Ectopic expression of nicotianamine synthase genes results in improved iron accumulation and increased nickel tolerance in transgenic tobacco



    1. Institut für Pflanzengenetik und Kulturpflanzenforschung (IPK), Corrensstr. 3, D-06466 Gatersleben, Germany
    Search for more papers by this author

    1. Institut für Pflanzengenetik und Kulturpflanzenforschung (IPK), Corrensstr. 3, D-06466 Gatersleben, Germany
    Search for more papers by this author
  • U. W. STEPHAN,

    1. Institut für Pflanzengenetik und Kulturpflanzenforschung (IPK), Corrensstr. 3, D-06466 Gatersleben, Germany
    Search for more papers by this author
  • R. HELL,

    1. Institut für Pflanzengenetik und Kulturpflanzenforschung (IPK), Corrensstr. 3, D-06466 Gatersleben, Germany
    Search for more papers by this author
    • *

      Present address, Heidelberger Institut für Pflanzenwissenschaften, Universität Heidelberg, Im Neuenheimer Feld 360, D-69120 Heidelberg, Germany.


    Corresponding author
    1. Institut für Pflanzengenetik und Kulturpflanzenforschung (IPK), Corrensstr. 3, D-06466 Gatersleben, Germany
      Dr H. Bäumlein. Fax: +49 39482 5500; e-mail:
    Search for more papers by this author

Dr H. Bäumlein. Fax: +49 39482 5500; e-mail:


Heavy metals are essential for basic cellular processes but toxic in higher concentrations. This requires the precise control of their intracellular concentrations, a process known as homeostasis. The metal-chelating, non-proteinogenous amino acid nicotianamine (NA) is a key component of plant metal assimilation and homeostasis. Its precise function is still unknown. Therefore, this article aims to contribute new information on the in vivo function of NA and to evaluate its potential use for plant nutrition and crop fortification. For this purpose, a nicotianamine synthase gene of Arabidopsis thaliana was ectopically expressed in transgenic tobacco plants. The presence of extra copies of the nicotianamine synthase gene co-segregated with up to 10-fold elevated levels of NA in comparison with wild type. The increased NA level led to: (a) a significantly increased iron level in leaves of adult plants; (b) the accumulation of zinc and manganese, but not copper; (c) an improvement of the iron use efficiency in adult plants grown under iron limitation; and (d) an enhanced tolerance against up to 1 m m nickel. Taken together, the data predict that NA may be a useful tool for improved plant nutrition on adverse soils and possibly for enhanced nutritional value of leaf and seed crops.


A sufficient iron supply for human nutrition, the breeding of iron-efficient plants for growth on adverse soils and the sustainable renaturation of contaminated soils by phytoremediation are challenging tasks for applied plant science dealing with metal metabolism. Some metals are essential components of various cellular processes, like photosynthesis and respiration; however, an excess supply is often toxic. Therefore, plants have evolved regulatory systems to maintain a precise equilibrium of metal concentrations, which is known as metal homeostasis (Clemens, Palmgren & Krämer 2002; Hell & Stephan 2003; Schmidt 2003; Curie & Briat 2003). Based on the statement of the Word Health Organization ( that iron deficiency is one of the most common nutritional disorders, it is highly desirable to improve the nutritional value of crop plants by increased concentrations of iron but also other essential metals. Related to this aim is the breeding of crop plants which thrive on low iron soils, since approximately one-third of the world soils are considered to be iron deficient. Various approaches have been applied to reach these goals. These approaches include, among others, for instance the ectopic expression of a yeast FRE gene (Samuelsen et al. 1998) and the overexpression of ferritin genes leading to an increased synthesis of the iron storage protein ferritin in cytoplasm and in plastids of transgenic tobacco (Van Wuytswinkel et al. 1999; Vansuyt, Mench & Briat 2000) and rice (Goto et al. 1999; Vasconcelos et al. 2003). Moreover, an improvement of the iron status could be achieved by the engineering of phytosiderophore synthesis via the overexpression of nicotianamine aminotransferase (NAAT) genes in cereal crop plants and tobacco (Takahashi et al. 2001). Finally, it has been demonstrated that the overexpression of the FRO2 ferric chelate reductase confers tolerance to growth on low iron (Connolly et al. 2003).

