Arabidopsis LHT1 Is a High-Affinity Transporter for Cellular Amino Acid Uptake in Both Root Epidermis and Leaf Mesophyll

Identification of Mutants Defective in Amino Acid Uptake

Arabidopsis is able to use various amino acids as the sole source of nitrogen when grown axenically. To identify transporters for amino acid uptake from soil, various mutant lines with reduced expression of amino acid transporters were grown on media containing amino acids as a sole nitrogen source. The screen included insertion lines for AAP3, AAP6, AAP8, LHT1, and various RNA interference (RNAi) lines (AAP2 and AAP4), several of which, according to GENEVESTIGATOR (https://www.genevestigator.ethz.ch/; Zimmermann et al., 2004), had been found to be expressed in roots. Only lines carrying insertions in LHT1 showed reduced growth compared with the wild type, suggesting that LHT1 may be involved in amino acid uptake (Figure 1 ). Two independent homozygous Arabidopsis lines carrying T-DNA insertions in LHT1 were identified in the SALK T-DNA insertion collection (Alonso et al., 2003). Sequence analysis confirmed that lht1-1 and lht1-2 contain T-DNA insertions in the third exon and fourth intron of the LHT1 coding region, respectively (see Supplemental Figure 1 online). PCR of genomic DNA with primers on both sides of the insertion (lht1-1) and a left border primer and a downstream primer confirmed that the insertions disrupt the locus (see Supplemental Figure 2 online). Both RT-PCR and RNA gel blot analysis failed to detect LHT1 mRNA, suggesting that both lht1-1 and lht1-2 are null mutants (Figure 2B ; see Supplemental Figure 3 online).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1.

Growth Comparison of T-DNA Insertion Lines, Wild-Type, and P_35S-LHT1 Plants on Amino Acids as Sole Nitrogen Source, and Uptake Rates of ¹⁴C-Labeled Amino Acids.

(A) Growth of the T-DNA insertion line lht1/lht1 and P_35S-LHT1 plants on 5 mM nitrate as nitrogen source was comparable to the growth of wild-type plants.

(B) On 5 mM Asp as sole nitrogen source, growth of the T-DNA insertion line was strongly reduced and P_35S-LHT1 plants grew larger than wild-type plants. Heterozygous plants (lht1/LHT1) showed an intermediate growth reduction. No differences were observed during the first 10 d, growth differences were visible after 15 d, and photographs were taken after 21 d.

(C) Growth comparison on Asn, GABA, Glu, and Gln. On 5 mM of the amino acids indicated, no difference was observed with Gln, reduced growth of the T-DNA insertion line was visible with Glu and GABA, and the T-DNA insertion line grew better than the wild type and overexpressors with Asn, possibly as a result of a toxic effect. Photographs were taken after 21 d.

(D) Uptake rates for Asp, Glu, and Gln. In short-time uptake studies, homozygous T-DNA insertion lines showed a reduced uptake rate for Asp (lht1/lht1, 0.31 pmol·h⁻¹·seedling⁻¹; Col-0, 1.69 pmol·h⁻¹·seedling⁻¹). Uptake rates were reduced to 18% in the T-DNA insertion line. Uptake rates were enhanced up to 253% of the wild type for Asp in P_35S-LHT1 plants (7.28 pmol·h⁻¹·seedling⁻¹). Error bars represent se (n = 3). Linear uptake rates for radiolabeled Glu were 1.06 pmol·h⁻¹·seedling⁻¹ for lht1/lht1 and 3.83 pmol·h⁻¹·seedling⁻¹ for Col-0, and rates for Gln were 1.75 pmol·h⁻¹·seedling⁻¹ for lht1/lht1 and 10.8 pmol·h⁻¹·seedling⁻¹ for Col-0 (1n = 10 seedlings, n = 3, error bars indicate se). Overexpression enhanced the uptake rate in all cases but was less pronounced for Gln. Uptake rates of ¹⁴C-labeled amino acids determined for P_35S-LHT1 seedlings were 12.0 pmol·h⁻¹·seedling⁻¹ for Glu and 22.2 pmol·h⁻¹·seedling⁻¹ for Gln.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2.

Organ-Specific Expression of LHT1, and Phenotypes of T-DNA Insertion Lines, lht1-RNAi Plants, and LHT1-Overexpressing Plants.

(A) Two homozygous T-DNA insertion lines (lht1-1 = SALK_034566, lht1-2 = SALK_026389) were significantly smaller than the wild type when grown in soil. To verify the effect of the disruption of the LHT1 locus, transgenic plants expressing a hairpin construct directed against LHT1 transcripts were generated. Two lines (lht1-RNAi-3 and lht1-RNAi-4) showed the same reduced growth as both homozygous T-DNA insertion lines. Overexpression of LHT1 (P_35S-LHT1) did not result in a significant phenotype.

(B) The reduction of transcript levels of LHT1 or overexpression in 35S plants was analyzed by RNA gel blot analysis on RNA extracted from whole plants. No transcripts of LHT1 were detectable in homozygous lines and RNAi lines.

(C) Total RNA was used as a loading control.

(D) LHT1 transcripts in Col-0 were detected in all organs except siliques, with the strongest signals in roots and leaves and weaker signals in stems, siliques, and flowers. Plants for RNA extraction were grown in fertilized soil in the greenhouse.

(E) Total RNA from each organ was used as a loading control.

When grown on either 5 mM Asp or 5 mM Glu (Figures 1B and 1C) as the sole nitrogen source, development of lht1-1 was retarded, whereas on 5 mM γ-aminobutyrate (GABA), a slight reduction in growth and development was observed relative to the wild type. Conversely, Asn, which had a toxic effect on wild-type plants, was less toxic for the lht1 mutants. Other proteinogenic amino acids were tested but did not sustain any growth of wild-type plants at concentrations of 5 mM (data not shown). The role of LHT1 as an amino acid uptake system in roots (at least for the amino acids tested) was confirmed and quantified by tracer uptake studies in whole seedlings. The uptake rates were reduced by 85% (Gln), 73% (Glu), and 82% (Asp) in lht1-1 relative to the wild type (Figure 1D). Total uptake rate was highest for Gln, with 1.8 pmol·h⁻¹·seedling⁻¹ in lht1-1 and 10.8 pmol·h⁻¹·seedling⁻¹ in the wild type. The uptake rates were lowest for Asp, with 0.3 pmol·h⁻¹·seedling⁻¹ in lht1-1 and 1.7 pmol·h⁻¹·seedling⁻¹ for wild-type plants.

