TY - JOUR
AU1 - Peleg-Grossman, Smadar
AU2 - Volpin, Hanne
AU3 - Levine, Alex
AB - Abstract The symbiotic relationships between legumes and rhizobacteria involve extensive signalling between the two organisms. Studies using genetic, biochemical, and pharmacological approaches have demonstrated the involvement of calcium and reactive oxygen species in the establishment of symbiotic interactions. In the early stage of the interactions rhizobia grow as infection thread within host root hairs and are internalized into the plant cells via endocytosis. It is shown here that inoculation of Medicago truncatula roots with Sinorhizobium meliloti induced a battery of vesicle trafficking genes, including the phosphatidylinositol 3-kinase (PI3K) gene that stimulated plasma membrane endocytosis and the production of reactive oxygen species (ROS). Inhibition of the PI3K suppressed the membrane endocytosis and subsequent oxidative burst and prevented root hair curling and formation of infection threads. Similar effects were produced by inhibition of PtdIns-specific phospholipase C (PI-PLC). However, neither inhibition of PI3K nor PI-PLC signalling blocked cytosolic Ca2+ influx or early nodulin (ENOD) gene expression. By contrast, the inhibitors induced ENODs transcription in the absence of Rhizobium, suggesting that the expression of ENODs responds to plasma membrane perturbations. In summary, the results show a major reprogramming of intracellular vesicle trafficking during the early stages of symbiotic interactions that co-ordinate the host responses. Activation of parallel signalling pathways leading to Cacyt2+ influx and ROS production that regulate the root hair curling and ENODs expression are also shown. Endocytosis, phosphatidylinositol-3 kinase, reactive oxygen species Introduction An exclusive feature of plants from the legume family is their ability to establish a symbiotic relationship with rhizobacteria. The symbiotic interactions are initiated by flavonoids exuded by the legume roots that induce biosynthesis of lipo-chito-oligosaccharide signalling molecules in Rhizobium species, termed nodulation factors (NFs), which mediate the nodulation process in the receptive host (Smit et al., 2005). The NF perception in legume roots induces expression of nodulation genes and causes curling of root hairs that entrap the rhizobacteria, leading to formation of infection threads and development of a new organ, a nodule, in the cortex. When the progressing infection thread enters the cortical cells, the bacteria bud off into the plant cytoplasm and become enveloped by the host plant membrane (Oldroyd and Downie, 2004). The NF-induced signal transduction in the legumes has been studied using genetic, biochemical and pharmacological approaches (Catoira et al., 2000; Charron et al., 2004). The root cell plasma membrane has emerged as a critical organelle in the regulation of symbiotic interactions. It is positioned at the cross-roads between the signals from the two organisms and harbours the NF receptors. Binding of NF to the plasma membrane receptors activates intracellular calcium influx. A major path leading to cytosolic Ca2+ influx in plants in response to various stimuli, including NFs, is mediated by inositol 1,4,5 triphosphate (InsP3), which is generated by the activity of a phosphatidylinositide-specific PLC (PI-PLC) on the cytosolic side of the plasma membrane (DeWald et al., 2001). In vitro analysis of isolated plasma membranes from Vigna unguiculata showed that NFs induced the PI-PLC activity, suggesting its participation in the NF-triggered signal transduction (Kelly and Irving, 2001). The PI-PLC hydrolyses the glycerophosphate ester linkage of phospholipids to diacylglycerol and phosphorylated head groups, including the PtdIns head group (Laxalt and Munnik, 2002). Recent pharmacological evidence implicated the PI-PLC also in induction of early nodulin (ENOD) gene expression (Pingret et al., 1998; Charron et al., 2004). The Ca2+-dependent signalling in plant–microbe interactions is closely associated with the generation of reactive oxygen species (ROS). The opposite is also true, as H2O2 treatment induced cytosolic Ca2+ influx (Levine et al., 1996; Klusener et al., 2002). ROS production is a major mechanism of plant defence to pathogens (Lamb and Dixon, 1997). Lately, it was shown that ROS are produced during symbiotic interactions as well (Santos et al., 2001; Ramu et al., 2002). Pharmacological analysis of ROS production during symbiosis showed that it was inhibited by diphenyleneiodonium (DPI), implicating the NADPH oxidase (gp91phox) or a similar flavin-containing enzyme, in agreement with the oxidative burst in plant–pathogen interaction (Lamb and Dixon, 1997). During pathogen attack, the production of ROS occurs in two phases: first within minutes of bacteria inoculation, followed by a sustained oxidative burst which triggers hypersensitive cell death (Levine et al., 1994). During symbiotic interactions, the oxidative burst seems to be suppressed at later stages (Santos et al., 2001; Shaw and Long, 2003b). The NF-triggered ROS production in M. truncatula roots induced the peroxidase rip1 gene that is involved in the modulation of the root hair growth and infection thread formation (Ramu et al., 2002). After establishment of the infection thread, the rhizobacteria grow down inside the root hair and are released into the cortical cells surrounded by the plant membrane by an endocytotic process (Morrison and Verma, 1987; Hernandez et al., 2004; Puppo et al., 2005). Endocytosis is part of intracellular vesicle trafficking that delivers and retrieves materials to/from various organelles membranes in all eukaryotes (Samaj et al., 2004). The membrane components bud off the endoplasmic reticulum or the plasma membranes, forming vesicles that fuse with target membranes, releasing their contents. During the endocytosis of rhizobia the symbiotic micro-organism is enclosed by a host-derived membrane (Hernandez et al., 2004). This vesicle trafficking is regulated by small GTPases of the Rab family and by phosphatidylinositide signalling molecules (Jurgens, 2004; Wang, 2004). The intracellular vesicle trafficking is highly conserved among eukaryotes. A key enzyme in the initiation of endocytosis is the phosphatidylinositol-3 kinase (PI3K), which regulates many important cellular functions (Meijer and Munnik, 2003). The differentially phosphorylated phosphatidylinositides serve as the membrane-attachment sites for proteins with pleckstrin homology (PH) domains, which function in the membrane trafficking. Recently, proteins containing the PH domain were identified in Arabidopsis, and a PtdIns transfer protein with a nodulin domain was shown to function in the root hair membrane growth (Meijer and Munnik, 2003; Vincent et al., 2005). Here, the role of the vesicle trafficking system in the establishment of symbiotic interactions between the model legume Medicago truncatula and live rhizobacteria, Sinorhizobium meliloti, was studied using pharmacological and cell biology methods. The induction of genes associated with the initiation of vesicle trafficking, such as PI3K as well as other trafficking genes that corresponded to increased plasma membrane endocytosis and ROS production in host roots are shown. Inhibition of PI3K or PI-PLC activity suppressed membrane endocytosis and the oxidative burst, resulting in reduced curling of root hairs and infection thread formation. It is also shown that perturbations in PtdIns signalling affect ENODs gene expression, while ROS are required for DMI3 transcription. Materials and methods Biological material and plant treatment Medicago truncatula seeds were germinated and grown as described by Santos et al. (2001), except that the seeds were scarified for 5 min by exposure to concentrated sulphuric acid. For the pharmacological experiments the concentrations of the inhibitors that were added to the medium are listed in the Table 1. Sinorhizobium meliloti was grown as described in Santos et al. (2001). The Rhizobium expressing a variant of green fluorescent protein described by Jamet et al. (2003) was a generous gift from Alain Puppo (CNRS, Sofia-Antipolis, France). Table 1. List of inhibitors and their effects on Rhizobium infection Name  Concentration  Principle target  Inhibition  Effects  Wortmannin  25 μM  PtdIns-3 kinase  +  Root hair curling        +  Endocytosis        +  Nodulation        +  ROS production        –  Ca flux        –  Enod expression  U-73122  10 μM  Phospholipase C  +  Root hair curling        +  ROS production        –  Ca flux        –  Enod expression  LY294002  100 μM  PtdIns-3 kinase  +  Root hair curling        +  ROS production  TMB-8  50 μM  IP3 receptor store channels  –  Root hair curling        –  Rhizobial growth  Verapamil  20 μM  Voltage-operated calcium channels  +  Root hair curling        +  Rhizobial growth        +  Enod expression        –  DMI3 expression  DPI  12.5–20 μM  NADPH oxidase  –  Root hair curling        –  Enod expression        +  DMI3 expression  Name  Concentration  Principle target  Inhibition  Effects  Wortmannin  25 μM  PtdIns-3 kinase  +  Root hair curling        +  Endocytosis        +  Nodulation        +  ROS production        –  Ca flux        –  Enod expression  U-73122  10 μM  Phospholipase C  +  Root hair curling        +  ROS production        –  Ca flux        –  Enod expression  LY294002  100 μM  PtdIns-3 kinase  +  Root hair curling        +  ROS production  TMB-8  50 μM  IP3 receptor store channels  –  Root hair curling        –  Rhizobial growth  Verapamil  20 μM  Voltage-operated calcium channels  +  Root hair curling        +  Rhizobial growth        +  Enod expression        –  DMI3 expression  DPI  12.5–20 μM  NADPH oxidase  –  Root hair curling        –  Enod expression        +  DMI3 expression  View Large Analysis of endocytosis Root sections were incubated with a styryl membrane probe FM1-43 (Molecular Probes, Eugene, OR) for 16 min., washed twice in water, and photographed every 5 min. under an epi-fluorescent microscope (Olympus IX70) with a narrow excitation/broad band emission filter (Ex/Em cube XF25: 485DF22/OG535) from Omega Optical, Inc. (Brattleboro, VT). Pictures were taken with a Coolpix 950 camera (Nikon Corporation, Tokyo) using identical exposure settings for all shots in a series, as described by Mazel et al. (2004). The results were quantitated using ImagePro software (Media Cybernetics, Silver Spring, MD). Statistical significance analysis (P value) was assessed by the Student's t test. ROS and Ca2+ detection in plants Seedlings were taken from agar plates after 3 d, washed, and transferred to new plates with nitrogen-free medium with or without the inhibitors. ROS were detected with the epi-fluorescent microscope 16 h later by the addition of 10 μM 2′,7′-dichlorofluorescein as described by Mazel et al. (2004). Similar results were also obtained after 24 h (not shown). Ca2+ was detected by the incubation of seedlings with 25 μM FLUO4-AM (Invitrogen) essentially as described by DeWald et al. (2001), except that loading was done for 2 h at room temperature and the dye was washed out for 30 min in water, and the roots were photographed with the epi-fluorescent microscope using the same set-up as for the detection of ROS. The roots were also viewed in a MRC-1024 confocal microscope (Bio-Rad) with argon laser and 535±10 nm emission filter. The quantitation of the ROS and Ca2+ signals was done using the ImagePro software. Roots were photographed using the specific narrow-band 535±10 nm emission filter, and the pixels of mean density were collected from representative images for the statistical analysis (n=12). RT-PCR assay Total RNA was extracted from M. truncatula roots before and after S. meliloti inoculation. The roots were frozen in liquid nitrogen and total RNA was extracted using Tri Reagent (Molecular Research Center, Inc) and transcribed into cDNA using oligo-dT as a primer with SuperScript II reverse transcriptase (Invitrogen). The cDNA was amplified by PCR using Taq polymerase and the following primers: ENOD12: forward 5′-TTTTTCTTTTCTGCTCTTGTCCTTG-3’ and reverse, 5′-AATGGATGTTATCTTCTGCTGGAGGA-3′. ENOD16: forward, 5′-CTTCCATCACGACGTGCATTC-3’; and reverse, 5′-AGCAGCACCACCTTTATTACC-3′. ENOD20: forward, 5′-TGGGCCTCTAACTACCAATTC-3’; and reverse, 5′-AGCACCACCACCTTTACTAC-3′. rip1: forward, 5′-GTCGAATCTCGCCTTGT-3’; and reverse, 5′-GGCCCTGTTGTATCTTGTGC-3′. MTn21: forward, 5′-CTCCGTTCGCCATTATCTTC-3’; and reverse, 5′-CACCGCCATCACAAAAGTC-3′. PI3K: forward, 5′-TGAGAGAAAGTCCATCCAGAG-3’; and reverse, 5′-CCACATACCAGCGAAGAAAG-3′. The EF1a gene, forward 5′-TCACATCAACATTGTGGTCATTGGC-3’; and reverse, 5′-TTGATCTGGTCAAGAGCCTCAAG-3′, was used to normalize the amount of RNA. Expressed sequence tag (EST) analysis of vesicle trafficking associated genes in M. truncatula roots Gene Indices (version 8.0) were downloaded from TIGR (http://www.tigr.org/tigr-scripts/tgi/T_index.cgi?species=medicago). Gene Indices with the following keywords in their annotation (Arf, Rab, Dynamin, SNARE, phosphatidylinositol kinase/phosphatase and vesicle sorting) were selected for digital expression analysis based on EST libraries related to rhizobial, mycorrhizal, and root pathogen infections, as well as their respective controls. The expression levels are presented as normalized clone counts in a profile matrix that consists of 50 rows for the relevant M. truncatula gene over 14 columns representing the cDNA libraries. Each cell of the matrix represents the relative number of clones (actual clone counts were normalized relative to the total number of clones represented in each of the libraries) observed for the corresponding Gene Index in the library analysed. The following cDNA libraries were used for the analysis: Control: developing root Medicago truncatula, KV0, MtBA.Rhizobium infected: KV1(24 h), KV2 (948 h), KV3 (72 h), MtBB (96 h), NF-NR and GVN (mature nodules), GVSN (senescent nodules). Mycorrhiza: MTAMP, MtBC, MTGIM. Pathogen: DSIR (Phytophthora medicaginis). Results Induction of endocytosis in M. truncatula roots by S. meliloti To examine the possible involvement of intracellular vesicle trafficking in the early steps of symbiotic interactions, the expression of the PtdIns-3 kinase (PI3K) gene was analysed, which is the key enzyme in the regulation of membrane trafficking in eukaryotes (Leevers et al., 1999; Jurgens, 2004). Inoculation of M. truncatula roots with S. meliloti caused a strong induction of the PI3K gene (Fig. 1A). Treatment with wortmannin, which is a specific inhibitor of the PtdIns-3 (and/or –4) kinases did not inhibit the PI3K gene, indicating that its transcription is not self-regulated by vesicle trafficking. To analyse the expression of a large proportion of the trafficking genes during the early, intermediate, and late stages of the symbiotic interactions with S. meliloti, the available EST libraries of M. truncatula were investigated for trafficking-associated gene expression (Fig. 1B). Some of the genes (e.g. homologues of dynamin, vesicle-associated-membrane proteins [VAMPs or v-SNAREs], small GTPases of the Rab and ARF families, and phosphatidylinositol kinases) were highly expressed, especially in plants inoculated with S. meliloti. Interestingly, while some trafficking genes were also induced by infection with symbiotic mycorrhizal fungi, none of the genes was induced by infection with the pathogenic Phytophthora fungus. Fig. 1. View largeDownload slide Expression of vesicle trafficking-associated genes in M. truncatula roots during symbiotic interactions. (A) Semi-quantitative RT-PCR analysis of PI3K gene expression in M. truncatula roots during 4 d after inoculation (inoc) with S. meliloti. The top panel shows the PI3K gene expression 4 d after inoculation. Where indicated, the seedlings were pretreated for 24 h with 25 μM wortmannin (WM). (B) Compilation of expressed sequence tags (ESTs) in cDNA libraries of vesicle trafficking genes in M. truncatula roots from uninoculated controls (Dev_root, KV0, MtBA(N-starved), inoculated with S. meliloti (KV1, KV2, KV3, MtBB, Nod_root, GVN, GVSN), inoculated with mycorrhiza (MTAMP, MtBC, MTGIM), or with Phytophthora medicaginis (DSIR). Fig. 1. View largeDownload slide Expression of vesicle trafficking-associated genes in M. truncatula roots during symbiotic interactions. (A) Semi-quantitative RT-PCR analysis of PI3K gene expression in M. truncatula roots during 4 d after inoculation (inoc) with S. meliloti. The top panel shows the PI3K gene expression 4 d after inoculation. Where indicated, the seedlings were pretreated for 24 h with 25 μM wortmannin (WM). (B) Compilation of expressed sequence tags (ESTs) in cDNA libraries of vesicle trafficking genes in M. truncatula roots from uninoculated controls (Dev_root, KV0, MtBA(N-starved), inoculated with S. meliloti (KV1, KV2, KV3, MtBB, Nod_root, GVN, GVSN), inoculated with mycorrhiza (MTAMP, MtBC, MTGIM), or with Phytophthora medicaginis (DSIR). To analyse the plasma membrane trafficking in M. truncatula roots after inoculation with S. meliloti, the seedlings were labelled with a styryl membrane tracker dye, FM1-43 that coats the plasmalemma (Emans et al., 2002). Since the styryl dyes are lipophilic and do not diffuse through membrane they initially label the plasma membrane and become internalized via endosomes by a process called bulk-flow membrane endocytosis (Bolte et al., 2004; Jurgens, 2004; Li et al., 2005). The membrane