Food Security in Nutrient-Stressed Environments: Exploiting Plantsí Genetic Capabilities (2002)

24. Genetic control of root exudation pp. 215-226
Z. Rengel
Soil Science and Plant Nurition, Faculty of Agriculture, The University of Western Ausralia, Nedlands WA 6907.

The literature on genetics of root exudation and on genotypic differences in qualitative and quantitative composition of root exudates in crop and native plant species was critically assessed. Differences in exadation have been reported for genotypes that differ in tolerance to nutrient deficiencies, ion toxicities, and pathogen attack. The exudation profile of a limited number of genotypes (frequently only two genotypes with the contrasting response to the environmental stress) have been reported to date. Little is known about the variability in larger samples of the germplasm or about actual genetics behind differential qualitative and quantitative composition of root exudates. Changing the exudation profile of a given genotype may be achieved by manipulating the biosynthetic capacity and by increasing the capacity of the plasma membrane to transport the specific compound out into the rhizosphere. Overexpression of the bacterial citrate synthase gene in the cytoplasm of tobacco plants resulted in exudation of large quantities of citrate into the rhizosphere and partial alleviation of the aluminium (Al) toxicity stress. A similar strategy of transforming plants with citrate synthase gene is being tried as a way of improving plant capacity to extract phosphorus (P) from soils with notoriously low P availability.

More research into the genetic basis of qualitative and quantitative differences in root exudation is warranted. Understanding the genetic control of root exudation, followed by manipulation of qualitative and quantitative composition of root exudates. will result in better adaptation of plants to environmental conditions and a greater yield of crops.

Understanding of the role that root exudates play in increasing plant adaptation to a given set of environmental conditions is sketchy at present. However, it is generally accepted that the type of exudates plants release into the rhizosphere has played a significant role in the distribution of plant species in various ecosystems. Calcicole plants exude mainly di- and tricarboxylic acids (with the former being good extractors of P and the latter good extractors of Fe and Mn from calcareous soils), while calcifuge plants exude mostly monocarboxylic acids (poor in mobilising P or Fe from calcareous soils) (Ström, 1997; Ström et al., 1994; Tyler and Ström, 1995). Similar distinctions also exist between crop plants (cf. Zhang et al., 1997).

Effective exudation of organic substances into the rhizosphere relies on at least three processes: 1. signalling sequence: 2. effective biosynthetic machinery producing a relatively large amount of potential exudate compounds, and 3, a membrane transporter that allows transfer of organic compounds into the rhizosphere.

Regulation of the complete exudation sequence and the underlying genetics is only in the initial stages of being unravelled at present. The two most popular strategies in increasing root exudation (overexpressing genes coding for critical enzymes in the biosynthesis of carboxylate anions, and engineering plants to increase exudation of phosphatase and phytase) (Delhaize, 19951 are promising approaches currently tested in several laboratories. In addition, genotypir differences among existing crop germplasm, including wild relatives should also he utilized (Cakmak et al., 1996a) and, particularly for cereals, rye as the cereal most tolerant to a wide range of nutrient deficiency and ion toxicity stresses (see Cakmak et al., 1997). Using genotypes from the local crop germplasm in the breeding programme would be advantageous because development of cultivars with tolerance to rhizosphere-related stresses could be expedited without severely disrupting the broad adaptation already selected and proved to work effectively, including regulation of relevant biochemical reactions involved in root exudation at an appropriately enhanced level.

This review will concentrate on the genetics underlying the capacity of plants to exude various organic compounds into the rhizosphere. More comprehensive treatises on other aspects dealing with root rxadates can be found elsewhere (e.g., Crowley and Rengel, 1999; Dinkelaker et al., 1995; 1997; Jones., 1998; Rengel, 1999; Uren and Reisenauer, 1988).

Role of exudates in resistance to ion toxicity

Tolerant genotypes of various plant species exude organic acid anions when challenged with Al toxicity. An Al-tolerant Phaseolus vulgaris genotype exuded 70-fold more citrate under Al toxicity than under control conditions and also 10 times more citrate than an Al-sensitive genotype grown with Al. In contrast. Al stress did not cause an increase in citrate release in the Al-sensitive genotype (Miyasaka et al., 1991). Under Al toxicity, an Al-tolerant maize genotype exudes citrate and relatively smaller amounts of malate in a dose-dependent manner (Jorge and Arruda, 19971. Under similar conditions, wheat genotypes tolerant to Al toxicity exude mostly malate (Basu et al., 1994; Delaize et al., 1993: Pellet et al., 1996: Ryan et al., 1995a, b). but sometimes succinate and oxalate as well (Christiansen-Weniger et al., 1992). Taro (Ma and Miyasaka, 1998) and buckwheat seedlings exude oxalic acid under Al toxicity (Zheng et al., 1998a, b), while radish, canola, and oats exude citric and malic acids (Zheng et al., 1998a). and Arabidopsis exudes either malate or citrate (Larsen et al., 1998).