In a second experiment, [¹⁴C]l-Gln uptake from the growth medium into roots and shoots was analyzed separately. The results show that both uptake via the root and transport to the shoot were strongly reduced in the mutants (Figure 3A ). The remaining transport activity in lht1 may be mediated by other members of the LHT subfamily, although LHT2 was not found in roots but rather in the tapetum (Lee and Tegeder, 2004). To test whether partial inhibition of LHT1 may already limit amino acid uptake, heterozygous LHT1/lht1-1 plants were analyzed. The heterozygous plants displayed reduced LHT1 mRNA levels and were retarded in growth on Asp as the sole N source, demonstrating that LHT1 activity is limiting (Figure 1B; see Supplemental Figure 2 online). Thus, the analysis of insertion mutants suggests that LHT1 functions as a nonselective amino acid transporter and represents one of the major uptake systems from the growth medium, at least for the amino acids tested.

• Download figure
• Open in new tab
• Download powerpoint

Figure 3.

Feeding of [¹⁴C]Gln to Roots, Protoplast Uptake with [¹⁴C]Pro, and Size Distribution of Protoplasts.

(A) [¹⁴C]Gln was applied to the agar, seedlings were removed, and roots and shoots were analyzed separately. Both in roots and shoots of the mutants, uptake of [¹⁴C]Gln was strongly reduced. Error bars represent sd (n = 3).

(B) Time-dependent uptake of [¹⁴C]Pro in mesophyll protoplasts of wild-type (closed circles) and lht1/lht1 (open circles) plants. Mesophyll protoplasts were isolated from 10-week-old plants grown under short-day conditions. Error bars represent sd (n = 3).

(C) Size distribution of mesophyll protoplasts isolated from wild-type plants and mutants. Values represent numbers of protoplasts between the indicated sizes as percentages of the total number (wild type, n = 538; lht1, n = 317).

Broad Selectivity of Yeast-Expressed LHT1

The LHT1 clone used in the original study complemented the His auxotrophy of the yeast strain JT16 (his4, hip1, can1, ura3), a mutant with reduced transport activity for basic amino acids (Chen and Bush, 1997). The substrate specificity, as determined by uptake measurements with radiolabeled amino acids and subtraction of the endogenous background mediated by a large number of endogenous transporters, had indicated that Lys and His were the best substrates, but other substrates, such as Glu, Leu, and several others, were also transported. To analyze the selectivity of LHT1, LHT1 cDNA was isolated by PCR from ecotype Columbia (Col-0) and expressed in the multiple–amino acid permease knockout yeast strain 22Δ8AA (Fischer et al., 2002), which is unable to use Asn, Asp, citrulline, GABA, Glu, or Pro as the sole nitrogen source. The yeast mutant 22Δ8AA was transformed with vectors expressing three clones of LHT1 isolated by RT-PCR under the control of the PMA1 promoter; AAP6 served as a positive control. On media containing 3 mM citrulline, Glu, Asp, GABA, or Pro as the sole nitrogen source, independent transformants expressing LHT1.1 were able to grow, in contrast with LHT1.3, which carried a T402G mutation caused by a PCR error that led to a premature stop codon (Figure 4A ). Cells transformed with the control plasmids grew only on 10 mM NH₄SO₄ and weakly on 1 mM Arg. On media with citrulline as the sole nitrogen source, only yeast transformed with functional LHT1 clones showed significant growth. These data demonstrate directly that LHT1 is able to mediate the transport of a variety of amino acids.

• Download figure
• Open in new tab
• Download powerpoint

Figure 4.

Biochemical Characterization of LHT1.

(A) Comparison of growth of the yeast mutant strain 22Δ8AA mediated by three PCR clones of LHT1 on different amino acids as sole nitrogen source compared with the general amino acid permease AAP6. Yeast transformed with empty vector was used as a control. LHT1.1 and LHT1.2 mediated growth comparable to AAP6 on 3 mM Asp, Pro, or Glu and mediated faster growth on 3 mM citrulline or GABA. LHT1.3, carrying a T402G mutation leading to a premature stop codon attributable to a PCR error, served as a control and did not mediate growth. Plates were scanned after 5 d except citrulline plates (7 d).

(B) Kinetic properties of LHT1. Linear uptake rates of l-[¹⁴C]Pro, l-[¹⁴C]Glu, and l-[¹⁴C]His were determined with concentrations ranging from 1 μM to saturation. Data represent six repeats for each time point (1, 2, and 4 min) and concentration, and error bars indicate se. The calculated K_m value of LHT1 for Pro is 10 ± 0.5 μM, and V_max ∼ 82 nmol Pro·g⁻¹ protein· min⁻¹; for Glu, K_m is 13.6 ± 0.7 μM and V_max ∼ 100 nmol Glu·g⁻¹ protein·min⁻¹; and for His, K_m is 360 ± 28 μM and V_max ∼ 105 nmol His· g⁻¹ protein·min⁻¹.

(C) Linearity and concentration-dependent velocity of Pro transport. These data were used for the K_m calculation of Pro uptake.

(D) Specificity of transport. Uptake of l-[¹⁴C]Pro (10 μM) into yeast cells (22Δ8AA-LHT1) was monitored in the presence of a 10-fold excess (100 μM) of final competitor concentration. Numbers indicate percentage uptake compared with control. l-[¹⁴C]Pro uptake without competitor was taken as 100%. n = 6, and error bars indicate se.