internalization can be traced and quantified by measuring the dye uptake as the amount of fluorescence that remains after washing out the surface-bound dye (Parton et al., 2000). Such periodic measurements reflect the rate of plasma membrane endocytosis (Parton et al., 2001). The movement of the membrane dye from the plasmalemma into the intracellular membranes was analysed in root slices from zone 1 (above the root tips) at 5 min intervals. In the unstimulated root cells only limited internal fluorescence was seen after washing the roots 20 min after incubation, indicating slow constitutive endocytosis, in line with the limited housekeeping bulk-flow membrane endocytosis (Fischer-Parton et al., 2000; Wiederkehr et al., 2000; Emans et al., 2002). The membrane internalization was strongly increased after S. meliloti inoculation, and the styryl membrane dye was rapidly dispersed inside the root cortex cells, following rhizobial inoculation (Fig. 2, middle inset), as opposed to little internal fluorescence in control roots (Fig. 2, left inset). Inhibition of PI3K activity with wortmannin, which was shown to block the non-specific bulk-flow endocytosis in plants (Jurgens, 2004), suppressed membrane internalization, resulting in retention of the FM1-43 dye within the plasma membrane for an extended time (Fig. 2, right inset). The arrest of FM1-43 dispersal from the plasmalemma is also seen in conserved sharp contours of cellular periphery, as compared with dye internalization and dispersal in the inoculated roots (compare insets in Fig. 2). Fig. 2. View largeDownload slide Induction of plasma membrane internalization by rhizobacteria. Cross-sections of Medicago truncatula root slices stained with a membrane tracker probe, FM1-43, and analysed by epi-fluorescent microscopy. 3-d-old seedlings were transferred to new plates with or without 25 μM wortmannin (WM). Seedlings were inoculated with S. meliloti 24 h later. The internalization of the plasma membrane was assayed by staining the outer membrane with the FM1-43 probe for 20 min, washing and photographing every 5 min (shown are epi-fluorescent microscope images after 20 min). Note that in the uninfected plants (Cont) most of the dye remained in the plasma membrane, while in the infected cells (Rhi) the internalization of the probe resulted in loss of the sharp membrane contours (Fischer-Parton et al., 2000). The wortmannin treatment stopped the membrane endocytosis, resulting in even sharper than the control-treated dye contour delineating the cell boundaries for extended period of time. The dye accumulation in the cytoplasm was quantified after 20 min by washing and analysing the mean density of fluorescent signal expressed as relative units (R.U.) using ImagePro software package (n=12). Bar=30 μm. Error bars indicate the standard error of the mean. Fig. 2. View largeDownload slide Induction of plasma membrane internalization by rhizobacteria. Cross-sections of Medicago truncatula root slices stained with a membrane tracker probe, FM1-43, and analysed by epi-fluorescent microscopy. 3-d-old seedlings were transferred to new plates with or without 25 μM wortmannin (WM). Seedlings were inoculated with S. meliloti 24 h later. The internalization of the plasma membrane was assayed by staining the outer membrane with the FM1-43 probe for 20 min, washing and photographing every 5 min (shown are epi-fluorescent microscope images after 20 min). Note that in the uninfected plants (Cont) most of the dye remained in the plasma membrane, while in the infected cells (Rhi) the internalization of the probe resulted in loss of the sharp membrane contours (Fischer-Parton et al., 2000). The wortmannin treatment stopped the membrane endocytosis, resulting in even sharper than the control-treated dye contour delineating the cell boundaries for extended period of time. The dye accumulation in the cytoplasm was quantified after 20 min by washing and analysing the mean density of fluorescent signal expressed as relative units (R.U.) using ImagePro software package (n=12). Bar=30 μm. Error bars indicate the standard error of the mean. The role of phosphatidylinositides in root hair curling and rhizobacteria infection Root hair deformation and curling is one of the earliest physiological responses in host plants after Nod-Factor perception (De Jong et al., 1993). The root hairs curl in response to NFs, and form a hook that encloses the bacteria, which then produce an infection thread within the hair. The deposition of cell wall material for the thread formation is an exocytotic process that also depends on intracellular vesicle trafficking (Hernandez et al., 2004; Samaj et al., 2004). To test the role of vesicle trafficking in the curling response, the seedlings were pre-treated with two chemically unrelated PI3K inhibitors, wortmannin or LY294002 (Mueller-Roeber and Pical, 2002). One hundred individual root hairs from five plants from independent infections were scored. About 80% showed hair deformation exhibiting curling (Fig. 3). Moreover, most of the roots inoculated with fluorescently-labelled S. meliloti exhibited bacterial presence in root hairs that appeared as infection threads after three days (Fig. 4B). Both, the root hair curling and the infection threads were abolished by the PI3K inhibitors (compare the representative inset pictures in Fig. 3B, C). Fig. 3. View largeDownload slide The effect of PI-PLC and PI3K inhibitors on root hair curling. M. truncatula seedlings were germinated on N-free medium and transferred to new plates without (A, B) or with 25 μM wortmannin (C), 10 μM U-73122 (D), or 100 μM LY294002 (histogram). Seedlings were inoculated with S. meliloti after 24 h in the lower-mid root hair zone one. Cont, control; Rhi, Rhizobium inoculation; WM, wortmannin. Pictures were taken 4 d after the inoculation. One hundred root hairs were counted for each treatment. Bar=100 μm. The experiments were repeated at least three times with similar results. Error bars indicate the standard error of the mean. Note the extreme curling of the hairs only in the positive control treatment shown in (B). Fig. 3. View largeDownload slide The effect of PI-PLC and PI3K inhibitors on root hair curling. M. truncatula seedlings were germinated on N-free medium and transferred to new plates without (A, B) or with 25 μM wortmannin (C), 10 μM U-73122 (D), or 100 μM LY294002 (histogram). Seedlings were inoculated with S. meliloti after 24 h in the lower-mid root hair zone one. Cont, control; Rhi, Rhizobium inoculation; WM, wortmannin. Pictures were taken 4 d after the inoculation. One hundred root hairs were counted for each treatment. Bar=100 μm. The experiments were repeated at least three times with similar results. Error bars indicate the standard error of the mean. Note the extreme curling of the hairs only in the positive control treatment shown in (B). Fig. 4. View largeDownload slide The effect of wortmannin on the formation of infection threads. Seedlings were grown on N-free medium for 3 d and then transferred to new plates containing 25 μM wortmannin (C) or buffer control (A, B). After 24 h the seedlings were inoculated with S. meliloti that expressed the fluorescently-labelled variant of GFP (B, C). Note the fluorescently-labelled nodule formed within the non-treated inoculated plant (B) 7 d after the inoculation. Bright field images are shown on the right. Bar=100 μm. Fig. 4. View largeDownload slide The effect of wortmannin on the formation of infection threads. Seedlings were grown on N-free medium for 3 d and then transferred to new plates containing 25 μM wortmannin (C) or buffer control (A, B). After 24 h the seedlings were inoculated with S. meliloti that expressed the fluorescently-labelled variant of GFP (B, C). Note the fluorescently-labelled nodule formed within the non-treated inoculated plant (B) 7 d after the inoculation. Bright field images are shown on the right. Bar=100 μm. The root hair curling was also suppressed by pre-treatment of M. truncatula with a PI-PLC inhibitor, U-73122 (inset in Fig. 3D), in agreement with its effect on NF-induced root hair curling (Kelly and Irving, 2001). Since some earlier works reported extensive toxicity of U-73122 in root cells (den Hartog et al., 2001), the viability of root cells, including root hairs, was tested by staining with Evan's blue. Less than 7% cell death (data not shown) was observed. Moreover, the U-73122 inhibitor did not impair Ca2+ influx into the cytosol or gene induction (Figs. 7, 9A), both phenomena that are associated with viability. Induction of oxidative burst by S. meliloti is mediated by PI-PLC and PI3K A common early response of plant cells to infection by pathogenic bacteria is the production of ROS that is mediated by a