How do some plants tolerate low levels of soil zinc? Mechanisms of zinc efficiency in crop plants

Authors

  • Gökhan Hacisalihoglu,

    1. US Plant, Soil and Nutrition Laboratory, US Department of Agriculture, Agricultural Research Service, Cornell University, Ithaca, New York 14853, USA
    Search for more papers by this author
  • Leon V. Kochian

    Corresponding author
    1. US Plant, Soil and Nutrition Laboratory, US Department of Agriculture, Agricultural Research Service, Cornell University, Ithaca, New York 14853, USA
      Author for correspondence: Leon V. Kochian Tel: +1 607 255 2454 Fax: +1 607 255 2459 Email: LVK1@cornell.edu
    Search for more papers by this author

Author for correspondence: Leon V. Kochian Tel: +1 607 255 2454 Fax: +1 607 255 2459 Email: LVK1@cornell.edu

Summary

Researchers are beginning to understand how some plant genotypes can maintain reasonable growth and yields under low soil Zn, a trait termed zinc efficiency (ZE). Several studies have shown no correlation between ZE and root Zn uptake, Zn translocation to shoot, or shoot Zn accumulation. Furthermore, it has not been possible to conclusively link differences in leaf subcellular Zn compartmentation with ZE. However, biochemical Zn utilization, including the ability to maintain the activity of Zn requiring enzymes in response to Zn deficiency may be a key component of ZE. The next logical step in investigations of this trait is research on the genetic and molecular basis for ZE, in order to better understand the relationship between Zn utilization and ZE, and to identify the gene(s) controlling ZE. Progress in this research area could provide the knowledge to facilitate the engineering of Zn-efficient plant varieties, which could help both crop production on marginal soils as well as possibly improve the micronutrient density of food crops to help address significant human nutrition problems related to micronutrient deficiency.

Introduction

Certainly one of the most important adaptive responses for crop plants involves their ability to deal with soil-mediated abiotic stresses involving either deficient or excessive levels of mineral and metal ions in the soil. Within the broad category of mineral-related abiotic stresses, zinc (Zn) deficiency is one of the most widespread limiting factors to crop production, affecting 30% of the world's soils, including many agricultural lands in Australia, South-east Asia, Turkey and the USA (Sillanpaa, 1990, Fig. 1). Zn deficiency is common in a wide range of soil types, including high pH calcareous soils, sandy soils, and high phosphorus-containing fertilized soils (Marschner, 1995). Many plant species such as bean, maize, wheat, rice and tomatoes are considered less tolerant to Zn deficiency stress and exhibit significant yield losses due to Zn deficiency compared to more tolerant species such as peas, carrots, and rye (Chapman, 1966).

Figure 1.

Geographic distribution of severe- (red areas) and moderate- (green areas) Zn deficient soils in the world. Adapted with permission from Alloway (2001).

Correction of Zn deficiency via fertilization is not always successful due to agronomic and economic factors. Some of those factors include reduced availability of Zn due to topsoil drying, subsoil constraints, disease interactions, and cost of fertilizer in developing countries (Graham & Rengel, 1993). As a consequence of coping with low Zn availability, certain plant genotypes are able to grow and yield well under Zn deficiency, which has been termed Zn efficiency (ZE) (Graham & Rengel, 1993). ZE is genetically based and the physiological and molecular mechanisms underlying ZE are just beginning to be understood.

Zn deficiency stress and the general topic of mineral nutrient efficiency have been the subject of several other reviews to which the reader is referred (Graham & Rengel, 1993; Kochian, 1993; Welch, 1995; Rengel, 1999; Cakmak, 2000). This review will summarize our perception of the current understanding of the physiological mechanisms underlying Zn efficiency (ZE) in different plant systems, and will also look at the prospects for future research utilizing molecular and genetic approaches to improve plant ZE.

