Target of rapamycin, PvTOR, is a key regulator of arbuscule development during mycorrhizal symbiosis in Phaseolus

Target of rapamycin (TOR) is a conserved central growth regulator in eukaryotes that has a key role in maintaining cellular nutrient and energy status. Arbuscular mycorrhizal (AM) fungi are mutualistic symbionts that assist the plant in increasing nutrient absorption from the rhizosphere. However, the role of legume TOR in AM fungal symbiosis development has not been investigated. In this study, we examined the function of legume TOR in the development and formation of AM fungal symbiosis. RNA-interference-mediated knockdown of TOR transcripts in common bean (Phaseolus vulgaris) hairy roots notably suppressed AM fungus-induced lateral root formation by altering the expression of root meristem regulatory genes, i.e., UPB1, RGFs, and sulfur assimilation and S-phase genes. Mycorrhized PvTOR-knockdown roots had significantly more extraradical hyphae and hyphopodia than the control (empty vector) roots. Strong promoter activity of PvTOR was observed at the site of hyphal penetration and colonization. Colonization along the root length was affected in mycorrhized PvTOR-knockdown roots and the arbuscules were stunted. Furthermore, the expression of genes induced by AM symbiosis such as SWEET1, VPY, VAMP713, and STR was repressed under mycorrhized conditions in PvTOR-knockdown roots. Based on these observations, we conclude that PvTOR is a key player in regulating arbuscule development during AM symbiosis in P. vulgaris. These results provide insight into legume TOR as a potential regulatory factor influencing the symbiotic associations of P. vulgaris and other legumes.


Results
PvTOR is expressed during AM fungal symbiosis in wild-type P. vulgaris. To determine the role of PvTOR in AM fungal symbiosis in the legume, P. vulgaris, we examined the temporal and spatial expression of PvTOR in wild-type P. vulgaris. Quantitative RT-PCR (RT-qPCR) analysis revealed that PvTOR expression varies among organs and tissues (Fig. 1A). PvTOR transcript levels were higher in vegetative organs (hypocotyl, leaf, stem, and root) than in reproductive organs (flower and young pod). To analyze PvTOR expression during the AM fungal symbiosis, we inoculated 5-day-old wild-type seedlings with Rhizophagus irregularis under low Pi (10 µM phosphate (P)) conditions. Uninoculated plants grown under identical conditions were used as controls. Inoculated and uninoculated roots were harvested at various time points and a small portion of each sample was stained with trypan blue to measure the overall percentage of root length colonization (RLC) 32 by AM fungi. All inoculated roots (1, 3, and 6 wpi [weeks post-inoculation] with R. irregularis) were colonized successfully, whereas the uninoculated control roots were free of such colonization (Fig. 1B).
We used the remaining portion of each root sample for RT-qPCR analysis of the expression levels of PvTOR and PvPT-4, a previously identified AM fungi-induced gene. The PvTOR mRNA levels were sharply increased at 3 dpi (days post inoculation) in inoculated roots and remained higher than in the uninoculated roots throughout the rest of the experiment (Fig. 1C). As expected, PvPT-4 expression was induced only in inoculated roots in all tested samples (Fig. 1D). Together, these results indicate that PvTOR is expressed in a majority of P. vulgaris tissues and is upregulated under AM fungi symbiotic conditions. The tissue specificity of PvTOR expression changes in response to AM fungi. The spatial expression patterns of PvTOR during AM fungal symbiosis were investigated using a promoter activity assay in transgenic roots. To create the reporter construct, the 1-kb region upstream of the translation start codon of PvTOR was fused to the chimeric reporter β-glucuronidase (GUS)-enhanced green fluorescent protein (PvTOR pro ::GUS-GFP). The PvTOR pro ::GUS-GFP reporter construct was transfected into P. vulgaris by hairy root transformation and the plants were then inoculated with AM fungi. In uninoculated transgenic roots, GUS staining showed that PvTOR was expressed primarily in the root tip ( Fig. 2A). After inoculation with AM fungi, the promoter activity was enhanced in the root tip and could also be detected in the elongation and maturation zones of the root (Fig. 2B). www.nature.com/scientificreports/ To further evaluate PvTOR promoter activity during AM fungal invasion, we used confocal microscopy to examine PvTOR pro ::GUS-GFP transgenic roots. In this assay, R. irregularis was stained with wheat germ agglutinin (WGA) conjugated to Alexa Fluor 633, which produces far-red fluorescence. Once extraradical hyphae of the AM fungi came into contact with the root epidermis (Fig. 2C), PvTOR promoter activity (GFP fluorescence) was detected at the site of fungal penetration (Fig. 2D-E). PvTOR promoter activity was also observed along the mycorrhized root at 2 wpi ( Fig. 2F-H). However, neither GUS staining nor GFP fluorescence was observed in uninoculated roots or in R. irregularis (data not shown). Taken together, these results show that the PvTOR promoter was active during the course of AM symbiosis.
