composition, with a “full-length” version in the nodulating clade and shorter SYMRK genes in plants outside this lineage. Only the most complete SYMRK version was fully functional in nodulation, suggesting this gene played a central role in the recruitment event associated with the evolution of intracellular root symbioses with bacteria.

Introduction

Nitrogen limits plant growth in many terrestrial ecosystems. Evolutionary adaptations to this constraint include symbiotic associations with bacteria that are capable of converting atmospheric nitrogen into ammonium. Extracellular associations of plants with diverse groups of nitrogen-fixing bacteria are phylogenetically widespread, but only a small group evolved the ability to accommodate bacteria endosymbiotically inside cell wall boundaries. Bacterial symbionts are confined within tubular structures called infection threads, which are surrounded by a host-derived membrane that is continuous with the plasma membrane, and bound by plant cell wall–like material [1,2]. The bulk of host plants including all actinorhizal species retain the bacterial symbionts within these structures during the nitrogen-fixing stage of the symbiosis [1,3]. In the most advanced forms found exclusively among legumes (Fabales) and Gunnera [4], symbiotic bacteria are delimited from the host cell cytoplasm only by a plant-derived membrane in the mature stage of the symbioses. In the respective legumes, they develop into bacteroids contained in organelle-like symbiosomes, where nitrogen fixation takes place (for a recent review, see [5]). Bacterial endosymbioses in both legumes and actinorhizal plants are typically associated with the formation of novel plant organs, so-called nodules, which are root-derived in the majority of cases [6]. Nitrogen-fixing root nodule symbiosis (RNS) occurs in two major forms. Actinorhiza hosts belong to three eurosid orders (Figure 1) and nodulate with Gram-positive actinobacteria of the genus Frankia [7]. Legumes, on the contrary, enter specific interactions with members of a diverse group of Gram-negative bacteria, termed rhizobia. For almost a century, the extreme diversity in organ structure, infection mechanisms, and bacterial symbionts among nodulating plants obscured the fact that the nodulating clade is monophyletic, which was revealed by molecular phylogeny relatively recently [8]. The restriction of endosymbiotic root nodulation to a monophyletic group of four angiosperm orders (Figure 1) is coincident with a patchy occurrence within this clade. These observations led to the hypothesis that a genetic change acquired by a common ancestor may predispose members of this lineage to evolve nodulation endosymbiosis [8].

Figure 1
SYMRK Exon-Intron Structure and Root Endosymbiotic Abilities of Angiosperm Lineages

The molecular adaptations underlying the evolution of plant-bacterial endosymbioses are still a mystery, despite a substantial biotechnological interest in understanding the genetic differences between nodulating and non-nodulating plants. While the molecular communication between legumes and rhizobia has been studied in some detail, important clues are expected from the genetic analysis of the yet underexplored Actinorhiza.

Bacterial signalling molecules and corresponding plant receptors involved in RNS are known only for the legume–rhizobium interaction. Frankia signals may be biochemically distinct from rhizobial chito-oligosaccharide nodulation factors [9], which would suggest an independent mechanism of host–symbiont recognition in Actinorhiza.

Phenotypic analysis of legume mutants has revealed a genetic link between RNS and Arbuscular Mycorrhiza (AM), which is a phosphate-scavenging association between plant roots and fungi belonging to the phylum Glomeromycota [10]. AM is widespread among land plants, where forms of AM are found in representatives of all major lineages. Fossil evidence for ancient AM-like associations [11] suggests a role of this symbiosis in the colonization of land about 450 million years ago.

The link of plant–fungal and plant–bacterial endosymbioses in legumes, which involves at least seven genes [1216] termed “common symbiosis genes” [17], inspired the idea that during the evolution of bacterial endosymbiosis, genes were recruited from the pre-existing AM genetic program [18]. However, the molecular steps involved are not clear.

Results

To gain insight into the evolution of nitrogen-fixing root nodulation, we analysed common symbiosis genes across angiosperm lineages with different symbiotic abilities. Many, including the calcium/calmodulin kinase gene CCaMK [14,19], or genes encoding the predicted cation channels CASTOR and POLLUX [12,20,21], are conserved in overall domain structure. We discovered exceptional diversification among genes encoding the symbiosis receptor kinase SYMRK in different species (Figure 1). While putative SYMRK kinase domains are conserved and contain characteristic sequence motifs discriminating them from related kinases (Figure S1), the predicted extracellular portion of SYMRK occurs in at least three versions of domain composition (Figure 1 and Table 1). The longest SYMRK version is present in all tested eurosids, including nodulating and non-nodulating lineages. Comprising 15 exons, it encodes three leucine-rich repeat (LRR) motifs and an extended N-terminal domain of unknown function (NEC-domain, Figure 1 and Table 1). Outside of the eurosid clade, which encompasses all nodulating groups, one or more exons are absent from SYMRK coding sequences (Figure 1 and Table 1).

Table 1
SYMRK Homologs, Conceptual SYMRK Proteins, and the Closest Arabidopsis Sequences

Genetic evidence indicates that SYMRK acts near a point of molecular convergence of AM and legume-rhizobium signalling [16,22]. The presumed ability of its diverged extracellular domain to perceive symbiosis-related signals [16] renders it a prime target for investigating the molecular adaptations underlying the evolution of RNS.

The homogenous occurrence of “full-length” SYMRK genes among legumes, actinorhizal plants, and non-nodulating eurosids raises the intriguing possibility that SYMRK is involved in the proposed genetic predisposition [8] of this clade to evolve nodulation. An important prediction following from this hypothesis is the common requirement of a full-length SYMRK version for all types of RNS. Furthermore, also non-nodulating members of this monophyletic clade may carry nodulation-competent versions of SYMRK. To test this concept, we analysed the functional capabilities of “full-length” SYMRK genes from symbiotically diverse eurosids.