Overview of zinc deficiency stress and zinc efficiency in plants

Soil Zn as an essential micronutrient and plant responses to Zn deficiency

Zinc (Zn) is an element required by virtually all organisms as it is a critical component of many enzymes and proteins (Marschner, 1995). Both the total Zn concentration and the free Zn2+ activity can vary greatly in soil solution, depending on a number of soil factors (e.g. pH, soil moisture, organic matter content, etc.). In a calcareous soil, the free Zn2+ activity might be as low as 10−9–10−11 m, which can be too low to support optimal crop growth (Barber, 1984; Kochian, 1993; Welch, 1995). Therefore, the low Zn availability on these types of soils limits crop production in many countries, including Australia, India, Turkey and the USA (Sillanpaa, 1990).

Characteristic symptoms of Zn deficiency includes chlorosis on young leaves, reduced leaf size (i.e. little leaf), and stunted, thin stems (Fig. 2a). Under severe Zn deficiency, older leaves show wilting and curling with extensive chlorosis and stunted growth (Marschner, 1995).

Figure 2.

Zn deficiency stress and screening for Zn-efficiency in common bean. Several bean lines were grown in chelate buffer nutrient solution for 21 days under controlled environmental conditions. (a) bean lines grown on nutrient solution with low Zn2+ (1 pm) activity; (b) same bean lines grown on sufficient Zn (150 pm Zn2+ activity; and (c) a Zn-inefficient and a Zn-efficient bean line grown on low Zn2+ (1 pm). Note severe Zn deficiency symptoms (e.g. stunted plants, smaller leaves and chlorosis) in the Zn-inefficient line on the left.

Genotypic variation in Zn efficiency

The application of Zn fertilizers is not a totally successful strategy in alleviating Zn deficiency because of agronomic (i.e. subsoil constraints, disease interactions), economic (i.e. unavailability of Zn fertilizers for poor farmers), and environmental (i.e. pollution associated with excessive fertilizer use) factors (Graham & Rengel, 1993; Hacisalihoglu, 2002). A more efficient and sustainable solution to Zn deficiency limitations to crop production is the development and use of Zn-efficient plant genotypes that can more effectively function under low soil Zn conditions, which would reduce fertilizer inputs and protect the environment as well. It has been well documented that certain plant species, as well as genotypes within certain species, exhibit a significant genetic-based variation in their tolerance to Zn deficiency. This is depicted in Fig. 2, which shows differences in the severity of Zn deficiency symptoms between zinc efficient and inefficient cultivars of common bean (Phaseolus vulgaris). As mentioned previously, the precise mechanisms underlying ZE are not yet clear but a number of studies have demonstrated differential ZE in several crop species, including bean (Ambler & Brown, 1969; Hacisalihoglu, 2002; Hacisalihoglu et al., 2003b), wheat (Graham & Rengel, 1993; Cakmak et al., 1997a), rice, and tomato (Bowen, 1987).

Although differential ZE between genotypes of the same plant species has been widely reported, limited information is available on the genetics of zinc efficiency. In soybean it has been suggested that a few genes may be responsible for ZE, based on measurements of foliar Zn concentrations (Hartwig et al., 1991). In a study using the bean genotypes Matterhorn (Zn-efficient) and T-39 (Zn-inefficient), Singh & Westermann (2002) reported that a single dominant gene controlled resistance to Zn deficiency. Recently, the genetics of ZE in hydroponically grown beans was investigated and the results suggested that ZE in bean was heritable and was also conferred by a single dominant gene (Hacisalihoglu, Hoekenga & Kochian, unpublished results). Both of these studies in bean used visual Zn deficiency symptoms in the shoot to score the ZE phenotype. These limited number of studies, which are all fairly narrow in scope (assessing small segregating populations), have yielded some preliminary evidence regarding the genetics of ZE in a few crop species. Given that the focus of much of the ZE research has been on physiological studies on cereal ZE, future investigations of the genetics of ZE in cereals as well as studies integrating genetic and physiological analyses of ZE mechanisms will certainly be useful. Furthermore, given the breadth and depth of genetic resources in the model plant species Arabidopsis thaliana, genetic analyses of ZE using either recombinant inbred lines or mutagenized populations of Arabidopsis certainly could yield important new information on this trait.

