• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of pnasPNASInfo for AuthorsSubscriptionsAboutThis Article
Proc Natl Acad Sci U S A. Jan 30, 2007; 104(5): 1720–1725.
Published online Jan 22, 2007. doi:  10.1073/pnas.0608136104
PMCID: PMC1785290
Plant Biology

A Medicago truncatula phosphate transporter indispensable for the arbuscular mycorrhizal symbiosis


The arbuscular mycorrhizal (AM) symbiosis is a mutualistic endosymbiosis formed by plant roots and AM fungi. Most vascular flowering plants have the ability to form these associations, which have a significant impact on plant health and consequently on ecosystem function. Nutrient exchange is a central feature of the AM symbiosis, and AM fungi obtain carbon from their plant host while assisting the plant with the acquisition of phosphorus (as phosphate) from the soil. In the AM symbiosis, the fungus delivers Pi to the root through specialized hyphae called arbuscules. The molecular mechanisms of Pi and carbon transfer in the symbiosis are largely unknown, as are the mechanisms by which the plant regulates the symbiosis in response to its nutrient status. Plants possess many classes of Pi transport proteins, including a unique clade (Pht1, subfamily I), members of which are expressed only in the AM symbiosis. Here, we show that MtPT4, a Medicago truncatula member of subfamily I, is essential for the acquisition of Pi delivered by the AM fungus. However, more significantly, MtPT4 function is critical for AM symbiosis. Loss of MtPT4 function leads to premature death of the arbuscules; the fungus is unable to proliferate within the root, and symbiosis is terminated. Thus, Pi transport is not only a benefit for the plant but is also a requirement for the AM symbiosis.

Keywords: biotrophic, membrane, mineral nutrition, mutualism, root

Plants and their arbuscular mycorrhizal (AM) fungal symbionts have coexisted for >400 million years (1), and the association is widespread in terrestrial ecosystems. AM symbiosis provides multiple benefits for the plant, not only enhanced phosphorus and nitrogen nutrition but also tolerance to pathogens and abiotic stresses (2, 3). For AM fungi, formation of a symbiosis is an obligate requirement. Their large spores contain reserves sufficient to support growth of a hyphal germ tube, but they must form an association with a plant to acquire additional carbon to complete their life cycle (4). Development of symbiosis, in particular intraradical development of the fungus in the root cortex, is reduced if Pi availability is high (57). This suggests that the plant has a mechanism of regulating fungal growth, presumably to avoid the unnecessary allocation of carbon resources. The molecular basis is unknown but could occur by the control of carbon allocation (8).

In the AM symbiosis, the fungal hyphae grow through the intercellular spaces of the root and subsequently invade the inner cortical cells, developing branched hyphae, called arbuscules, within the cells (9, 10). As each arbuscule forms, the plant cell envelops it in a membrane, the periarbuscular membrane, and the result is an extensive plant–fungal interface specialized for nutrient exchange (1114). It is predicted that Pi and carbon transfer occur at the arbuscule/cortical cell interface, although direct evidence for carbon transfer at this location is lacking (14). Current data suggest that Pi is translocated through the AM fungal hyphae as polyphosphate (polyP), and after hydrolysis, in the arbuscule, Pi is exported from the AM fungus to the periarbuscular space (1517). The import of Pi across the periarbuscular membrane, into the root cell is then mediated by plant transporters. Currently, two classes of Pi transporters are implicated in Pi transport in AM symbiosis: mycorrhiza-induced Pi transporters of Pht1, subfamily III; and mycorrhiza-specific Pi transporters of Pht1, subfamily I. Members of subfamily III, such as StPT3 of potato, ZEAma;Pht1;6 of maize, and LjPT3 of Lotus japonicus, are expressed in roots but show increased expression in symbiosis (1821). Partial down-regulation of LjPT3 reduced symbiotic Pi transfer ≈2-fold, indicating that this transporter contributes to the symbiotic Pi transport process (21). By contrast, members of subfamily 1, such as OsPT11 of rice, MtPT4 of M. truncatula, and LePT4 of tomato, are expressed exclusively in AM symbiosis (19, 22, 23), and their functions have not been described. All plants examined, except for Arabidopsis, a nonmycorrhizal plant species, possess at least one transporter in this subfamily. However, there is variation between plant species; rice contains one subfamily I Pi transporter (22), whereas potato and tomato each have two; and in tomato, it is proposed that they are functionally redundant (19). MtPT4, a member of subfamily I, is located in the periarbuscular membrane of M. truncatula, further suggesting a role in symbiotic Pi transfer (23).

