Allelic differences in Medicago truncatula NIP/LATD mutants correlate with their encoded proteins' transport activities in planta.

Medicago truncatula NIP/LATD gene, required for symbiotic nitrogen fixing nodule and root architecture development, encodes a member of the NRT1(PTR) family that demonstrates high-affinity nitrate transport in Xenopus laevis oocytes. Of three Mtnip/latd mutant proteins, one retains high-affinity nitrate transport in oocytes, while the other two are nitrate-transport defective. To further examine the mutant proteins' transport properties, the missense Mtnip/latd alleles were expressed in Arabidopsis thaliana chl1-5, resistant to the herbicide chlorate because of a deletion spanning the nitrate transporter AtNRT1.1(CHL1) gene. Mtnip-3 expression restored chlorate sensitivity in the Atchl1-5 mutant, similar to wild type MtNIP/LATD, while Mtnip-1 expression did not. The high-affinity nitrate transporter AtNRT2.1 gene was expressed in Mtnip-1 mutant roots; it did not complement, which could be caused by several factors. Together, these findings support the hypothesis that MtNIP/LATD may have another biochemical activity.

Most legumes are able to thrive in nitrogen (N) depleted soils because they can assimilate N via N fixing root nodules, symbiotic organs formed in partnership with rhizobia. Nodulation begins with an exchange of signals between the symbiotic partners, followed by plant cell divisions in the root cortex and pericycle and subsequent rhizobial invasion of the root through plant-derived infection threads. The infection threads in newly divided cells are encased by thin cell walls that can breach, forming an infection droplet. At the site of droplet formation, the plasma membrane separating the bacteria from the plant cytoplasm continues to proliferate, and individual rhizobia are endocytosed into the plant host cell cytoplasm, forming symbiosomes. Within effective nodules, bacteria establish a long-term infection within plant tissues, and additional differentiation of both plant and bacterial components occurs before N fixation commences. For more details, consult recent reviews.-

In Medicago truncatula, the NIP/LATD gene is essential for establishment of an effective N fixing symbiosis.- However, MtNIP/LATD is not required for the initial stages of rhizobial invasion into host roots, suggesting a role in a plant checkpoint that occurs between bacterial invasion and the establishment of an intracellular infection. Three M. truncatula mutants defective in NIP/LATD have been studied. The Mtnip-3 mutant has mild defects in lateral root (LR) elongation and develops symbiotic nodules with significantly reduced N fixation, whereas the Mtnip-1 mutant exhibits severe LR defects and defective nodules with rhizobia in infection threads but only rare rhizobial release into symbiosomes., The Mtlatd mutant has the most severe phenotype of the three alleles, with stunted root architecture and nodules similar to those found in Mtnip-1.

MtNIP/LATD encodes a protein in the NRT1(PTR) family, suggesting that MtNIP/LATD functions in small molecule transport. Members of the NRT1(PTR) family transport di- and tri-peptides, hormones, glucosinolates and other compounds.- Many NRT1(PTR) transporters are low affinity nitrate transporters. High affinity nitrate transporters are mostly found in the evolutionarily distinct NRT2 transporter family, although two dual affinity nitrate transporters in the NRT1(PTR) family have been described, Arabidopsis thaliana NRT1.1(CHL1) and M. truncatula NRT1.3. MtNIP/LATD protein was found to transport nitrate with a Km of 160 μM in Xenopus laevis oocytes, indicating that it is a high-affinity nitrate transporter.Mtnip-1 and Mtnip-3 have missense mutations, causing amino acid sequence changes A497V and E171K respectively, while Mtlatd has a nonsense mutation, W341Stop, in the MtNIP/LATD gene. Mtnip-1 and Mtlatd proteins were found not to transport nitrate in oocytes, but Mtnip-3 transported nitrate indistinguishably from wild type. This suggests that Mtnip-3 may be defective in transport of another compound, or could be defective in a different activity that is responsible for the phenotypes observed in the Mtnip-3 mutant., It is also possible that Mtnip-3 is capable of transporting nitrate in oocytes, but for some reason is not able to do so in planta. Here we further examine transport properties of the mutant proteins in planta. We also examine whether the phenotypes of a Mtnip/latd mutant can be rescued by expression of a high-affinity nitrate transporter.