PvTOR knockdown alters P. vulgaris root growth and the expression of root meristem regulatory genes. Myc-lipochitooligosaccharides produced by AM fungi during mycorrhizal colonization induce lateral root (LR) branching in M. truncatula 33 and increase LR density in P. vulgaris 34 . In Arabidopsis, TOR specifically regulates the proliferation of primary root meristem cells 23 . To clarify the involvement of PvTOR in AM fungi-induced changes in LR development, we used RNAi to knockdown PvTOR expression in transgenic hairy roots of P. vulgaris. Quantitative RT-PCR assays confirmed that PvTOR expression was reduced at least sixfold in the PvTOR-RNAi roots relative to control roots transformed with empty pTdT-DC-RNAi vector (Fig. 3A).
To analyze the effects of PvTOR knockdown on root growth, we measured the primary root growth and LR density of PvTOR-RNAi plants with and without inoculation with R. irregularis. We observed that the primary roots of PvTOR-RNAi plants were shorter than those of control plants at all observed time points, regardless of the presence or absence of R. irregularis (Fig. S1). In control plants, the LR density was much higher for inoculated roots than for uninoculated roots at all time points. By contrast, PvTOR-RNAi roots showed only a marginal RT-qPCR analysis of PvTOR transcript levels in different vegetative and reproductive organs. (B) AM fungal colonized roots were stained and assessed for percent mycorrhizal root length colonization under a light microscope at different time points. The remaining root portions were used for RT-qPCR analysis to measure (C) PvTOR and (D) PT-4 expression in AM roots at different time points. Transcript accumulation was normalized to the expression of EIF4a and IDE, which were used as reference genes. For C and D, the statistical significance of differences between uninoculated and Ri-inoculated roots was determined using an unpaired two-tailed Student's t-test (*P < 0.05; **P < 0.01; ***P < 0.001). For A-D, error bars refer to the SE of the mean of three biological replicates (n > 9). Ri, R. irregularis; dpi, days post inoculation; wpi, week(s) post inoculation. www.nature.com/scientificreports/ increase in LR density at the early time points after inoculation (i.e., 6 dpi; Fig. 3B). Quantitative measurements of the root mass confirmed these observations; a marginal increase of LR mass was observed in inoculated PvTOR-RNAi roots relative to uninoculated PvTOR-RNAi roots and, in general, LR mass was significantly decreased in PvTOR-RNAi roots relative to control roots, whether inoculated or not (Fig. S2). Next, we quantified the transcription of key genes involved in the TOR-mediated regulation of root meristem activity 23 . The transcription factor gene UPB1, whose overexpression inhibits root meristem expansion through redox control 35 , was transcriptionally upregulated in PvTOR-RNAi relative to control inoculated root meristems (Fig. 3C). Other genes were upregulated in inoculated roots relative to uninoculated roots, to levels significantly higher in control roots but only marginally higher in PvTOR-RNAi roots. These genes included those encoding ROOT MERISTEM GROWTH FACTORS (RGF6 and RGF9), SULFUR (Fig. 3 C).  (C) RT-qPCR analysis of the transcript levels of root meristem regulatory genes such as UPB1, RGF6, RGF9, GSS, APK2, SIR, APS1, APR3, ORC5, ETG1, MCM7, and CDC6 in control and PvTOR-RNAi transgenic roots at 6 dpi with Ri. Quantitative RT-PCR was performed on cDNA of root meristem RNA samples. Transcript accumulation was normalized to the expression of EIF4a and IDE, which were used as reference genes. For A, an unpaired two-tailed Student's t-test was used to assess statistical significance (***P < 0.001). For B, the statistical significance of differences was determined using Tukey's test followed by two-way ANOVA the results were statistically significant at p < 0.05 except for the Control (UI) vs. PvTOR-RNAi (UI), Control (UI) vs. PvTOR-RNAi (Ri-I), PvTOR-RNAi (UI) vs. PvTOR-RNAi (Ri-I) at 3 day samples and Control (UI) vs. PvTOR-RNAi (Ri-I) at 6 day samples. For D, Tukey's test followed by two-way ANOVA was used to asses statistical significance and the mean values of each gene with unlike letters were significantly different (P < 0.05). Error bars refer to the SD of the mean of three biological replicates (n > 9 for A & D, n > 30 for B). Ri, R. irregularis. www.nature.com/scientificreports/

Knockdown of PvTOR increases the length of AM extraradical hyphae.
