* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Download Physiologia Plantarum
Survey
Document related concepts
Plant stress measurement wikipedia , lookup
History of botany wikipedia , lookup
Plant defense against herbivory wikipedia , lookup
Ornamental bulbous plant wikipedia , lookup
Plant secondary metabolism wikipedia , lookup
Historia Plantarum (Theophrastus) wikipedia , lookup
Evolutionary history of plants wikipedia , lookup
Plant reproduction wikipedia , lookup
Plant breeding wikipedia , lookup
Plant physiology wikipedia , lookup
Plant morphology wikipedia , lookup
Perovskia atriplicifolia wikipedia , lookup
Plant use of endophytic fungi in defense wikipedia , lookup
Plant ecology wikipedia , lookup
Plant nutrition wikipedia , lookup
Sustainable landscaping wikipedia , lookup
Transcript
Copyright C Physiologia Plantarum 2002 ISSN 0031-9317 PHYSIOLOGIA PLANTARUM 115: 125–136. 2002 Printed in Denmark – all rights reserved Expression of nitrate transporter genes in tomato colonized by an arbuscular mycorrhizal fungus Ulrich Hildebrandta, Elmon Schmelzerb and Hermann Bothea,* a Botanisches Institut, Universität zu Koeln, Gyrhofstr. 15, D-50923 Köln, Germany Max-Planck Institut für Züchtungsforschung, Carl-von-Linné-Weg 10, D-50829 Köln, Germany *Corresponding author, e-mail: hermann.bothe/uni-koeln.de b Received 25 July 2001; revised 15 November 2001 PCR amplifications using tomato DNA and degenerate oligonucleotide primers allowed identification of a new putative nitrate transporter, termed NRT2;3. Its sequence showed typical motifs of a high affinity nitrate transporter of the Major Facilitator Superfamily (MFS). The formation of its mRNA was positively controlled by nitrate, and negatively by ammonium, but not by glutamine. In situ hybridization experiments showed that this transporter was mainly expressed in rhizodermal cells. Results from expression studies with two other nitrate transporters, LeNRT1;1 and LeNRT2;1, were essentially in accord with data of the literature. In roots colonized by the arbuscular mycorrhizal fungus Glomus intraradices Sy167, transcript formation of NRT2;3 extended to the inner cortical cells where the fungal structures, arbuscules and vesicles, were concentrated. Northern analyses indicated that the expression of only NRT2;3 among the transporters assayed was higher in AMF colonized tomato roots than in non-colonized controls. AMF-colonization caused a significant expression of a nitrate reductase gene of G. intraradices. The results may mean that AMF-colonization positively affects nitrate uptake from soil and nitrate allocation to the plant partner, probably mediated preferentially by LeNRT2;3. In addition, part of the nitrate taken up is reduced by the fungal partner itself and may then be transferred, when in excess, as glutamine to the plant symbiotic partner. Introduction More than 80% of the higher plants are capable of forming a symbiosis with arbuscular mycorrhizal fungi (AMF). The hyphae of the fungi exploit water and nutrients from soil particles more efficiently than the roots, and these components are effectively transferred from the fungi to the plant (Smith and Read 1997). This is undoubtedly the case for phosphate nutrition (Pearson and Jakobsen 1993), and likely happens also for Kπ, Zn2π, Cu2π and other elements (Kothari et al. 1991). As regard to phosphate, studies have reached the molecular level. A specific AMF phosphate translocator was shown to be expressed upon mycorrhizal colonization, whereas the plant phosphate translocator counterparts were repressed (Harrison and van Buuren 1995, Liu et al. 1998). For nitrogen, physiological data on nitrate uptake are somewhat equivocal (Hawkins and George 1999) but determinations of the 15NO3– and 15NH4π uptake rates indicated a significant contribution of AMF also to the N-budget of the plants (Johansen et al. 1992, Frey and Schuepp 1993, Tobar et al. 1994, Bago et al. 1996, Subramanian and Charest 1998). Expression studies using Physiol. Plant. 115, 2002 gene probes specific for either fungal or maize nitrate reductase indicated that transcript levels of the plant enzyme are lowered and those of the fungal counterpart are enhanced upon mycorrhizal colonization (Kaldorf et al. 1998). This was shown both by Northern analysis and in situ hybridization experiments. The expression of transporters catalysing the uptake of nitrate from the soil across the plasmalemma of either plants or fungi has not yet been investigated in the AMF symbiosis. However, it has just been published that a putative nitrate transporter from Medicago truncatula is downregulated in roots when colonized by some, but not all AMF (Burleigh 2001). Nitrate uptake in plants is complex. This anion is transferred across the plasmalemma of the root cells by a 2 Hπ/NO3– symport mechanism (Ullrich and Nowacky 1991) with the involvement of at least three kinetically distinct transport systems (recent reviews: Crawford and Glass 1998, Forde 2000). Two high affinity transport systems (the so-called HATSs) operate at concentrations ⬍100 mM. One of these is expressed in the absence of 125 nitrate and is therefore considered as being constitutive (cHATS). Its activity is, however, enhanced several fold by nitrate treatment at least in some plants. The other high affinity system (iHATS) is strongly induced by incubating plants with nitrate and is negatively feedbackregulated by products of nitrate assimilation, in particular by ammonium and glutamine (Crawford and Glass 1998, Forde 2000). The low affinity transport system (LATS) operates at NO3– concentrations ⬎1 mM, surprisingly displays linear NO3– uptake kinetics but is also considered as an active Hπ-dependent system (Crawford and Glass 1998, Forde 2000). Recently, a flurry of reports came out on the cloning and molecular characterization of genes encoding both high and low affinity transporters (Crawford and Glass 1998, Forde 2000). The nitrate transporters characterized so far are 500–600 amino acids in length and possess two sets of six transmembrane helices connected by a cytosolic loop. Low affinity transporters are encoded by the NRT1 gene family. These have been characterized in a variety of plants including