Exudation of initiate by wheat (Basu et al., 1994; Delhaize et al., 1993: Ryan et al., 1995a, b) and citrate by maize roots (Pellet et al., 1995) is confined to the root tips. Calculations showed that the concentration of exuded organic acid anions at the root tip might he sufficiently high to complex and therefore detoxify Al ions around the root tips, especially if mucilage that surrounds the root tip. and increases the thickness of the unstirred layer at the root tip surface, is considered (Pellet et al., 1995; Ryan et al., 1995b). Therefore, exudation of a relatively small amount of carboxylate (organic acid) anions, concentrated in a small soil volume at the root tip, is an energetically effective means of overcoming inhibition of growth due to Al toxicity. Root tips would need to be protected for a relatively short period of time daring which root rip cells would mature and become less sensitive to Al (Basu et al., 1994), even though exudation of carboxylate anions may continue for the duration of the Al stress (cf. Zheng et al., 1998a).

The three examples where the genetics behind the enhanced exudation process is, at least, partially known are malate exudation in wheat, citrate exudation in transgenic tobacco and papaya, and malate. citrate, and/or pyruvate exudation by Arabidopsis mutant plants.

Malate exudation in wheat

Isogenic wheat lines segregating at the Alt1 locus differ in tolerance to Al: when challenged with toxic Al concentrations, the Al-tolerant isogenic line exudes large amounts of malate, while the Al-sensitive line does not (Delhaize et al., 1993; Ryan et al., 1995a). However, the two isogenic lines do not differ in concentrations of malate in their root tissue (Ryan et al., 1995a). indicating that the difference between them resides in the transport component of the exudation sequence. So, it appears likely that the Alt1 locus codes for a protein that represents a malate channel or a non-specific anion channel embedded in the plasma membrane, or a regulator of such channels. Given the fast reaction to Al exposure (Delhaize et al., 1993; Ryan et al., 1997), it is likely that the Alt1 locus is expressed constitutively. Electrophysiological studies have indicated that the Al-regulated plasma membrane anion channel is insensitive to La3+ and might be present in root tip cells, but not in more mature root cells (Ryan et al., 1997). These characteristics are in accordance with Al-induced malate exudation by intact root tips (Basu et al., 1994; Delhaize et al., 1993; Pellet et al., 1996; Ryan et al., 1995a). The potential difference in the presence and operation of the Al-regulated anion channel between root tip cells of Al-tolerant and Al-sensitive isogenic lines has yet to be tested. Similarly, the channel conductance was only measured for Cl- (Ryan et al., 1997); it has yet to be shown that the channel can facilitate transfer of carboxylate anions (malate in particular). Finally. a gene coding for a putative malate/anion channel in the wheat root plasma membrane has yet to be cloned, with studies of the potential difference in the genetic code responsible for either the presence or a lack of regulation by Al yielding a potential breakthrough in understanding of the membrane efflux and root exudation into the rhizosphere.

At present, the exact function of the gene products of the Alt1 locus is not certain (possibly a regulator of the plasma membrane malate-conductive channel allowing exudation of malate into the rhizosphere, or maybe the channel protein itself altered to become sensitive to Al). Despite the fact that the exact function of the gene is not known, it appears logical to suggest that transfer of that gene into Al-sensitive wheat genotypes may result in increased tolerance to Al, especially given the additive effects of Al tolerance genes (cf. Rengel, 1992). Such a transfer within the wheat germplasm can be done via traditional breeding techniques. In addition to such traditional techniques, molecular biology approaches can be used in transferring any genes associated with tolerance to Al from any plant or bacterial species into higher plant species. and especially crops. As reliable transformation systems are becoming common for all major crops, improvements in Al tolerance to crops by horizontal transfer of foreign genes can be expected. In case the crop genotype that is amenable to transformation is not the one that has desirable commercial characteristics, the transformed genotype will serve as a parent in the standard backcrossing programme aimed at transferring the Al-tolerance trait into commercially important breeding material.

Exudation of citrate by transgenic tobacco and papaya

The citrate synthase gene has been transferred from the bacterium Pseudomonas aeruginosa into tobacco and papaya and recorded an increased accumulation of citrate in the cytosol (de la Fuente et al., 1997). This accumulation of citrate was accompanied by increased citrate efflux into the rhizosphere and a somewhat increased tolerance of transformed plants to Al.