To examine the effect of PvTOR knockdown on the establishment of AM symbiosis, composite P. vulgaris plants (i.e., plants induced to form hairy roots by transformation with Agrobacterium rhizogenes) were inoculated with R. irregularis and monitored from 1 to 6 wpi. Light microscopy observations of inoculated root surfaces revealed that PvTOR-RNAi roots had more extensive extraradical hyphae (ERH) than control roots and that this difference increased over time (Fig. 4A,B).
To further confirm this observation, we measured the lengths of the ERH. The mean length of ERH on PvTOR-RNAi roots was 2.5-fold greater than that of ERH on control roots at 6 wpi (Fig. 4C). It is worth noting that in Medicago truncatula, knockdown of AM symbiosis-induced SUCROSE SYNTHASE 1 (SUCS1) impaired fungal colonization specifically, resulting in fewer radical hyphae and vesicles during AM symbiosis 36 . Further, MtSWEET1b is reported to be responsible for arbuscule maintenance by transporting sugar across the peri arbuscular membrane 37 . We identified the homologues of sugar transporter genes in P. vulgaris based on M. truncatula sequences. The phylogenetic alignment of the sequences showed that MtSWEET1b (Medtr3g089125), PvSWEET1 (Phvul.009G134300) and MtSWEET6 (Medtr3g080990), PvSWEET6 (Phvul.006G000600) were grouped together in the same clade (Fig. S3). With the exception of SWEET1, the expression levels of sugar metabolism and sugar transport genes-namely, SUCS1, SUCS2, CYTOSOLIC INVERTASE 1 (CINV1), and SWEET6-related-were found to be induced in AM fungi inoculated PvTOR-RNAi roots than in mycorrhized control roots (Fig. 4D). Subsequent observations on hyphopodia revealed that the longer ERH on PvTOR-RNAi roots were associated with a significantly higher number of hyphopodia compared to control roots (Fig. 5A, Fig. S4).
We further examined the effect of PvTOR on AM fungal colonization by quantifying root length colonization (RLC) at various time points in control and PvTOR-RNAi roots. Total RLC, arbuscule RLC, and vesicle number were found to increase from 1 to 6 wpi in inoculated control roots. Inoculated PvTOR-RNAi roots showed significantly decreased total RLC (Fig. 5B) and significantly increased vesicle numbers relative to the control at all tested time points (Fig. S5). Furthermore, arbuscule RLC was more than eightfold lower in PvTOR-RNAi roots than in the control (Fig. 5B). Interestingly, the majority of arbuscules present in PvTOR-RNAi roots were RT-qPCR analysis of the transcript levels of sugar metabolism and sucrose transport genes in control and PvTOR-RNAi transgenic roots at 1 wpi with R. irregularis. Data are the fold-change expression relative to uninoculated transgenic roots. Transcript accumulation was normalized to the expression of EIF4a and IDE, which were used as reference genes. The statistical significance of differences between AM control and AM PvTOR-RNAi roots was determined using an unpaired two-tailed Student's t-test (*P < 0.05; **P < 0.01). Error bars refer to the SD of the mean of three biological replicates. www.nature.com/scientificreports/ abnormal ( Fig. 5C-D). These results suggest that PvTOR knockdown causes an increase in ERH, hyphopodia, and vesicles but a decrease in total RLC and arbuscule RLC.
