Polyamine Resistance Is Increased by Mutations in a Nitrate Transporter Gene NRT1.3 (AtNPF6.4) in Arabidopsis thaliana

Polyamines are small basic compounds present in all living organisms and act in a variety of biological processes. However, the mechanism of polyamine sensing, signaling and response in relation to other metabolic pathways remains to be fully addressed in plant cells. As one approach, we isolated Arabidopsis mutants that show increased resistance to spermine in terms of chlorosis. We show here that two of the mutants have a point mutation in a nitrate transporter gene of the NRT1/PTR family (NPF), NRT1.3 (AtNPF6.4). These mutants also exhibit increased resistance to putrescine and spermidine while loss-of-function mutants of the two closest homologs of NRT1.3, root-specific NRT1.1 (AtNPF6.3) and petiole-specific NRT1.4 (AtNPF6.2), were shown to have a normal sensitivity to polyamines. When the GUS reporter gene was expressed under the control of the NRT1.3 promoter, GUS staining was observed in leaf mesophyll cells and stem cortex cells but not in the epidermis, suggesting that NRT1.3 specifically functions in parenchymal tissues. We further found that the aerial part of the mutant seedling has normal levels of polyamines but shows reduced uptake of norspermidine compared with the wild type. These results suggest that polyamine transport or metabolism is associated with nitrate transport in the parenchymal tissue of the shoot.


INTRODUCTION
Polyamines such as putrescine and spermidine are small basic compounds present in all living organisms and play a role in a variety of biological processes including DNA replication, protein synthesis, and ion channel modulation (Igarashi and Kashiwagi, 2010b;Pegg and Casero, 2011). The cellular content of polyamines is highly regulated by biosynthesis, oxidative degradation, transport, and, in plants, conjugation to hydroxycinnamic acids. Both in animals and plants, most of the mRNAs of S-adenosylmethionine decarboxylase (AdoMetDC), which is a key enzyme providing the aminopropyl group for the synthesis of spermidine and spermine, contain small upstream open reading frames (uORFs) that mediate polyamine-dependent repression of the main ORF translation (Raney et al., 2002;Hanfrey et al., 2005). Plants and some bacteria have a structural isomer of spermine, thermospermine (Takano et al., 2012). We have found that expression of ACAULIS5 (ACL5), which encodes thermospermine synthase in Arabidopsis, is under negative feedback control by thermospermine (Kakehi et al., 2008;Tong et al., 2014). In animals and fungi, degradation of ornithine decarboxylase (ODC), which catalyzes the conversion of ornithine into putrescine, a rate-limiting step in polyamine biosynthesis, is directed by ODC antizyme (OAZ) whose synthesis is induced by higher cellular polyamine levels (Matsufuji et al., 1995;Ivanov et al., 2000).
As for transport systems, antiporters of putrescine-ornithine and cadaverine-lysine as well as putrescine-specific and spermidine-preferential uptake carriers have been characterized in Escherichia coli (Kashiwagi and Igarashi, 2011). In yeast, GAP1 catalyzes the uptake of putrescine and spermidine together with the uptake of amino acids (Uemura et al., 2005). AGP2 acts as a high-affinity amino acid permease and selectively catalyzes the uptake of spermidine (Aouida et al., 2005). Three additional proteins, DUR3, SAM3, and tonoplast-localized UGA4 also function in polyamine uptake, while five excretion proteins, TPO1 to TPO5, have been identified (Igarashi and Kashiwagi, 2010a). In human, while multiple systems for uptake of polyamines have been identified, the organic cation transporter2 (OCT2) has been shown to function as a common uptake carrier (Higashi et al., 2014). In plants, earlier studies using petals of Saintpaulia ionantha revealed that putrescine uptake is dependent on the external pH (Bagni and Pistocchi, 1985). In carrot cell cultures, the uptake of putrescine and spermidine is remarkably rapid, reaching a maximum within only 1 min (Pistocchi et al., 1987;Antognoni et al., 1993). A genetic study of an Arabidopsis wild-type accession that is resistant to paraquat identified RMV1 as a polyamine uptake transporter (Fujita et al., 2012). In rice, OsPUT1 to OsPUT3 have been identified as spermidine-preferential transporters. AtPUT1 to AtPUT3 are the orthologous proteins in Arabidopsis and function as high affinity spermidine uptake transporters (Mulangi et al., 2012), among which AtPUT3 is identical to RMV1 (Fujita and Shinozaki, 2014).
While cellular polyamine levels are maintained by the abovedescribed regulatory systems, they may also be interrelated with other metabolic pathways. In poplar cell cultures, spermidine and spermine levels positively correlate with most amino acids (Mattoo et al., 2010). Increased polyamine biosynthesis may result in increased assimilation of both nitrogen and carbon by the cells (Majumdar et al., 2016). On the other hand, exogenous polyamines cause a rapid production of nitric oxide (NO) in Arabidopsis (Tun et al., 2006). A genetic screen of Arabidopsis mutants with increased tolerance to norspermidine identified a dominant mutant, par1-1D, with increased expression of QSO2 (Alejandro et al., 2007). QSO2 encodes a quiescin-sulfhydryl oxidase and may activate K + efflux systems involved in xylem loading in roots thereby reducing the accumulation of toxic cations such as Na + and polyamines. We are interested in how plant cells perceive and respond to external polyamines. To gain insight into sensing, signaling, and responsive mechanisms to polyamines, we isolated Arabidopsis mutants that show increased resistance to spermine. Among them, two mutants were found to be alleles of NRT1.3 encoding a member of the low-affinity nitrate transporter family.