As in wheat, where exudation appears to be regulated by the membrane transport step (Ryan et al., 1995a), in transgenic tobacco plants transformed with the P. aeruginosa citrate synthase gene, saturation of the membrane transport systems regulated exudation of citrate into the rhizosphere because biosynthesis of citrate in transgenic lines increased up to 10-fold, but citrate exudation increased only up to 4-fold, compared to non-transformed parental lines (dc la Fuente et al., 1997). The signalling sequence that existed in the parental genotype was adequate for transferring the toxicity signal via the biosynthetic machinery to the membrane efflux transporter to close the sequence. However, the report by de la Fuente et al. (1997), with the amount of citrate synthesised in the cytosol of transgenic tobacco lines representing up to 86% of the fresh weight of roots, casts some doubt on the validity of the results reported. Moreover, increased citrate exudation by transgenic plants compared to the parental line offered increased protection from Al applied at concentrations of 75 to 200 mM, but not to Al applied at 50 mM (de la Fuente et al., 1997). This makes mechanistic explanations hard to come by.

Despite all the problems, the report by de la Fuente et al. (1997) showed in principle that transformation of plants with foreign genes to increase tolerance to Al may be possible. That report may therefore represent an important step toward more widespread transformation of crop plants with genes known to be associated with Al tolerance in other plant species. Transferring a single gene of bacterial citrate synthase (de la Fuente et al., 1997) resulted in a number of events. Increased biosynthesis of citrate in the cytosol had to be accompanied by adjustments in the rate of biosynthesis of precursors of citrate and products downstream from citrate in the biosynthetic pathways, and by adjustments in the mitochondria-cytosol traffic [218] of precursors and products. Citrate synthesis could occur in the mitochondria and the cytosol, but most of it is in mitochondria under physiological conditions. Functional regulation (and de novo biosynthesis?) of membrane transporters involved in transfer of citrate into the chizosphere was also important. Given that no prior knowledge existed about the possible capacity of tobacco and papaya to increase exudation of citrate or any other carboxylate anion under stress conditions (e.g., ion toxicity or nutrient deficiency), de la Fuente et al. (1997) took a crude shot-gun approach that appears to have worked amazingly well. However, transgenic tobacco plants created by de la Fuente et al. (1997) exuded more citrate than non-transformed parents regardless of the presence or absence of Al from the medium. Further work will need to concentrate on creating transgenic plants that will increase exudation of citrate only after exposure to Al toxicity, thus minimizing expenditure of carbon in adaptation to the environment.

Exudation of organic acid anions by Arabidopsis mutants

Arabidopsis mutants tolerant to Al appeared at the low frequency of 6 in 100,000. Five of these mutations mapped to a single locus on chromosome 1; these mutants exuded more malate, citrate, and/or pyruvate than the wild type (Larsen et al., 1998). Such an altered pattern of carboxylate anion exudation does not appear to be induced by Al [in contrast to wheat (Basu et al., 1994; Delhaize et al., 1993; Pellet et al., 1996; Ryan et al., 1995a, b), and maize (Pellet et al., 1995)]; rather, it is constitutively expressed. The case of Arabidopsis mutants, with malate, citrate and/or pyruvate (organic acid anions that are accumulated in the cytosol upon transfer of citrate from mitochondria) being exuded from roots, points to a model in which increased exudation of carboxylate (organic acid) anions may simply be a consequence of their accumulation in the cytosol.

The genetics of carboxylate anion exudation related to Al-toxicity stress appears simple, i.e., only slightly different from the normal genetics coding for compounds and processes existing in plants not subjected to stress. Therefore, any future attempts to transform plants with a small number of genes (even a single one) should have a relatively high chance of success. There are several research groups in the world working on transforming crop plants (e.g., wheat) with citrate synthase to increase biosynthesis and exudation of citrate. This is expected to increase tolerance to Al toxicity and P deficiency (citrate is effective not only in chelating Al into non-toxic forms, but also in solubilizing water-insoluble P complexes). Successful reports are expected to appear soon.