SYMRK Is Required for Actinorhiza and AM in Datisca glomerata

To investigate SYMRK function in Actinorhiza, we reduced root mRNA levels of D. glomerata (Datisca) SYMRK (DgSYMRK) via RNA-interference (RNAi). Quantitative PCR following reverse transcription showed a 36%–99% reduction of DgSYMRK transcript levels in knockdown roots (n = 16) compared with vector control roots (n = 16). Eight weeks after inoculation with Frankia bacteria, no nodules were detected on DgSYMRK RNAi roots (Figure 2A and and2B),2B), except for small, primordial swellings on 16% of independent transformed roots (9/55). Nonsilenced control roots of the same plants and roots transformed with a binary vector lacking the silencing cassette (transgenic control roots) showed wild type–like nodules with lobed structure typical for Datisca (Figure 2A and and2B).2B). This result demonstrates that SYMRK is essential for Actinorhiza development in Datisca. In conjunction with the well-documented role of legume SYMRK in the interaction with rhizobia [16,23], SYMRK thus represents a common genetic requirement for the two types of bacterial root endosymbiosis.

Figure 2
Nodulation and AM Formation Are Impaired in DgSYMRK Knockdown Roots

To test whether DgSYMRK is also required for AM, we inspected DgSYMRK RNAi roots for AM formation with the fungus Glomus intraradices (Glomus). Datisca wild-type roots of the same plants used for hairy root induction and independent transgenic control roots formed AM, with dense arbuscular colonization of inner cortical cells (Figure 2C–2F). In contrast, symbiotic development in DgSYMRK RNAi roots was strongly impaired. In 82% of independent transformed roots, no fungal infection was observed, despite the presence of extensive extraradical mycelium (Figure 2G), with those roots exhibiting strong reduction levels of DgSYMRK being nonsymbiotic concerning both nodulation and AM formation. Occasional infection attempts occurred but typically were aborted in the outer cell layers (Figure 2G and and2H).2H). We conclude that similar to the situation in legumes, SYMRK of the actinorhizal plant Datisca is involved in both bacterial and fungal endosymbioses.

SYMRK Does Not Mediate Specificity in Legume–Rhizobium Recognition

To determine whether SYMRK plays a role in the specific recognition of rhizobia by legume hosts, we tested whether L. japonicus (Lotus) SYMRK (LjSYMRK) can mediate nodulation in another legume, which interacts with a different rhizobial partner. The specific symbiont of Lotus is Mesorhizobium loti, whereas Medicago truncatula (Medicago) interacts with Sinorhizobium meliloti. Medicago dmi2 5P mutants exhibit a deletion in exon three of the SYMRK ortholog DMI2, leading to a frameshift and premature stop codon. Dmi2 5P plants form no infection threads or nodules upon inoculation with either rhizobial strain. Transgenic roots of these plants, and of wild-type control plants carrying LjSYMRK, formed infection threads and indeterminate, pink nodules typical for Medicago [24] with S. meliloti (Figure S2 and Table 2). LjSYMRK can therefore fully restore nodulation of Medicago with S. meliloti, indicating that SYMRK is not directly involved in determining legume–rhizobium specificity.

Table 2
Restoration of Root Symbioses in Medicago dmi2 and Lotus symrk Mutants Transformed with Different SYMRK Versions

Medicago dmi2 5P mutants are also impaired in AM. No arbuscules were observed within 2 wk of co-cultivation, with fungal infection being aborted at the root surface or after entry into epidermal cells (Figure S2 and Table 2). LjSYMRK restored the AM defect in transgenic roots of this line (Figure S2 and Table 2), demonstrating that SYMRK is sufficiently similar to DMI2 to support both fungal and bacterial endosymbioses in Medicago.

“Full-Length” SYMRK Versions from Symbiotically Distinct Eurosids Can Support Both AM and RNS in Lotus

To analyze the symbiotic capabilities of “full-length” eurosid SYMRK genes from a legume (MtDMI2), an actinorhizal plant (DgSYMRK), and the non-nodulating, AM-forming Tropaeolum majus (Tropaeolum; Brassicales) (TmSYMRK), we tested their potential to function in the Lotus symbiosis signalling context. We introduced these genes, under the control of the Lotus SYMRK promoter region, into roots of Lotus line SL1951–6 (symrk-10), which carries a symrk mutant allele encoding a kinase-dead SYMRK version [25,26]. Upon inoculation with Glomus, symrk-10 roots form no AM, and fungal infections are typically associated with aberrant hyphal swellings and are aborted after entry into epidermal cells (Figure 3A and and3B,3B, and Table 2). Interaction with M. loti is blocked at an early stage, and no infection threads or nodules form (Figure 3C and and3D,3D, and Table 2). In symrk-10 roots transformed with MtDMI2, DgSYMRK or TmSYMRK both AM and nodulation were restored, the latter involving the formation of infection threads and pink, bacteria-containing nodules (Table 2 and Figure 3) that were indistinguishable from wild-type nodules. In conclusion, consistent with a role of SYMRK in the predisposition to evolve RNS, we could not detect a functional diversification of the eurosid SYMRK version linked to features differentiating actinorhizal or legume nodulation, or to the specific recognition of bacterial symbionts. Thus, other factors, such as nod factor receptor kinases [22,27,28] or yet-unknown additional components, are likely accountable for the fine-tuning of recognition specificity in plant–bacterial endosymbioses within the eurosids.

Figure 3
Datisca, Medicago, and Tropaeolum SYMRK Restore Fungal and Bacterial Endosymbioses in Lotus symrk-10 Mutants

SYMRK Versions of Reduced Length Restore AM but Not RNS in Lotus

SYMRK from the non-nodulating eudicots Papaver rhoeas (poppy) and Lycopersicon esculentum (tomato) represent intermediate length and domain composition (Figure 1 and Table 1). To explore the symbiotic capabilities of this version, we introduced the two-LRR encoding LeSYMRK genomic sequence fused to the LjSYMRK promoter into Lotus symrk-10 transgenic roots. LeSYMRK selectively restored AM symbiosis, whereas neither infection threads nor nodules developed upon inoculation with M. loti (Table 2 and Figure S3E–S3L).