Physiological analysis of ZE and proposed mechanisms

In recent years, based on studies from several laboratories, we are just beginning to gain an understanding of the physiology of Zn efficiency mechanisms in plants. A variety of research approaches involving whole plant and cellular physiology, biochemistry and genetics have been used to study how plants cope with low available Zn. As depicted in Fig. 3, there are several key sites within the crop plant that could be involved in zinc efficiency. These include: root processes that increase the bioavailability of soil Zn for root uptake; enhanced root uptake and translocation of Zn from the root to the shoot; altered subcellular compartmentation of Zn in shoot cells, such that more Zn is maintained in the cytoplasm; and enhanced or more efficient biochemical utilization of Zn in cells of the shoot. All four of these possible sites for control of ZE have been investigated to varying degrees in research from a number of different laboratories. In the following section, the research on each of these possible regulatory sites for crop ZE will be discussed and the published findings evaluated.

Figure 3.

An overview of potential Zn efficiency mechanisms in plant species such as bean and wheat.

Root-localized zinc efficiency mechanisms

Root-based alterations in soil Zn availability Many of the earlier studies on ZE focused, quite naturally, on comparisons of root systems in efficient and inefficient genotypes of a specific plant species (often wheat) with regards to their ability to ‘mine’ Zn from the soil by increasing Zn availability in the rhizosphere (see for example, Dong et al., 1995; Cakmak et al., 1996a; Rengel et al., 1998). Several chemical (soil pH, redox potential or nutrient interactions), physical (organic matter content, soil texture, or clay content and type) and biological factors (mycorrhizae formation, phytosiderophore release) operating at the root–soil interface may affect availability and absorption of zinc and subsequently, zinc efficiency.

Root architecture varies among plant species and cultivars within plants species and has been implicated in influencing plant Zn availability and zinc efficiency (Dong et al., 1995). As a consequence, thinner roots with increased root surface area may increase the availability of Zn along with other nutrients due to a more thorough exploration of the soil (Rengel & Graham, 1995). Furthermore, roots of some plant species are colonized by mycorrhizal fungi that increase the uptake of diffusion-limited nutrients from soil including phosphorus, zinc and copper. In particular, vesicular-arbuscular mycorrhizae (VAM) have been cited to benefit plants by expanding the volume of soil explored by root systems and consequently increasing Zn uptake (Kothari et al., 1991). However, as genetic differences in ZE found in the field also can be seen in hydroponically grown plants that form no mycorrhizal associations, the role and importance of mycorrhizae in ZE is unclear.

With regards to root-mediated alterations in rhizosphere chemistry, this could involve changes in rhizosphere pH or the release of organic ligands that could chelate soil Zn and increase its availability. As soil pH increases, Zn availability for plants decreases and when soil pH rises above pH 5.5, Zn is adsorbed by soil constituents such as hydrous oxides of aluminium, iron and manganese or is precipitated by specific compounds (Marschner, 1995). Therefore, root-mediated processes that can lower the rhizosphere pH increase the plant Zn availability by solubilizing Zn from organic and inorganic soil solid phases. However, there have been no definitive studies implicating root-mediated changes in rhizosphere pH in ZE.