Role of exudates in response to nutrient deficiency

Breeding wheat and possibly other cereals for greater tolerance to micronutrient deficiency is possible by transferring rye genes controlling these tolerance traits. Tolerance to Fe, Zn and Mn deficiency traits is reliant, at least to some extent, on root exudation (Cakmak et al., 1994; Marschner and Römheld, 1994; Rengel, 1997, 1999; Rengel et al., 1998; Timonin, 1946). Individual genes conferring tolerance to nutrient deficiencies have not been deciphered yet, but transferring rye chromosomes 1R and 7R into wheat has increased its tolerance to Zn deficiency (Cakmak et al., 1997), and transferring chromosomes 2R and 7R increased tolerance to Mn and Fe deficiency in wheat (Schlegel et al., 1997). In contrast, there is no report linking tolerance to copper (Cu) deficiency with root exudation. However, transferring the rye chromosome 5R (Schlegel et al., 1997) or transferring just a small piece of chromatin from the long arm of the chromosome 5R to wheat increases its tolerance to Cu deficiency (Graham et al., 1987; Schlegel et al., 1993). It is known that genes for mugineic acid synthase and 3-hydroxymugineic acid synthase, the enzymes involved in biosynthesis of common phytosiderophores, are located on the rye chromosome 5R (Mon et al., 1990), but they map outside the region conferring Cu deficiency tolerance to the wheat-rye addition line (Schlegel et al., 1993). These findings are in accordance with the fact that phytosiderophores do not appear to be involved in uptake and transport of Cu (see Schlegel et al., 1993), but a possibility cannot be excluded that the function of the gene products coded by the genes conferring tolerance to Cu deficiency may in some way be connected with root exudation.

Iron deficiency

Generally, there are two types of mechanisms plants employ under Fe deficiency (for reviews see Crowley and Rengel, 1999; Jolley et al., 1996; Marschner and Römheld, 1994; Rengel, 1999). Only strategy II (employed by by grasses), which involves increased [219] exudation of phytosiderophores into the rhizosphere, will be considered here.

Graminaceous species acquire Fe by releasing phytosiderophores and taking up ferrated-phytosiderophore complexes through a specific uptake system that is strongly activated under Fe deficiency (Römheld. 1991: von Wiren et al., 1994. 1995). The rate of phytosiderophore release differs between crops and between genotypes of a particular crop. The rate of release of phytosiderophores is positively related to the tolerance of a species to Fe deficiency. The relative effectiveness in phytosiderophore release, and thus tolerance to Fe deficiency, decreases in the order: barley > maize > sorghum (Römheld and Marschner, 1990) or oat > maize (see Mori, 1994). Under Fe-deficiency conditions, the Fe-deficiency-tolerant genotypes of wheat (Hansen et al., 1995, 1996) and oats (Hansen and Jolley, 1995; Jolley and Brown, 1989) exuded more phytosiderophores than genotypes sensitive to Fe deficiency, making a phytosiderophore-exudation test suitable for screening genotypes in the breeding programme.

The release of phytosiderophores and the subsequent uptake of the Fe-phytosiderophore complex are under different genetic control (Römheld and Marschner. 1990), a was found that the Fe-deficiency-sensitive yellow-stripe mutant of maize (ysl) maintains a rate of phytosiderophore release similar to that of cultivars tolerant to Fe deficiency (e.g., Alice), with sensitivity of ys1 to Fe deficiency being the result of a defect in the uptake system for the Fe-phytosiderophore complex. Up to a 20-fold lower uptake rate in ys1 than Alice maize (von Wiren et al., 1994) is due to the mutation that affects a high-affinity uptake component, leading to a decrease in activity and/or a number of Fe-phytosiderophore transporters (decreased Imax, unchanged Km) (von Wiren et al., 1995).

A number of genes involved in biosynthesis of phytosiderophores and their precursor nicotianamine have been cloned (Mon. 1997), the most recent being a gene family coding for nicotianamine synthase (Higuchi et al., 1999). Nicotianamine synthase catalyses the polymerisation of three S-adenosylmethionine molecules, with the release of adenine and the azetidine ring formation. Since only strategy Il plants (graminaceous monocots) exude phytosiderophores under Fe and Zn deficiency (for references sec Rengel. 1999). it is interesting to note that nicotianamine synthase is regulated by Fe supply in these plants Higuchi et al., 1999). but not in other species (Higuchi et al., 1996), in which nicotianamine may only be a putative chelator involved in long-distance transport of divalent cations in plants (see von Viren et al., 1999; Welch, 1995).

Zinc deficiency

While exudation of phytosiderophores has primarily been associated with the response of graminaceous monocots to Fe deficiency (Cakmak et al., 1994: Jolley et al., 1996; Marschner and Römheld, 1994; Rengel, 1999; Römheld and Marschner, 1990) similar effect can he initiated as a response of wheat to Zn deficiency (Cakmak et al., 1994, 1996b; Rengel et al., 1999; Walter et al., 1994; Zhang et al., 1989). More phytosiderophores being exuded under Fe than under Zn deficiency (see Rengel et al., 1998; Walter et al., 1994) may indicate that the triggering process is more direct in the case of Fe deficiency, while a more complicated and/or slower sequence of events is necessary to trigger exudation of phytosiderophores under Zn deficiency. Moreover, wheat genotype considered tolerant to both Zn deficiency (Cakmak et al., 1994: 1996b; Hopkins et al., 1998; Walter et al., 1994; Rengel et al., 1998) and Fe deficiency (Hansen et al., 1995/1996) exude considerably larger amounts of phytosiderophores into the rooting medium than genotypes sensitive to Zn and Fe deficiencies, in order to increase mobilization of Zn and Fe from sparingly soluble sources.