A reduced SYMRK version is present in the AM-forming, non-nodulating monocots Oryza sativa (rice) and Zea maize (maize), encoding two LRRs only and a short N-terminal region, mainly covered by a single exon aligning with exon four of LjSYMRK (Figure 1 and Figure S1). To determine whether the short SYMRK version of rice, OsSYMRK, is sufficient for endosymbiosis formation in Lotus, we introduced the OsSYMRK genomic sequence controlled by the LjSYMRK promoter into symrk-10 roots. AM formation was fully restored in these roots, whereas nodulation with M. loti was impaired (Table 2 and Figure S3M–S3Z). OsSYMRK-containing symrk-10 roots inoculated with M. loti exhibited primordial swellings generally devoid of bacteria (Figure S3U–S3X). In rare cases, infection threads and small round nodules were observed, which contained bacterial colonies (Figure S3Y and S3Z). Hence, similar to LeSYMRK, OsSYMRK is compromised in supporting bacterial endosymbiosis in Lotus.

Discussion

A Molecular Link between the Two Types of RNS

In legumes, SYMRK is indispensable for root endosymbiosis with rhizobia and AM fungi [16,23]. We show here that this endosymbiosis gene is also required for nodulation in the actinorhizal plant Datisca. SYMRK, which is likewise essential for Actinorhiza formation of the tree species Casuarina glauca (Fagales) [29], represents the first known plant gene required for Actinorhiza, indicating a shared genetic basis of the two different types of RNS. A future task will be to determine whether further endosymbiosis genes acting in concert with SYMRK in legumes are also required for Actinorhiza.

An Ancient Genetic Program for AM among Angiosperms

The ability of different SYMRK versions from both dicot and monocot lineages to restore AM in Lotus indicates a homologous nature of the AM genetic program in angiosperms. This is consistent with the observation that loss-of-function mutations in the rice version of the legume symbiosis gene CCaMK results in loss of AM symbiosis [30]. In Arabidopsis, the absence of root symbiotic capability is accompanied by a deletion of several symbiosis genes, including SYMRK and CCaMK [21,31,32].

A Role of SYMRK in the Predisposition to Evolve RNS

Our survey of SYMRK sequences across angiosperms revealed at least three structurally distinct versions, and we could show that this polymorphism is functionally related to the root symbiotic capabilities of host plants. The variation in SYMRK domain composition is exceptional among the known common symbiosis genes. The congruence between the phylogenetic distribution of the “full-length” SYMRK version with the nodulating clade strongly suggests a link between SYMRK sequence evolution and the acquisition of endosymbiotic root nodulation with bacteria. An attractive hypothesis is that SYMRK sequence divergence was a critical step in mediating the recruitment of the otherwise conserved common symbiosis pathway from the pre-existing AM genetic program. Recruitment was proposed to account for the genetic link of AM and nodulation in legumes [17,18] and would make root–bacterial endosymbiosis as a whole a fascinating example for novel traits evolving on the basis of pre-existing genetic patterns.

A common feature associated with endosymbiotic bacterial infection in both actinorhizal [33] and legume hosts [34] is the formation of intracellular pre-infection threads (PITs) in host cells. These cytoplasmic structures resemble the pre-penetration apparatus (PPA) preceding fungal infection during AM formation [35]. Forming in anticipation of bacterial symbionts, PITs are thought to coordinate the uptake of bacteria and determine the spatial progression of infection through the host cell [33,34]. A similar role in guiding fungal transition through host cells in AM has been demonstrated for PPAs [35]. These developmental similarities in AM, Actinorhiza, and legume-rhizobium infection may reflect a common genetic program for endosymbiosis establishment and symbiont uptake in all three types of interactions. In AM, PPAs are not formed in mutants that are defective in certain common symbiosis genes [35]. It is therefore possible that a recruitment of AM symbiosis genes during the evolution of RNS facilitated the induction of intracellular accommodation structures in response to bacteria.

SYMRK Domain Function and Evolution

Repetitive LRR modules have been implicated in the determination and evolution of novel recognition specificities of receptor proteins [3638]. Interestingly, adaptive changes reflecting positive selective constraints can be traced in LRR– and NEC–encoding regions of SYMRK genes from different Medicago species, but these do not correlate with shifts in rhizobial specificity [39]. Our functional comparison of eurosid SYMRK versions indicates that SYMRK is not involved in determining recognition specificity in nodulation. However, an extended SYMRK version containing a set of three LRR motifs, as present in eurosid SYMRK genes, is required for fully supporting nodulation symbiosis of Lotus with M. loti. Shorter SYMRK versions from tomato or rice only suffice for AM. These functional differences may be caused by individual amino acid sequence polymorphisms, or alternatively, exons that are specifically required for bacterial endosymbiosis may be lacking in rice and tomato SYMRK versions.

At an overall structural level, exon acquisition from other genes encoding LRR or NEC-like domains [23,40] or, alternatively, retainment of exons in eurosid SYMRK genes, may have been an integral genetic factor in the evolution of bacterial endosymbiosis in angiosperms. The observation of small nodule-like structures on Lotus symrk mutant roots transformed with the OsSYMRK construct is counterintuitive, considering that the LeSYMRK version, which resembles the legume version more closely, does not support such developmental responses. One possible explanation may be that the nonmatching NEC region of LeSYMRK negatively interferes with nodulation, but not AM signalling in Lotus.

The NEC domain encoded by Lotus SYMRK exons two and three, upstream of the conserved LRR flanking region (CEC), is present across eudicot plants (Figure 1). Its function outside the nodulating group is unknown. The proposed involvement of SYMRK in processes such as reduction of the touch sensitivity of root hairs [41] may rely on this domain thereby imposing selective constraints. The NEC domain shows possible overall relatedness but only a low level of similarity to sequences present in the rice genome, and to sequences other than SYMRK candidates in genomes of dicots like Arabidopsis [23]. The apparent divergence observed among these potentially homologous sequences of yet unknown function is consistent with a hypothetical role as a receptor domain.