PvTOR is indispensable for arbuscule maturation. Given that PvTOR knockdown affected AM-symbiosis, we examined this phenotype in more detail. We inspected control and PvTOR-RNAi roots for fungal structures at 1 and 3 wpi with R. irregularis. Closer observation revealed that during the initial phases of symbiosis, such as hyphopodia formation and fungal entry into the cortex (intraradical hyphae; IRH), occurred normally in PvTOR-RNAi roots (Fig. 5C-D; Fig. S4). By contrast, although arbuscule development was initiated in PvTOR-RNAi roots, the arbuscules were stunted and clumped ( Fig. 6B-C). Furthermore, the PvTOR-RNAi arbuscules were smaller and less densely branched (Fig. 7B, E) than those of the controls (Fig. 6A, 7A, D). As another measure of arbuscule development, roots were stained with WGA conjugated with Alexa Fluor 488 to label fungal structures and the fluorescence intensity of arbuscule populations was measured using pixel intensity. The majority of arbuscules in PvTOR-RNAi transgenic roots showed fluorescence intensities ranging from 11-30 a.u. compared to intensities of 30-70 a.u. in controls (Fig. 7C). To precisely determine the arbuscule size range in control and PvTOR-RNAi roots, we measured the cross-sectional areas of arbuscule populations and sorted them into size categories. The majority of arbuscules in control roots had an area of 400-600 µm 2 , while in PvTOR-RNAi roots the area was significantly reduced, with most of arbuscules in the 100-200 µm 2 range (Fig. 7F). Over 90% of arbuscules on PvTOR-RNAi roots were stunted and this phenotype persisted at all observed time points (Fig. 6D; Fig. S6). By contrast, the control arbuscules were well developed and highly branched, completely filling the arbuscule-containing cortical cells (Fig. 6A, 7A,D). Interestingly, the number of AM vesicles was significantly higher in the PvTOR-RNAi roots than in the control (Fig. S5).
To test whether PvTOR is required for the uptake of P by the AM fungi, we measured P concentrations in inoculated and uninoculated PvTOR-RNAi and control roots at 3 and 6 wpi. A portion of the root samples was stained to determine AM fungal colonization and the total % RLC was approximately the same as previously shown in Fig. 5B. However, in inoculated plants, the total P concentrations were significantly lower in PvTOR-RNAi roots i.e., 40.2% at 3wpi and 46.9% at 6wpi; and 53.2% at 3wpi and 61.7% at 6wpi in shoots relative to the controls. The same was true of uninoculated plants; P concentrations were lower in PvTOR-RNAi roots www.nature.com/scientificreports/ and shoots than in the corresponding control samples (Fig. S7), confirming that in the absence of PvTOR, P uptake and transport are disturbed. Therefore, we conclude that PvTOR is essential for mycorrhizal P uptake in P. vulgaris.

Activation of the common symbiosis pathway and expression of AM fungi-induced genes are altered in PvTOR-knockdown roots. In M. truncatula, induction of GRAS-type transcription factors
specific to mycorrhizal signaling depends on activation of the common symbiosis signaling pathway (CSSP) 38 .
To confirm that this is also the case for R. irregularis-inoculated transgenic roots of P. vulgaris, we used RT-qPCR to analyze transcript accumulation of the P. vulgaris CSSP genes, SYMRK (Phvul.002G143400), CCAMK (Phvul.011G186900), and IPD3 (Phvul.002G128600), and of genes encoding the AM symbiosis-specific GRAStype TFs, NODULATION SIGNALING PATHWAY 2 (NSP2) (Phvul.008G165200) REDUCED ARBUSCULAR MYCORRHIZATION 1 (RAM1) (Phvul.001G089900) and RAM2 a GLYCEROL-3-PHOSPHATE ACYL TRANS-FERASE (GPAT) (Phvul.007G233600). Transcripts for both CCAMK and IPD3 were less abundant in PvTOR-RNAi roots, while there was no difference in SYMRK expression relative to controls (Fig. 8A). Transcripts for RAM1 and RAM2 were significantly more abundant in PvTOR-RNAi roots than in the control. AM specific markers PvPT4 and P. vulgaris homologues Phvul.003G143400 and Phvul.010G050900 of Oryza sativa AM1 39 and M. truncatula H + ATPase, HA1 40 were found to be induced in PvTOR-RNAi roots confirming successful colonization of the symbiont (Fig. 8B). Other genes induced by AM symbiosis include VAPYRIN (VPY) 41 and VAMP (SNARE) 42 in M. truncatula and STR1 and STR2 in Oryza sativa 43,44 . Mutation of these genes limits arbuscule growth of the AM fungi, resulting The statistical significance of differences between control and PvTOR-RNAi roots was determined using an unpaired two-tailed Student's t-test (***P < 0.001). Error bars refer to the SD of the mean of three biological replicates (n > 30 for D). a, arbuscule; ih, intercellular hyphae; sa, stunted arbuscule, iah, intracellular hyphae; wpi, week(s) post inoculation, Ri, R. irregularis. www.nature.com/scientificreports/ in a small and stunted arbuscule phenotype 43,44 . We measured the expression of the P. vulgaris homologs of these genes in inoculated roots and found that-with the exception of STR2-their transcript levels were significantly lower in PvTOR-RNAi roots than in control roots (Fig. 8B). Together, these results suggest that knockdown of PvTOR disrupts the expression of genes involved in arbuscule development but not those encoding AM symbiosis-specific GRAS-type TFs.