Using both transgenic lines in which MtPT4 is down-regulated by RNAi and MtPT4 loss-of-function mutants identified by target-induced local lesions in genomes (TILLING), we demonstrate that MtPT4 is essential for symbiotic Pi transport and development of AM symbiosis.

Results and Discussion

MtPT4 Is Required for Enhanced Pi Content and Growth Resulting from AM Symbiosis.

Transgenic M. truncatula plants expressing an RNAi construct designed to the 3′ end of the MtPT4 coding region were generated. In two independent transgenic lines designated MtPT4 RNAi 1 and RNAi 2, MtPT4 was effectively silenced, and MtPT4 transcripts and protein were not detectable [Fig. 1A and supporting information (SI) Fig. 6 A–F]. In contrast, two other independent transformants, MtPT4 RNAi 3 and 4, showed MtPT4 transcript and protein levels comparable to those of the vector control lines (V1 and V2), indicating that silencing of MtPT4 had not occurred in these lines (Fig. 1A). These lines were designated RNAi 3 (ND, for non-down-regulated) and RNA:4 (ND) and were used as additional controls. The region of MtPT4 targeted by the RNAi construct shares <42% identity at the nucleotide level with other phosphate transporters of the PHT1 family. Consistent with this sequence divergence, the expression of other PHT1 family members was not altered in the RNAi lines (SI Fig. 6G).

Fig. 1.
Molecular and physiological analyses of the MtPT4 RNAi transgenic plants and Mtpt4-1 mutants. (A) MtPT4 transcript (Top) and protein levels (Bottom) in homozygous vector control (V1 and V2) and MtPT4 RNAi transgenic lines (R 1, 2, 3, and 4) assayed by ...

The transgenic plants were inoculated with an AM fungus, Glomus versiforme, and grown under low-Pi conditions to reveal increases in shoot P content and biomass resulting from the symbiosis. Under these conditions, the shoot P content and shoot mass of the mock-inoculated plants were similar in all transgenic lines (Fig. 1 B and C). A significant increase in shoot P levels (average increase of 230%) and shoot mass (average increase of 500%) associated with the mycorrhizal treatment was apparent in the vector and MtPT4 RNAi (ND) control plants. By contrast, MtPT4 RNAi 1 and 2 plants inoculated with G. versiforme did not differ significantly from their mock-inoculated controls. These data indicate that MtPT4 is required for the increase in Pi content and growth associated with AM symbiosis and suggest that other Pi transporters do not compensate for the loss of MtPT4. Furthermore, additional mycorrhiza-inducible Pi transporters, which, given their existence in other plants, are predicted to exist in M. truncatula also, must play a minor or different role in AM symbiosis.

Loss of MtPT4 Function Affects the Morphology and Intra- and Extraradical Growth of the AM Fungus.

The root systems of these plants were analyzed to determine the extent of development of the symbiosis. The control and RNAi (ND) lines showed an average of 45% root length colonized (RLC), and fungal structures typical of the symbiosis were apparent, including intercellular hyphae, arbuscules, vesicles, and extraradical hyphae. By contrast, total fungal development in MtPT4 RNAi 1 and 2 plants was significantly lower, with only 12% and 8% RLC, respectively. In L. japonicus, a 1.5-fold reduction in arbuscule frequency was observed in a LjPT3 knockdown line; however, fungal morphology and development of the fungus within the root were not altered (21). In contrast, in the MtPT4 RNAi 1 and 2 plants, there was a striking difference in the morphology of the fungus relative to the controls. Arbuscules were infrequent, and those that were present were degenerating; moreover, the intercellular hyphae contained septa, which are morphological signatures of death in AM fungi (SI Fig. 7) (11, 12). This phenotype was unexpected and suggested that loss of MtPT4 function also affected development of G. versiforme within the root.