It will be a future challenge to determine the contribution of individual SYMRK LRR units as well as of the NEC domain and to resolve at the amino acid level the features of SYMRK proteins involved in conferring endosymbiotic nodulation capacity.

Additional Components Required for Nodulation

The diversity and scattered occurrence of nodulation symbioses within the eurosid lineage suggest multiple independent origins [42]. Only a subset of the plant species carrying the “full-length” version of SYMRK develop root nodules, yet SYMRK of the non-nodulating Tropaeolum proved competent to support nodulation in Lotus. Hence, there must be additional genetic features distinguishing the nodulators. Candidate genes include those that express the legume LysM receptor kinases NFR1 and NFR5 [22,27,43], which are required for responsiveness to rhizobial lipo-chito-oligosaccharide nodulation factors, but not for AM formation. A potential relevance of LysM receptors in Actinorhiza, or the identity of alternative receptors perceiving yet unknown Frankia signals, remains to be determined.

Materials and Methods

Isolation of SYMRK homologues.

We used a PCR strategy employing degenerate primers to obtain SYMRK sequence information from diverse angiosperms, for which no genome or root-derived expressed sequence tag sequences were available. Degenerate primers for the isolation of SYMRK genes were positioned in regions of the coding sequence conserved among SYMRK candidates, but not in other similar O. sativa (rice) and A. thaliana (Arabidopsis) sequences. For primer sequences, see Table S1.

λ Zap cDNA libraries were available for isolation of Ly. esculentum (tomato) and Alnus glutinosa (alder) SYMRK. A cosmid clone carrying the LeSYMRK genomic region was isolated from a pooled tomato Cf2/9 library (kind gift of J.D.G. Jones, The Sainsbury Laboratory, United Kingdom) and shotgun sequenced.

For rapid amplification of cDNA ends (RACE) reactions, total RNA was extracted from roots of uninoculated seedlings or young plants and DNaseI treated. RT and 5′/3′RACE reactions were done using the SMART RACE kit (Clontech), following nested degenerate PCR reactions ([10 s 94 °C, 10 s 52 °C, 30 s 72 °C] × 35, 5 min 72 °C) to obtain initial sequence information.

Construct generation for mutant complementation and Datisca SYMRK RNAi experiments.

For hairy root complementation assays, SYMRK cDNAs were amplified from root cDNA preparations (Superscript II, Invitrogen) of the respective species. Binary transformation vectors were pCAMBIA 1302 or pK7WG2D,1 [44] derivatives. L. japonicus (Lotus), Me. truncatula (Medicago), D. glomerata (Datisca), and T. majus (Tropaeolum) SYMRK coding sequences were amplified from complementary DNA using primers LjSYMRK_EC_f with LjSYMRK_PK_r, MtSYMRK_EC_f with MtSYMRK_PK_r, DgSYMRK_EC_f with DgSYMRK_PK_r and TmSYMRK_EC_f with TmSYMRK_PK_r (Table S1), respectively. An MtSYMRK genomic segment containing intron one was amplified from total DNA using primers MtSYMRK_EC_f with MtSYMRK_01_r (Table S1) and ligated to the 3′ fragment of the MtSYMRK cDNA following BstB1 digestion of both. The genomic sequences of tomato and rice SYMRK were amplified from total DNA using primers LeSYMRK_EC_f with LeSYMRK_PK_r and OsSYMRK_EC_f with OsSYMRK_PK_r (Table S1), respectively. A 285-bp fragment amplified with primers polyA_NOS_f and polyA_NOS_r (Table S1) from pJawohl8 RNAi (kind gift of P. Schulze-Lefert, Max Planck Institute for Plant Breeding Research, Cologne, Germany) was used as terminater in all constructs. SYMRK genes or coding sequences were under control of 4,970 bp of genomic sequence preceding the LjSYMRK translation start site. For pK7WG2D constructs, the cauliflower mosaic virus (CaMV) 35S promoter driving the Gateway-cassette was deleted Sal1(10124)-Sal1(38).

For Datisca RNAi experiments, a pRedRoot- [45] based binary vector encoding DsRED1 for visible selection was equipped with a pKANNIBAL [46] CaMV 35S promoter-driven hairpin construct containing 367-bp of 3′ coding and untranslated sequence of DgSYMRK in forward and reverse orientation, amplified with primers DgSYMRK_RNAi_f and DgSYMRK_RNAi_r (Table S1).

Biological material and phenotyping assays.

L. japonicus wild-type ecotype B-129 Gifu and derived mutant line symrk-10 [25] were inoculated with M. loti R7A as described [16]. Growth conditions were 24 °C constant at 16-h-light/8-h-dark cycles. For infection of Agrobacterium rhizogenes–induced transgenic roots in Lotus, M. loti MAFF 303099 expressing DsRED was applied at a final optical density at 600 nm (OD600) of 0.02 in translucent plastic boxes containing 300 ml Seramis (Mars) and 150 ml liquid Fahraeus Plant (FP) medium [47]. Ag. rhizogenes-transformed Me. truncatula A17 wild type and dmi2 5P (kind gift of G. Oldroyd, John Innes Centre, United Kingdom) plants were inoculated with S. meliloti 1021 carrying pBHR-mRFP [48] (OD600 of 0.02) in planting pots containing Seramis, and fertilized with FP medium supplemented with 50 μM KNO3 two times per week. For nodulation assays and simultaneous observation of infection thread (IT) formation in Lotus and Medicago, plants were harvested 4 and 5 wk post inoculation, respectively. Prior to inoculation, roots showing no eGFP fluorescence were removed. For AM phenotyping of Lotus and Medicago, plants were co-cultivated with G. intraradices BEG195 and harvested after 3 or 2 wk of co-cultivation, respectively. Root systems were stained with acidic ink as described [49]. Prior to staining of Ag. rhizogenes–induced root systems, roots showing no eGFP fluorescence were removed. Roots were scored AM-positive (AM+) if symbiotic structures (arbuscules and vesicles) were present, as AM-negative (AM–) if no arbuscules were present. Occasional Lotus symrk mutant roots showing vesicles not accompanied by arbuscules were scored AM–. Where complemented Lotus symrk mutant roots exhibited aborted infection sites in co-occurrence with successful infection and colonization events involving arbuscule and vesicle formation, roots were scored AM+. Datisca seeds and Frankia inoculum were a kind gift from K. Pawlowski (Department of Botany, Stockholm University, Sweden). Datisca was inoculated with compatible Frankia by potting in substrate with ~1 g/l crushed nodules and with G. intraradices BEG195 by adding substrate extracted from pots of inoculated Allium schoenoprasum plants. Growth conditions were 16 h light/8 h dark at 22 °C and 60% relative humidity. Seeds of T. majus and P. rhoeas were purchased at Notcutts Garden Centres (UK). The ability to develop AM with G. intraradices was confirmed for all species involved in the study.