Discussion
In plants, TOR plays a central regulatory role in modulating multiple cellular activities, including embryogenesis, meristem activation, root and leaf growth, flowering, and life span determination, as well as controlling photosynthesis, autophagy, and senescence 23,45,46 . Information about the role of this important protein in the regulation of symbiotic interactions is quite fragmentary. We previously showed that in the model legume P. vulgaris, TOR is involved in regulating rhizobial symbiosis, including infection thread progression and nodule organogenesis 30 .
The focus of our current study was to improve our understanding of the role of PvTOR in regulating interactions with another important endosymbiont, AM fungi. To this end, we analyzed the expression pattern of PvTOR in P. vulgaris, in the presence and absence of AM fungi, and studied the effect of PvTOR knockdown on AM symbiosis. In terms of PvTOR expression, our key finding was that PvTOR promoter activity, assayed in P. vulgaris hairy roots transformed with a PvTOR pro ::GUS-GFP reporter construct, increased in response to AM inoculation. The promoter activity in root cortical cells increased in coordination with the ramifying AM fungal mycelia, from the hyphopodium to the ERH to the IRH, an indication of mycorrhiza specific PvTOR promoter expression.
Transformation of P. vulgaris hairy roots with a pTdT-PvTOR RNAi construct resulted in an almost six-fold reduction in PvTOR transcript abundance. This PvTOR knockdown inhibited the proliferation of lateral roots that is normally caused by AM inoculation 47 . The TOR signaling pathway is also known to regulate the expression of genes involved in the proliferation of progenitor cells for root meristem activation and growth in Arabidopsis 23,45 . An analysis of the expression of root meristem regulatory genes (RGF, sulfur assimilation, and S-phase genes) in uninoculated PvTOR-RNAi hairy roots revealed significantly lower transcript levels relative to PvTOR-RNAi hairy roots inoculated with AM fungi at 6 dpi. We also observed increased expression of the transcription factor gene UPB1, whose overexpression is known to inhibit root meristem expansion through redox control 35 . We propose that UPB1 is one of the genes regulated by PvTOR, though further experimental studies are required to verify how TOR and redox regulatory signaling execute the cell proliferation through UPB1 transcription factor.
The ERH perform two main functions: they increase the surface area for mineral and water absorption from the soil and transport to the host via the arbuscule-cortical cell interface and they provide structures capable of colonizing new roots via hyphopodia 48 . Our observation shows that ERH became more extensive when the arbuscules were truncated by PvTOR knockdown.
AM fungal development in the host is governed by the nutritional status of the host plant. In response to AM colonization, the sink strength of host plant roots increases, allowing more sucrose to be unloaded from the phloem and exported toward the arbuscular cells. This increase in sink strength involves the activity of sucrose-cleaving enzyme invertases (INV) and SUCROSE SYNTHASES (SUCS) and the tight regulation of sucrose transporters 49,50 . AM-inoculated PvTOR-RNAi roots showed high expression levels of the sucrose synthase genes SUCS1 and SUCS2 and the CYTOSOLIC INVERTASE gene CINV1. The most important finding in these tissues was the transcript downregulation of the sugar transporter PvSWEET1, a homologue of MtSWEET1b. In M. truncatula, SWEET1b is localized to the periarbuscular membrane and is responsible for providing the arbuscule containing cells with the required carbon source for the growth and differentiation of arbuscules. However, the gene encoding the sugar transporter PvSWEET6 was induced in AM-inoculated PvTOR-RNAi roots by two-fold compared to AM-inoculated controls. Hence, PvSWEET6 could be the potential sugar transporter involved in the AM-specific source-to-sink sucrose transport in PvTOR-RNAi plants. Due to the stunted nature of arbuscules in PvTOR-RNAi roots, the available carbon source might be taken up by intraradicular hyphae via an unknown mechanism, as previously proposed by Bago and colleagues 51 . The increase in vesicle numbers could be to store the extra carbon received from the host.