plasma membrane NADPH oxidase (gp91phox) (Levine et al., 1994; Torres et al., 2002). Recent studies, showed that the oxidative burst was also induced in the legume roots by infection of Rhizobium (Santos et al., 2001; Herouart et al., 2002; Pauly et al., 2006). The generation of ROS in the cortical cells of M. truncatula roots after inoculation with S. meliloti (Fig. 5B) was confirmed, and the involvement of the NADPH oxidase was tested using a common suicide flavin inhibitor, diphenyleneiodonium (DPI) that inhibits flavoproteins, such as gp91phox (Mithoefer et al., 1997; Morre, 2002). The DPI treatment abolished ROS production in the roots (compare the main image of Fig. 5B with inset). The DPI treatment also suppressed the root hair curling (Fig. 5C) and infection thread formation (compare the main image of Fig. 5D with inset), indicating the involvement of ROS metabolism in the early steps of symbiotic interactions. Fig. 5. View largeDownload slide ROS production in M. truncatula roots inoculated with S. meliloti. M. truncatula seedlings were germinated on N-free medium and after 3 d transferred to new plates with (B inset, C, D inset) or without 12.5 μM diphenyleneiodonium (A, B, D). After another 24 h the seedlings were inoculated with S. meliloti (B–D, including insets). (A, B) ROS production in control uninoculated (A) or inoculated with S. meliloti (B) after 16 h, observed by epi-fluorescent microscopy. (C) The effect of DPI treatment on root hair curling. (D) Infection thread formation in M. truncatula roots inoculated with fluorescently-labelled S. meliloti. Inset shows inoculation in the presence of 12.5 μM DPI. Images were taken 4 d after inoculation. The set of experiments was repeated three times with similar results. Fig. 5. View largeDownload slide ROS production in M. truncatula roots inoculated with S. meliloti. M. truncatula seedlings were germinated on N-free medium and after 3 d transferred to new plates with (B inset, C, D inset) or without 12.5 μM diphenyleneiodonium (A, B, D). After another 24 h the seedlings were inoculated with S. meliloti (B–D, including insets). (A, B) ROS production in control uninoculated (A) or inoculated with S. meliloti (B) after 16 h, observed by epi-fluorescent microscopy. (C) The effect of DPI treatment on root hair curling. (D) Infection thread formation in M. truncatula roots inoculated with fluorescently-labelled S. meliloti. Inset shows inoculation in the presence of 12.5 μM DPI. Images were taken 4 d after inoculation. The set of experiments was repeated three times with similar results. To examine the signalling steps leading to the Rhizobium-induced oxidative burst, the involvement of PI-PLC, which activates the NADPH oxidase during pathogenesis (Legendre et al., 1993; Pappan et al., 1998; Laxalt and Munnik, 2002) was tested. Inhibition of the PI-PLC activity with U-73122 suppressed the oxidative burst in M. truncatula roots (Fig. 6). The production of ROS was also inhibited by PI3K inhibitors, wortmannin or LY294002 (Fig. 6), in agreement with the involvement of PI3K in the activation of gp91phox in neutrophils during phagocytosis. Fig. 6. View largeDownload slide The effect of PI3K and PI-PLC inhibitors on ROS production. ROS production was measured in M. truncatula roots inoculated with S. meliloti (Rhi) that were pretreated with 100 μM LY294002, 25 μM wortmannin, or with 10 μM U-73122. The accumulation of ROS was assayed using H2DCFDA dye and quantified using ImagePro software package (n=12). Insets show representative images of seedlings treated in a similar way with U-73122 or wortmannin (WM). All experiments were repeated several times with very similar results. Bars=U-73122, 100 μm; WM, 250 μm. Error bars indicate the standard error of the mean. Fig. 6. View largeDownload slide The effect of PI3K and PI-PLC inhibitors on ROS production. ROS production was measured in M. truncatula roots inoculated with S. meliloti (Rhi) that were pretreated with 100 μM LY294002, 25 μM wortmannin, or with 10 μM U-73122. The accumulation of ROS was assayed using H2DCFDA dye and quantified using ImagePro software package (n=12). Insets show representative images of seedlings treated in a similar way with U-73122 or wortmannin (WM). All experiments were repeated several times with very similar results. Bars=U-73122, 100 μm; WM, 250 μm. Error bars indicate the standard error of the mean. The role of phospholipase C and PI3K on cytosolic Ca2+ influx ROS production in plants, as well as in animals, is closely linked to Ca2+ signalling. A rapid rise in the cytosolic calcium (Cacyt2+) occurs within minutes of NF perception, and is followed by root hair deformation (Catoira et al., 2000; Walker et al., 2000; Shaw and Long, 2003a). Pharmacological studies using Ca2+ channel antagonists and measurements of Ca2+ fluxes showed that NF induced a rapid initial rise in Cacyt2+, followed by Ca2+ spiking (Smit et al., 2005). Since the majority of studies used purified NF, which bypasses the lag period for NF production in S. meliloti a preliminary time-course analysis was performed for the optimal timing of Ca2+ influx in M. truncatula roots. Based on these results the Ca2+ fluxes were measured 1 h after S. meliloti inoculation. The increased Cacyt2+ detected in M. truncatula roots after S. meliloti inoculation (Fig. 7) is in agreement with the NF-induced Ca2+ influx (Kelly and Irving, 2001; Smit et al., 2005). The rise in Cacyt2+ has been associated with the PI-PLC activity that produces the InsP3 second messenger through the hydrolysis of PtdIns(4,5)P2 (Charron et al., 2004; Hetherington and Brownlee, 2004). To probe the role of PI-PLC in the Cacyt2+ increase the PI-PLC was inhibited with U-73122. The inhibitor did not interfere with rise in Cacyt2+ (Fig. 7). Elevated Cacyt2+ was also observed in M. truncatula roots treated with wortmannin, which even increased the Cacyt2+ (Fig. 7). Fig. 7. View largeDownload slide Cytoplasmic Cacyt2+ influx during S. meliloti inoculation. Calcium influx during S. meliloti inoculation was assayed using a Cacyt2+-specific dye, FLUO4-AM. Seedlings were pre-treated with 10 μM U-73122 (U-7) or 25 μM wortmannin (WM) and inoculated with S. meliloti. The fluorescent signal was analysed by epi-fluorescent (white bars) and confocal (black bars, constructed from merged 20 sections of 0.5 μm) microscopy 1 h after inoculation, and quantified using ImagePro software (n=12). Error bars indicate standard error of the mean. Representative images are shown on the right. Bars= U-73122, 100 μm; WM, 50 μm. Fig. 7. View largeDownload slide Cytoplasmic Cacyt2+ influx during S. meliloti inoculation. Calcium influx during S. meliloti inoculation was assayed using a Cacyt2+-specific dye, FLUO4-AM. Seedlings were pre-treated with 10 μM U-73122 (U-7) or 25 μM wortmannin (WM) and inoculated with S. meliloti. The fluorescent signal was analysed by epi-fluorescent (white bars) and confocal (black bars, constructed from merged 20 sections of 0.5 μm) microscopy 1 h after inoculation, and quantified using ImagePro software (n=12). Error bars indicate standard error of the mean. Representative images are shown on the right. Bars= U-73122, 100 μm; WM, 50 μm. To analyse the role of Cacyt2+ influx in S. meliloti infection, the M. truncatula roots were treated with different Ca2+ channel antagonists for 1 d, followed by Rhizobium inoculation, and bacterial growth was analysed after 4 d. The appearance of the fluorescently-labelled S. meliloti in the root hairs was suppressed by verapamil, which blocks the voltage-dependent channels in the plasma membrane, but not by TMB-8, which blocks the Ca2+ release from internal stores (Fig. 8). Furthermore, the effects of the Ca2+ channel blockers on the S. meliloti infection were collimated with root hair deformation (compare the absolutely straight hairs of the verapamil-pretreated roots with TMB-8 treatment or the control inoculations). Fig. 8. View largeDownload slide The effect of calcium channel blockers on the S. meliloti infection. (A–C) M. truncatula were germinated on N-free medium and transferred to new plates after 3 d (A). After another 1 d on new plates the seedlings were inoculated with fluorescently-labelled S. meliloti (B, C). The roots were observed under bright field (A, B) or fluorescent (C) microscopy 1 d after inoculation. (D–I) The seedlings were grown as above but transferred to plates supplemented with 50 μM TMB-8 (D–F), or with 20 μM Verapamil (G–I). The seedlings in E, F and in H, I were inoculated with S. meliloti and observed as above under bright field (BF) or fluorescent light (FL). The experiments were repeated at least three times with very similar results. Bars=150 μm. Fig. 8. View largeDownload slide The effect of calcium