tomato in which at least two different NRT1 genes occur. Transcript accumulation of LeNRT1;2 was reported to be restricted to root hairs and to be nitrate inducible whereas that of LeNRT1;1 is also found in other root cells and is constitutively expressed (Lauter et al. 1996). The NRT2 family of genes encodes high affinity transporters with high sequence similarities in plants, fungi and algae (Crawford and Glass 1998, Forde 2000) and its expression is under feedback repression by N-metabolites resulting from nitrate reduction. The NRT2 genes can also occur in multiple forms in organisms. In tomato (Ono et al. 2000), the two genes LeNRT2;1 and LeNRT2;2 are highly similar in the coding region, but have significant nucleotide differences in the 3ƒ-untranslated part. In root hairs of tomato, NRT2;1 seemingly is the predominantly expressed gene, and mRNA formation of both genes appears to be co-ordinatively regulated (Ono et al. 2000). In barley, so far four different NRT2 genes have been described, but it is predicted that the genome of this plant might contain at least seven members of the NRT2 gene family (Trueman et al. 1996, Vidmar et al. 2000). The Arabidopsis genome project revealed the presence of two further NRT2 genes (AtNRT2;3 and 2;4; Forde 2000) in addition to the already characterized two genes AtNRT2;1 and 2;2 (Filleur et al. 1999, Zhuo et al. 1999). Thus the discovery of further genes of both the NRT1 and NRT2 families would not be unexpected for any plant. In the present study, the sequences available in the databanks were screened for conserved motifs of nitrate transporters, which were used to develop primers suitable for the amplification of NRT segments by PCR. This approach resulted in the identification of a new putative high affinity nitrate transporter (LeNRT2;3) with specific regulatory properties in its expression not yet described for any other of the NRT gene families. This publication mainly describes the results obtained on the differential expression of NRT genes in AMF colonized and control tomato plants. 126 Materials and methods Plant and fungal material Tomato (Lycopersicon esculentum Mill., var. Tamina GS, Schmitz and Laux, D-Hilden, Germany) and the AMF isolate Glomus intraradices Schenck & Smith INVAM Sy167 were the symbiotic partners in the present study. The greenhouse experiments were performed with surface-sterilized tomato seeds and G. intraradices inoculum produced with Tagetes patula L. as a host plant (Kaldorf et al. 1998). To ensure a high degree of mycorrhizal colonization, tomato was grown under phosphate limitation. Plants were watered daily. The plants were supplied twice weekly with a 1: 1 diluted Hoagland’s nutrient solution (7.5 mM sodium nitrate, no phosphate, 30 ml per plant) and, additionally, once a month with this solution plus 0.5 mM KH2PO4 as described in detail by Kaldorf et al. (1998). The degree of mycorrhizal colonization in root pieces was determined microscopically by a modified version of the gridline intersect method as described by Schmitz et al. (1991). Extraradical mycelia of G. intraradices Sy167 were isolated from pot cultures with two compartments, one for growth of roots and mycorrhizal structures (root compartment) and the other (hyphal compartment) filled with quartz sand (2–4 mm). Both were separated from each other by a 30-mm nylon mesh (Hydrobios, Kiel, Germany) penetrable only by hyphae, but not by roots. Three months after inoculation, hyphae were harvested from the quartz sand by differential wet-sieving with sieves of varying pore sizes (1 mm, 0.08 mm). The mycelia retained on the 0.08 mm sieve were dried on a paper towel for some seconds and immediately frozen in liquid nitrogen prior to use. Isolation of nucleic acids, Southern and Northern analyses Genomic DNA was isolated from tomato leaves as described by Raeder and Broda (1985) and RNA according to Nagy et al. (1988). For Southern experiments, DNA samples were digested with different restriction enzymes at 37æC. DNA fragments (20 mg per lane) were separated on an 0.8% agarose gel, transferred onto Biodyne B (Pall) nylon membranes using a vacuum blotting manifold (2016 VacuGeneXL Vacuum Blotting System, Pharmacia LKB) according to the manufacturer’s protocol and fixed by UV-exposure. Hybridizations were carried out with Digoxigenin labelled DNA-probes synthesized by incorporation of DIG-dUTP (from Roche, D-Mannheim, Germany) by PCR. Hybridization (at 42æC, using DIG-Easy-Hyb, Roche, Germany) and filter washing were carried out as recommended by the manufacturer (Roche). The DIG-labelled hybridized DNA was detected immunologically by alkaline phosphatase conjugated antibodies with CSPD or NBT/BCIP as substrate. For Northern analyses, total RNA (20 mg laneª1) was Physiol. Plant. 115, 2002 separated on 1.2% formaldehyde agarose gels in the presence of ethidium bromide. Chemiluminescent and colourimetric Northern blot analyses were carried out at least three times with identical results in each case using Digoxigenin-labelled riboprobes. The vectors pCRII or pGEM-Teasy, containing the cloned PCRproducts were linearized and used as template for in vitro transcription with Sp6 or T7 RNA polymerase producing DIG-labelled antisense-RNA probes. Signal intensities were quantified with the Computer Program NIH Image 1.62. After this hybridization with the gene probe for one of the nitrate transporters or nitrite reductase, filters were stripped for a subsequent hybridization with an 18S-rRNA probe. This probe was obtained by amplifying the nucleotides 1575–1783 (X51576) of the tomato 18S-rRNA by PCR. Synthesis of cDNA from tomato cDNA from tomato root or shoot total RNA (2 mg per reaction) was synthezised using the ‘SUPERSCRIPT Preamplification System for First Strand cDNA Synthesis’ (Gibco BRL, Karlsruhe, Germany) according to the manufacturer’s protocol. Subsequently, the obtained cDNA was digested with RNase H (2 U per ml, Gibco BRL) and purified using the QIAquick PCR Purification Kit (Qiagen, Hilden, Germany). PCR protocols to amplify segments of nitrate transporters A PCR-based approach with a set of degenerate primers was performed to amplify DNA segments of tomato NRT1 and NRT2 homologues (PNT1: 5ƒ-ACGTATTTG ACGGGAACNATGCAYYT-3ƒ and PNT2: 5ƒ-GCGCA AA TTCCMTANCCCCAYTC-3ƒ, for members of the NRT1-family; or PNT3: 5ƒ-TTCGTGTCGTGYCARR AYTGGATG-3ƒ and PNT4: 5ƒ-CATGGATCCCC AYT GNGGRAARTG-3ƒ for members of the NRT2-family). Using tomato DNA as a template, PCR mainly yielded amplification products of approximately 1.3 kb (PNT1/ PNT2) and 1.05 kb (PNT3/PNT4) in length applying a ‘touch-down’ method: 94æC, 4 min (94æC, 30 s; 54æC [ª 2æC/cycle], 30 s; 72 æC, 60 s) for 7 cycles and (94æC, 30 s; 50æC, 30 s; 72æC, 60 s) for 33 cycles and 72æC, 5 min (TaqPolymerase and buffers by Promega Corp., Madison, USA). With tomato root cDNA as a template, RACEPCR was performed to obtain the 3ƒ-UTR’s of members of the NRT1 and NRT2-family (Frohman et al. 1988); applying the following primers: RACE1: 5ƒ-CCACGAG TCGACTCTAGAGCTCGGATCC TTTTTTTTTTT TTTTTT-3ƒ; RACE2: 5ƒ-CCACGAGTCGACTCTAG3ƒ; RACE3: 5ƒ-CTCTAGAGCTCGGATCC-3ƒ as well as the gene specific primers PNT5–7: GCGTGTACACTTCCAGTAATGTTAGT for LeNRT2;3, PNT5–6: GGATGCACTCTTCCTGTTACATTTTG for Le NRT2;2 and PNT2A: CAAAGTAACGGGGGATAAACCATG for LeNRT1;1. For the production of specific DIG-and [35S]-labelled Physiol. Plant. 115, 2002 riboprobes and in order to increase their specificity, the following primers were used to produce shortened PCRproducts of the cloned segments devoid of the poly A/T stretch of the nitrate transporter gene segments: RTHAT1B: ATGAGGAAGTGATTACAAATG (LeNRT2;3); RTHAT2B: TTCCAGCGTATTGACAAGC (LeNRT2;2); RLEMIT1: CTTTGGAAATTGCTCATTCGA (LeNRT1;1). The primers PL1 and PL2 (Kaldorf et al. 1998) were used to amplify a 544-bp segment of the tomato nitrate reductase gene whereas the primers TNIRF2 (CAGAARTGTMGNATGATGTGG) and TNIRR1 (TTCGGGCANCCNGTCCARTGCAT) served to amplify a 533-bp cDNA-segment of a nitrite reductase gene. The obtained PCR-products were cloned into the vector pGEM-Teasy (Promega, Madison, USA). Their subsequent sequencing was carried out with an ABI PRISM (Ready Reaction Dye Terminator Cycle Sequencing Kit (PE-Biosystems, Weiterstadt, Germany) and an Applied Biosystems ABI PRISM (310 Genetic Analyser. The NCBI (Bethesda, MD, USA) BLAST-server was used for sequence comparisons. In situ hybridization experiments Fixation of tomato root pieces with 1% glutaraldehyde, embedding in Fibrowax (Plano, Twetzlar, Germany), and sectioning of the Fibrowax embedded tissue were performed as described (Schmelzer et al. 1989). Longitudinal sections (12 mm) were adhered to poly -lysine coated microscope slides. The sections were treated with pronase and fixed with 1% glutaraldehyde. For the in situ hybridization experiments, sections were treated exactly as previously described (Kaldorf et al. 1998). The vectors pCRII or pGEM-Teasy, containing the cloned PCRproducts, were linearized and used as templates for in vitro transcription with Sp6 or T7 RNA polymerase producing 35S-labelled sense and antisense-RNA probes. Signals were detected by microautoradiography (Schmelzer et al. 1988) and quantified by the image analysis program Image-ProR-Plus 4.1 (Media Cybernetics, Silver Spring, MD, USA). Each in situ hybridization experiment was performed at least three times with identical results. Determination of nitrate reductase activity NADH and NADPH-dependent NR activity of crude extracts from leaves and roots of tomato and of the extraradical mycelium of G. intraradices was measured as described by Lohaus et al. (1998) and Kaldorf et al. (1998). Results Identification of a new high affinity transporter, LeNRT2;3, from tomato The sequences deposited into the databanks allowed to develop primers for NRT1, NRT2, and nitrite reductase 127 (NII). For nitrate reductase (NIA gene), primers were available from the previous study (Kaldorf et al. 1998). Using the primers PNT1 and PNT2 (see Materials and methods) for the low affinity transporter NRT1 and genomic DNA as template, one PCR product of the right size and composition was obtained in the case of Arabidopsis, whereas several products were detected for tomato DNA (Hildebrandt 2001). To differentiate NRT1 gene PCR-products from artifacts in tomato, a digoxigenin-labelled segment located internally in the one Arabidopsis PCR-product (nucleotides 441–739, L10357) was synthesized and used to detect a NRT1 gene segment from tomato by heterologous probing. Cloning and sequencing verified its identity as NRT1 gene segment from tomato. In the case of NRT2 gene segments, the primers PNT3 and PNT4 provided several PCRproducts of similar size (1.0–1.1 kb) with genomic DNA from Arabidopsis, tomato or Nicotiana plumbaginifolia as template. After extraction from the gels, cloning and sequencing, two different clones (termed T1–6 and T1– 7) were identified in the case of tomato (for the other plants, see Hildebrandt 2001). Partial sequencing of these clones revealed an identity of only 85% among them in this DNA area. The clone T1–6 was 99% identical (674 bp determined) whereas T1–7 showed 85% identity (in 396 bp) to both LeNRT2;1 and LeNRT2;2 sequences of the databanks. To obtain more specific probes for differentiating among them, the untranslated, less conserved 3ƒ-terminal region was employed for getting PCR-products by the RACE-technique (Frohman et al. 1988) with the use of specific primers (for their composition see Materials and methods). This approach provided a 470-bp segment of LeNRT1;1, a 374-bp PCR product of LeNRT2;2 and a further one, with 431 bp, of a gene termed LeNRT2;3 with distinct sequence dissimilarities. In the case of the latter two, the sequence identities in the putatively transcribed region were only 77% on the DNA and 86% on the amino acid level suggesting that a new gene had been found with LeNRT2;3. To demonstrate that these clones were indeed from tomato, genomic DNA from this plant was used for Southern blot analysis with probes of the 374 bp PCR product of LeNRT2;2 and of the 431 bp of LeNRT2;3. Indeed, two completely different hybridization patterns were obtained (Fig. 1). To verify that LeNRT2;3 was, in fact, a new nitrate transporter homologue, a cDNA clone with the complete coding sequence, obtained by PCR, revealed the following structure: The length was 1646 bp with an ORF of 531 amino acids and with 12 putative