The manipulation of the cellular nicotianamine (NA) concentration seems to be another promising approach to increase the iron concentration in planta. Nicotianamine, a chelator of iron and other heavy metals, plays a key role in iron uptake, phloem transport and cytoplasmic distribution and ensures iron solubility in the weakly alkaline environment of the cytoplasm (Stephan et al. 1996; Von Wiren et al. 1999; Hell & Stephan 2003; Takahashi et al. 2003). Most of our knowledge on the physiological function of NA is based on investigations of the tomato mutant chloronerva (Scholz et al. 1992; Stephan & Scholz 1993). The cloning of nicotiamine synthase (NAS) genes (Herbik et al. 1999; Ling et al. 1999; Higuchi et al. 1999) provides new tools for the modulation of endogenous NA concentrations in plant tissues. Initial experiments based on the expression of Arabidopsis NAS genes in tobacco (Douchkov et al. 2002) and of the barley gene NASHOR1 in wheat (G. Hensel, IPK Gatersleben, personal comm.) demonstrate the principal applicability of this strategy for plant improvement.

Here we report on the increase of iron concentrations in transgenic tobacco lines overexpressing the Arabidopis AtNAS1 gene with beneficial consequences for plant iron content and iron use efficiency. Furthermore, it is demonstrated that the increased NA concentration enables the plants to tolerate toxic concentrations of nickel.


Plant growth

Tobacco seeds were surface-sterilized with 70% ethanol and allowed to germinate on Murashige and Skoog (MS) agar plates. A light and temperature cycle of 16/8 h (light intensity 330 µE m−2 s−1) and 25/18 °C was used with a relative humidity of approximately 65%. Two-week-old plants were transferred to trays filled with Hoagland medium containing defined iron concentrations. A concentration of 40 µm Fe-EDTA [ethylenediaminetetraacetic acid iron (III) sodium salt] (Arabidopsis) and 10 µm Fe-EDTA (tobacco) has been applied for normal iron supply; whereas 1 µm Fe-EDTA is considered as limited iron supply. All vessels for low iron growth conditions were prewashed with 3 m HCl and rinsed with de-ionized water before use. The plants were grown in trays for a period 25 d before analysis. Arabidopsis plants were placed on top of a rock-wool bundle fixed in a hole with the lower part dipped into a tray of Hoagland solution containing 40 µm Fe-EDTA. The tray was covered by a glass. Germination and further growth was at a day/night regime of 8/16 h (short day) and 21/19 °C and a light intensity of 330 µE m−2 s−1. The nutrient solution was aerated and renewed weekly. After 3 weeks the plants were transferred to fresh medium containing 0, 40 or 100 µm Fe EDTA. After 2 weeks the plants were harvested for analysis. For the nickel tolerance experiments tobacco seeds were allowed to germinate on MS-medium (Duchefa, Haarlem, The Netherlands) supplemented by the indicated concentrations (600 and 1000 µm) of nickel chloride. Further details of growth conditions have been described previously (Douchkov 2003).

Cloning techniques

Basic molecular biology methods were performed according to Sambrook, Maniatis & Fritsch (1989) and have been described in detail (Douchkov 2003). The intron-free ATNAS1 gene was PCR-amplified using primers containing BamHI sites. The BamHI fragment was cloned into pCRII (Invitrogen Corp., Carlsbad, CA, USA), resequenced and recloned in sense orientation into the BamHI site of the binary vector pBinAR19. These constructs have been used for Agrobacterium-mediated gene transfer into tobacco as described (Bäumlein et al. 1991).

Determination of metal ion concentrations

The concentration of iron and nickel was determined according to Stephan et al. (1996). Dry plant material was dissolved in 65% HNO3 at 170 °C for 3 h in a high pressure block (Pressure digester DAB III; Berghof GmbH, Eningen, Germany), and the samples were measured using an atomic absorption spectrometer SpectrAA 10 Plus (Varian, Inc., Mulgrave, Australia).