channel blockers on the S. meliloti infection. (A–C) M. truncatula were germinated on N-free medium and transferred to new plates after 3 d (A). After another 1 d on new plates the seedlings were inoculated with fluorescently-labelled S. meliloti (B, C). The roots were observed under bright field (A, B) or fluorescent (C) microscopy 1 d after inoculation. (D–I) The seedlings were grown as above but transferred to plates supplemented with 50 μM TMB-8 (D–F), or with 20 μM Verapamil (G–I). The seedlings in E, F and in H, I were inoculated with S. meliloti and observed as above under bright field (BF) or fluorescent light (FL). The experiments were repeated at least three times with very similar results. Bars=150 μm. The effect of PtdIns signalling on the early nodulin gene expression Finally, the involvement of the PtdIns signalling on the nodulation-related gene expression was analysed. Surprisingly, growth of M. truncatula seedlings on medium supplemented with the PI-PLC or the PI3K inhibitors caused induction of ENODs expression in the absence of rhizobia or NF (Fig. 9), despite their inhibitory effect on the root hair curling (Fig. 3). The expression of ENODs after S. meliloti inoculation was tested according to the peak induction times of the specific ENODs. Treatment with PtdIns inhibitors induced the early genes, such as the ENOD12 and DMI1, as well as the intermediate ones, such as ENOD16 and ENOD20, but not the late nodulin-like Mtn21 gene (Fig. 9B). U-73122 treatment did not inhibit the RIP1 or ENOD12 expression, but superinduced the DMI1 gene (Fig. 9A). These results suggest that legumes may sense perturbations in the membrane trafficking and respond by induction of the nodulin genes. To analyse the signalling steps downstream of the PI-PLC and the PI3K, such as Ca2+ influx and ROS production, the Ca2+ channel was blocked with verapamil and the NADPH oxidase was blocked with DPI, respectively. Interestingly, verapamil inhibited ENOD12 but not DMI3 gene expression, while DPI suppressed the DMI3 but not ENOD12 (Fig. 9C), suggesting induction of these genes by separate pathways (Fig. 10). Fig. 9. View largeDownload slide Analysis of nodulin genes expression during S. meliloti inoculation. (A) Semi-quantitative RT-PCR analysis of rip1 and MtENOD12 gene expression 2 d after inoculation of M. truncatula seedlings with S. meliloti (inoc). Seedlings were transferred as described in the legend to Fig. 5 into medium containing 10 μM U-73122. (B) Analysis of nodulin gene expression in the presence or absence of 25 μM wortmannin. RIP1, MtENOD12 and DMI3 were assayed 2 d after inoculation, MtENOD16 and MtENOD20 were assayed after 4 d, and MtN21 expression was after 7 d. (C) The involvement of Ca2+ and ROS signalling on expression of MtENOD12 and DMI3 genes. Seedlings were transferred into 20 μM verapamil (Ver), or into 20 μM DPI (DPI), and inoculated with S. meliloti 24 h later. Total RNA was extracted 2 d after inoculation. EF1α RNA was used as control for the normalization and quality of RNA. All experiments were repeated at least three times with very similar results. Fig. 9. View largeDownload slide Analysis of nodulin genes expression during S. meliloti inoculation. (A) Semi-quantitative RT-PCR analysis of rip1 and MtENOD12 gene expression 2 d after inoculation of M. truncatula seedlings with S. meliloti (inoc). Seedlings were transferred as described in the legend to Fig. 5 into medium containing 10 μM U-73122. (B) Analysis of nodulin gene expression in the presence or absence of 25 μM wortmannin. RIP1, MtENOD12 and DMI3 were assayed 2 d after inoculation, MtENOD16 and MtENOD20 were assayed after 4 d, and MtN21 expression was after 7 d. (C) The involvement of Ca2+ and ROS signalling on expression of MtENOD12 and DMI3 genes. Seedlings were transferred into 20 μM verapamil (Ver), or into 20 μM DPI (DPI), and inoculated with S. meliloti 24 h later. Total RNA was extracted 2 d after inoculation. EF1α RNA was used as control for the normalization and quality of RNA. All experiments were repeated at least three times with very similar results. Fig. 10. View largeDownload slide Proposed scheme of signal transduction in M. truncatula roots during S. meliloti inoculation. NF binding to specific receptor (DMI2) induces Ca2+ influx and activates PtdIns-dependent PLC (PI-PLC) and PI3K which regulate the NADPH oxidase-mediated production of reactive oxygen species (ROS). Activity of both (Ca2+ and ROS) pathways is required for root hair curling and formation of infection threads, while the induction of ENOD genes occurred in response to perturbations in PtdIns signalling. Expression of ENOD12 depended on Cacyt2+ influx, while the constitutive expression of DMI3 depended on ROS production, as was RIP1 (described in Ramu et al., 2002). Fig. 10. View largeDownload slide Proposed scheme of signal transduction in M. truncatula roots during S. meliloti inoculation. NF binding to specific receptor (DMI2) induces Ca2+ influx and activates PtdIns-dependent PLC (PI-PLC) and PI3K which regulate the NADPH oxidase-mediated production of reactive oxygen species (ROS). Activity of both (Ca2+ and ROS) pathways is required for root hair curling and formation of infection threads, while the induction of ENOD genes occurred in response to perturbations in PtdIns signalling. Expression of ENOD12 depended on Cacyt2+ influx, while the constitutive expression of DMI3 depended on ROS production, as was RIP1 (described in Ramu et al., 2002). Discussion Endocytosis has been proposed to play an important role in the evolution of the endosymbiosis in eukaryotes, leading to emergence of specialized organelles, such as mitochondria and plastids (Samaj et al., 2004). Groundwork analysis of the signalling mechanisms of symbiotic interactions in legumes by pharmacological approach with purified NFs showed involvement of G-proteins and PtdIns-PLC. The early host plant responses are mediated by phosphoinositides and Ca2+ second messengers and production of ROS (Pingret et al., 1998; Santos et al., 2001; Engstrom et al., 2002; Charron et al., 2004). The increase in plasma membrane endocytosis in response to S. meliloti inoculation is in line with the NF-induced actin reorganization, a process that involves vesicle trafficking (Miller et al., 1999; Samaj et al., 2004). Moreover, it was shown that changes in plant cytoskeleton are mediated by PtdIns (DeWald et al., 2001; Anthony et al., 2004). It is shown here that vesicle trafficking plays an important role in the symbiotic interactions of M. truncatula inoculated with S. meliloti, as recently shown in the non-symbiotic interaction with pathogens (Wick et al., 2003). These studies allow the extension of comparative analysis between symbiotic and pathogenic plant–microbe interactions. Induction and function of endocytosis in symbiotic interactions During the later stages of symbiotic interactions the rhizobacteria become internalized in the cytoplasm of individual cortical cells of the host by endocytosis from unwalled infection droplets (Hernandez et al., 2004; Puppo et al., 2005). Endocytosis also functions in bacterial entry into animal cells (Meresse et al., 1999). The analysis of trafficking gene expression in M. truncatula inoculated with S. meliloti showed the induction of endocytosis-associated genes that were shown to function in the bacterial entry step in animal cells (Stein et al., 2005). In plants, PI3K was recently shown to function in autophagy of tobacco mosaic virus (Ivashuta et al., 2005). The autophagy process, of the cytoplasm-to-vacuole, is analogous to the endocytosis from extracellular space to vacuole (Herman and Larkins, 1999). Inoculation of S. meliloti induced PI3K expression in M. truncatula as well as other vesicle trafficking genes. Particularly notable is induction of several specific small (Rab- and Arf-like) and large (dynamin-like) GTPases by rhizobial and/or mycorrhizal infections (Fig. 1B). Interestingly, many of the genes exhibited transient expression in the early stages of interaction, and returned to normal levels after 96 h, in agreement with a recent microarray-based transcript analysis in M. truncatula (Lohar et al., 2006). The large-scale analysis also revealed the induction of specific PtdIns effectors and SNAREs that function in vesicle targeting. These results are in agreement with previous small-scale analysis of trafficking genes that found induction of PI3K and Rab7 GTPase in soybean during nodule organogenesis (Cheon et al., 1993; Hong and Verma, 1994; Son et al., 2003). Inhibition of PI3K activity blocked endocytotic internalization of the host plasma membrane, resulting in suppression of the root hair curling and formation of infection threads (Figs 3, 4). These