When the role of root exudates in ZE is looked at, there have been a number of studies that have focused on phytosiderophore release. Phytosiderophores (PS) are nonprotein amino acids that chelate a number of micronutrients, including Fe and Zn and are released from the roots of grasses under iron deficiency (Marschner, 1995). It is widely accepted that PS release and root absorption of Fe(III)-PS are key factors in Fe nutrition in grasses. Several researchers have suggested that phytosiderophores also can play a role in grass Zn nutrition and thus, quite possibly, Zn efficiency. With regard to the role of PS in plant Zn nutrition, Zhang et al. (1991) reported that Zn deficient graminaceous species released pytosiderophores and thus increased the mobilization of Zn and Fe in soil. Evidence in support of the possibility that roots of grass species could absorb the Zn-PS complex was presented by von Wiren et al. (1996), who investigated Zn2+ and Zn-PS uptake in roots of the maize yellow-stripe 1 (ys1) mutant, which is defective in Fe-PS uptake. These researchers showed that Zn uptake from Zn-PS was significantly reduced in the ys1 mutant, compared with wild type maize. Thus, it was speculated that two root Zn uptake pathways exist in grasses, one that involves free Zn2+, and a second based on Zn-PS uptake. The literature on the role of root PS release in ZE is both contradictory and somewhat controversial. Cakmak et al. (1996b) reported that Zn-efficient bread wheat genotypes had higher PS release than Zn-inefficient durum wheat genotypes under Zn deficiency. However, in another study, they found that phytosiderophore release did not correlate with differences in zinc efficiency of efficient and inefficient bread wheat genotypes under Zn deficiency conditions (Erenoglu et al., 1996). In a study with bread and durum wheat cultivars, Rengel & Romheld (2000) reported that Zn-efficient bread wheat cultivars released more pytosiderophores than Zn-inefficient cultivars. There have been several other studies that contradict the role of PS in both Zn nutrition and zinc efficiency. Both Gries et al. (1995) and Pedler et al. (2000) found no significant Zn-deficiency induced PS release in barley and wheat cultivars, that had been reported by others, in response to Zn deficiency. Furthermore, in the work by Pedler et al. (2000), they found no differences in PS release in barley and wheat cultivars that have been reported to differ significantly in ZE. They did find that Fe deficiency induced a large PS release in all barley and wheat genotypes studied. These researchers suggested that the previous observations of Zn-deficiency induced PS release may be dependent on the growth methods used, and might be explained by an induced physiological deficiency of Fe, and not Zn. In a recent review on this topic, Rengel (2001) indicated that further work is needed to arrive at any definitive conclusions about the possible role of PS release in ZE. From our point of view, when all of the available literature on the role of root-based changes in soil Zn availability in ZE is considered, it does not appear that these processes play a key role in ZE. Certainly the above mentioned observation that genotypic differences in ZE determined on Zn deficient soils are also seen when the same plants are grown under hydroponic culture where Zn availability is not an issue, argues against an important role for rhizosphere Zn availability in ZE. Although, root exudation does not seem to have a major importance in ZE, it cannot be ruled out, and a final determination on the role of rhizosphere bioavailability of Zn in ZE awaits the results of further study (Rengel, 2001).

Root Zn uptake and translocation to the shoot Zn first enters the root cell wall free space by diffusion and then moves across the plasma membrane via the action of ion transport proteins. Zn is mainly taken up by plant roots as the Zn2+ ion, although in grasses a second pathway involving the uptake of Zn-PS could also occur (Kochian, 1993; von Wiren et al., 1996). It appears that root Zn2+ uptake is mediated by two different transport systems: a high-velocity, low-affinity system (Km = 2–5 µm), and a low velocity, high affinity system (Km = 0.6–2 nm) that probably is the dominant transport system under low soil Zn conditions (Hacisalihoglu et al., 2001; Hacisalihoglu, 2002).