To confirm this, we used TILLING to identify an ethane methyl sulfonate mutant allele of MtPT4, mtpt4-1. mtpt4-1 contains a G-to-A transition within a transmembrane domain-coding region, resulting in the replacement of glycine at position 172 with arginine. Structure–function analyses of transporter proteins have shown that in transmembrane domains, the replacement of noncharged residues such as glycine, with charged amino acids, leads to protein instability and loss of transport activity (24, 25). Consistent with this, MtPT4 protein was not detected in mycorrhizal roots of mtpt4-1 (SI Fig. 8 A–F). Furthermore, the transport activity of the mtpt4-1 mutant transporter protein was assayed in yeast and, although the protein is expressed, Pi transport activity was not stimulated, indicating that the mutation impacts transport activity (SI Fig. 8G). The mtpt4-1 mutant is effectively a null mutant.

mtpt4-1 and WT (A17) plants were inoculated with G. versiforme, harvested at 54 days postinoculation (dpi), and stained with a vital stain followed by a cell-wall stain to evaluate both live and total fungal structures in the roots (26). In A17 roots, colonization levels assessed by cell-wall and vital staining were 71% and 65% RLC, respectively, indicating that G. versiforme was alive and proliferating within the roots (Fig. 1D). In addition, development of the extraradical hyphae was extensive (Fig. 1E). In contrast, in mtpt4-1 roots, infection units were short and consisted mostly of degenerate arbuscules and septate hyphae, which were dead. Colonization levels assessed by cell-wall and vital staining were 23% and 2% RLC, respectively (Fig. 1D). Furthermore, extraradical hyphae did not develop (Fig. 1E). These data indicate that the G. versiforme can invade mtpt4-1 roots, but the difference in colonization levels assessed by total and vital staining indicates that fungal growth is not sustained in the mtpt4-1 mutant. The low colonization and collapsed arbuscule phenotypes of mtpt4-1 (Fig. 2) are the same as those of the MtPT4 RNAi 1 and 2 lines (SI Fig. 7). Furthermore, in an F2 population (59 individuals) derived from a cross of mtpt4-1 and A17, the mtpt4-1 genotype shows perfect cosegregation with the collapsed degenerating arbuscule phenotype (SI Fig. 9).

Fig. 2.
Laser-scanning confocal microscope images of G. versiforme arbuscules in roots of M. truncatula WT (A) and mtpt4-1 (B and C). Images are from roots at 146 hpi (taken from the experiment described in Fig. 4) but are representative of the majority of arbuscules ...

The Absence of MtPT4 Leads to the Accumulation of polyP in the Arbuscules.

The classical approach to examine Pi transfer in AM symbiosis uses radioisotopes to monitor Pi delivery by the fungus (2, 27), and our initial intent was to further support the physiological studies shown in Fig. 1 with such analyses. However, during the course of the phenotypic studies, it became apparent that the mycorrhizal phenotype of the MtPT4 RNAi and mtpt4-1 mutants was unexpectedly severe. As revealed by the experiments described in Fig. 1, growth of the fungus is not sustained in MtPT4 RNAi and mtpt4-1 roots, and extraradical hyphae do not develop (Fig. 1E). Without a living extraradical mycelium in mutant lines, the results of radioisotope transfer experiments are inconclusive. As an alternative, we used polyP staining (16) to determine whether loss of MtPT4 function affected the distribution of polyP in the fungus within the root. In the cortex of vector control and A17 roots, G. versiforme intercellular hyphae stained strongly for polyP, whereas the arbuscules showed weak staining, consistent with the hydrolysis and export of Pi to the plant. In MtPT4 RNAi 1, RNAi 2, and mtpt4-1 roots, G. versiforme intercellular hyphae, and arbuscules showed strong staining for polyP, indicating that polyP accumulates in the arbuscules in these lines (Fig. 3 and SI Fig. 10). The polyP staining patterns are consistent with the shoot P data (Fig. 1B) and suggest that loss of MtPT4 function results in a block in symbiotic Pi transfer from the arbuscule to the cortical cell.