The kinetic properties of LHT1 were determined by uptake experiments with ¹⁴C-labeled Pro, His, and Glu in the yeast mutant 22Δ8AA (Figure 4B). The K_m value of LHT1 for Pro was ∼10 μM, with a V_max of ∼80 nmol·g protein⁻¹·min⁻¹; for Glu, the K_m was ∼14 μM and the V_max was ∼100 nmol·g protein⁻¹·min⁻¹; for His, the K_m was ∼360 μM and the V_max was ∼105 nmol·g protein⁻¹·min⁻¹. Pro uptake kinetics were linear over time (Figure 4C), as were the other kinetics for the determination of K_m values (data not shown). Although Glu transport was also significant compared with the previous study (Chen and Bush, 1997), the significant Pro transport activity observed here was not detected in the previous work. The K_m values for Pro and Glu are in the same range as the K_m value for Pro determined for LHT2 (13 μM) (Lee and Tegeder, 2004), and both LHT1 and LHT2 show a higher affinity toward these amino acids compared with the AAPs (Fischer et al., 2002; Okumoto et al., 2002).

To determine the specificity in further detail, [¹⁴C]Pro uptake was determined at a 10-fold excess of unlabeled proteinogenic amino acids. Under the conditions tested (i.e., with a Pro concentration at the K_m for Pro [10 μM]), the strongest competitors for Pro uptake were Cys and Phe; the weakest competitors were the basic amino acids His, Lys, and Arg (Figure 4D).

Finally, to demonstrate that amino acids, which are competitors for Pro uptake, are also direct substrates for LHT1, uptake studies were performed at a concentration of 10 μM. All tested amino acids showed linear uptake kinetics, but as already shown in the competition experiments, His was transported with the lowest velocity at a concentration of 10 μM (Figure 5 ).

• Download figure
• Open in new tab
• Download powerpoint

Figure 5.

Comparison of Linear Uptake Rates of Six Amino Acids at a Concentration of 10 μM, CCCP Sensitivity, and Energy Dependence of Transport.

¹⁴C-labeled Gly, Ala, Val, Glu, Pro, and His were used at a concentration of 10 μM, and the velocity of transport was determined. Closed circles, pDR196-LHT1; open circles, addition of CCCP (10 μM) to yeast expressing LHT1 after 1 min; open triangles, LHT1 without glucose; closed triangles, vector control (pDR196). The addition of CCCP (arrows) abolished the uptake activity of LHT1. Yeast cells expressing LHT1 without glucose displayed no uptake activity above the vector control. The uptake velocity of His was a factor 4- to 12-fold lower than the uptake rates for other amino acids, confirming the lower affinity of LHT1 toward His.

The energy dependence of the uptake was shown by omitting glucose from the uptake medium in the experiments using [¹⁴C]Gly and [¹⁴C]Ala as tracers, leading to a reduction of uptake activity to background levels. The addition of the protonophore CCCP (10 μM) 1 min after the start of the uptake experiment abolished amino acid uptake. These data suggest proton cotransport as the uptake mechanism (Figure 5), although secondary effects like the change in cytosolic pH and the breakdown of ATP synthesis may affect transport activity.

We also compared the properties of LHT1 in the presence and absence of sorbitol, a major difference to the protocol used compared with the previous study (Chen and Bush, 1997), but could not detect significant effects of sorbitol (data not shown).

Thus, LHT1 functions as a general amino acid permease similar to other AAPs (Okumoto et al., 2002); however, compared with all other known amino acid transporters, it displays a much higher affinity for amino acids. Both the high affinity and the broad selectivity are compatible with a function in the uptake of amino acids from soil and with the in planta results observed in lht1 mutants.

Expression Pattern of LHT1 and Subcellular Localization

RNA gel blot analysis showed strong expression in roots, leaves, stems, flowers, and siliques (Figure 2D) (Chen and Bush, 1997). To analyze the expression pattern in more detail, promoter β-glucuronidase (GUS) studies were performed. A 2.7-kb promoter fragment containing the upstream region (up to the next open reading frame [ORF]) and including the first intron of LHT1 was fused translationally to the GUS gene (P_LHT1-GUS), and Arabidopsis plants were transformed. Analysis of 14 independent lines showed a similar GUS pattern. Histochemical GUS studies showed that LHT1 is expressed in roots already at early stages of development, mainly in the rhizodermis of emerging roots and in lateral roots, but not in the main root (Figures 6A and 6F). In older plants, the tips of all roots show GUS staining (Figure 6B). Expression in young leaves is weaker compared with older leaves and is detected first at the hydathodes. At later stages of leaf development, expression is found in the epidermis, whereas in mature leaves expression was found in all mesophyll cells (Figures 6D, 6E, and 6I). Expression was absent from vascular bundles (Figures 6C to 6E). In flowers, the sepals and petals were also stained, but no expression was detected in siliques and seeds (Figures 6G and 6H). The GUS expression in roots supports the assumed role of LHT1 in the uptake of amino acids from the soil. In leaves, a strong increase in GUS activity is visible during the progression of plant age (Figures 6I to 6K). The leaf expression suggests that LHT1 serves additional functions, potentially the cellular uptake of amino acids delivered from the root via the xylem or in cell-to-cell transport in source leaves. All of the GUS data fit well to the available expression data on public microarray resources and are in good agreement with the observed root expression in seedlings and the developmental change in the expression pattern.

• Download figure
• Open in new tab
• Download powerpoint

Figure 6.

Tissue-Specific GUS Activity under the Control of the LHT1 Promoter.

(A) In seedlings (5 d old), strong GUS expression is visible in the lateral roots and only weak signals are visible in the cotyledons.

(B) In older plants grown in soil, GUS expression can be found at all root tips.

(C) No expression is visible in the veins of leaves.

(D) Cross section of a young leaf shows staining exclusively in the lower and upper epidermis.

(E) Cross section of an old leaf displays staining in all mesophyll cells and in the epidermis.

(F) In roots, staining is also visible in the emerging lateral roots.

(G) In flowers, GUS signal can be found in sepals and pedicels.

(H) No signal is visible in siliques or seeds.

(I) to (K) Staining of whole plants grown in fertilized soil after 24 (I), 31 (J), and 42 (K) d. All plants were stained for 8 h at 37°C. Signal intensity increased in leaves of older plants.