Lotus transformation.

Transgenic roots on Lotus symrk-10 mutants were induced using Ag. rhizogenes strains AR1193 [50] and LBA1334 [51] as described by Díaz et al. [52] (modified).

Medicago transformation.

Medicago seedlings were transformed as described at http://www.isv.cnrs-gif.fr/embo01/manuels/index.html (modified), using strain Ag. rhizogenes AR1193 [50].

Datisca transformation.

Twelve-wk-old Datisca plants were inoculated with Ag. rhizogenes strain LBA1334 [51] carrying the silencing construct by stem injection, and roots emerging at infection sites were covered with substrate. Three-wk post inoculation roots were inspected for DsRED1 fluorescence. Nonfluorescent roots were removed, and plants were repotted and grown for 8 wk. After determination of the nodulation phenotype, individual fluorescent roots were divided into two halves. One half was stained for mycorrhiza visualization, the second used for total RNA extraction (RNeasy Plant Kit, Qiagen). Quantitative RT-PCR was performed with GeneAmp5700 (Applied Biosystems) using the SuperScript III Platinum Two-Step qRT-PCR-Kit (Invitrogen). A 123-bp DgSYMRK fragment was amplified using primers DgqPCR_SYMRK_f with DgqPCR_SYMRK_r (Table S1). As control, polyubiquitin cDNA was amplified using primers DgqPCR_Ubi_f with DgqPCR_Ubi_r (Table S1). Representative fragments were sequenced for identity confirmation.

Supporting Information

Figure S1

SYMRK Kinase Regions Share Several Defining Conserved Amino Acid Motifs, Which Are Absent in Similar Sequences in Arabidopsis and Rice:

Black shading indicates amino acid residues identical in all sequences, residues found in at least 50% of the sequences are shaded gray. Bars delimit predicted SYMRK protein domains. Dark blue, conserved extracellular region (CEC); black, LRRs; gray, imperfect LRR; white, juxtamembrane regions; brown, transmembrane region; green, protein kinase domain. Light blue bars with stars mark some of the regions conserved among SYMRK candidates, but not in other homologous sequences in rice and A. thaliana. Locus tags are indicated for similar sequences not regarded as SYMRK candidates. Sequences aligning with regions upstream of exon 4 of LjSYMRK are not included.

(96 KB DOC)

Figure S2

LjSYMRK Restores Nodulation and AM Formation in Medicago dmi2 5P Mutants:

Transformation assay and selection were as in Figure 3. (A, B, K, and L) Me. truncatula (Medicago) dmi2 5P roots transformed with the respective control vector lacking an LjSYMRK expression cassette.

(C–F, M, and N) Medicago wild-type and (G–J, O and P) dmi2 5P roots transformed with the LjSYMRK coding sequence controlled by the LjSYMRK promoter.

(A–J) Roots inoculated with S. meliloti expressing DsRED for 5 wk. (A and B) Transgenic dmi2 5P roots carrying the control vector, showing no nodules. (C and D) Nodules on transgenic and nontransgenic roots of a wild-type plant transformed with LjSYMRK and (E and F) individual nodule containing DsRED expressing S. meliloti bacteria. (G and H) dmi2 5P root system with nodule formation confined to roots transformed with LjSYMRK. (I and J) Nodule on an LjSYMRK containing dmi2 5P root showing bacterial DsRED expression.

(K–P) Roots co-cultivated with G. intraradices for 2 wk. (K and L) Transgenic dmi2 5P control roots lacking hyphal proliferation and arbuscule formation in the inner root cortex. Hyphal swellings in the root periphery (L and arrow in K) indicate abortion of fungal infections. Longer co-cultivation for 3 wk or more allowed for successful fungal infections of mutant roots, which was similarly reported for other dmi2 mutant lines [58]. (M and N) Wild-type and (O and P) dmi2 5P roots transformed with LjSYMRK showing dense fungal colonization of the root inner cortex accompanied by arbuscule formation.

Scale bars: (A–D and G–H) 2 mm; (E–F and I–J) 0.5 mm; (K, M, and O) 0.1 mm; (L, N, and P) 0.02 mm.

(1.8 MB PDF)

Figure S3

Tomato and Rice SYMRK Restore AM Symbiosis in Lotus symrk-10 Mutants, but Cannot or Only Partially Complement Bacterial Endosymbiosis Formation:

Transformation assay and selection were as in Figure 3. (A–D) Lotus symrk-10 roots transformed with the respective control vector lacking a SYMRK expression cassette. (E–L) Lotus wild-type (E–H) and symrk-10 (I–L) roots transformed with LeSYMRK. (M–Z) Lotus wild-type (M–R) and symrk-10 (S–Z) roots transformed with OsSYMRK.