findings implicate vesicle trafficking in the early steps of the symbiotic interactions, even before the endocytotic internalization of the infection droplets, probably at the stage of signal exchange between the interacting organisms. Ca2+ signalling in S. meliloti infection Perception of NFs causes a biphasic Cacyt2+ influx in the host root cells. The first phase occurs within seconds after binding of NFs to the plasma membrane receptor, and is followed by Ca2+ spiking (Walker et al., 2000; Shaw and Long, 2003a; Oldroyd and Downie, 2004). The phases differ in their regulation and function. The role of Ca2+ fluxes in symbiotic interactions has been studied using purified NFs and a range of pharmacological agents that modify Ca2+ channel activity (Engstrom et al., 2002). These studies showed that Ca2+ spiking was essential for the induction of ENOD11 (Charron et al., 2004). It is notable that neither verapamil nor TMB-8 prevented NF-induced Ca2+ spiking in M. truncatula (Engstrom et al., 2002). In soybean protoplasts, verapamil inhibited the initial Ca2+ influx without blocking Ca2+ spiking, suggesting separate pathways of the two Ca2+ fluxes (Yokoyama et al., 2000). It is shown that verapamil suppressed the rhizobium-induced root hair curling response (Fig. 8), and the induction of ENOD12 (Fig. 9C), while TMB-8 was ineffective in blocking either action. The hair curling was also suppressed by inhibition of the PI3K (Fig. 3D) or the PI-PLC activities (Fig. 3D), also without inhibiting the Ca2+ influx (Fig. 7), or suppression of ENOD12 expression (Fig. 9). These results indicate an early split in rhizobia-induced signal transduction pathways (Fig. 10). ROS signalling in S. meliloti infection Production of reactive oxygen species (ROS) is a universal plant response to invading micro-organisms, and is considered a hallmark of the hypersensitive reaction to pathogens (Levine et al., 1996). The pathogen perception is mediated by elicitors and involves recognition of pathogen-associated molecular patterns (Ausubel, 2005), leading to the activation of plant NADPH oxidase, gp91phox, which is a close homologue of the neutrophil (phagocyte) enzyme (Keller et al., 1998; Torres et al., 1998). The plant gp91phox has been associated with almost every type of plant–pathogen interaction (Lamb and Dixon, 1997). Recently, oxidative burst was also detected in the early stages of symbiotic interactions, suggesting that legumes respond to NFs secreted by rhizobacteria in a manner similar to elicitors (Santos et al., 2001; Ramu et al., 2002). Recently, it was shown that the gp91phox complex in the plasma membrane is inactive, and becomes activated in endosomes (Simonsen and Stenmark, 2001). Although detailed analysis of the gp91phox activation in plants is not available, proteins with PX (phox) domains that mediate the gp91phox internalization have been identified in Arabidopsis (Meijer and Munnik, 2003). The suppression of ROS production by DPI and by PI3K inhibitors suggests that a similar activation mechanism may operate during the symbiotic interactions (Fig. 5). This suggestion is also supported by wortmannin-caused inhibition of NADPH oxidase activity in stomata (Park et al., 2003). The S. meliloti-induced ROS production depended on the activities of both PI3K and PI-PLC (Fig. 6). This is in agreement with the effect of U-73122 in tomato cells challenged with a Cladosporium fulvum elicitor AVR4 (Laxalt and Munnik, 2002). Interestingly, neither wortmannin nor LY294002 (inhibitors of PI3K), nor U-73122 (PI-PLC inhibitor) affected the Cacyt2+ influx (Fig. 7), supporting the early split in symbiotic signal transduction pathways: one branch leading to the oxidative burst via PI-PLC and PI3K, with another mediating the Cacyt2+ influx. Both pathways, however, participate in root hair curling and the rhizobial infection, as shown by the effect of verapamil on these responses. ENOD gene expression Establishment of symbiotic interactions in legumes with compatible rhizobacteria involves substantial changes in host gene expression, particularly induction of nodulation genes, i.e. nodulins (ENODs). It is shown that ENODs expression is not co-ordinated as a group and is not linked with other signalling events during the symbiotic interactions. This conclusion is based on the inhibition of oxidative burst and of root hair curling, but not of ENODs transcription by PI-PLC and PI3K inhibitors. Moreover, individual nodulins were differentially affected by Cacyt2+ influx and oxidative burst. The voltage-gated Ca2+ channel activity was necessary for ENOD12 but not for DMI3 gene expression. On the other hand, inhibition of oxidative burst by DPI suppressed the DMI3 but not ENOD12 (Fig. 9C). These results are in contrast with an earlier report that U-73122 blocked the induction of the ENOD12 by NF (Pingret et al., 1998). It is possible that S. meliloti induced an additional, NF-independent signalling pathway. Alternatively, inhibition of ENOD12 expression, observed by Pingret et al., could be a result of cell death, as 80% of the root hairs showed propidium iodide staining. Root cell viability was tested by Evan's blue staining (because propidium iodide stains rhizobacteria and cell walls) and cell death was only observed in a very few (<5%) root hairs. The viability of treated plants in this work can also be inferred from the ENODs induction (Fig. 9). These results were not due to ineffective U-73122 application, as it abolished the root hair curling (Fig. 3), and suppressed the oxidative burst (Fig. 5). Interestingly, the suppression of membrane trafficking by PI-PLC or PI3K inhibitors induced the expression of ENOD genes in the absence of S. meliloti (Fig. 9), suggesting that root cells sense the rhizobacteria-induced perturbations in membrane trafficking and respond by the induction of ENOD genes. Interestingly, the early nodulins (ENOD16 and ENOD20) encode putative arabinogalactan proteins (Vernoud et al., 1999), and ENOD12 is a hydroxyproline rich protein that is associated with the infection threads. These genes may be induced by the membrane perturbations, caused by S. meliloti. Phospholipid signalling was recently shown to mediate nodulation in M. truncatula (Charron et al., 2004). Concluding remarks The results, summarized in Fig. 10, show that membrane trafficking plays an important role in the symbiotic interactions by regulating ROS production, which in turn mediates root hair curling (Fig. 5C) and infection threads formation (Fig. 5D). ROS were also shown to mediate the induction of RIP1 gene during symbiotic interaction (Ramu et al., 2002). Interestingly, it was recently shown that Medicago plants that express the DMI3 protein lacking its autoregulatory domain are not infected by rhizobia, suggesting that the autoinhibition of DMI3 kinase activity is essential for infection (Gleason et al., 2006). These results show that DMI3 transcription was strongly suppressed by plant treatment with the NADPH oxidase inhibitor, DPI (Fig. 9C). Moreover, DPI treatment also suppressed root hair curling and infection thread formation (Fig. 5C, D, respectively). Since dmi3 mutants are capable of reorientating root hair growth in response to Nod factors (Riely et al., 2006), these data indicate that the DMI pathway is not coupled to the root hair curling response. The PtdIns-associated signalling described here adds another layer to the regulation of symbiotic interactions, in addition to Ca2+ influx and ROS production. Interestingly, the expression of ENODs seems to be controlled by signals not directly related to the morphological changes leading to root hair curling. These data also support the alternative to the defence role of the oxidative burst, whereby ROS act to modulate the root hair growth and establishment of infection threads, as suggested by Shaw and Long (2003b), probably by peroxide action on cell wall proteins. It is proposed that the PtdIns signalling may function in the co-ordination of both the early (before rhizobial entry into roots), as well as the late (endocytosis of infection droplets) steps of the symbiotic interactions. Abbreviations Abbreviations Cacyt2+ cytosolic calcium DPI diphenyleneiodonium PtdIns-3K (PI3K) phosphatidylinositol-3 kinase InsP3 inositol triphosphate PI-PLC phosphoinositide-specific phospholipase C DCFDA 2′,7′-dichlorofluorescein diacetate We thank Y Kapulnik (Volcani Center) for M. truncatula seeds and comments on the manuscript and Naomi Book-Melamed for confocal microscopy analysis. This research was supported by the Israel Science Foundation and by an EMBO fellowship ASTF 58-2006 to SP-G. References Anthony RG,  Henriques R,  Helfer A,  Mesaros T,  Rios G,  Testerink C,  Munnik T,  Deak M,  Koncz C,  Bogre L.  