Because amino acids are transported not only across the plasma membrane but essentially in and out of all cellular compartments, the cellular localization of the transporter was analyzed. Fluorescence microscopy of N-terminal P_35S-green fluorescent protein (GFP)-LHT1 fusions transiently expressed in protoplasts indicates that LHT1 is located at the plasma membrane (Figure 7 ). Moreover, LHT1 had been identified in a plasma membrane proteomics analysis (Alexandersson et al., 2004). Thus, LHT1 seems to be responsible for amino acid uptake from the soil and for cellular retrieval from the apoplasm.

• Download figure
• Open in new tab
• Download powerpoint

Figure 7.

Subcellular Localization Using P_35S-LHT1-GFP Fusion Protein.

LHT1 mRNA was translationally fused with GFP and expressed transitionally in Arabidopsis protoplasts (P_35S-LHT1-GFP). Representative fluorescence microscopic images of a protoplast expressing LHT1-GFP fusion protein are shown. GFP fluorescence was detected at the plasma membrane. Protoplasts expressing GFP protein under the control of the CaMV 35S promoter were used as a control (P_35S-GFP). GFP fluorescence in control protoplasts was detected throughout the cytosol. The fluorescence (left) and bright-field (middle) images are overlaid at right.

A Second Role of LHT1 in Leaves

To test for the function of LHT1 in other tissues in addition to roots (i.e., in leaves), plants were grown in the presence of inorganic nitrogen sources in soil. On fertilized soil, homozygous lht1 lines were significantly smaller compared with the wild type (Figure 2A). A similar phenotype was observed when the expression of LHT1 was inhibited by RNAi (Figures 2A and 2B; see Supplemental Figure 1 online). In the heterozygous lines, in which expression of LHT1 mRNA was reduced by ∼50%, growth of the plants in soil was similar to that in the wild type (see Supplemental Figure 2 online). Because nitrogen supply should not be limiting on fertilized soil, the growth phenotype is most likely attributable to a lack of amino acid transport activity at a different site in the plant. Because LHT1 is the only amino acid transporter found to be expressed in leaf cells outside of the vascular system to date, the second crucial role of LHT1 is potentially in the cellular uptake of amino acids in leaf cells. The long-distance transport of amino acids between roots and leaves and the other organs is complex, because amino acids are both transported in the phloem and xylem and cycle in the plant (Cooper and Clarkson, 1989). A defect in cycling is expected to limit nitrogen use and thus growth, and one would expect amino acids to accumulate in the apoplasm of leaves. A function in leaf cell uptake is supported by increased amino acid content in the apoplasm of lht1-1 compared with the wild type (Table 1 ). Given the complexity of cycling and the involvement of multiple transport steps, it is not surprising that the overall amino acid composition of the leaves was altered compared with the wild type and that the total protein content was increased in lht1-1 leaves (Table 2 ). Anion and cation analyses showed that nitrate levels were reduced and free ammonium was increased in lht1 leaves (Table 2).

Reduced Amino Acid Uptake in lht1 Mesophyll Protoplasts

To test directly whether the loss of one amino acid transporter in mesophyll cells could lead to the increased level of amino acids in the apoplasm of lht1 mutants, analyses of the uptake of radiotracers into protoplasts isolated from mature leaves of the mutants were performed. In both mutant and wild-type protoplasts, uptake of [¹⁴C]l-Pro was linear over 60 min, but protoplasts isolated from lht1/lht1 plants showed a drastic reduction in Pro uptake velocity (∼10% compared with the wild type) (Figure 3B). Because the mutant plants are significantly smaller and uptake capacity depends on the cell surface, protoplast diameters were measured. No difference in size distribution or average size of the protoplast of a typical preparation was found when comparing wild-type and mutant protoplasts (Figure 3C). These data suggest that LHT1 is important for amino acid uptake from the apoplasm into leaf mesophyll cells.

Complementation of the Soil Phenotype by Reexpression of LHT1 in Leaves of lht1-1 Mutants

To determine whether the two potential phenotypes (i.e., the defect in root uptake and the defect in leaf uptake) are independent, a root-specific lht1 knockout was generated by expression of LHT1 in the mutant under the control of the potato stem/leaf-specific promoter St LS1 (Eckes et al., 1986). The St LS1 promoter is active in cells with fully developed chloroplasts, and reporter gene analysis showed the absence of promoter activity in all tissues without green chloroplasts (Stockhaus et al., 1989). lht1 plants transformed with this construct developed normally in fertilized soil (Figure 9A), whereas the mutants were strongly retarded in growth. The recovery of the phenotype in soil supports the hypothesis that the phenotype in fertilized soil is attributable to reduced cell-to-cell transport and cycling of amino acids. On the other hand, the phenotype of the lht1 mutant observed on media containing Asp as the sole nitrogen source (Figure 1B) could not be rescued by the expression of LHT1 under the control of the St LS1 promoter (Figures 9B to 9F), indicating that the expression of LHT1 in root tissue is necessary for Asp uptake.

Regulation of LHT1 Expression by Nitrogen

Inorganic nitrogen uptake is highly regulated by the availability of nitrogen sources (Gazzarrini et al., 1999; von Wiren et al., 2000). To analyze whether the expression of LHT1 is regulated depending on the nitrogen supply or nitrogen status of the plant, seedlings were grown on various N sources and mRNA levels in the wild type were determined. In parallel, enzymatic GUS activity was determined in extracts from P_LHT1-GUS plants grown under the same conditions. Expression of LHT1 was found to be inducible by amino acids. On control medium (2MS; see Methods) with 40 mM ammonium nitrate, LHT1 expression was low. In the presence of other amino acids as sole N sources (5 mM), but also on 5 mM nitrate, expression of LHT1 was induced (Figure 8 ). The GUS activity determined in parallel experiments with P_LHT1-GUS correlates with this upregulation of LHT1 mRNA levels. The data show that nitrogen supply and composition are sensed by the plant and that LHT1 expression is regulated accordingly. The regulation of expression may be a means of adapting to the local changes in organic and inorganic N supply in the root area. Further studies will be required to define the exact nature of the signal and signaling cascade and the integration with other nitrogen transporters.