Determination of nicotianamine concentration

Plant material was crushed in liquid nitrogen and the powder was mixed with water and homogenized after thawing with a Potter-Elvehjem tissue grinder (Corning Inc., Lindfield, NSW, Australia). The homogenate was stirred for 1 h, deproteinized by heating to 80 °C, centrifuged at 48 000 g and lyophilized. The lyophilizate was dissolved in 0.2 m Na-citrate buffer (pH 3.05) and the NA concentration was determined after post-column derivatization with ninhydrin at 570 nm using an amino acid analyser S432 (Sykam GmbH, Gilching, Germany) as described by Schmidke & Stephan (1995).

Statistical analysis

For the statistical treatment of the data the Kruskall–Wallis one-way analysis of variance of ranks (P = 0.05%) was used.


Elevated nicotianamine concentration in Arabidopsis under iron limitation

Contradicting results have been reported concerning the dependence of the NA concentration on the concentration of externally supplied iron (Higuchi et al. 2001; Pich et al. 2001; Douchkov et al. 2002). Therefore, wild-type Arabidopsis plants were grown in hydroponic culture in the presence of 0, 1, 40 and 100 µm iron. The first two conditions resulted in chlorotic plants. The determination of NA concentrations clearly demonstrated that Arabidopsis responds to limited iron conditions with an increased concentration of NA (Fig. 1 upper two panels). The endogenous iron concentration of young leaves of these plants and the average plant dry weight are given in the two lower panels. The observed increase in NA under limited iron supply conditions led to the expectation that the iron efficiency of plants could be improved by a constitutively increased NA concentration. To achieve this goal various NAS genes were ectopically expressed in transgenic tobacco plants.

Figure 1.

NA concentrations given in nmol g−1 FW in shoots and roots of Arabidopsis thaliana plants grown under different iron concentrations (upper two panels). Fe concentrations (µmol g−1 DW) in shoots of Arabidopsis thaliana grown under different iron concentrations (lower panel left). Average plant dry weight (mg) of Arabidopsis thaliana plants grown under different iron concentrations (lower panel, right). The bars represent the standard error of three independent extractions.

Ectopic expression of a NA-synthase gene results in higher NA concentrations

The AtNAS1 gene (At5g04950) was placed in sense orientation under the control of the CaMV-35S promoter and used for the transformation of tobacco. A total of 40 independent transformants were generated. After determination of the NA concentrations in leaves, three lines with the highest NA concentration were selected (S2, S25, S38) for further physiological experiments. The NA concentrations in young leaves of plants of the T2 generation were determined. All three lines contained an average of about 600 nmol NA g−1 fresh weight (FW), representing an approximately six-fold increase in NA concentration in comparison with wild-type plants (Fig. 2). Analogous experiments based on the genes AtNAS3 (At1g09240) from Arabidopsis and NASHOR1 (Herbik et al. 1999) from barley did not result in an increased NA concentration.

Figure 2.

Average NA concentrations in comparable young leaves (approximately 5 cm length) of 35S::AtNAS1 T2 tobacco plants of lines S2, S25, S38 and wild-type (WT) control plants grown on Hoagland medium supplemented with 1 µm Fe-EDTA. Each column represents the mean value of 10 individual plants per line. The bars represent the standard error for the corresponding data set. Plants were grown Hoagland solution containing 1 µm Fe EDTA (upper panel). Iron concentrations (µmol g−1 DW) in young leaves of T235S::AtNAS1 plants of the lines S2, S25, S38 compared to wild-type (WT) grown under 1 µm Fe EDTA conditions. Each column represents an individual plant (lower panel).