A protein kinase target of PDK1 signaling pathway involved in root hair growth in Arabidopsis,  EMBO Journal ,  2004, vol.  23 (pg.  572- 581) Google Scholar CrossRef Search ADS PubMed  Ausubel FM.  Are innate immune signaling pathways in plants and animals conserved?,  Nature Immunology ,  2005, vol.  6 (pg.  973- 979) Google Scholar CrossRef Search ADS PubMed  Bolte S,  Talbot C,  Boutte Y,  Catrice O,  Read ND,  Satiat-Jeunemaitre B.  FM-dyes as experimental probes for dissecting vesicle trafficking in living plant cells,  Journal of Microscopy ,  2004, vol.  214 (pg.  159- 173) Google Scholar CrossRef Search ADS PubMed  Catoira R,  Galera C,  de Billy F,  Penmetsa RV,  Journet E-P,  Maillet F,  Rosenberg C,  Cook D,  Gough C,  Denarie J.  Four genes of Medicago truncatula controlling components of a Nod Factor transduction pathway,  The Plant Cell ,  2000, vol.  12 (pg.  1647- 1666) Google Scholar CrossRef Search ADS PubMed  Charron D,  Pingret J-L,  Chabaud M,  Journet E-P,  Barker DG.  Pharmacological evidence that multiple phospholipid signaling pathways link rhizobium nodulation factor perception in Medicago truncatula root hairs to intracellular responses, including Ca2+ spiking and specific ENOD gene expression,  Plant Physiology ,  2004, vol.  136 (pg.  3582- 3593) Google Scholar CrossRef Search ADS PubMed  Cheon CI,  Lee NG,  Siddique ABM,  Bal AK,  Verma DPS.  Roles of plant homologs of Rab1p and Rab7p in the biogenesis of the peribacteroid membrane, a subcellular compartment formed de novo during root-nodule symbiosis,  EMBO Journal ,  1993, vol.  12 (pg.  4125- 4135) Google Scholar PubMed  De Jong AJ,  Heidstra R,  Spaink HP, et al.  Rhizobium lipooligosaccharides rescue a carrot somatic embryo mutant,  The Plant Cell ,  1993, vol.  5 (pg.  615- 620) Google Scholar CrossRef Search ADS PubMed  den Hartog M,  Musgrave A,  Munnik T.  Nod factor-induced phosphatidic acid and diacylglycerol pyrophosphate formation: a role for phospholipase C and D in root hair deformation,  The Plant Journal ,  2001, vol.  25 (pg.  55- 65) Google Scholar CrossRef Search ADS PubMed  De Wald DB,  Torabinejad J,  Jones CA,  Shope JC,  Cangelosi AR,  Thompson JE,  Prestwich GD,  Hama H.  Rapid accumulation of phosphatidylinositol 4,5-bisphosphate and inositol 1,4,5-trisphosphate correlates with calcium mobilization in salt-stressed Arabidopsis,  Plant Physiology ,  2001, vol.  126 (pg.  759- 769) Google Scholar CrossRef Search ADS PubMed  Emans N,  Zimmermann S,  Fischer R.  Uptake of a fluorescent marker in plant cells is sensitive to brefeldin A and wortmannin,  The Plant Cell ,  2002, vol.  14 (pg.  71- 86) Google Scholar CrossRef Search ADS PubMed  Engstrom EM,  Ehrhardt DW,  Mitra RM,  Long SR.  Pharmacological analysis of Nod Factor-induced calcium spiking in Medicago truncatula. Evidence for the requirement of type IIA calcium pumps and phosphoinositide signaling,  Plant Physiology ,  2002, vol.  128 (pg.  1390- 1401) Google Scholar CrossRef Search ADS PubMed  Fischer-Parton S,  Parton RM,  Hickey PC,  Dijksterhuis J,  Atkinson HA,  Read ND.  Confocal microscopy of FM4-64 as a tool for analysing endocytosis and vesicle trafficking in living fungal hyphae,  Journal of Microscopy ,  2000, vol.  198 (pg.  246- 259) Google Scholar CrossRef Search ADS PubMed  Gleason C,  Chaudhuri S,  Yang T,  Munoz A,  Poovaiah BW,  Oldroyd GED.  Nodulation independent of rhizobia induced by a calcium-activated kinase lacking autoinhibition,  Nature ,  2006, vol.  441 (pg.  1149- 1152) Google Scholar CrossRef Search ADS PubMed  Herman EM,  Larkins BA.  Protein storage bodies and vacuoles,  The Plant Cell ,  1999, vol.  11 (pg.  601- 614) Google Scholar CrossRef Search ADS PubMed  Hernandez LE,  Escobar C,  Drobak BK,  Bisseling T,  Brewin NJ.  Novel expression patterns of phosphatidylinositol 3-hydroxy kinase in nodulated Medicago spp. plants,  Journal of Experimental Botany ,  2004, vol.  55 (pg.  957- 959) Google Scholar CrossRef Search ADS PubMed  Herouart D,  Baudouin E,  Frendo P,  Harrison J,  Santos R,  Jamet A,  Van de Sype G,  Touati D,  Puppo A.  Reactive oxygen species, nitric oxide and glutathione: a key role in the establishment of the legume–Rhizobium symbiosis?,  Plant Physiology and Biochemistry ,  2002, vol.  40 (pg.  619- 624) Google Scholar CrossRef Search ADS   Hetherington AM,  Brownlee C.  The generation of Ca2+ signals in plants,  Annual Review of Plant Botany ,  2004, vol.  55 (pg.  401- 427) Google Scholar CrossRef Search ADS   Hong Z,  Verma D.  A phosphatidylinositol 3-kinase is induced during soybean nodule organogenesis and is associated with membrane proliferation,  Proceedings of the National Academy of Sciences, USA ,  1994, vol.  91 (pg.  9617- 9621) Google Scholar CrossRef Search ADS   Ivashuta S,  Liu J,  Liu J,  Lohar DP,  Haridas S,  Bucciarelli B,  VandenBosch KA,  Vance CP,  Harrison MJ,  Gantt JS.  RNA interference identifies a calcium-dependent protein kinase involved in Medicago truncatula root development,  The Plant Cell ,  2005, vol.  17 (pg.  2911- 2921) Google Scholar CrossRef Search ADS PubMed  Jamet A,  Sigaud S,  Van de Sype G,  Puppo A,  Herouart D.  Expression of the bacterial catalase genes during Sinorhizobium meliloti–Medicago sativa symbiosis and their crucial role during the infection process,  Molecular Plant–Microbe Interactions ,  2003, vol.  16 (pg.  217- 225) Google Scholar CrossRef Search ADS PubMed  Jurgens G.  Membrane trafficking in plants,  Annual Review of Cell Development Biology ,  2004, vol.  20 (pg.  481- 504) Google Scholar CrossRef Search ADS   Keller T,  Damude HG,  Werner D,  Doerner P,  Dixon RA,  Lamb C.  A plant homolog of the neutrophil NADPH oxidase gp91-phox subunit gene encodes a plasma membrane protein with Ca2+ binding motifs,  The Plant Cell ,  1998, vol.  10 (pg.  255- 266) Google Scholar PubMed  Kelly MN,  Irving HR.  Nod factors stimulate plasma membrane delimited phospholipase C activity in vitro,  Physiologia Plantarum ,  2001, vol.  113 (pg.  461- 468) Google Scholar CrossRef Search ADS   Klusener B,  Young JJ,  Murata Y,  Allen GJ,  Mori IC,  Hugouvieux V,  Schroeder JI.  Convergence of calcium signaling pathways of pathogenic elicitors and abscisic acid in Arabidopsis guard cells,  Plant Physiology ,  2002, vol.  130 (pg.  2152- 2163) Google Scholar CrossRef Search ADS PubMed  Lamb C,  Dixon RA.  The oxidative burst in plant disease resistance,  Annual Review of Plant Physiology and Plant Molecular Biology ,  1997, vol.  48 (pg.  251- 275) Google Scholar CrossRef Search ADS PubMed  Laxalt AM,  Munnik T.  Phospholipid signalling in plant defence,  Current Opinion in Plant Biology ,  2002, vol.  5 (pg.  332- 338) Google Scholar CrossRef Search ADS PubMed  Leevers SJ,  Vanhaesebroeck B,  Waterfield MD.  Signalling through phosphoinositide 3-kinases: the lipids take centre stage,  Current Opinion in Cell Biology ,  1999, vol.  11 (pg.  219- 225) Google Scholar CrossRef Search ADS PubMed  Legendre L,  Yueh YG,  Crain R,  Haddock N,  Heinstein PF,  Low PS.  Phospholipase C activation during elicitation of the oxidative burst in cultured plant cells,  Journal of Biological Chemistry ,  1993, vol.  268 (pg.  24559- 24563) Google Scholar PubMed  Levine A,  Pennell R,  Alvarez M,  Palmer R,  Lamb CJ.  Calcium-mediated apoptosis in a plant hypersensitive disease resistance response,  Current Biology ,  1996, vol.  6 (pg.  427- 437) Google Scholar CrossRef Search ADS PubMed  Levine A,  Tenhaken R,  Dixon R,  Lamb C.  H2O2 from the oxidative burst orchestrates the plant hypersensitive disease resistance response,  Cell ,  1994, vol.  79 (pg.  583- 593) Google Scholar CrossRef Search ADS PubMed  Li J,  Yang H,  Peer WA, et al.  Arabidopsis H+-PPase AVP1 regulates auxin-mediated organ development,  Science ,  2005, vol.  310 (pg.  121- 125) Google Scholar CrossRef Search ADS PubMed  Lohar DP,  Sharopova N,  Endre G,  Penuela S,  Samac D,  Town C,  Silverstein KAT,  VandenBosch KA.  Transcript analysis of early nodulation events in Medicago truncatula,  Plant Physiology ,  2006, vol.  140 (pg.  221- 234) Google Scholar CrossRef Search ADS PubMed  Mazel A,  Leshem Y,  Tiwari BS,  Levine A.  Induction of salt and osmotic stress tolerance by overexpression of an intracellular vesicle trafficking protein AtRab7 (AtRabG3e),  Plant Physiology ,  2004, vol.  134 (pg.  118- 128) Google Scholar CrossRef Search ADS PubMed  Meijer HJG,  Munnik T.  Phospholipid-based signaling in plants,  Annual Review of Plant Biology ,  2003, vol.  54 (pg.  265- 306) Google Scholar CrossRef Search ADS PubMed  Meresse S,  Steele-Mortimer O,  Finlay BB,  Gorvel J-P.  The rab7 GTPase controls the maturation of Salmonella typhimurium-containing vacuoles in HeLa cells,  EMBO Journal ,  1999, vol.  18 (pg.  4394- 4403) Google Scholar CrossRef Search ADS PubMed  Miller DD,  de Ruijter NCA,  Bisseling T,  Emons AMC.  