• Download figure
• Open in new tab
• Download powerpoint

Figure 8.

GUS Activity under the Control of the LHT1 Promoter and LHT1 mRNA Levels in Seedlings Grown on Various N Sources.

(A) LHT1 promoter-GUS seeds were sown on the indicated amino acids or nitrate as N source (all at 5 mM), and GUS activity was determined after 14 d as described. GUS activity was induced on all amino acids and nitrate as N sources compared with MS medium containing 40 mM ammonium nitrate. Induction was lowest on plates containing 5 mM Gln (1n = five seedlings, n = 3, error bars represent se). 4MU, 4-methylumbelliferone.

(B) Corresponding expression levels of LHT1 determined by RNA gel blot analysis. Wild-type (Col-0) seedlings were grown as described above, and RNA was extracted. The level of LHT1 mRNA is in good correlation with the determined GUS activity and was lowest for plants grown on MS medium (40 mM ammonium nitrate). On Gln, only weak induction of LHT1 is visible.

(C) Total RNA was used as a loading control.

Overexpression of LHT1 Leads to Enhanced Amino Acid Uptake in Roots

Given that heterozygous lht1 mutants showed a reduction in amino acid uptake, we tested whether LHT1 activity limits organic nitrogen uptake in a noncultivated plant species, a question relevant with respect to the finding that plant species differ in their capacity to take up amino acids (Lipson and Nasholm, 2001). To test whether LHT1 is limiting for amino acid uptake, LHT1 was overexpressed under the control of the CaMV 35S promoter. The overexpressing lines grew larger than wild-type plants on media containing Asp as the sole nitrogen source, whereas the toxicity observed in the presence of Asn was increased (Figures 1B and 1C). The uptake rates of ¹⁴C-labeled Asp, Glu, and Gln were also strongly increased in the P_35S-LHT1 plants, consistent with the observation of improved growth under limiting N conditions (Figures 1B and 1D). Partial growth inhibition was observed in heterozygous lines that expressed less LHT1 mRNA on Asp-containing media (Figure 1B), establishing a direct correlation to the growth phenotype under limiting N conditions and the LHT1 mRNA levels. The overexpression lines showed no obvious phenotypic differences compared with the wild type under greenhouse conditions, and amino acid levels in the apoplasmic wash fluid of P_35S-LHT1 plants were not altered significantly (Table 3 ), suggesting that LHT1 does not limit xylem unloading in leaves. The data show that overexpression of an amino acid transporter can improve plant growth under limiting N conditions.

LHT1 Is Involved in Amino Acid Uptake into Roots

Apart from the inorganic nitrogen forms, ammonium and nitrate, the soil may also contain significant amounts of organic nitrogen derived from the decomposition of organic matter. It had been shown that plants have the capacity to take up amino acids, and putative amino acid transporters have been localized to epidermal root cells (Soldal and Nissen, 1978; Bick et al., 1998).

In this work, LHT1, a transporter identified previously by Chen and Bush (1997), transports a broad spectrum of amino acids and is expressed in the rhizodermis of roots, making it a candidate for an uptake carrier for amino acids. The growth of mutants defective in LHT1 function is retarded, and radiotracer studies provide direct evidence for reduced uptake activity.

Moreover, the growth of lht1 mutants on Asp as the sole nitrogen source was not complemented when the LHT1 gene was expressed only in leaf tissue (Figures 9B to 9F), whereas the phenotype was complemented by expressing the LHT1 gene under the control of the 35S promoter (Figure 1B). These results indicate that LHT1 is responsible for the uptake of amino acids into root tissue, which may occur either at the root surface or from the apoplasm of the cortex.

• Download figure
• Open in new tab
• Download powerpoint

Figure 9.

Expression of LHT1 under the Control of the Potato LS1 Promoter in the lht1 Mutant Complements the Growth Phenotype in Soil but Not on Media Supplemented with Asp as the Sole Nitrogen Source.

(A) lht1 mutants transformed with LHT1 cDNA under the control of a leaf-specific promoter, P_St-LS1-LHT1, grew normally in fertilized soil, whereas lht1 mutants were retarded in growth.

(B) to (F) Expression of LHT1 under P_St-LS1-LHT1 does not complement the growth phenotype of the lht1 mutant in axenic culture on media containing 5 mM Asp as the sole nitrogen source.

(B) Root length of seedlings after growth on agar plates containing 5 mM Asp and glufosinate for 28 d. Expression of P_St-LS1-LHT1 in lht1 mutants did not improve growth compared with lht1 mutants transformed with the vector control.

(C) and (D) Two independent lines of lht1/P_St-LS1-LHT1 plants grown on 5 mM Asp.

(E) and (F) Two independent lines of lht1 plants expressing vector control grown under the same conditions as the lines in (C) and (D). Note that no phenotypic difference can be observed between (C) and (D), and (E) and (F). Bar = 1 cm.

The kinetic properties of LHT1 and its high affinity toward amino acids are consistent with ecological data. Free amino acid levels in soils are 20 to 100 μM (Lipson and Nasholm, 2001). The function in cellular uptake is further supported by the finding that LHT1 is sensitive to protonophores and that the activity depends on the presence of glucose as an energy source (Chen and Bush, 1997; this work). Together, the biochemical properties of LHT1 and its localization in lateral roots are all in agreement with a function in amino acid uptake from the soil. The direct correlation of the expression level of LHT1, amino acid uptake, and growth is supported by the reduced growth of heterozygous plants on Asp as a nitrogen source and the better growth of LHT1-overexpressing plants. The capacity to take up amino acids even in a species naturalized over a broad latitudinal and climatic range may increase competitiveness, especially under limiting N conditions. It has been shown that the natural variance in uptake capacity varies within the species of a family, dependent probably on adaptation to different climatic environments, resulting in a higher or lower concentration of organic N in soil (Raab et al., 1999). This can be highly relevant for plant growth because the ability to use limiting nutrients such as amino acids is an important factor for the competitive productivity of a plant species. In arctic regions with high amounts of organic N in the soil, the use of limiting N determines the dominance of a species (McKane et al., 2002). Moreover, plants compete with microorganisms for organic nitrogen in soil (Lipson et al., 1999), and high-affinity uptake mechanisms and specialization can help them to survive in N-limited regions.