Higher NA concentrations result in an increased iron concentration in leaves

Young leaves of individual plants of the three selected lines (S2, S25, S38) grown under limited iron supply were analysed to test whether the increased NA concentration had an effect on the iron concentration under limited iron supply. A statistical analysis (Kruskall-Wallis one-way analysis of variance of ranks) revealed a significantly higher iron concentration (P = 0.05% level) for lines S2 and S25 (Fig. 2). Line 38 failed in this test; however, the majority of individual plants exhibited higher iron concentrations than the wild-type controls. In addition to iron the concentrations of copper, zinc and manganese were determined. Interestingly, the concentrations of zinc and manganese were found to have significantly higher values in all three transgenic tobacco lines, including line 38. The effect of the increased NA concentrations on copper accumulation was not conclusive (Fig. 3).

Figure 3.

Average zinc, copper and manganese concentrations (µmol g−1 DW) in young leaves of 35S::AtNAS1 T2 plants. Each column represents the mean value of 10 individual plants per line (S2, S25, S38). The bars represent the standard error for the corresponding data set.

Higher NA concentrations leads to increased tolerance against iron deficiency-induced chlorosis

Iron limiting conditions induced characteristic deficiency symptoms including chlorosis. When grown under iron limitation, all three transgenic plant lines exhibited visibly less chlorosis compared to wild type (Fig. 4). This was confirmed by the analysis of the chorophyll content of comparable young leaves of the S2, S25 and S38 transformants. They contained, respectively, 50, 100 and 150% higher chlorophyll concentrations compared with controls (Fig. 4).

Figure 4.

Phenotypes of representative plants of the lines S2, S25, S38 compared to wild type. The transgenic lines with an increased NA concentration exhibit a less serious chlorosis than the wild type plants under iron limitation (1 µm Fe EDTA) (upper panel). Average chlorophyll concentrations in young leaves of 35S::AtNAS1 T2 tobacco plants grown under iron limitation (1 µm Fe EDTA). Each column represents the mean value of 10 individuals per line (S2, S25, S38) and wild type (WT). The bars represent the standard deviation for the corresponding data set (lower panel).

Functional correlation between transgene transcript, NA and iron concentrations

To unequivocally demonstrate that the increased NA concentration is the direct result of the ectopic AtNAS1 gene expression, segregating T2 sister plants of the three transgenic lines were analysed for the presence of the transgene transcript, NA and iron. As shown in Fig. 5, there was a clear correlation between the presence or absence of the transcript and wild-type levels of NA or five- to 10-fold increased NA concentrations. A parallel analysis of the iron concentration further corroborated that, as a consequence of this, the ectopic expression of the AtNAS1 gene of Arabidopsis in tobacco resulted in a significant increase of the iron concentration.

Figure 5.

Nicotianamine concentrations (upper panel), AtNAS1 northern signal (middle panel) and iron concentration (lower panel) of segregating 35S::AtNAS1 T2 plants. The numbers indicate the individual plants of the progeny of the three lines S2, S25, S38 grown on Hoagland medium supplemented with 1 µm Fe-EDTA. The presence of the Northern signal correlates with the NA concentration. The group of transgenic plants contain a significantly higher iron concentration.

Enhanced heavy metal tolerance of overexpressor lines

The potential influence of the higher NA concentration on plant growth in the presence of heavy metals was tested on media supplemented by cadmium and nickel and with elevated concentrations of copper and manganese. No clear difference between overexpressing lines and wild-type plants could be found for copper, manganese and cadmium. However, a remarkably strong protective effect against nickel toxicity was observed. When segregating sister plants of the three AtNAS1 overexpressing lines were grown on media containing 600 and 1000 µm nickel chloride, the transgenic plants tolerated nickel concentrations that were toxic to wild-type plants and to plants of segregating wild-type offspring as shown for lines S25 and S2 (Fig. 6). Similar results were obtained for line S38. This observation was confirmed by the three- to seven-fold higher chlorophyll contents of the three transgenic lines in comparison with wild type in the presence of 1 m m nickel. Surprisingly, the endogenous nickel content of these plants was not greatly changed according to atomic absorption analysis (Table 1), suggesting that the plants are protected from nickel toxicity by the limitation of uptake. The iron concentrations were not greatly different between controls and the overexpressing lines under these conditions.