(A, B, E, F, I, J, M, N, S, and T) Roots co-cultivated with G. intraradices for three weeks. (A and B) Transgenic symrk-10 control root with extraradical mycelium but no intraradical fungal colonization or arbuscule formation. Swollen hyphal structures indicative of aborted fungal infections can be observed within epidermal cells (B and arrow in A). (E and F) Wild-type and (I and J) symrk-10 roots transformed with LeSYMRK, showing fungal colonization of the inner root cortex (E, I) and arbuscule formation in inner cortical cells (F, J). (M and N) Wild-type and (S and T) symrk-10 mutant roots transformed with OsSYMRK, similarly showing cortical AM colonization (M, S) and arbuscule formation (N, T).

(C, D, G, H, K, L, O–R, and U–Z) Root systems inoculated with M. loti expressing DsRED for 4 wk. (C and D) symrk-10 root system with transgenic control roots, showing no nodules. (G and H) and (O–R) Wild-type root systems with M. loti–containing pink nodules on nontransgenic and on transgenic roots carrying LeSYMRK or OsSYMRK, respectively, indicating that these transgenes do not impair nodulation in transgenic wild-type roots. (K and L) symrk-10 root system transformed with LeSYMRK, showing no nodules. In a single case, one nodule primordium was observed. (U–Z) symrk-10 root system transformed with OsSYMRK, showing no fully developed nodules, but nodule primordia which are mostly noncolonized by bacteria, the latter proliferating on the primordial surface (W and X). In rare cases, small nodules were observed that contained bacteria, but, with one exception, showed no pinkish coloration in white light (Y and Z).

Scale bars: (A, E, I, M, and S) 0.1 mm; (B, F, J, N, and T) 0.02 mm; (C, D, G, H, K, L, O, P, U, V) 2 mm; (Q, R, and W–Z) 0.5 mm.

(3.8 MB PDF)

Table S1

Primer Sequences:

(106 KB DOC)

Accession Numbers

Sequences of SYMRK homologs were deposited at the EMBL Nucleotide Sequence Database (http://www.ebi.ac.uk/embl/) under accession numbers AY935263 (Al. glutinosa); AM271000, AM931079 (D. glomerata coding and genomic sequence, respectively); AY935267 (Lupinus albus); AY935265 (T. majus); AY935266, AY940041 (Ly. esculentum coding and genomic sequence, respectively); AM270999 (P. rhoeas); AM851092 (Po. trichocarpa). The GenBank (http://www.ncbi.nlm.nih.gov/Genbank/index.html) accession number for pCAMBIA 1302 is AF234298.

Acknowledgments

We are grateful to K. Pawlowski (Department of Botany, Stockholm University, Sweden) for supplying D. glomerata seeds, Frankia inoculum, and an Al. glutinosa cDNA library, and for discussion. We thank M. Bucher (Institute of Plant Sciences, ETH Zürich, Switzerland) and J.D.G. Jones (The Sainsbury Laboratory, United Kingdom) for access to tomato cDNA and cosmid libraries, respectively, and P. Schulze-Lefert (Max Planck Institute for Plant Breeding Research, Cologne, Germany) for supplying the plasmid pJawohl8 RNAi. We gratefully obtained Me. truncatula dmi2 mutant seeds from G. Oldroyd (John Innes Centre, United Kingdom) and Me. truncatula A17 wild-type seed material from T. Wang (John Innes Centre, United Kingdom). M. Hayashi (University of Munich, Germany) and S. Goormachtig (Ghent University, Belgium) kindly provided DsRED expressing strains of M. loti MAFF and Sinorhizobium meliloti 1021, respectively. We thank J.A. Gossmann for making the Tropaeolum SYMRK complementation construct, and G. Strobel for technical assistance.

Abbreviations

AM
Arbuscular Mycorrhiza
LRR
leucine-rich repeat
NEC
N-terminal extracellular region of predicted SYMRK proteins
PIT
pre-infection thread
PPA
pre-penetration apparatus
RNS
root nodule symbiosis
RACE
rapid amplification of cDNA ends
RNAi
RNA interference

Footnotes

¤ Current address: Quintiles, Budapest, Hungary

Author contributions. KM, GG and MP conceived and designed the experiments, performed the experiments, and analyzed the data. KM and MP wrote the paper.

Funding. Research at the Sainsbury Laboratory is funded by the Gatsby Charitable Foundation.

Competing interests. The authors have declared that no competing interests exist.