Fig. 3.
Toluidine blue staining at pH 1 shows polyP as dark-red metachromatic granules in the intercellular hyphae (indicated by arrowheads) of G. versiforme in the V1 (A) and MtPT4 RNAi 2 roots (B). Arbuscules (indicated by arrows) stain strongly in MtPT4 RNAi ...

MtPT4 Is Required to Maintain Arbuscule Life and to Sustain Development of the AM Fungus.

The altered morphology of the arbuscules and the failure of the fungus to become established in the MtPT4 RNAi and mtpt4-1 roots suggested a possible alteration in arbuscule development or lifespan in the mutant lines. Arbuscule development requires the differentiation of both symbionts and is partially controlled by the plant (9). In M. truncatula, G. versiforme arbuscules develop by repeated dichotomous branching and grow until they fill the cortical cell. At some point after maturity, the arbuscules collapse, degenerate, and die. The mechanisms underlying arbuscule turnover are unknown but do not involve plant cell death. Each invaded plant cell remains alive throughout the life of the arbuscule and can host successive arbuscules (11, 12, 28). Studies of arbuscule lifespan are limited, but where examined, the lifespan is suggested to be ≈8.5 days (29). Consequently, in mycorrhizal roots, arbuscule formation and decay are ongoing, and the roots contain a population of developing and degenerating arbuscules.

To evaluate arbuscule development, MtPT4 RNAi, mtpt4-1, and their respective controls were inoculated with pregerminated G. versiforme spores that had been primed to colonize roots by preexposure to root exudates (3032). Roots were harvested at 49, 74, and 146 h postinoculation (hpi) and 11 dpi and stained to reveal the fungus. Individual infected roots, with a distinct infection event arising from a single fungal entry point, referred to as an infection unit (11), were analyzed (SI Fig. 11). Infection unit lengths were measured and, to obtain a quantitative measure of arbuscule development, the length of individual arbuscules in the infection unit was measured. The data from the two independent vector controls (V1 and V2) and the two independent MtPT4 RNAi lines (RNAi 1 and 2) were combined. At 49 hpi, infection unit length in MtPT4 RNAi lines and mtpt4-1 did not differ significantly from their respective controls, indicating that loss of MtPT4 does not affect the initial rate of growth of the fungus within the root cortex (SI Table 1). At 49 hpi, the populations of arbuscules in the vector and MtPT4 RNAi plants were similar, although the vector control plants showed a trend toward smaller arbuscules compared with MtPT4 RNAi plants (significant for the 25- to 35- and 45- to 55-μm size classes) (Fig. 4). Although not significant, the pattern was similar for A17 and mtpt4-1 plants. At 74 hpi, there were no significant differences between the arbuscule populations in the vector and MtPT4 RNAi plants or in mtpt4-1 and its respective A17 control. Relative to the 49 hpi timepoint, the median arbuscule size in all populations had increased to ≈42 μm, indicating that the arbuscules were growing. At 146 hpi, significant differences in the arbuscule populations became apparent (Fig. 4). The arbuscule populations in the vector and A17 roots showed median arbuscule sizes of 43 and 44 μm, respectively, and an increasing proportion of the population was represented in the large-size classes, indicating that the arbuscules continue to grow. In contrast, in the MtPT4 RNAi and mtpt4-1 plants, the arbuscule populations showed a shift in the opposite direction. In each case, there was a strongly significant increase in the proportion of arbuscules in the 15- to 25-μm size class and a corresponding significant decrease in arbuscules in the 55- to 65-μm class. For mtpt4-1, a significant decrease in the 45- to 55-μm size class was observed also. The median size of arbuscules fell to 32 and 28 μm in the MtPT4 RNAi and mtpt4-1 plants, respectively. Morphological analyses indicated that this shift in the arbuscule population was associated with arbuscule collapse and degeneration. Additional quantification demonstrated that, in MtPT4 RNAi and mtpt4-1 roots, a significantly higher percentage of degenerating arbuscules was present even in the youngest regions of the infection unit (SI Table 2). To confirm that this trend was maintained, the analysis of two plant lines was further extended to 11 dpi. At 11 dpi in V1 roots, 73% of the arbuscules were represented in size classes of >35 μm. By contrast, in MtPT4 RNAi 1 roots, the arbuscule population was in significant decline, and 72% of the arbuscules were represented in size classes <35 μm (data not shown).