There are at least six different families of transporters that could facilitate plant Zn transport, including ZIP (Zrt- and Irt-related proteins), CDF (Cation Diffusion Facilitator proteins), P-type ATPase (metal transporting ATPases), NRAMP (natural resistance-associated macrophage proteins), and CAX (calcium and other divalent cation exchange antiporters) proteins. Recent studies suggest that the ZIP family of micronutrient transporters probably include some of the important Zn transporters that facilitate Zn entry into plant cells. Collectively, over 85 ZIP family members have been identified in different organisms, including 18 different ZIP genes in Arabidopsis; other members of this gene family have been identified in rice, the metal-accumulator plant Thlaspi caerulescens, animals, insects, fungi, and archae (Guerinot, 2000). In Arabidopsis, it is believed that the ZIP1-4 genes are involved in plant Zn transport but it is not known which ZIP transporters actually mediate root Zn uptake from the soil (Guerinot, 2000).

The long distance transport of Zn is known to occur primarily via the flux of Zn2+ ions in the xylem stream; Zn concentrations in xylem sap have been shown to range from 2 to nearly 100 µm (Kochian, 1993; Marschner, 1995). Leaf Zn concentrations below 10–15 µg Zn g−1 dry weight are considered Zn deficient; whereas 20–100 µg Zn g−1 leaf dry weight is sufficient for normal plant growth (Marschner, 1995). Under low Zn conditions, zinc is not usually effectively mobilized from the older leaves, and because of this, it is generally believed that the phloem mobility of Zn is fairly low. This has been suggested to explain why zinc deficiency symptoms appear primarily on young leaves (Marschner, 1995). However, a recent study in wheat has suggested that under certain conditions, Zn may be able to move fairly readily in the phloem (Haslett et al., 2001).

With regards to Zn translocation to the shoot, the relevant literature on this topic suggests that this transport process also does not correlate with Zn efficiency. In a field study of 37 bread wheat and 3 durum wheat cultivars differing in ZE, no significant differences in shoot Zn concentration were found for the different cultivars grown on either Zn-deficient or -sufficient soils (Kalayci et al., 1999). In a recent study from our laboratory, where 65Zn translocation was directly measured in both wheat and bean cultivars differing in ZE, again no correlation between Zn translocation to the shoot and differential ZE was found (Hacisalihoglu, 2002; Hacisalihoglu et al., 2003a).

Shoot-localized Zn efficiency mechanisms

As detailed in the root-localized mechanisms section, it does not appear that there is strong evidence to implicate root-mediated processes in crop ZE. This contention is supported by the observation, often made in the literature, that when efficient and inefficient genotypes are grown in Zn deficient soil or hydroponic media, no significant differences in shoot Zn concentration are seen, even when dramatic differences in Zn deficiency symptoms are seen in contrasting genotypes (for example, see Cakmak et al., 1999; Kalayci et al., 1999; Torun et al., 2000; Hacisalihoglu et al., 2003a). These findings suggest that the transport of Zn from the soil to the root and then to the shoot, are not the most important factors in differential ZE. The obvious implication from these observations is that crop ZE must be a shoot-mediated trait. Evidence in support of this comes from recent grafting experiments in our laboratory with efficient and inefficient bean cultivars. Only when the shoot of the efficient bean cultivar was grafted on the root system from either the inefficient or efficient cultivars was ZE observed (unpublished results). The two most likely possibilities for a shoot-mediated mechanism for ZE are: alterations in subcellular Zn compartmentation and homeostasis such that the efficient genotypes maintain higher levels of Zn in the cytoplasm of leaf cells; and more efficient biochemical utilization of cellular Zn, such that Zn-requiring macromolecules can efficiently incorporate Zn as a cofactor under low Zn conditions.

Subcellular Zn compartmentation and homeostasis All organisms have homeostatic mechanisms to maintain the required concentrations of essential nutrients for optimal biochemical and physiological functioning. It is clear that cellular Zn homeostasis is quite complex and must be controlled by a combination of transport, chelation, trafficking and sequestration (Clemens, 2001). The study of subcellular Zn compartmentation is not a trivial technical undertaking and thus limited information exists concerning this topic.