An increase in ERH was also found to increase the number of hyphopodia. The GRAS-type transcription factors RAM1 52 and RAM2 53 function in hyphopodia formation. Similarly, in P. vulgaris, there was an increase in transcript accumulation of NSP2, RAM1, and RAM2 in mycorrhized PvTOR-RNAi roots, providing molecular confirmation for the increased hyphopodia phenotype.
Analyses of the stunted arbuscule phenotype of PvTOR-RNAi roots revealed that the expression levels of VPY 41,54 , VAMP 55 , and STR 43,44 were supressed. STR and STR2 function as a heterodimer in the periarbuscular membrane that may possibly export a nutrient signal molecule essential for arbuscule development, as in M. truncatula 43 . In PvTOR-RNAi roots, STR expression was significantly reduced whereas STR2 expression increased during AM symbiosis, implying that in the absence of STR, STR/STR2 heterodimer complex formation is affected and could account for the stunted arbuscules in PvTOR-RNAi roots.
Taken together, our study suggests that PvTOR permits arbuscule maturation during AM symbiosis in P. vulgaris by modulating AM-specific sugar transporters and arbuscule-specific genes. Though there was a surge in mycelial growth and hypopodial numbers, the P content of the PvTOR-RNAi plants did not differ from that of the controls. To improve our understanding of the involvement of PvTOR in the common symbiotic signaling pathway of legumes, future studies should examine the downstream proteins, subcellular localization of TOR, and physiological mechanism of legume TOR regulation during symbiotic interactions.

Methods
Plant materials, inoculation, and growth conditions. Seeds of Phaseolus vulgaris L. cv. Negro Jamapa were obtained from Instituto de Biotecnología, UNAM, Mexico. The seeds were surface-sterilized 56 , germinated in the dark on wet filter paper for two days at 28 °C, transferred to sterile vermiculite, and grown under a 16-h photoperiod at 28 ± 1 °C. All the experiments involving plants were carried out in accordance with appropriate guidelines. Five-day-old plants were inoculated with Rhizophagus irregularis (800 spores/plant) [Symplanta, Darmstadt, Germany] and irrigated twice weekly with half-strength B&D solution 57 containing a low concentration of potassium phosphate (10 µM, K 2 HPO 4 ) to promote AM colonization 13 . At different time points, a portion of root sample (50% of total root volume) was excised from each plant, immediately frozen in liquid nitrogen, www.nature.com/scientificreports/ and stored at -80 °C for RNA extraction. The remaining portion of the root samples was stained to determine the percent root length colonization of AM fungi 32 . A set of plants grown separately under identical conditions but without R. irregularis inoculation served as controls.

Plasmid construction and generation of composite plants. To develop the RNAi construct of
PvTOR, a fragment corresponding to the non-conserved region of the C terminus and 3′-UTR of PvTOR (Phvul.002G049900) was amplified from the cDNA isolated from the root tips of 2-day-old germinated P. vulgaris, using specific oligonucleotides (Table S1). The PCR product was recombined with pTdT-DC-RNAi vector using the Gateway system (Invitrogen, Carlsbad, California, USA). The correct orientation of the clone was confirmed by sequencing the insert of the plasmid. Empty pTdT-DC-RNAi vector was used as the control. Upstream of the TOR translation start site, a 1-kb promoter fragment was amplified from P. vulgaris genomic DNA using specific primers (Table S1) and cloned into the pENTR/SD/D-TOPO vector (Invitrogen, Carlsbad, California, USA). The Gateway LR reaction was performed between the entry vector pENTR/SD/D-TOPO-PvTOR and the destination vector pBGWSF7.0 according to the manufacturer's instructions (Invitrogen). The Agrobacterium rhizogenes/K599 strain carrying the corresponding constructs was used to initiate hairy root formation on P. vulgaris tissues and form composite plants after transformation 56 . Transgenic hairy roots expressing the PvTOR-RNAi vector or PvTOR pro ::GFP-GUS were selected under an epifluorescence stereomicroscope based on red fluorescent protein (RFP) and green fluorescent protein (GFP) expression, respectively. RFP fluorescence was excited at 561 nm by a solid-state laser and emission was filtered using a band pass filter of 640/50 nm. GFP fluorescence was excited with a blue argon ion laser (488 nm) and emitted fluorescence was collected from 510 to 540 nm.