Fig. 4.
Analysis of arbuscule development in M. truncatula RNAi lines and mtpt4-1 inoculated with G. versiforme. Frequency graphs showing arbuscule size (length) distribution in arbuscule populations in vector (dotted line) and MtPT4 RNAi (solid lines) plants ...

Although the quantitative shift in the arbuscule populations was measured at 146 hpi, qualitative differences in the arbuscules were apparent at an earlier stage. Septa, an early morphological feature associated with arbuscule degeneration (11, 12), were visible first at 76 hpi in MtPT4 RNAi roots and 49 hpi in mtpt4-1 roots. The phenotype in mtpt4-1 is marginally more severe than the RNAi lines. Overall, these data indicate that, in plants lacking MtPT4, the initial development of arbuscules proceeds normally, but death of the arbuscule is triggered prematurely, as rapidly as 49 hpi, and further development of AM symbiosis cannot occur. Vital staining of WT and mtpt4-1 roots at 76 hpi confirmed that arbuscules with turgid branches without septa were alive, whereas collapsed arbuscules containing septa were dead. Regardless of the state of the arbuscules, the invaded root cortical cells remained alive in both mutant and WT plants (Fig. 5). There are no signs of hypersensitive cell death, a typical plant strategy to eliminate an unwanted microorganisms. Therefore, the premature arbuscule collapse and death in the mtpt4 mutants likely occur by the acceleration of the normal mechanism of arbuscule turnover, the molecular basis of which is still unknown.

Fig. 5.
Vital staining of WT and mtpt4-1 roots colonized with G. versiforme. Cell-wall and vital staining reveal total fungal structures (fluorescent green staining) and living fungal structures (purple/blue precipitate), respectively. Vital staining identifies ...

Phosphate Transfer Through MtPT4 Signals the Presence of a Beneficial Fungal Symbiont.

In summary, the data indicate that MtPT4, a mycorrhiza-specific Pi transporter, is required for Pi transfer in symbiosis, but more significantly, Pi transport by MtPT4 is clearly essential for AM symbiosis. Without MtPT4 function, the arbuscules die prematurely, and fungal growth is not sustained. Currently, it is unclear whether rapid death of the arbuscule is a fungal response to the local conditions, or whether it is triggered by the plant. The accumulation of polyP in the arbuscule may be toxic; however, this seems unlikely, because in WT roots, both the extra- and intraradical hyphae can accumulate significant levels of polyP (33, 34). Alternatively, arbuscule death could be triggered by the plant, either actively by acceleration of the normal mechanism of arbuscule turnover or simply as a consequence of inadequate carbon flow from the plant. A link between carbon and phosphate exchange in AM symbiosis has not been shown directly but has received extensive discussion, particularly relating to the evolutionary stability of AM symbiosis (14, 35). In a recent article (8), it was proposed that the mechanisms of carbon allocation that support lateral root proliferation in nutrient-rich patches might be used in symbiosis. In this way, Pi delivery by the fungus would trigger increased carbon allocation specifically to the region of the root containing the fungus (8).

We propose that the import of Pi by MtPT4 serves, either directly or indirectly, as a signal to the plant cell of the presence of a beneficial fungal symbiont. In the absence of this signal, death of the arbuscule ensues, and proliferation of the fungus is prevented. In further support of this, symbiosis does not develop under extremely low Pi conditions, and infection is stimulated dramatically with the application of small amounts of Pi (6, 36). Furthermore, although AM fungi can provide benefits other than enhanced Pi (37, 38), radioisotope experiments suggest that some symbiotic Pi transfer occurs in all AM symbioses (2). Both of these points are consistent with our proposal that Pi flow across the periarbuscular membrane is essential for symbiosis.

If, as suggested by our data, Pi delivery to the cortical cell serves as a signal to permit continued development of the arbuscule and consequently to sustain fungal existence within the root, then the plant has a robust mechanism to identify and discard an ineffective fungal symbiont before a significant commitment of resources.

Materials and Methods

Transgenic M. truncatula Lines.