The data showing enhanced amino acid uptake by plants overexpressing LHT1 open new research areas, such as whether the overexpressing plants are more competitive in wild populations and can outcompete other plants and microorganisms under limiting N supply. In addition, the enhanced uptake capacity for organic N may be one step toward improving the nitrogen use efficiency of crop plants, expanding agriculture into areas where cultivation has not been possible and decreasing nitrogen leaching and ground water contamination.

In contrast with Arabidopsis, most plants are mycorrhized. Mycorrhiza contribute to nitrogen supply and transfer nitrogen across the fungus–root interface, presumably in the form of amino acids (Wipf et al., 2002a). Thus, it is conceivable that in mycorrhized species, LHT1 homologs may be involved in the transfer of fungus-derived amino acids to the plant. In addition, it has been shown that mineralization of organic N is not a prerequisite for the uptake of nitrogen in boreal forests (Nasholm et al., 1998) and that intact amino acids are taken up directly from the soil (Meline and Nilsson, 1953). Trees, although most are mycorrhized, are able to use supplied organic N as a sole nitrogen source, and no growth differences are observed on amino acids compared with inorganic forms of nitrogen (Ohlund and Nasholm, 2001). LHT1 homologs may be important for amino acid uptake in these species as well, because database searches show that LHTs are present not only in other dicotyledons but also in monocotyledonous species (rice [Oryza sativa], Os HT [Liu et al., 2005]; easter lily [Lilium longiflorum], accession number BAA04838.1). The availability of LHT1 as a probe may provide a useful starting point for studying N efficiency in ecosystems in which organic N plays an important role in plant growth and species development.

A Potential Role for LHT1 in Apoplasmic Retrieval in Leaves and Amino Acid Cycling

Apart from its role in amino acid uptake from the substrate, LHT1 seems to play a second role in amino acid distribution. In soil fertilized with inorganic nitrogen, plant growth was severely inhibited. Because the major nitrogen sources, ammonium and nitrate, were present under these conditions, it is not likely that the inhibition of growth is attributable to reduced amino acid uptake. The expression of LHT1 in leaf mesophyll and epidermis suggested a role for LHT1 in the uptake of amino acids into leaf cells from the apoplasm. This hypothesis is consistent with the finding that the apoplasmic wash fluids of the lht1 mutants are enriched in amino acid content (Table 1). Mesophyll protoplasts isolated from leaves of the mutants display a strong reduction in amino acid uptake (Figure 3B). Moreover, reintroduction of LHT1 into the lht1 mutant under the control of the stem/leaf-specific promoter St LS1 was sufficient to suppress the growth deficiency phenotype.

Amino acids are transported from leaves to roots in the phloem and from roots to shoots via the xylem. On average, each 1 mol of N absorbed cycles once before being used for growth (Cooper and Clarkson, 1989). Little exchange has been observed between N in the transport pools and that in the bulk of shoots and roots (Atkins, 2000). Thus, inhibition of uptake from the apoplasm reduces cell-to-cell transport and the cycling of amino acids within the plant. A defect in amino acid cycling is expected to limit nitrogen use and thus growth. Because extracellular pathogens such as epiphytic fungi and bacteria are thought to use amino acids for their nutrition, the presence of a retrieval system for amino acids from the apoplasm may also represent an important passive defense mechanism to keep the amino acid levels in the apoplasm low (Struck et al., 2004). It is thus suggested that LHT1 plays a role in taking up amino acids delivered to the mesophyll (e.g., via the xylem) and in keeping amino acid levels low in the cell wall space around leaf cells. The fact that the amino acid levels in the apoplasmic wash fluids of the overexpressing plants is not further reduced compared with those in control plants (Table 3) suggests that the available LHT activity is not limiting in the wild type. The ∼30% increase in amino acid accumulation in the apoplast of lht1 mutants may be considered low in that LHT1 would be the only importer for amino acids from the apoplasm; however, it is expected that other processes, such as phloem loading and regulatory effects, limit the extracellular levels in lht1 mutants. The observation that a mesophyll amino acid transporter is required for normal plant growth suggests that cycling is of central importance. Pulse-labeling studies in these mutants may provide novel insights into the complex cycling processes (Atkins, 2000).

The expression profile of LHT1 in leaves is compatible with the uptake into the mesophyll and epidermal cells, whereas other transporters are probably involved directly in the uptake of xylem-borne amino acids around the xylem, such as AAP6 (Okumoto et al., 2002). Nevertheless, the phenotype conferred by lht1 in fertilized soil and the altered levels of various amino acids (Table 2) show that this disruption interferes with a complex network of amino acid transport and metabolism. At least 53 putative transporters for amino acids have been identified. Most transporters analyzed to date show broad selectivity. Considering the large number of putative amino acid transporters, combined with possible colocalization, spatial and temporal overlapping expression, or specialized function in subcellular compartments, concerted effort will be required to analyze the complete cycling system. In some cases, the disruption of transporter genes leads to no visible phenotype, as in AAP3 (Okumoto et al., 2004). Double or triple knockouts may help determine how these transporters function in the network of amino acid distribution within the plant. With 53 putative amino acid transporters (Wipf et al., 2002b), however, a combinatorial analysis of all triple knockout mutants would involve the analysis of 23,426 combinations ( n = 53, k = number of knock out combinations). This demonstrates that a complete understanding of amino acid transport requires concerted action to examine the functions of all amino acid transporters. In addition, novel approaches will be required to obtain insights regarding the spatial and temporal distribution of amino acids in mutant leaves (Looger et al., 2005).