Figure 6.

Growth of 35S::AtNAS1 T2 tobacco plants on MS medium containing phytotoxic concentrations of nickel. Plants of the T2 generation of lines S25 and S2 were grown on MS-medium containing 600 µm (left panels) and 1000 µm (right panels) nickel chloride for 10 d. Transgenic plants (OX) and segregating wild-type (WT)-like sister plants (WT) are compared. Similar results were obtained with segregating plants of line S38.

Table 1.  Comparison of nickel concentrations of 35S::AtNAS1 tobacco plants (lines S2, S25, S38) with segregating sister plants (WT) grown on MS medium agar
Line600 µm Ni2+ medium
(µmol NI2+ g−1 DW)
100 µm NI2+ medium
(µmol NI2+ g−1 DW)
  1. Seeds were germinated on the nickel -containing media (600 and 1000 µm NiCl2) and the nickel concentration was determined in 4-week-old seedlings. Care was taken not to contaminate the sample with traces of the medium. The nickel concentrations in plants are given in µmol g−1 DW (three independent measurements).

WT102 ± 5138 ± 6
S2 78 ± 4153 ± 5
S25 80 ± 6168 ± 5
S38 57 ± 4158 ± 7


Increased NA concentration counteracts iron limitation in wild-type plants

Contradictory data have been reported about the influence of the iron status on the NA level. In roots of strategy II plants, the activity of nicotianamine synthase is increased under iron deficiency, leading to the enhanced synthesis of phytosiderophores (Higuchi et al. 2001). In barley the NASHOR1 gene transcript level exhibited both root specificity and inducibility under iron limitation (Douchkov 2003). In contrast, Pich et al. (2001) described a positive correlation between the NA concentration and the endogenous iron level in tomato. Since the Fe-NA complexes are poor Fenton reagents (Von Wirén et al. 1999), these authors suggested a protective function of high NA levels against oxidative damage. The data presented here for Arabidopsis demonstrated an inverse correlation of the NA concentration both in shoots and roots with the exogenous iron concentration. The most obvious interpretation would be that the plant reacts with an increased NA concentration to counteract limited iron supply conditions. The observed elevation of the NA concentration could be the consequence of altered synthesis or degradation rates, but in any case it seemed to be of interest to test the physiological consequences of an artificially increased endogenous NA.

Ectopic expression of NA -genes leads to an increased endogenous NA level

An exogenous supply of NA to iron-starved sunflower plants leads to re-greening and better growth, probably due to the optimization of iron uptake and/or distribution within the plant (Scholz et al. 1988a). Together with the data mentioned in the previous section, this observation led to the assumption that an increase of the endogenous NA level could result in plants with an improved higher iron efficiency. To reach this goal, several different NAS-genes were used for ectopic expression in transgenic tobacco, including the NASHOR1 gene of barley (Herbik et al. 1999) and the Arabidopsis genes AtNAS3 (At1g09240) and AtNAS1 (At5g04950). Although the monocot NASHOR1 gene was chosen to avoid co-suppression effects in the dicot tobacco, no increase of the NA concentration could be detected as a consequence of NASHOR1 gene expression (Douchkov 2003). This is in contrast to recently reported data (Takahashi et al. 2003). These authors described a two- to three-fold increase of iron concentration in tobacco plants that express the barley gene HvNAS1, another member of the gene family, although NA concentrations in these lines have not been reported. Similar to the experiments with the barley NASHOR1 gene no significant increase of the NA concentration could be observed by overexpression of the AtNAS3-gene, although more than 10 independent transgenic tobacco lines were analysed (Douchkov 2003). The reason for the failure of both of these constructs is not yet known, but effects on message stability and translation efficiency could be responsible. It should also be considered that the enzymatic properties of the encoded NAS proteins have not been analysed yet. In contrast to these results, a strong increase in the NA concentration could be observed by the overexpression of the gene AtNAS1. Concentrations of up to 1300 nmol NA g−1 FW were measured in lines with the highest expression of the AtNAS1 gene. These are among the highest NA concentrations ever reported in plants (Stephan & Rudolph 1984). The three lines (S2, S25, S38) with the highest NA concentrations were selected for further experiments. To exclude any effects of the transformation procedure, segregating sister plants of the T2 generation of all three lines were analysed in parallel for the presence and expression of the transgene and the level of NA. Based on the clear correlation between transcript and NA levels, it was unequivocally shown that the expression of the AtNAS1 gene in tobacco leads to a five- to 10-fold increase in the endogenous NA concentration. The NA levels of the overexpressing plants can be roughly grouped into two classes. Further segregation analysis revealed that this is a gene copy effect. Plants containing the highest NA level were homozygous for the transgene and plants with intermediary levels were heterozygous as analysed by the segregation of the kanamycin resistance (not shown).