One of the ways to study mineral ion compartmentation in plant cells is via radiotracer efflux analysis or compartmental analysis. This approach has been a widely used technique to approximate plant subcellular ion fluxes, membrane kinetic parameters, and ionic concentrations within subcellular compartments. Santa Maria & Cogliatti (1988) found that 65Zn efflux in wheat roots could be fitted to a triple exponential function and c. 8% of the root Zn was hypothesized to be in the cytoplasm, 76% in the vacuole, with the rest in the cell wall. When this approach was applied to leaf tissue from Zn-efficient and -inefficient wheat cultivars grown under low Zn conditions, leaf compartmentation studies showed that both the efficient and inefficient genotypes had similar Zn content for the vacuole (83–85%) and cytoplasm (9–11%) (Hacisalihoglu et al., 2003a). The results from this single study suggest that subcellular Zn compartmentation may not play a role in ZE, but clearly more work is needed in this area.

It is possible that insights into this topic may come from recent work in bacteria and other organisms. In bacteria, there are very low levels of free Zn2+ in the cytosol as Zn is always held tightly by organic ligands and its trafficking is controlled by chelators, chaperones and proteins that can effectively bind free Zn2+ at femtomolar (10−15m) activities (Outten & O’Halloran, 2001). In view of the fact that the cellular activity of free Zn2+ is very low, the question of how zinc gets distributed in the cell becomes a very important issue. Metallothionein is a cellular protein reservoir for zinc and could deliver zinc to sites where it is required. Therefore, metallothioneins could be a chaperone, functioning as a zinc donor or acceptor (Maret & Vallee, 1998). Another possible component of Zn homeostasis could involve regulation of the expression of genes associated with Zn nutrition by changes in cellular Zn status. This could be controlled collectively by a variety of response elements in the promoters of specific genes (DeMoor & Koropatnick, 2000). For Zn homeostasis, metal responsive elements (MRE) could be involved, which are cis-acting DNA elements involved in the control of gene expression in relation to changes in plant metal status. In mammals, the metal response element-binding transcription factor-1 (MTF1) mediates the regulation of several genes involved in Zn homeostasis, including responses to both Zn deficiency and toxicity in eukaryotes. MTF1 has been suggested to be involved in the regulation of the free Zn concentration in the cell by allowing zinc to bind to metal responsive elements (MRE) and initiate metallothionein gene transcription (Andrews, 2001). Furthermore, Shuy et al. (1999) demonstrated that MTs are a part of the Zn-scavenging mechanism for cell survival under extreme Zn deficiency in mouse cells. In a study with mouse cells treated with Cd or Zn, the researchers examined the subcellular distribution and DNA binding activity of MTF1. They found that when cellular Zn levels were increased, this was associated with the rapid translocation of MTF1 from the cytosol to the nucleus, where it presumably functions to activate the expression of genes involved in Zn homeostasis (Smirnova et al., 2000).

The homeostasis, regulation and trafficking of cellular Zn are still very poorly understood aspects of Zn nutrition and homeostasis. There are clearly a number of proteins which appear to have the capacity to bind Zn with a high affinity and which may play a role in Zn homeostasis and trafficking. A number of these proteins also appear to be involved in the regulation of the expression of genes involved in Zn metabolism (Berg, 1990). A potentially key player in this scenario could be a plant homologue of MTF1, which regulates the transcription of MT genes in eukaryotes and appears to play an important role in the co-ordination of cellular Zn homeostasis. At present we know much less about the molecular mechanisms of these Zn sensors and there is a clear need for future research in this field, especially in relation to Zn homeostasis and efficiency in plants.