Plant Material, Growth, Transformation, and Transport Measurements

Arabidopsis thaliana plants (ecotype Col-0) were grown in growth chambers (14 h of light, 21.5°C) on Murashige and Skoog (MS) medium (40 mM ammonium nitrate; Murashige and Skoog, 1962) supplemented with 2% glucose (2MS) or 2MS without nitrogen (2MS-N) supplemented with 5 mM of each nitrogen source. Plant material used for RNA gel blot analysis, amino acid HPLC, and promoter-GUS expression was cultured in the greenhouse (16 h of light, 25°C) in soil. Transformation was performed under vacuum infiltration with Agrobacterium tumefaciens strains pGV2260 and pGV3101 (Bechtold and Pelletier, 1998). The suspension culture of tobacco (Nicotiana tabacum) Bright Yellow-2 cell line was grown as described (Merkle et al., 1996). Protoplasts from tobacco Bright Yellow-2 cultures were prepared as described (Merkle et al., 1996). Transient transformation of the protoplasts with polyethylene glycol was performed according to the protocol of Negrutiu et al. (1992). For confocal laser scanning microscopy, protoplasts were incubated overnight at 24°C in the dark after transformation and observed with a Leica DM RE microscope.

T-DNA insertion lines (Alonso et al., 2003) provided by the Salk Institute Genomic Analysis Laboratory were ordered at the ABRC. The T-DNA insertions of SALK_034566 (exon 3) and SALK_026389 (intron 4) were confirmed by sequencing PCR products with pROK2-specific left border primer.

For transport measurements, seedlings were grown for 14 d on 2MS (2% sucrose, 40 mM ammonium nitrate) and transferred in 1 mL of 2MS-N supplemented with 9.25 kBq of l-[U-¹⁴C]Asp (7.66 GBq/mmol), l-[U-¹⁴C]Glu (8.81 GBq/mmol), or l-[U-¹⁴C]Gln (8.85 GBq/mmol). Samples for linear uptake rates were removed after 90 and 180 min, transferred to 8 mL of ice-cold buffer (2MS-N, 10 mM Asp, 10 mM Glu, and 10 mM Gln), filtered, and washed twice with 8 mL of buffer. The seedlings were transferred to 1 mL of tissue solubilizer (p-diisobutyl-cresoxyethoxyethyl dimethylbenzylammonium hydroxide; Packard Instrument), incubated overnight, and neutralized with 1 mL of acetic acid, and radioactivity was determined by liquid scintillation spectrometry. The measurements represent means of three parallel experiments, and each sample consisted of 10 seedlings.

To determine the uptake of Gln into roots and shoots separately, [¹⁴C]Gln (92.5 kBq in 4 mL of water) was applied carefully to the plates without contacting the leaves and allowed to diffuse for 60 min. Seedlings were removed at the indicated time points, and roots and shoots were separated and treated as described above.

Protoplast Work: Mesophyll Protoplasts, Viability Assay, Size Measurements, and Uptake

Mesophyll protoplasts were prepared from 10-week-old plants grown under short-day conditions (8 h of light, 16 h of dark). Two pots of wild-type plants and six to eight pots for lht1 plants were used. Abaxial sides of leaves were peeled and collected in glass Petri dishes in mannitol solution (0.5 M mannitol). The mannitol solution was replaced with cell wall–digesting medium (1% [w/v] Cellulase [Onozuka R10], 0.25% Macerozyme (Duchefa), 0.4 M mannitol, 8 mM CaCl₂, and 10 mM MES, pH 5.5) and incubated for 3 h at 28°C without shaking. The resulting suspension was carefully filtered subsequently through nylon mesh (120 and 80 μm). Protoplasts were sedimented and washed twice at 100g for 5 min in 0.16 M mannitol/0.13 M CaCl₂ and finally resuspended in ∼3 mL of uptake medium (0.4 M mannitol, 8 mM CaCl₂, and 10 mM MES, pH 5.5). The density of the suspension was adjusted to 10⁶ cells/mL. The viability of the preparation was determined with fluorescein diacetate (Sigma-Aldrich) as described by Lasat et al. (1998). The viability of a typical preparation ranged from 89 to 93%.

To measure the diameter of the protoplasts, diameters were determined from digital images of the suspension and measured with the freeware tool ImageJ (http://rsb.info.nih.gov/ij/) using the squares of a hemacytometer (250 μm × 250 μm) as the calibration standard.

Uptake experiments were performed as described previously with slight modifications (Brown et al., 1997; Cosio et al., 2004). Protoplasts were preincubated with ³H₂O (168 kBq/mL) for 30 min at 22°C to allow ³H₂O to reach equilibrium inside and outside the protoplasts, enabling the calculation of the protoplast volume as described (Fieuw and Willenbrink, 1991; Brown et al., 1997).

Uptake experiments were started by adding [¹⁴C]Pro (84 kBq/mL final concentration) in uptake buffer. At the time points indicated, 100 μL of the suspension was removed and placed on top of a discontinuous gradient of 150 μL of silicon oil (AR200; Fluka) on top of 30 μL of 40% Percoll medium (uptake solution with 40% Percoll, pH 5.5) in a 400-μL microcentrifuge tube. The samples were centrifuged (20 s, 10,000g) to wash the protoplasts through the oil and separate from the feeding solution. The tips containing the protoplasts were cut off and placed in scintillation vials, and radioactivity was counted via liquid scintillation spectrometry. One microliter of the supernatant was taken in parallel for the determination of radioactivity. Uptake of [¹⁴C]Pro was calculated as the increase of ¹⁴C in the sample in relation to the amount of ³H as described (Brown et al., 1997). The formula used was {(¹⁴C [dpm] in protoplast pellet/³H₂O [dpm] in pellet × ¹⁴C [dpm] in 1 μL of supernatant/³H₂O [dpm] in 1 μL of supernatant)} × external Pro concentration (nmol/μL) to calculate the amount of Pro taken up (nmol) and the protoplast volume (μL).