transmembrane helices, including a distinct connecting cytosolic loop between domains 6 and 7. The hydropathy plot (data not shown) closely resembled that of GmNRT2 (Amarasinghe et al. 1998). The sequence identity, on the amino acid level, was 89% to LeNRT2;1, 88% to LeNRT2;2 and at least 72% to the nitrate transporters from other plants GmNRT2, AtNRT2;1, AtNRT2;2, OsNRT2. The motifs AGWGNMG typical for nitrate transporters (Trueman et al. 1996) and 128 GAICDMLGPR, specific to members of MFS (Forde 2000) were present; and the conserved protein kinase C recognition motifs were exactly as described in Forde (2000). The sequence is available under the accession number AY038800. The expression of the different transporters in tomato tissues For the hybridization experiments, digoxigenin labelled probes were generated for NRT1;1 (432 bp), NRT2;2 (374 bp), NRT2;3 (431 bp), nitrate reductase (NIA, 544 bp) and nitrite reductase (NII, 533 bp) by PCR. In Southern experiments using the PCR products, signal intensities were at least 103 weaker in heterologous hybridization with any combination of the three transporters NRT1;1, NRT2;2 or NRT2;3 than in homologous probing (Hildebrandt 2001). Northern experiments showed that the gene expression of these three transporters as well as of NIA and NII underwent diurnal fluctuations (Hildebrandt 2001) as known from data of others (Forde 2000, Ono et al. 2000). In roots, NIA expression was maximal at 3 pm and that of NII at 6 pm. The expression of LeNRT1;1 was highest in the morning until 3 pm and declined in the evening, whereas Northern blot intensities were still high in the evening in the case of NRT2;3. To have standardized material, 12to 14-week-old plants were incubated for 5 days with the different N-sources prior to harvest between 5 and 7 pm, and stored in the deep-freezer until the isolation of RNA. In Northern experiments with total RNA isolated from roots of this material (Fig. 2), formation of NRT1;1, NRT2;2 and NRT2;3 and, as a control, of NII transcripts, were positively regulated by nitrate but negatively regulated by ammonium in this case, except for NRT1;1. Incubation with both nitrate and ammonium resulted in expression levels being intermediate of the application with nitrate or ammonium alone. Glutamine depressed the transcript levels of NII, NRT1;1 and NRT2;2, but remarkably not of NRT2;3 (Fig. 2, lane 2). This lane indicates that glutamine even appeared to induce its level. The Northern hybridization data of Fig. 2 were reproducibly observed in different experiments using two independent plant batches. In the same root material, nitrate reductase activity, determined by the formation of nitrite from nitrate, was highest in nitrate incubated roots (about 1 mmol hª1 gª1 FW). This rate was only 1/4 as high in glutamine- and 1/10 in ammonium-incubated roots (Hildebrandt 2001). In roots, intensive hybridization bands were detectable in the case of the new high affinity transporter NRT2;3 and weaker ones with NRT2;2, whereas no signals were obtained with RNA from leaves. In contrast, bands for NRT1;1 were seen in roots and leaves, albeit with lower signal intensities in leaves than in roots. To verify this by a more sensitive approach, total RNA was isolated from roots, flowers, leaves and young fruits of tomato and translated to cDNA using the primer RACE1. The Physiol. Plant. 115, 2002 Fig. 1. Genomic Southern blot analysis of tomato nitrate transporter genes LeNRT2;2 and LeNRT2;3. Tomato genomic DNA (20 mg laneª1) was digested (6 h, 37æC) with EcoRI (lane 1), EcoRV (lane 2) and HindIII (lane 3), fractionated on a 0.8% agarose gel and transferred onto a nylon membrane. Blot A was probed with a partial LeNRT2;3 cDNA clone (431 bp) and blot B with a 374-bp LeNRT2;2 cDNA clone. Blot B showed a very similar pattern as described for LeNRT2;2 (Ono et al. 2000). RT-PCR experiments confirmed that NRT1;1 was expressed in all parts of the plant. This approach also confirmed that NRT2;1 was not expressed in any of the aerial parts of the plants, even when nested PCR was employed. For NRT2;3, a weak signal was detectable, however, only in the case of young fruits in this nested PCR experiment (not shown, but see Hildebrandt 2001). Antisense and sense probes of NRT1;1 and NRT2;3 were developed to analyse mRNA transcript formation in young tomato roots by in situ hybridization experiments. In order to get high expression levels of the nitrate transporters, 12-week-old tomato plants were kept in N-deficient medium for 1 week and then incubated in Hoagland solution supplemented with 7.5 mM nitrate for 6 h. The differences in the labelling on the microscopic slides obtained with the antisense and sense probe indicated that the transcript level of NRT1;1 was high in all young root cells close to the tip, including rhizodermal and inner cells, but was restricted to the inner cortical cells in root segments at the hair zone (Fig. 3I). Transcript formation of NRT2;3 was also high in all cells close to the root tip. In more differentiated root tissue, next to the root hair zone, silver grain deposition due to NRT2;3 expression was confined to rhizodermal Fig. 2. The formation of tomato root mRNA of nitrate transporters and nitrite reductase in dependence of the N-source in the medium. Twelve-week-old tomato plants were incubated with 1: 1 diluted Hoagland nutrient solution, pH 6.0, supplemented with the different N-sources for 6 days. Lane 1; no extra nitrogen, only 0.042 mM from (NH4)6Mo7O27. H2O, lane 2: plus 7.5 mM glutamine, lane 3: 3.75 mM nitrate and 3.75 mM glutamine, lane 4: 7.5 mM ammonium, lane 5: 3.75 mM ammonium and 3.75 mM nitrate and lane 6: 7.5 mM nitrate. For these Northern blots, root total RNA (20 mg laneª1, isolated from three plants per lane) was hybridized to probes for LeNRT1;1, LeNRT2;2, LeNRT2;3 and tomato NII (Ω nitrite reductase). Details are described in Materials and methods. Physiol. Plant. 