Improved metal nutrient content and iron efficiency of plants with increased NA levels

The observation of improved growth after exogenous supply of NA to iron-starved plants (Scholz et al. 1988a) gave rise to the hypothesis that an increase of the endogenous NA concentration should lead to changes in iron metabolism. Sister plants with or without the 35S::AtNAS1 transgene were analysed for their iron and metal micronutrient contents. The positive effect on concentrations of iron, zinc and manganese was small, but significant, whereas copper concentrations were not affected. This indicates that NA interferes either directly by chelation or indirectly by regulation of iron, zinc and manganese homeostasis. In contrast, the control of intracellular copper concentrations appears to be independent of NA.

A beneficial effect of NA on iron metabolism was further supported by comparisons of the transgenic plants and segregating wild types. The transgenic plants performed better and were visibly greener under iron-deficient growth conditions. The visual observation was further confirmed by the higher tolerance to iron limitation-induced chlorosis as measured by a significantly higher chlorophyll content of the transgenic plants. The increase of the endogenous NA concentration is accompanied by a significant increase of the iron concentration (lines S2 and S25). Thus, in accordance with physiological observations (Stephan & Scholz 1993), the improved growth and chlorosis tolerance of the transgenic plant lines obviously results from the increased NA concentration.

The increase in iron concentration in the leaves by a factor of about 1.5-fold due to the ectopic expression of the AtNAS1 gene was in the range observed by other approaches using FRE, ferritin and phytosiderophore synthesis genes (Samuelsen et al. 1998; Van Wuytswinkel et al. 1999; Goto et al. 1999; Vansuyt et al. 2000; Takahashi et al. 2001, 2003; Vasconcelos et al. 2003). It is tempting to speculate that a combination of NAS overexpression with one of these independent mechanisms could lead to a synergistic enhancement of iron concentrations.

Taken together, the data allow the conclusion that the ectopic expression of the AtNAS1 gene of Arabidopsis in tobacco results in a significant increase in the concentration of NA and, as a consequence of this, also in the concentration of iron and some other mineral nutrients in leaves. Thus, in addition to the efforts described above, this approach provides an interesting alternative to increase the iron efficiency of plants.

Increased tolerance against toxic nickel concentrations

The effect of elevated NA concentrations on the iron, zinc and manganese levels was significant but moderate. However, the transgenic plants clearly outperformed the segregating sister plants in terms of nickel tolerance. Phytotoxic concentrations of up to 1 m m nickel were tolerated by all three transgenic lines tested.

Nickel is chemically related to iron and cobalt. Plants cannot complete their life cycle without an adequate nickel supply. Nickel is the metal component required for the activity of several enzymes, where it is co-ordinated either to N- and O-ligands as in urease, S-ligands as in hydrogenases or to N-ligands of tetrapyrrol structures. Nickel toxicity is of concern in relation to the application of sewage sludge that is often high in nickel (for an overview see Marschner 1995). Nickel is a potent environmental pollutant in industrialized countries and has recently been found to be involved in a hypoxia-inducible nickel carcinogenesis pathway (Salnikow et al. 2003).