Biochemical utilization of Zn The critical Zn deficiency level in leaves is considered to be 15–20 µg Zn g−1 dry weight (Marschner, 1995). Given the fact that Zn-efficient and Zn-inefficient plants have similar Zn concentrations in their leaves under conditions where only the Zn-inefficient plants exhibit severe Zn deficiency symptoms, internal utilization of Zn has to be considered as an important potential Zn efficiency mechanism (Rengel & Graham, 1995). Zn is an essential component of more than 300 enzymes, thus, the biochemical processes involved in associating cellular Zn with these metalloenzymes could be part of the ZE trait. Despite the large number of Zn requiring enzymes in eukaryotes, only a few of these enzymes have been studied in relation to ZE. Carbonic anhydrase (CA) is a catalytic Zn requiring metalloenzyme that reversibly catalyses the hydration of CO2 to yield HCO3. Rengel (1995) has suggested that Zn-efficient wheat genotypes could maintain a somewhat higher CA activity than in Zn-inefficient genotypes under Zn deficiency. From these findings, the authors suggested that Zn-efficient wheat genotypes benefited from higher CA levels that could help maintain higher photosynthesis rates and dry matter production. It was shown in rice that leaf carbonic anhydrase activity decreased with Zn deficiency, and this led the authors to suggest that CA assists in maintaining significant levels of photosynthetic activity under Zn deficiency by mediating the movement of CO2 between stomata and the CO2 fixation sites in the mesophyll chloroplasts (Sasaki et al., 1998).

In a systematic investigation of the role of root and shoot processes in bean and wheat ZE, Hacisalihoglu and colleagues (Hacisalihoglu et al., 2001, 2003a, 2003b; Hacisalihoglu, 2002) found that of all the root and shoot localized processes studied (including root Zn influx, Zn translocation to the shoot, shoot Zn compartmentation), the only one that correlated with ZE was the activity of several shoot-localized Zn-requiring enzymes. That is, when bean and wheat genotypes differing in ZE were grown under low Zn conditions, which caused significant Zn deficiency symptoms only in the inefficient cultivars, significantly higher levels of the Zn requiring enzymes carbonic anhydrase and Zn/Cu superoxide dismutase were found in the efficient wheat and bean genotypes. These data, together with the earlier findings by other researchers, provide some circumstantial evidence in support of biochemical Zn utilization as a possible factor in ZE.

Zn deficiency may also inhibit the activities of Zn requiring antioxidant enzymes and cause necrotic symptoms due to reduced scavenging of the active oxygen free radical species. Cu/Zn superoxide dismutase (SOD) is a 32-kDa Zn requiring enzyme that detoxifies superoxide radicals and is located in the stroma of chloroplasts (Marschner, 1995). In the work by Hacisalihoglu et al. (2003a) described above, Cu/ZnSOD was used to represent a large class of Zn requiring enzymes, in studies on biochemical Zn utilization. However, it is possible that this enzyme also plays a direct role in tolerance to Zn deficiency stress. Zn deficiency may be one of a number of abiotic stresses that cause increased production of active oxygen species and/or decreased activity of antioxidative enzymes such as Cu/ZnSOD. Cakmak et al. (1997b) studied different rye and wheat cultivars and showed that Zn efficiency was positively correlated with the activity of Cu/ZnSOD. Recent work with black gram (Vigna mungo) suggests that activities of Cu/ZnSOD and CA are positively correlated with Zn supply and could be used as indicators of Zn deficiency (Pandey et al., 2002). On the molecular level, it has been found that Cu/ZnSOD gene expression was higher in Zn-efficient wheat genotypes compared with inefficient ones grown under low-Zn conditions (Hacisalihoglu et al., 2003a). The same study also demonstrated that Zn-efficient wheat genotypes maintained significantly higher Cu/ZnSOD enzyme activity. These findings allow one to speculate that alterations in the Zn-dependent regulation of the expression of key Zn requiring enzymes such as Cu/ZnSOD could play a role in ZE. This would be a distinctly different mechanism than one based on the biochemical utilization of Zn with regards to its role as a key cofactor for Zn-requiring proteins.