115, 2002 129 cells (Fig. 3II). The preparation method did not preserve root hairs, therefore any expression could not be studied there. Since hybridization with the sense probe as control gave high background labelling in all cells, silver grains in the antisense and sense probed samples were counted in 100 mm2 areas using a computer imaging program (Table 1A). The data obtained from these quantifications confirmed within the statistic significance the conclusions just drawn from the visual inspections of the in situ hybridizations on the microscopic slides. Comparison of the expression of nitrate transporters in AM colonized and control plants For these experiments, 14-week-old plants were used with a degree of mycorrhizal colonization of 78 ∫ 3% and contents of arbuscules of 45 ∫ 8%, vesicles of 67 ∫ 5% and intraradical hyphae of 77 ∫ 3%. Control plants had at best 4% of all these structures and were therefore considered as non-mycorrhizal. Plants had about the same developmental stage (beginning of the formation of red berries in the lowest inflorescence), but flowers had developed 1 week earlier in AMF plants than in the controls. The day before harvest the plants had been supplemented with 30 ml of a 1 : 1 water-diluted Hoagland solution. For NRT1;1, no distinct signals were obtained in mycorrhizal roots and non-mycorrhizal controls in the in situ hybridization experiments under these conditions. Completely different results were obtained for NRT2;3. In mycorrhizal roots, transcripts of NRT2;3 were clearly detectable in the inner cortical cells in tissues close to the root hair zone (Fig. 4A) whereas Fig. 3. The distribution of LeNRT1;1 and LeNRT2;3 mRNA in tomato root tissue. LeNRT1;1 mRNA accumulates in epidermal cells of young roots and in cortical cells of older root tissues whereas LeNRT2;3 mRNA was detectable only in the cells of the rhizodermis as shown by in situ hybridization. Twelve-week-old non-mycorrhizal tomato plants were grown for 1 week in a N-deficient nutrient solution. Six hours before harvest the plants were transferred into a 7.5-mM NO3–-containing nutrient solution. The bright band seen in B is due to artificial folding of the tissue. (I) In situ hybridization of antisense (A and C) and sense (B and D) LeNRT1;1 probe to longitudinal sections of tomato root tissues close to the root tip (A and B) and to regions close to the side root formation zone (C and D). (II) In situ hybridization of antisense (E and G) and sense (F and H) LeNRT2;3 probe to longitudinal sections of tomato root tissues close to the root tip (E and F) and to regions close to the side root formation zone (G and H). 130 Physiol. Plant. 115, 2002 Table 1. Quantification of the signals of the in situ hybridization in root cortical and rhizodermal cells with the probes for LeNRT1;1 (low affinity transporter) and LeNRT2;3 (high affinity transporter). For each value, the number of silver grains was counted in an area of 100 mm2 (n Ω 10, standard deviations given). For 12-week-old plants, counts for both NRT1;1 and NRT2;3 were done in the rhizodermis and inner cortex of the differentiating zone close to the root tip and at the height of the root hair zone in non-mycorrhizal plants. For 14-weekold plants, silver grain depositions due to NRT2;3 were determined at the height of the root hair zone in both mycorrhizal and nonmycorrhizal plants. Antisense probe Rhizodermis Sense probe Cortex Background Rhizodermis Cortex Background 12-week-old plants NRT 1;1 Differentiating zone Root hair zone 75 ∫ 26 10 ∫ 3 8 ∫4 50 ∫ 14 8 ∫4 3 ∫2 15 ∫ 6 8 ∫5 15 ∫ 11 10 ∫ 5 2 ∫1 3 ∫2 NRT2;3 Differentiating zone Root hair zone 43 ∫ 18 49 ∫ 5 9 ∫5 7 ∫4 2 ∫2 4 ∫1 5 ∫4 4 ∫2 7 ∫6 8 ∫2 2 ∫1 4 ∫2 63 ∫ 36 10 ∫ 3 35 ∫ 14 30 ∫ 6 32 ∫ 14 18 ∫ 4 23 ∫ 9 16 ∫ 3 11 ∫ 2 20 ∫ 4 14-week-old plants, root hair zone NRT2;3 Mycorrhizal 120 ∫ 25 Nonmycorrhizal 111 ∫ 32 Fig. 4. Comparison of the mRNA transcript distribution of LeNRT2;3 in mycorrhizal and non-mycorrhizal tomato. (I) In situ hybridization of antisense (A and C) and sense (B and D) probe of LeNRT2;3 to longitudinal sections of mycorrhizal (A and B) and non-mycorrhizal root tissues (C and D) at the root hair zone. (II) In situ hybridization of LeNRT2;3 antisense (A) and sense (B) probe to longitudinal sections of non-mycorrhizal tissue at the root tip. Physiol. Plant. 115, 2002 131 only background signals were obtained in roots of control plants (Fig. 4B). In AMF colonized roots, fungal structures (arbuscules, vesicles) were also seen in this area but without any specific enrichment of the silver grain distribution in their vicinity. The signal density was also high in the outer cortical and in the rhizodermal cells. These statements are corroborated by the statistical evaluation of the computer counts (Table 1B). A 3to 4-fold higher intensity of silver grain deposition was detectable in the cortical cells of mycorrhizal roots than in those of the control plants. In situ hybridization experiments performed with sections of aerial parts (leaves, stems, flowers, young fruits) of AMF colonized and control plants did not provide specific silver grain depositions for both NRT2;3 and NRT2;1 (not documented). Northern analyses were an alternative to reveal any difference in the mRNA formations of nitrate transporters in AMF and control plants. Hybridization signal intensities obtained with isolated RNA and the antisense probes for NRT1;1 and NRT2;2 were essentially the same in both root materials (Fig. 5). However, distinct differences were found both in the visible appearance and in the densitometrical determination in the case of NRT2;3 (Fig. 5). This difference in the Northern experiments was particularly impressive when three different RNA concen- trations were blotted (Fig. 6). The signal intensity almost linearly increased with the amount of RNA blotted. More noteworthy, NRT2;3 transcript amounts in AMF roots were almost 2-fold higher than in non-colonized controls (Fig. 6). The expression of nitrate and nitrite reductase genes in AMF colonized tomato RNA samples from 14-week-old mycorrhizal tomato roots (⬎ 80% mycorrhizal colonization) and controls (⬍ 2% colonization) were hybridized with the digoxigeninlabelled probes for either nitrate or nitrite reductases in Northern blot experiments. Statistically significant differences were not detected for mRNA formation in roots or shoots in three different experiments analysing in total 12 mycorrhizal and 12 control plants (Hildebrandt 2001). In contrast, significant