Mutant Isolation, Mapping, and Next Generation Sequencing
For screening of mutants resistant to polyamines, 49,000 M2 seeds of ethyl methanesulfonate (EMS)-mutagenized Col-0 were sown on MS agar plates containing 3 mM spermine and 28 resistant plant lines were selected. Among them, three lines that reproducibly showed increased resistance to spermine in the progeny were named spermine resistant1 (sper1), sper2, and sper3, respectively.
For chromosomal mapping of the mutant loci, F2 seeds from the cross between the mutants and Ler were sown on MS agar plates containing 3 mM spermine and the DNA was extracted from each individual that was resistant to spermine. Totally ca. 80-100 seedlings were selected for each mutant. PCR-based mapping was performed using simple sequence length polymorphism (SSLP) markers (Bell and Ecker, 1994) and cleaved amplified polymorphic sequence (CAPS) markers (Konieczny and Ausubel, 1993). Additional markers were designed according to the TAIR database 1 and the primers used are shown in Supplementary Table S1.
Genome DNA sequences of the mutants were determined by next-generation sequencing with the SOLiD platform. The multiplex libraries were constructed using the SOLiD barcoding and sequenced on a single SOLiD slide (Tabata et al., 2013). The resulting sequence reads were aligned against the A. thaliana Col-0 genome reference TAIR9 (TAIR_chr_all.fas) using BioScope 1.3 software (Life Technologies) with default parameters. Single nucleotide polymorphisms (SNPs) were called using a diBayes SNP caller, a component of BioScope 1.3 and validated as described in Tabata et al. (2013).
The constructs were introduced into Agrobacterium tumefaciens C58C1 by electroporation. Transformation of Arabidopsis with the GUS construct was performed by the floral dip method (Harrison et al., 2006) using 0.05% Silwet L-77 in the dipping solution.

Measurement of Chlorophyll
Chlorophyll was extracted from 50 mg of the aerial part of 10day-old seedlings in 1 mL of N,N-dimethyl formamide at 4 • C overnight in the dark and assayed as described (Porra et al., 1989).  (B) Chlorophyll content of 10-day-old seedlings grown in MS agar plates with 2 mM spermine. Error bars represent the SE (n = 3). Asterisks indicate statistically significant difference between groups assessed by ANOVA with Tukey-Kramer HSD (P < 0.05).