There is usually a good relationship between root exudation of phytosiderophores and differential tolerance to Z deficiency when durum (sensitive) and bread wheats (tolerant) are compared (Cakmak et al., 1994, 1996b; Rengel, 1999: Rengel et al., 998). However, the positive correlation between phytosiderophore exudation and tolerance of bread wheat genotypes to Zn deficiency may not always hold, as shown for seven (Erenoglu et al., 1996) and five bread wheat genotypes (Cakmak et al., 1998).

Phosphorus deficiency

Under P deficiency, plants exude a wide range of organic and inorganic compounds to increase mobilization of P from sparingly soluble sources (for references see Rengel. 1999). Similar amounts of exudates were collected from roots with or without arbuscular mycorrhizae (AM) (Araizeh et al., 1995). indicating that an extra carbon cost of supporting AM colonization did not adversely affect the capacity of plants to partition a portion of assimilates to root exudation.

Species and genotypes within species that differ in tolerance to P deficiency also differ in the solubilizing activity of their root exudates (e.g., for acid phosphatase see Caradus, 1995; for unspecified Fe-P solubilizing activity of root exudates see Subbarao et al., 1997). A comparison of durum and flax genotypes of unknown levels of tolerance to P deficiency has shown large differences in the amounts of organic acids released into the rhizosphere (Cieslinski et al., 1997).

The activity of both phytase and acid phosphatase increased in root exudates from a number of species and genotypes within species grown under low-P supply (e.g., Ascencio, 1997; Asmat, 1997; Li et al., 1997). However, kinetic properties of enzymes from different species are different (Ascencio, 1997), indicating the importance of optimizing enzyme properties, rather than just increasing enzyme synthesis and exudation in any future attempt to manipulate phosphatase activity to increase adaptation of genotypes to low-P environments.

Acid phosphatase is exuded from roots under P deficiency, with the relative rate of increase in exudation greatly varying among various plant species (Tadano et al., 1993). The P-deficiency-tolerant genotypes of bean (Helal, 1990) and barley (Asmar et al., 1995) had a greater activity of extracellular phosphatases in the rhizosphere soil than genotypes sensitive to P deficiency. In rice, genotypes differed in acid phosphatase exudation under P deficiency (Ni et al., 1996).

There is a wide variability in both the absolute amounts and the type of organic acid (carboxylate) anions exuded by roots of Australian native plants [various species from cluster-root-forming (Lamont, 1982; Skene, 1998) genera from the family Proteaceae: Hakea, Banksia, Dryandra, etc.] adapted to low-fertility soils where deficiency of available P and micronutrients is common (Dinkelaker et al., 1995, 1997; Grierson, 1992; Roelofs, Rengel, Lambers and Kuo, unpublished results). L-malate, malonate, lactate, acetate, citrate, succinate, and trans-aconitate accounted for the largest proportions of carboxylate anions exuded by entire root systems of most of the species tested. In contrast, for cluster roots of all the species tested, L-malate, malonate, lactate, citrate, and trans-aconitate represented large proportions of total carboxylate anion exudation (Roelofs, Rengel, Lambers and Kuo, unpublished results). Qualitative and quantitative differences in carboxylate anion exudation among Proteaceae species may correspond with the distribution of these species in the south-west Australia, and may contribute to the species richness in the kwongan vegetation (cf. Cowling and Lamont, 1998). The hypothesis is that species use P from various soil pools and thus, instead of competing among themselves, complement each other and coexist in the same ecosystem.


The knowledge on the genetics underlying root exudation of organic compounds is expanding rapidly. Such knowledge will allow selection of superior genotypes and deliberate breeding of crop plants for increased tolerance to nutrient deficiency and ion toxicity, efficiency of nodulation, tolerance to various diseases, etc. Traditional breeding techniques will be combined with transformation of plants with genes coding for specific enzymes in the biosynthesis of organic compounds and/or those involved in regulation of efflux of organic compounds across the membrane. All these developments will allow better adaptation of crop genotypes to environment and more sustainable and profitable farming in the future.



Mineral Uptake