differences were seen in the Northern blots with the Glomus riboprobe for nitrate reductase (Fig. 7). The signal intensity with RNA from AMF colonized roots was approximately five times higher than that in the controls. The finding that low intensity signals were detected in the hybridizations of RNA from non-mycorrhizal plants with the Glomus probe probably reflects the lack of absolute specificity of the probe. When the Glomus NIA probe was hybridized with the tomato NIA probe in control experiments, a low level of cross-hybridization Fig. 5. Northern blot analysis of nitrate transporter genes (LeNRT1;1, LeNRT2;2 and LeNRT2;3) in mycorrhizal and nonmycorrhizal tomato roots. Total RNA was isolated from 14-week-old tomato plants and hybridized with the digoxigeninlabelled probes of the nitrate transporters (Fig. 5A). Subsequently, the filters were stripped and hybridized to a 18S-rRNA riboprobe for control of equal loading. Signal intensities were quantified densitometrically. In Fig. 5B the signal intensities obtained with RNA from nonmycorrhizal roots were arbitrarily set to 10 in each case, which allows a direct comparison with the values obtained for mycorrhizal samples. NM Ω nonmycorrhizal sample, M Ω mycorrhizal samples, Wi Ω non-mycorrhizal plants incubated with 7.5 mM nitrate for 6 h prior to the isolation of the mRNA from the roots. 132 Physiol. Plant. 115, 2002 Fig. 6. Northern blot analysis of the specific nitrate transporter, LeNRT2;3, in dependence of the amount of RNA blotted. (A) amount of RNA loaded, (B) Signal intensity obtained by hybridizing with digoxigenin labelled LeNRT2;3 probe, (C) Densitometric quantification of the signal. was detectable (⬍ 5% signal intensity compared to that in homologous hybridization). The formation of nitrite from nitrate catalysed by nitrate reductase was determined in crude extracts from roots and leaves (Table 2). The activity was both NADPHand NADH-dependent with no large differences in the effectiveness of both donors. The slightly higher activity in AMF plants may be due to the enzyme contribution by the fungal partner. Table 2 also indicates that the enzyme was measurable in extraradical hyphae, which had been grown separated from the roots in the compartment system (George et al. 1992). The RFLP-analysis of DNA from spores isolated from the root-free compartment as described in Hildebrandt et al. 2001, verified that the tomato plants had indeed been grown with Glomus intraradices Sy167 and not with any cross-contaminant. Discussion The present investigation confirmed that the nitrate transporter composition of plants can be complex (Crawford and Glass 1998, Forde 2000). For tomato, the two described low affinity transporters NRT1;1 and NRT1;2 (Lauter et al. 1996) and one (NRT2 : 2) of the two closely related high affinity transporters (Ono et al. 2000) were also detected in the present, PCR-based assay. The gene LeNTR2;3 was not yet described and likely encodes a high affinity transporter. It possesses the structure of the Major Facilitator Superfamily (MFS) and also the typical signature motifs of the high affinity Fig. 7. Northern blot analysis of Glomus intraradices Sy167 nitrate reductase mRNA levels in mycorrhizal and non-mycorrhizal tomato Fourteen week old plants were used and a pool of 20 mg total root RNA from 3 plants laneª1 were blotted. After stripping, the filters were hybridized with a probe for 18S-rRNA as a control for equal loading. The data below the figure represent mean values and standard deviations from the two different lanes. Physiol. Plant. 115, 2002 nitrate transporters within this superfamily (Trueman et al. 1996). In the dendrogram (Fig. 8), it clusters in between the characterized high affinity nitrate transporters of Glycine max (Amarasinghe et al. 1998) and Nicotiana plumbaginifolia (Krapp et al. 1998). As stated for all nitrate transporters (Forde 2000), the tree of sequence relationship does not give hints on the substrate specificity of an uncharacterized member. The expression of NRT2;3 is controlled by nitrate. This is a special feature of nitrate transporters since others, like the sulphate or phosphate transporters, are not induced by their substrates but are only de-repressed by their deficiency (Forde 2000). The expression of LeNRT2;3 is negatively controlled by ammonium but, remarkably, not by glutamine. All other nitrate transporters from any plant described so far are controlled by the availability of glutamine. The detection of a new high affinity transporter, NRT2;3, is not unexpected, since about 70% of the PCR subclones obtained from a root cDNA library of mature tomato could be identified as LeNRT2;2, whereas LeNRT2;1 appeared to be almost exclusively confined to root hairs (Ono et al. 2000). The in situ hybridization experiments in roots often suffer from a large background level of the silver grain depositions. In the present study, depositions were too massive close to the root tip (calyptra) as to detect specific differences there. In slightly older root tissues, the data obtained in general matched with results of in situ hybridization experiments published for other plants so far. The NRT1;1 probe of tomato is closely related to Arabidopsis NRT1;1 (Ω CHL1, Huang et al. 1996) which was found to be expressed in the rhizodermal cells in young root tissue, and only in the cortex and endodermis more distant from the tip (Huang et al. 1996). In line with this, LeNRT1;1 transcripts were also detected in the cortical cells in more differentiated root tissue in the present study. Rhizodermal cells, however, did not show a specific enrichment in older tissue in the present study. In contrast, NRT1;2 of Arabidopsis (Huang et al. 1999) and its homologue from rice OsNRT1 (Lin et al. 2000) remained confined to the rhizodermis. With respect to the high affinity transporters, in situ hybridization experiments have only been performed with NRT2;1 from Nicotiana plumbaginifolia (Krapp et al. 1998) which resides in all layers of the root tip and is 133 Table 2. Specific activities of nitrate reductase in AMF colonized and control plants of tomato. The data are given in mmol gª1 FW hª1. Means ∫ are given for 16 independent