GUS Staining
Samples were prefixed for 20 min in ice-cold 90% acetone (v/v), vacuum infiltrated for 20 min with the GUS staining solution containing 50 mM sodium phosphate buffer pH7.2, 2 mM potassium ferricyanide, 2 mM potassium ferrocyanide, 0.1% Triton X-100 (v/v) and 1 mM X-glucuronide, and incubated at 37 • C overnight (Jefferson et al., 1987). The stained samples were cleared with 70% ethanol and observed under a light microscope. For sectioning, GUS-stained tissues were dehydrated in a graded ethanol series and embedded in Technovit 7100 (Heraeus Kulzer, Wehrheim, Germany) according to the manufacturer's instructions. Sections were cut to 10 µm.

Transient Expression of GFP in Tobacco
Agrobacterium cells containing the NRT1.3-GFP construct were cultured in Luria broth (LB) with 50 µg/ml of kanamycin and hygromycin at 28 • C overnight. Agrobacterium cells containing the p19 plasmid as a suppressor of silencing (Silhavy et al., 2002) were cultured in LB with 50 µg/ml kanamycin. Both cultures were centrifuged, dissolved in 10 ml infiltration buffer (10 mM MES, 10 mM MgCl 2 , 100 µM acetosyringone), and mixed together. The suspension was adjusted to a final OD 600 of 0.8 for infiltration into leaves of 4-week-old Nicotiana benthamiana. At 2 days after infiltration, the GFP fluorescence signals were observed in the epidermal cells under confocal laser scanning microscopy (Eclipse C1, Nikon, Tokyo, Japan).

HPLC Analysis
Cellular polyamine levels were measured by HPLC. Acidsoluble free polyamines were extracted with perchloric acid from the aerial part of 10-day-old seedlings and benzoylated as described previously (Tong et al., 2014). The samples were analyzed using a reverse phase HPLC system equipped with TSKgel ODS-80Ts column (Toso, Tokyo, Japan).