To examine transport properties of the mutant MtNIP/LATD alleles in planta, we expressed the alleles separately in the A. thaliana chl1–5 mutant, with a deletion spanning the AtNRT1.1(CHL1) gene, encoding a major dual affinity nitrate transporter. The Atchl1–5 mutant was originally isolated on the basis of its resistance to chlorate, an herbicide and nitrate analog that is transported through the AtNRT1.1(CHL) nitrate transporter into roots where it is converted to phytotoxic chlorite. Therefore, the expression of functional nitrate transporters in Atchl1–5 results in the loss of chlorate resistance and a reduction of plant vigor after chlorate treatment. The two mutant alleles encoding missense mutations in their MtNIP/LATD were investigated in Atchl1–5 for their ability to restore sensitivity, and plant vigor was determined by documenting overall plant size, mass and chlorophyll content. Two Atchl1–5 lines independently transformed with a constitutively expressed Mtnip-1 and three independent Atchl1–5 lines expressing Mtnip-3 cDNA were selected for further analysis, based on similar, robust Mtnip-1 or Mtnip-3 mRNA expression compared with wild-type MtNIP/LATD expression in control Atchl1–5 lines expressing wild-type MtNIP/LATD.

Atchl1–5 plants expressing Mtnip-1 or Mtnip-3 and controls were treated with chlorate, as described previously.,Mtnip-3 cDNA expression in Atchl1–5 restored chlorate sensitivity (representative plants are in Fig. 1, far right; shown in color in ) and resulted in a dramatic reduction in plant size, as did the expression of positive controls, AtNRT1.1 cDNA and MtNIP/LATD cDNA (Fig. 1; ). In contrast, Mtnip-1 expressing Atchl1–5 plants were found to be chlorate resistant and were indistinguishable from Atchl1–5 plants (Fig. 1; ). The vigor of Atchl1–5 plants expressing test genes in trans was determined by measuring fresh weight (Fig. 2A) and chlorophyll content (Fig. 2B) after treatment with chlorate. The fresh weights of the three lines of Atchl1–5 expressing Mtnip-3 cDNA were indistinguishable (lines 1 and 2) or slightly larger than (line 3) the weights of Atchl1–5 plants transformed with AtNRT1.1 or MtNIP/LATD expression constructs (Fig. 2A). In contrast, the masses of plants from the two independent Atchl1–5/Mtnip-1 lines were indistinguishable from those of negative control Atchl1–5 plants, indicating that they retained chlorate resistance (Fig. 2A). Chlorophyll content was also measured in extracts from chlorate-treated plants expressing the control or test genes. The chlorophyll content of Atchl1–5/Mtnip-3 lines 1 and 2 were similar to those of the Atchl1–5 lines expressing the positive control AtNRT1.1 or wild-type MtNIP/LATD and to wild type Col-0, while the third line, Atchl1–5 line 3, had a slightly higher chlorophyll content than the positive controls (Fig. 2B). Both Atchl1–5/Mtnip-1 plants had higher chlorophyll content than Atchl1–5 plants expressing AtNRT1.1, MtNIP/LATD or Mtnip-3, but slightly lower than Atchl1–5 control plants (Fig. 2B). Collectively, these data indicate that MtNIP/LATD and Mtnip-3 proteins transport chlorate in planta (Figs. 1F and2A), although transport by Mtnip-3 protein may be slightly less efficient than wild-type MtNIP/LATD (Fig. 2B). Similarly, the data indicate that Mtnip-1 protein is non-functional (Figs. 1E and 2A) or retains a very small fraction of transport activity (Fig. 2B).