In summary, considerably more work is needed before definitive mechanisms for ZE can be described. Based on the published information that is currently available, which appears to indicate that ZE is quite possibly a shoot-localized trait, we are suggesting that biochemical Zn utilization could be a key player in the mechanistic basis for crop ZE (Rengel, 1995, 1999; Cakmak, 2000; Hacisalihoglu et al., 2003a). Furthermore, it is possible that the regulation of expression of genes encoding Zn-requiring enzymes could also be a component of ZE mechanisms in wheat. These findings suggest that a currently unidentified mechanism could be operating in Zn-efficient genotypes to allow them to more effectively utilize low levels of cytoplasmic Zn for biochemical processes such as enzyme synthesis and activation under conditions of plant growth on low Zn soils. Future studies should focus on possible mechanisms by which Zn is associated with key enzymes to facilitate enzyme activation.

Conclusions and future perspectives

Zn deficiency stress in plants has been studied for more than 40 years and based on research findings from the past decade, this field has reached an interesting stage. A number of recent findings have begun to shed light on the physiological basis of ZE. Potential mechanisms including root Zn uptake and translocation, Zn sequestration in leaves, and biochemical utilization of Zn have been studied; however, a multitude of questions concerning ZE mechanisms still remain. The current state of this field does not point to any single simple mechanism for plant Zn efficiency. Root Zn uptake has often been considered one of the key components of ZE; however, to date, a consistent correlation between root Zn uptake or other root processes and Zn efficiency has not been observed in the literature. Several studies have suggested that ZE may be a shoot-localized trait. With regards to candidate mechanisms for shoot-based ZE, a positive correlation between biochemical utilization involving Zn requiring enzymes and ZE has been observed in wheat, bean and black gram. Based on studies that to date have only focused on two Zn requiring enzymes (carbonic anhydrase and Cu/ZnSOD), Zn-efficient genotypes appear to maintain higher activities of these key Zn-requiring enzymes when grown under low-Zn conditions. It is possible that Zn-efficient genotypes utilize Zn more efficiently through enzymes like Cu/ZnSOD and carbonic anhydrase as a result of the ZE trait. However, at present we do not have a complete understanding of this association and more information is needed with regard to the precise role of the different components of biochemical Zn utilization. As more data become available, our understanding and perception of the mechanistic basis for ZE is surely going to require revision.

Genotypes with high Zn efficiency occur in many plant species but the genetics of this trait remains elusive. Certainly research on ZE will benefit from molecular and genetic investigations of ZE. The model plant Arabidopsis with its complete genome sequence, and other developing genomic resources including gene knockout lines and activation-tagged mutants, offer a number of possible strategies to dissect and examine ZE. For example, a scenario involving screening of activation-tagged Arabidopsis lines for gain of function mutants exhibiting increased ZE could set the stage for a molecular genetic analysis of this important trait. Furthermore, identification and characterization of the genes conferring ZE should provide important new insights into the molecular and physiological basis of crop plant ZE. Additionally, identification of the genes underlying ZE could provide the molecular tools to facilitate the engineering of more Zn-efficient crops. Another goal of related research could be to engineer food crops that accumulate elevated levels of Zn in the edible portions and thus, have higher nutritional value. This may not only lead to maximal plant and human nutrition with minimal fertilization, but also help to nourish the world population, which is expected to reach 10 billion by the year 2070. Ultimately, an interdisciplinary approach integrating molecular, genetic, physiological, and biochemical approaches holds much promise for improving both plant and human nutrition.

Acknowledgements

We apologize to all colleagues whose work we were unable to cite due to space limitations. We wish to thank Drs Jon Hart and Ross Welch of USDA-ARS, Cornell University, Ithaca, NY, USA, and Dr Ismail Cakmak of Sabanci University, Turkey, for valuable discussions and their insights into zinc efficiency; Dr Brian Alloway of International Zn Association, Belgium for providing Fig. 1; and Gabriela Treso for critical reading of the manuscript. The work of Gokhan Hacisalihoglu was funded in part by the Republic of Turkey.

Ancillary