measurements for the plant material. For the experiments with the extraradical hyphae was n Ω 6. Source of plant and fungal material Specific activity NADH-dependent Leaves, mycorrhizal plants Leaves, control plants Roots, mycorrhizal plants Roots, control plants Extraradical hyphae of Glomus intraradices Sy167 1.95 1.68 0.49 0.32 1.13 restricted to the lateral root primordia and rhizodermis in older roots which was also seen for the new tomato NRT2;3 in the present study. Most interestingly, the expression of LeNRT2;3 extended to the inner cortical cells when the roots were colonized by AMF structures. A similar pattern was described for a hexose transporter in AMF colonized and control plants of Medicago truncatula (Harrison 1996). For a phosphate transporter (LePT1), a specific enrichment was detected even in the arbuscule containing cells (Rosewarne et al. 1999). The ∫ 0.11 ∫ 0.09 ∫ 0.10 ∫ 0.11 ∫ 0.16 Specific activity NADPH-dependent 1.89 1.47 0.33 0.23 1.16 ∫ 0.14 ∫ 0.13 ∫ 0.08 ∫ 0.08 ∫ 0.20 methods employed in the present study did not permit such a fine resolution within the tissue structures. However, they allowed to conclude that AMF colonization caused an increase in the transcripts of different translocators in the interior of the root parenchyma but never beyond the endodermis into the stele. The data from the situ hybridization experiments with LeNRT2;3 are in line with the results from Northern hybridizations which also show an enrichment of its transcripts in AMF tomato but not in control plants. Fig. 8. Phylogenetic tree of amino acid sequences of different high affinity nitrate transporters with the position of the new LeNRT2;3. The phylogram was constructed using 4.0b6 (Swofford 1999) applying the mean character distance matrix and the neighbour joining tree construction method. The bar beneath the tree represents distance between amino acid sequences indicating the number of substitution events. Accession numbers of the used sequences are given in brackets. 134 Physiol. Plant. 115, 2002 In contrast to the previous results obtained with AMF colonized maize (Kaldorf et al. 1998), plant nitrate reductase transcripts and activities were not significantly reduced by the AMF colonization in the tomato roots currently assayed. This may be a plant specific trait and/ or may also be variable with the plant/AMF growth state. Consistently with the previous data obtained with maize (Kaldorf et al. 1998), fungal nitrate reductase transcripts were synthesized in significant amounts in AMF colonized tomato roots. This might reflect the fact that the fungal partner takes over a major portion of nitrate reduction with the colonization of the roots. Such a statement can also be reconciled with the findings that the transcript levels of a plant specific nitrate transporter, LeNRT2;3, were higher in AM colonized roots than in controls. Assuming that enzyme synthesis parallels transcript formation, the results may reflect that the total nitrate acquisition is drastically improved by the AMF colonization under the experimental conditions employed. Both symbiotic partners may benefit from the improved nitrate reduction. The observation that NRT2;3 transcripts are not repressed by glutamine may fit in this context. We venture to speculate from all these findings that the fungal partner transfers the excess of nitrate and glutamine not needed for its own growth to the plant partner. In the present investigation, the expression of the fungal counterparts of the nitrate transporters could not be studied. Despite of many attempts, the fungal transporters were not found by either heterologous probing or by PCR amplifications of the fungal DNA using degenerate oligonucleotide primers. PCR amplifications were successful with fungal nitrate reductase (Kaldorf et al. 1998) which was, however, the only case. The sequence comparisons (Kaldorf et al. 1994) showed that the Glomus nitrate reductase gene has fewer similarities to the enzyme from other fungi (Neurospora, Aspergillus) than to that of higher plants. Other AMF enzymes are seemingly evolutionarily more distant from those of other fungi which may be the reason that only few AMF genes (⬍ 100) have been identified and sequenced until now. Differential display techniques could be a tedious avenue to identify AMF nitrate transporters. Alternatively, genomic sequencing upstream and downstream of the NIA gene could be rewarding, since the genes coding for nitrate uptake and assimilation cluster in several organisms including fungi (Marzluf 1997, Jargeat et al. 2000). Acknowledgements – This work was kindly supported by grants from the Deutsche Forschungsgemeinschaft. References Amarasinghe BHRR, de Bruxelles GL, Braddon M, Onyeocha I, Forde BD, Udvardi MK (1998) Regulation of GmNRT2 expression and nitrate transport activity in roots of soybean (Glycine max). Planta 206: 44–52 Bago B, Vierheilig H, Piché Y, Aczon-Aguilar C (1996) Nitrate dePhysiol. Plant. 115, 2002 pletion and pH-changes induced by extraradical mycelium of the arbuscular mycorrhizal fungus Glomus intraradices grown in monoxenic culture. New Phytol 133: 273–280 Burleigh SH (2001) Relative quantitative RT-PCR to study the expression of plant nutrient transporters in arbuscular mycorrhizas. Plant Sci 160: 899–904 Crawford NM, Glass ADM (1998) Molecular and physiological aspects of nitrate uptake in plants. Trends Plant Sci 3: 389–395 Filleur S, Daniel-Vedele F (1999) Expression analysis of a highaffinity nitrate transporter isolated from Arabidopsis thaliana by differential display. Planta 207: 461–469 Forde BG (2000) Nitrate transporters in plants: structure, function and regulation. Biochim Biophys Acta 1465: 219–235 Frey B, Schuepp H (1993) Acquisition of nitrogen by external hyphae of arbuscular mycorrhizal fungi associated with Zea mays L. New Phytol 124: 221–250 Frohman MA, Dush MK, Martin GR (1988) Rapid production of full-length cDNA’s from rare transcripts: amplification using a single gene-specific oligonucleotide primer. Proc Natl Acad Sci USA 85: 8998–9002 George E, Häusler K, Vetterlein D, Gorgus E, Marschner H (1992) Water and nutrient translocation by hyphae of Glomus mosseae. Can J Bot 70: 2130–2137 Harrison MJ (1996) A