References

  • Berg RH. Frankia forms infection threads. Can J Bot. 1999;77:1327–1333.
  • Brewin NJ. Plant cell wall remodelling in the rhizobium-legume symbiosis. Crit Rev Plant Sci. 2004;23:293–316.
  • Sprent J. Nodulation in legumes. Kew: Royal Botanical Gardens; 2001.
  • Uheda E, Silvester WB. The role of papillae during the infection process in the Gunnera-Nostoc symbiosis. Plant Cell Physiol. 2001;42:780–783. [PubMed]
  • Prell J, Poole P. Metabolic changes of rhizobia in legume nodules. Trends Microbiol. 2006;14:161–168. [PubMed]
  • Pawlowski K, Sprent J. Comparison between actinorhizal and legume symbiosis. In: Pawlowski K, Newton WE, editors. Nitrogen-fixing actinorhizal symbioses. Dordrecht: Springer Netherlands; 2008. pp. 261–288.
  • Swensen SM, Benson DR. Evolution of actinorhizal host plants and Frankia endosymbionts. In: Pawlowski K, Newton WE, editors. Nitrogen-fixing actinorhizal symbioses. Dordrecht: Springer Netherlands; 2008. pp. 73–104.
  • Soltis DE, Soltis PS, Morgan DR, Swensen SM, Mullin BC, et al. Chloroplast gene sequence data suggest a single origin of the predisposition for symbiotic nitrogen fixation in angiosperms. Proc Natl Acad Sci U S A. 1995;92:2647–2651. [PMC free article] [PubMed]
  • Ceremonie H, Debelle F, Fernandez MP. Structural and functional comparison of Frankia root hair deforming factor and rhizobia Nod factor. Can J Bot. 1999;77:1293–1301.
  • Schüßler A, Schwarzott D, Walker C. A new fungal phylum, the Glomeromycota: phylogeny and evolution. Mycological Research. 2001;105:1413–1421.
  • Remy W, Taylor TN, Haas H, Kerp H. Four hundred-million year-old vesicular arbuscular mycorrhizae. Proc Natl Acad Sci U S A. 1994;91:11841–11843. [PMC free article] [PubMed]
  • Imaizumi-Anraku H, Takeda N, Charpentier M, Perry J, Miwa H, et al. Plastid proteins crucial for symbiotic fungal and bacterial entry into plant roots. Nature. 2005;433:527–531. [PubMed]
  • Kanamori N, Madsen LH, Radutoiu S, Frantescu M, Quistgaard EM, et al. A nucleoporin is required for induction of Ca2+ spiking in legume nodule development and essential for rhizobial and fungal symbiosis. Proc Natl Acad Sci U S A. 2006;103:359–364. [PMC free article] [PubMed]
  • Levy J, Bres C, Geurts R, Chalhoub B, Kulikova O, et al. A putative Ca2+ and calmodulin-dependent protein kinase required for bacterial and fungal symbioses. Science. 2004;303:1361–1364. [PubMed]
  • Saito K, Yoshikawa M, Yano K, Miwa H, Uchida H, et al. NUCLEOPORIN85 is required for calcium spiking, fungal and bacterial symbioses, and seed production in Lotus japonicus. Plant Cell. 2007;19:610–624. [PMC free article] [PubMed]
  • Stracke S, Kistner C, Yoshida S, Mulder L, Sato S, et al. A plant receptor-like kinase required for both bacterial and fungal symbiosis. Nature. 2002;417:959–962. [PubMed]
  • Kistner C, Parniske M. Evolution of signal transduction in intracellular symbiosis. Trends Plant Sci. 2002;7:511–518. [PubMed]
  • LaRue T, Weeden N. The symbiosis genes of the host. In: Kiss GB, Endre G, editors. Proceedings of the 1st European Nitrogen Fixation Conference. Szeged, Hungary: Officina Press; 1994. pp. 147–151.
  • Patil S, Takezawa D, Poovaiah BW. Chimeric plant calcium/calmodulin-dependent protein kinase gene with a neural visinin-like calcium-binding domain. Proc Natl Acad Sci U S A. 1995;92:4897–4901. [PMC free article] [PubMed]
  • Ane JM, Kiss GB, Riely BK, Penmetsa RV, Oldroyd GE, et al. Medicago truncatula DMI1 required for bacterial and fungal symbioses in legumes. Science. 2004;303:1364–1367. [PubMed]
  • Zhu H, Riely BK, Burns NJ, Ane JM. Tracing non-legume orthologs of legume genes required for nodulation and arbuscular mycorrhizal symbioses. Genetics. 2006;172:2491–2499. [PMC free article] [PubMed]
  • Radutoiu S, Madsen LH, Madsen EB, Felle HH, Umehara Y, et al. Plant recognition of symbiotic bacteria requires two LysM receptor-like kinases. Nature. 2003;425:585–592. [PubMed]
  • Endre G, Kereszt A, Kevei Z, Mihacea S, Kalo P, et al. A receptor kinase gene regulating symbiotic nodule development. Nature. 2002;417:962–966. [PubMed]
  • Timmers AC, Auriac MC, Truchet G. Refined analysis of early symbiotic steps of the Rhizobium-Medicago interaction in relationship with microtubular cytoskeleton rearrangements. Development. 1999;126:3617–3628. [PubMed]
  • Perry JA, Wang TL, Welham TJ, Gardner S, Pike JM, et al. A TILLING reverse genetics tool and a web-accessible collection of mutants of the legume Lotus japonicus. Plant Physiol. 2003;131:866–871. [PMC free article] [PubMed]
  • Yoshida S, Parniske M. Regulation of plant symbiosis receptor kinase through serine and threonine phosphorylation. J Biol Chem. 2005;280:9203–9209. [PubMed]
  • Madsen EB, Madsen LH, Radutoiu S, Olbryt M, Rakwalska M, et al. A receptor kinase gene of the LysM type is involved in legume perception of rhizobial signals. Nature. 2003;425:637–640. [PubMed]
  • Radutoiu S, Madsen LH, Madsen EB, Jurkiewicz A, Fukai E, et al. LysM domains mediate lipochitin-oligosaccharide recognition and Nfr genes extend the symbiotic host range. Embo J. 2007;26:3923–3935. [PMC free article] [PubMed]
  • Gherbi H, Markmann K, Svistoonoff S, Estevan J, Autran D, et al. SymRK defines a common genetic basis for plant root endosymbioses with AM fungi, rhizobia and Frankia bacteria. Proc Natl Acad Sci U S A. 2008. In press. [PMC free article] [PubMed]
  • Chen C, Gao M, Liu J, Zhu H. Fungal symbiosis in rice requires an ortholog of a legume common symbiosis gene encoding a Ca2+/calmodulin-dependent protein kinase. Plant Physiol. 2007;145:1619–1628. [PMC free article] [PubMed]
  • Kevei Z, Seres A, Kereszt A, Kalo P, Kiss P, et al. Significant microsynteny with new evolutionary highlights is detected between Arabidopsis and legume model plants despite the lack of macrosynteny. Mol Genet Genomics. 2005;274:644–657. [PubMed]