Nickel tolerance and the capability to hyperaccumulate this metal has been shown to be correlated with the enhanced production of histidine in the hyperaccumulating crucifere Alyssum (Krämer et al. 1996, 2000) but not in Thlaspi goesingense (Persans et al. 1999; Persans, Nieman & Salt 2001). In the latter species the majority of leaf nickel was associated with the cell wall, while the remaining nickel was associated with citrate and histidine, primarily localized in the vacuolar and cytoplasm, respectively (Krämer et al. 2000; Küpper et al. 2001). Furthermore, enhanced chelation of Ni(II) by NA in the xylem as response to toxic levels of external nickel has been suggested as the physiological basis of nickel tolerance in the hyperaccumulator Thlaspi caerulescens (Vacchina et al. 2003).

The NA-dependent tolerance to high nickel concentrations described here obviously follows another pathway. Surprisingly, no great change of the endogenous nickel concentration could be found. At a concentration of 600 µm nickel in the medium, the endogenous nickel concentration was slightly reduced, leading to the preliminary conclusion that the excessive external nickel was excluded from uptake. No clear difference of the endogenous nickel concentration could be detected under the exogenous nickel concentration of 1000 µm, suggesting that the effects of this treatment are more likely due to a decrease in the toxicity of the metal in vivo, possibly through chelation.

One possible explanation for the unchanged concentration of nickel in the transgenic lines could be related to the implication of NA in cell-to-cell transport of iron and possibly other micronutrients. Yellow-stripe (YS)-like transporters were originally discovered as iron-phytosiderophore transporters in roots of maize (Curie et al. 2001). The expression of YS-like genes in shoots of strategy 1 and 2 plants led to the assumption that the structurally related iron-NA complexes also could be transported. This was recently demonstrated for YS-proteins from maize (Schaaf et al. 2004; Roberts et al. 2004). The 35S CaMV promoter used in this study most likely led to elevated NA concentrations also in roots. The excess NA might have caused preferential loading of the YS-like transporters by its natural micronutrient substrates and exclusion of nickel.

Moreover, the increased endogenous NA concentration should result in an increased cellular pool of the Fe-NA complex, which is the only soluble form of Fe2+ in the weakly alkaline environment of the cytoplasm. As part of this complex iron is available for metabolic processes including the interaction with a still hypothetical cellular iron sensor. Its saturation could cause the down-regulation of iron uptake processes such as rhizodermal reductase activity and proton extrusion and finally lead to the prevention of the accumulation of nickel to toxic concentrations. Minor amounts nevertheless leaking in due to the high external nickel supply might be detoxified internally as NA complex (Schaaf et al. 2004).

Yet another possible explanation is based on the stability of different metal–NA-complexes. The order of affinity follows the Irving-Williams rule: Cu2+ > Ni2+ > Co2+ > Zn2+ > Fe2+ > Mn2+ (Scholz et al. 1988a; Stephan & Scholz 1993). It is therefore tempting to suggest that heavy metals which form NA complexes with higher stability than that with Fe2+ could displace the iron from the NA complex causing iron deficiency. This effect could be overcome by the five- to 10-fold higher NA levels in the AtNAS1 gene overexpressing plants.

The importance of nicotianamine for heavy metal metabolism has recently been demonstrated by micro-array analysis. Comparative analysis of gene expression in roots of Arabidopsis thaliana and Arabidopsis halleri identified nicotianamine as a potential metal hyperaccumulation factor. These studies detected nicotianamine synthase genes among the most abundantly transcribed messages in the hyperaccumlator A. halleri (Weber et al. 2004; Becher et al. 2004).

The observed effect of elevated NA concentrations for protection against otherwise toxic nickel concentrations clearly needs further investigations but may open an unexpected new connection between metal homeostasis and the regulation of micronutrient uptake.


The work was supported by grants from the Deutsche Forschungsgemeinschaft (STE 594/4-3). We would like to acknowledge the dedicated assistance of Elke Liemann with plant transformation and Wally Wendt with metal and NA analyses.