sper3-1 and sper3-2 Are Resistant to Polyamines
We identified three mutants that showed increased resistance to 3 mM spermine in an M2 population derived from EMSmutagenized seeds of the accession Col-0 of A. thaliana and tentatively named sper1, sper2, and sper3. Because rough mapping data showed that sper1 is located on chromosome 4 while sper2 and sper3 are on chromosome 3 and, as shown below, sper3 and sper2 were found to be different alleles of the same gene, these alleles were designated sper3-1 and sper3-2, respectively. Under normal growth conditions, sper3-1 and sper3-2 have no obvious morphological abnormalities compared with the wild type ( Figure 1A). However, when grown on MS agar plates containing different concentrations of spermine, sper3-1 and sper3-2 seedlings retained green leaves at 3 mM spermine while wild-type seedlings showed chlorosis at 1 mM spermine ( Figure 1B). These mutant seedlings were also resistant to 25 mM putrescine and 5 mM spermidine while they were no more resistant to basic amino acids, lysine and arginine, than wild-type seedlings ( Figure 1C and not shown). F1 seedlings from a backcross of sper3-1 to its wild-type parent Col-0 showed intermediate sensitivity to spermine in terms of the leaf development (Figure 2A) and the chlorophyll content (Figure 2B), indicating that sper3-1 represents a semi-dominant allele. Similarly, the seedlings heterozygous for sper3-2 showed intermediate sensitivity to spermine (Figure 2B), indicating that sper3-2 also has a semidominant effect.
sper3-1 and sper3-2 Are Alleles of NRT1.3 Fine mapping using F2 plants from a cross between sper3-1 and the accession Ler delimited the locus to a 200-kb region on chromosome 3 ( Figure 3A). Next-generation sequencing data of this region of the sper3-1 genomic DNA revealed a G-to-A point mutation in NRT1.3 (AtNPF6.4; Figure 3B), which belongs to the NRT1/PTR low-affinity nitrate transporter family (Tsay et al., 2007;Supplementary Figure S1). The G-to-A base substitution in sper3-1 changes glutamate 46, which is conserved in the family, to lysine ( Figure 3C). On the other hand, sequencing of the sper3-2 genomic DNA revealed a C-to-T base substitution that changes alanine 433 of NRT1.3 (Figure 3B), which is also conserved in the family, to valine ( Figure 3D).
To confirm further that polyamine resistance is indeed caused by a mutation in NRT1.3, we examined whether a T-DNA insertion mutant of NRT1.3 shows increased resistance to spermine. The results revealed that the homozygous knockout allele designated sper3-3 ( Figure 3B) had no detectable levels of full-length NRT1.3 mRNA (Supplementary Figure S2) and clearly showed increased resistance to 3 mM spermine ( Figure 4A). Unlike sper3-1/+ and sper3-2/+, heterozygous sper3-3/+ seedlings showed no increased resistance to spermine as shown in the chlorophyll content (Figure 4B), indicating that sper3-3 is a recessive allele. FIGURE 4 | Growth response of sper3-3 and other knockout mutants of the AtNPF6 subfamily to spermine. (A) Phenotype of 7-day-old seedlings grown in MS agar plates with (+) or without (−) 2 mM spermine (Spm). chl1-5, nrt1.4-2, and nrt1.2-1 (ait1-1) represent loss-of-function mutants of AtNPF6.3, AtNPF6.2, and AtNPF4.6, respectively. (B) Chlorophyll content of 7-day-old seedlings grown in MS agar plates with 2 mM spermine. Error bars represent the SE (n = 3). Asterisks indicate statistically significant difference between the wild-type and mutants ( * * P < 0.01 Student's t-test). Figure S3A). When wild-type seedlings grown for 7 days on MS agar plates were incubated for 2 or 24 h in MS solutions with 100 µM spermine, no significant changes in the level of the NRT1.3 mRNA were detected (Supplementary Figure  S3B). Although NRT1.3 expression is induced by NO 3 − in shoots (Okamoto et al., 2003), its tissue expression pattern has not been investigated in previous studies. Thus, we generated transgenic plants carrying the NRT1.3 promoter-GUS fusion gene and examined the GUS expression in more than eight independent plant lines. GUS staining was detected in the hypocotyl, cotyledons, and leaves but not in the root of seedlings ( Figure 5A). In adult plants, inflorescence stems, pedicels, and sepals of flowers were strongly stained (Figures 5B,C). Tissue sections revealed that GUS expression was restricted to mesophyll cells of leaves and cortical cells of stems but not in epidermal cells (Figures 5D,E), suggesting that NRT1.3 specifically functions in parenchymal tissues.

NRT1.3 expression was detected by RT-PCR in all organs examined with the highest expression in flowers (Supplementary
To examine cellular localization of NRT1.3, an NRT1.3-GFP translational fusion gene was constructed and transiently expressed under the CaMV 35S promoter in tobacco leaves by agro-infiltration. The NRT1.3-GFP fluorescence was detected mainly along the plasma membrane ( Figure 5F).

Accumulation and Uptake of Polyamines in sper3-1
To examine whether or not sper3-1 affects polyamine transport or metabolism, cellular polyamine levels in the aerial part of sper3-1 seedlings were quantified by HPLC. The results revealed that the levels of all four polyamines, putrescine, spermidine, spermine, and thermospermine, were not affected by sper3-1 ( Figure 6A). We next investigated polyamine uptake activity in sper3-1 by using norspermidine as a tracer because it is not present in wild-type plants. Seedlings grown for 7 days on MS agar plates were placed on the plate containing 100 µM norspermidine and the aerial part was subjected to polyamine extraction and HPLC. The results showed that the uptake level of norspermidine was initially indistinguishable between the wild type and sper3-1 but significantly declined after 24 h incubation in sper3-1 ( Figure 6B).