A DNA fragment corresponding to nucleotides 1363–1587 of the MtPT4 cDNA (relative to the ATG start codon) was cloned into pHellsgate 8 (39). The construct was introduced into Agrobacterium tumefaciens (LBA4404) and used to transform M. truncatula (A17) (40). Plants were transformed with the nonrecombinant pHellsgate 8 as vector controls. Primary transformants were selfed, and lines that segregated 3:1 for the presence of the NPTII gene (n = 100–120 plants) in the T2 generation were selected. Plants homozygous for a single T-DNA insertion were identified by progeny testing and the presence of the T-DNA verified by PCR.

M. truncatula mtpt4 Mutants.

A M. truncatula mutant, mtpt4-1, was identified from an ethane mehtyl sulfonate mutant population of M. truncatula cv. Jemalong, genotype A17 by TILLING (41) by using FIR1 primers forward (5′-CTGGCATGGGATTCTTCACT-3′) and reverse (5′-CGGTTTCAGGCATTTTCATT-3′). The TILLING population consisted of 4,032 independently derived mutant M. truncatula germ lines. Mutant alleles were recovered at an average rate of one SNP per 377 germ lines per kbp interrogated, which corresponds to an average of ≈1,330 lesions per genome (germ line). mtpt4-1 (G172R) has a G-to-A transition that alters glycine (position 172) to arginine.

Cosegregation Analysis.

Individual F2 plants (59) were assayed in parallel for the zygosity of the mutation and the mtpt4-1 phenotype. (Details are provided in SI Fig. 9 and SI Text.)

Plant Growth Conditions.

Unless otherwise indicated, plants were grown in a growth chamber under a 16-h-light (25°C)/8-h-dark (22°C) regime in sterile Turface (Profile Products, Buffalo Grove, IL) and inoculated with surface-sterilized G. versiforme spores, as described (42). Plants were fertilized twice weekly with ½ strength Hoagland's solution (43) containing 20 μM Pi. Roots were prepared for staining according to Grace and Stribley (44) and stained in 0.2 μg/ml WGA-Alexafluor 488 (Molecular Probes, Eugene, OR). Percent RLC was monitored by the modified gridline intersect method (45).

To monitor mycorrhiza-associated growth responses (Fig. 1), seedlings were planted into an acid-washed sand/Turface mixture (ratio 2:1) and inoculated with 1,000 G. versiforme spores. Plants were harvested at 72 dpi, weighed, and the total phosphorus level was measured according to Ames (46). To monitor development of AM symbiosis in mtpt4-1 (reported Fig. 1 D and E), 10-day-old plants were inoculated with 800 G. versiforme spores and harvested at 54 dpi. To estimate extraradical hyphal length, root pieces with living hyphae attached were floated in liquid to extend the hyphae from the root. Measurements were made by using an Olympus (Melville, NY) SZX12 stereoscope. To follow arbuscule development and turnover, a synchronized inoculation system was used (32). Thirty-five-day-old plants were placed in contact with 700 G. versiforme spores that had received a 21-day exposure to root exudates. Plants were harvested at 49, 74, and 146 hpi and 11 dpi (three plants per line at each timepoint) and stained with WGA-Alexafluor 488. By using Metamorph software (Metamorph Universal Imaging, West Chester, PA), arbuscule size was determined by measuring the length of all visible arbuscules in independent infection units (n = 5 infection units for each line), with an average of 21, 42, and 66 arbuscules per infection unit at 49, 74, and 146 h, respectively. At 11 dpi, analyses were limited to observation of 28 arbuscules per infection unit. An infection unit is the infection event arising from a single entry point into the root (11). The oldest hyphae are near the appressorium and entry point, whereas the youngest hyphae are at the growing hyphal fronts (SI Fig. 11).

Staining Procedures.

Vital staining was according to Schaffer and Peterson (26) followed by WGA-Alexafluor 488 staining. PolyP staining was according to Ezawa et al. (16).

Gene Expression Analyses.

RNA extraction and RT-PCR protocols followed standard procedures. The amplification of the elongation factor EF1α was used as a control. MtPT4 primers, qMtPT4-L4 5′-GACACGAGGCGCTTTCATAGCAGC-3′; qMtPT4-R4 5′-GTCATCGCAGCTGGAACAGCACCG-3′. EF1α primers P1: 5′-GTCAAAACATGGTTGCTGCACAAGC-3′ and P2: 5′-TTAGGTCACAAGGCAGATTGCAGG-3′.