  • Zhu H, Choi HK, Cook DR, Shoemaker RC. Bridging model and crop legumes through comparative genomics. Plant Physiol. 2005;137:1189–1196. [PMC free article] [PubMed]
  • Berg RH. Cytoplasmatic bridge formation in the nodule apex of actinorhizal root nodules. Can J Botany. 1999;77:1351–1357.
  • van Brussel AAN, Bakhuizen R, Spaink HP, Tak T, Lugtenberg BJJ, et al. Induction of pre-infection thread structures in the leguminous host plant by mitogenic lipo-oligosaccharides of Rhizobium. Science. 1992;257:70–72. [PubMed]
  • Genre A, Chabaud M, Timmers T, Bonfante P, Barker DG. Arbuscular mycorrhizal fungi elicit a novel intracellular apparatus in Medicago truncatula root epidermal cells before infection. Plant Cell. 2005;17:3489–3499. [PMC free article] [PubMed]
  • Parniske M, Hammond-Kosack KE, Golstein C, Thomas CM, Jones DA, et al. Novel disease resistance specificities result from sequence exchange between tandemly repeated genes at the Cf-4/9 locus of tomato. Cell. 1997;91:821–832. [PubMed]
  • Shiu SH, Karlowski WM, Pan R, Tzeng YH, Mayer KF, et al. Comparative analysis of the receptor-like kinase family in Arabidopsis and rice. Plant Cell. 2004;16:1220–1234. [PMC free article] [PubMed]
  • van der Hoorn RA, Wulff BB, Rivas S, Durrant MC, van der Ploeg A, et al. Structure-function analysis of Cf-9, a receptor-like protein with extracytoplasmic leucine-rich repeats. Plant Cell. 2005;17:1000–1015. [PMC free article] [PubMed]
  • De Mita S, Santoni S, Ronfort J, Bataillon T. Adaptive evolution of the symbiotic gene NORK is not correlated with shifts of rhizobial specificity in the genus Medicago. BMC Evol Biol. 2007;7:210. [PMC free article] [PubMed]
  • Dolan J, Walshe K, Alsbury S, Hokamp K, O'Keeffe S, et al. The extracellular Leucine-Rich Repeat superfamily; a comparative survey and analysis of evolutionary relationships and expression patterns. BMC Genomics. 2007;8:320. [PMC free article] [PubMed]
  • Esseling JJ, Lhuissier FG, Emons AM. A nonsymbiotic root hair tip growth phenotype in NORK-mutated legumes: implications for nodulation factor-induced signaling and formation of a multifaceted root hair pocket for bacteria. Plant Cell. 2004;16:933–944. [PMC free article] [PubMed]
  • Doyle JJ, Doyle JL, Ballenger JA, Dickson EE, Kajita T, et al. A phylogeny of the chloroplast gene rbcL in the Leguminosae: taxonomic correlations and insights into the evolution of nodulation. Am J Botany. 1997;84:541–554. [PubMed]
  • Limpens E, Franken C, Smit P, Willemse J, Bisseling T, et al. LysM domain receptor kinases regulating rhizobial Nod factor-induced infection. Science. 2003;302:630–633. [PubMed]
  • Karimi M, Inze D, Depicker A. GATEWAY vectors for Agrobacterium-mediated plant transformation. Trends Plant Sci. 2002;7:193–195. [PubMed]
  • Limpens E, Ramos J, Franken C, Raz V, Compaan B, et al. RNA interference in Agrobacterium rhizogenes-transformed roots of Arabidopsis and Medicago truncatula. J Exp Bot. 2004;55:983–992. [PubMed]
  • Wesley SV, Helliwell CA, Smith NA, Wang MB, Rouse DT, et al. Construct design for efficient, effective and high-throughput gene silencing in plants. Plant J. 2001;27:581–590. [PubMed]
  • Fahraeus G. The infection of clover root hairs by nodule bacteria studied by a simple glass slide technique. J Gen Microbiol. 1957;16:374–381. [PubMed]
  • Smit P, Raedts J, Portyanko V, Debelle F, Gough C, et al. NSP1 of the GRAS protein family is essential for rhizobial Nod factor-induced transcription. Science. 2005;308:1789–1791. [PubMed]
  • Demchenko K, Winzer T, Stougaard J, Parniske M, Pawlowski K. Distinct roles of Lotus japonicus SYMRK and SYM15 in root colonization and arbuscule formation. New Phytol. 2004;163:381–392.
  • Stougaard J, Abildsten D, Marcker KA. The Agrobacterium rhizogenes pRi TL-DNA segment as a gene vector system for transformation of plants. Mol Gen Genet. 1987;207:251–255.
  • Offringa IA, Melchers LS, Regensburg-Tuink AJG, Costantino P, Schilperoort RA, et al. Complementation of Agrobacterium tumefaciens tumor-inducing aux mutants by genes from the TR-region of the Ri plasmid of Agrobacterium rhizogenes. Proc Natl Acad Sci U S A. 1986;83:6935–6939. [PMC free article] [PubMed]
  • Diaz CL, Gronlund M, Schlaman HRM, Spaink HP. Induction of hairy roots for symbiotic gene expression studies. In: Márquez AJ, editor. Lotus japonicus handbook. Berlin: Springer; 2005. pp. 261–277.
  • The Angiosperm Phylogeny Group. An update of the Angiosperm Phylogeny Group classification for the orders and families of flowering plants: APG II. Bot J Linn Soc. 2003;141:399–436.
  • Zhang LB, Simmons MP, Kocyan A, Renner SS. Phylogeny of the Cucurbitales based on DNA sequences of nine loci from three genomes: implications for morphological and sexual system evolution. Mol Phylogenet Evol. 2006;39:305–322. [PubMed]
  • Paszkowski U, Jakovleva L, Boller T. Maize mutants affected at distinct stages of the arbuscular mycorrhizal symbiosis. Plant J. 2006;47:156–173. [PubMed]
  • Kikuchi S, Satoh K, Nagata T, Kawagashira N, Doi K, et al. Collection, mapping, and annotation of over 28,000 cDNA clones from japonica rice. Science. 2003;301:376–379. [PubMed]
  • Shiu SH, Bleecker AB. Receptor-like kinases from Arabidopsis form a monophyletic gene family related to animal receptor kinases. Proc Natl Acad Sci U S A. 2001;98:10763–10768. [PMC free article] [PubMed]
  • Morandi D, Prado E, Sagan M, Duc G. Characterisation of new symbiotic Medicago truncatula (Gaertn.) mutants, and phenotypic or genotypic complementary information on previously described mutants. Mycorrhiza. 2005;15:283–289. [PubMed]

Articles from PLoS Biology are provided here courtesy of Public Library of Science
PubReader format: click here to try

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...