Figure 1.Mtnip-3, but not Mtnip-1, complements the chlorate-insensitivity phenotype of the Arabidopsis chl1–5 mutant. Arabidopsis control and test plants were treated with chlorate, a nitrate analog that can be converted to toxic chlorite after uptake., Two independent Atchl1–5/Mtnip-1 lines and three independent Atchl1–5/Mtnip-3 lines were tested; representative plants are shown. The genotype and transgene in each plant is indicated. Bar = 1 cm. The Mtnip-3 gene was able to confer chlorate sensitivity on Arabidopsis chl1–5 plants, similar to the MtNIP/LATD and AtNRT1.1 genes.

Figure 2. Fresh weight and chlorophyll content of chlorate treated Arabidopsis plants. The fresh weights and chlorophyll content of the Arabidopsis Col-0, Atchl1–5 and Atchl1–5 plants transformed with constructs and treated with chlorate, as in Figure 1, were measured. Asterisks mark values that are significantly different at the 1% level from the Atchl1–5 value, using the paired t-test. (A) The fresh weights of Atchl1–5 plants transformed with the Mtnip-1 gene were indistinguishable from Atchl1–5 plants, and the fresh weights of two of the Atchl1–5 lines transformed with Mtnip-3 were indistinguishable from those of Col-0 or Atchl1–5 transformed with either MtNIP/LATD and AtNRT1.1 genes. The third Atchl1–5/Mtnip-3 line has similar, but not identical, fresh weight to the other two. (B) The chlorophyll contents of Atchl1–5 plants transformed with the Mtnip-1 gene were similar to Atchl1–5 plants, and the chlorophyll contents of two of the Atchl1–5 lines transformed with Mtnip-3 were similar to those of Col-0 and to chlorophyll contents of Atchl1–5 transformed with either MtNIP/LATD and AtNRT1.1 genes. The third Atchl1–5/Mtnip-3 line had intermediate chlorophyll content.

Previously, we tested the expression of two NRT1(PTR) genes, the aforementioned AtNRT1.1(CHL1) gene and AgDCAT, encoding a dicarboxylate transporter, for their abilities to rescue Mtnip-1 phenotypes when expressed in Mtnip-1 roots. We found that expression of AtNRT1.1 in Mtnip-1’s roots partially rescued the root architecture phenotype, supporting the hypothesis that MtNIP/LATD in M. truncatula transports nitrate at high affinity. However, because AtNRT1.1 also transports auxin and acts as a nitrate sensor, it cannot be ruled out that one of these other activities is responsible for the phenotype of Mtnip-1 roots constitutively expressing AtNRT1.1. AgDCAT expression in Mtnip-1 was without effect on root or nodule phenotypes. Because MtNIP/LATD was shown to transport nitrate with high affinity in oocytes, here we tested the MtNIP/LATD promoter-driven expression of AtNRT2.1, encoding a high-affinity nitrate transporter, in Mtnip-1 roots to determine if it complemented the phenotypic defects. We found that AtNRT2.1 expression had no effect on the Mtnip-1 phenotypes, with neither root architecture nor nodulation defects altered (). While this might suggest that the important biological role of MtNIP/LATD is not in nitrate transport, non-complementation of Mtnip-1’s phenotype by AtNRT2.1 may be due to other factors. One possible factor is the requirement for proteins in the AtNRT2 family to interact with proteins in the NAR2.1, also called NRT3, family for functional high-affinity nitrate transport. Another is the apparent post-transcriptional control of AtNRT2.1. Thus, this negative result may not be informative.