MtPT4 Immunodetection.

Microsomal proteins preparation, Western blot analysis, and immunolocalization of the MtPT4 protein were performed as described (23).

Statistical Analyses.

The effects of loss of MtPT4 function on shoot P, shoot growth, and arbuscule populations were tested on 5, 5, and 12 independently grown plants per line, respectively. Data were analyzed by using ANOVA to test for the effects of plant genotype on the parameters. When a significant effect was detected (P < 0.02), pairwise comparisons (Student's t test) were performed by using SAS 9.1 software (SAS Institute, Cary, NC) to identify which genotypes showed statistically significant differences (P < 0.05). All data presented are means ± SE, except Fig. 1E, where the SD is shown. The statistical significance (P value) is reported in the legends of the Figs. 115, SI Figs. 6–11, and SI Tables 1 and 2.

For analyses of arbuscule populations, arbuscules were grouped into size classes ranging from 15 to 75 μm in 10-μm increments, and for each individual infection unit, the distribution was expressed as a percentage of arbuscules in the infection unit. For the two independent vector control and MtPT4 RNAi lines, the results presented in Fig. 4 are the pooled data of individual lines 1 and 2. To fulfill the assumption of a normal distribution, the percentages were transformed by using the ARCSIN transformation before performing ANOVA.

Student's t test was used for other pairwise comparisons (SI Tables 1 and 2; Fig. 1 D and E).

Supplementary Material

Supporting Information:


We thank Tom Kraft for assistance with the genetic analysis of the mutants and Melina Lopez-Meyer for assistance with the setup of the arbuscule turnover experiment. Financial support for the MtPT4 project was provided by the U.S. National Science Foundation (Grant IBN-0343975, to M.J.H.). The M. truncatula TILLING project was supported by the U.S. Department of Agriculture (Grant CRIS 3640-21000-019-025, to D.R.C).


arbuscular mycorrhizal
root length colonized
days postinoculation
hours postinoculation
target-induced local lesions in genomes.


The authors declare no conflict of interest.

This article is a PNAS direct submission.

This article contains supporting information online at www.pnas.org/cgi/content/full/0608136104/DC1.