The role of actin in root hair morphogenesis: studies with lipochito-oligosaccharide as a growth stimulator and cytochalasin as an actin perturbing drug,  The Plant Journal ,  1999, vol.  17 (pg.  141- 154) Google Scholar CrossRef Search ADS   Mithoefer A,  Daxberger A,  Fromhold-Treu D,  Ebel J.  Involvement of an NAD(P)H oxidase in the elicitor-inducible oxidative burst of soybean,  Phytochemistry ,  1997, vol.  45 (pg.  1101- 1107) Google Scholar CrossRef Search ADS   Morre J.  Preferential Inhibition of the plasma membrane NADH oxidase (NOX) activity by diphenyleneiodonium chloride with NADPH as donor,  Antioxidants and Redox Signaling ,  2002, vol.  4 (pg.  207- 212) Google Scholar CrossRef Search ADS PubMed  Morrison N,  Verma DPS.  A block in the endocytosis of Rhizobium allows cellular-differentiation in nodules, but affects the expression of some peribacteroid membrane nodulins,  Plant Molecular Biology ,  1987, vol.  9 (pg.  185- 196) Google Scholar CrossRef Search ADS PubMed  Mueller-Roeber B,  Pical C.  Inositol phospholipid metabolism in Arabidopsis. Characterized and putative isoforms of inositol phospholipid kinase and phosphoinositide-specific phospholipase C,  Plant Physiology ,  2002, vol.  130 (pg.  22- 46) Google Scholar CrossRef Search ADS PubMed  Oldroyd GED,  Downie JA.  Calcium, kinases and nodulation signalling in legumes,  Nature Cell Biology ,  2004, vol.  5 (pg.  566- 576) Google Scholar CrossRef Search ADS   Pappan K,  Austin-Brown S,  Chapman KD,  Wang X.  Substrate selectivities and lipid modulation of plant phospholipase D[alpha], -[beta], and -[gamma],  Archives of Biochemistry and Biophysics ,  1998, vol.  353 (pg.  131- 140) Google Scholar CrossRef Search ADS PubMed  Park K-Y,  Jung J-Y,  Park J,  Hwang J-U,  Kim Y-W,  Hwang I,  Lee Y.  A role for phosphatidylinositol 3-phosphate in abscisic acid-induced reactive oxygen species generation in guard cells,  Plant Physiology ,  2003, vol.  132 (pg.  92- 98) Google Scholar CrossRef Search ADS PubMed  Parton RM,  Dyer AF,  Read ND,  Trewavas AJ.  Apical structure of actively growing fern rhizoids examined by DIC and confocal microscopy,  Annals of Botany ,  2000, vol.  85 (pg.  233- 245) Google Scholar CrossRef Search ADS   Parton RM,  Fischer-Parton S,  Watahiki MK,  Trewavas AJ.  Dynamics of the apical vesicle accumulation and the rate of growth are related in individual pollen tubes,  Journal of Cell Science ,  2001, vol.  114 (pg.  2685- 2695) Google Scholar PubMed  Pauly N,  Pucciariello C,  Mandon K,  Innocenti G,  Jamet A,  Baudouin E,  Herouart D,  Frendo P,  Puppo A.  Reactive oxygen and nitrogen species and glutathione: key players in the legume–Rhizobium symbiosis,  Journal of Experimental Botany ,  2006, vol.  57 (pg.  1769- 1776) Google Scholar CrossRef Search ADS PubMed  Pingret J-L,  Journet E-P,  Barker DG.  Rhizobium Nod Factor signaling: evidence for a G protein-mediated transduction mechanism,  The Plant Cell ,  1998, vol.  10 (pg.  659- 672) Google Scholar PubMed  Puppo A,  Groten K,  Bastian F,  Carzaniga R,  Soussi M,  Lucas MM,  de Felipe MR,  Harrison J,  Vanacker H,  Foyer CH.  Legume nodule senescence: roles for redox and hormone signalling in the orchestration of the natural aging process,  New Phytologist ,  2005, vol.  165 (pg.  683- 701) Google Scholar CrossRef Search ADS PubMed  Ramu SK,  Peng HM,  Cook DR.  Nod factor induction of reactive oxygen species production is correlated with expression of the early nodulin gene Rip1 in Medicago truncatula,  Molecular Plant–Microbe Interactions ,  2002, vol.  15 (pg.  522- 528) Google Scholar CrossRef Search ADS PubMed  Riely BK,  Mun J-H,  Ane J-M.  Unravelling the molecular basis for symbiotic signal transduction in legumes,  Molecular Plant Pathology ,  2006, vol.  7 (pg.  197- 207) Google Scholar CrossRef Search ADS PubMed  Samaj J,  Baluska F,  Voigt B,  Schlicht M,  Volkmann D,  Menzel D.  Endocytosis, actin cytoskeleton, and signaling,  Plant Physiology ,  2004, vol.  135 (pg.  1150- 1161) Google Scholar CrossRef Search ADS PubMed  Santos R,  Herouart D,  Sigaud S,  Touati D,  Puppo A.  Oxidative burst in alfalfa-Sinorhizobium meliloti symbiotic interaction,  Molecular Plant–Microbe Interactions ,  2001, vol.  14 (pg.  86- 89) Google Scholar CrossRef Search ADS PubMed  Shaw S,  Long S.  Nod factor elicits two separable calcium responses in Medicago truncatula root hair cells,  Plant Physiology ,  2003, vol.  131 (pg.  976- 984) Google Scholar CrossRef Search ADS PubMed  Shaw SL,  Long SR.  Nod factor inhibition of reactive oxygen efflux in a host legume,  Plant Physiology ,  2003, vol.  132 (pg.  2196- 2204) Google Scholar CrossRef Search ADS PubMed  Simonsen A,  Stenmark H.  PX domains: attracted by phosphoinositides,  Nature Cell Biology ,  2001, vol.  3 (pg.  E179- E182) Google Scholar CrossRef Search ADS PubMed  Smit P,  Raedts J,  Portyanko V,  Debelle F,  Gough C,  Bisseling T,  Geurts R.  NSP1 of the GRAS protein family is essential for rhizobial Nod Factor-induced transcription,  Science ,  2005, vol.  308 (pg.  1789- 1791) Google Scholar CrossRef Search ADS PubMed  Son O,  Yang HS,  Lee HJ, et al.  Expression of srab7 and SCaM genes required for endocytosis of Rhizobium in root nodules,  Plant Science ,  2003, vol.  165 (pg.  1239- 1244) Google Scholar CrossRef Search ADS   Stein MP,  Cao CH,  Tessema M,  Feng Y,  Romero E,  Welford A,  Wandinger-Ness A.  Interaction and functional analyses of human VPS34/p150 phosphatidylinositol 3-kinase complex with Rab7,  Methods in Enzymology ,  2005, vol.  403 (pg.  628- 649) Google Scholar PubMed  Torres MA,  Dangl JL,  Jones JDG.  Arabidopsis gp91phox homologues AtrbohD and AtrbohF are required for accumulation of reactive oxygen intermediates in the plant defence response,  Proceedings of the National Academy of Sciences, USA ,  2002, vol.  99 (pg.  517- 522) Google Scholar CrossRef Search ADS   Torres MA,  Onouchi H,  Hamada S,  Machida C,  Hammond-Kosack KE,  Jones JDG.  Six Arabidopsis thaliana homologues of the human respiratory burst oxidase (gp91phox),  The Plant Journal ,  1998, vol.  14 (pg.  365- 370) Google Scholar CrossRef Search ADS PubMed  Vernoud V,  Journet EP,  Barker DG.  MtENOD20, a Nod factor-inducible molecular marker for root cortical cell activation,  Molecular Plant–Microbe Interactions ,  1999, vol.  12 (pg.  604- 614) Google Scholar CrossRef Search ADS   Vincent P,  Chua M,  Nogue F,  Fairbrother A,  Mekeel H,  Xu Y,  Allen N,  Bibikova TN,  Gilroy S,  Bankaitis VA.  A Sec14p-nodulin domain phosphatidylinositol transfer protein polarizes membrane growth of Arabidopsis thaliana root hairs,  Journal of Cell Biology ,  2005, vol.  168 (pg.  801- 812) Google Scholar CrossRef Search ADS PubMed  Walker SA,  Viprey V,  Downie JA.  Dissection of nodulation signaling using pea mutants defective for calcium spiking induced by Nod factors and chitin oligomers,  Proceedings of the National Academy of Sciences, USA ,  2000, vol.  97 (pg.  13413- 13418) Google Scholar CrossRef Search ADS   Wang X.  Lipid signaling,  Current Opinion in Plant Biology ,  2004, vol.  7 (pg.  329- 336) Google Scholar CrossRef Search ADS PubMed  Wick P,  Gansel X,  Oulevey C,  Page V,  Studer I,  Durst M,  Sticher L.  The expression of the t-SNARE AtSNAP33 is induced by pathogens and mechanical stimulation,  Plant Physiology ,  2003, vol.  132 (pg.  343- 351) Google Scholar CrossRef Search ADS PubMed  Wiederkehr A,  Avaro S,  Prescianotto-Baschong C,  Haguenauer-Tsapis R,  Riezman H.  The F-box protein Rcy1p is involved in endocytic membrane traffic and recycling out of an early endosome in Saccharomyces cerevisiae,  Journal of Cell Biology ,  2000, vol.  149 (pg.  397- 410) Google Scholar CrossRef Search ADS PubMed  Yokoyama T,  Kobayashi N,  Kouchi H,  Minamisawa K,  Kaku H,  Tsuchiya K.  A lipochito-oligosaccharide, Nod factor, induces transient calcium influx in soybean suspension-cultured cells,  The Plant Journal ,  2000, vol.  22 (pg.  71- 78) Google Scholar CrossRef Search ADS PubMed  © The Author [2007]. Published by Oxford University Press [on behalf of the Society for Experimental Biology]. All rights reserved. For Permissions, please e-mail: journals.permissions@oxfordjournals.org
TI - Root hair curling and Rhizobium infection in Medicago truncatula are mediated by phosphatidylinositide-regulated endocytosis and reactive oxygen species
JF - Journal of Experimental Botany
DO - 10.1093/jxb/erm013
DA - 2007-04-09
UR - https://www.deepdyve.com/lp/oxford-university-press/root-hair-curling-and-rhizobium-infection-in-medicago-truncatula-are-uIb3HN5Xyi
SP - 1637
EP - 1649
VL - 58
IS - 7
DP - DeepDyve
ER -