DNA Work

Primers

Sequence-specific primers are shown in uppercase letters, and restriction sites are underlined. For yeast expression: PstI 5′ primer, 5′-ctgcagACCATGGTAGCTCAAGCTCC-3′; SalI 3′ primer, 5′-gtcgacTTATGAGTAAAACTTGTATCC-3′. For GFP fusion: blunt 5′ primer, 5′-ccccctgcagACCATGGTAGCTCAAGCTCC-3′; BamHI 3′ primer, 5′-ggatcccTGAGTAAAACTTGTATCCTTTTGC-3′. For pRTRNAi vector: intron AAP6 5′ primer, 5′-ctcCTAGGTCAGATTCGCTATC-3′; intron AAP6 3′ primer, 5′-tctagagtcgacccccccgggTTCCTCCTGTTTATCTTTTAACATC-3′. For RNAi constructs: XhoI 5′ primer, 5′-ctcgagGCAGCAGCGAGACAAAAAG-3′; blunt 3′ primer, 5′-GATTTGCCTCCAGTGACC-3′. For GUS fusion, XbaI 5′ primer, 5′-tctagaCCTTCATGTTCTAAACTATATG-3′; EcoRI 3′ primer, 5′-GTTTTCGTAAGAATTCAATTAC-3′; EcoRI 5′ primer, 5′-GTAATTGAATTCTTACGAAAAC-3′; blunt 3′ primer, 5′-atcGCTGCTGCTAATTTCTCATC-3′. For St LS1-promotor fusion: pAH1, 5′-gagagcaaaagaagaaaaaggctcgagtttggatcccctgctttaatgagatatgcg-3′; pAH2, 5′-ctatagggcgaattgggaagcttatatctgcagaattc-3′. For LHT1: XhO1, 5′-gagctcgagatggtagctcaagctcctcatgatg-3′; BamH1, 5′-gagggatccTTATGAGTAAAACTTGTATCCTTTTGC-3′.

RNA Work

Total RNA was isolated from mature leaves, stems, and other organs, and 20 μg of total RNA was separated on 1.2% formaldehyde agarose gels (Riesmeier et al., 1994). Hybridization was performed at 65°C in 0.25 M sodium phosphate, pH 7.2, 7% SDS, 1 mM EDTA, and 1% BSA for 16 h using the cDNA of LHT1 as a probe. Filters were washed twice with 2× SSC/0.1% SDS (1× SSC is 0.15 M NaCl and 0.015 M sodium citrate) and 0.2× SSC/0.1% SDS for 20 min at 65°C and exposed to x-ray film (Hyperfilm Amersham)

Yeast, Transformation, Competition, and Transport Measurements

The yeast strain 22Δ8AA (Fischer et al., 2002) was transformed (Dohmen et al., 1991) with three independent cDNA clones of LHT1, one cDNA clone of AAP6, and the vector pDR196. Transformants were selected on media lacking uracil supplemented with 10 mM ammonia sulfate. Complementation was performed on Arg (1 mM), Asp, citrulline, GABA, Glu, and Pro (3 mM each) as sole nitrogen source. For K_m determination, competition experiments, and linear uptake rates, yeast cells were grown to A₆₀₀ = 0.5, washed, and resuspended in ice-cold buffer (50 mM NaPi, pH 4.5, and 0.6 M sorbitol) to a final A₆₀₀ = 5. Yeast cells (200 μL) were preincubated at 30°C with 10 μL of glucose (1 M) before the start of the experiment. Samples for linear uptake rates were removed after 60, 120, and 240 s and transferred to 4 mL of ice-cold buffer, filtered on glass fiber filters, and washed twice with 4 mL of buffer. Radioactivity was determined by liquid scintillation spectrometry. Endogenous uptake activity of yeast transformed with pDR196 was subtracted as background. Competition experiments were performed with 18.5 kBq of l-[U-¹⁴C]Pro (8.58 GBq/mmol) and a 10-fold (100 μM) excess of the respective amino acids. The measurements represent means of three parallel experiments.

Analytical Methods

Liquid nitrogen–ground leaves were freeze-dried before amino acid extraction. Five milligrams of powder was extracted once in 80% methanol and once in 20% methanol. Both extracts were pooled and lyophilized. Pellets were resuspended in lithium buffer (0.7% lithium acetate and 0.6% LiCl; Pickering Laboratories) with 0.2 mM norleucine as an internal standard. Amino acids were separated by HPLC on a cation-exchange column (high-efficiency fluid column, 3 mm × 150 mm; Pickering Laboratories) using lithium buffer as an eluant. Amino acids were derivatized with ninhydrin before photometric detection.

Apoplasmic wash fluids were collected using the infiltration–centrifugation method (Lohaus et al., 2001) as described by Pilot et al. (2004). For one run, rosette leaves corresponding to 4 g fresh weight (leaves of ∼100 mutant plants and ∼30 wild-type plants) were collected.

Anion and cations were extracted from 5 mg of freeze-dried material with 1 mL of deionized water at 95°C for 5 min. Ion composition of leaf material was determined by HPLC (DX120 ion chromatography; Dionex) with an IonPac CS12A column (elunt, 20 M methansulfonic acid) for cations and an IonPac AS9-HC column (eluant, 9 mM Na₂CO₃) for anions using a suppressed conductivity detection system and an AS40 automated sampler (Dionex).

Soluble protein concentrations were determined according to Bradford (1976).

Enzymatic GUS Assay

GUS activity was determined using a modified fluorometric assay according to Jefferson et al. (1987) in a microplate fluorescence reader (Tecan-Safire). Proteins were extracted from five seedlings grown for 14 d on 2MS-N supplemented with the respective N source, and 20 μg of protein was used in the assay with 4-methylumbelliferyl-β-glucoronide as a substrate. The linearity of the reaction was monitored over 30 min, and a calibration curve using 4-methylumbelliferone (0 to 75 μM) was used for quantification.

Accession Numbers

Sequence data discussed in this article can be found in the GenBank/EMBL data libraries under the following accession numbers: LHT1, At5g40780, U39782; AAP6, At5g49630; LHT2, At1g24400; potato AAP1, Y09825.

Supplemental Data

The following materials are available in the online version of this article.

• Supplemental Figure 1. Maps of Genomic Structure, T-DNA Insertion, and DNA Constructs for GUS, RNAi, and GFP.

• Supplemental Figure 2. Verification of T-DNA Insertion and Reduced RNA Levels in Heterozygous Plants.

• Supplemental Figure 3. Absence of LHT1 Transcripts in lht1/lht1 Plants Compared with Col-0 Determined by RT-PCR.