1. Redeker D, Kodner R, Graham L. Science. 2000;289:1920–1921. [PubMed]
2. Smith SE, Smith FA, Jakobsen I. Plant Physiol. 2003;133:16–20. [PMC free article] [PubMed]
3. Newsham KK, Fitter AH, Watterson AR. J Ecol. 1995;83:991–1000.
4. Bago B, Pfeffer PE, Shachar-Hill Y. Plant Physiol. 2000;124:949–957. [PMC free article] [PubMed]
5. Braunberger PG, Miller MH, Peterson RL. New Phytol. 1991;119:107–113.
6. Koide R, Li M. New Phytol. 1990;114:59–74.
7. Thomson BD, Robson AD, Abbott LK. New Phytol. 1986;103:751–765.
8. Fitter AH. New Phytol. 2006;172:3–6. [PubMed]
9. Parniske M. Curr Opin Plant Biol. 2004;7:414–421. [PubMed]
10. Harrison MJ. Annu Rev Microbiol. 2005;59:19–42. [PubMed]
11. Cox G, Sanders F. New Phytol. 1974;73:901–912.
12. Bonfante-Fasolo P. In: VA Mycorrhizae. Powell CL, Bagyaraj DJ, editors. Boca Raton, FL: CRC; 1984. pp. 5–33.
13. Gianinazzi-Pearson V, Smith SE, Gianinazzi S, Smith FA. New Phytol. 1991;117:61–74.
14. Smith SE, Smith FA. New Phytol. 1990;114:1–38.
15. Cox G, Moran KJ, Sanders F, Nockolds C, Tinker PB. New Phytol. 1980;84:649–659.
16. Ezawa T, Cavagnaro TR, Smith SE, Smith FA, Ohtomo R. New Phytol. 2003;161:387–392.
17. Kohjima T, Saito M. Mycol Res. 2004;108:610–615. [PubMed]
18. Rausch C, Daram P, Brunner S, Jansa J, Laloi M, Leggewie G, Amrhein N, Bucher M. Nature. 2001;414:462–466. [PubMed]
19. Nagy F, Karandashov V, Chague W, Kalinkevich K, Tamasloukht M, Xu GH, Jakobsen I, Levy AA, Amrhein N, Bucher M. Plant J. 2005;42:236–250. [PubMed]
20. Nagy R, Vasconcelos M, Zhao S, McElver J, Bruce W, Amrhein N, Raghothama K, Bucher M. Plant Biol. 2006;8:186–197. [PubMed]
21. Maeda D, Ashida K, Iguchi K, Chechetka S, Hijikata A, Okusako Y, Deguchi Y, Izui K, Hata S. Plant Cell Physiol. 2006;47:807–817. [PubMed]
22. Paszkowski U, Kroken S, Roux C, Briggs SP. Proc Natl Acad Sci USA. 2002;99:13324–13329. [PMC free article] [PubMed]
23. Harrison MJ, Dewbre GR, Liu J. Plant Cell. 2002;14:2413–2429. [PMC free article] [PubMed]
24. Galic S, Schneider HP, Broer A, Deitmer JW, Broer S. Biochem J. 2003;376:413–422. [PMC free article] [PubMed]
25. Chen LY, Pan CJ, Chou JY. Hum Mol Genet. 2002;11:3199–3207. [PubMed]
26. Schaffer GF, Peterson LR. Mycorrhiza. 1993;4:29–35.
27. Pearson JN, Jakobsen I. New Phytol. 1993;124:489–494.
28. Gianinazzi-Pearson V, Gianinazzi S. In: Cell to Cell Signals in Plant, Animal and Microbial Symbiosis. Scannerini S, Smith DC, Bonfante-Fasolo P, Gianinazzi-Pearson V, editors. Berlin: Springer; 1988. pp. 73–84.
29. Alexander T, Toth R, Meier R, Weber HC. Can J Bot. 1989;67:2505–2513.
30. Tamasloukht M, Sejalon-Delmas N, Kluever A, Jauneau A, Roux C, Becard G, Franken P. Plant Physiol. 2003;131:1468–1478. [PMC free article] [PubMed]
31. Akiyama K, Matsuzaki K-I, Hayashi H. Nature. 2005;435:824–827. [PubMed]
32. Lopez-Meyer M, Harrison MJ. Federico Sánchez CQ, López-Lara IM, Geiger O, editors. Mol Plant–Microbe Interact. 2006;Vol 5:546–551.
33. Viereck N, Hanson PE, Jakobsen I. New Phytol. 2004;162:783–794.
34. Ohtomo R, Saito M. New Phytol. 2005;167:571–578. [PubMed]
35. Johnson NC, Graham JH, Smith AF. New Phytol. 1997;135:575–587.
36. Bolan NS, Robson AD, Barrow NJ. Soil Biol Biochem. 1984;16:419–420.
37. Hodge A, Campbell CD, Fitter AH. Nature. 2001;413:297–299. [PubMed]
38. Govindarajulu M, Pfeffer PE, Jin HR, Abubaker J, Douds DD, Allen JW, Bucking H, Lammers PJ, Shachar-Hill Y. Nature. 2005;435:819–823. [PubMed]
39. Helliwell CA, Wesley SV, Wielopolska AJ, Waterhouse PM. Funct Plant Biol. 2002;29:1217–1225.
40. Trieu AT, Harrison MJ. Plant Cell Rep. 1996;16:6–11. [PubMed]
41. McCallum CM, Comai L, Greene EA, Henikoff S. Plant Physiol. 2000;123:439–442. [PMC free article] [PubMed]
42. Liu JY, Blaylock LA, Harrison MJ. Can J Bot. 2004;82:1177–1185.
43. Arnon DI, Hoagland DR. Soil Sci. 1940;50:463–483.
44. Grace C, Stribley DP. Mycol Res. 1991;95:1160–1162.
45. McGonigle TP, Miller MH, Evans DG, Fairchild GL, Swan JA. New Phytol. 1990;115:495–501.
46. Ames BN. Methods Enzymol. 1966;8:115–118.

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...