sugar transporter from Medicago truncatula: altered expression pattern in roots during vesicular arbuscular (VA) mycorrhizal associations. Plant J 9: 491–503 Harrison MJ, van Buuren M (1995) A phosphate transporter from the mycorrhizal fungus Glomus versiforme. Nature 378: 626–629 Hawkins H-J, George E (1999) Effect of plant nitrogen status on the contribution of arbuscular mycorrhizal hyphae to plant nitrogen uptake. Physiol Plant 105: 694–700 Hildebrandt U (2001) Untersuchungen zur molekularbiologischen Charakterisierung der Nitratassimilation in der Symbiose zwischen dem arbuskulären Mykorrhiza-Pilz Glomus und Tomate. PhD thesis The University of Cologne, http://www.ub.unikoeln.de/ediss/archiv/11v4094.pdf Hildebrandt U, Janetta K, Ouziad F, Renne B, Nawrath K, Bothe H (2001) Arbuscular mycorrhizal colonization of halophytes in Central European salt marshes. Mycorrhiza 10: 175–183 Huang N-C, Chiang C-S, Crawford NM, Tsay Y-F (1996) CHL 1 encodes a component of the low-affinity nitrate uptake system in Arabidopsis and shows cell type-specific expression in roots. Plant Cell 8: 2183–2191 Huang N-C, Liu K-H, Lo H-J, Tsay Y-F (1999) Cloning and functional characterization of an Arabidopsis nitrate transporter gene that encodes a constitutive component of low affinity uptake. Plant Cell 11: 1381–1392 Jargeat P, Gay G, Debaud JC, Marmeisse R (2000) Transcription of a nitrate redcutase gene isolated from the symbiotic basidiomycete fungus Hebeloma cylindrosporum does not require induction by nitrate. Mol Gen Genet 263: 948–956 Johansen A, Jakobsen I, Jensen ES (1992) Hyphal transport of 15Nlabelled nitrogen by a vesicular-arbuscular mycorrhizal fungus and its effect on depletion of inorganic soil N. New Phytol 122: 281–288 Kaldorf M, Schmelzer E, Bothe H (1998) Expression of maize and fungal nitrate reductase genes in arbuscular mycorrhiza. Mol Plant Microbe Interact 11: 439–448 Kaldorf M, Zimmer W, Bothe H (1994) Genetic evidence for the occurrence of assimilatory nitrate reductase in arbuscular mycorrhizal and other fungi. Mycorrhiza 5: 23–28 Kothari SK, Marschner H, Römheld V (1991) Contribution of the VA mycorrhizal hyphae in acquisition of phosphorus and zinc by maize grown in a calcareous soil. Plant Soil 131: 177–185 Krapp A, Fraisier V, Scheible W-R, Quesada A, Gojon A, Stitt M, Caboche M, Daniel-Vedele F (1998) Expression studies of Nrt2:1Np, a putative high-affinity nitrate transporter: evidence for its role in nitrate uptake. Plant J 14: 723–731 Lauter F-R, Ninnemann O, Bucher M, Riesmeier JW, Frommer WB (1996) Preferential expression of an ammonium transporter and of two putative nitrate transporters in root hairs of tomato. Proc Natl Acad Sci USA 93: 8139–8144 Lin C-M, Koh S, Stacey G, Yu S-M, Lin T-Y, Tsay Y-F (2000) Cloning and functional characterization of a constitutively expressed nitrate transporter gene OsNRT1 from rice. Plant Physiol 122: 379–388 Liu H, Trieu AT, Blaylock LA, Harrison MJ (1998) Cloning and 135 characterization of two phosphate transporters from Medicago truncatula roots: regulation in response to phosphate and to colonization by arbuscular mycorrhizal (AM) fungi. Mol Plant Microbe Interact 11: 14–22 Lohaus G, Büker M, Hussmann M, Heldt H-W (1998) Transport of aminoacids with special emphasis on the synthesis and transport of asparagine in Illinois Low Protein and Illinois High Protein strains of maize. Planta 205: 181–188 Marzluf GA (1997) Genetic regulation of nitrogen metabolism in the fungi. Microbiol Mol Biol Rev 61: 17–32 Nagy F, Kay S, Chua N-H (1988) Analysis of gene expression in transgenic plants. In: Gelvin SB, Schilperoot RA (eds) Plant Molecular Biology Manual. Kluwer, Dordrecht, Vol 4, pp 1–29 Ono F, Frommer WB, von Wirén N (2000) Coordinated diurnal regulation of low-and high-affinity nitrate transporters in tomato. Plant Biol 2: 17–23 Pearson JN, Jakobsen I (1993) Symbiotic exchange of carbon and phosphorus between cucumber and three mycorrhizal fungi. New Phytol 124: 481–488 Raeder U, Broda P (1985) Rapid preparation of DNA from filamentous fungi. Lett Appl Microbiol 1: 17–20 Rosewarne GM, Barker SJ, Smith SE, Smith FA, Schachtman DP (1999) A Lycopersicon esculentum phosphate transporter (LEPT1) involved in phosphorus uptake from a vesicular-arbuscular mycorrhizal fungus. New Phytol 144: 507–516 Schmelzer E, Jahnen W, Hahlbrock K (1988) In situ localization of light-induced chalcone synthase mRNA, chalcone synthase, and flavonoid end products in epidermal cells of parsley leaves. Proc Natl Acad Sci USA 85: 2989–2993 Schmelzer E, Krüger-Lebus S, Hahlbrock K (1989) Temporal and spatial patterns of gene expression around sites of attempted fungal infection in parsley leaves. Plant Cell 1: 993–1001 Schmitz O, Danneberg G, Hundeshagen B, Klingner A, Bothe H (1991) Quantification of vesicular-arbuscular mycorrhiza by biochemical parameters. J Plant Physiol 139: 106–114 Smith SE, Read DJ (1997) Mycorrhizal Symbiosis 2nd edn. Academic Press, San Diego, CA Subramanian KS, Charest C (1998) Arbuscular mycorrhizae and nitrogen assimilation in maize after drought and recovery. Physiol Plant 102: 285–296 Swofford DL (1999) PAUP*. Phylogenetic analysis using parsimony (* and other methods), Version 4. Sinauer Associates, Sunderland, MA Tobar R, Azcon R, Barea JM (1994) Improved nitrogen uptake and transport from 15N- labelled nitrate by external hyphae of arbuscular mycorrhiza under water-stressed conditions. New Phytol 126: 119–122 Trueman LJ, Richardson A, Forde BG (1996) Molecular cloning of higher plant homologues of the high-affinity nitrate transporters of Chlamydomonas reinhardtii and Aspergillus nidulans. Gene 175: 223–231 Ullrich W, Nowacky A (1981) Nitrate-dependent membrane potential changes and their induction in Lemna gibba G1. Plant Sci Lett 22: 211–217 Vidmar JJ, Zhuo D, Siddiqi MY, Glass ADM (2000) Isolation and characterization of HvNRT2.3 and HvNRT2.4 cDNAs encoding high-affinity nitrate transporters from roots of barley. Plant Physiol 122: 783–792 Zhuo D, Okamoto M, Vidmar JJ, Glass ADM (1999) Regulation of a putative high-affinity nitrate transporter (Nrt2;1At) in roots of Arabidopsis thaliana. Plant J 17: 563–568 Edited by J. K. Schjørring 136 Physiol. Plant. 115, 2002