In conclusion, previously we found that MtNIP/LATD and Mtnip-3 proteins transport nitrate in oocytes, while the Mtnip-1 and Mtlatd proteins did not. Here, we show that Mtnip-3 expression, like that of MtNIP/LATD, complements the chlorate resistance of the Arabidopsis chl1–5 mutant, while Mtnip-1 does not. This new data demonstrates that Mtnip-3 transports the nitrate analog chlorate in planta, thus, strongly suggesting that Mtnip-3 protein is proficient at nitrate transport in planta. This result is consistent with the previous observation that Mtnip-3 transports nitrate in oocytes. However, the Mtnip-3 mutant has root architecture and nodulation defects in M. truncatula., Together, these findings suggest that Mtnip-3 is defective in a second, so far unknown, biochemical activity found in MtNIP/LATD. Finally, we found that AtNRT2.1 does not complement any of the phenotypes found in Mtnip-1. However, complementation of Mtnip-1 by AtNRT2.1 may require co-expression of AtNAR2.1 22 or post-transcriptional modulation, and thus non-complementation by AtNRT2.1 alone may not be informative.

Materials and Methods

A. thaliana plant constructs and chlorate tests

Mtnip-1 and Mtnip-3 cDNA were amplified from oocyte expression vectors used previously using primers NIPC2F (TGAACCATGGAGTACACAAACAGTGATGATGCTAC) and NIPCODBst1R (AAAAAGGTCACCTATGAAGTAGGCAACTCCCTGT) and subsequently cloned into NcoI and BstEII site of pMS004 to create pYM001 and pYM002. Then BamHI/BstEII fragments from pYM001 and pYM002 were cloned into pCAMBIA2301 to create pYM003 (pCAMBIA2301-pAtEF1α-Mtnip-1) and pYM004 (pCAMBIA2301-AtEF1α-Mtnip-1), yielding binary vectors with the mutant alleles under the control of the constitutive AtEF1α promoter. pYM003 and pYM004 were transformed into Agrobacterium tumefaciens GV3101(pMP90) cells and then transformed into Atchl1–5 mutant by the floral dip method. Homozygous plants for each line were selected on half strength MS medium (Research Products International, Prospect IL) containing 25 mg/L kanamycin. RNA was extracted from each line (Qiagen, Valencia CA, RNeasy kit), contaminating DNA removed (Life Technologies, Grand Island, NY, Ambion TURBO DNA-free kit; then Qiagen RNeasy MinElute kit) and cDNA synthesized (Life Technologies, Invitrogen SuperScript First-Strand Synthesis system). The cDNA was subjected to semi-quantitative RT-PCR for levels of MtNIP/LATD or mutant allele mRNA expression using primers NipC2F (TGAACATGGAGTACACAAACAGTGATGATGCTAC) and NipCSeq2R (TTCTTGGTTGCCACCACAAC). Independent lines having similar expression levels were selected for further analysis (data not shown). Chlorate sensitivity tests were performed as described.,Mtlatd, because it has a nonsense mutation in the middle of the MtNIP/LATD coding region, was not tested for its ability to complement the Atchl1–5 mutant.

AtNRT2.1 expression in M. truncatula

A 1.7 Kbp AtNRT2.1 cDNA was amplified from an AtNRT2.1 cDNA clone obtained from Dr. Anthony Glass using primers NRT2.1BamHIF (CTAGGGATCCATGGGTGATTCTACTGGTGAGCCG) and NRT2.1BstEIIR (ATGAGGTAACCTCAAACATTGTTGGGTGTGTTCTCAGGCGG). Subsequently, the 1.7 Kbp AtNRT2.1 cDNA was cloned into BglII and BstEII sites of pMS014, a vector containing the 3 kb MtNIP/LATD promoter with an EcoRI site at its 5′ end and a BglII site at its 3′ end, to create pMS016. Subsequently the EcoRI/BstEII fragment was cloned into pCAMBIA2301 creating pMS020 (pCAMBIA2301-pMtNIP/LATD-AtNRT2.1). pMS020 was transformed into Agrobacterium rhizogenes MSU440 by the freeze/thaw method. Hairy root transformation of M. truncatula and nodulation studies were performed as described.