PvTOR promoter analysis. The PvTOR pro ::GFP-GUS promoter construct was transfected into common bean cv. Negro Jamapa by hairy root transformation, and the resulting transgenic composite plants were inoculated with R. irregularis (~ 800 spores per plant). The roots were harvested at 5-14 dpi and stained either for promoter fusion GUS activity according to Jefferson 58 or WGA (Wheat Germ Agglutinin) conjugated to Alexa Fluor 633 (Invitrogen, Carlsbad, California, USA) to visualize fungal structures 59 in RED fluorescence, using a ZEISS LSM-510 confocal laser-scanning microscope. The PvTOR promoter activity was monitored in transgenic roots expressing PvTOR pro ::GUS-GFP. WGA-Alexa Fluor 633 (red channel) was excited with an argon ion laser (633 nm), and emitted fluorescence was collected from 652 to 752 nm.
RNA extraction and quantitative real-time PCR analysis. Total RNA was isolated from P. vulgaris roots using TRIzol reagent, according to the manufacturer's recommendations (Thermo Scientific, Waltham, USA). Genomic DNA contamination from RNA samples was eliminated by incubating the samples with RNasefree DNase (1 U µl -1 ) at 37 °C for 15 min and then at 65 °C for 10 min. RNA integrity and concentration were determined by electrophoresis and NanoDrop ND-2000 (Thermo Scientific, Wilmington, USA) spectrophotometry, respectively.
Quantitative real-time PCR was performed using an iScript One-step RT-PCR Kit with SYBR Green (Bio-Rad, Hercules, California, USA), following the manufacturer's instructions, in an iQ5 Multicolor Real-time PCR Detection System (Bio-Rad, Hercules, California, USA). Each reaction was set up using 40 ng of RNA as template. A control sample, which lacked reverse transcriptase (RT), was included to confirm the absence of contaminant DNA. Relative gene expression levels were calculated using the formula 2 -ΔCT , where cycle threshold value (ΔCT) is the CT of the gene of interest minus the CT of the reference gene. P. vulgaris EIF4a (Phvul.010G136300) and IDE were used as control genes, as previously described 23,60 . The relative expression values, normalized with two reference genes, were calculated according to Vandesompele and colleagues 61 . The data are averages of three biological replicates and each sample was assessed in triplicate. The expression of genes listed in Table S1 was quantified using gene-specific oligonucleotides.

Root growth parameters. Composite plants grown in pots of vermiculite and irrigated with B&D medium
were used to determine root growth parameters and superoxide accumulation in transgenic roots. Transgenic roots expressing RFP were selected at various intervals from both uninoculated and R. irregularis-inoculated plants and root growth parameters were recorded. Lateral root density was calculated using the formula: D = LR/ L´, where D = density of lateral roots; LR = number of lateral roots; and L´ = length of the main root between the first and last lateral root 62 .
Quantification of mycorrhizal colonization and microscopy analysis. The AM fungi-inoculated roots were sampled at 1, 3, and 6 wpi and stained with trypan blue using the modified histochemical staining method 32 or WGA-Alexa Fluor 488 59 to measure the mycorrhizal colonization. Using a light microscope (DMLB Bright-field Microscope; Leica, Wetzlar, Germany), trypan blue-stained root samples were analyzed to visualize fungal structures (extraradical hyphae, hyphopodia, intraradical hyphae, vesicles, and arbuscules) and assess the root length colonization (percent RLC) as per McGonigle and colleagues 32 . Lengths of extra-radical hyphae (ERH) were determined according to a published protocol 63 with some modifications 64 . Arbuscule size was measured using images obtained using a ZEISS LSM-510 confocal laser scanning microscope (ZEISS, Oberkochen, Germany). Z-stacks of Alexa Fluor 488-stained mycorrhized roots were generated from 12-18 serial images taken at increments of 1.25 µM, and analyzed using the LSM 5 tool. Alexa Fluor 488 (green channel) was excited with an argon ion laser (488 nm) and emitted fluorescence was collected from 510 to 540 nm.