DISCUSSION
In this study, we found that alleles of NRT1.3 encoding a member of the NRT1/PTR family of nitrate transport proteins are responsible for the mutants with increased resistance to polyamines. The NRT1/PTR family in Arabidopsis consists of 53 members and includes both nitrate and oligopeptide transporters. The two closest homologs of NRT1.3, NRT1.4, and NRT1.1 share 51 and 47% amino acid sequence identities with NRT1.3, respectively. NRT1.1 is a dual-affinity transporter involved in both low-and high-affinity nitrate uptake, depending on cellular nitrate concentrations (Wang et al., 1998;Liu et al., 1999), and also facilitates the uptake of auxin in the absence of nitrate (Krouk et al., 2010). In contrast, NRT1.4 has been shown to be a low-affinity nitrate transporter (Chiu et al., 2004). Some other members are known to transport small organic anions as well as nitrate. NRT1.2 transports nitrate and abscisic acid Kanno et al., 2012). NRT1.7, NRT1.9, and NRT1.10 (GTR2) play a role in the uptake of both nitrate and glucosinolates (Fan et al., 2009;Nour-Eldin et al., 2012). Although NRT1.3 remains to be characterized biochemically, its putative ortholog in Medicago truncatula, MtNRT1.3, has been shown to be a dual affinity transporter and also an ABA transporter (Morère-Le Paven et al., 2011;Pellizzaro et al., 2014). Given the cationic nature of polyamines, it is unlikely that NRT1.3 has the ability to transport polyamines, although its possibility cannot be excluded completely.
Nitrate is taken up from soil by the root tissue but its assimilation can take place in both the root and the shoot. NRT1.1 and NRT1.2 are involved in the uptake of nitrate from soil (Tsay et al., 1993;Wang et al., 1998;Huang et al., 1999;Liu et al., 1999). While NRT1.5 is expressed in root pericycle cells close to xylem and acts in loading nitrate into xylem vessels for rootto-shoot transport Chen et al., 2012), NRT1.8, which is expressed in root xylem parenchyma cells, and NRT1.9, which is expressed in companion cells of the root phloem, are also involved in regulating root-to-shoot nitrate translocation (Li et al., 2010;Wang and Tsay, 2011). NRT1.7 is expressed in the phloem of the leaf vein, indicating that nitrate can be remobilized from older leaves to younger leaves (Fan et al., 2009). NRT1.4 is expressed in the leaf petiole and may be responsible for nitrate homeostasis in leaves (Chiu et al., 2004). NRT1.6 has been shown to function in delivering nitrate to the developing embryo in the seed (Almagro et al., 2008). We confirmed here that NRT1.3 is highly expressed in flowers, stems, and leaves. Our results of GUS expression ( Figure 5) suggest that, complementarily to other members of the NRT1/PTR family, NRT1.3 may be expressed in mesophyll cells of the leaf and the cortex of the stem. However, because all sper3 alleles show wild-type phenotype under normal growth condition (Figure 1A), other nitrate transporters may function redundantly in these parenchymal tissues. Interestingly, the NRT1.3 promoter-GUS expression was not detected in epidermal cells. This might be related to the lack of chloroplasts in these cells. In shoot organs, nitrate assimilation is generally coupled to photosynthetic electron flow because the process of nitrite reduction occurs in the chloroplast stroma (Foyer and Noctor, 2002). The induction of NRT1.3 expression by NO 3 − is detected only in shoots (Okamoto et al., 2003). Furthermore, NRT1.3 expression is up-regulated by light while NRT1.1 and NRT1.5 by sugar (Lejay et al., 2008). Taken together, as a member of the NRT1/PTR family, NRT1.3 may play a role in the final step of nitrate supply to photosynthesizing cells. The strong fluorescence of NRT1.3-GFP at the cell periphery FIGURE 6 | Polyamine content of wild-type and sper3-1 seedlings. (A) Level of putrescine (Put), spermidine (Spd), spermine (Spm), and thermospermine (Tspm) in grown in MS agar plates. Polyamines were extracted and analyzed by HPLC as described in section "Materials and Methods." (B) Level of norspermidine (Nspd) in above-ground parts of seedlings. Wild-type and sper3-1 seedlings were grown for 7 days in MS agar plates, transferred to plates containing 100 µM norspermidine, and incubated for indicated hours. Error bars represent the SE (n = 3). An asterisk indicates statistically significant difference between wild-type and sper3-1 ( * P < 0.05 Student's t-test).
( Figure 5F) suggests that NRT1.3 is localized to the plasma membrane. However, its localization to endomembranes might also be possible. NRT2.7, one of the seven members of the NRT2 family, has been shown to be localized to the vacuolar membrane in seeds (Chopin et al., 2007).
We detected a reduced uptake of norspermidine in sper3-1 ( Figure 6B). This raises a possibility that polyamine transport is associated or at least coincided with nitrate transport in parenchymal tissues and its defect may cause the reduced uptake of polyamines. sper3 mutants might have higher accumulation of nitrate in the root tissue and/or in the apoplast of shoot organs which could interfere with polyamine uptake. A recent study has revealed that mutants of a nitrate efflux channel, SLAH3, were more sensitive to ammonium, suggesting that the efflux of nitrate by SLAH3 alleviates ammonium toxicity in roots (Zheng et al., 2015). Although the sensitivity to polyamines in these mutants remains to be examined, SLAH3 might also serve to alleviate polyamine toxicity. On the other hand, considering that the initial uptake efficiency of norspermidine in sper3-1 was not different from that in the wild type (Figure 6B), it is also possible that the incorporated polyamines can be metabolized more rapidly under reduced availability of nitrate in shoot parenchymal tissues of sper3.
According to the crystal structure of NRT1.1, it forms a homodimer in the inward-facing conformation (Parker and Newstead, 2014;Sun et al., 2014). If this is also the case with NRT1.3, the semi-dominant trait of sper3-1 and sper3-2 might be due to a dominant negative effect by the presumed dimerization with the wild-type protein rather than a haplo-insufficiency. The NRT1/PTR family protein is also known as a proton-coupled transporter. In NRT1.1, H356 in the 7th transmembrane domain (TM7) is a binding site of nitrate, while the ExxER motif in TM1 plays a role in proton binding and co-transport (Steiner et al., 1995). This motif interacts with K164-E476 salt bridge involved in nitrate uptake (Doki et al., 2013;Parker and Newstead, 2014). The E46K amino acid substitution in sper3-1 occurs in the second E of the ExxER motif ( Figure 3C), confirming the importance of this amino acid. Based on the phenotype of sper3-1 and sper3-2, the A433V amino acid substitution in sper3-2 may affect the function of NRT1.3 in a similar way to sper3-1. These alleles will provide useful tools for further biochemical study of this nitrate transporter.

CONCLUSION
This study suggests that nitrate transport in the parenchymal tissue of the shoot is closely interrelated with polyamine uptake or metabolism. Although mutants of NRT1.1, NRT1.2, and NRT1.4 showed normal sensitivity to spermine, involvement of some additional members in this interrelation may be possible, given the presence of multiple members of the NRT1/PTR family with various expression patterns. As a future study, it would be interesting to generate double and multiple mutants of these nitrate transporters and examine their sensitivity to polyamines.

AUTHOR CONTRIBUTIONS
WT, AI, HM, and TT conceived and designed the experiments. WT, AI, RT, and TT performed the experiments. WT, HM, RT, and TT analyzed the data. SS, KY, MY, MH, and SS contributed materials and analysis tools. WT and TT wrote the paper.